Journal of Tropical Pediatrics

2018 Journal of Tropical Pediatrics

doi: 10.1093/tropej/fmy051

Childhood anaemia in Malawian Children Childhood anaemia remains a significant public health problem, with nearly 300 million children under the age of 5 years being anaemic globally, 28.5% of which living in sub-Saharan Africa (SSA). Malawi has one of the highest prevalence rates of anaemia across nations, with studies reporting that 63% of children aged 6-<59 months in this country are anaemic and up to 3% severely anaemic. Recognizing the multifactorial aetiology of anaemia, Peter Austin Morton Ntenda and colleagues, from the School of Public Health, Taipei Medical University, in Taiwan, aimed to explore whether individual level and community-level factors both could influence childhood anaemia, severe anaemia, and haemoglobin (Hb) concentration among a cohort of 2597 Malawian children. Authors report that anaemia was more frequent among younger ages, malnourished children and those with a recent history of fever. Individual-level factors appeared to have stronger effects than community-level factors on childhood anemia, severe anemia and Hb concentration in Malawi, although community-level female education seemed to play also an important role in childhood anemia. They conclude that public health interventions targeting childhood anaemia should focus on food insecurity, malaria infection, iron and other micronutrient deficiencies, household living standards, as well as community female education. See page 267 Predictors of Poor Outcome in Neonates with bacterial Meningitis in India Acute bacterial meningitis (ABM) is a devastating disease, particularly in the neonatal period. Mortality rates from developed countries range from 3% to 13%, compared with 30–40% in developing countries, and up to 20–30% survivors will develop neurological sequelae. In this issue of the Journal of Tropical Pediatrics, Mala Kumar et al, from King George’s Medical University, Lucknow, India, aimed to characterize outcomes among ABM patients and describe predictors of poor outcome. Eleven percent of the ABM neonatal cases died, and 27% were discharged with an abnormal neurological examination (NE). Risk factors for a poor outcome included prolonged shock, coma, seizures, requiring mechanical ventilation or orogastric feeding and electroencephalography (EEG) abnormalities. Abnormal NE, EEG or brainstem-evoked reflex audiometry at discharge predicted neurodevelopmental delays at three months. Authors conclude that mortality and morbidity of neonates with ABM were similar to that in developed countries. Outcome depended on severity of the disease on admission and NE at discharge. See page 297 Intussusception rate in young children before introduction of rotavirus vaccine in North India The last two decades have witnessed a significant effort to develop a safe and effective rotavirus vaccine to prevent the significant morbidity and mortality associated with rotavirus infection, particularly in developing countries. Historic safety concerns related to an increased risk of intussusception in relation to rotavirus vaccines have triggered particular vigilance in relation to this rare but life-threatening adverse event. Madhu Gupta and colleagues, from the Post Graduate Institute of Medical Education and Research, in Chandigarh, India, developed an intussusception surveillance system, so as to define baseline data on intussusception rates prior to the introduction in Northern India of this vaccine. Authors estimated a baseline annual incidence rate of 20/100 000 in infants, and 5/100 000 in children <5 years of age, in Chandigarh. Maintenance of such a surveillance system will allow to monitor changes in incidence following rotavirus vaccine introduction. See page 326 © The Author(s) [2018]. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

2018 Journal of Tropical Pediatrics

doi: 10.1093/tropej/fmx056pmid: 28985403

Abstract This study aimed to assess the maternal anthropometric parameters of human immunodeficiency virus (HIV)-infected and HIV-uninfected mothers as well as to assess the neonatal anthropometric parameters of premature infants in relation to maternal anthropometric parameters (weight, height and mid-upper-arm circumference), HIV status and anti-retroviral therapy (ART) regimen. Study participants included HIV-infected and HIV-uninfected mothers who gave birth to premature infants. All HIV-infected mothers received ART. The incidence of intra-uterine growth restriction (IUGR) among premature infants was high. Maternal anthropometric parameters, HIV status and ART exposure showed no association with IUGR in this study. Sufficient maternal ART exposure may positively influence head circumference at birth, which might determine the neurodevelopmental outcome of these infants. anthropometry, maternal anthropometry, HIV, premature infants, preterm, infectious diseases INTRODUCTION A premature infant is defined as an infant born at <37 weeks’ gestation [1]. Fifteen million premature births occur worldwide of which 60% of these deliveries occur in low- and middle-income countries [2]. Premature delivery is considered a significant global perinatal health problem with a rising incidence in Southern Africa [3]. In a meta-analysis by Beck and colleagues, the worldwide prevalence of premature birth was 9.6%, and 85% of these preterm births were from the African and Asian continents [3]. Evidence from developing countries suggests that human immunodeficiency virus (HIV)-infected mothers have an increased risk of giving birth to premature infants, and the occurrence is directly associated with the clinical stage of the disease [4–6]. Research shows that HIV-infected women were more likely to give birth to low birth weight (LBW) and intra-uterine growth restricted premature infants [2, 5, 6]. The role of anti-retroviral therapy (ART) in intra-uterine growth restriction (IUGR) is uncertain. Studies conducted in the developed world showed no association between in utero exposure to ART and IUGR [5, 7–9]; however, results obtained from the developing world are conflicting [9–12]. Maternal anthropometric parameters during pregnancy are key factors in determining foetal growth and anthropometrical parameters at birth. In South Africa, poor household food security and the effects of HIV/AIDS Wasting Syndrome, or the combination thereof, may contribute to maternal undernutrition [13]. In the undernourished mother, the supply of maternal foetal nutrients are compromised, thereby restricting foetal growth. Foetal growth is more affected by chronic maternal undernutrition than nutrient restriction during the pregnancy period [14]. In addition to undernutrition, the incidence of overweight and obesity is rising dramatically in South Africa [11]. In a small subset of people, maternal overnutrition may increase the risk of delivering a premature IUGR infant, though the exact mechanisms are poorly understood [15]. An observational study found that obese women had a higher rate of IUGR deliveries as well as an increased frequency of admissions to the neonatal intensive care unit [16]. Maternal overnutrition impedes placental growth and ultimately leads to growth faltering, which significantly increases neonatal morbidity and mortality [17]. Highly active antiretroviral therapy use is often associated with overnutrition and metabolic and endocrine abnormalities [18, 19]. South Africa is a country where the incidence of HIV infection and premature births is high and therefore the need existed to [4] to assess the maternal anthropometric parameters of HIV-infected and HIV-uninfected mothers as well as to assess the neonatal anthropometric parameters of premature infants in relation to maternal anthropometric parameters [weight, height and mid-upper-arm circumference (MUAC)], HIV status and ART regimen. METHODOLOGY A cross-sectional study with an analytical component was conducted in the postnatal wards of Kalafong Hospital, Gauteng, South Africa from August 2014 to April 2015. Consecutive sampling was used to recruit participants. The study population included postnatal lactating HIV-infected mothers receiving ART and HIV-uninfected mothers who gave birth to premature infants. All mothers received accommodation in the ward as per standard protocol. The sample size was estimated by a statistician based on data of live births from the institution. The sample size was calculated by a one-way ANOVA calculation to obtain an effect size root mean square error = 0.55. Consecutive consenting postnatal mother–premature infant pairs were included in the study. The study population consisted of HIV-infected and HIV-uninfected mothers who gave birth to premature infants. All HIV-infected mothers received ART. An HIV positive status was established by a routine confirmed positive rapid HIV test during the prenatal period, and an HIV negative status was confirmed with a routine enzyme-linked immunosorbent assay (ELISA) test in the postnatal period. Gestational age was estimated by using the Ballard score, as early sonars are rarely performed. Standard means of medical treatment was continued during the study period. A researcher-administered questionnaire, tested for face and content validity during a pilot project, was used to obtain demographic and clinical information from the mother. Information regarding HIV status, ART regimen and CD4 count was collected. Maternal anthropometric information was obtained on Day 7 postnatally. This included weight, height and the MUAC. Maternal undernutrition was classified as a body mass index (BMI) < 20.3 [20] and/or MUAC < 214 mm [21] and maternal overnutrition was defined as a BMI > 25 kg/m2 [20]. Birth infantile anthropometric information was obtained from the patient folder. This included birth weight, length and head circumference (HC). A trained dietitian repeated the length and HC measurements on Day 7 postnatally according to standardized procedures [20]. IUGR was defined as a weight, length or HC <10th percentile [1]. Symmetrical IUGR (s-IUGR) is defined as a weight, length and HC <10th percentile and a ponderal index (PI) ≥2.0 and asymmetrical IUGR (a-IUGR) as weight <10th percentile with the sparing of length and HC and a PI <2.0 [22, 23]. Medical information was noted from the hospital file. Data captured on Microsoft Excel 2013® was exported to Statistica version 12 [StatSoft Inc. (2015) STATISTICA (data analysis software system), www.statsoft.com] for data analysis. Statistical analysis was done with the assistance of Tygerberg Biostatistics Unit, Stellenbosch University. Descriptive statistics were used to report the demographics of participants. When groups were compared, ANOVA and t-tests were used. Statistical significance was defined as p < 0.05. The study was granted ethics approval from the human research ethics committee of Stellenbosch University (S13/09/165) and the research ethics committee of the Faculty of Health Sciences at the University of Pretoria (191/2014). The study was conducted in accordance with Good Clinical Practice Guidelines. RESULTS A total of 90 mother–infant pairs were screened for inclusion in the investigation. Sixteen pairs were excluded of which 12 mothers were discharged before data collection procedures were completed, one mother was not on ART at the time of data collection, two mothers were excluded owing to an ELISA test not performed and one mother declined consent. The final sample consisted out of 74 mother–infant pairs. Thirty-eight (51%) mothers were HIV-infected and 36 (49%) were HIV-uninfected. All HIV-infected mothers received a fixed dose combination ART, which consists of tenofovir, emtricitabine and efavirenz [24]. The mean ART treatment period was 8.9 months (SD ± 15.4; 1 day to 7 years). In this group, the mean CD4 was 367 cells/mm3 (SD ± 165; CI = 51–720) cells/mm3. There were no significant differences in the demographic characteristics of mothers or infants. The mean gestational age of HIV-exposed infants was 31.8 weeks (SD ± 3.2) and of HIV-unexposed was 32.4 weeks (SD ± 2.86). The mean birthweight was 1468 g (SD ± 458.8 g), length 40.7 cm (SD ± 5.0 cm) and HC 28.7 cm (SD ± 3.1 cm). The demographic characteristics are depicted in Table 1. Table 1 Demographic characteristics of the sample Demographics Mothers HIV-infected HIV-uninfected p-value* n=38 (51%) n=36 (49%) Ethnicity p=0.3 African 38 (100%) 35 (97%) Caucasian 0 (0%) 1 (3%) Age in years mean (SD) 29 (5.57) 27 (5.11) p=0.18 Education level (%) p=0.76 No formal education 7 (19%) 7 (18%) >Grade 10 12 (33%) 8 (21%) ≥Grade 12 11 (31%) 14 (3%7) Tertiary level 8 (22%) 7 (18%) Demographics Mothers HIV-infected HIV-uninfected p-value* n=38 (51%) n=36 (49%) Ethnicity p=0.3 African 38 (100%) 35 (97%) Caucasian 0 (0%) 1 (3%) Age in years mean (SD) 29 (5.57) 27 (5.11) p=0.18 Education level (%) p=0.76 No formal education 7 (19%) 7 (18%) >Grade 10 12 (33%) 8 (21%) ≥Grade 12 11 (31%) 14 (3%7) Tertiary level 8 (22%) 7 (18%) Significance defined as *p = 0.05. Table 1 Demographic characteristics of the sample Demographics Mothers HIV-infected HIV-uninfected p-value* n=38 (51%) n=36 (49%) Ethnicity p=0.3 African 38 (100%) 35 (97%) Caucasian 0 (0%) 1 (3%) Age in years mean (SD) 29 (5.57) 27 (5.11) p=0.18 Education level (%) p=0.76 No formal education 7 (19%) 7 (18%) >Grade 10 12 (33%) 8 (21%) ≥Grade 12 11 (31%) 14 (3%7) Tertiary level 8 (22%) 7 (18%) Demographics Mothers HIV-infected HIV-uninfected p-value* n=38 (51%) n=36 (49%) Ethnicity p=0.3 African 38 (100%) 35 (97%) Caucasian 0 (0%) 1 (3%) Age in years mean (SD) 29 (5.57) 27 (5.11) p=0.18 Education level (%) p=0.76 No formal education 7 (19%) 7 (18%) >Grade 10 12 (33%) 8 (21%) ≥Grade 12 11 (31%) 14 (3%7) Tertiary level 8 (22%) 7 (18%) Significance defined as *p = 0.05. The mean maternal BMI was 26.7 kg/m2 (SD ± 4.88 kg/m2, CI = 25.7–27.9) and the mean MUAC was 289 mm (SD ± 3.93 mm; CI = 28.0–29.8). Under- and overnutrition prevalence was low with two (3%) mothers classified as underweight and 14 (19%) as obese. Twenty-seven (36%) mothers’ anthropometric parameters fell within normal ranges and 31 (42%) mothers’ anthropometric parameters were classified as overweight. The mean BMI and MUAC did not differ between HIV-infected and -uninfected mothers (p = 0.89 and p = 0.71, respectively). Furthermore, ART exposure duration had no significant effect on maternal anthropometrical parameters (Figure 1). Fig. 1. View largeDownload slide Maternal anthropometric data, HIV status and ART exposure period. Maternal BMI (A) and MUAC (B). HIV: human immunodeficiency virus, ART: antiretroviral Therapy, IUGR: intrauterine growth restriction, s-IUGR: symmetrical-IUGR, a-IUGR: asymmetrical IUGR. *p = 0.05. Fig. 1. View largeDownload slide Maternal anthropometric data, HIV status and ART exposure period. Maternal BMI (A) and MUAC (B). HIV: human immunodeficiency virus, ART: antiretroviral Therapy, IUGR: intrauterine growth restriction, s-IUGR: symmetrical-IUGR, a-IUGR: asymmetrical IUGR. *p = 0.05. Fig. 2. View largeDownload slide Anthropometric parameters <10 percentile of HIV-exposed premature infants according to maternal ART period. ART: antiretroviral therapy, HC: head circumference, **p < 0.01. Fig. 2. View largeDownload slide Anthropometric parameters <10 percentile of HIV-exposed premature infants according to maternal ART period. ART: antiretroviral therapy, HC: head circumference, **p < 0.01. IUGR was present in 40 (54%) of the premature infants. Of these, 13 (33%) were s-IUGR (PI ≥ 2.0) and 27 (67%) were a-IUGR (PI < 2.0), indicating wasting at birth. Similar incidence of IUGR was found in infants born to HIV-infected and HIV-uninfected mothers. Furthermore, no significant differences (p = 1.00) existed between infants with s-IUGR and a-IUGR according to maternal HIV status. Maternal anthropometric classifications (p = 0.79) and maternal HIV-infection and -ART regimen (p = 0.82) had no effect on the prevalence of IUGR (Table 2). No differences existed for birthweight (p = 0.18) and length (p = 0.15) or HC (p = 0.27) between infants born to HIV-infected and HIV-uninfected mothers. HIV exposure did not influence birthweight (p = 0.18), length (p = 0.15) or HC (p = 0.27). Maternal ART exposure time had no effect on infantile weights and lengths (p = 0.764 and p = 0.647, respectively); however, infants of mothers who received ART for >20 weeks showed significantly less restrictions related to their HC measurements (p = 0.003). HIV exposed infants whose anthropometrics fell <10th percentile are depicted in Figure 2. Maternal ART exposure had no effect on birthweight categories (Figure 3). Similarly, maternal CD4 cell counts had no effect on the incidence of IUGR. Table 2 Neonatal anthropometric parameters according to maternal HIV status, treatment regimen and anthropometric parameters Maternal characteristics IUGR (n=40) Non-IUGR (n=34) p-value* n (%) n (%) HIV status p=0.72 HIV-infected on ART 20 (50) 18 (53)  ART <4 weeks 5 (25) 2 (11)  ART 4–20 weeks 7 (35) 6 (33)  ART >20 weeks 8 (40) 10 (56) HIV-uninfected 20 (50) 16 (47) Maternal nutritional status p=0.79 Undernourished 1 (1) 1 (1) Normal 16 (22) 11 (15) Overweight 16 (22) 15 (20) Obese 7 (9) 7 (9) Maternal characteristics IUGR (n=40) Non-IUGR (n=34) p-value* n (%) n (%) HIV status p=0.72 HIV-infected on ART 20 (50) 18 (53)  ART <4 weeks 5 (25) 2 (11)  ART 4–20 weeks 7 (35) 6 (33)  ART >20 weeks 8 (40) 10 (56) HIV-uninfected 20 (50) 16 (47) Maternal nutritional status p=0.79 Undernourished 1 (1) 1 (1) Normal 16 (22) 11 (15) Overweight 16 (22) 15 (20) Obese 7 (9) 7 (9) ART: antiretroviral therapy, HIV: human immunodeficiency virus, IUGR: intra-uterine growth restriction; *p = 0.05. Table 2 Neonatal anthropometric parameters according to maternal HIV status, treatment regimen and anthropometric parameters Maternal characteristics IUGR (n=40) Non-IUGR (n=34) p-value* n (%) n (%) HIV status p=0.72 HIV-infected on ART 20 (50) 18 (53)  ART <4 weeks 5 (25) 2 (11)  ART 4–20 weeks 7 (35) 6 (33)  ART >20 weeks 8 (40) 10 (56) HIV-uninfected 20 (50) 16 (47) Maternal nutritional status p=0.79 Undernourished 1 (1) 1 (1) Normal 16 (22) 11 (15) Overweight 16 (22) 15 (20) Obese 7 (9) 7 (9) Maternal characteristics IUGR (n=40) Non-IUGR (n=34) p-value* n (%) n (%) HIV status p=0.72 HIV-infected on ART 20 (50) 18 (53)  ART <4 weeks 5 (25) 2 (11)  ART 4–20 weeks 7 (35) 6 (33)  ART >20 weeks 8 (40) 10 (56) HIV-uninfected 20 (50) 16 (47) Maternal nutritional status p=0.79 Undernourished 1 (1) 1 (1) Normal 16 (22) 11 (15) Overweight 16 (22) 15 (20) Obese 7 (9) 7 (9) ART: antiretroviral therapy, HIV: human immunodeficiency virus, IUGR: intra-uterine growth restriction; *p = 0.05. Fig 3. View largeDownload slide Birthweight according to maternal ART exposure time and HIV-uninfected mothers. ART: antiretroviral Therapy, LBW: low birth weight, VLBW: very low birth weight, ELBW: extremely low birth weight, p = 0.63. Fig 3. View largeDownload slide Birthweight according to maternal ART exposure time and HIV-uninfected mothers. ART: antiretroviral Therapy, LBW: low birth weight, VLBW: very low birth weight, ELBW: extremely low birth weight, p = 0.63. DISCUSSION Early recognition of HIV infection is a key strategy in decreasing mother to child transmission. It is evident that the current Prevention of Mother to Child Transmission of HIV programme is effective when comparing the mean ART treatment period and mean gestation period. Our results did not show an association between maternal HIV-infection and adverse pregnancy outcomes, such as premature birth and IUGR. This is conflicting to numerous studies from Sub-Saharan Africa that reported significant results [4–6]. The short- and long-term effects of ART on in utero development are unknown, particularly in developing countries. While some studies found that maternal antiretroviral drugs compromised neonatal anthropometric classification [5, 7–9], others, including ours, did not [9–12]. Anthropometrics (birthweight, length and HC) and birthweight categories were not influenced by maternal HIV-status or ART use. While ART may increase the risk for LBW [5, 7–9], the importance of maternal ART provision is imperative to prevent Maternal to Child Transmission of HIV. However, other strategies to prevent LBW should be optimized, as LBW infants are at higher risk of mortality compared with a term appropriate-for-gestation infant [25]. Mothers had a mean BMI of 26.7 kg/m2, which suggests a risk for overweight among childbearing women. The prevalence of IUGR was high (53%) in this group. The combination of mothers at risk of overweight and infants being born with IUGR have important implications for HIV-infected and HIV-uninfected women of reproductive age, as it indicates possible metabolic disturbances for them and their offspring. The role of ART may further disturb clinical anthropometric and metabolic parameters. ART is associated with metabolic aberrations, including central obesity, dyslipidaemia and insulin resistance [18]. Although maternal anthropometrics were not influenced by ART exposure time, the possibility of pre-pregnancy or future morphological and/or metabolic effects should not be excluded. Epidemiological studies and animal models associate LBW with risk of adult obesity and metabolic syndrome. Maternal overnutrition, whether from obesity, high energy and fat diets or excessive weight gain in pregnancy, has delivered varied results concerning birth weight. However, in the adult offspring, obesity and metabolic abnormalities are highly prevalent, indicating evidence of metabolic programming [26–29]. The intrauterine exposure to endocrine disrupting chemicals (or obesogens), present in overweight mothers, alter the developmental programming of adipogenesis of the foetus through gene expression. The developing foetus responds by producing structural and functional changes in tissues and organ systems, known as foetal programming, which result in increased plasticity of adipocytes [30]. A third (32%) of infants with IUGR presented with low HC measurements. Subanalyses demonstrated a higher incidence of HC restriction among women with ART exposure <20 weeks and insufficient ART exposure. This holds important implications, as an impaired HC at birth is a strong indicator of impaired neurodevelopmental outcomes in childhood [31]. Studies that explored prenatal exposure to ART did not find differences in infant neurodevelopmental outcomes [32, 33]. The results indicate the possibility that increased ART exposure might protect infants against impaired brain development, as seen in studies on longer ART duration and association with reduction of some neurologic impairment in children [34]. CONCLUSION The incidence of IUGR among premature infants was high. Maternal anthropometric parameters, HIV status and ART exposure showed no association with IUGR, nor the type of IUGR in this study. Sufficient maternal ART exposure may positively influence HC at birth, which might determine the neurodevelopmental outcome of these infants. ACKNOWLEDGEMENTS Gratitude is extended to the Department of Paediatrics at Kalafong Hospital as well as the team that assisted with data collection for the support and assistance in the research project. The Tygerberg Biostatistics Unit of Stellenbosch University is acknowledged for assistance with the statistical analyses. FUNDING This work was supported by the Harry Crossley Foundation. REFERENCES 1 Goldenberg RL , Culhane JF , Iams JD , et al. Epidemiology and causes of preterm birth . Lancet 2008 ; 371 : 75 – 84 . Google Scholar CrossRef Search ADS PubMed 2 Vogel JP , Lee AC , Souza JP. Maternal morbidity and preterm birth in 22 low- and middle income countries: a secondary analysis of the WHO global survey dataset . BMC Pregnancy Childbirth 2014 ; 14 : 56 – 70 . Google Scholar CrossRef Search ADS PubMed 3 Beck S , Wojdyla D , Say L , et al. Worldwide incidence of premature birth: a systematic review of maternal mortality and morbidity . Bull World Health Org 2010 ; 88 : 31 – 8 . Google Scholar CrossRef Search ADS PubMed 4 Rollins N , Coovadia H , Bland RM , et al. Pregnancy outcomes in HIV-infected and –uninfected women in rural and urban South Africa . J Acquir Immune Defic Syndr 2007 ; 44 : 321 – 8 . Google Scholar CrossRef Search ADS PubMed 5 Suy A , Martinez E , Coll O , et al. Increased risk of pre-eclampsia and fetal death in HIV-infected pregnant women receiving highly active anti-retroviral theray . AIDS 2006 ; 20 : 59 – 66 . Google Scholar CrossRef Search ADS PubMed 6 Stiehm ER. Newborn factors in maternal-infant transmission of paediatric HIV infection . J Nutr 1996 ; 126(Suppl 10) : 2632S – 6S . Google Scholar CrossRef Search ADS 7 Powis KM , Smeaton L , Ogwu A , et al. Effects of in utero antiretroviral exposure on longitudinal growth of HIV-exposed infants in Botswana . J Acquir Immune Def Syndr 2011 ; 56 : 131 – 8 . Google Scholar CrossRef Search ADS 8 Briand N , Mandelbrot L , Le Chenadec J , et al. No relation between in-utero exposure to HAART and intrauterine growth retardation . AIDS 2009 ; 23 : 1235 – 43 . Google Scholar CrossRef Search ADS PubMed 9 Lambert JS , Watts DH , Mofenson N , et al. Risk factors for preterm birth, low birth weight, and intra-uterine growth retardation in infants born to HIV Infected pregnant women receiving ziduvidine . AIDS 2000 ; 14 : 1398 – 9 . Google Scholar CrossRef Search ADS 10 El Beitune P , Duarte G , Machado AA , et al. Effect of antiretroviral drugs on maternal CD4 lymphocyte counts, HIV-1 RNA levels and anthropometric parameters of their neonates . Clinics 2005 ; 50 : 207 – 12 . Google Scholar CrossRef Search ADS 11 Ryder RW , Temmerman M. The effects of HIV-1 infection during pregnancy and the perinatal period on maternal and infant health in Africa . AIDS 1991 ; 5(Suppl 1) : S75 – 85 . Google Scholar PubMed 12 Temmerman M , Chomba EM , Poit P. HIV-1 and reproductive health in Africa . Int J Gynecol Obstet 1994 ; 44 : 107 – 11 . Google Scholar CrossRef Search ADS 13 Kimani-Murage EW. Exploring the paradox: double burden of malnutrition in rural South Africa . Glob Health Action 2013 ; 6 : 193 – 205 . 14 Thureen PJ , Anderson MS , Hay WW. The small-for-gestational-age infant . Neoreviews 2001 ; 2 : 139 – 149 . Google Scholar CrossRef Search ADS 15 Brett KE , Ferraro ZM , Yockell-Lelievre J , et al. Maternal-fetal nutrient transport in pregnancy pathologies: the role of the placenta . Int J Mol Sci 2014 ; 15 : 16153 – 85 . Google Scholar CrossRef Search ADS PubMed 16 Radulescu L , Munteanu O , Popa F , et al. The implications and consequences of maternal obsesity on fetal intrauterine growth restriction . J Med Life 2013 ; 6 : 292 – 8 . Google Scholar PubMed 17 Wu G , Brazer FW , Cudd TA , et al. Maternal nutrition and fetal development . J Nutr 2004 ; 134 : 2169 – 72 . Google Scholar CrossRef Search ADS PubMed 18 Anuurad E , Semrad A , Berglund L. Human immunodeficiency virus and highly active antiretroviral therapy-associated metabolic disorders and risk factors for cardiovascular disease. Metab . Syndr Relat Disord 2009 ; 7 : 401 – 10 . Google Scholar CrossRef Search ADS 19 Hochter B , Slowinski T , Bauer C , et al. The advanced fetal programming hypothesis . Nephrol Dial Transplant 2001 ; 16 : 1298 – 9 . Google Scholar CrossRef Search ADS PubMed 20 WHO . 1995 . Infant and young child. Exclusive breastfeeding. Department of child and adolescent health and development, nutrition. http://www.who.int/child-adolescent-health/NUTRITION/infant_exclusive.html (22 December 2015, date last accessed). 21 Classification of Acute Malnutrition in Adults . http://www.foodsecurity.nrc-handbooks.org/assets/nutrition-assessment-table-6.pdf (July 2015, date last accessed). 22 Rosenberg A. The IUGR newborn . Semin Perinatol 2008 ; 32 : 219 – 24 . Google Scholar CrossRef Search ADS PubMed 23 Akram DS , Arif F. Ponderal index of low birth weight babies- a hospital-based study . J Pak Med Assoc 2005 ; 55 : 229 – 31 . Google Scholar PubMed 24 National Department of Health, South Africa . Fixed-dose combination for adults accessing antiretroviral therapy. http://www.sahivsoc.org/upload/documents/Fixed%20dose%20combination%20for%20adults%20accessing%20antiretroviral%20therapy.pdf (June 2015, date last accessed). 25 Wei R , Msamanga GI , Spiegelman D , et al. Associations between low birth weight and infant mortality in children born to human immunodeficiency virus-1 infected mothers in Tanzania . Pediatr Infect Dis J 2004 ; 23 : 530 – 5 . Google Scholar CrossRef Search ADS PubMed 26 Desai M , Beall M , Ross MG. Developmental origins of obesity: programmed adipogenesis . Curr Diab Rep 2013 ; 1 : 29 – 33 . 27 Howie GJ , Sloboda DM , Kamal T , et al. Maternal nutritional history predicts obesity in adult offspring independent of postnatal diet . J Physiol 2009 ; 587 : 905 – 15 . Google Scholar CrossRef Search ADS PubMed 28 Ferezou-Viala J , Roy AF , Serougne C , et al. Long-term consequences of maternal high-fat feeding on hypothalamic leptin sensitivity and diet-induced obesity in the offspring . Am J Physiol Regul Integr Comp Physiol 2007 ; 293 : R1056 – 62 . Google Scholar CrossRef Search ADS PubMed 29 Samuelsson AM , Matthews PA , Argenton M , et al. Diet-induced obesity in female mice leads to offspring hyperphagia, adiposity, hypertension, and insulin resistance: a novel murine model of developmental programming . Hypertension 2008 ; 512 : 383 – 92 . Google Scholar CrossRef Search ADS 30 Janesick A , Blumberg B. Endocrine disrupting chemicals and the developmental programming of adipogenesis and obesity . Birth Defects Res C Embryo Today 2011 ; 93 : 34 – 50 . Google Scholar CrossRef Search ADS PubMed 31 Neubauer V , Griesmaier E , Pehbock-Walser N , et al. Poor postnatal head growth in very preterm infants is associated with impaired neurodevelopment outcome . Acta Paediatr 2013 ; 102 : 883 – 8 . Google Scholar CrossRef Search ADS PubMed 32 Alimenti A , Forbes JC , Oberlander TF , et al. A prospective controlled study of neurodevelopment in hiv-uninfected children exposed to combination antiretroviral drugs in pregnancy . Pediatrics 2006 ; 118 : e1139 – 45 . Google Scholar CrossRef Search ADS PubMed 33 Lindsey JC , Malee KM , Brouwers P , et al. Neurodevelopmental functioning in hiv-infected infants and young children before and after the introduction of protease inhibitor-based highly active antiretroviral therapy . Pediatrics 2007 ; 119 : e681 – 93 . Google Scholar CrossRef Search ADS PubMed 34 Brahmbhatt H , Boivin M , Ssempijja V , et al. Neurodevelopmental benefits of antiretroviral therapy in Ugandan children aged 0-6 years with HIV . J Acquir Immun Defic Syndr 2014 ; 67 : 316 – 22 . Google Scholar CrossRef Search ADS © The Author [2017]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

2018 Journal of Tropical Pediatrics

doi: 10.1093/tropej/fmx057pmid: 28977653

Abstract Objective The objective of this study was to compare efficacy of continuous positive airway pressure (CPAP) and heated humidified high-flow nasal cannula (HHHFNC) as noninvasive respiratory support in post-extubation period in very low birth weight (VLBW) infants. Method This retrospective study enrolled 136 neonates, ≤32 weeks gestation and ≤1500 grams birth weight, requiring noninvasive respiratory support during post-extubation period. Results There was no significant difference in post-extubation failure in HHHFNC group when compared with CPAP group (p > 0.05) but post-extubation complication was significantly higher in CPAP group (p < 0.05) including nasal septal trauma and pneumothorax. Conclusions In neonates ≤32 weeks of gestational age, HHHFNC showed similar efficacy, and better safety profile than nasal-CPAP when used during post-extubation period for respiratory support. heated humidified high-flow nasal cannula (HHHFNC), continuous positive airway pressure (CPAP), respiratory support, post-extubation period, complications INTRODUCTION Respiratory failure remains a major problem in the neonatal intensive care unit (NICU), especially in preterm infants. The past decade has seen dramatic changes in respiratory care of preterm infants with the implementation of labor room continuous positive airway pressure (CPAP), selective use of surfactant for respiratory distress syndrome (RDS), caffeine for early extubation and noninvasive ventilation with the objective of minimizing the lung injury. To avoid ventilator-induced lung injury, ventilator support is minimized by early application of noninvasive respiratory support [1]. Nasal CPAP, nasal intermittent positive pressure ventilation and bi-level positive airway pressure (BiPAP) are three most commonly used noninvasive respiratory supports [2, 3]. Nasal CPAP is associated with complicated fixation techniques, positional problem, nasal trauma and apparent neonatal agitation [4, 5], whereas high flow through nasal cannula has inadequately warmed and humidified gas, which increases the risk of mucosal injury and nosocomial infection [6–8]. Heated humidified high-flow nasal cannula (HHHFNC) is gaining popularity in clinical practice owing to technical ease of its use without sealing. Retrospective and observational studies suggest that HHHFNC can be used as noninvasive respiratory support of premature infant with respiratory distress [7, 9–11]. Various randomized controlled trials (RCTs) demonstrated no significant difference in HHHFNC and CPAP during post-extubation respiratory support [12–16]. A recent systematic review revealed HHHFNC has similar rates of efficacy to other forms of noninvasive respiratory support in preterm infant as post-extubation support and was also associated with less complications like nasal trauma [17]. HHHFNC has gained popularity in developed countries but still there is no concrete evidence over its use in low- and middle-income countries, as there are less trials that have sought the role of HHHFNC in these countries like India. Hence, we conducted this study to see the efficacy of HHHFNC in comparison with CPAP for post-extubation respiratory support and thus generating evidence over the use of HHHFNC in low- and middle-income countries. MATERIAL AND METHODS This retrospective and prospective observational cohort study was conducted at a tertiary level NICU of Deep Hospital, Ludhiana, North India. Routine use of HHHFNC in infants commenced in our unit from 1 July 2013. Infants were eligible if they were born at a gestational age of <32 weeks or birth weight ≤1500 g either in our hospital or transferred in our unit before 48 h of post-natal age and needed noninvasive respiratory support after a period of mechanical ventilation with an endotracheal tube. Infants placed on CPAP between 1 January 2013 and 30 June 2013 were assigned to the CPAP group, and infants placed on HHHFNC from 1 July 2013 to 30 December 2013 were assigned to the HHHFNC group. We compared the safety and efficacy of the CPAP and HHHFNC in both the groups. Infants with lethal congenital anomalies/malformations (Pierre-Robin, Treacher Collins, Goldenhar, choanal atresia, cleft lip/palate), fatal chromosomal anomalies, gastrointestinal malformation (intestinal atresia, omphalocele, gastroschisis or diaphragmatic hernia), severe asphyxia (hypoxic ischemic encephalopathy stage III) and cyanotic congenital heart disease were excluded from the study. The records of infants admitted in Deep hospital have their medical information recorded in medical records. This information was used for retrospective data collection. The neonates who fulfilled predefined extubation criteria (requiring minimal ventilator settings on Synchronised intermittent mandatory ventilation/Pressure support ventilation mode [peak inspiratory pressure = 10–12 mmH2O, FiO2 = 0.25–0.30% with SPO2 = 92–95%, Rate = 20–25] for at least 12 h, clinically stable, maintaining perfusion, normal metabolic milieu) were extubated to either HHHFNC (1 July to 30 December 2013) or CPAP (1 January to 30 June 2013) or room air as per the discretion of the treating physician during the study period duration. We used AIRVOTM 2 high flow system (Fisher and Paykel) as HHHFNC and Fischer and Paykel HHHFNC prongs, and the diameter of the nasal prongs was <50% of the neonate nares, thus allowing leak. In HHHFNC group, initial flow rate was started at 5 l/min and initial FiO2 was 30%. FiO2 was titrated to maintain SpO2 between 90% and 95%. Flow rate was increased to a maximum of 7 l/min above the starting flow rate. In CPAP group, infants were started on a bubble CPAP generator (Fisher and Paykel Healthcare, Inc.) using short bi-nasal prongs (Hudson). Initial CPAP pressure was 5 cm of H2O and FiO2 of 30%. FiO2 was adjusted to maintain SpO2 between 90% and 95%. Flow rate or CPAP pressure was increased by 1 (maximum flow of 7 l/min or CPAP pressure of 7 cmH2O) if FiO2 increased by 10% above the starting FiO2 or pCO2, increased by 10 mmHg above the baseline value or increased respiratory distress score (Silverman Anderson score) by 2 from the baseline or >5 over a duration of 1 h or decreased lung expansion on the chest radiograph. Infants who were ventilated >7 days or who were intubated more than three times during the ventilation period owing to any sentinel events were given peri-extubation single dose of injection dexamethasone 0.15 mg/kg. Injection caffeine (loading dose at 20 mg/kg/dose of caffeine citrate and 5–7.5 mg/kg/dose as a maintenance dose) was used as a standard practice in unit to facilitate extubation and for prevention of apnea of prematurity. Infants who required reintubation within 48 h were considered to have extubation failure (criteria for reintubation were recurrent or severe apnea, more than four episodes per hour or need for bag and mask ventilation for any apnea, FiO2 requirement >60% with Silverman score >6, pH < 7.20 and PaCO2 > 60 mmHg, severe metabolic acidosis, arterial base deficit >−10 and shock requiring inotropic support). Baseline characteristics of the study population were recorded in predesigned proforma. Late onset sepsis (after 72 h of life) was defined by (a) positive blood, Cerebrospinal fluid or urine culture or (b) clinical signs of sepsis with C reactive protein > 10 mg/l. Infant was considered to have antenatal steroid (ANS) exposure if mother was administered at least one dose of steroid 12 h before delivery. Definitions of small for gestational age, prolonged preterm premature rupture of membranes, chorioamnionitis, gestational diabetes mellitus, pregnancy-induced hypertension, RDS, patent ductus arteriosus, early onset sepsis (<72 h of life) and indications for surfactant, early rescue surfactant (within 2 h of life), late rescue surfactant (after 2 h of life) and ibuprofen/paracetamol were as per standard protocol. Post-extubation complications defined were air leak, nasal septal trauma, CPAP belly or nasal bridge damage. STUDY OUTCOMES The primary outcome of this study was extubation failure within 48 h after extubation. Secondary outcomes were incidence of pneumothorax, predefined post-extubation complications and requirement of post-extubation respiratory support up to 72 h. STATISTICAL METHODS All the data were entered in excel sheet and analyzed. Chi-square and Fisher exact test were used to compare categorical data. Continuous variables were analyzed by Student’s two-sample t-test and Mann–Whitney U-test as appropriate. p value < 0.05 was considered as statistically significant. All p values are two sided and have not been adjusted for multiple comparison. Data were analyzed by using IBM SPSS V.18 software. RESULTS In this study, 136 infants in both the group fulfilled the inclusion criteria of the study, 56 infants were in the CPAP group and rest were in the HHHFNC group. In this study, baseline characteristics of the included infants during 6 month period of first and second half of the year were compared. There was no difference in the baseline characteristics between the two groups (Table 1). Table 1 Baseline characteristics of the study population Baseline Characteristics CPAP group (n=56) HHHFNC group (n=80) p value Mean birth weight (g) 1263.2±224.7 1187.00±219.4 0.870 Mean birth gestational age (weeks) 28.7±2.0 28.8±1.9 0.331 Infants <28 weeks 16 (28.5%) 20 (25%) 1.000 Extremely low birth weight 8 (14.2%) 20 (20%) 0.672 Extramural 28 (50%) 38 (47.5%) 0.247 Male 44 (78.6%) 56 (70%) 0.704 Caesarean section 28 (50%) 32 (40%) 0.728 ANS 50 (89.2%) 72 (90%) 0.892 Early onset sepsis 20 (35.7%) 44 (55%) 0.315 Early rescue surfactant 32 (57.1%) 38 (47.5%) 0.268 Any sentinel events(Air leak, accidental extubation, tube block) 0 (0%) 0 (0%) – Late-onset sepsis 28 (50%) 60 (75%) 0.163 Hemodynamic significant PDA requiring treatment 24 (42.8%) 24 (30.0%) 0.487 Baseline Characteristics CPAP group (n=56) HHHFNC group (n=80) p value Mean birth weight (g) 1263.2±224.7 1187.00±219.4 0.870 Mean birth gestational age (weeks) 28.7±2.0 28.8±1.9 0.331 Infants <28 weeks 16 (28.5%) 20 (25%) 1.000 Extremely low birth weight 8 (14.2%) 20 (20%) 0.672 Extramural 28 (50%) 38 (47.5%) 0.247 Male 44 (78.6%) 56 (70%) 0.704 Caesarean section 28 (50%) 32 (40%) 0.728 ANS 50 (89.2%) 72 (90%) 0.892 Early onset sepsis 20 (35.7%) 44 (55%) 0.315 Early rescue surfactant 32 (57.1%) 38 (47.5%) 0.268 Any sentinel events(Air leak, accidental extubation, tube block) 0 (0%) 0 (0%) – Late-onset sepsis 28 (50%) 60 (75%) 0.163 Hemodynamic significant PDA requiring treatment 24 (42.8%) 24 (30.0%) 0.487 Table 1 Baseline characteristics of the study population Baseline Characteristics CPAP group (n=56) HHHFNC group (n=80) p value Mean birth weight (g) 1263.2±224.7 1187.00±219.4 0.870 Mean birth gestational age (weeks) 28.7±2.0 28.8±1.9 0.331 Infants <28 weeks 16 (28.5%) 20 (25%) 1.000 Extremely low birth weight 8 (14.2%) 20 (20%) 0.672 Extramural 28 (50%) 38 (47.5%) 0.247 Male 44 (78.6%) 56 (70%) 0.704 Caesarean section 28 (50%) 32 (40%) 0.728 ANS 50 (89.2%) 72 (90%) 0.892 Early onset sepsis 20 (35.7%) 44 (55%) 0.315 Early rescue surfactant 32 (57.1%) 38 (47.5%) 0.268 Any sentinel events(Air leak, accidental extubation, tube block) 0 (0%) 0 (0%) – Late-onset sepsis 28 (50%) 60 (75%) 0.163 Hemodynamic significant PDA requiring treatment 24 (42.8%) 24 (30.0%) 0.487 Baseline Characteristics CPAP group (n=56) HHHFNC group (n=80) p value Mean birth weight (g) 1263.2±224.7 1187.00±219.4 0.870 Mean birth gestational age (weeks) 28.7±2.0 28.8±1.9 0.331 Infants <28 weeks 16 (28.5%) 20 (25%) 1.000 Extremely low birth weight 8 (14.2%) 20 (20%) 0.672 Extramural 28 (50%) 38 (47.5%) 0.247 Male 44 (78.6%) 56 (70%) 0.704 Caesarean section 28 (50%) 32 (40%) 0.728 ANS 50 (89.2%) 72 (90%) 0.892 Early onset sepsis 20 (35.7%) 44 (55%) 0.315 Early rescue surfactant 32 (57.1%) 38 (47.5%) 0.268 Any sentinel events(Air leak, accidental extubation, tube block) 0 (0%) 0 (0%) – Late-onset sepsis 28 (50%) 60 (75%) 0.163 Hemodynamic significant PDA requiring treatment 24 (42.8%) 24 (30.0%) 0.487 There was no difference in the incidence of post-extubation failure in both the groups (p > 0.05). The incidence of post-extubation complication was significantly higher in CPAP group when compared with HHHFNC group (p < 0.001). Of 12 infants, eight infants developed nasal septal damage and four infants landed up with pneumothorax. There was no significant difference in the duration of post-extubation respiratory support up to 72 h in both the groups (Table 2). Table 2 Primary and secondary outcomes Outcomes CPAP group (n = 56) HHHFNC group (n = 80) p value Post-extubation failure 2 (3.5%) 0 (0%) 0.088 Post-extubation complications 12 (21.4%) 0 (0%) <0.001 Nasal trauma 8 (14.28%) 0 (0%) 0.0005 Pneumothorax 4 (7.14%) 0 (0%) 0.015 Post-extubation respiratory support up to 72 h 40 (71.4%) 60 (75%) 0.642 Outcomes CPAP group (n = 56) HHHFNC group (n = 80) p value Post-extubation failure 2 (3.5%) 0 (0%) 0.088 Post-extubation complications 12 (21.4%) 0 (0%) <0.001 Nasal trauma 8 (14.28%) 0 (0%) 0.0005 Pneumothorax 4 (7.14%) 0 (0%) 0.015 Post-extubation respiratory support up to 72 h 40 (71.4%) 60 (75%) 0.642 Table 2 Primary and secondary outcomes Outcomes CPAP group (n = 56) HHHFNC group (n = 80) p value Post-extubation failure 2 (3.5%) 0 (0%) 0.088 Post-extubation complications 12 (21.4%) 0 (0%) <0.001 Nasal trauma 8 (14.28%) 0 (0%) 0.0005 Pneumothorax 4 (7.14%) 0 (0%) 0.015 Post-extubation respiratory support up to 72 h 40 (71.4%) 60 (75%) 0.642 Outcomes CPAP group (n = 56) HHHFNC group (n = 80) p value Post-extubation failure 2 (3.5%) 0 (0%) 0.088 Post-extubation complications 12 (21.4%) 0 (0%) <0.001 Nasal trauma 8 (14.28%) 0 (0%) 0.0005 Pneumothorax 4 (7.14%) 0 (0%) 0.015 Post-extubation respiratory support up to 72 h 40 (71.4%) 60 (75%) 0.642 DISCUSSION The results of our study show that HHHFNC is equally efficacious to CPAP for post-extubation. There was also significant reduction in post-extubation complications with application of HHHFNC including nasal trauma and pneumothorax. The results of our study are similar to published RCTs [13, 14, 18]. This observation further supports the early retrospective and observational studies and recent systematic review [7, 9–11, 17]. According to these studies, HHHFNC can be applied safely and effectively as noninvasive respiratory management of premature infant. A recent randomized noninferiority trial revealed HHHFNC had efficacy and safety similar to Nasal continuous positive airway pressure (n-CPAP)/BiPAP when used as a primary approach in mild to moderate RDS [19]. In our study, duration of post-extubation respiratory support up to 72 h was not statistically significantly different in the groups, whereas in others conducted, RCT infants who were on CPAP had significantly shorter duration of post-extubation respiratory support [13, 14, 18]. In our study, overall adverse events were higher in CPAP group compared with HHHFNC group in contrast to adverse events similar in both groups in multi-centric trial [18]. The results of adverse events were similar to recently published meta-analysis that showed HHHFNC resulted in significant reduction in incidence of nasal trauma when compared with CPAP [20]. In this study, we reported that CPAP leads to increase in pneumothorax in post-extubation period. The results of the study are similar to recently published Cochrane meta-analysis [17]. The strength of the study includes adequate sample size and strict study protocol and limitation of our study includes its retrospective nature. Thus, HHHFNC group had similar efficacy and less adverse events during post-extubation period in very low birth weight (VLBW) infants. However, additional prospective research is needed to better define the utility and safety of HHFNC compared with NCPAP. CONCLUSION The results of our study clearly demonstrated the efficacy and safety of HHHFNC in VLBW preterm infants in post-extubation period. Owing to limited evidence from low- and middle-income countries, a large and well-designed multi-centric RCT from these countries will give good evidence for use of HHHFNC in these countries because neonatal population and NICU care of the developing countries is different from developed countries, thus making generalization of results from developed countries difficult. REFERENCES 1 DiBlasi RM. Neonatal noninvasive ventilation techniques: do we really need to intubate? Respir Care 2011 ; 56 : 1273 – 94 . Google Scholar CrossRef Search ADS PubMed 2 Roberts CT , Davis PG , Owen LS. Neonatal non-invasive respiratory support: synchronised NIPPV, non-synchronised NIPPV or bi-level CPAP: what is the evidence in 2013 . Neonatology 2013 ; 104 : 203 – 9 . Google Scholar CrossRef Search ADS PubMed 3 Amatya S , Rastogi D , Bhutada A , et al. Weaning of nasal CPAP in preterm infants: who, when and how? A systematic review of the literature . World J Pediatr 2015 ; 11 : 7 – 13 . Google Scholar CrossRef Search ADS PubMed 4 Bonner KM , Mainous RO. The nursing care of the infant receiving bubble CPAP therapy . Adv Neonatal Care 2008 ; 8 : 78 – 95 . quiz 96–97. Google Scholar CrossRef Search ADS PubMed 5 McCoskey L. Nursing care guidelines for prevention of nasal breakdown in neonates receiving nasal CPAP . Adv Neonatal Care 2008 ; 8 : 116 – 24 . Google Scholar CrossRef Search ADS PubMed 6 Waugh JB , Granger WM. An evaluation of 2 new devices for nasal high-flow gas therapy . Respir Care 2004 ; 49 : 902 – 6 . Google Scholar PubMed 7 Woodhead DD , Lambert DK , Clark JM , et al. Comparing two methods of delivering high-flow gas therapy by nasal cannula following endotracheal extubation: a prospective, randomized, masked, crossover trial . J Perinatol 2006 ; 26 : 481 – 5 . Google Scholar CrossRef Search ADS PubMed 8 Kopelman AE , Holbert D. Use of oxygen cannulas in extremely low birthweight infants is associated with mucosal trauma and bleeding, and possibly with coagulase negative staphylococcal sepsis . J Perinatol 2003 ; 23 : 94 – 7 . Google Scholar CrossRef Search ADS PubMed 9 Holleman-Duray D , Kaupie D , Weiss MG. Heated humidified high flow nasal cannula: use and a neonatal early extubation protocol . J Perinatol 2007 ; 27 : 776 – 81 . Google Scholar CrossRef Search ADS PubMed 10 Miller SM , Dowd SA. High flow nasal cannula and extubation success in the premature infant: a comparison of two modalities . J Perinatol 2010 ; 30 : 805 – 8 . Google Scholar CrossRef Search ADS PubMed 11 Shoemaker MT , Pierce MR , Yoder BA , et al. High flow nasal cannula versus nasal CPAP for neonatal respiratory disease: a retrospective study . J Perinatol 2007 ; 27 : 85 – 91 . Google Scholar CrossRef Search ADS PubMed 12 Campbell DM , Shah PS , Shah V , et al. Nasal continuous positive airway pressure from high flow cannula versus infant flow for preterm infants . J Perinatol 2006 ; 26 : 546 – 9 . Google Scholar CrossRef Search ADS PubMed 13 Collins CL , Holberton JR , Barfield C , et al. A randomized controlled trial to compare heated humidified high-flow nasal cannulae with nasal continuous positive airway pressure postextubation in premature infants . J Pediatr 2013 ; 162 : 949 – 54 . Google Scholar CrossRef Search ADS PubMed 14 Manley BJ , Owen LS , Doyle LW , et al. High-flow nasal cannulae in very preterm infants after extubation . N Engl J Med 2013 ; 369 : 1425 – 33 . Google Scholar CrossRef Search ADS PubMed 15 Collaborative Group for the Multicenter Study on Heated Humidified High flow Nasal Cannula Ventilation . Efficacy and safety of heated humidified high flow nasal cannula for prevention of extubation failure in neonates , [in Chinese]. Zhonghua Er Ke Za Zhi 2014 ; 52 : 271 – 6 . PubMed 16 Mostafa-Gharehbaghi M , Mojabi H. Comparing the effectiveness of nasal continuous positive airway pressure (NCPAP) and high flow nasal cannula (HFNC) in prevention of post extubation assisted ventilation . Zahedan J Res Med Sci 2015 ; 17 : e984 . Google Scholar CrossRef Search ADS 17 Wilkinson D , Andersen C , O'Donnell CP , et al. High flow nasal cannula for respiratory support in preterm infants . Cochrane Database Syst Rev 2016 ; 2 : CD006405 . Google Scholar PubMed 18 Yoder BA , Stoddard RA , Li M , et al. Heated, humidified high-flow nasal cannula versus nasal CPAP for respiratory support in neonates . Pediatrics 2013 ; 131 : e1482 – 90 . Google Scholar CrossRef Search ADS PubMed 19 Lavizzari A , Colnaghi M , Ciuffini F , et al. Heated, humidified high flow nasal cannula v/s nasal continuous positive airway pressure for respiratory distress syndrome: a randomised clinical non inferiority trial . JAMA Pediatrics 2017 , in press. 20 Fleeman N , Mahon J , Bates V , et al. The clinical effectiveness and cost-effectiveness of heated humidified high flow nasal cannula compared with usual care for preterm infants: systematic review and economic evaluation . Health Technol Assess 2016 ; 20 : 1 – 68 . Google Scholar CrossRef Search ADS PubMed © The Author [2017]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

2018 Journal of Tropical Pediatrics

doi: 10.1093/tropej/fmx059pmid: 28977637

Abstract Background The purpose of this article was to examine individual- and community-level factors associated with childhood anemia, severe anemia, and hemoglobin (Hb) concentration in Malawi. Methods Using data from the 2010 Malawi demographic and health survey (MDHS), the multilevel regression models were constructed to analyze 2597 children aged 6–59 months living in 849 communities. Results The results showed that both childhood anemia and severe anemia were negatively associated with child’s age, no fever in the previous 2 weeks and height-for-age, and positively associated with residing in poor household. Childhood anemia was negatively associated with community female education. Child’s age, no fever in the previous 2 weeks and maternal Hb levels were positively associated with child Hb concentration, while residing in poorest households was negatively associated with children’s Hb concentration. Conclusion Comprehensive public health strategies aimed at reducing childhood anemia need to focus more on the significant characteristics addressed in this study. childhood anemia, multilevel study, Malawi BACKGROUND Anemia is defined as a condition in which the number of red blood cells, hemoglobin (Hb) concentration or packed-cell volume, is insufficient to meet the body’s physiologic needs [1–4]. A global estimate of childhood anemia indicated that 293.1 million children under the age of 5 years are anemic, and 28.5% of these children reside in sub-Saharan Africa (SSA) [5, 6]. Malawi has one of the highest prevalence rates of anemia across nations [7]. Studies have reported that 63% of children aged 6∼59 months in Malawi are anemic and 3% are severe anemic [8]. The World Health Organization (WHO) statement on anemia underlines its multifactorial etiology [9]. Approximately 50% cases of anemia are considered to be due to an iron deficiency [5, 6, 10], although other factors such as nutritional deficiencies [6, 11, 12], acute and chronic inflammation [12–14], and inherited or acquired disorders that affect Hb synthesis [12, 15], can all be etiologies of anemia. Previous studies also had demonstrated that factors such as a child’s biological characteristics [5, 16–19], maternal characteristics [16–19], household economic status [17, 19], and community characteristics [19] have impacts on childhood anemia. Despite the individual risk factors of childhood anemia have been well explored in prior studies [3, 20–23], few studies have simultaneously considered the influence of contextual- and individual-level factors on childhood anemia, particularly in Malawi. Previously, it was reported that the community constitutes a key component of socioeconomic challenges to good health [24–27]. Therefore, this study aimed to examine whether individual-level and community-level factors both can influence childhood anemia, severe anemia, and Hb concentration. MATERIALS AND METHODS Study design Data came from the 2010 Malawi Demographic and Health Survey (MDHS). Methods used in this study have been described in details elsewhere [8]. Using a stratified two-stage cluster design, the 2010 MDHS produced a nationally representative sample. The first stage selected 849 communities (clusters) and the second stage selected 27 345 households. Analyses were limited to the children whose households were selected for hemoglobin estimation. A random procedure was conducted to select one child per household to avoid the clustering effects, which generated a final sample size of 2597. Data collection Using face-to-face interviews, data were collected from women aged 15–49 with children below the age of 5 years prior to the survey. Anemia test from children aged 6–59 months was done through finger prick blood testing using the HemoCue blood hemoglobin (Hb) system. The HemoCue system is based on the conversion of Hb to cyanmethemoglobin and its detection by measuring absorption in a spectrophotometer [28]. Measures Outcome variables Childhood anemia, severe anemia, and hemoglobin concentration were the outcomes. Using WHO recommendations, childhood anemia was defined as children aged 6–59 months with an Hb level of <11 g/dL, while severe anemia was defined as children aged 6–59 months with an Hb level of <7.0 g/dL [1]. Hb concentration was a continuous variable measured in g/dL. Independent variables Individual/maternal/household-level factors Child-specific factors included child’s sex, child’s age (in months), birth order, history of fever and diarrhea in the past 2 weeks, stunting, underweight, and vitamin A supplements and deworming in the past 6 months. Maternal and household characteristics include age (in years), educational attainment, maternal anemic status, and household wealth status (Table 1). We defined stunting and underweight as children with height-for-age and weight-for-age Z-score < −2 relative to WHO standards [29, 30]. Vitamin A Supplement was defined as whether the child had received a high dose of vitamin A supplement in the past 6 months [31]. Fever diagnosis was based on the self-reports by mothers about symptoms that had occurred within 2 weeks prior to the survey date [8, 32]. Diarrheal disease was defined as the passage of three or more loose or liquid stools during a 24-h period [33]. Mothers with hemoglobin level of <12 g/dL were regarded to be anemic [1]. The wealth index was constructed using data on a household’s ownership of selected assets, such as televisions, materials used for constructing the house etc., through the principal component analysis [34, 35]. Table 1 Description of study variables Characteristics Measure Outcome variables  Childhood anemia 1 = if the child has a hemoglobin level of < 11.0 g/dL, and 0 = otherwise  Severe anemia 1 = if the child has a hemoglobin level of < 7.0 g/dL, and 0 = otherwise  Hemoglobin concentration Continuous variable measured in grams per deciliter (g/dL) Individual-level variables  Sex of the child (Ref: female) 1 = female, and 0 = male  Age of the child Continuous variable measured in months  Birth order of child Continuous variable  Diarrhea in the past 2 weeks (Ref: yes) 1 = if the child had diarrhea, and 0 = otherwise  Fever in the past 2 weeks (Ref: yes) 1 = if the child had a fever, and 0 = otherwise  History of stunting (Ref: stunted)a 1 = if the child is stunted, and 0 = otherwise  History of wasting (Ref: wasted)b 1 = if the child is wasted, and 0 = otherwise  Vitamin A Supplements in previous 6 months (Ref: yes) 1 = if the child received vitamin supplements A in the past 6 months, and 0 = otherwise  Deworming in previous 6 months (Ref: yes) 1 = if the child received deworming medicine in the past 6 months, and 0 = otherwise  Mother’s age group (Ref: 35∼49) Continuous variable  Mother’s educational level (Ref: Secondary and above) 0 = none, 1 = primary, 2 = secondary and above  Maternal anemia status (Ref: anemic) Continuous variable  Household wealth index (Ref: Richest) 0 = poorest, 1 = poor, 2 = middle, 3 = rich, 4 = richest Community-level variables  Place of residence (Ref: urban) 1 = rural area; 0 = urban  Geographical region (Ref: southern) 0 = northern, 1 = central and 2 = southern  Community wealth Mean percentage of richest (rich, richest) households in the community—upper 40% of household wealth index  Community female education Mean percentage of women in the community with primary education and above  Community water supplyc Mean percentage of households in the community with access to improved drinking water sources  Community sanitation servicesd Mean percentage of households in the community with access to improved sanitation Characteristics Measure Outcome variables  Childhood anemia 1 = if the child has a hemoglobin level of < 11.0 g/dL, and 0 = otherwise  Severe anemia 1 = if the child has a hemoglobin level of < 7.0 g/dL, and 0 = otherwise  Hemoglobin concentration Continuous variable measured in grams per deciliter (g/dL) Individual-level variables  Sex of the child (Ref: female) 1 = female, and 0 = male  Age of the child Continuous variable measured in months  Birth order of child Continuous variable  Diarrhea in the past 2 weeks (Ref: yes) 1 = if the child had diarrhea, and 0 = otherwise  Fever in the past 2 weeks (Ref: yes) 1 = if the child had a fever, and 0 = otherwise  History of stunting (Ref: stunted)a 1 = if the child is stunted, and 0 = otherwise  History of wasting (Ref: wasted)b 1 = if the child is wasted, and 0 = otherwise  Vitamin A Supplements in previous 6 months (Ref: yes) 1 = if the child received vitamin supplements A in the past 6 months, and 0 = otherwise  Deworming in previous 6 months (Ref: yes) 1 = if the child received deworming medicine in the past 6 months, and 0 = otherwise  Mother’s age group (Ref: 35∼49) Continuous variable  Mother’s educational level (Ref: Secondary and above) 0 = none, 1 = primary, 2 = secondary and above  Maternal anemia status (Ref: anemic) Continuous variable  Household wealth index (Ref: Richest) 0 = poorest, 1 = poor, 2 = middle, 3 = rich, 4 = richest Community-level variables  Place of residence (Ref: urban) 1 = rural area; 0 = urban  Geographical region (Ref: southern) 0 = northern, 1 = central and 2 = southern  Community wealth Mean percentage of richest (rich, richest) households in the community—upper 40% of household wealth index  Community female education Mean percentage of women in the community with primary education and above  Community water supplyc Mean percentage of households in the community with access to improved drinking water sources  Community sanitation servicesd Mean percentage of households in the community with access to improved sanitation Note: g/dL = grams per deciliter; WHO = World Health Organization. a Height-for-age < −2 standard deviation of the WHO reference group [29, 30]. b Weight-for-age < −2 standard deviation of the WHO reference group [29, 30]. c Improved drinking water (piped water into dwelling, piped water to yard/plot, public tap or standpipe, tubewell or borehole, protected dug well, protected spring and rainwater) [36]. d Improved sanitation (flush toilet, piped sewer system, septic tank, flush/pour flush to pit latrine, ventilated improved pit latrine, pit latrine with slab, and composting toilet) [36]. Table 1 Description of study variables Characteristics Measure Outcome variables  Childhood anemia 1 = if the child has a hemoglobin level of < 11.0 g/dL, and 0 = otherwise  Severe anemia 1 = if the child has a hemoglobin level of < 7.0 g/dL, and 0 = otherwise  Hemoglobin concentration Continuous variable measured in grams per deciliter (g/dL) Individual-level variables  Sex of the child (Ref: female) 1 = female, and 0 = male  Age of the child Continuous variable measured in months  Birth order of child Continuous variable  Diarrhea in the past 2 weeks (Ref: yes) 1 = if the child had diarrhea, and 0 = otherwise  Fever in the past 2 weeks (Ref: yes) 1 = if the child had a fever, and 0 = otherwise  History of stunting (Ref: stunted)a 1 = if the child is stunted, and 0 = otherwise  History of wasting (Ref: wasted)b 1 = if the child is wasted, and 0 = otherwise  Vitamin A Supplements in previous 6 months (Ref: yes) 1 = if the child received vitamin supplements A in the past 6 months, and 0 = otherwise  Deworming in previous 6 months (Ref: yes) 1 = if the child received deworming medicine in the past 6 months, and 0 = otherwise  Mother’s age group (Ref: 35∼49) Continuous variable  Mother’s educational level (Ref: Secondary and above) 0 = none, 1 = primary, 2 = secondary and above  Maternal anemia status (Ref: anemic) Continuous variable  Household wealth index (Ref: Richest) 0 = poorest, 1 = poor, 2 = middle, 3 = rich, 4 = richest Community-level variables  Place of residence (Ref: urban) 1 = rural area; 0 = urban  Geographical region (Ref: southern) 0 = northern, 1 = central and 2 = southern  Community wealth Mean percentage of richest (rich, richest) households in the community—upper 40% of household wealth index  Community female education Mean percentage of women in the community with primary education and above  Community water supplyc Mean percentage of households in the community with access to improved drinking water sources  Community sanitation servicesd Mean percentage of households in the community with access to improved sanitation Characteristics Measure Outcome variables  Childhood anemia 1 = if the child has a hemoglobin level of < 11.0 g/dL, and 0 = otherwise  Severe anemia 1 = if the child has a hemoglobin level of < 7.0 g/dL, and 0 = otherwise  Hemoglobin concentration Continuous variable measured in grams per deciliter (g/dL) Individual-level variables  Sex of the child (Ref: female) 1 = female, and 0 = male  Age of the child Continuous variable measured in months  Birth order of child Continuous variable  Diarrhea in the past 2 weeks (Ref: yes) 1 = if the child had diarrhea, and 0 = otherwise  Fever in the past 2 weeks (Ref: yes) 1 = if the child had a fever, and 0 = otherwise  History of stunting (Ref: stunted)a 1 = if the child is stunted, and 0 = otherwise  History of wasting (Ref: wasted)b 1 = if the child is wasted, and 0 = otherwise  Vitamin A Supplements in previous 6 months (Ref: yes) 1 = if the child received vitamin supplements A in the past 6 months, and 0 = otherwise  Deworming in previous 6 months (Ref: yes) 1 = if the child received deworming medicine in the past 6 months, and 0 = otherwise  Mother’s age group (Ref: 35∼49) Continuous variable  Mother’s educational level (Ref: Secondary and above) 0 = none, 1 = primary, 2 = secondary and above  Maternal anemia status (Ref: anemic) Continuous variable  Household wealth index (Ref: Richest) 0 = poorest, 1 = poor, 2 = middle, 3 = rich, 4 = richest Community-level variables  Place of residence (Ref: urban) 1 = rural area; 0 = urban  Geographical region (Ref: southern) 0 = northern, 1 = central and 2 = southern  Community wealth Mean percentage of richest (rich, richest) households in the community—upper 40% of household wealth index  Community female education Mean percentage of women in the community with primary education and above  Community water supplyc Mean percentage of households in the community with access to improved drinking water sources  Community sanitation servicesd Mean percentage of households in the community with access to improved sanitation Note: g/dL = grams per deciliter; WHO = World Health Organization. a Height-for-age < −2 standard deviation of the WHO reference group [29, 30]. b Weight-for-age < −2 standard deviation of the WHO reference group [29, 30]. c Improved drinking water (piped water into dwelling, piped water to yard/plot, public tap or standpipe, tubewell or borehole, protected dug well, protected spring and rainwater) [36]. d Improved sanitation (flush toilet, piped sewer system, septic tank, flush/pour flush to pit latrine, ventilated improved pit latrine, pit latrine with slab, and composting toilet) [36]. Community-level factors We included six variables. Two variables indicate an area of residence, i.e., place of residence and geographical region. Four continuous variables assessed community wealth, community female education, community water supply and community sanitation services. We defined a community based on the primary sample unit in the DHS data. All continuous community-level factors were constructed by aggregating individual-level data to the community level (Table 1). The WHO and UNICEF Joint Monitoring Programme (JMP) for water supply and sanitation guidelines were used to define safe drinking water and improved sanitation [36]. Statistical analyses Two-level multilevel multivariate regression models were constructed for the analyses. Generalized linear mixed models were constructed for the anemia and the severe anemia outcomes, while the linear mixed-effect models were constructed for the Hb concentration outcomes [37]. Because children living in the same community may be more similar to each other than individuals from different communities, the multilevel models were used to adjust the correlated individual responses nested under the same community. Intraclass correlations (ICC) and percentage change in variation (PCV) were reported to assess the extent of which community variances were explained in each model. Model fits were assessed using deviation information criterion, while variance inflation factor was used to examine for multicollinearity. Ethics statement The protocol for blood sample collection and the questionnaires was reviewed and approved by the Malawi National Health Sciences Research Committee, the Institutional Review Board of ICF Macro, and the Centers for Disease Control in Atlanta. Informed consent was obtained at the beginning of each interview, and the authors sought permission from the DHS program for the use of the data. RESULTS Sample characteristics The mean Hb concentration was 10.47 g/dL. Figure 1 shows the distribution of children’s Hb level. The prevalence of anemia and severe anemia were estimated at 63% and 3%, respectively. Table 2 shows the descriptive statistics. The prevalence of anemia was observed to be highest among younger, chronically undernourished, lower socioeconomic children and children with a history of diarrhea and fever and without proper parental care practices. Table 2 Characteristics of children aged 6–59 months with anemia and severe anemia in Malawi, n = 2597 Characteristics Total Anemiaa Severe anemiab No Yes p-value No Yes p-value Individual-level factors Child age, months, Mean±SD* 31.51±14.91 36.42±13.69 28.62±14.85 <.0001 31.77±14.91 22.68±12.09 <.0001  Sex of the child, n (%)**   Male 1293 (49.79) 471 (36.43) 822 (63.57) 0.4668 1254 (96.98) 39 (3.02) 0.6109   Female 1304 (50.21) 493 (37.81) 811 (62.19) 1269 (97.32) 35 (2.68)  Birth order, Mean±SD* 3.67±2.32 3.62±2.25 3.70±2.35 0.7407 3.67±2.31 3.78±2.46 0.4648  Height-for-Age, Mean ± SD* −1.84±1.51 −1.75±1.40 −1.89±1.57 <.0001 −1.83±1.50 −1.99±1.07 0.8754  Weight-for-Age Mean ± SD* −0.84±1.08 −0.76±1.04 −0.88±1.10 0.0043 −0.83±1.25 1.05±1.20 0.4251  Vitamin A in the past 6 months, n (%)**   No 323 (12.44) 101 (31.27) 222 (68.73) 0.0200 315 (97.52) 8 (2.48) 0.6671   Yes 2274 (87.57) 863 (37.95) 1411 (62.05) 2208 (97.10) 66 (2.90)  Deworming in previous 6 months, n (%)**   No 698 (26.88) 219 (31.38) 479 (68.62) 0.0002 671 (96.13) 27 (3.87) 0.0585   Yes 1899 (73.12) 745 (39.23) 1154 (60.77) 1852 (97.53) 47 (2.47)  Diarrhea, n (%)**   No 2166 (83.40) 852 (39.34) 1314 (60.66) <.0001 2110 (97.41) 56 (2.59) 0.0698   Yes 431 (16.60) 112 (25.99) 319 (74.01) 413 (95.82) 18 (4.18)  Fever, n (%)**   No 1662 (64.00) 709 (42.60) 954 (57.40) <.0001 1627 (97.89) 35 (2.11) 0.0024   Yes 935 (36.00) 256 (27.38) 679 (72.62) 896 (95.83) 39 (4.17)  Mother’s age, (years) Mean ± SD* 28.85±6.92 29.26±6.65 28.61±7.07 0.0161 28.84±6.87 29.14±8.31 0.9135  Mother’s education, n (%)**   No formal education 422 (16.25) 129 (30.57) 293 (69.43) 0.0005 409 (96.92) 13 (3.08) 0.8923   Primary 1817 (69.97) 677 (37.26) 1140 (62.74) 1765 (97.14) 52 (2.86)   Secondary and above 358 (13.79) 158 (44.13) 200 (55.87) 349 (97.49) 9 (2.51)  Maternal hemoglobin level, Mean ± SD* 12.78±1.75 13.10±1.75 12.59±1.72 <.0001 12.78±1.75 12.65±1.54 0.0200  Household wealth, n (%)**   Poorest 482 (18.56) 141 (29.25) 341 (70.75) <.0001 462 (95.85) 20 (4.15) 0.0581   Poor 603 (23.22) 205 (34.00) 398 (66.00) 583 (96.68) 20 (3.32)   Middle 590 (22.72) 209 (35.42) 381 (64.58) 571 (96.78) 19 (3.32)   Rich 513 (19.75) 211 (41.13) 302 (58.87) 503 (98.05) 10 (1.95)   Richest 409 (15.75) 198 (48.41) 211 (51.59) 404 (98.78) 5 (1.22) Community-level factors  Place of residence n (%)**   Urban 285 (10.97) 136 (47.72) 149 (52.28) <.0001 280 (98.25) 5 (1.75) 0.2390   Rural 2312 (89.03) 828 (35.81) 1484 (64.19) 2243 (97.02) 69 (2.98)  Geographical region n (%)**   Northern 427 (16.44) 186 (43.56) 241 (56.44) 0.0082 418 (97.89) 9 (2.11) 0.1440   Central 984 (37.89) 361 (36.69) 623 (63.31) 948 (96.34) 36 (3.66)   Southern 1186 (45.67) 417 (35.16) 769 (64.84) 1157 (97.55) 29 (2.45) Community wealthcMean ± SD* 37.85±29.32 37.83±28.32 30.35±29.16 <.0001 33.25±26.38 28.87±22.65 0.2174 Community female educationdMean ± SD* 84.48±14.77 85.89±13.46 82.11±15.15 <.0001 83.60±14.64 80.75±15.21 0.1382 Community water supplyeMean ± SD* 79.11±25.22 80.48±25.09 78.30±25.27 0.0347 79.04±25.35 81.60±20.49 0.4992 Community sanitation servicesfMean ± SD* 10.01±16.03 11.06±17.11 9.40±15.33 0.0040 9.96±16.03 11.85±16.31 0.1929 Characteristics Total Anemiaa Severe anemiab No Yes p-value No Yes p-value Individual-level factors Child age, months, Mean±SD* 31.51±14.91 36.42±13.69 28.62±14.85 <.0001 31.77±14.91 22.68±12.09 <.0001  Sex of the child, n (%)**   Male 1293 (49.79) 471 (36.43) 822 (63.57) 0.4668 1254 (96.98) 39 (3.02) 0.6109   Female 1304 (50.21) 493 (37.81) 811 (62.19) 1269 (97.32) 35 (2.68)  Birth order, Mean±SD* 3.67±2.32 3.62±2.25 3.70±2.35 0.7407 3.67±2.31 3.78±2.46 0.4648  Height-for-Age, Mean ± SD* −1.84±1.51 −1.75±1.40 −1.89±1.57 <.0001 −1.83±1.50 −1.99±1.07 0.8754  Weight-for-Age Mean ± SD* −0.84±1.08 −0.76±1.04 −0.88±1.10 0.0043 −0.83±1.25 1.05±1.20 0.4251  Vitamin A in the past 6 months, n (%)**   No 323 (12.44) 101 (31.27) 222 (68.73) 0.0200 315 (97.52) 8 (2.48) 0.6671   Yes 2274 (87.57) 863 (37.95) 1411 (62.05) 2208 (97.10) 66 (2.90)  Deworming in previous 6 months, n (%)**   No 698 (26.88) 219 (31.38) 479 (68.62) 0.0002 671 (96.13) 27 (3.87) 0.0585   Yes 1899 (73.12) 745 (39.23) 1154 (60.77) 1852 (97.53) 47 (2.47)  Diarrhea, n (%)**   No 2166 (83.40) 852 (39.34) 1314 (60.66) <.0001 2110 (97.41) 56 (2.59) 0.0698   Yes 431 (16.60) 112 (25.99) 319 (74.01) 413 (95.82) 18 (4.18)  Fever, n (%)**   No 1662 (64.00) 709 (42.60) 954 (57.40) <.0001 1627 (97.89) 35 (2.11) 0.0024   Yes 935 (36.00) 256 (27.38) 679 (72.62) 896 (95.83) 39 (4.17)  Mother’s age, (years) Mean ± SD* 28.85±6.92 29.26±6.65 28.61±7.07 0.0161 28.84±6.87 29.14±8.31 0.9135  Mother’s education, n (%)**   No formal education 422 (16.25) 129 (30.57) 293 (69.43) 0.0005 409 (96.92) 13 (3.08) 0.8923   Primary 1817 (69.97) 677 (37.26) 1140 (62.74) 1765 (97.14) 52 (2.86)   Secondary and above 358 (13.79) 158 (44.13) 200 (55.87) 349 (97.49) 9 (2.51)  Maternal hemoglobin level, Mean ± SD* 12.78±1.75 13.10±1.75 12.59±1.72 <.0001 12.78±1.75 12.65±1.54 0.0200  Household wealth, n (%)**   Poorest 482 (18.56) 141 (29.25) 341 (70.75) <.0001 462 (95.85) 20 (4.15) 0.0581   Poor 603 (23.22) 205 (34.00) 398 (66.00) 583 (96.68) 20 (3.32)   Middle 590 (22.72) 209 (35.42) 381 (64.58) 571 (96.78) 19 (3.32)   Rich 513 (19.75) 211 (41.13) 302 (58.87) 503 (98.05) 10 (1.95)   Richest 409 (15.75) 198 (48.41) 211 (51.59) 404 (98.78) 5 (1.22) Community-level factors  Place of residence n (%)**   Urban 285 (10.97) 136 (47.72) 149 (52.28) <.0001 280 (98.25) 5 (1.75) 0.2390   Rural 2312 (89.03) 828 (35.81) 1484 (64.19) 2243 (97.02) 69 (2.98)  Geographical region n (%)**   Northern 427 (16.44) 186 (43.56) 241 (56.44) 0.0082 418 (97.89) 9 (2.11) 0.1440   Central 984 (37.89) 361 (36.69) 623 (63.31) 948 (96.34) 36 (3.66)   Southern 1186 (45.67) 417 (35.16) 769 (64.84) 1157 (97.55) 29 (2.45) Community wealthcMean ± SD* 37.85±29.32 37.83±28.32 30.35±29.16 <.0001 33.25±26.38 28.87±22.65 0.2174 Community female educationdMean ± SD* 84.48±14.77 85.89±13.46 82.11±15.15 <.0001 83.60±14.64 80.75±15.21 0.1382 Community water supplyeMean ± SD* 79.11±25.22 80.48±25.09 78.30±25.27 0.0347 79.04±25.35 81.60±20.49 0.4992 Community sanitation servicesfMean ± SD* 10.01±16.03 11.06±17.11 9.40±15.33 0.0040 9.96±16.03 11.85±16.31 0.1929 a Hb < 11g/dL. b Hb < 7 g/dL; Hb = hemoglobin; SD = standard deviation. c Percentage of households in the community categorized as rich (upper 40% of quintiles). d Percentage of female in the community with primary and above education. e Percentage of household in the community with access to improved water sources. f Percentage of household in the community with access to improved sanitation facilities. * Groups were significantly different by univariate logistic regression, p < 0.05. ** Groups were significantly different by chi-square test, p < 0.05. Table 2 Characteristics of children aged 6–59 months with anemia and severe anemia in Malawi, n = 2597 Characteristics Total Anemiaa Severe anemiab No Yes p-value No Yes p-value Individual-level factors Child age, months, Mean±SD* 31.51±14.91 36.42±13.69 28.62±14.85 <.0001 31.77±14.91 22.68±12.09 <.0001  Sex of the child, n (%)**   Male 1293 (49.79) 471 (36.43) 822 (63.57) 0.4668 1254 (96.98) 39 (3.02) 0.6109   Female 1304 (50.21) 493 (37.81) 811 (62.19) 1269 (97.32) 35 (2.68)  Birth order, Mean±SD* 3.67±2.32 3.62±2.25 3.70±2.35 0.7407 3.67±2.31 3.78±2.46 0.4648  Height-for-Age, Mean ± SD* −1.84±1.51 −1.75±1.40 −1.89±1.57 <.0001 −1.83±1.50 −1.99±1.07 0.8754  Weight-for-Age Mean ± SD* −0.84±1.08 −0.76±1.04 −0.88±1.10 0.0043 −0.83±1.25 1.05±1.20 0.4251  Vitamin A in the past 6 months, n (%)**   No 323 (12.44) 101 (31.27) 222 (68.73) 0.0200 315 (97.52) 8 (2.48) 0.6671   Yes 2274 (87.57) 863 (37.95) 1411 (62.05) 2208 (97.10) 66 (2.90)  Deworming in previous 6 months, n (%)**   No 698 (26.88) 219 (31.38) 479 (68.62) 0.0002 671 (96.13) 27 (3.87) 0.0585   Yes 1899 (73.12) 745 (39.23) 1154 (60.77) 1852 (97.53) 47 (2.47)  Diarrhea, n (%)**   No 2166 (83.40) 852 (39.34) 1314 (60.66) <.0001 2110 (97.41) 56 (2.59) 0.0698   Yes 431 (16.60) 112 (25.99) 319 (74.01) 413 (95.82) 18 (4.18)  Fever, n (%)**   No 1662 (64.00) 709 (42.60) 954 (57.40) <.0001 1627 (97.89) 35 (2.11) 0.0024   Yes 935 (36.00) 256 (27.38) 679 (72.62) 896 (95.83) 39 (4.17)  Mother’s age, (years) Mean ± SD* 28.85±6.92 29.26±6.65 28.61±7.07 0.0161 28.84±6.87 29.14±8.31 0.9135  Mother’s education, n (%)**   No formal education 422 (16.25) 129 (30.57) 293 (69.43) 0.0005 409 (96.92) 13 (3.08) 0.8923   Primary 1817 (69.97) 677 (37.26) 1140 (62.74) 1765 (97.14) 52 (2.86)   Secondary and above 358 (13.79) 158 (44.13) 200 (55.87) 349 (97.49) 9 (2.51)  Maternal hemoglobin level, Mean ± SD* 12.78±1.75 13.10±1.75 12.59±1.72 <.0001 12.78±1.75 12.65±1.54 0.0200  Household wealth, n (%)**   Poorest 482 (18.56) 141 (29.25) 341 (70.75) <.0001 462 (95.85) 20 (4.15) 0.0581   Poor 603 (23.22) 205 (34.00) 398 (66.00) 583 (96.68) 20 (3.32)   Middle 590 (22.72) 209 (35.42) 381 (64.58) 571 (96.78) 19 (3.32)   Rich 513 (19.75) 211 (41.13) 302 (58.87) 503 (98.05) 10 (1.95)   Richest 409 (15.75) 198 (48.41) 211 (51.59) 404 (98.78) 5 (1.22) Community-level factors  Place of residence n (%)**   Urban 285 (10.97) 136 (47.72) 149 (52.28) <.0001 280 (98.25) 5 (1.75) 0.2390   Rural 2312 (89.03) 828 (35.81) 1484 (64.19) 2243 (97.02) 69 (2.98)  Geographical region n (%)**   Northern 427 (16.44) 186 (43.56) 241 (56.44) 0.0082 418 (97.89) 9 (2.11) 0.1440   Central 984 (37.89) 361 (36.69) 623 (63.31) 948 (96.34) 36 (3.66)   Southern 1186 (45.67) 417 (35.16) 769 (64.84) 1157 (97.55) 29 (2.45) Community wealthcMean ± SD* 37.85±29.32 37.83±28.32 30.35±29.16 <.0001 33.25±26.38 28.87±22.65 0.2174 Community female educationdMean ± SD* 84.48±14.77 85.89±13.46 82.11±15.15 <.0001 83.60±14.64 80.75±15.21 0.1382 Community water supplyeMean ± SD* 79.11±25.22 80.48±25.09 78.30±25.27 0.0347 79.04±25.35 81.60±20.49 0.4992 Community sanitation servicesfMean ± SD* 10.01±16.03 11.06±17.11 9.40±15.33 0.0040 9.96±16.03 11.85±16.31 0.1929 Characteristics Total Anemiaa Severe anemiab No Yes p-value No Yes p-value Individual-level factors Child age, months, Mean±SD* 31.51±14.91 36.42±13.69 28.62±14.85 <.0001 31.77±14.91 22.68±12.09 <.0001  Sex of the child, n (%)**   Male 1293 (49.79) 471 (36.43) 822 (63.57) 0.4668 1254 (96.98) 39 (3.02) 0.6109   Female 1304 (50.21) 493 (37.81) 811 (62.19) 1269 (97.32) 35 (2.68)  Birth order, Mean±SD* 3.67±2.32 3.62±2.25 3.70±2.35 0.7407 3.67±2.31 3.78±2.46 0.4648  Height-for-Age, Mean ± SD* −1.84±1.51 −1.75±1.40 −1.89±1.57 <.0001 −1.83±1.50 −1.99±1.07 0.8754  Weight-for-Age Mean ± SD* −0.84±1.08 −0.76±1.04 −0.88±1.10 0.0043 −0.83±1.25 1.05±1.20 0.4251  Vitamin A in the past 6 months, n (%)**   No 323 (12.44) 101 (31.27) 222 (68.73) 0.0200 315 (97.52) 8 (2.48) 0.6671   Yes 2274 (87.57) 863 (37.95) 1411 (62.05) 2208 (97.10) 66 (2.90)  Deworming in previous 6 months, n (%)**   No 698 (26.88) 219 (31.38) 479 (68.62) 0.0002 671 (96.13) 27 (3.87) 0.0585   Yes 1899 (73.12) 745 (39.23) 1154 (60.77) 1852 (97.53) 47 (2.47)  Diarrhea, n (%)**   No 2166 (83.40) 852 (39.34) 1314 (60.66) <.0001 2110 (97.41) 56 (2.59) 0.0698   Yes 431 (16.60) 112 (25.99) 319 (74.01) 413 (95.82) 18 (4.18)  Fever, n (%)**   No 1662 (64.00) 709 (42.60) 954 (57.40) <.0001 1627 (97.89) 35 (2.11) 0.0024   Yes 935 (36.00) 256 (27.38) 679 (72.62) 896 (95.83) 39 (4.17)  Mother’s age, (years) Mean ± SD* 28.85±6.92 29.26±6.65 28.61±7.07 0.0161 28.84±6.87 29.14±8.31 0.9135  Mother’s education, n (%)**   No formal education 422 (16.25) 129 (30.57) 293 (69.43) 0.0005 409 (96.92) 13 (3.08) 0.8923   Primary 1817 (69.97) 677 (37.26) 1140 (62.74) 1765 (97.14) 52 (2.86)   Secondary and above 358 (13.79) 158 (44.13) 200 (55.87) 349 (97.49) 9 (2.51)  Maternal hemoglobin level, Mean ± SD* 12.78±1.75 13.10±1.75 12.59±1.72 <.0001 12.78±1.75 12.65±1.54 0.0200  Household wealth, n (%)**   Poorest 482 (18.56) 141 (29.25) 341 (70.75) <.0001 462 (95.85) 20 (4.15) 0.0581   Poor 603 (23.22) 205 (34.00) 398 (66.00) 583 (96.68) 20 (3.32)   Middle 590 (22.72) 209 (35.42) 381 (64.58) 571 (96.78) 19 (3.32)   Rich 513 (19.75) 211 (41.13) 302 (58.87) 503 (98.05) 10 (1.95)   Richest 409 (15.75) 198 (48.41) 211 (51.59) 404 (98.78) 5 (1.22) Community-level factors  Place of residence n (%)**   Urban 285 (10.97) 136 (47.72) 149 (52.28) <.0001 280 (98.25) 5 (1.75) 0.2390   Rural 2312 (89.03) 828 (35.81) 1484 (64.19) 2243 (97.02) 69 (2.98)  Geographical region n (%)**   Northern 427 (16.44) 186 (43.56) 241 (56.44) 0.0082 418 (97.89) 9 (2.11) 0.1440   Central 984 (37.89) 361 (36.69) 623 (63.31) 948 (96.34) 36 (3.66)   Southern 1186 (45.67) 417 (35.16) 769 (64.84) 1157 (97.55) 29 (2.45) Community wealthcMean ± SD* 37.85±29.32 37.83±28.32 30.35±29.16 <.0001 33.25±26.38 28.87±22.65 0.2174 Community female educationdMean ± SD* 84.48±14.77 85.89±13.46 82.11±15.15 <.0001 83.60±14.64 80.75±15.21 0.1382 Community water supplyeMean ± SD* 79.11±25.22 80.48±25.09 78.30±25.27 0.0347 79.04±25.35 81.60±20.49 0.4992 Community sanitation servicesfMean ± SD* 10.01±16.03 11.06±17.11 9.40±15.33 0.0040 9.96±16.03 11.85±16.31 0.1929 a Hb < 11g/dL. b Hb < 7 g/dL; Hb = hemoglobin; SD = standard deviation. c Percentage of households in the community categorized as rich (upper 40% of quintiles). d Percentage of female in the community with primary and above education. e Percentage of household in the community with access to improved water sources. f Percentage of household in the community with access to improved sanitation facilities. * Groups were significantly different by univariate logistic regression, p < 0.05. ** Groups were significantly different by chi-square test, p < 0.05. Fig. 1. View largeDownload slide Distribution of hemoglobin concentration in children (6–59 months; n = 2597). Fig. 1. View largeDownload slide Distribution of hemoglobin concentration in children (6–59 months; n = 2597). Figure 2 shows the scatter plots of the relationship between children’s Hb levels and a selected subset of variables. Apart from the community safe drinking water, child's age, height-for-age, and community female education were positively associated with increased children’s Hb levels. Fig. 2. View largeDownload slide Scatter plots of children’s hemoglobin concentration against (A) age of child, (B) child’s height-for-age z-score, (C) mean percentage of community female education, and (D) mean percentage of community safe drinking water (n = 2597). Fig. 2. View largeDownload slide Scatter plots of children’s hemoglobin concentration against (A) age of child, (B) child’s height-for-age z-score, (C) mean percentage of community female education, and (D) mean percentage of community safe drinking water (n = 2597). Effects of individual-level and community-level factors on childhood anemia Table 3 showed the results of the multilevel analyses (Model 4). For Hb concentration, the results showed that age of the child (β = 0.2680), having no history of fever (β = 3.2172) and mothers’ Hb level (β = 0.0118) were positively associated with child Hb concentration, while residing in the poor households (β = −4.8415) and residing in communities with lower percentage of household with access to safe drinking water (β = −0.05863) were negatively associated with child Hb concentration. Table 3 Multilevel analysis of determinants of Hb concentration, anemia, and severe anemia in children aged 6–59 months in Malawi, n = 2597 Characteristics Hb concentration (g/dL) Anemia <11.0 g/dL Severe anemia <7.0 g/dL Coeff (SE) Coeff (SE) Coeff (SE) Individual-level factors  Sex of the child   Male/female 0.4025 (1.1756) 0.7321 0.03709 (0.09055 0.6822 0.2284 (0.3338) 0.4940  Child’s age (months) 0.2680 (0.04411) <.0001 −0.03816 (0.00352) <.0001 −0.07556 (0.01482) <.0001  Birth order −0.3100 (0.4937) 0.5302 0.03827 (0.03800) 0.3141 −0.2121 (0.1359) 0.1186  Diarrhea   No/yes 2.2290 (1.6806) 0.1849 −0.09258 (0.1356) 0.4950 0.3975 (0.4353) 0.3613  Fever   No/yes 3.2172 (1.2764) 0.0118 −0.5333 (0.1004) <.0001 −0.7595 (0.3525) 0.0313  Height-for-Age 0.003629 (0.00406) 0.3713 −0.00090 (0.00034) 0.0045 −0.00357 (0.00120) 0.0028  Weight-for-Age 0.002160 (0.00485) 0.6561 −0.00028 (0.00038) 0.4608 −0.00040 (0.00132) 0.7611  Vitamin A supplements   No/yes −2.3325 (1.8849) 0.2161 0.3062 (0.1520) 0.0442 −0.8281 (0.5742) 0.1494  Deworming past 6 months   No/yes 0.2403 (1.4736) 0.8705 −0.1470 (0.1156) 0.2037 0.05171 (0.3981) 0.8967  Mother’s age (years) 0.06043 (0.1654) 0.7149 −0.01366 (0.01275) 0.2841 0.09761 (0.0437) 0.0256  Mother’s education   No formal education −0.3020 (2.5229) 0.9047 0.2593 (0.1945) 0.1826 −0.04845 (0.7058) 0.9453   Primary 1.7588 (1.9060) 0.3562 0.07648 (0.1431) 0.5932 −0.01523 (0.5739) 0.9788  Maternal Hb level g/dL 0.0118 (0.9988) <.0001 −0.4397 (0.08304) <.0001 0.03039 (0.2682) 0.9098  Household wealth   Poorest −4.8415 (2.4753) 0.0497 0.3813 (0.1888) 0.0436 2.2544 (0.8753) 0.0101   Poor −2.8902 (2.3313) 0.2152 0.2362 (0.1756) 0.1787 1.7866 (0.8496) 0.0356   Middle −2.8914 (2.2707) 0.2031 0.2465 (0.1706) 0.1487 1.9761 (0.8362) 0.0182   Rich −1.5191 (2.1490) 0.4797 0.1595 (0.1603) 0.3201 1.3131 (0.8526) 0.1237 Community-level factors  Place of residence   Urban/Rural −0.3966 (2.7322) 0.8846 0.04310 (0.2026) 0.8316 −0.1990 (1.0815) 0.8540  Geographical region   Northern −1.2210 (1.9894) 0.5394 −0.05089 (0.1481) 0.7312 0.09739 (0.8002) 0.9031   Central 0.2794 (1.4080) 0.8427 −0.08515 (0.1070) 0.4263 0.5541 (0.5236) 0.2901  Community wealtha 0.07503 (0.04151) 0.0709 −0.0058 (0.00312) 0.0614 −0.00294 (0.01561) 0.8507  Community educationb 0.07394 (0.05281) 0.1617 −0.0099 (0.00411) 0.0159 0.004458 (0.01973) 0.8213  Community water supplyc −0.05863 (0.02680) 0.0288 −0.0003 (0.00205) 0.8813 0.01865 (0.01184) 0.1155  Community sanitation servicesd −0.08770 (0.04745) 0.0647 0.0033 (0.00353) 0.3517 0.02176 (0.01720) 0.2061 Measures of variation  Community-level variance   Area variance [τ(SE)] 38.5825 (17.6723) 0.0080 0.1696 (0.08952) 0.0113 16.3708 (5.2432) 0.0009   ICC (%) 2.28 4.90 83.2   PCV (%) 36.38 33.23 39.58  Model fit statistics   DIC (-2log likelihood) 24986.4 3096.15 570.10 Characteristics Hb concentration (g/dL) Anemia <11.0 g/dL Severe anemia <7.0 g/dL Coeff (SE) Coeff (SE) Coeff (SE) Individual-level factors  Sex of the child   Male/female 0.4025 (1.1756) 0.7321 0.03709 (0.09055 0.6822 0.2284 (0.3338) 0.4940  Child’s age (months) 0.2680 (0.04411) <.0001 −0.03816 (0.00352) <.0001 −0.07556 (0.01482) <.0001  Birth order −0.3100 (0.4937) 0.5302 0.03827 (0.03800) 0.3141 −0.2121 (0.1359) 0.1186  Diarrhea   No/yes 2.2290 (1.6806) 0.1849 −0.09258 (0.1356) 0.4950 0.3975 (0.4353) 0.3613  Fever   No/yes 3.2172 (1.2764) 0.0118 −0.5333 (0.1004) <.0001 −0.7595 (0.3525) 0.0313  Height-for-Age 0.003629 (0.00406) 0.3713 −0.00090 (0.00034) 0.0045 −0.00357 (0.00120) 0.0028  Weight-for-Age 0.002160 (0.00485) 0.6561 −0.00028 (0.00038) 0.4608 −0.00040 (0.00132) 0.7611  Vitamin A supplements   No/yes −2.3325 (1.8849) 0.2161 0.3062 (0.1520) 0.0442 −0.8281 (0.5742) 0.1494  Deworming past 6 months   No/yes 0.2403 (1.4736) 0.8705 −0.1470 (0.1156) 0.2037 0.05171 (0.3981) 0.8967  Mother’s age (years) 0.06043 (0.1654) 0.7149 −0.01366 (0.01275) 0.2841 0.09761 (0.0437) 0.0256  Mother’s education   No formal education −0.3020 (2.5229) 0.9047 0.2593 (0.1945) 0.1826 −0.04845 (0.7058) 0.9453   Primary 1.7588 (1.9060) 0.3562 0.07648 (0.1431) 0.5932 −0.01523 (0.5739) 0.9788  Maternal Hb level g/dL 0.0118 (0.9988) <.0001 −0.4397 (0.08304) <.0001 0.03039 (0.2682) 0.9098  Household wealth   Poorest −4.8415 (2.4753) 0.0497 0.3813 (0.1888) 0.0436 2.2544 (0.8753) 0.0101   Poor −2.8902 (2.3313) 0.2152 0.2362 (0.1756) 0.1787 1.7866 (0.8496) 0.0356   Middle −2.8914 (2.2707) 0.2031 0.2465 (0.1706) 0.1487 1.9761 (0.8362) 0.0182   Rich −1.5191 (2.1490) 0.4797 0.1595 (0.1603) 0.3201 1.3131 (0.8526) 0.1237 Community-level factors  Place of residence   Urban/Rural −0.3966 (2.7322) 0.8846 0.04310 (0.2026) 0.8316 −0.1990 (1.0815) 0.8540  Geographical region   Northern −1.2210 (1.9894) 0.5394 −0.05089 (0.1481) 0.7312 0.09739 (0.8002) 0.9031   Central 0.2794 (1.4080) 0.8427 −0.08515 (0.1070) 0.4263 0.5541 (0.5236) 0.2901  Community wealtha 0.07503 (0.04151) 0.0709 −0.0058 (0.00312) 0.0614 −0.00294 (0.01561) 0.8507  Community educationb 0.07394 (0.05281) 0.1617 −0.0099 (0.00411) 0.0159 0.004458 (0.01973) 0.8213  Community water supplyc −0.05863 (0.02680) 0.0288 −0.0003 (0.00205) 0.8813 0.01865 (0.01184) 0.1155  Community sanitation servicesd −0.08770 (0.04745) 0.0647 0.0033 (0.00353) 0.3517 0.02176 (0.01720) 0.2061 Measures of variation  Community-level variance   Area variance [τ(SE)] 38.5825 (17.6723) 0.0080 0.1696 (0.08952) 0.0113 16.3708 (5.2432) 0.0009   ICC (%) 2.28 4.90 83.2   PCV (%) 36.38 33.23 39.58  Model fit statistics   DIC (-2log likelihood) 24986.4 3096.15 570.10 Note: Hb, hemoglobin; Coeff, beta coefficient; [τ (SE)], community-level variance; SE, standard error; ICC, intraclass correlations; PCV, proportional change in variance; DIC, deviation information criterion. The Bold texts indicate a statistically significant association at a p-value less than 0.05. a Percentage of households in the community categorized as rich (upper 40% of quintiles). b Percentage of female in the community with primary and above education. c Percentage of household in the community with access to safe water supply. d  Percentage of household in the community with access to improved sanitation facilities. Table 3 Multilevel analysis of determinants of Hb concentration, anemia, and severe anemia in children aged 6–59 months in Malawi, n = 2597 Characteristics Hb concentration (g/dL) Anemia <11.0 g/dL Severe anemia <7.0 g/dL Coeff (SE) Coeff (SE) Coeff (SE) Individual-level factors  Sex of the child   Male/female 0.4025 (1.1756) 0.7321 0.03709 (0.09055 0.6822 0.2284 (0.3338) 0.4940  Child’s age (months) 0.2680 (0.04411) <.0001 −0.03816 (0.00352) <.0001 −0.07556 (0.01482) <.0001  Birth order −0.3100 (0.4937) 0.5302 0.03827 (0.03800) 0.3141 −0.2121 (0.1359) 0.1186  Diarrhea   No/yes 2.2290 (1.6806) 0.1849 −0.09258 (0.1356) 0.4950 0.3975 (0.4353) 0.3613  Fever   No/yes 3.2172 (1.2764) 0.0118 −0.5333 (0.1004) <.0001 −0.7595 (0.3525) 0.0313  Height-for-Age 0.003629 (0.00406) 0.3713 −0.00090 (0.00034) 0.0045 −0.00357 (0.00120) 0.0028  Weight-for-Age 0.002160 (0.00485) 0.6561 −0.00028 (0.00038) 0.4608 −0.00040 (0.00132) 0.7611  Vitamin A supplements   No/yes −2.3325 (1.8849) 0.2161 0.3062 (0.1520) 0.0442 −0.8281 (0.5742) 0.1494  Deworming past 6 months   No/yes 0.2403 (1.4736) 0.8705 −0.1470 (0.1156) 0.2037 0.05171 (0.3981) 0.8967  Mother’s age (years) 0.06043 (0.1654) 0.7149 −0.01366 (0.01275) 0.2841 0.09761 (0.0437) 0.0256  Mother’s education   No formal education −0.3020 (2.5229) 0.9047 0.2593 (0.1945) 0.1826 −0.04845 (0.7058) 0.9453   Primary 1.7588 (1.9060) 0.3562 0.07648 (0.1431) 0.5932 −0.01523 (0.5739) 0.9788  Maternal Hb level g/dL 0.0118 (0.9988) <.0001 −0.4397 (0.08304) <.0001 0.03039 (0.2682) 0.9098  Household wealth   Poorest −4.8415 (2.4753) 0.0497 0.3813 (0.1888) 0.0436 2.2544 (0.8753) 0.0101   Poor −2.8902 (2.3313) 0.2152 0.2362 (0.1756) 0.1787 1.7866 (0.8496) 0.0356   Middle −2.8914 (2.2707) 0.2031 0.2465 (0.1706) 0.1487 1.9761 (0.8362) 0.0182   Rich −1.5191 (2.1490) 0.4797 0.1595 (0.1603) 0.3201 1.3131 (0.8526) 0.1237 Community-level factors  Place of residence   Urban/Rural −0.3966 (2.7322) 0.8846 0.04310 (0.2026) 0.8316 −0.1990 (1.0815) 0.8540  Geographical region   Northern −1.2210 (1.9894) 0.5394 −0.05089 (0.1481) 0.7312 0.09739 (0.8002) 0.9031   Central 0.2794 (1.4080) 0.8427 −0.08515 (0.1070) 0.4263 0.5541 (0.5236) 0.2901  Community wealtha 0.07503 (0.04151) 0.0709 −0.0058 (0.00312) 0.0614 −0.00294 (0.01561) 0.8507  Community educationb 0.07394 (0.05281) 0.1617 −0.0099 (0.00411) 0.0159 0.004458 (0.01973) 0.8213  Community water supplyc −0.05863 (0.02680) 0.0288 −0.0003 (0.00205) 0.8813 0.01865 (0.01184) 0.1155  Community sanitation servicesd −0.08770 (0.04745) 0.0647 0.0033 (0.00353) 0.3517 0.02176 (0.01720) 0.2061 Measures of variation  Community-level variance   Area variance [τ(SE)] 38.5825 (17.6723) 0.0080 0.1696 (0.08952) 0.0113 16.3708 (5.2432) 0.0009   ICC (%) 2.28 4.90 83.2   PCV (%) 36.38 33.23 39.58  Model fit statistics   DIC (-2log likelihood) 24986.4 3096.15 570.10 Characteristics Hb concentration (g/dL) Anemia <11.0 g/dL Severe anemia <7.0 g/dL Coeff (SE) Coeff (SE) Coeff (SE) Individual-level factors  Sex of the child   Male/female 0.4025 (1.1756) 0.7321 0.03709 (0.09055 0.6822 0.2284 (0.3338) 0.4940  Child’s age (months) 0.2680 (0.04411) <.0001 −0.03816 (0.00352) <.0001 −0.07556 (0.01482) <.0001  Birth order −0.3100 (0.4937) 0.5302 0.03827 (0.03800) 0.3141 −0.2121 (0.1359) 0.1186  Diarrhea   No/yes 2.2290 (1.6806) 0.1849 −0.09258 (0.1356) 0.4950 0.3975 (0.4353) 0.3613  Fever   No/yes 3.2172 (1.2764) 0.0118 −0.5333 (0.1004) <.0001 −0.7595 (0.3525) 0.0313  Height-for-Age 0.003629 (0.00406) 0.3713 −0.00090 (0.00034) 0.0045 −0.00357 (0.00120) 0.0028  Weight-for-Age 0.002160 (0.00485) 0.6561 −0.00028 (0.00038) 0.4608 −0.00040 (0.00132) 0.7611  Vitamin A supplements   No/yes −2.3325 (1.8849) 0.2161 0.3062 (0.1520) 0.0442 −0.8281 (0.5742) 0.1494  Deworming past 6 months   No/yes 0.2403 (1.4736) 0.8705 −0.1470 (0.1156) 0.2037 0.05171 (0.3981) 0.8967  Mother’s age (years) 0.06043 (0.1654) 0.7149 −0.01366 (0.01275) 0.2841 0.09761 (0.0437) 0.0256  Mother’s education   No formal education −0.3020 (2.5229) 0.9047 0.2593 (0.1945) 0.1826 −0.04845 (0.7058) 0.9453   Primary 1.7588 (1.9060) 0.3562 0.07648 (0.1431) 0.5932 −0.01523 (0.5739) 0.9788  Maternal Hb level g/dL 0.0118 (0.9988) <.0001 −0.4397 (0.08304) <.0001 0.03039 (0.2682) 0.9098  Household wealth   Poorest −4.8415 (2.4753) 0.0497 0.3813 (0.1888) 0.0436 2.2544 (0.8753) 0.0101   Poor −2.8902 (2.3313) 0.2152 0.2362 (0.1756) 0.1787 1.7866 (0.8496) 0.0356   Middle −2.8914 (2.2707) 0.2031 0.2465 (0.1706) 0.1487 1.9761 (0.8362) 0.0182   Rich −1.5191 (2.1490) 0.4797 0.1595 (0.1603) 0.3201 1.3131 (0.8526) 0.1237 Community-level factors  Place of residence   Urban/Rural −0.3966 (2.7322) 0.8846 0.04310 (0.2026) 0.8316 −0.1990 (1.0815) 0.8540  Geographical region   Northern −1.2210 (1.9894) 0.5394 −0.05089 (0.1481) 0.7312 0.09739 (0.8002) 0.9031   Central 0.2794 (1.4080) 0.8427 −0.08515 (0.1070) 0.4263 0.5541 (0.5236) 0.2901  Community wealtha 0.07503 (0.04151) 0.0709 −0.0058 (0.00312) 0.0614 −0.00294 (0.01561) 0.8507  Community educationb 0.07394 (0.05281) 0.1617 −0.0099 (0.00411) 0.0159 0.004458 (0.01973) 0.8213  Community water supplyc −0.05863 (0.02680) 0.0288 −0.0003 (0.00205) 0.8813 0.01865 (0.01184) 0.1155  Community sanitation servicesd −0.08770 (0.04745) 0.0647 0.0033 (0.00353) 0.3517 0.02176 (0.01720) 0.2061 Measures of variation  Community-level variance   Area variance [τ(SE)] 38.5825 (17.6723) 0.0080 0.1696 (0.08952) 0.0113 16.3708 (5.2432) 0.0009   ICC (%) 2.28 4.90 83.2   PCV (%) 36.38 33.23 39.58  Model fit statistics   DIC (-2log likelihood) 24986.4 3096.15 570.10 Note: Hb, hemoglobin; Coeff, beta coefficient; [τ (SE)], community-level variance; SE, standard error; ICC, intraclass correlations; PCV, proportional change in variance; DIC, deviation information criterion. The Bold texts indicate a statistically significant association at a p-value less than 0.05. a Percentage of households in the community categorized as rich (upper 40% of quintiles). b Percentage of female in the community with primary and above education. c Percentage of household in the community with access to safe water supply. d  Percentage of household in the community with access to improved sanitation facilities. For childhood anemia, the results showed that child’s age (β = −0.03816), having no history of fever (β = −0.4892), height-for-age (β = −0.00090), maternal Hb level (β = −0.4397) and residing in communities with higher percentage of women with primary and above education (β = −0.0099) were negatively associated with childhood anemia, whereas residing in the poorest households (β = 0.3813) and having no vitamin A supplements (β = 0.3062) were positively associated with childhood anemia. For severe anemia, the results showed that child’s age (β = −0.07556), having no history of fever (β = −0.7595) and height-for-age (β = −0.00357) were negatively associated with childhood anemia, while maternal age (β = 0.09761) and residing in poor households were positively associated with severe anemia. The ICCs show that about 2%, 5% and 83% of community-level variance were unexplained for Hb concentration, childhood anemia and severe anemia, respectively, which shows the community-level characteristics included in this study cannot explain most of the community-level variance especially in severe anemia. The PCVs show that 36%, 33% and 40% of the variance in Hb concentration, anemia and severe anemia across communities were respectively explained by individual-level and community-level factors. DISCUSSION This is the first national and multilevel study that has ever been conducted to assess risk factors for childhood anemia in Malawi. We have found out that the likelihood of anemia increased in younger children. This result is consistent with previous findings [3, 4, 38–40]. Anemia is common among children around the time of the growth spurt, particularly between 6 and 23 months. If exogenous iron supplement is inadequate during this time, infant anemia occurs easily [38, 41–43]. Infants are known to be susceptible to infections via contaminated water and food, which can also affect their ability to absorb iron. We also found that fever and stunting were positively associated with our outcomes. Similar findings were reported in previous studies [3, 18, 19, 38, 44]. Fever in under 5 children in Malawi is a common symptom of acute and chronical inflammatory diseases especially malaria [12, 18]. Previous studies have reported that malaria increases extravascular hemolysis of red blood cells with a concomitant failure of the bone marrow to increase red cell production to compensate for these losses [45, 46, 47]. On the other hand, stunting indicates chronic food shortage and long-term effects of micronutrient deficiencies, which are associated with low concentrations of Hb [48]. In our study, we have shown that supplement of the high dose of vitamin A can be a solution to childhood anemia [49–52]. Our findings suggested that infants born to anemic mothers from economically poor households are more likely to develop anemia, which is in line with previous studies [18, 19, 38, 42]. Poor family wealth is an indicator of conditions of extreme social deprivation; hence, caregivers might face difficulties in providing nutritious food to their children [4, 18, 42]. Additionally, maternal anemia status may lead to poor stores of iron, zinc, vitamin A and B12 and folate in breast milk, which further impact child anemia status [53]. Our study showed an important result that community-level woman’s education was significantly associated with childhood anemia. Possible explanations may be that higher community education provides a context where women are more likely to obtain knowledge and material resources that can benefit a child's health [54]. Thus, our findings provide important evidence that contextual factors sometimes can have significant impacts on a child's health above and beyond individual-level factors. Surprisingly communities with safe drinking water were negatively associated with Hb concentration. Future studies need to explore more about the relationship between water sources and Hb concentration. Regarding the ICCs in our study, 7%, 89% and 5% of the total variance of childhood anemia, severe anemia and children’s hemoglobin could be attributed to community-level factors, respectively. Previous research on multilevel analyses have reported similar small ICCs; however, researchers also indicated that an ICC at or above 2% is suggestive of significant group-level variance that required a multilevel study design [55, 56]. The use of a national representative sample is a main strength of this study. However, our study has some limitations. First, the use of a cross-sectional study design did not allow us to establish cause-and-effect relationships. Second, our study relied on Hb as the measure of anemia; further studies should consider other red blood cell indices [57]. Third, there might be recall bias, as the occurrences of diarrhea and fever were obtained by self-reports. Our results, like other survey studies, are prone to the interviewer bias (i.e., social desirability effect). Last, we are unable to adjust all confounding factors because this study is based on a secondary data analysis. CONCLUSIONS Our study results indicated that individual-level factors have stronger effects than community-level factors on childhood anemia, severe anemia and Hb concentration in Malawi. However, community-level female education still plays an important role in childhood anemia. Public health interventions targeting childhood anemia should focus on food insecurity, malaria infection, iron and other micronutrient deficiencies, household living standards, as well as community female education. ACKNOWLEDGEMENT The authors are sincerely grateful to National Statistical Office (NSO) and the Community Health Sciences Unit (CHSU) of Malawi for data collection. We give thanks to the MEASURE DHS for providing us with the population-based dataset through their archives which can be downloaded from http://dhsprogram.com/data/available-datasets.cfm. FUNDING The Ministry of Science and Technology, Taiwan [grant number: Most 103-2410-H-038-002]. REFERENCES 1 WHO . Haemoglobin concentrations for the diagnosis of anaemia and assessment of severity. Vitamin and mineral nutrition information system . Geneva : World Health Organization , 2011 (WHO/NMH/NHD/MNM/11.1). http://www.who.int/vmnis/indicators/haemoglobin.pdf (7 February 2016, date last accessed). 2 Gutema B , Adissu W , Asress Y , et al. Anemia and associated factors among school-age children in Filtu Town, Somali region, Southeast Ethiopia . BMC Hematol 2014 ; 14 : 13 . Google Scholar CrossRef Search ADS PubMed 3 Iannotti LL , Delnatus JR , Odom AR , et al. Determinants of anemia and hemoglobin concentration in Haitian school-aged children . Am J Trop Med Hyg 2015 ; 93 : 1092 – 8 . Google Scholar CrossRef Search ADS PubMed 4 Gao W , Yan H , Duolao W , et al. Severity of anemia among children under 36 months old in Rural Western China . PLoS One 2013 ; 8 : 2 – 7 . 5 Simbauranga RH , Kamugisha E , Hokororo A , et al. Prevalence and factors associated with severe anaemia amongst under-five children hospitalized at Bugando Medical Centre, Mwanza, Tanzania . BMC Hematol 2015 ; 15 : 13 . Google Scholar CrossRef Search ADS PubMed 6 Magalhães RJS , Clements ACA. Mapping the risk of anaemia in preschool-age children: the contribution of malnutrition, malaria, and helminth infections in West Africa . PLoS Med 2011 ; 8 : e1000438 . Google Scholar CrossRef Search ADS PubMed 7 WHO . The Global Prevalence of Anaemia in 2011 . Geneva : World Health Organization , 2015 . http://apps.who.int/iris/bitstream/10665/177094/1/9789241564960_eng.pdf (5 February 2016, date last accessed). 8 National Statistical Office (NSO), ICF Macro . Malawi demographic and health survey 2010. Zomba, Malawi: NSO and Calverton, MD: ICF Macro, 2011 . http://www.dhsprogram.com/pubs/pdf/FR247/FR247.pdf (11 August 2016, date last accessed). 9 WHO/UNICEF . Joint statement by the World Health Organization and the United Nations Children’s Fund. Focusing on anaemia, towards an integrated approach for effective anaemia control. http://www.who.int/nutrition/publications/micronutrients/WHOandUNICEF_statement_anaemia_en.pdf?ua=1 (7 February 2016, date last accessed). 10 Shill KB , Karmakar P , Kibria MG , et al. Prevalence of iron-deficiency anaemia among university students in Noakhali Region, Bangladesh . J Heal Popul Nutr 2014 ; 32 : 103 – 10 . 11 Semba RD , Bloem MW. The anemia of vitamin A deficiency: epidemiology and pathogenesis . Eur J Clin Nutr 2002 ; 56 : 271 – 81 . Google Scholar CrossRef Search ADS PubMed 12 Calis JCJ , Phiri KS , Faragher EB , et al. Severe anemia in Malawian children . N Engl J Med 2008 ; 358 : 888 – 99 . Google Scholar CrossRef Search ADS PubMed 13 Williams TN , Uyoga S , Macharia A , et al. Bacteraemia in Kenyan children with sickle-cell anaemia: a retrospective cohort and case-control study . Lancet 2009 ; 374 : 1364 – 70 . Google Scholar CrossRef Search ADS PubMed 14 Crawley J. Reducing the burden of anemia in infants and young children in malaria-endemic countries of Africa: from evidence to action . Am J Trop Med Hyg 2004 ; 71 : 25 – 34 . Google Scholar CrossRef Search ADS PubMed 15 Morris CR , Singer ST , Walters MC. Clinical hemoglobinopathies: iron, lungs and new blood . Curr Opin Hematol 2006 ; 13 : 407 – 18 . Google Scholar PubMed 16 Zuffo CRK , Osorio MM , Taconeli CA , et al. Prevalence and risk factors of anemia in children . J Pediatr 2016 ; 92 : 353 – 60 . http://dx.doi.org/10.1016/j.jped.2015.09.007. Google Scholar CrossRef Search ADS 17 Cardoso MA , Scopel KKG , Muniz PT , et al. Underlying factors associated with anemia in amazonian children: a population-based, cross-sectional study . PLoS One 2012 ; 7 : e36341 . Google Scholar CrossRef Search ADS PubMed 18 Ngwira A , Kazembe LN. Bayesian random effects modelling with application to childhood anaemia in Malawi . BMC Public Health 2015 ; 15 : 161 . Google Scholar CrossRef Search ADS PubMed 19 Ngnie-Teta I , Receveur O , Kuate-Defo B. Risk factors for moderate to severe anemia among children in Benin and Mali: insights from a multilevel analysis . Food Nutr Bull 2007 ; 28 : 76 – 89 . Google Scholar CrossRef Search ADS PubMed 20 Siekmans K , Receveur O , Haddad S. Can an integrated approach reduce child vulnerability to anaemia? Evidence from three African countries . PLoS One 2014 ; 9 : e90108 . Google Scholar CrossRef Search ADS PubMed 21 Oladeinde BH , Omoregie R , Olley M , et al. Malaria and anemia among children in a low resource setting in Nigeria . Iran J Parasitol 2012 ; 7 : 31 – 7 . Google Scholar PubMed 22 Yang W , Li X , Li Y , et al. Anemia, malnutrition and their correlations with socio-demographic characteristics and feeding practices among infants aged 0-18 months in rural areas of Shaanxi province in northwestern China: a cross-sectional study . BMC Public Health 2012 ; 12 : 1127 . Google Scholar CrossRef Search ADS PubMed 23 Foote EM , Sullivan KM , Ruth LJ , et al. Determinants of anemia among preschool children in Rural, Western Kenya . Am J Trop Med Hyg 2013 ; 88 : 757 – 64 . Google Scholar CrossRef Search ADS PubMed 24 Uthman OA. A multilevel analysis of individual and community effect on chronic childhood malnutrition in rural Nigeria . J Trop Pediatr 2009 ; 55 : 109 – 15 . Google Scholar CrossRef Search ADS PubMed 25 Merlo J , Yang M , Chaix B , et al. A brief conceptual tutorial on multilevel analysis in social epidemiology: investigating contextual phenomena in different groups of people . J Epidemiol Community Health 2005 ; 59 : 729 – 36 . Google Scholar CrossRef Search ADS PubMed 26 Roux AVD. A glossary for multilevel analysis . J Epidemiol Community Health 2002 ; 56 : 588 – 94 . Google Scholar CrossRef Search ADS PubMed 27 Ntenda PAM , Chuang KY , Tiruneh FN , et al. Analysis of the effects of individual and community level factors on childhood immunization in Malawi . Vaccine 2017 ; 35 : 1907 – 17 . Google Scholar CrossRef Search ADS PubMed 28 Wk B , He K , Lh B. Hemocue–a new haemoglobinometer in the clinic . Eur J Anaesthesiol 1991 ; 8 : 55 – 8 . Google Scholar PubMed 29 WHO, Multicentre Growth Reference Study Group . WHO child growth standards: length/height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass index-for-age: methods and development . Geneva : World Health Organization , 2006 . http://www.who.int/childgrowth/standards/technical_report/en/ (7 February 2016, date last accessed). 30 Onis M. WHO child growth standards: length/height-for-age, weight-for-age, weight-for-length, weight-forheight and body mass index-for-age: methods and development: methods and development. Geneva, Switzerland . Acta Pædiatrica 2006 ; 95 : 76 – 85 . 31 WHO . Guideline: vitamin A supplementation in infants and children 6 – 59 months of age. 2011 . http://apps.who.int/iris/bitstream/10665/44664/1/9789241501767_eng.pdf. 32 NHS . Choices, your health, your choices, fever in children. 2015 . http://www.nhs.uk/conditions/feverchildren/Pages/Introduction.aspx (8 May 2017, date last accessed). 33 World Health Organization (WHO), United Nations Children’s Fund (UNICEF) . Diarrhoea: why children are still dying and what can be done. 2009 . http://apps.who.int/iris/bitstream/10665/44174/1/9789241598415_eng.pdf (8 May 2017, date last accessed). 34 Rutstein SO , Johnson K. The DHS wealth index. DHS Comparative Reports No. 6. Calverton, MD: ORC Macro, 2004 . 35 The DHS Program . Demographic and health survey, wealth index. ICF, 530 Gaither Road, Suite 500, Rockville, MD 20850. http://www.dhsprogram.com/topics/wealth-index/Index.cfm. 36 World Health Organization (WHO), United Nations Children’s Fund (UNICEF) . WHO and UNICEF Joint Monitoring Programme (JMP) for water supply and sanitation: types of drinking-water sources and sanitation. Geneva, Switzerland. http://www.wssinfo.org/definitions-methods/watsan-categories/ (28 December 2015, date last accessed). 37 SAS Institute Inc . Base SAS® 9.4 Procedures Guide: Statistical Procedures . 2nd edn. Cary, NC : SAS Institute Inc. USA , 2013 , 535 . 38 Zhao A , Zhang Y , Peng Y , et al. Prevalence of anemia and its risk factors among children 6–36 months old in Burma . Am J Trop Med Hyg 2012 ; 87 : 306 – 11 . Google Scholar CrossRef Search ADS PubMed 39 Austin AM , Fawzi W , Hill AG. Anaemia among Egyptian children between 2000 and 2005: trends and predictors . Matern Child Nutr 2012 ; 8 : 522 – 32 . Google Scholar CrossRef Search ADS PubMed 40 Semedo RM , Santos MM , Baião MR , et al. Prevalence of anaemia and associated factors among children below five years of age in Cape Verde, West Africa . J Health Popul Nutr 2014 ; 32 : 646 – 57 . Google Scholar PubMed 41 Kamal MM , Hasan MM , Davey R. Determinants of childhood morbidity in Bangladesh: evidence from the Demographic and Health Survey 2011 . BMJ Open 2015 ; 5 : e007538. Google Scholar CrossRef Search ADS PubMed 42 Woldie H , Kebede Y , Tariku A. Factors associated with anemia among children aged 6-23 months attending growth monitoring at Tsitsika Health Center, Wag-Himra Zone, Northeast Ethiopia . J Nutr Metab 2015 ; 2015 : e928632. Google Scholar CrossRef Search ADS 43 Osório MM , Lira PI , Batista-Filho M , et al. Prevalence of anemia in children 6-59 months old in the state of Pernambuco, Brazil . Rev Panam Salud Publica 2001 ; 10 : 101 – 7 . Google Scholar CrossRef Search ADS PubMed 44 Konstantyner T , Roma Oliveira TC , de Aguiar Carrazedo Taddei JA. Risk factors for anemia among Brazilian infants from the 2006 National Demographic Health Survey . Anemia 2012 ; 2012 : e850681 . Google Scholar CrossRef Search ADS 45 Kai OK , Roberts DJ. The pathophysiology of malarial anaemia: where have all the red cells gone ? 2008 ; 4 : 4 – 7 . 46 Phillips RE , Pasvol G. Anaemia of Plasmodium falciparum malaria . Baillieres Clin Haematol 1992 ; 5 : 315 – 30 . Google Scholar CrossRef Search ADS PubMed 47 Weiss G , Goodnough LT. Anemia of chronic disease . N Engl J Med 2005 ; 352 : 1011 – 23 . Google Scholar CrossRef Search ADS PubMed 48 Khan JR , Awan N , Misu F. Determinants of anemia among 6–59 months aged children in Bangladesh: evidence from nationally representative data . BMC Pediatr 2016 ; 16 : 3 . Google Scholar CrossRef Search ADS PubMed 49 Mwanri L , Worsley A , Ryan P , et al. Supplemental vitamin A improves anemia and growth in anemic school children in Tanzania . J Nutr 2000 ; 130 : 2691 – 6 . Google Scholar CrossRef Search ADS PubMed 50 Garg A , Abrol P , Tewari AD , et al. Effect of vitamin A supplementation on hematopoisis in children with anemia . Indian J Clin Biochem 2005 ; 20 : 85 – 6 . Google Scholar CrossRef Search ADS PubMed 51 Gebremedhin S. Effect of a single high dose vitamin A supplementation on the hemoglobin status of children aged 6 – 59 months: propensity score matched retrospective cohort study based on the data of Ethiopian Demographic and Health Survey 2011 . BMC Pediatr 2014 ; 14 : 79 . Google Scholar CrossRef Search ADS PubMed 52 Zimmermann MB , Biebinger R , Rohner F , et al. Vitamin A supplementation in children with poor vitamin A and iron status increases erythropoietin and hemoglobin concentrations without changing total body iron . Am J Clin Nutr 2006 ; 84 : 580 – 6 . Google Scholar CrossRef Search ADS PubMed 53 Wang J , Wang H , Chang S , et al. The influence of malnutrition and micronutrient status on anemic risk in children under 3 years old in poor areas in China . PLoS One 2015 ; 10 : e0140840 . Google Scholar CrossRef Search ADS PubMed 54 National Research Council and Institute of Medicine . Children’s Health, the Nation’s Wealth: Assessing and Improving Child Health. Committee on Evaluation of Children’s Health. Board on Children, Youth, and Families, Division of Behavioral and Social Sciences and Education . Washington, DC . The National Academies Press , 2004 , 28 – 44 . 55 Magadi MA. Household and community HIV/AIDS status and child malnutrition in sub-Saharan Africa: evidence from the demographic and health surveys . Soc Sci Med 2011 ; 73 : 436 – 46 . Google Scholar CrossRef Search ADS PubMed 56 Theall KP , Scribner R , Broyles S , et al. Impact of small group size on neighbourhood influences in multilevel models . J Epidemiol Community Health 2011 ; 65 : 688 – 95 . Google Scholar CrossRef Search ADS PubMed 57 Coyer SM. Anemia: diagnosis and management . J Pediatr Heal Care 2005 ; 19 : 380 – 5 . Google Scholar CrossRef Search ADS © The Author [2017]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

2018 Journal of Tropical Pediatrics

doi: 10.1093/tropej/fmx060pmid: 28985427

Abstract Chronic abdominal pain has many etiologies, one of them being parasites. The aim of this study was to find an association between chronic abdominal pain in children and Blastocystis hominis (Bh). Clinical files of patients with Bh and functional abdominal pain were reviewed. A comparison was made between patients who showed an improvement of their symptoms and those who did not. Out of the 138 patients who had functional abdominal pain and Bh, 37 patients did not receive any treatment (26.8%), while 101 received it and were treated with different antimicrobial agents (73.2%); regarding the improvement of symptoms, a statistically significant difference (p < 0.001) was observed. Chronic abdominal pain in children has different etiologies; however, we have documented through this work that it is appropriate to provide antimicrobial treatment for patients with Bh and chronic abdominal pain. Blastocystis hominis, chronic abdominal pain, irritable bowel syndrome, children, functional abdominal pain BACKGROUND Functional abdominal pain is one of the most frequent causes for consultation in pediatric routine clinical practice. However, its concept per se is unclear as well as its pathophysiological mechanisms [1]. In 1958, J. Apley defined chronic abdominal pain as at least three episodes of abdominal pain, severe enough to affect their daily activities over a period of 3 months [2]. This concept has been modified through time and, in the Rome III Criteria, it is considered within the functional gastrointestinal disorders (FGID) defined as an episodic or continuous abdominal pain for which there is no evidence of an inflammatory, anatomic, metabolic or neoplastic disease to explain the subject’s symptoms, occurring at least once a week during the last 2 months before its diagnosis [3]. Some parasites such as Giardia lamblia and Entamoeba histolytica are well described as a cause of chronic abdominal pain in children, and Blastocystis hominis (Bh) is a controversial parasite, as it has been found in asymptomatic individuals [4, 5] and its actual role in abdominal pain is still unclear. Blastocystis hominis has been identified as a potential pathogenic agent responsible for FGID, namely, irritable bowel syndrome [6–8]. Although its pathophysiological mechanism is not yet established, an alteration of the intestinal permeation owing to pro-inflammatory cytokines leading to visceral inflammation and hypersensitivity has been proposed [9]. However, the actual role of Bh as a cause of functional abdominal pain in children has not been studied. Therefore, the aim of this study was to find an association between functional abdominal pain and Bh. MATERIAL AND METHODS This was a retrospective, comparative and analytical study. A search was conducted in the electronic capture system of the National Institute of Pediatrics for all stool analyses positive for Bh between January 2003 and October 2015. The reasons for requesting an exam for these samples were screened and only those samples matching a gastroenterological diagnosis were selected. The clinical files of patients matching a referral diagnosis of chronic abdominal pain were reviewed. Functional abdominal pain Clinical files of referred patients with functional abdominal pain were reviewed and classified with functional abdominal pain if they had abdominal pain twice a week in the last 2 months before diagnosis. Stool analyses All stool analyses were conducted at the Parasitology Laboratory, using the FAUST technique. They were considered positive if any of the sent samples were positive for Bh. Response to treatment Treatment response was considered adequate when the patient had no symptoms in the control consultation visit and when three control stool analyses were negative. Files meeting these criteria were used for extracting information about gender, age, treatment, response to established management and control stool analyses results. Those cases where chronic abdominal pain criteria were not met and with incomplete information in the clinical files were excluded. STATISTICAL ANALYSIS The statistical analysis was conducted using the SPSS V21.0 software, through descriptive statistics for demographic endpoints and a chi-squared test for comparing proportions. A p value <0.05 was considered significant. RESULTS A total of 9637 samples in the database were positive for this microorganism, 2425 (28%) of which were sent with diagnoses related to gastrointestinal origin pathologies. A total of 322 samples (13%), corresponding to 166 patients, had a diagnosis of functional abdominal pain; only those of 138 patients were included, as 28 of them did not meet the inclusion criteria. Of the 138 patients with functional abdominal pain, 80 were female (58%) and 58 were male (42%). The mean age of patients was 10.96 (SD + 3.5). Regarding their management, 37 patients did not receive antimicrobial treatment (6.8%), while 101 received different types of antimicrobial agents (73.2%). Of the 101 patients who received treatment, 69 improved, while 32 continued presenting symptoms; of the 37 patients who did not receive treatment, 13 improved, while 24 continued presenting symptoms. When comparing the improvement of symptoms of patients who received treatment vs. that of patients who did not receive antimicrobial therapy, a statistically significant difference (p < 0.001) was found (Fig. 1). Fig. 1. View largeDownload slide Response to treatment comparing patients who receive treatment with those who did not receive. Fig. 1. View largeDownload slide Response to treatment comparing patients who receive treatment with those who did not receive. Table 1 shows a summary of received treatments and response to them. The mostly used treatment was metronidazole (26% patients), followed by the combination of tinidazole and mebendazole (24.6% patients). When comparing the type of treatment and symptoms improvement, secnidazole and the combination of tinidazole/mebendazole showed a similar statistical difference (p < 0.001), while nitazoxanide had less statistical difference, but still significant (p = 0.04). Metronidazole, albendazole, rifaximin and sulfamethoxazole/trimethoprim had no statistically significant differences (Fig. 2). Table 1 Antimicrobial therapy and improvement of symptoms Antibiotic Improvement of symptoms p value No Yes Albendazole 0 1 0.185 Tinidazole/mebendazole 9 25 0.001 Metronidazole 16 20 0.08 Nitazoxanide 1 8 0.004 Sulfamethoxazole/trimethoprim 1 0 0.465 Secnidazole 3 15 0.001 Rifaximin 1 0 0.465 Antibiotic Improvement of symptoms p value No Yes Albendazole 0 1 0.185 Tinidazole/mebendazole 9 25 0.001 Metronidazole 16 20 0.08 Nitazoxanide 1 8 0.004 Sulfamethoxazole/trimethoprim 1 0 0.465 Secnidazole 3 15 0.001 Rifaximin 1 0 0.465 Table 1 Antimicrobial therapy and improvement of symptoms Antibiotic Improvement of symptoms p value No Yes Albendazole 0 1 0.185 Tinidazole/mebendazole 9 25 0.001 Metronidazole 16 20 0.08 Nitazoxanide 1 8 0.004 Sulfamethoxazole/trimethoprim 1 0 0.465 Secnidazole 3 15 0.001 Rifaximin 1 0 0.465 Antibiotic Improvement of symptoms p value No Yes Albendazole 0 1 0.185 Tinidazole/mebendazole 9 25 0.001 Metronidazole 16 20 0.08 Nitazoxanide 1 8 0.004 Sulfamethoxazole/trimethoprim 1 0 0.465 Secnidazole 3 15 0.001 Rifaximin 1 0 0.465 Fig. 2. View largeDownload slide Improvement of symptoms according to the antimicrobial therapy. Note: A = albendazole; M = mebendazole/tinidazole; MET = metronidazole; NIT = nitazoxanide; RIF = rifaximin; S = secnidazole; TMP/SMX = trimethroprim/sulfamethoxazole. Fig. 2. View largeDownload slide Improvement of symptoms according to the antimicrobial therapy. Note: A = albendazole; M = mebendazole/tinidazole; MET = metronidazole; NIT = nitazoxanide; RIF = rifaximin; S = secnidazole; TMP/SMX = trimethroprim/sulfamethoxazole. DISCUSSION Functional abdominal pain is one of the most frequent causes for overall pediatric consultation (2–4%). However, it has only been acknowledged as a global health problem until recently. It is estimated that 13–17% of school-age children suffer from it [10]. The incidence of FGID in children and adolescents is approximately 23% [11] and it is considered more common among adolescents (7–25%), with a higher frequency in women than in men [10]; this is consistent with the findings in our study, in which the female population represented 58% of the studied sample, with a mean age of 10.96 years. FGID represent a serious problem considering the parents anguish, multiple hospitalizations, economic expenditure and numerous exams required to rule out organic pathologies. The pathophysiology of FGIDs in children is still unclear. However, it is possible that there are several related factors such as a disrupted intestinal reactivity in response to luminal or psychological stimuli, visceral afferent hypersensitivity and intestinal hypersensitivity to pain [12]. Low-grade persisting inflammation plays an important role in the pathophysiology of this disease and may occur in the deep layers of intestinal tissue, as proven by Torn Blom et al. [13] who showed the presence of periganglionic lymphocytic infiltration in the myenteric plexus. As previously commented, Bh has been proposed as an etiological agent for FGIDs, mainly for irritable bowel syndrome, as well as for extraintestinal manifestations such as chronic angioedema [14]. This agent is an intestinal unicellular and polymorphic eukaryote parasite, one of the intestinal protozoans mostly found in human and mammal gastrointestinal tracts, with a natural habitat in the colon and cecum [15]. In a study published in 2010, the prevalence of Bh in the mountain chain of Veracruz, Mexico was 80% [16]. As well as other parasites, its presence is associated to a limited access to health services, bad personal hygiene, exposure to domestic animals and consumption of contaminated water and food [17]. Reports are contradictory regarding the pathogenicity of Blastocystis. Its pathogenic mechanisms include different proteases, proteolytic enzymes and pro-inflammatory cytokines, such as Interleukin 6 and tumor necrosis factor-α, with decay of secretory IgA [6, 18, 19]. Some studies, such as the one of Azizian et al. [6], have shown that the presence of disease and its severity vary according to the parasitic load and subtype of Blastocystis, with a variability in the plasma concentration of cytokines and enzymes, all involved in the development of irritable bowel syndrome [19]. Clinical manifestations in symptomatic individuals are unspecific: abdominal pain, diarrhea, nausea, vomit, decreased appetite, flatulence [4]. However, these are also present in asymptomatic people [20]. Our study documented the presence of Bh in patients diagnosed with chronic abdominal pain followed-up in the outpatient clinic, constituting 13% of samples positive for Bh. This is quite similar with the findings of Gijsbers et al. [21] who associate chronic abdominal pain with Bh in 18% of their patients. Whether treatment should be given for Bh continues to be a complex subject considering the debate about its true pathogenicity [22]. In the analysis of the 138 patients included in this study, we found a statistical significant difference in favor of providing treatment with an antimicrobial agent, showing an improvement of the symptoms after completion of the regime. There are reports about an adequate response to treatment with metronidazole or sulfamethoxazole/trimethoprim, although some authors consider that clinical improvement may be owing to eradication of secondary pathogens [15]. The mostly prescribed treatment in our group of patients was metronidazole, which achieved symptoms elimination in only 55% of subjects, consistent with the published literature about an increase of resistance of Bh to this antiparasitic drug [23]. Even though the variety of treatments was wide, tinidazole and mebendazole, secnidazole and nitazoxanide showed the best results. Overall, it can be said that in the face of Bh in stool exams of patients with chronic abdominal pain, the use of an antimicrobial treatment improves the symptoms. Dinleyici et al. [24] carried out a randomized clinical trial in patients with gastrointestinal symptoms and Bh in feces, finding statistically significant differences in favor of establishing a treatment; this difference increased when the authors added Saccharomyces boulardii to the treatment with metronidazole. For being a retrospective study, our study has limitations, including different types of bias, in addition to an initial search conducted only in patients with Bh and not those with functional abdominal pain. However, our study raises the possibility that Bh may play an etiological role in some cases of functional abdominal pain and opens a window for prospective studies and even randomized clinical trials to prove this hypothesis. CONCLUSIONS The etiology of chronic abdominal pain is varied; however, we consider according with our result that it is mandatory to conduct stool exams, and in cases where Bh is present, it is important to provide treatment attempting to eradicate it before initiating expensive diagnostic approaches or classifying a patient as having functional abdominal pain. REFERENCES 1 Korterink JJ , Rutten JM , Venmans L , et al. Pharmacologic treatment in pediatric functional abdominal pain disorders: a systematic review . J Pediatr 2015 ; 166 : 424 – 31 . Google Scholar CrossRef Search ADS PubMed 2 Apley J , Naish N. Recurrent abdominal pains: a field survey of 1000 school children . Arch Dis Child 1958 ; 33 : 165 – 70 . Google Scholar CrossRef Search ADS PubMed 3 Rasquin-Weber A , Hyman PE , Cucchiara S , et al. Childhood functional gastrointestinal disorders . Gut 1999 ; 45(Suppl 2) : II60 – 8 . Google Scholar PubMed 4 Basak S , Rajurkar MN , Mallick SK. Detection of Blastocystis hominis: a controversial human pathogen . Parasitol Res 2014 ; 113 : 261 . Google Scholar CrossRef Search ADS PubMed 5 Ibarra C , Herrera V , Pérez de Arce E , et al. Parasitosis and irritable bowel syndrome . Rev Chilena Infectol 2016 ; 33 : 268 – 74 . Google Scholar CrossRef Search ADS PubMed 6 Azizian M , Basati G , Abangah G , et al. Contribution of Blastocystis hominis subtypes and associated inflammatory factors in development of irritable bowel syndrome . Parasitol Res 2016 ; 115 : 2003 – 9 . Google Scholar CrossRef Search ADS PubMed 7 Fouad SA , Basyoni MM , Fahmy RA , et al. The pathogenic role of different Blastocystis hominis genotypes isolated from patients with irritable bowel syndrome . Arab J Gastroenterol 2011 ; 12 : 194 – 200 . Google Scholar CrossRef Search ADS PubMed 8 Jimenez-Gonzalez DE , Martinez-Flores WA , Reyes-Gordillo J , et al. Blastocystis infection is associated with irritable bowel syndrome in a Mexican patient population . Parasitol Res 2012 ; 110 : 1269 – 75 . Google Scholar CrossRef Search ADS PubMed 9 Poirier P , Wawrzyniak I , Vivarès CP , et al. New insights into Blastocystis spp.: a potential link with irritable bowel syndrome . PLoS Pathog 2012 ; 8 : e1002545 . Google Scholar CrossRef Search ADS PubMed 10 Lorena RG , Rossana FH , Daniela MO. Dolor Abdominal Crónico en pediatría . Rev Chil Pediatr 2012 ; 83 : 279 – 89 . Google Scholar CrossRef Search ADS 11 Lewis ML , Palsson OS , Whitehead WE , et al. Prevalence of functional gastrointestinal disorders in children and adolescents . J Pediatr 2016 ; 177 : 39 – 43 . Google Scholar CrossRef Search ADS PubMed 12 Thompson WG , Heaton KW , Smyth GT , et al. Irritable bowel syndrome in general practice: prevalence, characteristics, and referral . Gut 2000 ; 46 : 78 – 82 . Google Scholar CrossRef Search ADS PubMed 13 Törnblom H , Lindberg G , Nyberg B , et al. Full-thickness biopsy of the jejunum reveals inflammation and enteric neuropathy in irritable bowel syndrome . Gastroenterology 2002 ; 123 : 1972 – 9 . Google Scholar CrossRef Search ADS PubMed 14 Verma R , Delfanian K. Blastocystis hominis associated acute urticaria . Am J Med Sci 2013 ; 346 : 80 – 1 . Google Scholar CrossRef Search ADS PubMed 15 Abdulsalam AM , Ithoi I , Al-Mekhlafi HM , et al. Prevalence, predictors and clinical significance of Blastocystis sp. in Sebha, Libya . Parasit Vectors 2013 ; 6 : 86. Google Scholar CrossRef Search ADS PubMed 16 Martínez I , Gutiérrez M , Ruiz L , et al. Blastocystis hominis y su relación con el estado nutricional de escolares en una comunidad de la sierra de Huayacocotla, Veracruz, México . Rev Biomed 2010 ; 21 : 77 – 84 . 17 Parija SC , Jeremiah SS. Blastocystis: taxonomy, biology and virulence . Trop Parasitol 2013 ; 3 : 17 – 25 . Google Scholar CrossRef Search ADS PubMed 18 Roberts T , Stark D , Harkness J. Update on the pathogenic potential and treatment options for Blastocystis sp . Gut Pathog 2014 ; 6 : 17. Google Scholar CrossRef Search ADS PubMed 19 Stensvold CR , Alfellani M , Clark CG. Levels of genetic diversity vary dramatically between Blastocystis subtypes . Infect Genet Evol 2012 ; 12 : 263 – 73 . Google Scholar CrossRef Search ADS PubMed 20 Bálint A , Dóczi I , Bereczki L , et al. Do not forget the stool examination!-cutaneous and gastrointestinal manifestations of Blastocystis sp. infection . Parasitol Res 2014 ; 113 : 1585 – 90 . Google Scholar CrossRef Search ADS PubMed 21 Gijsbers CF , Schweizer JJ , Büller HA. Protozoa as a cause of recurrent abdominal pain in children . J Pediatr Gastroenterol Nutr 2013 ; 57 : 603 – 6 . Google Scholar CrossRef Search ADS PubMed 22 Stensvold CR , Nielsen HV , Mølbak K , et al. Pursuing the clinical significance of Blastocystis–diagnostic limitations . Trends Parasitol 2009 ; 25 : 23 – 9 . Google Scholar CrossRef Search ADS PubMed 23 Coyle CM , Varughese J , Weiss LM. Blastocystis: to treat or not to treat . Clin Infect Dis 2012 : 54 : 105 – 10 . Google Scholar CrossRef Search ADS PubMed 24 Dinleyici EC , Eren M , Dogan N , et al. Clinical efficacy of Saccharomyces boulardii or metronidazole in symptomatic children with Blastocystis hominis infection . Parasitol Res 2011 ; 108 : 541 – 5 . Google Scholar CrossRef Search ADS PubMed © The Author [2017]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

2018 Journal of Tropical Pediatrics

doi: 10.1093/tropej/fmx062pmid: 28977620

Abstract Introduction Infantile tremor syndrome (ITS) is characterized by anaemia, skin depigmentation, tremors and developmental delay. The lack of sufficient literature on ITS and its conflicting association with vitamin B12 deficiency made us present this article. Objective The objective of this study is to describe demographic, clinical and laboratory profile and outcome of ITS. Methods This is a retrospective chart review of all children presenting with typical features of ITS attending a tertiary paediatric centre in southern India between January 2014 and January 2017. All children with pallor, skin depigmentation and developmental delay, with/without tremors, were included. Anaemia, developmental delay and tremors secondary to non-nutritional causes like metabolic causes were excluded. Results Of 70 children, 66 were exclusively breastfed and 46 mothers were vegetarians. Mean age of presentation was 13.2 months. Developmental delay was noted in 64, regression in 6 and tremors in 40. Vitamin B12 levels were low in 62 cases. Conclusion ITS should be considered in children <3 years with anaemia, developmental delay/regression and skin depigmentation, with/without tremors. ITS can be seen in < 3 months of age and in high socio-economic status. infantile tremor syndrome, breastfeeding, vegetarian diet, vitamin B12, weaning INTRODUCTION Infantile tremor syndrome (ITS) is a clinical condition characterized by tremors, anaemia, skin depigmentation and developmental delay in children between age 5 months and 3 years [1, 2]. It is most commonly reported from Indian subcontinent, Southeast Asian countries and Africa. Exact incidence is not known. In India, it accounts for 0.2–2% paediatric hospital admissions [3–6]. Currently no data are available on prevalence of ITS. ITS is not only associated with vitamin B12 deficiency, but it is also described in other micronutrient deficiencies and in non-vegan families. Recent review concluded that ITS is nutritional deficiency syndrome with laboratory evidence of vitamin B12 deficiency [6]. However, Rajesh Gupta et al. reported only 2 of 12 children had low vitamin B12 level. Since number of patients in above studies are <50 in number, cause for ITS is not clear [7]. In view of paucity of sufficient literature and conflicting result in association with vitamin B12 deficiency in available literature, we planned for this study. In all patients, detailed history, including birth history and diet history of child and mother, was taken and physical examination was carried out. Nutritional status was classified according to Indian academy of paediatrics [8]. A complete haematological work up (haemoglobin, total leukocyte count, platelet count, MCV, MCHC, MCH, peripheral blood film) using automated five-part differential haematology cell counter from Mindray 5200 BC (China) was carried out before the administration of any form of haematinics. Vitamin B12 level was done for all children. Vitamin B12 was measured by chemiluminescence method. Normal range of serum vitamin B12 is 200–800 pg/ml. The cut off value to consider vitamin B12 deficiency was taken as <200 pg/ml. Chest radiograph, urine examination, blood cultures were done in cases associated with lower respiratory tract infection, urinary tract infection or sepsis. CT/MRI scan of brain and metabolic work up was done as and when relevant. Institutional ethical clearance was taken. RESULTS The total number of cases seen during this period was 70. The mean age of presentation was 13.2 months (standard deviation of 5.6), ranging from 3 months to 30 months; three were less than 6 months. There were 38 males and 32 females, 24 were from urban areas and 46 were from rural areas. Nine children belonged to upper class, 2 to upper middle, 42 to lower middle and 17 to upper lower socio-economic status according to modified Kuppuswamy socio-economic classification. Two babies were preterm with low birth weight. History of birth asphyxia was present in three children. One child had a history of neonatal meningitis. Clinical features are mentioned in Table 1. Three children had convulsions, four had lower respiratory tract infection and three children had acute gastroenteritis. All children were exclusively breastfed till the time of admission, except in four children, in addition to breastfeeding, ragi porridge was started but in insufficient amounts. Only 14 mothers were non-vegetarians. There was grade 1 malnutrition in 32, grade II in 13 and grade III in 10 as per IAP classification of malnutrition. All patients were haemodynamically stable at the time of admission except one child, who was in septic shock. All had pallor and knuckle depigmentation (Fig. 1). Other nutritional deficiency findings (vitamin A deficiency, chelitis) were present in 10. Table 1 Clinical findings in children with ITS Clinical findings Number of cases (70) Developmental delay 64 (91.0%) Regression 06 (08.5%) Tremors 40 (64.2%) Predominantly breastfed 66 (94.2%) Mothers vegetarian diet 46 (65.7%) Protein energy malnutrition 55 (78.0%) Pallor 70 (100%) Knuckle pigmentation 70 (100%) Hypopigmented and sparse hair 53 (75.0%) Pedal oedema 02 (02.8%) Lethargy/apathy 38 (54.2%) Hypertonia 29 (41.4%) Hypotonia 13 (18.5%) Hepatomegaly 02 (02.8%) Microcephaly 20 (28.5%) Maternal anaemia 40 (57.1%) Clinical findings Number of cases (70) Developmental delay 64 (91.0%) Regression 06 (08.5%) Tremors 40 (64.2%) Predominantly breastfed 66 (94.2%) Mothers vegetarian diet 46 (65.7%) Protein energy malnutrition 55 (78.0%) Pallor 70 (100%) Knuckle pigmentation 70 (100%) Hypopigmented and sparse hair 53 (75.0%) Pedal oedema 02 (02.8%) Lethargy/apathy 38 (54.2%) Hypertonia 29 (41.4%) Hypotonia 13 (18.5%) Hepatomegaly 02 (02.8%) Microcephaly 20 (28.5%) Maternal anaemia 40 (57.1%) Table 1 Clinical findings in children with ITS Clinical findings Number of cases (70) Developmental delay 64 (91.0%) Regression 06 (08.5%) Tremors 40 (64.2%) Predominantly breastfed 66 (94.2%) Mothers vegetarian diet 46 (65.7%) Protein energy malnutrition 55 (78.0%) Pallor 70 (100%) Knuckle pigmentation 70 (100%) Hypopigmented and sparse hair 53 (75.0%) Pedal oedema 02 (02.8%) Lethargy/apathy 38 (54.2%) Hypertonia 29 (41.4%) Hypotonia 13 (18.5%) Hepatomegaly 02 (02.8%) Microcephaly 20 (28.5%) Maternal anaemia 40 (57.1%) Clinical findings Number of cases (70) Developmental delay 64 (91.0%) Regression 06 (08.5%) Tremors 40 (64.2%) Predominantly breastfed 66 (94.2%) Mothers vegetarian diet 46 (65.7%) Protein energy malnutrition 55 (78.0%) Pallor 70 (100%) Knuckle pigmentation 70 (100%) Hypopigmented and sparse hair 53 (75.0%) Pedal oedema 02 (02.8%) Lethargy/apathy 38 (54.2%) Hypertonia 29 (41.4%) Hypotonia 13 (18.5%) Hepatomegaly 02 (02.8%) Microcephaly 20 (28.5%) Maternal anaemia 40 (57.1%) Fig. 1. View largeDownload slide Clinical photo showing hypopigmented sparse hair, knuckle pigmentation and apathy. Fig. 1. View largeDownload slide Clinical photo showing hypopigmented sparse hair, knuckle pigmentation and apathy. Haematological results in children with ITS are shown in Table 2. Serum vitamin B12 levels of babies were analysed in 66 of 70; it was found to be low in 62, normal in 4. Vitamin B12 levels were low in all children without tremors. Serum vitamin B12 levels were performed in nine mothers and in all it was low. Chest radiograph was suggestive of pneumonia in four children. CT scan of brain was done in 11 children; four were normal and the other seven had features of cerebral atrophy. Table 2 Laboratory findings in children with ITS Laboratory findings Number of cases (70) Haemoglobin (Hb < 11 mg/dl) 70 (100%) Leucopenia (TLC < 4000/mm3) 22 (31.4%) Thrombocytopenia (Platelet count <1,50,000/mm3) 26 (37.1%) High MCV (>100 fl) 40 (57.1%) High MCH (>32 pg) 42 (60.0%) High MCHC (>36 g/dl) 16 (34.2%) Macrocytic anaemia 45 (64.2%) Microcytic anaemia 06 (8.50%) Dimorphic anaemia 05 (7.10%) Normocytic normochromic anaemia 14 (20.0%) Low child serum Vit B12 (<200 pg/ml) (n=66) 62 (93.9%) Low maternal Vit B12 (<200 pg/ml) (n=9) 09 (100%) Cerebral atrophy (n=11) 07 (63.6%) Laboratory findings Number of cases (70) Haemoglobin (Hb < 11 mg/dl) 70 (100%) Leucopenia (TLC < 4000/mm3) 22 (31.4%) Thrombocytopenia (Platelet count <1,50,000/mm3) 26 (37.1%) High MCV (>100 fl) 40 (57.1%) High MCH (>32 pg) 42 (60.0%) High MCHC (>36 g/dl) 16 (34.2%) Macrocytic anaemia 45 (64.2%) Microcytic anaemia 06 (8.50%) Dimorphic anaemia 05 (7.10%) Normocytic normochromic anaemia 14 (20.0%) Low child serum Vit B12 (<200 pg/ml) (n=66) 62 (93.9%) Low maternal Vit B12 (<200 pg/ml) (n=9) 09 (100%) Cerebral atrophy (n=11) 07 (63.6%) Table 2 Laboratory findings in children with ITS Laboratory findings Number of cases (70) Haemoglobin (Hb < 11 mg/dl) 70 (100%) Leucopenia (TLC < 4000/mm3) 22 (31.4%) Thrombocytopenia (Platelet count <1,50,000/mm3) 26 (37.1%) High MCV (>100 fl) 40 (57.1%) High MCH (>32 pg) 42 (60.0%) High MCHC (>36 g/dl) 16 (34.2%) Macrocytic anaemia 45 (64.2%) Microcytic anaemia 06 (8.50%) Dimorphic anaemia 05 (7.10%) Normocytic normochromic anaemia 14 (20.0%) Low child serum Vit B12 (<200 pg/ml) (n=66) 62 (93.9%) Low maternal Vit B12 (<200 pg/ml) (n=9) 09 (100%) Cerebral atrophy (n=11) 07 (63.6%) Laboratory findings Number of cases (70) Haemoglobin (Hb < 11 mg/dl) 70 (100%) Leucopenia (TLC < 4000/mm3) 22 (31.4%) Thrombocytopenia (Platelet count <1,50,000/mm3) 26 (37.1%) High MCV (>100 fl) 40 (57.1%) High MCH (>32 pg) 42 (60.0%) High MCHC (>36 g/dl) 16 (34.2%) Macrocytic anaemia 45 (64.2%) Microcytic anaemia 06 (8.50%) Dimorphic anaemia 05 (7.10%) Normocytic normochromic anaemia 14 (20.0%) Low child serum Vit B12 (<200 pg/ml) (n=66) 62 (93.9%) Low maternal Vit B12 (<200 pg/ml) (n=9) 09 (100%) Cerebral atrophy (n=11) 07 (63.6%) All children received vitamin B12 injection intravenously for 14 days, with folic acid and iron supplementation given in case of iron-deficiency anaemia. Antibiotics were given as and when indicated. Three children required blood transfusion. For all children, diet advice and infant stimulation was given. Forty children had tremors and 20 of them received propranolol. The mean duration for control of tremors in our cases was 35.3 days. One child expired because of septic shock, despite good supportive care with mechanical ventilation, inotropes and antibiotics. Remaining children improved with treatment. DISCUSSION In the present study, equal sex distribution was found. Some studies reported that boys are more commonly affected than girls, while few studies showed equal sex distribution in occurrence of ITS [9, 10]. Present study found 82.8% babies were between 6 and 18 months of age. Bajpai et al. found 94.1% and Sachdev et al. found 89.2% children between 6 and 18 months [11, 12]. The youngest child in our study was 3 months. This was the earliest presentation described in the literature and could be explained by underlying maternal nutritional deficiency. The reason for this age of presentation may be improper introduction of complementary food and predominant breastfeeding; large number of malnourished individuals were also seen in this age group. Earlier studies reported ITS mainly in low socio-economic class. However in this study, nine belong to upper class. This may be because of lack of awareness of timely introduction of complementary food. Most of the children had grade 1 malnutrition, and none of the children were marasmic, on the contrary, they appeared plump; only two children had pitting pedal oedema, and evidence of other vitamin deficiency was present in 10 cases. Similar changes have been consistently reported by others [13]. In this study, neurological and haematological manifestations were present in all children. In the present study, tremors were seen in 40 cases. Before onset of tremors, seven had preceding illness, either respiratory tract infection or gastroenteritis. The preceding or accompanying illness may precipitate the nutritional deficiency and lead to acute manifestation of ITS. Any occurrence of systemic illness could initiate a reaction, resulting in functional disturbance of motor activity and manifested as tremors [11]. In one case, tremor developed following DPT vaccine. Usually these children are listless, apathic and disinterested in surroundings. Hypertonia was present in 29 cases, and exaggerated deep tendon reflexes were present in 29 cases. In the literature, hypotonia is most commonly described, but in the present study, hypertonia is more commonly seen than hypotonia; this may be because of development of spastic quadriparesis as a late presentation, and microcephaly was observed in all these cases. In the present study, serum vitamin B12 levels were low in 93.9% of cases. Only four children had normal vitamin B12 level, and one had evidence of vitamin B12 deficiency such as macrocytosis and peripheral smear suggestive of megaloblastic anaemia and low maternal serum vitamin B12. Other three children had iron-deficiency anaemia. The findings of earlier authors were anaemia in ITS may be due to vitamin B12 deficiency or iron deficiency alone or dual deficiency or folate deficiency may be present. In G. Garewal et al.’s study, in addition to vitamin B12 deficiency, folate and iron deficiency was found [14]. The aetiology of ITS is still elusive. Aetiological possibilities are nutritional, viral infections and degenerative hypothesis, but none have been conclusively proven [1]; nutritional theory is the most accepted. Recent review concluded that ITS is because of vitamin B12 deficiency [6]. However, Rajesh Gupta et al. reported only 2 of 12 children had low vitamin B12 level [7]. In this study, all children had evidence of vitamin B12 deficiency, except three children. Epidemiologically, ITS occurs in exclusively breastfed infants of vegan mothers, pointing to vitamin B12 deficiency [15], which showed similarity with the present study. Skin depigmentation seen in ITS is also a well-known sign of vitamin B12 deficiency [16]. It is usual to find other associated nutritional deficiencies like protein, vitamin A, D, C and B-complex and other micronutrients [1, 17–19]. In this study, other nutritional deficiency findings were present in 10. Management of ITS being largely empirical, it includes vitamin B12, folic acid, iron, calcium, zinc, magnesium and high protein diet [20]. All our cases received vitamin B12, folic acid and iron (in case of iron deficiency anaemia) apart from dietary advice. Three children required blood transfusion because of severe anaemia. Infant stimulation is also important to reduce long-term neurodevelopmental sequelae. For management of tremors, many drugs have been tried, including phenobarbitone, chlorpromazine, carbamazepine and propranolol [9, 21]. Patients in the present study were treated with propranolol and favourable response was obtained. The mean duration of tremor control in our cases was 35.3 days with range from 7 to 75 days. Other studies reported mean duration of tremors control being 50.5 days with a range from 3 to 225 days. All of them responded well both haematologically and neurologically except one child who expired because of septic shock. One child developed infantile spasms during recovery. Skin depigmentation and hair colour changes take months to clear. Children in pretremor stage improved gradually, showing interest in surroundings, apathy decreases and mental dullness takes months to years to come back to normal. In the present series of cases, it took 8–12 months. This therapeutic response favours a vital role of vitamin B12 deficiency in causation of this syndrome. Based on above findings, we considered vitamin B12 deficiency is a possible aetiology for ITS. We suggest in vivo experiments in mice/primates using magnetic resonance spectroscopy to clearly identify major aetiologic factors. Inadequate calories because of improper introduction of complementary foods, poor dietary habits, predominant breastfeeding, mothers living on diet devoid of animal products are likely cause for low vitamin B12 levels. Improvement in nutritional status, living conditions and better weaning practices could explain the reducing rates of ITS over the years. We recommend fortified foods with vitamin B12 for vegan mothers. It is necessary to check vitamin B12 levels in mothers if clinically suspected in last trimester and to supplement accordingly to prevent ITS in babies and to improve weaning practices. Study limitation is that this is a retrospective case review. However, this is the largest number of cases reported. CONCLUSION ITS needs to be considered in any child <3 years presenting with anaemia, developmental delay/regression, skin depigmentation and sparse hair, with or without tremors. ITS should be excluded in any child presenting with neuroregression, as it is a treatable condition to prevent long-term neurodevelopmental sequelae. This study adds that ITS can be seen in <3 months of age, in high socio-economic status and it can mimic neurodegenerative diseases. ACKNOWLEDGEMENTS We acknowledge department of pathology for analysis of complete blood count and peripheral smear. REFERENCES 1 Kalra V , Infantile tremor syndrome. In: Ghai OP (ed.). Essential Pediatrics . 7th edn. New Delhi : CBS Publishers and Distributions Pvt. Ltd , 2009 , 558 – 9 . 2 Sharada B , Bhandari B. Infantile tremor syndrome . Indian Pediatr 1987 ; 24 : 415 – 21 . Google Scholar PubMed 3 Gupte S , Pal M , Gupta SK ,et al. Infantile tremor syndrome (ITS). In Gupte S (ed.). Textbook of Paediatric Nutrition . New Delhi : Peepee , 2006 , 255 – 65 . 4 Holla RG , Prasad AN. Infantile tremor syndrome . Med J Armed Forces India 2010 ; 66 : 186 – 7 . Google Scholar CrossRef Search ADS PubMed 5 Singla DA , Sharma M , Sharma S , et al. The changing profile of infantile tremor syndrome in hilly terrain of India . J Neurosci Rural Pract 2015 ; 6 : 610 – 2 . Google Scholar CrossRef Search ADS PubMed 6 Goraya JS , Kaur S. Infantile tremor syndrome: a review and critical appraisal of its etiology . J Pediatr Neurosci 2016 ; 11 : 298 – 304 . Google Scholar CrossRef Search ADS PubMed 7 Gupta R , Mandliya J , Sonker P , et al. Infantile tremor syndrome: role of vitamin B12 revisited . J Pediatr Neurosci 2016 ; 11 : 305 – 8 . Google Scholar CrossRef Search ADS PubMed 8 Ghai OP. Nutrition. In Ghai OP , Paul VK , Bagga A (eds). Ghai Essential pediatrics . 7th edn. New Delhi : Dr.Ghai , 2007 , 64 9 Ratageri VH , Shepur TA , Patil MM , et al. Scurvy in infantile tremor syndrome . Indian J Pediatr 2005 ; 72 : 883 – 4 . Google Scholar CrossRef Search ADS PubMed 10 Chaparrwal BL , Singh SD , Mehta S , et al. Magnesium level in serum and CSF in meningo-encephalitic syndrome . Indian J Paediatr 1971 ; 38 : 331 – 3 . Google Scholar CrossRef Search ADS 11 Bajpai PC , Tandon PN , Sharma NL , et al. Infantile tremor syndrome . Acta Neurol Scand 1965 ; 41 : 473 – 86 . Google Scholar CrossRef Search ADS PubMed 12 Sachdev K , Manchanda SS , Lal H. The syndrome of tremor, mental regression and anaemia in infants and young children . Indian Pediatr 1965 ; 2 : 239. Google Scholar PubMed 13 Garg BK , Srivastava JR. Infantile tremor syndrome . Ind J Pediatr 1969 ; 36 : 213 – 18 . Google Scholar CrossRef Search ADS 14 Garewal G , Narang A , Das KC. Infantile tremor syndrome: a vitamin B12 deficiency syndrome in infants . J Trop Pediatr 1988 ; 34 : 174 – 8 . Google Scholar CrossRef Search ADS PubMed 15 Goraya JS , Kaur S. Infantile tremor syndrome—down but not out . Indian Paediatr 2015 ; 52 : 249 – 50 . Google Scholar CrossRef Search ADS 16 Simsek OP , Gonc N , Gumruk F , et al. A child with vitamin B12 deficiency presenting with pancytopenia and hyperpigmentation . J Pediatr Hematol Oncol 2004 ; 26 : 834 – 6 . Google Scholar PubMed 17 Charles C , Garg BK , Srivastava JR. Infantile tremor syndrome . Indian J Pediatr 1969 ; 36 : 213 – 28 . Google Scholar CrossRef Search ADS PubMed 18 Vora RM , Tullu MS , Bartakke SP , et al. Infantile tremor syndrome and zinc deficiency . Indian J Med Sci 2002 ; 56 : 69 – 72 . Google Scholar PubMed 19 Ratageri VH , Shepur TA , Patil MM , et al. Scurvy in infantile tremor syndrome . Indian J Paediatr 2005 ; 72 : 883 – 4 . Google Scholar CrossRef Search ADS 20 Ghai OP , Gupta P. Infantile tremor syndrome. In: Ghai OP , Gupta P , Paul VK (eds). Ghai Essential Paediatrics . 8th edn. New Delhi : Dr. Ghai , 2013 , 580 – 1 . 21 Murali MV , Sharma PP , Koul PB , et al. Carbamazepine therapy for infantile tremor syndrome . Indian Paediatr 1993 ; 30 : 72 – 4 . © The Author [2017]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

2018 Journal of Tropical Pediatrics

doi: 10.1093/tropej/fmx065pmid: 28977585

Abstract Background Human immunodeficiency virus (HIV) and malaria are leading causes of morbidity and mortality among under-fives in sub-Saharan Africa. HIV infection could affect development of antimalarial immunity by impaired parasite clearance with predisposition to higher malaria parasitaemia. Objective The objective of this study is to assess asymptomatic malaria parasite density (AMPD) in HIV-1-infected under-fives in a holoendemic zone. Methods HIV-1-positive and -negative children <5 years on follow-up care were recruited and AMPD and CD4 counts were determined. Results A total of 358 children were studied. Significantly higher malaria parasitaemia was found in HIV-infected individuals (118.7 vs. 87.3 parasite/μl, p = 0.021). Disparity in AMPD was most pronounced at infancy with similar distribution at all age brackets and consistently higher parasitaemia in the subjects. Conclusion Parasitaemia is higher in HIV-infected than uninfected children. The burden is highest at infancy. Acquisition of antimalarial immunity is similar in both groups. Parasitaemia is not significantly affected by clinical disease stage or worsening immunosuppression. asymptomatic, parasite density, under-fives, immunity INTRODUCTION Human immunodeficiency virus (HIV) and Falciparum malaria infection are leading causes of morbidity and mortality in sub-Saharan Africa [1]. The epidemiologic overlap has been the source of concern, as coinfection could further increase the health burden [2]. Falciparum malaria is a major public health problem, with one child dying every minute in Africa [3], especially among under-fives owing to naive immunity [4–6]. In endemic areas, the transfer of maternal antibodies, repeated early childhood exposure and clinical episodes generate and sustain partial antimalarial immunity, which ameliorates infection but does not prevent asymptomatic malaria parasitaemia [7], i.e. the presence of malaria parasites in the blood without symptoms [8]. CD4+ T-cells are essential for developing antimalarial immunity [9]; they help B cells produce antibodies and indirectly support the control of parasitaemia through cytokine production and macrophage activation [10]. In HIV-infected children, malaria-specific immunity development and parasite clearance are impaired, resulting in increased predisposition to parasitaemia [9, 11–13]. Asymptomatic malaria parasitaemia is a predictive parameter for transmission and assessment of immunity in exposed populations [8, 14, 15]. This study highlights antimalarial immunity in a holoendemic region where sustained exposure generates immunity in childhood. Withworth et al. [2] found twice as many parasitaemia episodes, increasing with worsening immunosuppression in HIV-infected adults with symptomatic malaria in a moderate endemic area. The findings were collaborated by French et al. [16]. Coinfected pregnant adults living in stable and unstable malaria regions also showed increased parasitaemia [17–21]. Studies of asymptomatic malaria in HIV-infected individuals are few and limited in children from holoendemic regions having almost half of global malaria cases [22]. This cross-sectional study examines asymptomatic parasitaemia in HIV-infected children <5 years and the role of CD4 T-cells on parasite density. METHODS The study was conducted at the paediatric HIV clinic of University of Benin Teaching Hospital (UBTH), between March and July 2012. Malaria transmission is holoendemic; intense and stable transmission occurs all year round with peaks during the rainy season [23]. The HIV diagnosis for afebrile children 6 weeks to <18 months was by HIV DNA polymerase chain reaction, and those ≥18 months had HIV rapid antibody serial testing algorithm II using standard protocols [24]. Controls were age- and sex-matched HIV-1-negative children attending clinics for non-chronic medical conditions. Children with history of malaria treatment in the preceding 2 weeks and those on chemoprophylaxis were excluded. Haemoglobin genotype was done to exclude sickle cell disease in subjects and controls. Information on co-trimoxazole (CTX) prophylaxis is according to the national guideline; house screening and use of insecticide-treated nets (ITNs) were noted. Ethical approval was obtained from UBTH Ethics and Research Committee (ADM/E.22 A/VOL, VII/390), while informed, written consent was from parent(s)/guardian(s) of subjects and controls. Pre-test/Post-test counselling was done before and following HIV testing. The clinical (World Health Organization, WHO) staging of HIV/AIDS of the subjects was noted. Blood samples were collected into ethylenediaminetetraacetic acid (EDTA) anticoagulated containers and concomitant samples for CD4 count. Thick and thin blood smears were stained with 2% Giemsa for 30 min and read by WHO-certified microscopists who were uninvolved in direct patient care and blinded to the clinical status of the study participants. Asexual malaria parasitaemia was recorded as present or absent. Parasites/μl were calculated from thick blood smears by counting asexual parasites per 200 leukocytes, assuming a leukocyte count of 8000/ml. A blood smear was considered negative on examination of 100 high-power fields without parasites. Thin smears were for determination of parasite species [24]. CD4 counts was analysed using Flow Cytometer Cyflow SL Green® (Partec). The CD4 counts were documented according to the Centres for Disease Control and Prevention immunologic classification [24]. HIV screening test was carried out for the controls using HIV rapid testing serial algorithm II for HIV diagnosis according to national guidelines [25, 26]. Statistical analysis Clinical and laboratory data management and analysis were by IBM SPSS statistic 20. The test of normality (Kolmogorov–Smirnov) showed that malaria parasite density for microscopy positive slides was not normally distributed, hence the use of non-parametric tests for analysis. Mann–Whitney U test was to compare the mean parasite densities between cohorts. While Kruskal–Wallis test was for comparison of more than two means, the level of significance was at p < 0.05 and confidence level of 95%. RESULTS General characteristics In all, 179 HIV-1 positive and 179 HIV-1 negative sex- and age-matched controls were studied. The subjects consisted of 96 (53.6%) males and 83 (46.4%) females at the ratio of 1.16:1. The age range of the study population was 2–59 months and their age distribution is shown in Table 1. Table 1 Age distribution of the subjects and controls Age group (Months) Subjects n (%) Controlsn (%) 2 to < 12 9 (5) 9 (5) 12 to < 24 13(7.3) 13(7.3) 24 to < 36 25(14) 25(14) 36 to < 48 35(19.5) 35(19.5) 48 to 59 97(54.2) 97(54.2) Age group (Months) Subjects n (%) Controlsn (%) 2 to < 12 9 (5) 9 (5) 12 to < 24 13(7.3) 13(7.3) 24 to < 36 25(14) 25(14) 36 to < 48 35(19.5) 35(19.5) 48 to 59 97(54.2) 97(54.2) Table 1 Age distribution of the subjects and controls Age group (Months) Subjects n (%) Controlsn (%) 2 to < 12 9 (5) 9 (5) 12 to < 24 13(7.3) 13(7.3) 24 to < 36 25(14) 25(14) 36 to < 48 35(19.5) 35(19.5) 48 to 59 97(54.2) 97(54.2) Age group (Months) Subjects n (%) Controlsn (%) 2 to < 12 9 (5) 9 (5) 12 to < 24 13(7.3) 13(7.3) 24 to < 36 25(14) 25(14) 36 to < 48 35(19.5) 35(19.5) 48 to 59 97(54.2) 97(54.2) Most of the subjects (76; 42.5%) were in WHO clinical Stage 1, 44 (24.6%) in Stage 2, 46 (25.7%) in Stage 3, while 13 (7.3%) were in Stage 4. Majority of the subjects [152 (84.9%)] were in immunologic Categories 1 and 2, while 27 (15.1%) were in Category 3. Plasmodium falciparum was the only parasite specie identified. In all, 61 (34.1%) of the subjects and 32 (17.9%) of the controls had positive parasite smears (χ2 = 13.166, p = 0.001). Asymptomatic mean parasite density The mean parasite density (for microscopy-positive slides only) in subjects of 118.7 ± 70.5 parasite/µl was significantly higher than 87.3 ± 36.9 noted in controls (Mann–Whitney U = 705.5, p = 0.029) (Table 2). Table 2 Mean parasite density by age (groups) in subjects and controls Age group (months) Subject mpd (SD) Controls mpd (SD) Mann-Whitney U P 2 to  <12 148.8 (93.3) 90.0 1.000 0.380 12 to < 24 84.7 (55.0) 65.0 (5.0) 6.000 0.437 24 to < 36 127.9 (54.8) 96.7 (44.5) 21.500 0.187 36 to < 48 107.6 (42.4) 81.3 (27.7) 22.000 0.189 48 to < 59 122.2 (83.3) 89.5 (41.4) 173.000 0.346 118.7 (70.5) 87.3 (36.9) 705.500 0.029 Age group (months) Subject mpd (SD) Controls mpd (SD) Mann-Whitney U P 2 to  <12 148.8 (93.3) 90.0 1.000 0.380 12 to < 24 84.7 (55.0) 65.0 (5.0) 6.000 0.437 24 to < 36 127.9 (54.8) 96.7 (44.5) 21.500 0.187 36 to < 48 107.6 (42.4) 81.3 (27.7) 22.000 0.189 48 to < 59 122.2 (83.3) 89.5 (41.4) 173.000 0.346 118.7 (70.5) 87.3 (36.9) 705.500 0.029 Kruskal–Wallis test subjects: χ2 = 3.255, p = 0.516. Controls: χ2 = 2.531, p = 0.639, df = 4. mpd, mean parasite density/µl; SD, standard deviation. Table 2 Mean parasite density by age (groups) in subjects and controls Age group (months) Subject mpd (SD) Controls mpd (SD) Mann-Whitney U P 2 to  <12 148.8 (93.3) 90.0 1.000 0.380 12 to < 24 84.7 (55.0) 65.0 (5.0) 6.000 0.437 24 to < 36 127.9 (54.8) 96.7 (44.5) 21.500 0.187 36 to < 48 107.6 (42.4) 81.3 (27.7) 22.000 0.189 48 to < 59 122.2 (83.3) 89.5 (41.4) 173.000 0.346 118.7 (70.5) 87.3 (36.9) 705.500 0.029 Age group (months) Subject mpd (SD) Controls mpd (SD) Mann-Whitney U P 2 to  <12 148.8 (93.3) 90.0 1.000 0.380 12 to < 24 84.7 (55.0) 65.0 (5.0) 6.000 0.437 24 to < 36 127.9 (54.8) 96.7 (44.5) 21.500 0.187 36 to < 48 107.6 (42.4) 81.3 (27.7) 22.000 0.189 48 to < 59 122.2 (83.3) 89.5 (41.4) 173.000 0.346 118.7 (70.5) 87.3 (36.9) 705.500 0.029 Kruskal–Wallis test subjects: χ2 = 3.255, p = 0.516. Controls: χ2 = 2.531, p = 0.639, df = 4. mpd, mean parasite density/µl; SD, standard deviation. Asymptomatic parasitaemia in subjects was higher than in controls of all age brackets. Subjects 2 to <12 months (infancy) had the highest density of parasites (148.8 parasites/μl), and 12 to < 24 months age bracket had the least (84.7 parasites/μl). The differences in parasitaemia by age brackets of subjects and controls were not statistically significant (Table 2). The changes in asymptomatic malaria parasite density (AMPD) by age groups of subjects and controls are shown in Fig. 1. Fig. 1. View largeDownload slide Line graph depicting the mean parasite densities by age groups of the subjects and controls. Fig. 1. View largeDownload slide Line graph depicting the mean parasite densities by age groups of the subjects and controls. Parasitaemia by gender The AMPD of male subjects was 118.6 ± 70.5 parasites/µl as against 119.0 ± 54.0 in female subjects, p = 0.84. For the controls, it was 95.5 ± 36.4 parasites/µl in males and 78.0 ± 36.4 in females, p = 0.84. Parasitaemia by WHO stage and immune category The AMPD was least (111.6 parasite/µl) among subjects in WHO Stage 1 and highest (131.2 parasite/µl) among those in WHO Stage 4. Parasitaemia increased with worsening WHO clinical staging; however, this relationship was not statistically significant (Kruskal–Wallis test, χ2 = 1.514, p = 0.679) (Table 3). Table 3 Mean parasite density by WHO clinical staging of subjects WHO stage Subjects mpd (±SD) 1 76 111.6 (70.9) 2 44 113.1 (66.6) 3 46 129.0 (71.7) 4 13 138.2 (92.2) Total 179 118.7 (70.5) WHO stage Subjects mpd (±SD) 1 76 111.6 (70.9) 2 44 113.1 (66.6) 3 46 129.0 (71.7) 4 13 138.2 (92.2) Total 179 118.7 (70.5) Kruskal–Wallis test χ2= 1.514, p = 0.679, df = 3. Table 3 Mean parasite density by WHO clinical staging of subjects WHO stage Subjects mpd (±SD) 1 76 111.6 (70.9) 2 44 113.1 (66.6) 3 46 129.0 (71.7) 4 13 138.2 (92.2) Total 179 118.7 (70.5) WHO stage Subjects mpd (±SD) 1 76 111.6 (70.9) 2 44 113.1 (66.6) 3 46 129.0 (71.7) 4 13 138.2 (92.2) Total 179 118.7 (70.5) Kruskal–Wallis test χ2= 1.514, p = 0.679, df = 3. The AMPD was lowest among subjects in immune Category 1 and highest in immune Category 2. Association between parasitaemia and immune category was not statistically significant. (Kruskal–Wallis test, χ2 = 0.970, p = 0.616) (Table 4). Table 4 Mean parasite density by immune category of subjects Immune category Subjects Mean parasite density (±SD) 1 80 100.8 (49.0) 2 72 131.7 (72.3) 3 27 128.8 (91.5) Total 179 118.7 (70.5) Immune category Subjects Mean parasite density (±SD) 1 80 100.8 (49.0) 2 72 131.7 (72.3) 3 27 128.8 (91.5) Total 179 118.7 (70.5) Kruskal–Wallis test χ2 =0.970, p = 0.616, df = 2. Table 4 Mean parasite density by immune category of subjects Immune category Subjects Mean parasite density (±SD) 1 80 100.8 (49.0) 2 72 131.7 (72.3) 3 27 128.8 (91.5) Total 179 118.7 (70.5) Immune category Subjects Mean parasite density (±SD) 1 80 100.8 (49.0) 2 72 131.7 (72.3) 3 27 128.8 (91.5) Total 179 118.7 (70.5) Kruskal–Wallis test χ2 =0.970, p = 0.616, df = 2. Parasitaemia with antiretroviral therapy (ART) use Subjects on ART had higher AMPD (123.4 ± 73.7) than those not on treatment (88.1 ± 32.1) although the relationship was not significant (Mann–Whitney U = 165.500, p = 0.320). Mean parasite densities were higher in subjects receiving Zidovudine, Nevirapine, Lamivudine, Efavirenz and Stavudine (Table 5). None of the subjects on Ritonavir/Lopinavir and Abacavir had any parasitaemia. Table 5 Mean parasite density by ART use ART n (%) mpd (±SD) Mann-Whitney U P Zidovudine  Yes 144 (80.4) 118.9 (69.4)  No 35 (19.6) 117.9 (78.8) 266.500 0.873 Nevirapine  Yes 148 (82.7) 123.6 (75.1)  No 31 (17.9) 94.0 (31.0) 223.000 0.533 Lamivudine  Yes 151 (84.4) 123.4 (73.7)  No 28 (15.6) 88.13 (32.1) 165.500 0.320 Efavirenz  Yes 6 (3.4) 183.0 (113.6)  No 173 (96.6) 115.4 (67.5) 43.500 0.147 Stavudine  Yes 12 (6.7) 139.0 (91.9)  No 167 (93.3) 118.1 (70.5) 50.000 0.715 ART  Yes 155 (86.6) 123.4 (73.7)  No 24 (13.4) 88.1 (32.1) 165.500 0.320 ART n (%) mpd (±SD) Mann-Whitney U P Zidovudine  Yes 144 (80.4) 118.9 (69.4)  No 35 (19.6) 117.9 (78.8) 266.500 0.873 Nevirapine  Yes 148 (82.7) 123.6 (75.1)  No 31 (17.9) 94.0 (31.0) 223.000 0.533 Lamivudine  Yes 151 (84.4) 123.4 (73.7)  No 28 (15.6) 88.13 (32.1) 165.500 0.320 Efavirenz  Yes 6 (3.4) 183.0 (113.6)  No 173 (96.6) 115.4 (67.5) 43.500 0.147 Stavudine  Yes 12 (6.7) 139.0 (91.9)  No 167 (93.3) 118.1 (70.5) 50.000 0.715 ART  Yes 155 (86.6) 123.4 (73.7)  No 24 (13.4) 88.1 (32.1) 165.500 0.320 Table 5 Mean parasite density by ART use ART n (%) mpd (±SD) Mann-Whitney U P Zidovudine  Yes 144 (80.4) 118.9 (69.4)  No 35 (19.6) 117.9 (78.8) 266.500 0.873 Nevirapine  Yes 148 (82.7) 123.6 (75.1)  No 31 (17.9) 94.0 (31.0) 223.000 0.533 Lamivudine  Yes 151 (84.4) 123.4 (73.7)  No 28 (15.6) 88.13 (32.1) 165.500 0.320 Efavirenz  Yes 6 (3.4) 183.0 (113.6)  No 173 (96.6) 115.4 (67.5) 43.500 0.147 Stavudine  Yes 12 (6.7) 139.0 (91.9)  No 167 (93.3) 118.1 (70.5) 50.000 0.715 ART  Yes 155 (86.6) 123.4 (73.7)  No 24 (13.4) 88.1 (32.1) 165.500 0.320 ART n (%) mpd (±SD) Mann-Whitney U P Zidovudine  Yes 144 (80.4) 118.9 (69.4)  No 35 (19.6) 117.9 (78.8) 266.500 0.873 Nevirapine  Yes 148 (82.7) 123.6 (75.1)  No 31 (17.9) 94.0 (31.0) 223.000 0.533 Lamivudine  Yes 151 (84.4) 123.4 (73.7)  No 28 (15.6) 88.13 (32.1) 165.500 0.320 Efavirenz  Yes 6 (3.4) 183.0 (113.6)  No 173 (96.6) 115.4 (67.5) 43.500 0.147 Stavudine  Yes 12 (6.7) 139.0 (91.9)  No 167 (93.3) 118.1 (70.5) 50.000 0.715 ART  Yes 155 (86.6) 123.4 (73.7)  No 24 (13.4) 88.1 (32.1) 165.500 0.320 Parasitaemia with duration of ART use The AMPD was highest (128.4 ± 77.8) among those that received ART for <6 months, while the least parasite density (116.0 ± 45.7) was in those on ART for 18 to < 24 months. The lowest AMPD (88.1 ± 32.1) was among those that are ART naive. This relationship between AMPD and duration of ART was not statistically significant (Table 6). Table 6 Mean parasite density by duration of ART use ART duration (months) n Mean parasite density(±SD) Parasite/µl Nil 25  88.1 (32.1) 0 to < 6 22 128.4 (77.8) 6 to < 12 34 123.2 (82.7) 12 to < 18 9 124.6 (77.3) 18 to < 24 16 116.0 (45.7) ≥24 73 121.4 (76.9) ART duration (months) n Mean parasite density(±SD) Parasite/µl Nil 25  88.1 (32.1) 0 to < 6 22 128.4 (77.8) 6 to < 12 34 123.2 (82.7) 12 to < 18 9 124.6 (77.3) 18 to < 24 16 116.0 (45.7) ≥24 73 121.4 (76.9) Kruskal–Wallis test χ2 = 1.799, p = 0.876, df = 5. Table 6 Mean parasite density by duration of ART use ART duration (months) n Mean parasite density(±SD) Parasite/µl Nil 25  88.1 (32.1) 0 to < 6 22 128.4 (77.8) 6 to < 12 34 123.2 (82.7) 12 to < 18 9 124.6 (77.3) 18 to < 24 16 116.0 (45.7) ≥24 73 121.4 (76.9) ART duration (months) n Mean parasite density(±SD) Parasite/µl Nil 25  88.1 (32.1) 0 to < 6 22 128.4 (77.8) 6 to < 12 34 123.2 (82.7) 12 to < 18 9 124.6 (77.3) 18 to < 24 16 116.0 (45.7) ≥24 73 121.4 (76.9) Kruskal–Wallis test χ2 = 1.799, p = 0.876, df = 5. Parasitaemia and CTX prophylaxis Up to 46 of 154 subjects on CTX prophylaxis had positive result while 15 of 25 not receiving prophylaxis had positive malaria parasitaemia, a 3-fold risk (χ2 = 8.692, p = 0.003). The AMPD among subjects on prophylaxis of 127.5 ± 75.0 parasites/µl was not significantly higher than 91.9 ± 47.1 among those not on CTX, Mann–Whitney U = 242.500, p = 0.086. Parasitaemia and the use of ITN and window net The AMPD of infected children sleeping under ITNs was not significantly higher than those not sleeping under ITN; also for Children living in screened houses, AMPD was not significantly different from those living in unscreened houses (Table 7). Table 7 Mean parasite density of subjects by use of ITN and window net n (%) Mpd (±SD) Mann-Whitney U Window netting  Yes 114(63.7) 129.4 (71.3) 302.500; p=0.099  No 65 (36.3) 96.8 (65.0) ITN use  Yes 20 (11.2) 161.5 (101.2) 20.500; p=0.137  No 159 (88.8) 112.3 (63.5) n (%) Mpd (±SD) Mann-Whitney U Window netting  Yes 114(63.7) 129.4 (71.3) 302.500; p=0.099  No 65 (36.3) 96.8 (65.0) ITN use  Yes 20 (11.2) 161.5 (101.2) 20.500; p=0.137  No 159 (88.8) 112.3 (63.5) Table 7 Mean parasite density of subjects by use of ITN and window net n (%) Mpd (±SD) Mann-Whitney U Window netting  Yes 114(63.7) 129.4 (71.3) 302.500; p=0.099  No 65 (36.3) 96.8 (65.0) ITN use  Yes 20 (11.2) 161.5 (101.2) 20.500; p=0.137  No 159 (88.8) 112.3 (63.5) n (%) Mpd (±SD) Mann-Whitney U Window netting  Yes 114(63.7) 129.4 (71.3) 302.500; p=0.099  No 65 (36.3) 96.8 (65.0) ITN use  Yes 20 (11.2) 161.5 (101.2) 20.500; p=0.137  No 159 (88.8) 112.3 (63.5) DISCUSSION This study examines AMPD in HIV-infected children <5 years living in the holoendemic malaria region of Southern Nigeria. The study reveals significantly higher parasitaemia among HIV-infected compared with uninfected children. Malaria parasitaemia in infected children was higher at every age group, and the disparity was highest at infancy. Age was not a significant predictor of parasitaemia. The parasite density increased albeit non-significantly with worsening clinical staging and immunosuppression. Increased parasitaemia among HIV-infected children could be owing to the immune deficiencies and dysfunction attributable to HIV infection. HIV-related immunodeficiency is the probable cause, as the parasite may thrive better in infected than uninfected group. Malaria parasitaemia influences development of partial immunity [27], protection against clinical disease and ameliorates new infections in persons living in endemic areas [2, 28]. HIV-infected children therefore are at higher risk of having malaria parasitaemia, making HIV infection a risk factor for enhanced asymptomatic malaria parasitaemia [13–15]. The asymptomatic parasitaemia of 118.7 parasites/µl amongst HIV-infected children is higher than 85.2 parasites/µl reported by Adetifa et al. [29] in HIV-positive children 1–5 years in Lagos, South-West Nigeria. The reason for the difference in parasite density is not readily apparent. The AMPD reported in this study was smaller than 1922 parasites/µl reported by Owusu-Agyei et al. in Ghana among HIV-positive children 1–2 years and 1903 among 3–4-year-olds [15]. The study in Ghana reported geometric mean, while arithmetic mean was quoted in our study; this could have accounted for the large difference in parasitaemia. The pattern of parasitaemia in infected and uninfected children showed striking similarities with increasing age (Fig. 1). The narrow spectrum of parasitaemia may be owing to asymptomatic infection and small parasite densities recorded. Symptomatic malaria occurs with higher malaria parasitaemia [7]. Infancy was the peak of parasitaemia in both groups followed by a nadir at 12 to 24 months, to the staggered increase at 48 to <59 months. It seems probable therefore that infected children develop antimalarial immunity timely as non-infected peers albeit with higher parasitaemia. This is an area needing further exploration. Asymptomatic parasitaemia is necessary for development of specific antimalarial immunity and maintenance of the transmission cycle [7, 13–15, 30]. It can therefore be speculated that HIV infection does not distort the acquisition of antimalarial immunity. The finding of the highest parasitaemia and the worst disparity at infancy is an area needing further exploration. Comparative studies on HIV-infected/uninfected pregnant women have consistently reported higher parasitaemia in the infected groups. This may be partly owing to transient loss of acquired antimalarial immunity, leading to higher malaria and placental parasitaemia at delivery [19, 20]. The insufficient or defective transfer of antimalarial immune factors transplacentally from HIV-infected mothers could predispose their infants to higher AMPD as seen in this study. That uninfected infants had lower parasitaemia underscores this proposition. Therefore, infancy remains the critical point of enhanced parasitaemia in infected children. Appropriate vector avoidance is therefore recommended for malaria control. Malaria parasitaemia with HIV clinical disease stage and worsening immunosuppression is one of the objectives of this study. While parasitaemia increased with clinical staging, there was no consistent trend with immune category and the relationship was not significant. Clinical disease stage and immune category do not affect antimalarial immunity development among children <5 years. This finding differs from some studies on clinical (symptomatic) malaria [9, 17]. Apart from CD4+ T-cells, there are other factors required for development and sustenance of antimalarial immunity. Malaria parasitaemia was uninfluenced by ART, and there was increased parasitaemia among those on Zidovudine, Nevirapine, Lamivudine, Efavirenz and Stavudine. The duration of ART did not significantly affect parasitaemia. The reason for the non-association may be from lack of antimalarial properties by most ARTs. Their use, however, is known to enhance immune reconstitution and subsequently improve parasite clearance with attendant lower parasitaemia [31]. This is an expected occurrence with longer duration of ART administration. Our findings may be partly owing to the few number of children on protease inhibitors (Pi) (1.1%), which has distinct antimalarial properties [32]. The least parasitaemia, however, was among those that are ART naive. Parasitaemia in this study was unaffected by window netting or use of ITNs. The lack of association between ITN use and malaria parasite density is surprising, as ITN is a known factor for its beneficial vector avoidance property [31]. Although the fraction of infected children using ITNs is small (11.2%), infrequent use maybe contributory. Infected children on CTX had significantly lower prevalence of malaria parasitaemia as has been reported by other researchers and collaborated by this study. However, the finding of non-statistical higher parasitaemia among infected children on CTX is worrisome. This may be part of the higher AMPD associated with HIV infection; however, evidence abounds that CTX reduces symptomatic malaria incidence by 76%. ART and ITNs substantially reduce malaria frequency in HIV-infected adults [31]. In conclusion, malaria parasitaemia is higher in asymptomatic HIV-infected compared with uninfected children <5 years. In all age brackets, higher parasitaemia was maintained and a similar pattern is observed in both groups. Acquisition of antimalarial immunity in both groups may be similar albeit with higher parasitaemia in HIV-infected children. ACKNOWLEDGEMENT To all the children that attend the Paediatric HIV clinic in UBTH Benin City and all the staff that made this work possible, thank you for your support and immense contribution. FUNDING The study is self-funded with technical support of University of Benin Teaching Hospital, Benin City, Edo State, Nigeria. References 1 Joint United Nations Programme on HIV/AIDS (UNAIDS)/World Health Organization (WHO) . AIDS epidemic update: 2009. UNAIDS. Geneva: UNAIDS/WHO, 2009 . 2 Whitworth J , Morgan D , Quigley M , et al. Effect of HIV-1 and increasing immunosuppression on malaria parasitaemia and clinical episodes in adults in rural Uganda: a cohort study . Lancet 2000 ; 356 : 1051 – 6 . Google Scholar CrossRef Search ADS PubMed 3 Krogstad DJ. Malaria as a re-emerging disease . Epidemiol Rev 1996 ; 18 : 77 – 89 . Google Scholar CrossRef Search ADS PubMed 4 Samba E. Preface: the economic burden of malaria . Am J Trop Med Hyg 2001 ; 64 : ii . Google Scholar CrossRef Search ADS PubMed 5 Breman JG , Egan A , Keusch GT. The intolerable burden of malaria: a new look at the numbers . Am J Trop Med Hyg 2001 ; 64 (suppl): iv – vii . Google Scholar CrossRef Search ADS PubMed 6 Najera JA , Liese BH , Hammer J. Malaria In: Jamison DJ , Mosley WH , Measham AR , et al. (eds). Disease Control Priorities in Developing Countries . New York : Oxford University Press , 1993 , 281 – 302 . 7 Bruce-Chwatt LJ. Malaria in African infants and children in southern Nigeria . Ann Trop Med Parasitol 1952 ; 46 : 173 – 200 . Google Scholar CrossRef Search ADS PubMed 8 Vinetz JM , Gilman RH. Asymptomatic plasmodium parasitaemia and the ecology of malaria transmission . Am J Trop Med Hyg 2002 ; 66 : 639 – 40 . Google Scholar CrossRef Search ADS PubMed 9 Chandramohan D , Greenwood BM. Is there an interaction between human immunodeficiency virus and Plasmodium falciparum? Int J Epidemiol 1998 ; 27 : 296 – 301 . Google Scholar CrossRef Search ADS PubMed 10 Brown J , Greenwood BM , Terry RJ. Cellular mechanisms involved in recovery from acute malaria in Gambian children . Parasite Immunol 1986 ; 8 : 551 – 64 . Google Scholar CrossRef Search ADS PubMed 11 Atzori C , Bruno A , Chichino G , et al. HIV-1 and parasitic infections in rural Tanzania . Ann Trop Med Parasitol 1993 ; 87 : 585 – 93 . Google Scholar CrossRef Search ADS PubMed 12 Missinou MA , Lell B , Kremsner PG. Uncommon asymptomatic Plasmodium falciparum infections in Gabonese children . Clin Infect Dis 2003 ; 36 : 1198 – 202 . Google Scholar CrossRef Search ADS PubMed 13 World Health Organization . Malaria and HIV-1/AIDS interactions and implications: conclusions of a technical consultation convened by WHO, June 23–25, 2004. Report no. WHO/HIV/2004.08. Geneva: WHO, 2004 . 14 Alves FP , Durlacher RR , Menezes MJ , et al. High prevalence of asymptomatic plasmodium vivax and plasmodium falciparum infections in native Amazonian populations . Am J Trop Hyg 2002 ; 66 : 641 – 8 . Google Scholar CrossRef Search ADS 15 Owusu-Agyei S , Smith T , Beck H-P , et al. Molecular epidemiology of Plasmodium falciparum infections among asymptomatic inhabitants of a holoendemic malarious area in northern Ghana . Trop Med Int Health 2002 ; 5 : 421 – 8 . Google Scholar CrossRef Search ADS 16 French N , Nakiyingi J , Lugada E , et al. Increasing rates of malarial fever with deteriorating immune status in HIV-1-infected Ugandan adults . Aids 2001 ; 15 : 899 – 906 . Google Scholar CrossRef Search ADS PubMed 17 Patnaik P , Jere CS , Miller WC , et al. Effects of HIV-1 serostatus, HIV-1 RNA concentration, and CD4 cell count on the incidence of malaria infection in a cohort of adults in rural Malawi . J Infect Dis 2005 ; 192 : 984 – 91 . Google Scholar CrossRef Search ADS PubMed 18 Cohen C , Karstaedt A , Frean J , et al. Increased prevalence of severe malaria in HIV-infected adults in South Africa . Clin Infect Dis 2005 ; 41 : 1631 – 7 . Google Scholar CrossRef Search ADS PubMed 19 Steketee RW , Wirima JJ , Bloland PB , et al. Impairment of a pregnant woman's acquired ability to limit Plasmodium falciparum by infection with human immunodeficiency virus type-1 . Am J Trop Med Hyg 1996 ; 55 : 42 – 9 . Google Scholar CrossRef Search ADS PubMed 20 Verhoeff FH , Brabin BJ , Hart CA , et al. Increased prevalence of malaria in HIV-infected pregnant women and its implications for malaria control . Trop Med Int Health 1999 ; 4 : 5 – 12 . Google Scholar CrossRef Search ADS PubMed 21 Grimwade K , French N , Mbatha D , et al. HIV infection as a cofactor for severe falciparum malaria in adults living in a region of unstable malaria transmission in South Africa . Aids 2004 ; 18 : 547 – 54 . Google Scholar CrossRef Search ADS PubMed 22 Hay SL , Guerra CA , Tatem AJ , et al. The global distribution and population at risk of malaria: past, present, and future . Lancet Infect Dis 2004 ; 4 : 327 – 36 . Google Scholar CrossRef Search ADS PubMed 23 Federal Ministry of Health . Malaria in Nigeria . Nig Bull Epidemiol 1991 : 2 – 19 . 24 World Health Organization . Basic Malaria Microscopy, Part 1, Learners’Guide . A WHO Teaching Material. Switzerland : WHO , 1991 , 67 – 8 . 25 The Panel on Antiretroviral Therapy and Medical Management of HIV-Infected Children. Guidelines for the use of antiretroviral agents in pediatric HIV infection, 1 – 219 . http://aidsinfo.nih.gov/ContentFiles/PediatricGuidelines.pdf. 26 National Guidelines for Paediatric HIV & AIDS Treatment and Care . Nigeria : Federal Ministry of Health , 2010 , 1 – 233 . 27 Staalsoe T , Hviid L. The role of variant-specific immunity in asymptomatic malaria infections: maintaining a fine balance . Parasitol Today 1998 ; 14 : 177 – 8 . Google Scholar CrossRef Search ADS PubMed 28 Farnert A , Arez AP , Correia AT , et al. Sampling and storage of blood and the detection of malaria parasites by polymerase chain reaction . Trans R Soc Trop Med Hyg 1999 ; 93 : 50 – 3 . Google Scholar CrossRef Search ADS PubMed 29 Adetifa IM , Akinsulie AO , Temiye EO , et al. Effect of antiretroviral therapy on asymptomatic malaria parasitaemia in HIV-1 infected children . Nig Postgrad Med J 2008 ; 15 : 141 – 5 . 30 Orogade AA , Ogala WN , Aikhionbare HA. Asymptomatic malaria parasitaemia—a suitable index for evaluation of malaria vector control measures . Nig J Paediatr 2002 ; 29 : 23 – 6 . Google Scholar CrossRef Search ADS 31 Mermin J , Ekwaru JP , Liechty AL , et al. Effect of co-trimoxazole prophylaxis, antiretroviral therapy, and insecticide-treated bednets on the frequency of malaria in HIV-1-infected adults in Uganda . Lancet 2006 ; 367 : 1256 – 61 . Google Scholar CrossRef Search ADS PubMed 32 Parikh S , Gut J , Istvan E , et al. Antimalarial activity of human immunodeficiency virus type 1 protease inhibitors . Antimicrob Agents Chemother 2005 ; 49 : 2983 – 5 . Google Scholar CrossRef Search ADS PubMed © The Author [2017]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

2018 Journal of Tropical Pediatrics

doi: 10.1093/tropej/fmx066pmid: 29036732

Abstract Background The mortality of neonatal pyogenic meningitis is reduced to 10–15%, but morbidity is unchanged. Methods Primary objective is to determine the outcome, i.e. death or abnormal neurological examination (NE) at discharge and abnormal developmental quotient (DQ) at 3 months. Secondary objective is to find predictors of poor outcome. Results In all, 89 neonates enrolled, 10 expired and 24 neonates had abnormal NE at discharge. A total of 59 neonates came for follow up, 13 had DQ < 70. Prolonged shock (odds ratio, OR: 8.28; p = 0.001), coma (OR: 4.3; p = 0.001), seizures (OR: 14; p = 0.012), mechanical ventilation (OR: 18.55; p = 0.00), orogastric feeding (OR: 2.78; p = 0.042) and electroencephalography (EEG; OR: 9.6; p = 0.00) predicted poor short-term outcome. Abnormal NE at discharge (OR: 15.6; p = 0.001), EEG (OR = 10.60; p = 0.00) and brainstem-evoked reflex audiometry (OR = 37.20, p = 0.00) predicted a low DQ at 3 months. Mortality and morbidity of neonates with Pyogenic Meningitis (PM) were similar to that in developed countries. Outcome depended on severity of the disease and NE at discharge. neonatal meningitis, neurodevelopmental outcome, predictors INTRODUCTION Pyogenic meningitis is a devastating disease at any time of life. When it occurs in a neonate, the consequences are even more sinister. Mortality rates from developed countries range from 3% to 13% compared with 30–40% in developing countries, and 20–30% may have neurological sequelae [1–5, 6]. A number of predictors of poor outcome in neonates with PM have been ascertained like low birth weight or preterm birth, leukopenia (WBC < 5000/μl) and neutropenia at presentation [1, 7, 8], requirement for mechanical ventilatory support or inotropes [8, 9] and delayed sterilization of the cerebrospinal fluid (CSF) [10]. Most of these observations are from the West. Certain factors like the socio-economic status of the neonates, place of residence (urban/rural) and having received inadequate or irrational antibiotics with poor documentation before admission may be additional factors affecting outcome in resource-limited set-ups. Therefore, this study was carried out to ascertain the proportion of neonates with PM who die or survive with morbidity and to ascertain predictors of poor outcome in our set-up. METHODS Ours is a prospective observational study done in a public-sector tertiary-level neonatal unit of North India. The study was done over a period of 1 year after taking ethical approval from the ethics committee of the university. All consecutive term neonates admitted in the unit with clinical symptoms and signs of sepsis and diagnosed as PM formed the study subjects. Criteria for diagnosis of PM was neonates with clinical signs and symptoms of sepsis with a CSF showing >21 cells/cumm (> 6 polymorphs) or sugar < 36 mg/dl or protein > 129 mg/dl or positive CSF culture [11]. If the lumbar puncture (LP) was traumatic, the CSF was sent for culture and sugar estimation. LP was repeated after 48 h for cells, proteins, sugar and Gram stain and culture. Preterm neonates, neonates with birth asphyxia (Apgar score at 5 min <4 or Hypoxic Ischemic Encephalopathy (HIE) Stages II or III), major congenital malformations, metabolic disorders, chromosomal disorders, jaundice in exchange transfusion range and whose parents did not give consent were excluded. Baseline characteristics, clinical status at admission and during hospital stay and outcome were noted. Neonates were managed according to the unit protocol under supervision of the treating consultant. Cranial ultrasound (USG) and electroencephalography (EEG) were done before discharge. At the time of discharge a detailed neurological examination (NE) was done using the Amiel-Tison method [12]. The neonates were followed up in the out patient department. At 3 months of age, brainstem-evoked reflex audiometry (BERA), NE and neurodevelopmental assessment were done. Neurodevelopmental assessment was done by a psychologist using a tool by J Bharathraj [13]. Infants were classified as developmentally delayed if developmental quotient (DQ) was <70 [14]. EEG with moderate abnormalities (normal background with two or more abnormalities, asymmetry, asynchrony and mild voltage depression with additional abnormality) and marked abnormalities (burst suppression or marked depression/isoelectric background activity) were considered as abnormal [15]. Cranial USG was considered as showing a complication of PM when there was ventriculitis, parenchymal involvement or post-infectious hydrocephalus [16]. STATISTICAL ANALYSIS Variables of neonates with favourable and poor short-term and long-term outcomes were compared to detect predictors of adverse outcome. Analysis was performed using SPSS version 16.0 software. Univariate analysis was done using chi-square test for dichotomous variables and Student’s t-test for continuous variables. RESULTS During the study period, 1840 neonates were admitted in neonatal intensive care unit (NICU) and 180 had PM. Of these, 77 did not meet the inclusion criteria, as 47 were premature, 20 had birth asphyxia and 10 had congenital malformations. Of the remaining 103 neonates enrolled, 14 left against medical advice. So, 89 neonates with PM were finally analysed. Eighty-one neonates (91.0%) were hospital born and 21 were born by caesarean section; male to female ratio was 5.8:1, mean age at diagnosis was 11 days and mean duration of symptoms before admission was 3 days. Among the study subjects, 59 (66.2%) were from rural background and 67 (75%) patients belonged to low socio-economic status. The baseline characteristics of the study neonates are given in Table 1. Table 1 Baseline characteristics of neonates (n = 89) with pyogenic meningitis Variable No (%) Residence Urban 30 (33.7%) Rural 59 (66.2%) Socio-economic status Middle 22 (24.7%) Low 67 (75.2%) Place of delivery Home 8 (8.9%) Hospital 81 (91.0%) Mode of delivery Normal 68 (76.4%) Caesarean 21 (23.5%) Sex Male 76 (85.3%) Female 13 (14.6%) Duration of illness after birth <72 h 38 (42.6%) >72 h 51 (57.3%) Variable No (%) Residence Urban 30 (33.7%) Rural 59 (66.2%) Socio-economic status Middle 22 (24.7%) Low 67 (75.2%) Place of delivery Home 8 (8.9%) Hospital 81 (91.0%) Mode of delivery Normal 68 (76.4%) Caesarean 21 (23.5%) Sex Male 76 (85.3%) Female 13 (14.6%) Duration of illness after birth <72 h 38 (42.6%) >72 h 51 (57.3%) Socio-economic status means score: 11–25 (middle), 3–10 (low) by Kuppuswamy’s classification of socio-economic classes (2016). Table 1 Baseline characteristics of neonates (n = 89) with pyogenic meningitis Variable No (%) Residence Urban 30 (33.7%) Rural 59 (66.2%) Socio-economic status Middle 22 (24.7%) Low 67 (75.2%) Place of delivery Home 8 (8.9%) Hospital 81 (91.0%) Mode of delivery Normal 68 (76.4%) Caesarean 21 (23.5%) Sex Male 76 (85.3%) Female 13 (14.6%) Duration of illness after birth <72 h 38 (42.6%) >72 h 51 (57.3%) Variable No (%) Residence Urban 30 (33.7%) Rural 59 (66.2%) Socio-economic status Middle 22 (24.7%) Low 67 (75.2%) Place of delivery Home 8 (8.9%) Hospital 81 (91.0%) Mode of delivery Normal 68 (76.4%) Caesarean 21 (23.5%) Sex Male 76 (85.3%) Female 13 (14.6%) Duration of illness after birth <72 h 38 (42.6%) >72 h 51 (57.3%) Socio-economic status means score: 11–25 (middle), 3–10 (low) by Kuppuswamy’s classification of socio-economic classes (2016). As shown in Table 2, of 89 neonates with PM, 34 (38.2%) had poor short-term outcome, i.e. death (n = 10) or abnormal NE at discharge (n = 24). Prolonged shock (>24 h) (odds ratio, OR: 8.28; 95% confidence interval [CI]: 2.1–32.54; p = 0.001), prolonged seizures (>24 h) (OR: 14; 95% CI: 1.3–21.8; p = 0.012), prolonged coma (>24 h) (OR: 4.3; 95% CI: 1.73–10.71; p = 0.001), orogastric feeding > 3 days (OR: 2.78; 95% CI: 1.01–7.59; p = 0.042), prolonged mechanical ventilation (MV; >24 h) (OR: 18.55; 95% CI: 3.86–89.01; p = 0.00) and abnormal EEG (OR: 9.6; 95% CI: 3.08–29.85; p = 0.00) were associated with poor short-term outcome. Table 2 Predictors of short-term outcome in neonates with pyogenic meningitis (N = 89) Variable Good outcome (n = 55) Poor outcome (n = 34) p value n (%) n (%) Male sex 49 (89) 27 (79.4) 0.290 Rural residence 35 (63) 24 (70) 0.5 Low socio-economic status 42 (76.3) 25 (73.5) 0.76 Bulging fontanelle 10 (18.1) 10 (29.4) 0.217 Abnormal tone at admission 28 (50.9) 19 (55.8) 0.648 Sclerema 0 2 (5.8) 0.069 Prolonged shock (>24 h) 3 (5.4) 11 (32.3) OR: 8.28; 95% CI: 2.1–32.54; p = 0.001 Prolonged seizure (>24 h) 11 (12.5) 8 (23.5) OR: 14; 95% CI: 1.3–21.8; p = 0.012 Prolonged coma >24 h 15 (27.2) 21 (61.7) OR: 4.3; 95% CI: 1.73–10.71; p = 0.001 Orogastric feeding > 3 days 23 (41.8) 16 (47.0) OR: 2.78; 95% CI: 1.01–7.59; p = 0.042 Prolonged MV > 24 h 2 (3.6) 14 (41.1 OR: 18.55; 95% CI: 3.86–89.01; p = 0.00 Positive blood culture 14 (25.4) 7 (20.5 0.599 Abnormal EEG 7 (12.7) 14 (41.1) OR: 9.6; 95% CI: 3.08–29.85; p = 0.00 Abnormal cranial USG 6 (10.9) 1 (2.9 0.447 Variable Good outcome (n = 55) Poor outcome (n = 34) p value n (%) n (%) Male sex 49 (89) 27 (79.4) 0.290 Rural residence 35 (63) 24 (70) 0.5 Low socio-economic status 42 (76.3) 25 (73.5) 0.76 Bulging fontanelle 10 (18.1) 10 (29.4) 0.217 Abnormal tone at admission 28 (50.9) 19 (55.8) 0.648 Sclerema 0 2 (5.8) 0.069 Prolonged shock (>24 h) 3 (5.4) 11 (32.3) OR: 8.28; 95% CI: 2.1–32.54; p = 0.001 Prolonged seizure (>24 h) 11 (12.5) 8 (23.5) OR: 14; 95% CI: 1.3–21.8; p = 0.012 Prolonged coma >24 h 15 (27.2) 21 (61.7) OR: 4.3; 95% CI: 1.73–10.71; p = 0.001 Orogastric feeding > 3 days 23 (41.8) 16 (47.0) OR: 2.78; 95% CI: 1.01–7.59; p = 0.042 Prolonged MV > 24 h 2 (3.6) 14 (41.1 OR: 18.55; 95% CI: 3.86–89.01; p = 0.00 Positive blood culture 14 (25.4) 7 (20.5 0.599 Abnormal EEG 7 (12.7) 14 (41.1) OR: 9.6; 95% CI: 3.08–29.85; p = 0.00 Abnormal cranial USG 6 (10.9) 1 (2.9 0.447 Notes: Bold value means the P < 0.01 and statistically significant. Good outcome = discharged with normal NE. Poor outcome = expired or abnormal NE at discharge. Table 2 Predictors of short-term outcome in neonates with pyogenic meningitis (N = 89) Variable Good outcome (n = 55) Poor outcome (n = 34) p value n (%) n (%) Male sex 49 (89) 27 (79.4) 0.290 Rural residence 35 (63) 24 (70) 0.5 Low socio-economic status 42 (76.3) 25 (73.5) 0.76 Bulging fontanelle 10 (18.1) 10 (29.4) 0.217 Abnormal tone at admission 28 (50.9) 19 (55.8) 0.648 Sclerema 0 2 (5.8) 0.069 Prolonged shock (>24 h) 3 (5.4) 11 (32.3) OR: 8.28; 95% CI: 2.1–32.54; p = 0.001 Prolonged seizure (>24 h) 11 (12.5) 8 (23.5) OR: 14; 95% CI: 1.3–21.8; p = 0.012 Prolonged coma >24 h 15 (27.2) 21 (61.7) OR: 4.3; 95% CI: 1.73–10.71; p = 0.001 Orogastric feeding > 3 days 23 (41.8) 16 (47.0) OR: 2.78; 95% CI: 1.01–7.59; p = 0.042 Prolonged MV > 24 h 2 (3.6) 14 (41.1 OR: 18.55; 95% CI: 3.86–89.01; p = 0.00 Positive blood culture 14 (25.4) 7 (20.5 0.599 Abnormal EEG 7 (12.7) 14 (41.1) OR: 9.6; 95% CI: 3.08–29.85; p = 0.00 Abnormal cranial USG 6 (10.9) 1 (2.9 0.447 Variable Good outcome (n = 55) Poor outcome (n = 34) p value n (%) n (%) Male sex 49 (89) 27 (79.4) 0.290 Rural residence 35 (63) 24 (70) 0.5 Low socio-economic status 42 (76.3) 25 (73.5) 0.76 Bulging fontanelle 10 (18.1) 10 (29.4) 0.217 Abnormal tone at admission 28 (50.9) 19 (55.8) 0.648 Sclerema 0 2 (5.8) 0.069 Prolonged shock (>24 h) 3 (5.4) 11 (32.3) OR: 8.28; 95% CI: 2.1–32.54; p = 0.001 Prolonged seizure (>24 h) 11 (12.5) 8 (23.5) OR: 14; 95% CI: 1.3–21.8; p = 0.012 Prolonged coma >24 h 15 (27.2) 21 (61.7) OR: 4.3; 95% CI: 1.73–10.71; p = 0.001 Orogastric feeding > 3 days 23 (41.8) 16 (47.0) OR: 2.78; 95% CI: 1.01–7.59; p = 0.042 Prolonged MV > 24 h 2 (3.6) 14 (41.1 OR: 18.55; 95% CI: 3.86–89.01; p = 0.00 Positive blood culture 14 (25.4) 7 (20.5 0.599 Abnormal EEG 7 (12.7) 14 (41.1) OR: 9.6; 95% CI: 3.08–29.85; p = 0.00 Abnormal cranial USG 6 (10.9) 1 (2.9 0.447 Notes: Bold value means the P < 0.01 and statistically significant. Good outcome = discharged with normal NE. Poor outcome = expired or abnormal NE at discharge. However, male sex (27 vs. 49; p = 0.29), rural residence (24 vs. 35; p = 0.5), low socio-economic status (25 vs. 42; p = 0.7), bulging fontanelle (10 vs. 10; p = 0.217), abnormal tone on admission (19 vs. 28; p = 0.648), sclerema (2 vs. 0; p = 0.069), positive blood culture (7 vs. 14; p = 0.599) and abnormal cranial USG (1 vs. 6; p = 0.477) were not associated with poor short-term outcome. At 3 months of follow up, 59 of 79 (75%) neonates turned up and 13 (22%) had low DQ. Neonates with normal and abnormal DQ were compared. As shown in Table 2, abnormal NE at discharge (OR: 15.6; 95% CI: 3–80.6; p = 0.001), abnormal EEG (OR = 10.60; 95th CI = 2.46–45.54; p = 0.00) and abnormal BERA (OR = 37.20, 95th CI = 3.66–377.85; p = 0.00) were statistically significant for predicting low DQ at 3 months of age. None of the clinical parameters at admission, during hospital stay and laboratory variables including the CSF characteristics was statistically significant for predicting low DQ at 3 months of age. As shown in Table 3, among the various parameters of NE at discharge, abnormal fixation and tracking (OR = 7.60, 95th CI = 1.96–29.39, p = 0.002), abnormal social interaction (OR = 20.21, 95th CI = 0.906–450.70, p = 0.007), abnormal palmer grasp (OR = 13.75, 95th CI = 2.26–83.56, p = 0.001) and abnormal Moros (OR = 15.58, 95th CI = 3.01–80.66, p = 0.000) predicted a low DQ (<70) at 3 months of age. Table 3 Predictors of outcome at 3 months of age in neonates with pyogenic meningitis (n = 59) Variable Neonates with normal DQ (46) n (%) Neonates with abnormal DQ (13) n (%) p value Clinical/laboratory characteristics Prolonged seizures >24 h 6 (13) 3 (23) 0.081 Prolonged shock 5 (10.8) 2 (15.3) 0.657 Bulging fontanelle 8 (17.3) 3 (23) 0.642 Abnormal tone 23 (50) 9 (69.2) 0.219 Sclerema 0 1 (7.6) 0.058 Prolonged MV > 24 h 5 (10.8) 1 (7.6) 0.738 Prolonged ASM >24 h 15 (32.6) 7 (53.8) 0.162 Positive sepsis screen 37 (80.4) 8 (61.5) 0.157 Positive blood culture 7 (15.2) 4 (30.7) 0.290 Positive CSF culture 1 (2.1) 1 (7.6) 0.332 Abnormal EEG 11 (23.9) 10 (76.9) OR = 10.60, 95th CI = 2.46–45.54; p = 0.00 Abnormal cranial USG 3 (6.5) 5 (38.4) 0.311 Abnormal BERA 1 (2.1) 6 (46.1) OR = 37.20, 95th CI = 3.66–377.85; p = 0.0 Abnormal NE at discharge 12 (26) 11 (84) OR = 15.6, 95th CI = 3–80.6; p = 0.001 Abnormal fix and track 8 8 0.002 Abnormal ocular sign 0 0 – Abnormal response to voice 0 1 0.058 Abnormal social interaction 0 2 0.007 Abnormal crying 0 0 – Abnormal excitability 0 0 – Seizures at discharge 0 0 – Abnormal spontaneous activity 0.146 Mild abnormal 0 1 Gross abnormal 1 0 Spontaneous thumb abduction 1 1 0.332 Upper limb recoil Mild abnormal 5 3 0.300 Gross abnormal 1 1 Scarf sig Mild abnormal 2 1 0.776 Gross abnormal 1 0 Lower limb recoil Mild abnormal 2 1 0.776 Gross abnormal 1 0 Popliteal angle 1 0 0.592 Ventral incurvation 0 1 0.058 Dorsal incurvation 1 1 0.332 Righting reaction 0.742 Mild abnormal 3 1 Gross abnormal 2 0 Raise to sit Mild abnormal 2 3 0.082 Gross abnormal 2 0 Back to lie Mild abnormal 2 2 0.296 Gross abnormal 2 0 Palmar grasp 2 5 0.001 Automatic walking 2 2 0.162 Abnormal moros 12 11 0.00 Abnormal asymmetric tonic neck reflex 9 4 0.389 Variable Neonates with normal DQ (46) n (%) Neonates with abnormal DQ (13) n (%) p value Clinical/laboratory characteristics Prolonged seizures >24 h 6 (13) 3 (23) 0.081 Prolonged shock 5 (10.8) 2 (15.3) 0.657 Bulging fontanelle 8 (17.3) 3 (23) 0.642 Abnormal tone 23 (50) 9 (69.2) 0.219 Sclerema 0 1 (7.6) 0.058 Prolonged MV > 24 h 5 (10.8) 1 (7.6) 0.738 Prolonged ASM >24 h 15 (32.6) 7 (53.8) 0.162 Positive sepsis screen 37 (80.4) 8 (61.5) 0.157 Positive blood culture 7 (15.2) 4 (30.7) 0.290 Positive CSF culture 1 (2.1) 1 (7.6) 0.332 Abnormal EEG 11 (23.9) 10 (76.9) OR = 10.60, 95th CI = 2.46–45.54; p = 0.00 Abnormal cranial USG 3 (6.5) 5 (38.4) 0.311 Abnormal BERA 1 (2.1) 6 (46.1) OR = 37.20, 95th CI = 3.66–377.85; p = 0.0 Abnormal NE at discharge 12 (26) 11 (84) OR = 15.6, 95th CI = 3–80.6; p = 0.001 Abnormal fix and track 8 8 0.002 Abnormal ocular sign 0 0 – Abnormal response to voice 0 1 0.058 Abnormal social interaction 0 2 0.007 Abnormal crying 0 0 – Abnormal excitability 0 0 – Seizures at discharge 0 0 – Abnormal spontaneous activity 0.146 Mild abnormal 0 1 Gross abnormal 1 0 Spontaneous thumb abduction 1 1 0.332 Upper limb recoil Mild abnormal 5 3 0.300 Gross abnormal 1 1 Scarf sig Mild abnormal 2 1 0.776 Gross abnormal 1 0 Lower limb recoil Mild abnormal 2 1 0.776 Gross abnormal 1 0 Popliteal angle 1 0 0.592 Ventral incurvation 0 1 0.058 Dorsal incurvation 1 1 0.332 Righting reaction 0.742 Mild abnormal 3 1 Gross abnormal 2 0 Raise to sit Mild abnormal 2 3 0.082 Gross abnormal 2 0 Back to lie Mild abnormal 2 2 0.296 Gross abnormal 2 0 Palmar grasp 2 5 0.001 Automatic walking 2 2 0.162 Abnormal moros 12 11 0.00 Abnormal asymmetric tonic neck reflex 9 4 0.389 Notes: Bold value means the P < 0.01 and statistically significant. NE, neurological examination; MV, mechanical ventilation; ASM, altered sensorium; USG, ultra sound; BERA, brainstem-evoked reflex audiometry. Table 3 Predictors of outcome at 3 months of age in neonates with pyogenic meningitis (n = 59) Variable Neonates with normal DQ (46) n (%) Neonates with abnormal DQ (13) n (%) p value Clinical/laboratory characteristics Prolonged seizures >24 h 6 (13) 3 (23) 0.081 Prolonged shock 5 (10.8) 2 (15.3) 0.657 Bulging fontanelle 8 (17.3) 3 (23) 0.642 Abnormal tone 23 (50) 9 (69.2) 0.219 Sclerema 0 1 (7.6) 0.058 Prolonged MV > 24 h 5 (10.8) 1 (7.6) 0.738 Prolonged ASM >24 h 15 (32.6) 7 (53.8) 0.162 Positive sepsis screen 37 (80.4) 8 (61.5) 0.157 Positive blood culture 7 (15.2) 4 (30.7) 0.290 Positive CSF culture 1 (2.1) 1 (7.6) 0.332 Abnormal EEG 11 (23.9) 10 (76.9) OR = 10.60, 95th CI = 2.46–45.54; p = 0.00 Abnormal cranial USG 3 (6.5) 5 (38.4) 0.311 Abnormal BERA 1 (2.1) 6 (46.1) OR = 37.20, 95th CI = 3.66–377.85; p = 0.0 Abnormal NE at discharge 12 (26) 11 (84) OR = 15.6, 95th CI = 3–80.6; p = 0.001 Abnormal fix and track 8 8 0.002 Abnormal ocular sign 0 0 – Abnormal response to voice 0 1 0.058 Abnormal social interaction 0 2 0.007 Abnormal crying 0 0 – Abnormal excitability 0 0 – Seizures at discharge 0 0 – Abnormal spontaneous activity 0.146 Mild abnormal 0 1 Gross abnormal 1 0 Spontaneous thumb abduction 1 1 0.332 Upper limb recoil Mild abnormal 5 3 0.300 Gross abnormal 1 1 Scarf sig Mild abnormal 2 1 0.776 Gross abnormal 1 0 Lower limb recoil Mild abnormal 2 1 0.776 Gross abnormal 1 0 Popliteal angle 1 0 0.592 Ventral incurvation 0 1 0.058 Dorsal incurvation 1 1 0.332 Righting reaction 0.742 Mild abnormal 3 1 Gross abnormal 2 0 Raise to sit Mild abnormal 2 3 0.082 Gross abnormal 2 0 Back to lie Mild abnormal 2 2 0.296 Gross abnormal 2 0 Palmar grasp 2 5 0.001 Automatic walking 2 2 0.162 Abnormal moros 12 11 0.00 Abnormal asymmetric tonic neck reflex 9 4 0.389 Variable Neonates with normal DQ (46) n (%) Neonates with abnormal DQ (13) n (%) p value Clinical/laboratory characteristics Prolonged seizures >24 h 6 (13) 3 (23) 0.081 Prolonged shock 5 (10.8) 2 (15.3) 0.657 Bulging fontanelle 8 (17.3) 3 (23) 0.642 Abnormal tone 23 (50) 9 (69.2) 0.219 Sclerema 0 1 (7.6) 0.058 Prolonged MV > 24 h 5 (10.8) 1 (7.6) 0.738 Prolonged ASM >24 h 15 (32.6) 7 (53.8) 0.162 Positive sepsis screen 37 (80.4) 8 (61.5) 0.157 Positive blood culture 7 (15.2) 4 (30.7) 0.290 Positive CSF culture 1 (2.1) 1 (7.6) 0.332 Abnormal EEG 11 (23.9) 10 (76.9) OR = 10.60, 95th CI = 2.46–45.54; p = 0.00 Abnormal cranial USG 3 (6.5) 5 (38.4) 0.311 Abnormal BERA 1 (2.1) 6 (46.1) OR = 37.20, 95th CI = 3.66–377.85; p = 0.0 Abnormal NE at discharge 12 (26) 11 (84) OR = 15.6, 95th CI = 3–80.6; p = 0.001 Abnormal fix and track 8 8 0.002 Abnormal ocular sign 0 0 – Abnormal response to voice 0 1 0.058 Abnormal social interaction 0 2 0.007 Abnormal crying 0 0 – Abnormal excitability 0 0 – Seizures at discharge 0 0 – Abnormal spontaneous activity 0.146 Mild abnormal 0 1 Gross abnormal 1 0 Spontaneous thumb abduction 1 1 0.332 Upper limb recoil Mild abnormal 5 3 0.300 Gross abnormal 1 1 Scarf sig Mild abnormal 2 1 0.776 Gross abnormal 1 0 Lower limb recoil Mild abnormal 2 1 0.776 Gross abnormal 1 0 Popliteal angle 1 0 0.592 Ventral incurvation 0 1 0.058 Dorsal incurvation 1 1 0.332 Righting reaction 0.742 Mild abnormal 3 1 Gross abnormal 2 0 Raise to sit Mild abnormal 2 3 0.082 Gross abnormal 2 0 Back to lie Mild abnormal 2 2 0.296 Gross abnormal 2 0 Palmar grasp 2 5 0.001 Automatic walking 2 2 0.162 Abnormal moros 12 11 0.00 Abnormal asymmetric tonic neck reflex 9 4 0.389 Notes: Bold value means the P < 0.01 and statistically significant. NE, neurological examination; MV, mechanical ventilation; ASM, altered sensorium; USG, ultra sound; BERA, brainstem-evoked reflex audiometry. The flow diagram of the study is depicted in Fig. 1. Fig. 1. View largeDownload slide Flow chart of the study. Fig. 1. View largeDownload slide Flow chart of the study. Of the 24 neonates who had abnormal NE at discharge, 23 came for follow up at 3 months and 11 had abnormal DQ. Of the 55 neonates with normal NE at discharge, only 36 came for follow up and 2 had abnormal DQ. DISCUSSION We had 89 neonates with PM during the study period, 79 were discharged and 10 died. Of the 79 patients discharged, 24 had abnormal NE at the time of discharge. Therefore, of 89 neonates with PM, 34 had a poor short-term outcome and 13 (16.5%) of the survivors had a low DQ at 3 months. The literature describes a mortality rate of 3–13% in developed countries and 30–40% in developing countries for neonates with PM [1–5, 6]. In our study, mortality rate was 11.2%, which is similar to that of developed countries. We found that of the baseline characteristics like poor socio-economic status, mode of delivery and sex of baby did not predict a poor outcome. The seriousness of the illness as depicted by abnormal sensorium for >24 h (p = 0.001, OR = 4.30), prolonged shock (p = .000, OR = 8.28), prolonged MV (p = 0.000, OR = 18.55), prolonged seizures (p = 0.012, OR = 14) and orogastric feeding for >3 days (p = 0.042, OR = 2.78) predicted a poor short-term outcome including death. However, it did not predict an abnormal DQ at 3 months of age in the survivors. Our findings are similar to those found by Gill Klinger, Ben Hamouda H, Pekonen T and De Jonge et al. [8, 17–19]. In addition, some laboratory parameters like leukopenia (Gill Klinger, Philip AGS), pleocytosis < 500 (Ben Hamouda), Streptococcus pneumoniae (De Jonge), absolute neutrophil count < 1.0 × 103/ml, CSF protein > 0.3 gm/dl and C reactive protein > 20 mg% (Philip AGS) also predicted a poor outcome. In two recent studies, high CSF protein concentration prognosticated poor outcome in neonates with bacterial meningitis [20, 21]. However, we did not find any of the laboratory parameters as predictors of poor outcome of PM. In another study of neurodevelopmental outcome of neonatal meningitis by Tatishvili NA in Georgia, it was found that the outcome of bacterial meningitis depends on the starting point of the disease. Meningitis, which began earlier than 72 h of life, was characterized by severe prognosis [22]. In our study, the outcome of the neonates was the same whether it started within 72 h or after that. Our data suggest that the ultimate neurological prognosis of infants with PM should not be proclaimed with certainty at discharge. Almost half of the neonates assigned a poor outcome at discharge on the basis of abnormal NE had a normal DQ at 3 months. This appears to be a reassuring observation, but the ultimate outcome of these infants can be gauged only after long-term follow up when learning disabilities and subtle cognitive disabilities too can be picked up. Predictors of poor outcome (DQ < 70) at 3 months of age were an abnormal EEG (p = 0.000, OR = 10.60) and BERA (p = 0.000, OR = 37.20). Of the neurological signs at discharge, four could predict a low DQ at 3 months. These were abnormal fixation and tracking (p = 0.002, OR = 7.60), abnormal social interaction (p = 0.007, OR = 20.21), abnormal palmer grasp (p = 0.001, OR = 13.75) and abnormal Moro (p = 0.000, OR = 15.58). Abnormal EEG was a predictor of poor outcome at discharge and at 3 months of age. Results of our study are in agreement with the studies done previously. Klinger G reported that neonates with PM with normal or mildly abnormal EEGs had good outcomes, whereas those with moderate to markedly abnormal EEGs died or survived with adverse outcome [23]. Ter Horst reported that all infants with continuous low voltage or flat trace on EEG had an adverse outcome [24]. Lein also found hearing impairment during hospitalization to predict a poor outcome [21]. The strength of our study was that it was a prospective. The weakness was a small sample size, diagnosing PM without a positive CSF culture and high dropout rate of 25%. From our study, it is evident that neonatal meningitis still has a high mortality and morbidity. Abnormal NE at discharge (especially abnormal social interaction, fixation and tracking, Moros and Palmar grasp), EEG and abnormal BERA can help in prognosticating neonates with PM. References 1 Harvey D , Holt DE , Bedford H. Bacterial meningitis in the newborn: a prospective study of mortality and morbidity . Semin Perinatol 1999 ; 23 : 218 – 25 . Google Scholar CrossRef Search ADS PubMed 2 Louvois JD , Halket S , Harvey D. Neonatal meningitis in England and Wales: sequelae at 5 years of age . Eur J Pediatr 2005 ; 164 : 730 – 4 . Google Scholar CrossRef Search ADS PubMed 3 Bortolussi R , Krishnan C , Armstrong D , et al. Prognosis for survival in neonatal meningitis: clinical and pathologic review of 52 cases . Can Med Assoc J 1978 ; 118 : 165 – 8 . Google Scholar PubMed 4 Chang CJ , Chang WN , Huang LT , et al. Neonatal bacterial meningitis in southern Taiwan . Pediatr Neurol 2003 ; 29 : 288 – 294 . Google Scholar CrossRef Search ADS PubMed 5 Smith PB , Cotten CM , Garges HP , et al. A comparison of neonatal Gram-negative rod and Gram-positive cocci meningitis . J Perinatol 2006 ; 26 : 111 – 14 . Google Scholar CrossRef Search ADS PubMed 6 Philip AGS. Neonatal meningitis in the New Millennium . Neoreviews 2003 ; 4 : e73 . DOI: 10.1542/neo.4-3-e73. Google Scholar CrossRef Search ADS 7 May M , Daley AJ , Donath S , et al. Early onset neonatal meningitis in Australia and New Zealand, 1992-2002 . Arch Dis Child Fetal Neonatal Ed 2005 ; 90 : F324 – 7 . Google Scholar CrossRef Search ADS PubMed 8 Klinger G , Chin CN , Beyene J , et al. Predicting the outcome of neonatal bacterial meningitis . Pediatrics 2000 ; 106 : 477 – 82 . Google Scholar CrossRef Search ADS PubMed 9 Franco SM , Cornelius VE , Andrews BF. Long-term outcome of neonatal meningitis . Am J Dis Child 1992 ; 146 : 567 – 71 . Google Scholar PubMed 10 Greenberg RG , Benjamin DK Jr , Cohen-Wolkowiez M , et al. Repeat lumbar punctures in infants with meningitis in the neonatal intensive care unit . J Perinatol 2011 ; 31 : 425 – 9 . Google Scholar CrossRef Search ADS PubMed 11 Stoll BJ , Shane AL. Infections of the neonatal infant. In: Kliegman S , St Geme S (eds). Nelson text book of paediatrics . 21th edn . 917 . 12 Amiel-Tison C. Update of the Amiel-Tison neurologic assessment for the term neonate or at 40 weeks corrected age . Pediatr Neurol 2002 ; 27 : 196 – 212 . Google Scholar CrossRef Search ADS PubMed 13 Bharathraj J. Manual on Developmental Screening Test . Mysore : Swayamsiddha Prakashana , 1977 , 1 – 2 . 14 Bharathraj J. Indian Adaptation of Vineland Social Maturity Scale: Enlarged Version . Mysore : Swayamsiddha Prakashana , 1992 . 15 Bedford H , de Louvois J , Halket S , et al. Meningitis in infancy in England and Wales: follow-up at age 5 years . BMJ 2001 ; 323 : 1 – 5 . Google Scholar CrossRef Search ADS PubMed 16 Yikilmaz A , Taylor GA. Sonographic findings in bacterial meningitis in neonates and young infants . Pediatr Radiol 2008 ; 38 : 129 – 37 . Google Scholar CrossRef Search ADS PubMed 17 Ben Hamouda H , Ben Haj KA , Hamza MA , et al. Clinical outcome and prognosis of neonatal bacterial meningitis . Arch Pediatr 2013 ; 20 : 938 – 44 . Google Scholar CrossRef Search ADS PubMed 18 Pelkonen T , Roine I , Monteiro L , et al. Risk factors for death and severe neurological sequelae in childhood bacterial meningitis in sub-Saharan Africa . Clin Infect Dis 2009 ; 48 : 1107 – 10 . Google Scholar CrossRef Search ADS PubMed 19 De Jonge RC , van Furth AM , Wassenaar M , et al. Predicting sequelae and death after bacterial meningitis in childhood: a systematic review of prognostic studies . BMC Infect Dis 2010 ; 10 : 232 . Google Scholar CrossRef Search ADS PubMed 20 Tan J , Kan J , Qiu G , et al. Clinical prognosis in neonatal bacterial meningitis: the role of cerebrospinal fluid protein . PLoS One 2015 ; 10 : e0141620 . doi: 10.1371/journal.pone.0141620. eCollection 2015. Google Scholar CrossRef Search ADS PubMed 21 Lin MC , Chi H , Chiu NC , et al. Factors for poor prognosis of neonatal bacterial meningitis in a medical center in Northern Taiwan . J Microbiol Immunol Infect 2012 ; 45 : 442 – 7 . Google Scholar CrossRef Search ADS PubMed 22 Tatishvili NA , Sirbiladze TV , Kipiani TB , et al. Early predictors of neurodevelopmental outcome of neonatal bacterial meningitis [in Russian] . Georgian Med News 2005 ; 129 : 82 – 4 . 23 Klinger G , Chin C , Hiroshi O , et al. Prognostic value of EEG in neonatal bacterial meningitis . Pediatr Neurol 2001 ; 24 : 28 – 31 . Google Scholar CrossRef Search ADS PubMed 24 Ter Horst HJ , Van Olffen M , Remmelts HJ , et al. The prognostic value of amplitude integrated EEG in neonatal sepsis and/or meningitis . Acta Paediatr 2010 ; 99 : 194 – 200 . Google Scholar PubMed © The Author [2017]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

2018 Journal of Tropical Pediatrics

doi: 10.1093/tropej/fmx067pmid: 28977670

Abstract Congenital abnormalities are important causes of morbidity and mortality in children and significantly add to the burdens on healthcare in developing countries. Unfortunately, there remains a paucity of information on congenital birth defects in most developing countries. This is a 4-year prospective study that assessed the patterns and predictors of congenital anomalies among newborns at the Enugu State University Teaching Hospital, Nigeria. In total, 5830 deliveries were recorded, of which 38 had congenital anomalies, giving an incidence rate of 6.5/1000 live births. Fifty-two newborns were enrolled as nested controls. Factors significantly associated with congenital anomalies were low birth weight (p = 0.009), low socio-economic class (p = 0.011), lower maternal educational attainment (p = 0.009), parity of ≥ 5 (p = 0.002), febrile illness (p = 0.001) and the use of local concoction in index pregnancy (p = 0.009). More than half of the anomalies reported involved the musculoskeletal system. Occurrence of congenital anomalies may be prevented by curtailing risk factors identified in this study. congenital anomalies, newborn, incidence, pattern, Enugu INTRODUCTION Congenital malformations also referred to as birth defects are anomalies that occur in utero and are usually detected at birth [1]. These malformations can be structural or functional and can occur in isolation, in association, or as part of a syndrome. Congenital abnormalities are important cause of long-term morbidity, disability and mortality in children. They, most of the time, limit the newborns’ ability to adapt to extra-uterine life and may significantly reduce their quality of life. The impairments and handicaps that result from congenital malformations usually pose serious burdens on the health, social and financial well-being of the child, his/her family and the healthcare system. According to a 5-year mortality report released by the World Health Organization in 2005, >300 000 newborns die within the first month of life because of complications associated with congenital malformations [1]. A 23-year retrospective study done in south-south Nigeria estimated the prevalence of congenital anomalies in newborns to be 0.35% [2]. Although the exact cause of congenital malformations is usually unknown in about 50% of cases [3], it is however widely believed that most birth defects can be prevented by simple measures such as vaccination, folic acid supplementation, infection prevention and avoidance of certain indulgences like smoking, alcohol and use of certain drugs during pre-pregnancy and gestational periods [1]. In 2010, the World Health Organization in collaboration with other international organizations through the resolution on birth defects of the Sixty-third World Health Assembly recommended the promotion of primary prevention and improvement of the health of children with congenital anomalies. To achieve this, they proposed developing expertise on preventive strategies, developing and strengthening registration and surveillance systems and strengthening research and studies on aetiologies and diagnosis of congenital anomalies [1]. Despite these recommendations, there is a paucity of information on congenital birth defect in Nigeria and perhaps most developing countries because of little or no surveillance activities in this area. This 4-year passive surveillance study examined the pattern of congenital anomalies seen among newborns delivered in the Enugu State University Teaching Hospital (ESUTH). Secondarily, it also sought to explore maternal socio-demographic factors associated with the development of these anomalies in newborns. It is believed that the findings of this study will contribute to the body of knowledge on this subject and further strengthen practices that encourage prevention of birth defects and its associated medico-social consequences in children. METHODOLOGY Study area and site This was a prospective study carried out between the labour ward and the neonatal intensive care unit (NICU) of ESUTH, Parklane. The site is located within Enugu, the capital city of Enugu State, south-east Nigeria. ESUTH is a tertiary health institution that offers specialized medical services and serves as a referral centre to Private, General, Mission Hospitals and other delivery homes within Enugu and neighbouring States. The NICU offers 24-h services to newly delivered babies for observation and sick babies born within and outside the hospital. It is located in close proximity to the labour ward, which has an average of 144 deliveries per month. The special neonatal unit is manned by consultant neonatologists and resident doctors who are specialists in Paediatrics with further sub-specialist training in neonatology. Newborn participation and enrolment The study was carried out over a period of 4-year (January 2013–January 2017). Newborn babies of consenting mothers delivered in ESUTH that had congenital anomalies were consecutively enrolled and followed up with daily reviews till discharge or death. For every newborn with congenital anomalies enrolled, at least one newborn delivered around the same time matched as close as possible for anthropometric parameters was also enrolled. All anomalies encountered in enrolled newborns after delivery were documented. Other information collected included gestational age at delivery, anthropometric measurements as well as important obstetrics history. Some of these included maternal socio-demographic parameters, febrile illness during pregnancy, use of unprescribed medications and/or local concoction and pre-pregnant use of folic acid. The diagnosis of congenital malformations was based on clinical examination and laboratory and radiological investigations where appropriate. For the sake of this study, these diagnoses were made by a neonatologist who was not involved in this study. A second neonatologist was also brought in to confirm or refute the diagnosis in all cases. Data entry and analysis The above measures were documented at presentation in the relevant sections of the questionnaire and subsequently transferred into a Microsoft Excel Sheet. Distribution of the measures of outcome and predictor variables was analysed and recorded in percentages. Enrolees with significant missing information were excluded from the data analysis. The chi-square was used to assess variables significantly associated with congenital malformations in newborn. Data were analysed using IBM® SPSS version 18.0 (SPSS Inc, Chicago, IL). Statistical significance was set at p < 0.05. Ethical consideration Ethical clearance was obtained from the ESUTH Ethics Committee. Before recruitment of each subject, informed consent was obtained from every mother–newborn pair. Participation in the study was entirely voluntary, and no financial inducement whatsoever was involved. Participants were informed that voluntary withdrawal at any stage of interaction was guaranteed for them without any adverse effect to the mother or their baby. All information was handled with strict confidentiality. RESULT Characteristics of study respondents The surveillance for congenital abnormalities in newborn delivered at the ESUTH was carried out during a 4-year period. During this period, 5830 babies were delivered in the labour ward of the ESUTH giving an average of 122 births per month. Thirty-eight (∼0.7%) of these babies were born with one or more congenital anomalies giving an in-hospital incidence of rate of 6.5 per 1000 live births. Table 1 shows a summary of the characteristics of mothers and newborns enrolled for this study. In all, 90 newborns were enrolled with a male–female ratio of 1:4. Thirty-eight (48%) were born with one or more congenital malformation, while the remaining 52 (58%) had none. In total, 77% of the enrolled newborns were term deliveries and the remaining 23% were preterm (i.e. deliveries before 37 completed weeks of gestation). In total, 71% were born with a birth weight of ≥2.5 kg, while the rest were <2.5 kg at birth. In total, 11 of the 90 enrolled newborns died before discharge from the hospital and all were babies with congenital malformation. Majority (85%) of the mothers surveyed were <35 years old and 93% had parity of ≤ 4 at conception for the index baby. About two-thirds (63%) of the 90 respondents had post-secondary school education with the other 32 and 5% completing secondary and primary education, respectively. Table 1. Maternal and newborn parameters associated with occurrence of congenital deformity Features Congenital malformation present χ2 Total Yes No pa Gestational age at delivery (n, 90)  Term 69 (77) 29 (48) 40 (58) 0.005  Preterm 21 (23) 9 (43) 12 (57) 0.946 Gender of newborn (n, 90)  Male 53 (59) 23 (43) 30 (57) 0.073  Female 37 (41) 15 (41) 22 (59) 0.787 Birth weight (n, 90)  ≥2.5 kg 71 (79) 25 (35) 46 (65) 6.776  <2.5 kg 19 (21) 13 (68) 6 (32) 0.009 Weight of newborn for gestational age (n, 88)  Adequate 77 (88) 31 (40) 46 (60) 0.107  Small 11 (12) 5 (45) 6 (55) 0.743 Maternal age at conception for index pregnancy (n, 88)  ≥35 years 13 (15) 7 (54) 6 (46) 1.056  <35 years 75 (85) 29 (39) 46 (61) 0.304 Socio-economic class (n, 88)  High 25 (28) 5 (20) 20 (80) 9.076  Middle 33 (38) 13 (39) 20 (61) 0.011  Low 30 (34) 18 (60) 12 (40) Mother educational attainment (n, 87)  Post-secondary 55 (63) 16 (29) 39 (71) 10.85  Completed secondary 28 (32) 15 (54) 13 (46) 0.004  Completed primary or less 4 (5) 4 (100) 0 (0) Parity of mother at conception (n, 88)  ≤4 82 (93) 30 (37) 52 (63) 9.301  ≥5 6 (7) 6 (100) 0 (0) 0.002 Antenatal care in index pregnancy (n, 90)  Yes 76 (84) 34 (5) 42 (55) 1.266  No 14 (16) 4 (29) 10 (71) 0.260 Febrile±rash illness in index pregnancy (n, 86)  Yes 44 (51) 25 (57) 19 (43) 11.26  No 42 (49) 9 (21) 33 (79) 0.001 Local concoction use in index pregnancy (n, 90)  Yes 19 (21) 13 (68) 6 (32) 6.776  No 71 (79) 25 (35) 46 (65) 0.009 Self-medication in index pregnancy (n, 83)  Yes 23 (28) 7 (30) 16 (70) 0.650  No 60 (72) 24 (40) 36 (60) 0.420 Pre-pregnancy folic acid use in index pregnancy (n, 87)  Yes 8 (9) 2 (25) 6 (75) 0.850  No 79 (91) 33 (42) 46 (58) 0.357 Asphyxia in newborn at birth (n, 90)  Yes 15 (17) 13 (87) 2 (13) 14.58  No 75 (83) 25 (33) 50 (67) 0.001 Outcome of newborn (n, 86)  Alive 75 (87) 23 (31) 52 (69) 19.29  Died 11 (13) 11 (100) 0 (0) 0.000 Features Congenital malformation present χ2 Total Yes No pa Gestational age at delivery (n, 90)  Term 69 (77) 29 (48) 40 (58) 0.005  Preterm 21 (23) 9 (43) 12 (57) 0.946 Gender of newborn (n, 90)  Male 53 (59) 23 (43) 30 (57) 0.073  Female 37 (41) 15 (41) 22 (59) 0.787 Birth weight (n, 90)  ≥2.5 kg 71 (79) 25 (35) 46 (65) 6.776  <2.5 kg 19 (21) 13 (68) 6 (32) 0.009 Weight of newborn for gestational age (n, 88)  Adequate 77 (88) 31 (40) 46 (60) 0.107  Small 11 (12) 5 (45) 6 (55) 0.743 Maternal age at conception for index pregnancy (n, 88)  ≥35 years 13 (15) 7 (54) 6 (46) 1.056  <35 years 75 (85) 29 (39) 46 (61) 0.304 Socio-economic class (n, 88)  High 25 (28) 5 (20) 20 (80) 9.076  Middle 33 (38) 13 (39) 20 (61) 0.011  Low 30 (34) 18 (60) 12 (40) Mother educational attainment (n, 87)  Post-secondary 55 (63) 16 (29) 39 (71) 10.85  Completed secondary 28 (32) 15 (54) 13 (46) 0.004  Completed primary or less 4 (5) 4 (100) 0 (0) Parity of mother at conception (n, 88)  ≤4 82 (93) 30 (37) 52 (63) 9.301  ≥5 6 (7) 6 (100) 0 (0) 0.002 Antenatal care in index pregnancy (n, 90)  Yes 76 (84) 34 (5) 42 (55) 1.266  No 14 (16) 4 (29) 10 (71) 0.260 Febrile±rash illness in index pregnancy (n, 86)  Yes 44 (51) 25 (57) 19 (43) 11.26  No 42 (49) 9 (21) 33 (79) 0.001 Local concoction use in index pregnancy (n, 90)  Yes 19 (21) 13 (68) 6 (32) 6.776  No 71 (79) 25 (35) 46 (65) 0.009 Self-medication in index pregnancy (n, 83)  Yes 23 (28) 7 (30) 16 (70) 0.650  No 60 (72) 24 (40) 36 (60) 0.420 Pre-pregnancy folic acid use in index pregnancy (n, 87)  Yes 8 (9) 2 (25) 6 (75) 0.850  No 79 (91) 33 (42) 46 (58) 0.357 Asphyxia in newborn at birth (n, 90)  Yes 15 (17) 13 (87) 2 (13) 14.58  No 75 (83) 25 (33) 50 (67) 0.001 Outcome of newborn (n, 86)  Alive 75 (87) 23 (31) 52 (69) 19.29  Died 11 (13) 11 (100) 0 (0) 0.000 Notes: Bold values of p are statistically significant. a Fischer’s exact test used in estimation where applicable. Table 1. Maternal and newborn parameters associated with occurrence of congenital deformity Features Congenital malformation present χ2 Total Yes No pa Gestational age at delivery (n, 90)  Term 69 (77) 29 (48) 40 (58) 0.005  Preterm 21 (23) 9 (43) 12 (57) 0.946 Gender of newborn (n, 90)  Male 53 (59) 23 (43) 30 (57) 0.073  Female 37 (41) 15 (41) 22 (59) 0.787 Birth weight (n, 90)  ≥2.5 kg 71 (79) 25 (35) 46 (65) 6.776  <2.5 kg 19 (21) 13 (68) 6 (32) 0.009 Weight of newborn for gestational age (n, 88)  Adequate 77 (88) 31 (40) 46 (60) 0.107  Small 11 (12) 5 (45) 6 (55) 0.743 Maternal age at conception for index pregnancy (n, 88)  ≥35 years 13 (15) 7 (54) 6 (46) 1.056  <35 years 75 (85) 29 (39) 46 (61) 0.304 Socio-economic class (n, 88)  High 25 (28) 5 (20) 20 (80) 9.076  Middle 33 (38) 13 (39) 20 (61) 0.011  Low 30 (34) 18 (60) 12 (40) Mother educational attainment (n, 87)  Post-secondary 55 (63) 16 (29) 39 (71) 10.85  Completed secondary 28 (32) 15 (54) 13 (46) 0.004  Completed primary or less 4 (5) 4 (100) 0 (0) Parity of mother at conception (n, 88)  ≤4 82 (93) 30 (37) 52 (63) 9.301  ≥5 6 (7) 6 (100) 0 (0) 0.002 Antenatal care in index pregnancy (n, 90)  Yes 76 (84) 34 (5) 42 (55) 1.266  No 14 (16) 4 (29) 10 (71) 0.260 Febrile±rash illness in index pregnancy (n, 86)  Yes 44 (51) 25 (57) 19 (43) 11.26  No 42 (49) 9 (21) 33 (79) 0.001 Local concoction use in index pregnancy (n, 90)  Yes 19 (21) 13 (68) 6 (32) 6.776  No 71 (79) 25 (35) 46 (65) 0.009 Self-medication in index pregnancy (n, 83)  Yes 23 (28) 7 (30) 16 (70) 0.650  No 60 (72) 24 (40) 36 (60) 0.420 Pre-pregnancy folic acid use in index pregnancy (n, 87)  Yes 8 (9) 2 (25) 6 (75) 0.850  No 79 (91) 33 (42) 46 (58) 0.357 Asphyxia in newborn at birth (n, 90)  Yes 15 (17) 13 (87) 2 (13) 14.58  No 75 (83) 25 (33) 50 (67) 0.001 Outcome of newborn (n, 86)  Alive 75 (87) 23 (31) 52 (69) 19.29  Died 11 (13) 11 (100) 0 (0) 0.000 Features Congenital malformation present χ2 Total Yes No pa Gestational age at delivery (n, 90)  Term 69 (77) 29 (48) 40 (58) 0.005  Preterm 21 (23) 9 (43) 12 (57) 0.946 Gender of newborn (n, 90)  Male 53 (59) 23 (43) 30 (57) 0.073  Female 37 (41) 15 (41) 22 (59) 0.787 Birth weight (n, 90)  ≥2.5 kg 71 (79) 25 (35) 46 (65) 6.776  <2.5 kg 19 (21) 13 (68) 6 (32) 0.009 Weight of newborn for gestational age (n, 88)  Adequate 77 (88) 31 (40) 46 (60) 0.107  Small 11 (12) 5 (45) 6 (55) 0.743 Maternal age at conception for index pregnancy (n, 88)  ≥35 years 13 (15) 7 (54) 6 (46) 1.056  <35 years 75 (85) 29 (39) 46 (61) 0.304 Socio-economic class (n, 88)  High 25 (28) 5 (20) 20 (80) 9.076  Middle 33 (38) 13 (39) 20 (61) 0.011  Low 30 (34) 18 (60) 12 (40) Mother educational attainment (n, 87)  Post-secondary 55 (63) 16 (29) 39 (71) 10.85  Completed secondary 28 (32) 15 (54) 13 (46) 0.004  Completed primary or less 4 (5) 4 (100) 0 (0) Parity of mother at conception (n, 88)  ≤4 82 (93) 30 (37) 52 (63) 9.301  ≥5 6 (7) 6 (100) 0 (0) 0.002 Antenatal care in index pregnancy (n, 90)  Yes 76 (84) 34 (5) 42 (55) 1.266  No 14 (16) 4 (29) 10 (71) 0.260 Febrile±rash illness in index pregnancy (n, 86)  Yes 44 (51) 25 (57) 19 (43) 11.26  No 42 (49) 9 (21) 33 (79) 0.001 Local concoction use in index pregnancy (n, 90)  Yes 19 (21) 13 (68) 6 (32) 6.776  No 71 (79) 25 (35) 46 (65) 0.009 Self-medication in index pregnancy (n, 83)  Yes 23 (28) 7 (30) 16 (70) 0.650  No 60 (72) 24 (40) 36 (60) 0.420 Pre-pregnancy folic acid use in index pregnancy (n, 87)  Yes 8 (9) 2 (25) 6 (75) 0.850  No 79 (91) 33 (42) 46 (58) 0.357 Asphyxia in newborn at birth (n, 90)  Yes 15 (17) 13 (87) 2 (13) 14.58  No 75 (83) 25 (33) 50 (67) 0.001 Outcome of newborn (n, 86)  Alive 75 (87) 23 (31) 52 (69) 19.29  Died 11 (13) 11 (100) 0 (0) 0.000 Notes: Bold values of p are statistically significant. a Fischer’s exact test used in estimation where applicable. Predictors of congenital malformation in newborns Table 1 shows a cross-tabulation of maternal and newborn parameters considered in this study. Of parameters analysed only birth weight, socio-economic class, maternal educational attainment, parity, febrile illness with or without rash and use of local concoction in index pregnancy were significantly associated with congenital malformations in newborns. In total, 68% of newborns with birth weight <2.5 kg compared with 35% of newborns with birth weight ≥2.5 kg had congenital malformation (p = 0.009). Mothers in the low socio-economic class had 60% of the babies with congenital malformation compared with 39 and 20% for mother in the middle and high socio-economic class (p = 0.011). Similarly, it was noted that only 29% of mothers with post-secondary school education had a congenitally malformed baby compared with those with lower educational attainment (54% secondary education vs. 100% primary education or less; p = 0.009). Additionally, 37% of mothers with ≤4 children at the time of conception compared with 100% of mothers with ≥5 children at conception had congenitally malformed babies (p = 0.002). Furthermore, it was seen that 57% of mothers that had febrile illnesses and 68% that used local concoction during the index pregnancy compared with 21% without febrile illness and 35% of mother that did not use local concoction had babies with congenital anomalies (p = 0.001 and p = 0.009, respectively). Finally, mothers ≥35 years had more babies with congenital malformation (54%) than mothers that were ≤34 years (39%). This however did not attain statistical significance (p = 0.304). More babies with congenital anomalies (87%) than those without such anomalies (33%) had perinatal asphyxia (p = 0.001), and significantly, more babies with congenital anomalies died while in the NICU compared with babies without congenital anomalies (p = 0.000). Systems and organs involved in congenital malformations encountered in newborns Table 2 shows congenital malformation encountered in surveyed newborns occurred in isolation, in association or as known syndromes. A total of 133 clinically or radiological diagnosed anomalies were documented in the 38 newborns with congenital malformations. Seventeen (45%) of these newborns had anomalies that occurred in syndromic patterns, while the other 21 (55%) newborns had anomalies that occurred in isolations and/or associations that affected one or more systems. The syndromes encountered include Down syndrome, 8 (47%); Turner syndrome, 2 (12%); and congenital heart defects, 3 (17%). Others were Edward syndrome, 1 (6%); Potter’s syndrome, 1 (6%); osteogenesis imperfect, 1 (6%); and Smith–Lemli–Opitz, 1 (6%). The musculoskeletal system was the most affected with slightly >50% of deformities associated with this system. Twenty-two (16.5%) of abnormalities involved the digestive system, while the central nervous system (10.5%), the urogenital (8.0%) and integumentary system (8.3%) were approximately equally affected. Surprisingly, the cardiovascular system was the least involved system (6.7%) but was present in close to half of the newborns (45.5%) with congenital abnormalities that died after birth. Table 2 Deformitiesa encountered in newborns with congenital malformation Musculoskeletal system n, 67  Flat nasal bridge 6 (8.9) Right forearm shortening 1 (1.5)  Medial canthus hypertelorism 9 (13.4) Absent left forearm 1 (1.5)  Talipes deformities 8 (11.9) Absence of three medial fingers 2 (3.0)  Low set ears 9 (13.4) Syndactyly 1 (1.5)  Sanders feet 1 (1.5) Macrosomia 2 (3.0)  Toe misalignment 2 (3.0) Short neck 1 (1.5)  Receding chin 1 (1.5) Short webbed neck 5 (7.5)  Fixed flexing deformity of the wrist 2 (3.0) Left hemihypertrophy 1 (1.5)  Prominent philtrum 2 (3.0) Rocker bottom foot 1 (1.5)  High arched feet 1 (1.5) Anterior abdominal wall defect 1 (1.5)  Short lower limb 1 (1.5) Polydactyly 2 (3.0)  Malformed pinna 2 (3.0) Micro-opthalmia 2 (3.0)  Short sturdy fingers 2 (3.0) High arched palate 1 (1.5) Central nervous system n, 14  Microcephaly 3 (21.4) Sutural diastasis 1 (7.1)  Encephalocele 2 (14.3) Prominent occipitus 1 (7.1)  Meningocele 2 (14.3) Hydrocephalocele 1 (7.1)  Spinal bifidia occulta 2 (14.3) Macrocephaly 2 (14.3) Urogenital system n, 10  Hypospadia 1 (10.0) Clitomegaly 2 (20.0)  Micropenis 2 (20.0) Poorly developed scrotum 1 (10.0)  Bilateral undescended testis 1 (10.0) Cryptorchidism 1 (10.0)  Abnormal external genitalia 1 (10.0) Bilateral nephromegaly 1 (10.0) Digestive system n, 22  Tracheooesphageal fistula 2 (9.1) Imperforate anus 2 (9.1)  Duodenal atresia 2 (9.1) Macroglossia 2 (9.1)  Omphalocele 3 (13.6) Cleft lips 4 (18.2)  Micrognathia 7 (31.8) Cardiovascular system n, 9  Dextrocardia 1 (11.1) Patent ductus ateriosus 1 (11.1)  Ventricular septal defect 4 (44.4) Coarctation of the aorta 1(11.1)  Atrial septal defect 1 (11.1) Cyanotic congenital heart disease 1 (11.1) Integumentary system n, 11  Hypopigmentation of the face and neck 2 (18.2) Single palmar creases 3 (27.3)  Widely spaced nipples 1 (9.1) Redundant skin at back of neck 1 (9.1)  Slanting palpeberal fissure 4 (36.4) Total deformities encountered in newborns surveyed N=133 Musculoskeletal system n, 67  Flat nasal bridge 6 (8.9) Right forearm shortening 1 (1.5)  Medial canthus hypertelorism 9 (13.4) Absent left forearm 1 (1.5)  Talipes deformities 8 (11.9) Absence of three medial fingers 2 (3.0)  Low set ears 9 (13.4) Syndactyly 1 (1.5)  Sanders feet 1 (1.5) Macrosomia 2 (3.0)  Toe misalignment 2 (3.0) Short neck 1 (1.5)  Receding chin 1 (1.5) Short webbed neck 5 (7.5)  Fixed flexing deformity of the wrist 2 (3.0) Left hemihypertrophy 1 (1.5)  Prominent philtrum 2 (3.0) Rocker bottom foot 1 (1.5)  High arched feet 1 (1.5) Anterior abdominal wall defect 1 (1.5)  Short lower limb 1 (1.5) Polydactyly 2 (3.0)  Malformed pinna 2 (3.0) Micro-opthalmia 2 (3.0)  Short sturdy fingers 2 (3.0) High arched palate 1 (1.5) Central nervous system n, 14  Microcephaly 3 (21.4) Sutural diastasis 1 (7.1)  Encephalocele 2 (14.3) Prominent occipitus 1 (7.1)  Meningocele 2 (14.3) Hydrocephalocele 1 (7.1)  Spinal bifidia occulta 2 (14.3) Macrocephaly 2 (14.3) Urogenital system n, 10  Hypospadia 1 (10.0) Clitomegaly 2 (20.0)  Micropenis 2 (20.0) Poorly developed scrotum 1 (10.0)  Bilateral undescended testis 1 (10.0) Cryptorchidism 1 (10.0)  Abnormal external genitalia 1 (10.0) Bilateral nephromegaly 1 (10.0) Digestive system n, 22  Tracheooesphageal fistula 2 (9.1) Imperforate anus 2 (9.1)  Duodenal atresia 2 (9.1) Macroglossia 2 (9.1)  Omphalocele 3 (13.6) Cleft lips 4 (18.2)  Micrognathia 7 (31.8) Cardiovascular system n, 9  Dextrocardia 1 (11.1) Patent ductus ateriosus 1 (11.1)  Ventricular septal defect 4 (44.4) Coarctation of the aorta 1(11.1)  Atrial septal defect 1 (11.1) Cyanotic congenital heart disease 1 (11.1) Integumentary system n, 11  Hypopigmentation of the face and neck 2 (18.2) Single palmar creases 3 (27.3)  Widely spaced nipples 1 (9.1) Redundant skin at back of neck 1 (9.1)  Slanting palpeberal fissure 4 (36.4) Total deformities encountered in newborns surveyed N=133 a Deformities encountered in newborns occurred either individually, in association or in syndromic pattern. Table 2 Deformitiesa encountered in newborns with congenital malformation Musculoskeletal system n, 67  Flat nasal bridge 6 (8.9) Right forearm shortening 1 (1.5)  Medial canthus hypertelorism 9 (13.4) Absent left forearm 1 (1.5)  Talipes deformities 8 (11.9) Absence of three medial fingers 2 (3.0)  Low set ears 9 (13.4) Syndactyly 1 (1.5)  Sanders feet 1 (1.5) Macrosomia 2 (3.0)  Toe misalignment 2 (3.0) Short neck 1 (1.5)  Receding chin 1 (1.5) Short webbed neck 5 (7.5)  Fixed flexing deformity of the wrist 2 (3.0) Left hemihypertrophy 1 (1.5)  Prominent philtrum 2 (3.0) Rocker bottom foot 1 (1.5)  High arched feet 1 (1.5) Anterior abdominal wall defect 1 (1.5)  Short lower limb 1 (1.5) Polydactyly 2 (3.0)  Malformed pinna 2 (3.0) Micro-opthalmia 2 (3.0)  Short sturdy fingers 2 (3.0) High arched palate 1 (1.5) Central nervous system n, 14  Microcephaly 3 (21.4) Sutural diastasis 1 (7.1)  Encephalocele 2 (14.3) Prominent occipitus 1 (7.1)  Meningocele 2 (14.3) Hydrocephalocele 1 (7.1)  Spinal bifidia occulta 2 (14.3) Macrocephaly 2 (14.3) Urogenital system n, 10  Hypospadia 1 (10.0) Clitomegaly 2 (20.0)  Micropenis 2 (20.0) Poorly developed scrotum 1 (10.0)  Bilateral undescended testis 1 (10.0) Cryptorchidism 1 (10.0)  Abnormal external genitalia 1 (10.0) Bilateral nephromegaly 1 (10.0) Digestive system n, 22  Tracheooesphageal fistula 2 (9.1) Imperforate anus 2 (9.1)  Duodenal atresia 2 (9.1) Macroglossia 2 (9.1)  Omphalocele 3 (13.6) Cleft lips 4 (18.2)  Micrognathia 7 (31.8) Cardiovascular system n, 9  Dextrocardia 1 (11.1) Patent ductus ateriosus 1 (11.1)  Ventricular septal defect 4 (44.4) Coarctation of the aorta 1(11.1)  Atrial septal defect 1 (11.1) Cyanotic congenital heart disease 1 (11.1) Integumentary system n, 11  Hypopigmentation of the face and neck 2 (18.2) Single palmar creases 3 (27.3)  Widely spaced nipples 1 (9.1) Redundant skin at back of neck 1 (9.1)  Slanting palpeberal fissure 4 (36.4) Total deformities encountered in newborns surveyed N=133 Musculoskeletal system n, 67  Flat nasal bridge 6 (8.9) Right forearm shortening 1 (1.5)  Medial canthus hypertelorism 9 (13.4) Absent left forearm 1 (1.5)  Talipes deformities 8 (11.9) Absence of three medial fingers 2 (3.0)  Low set ears 9 (13.4) Syndactyly 1 (1.5)  Sanders feet 1 (1.5) Macrosomia 2 (3.0)  Toe misalignment 2 (3.0) Short neck 1 (1.5)  Receding chin 1 (1.5) Short webbed neck 5 (7.5)  Fixed flexing deformity of the wrist 2 (3.0) Left hemihypertrophy 1 (1.5)  Prominent philtrum 2 (3.0) Rocker bottom foot 1 (1.5)  High arched feet 1 (1.5) Anterior abdominal wall defect 1 (1.5)  Short lower limb 1 (1.5) Polydactyly 2 (3.0)  Malformed pinna 2 (3.0) Micro-opthalmia 2 (3.0)  Short sturdy fingers 2 (3.0) High arched palate 1 (1.5) Central nervous system n, 14  Microcephaly 3 (21.4) Sutural diastasis 1 (7.1)  Encephalocele 2 (14.3) Prominent occipitus 1 (7.1)  Meningocele 2 (14.3) Hydrocephalocele 1 (7.1)  Spinal bifidia occulta 2 (14.3) Macrocephaly 2 (14.3) Urogenital system n, 10  Hypospadia 1 (10.0) Clitomegaly 2 (20.0)  Micropenis 2 (20.0) Poorly developed scrotum 1 (10.0)  Bilateral undescended testis 1 (10.0) Cryptorchidism 1 (10.0)  Abnormal external genitalia 1 (10.0) Bilateral nephromegaly 1 (10.0) Digestive system n, 22  Tracheooesphageal fistula 2 (9.1) Imperforate anus 2 (9.1)  Duodenal atresia 2 (9.1) Macroglossia 2 (9.1)  Omphalocele 3 (13.6) Cleft lips 4 (18.2)  Micrognathia 7 (31.8) Cardiovascular system n, 9  Dextrocardia 1 (11.1) Patent ductus ateriosus 1 (11.1)  Ventricular septal defect 4 (44.4) Coarctation of the aorta 1(11.1)  Atrial septal defect 1 (11.1) Cyanotic congenital heart disease 1 (11.1) Integumentary system n, 11  Hypopigmentation of the face and neck 2 (18.2) Single palmar creases 3 (27.3)  Widely spaced nipples 1 (9.1) Redundant skin at back of neck 1 (9.1)  Slanting palpeberal fissure 4 (36.4) Total deformities encountered in newborns surveyed N=133 a Deformities encountered in newborns occurred either individually, in association or in syndromic pattern. DISCUSSION The overall incidence rate of congenital malformations in this study was 6.5 per 1000 live births. Though the incidence of congenital malformation in Nigeria has not been properly documented because of a lack of proper record-keeping [4], studies in this regard have revealed a wide variation in rates ranging between 2.2 per 1000 live births and 15.8 per 1000 live births in different zones of the country [5–8]. It is pertinent to note that these figures may be grossly underestimated because low autopsy rates and hospital-based data are limitations that influence accurate assessment of the magnitude of the problem in Nigeria [8]. The high rate of 15.8 per 100 live births was obtained from a study done in Lagos, south-west of Nigeria, and has been attributed to the multiethnic and industrial nature of Lagos suggesting genetic and environmental risk factors for the increased rate in the area [8]. In contrast to this, Rivers State, south-south Nigeria, an area also plagued by environmental pollution, recorded a low rate of 2.2 per 1000 births. There is need for more research into the association between congenital anomalies and environmental pollution in our environment in view of other risk factors such as smoking, alcohol, lack of adequate vaccination and folic acid supplementation, etc. This study demonstrated a strong association between increased parity and congenital malformations similar to findings in other studies [6, 8–10]. This is in contrast to a study that demonstrated more congenital malformations in primiparous mothers [11]. The difference may be explained by the fact that the effect of parity varies between different malformations [12]. For instance, the risk for polydactyly and syndactyly is higher than at low parity, whilst the opposite is true for conditions like hypospadias and omphalocele [12]. Mothers >35 years had more babies with congenital malformations than those <35 years. Although this was not statistically significant, it still reflects the well-known effect of advancing age on the incidence of congenital abnormalities [13]. For this reason, females who are >35 years of age need to be examined more carefully, as the risk of giving birth to a foetus with congenital malformations is greater [14]. It still remains unknown why advanced maternal age is an independent risk factor for foetal death. However, it has been suggested that the increasing age of mothers is associated with an increase in chromosomal meiotic errors [14]. Musculoskeletal abnormalities were observed to occur more frequently than other abnormalities in this study. This is in keeping with the findings of Christianson et al [15]. who noted higher musculoskeletal abnormalities than neural tube defect among black population. However, it is contrast to the findings of other local studies, which reported more digestive and central nervous system abnormalities [4, 6–8]. This difference may be because of the fact that musculoskeletal abnormalities may be perceived as being less severe with an increased likelihood of obtaining medical care, while paucity of diagnostic facilities and cost of surgical care may discourage presentation of the more severe digestive and central nervous system abnormalities. Although the cost of care for congenital abnormalities is expensive and unaffordable in developing countries like Nigeria, most birth defects can be prevented by simple measures such as vaccination, folic acid supplementation, infection prevention and avoidance of certain indulgences like smoking, alcohol and use of certain drugs during pre-pregnancy and gestational periods [1]. Effective implementation of these strategies may reduce mortality from congenital malformations, which are a significant cause of perinatal mortality. CONCLUSION This study reported a 6.5 per 1000 live birth incidence of congenital anomalies. The musculoskeletal system was most commonly affected. In decreasing order, the other systems involved were the digestive (16.5%), the central nervous (10.5%), the urogenital (8.0%), the integumentary (8.3%) and the cardiovascular (6.7%) systems. Among the significant associations identified with congenital malformations were birth weight, maternal socio-economic class, maternal educational attainment, parity, febrile illness and the use of local concoction in an index pregnancy. It was also found that babies with congenital anomalies have a higher tendency to either die in hospital or develop perinatal asphyxia compared with those without such anomalies. LIMITATIONS This study was conducted in resource-limited settings; hence, carrying out chromosomal studies to confirm diagnosis was not feasible. Also, some of the information provided by the respondents may have been affected by recall or intentional lie bias, especially for mothers whose babies had congenital anomalies and may not be psychologically disposed to grant interviews. FUNDING This work was completely sponsored by equal financial contributions from all authors. REFERENCES 1 World Health Organization fact sheet on congenital abnormalities . Available from http://www.who.int/mediacentre/factsheets/fs370/en/ and updated 2016 . 2 Ekanem TB , Okon DE , Akpantah AO , et al. Prevalence of congenital malformations in Cross River and Akwa Ibom states of Nigeria from 1980–2003 . Congenit Anom 2008 ; 48 : 167 – 70 . Google Scholar CrossRef Search ADS 3 Stevenson RE. The genetic basis of human anomalies. In: Stevenson RE , Hall JG , Goodman RM (eds). Human Malformations and Related Anomalies . Vol. 1 . New York, NY : Oxford University Press , 1993 , 115 . 4 Onankpa BO , Adamu A. Pattern and outcome of gross congenital malformations at birth amongst newborns admitted to a tertiary hospital in northern Nigeria . Niger J Paediatr 2014 ; 41 : 337 – 40 . Google Scholar CrossRef Search ADS 5 Ekanem TB , Bassey IE , Mesembe OE , et al. Incidence of congenital malformation in 2 major hospitals in Rivers state of Nigeria from 1990 to 2003/Incidence des malformations congenitales dans deux grands hopitaux de l'Etat de Rivers (Nigeria) de 1990 a 2003 . East Mediterr Health J 2011 ; 17 : 701 . Google Scholar CrossRef Search ADS PubMed 6 Onyearugha CN , Onyire BN. Congenital malformations as seen in a secondary healthcare institution in Southeast Nigeria . J Med Invest Pract 2014 ; 9 : 59 . 7 Mukhtar-Yola M , Ibrahim M , Belonwu R , et al. The prevalence and perinatal outcome of obvious congenital malformations among inborn babies at Aminu Kano Teaching Hospital, Kano . Niger J Paediatr 2005 ; 32 : 47 – 51 . 8 Iroha EO , Egri-Okwaji MT , Odum CU , et al. Perinatal outcome of obvious congenital malformation as seen at the Lagos University Teaching Hospital, Nigeria . Niger J Paediatr 2001 ; 28 : 73 – 7 . 9 Ambe JP , Madziga AG , Akpede GO , et al. Pattern and outcome of congenital malformations in newborn babies in a Nigerian teaching hospital . West Afr J Med 2010 ; 29 : 24 – 9 . Google Scholar PubMed 10 Jehangir W , Ali F , Jahangir T , et al. Prevalence of gross congenital malformations at birth in the neonates in a tertiary care hospital . APMC 2009 ; 3 : 47 – 50 . 11 Shawky RM , Elsedfy HH , Abolouz SK , et al. Prevalence of congenital malformations in a thousand consecutive Egyptian liveborn . Egypt J Med Hum Genet 2001 ; 2 : 43 – 53 . 12 Källén B. Epidemiology of Human Congenital Malformations . Cham, Heidelberg; New York, NY; Dodrecht; London : Springer , 2014 . Google Scholar CrossRef Search ADS 13 Nazer HJ , Cifuentes OL , Aguila RA , et al. The association between maternal age and congenital malformations . Rev Med Chil 2007 ; 135 : 1463 – 9 . Google Scholar PubMed 14 Khan A , Zuhaid M , Fayaz M , et al. Frequency of congenital anomalies in newborns and its relation to maternal health in a Tertiary Care Hospital in Peshawar, Pakistan . Int J Med Stud 2015 ; 3 : 19 – 23 . 15 Christianson AL , Howson CP , Modell B. March of Dimes: Global Report on Birth Defects, the Hidden Toll of Dying and Disabled Children. White Plains, New York: March of Dimes Birth Defects Foundation, 2006. © The Author [2017]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

2018 Journal of Tropical Pediatrics

doi: 10.1093/tropej/fmx068pmid: 29036682

Abstract Objectives To identify the perinatal risk factors for early-onset Group B Streptococcus (EOGBS) sepsis in neonates after inception of a risk-based maternal intrapartum antibiotic prophylaxis strategy in 2004. Design Case control study. Methods All newborn with early onset GBS sepsis (born between 2004 and 2013) were deemed to be “cases” and controls were selected in a 1:4 ratio. Results More than three per vaginal (PV) examinations [odds ratio (OR) 8.57, 95% confidence interval (CI) 3.10–23.6] was a significant risk factors. Peripartum fever (OR 3.54, 95% CI 1.3–9.67), urinary tract infection (OR 2.88, 95% CI 1.08–7.63), meconium-stained amniotic fluid (MSAF) (OR 2.52, 95% CI 1.18–5.37) and caesarean section (OR 1.99, 95% CI 1.16–3.43) were also found to be associated with EOGBS sepsis. Conclusion Multiple vaginal examinations are the strongest risk factors for peripartum Group B Streptococcal (GBS) sepsis. The association of MSAF and caesarean section indicates that foetal distress is an early symptom of perinatal GBS infection. Group B Streptococcus, infant, newborn, factors, risk, India, neonatal sepsis INTRODUCTION Group B Streptococcus is a Gram-positive coccus known to be one of the main causative agents of early-onset sepsis (EOS) in neonates. The reservoir of Group B Streptococcus is usually the gastrointestinal tract of the mother; hence, rectovaginal colonization of mothers results in vertical transmission at the time of delivery [1]. Intrapartum antibiotic prophylaxis (IAP) is given to reduce this transmission [2]. The Centre for Diseases Control (CDC) recommends universal screening of mothers between 35 and 37 weeks of gestation to identify colonized mothers and to provide antibiotics to all those who are colonized [3]. Another approach is to use a risk-based strategy and to provide IAP to those mothers who have certain risk factors. The incidence of Group B Streptococcal (GBS) sepsis in neonates was 0.17/1000 live births in our institution over a 10-year period (1988–97) [4]. At that time, intrapartum antibiotic (ampicillin) was only given to pregnant women with chorioamnionitis or GBS urinary tract infection (UTI). As the incidence of EOS with GBS appeared to be rising over the next 5 years, a case-control study was done in 2003 to identify the risk factors for EOS [5]. Following this study, the policy of IAP was revised in 2004, and it has since been given in the presence of the following risk factors: maternal chorioamnionitis, GBS UTI, spontaneous preterm labour, preterm premature rupture of membranes (PPROM), peripartum fever (≥38°C) and prolonged rupture of membranes (PROM) >18 h. The incidence of EOGBS decreased with this strategy from 0.68/1000 live births (1998–2003) to 0.55/1000 live births (2004–10) [6]. All newborns born to mothers with these risk factors also have a sepsis screen done. A study in our institution in 2012 showed rectovaginal colonization in 7.6% of pregnant women screened at 35–37 weeks gestation [7]. Since then, some obstetricians offer women an option of having a rectovaginal screening at 35–37 weeks with IAP given if screen is positive. More than 30 000 women attend the obstetric outpatient department per year, and therefore, universal screening is not feasible either at clinic or laboratory levels. Hence, few obstetricians offer screening to their patients. Therefore, <1% women undergo screening. As there was a reduction in incidence of EOS with GBS after introduction of IAP, but the incidence remained higher than most high-middle-income countries, this study was undertaken to identify any additional risk factors for EOGBS. This case-control study was done to identify the risk factors for EOGBS sepsis after initiation of risk-based IAP. To our knowledge, no study has reported on risk factors after providing IAP. Identification of persisting and new risk factors may help us modify the current risk-based IAP policy. METHODS Christian Medical College is a tertiary-level perinatal centre, which is a private, patient-paid/charitable hospital catering to three districts in South India. As this is the only referral centre for a large population, >50% of deliveries are high-risk deliveries. All babies admitted in the neonatology unit from 1 January 2004 to 31 December 2014 who were found to have invasive GBS infection within 72 h of birth were identified as the ‘cases’. Invasive disease was described as isolation of the organism from any sterile site (blood/cerebrospinal fluid). The two babies born just before and the two babies born after the ‘cases’ were deemed to be the controls. The records of the mother–baby dyad were retrieved from the Medical Records Department. Data on maternal risk factors and neonatal outcome were collected in a predesigned questionnaire. Data from the pre-IAP era (1998–2003) were also re-analysed using the same methodology and the results were compared between the two epochs. However, data from only three controls per ‘case’ could be obtained from the earlier era. Sample size In a previous study by Benitz et al. [8], the risk factors, which were most significant for the development of EOS in newborns, were intrapartum fever [odds ratio (OR) 4.05] and chorioamnionitis with an OR of 6.43. Therefore, with an estimated 10% of controls having intrapartum fever and 1:4 case:control ratio, we needed a minimum of 54 cases and 216 controls. Descriptive statistics were reported using frequency and percentage for categorical variables. Association between the outcome and categorical variables was analysed using chi-square/Fisher’s exact test. Maternal risk factors associated with neonatal GBS infection were analysed by binary logistic regression using stepwise method with 25% significance for unadjusted analysis and 5% level of significance for adjusted analysis. The values were reported using OR and 95% confidence interval (CI). SPSS 16.0 was used for statistical analysis. The study was conducted with approval from the institutional review board. RESULTS In the period 2004–14, after the initiation of a risk-based intrapartum antibiotic policy, 71 babies were found to have an EOGBS sepsis. On analysing the risk factors associated with developing EOGBS sepsis, more than three per vaginal (PV) examinations after rupture of membranes was found to be a significant factor associated with EOGBS, with an adjusted OR of 8.57 (3.10–23.6) (Table 1). Peripartum fever with OR of 3.54 (95% CI 1.30–9.67) and UTI with OR of 2.88 (95% CI 1.08–7.63) were both significantly associated with EOGBS sepsis. The risk factors for EOGBS sepsis before 2003 are also represented in Table 1 for comparison. Factors like preterm pre-labour rupture of membranes (PPROM) and PROM >18 or 24 h, that were significant risk factors before 2003, were not found to be significant after initiating IAP. In both the eras, there was a trend of babies with EOGBS infection being born by caesarean section (20.1 and 27.3% versus 42.6 and 42.9%). Table 1 Perinatal risk factors for early onset GBS sepsis after and before intrapartum antibiotics Risk factors Pre IAP era (1998–2003) Post IAP era (2004–2015) Cases Controls Unadjusted OR Adjusted OR Cases Controls Unadjusted OR Adjusted OR N = 47 N = 144 OR (95% CI) OR (95% CI) N = 71 N = 284 OR (95% CI) OR (95% CI) n (%) n (%) n (%) n (%) Prematurity<37 wks 9 (19.1%) 11 (7.6%) 2.91 (1.12–7.57) – 10 (14.3%) 42 (14.9%) 0.95 (0.45–2.01) – Low birth weight <2500 gm 8 (17%) 22 (15.3%) 1.16 (0.48–2.8) – 22 (31.9%) 59 (20.8%) 1.78 (0.9–3.19) – Chorioamnionitis 4 (8.5%) 0 5.53 – 0 6 (2.1%) – – Prolonged rupture of membranes >18 hours 14 (29.8%) 14 (9.7%) 3.93 (1.71–9.06) – 2 (2.8%) 17 (5.9%) 2.197 (0.4–9.7) – Prolonged rupture of membranes >24 hours 10 (21.3%) 7 (4.9%) 5.29 (1.88–14.84) 5.09 (1.44–17.93) 2 (2.8%) 8 (2.8%) 1.0 (0.2–4.8) – Prelabour rupture of membranes 11 (23.4%) 28 (19.4%) 1.26 (0.57–2.79) – 10 (14.1%) 54 (19%) 1.4 (0.6–2.9) – Peripartum fever 17 (36.2%) 21 (14.6%) 3.62 (1.69–7.78) 2.23 (0.57–5.73) 9 (14.5%) 12 (4.3%) 3.77 (1.5–9.41) 3.54 (1.30–9.67) More than 3 per vaginal examination after rupture of membranes 16 (34%) 5 (3.5%) 17.7 (5.9–52.9) 11.22 (3.38–37.22) 11 (16.7%) 8 (2.8%) 6.9 (2.65–17.94) 8.57 (3.10–23.6) Meconium stained amniotic fluid 10 (21.3%) 25 (17.4%) 1.36 (0.59–3.10) – 15 (23.8%) 33 (11.8%) 2.33 (1.180–4.63) 2.52 (1.18–5.37) Urinary tract infection 3 (6.4%) 0 – – 9 (13.4%) 16 (5.7%) 2.57 (1.08–6.10) 2.88 (1.08–7.63) Primigravida 26 (55.3%) 68 (47.2%) 1.45 (0.74–2.83) – 43 (61%) 145 (51.4%) 1.50 (0.88–2.56) – Caesarean 20 (42.6%) 29 (20.1%) 3.05 (1.49–6.21) 1.43 (0.55–3.71) 30 (42.9%) 77 (27.3%) 1.99 (1.16–3.43) – Risk factors Pre IAP era (1998–2003) Post IAP era (2004–2015) Cases Controls Unadjusted OR Adjusted OR Cases Controls Unadjusted OR Adjusted OR N = 47 N = 144 OR (95% CI) OR (95% CI) N = 71 N = 284 OR (95% CI) OR (95% CI) n (%) n (%) n (%) n (%) Prematurity<37 wks 9 (19.1%) 11 (7.6%) 2.91 (1.12–7.57) – 10 (14.3%) 42 (14.9%) 0.95 (0.45–2.01) – Low birth weight <2500 gm 8 (17%) 22 (15.3%) 1.16 (0.48–2.8) – 22 (31.9%) 59 (20.8%) 1.78 (0.9–3.19) – Chorioamnionitis 4 (8.5%) 0 5.53 – 0 6 (2.1%) – – Prolonged rupture of membranes >18 hours 14 (29.8%) 14 (9.7%) 3.93 (1.71–9.06) – 2 (2.8%) 17 (5.9%) 2.197 (0.4–9.7) – Prolonged rupture of membranes >24 hours 10 (21.3%) 7 (4.9%) 5.29 (1.88–14.84) 5.09 (1.44–17.93) 2 (2.8%) 8 (2.8%) 1.0 (0.2–4.8) – Prelabour rupture of membranes 11 (23.4%) 28 (19.4%) 1.26 (0.57–2.79) – 10 (14.1%) 54 (19%) 1.4 (0.6–2.9) – Peripartum fever 17 (36.2%) 21 (14.6%) 3.62 (1.69–7.78) 2.23 (0.57–5.73) 9 (14.5%) 12 (4.3%) 3.77 (1.5–9.41) 3.54 (1.30–9.67) More than 3 per vaginal examination after rupture of membranes 16 (34%) 5 (3.5%) 17.7 (5.9–52.9) 11.22 (3.38–37.22) 11 (16.7%) 8 (2.8%) 6.9 (2.65–17.94) 8.57 (3.10–23.6) Meconium stained amniotic fluid 10 (21.3%) 25 (17.4%) 1.36 (0.59–3.10) – 15 (23.8%) 33 (11.8%) 2.33 (1.180–4.63) 2.52 (1.18–5.37) Urinary tract infection 3 (6.4%) 0 – – 9 (13.4%) 16 (5.7%) 2.57 (1.08–6.10) 2.88 (1.08–7.63) Primigravida 26 (55.3%) 68 (47.2%) 1.45 (0.74–2.83) – 43 (61%) 145 (51.4%) 1.50 (0.88–2.56) – Caesarean 20 (42.6%) 29 (20.1%) 3.05 (1.49–6.21) 1.43 (0.55–3.71) 30 (42.9%) 77 (27.3%) 1.99 (1.16–3.43) – Table 1 Perinatal risk factors for early onset GBS sepsis after and before intrapartum antibiotics Risk factors Pre IAP era (1998–2003) Post IAP era (2004–2015) Cases Controls Unadjusted OR Adjusted OR Cases Controls Unadjusted OR Adjusted OR N = 47 N = 144 OR (95% CI) OR (95% CI) N = 71 N = 284 OR (95% CI) OR (95% CI) n (%) n (%) n (%) n (%) Prematurity<37 wks 9 (19.1%) 11 (7.6%) 2.91 (1.12–7.57) – 10 (14.3%) 42 (14.9%) 0.95 (0.45–2.01) – Low birth weight <2500 gm 8 (17%) 22 (15.3%) 1.16 (0.48–2.8) – 22 (31.9%) 59 (20.8%) 1.78 (0.9–3.19) – Chorioamnionitis 4 (8.5%) 0 5.53 – 0 6 (2.1%) – – Prolonged rupture of membranes >18 hours 14 (29.8%) 14 (9.7%) 3.93 (1.71–9.06) – 2 (2.8%) 17 (5.9%) 2.197 (0.4–9.7) – Prolonged rupture of membranes >24 hours 10 (21.3%) 7 (4.9%) 5.29 (1.88–14.84) 5.09 (1.44–17.93) 2 (2.8%) 8 (2.8%) 1.0 (0.2–4.8) – Prelabour rupture of membranes 11 (23.4%) 28 (19.4%) 1.26 (0.57–2.79) – 10 (14.1%) 54 (19%) 1.4 (0.6–2.9) – Peripartum fever 17 (36.2%) 21 (14.6%) 3.62 (1.69–7.78) 2.23 (0.57–5.73) 9 (14.5%) 12 (4.3%) 3.77 (1.5–9.41) 3.54 (1.30–9.67) More than 3 per vaginal examination after rupture of membranes 16 (34%) 5 (3.5%) 17.7 (5.9–52.9) 11.22 (3.38–37.22) 11 (16.7%) 8 (2.8%) 6.9 (2.65–17.94) 8.57 (3.10–23.6) Meconium stained amniotic fluid 10 (21.3%) 25 (17.4%) 1.36 (0.59–3.10) – 15 (23.8%) 33 (11.8%) 2.33 (1.180–4.63) 2.52 (1.18–5.37) Urinary tract infection 3 (6.4%) 0 – – 9 (13.4%) 16 (5.7%) 2.57 (1.08–6.10) 2.88 (1.08–7.63) Primigravida 26 (55.3%) 68 (47.2%) 1.45 (0.74–2.83) – 43 (61%) 145 (51.4%) 1.50 (0.88–2.56) – Caesarean 20 (42.6%) 29 (20.1%) 3.05 (1.49–6.21) 1.43 (0.55–3.71) 30 (42.9%) 77 (27.3%) 1.99 (1.16–3.43) – Risk factors Pre IAP era (1998–2003) Post IAP era (2004–2015) Cases Controls Unadjusted OR Adjusted OR Cases Controls Unadjusted OR Adjusted OR N = 47 N = 144 OR (95% CI) OR (95% CI) N = 71 N = 284 OR (95% CI) OR (95% CI) n (%) n (%) n (%) n (%) Prematurity<37 wks 9 (19.1%) 11 (7.6%) 2.91 (1.12–7.57) – 10 (14.3%) 42 (14.9%) 0.95 (0.45–2.01) – Low birth weight <2500 gm 8 (17%) 22 (15.3%) 1.16 (0.48–2.8) – 22 (31.9%) 59 (20.8%) 1.78 (0.9–3.19) – Chorioamnionitis 4 (8.5%) 0 5.53 – 0 6 (2.1%) – – Prolonged rupture of membranes >18 hours 14 (29.8%) 14 (9.7%) 3.93 (1.71–9.06) – 2 (2.8%) 17 (5.9%) 2.197 (0.4–9.7) – Prolonged rupture of membranes >24 hours 10 (21.3%) 7 (4.9%) 5.29 (1.88–14.84) 5.09 (1.44–17.93) 2 (2.8%) 8 (2.8%) 1.0 (0.2–4.8) – Prelabour rupture of membranes 11 (23.4%) 28 (19.4%) 1.26 (0.57–2.79) – 10 (14.1%) 54 (19%) 1.4 (0.6–2.9) – Peripartum fever 17 (36.2%) 21 (14.6%) 3.62 (1.69–7.78) 2.23 (0.57–5.73) 9 (14.5%) 12 (4.3%) 3.77 (1.5–9.41) 3.54 (1.30–9.67) More than 3 per vaginal examination after rupture of membranes 16 (34%) 5 (3.5%) 17.7 (5.9–52.9) 11.22 (3.38–37.22) 11 (16.7%) 8 (2.8%) 6.9 (2.65–17.94) 8.57 (3.10–23.6) Meconium stained amniotic fluid 10 (21.3%) 25 (17.4%) 1.36 (0.59–3.10) – 15 (23.8%) 33 (11.8%) 2.33 (1.180–4.63) 2.52 (1.18–5.37) Urinary tract infection 3 (6.4%) 0 – – 9 (13.4%) 16 (5.7%) 2.57 (1.08–6.10) 2.88 (1.08–7.63) Primigravida 26 (55.3%) 68 (47.2%) 1.45 (0.74–2.83) – 43 (61%) 145 (51.4%) 1.50 (0.88–2.56) – Caesarean 20 (42.6%) 29 (20.1%) 3.05 (1.49–6.21) 1.43 (0.55–3.71) 30 (42.9%) 77 (27.3%) 1.99 (1.16–3.43) – An increased number of cases were born through meconium-stained amniotic fluid (MSAF) [OR 2.52 (95% 1.18–5.37)]. As compared with the controls, a significant number of babies with sepsis had abnormal cardiotocogram (CTG) before delivery [33.8%, OR 2.17 (1.22–3.86)]. Abnormal CTG was defined as Category II or Category III CTG. There was also an increased need for resuscitation at birth—23 versus 3.5%, OR 2.58 (1.12–5.9) (Table 2). Table 2 Association of CTG abnormalities and resuscitation with GBS sepsis Variable Cases (N, %) Controls (N, %) OR (95% CI) Resuscitation 17 (23) 10 (3.5) 2.58 (1.12–5.90) Non-reassuring foetal status on CTG 24 (33.8) 54 (19) 2.17 (1.22–3.86) Variable Cases (N, %) Controls (N, %) OR (95% CI) Resuscitation 17 (23) 10 (3.5) 2.58 (1.12–5.90) Non-reassuring foetal status on CTG 24 (33.8) 54 (19) 2.17 (1.22–3.86) Table 2 Association of CTG abnormalities and resuscitation with GBS sepsis Variable Cases (N, %) Controls (N, %) OR (95% CI) Resuscitation 17 (23) 10 (3.5) 2.58 (1.12–5.90) Non-reassuring foetal status on CTG 24 (33.8) 54 (19) 2.17 (1.22–3.86) Variable Cases (N, %) Controls (N, %) OR (95% CI) Resuscitation 17 (23) 10 (3.5) 2.58 (1.12–5.90) Non-reassuring foetal status on CTG 24 (33.8) 54 (19) 2.17 (1.22–3.86) DISCUSSION Prevention of vertical transmission of Group B Streptococci from a colonized mother to her newborn has been one of the successful strategies adopted to reduce early-onset infection in newborn. This is ideally done by screening pregnant women in late pregnancy and giving intrapartum antibiotics to those found colonized. Alternatively, antibiotics can be given in labour to pregnant women who have certain risk factors, which predispose their newborn to EOGBS infection. Universal screening has been found to be cost-effective only when the incidence of EOGBS is >1.2/1000 live births [9]. This strategy might also be cost prohibitive and technically not feasible in most mid-low-income countries. Hence, many institutions use a risk-based intrapartum antibiotic policy. After a case-control study in 2003, we adopted a risk-based IAP policy since 2004. This study looks at the risk factors for EOGBS infections when an IAP policy is in place. This was done to identify any new risk factors and to assess the effect of IAP on existing risk factors. Prematurity (<37 weeks) was a risk factor in the pre-IAP epoch. This was similar to previous similar studies [8, 10, 11]. In the post-IAP epoch, it was not significantly associated with GBS infection. This is attributed to the successful IAP coverage of mothers who present with preterm labour or PPROM since 2004. Similarly, PROM >18 h has been identified to be significantly associated with GBS sepsis in many studies [8, 10]. A similar association was found in our pre-IAP epoch, which was no longer seen after initiating IAP. This again is probably because of IAP given in this setting. Peripartum fever continued to be a risk factor, despite a policy of IAP for women in labour having fever >38°C. A previous study by Al-Kadri et al. [12] also found an association between peripartum fever and EOGBS (OR 7.10, 2.50–20.71). Though we are not able to account for this continuing association, a possible explanation could be that the obstetricians’ compliance with giving IAP in fever is lower (compared with other risk factors), especially if they felt that fever could be attributed to another aetiology. We do not have data on overall compliance with the overall IAP policy and in individual risk factors throughout the period under study. Observationally, the compliance seems to be between 50 and 60%. It is logical to assume that improvement in compliance to IAP closer to 100% in mothers with risk factors would further reduce neonatal EOGBS incidence in the institution. More than three vaginal examinations after ROM showed the strongest association with EOGBS disease in both the pre- and post-IAP epochs. This indirectly points to a high colonization rate in our population. However, a study in 2012 showed maternal colonization rate of only 7.6%, which is much lower than most high-middle-income countries. We are hence unable to account for this discrepancy. Previous studies have not reported the magnitude of association of multiple PVs and EOGBS sepsis in neonates, though this is known to be associated with chorioamnionitis [13]. Hence, we would strongly encourage IAP to unscreened mothers who had multiple vaginal examinations in labour. Alternatively, this could be considered a major risk factor for EOGBS sepsis necessitating sepsis workup in the newborns. This could go a long way in reducing EOGBS sepsis in developing countries. Our data show that babies with GBS sepsis were more likely to be born through MSAF, have an abnormal CTG, be delivered by caesarean section and require resuscitation at birth. Studies have shown that GBS-colonized women had more foetal distress when compared with GBS-negative women [13]. It has been hypothesized that EOGBS sepsis is a spectrum of infection that also involves the foetus, which seems to be borne out by our data. Over the past few years, babies who have unexplained foetal distress and are asphyxiated/have respiratory depression at birth have had sepsis workup in our hospital and been covered with penicillin and gentamicin. The study proves that IAP definitely helps in the settings of preterm labour/PPROM as well as PROM in reducing EOGBS sepsis. It highlights the strong association of multiple vaginal examinations with GBS invasive disease. Where universal screening of mothers is not practised, this should be a guide to obstetricians to reduce unwanted vaginal examinations and also be the single most important risk factor to do a sepsis screen in symptomatic newborns. Our study thus can serve as a template for risk-based intrapartum treatment of mothers and newborn evaluation for EOS, especially in middle-low-income countries. Further studies are needed to evaluate the effect of infection in causing foetal distress/asphyxia in the developing world. References 1 Dillon HC , Gray E , Pass MA , et al. Anorectal and vaginal carriage of group B streptococci during pregnancy . J Infect Dis 1982 ; 145 : 794 – 9 . Google Scholar CrossRef Search ADS PubMed 2 Ohlsson A , Shah VS. Intrapartum antibiotics for known maternal Group B Streptococcal colonization . Cochrane Database Syst Rev 2014 ; 10 : CD007467 . 3 Verani JR , McGee L , Schrag SJ. Centers for Disease Control and Prevention: Prevention of perinatal group B streptococcal disease–revised guidelines from CDC, 2010. Morb Mortal Wkly Rep 2010; 59 : 1 – 36 . 4 Kuruvilla KA , Thomas N , Jesudasan MV , et al. Neonatal Group B Streptococcal bacteraemia in India: ten years’ experience . Acta Paediatr 1999 ; 88 : 1031 – 2 . Google Scholar CrossRef Search ADS PubMed 5 Sridhar S , Jolly C , Niranjan T , et al. Neonatal Group B Streptococcal infection in India—myth or reality? Poster presented at The XXIII annual convention of the National Neonatology Forum, Hyderabad. December 19–21, 2003 . 6 Sridhar S , Grace R , Nithya PJ , et al. Group B Streptococcal infection in a tertiary hospital in India—1998–2010 . Pediatr Infect Dis J 2014 ; 33 : 1091 – 2 . Google Scholar CrossRef Search ADS PubMed 7 Santhanam S , Jose R , Sahni RD , et al. Prevalence of Group B Streptococcal colonization among pregnant women and neonates in a tertiary hospital in India . J Turk Ger Gynecol Assoc 2017 ; (In press). doi: 10.4274/jtgga.2017.0032. 8 Benitz WE , Gould JB , Druzin ML. Risk factors for early-onset group streptococcal sepsis: estimation of odds ratios by critical literature review . Pediatrics 1999 ; 103 : e77 . Google Scholar CrossRef Search ADS PubMed 9 Mohle-Boetani JC , Schuchat A , Plikaytis BD , et al. Comparison of prevention strategies for neonatal Group B Streptococcal infection: a population-based economic analysis . JAMA 1993 ; 270 : 1442 – 8 . Google Scholar CrossRef Search ADS PubMed 10 Oddie S , Embleton ND. Risk factors for early onset neonatal Group B Streptococcal sepsis: case-control study . BMJ 2002 ; 325 : 308 . Google Scholar CrossRef Search ADS PubMed 11 Adair CE , Kowalsky L , Quon H , et al. Risk factors for early-onset Group B Streptococcal disease in neonates: a population-based case-control study . CMAJ Can Med Assoc J J Assoc Medicale Can 2003 ; 169 : 198 – 203 . 12 Al-Kadri HM , Bamuhair SS , Johani SMA , et al. Maternal and neonatal risk factors for early-onset Group B Streptococcal disease: a case control study . Int J Womens Health 2013 ; 5 : 729 – 35 . Google Scholar CrossRef Search ADS PubMed 13 Shi C , Qu S , Yang L , et al. Detection of maternal colonization of group B streptococcus in late pregnancy by real-time polymerase chain reaction and its effect on perinatal outcome . Zhonghua Fu Chan Ke Za Zhi 2010 ; 45 : 12 – 6 . Google Scholar PubMed © The Author [2017]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

2018 Journal of Tropical Pediatrics

doi: 10.1093/tropej/fmx071pmid: 29036724

Abstract Objective The objective of this article is to study the etiology of community-acquired pneumonia in children with airway malacia. Methods We retrospectively reviewed the medical records of 428 pneumonia patients. All patients underwent bronchoscopy, and bronchoalveolar lavage samples were processed for microbiological assessment. Results In a total of 428 cases reviewed, 60 were found to have airway malacia. Pathogens were identified in 44 of the 60 specimens (73.3%), with 32 being single-pathogen infections. The most common pathogen was respiratory syncytial virus (RSV; 20%). Mixed-pathogen infections were observed in 12 patients. Airway malacia patients were younger than those without malacia (10.5 vs. 50 months, respectively; p < 0.001). Compared with those without airway malacia, wheezing, cyanosis and admission to the pediatric intensive care unit were more common in children with airway malacia and their hospital stay was longer. Conclusion RSV was the most common pathogen in those with airway malacia. Airway malacia was found to aggravate infectious pneumonia. children, community-acquired pneumonia, airway malacia, pathogen INTRODUCTION Protracted or recurrent pneumonia poses a significant challenge to the pediatricians. Children with airway malacia often have protracted courses of airway infections because dynamic airway collapse during coughing results in impaired mucociliary clearance and retention of tracheobronchial secretions [1–3]. Zhang et al. found that underlying diseases, such as airway abnormalities, are associated with severe acute lower respiratory infections [4]. Furthermore, Gokdemir et al. [5] found that malacia disorders were the most common causes (7.0%) of persistent and recurrent pneumonia. Previous studies have focused on the association among recurrent wheeze, chronic cough and airway malacia, with few reports focusing on the etiologic organisms responsible for recurrent or persistent pneumonia with airway malacia. In a study by Boogaard et al. [5], recurrent lower respiratory tract infections were found in 60 of 96 (63%) patients with airway malacia, and in 77.9% of the children with malacia, at least one pathogen was cultured. A more recent investigation by De Baets et al. [6] reported that 56% of the children reviewed with persistent respiratory symptoms had a positive bronchoalveolar lavage culture. These findings suggest protracted bacterial infection as a possible cause of persistent respiratory symptoms. Little is known about the pathogens and clinical features in protracted and recurrent pneumonia patients with airway malacia. The aim of this study was to determine the pathogens and clinical manifestations of protracted and recurrent pneumonia in children with airway malacia, and to compare these with pneumonia in children with no airway malacia to provide a basis for a reasonable choice of antibiotic. MATERIALS AND METHODS Patients This study was retrospectively conducted on patients with protracted or recurrent pneumonia who were admitted to the Department of Respiratory Disease at the Children’s Hospital Soochow University, China, from January 2014 to December 2015. Protracted pneumonia is defined as a lower respiratory tract infection persisting for ≥30 days; recurrent pneumonia is defined as at least two pneumonia episodes within 1 year [5]. Patients with congenital heart disease, immune deficiency, neuromuscular disease or foreign body aspiration were excluded from the study. Bronchoscopy BAL fluid (BALF) samples were taken from all patients and assessed for respiratory pathogens. ‘Laryngomalacia’ is the inward collapse of the supraglottic structures of the glottis on inspiration, causing airway obstruction. ‘Tracheomalacia’ is a tracheal deformity at the end of expiration, maintained during spontaneous respiration but that can be altered by the passage of the bronchoscope or positive airway pressure. ‘Bronchomalacia’ is the appearance of a deformity in the right or left mainstem bronchi and/or their respective divisions at the lobar or segmental level. Tracheomalacia and/or bronchomalacia can be classified as mild (less than one-third invagination), moderate (one-third to one-half invagination) or severe (more than four-fifth invagination) [7]. For BALF collection, the bronchoscope was inserted into the bronchus and three samples in 1.0 ml/kg normal saline solution were taken. Indications for bronchoscopy include unexplained hemoptysis or chronic excitant cough, pulmonary atelectasis, local stridor, tracheal and bronchial pulmonary hypoplasia and deformity, pulmonary diffuse disease and protracted or recurrent pulmonary infections. Contraindications for bronchoscopy include severe arrhythmia, cardiac failure, severe respiratory failure, severe bleeding tendency and coagulation dysfunction [8]. Detection of pathogens Direct immunofluorescence was used to detect syncytial viral infection (respiratory syncytial virus, RSV), influenza virus A, influenza virus B, parainfluenza virus (PIV) I, PIV II, PIV III and adenovirus (ADV). All assay kits were purchased from Chemicon International, Inc. (Billerica, MA, USA), and all staining procedures were done according to the manufacturer’s instructions. Immunostained preparations were viewed using a Leica 020-518.500 fluorescence microscope (Leica Microsystems, Wetzlar, Germany). Detection of the human metapneumovirus gene The primer sequences for human metapneumovirus (HMPV) were 5ʹ-AACCGTGTACTAAGTGATGCACTC-3ʹ; antisense, 5ʹ-CATTGTTTGACCGGCCCCATAA-3ʹ. HMPV was assayed by fluorescent real-time polymerase chain reaction (RT-PCR) using the iCycler RT-PCR system (Bio-Rad, Hercules, CA, USA). The cyclic temperature settings were 94.0 °C, 30 s; 56.0 °C, 30 s; 72.0 °C, 30 s and were amplified by 40 cycles. Detection of the human bocavirus gene The primer sequences for the human bocavirus (HBoV) gene were HBoV-F: 5ʹ-TGACATTCAACTACCAACAACCTG-3ʹ; HBoV-R: 5ʹ-CAGATCCTTTTCCTCCTCCAATAC-3ʹ; and HBoV-probe: AGCACCACAAAACACCTCAGGGG-TAMRA. HBoV-DNA was detected by RT-PCR. The cyclic temperature settings were 94.0 °C, 30 s; 56.0 °C, 30 s; 72.0 °C, 30 s and were amplified by 40 cycles. Detection of the human rhinovirus gene The primer sequences for human rhinovirus (HRV) were HRV-F: TGG ACA GGG TGT GAA GAG C; HRV-R: CAA AGT AGT CGG TCC CAT CC; and HRV-probe: FAM-TCC TCC GGC CCC TGA ATG-TAMRA. HRV-DNA was detected by RT-PCR. The cyclic temperature settings were 95.0 °C, 5.0 min; 95.0 °C, 15 s; 60.0 °C, 30 s and were amplified by 40 cycles. Detection of bacteria We prepared cultures for quantitative analysis to assess the presence of common aerobic and anaerobic bacteria. Selected media (e.g. Columbia AGAR blood plate, chocolate tablet) were inoculated and placed in a carbon dioxide incubator (50 ml/l) (YAMATO, Ltd, Tokyo, Japan) at 35.0 °C for 18∼24 h. Bacteria were identified based on the characteristics of their colonies, Gram staining, microscopic performance and biochemical reaction. C-reactive protein detection Blood samples were taken from each patient and were measured using the XE–2000i Automated Hematology System (SYSMEX, Kobe, Japan); C-reactive protein (CRP) was detected by immune scattering turbidimetry using the HITACHI 7600-010 automatic biochemical analyzer (Hitachi, Ltd., Tokyo, Japan). Serology testing for Mycoplasma pneumoniae and Chlamydophila pneumonia The presence of specific IgM and IgG antibodies against Mycoplasma pneumoniae (MP) was investigated in serum samples of patients using a commercial ELISA kit (Serion ELISA classic M. pneumoniae IgG/IgM, Institute Virion/Serion, Germany). IgA and IgG antibodies against Chlamydophila pneumoniae were detected with Serion ELISA classic C. pneumoniae IgA/IgG kits. Statistical analyses All data were analyzed using PASW 20.0. Comparisons among groups were performed using the Chi-squared test. Fisher’s exact probability test was used to analyze data that did not meet the requirements for the Chi-squared test. Data that were not normally distributed were compared using the Mann–Whitney U test. Here, p < 0.05 was considered statistically significant. RESULTS Patient characteristics The records of 428 patients were examined for the study. In all, 60 suffered from airway malacia (Table 1); the remaining 368 patients were used as the control group. There were 43 (71.6%) males and 17 (28.3%) females. Seventeen patients (28.3%) were <6.0 months old, 16 (26.7%) were from 6 months to 1.0 year old, 18 (30%) were from 1.0 to 3.0 years old, 5 (8.3%) were from 3.0 to 5.0 years old and 4 (6.7%) were >5.0 years old. The control group (n = 368) was composed of 199 (54.1%) males and 169 (45.9%) females. Patients whose parents refused bronchoscopy and those patients with bronchoscopy contraindications were excluded from the study. The number of patients whose parents refused bronchoscopy was 45. Two patients were with bronchoscopy contraindications. The diagnoses in this group were recurrent/protracted pneumonia (n = 22), bronchial asthma (n = 9), Bordetella pertussis-like symptoms (n = 9), gastroesophageal reflux (n = 2), atelectasis (n = 2), pleural effusion (n = 1), respiratory failure (n = 1) and heart failure (n = 1). Table 1 Airway malacia in patients with community-acquired pneumonia Airway malacia Cases (n) Laryngomalacia 7 Mild tracheomalacia 11 Mild bronchomalacia 23 Moderate bronchomalacia 14 Severe bronchomalacia 1 Laryngotracheobronchomalacia 2 Laryngomalacia and bronchomalacia 1 Laryngotracheomalacia 1 Airway malacia Cases (n) Laryngomalacia 7 Mild tracheomalacia 11 Mild bronchomalacia 23 Moderate bronchomalacia 14 Severe bronchomalacia 1 Laryngotracheobronchomalacia 2 Laryngomalacia and bronchomalacia 1 Laryngotracheomalacia 1 Table 1 Airway malacia in patients with community-acquired pneumonia Airway malacia Cases (n) Laryngomalacia 7 Mild tracheomalacia 11 Mild bronchomalacia 23 Moderate bronchomalacia 14 Severe bronchomalacia 1 Laryngotracheobronchomalacia 2 Laryngomalacia and bronchomalacia 1 Laryngotracheomalacia 1 Airway malacia Cases (n) Laryngomalacia 7 Mild tracheomalacia 11 Mild bronchomalacia 23 Moderate bronchomalacia 14 Severe bronchomalacia 1 Laryngotracheobronchomalacia 2 Laryngomalacia and bronchomalacia 1 Laryngotracheomalacia 1 Pathogens Pathogens were identified in 44 of 60 (73.3%) specimens from patients with airway malacia, of which 32 were single-pathogen infections. RSV was the most common single pathogen affecting 12 cases (37.5%). The most common mixed-pathogen infection was MP + RSV (25.0%) in 12 patients (Table 2). Table 2 Pathogens identified from patients with airway malacia Pathogen Cases (n) Single-pathogen  RSV 12  MP 7  SP 5  PIV III 2  HRV 3  Enterobacter aerogenes 2  Haemophilus influenzae 1 Mixed-pathogen  MP + RSV 12  PIV III + MP 2  SP + HRV 1  HRV+ KP 1  Gram-positive coccus + MP 1  MP + HRV 1  HI + MP 1  SP + MP + HRV 1  Gram-positive coccus + MP + HBoV 1 Pathogen Cases (n) Single-pathogen  RSV 12  MP 7  SP 5  PIV III 2  HRV 3  Enterobacter aerogenes 2  Haemophilus influenzae 1 Mixed-pathogen  MP + RSV 12  PIV III + MP 2  SP + HRV 1  HRV+ KP 1  Gram-positive coccus + MP 1  MP + HRV 1  HI + MP 1  SP + MP + HRV 1  Gram-positive coccus + MP + HBoV 1 Table 2 Pathogens identified from patients with airway malacia Pathogen Cases (n) Single-pathogen  RSV 12  MP 7  SP 5  PIV III 2  HRV 3  Enterobacter aerogenes 2  Haemophilus influenzae 1 Mixed-pathogen  MP + RSV 12  PIV III + MP 2  SP + HRV 1  HRV+ KP 1  Gram-positive coccus + MP 1  MP + HRV 1  HI + MP 1  SP + MP + HRV 1  Gram-positive coccus + MP + HBoV 1 Pathogen Cases (n) Single-pathogen  RSV 12  MP 7  SP 5  PIV III 2  HRV 3  Enterobacter aerogenes 2  Haemophilus influenzae 1 Mixed-pathogen  MP + RSV 12  PIV III + MP 2  SP + HRV 1  HRV+ KP 1  Gram-positive coccus + MP 1  MP + HRV 1  HI + MP 1  SP + MP + HRV 1  Gram-positive coccus + MP + HBoV 1 Pathogens were identified in 249 of 368 specimens with no airway malacia (67.7%) of which 188 were single-pathogen infections. MP was the most common single pathogen affecting 153 cases (81.4%). The most common mixed-pathogen infections were MP + SP (14.8%) in nine patients and MP + Gram-positive coccus (14.8%) in nine patients (Table 3). Table 3 Pathogens identified from patients with no airway malacia Pathogen Cases (n) Single-pathogen  MP 153  SP 10  RSV 5  ADV 3  HBoV 2  PIV III 1  InFA 2  CP 1  PA 3  Enterobacter aerogenes 2  Haemophilus influenzae 3  Gram-positive coccus 1  Moraxella catarrhalis 1 Mixed-pathogen  MP + RSV 6  MP + InFA 3  MP + InFB 2  HRV + SP 1  HRV + KP 1  MP + PA 1  RSV + HBoV 3  Enterococcus + MP 2  MP + Escherichia coli 3  MP + HBoV 3  MP + SP 9  Gram-positive coccus + MP 9  MP + HI 2  RSV + HI 2  MP + CP 3  Neisseria bacteria + MP 1  MP + SA 2  MP + Stenotrophomonas maltophilia 1  SP + HI 1  MP + RSV + SP 2  MP + RSV + HI 1  MP + PinfIII + Enterobacter aerogenes 1 Pathogen Cases (n) Single-pathogen  MP 153  SP 10  RSV 5  ADV 3  HBoV 2  PIV III 1  InFA 2  CP 1  PA 3  Enterobacter aerogenes 2  Haemophilus influenzae 3  Gram-positive coccus 1  Moraxella catarrhalis 1 Mixed-pathogen  MP + RSV 6  MP + InFA 3  MP + InFB 2  HRV + SP 1  HRV + KP 1  MP + PA 1  RSV + HBoV 3  Enterococcus + MP 2  MP + Escherichia coli 3  MP + HBoV 3  MP + SP 9  Gram-positive coccus + MP 9  MP + HI 2  RSV + HI 2  MP + CP 3  Neisseria bacteria + MP 1  MP + SA 2  MP + Stenotrophomonas maltophilia 1  SP + HI 1  MP + RSV + SP 2  MP + RSV + HI 1  MP + PinfIII + Enterobacter aerogenes 1 Table 3 Pathogens identified from patients with no airway malacia Pathogen Cases (n) Single-pathogen  MP 153  SP 10  RSV 5  ADV 3  HBoV 2  PIV III 1  InFA 2  CP 1  PA 3  Enterobacter aerogenes 2  Haemophilus influenzae 3  Gram-positive coccus 1  Moraxella catarrhalis 1 Mixed-pathogen  MP + RSV 6  MP + InFA 3  MP + InFB 2  HRV + SP 1  HRV + KP 1  MP + PA 1  RSV + HBoV 3  Enterococcus + MP 2  MP + Escherichia coli 3  MP + HBoV 3  MP + SP 9  Gram-positive coccus + MP 9  MP + HI 2  RSV + HI 2  MP + CP 3  Neisseria bacteria + MP 1  MP + SA 2  MP + Stenotrophomonas maltophilia 1  SP + HI 1  MP + RSV + SP 2  MP + RSV + HI 1  MP + PinfIII + Enterobacter aerogenes 1 Pathogen Cases (n) Single-pathogen  MP 153  SP 10  RSV 5  ADV 3  HBoV 2  PIV III 1  InFA 2  CP 1  PA 3  Enterobacter aerogenes 2  Haemophilus influenzae 3  Gram-positive coccus 1  Moraxella catarrhalis 1 Mixed-pathogen  MP + RSV 6  MP + InFA 3  MP + InFB 2  HRV + SP 1  HRV + KP 1  MP + PA 1  RSV + HBoV 3  Enterococcus + MP 2  MP + Escherichia coli 3  MP + HBoV 3  MP + SP 9  Gram-positive coccus + MP 9  MP + HI 2  RSV + HI 2  MP + CP 3  Neisseria bacteria + MP 1  MP + SA 2  MP + Stenotrophomonas maltophilia 1  SP + HI 1  MP + RSV + SP 2  MP + RSV + HI 1  MP + PinfIII + Enterobacter aerogenes 1 Comparison of clinical features of CAP patients with and without airway malacia Airway malacia patients were younger than those without malacia (10.5 vs. 50 months, p < .001). The proportion of males with airway malacia was higher than that of females (73.7% vs. 26.7%, p = 0.005). Wheezing, cyanosis and admission to pediatric intensive care unit (PICU) were more common in children with airway malacia (68.3% vs. 30.9%, p < 0.001; 8.3% vs. 0.8%, p = 0.002; 11.7% vs. 3.5%, p = 0.013). The average hospital stay was longer in children with airway malacia (10 days vs. 8.0 days, p = 0.004). MP (81.4%) was the most common single pathogen in CAP patients without airway malacia while virus (56.3%) was the most common pathogen with airway malacia (Table 4). Table 4 Demographic and clinical features of community-acquired pneumonia with or without airway malacia Characteristics No airway deformity (n = 368), n (%) Airway deformity (n = 60) n (%) X2/Z test p Mean age (months) 50.0 (21.0–85.0) 10.5 (4.75–24.25) Z = 7.372 <0.001 Sex  Male 199 (54.1) 44 (73.3)  Female 169 (45.9) 16 (26.7) X2 = 7.796 0.005 Symptom  Cough 357 (97) 59 (98.3) 1.0  Wheezing 114 (30.9) 41 (68.3) X2 = 31.164 <0.001  Fever 252 (68.5) 24 (40) X2 = 18.269 <0.001  Dyspnea 26 (7.1) 8 (13.3) X2 = 2.27 0.132  Cyanosis 3 (0.8) 5 (8.3) 0.002 Physical examination  Lung wheezing rales 114 (30.9) 41 (68.3) X2 = 31.164 <0.001  Lung moisture rales 11 (39.2) 12 (9.3) X2 = 29.359 <0.001 Lab tests WBC (×109) 9.0445 (6.59–13.02) 10.83 (8.0795–15.65) Z = 2.769 0.006 CRP (mg/l) 1.04 (0.2–9.02) 1.03 (0.135–2.915) Z = 1.194 0.233 PICU admission 13 (3.5) 7 (11.7) 0.013 Hospital stay (days) 8.0 (6.0–11.0) 10.0 (7.0–16.0) Z = 2.912 0.004 Cellularity of BALF  Neutrophil (%) 40.0 (15.0–74.0) 39.0 (19.5–67.5) Z = 0.433 0.665  Lymphocyte (%) 3.0 (2.0–7.0) 4.0 (2.0–7.0) Z = 2.157 0.031  Macrophage (%) 51.0 (19.0–81.0) 55.0 (21.5–73.0) Z = 0.23 0.818  Eosinophil (%) 0.0 (0.0–1.0) 0 (0–1.0) Z = 0.151 0.88 Positive pathogen of BALF 249 (67.7) 44 (73.3) X2 = 0.768 0.381  One pathogen 188 (75.5) 32 (72.7) X2 = 0.104 0.747   Mycoplasma pneumoniae 153 (81.4) 6 (18.8) X2 = 10.138 <0.001   Bacteria 21 (11.2) 8 (25) 0.057   Virus 13 (6.9) 18 (56.3) X2 = 25.436 <0.001   Chlamydia pneumoniae 1 (0.5) 0 (0) 1.0  Two pathogens 57 (22.9) 10 (22.7) X2 = 0.054 0.816   MP + CP 3 (5.3) 0 (0) 1.0   MP + virus 15 (26.3) 6 (60) 0.097   MP + bacteria 32 (54.4) 2 (20) 0.292   Virus + bacteria 2 (3.5) 2 (20) 0.096   Two viruses 4 (7) 0 (0) 1.0   Two bacteria 1 (3.5) 0 (0) 1.0  Three pathogens 4 (3.6) 2 (4.5) 0.2   MP + virus + bacteria 3 (75) 2 (100) 0.146   MP + virus + virus 1 (25) 0 (0) 1.0 Characteristics No airway deformity (n = 368), n (%) Airway deformity (n = 60) n (%) X2/Z test p Mean age (months) 50.0 (21.0–85.0) 10.5 (4.75–24.25) Z = 7.372 <0.001 Sex  Male 199 (54.1) 44 (73.3)  Female 169 (45.9) 16 (26.7) X2 = 7.796 0.005 Symptom  Cough 357 (97) 59 (98.3) 1.0  Wheezing 114 (30.9) 41 (68.3) X2 = 31.164 <0.001  Fever 252 (68.5) 24 (40) X2 = 18.269 <0.001  Dyspnea 26 (7.1) 8 (13.3) X2 = 2.27 0.132  Cyanosis 3 (0.8) 5 (8.3) 0.002 Physical examination  Lung wheezing rales 114 (30.9) 41 (68.3) X2 = 31.164 <0.001  Lung moisture rales 11 (39.2) 12 (9.3) X2 = 29.359 <0.001 Lab tests WBC (×109) 9.0445 (6.59–13.02) 10.83 (8.0795–15.65) Z = 2.769 0.006 CRP (mg/l) 1.04 (0.2–9.02) 1.03 (0.135–2.915) Z = 1.194 0.233 PICU admission 13 (3.5) 7 (11.7) 0.013 Hospital stay (days) 8.0 (6.0–11.0) 10.0 (7.0–16.0) Z = 2.912 0.004 Cellularity of BALF  Neutrophil (%) 40.0 (15.0–74.0) 39.0 (19.5–67.5) Z = 0.433 0.665  Lymphocyte (%) 3.0 (2.0–7.0) 4.0 (2.0–7.0) Z = 2.157 0.031  Macrophage (%) 51.0 (19.0–81.0) 55.0 (21.5–73.0) Z = 0.23 0.818  Eosinophil (%) 0.0 (0.0–1.0) 0 (0–1.0) Z = 0.151 0.88 Positive pathogen of BALF 249 (67.7) 44 (73.3) X2 = 0.768 0.381  One pathogen 188 (75.5) 32 (72.7) X2 = 0.104 0.747   Mycoplasma pneumoniae 153 (81.4) 6 (18.8) X2 = 10.138 <0.001   Bacteria 21 (11.2) 8 (25) 0.057   Virus 13 (6.9) 18 (56.3) X2 = 25.436 <0.001   Chlamydia pneumoniae 1 (0.5) 0 (0) 1.0  Two pathogens 57 (22.9) 10 (22.7) X2 = 0.054 0.816   MP + CP 3 (5.3) 0 (0) 1.0   MP + virus 15 (26.3) 6 (60) 0.097   MP + bacteria 32 (54.4) 2 (20) 0.292   Virus + bacteria 2 (3.5) 2 (20) 0.096   Two viruses 4 (7) 0 (0) 1.0   Two bacteria 1 (3.5) 0 (0) 1.0  Three pathogens 4 (3.6) 2 (4.5) 0.2   MP + virus + bacteria 3 (75) 2 (100) 0.146   MP + virus + virus 1 (25) 0 (0) 1.0 Note: Red: Fisher’s exact test; M (P25–P75): Mann–Whitney U-test. Table 4 Demographic and clinical features of community-acquired pneumonia with or without airway malacia Characteristics No airway deformity (n = 368), n (%) Airway deformity (n = 60) n (%) X2/Z test p Mean age (months) 50.0 (21.0–85.0) 10.5 (4.75–24.25) Z = 7.372 <0.001 Sex  Male 199 (54.1) 44 (73.3)  Female 169 (45.9) 16 (26.7) X2 = 7.796 0.005 Symptom  Cough 357 (97) 59 (98.3) 1.0  Wheezing 114 (30.9) 41 (68.3) X2 = 31.164 <0.001  Fever 252 (68.5) 24 (40) X2 = 18.269 <0.001  Dyspnea 26 (7.1) 8 (13.3) X2 = 2.27 0.132  Cyanosis 3 (0.8) 5 (8.3) 0.002 Physical examination  Lung wheezing rales 114 (30.9) 41 (68.3) X2 = 31.164 <0.001  Lung moisture rales 11 (39.2) 12 (9.3) X2 = 29.359 <0.001 Lab tests WBC (×109) 9.0445 (6.59–13.02) 10.83 (8.0795–15.65) Z = 2.769 0.006 CRP (mg/l) 1.04 (0.2–9.02) 1.03 (0.135–2.915) Z = 1.194 0.233 PICU admission 13 (3.5) 7 (11.7) 0.013 Hospital stay (days) 8.0 (6.0–11.0) 10.0 (7.0–16.0) Z = 2.912 0.004 Cellularity of BALF  Neutrophil (%) 40.0 (15.0–74.0) 39.0 (19.5–67.5) Z = 0.433 0.665  Lymphocyte (%) 3.0 (2.0–7.0) 4.0 (2.0–7.0) Z = 2.157 0.031  Macrophage (%) 51.0 (19.0–81.0) 55.0 (21.5–73.0) Z = 0.23 0.818  Eosinophil (%) 0.0 (0.0–1.0) 0 (0–1.0) Z = 0.151 0.88 Positive pathogen of BALF 249 (67.7) 44 (73.3) X2 = 0.768 0.381  One pathogen 188 (75.5) 32 (72.7) X2 = 0.104 0.747   Mycoplasma pneumoniae 153 (81.4) 6 (18.8) X2 = 10.138 <0.001   Bacteria 21 (11.2) 8 (25) 0.057   Virus 13 (6.9) 18 (56.3) X2 = 25.436 <0.001   Chlamydia pneumoniae 1 (0.5) 0 (0) 1.0  Two pathogens 57 (22.9) 10 (22.7) X2 = 0.054 0.816   MP + CP 3 (5.3) 0 (0) 1.0   MP + virus 15 (26.3) 6 (60) 0.097   MP + bacteria 32 (54.4) 2 (20) 0.292   Virus + bacteria 2 (3.5) 2 (20) 0.096   Two viruses 4 (7) 0 (0) 1.0   Two bacteria 1 (3.5) 0 (0) 1.0  Three pathogens 4 (3.6) 2 (4.5) 0.2   MP + virus + bacteria 3 (75) 2 (100) 0.146   MP + virus + virus 1 (25) 0 (0) 1.0 Characteristics No airway deformity (n = 368), n (%) Airway deformity (n = 60) n (%) X2/Z test p Mean age (months) 50.0 (21.0–85.0) 10.5 (4.75–24.25) Z = 7.372 <0.001 Sex  Male 199 (54.1) 44 (73.3)  Female 169 (45.9) 16 (26.7) X2 = 7.796 0.005 Symptom  Cough 357 (97) 59 (98.3) 1.0  Wheezing 114 (30.9) 41 (68.3) X2 = 31.164 <0.001  Fever 252 (68.5) 24 (40) X2 = 18.269 <0.001  Dyspnea 26 (7.1) 8 (13.3) X2 = 2.27 0.132  Cyanosis 3 (0.8) 5 (8.3) 0.002 Physical examination  Lung wheezing rales 114 (30.9) 41 (68.3) X2 = 31.164 <0.001  Lung moisture rales 11 (39.2) 12 (9.3) X2 = 29.359 <0.001 Lab tests WBC (×109) 9.0445 (6.59–13.02) 10.83 (8.0795–15.65) Z = 2.769 0.006 CRP (mg/l) 1.04 (0.2–9.02) 1.03 (0.135–2.915) Z = 1.194 0.233 PICU admission 13 (3.5) 7 (11.7) 0.013 Hospital stay (days) 8.0 (6.0–11.0) 10.0 (7.0–16.0) Z = 2.912 0.004 Cellularity of BALF  Neutrophil (%) 40.0 (15.0–74.0) 39.0 (19.5–67.5) Z = 0.433 0.665  Lymphocyte (%) 3.0 (2.0–7.0) 4.0 (2.0–7.0) Z = 2.157 0.031  Macrophage (%) 51.0 (19.0–81.0) 55.0 (21.5–73.0) Z = 0.23 0.818  Eosinophil (%) 0.0 (0.0–1.0) 0 (0–1.0) Z = 0.151 0.88 Positive pathogen of BALF 249 (67.7) 44 (73.3) X2 = 0.768 0.381  One pathogen 188 (75.5) 32 (72.7) X2 = 0.104 0.747   Mycoplasma pneumoniae 153 (81.4) 6 (18.8) X2 = 10.138 <0.001   Bacteria 21 (11.2) 8 (25) 0.057   Virus 13 (6.9) 18 (56.3) X2 = 25.436 <0.001   Chlamydia pneumoniae 1 (0.5) 0 (0) 1.0  Two pathogens 57 (22.9) 10 (22.7) X2 = 0.054 0.816   MP + CP 3 (5.3) 0 (0) 1.0   MP + virus 15 (26.3) 6 (60) 0.097   MP + bacteria 32 (54.4) 2 (20) 0.292   Virus + bacteria 2 (3.5) 2 (20) 0.096   Two viruses 4 (7) 0 (0) 1.0   Two bacteria 1 (3.5) 0 (0) 1.0  Three pathogens 4 (3.6) 2 (4.5) 0.2   MP + virus + bacteria 3 (75) 2 (100) 0.146   MP + virus + virus 1 (25) 0 (0) 1.0 Note: Red: Fisher’s exact test; M (P25–P75): Mann–Whitney U-test. Comparison of clinical features of single- and mixed-pathogen infection in CAP patients without airway malacia Of those CAP patients without airway malacia, the proportion of males with mixed-pathogen infections was higher than that of females. CRP was higher in those with mixed-pathogen infections without airway malacia than in those with single-pathogen infections (Table 5). Table 5 Demographic and clinical features of single-pathogen and mixed-pathogen infections in community-acquired pneumonia without airway malacia Characteristics Single-pathogen infection (n = 188), n (%) Mixed-pathogen infection (n = 61) n (%) X2/Z test p Mean age (months) 49.5 (23.0–80.75) 53.0 (19.0–83.25) Z = 0.535 0.593 Sex  Male 93 (54.1) 44 (73.3)  Female 95 (45.9) 17 (26.7) X2 = 9.558 0.002 Symptom  Cough 183 (97.3) 59 (96.7) 0.681  Wheezing 60 (31.9) 21 (34.4) X2 = 0.132 0.716  Fever 127 (67.6) 46 (75.4) X2 = 1.341 0.247  Dyspnea 7 (3.7) 6 (9.8) 0.092  Cyanosis 3 (1.6) 0 (0) 1.0 Physical examination  Lung wheezing rales 58 (30.9) 21 (34.4) X2 = 0.272 0.602  Lung moisture rales 115 (61.2) 33 (54.1) X2 = 0.955 0.328 Lab tests WBC (×109) 8.93 (6.555–13.09) 8.86 (6.77–15.28) Z = 0.422 0.673 CRP (mg/l) 0.41 (0.1–1.65) 3.7045 (0.275–12.515) Z = 3.49 0.001 PICU admission 6 (3.2) 3 (4.9) 0.693 Hospital stay (days) 9.0 (7.0–12.0) 8.0 (6.0–10.0) Z = 1.6 0.11 Characteristics Single-pathogen infection (n = 188), n (%) Mixed-pathogen infection (n = 61) n (%) X2/Z test p Mean age (months) 49.5 (23.0–80.75) 53.0 (19.0–83.25) Z = 0.535 0.593 Sex  Male 93 (54.1) 44 (73.3)  Female 95 (45.9) 17 (26.7) X2 = 9.558 0.002 Symptom  Cough 183 (97.3) 59 (96.7) 0.681  Wheezing 60 (31.9) 21 (34.4) X2 = 0.132 0.716  Fever 127 (67.6) 46 (75.4) X2 = 1.341 0.247  Dyspnea 7 (3.7) 6 (9.8) 0.092  Cyanosis 3 (1.6) 0 (0) 1.0 Physical examination  Lung wheezing rales 58 (30.9) 21 (34.4) X2 = 0.272 0.602  Lung moisture rales 115 (61.2) 33 (54.1) X2 = 0.955 0.328 Lab tests WBC (×109) 8.93 (6.555–13.09) 8.86 (6.77–15.28) Z = 0.422 0.673 CRP (mg/l) 0.41 (0.1–1.65) 3.7045 (0.275–12.515) Z = 3.49 0.001 PICU admission 6 (3.2) 3 (4.9) 0.693 Hospital stay (days) 9.0 (7.0–12.0) 8.0 (6.0–10.0) Z = 1.6 0.11 Note: Red: Fisher’s exact test. Table 5 Demographic and clinical features of single-pathogen and mixed-pathogen infections in community-acquired pneumonia without airway malacia Characteristics Single-pathogen infection (n = 188), n (%) Mixed-pathogen infection (n = 61) n (%) X2/Z test p Mean age (months) 49.5 (23.0–80.75) 53.0 (19.0–83.25) Z = 0.535 0.593 Sex  Male 93 (54.1) 44 (73.3)  Female 95 (45.9) 17 (26.7) X2 = 9.558 0.002 Symptom  Cough 183 (97.3) 59 (96.7) 0.681  Wheezing 60 (31.9) 21 (34.4) X2 = 0.132 0.716  Fever 127 (67.6) 46 (75.4) X2 = 1.341 0.247  Dyspnea 7 (3.7) 6 (9.8) 0.092  Cyanosis 3 (1.6) 0 (0) 1.0 Physical examination  Lung wheezing rales 58 (30.9) 21 (34.4) X2 = 0.272 0.602  Lung moisture rales 115 (61.2) 33 (54.1) X2 = 0.955 0.328 Lab tests WBC (×109) 8.93 (6.555–13.09) 8.86 (6.77–15.28) Z = 0.422 0.673 CRP (mg/l) 0.41 (0.1–1.65) 3.7045 (0.275–12.515) Z = 3.49 0.001 PICU admission 6 (3.2) 3 (4.9) 0.693 Hospital stay (days) 9.0 (7.0–12.0) 8.0 (6.0–10.0) Z = 1.6 0.11 Characteristics Single-pathogen infection (n = 188), n (%) Mixed-pathogen infection (n = 61) n (%) X2/Z test p Mean age (months) 49.5 (23.0–80.75) 53.0 (19.0–83.25) Z = 0.535 0.593 Sex  Male 93 (54.1) 44 (73.3)  Female 95 (45.9) 17 (26.7) X2 = 9.558 0.002 Symptom  Cough 183 (97.3) 59 (96.7) 0.681  Wheezing 60 (31.9) 21 (34.4) X2 = 0.132 0.716  Fever 127 (67.6) 46 (75.4) X2 = 1.341 0.247  Dyspnea 7 (3.7) 6 (9.8) 0.092  Cyanosis 3 (1.6) 0 (0) 1.0 Physical examination  Lung wheezing rales 58 (30.9) 21 (34.4) X2 = 0.272 0.602  Lung moisture rales 115 (61.2) 33 (54.1) X2 = 0.955 0.328 Lab tests WBC (×109) 8.93 (6.555–13.09) 8.86 (6.77–15.28) Z = 0.422 0.673 CRP (mg/l) 0.41 (0.1–1.65) 3.7045 (0.275–12.515) Z = 3.49 0.001 PICU admission 6 (3.2) 3 (4.9) 0.693 Hospital stay (days) 9.0 (7.0–12.0) 8.0 (6.0–10.0) Z = 1.6 0.11 Note: Red: Fisher’s exact test. Comparison of clinical features of single- and mixed-pathogen infections in CAP with airway malacia There was no significant difference in clinical characteristics, such as age, fever, wheezing, dyspnea, cyanosis, admission to PICU and hospital stay, between patients with single-pathogen infections and those with mixed-pathogen infections in CAP with airway malacia (Table 6). Table 6 Demographic and clinical features of single-pathogen and mixed-pathogen infections in CAP patients with airway malacia Characteristics Single-pathogen infection (n = 32), n (%) Mixed-pathogen infection (n = 12), n (%) X2/Z test p Mean age (months) 10.5 (4.0–23.0) 7.5 (4.5–14.5) Z = 0.158 0.874 Sex  Male 24 (75) 9 (75)  Female 8 (25) 3 (25) 1.0 Symptom  Cough 31 (96.8) 12 (100) 1.0  Wheezing 20 (62.5) 10 (83.3) 0.282  Fever 12 (37.5) 4 (41.7) 1.0  Dyspnea 6 (18.8) 2 (16.7) 1.0  Cyanosis 3 (9.4) 2 (16.7) 0.603 Physical examination  Lung wheezing rales 18 (56.3) 8 (66.7) 0.733  Lung moisture rales 23 (71.8) 11 (91.7) 0.241 Lab tests WBC (×109) 10.05 (7.96–13.51) 13.07 (9.04–18.44) Z = 1.814 0.07 CRP (mg/l) 1.84 (0.115–3.2775) 1.94 (0.275–6.1675) Z = 0.514 0.607 PICU admission 5 (15.6) 2 (16.7) 1.0 Hospital stay (days) 9.0 (7.25–15.0) 12.5 (8.5–17.0) Z = 1.23 0.219 Characteristics Single-pathogen infection (n = 32), n (%) Mixed-pathogen infection (n = 12), n (%) X2/Z test p Mean age (months) 10.5 (4.0–23.0) 7.5 (4.5–14.5) Z = 0.158 0.874 Sex  Male 24 (75) 9 (75)  Female 8 (25) 3 (25) 1.0 Symptom  Cough 31 (96.8) 12 (100) 1.0  Wheezing 20 (62.5) 10 (83.3) 0.282  Fever 12 (37.5) 4 (41.7) 1.0  Dyspnea 6 (18.8) 2 (16.7) 1.0  Cyanosis 3 (9.4) 2 (16.7) 0.603 Physical examination  Lung wheezing rales 18 (56.3) 8 (66.7) 0.733  Lung moisture rales 23 (71.8) 11 (91.7) 0.241 Lab tests WBC (×109) 10.05 (7.96–13.51) 13.07 (9.04–18.44) Z = 1.814 0.07 CRP (mg/l) 1.84 (0.115–3.2775) 1.94 (0.275–6.1675) Z = 0.514 0.607 PICU admission 5 (15.6) 2 (16.7) 1.0 Hospital stay (days) 9.0 (7.25–15.0) 12.5 (8.5–17.0) Z = 1.23 0.219 Note: Red: Fisher’s exact test. Table 6 Demographic and clinical features of single-pathogen and mixed-pathogen infections in CAP patients with airway malacia Characteristics Single-pathogen infection (n = 32), n (%) Mixed-pathogen infection (n = 12), n (%) X2/Z test p Mean age (months) 10.5 (4.0–23.0) 7.5 (4.5–14.5) Z = 0.158 0.874 Sex  Male 24 (75) 9 (75)  Female 8 (25) 3 (25) 1.0 Symptom  Cough 31 (96.8) 12 (100) 1.0  Wheezing 20 (62.5) 10 (83.3) 0.282  Fever 12 (37.5) 4 (41.7) 1.0  Dyspnea 6 (18.8) 2 (16.7) 1.0  Cyanosis 3 (9.4) 2 (16.7) 0.603 Physical examination  Lung wheezing rales 18 (56.3) 8 (66.7) 0.733  Lung moisture rales 23 (71.8) 11 (91.7) 0.241 Lab tests WBC (×109) 10.05 (7.96–13.51) 13.07 (9.04–18.44) Z = 1.814 0.07 CRP (mg/l) 1.84 (0.115–3.2775) 1.94 (0.275–6.1675) Z = 0.514 0.607 PICU admission 5 (15.6) 2 (16.7) 1.0 Hospital stay (days) 9.0 (7.25–15.0) 12.5 (8.5–17.0) Z = 1.23 0.219 Characteristics Single-pathogen infection (n = 32), n (%) Mixed-pathogen infection (n = 12), n (%) X2/Z test p Mean age (months) 10.5 (4.0–23.0) 7.5 (4.5–14.5) Z = 0.158 0.874 Sex  Male 24 (75) 9 (75)  Female 8 (25) 3 (25) 1.0 Symptom  Cough 31 (96.8) 12 (100) 1.0  Wheezing 20 (62.5) 10 (83.3) 0.282  Fever 12 (37.5) 4 (41.7) 1.0  Dyspnea 6 (18.8) 2 (16.7) 1.0  Cyanosis 3 (9.4) 2 (16.7) 0.603 Physical examination  Lung wheezing rales 18 (56.3) 8 (66.7) 0.733  Lung moisture rales 23 (71.8) 11 (91.7) 0.241 Lab tests WBC (×109) 10.05 (7.96–13.51) 13.07 (9.04–18.44) Z = 1.814 0.07 CRP (mg/l) 1.84 (0.115–3.2775) 1.94 (0.275–6.1675) Z = 0.514 0.607 PICU admission 5 (15.6) 2 (16.7) 1.0 Hospital stay (days) 9.0 (7.25–15.0) 12.5 (8.5–17.0) Z = 1.23 0.219 Note: Red: Fisher’s exact test. Associations between severity of airway malacia and PICU stay and duration of hospitalization No statistically significant associations between severity of airway malacia and PICU admission and duration of hospitalization were observed in the airway malacia group (Table 7). Table 7 Associations between severity of airway malacia and admission to the PICU and duration of hospitalization Characteristics Mild malacia (n = 43) Moderate malacia (n = 14) Severe malacia (n = 3) Kruskal–Wallis/Fisher p PICU admission 6 (13.9) 0 (0) 1 (33.3) 2.064 >0.05 Duration of hospitalization 12.23 ± 6.34 9.5 ± 5.56 12.0 ± 11.36 3.565 >0.05 Characteristics Mild malacia (n = 43) Moderate malacia (n = 14) Severe malacia (n = 3) Kruskal–Wallis/Fisher p PICU admission 6 (13.9) 0 (0) 1 (33.3) 2.064 >0.05 Duration of hospitalization 12.23 ± 6.34 9.5 ± 5.56 12.0 ± 11.36 3.565 >0.05 Table 7 Associations between severity of airway malacia and admission to the PICU and duration of hospitalization Characteristics Mild malacia (n = 43) Moderate malacia (n = 14) Severe malacia (n = 3) Kruskal–Wallis/Fisher p PICU admission 6 (13.9) 0 (0) 1 (33.3) 2.064 >0.05 Duration of hospitalization 12.23 ± 6.34 9.5 ± 5.56 12.0 ± 11.36 3.565 >0.05 Characteristics Mild malacia (n = 43) Moderate malacia (n = 14) Severe malacia (n = 3) Kruskal–Wallis/Fisher p PICU admission 6 (13.9) 0 (0) 1 (33.3) 2.064 >0.05 Duration of hospitalization 12.23 ± 6.34 9.5 ± 5.56 12.0 ± 11.36 3.565 >0.05 DISCUSSION Current management of lower respiratory tract infections in children with airway malacia, which includes liberal use of antibiotics and physiotherapy, is not evidence based [9]; therefore, determining the correct pathogen is particularly important for determining the correct treatment. This is the first study examining the etiology of pneumonia with airway malacia in China. Of the 60 protracted and recurrent CAP cases with airway malacia, with a positive pathogen detection rate of 73.3%, we found RSV to be the most common pathogen, followed by SP and MP. However, in CAP patients without airway malacia, MP was the most common pathogen, the results of which are different from those of previous reports of 10–40% [10, 11]. In addition, RSV and MP incidences were higher than those in previous reports [12, 13]. De Baets et al. [6] studied 124 children with persistent respiratory symptoms and fiberoptic bronchoscopy with BAL were performed; they found that 47% of cases had structural abnormality of the central airways, 56% of BAL samples tested positive for bacterial culture with Moraxella catarrhalis (51%), HI (28%), SP (13%), Staphylococcus aureus (10%). The pathogen detection rate was higher than that in our study, and the pathogen was different. In those patients with tracheobronchomalacia, which lack