Clinical Features and Outcome of Children with Severe Lower Respiratory Tract Infection Admitted to a Pediatric Intensive Care Unit in South Africa

Clinical Features and Outcome of Children with Severe Lower Respiratory Tract Infection Admitted... Abstract Aim Severe acute lower respiratory tract infection (ALRTI) remains an important cause of childhood morbidity and mortality. Methods This is a 12-month retrospective cohort study of children (0–12 years) admitted to a pediatric intensive care unit (PICU) with ALRTI to investigate risk factors, clinical course and in-hospital survival. Results In total, 265 patients (median age = 4 months [2–12 months]) were identified. In all,102 (38.5%) had co-morbid disease. Twenty-seven (10.2%) were HIV-infected and 87 (32.8%) were HIV-exposed. In-hospital mortality was 34 (12.8%)—24 (9.1%) in PICU and 10 in the wards. Median duration of intensive care unit was 4.0 days (2.0–8.0) and hospital stay was 12.5 days (7.9–28.0). In total, 192 (72.5%) children required invasive ventilation and 42 (15.8%) required inotropic support. Risk factors for mortality included severe malnutrition (odds ratio [OR] = 8.25; 95% confidence interval [CI] = 1.47–46.21), informal housing (OR = 11.87; CI = 1.89–20.81) or inotropic support (OR = 44.35; CI = 8.20–239.92). HIV exposure or infection was associated with a longer duration of hospital stay (OR = 4.41; CI = 2.44–6.39). Conclusion Severe ALRTI is associated with a high mortality with several factors impacting on in-hospital survival. respiratory tract infection, intensive care unit, pediatric, survival, risk factor INTRODUCTION Acute lower respiratory tract infection (ALRTI), primarily pneumonia, is a leading cause of death in children under the age of 5 years, placing a considerable burden on health and economic resources particularly in low- and middle-income countries (LMICs) [1]. The World Health Organization (WHO) recognizes that pneumonia alone is responsible for up to 15% of deaths in children [2]. In Africa, pneumonia is the most important single cause of under 5-year mortality outside the neonatal period [3], with a disproportionate number of cases of severe disease relative to the childhood population. In South Africa, pneumonia remains one of the leading causes of death in hospital in children under 5 years [4, 5]. There are limited data on hospital outcome of severe ALRTI in African children. WHO estimated 19.2 million cases of severe ALRTI in 2010, with only a proportion of those being treated in a hospital setting [6]. The in-hospital case fatality rate ranges from 2.3% to 6% depending on severity of disease [6]. A study done in 1998 at Red Cross War Memorial Children’s Hospital (RCH) of HIV-infected and uninfected children admitted to a pediatric intensive care unit (PICU) with ALRTI reported a mortality rate of 29% and 14%, respectively [7]. Importantly this study preceded the roll-out of highly active antiretroviral therapy (ART) for children. The limited African data suggest a much higher case fatality rate than that reported by the WHO [6] and highlights the need for further research into the causes, risk factors and outcome of severe ALRTI in this setting. Several factors may predispose children to ALRTI and to severe disease. Reported risk factors include age less than 1 year [8–10], prematurity, low birth weight [11–13], severe or moderate malnutrition [14], micronutrient deficiency or co-morbid disease (especially congenital cardiac disease [15]) and immunosuppressive disease (especially HIV infection [16]). Increasing evidence suggests that HIV-exposed and uninfected children have a higher risk for ALRTI or opportunistic infections than unexposed children [17, 18]. A protective factor identified in many studies is exclusive breastfeeding for 6 months [19–21]; protection wanes with age, but exclusive breastfeeding may halve the risk of ALRTI in infancy. The aim of this study was to describe the clinical characteristics and outcomes of children with severe ALRTI admitted to a PICU in a tertiary hospital in South Africa and to identify risk factors associated with in-hospital mortality. METHODS Study design and setting A retrospective study of patients admitted to the PICU at RCH, South Africa, with a diagnosis of ALRTI was conducted. RCH is a tertiary hospital with 300 beds offering comprehensive care to pediatric patients in the Western Cape. The PICU is a 20-bedded multidisciplinary unit. Children under 12 years admitted to PICU with a primary diagnosis of ALRTI from January 2012 to December 2012 were included. Children who developed a nosocomial ALRTI (defined as developing after 48 h of hospital admission) or those for whom the medical records were incomplete were excluded from the study. Data source The PICU database (Ethics approval number 139/2011) was used as the primary source of information. This is a database of all patients admitted to the PICU capturing demographic, clinical and socioeconomic details routinely required during a PICU admission. The database was used to identify patients with a primary diagnosis of lower respiratory tract disease using the following subheadings: ‘pneumonia’, ‘lower respiratory tract infection’, ‘bronchopneumonia’, ‘bronchiolitis’ and ‘pulmonary tuberculosis’. Medical records and National Health Laboratory records were used to gain further clinical information regarding risk factors and investigation results. Terminology used in this study Malnutrition was defined according to the WHO classification using z-score cut-offs of <−2 and <−3 for moderate and severe malnutrition, respectively [22]. Prematurity was defined as birth before 37 completed weeks of gestation. Birthweight was categorized as low birthweight (<2.5 kg), very low birth weight (1.00–1.49 kg) and extremely low birthweight (<1.0 kg). Fenton’s correction scale was used to adjust for gestational age in ex-premature patients when assessing birthweight and current weight in terms of z-scores [23]. Feeding data were obtained from maternal self-report. Exclusive breastfeeding, defined by WHO, is the exclusive use of breast milk for nutrition up until age 6 months or until hospital admission if younger than 6 months of age. Mixed feeding was defined as a mixture of breast and formula milk or any other oral substance for nutrition. Ventilatory support was considered noninvasive if nasal cannula, high flow nasal cannula oxygen or continuous positive airway pressure was used or invasive if positive pressure ventilation or high-frequency oscillatory ventilation was used via an endotracheal tube or a tracheostomy. Data on HIV status were confirmed using results accessed via the National Health Laboratory Services. HIV infection was defined as a child who was polymerase chain reaction (PCR)-positive if under 18 months or antibody-positive if older. HIV exposure was defined as a child who tested HIV-negative but whose mother was recorded as being HIV-infected. Data on prevention of mother-to-child transmission (PMTCT) strategies used were obtained from records. Data on microbiological etiology were obtained from culture of blood, induced sputum, tracheal aspirate, bronchoalveolar lavage or pleural fluid for bacteria and fungi. Viral PCR panel testing for seven respiratory viruses was performed on nasopharyngeal aspirates and bronchoalveolar lavage specimens. Fungal PCR testing was performed on bronchoalveolar lavage specimens and tracheal aspirates. Tuberculosis microscopy, culture and PCR were performed on induced sputum and gastric washings. Cytomegalovirus (CMV) was measured in HIV-infected children with a quantitative viral load of log >4 considered positive for disease. Outcomes measured included duration of PICU stay, duration of in-hospital stay and survival to discharge. The pediatric index of mortality score (PIM2) was used to predict mortality, as per standard practice in 2012. PIM2 is a composite score of clinical factors, biochemical markers, ventilatory requirements and underlying risk factors at the time of PICU admission that has been validated for use in LMIC [24, 25]. The standardized mortality rate (SMR) is calculated by dividing the actual mortality by the predicted mortality score. The SMR is a marker of the quality of care or performance in the unit. Statistical analysis Data were analyzed using STATA 13 statistical software (STATA Corporation, TX, USA). Exploratory data analysis of categorical and continuous variables included frequency tables and histograms to determine distribution. Simple descriptive statistics were used to characterize the study population. Statistical tests included the chi-square test and Kruskal–Wallis comparison of medians and Student’s t-test for comparison of means. Risk factors tending toward statistical significance (p < 0.26) or those chosen on clinical reasoning were included in a multivariate Cox regression analysis. Factors considered to be competing risk factors for mortality were excluded (duration of PICU stay and invasive ventilatory requirements). HIV infection and breastfeeding were treated as potential confounding variables. Statistical tests were two-sided at α = 0.05. Ethical considerations Ethical approval: University of Cape Town Faculty of Health Science Human Research Ethics Committee (HREC 435/2013). RESULTS In total, 358 patients were identified; 265 complete patient data sets were included for analysis (Fig. 1). A total of 265 patients (median [interquartile range, IQR] age = 4 months [2–12 months]) were admitted with ALRTI; 157 (59.3%) were male. Co-morbidities were present in 102 (38.5%) cases, cardiac disease in 42 (15.9%) and tuberculosis in 7 (6.4%). Whereas only 27 (10.2%) patients were HIV-infected, 87 (32.8%) were HIV-exposed. Baseline characteristics are described in Table 1. Table 1 Baseline characteristics of children admitted with ALRTI, by hospital outcome including univariate analysis of characteristic on hospital outcome Characteristic All N (%) Survived N (%) Died N (%) Odds ratio (95% CI) p Value Gender  N (%) male 157 (59.3) 133 (57.6) 24 (70.6) 1.76 (0.80 to 3.86) 0.153 Age in months  Median (IQR)a 4 (2 to 12) 4 (2 to 12) 6.5 (2.5 to 12) 1.00 (0.98 to 1.01) 1 Birthweight WAZb  Median (IQR) −0.8 (−1.4 to 0) −0.8 (−1.4 to 0.1) −0.9 (−1.5 to 0.5) 0.88 (0.63 to 1.23) 0.458 Preterm (<37 weeks) 60 (22.6) 53 (22.9) 7 (20.6) 0.87 (0.36 to 2.11) 0.759 HIV status N (%)  Unexposed 178 (67.2) 155 (67.1) 23 (67.7) Reference  Exposed, not infected 60 (22.6) 52 (22.5) 8 (23.5) 1.04 (0.44 to 2.45) 0.935  Exposed, infected 27 (10.2) 24 (10.4) 3 (8.8) 0.84 (0.23 to 3.02) 0.792 PMTCTcN (%)  Antenatal received 44 (50.6) 36 (47.4) 8 (72.7) 1.14 (0.90 to 1.43) 0.281  Perinatal received 50 (57.5) 41 (54.0) 9 (81.8) 1.14 (0.90 to 1.44) 0.285  Postnatal—correct 54 (62.1) 45 (59.2) 9 (81.8) 0.96 (0.73 to 1.27) 0.781  Postnatal—incorrect dose 18 (20.7) 17 (22.4) 1 (9.1) Infant feeding N (%)  Exclusive breast-fedd 121 (46.2) 99 (43.0) 22 (68.8) Reference  Exclusive formula 61 (23.3) 57 (24.8) 4 (12.5) 0.31 (0.10 to 0.96) 0.043  Mixed feeding 80 (30.5) 74 (32.2) 6 (18.8) 0.36 (0.14 to 0.94) 0.038 Nutritional status  None/mild 202 (76.5) 185 (80.1) 17 (51.5) Reference  Moderate 27 (10.2) 22 (9.5) 5 (15.2) 2.47 (0.83 to 7.36) 0.104  Severe 35 (13.3) 35 (13.3) 11 (33.1) 4.99 (2.90 to 11.9) <0.001 Income bracket  ≥R4,000 per month 20 (7.5) 18 (7.8) 2 (5.9) Reference  <R4,000 per month 245 (92.5) 213 (92.2) 32 (94.1) 1.35 (0.30 to 6.10) 0.695 Housing  Formal 136 (52.3) 126 (55.5) 10 (44.5) Reference  Informal 124 (47.7) 101 (30.3) 23 (69.7) 2.87 (1.30 to 6.30) 0.009 Tobacco smoke exposure (self-report) 59 (25.1) 56 (26.8) 3 (11.5) 0.36 (0.10 to 1.23) 0.103 TB disease 17 (6.4) 14 (6.1) 3 (8.8) 1.50 (0.41 to 5.52) 0.542 No comorbid disease 164 (61.9) 148 (64.1) 16 (47.1) Reference Cardiac diagnosis 42 (15.9) 32 (13.9) 10 (29.4) 2.89 (1.20 to 6.95) 0.018 Other 59 (22.3) 51 (22.1) 8 (23.5) 1.45 (0.59 to 3.59) 0.421 At least one dose of pneumococcal conjugate vaccine administered 180 (67.9) 155 (67.1) 25 (73.5) 1.36 (0.61 to 3.06) 0.455 Characteristic All N (%) Survived N (%) Died N (%) Odds ratio (95% CI) p Value Gender  N (%) male 157 (59.3) 133 (57.6) 24 (70.6) 1.76 (0.80 to 3.86) 0.153 Age in months  Median (IQR)a 4 (2 to 12) 4 (2 to 12) 6.5 (2.5 to 12) 1.00 (0.98 to 1.01) 1 Birthweight WAZb  Median (IQR) −0.8 (−1.4 to 0) −0.8 (−1.4 to 0.1) −0.9 (−1.5 to 0.5) 0.88 (0.63 to 1.23) 0.458 Preterm (<37 weeks) 60 (22.6) 53 (22.9) 7 (20.6) 0.87 (0.36 to 2.11) 0.759 HIV status N (%)  Unexposed 178 (67.2) 155 (67.1) 23 (67.7) Reference  Exposed, not infected 60 (22.6) 52 (22.5) 8 (23.5) 1.04 (0.44 to 2.45) 0.935  Exposed, infected 27 (10.2) 24 (10.4) 3 (8.8) 0.84 (0.23 to 3.02) 0.792 PMTCTcN (%)  Antenatal received 44 (50.6) 36 (47.4) 8 (72.7) 1.14 (0.90 to 1.43) 0.281  Perinatal received 50 (57.5) 41 (54.0) 9 (81.8) 1.14 (0.90 to 1.44) 0.285  Postnatal—correct 54 (62.1) 45 (59.2) 9 (81.8) 0.96 (0.73 to 1.27) 0.781  Postnatal—incorrect dose 18 (20.7) 17 (22.4) 1 (9.1) Infant feeding N (%)  Exclusive breast-fedd 121 (46.2) 99 (43.0) 22 (68.8) Reference  Exclusive formula 61 (23.3) 57 (24.8) 4 (12.5) 0.31 (0.10 to 0.96) 0.043  Mixed feeding 80 (30.5) 74 (32.2) 6 (18.8) 0.36 (0.14 to 0.94) 0.038 Nutritional status  None/mild 202 (76.5) 185 (80.1) 17 (51.5) Reference  Moderate 27 (10.2) 22 (9.5) 5 (15.2) 2.47 (0.83 to 7.36) 0.104  Severe 35 (13.3) 35 (13.3) 11 (33.1) 4.99 (2.90 to 11.9) <0.001 Income bracket  ≥R4,000 per month 20 (7.5) 18 (7.8) 2 (5.9) Reference  <R4,000 per month 245 (92.5) 213 (92.2) 32 (94.1) 1.35 (0.30 to 6.10) 0.695 Housing  Formal 136 (52.3) 126 (55.5) 10 (44.5) Reference  Informal 124 (47.7) 101 (30.3) 23 (69.7) 2.87 (1.30 to 6.30) 0.009 Tobacco smoke exposure (self-report) 59 (25.1) 56 (26.8) 3 (11.5) 0.36 (0.10 to 1.23) 0.103 TB disease 17 (6.4) 14 (6.1) 3 (8.8) 1.50 (0.41 to 5.52) 0.542 No comorbid disease 164 (61.9) 148 (64.1) 16 (47.1) Reference Cardiac diagnosis 42 (15.9) 32 (13.9) 10 (29.4) 2.89 (1.20 to 6.95) 0.018 Other 59 (22.3) 51 (22.1) 8 (23.5) 1.45 (0.59 to 3.59) 0.421 At least one dose of pneumococcal conjugate vaccine administered 180 (67.9) 155 (67.1) 25 (73.5) 1.36 (0.61 to 3.06) 0.455 a IQR = interquartile range. b WAZ = weight for age Z score (adjusted for gestational age using Fenton’s correctional scale). c PMTCT = prevention of mother-to-child transmission. d Exclusive breastfeeding from the time of birth until admission to hospital. Table 1 Baseline characteristics of children admitted with ALRTI, by hospital outcome including univariate analysis of characteristic on hospital outcome Characteristic All N (%) Survived N (%) Died N (%) Odds ratio (95% CI) p Value Gender  N (%) male 157 (59.3) 133 (57.6) 24 (70.6) 1.76 (0.80 to 3.86) 0.153 Age in months  Median (IQR)a 4 (2 to 12) 4 (2 to 12) 6.5 (2.5 to 12) 1.00 (0.98 to 1.01) 1 Birthweight WAZb  Median (IQR) −0.8 (−1.4 to 0) −0.8 (−1.4 to 0.1) −0.9 (−1.5 to 0.5) 0.88 (0.63 to 1.23) 0.458 Preterm (<37 weeks) 60 (22.6) 53 (22.9) 7 (20.6) 0.87 (0.36 to 2.11) 0.759 HIV status N (%)  Unexposed 178 (67.2) 155 (67.1) 23 (67.7) Reference  Exposed, not infected 60 (22.6) 52 (22.5) 8 (23.5) 1.04 (0.44 to 2.45) 0.935  Exposed, infected 27 (10.2) 24 (10.4) 3 (8.8) 0.84 (0.23 to 3.02) 0.792 PMTCTcN (%)  Antenatal received 44 (50.6) 36 (47.4) 8 (72.7) 1.14 (0.90 to 1.43) 0.281  Perinatal received 50 (57.5) 41 (54.0) 9 (81.8) 1.14 (0.90 to 1.44) 0.285  Postnatal—correct 54 (62.1) 45 (59.2) 9 (81.8) 0.96 (0.73 to 1.27) 0.781  Postnatal—incorrect dose 18 (20.7) 17 (22.4) 1 (9.1) Infant feeding N (%)  Exclusive breast-fedd 121 (46.2) 99 (43.0) 22 (68.8) Reference  Exclusive formula 61 (23.3) 57 (24.8) 4 (12.5) 0.31 (0.10 to 0.96) 0.043  Mixed feeding 80 (30.5) 74 (32.2) 6 (18.8) 0.36 (0.14 to 0.94) 0.038 Nutritional status  None/mild 202 (76.5) 185 (80.1) 17 (51.5) Reference  Moderate 27 (10.2) 22 (9.5) 5 (15.2) 2.47 (0.83 to 7.36) 0.104  Severe 35 (13.3) 35 (13.3) 11 (33.1) 4.99 (2.90 to 11.9) <0.001 Income bracket  ≥R4,000 per month 20 (7.5) 18 (7.8) 2 (5.9) Reference  <R4,000 per month 245 (92.5) 213 (92.2) 32 (94.1) 1.35 (0.30 to 6.10) 0.695 Housing  Formal 136 (52.3) 126 (55.5) 10 (44.5) Reference  Informal 124 (47.7) 101 (30.3) 23 (69.7) 2.87 (1.30 to 6.30) 0.009 Tobacco smoke exposure (self-report) 59 (25.1) 56 (26.8) 3 (11.5) 0.36 (0.10 to 1.23) 0.103 TB disease 17 (6.4) 14 (6.1) 3 (8.8) 1.50 (0.41 to 5.52) 0.542 No comorbid disease 164 (61.9) 148 (64.1) 16 (47.1) Reference Cardiac diagnosis 42 (15.9) 32 (13.9) 10 (29.4) 2.89 (1.20 to 6.95) 0.018 Other 59 (22.3) 51 (22.1) 8 (23.5) 1.45 (0.59 to 3.59) 0.421 At least one dose of pneumococcal conjugate vaccine administered 180 (67.9) 155 (67.1) 25 (73.5) 1.36 (0.61 to 3.06) 0.455 Characteristic All N (%) Survived N (%) Died N (%) Odds ratio (95% CI) p Value Gender  N (%) male 157 (59.3) 133 (57.6) 24 (70.6) 1.76 (0.80 to 3.86) 0.153 Age in months  Median (IQR)a 4 (2 to 12) 4 (2 to 12) 6.5 (2.5 to 12) 1.00 (0.98 to 1.01) 1 Birthweight WAZb  Median (IQR) −0.8 (−1.4 to 0) −0.8 (−1.4 to 0.1) −0.9 (−1.5 to 0.5) 0.88 (0.63 to 1.23) 0.458 Preterm (<37 weeks) 60 (22.6) 53 (22.9) 7 (20.6) 0.87 (0.36 to 2.11) 0.759 HIV status N (%)  Unexposed 178 (67.2) 155 (67.1) 23 (67.7) Reference  Exposed, not infected 60 (22.6) 52 (22.5) 8 (23.5) 1.04 (0.44 to 2.45) 0.935  Exposed, infected 27 (10.2) 24 (10.4) 3 (8.8) 0.84 (0.23 to 3.02) 0.792 PMTCTcN (%)  Antenatal received 44 (50.6) 36 (47.4) 8 (72.7) 1.14 (0.90 to 1.43) 0.281  Perinatal received 50 (57.5) 41 (54.0) 9 (81.8) 1.14 (0.90 to 1.44) 0.285  Postnatal—correct 54 (62.1) 45 (59.2) 9 (81.8) 0.96 (0.73 to 1.27) 0.781  Postnatal—incorrect dose 18 (20.7) 17 (22.4) 1 (9.1) Infant feeding N (%)  Exclusive breast-fedd 121 (46.2) 99 (43.0) 22 (68.8) Reference  Exclusive formula 61 (23.3) 57 (24.8) 4 (12.5) 0.31 (0.10 to 0.96) 0.043  Mixed feeding 80 (30.5) 74 (32.2) 6 (18.8) 0.36 (0.14 to 0.94) 0.038 Nutritional status  None/mild 202 (76.5) 185 (80.1) 17 (51.5) Reference  Moderate 27 (10.2) 22 (9.5) 5 (15.2) 2.47 (0.83 to 7.36) 0.104  Severe 35 (13.3) 35 (13.3) 11 (33.1) 4.99 (2.90 to 11.9) <0.001 Income bracket  ≥R4,000 per month 20 (7.5) 18 (7.8) 2 (5.9) Reference  <R4,000 per month 245 (92.5) 213 (92.2) 32 (94.1) 1.35 (0.30 to 6.10) 0.695 Housing  Formal 136 (52.3) 126 (55.5) 10 (44.5) Reference  Informal 124 (47.7) 101 (30.3) 23 (69.7) 2.87 (1.30 to 6.30) 0.009 Tobacco smoke exposure (self-report) 59 (25.1) 56 (26.8) 3 (11.5) 0.36 (0.10 to 1.23) 0.103 TB disease 17 (6.4) 14 (6.1) 3 (8.8) 1.50 (0.41 to 5.52) 0.542 No comorbid disease 164 (61.9) 148 (64.1) 16 (47.1) Reference Cardiac diagnosis 42 (15.9) 32 (13.9) 10 (29.4) 2.89 (1.20 to 6.95) 0.018 Other 59 (22.3) 51 (22.1) 8 (23.5) 1.45 (0.59 to 3.59) 0.421 At least one dose of pneumococcal conjugate vaccine administered 180 (67.9) 155 (67.1) 25 (73.5) 1.36 (0.61 to 3.06) 0.455 a IQR = interquartile range. b WAZ = weight for age Z score (adjusted for gestational age using Fenton’s correctional scale). c PMTCT = prevention of mother-to-child transmission. d Exclusive breastfeeding from the time of birth until admission to hospital. Fig. 1. View largeDownload slide Summary of patients included in the study. Fig. 1. View largeDownload slide Summary of patients included in the study. Microbiologic investigations Overall, 960 investigations were performed. In all, 206 (77.7%) patients had at least one organism identified, with 86 (32.5%) having more than one organism (Table 2). Viruses were most commonly identified (117 [44.2%]), especially respiratory syncytial virus (9%), rhinovirus (7%) and adenovirus (5%). Of 27 patients tested, 18 had a CMV log > 4 (66.7%). There was no difference in microbiological findings between HIV-unexposed and HIV-infected groups (odds ratio [OR] = 0.99; 95% confidence interval [CI] = 0.65–1.53). Table 2 Organisms isolated and associated risk of in-hospital mortality Organism isolated Number Proportion (%) Odds ratio (95% CI) p Value None 59 22.30 Bacteriaa  Single bacterium isolated 25 9.20 4.01 (1.53 to 10.57) 0.005  More than one bacteria isolated 2 1.00 Virusb  Single virus isolated 91 34.20 0.43 (0.15 to 1.21) 0.112   More than one virus isolated 26 10.00 Fungal (Pneumocystis jirovecii)c 4 1.50 Mixed infections  Bacterial/viral 45 17.00 1.07 (0.38 to 3.03) 0.885  Bacterial/fungal 3 1.10  Viral/fungal 7 2.60  Bacterial/viral/fungal 3 1.10 Total 203 100 Organism isolated Number Proportion (%) Odds ratio (95% CI) p Value None 59 22.30 Bacteriaa  Single bacterium isolated 25 9.20 4.01 (1.53 to 10.57) 0.005  More than one bacteria isolated 2 1.00 Virusb  Single virus isolated 91 34.20 0.43 (0.15 to 1.21) 0.112   More than one virus isolated 26 10.00 Fungal (Pneumocystis jirovecii)c 4 1.50 Mixed infections  Bacterial/viral 45 17.00 1.07 (0.38 to 3.03) 0.885  Bacterial/fungal 3 1.10  Viral/fungal 7 2.60  Bacterial/viral/fungal 3 1.10 Total 203 100 a Bacteria were isolated from blood culture, bronchoalveolar lavages, culture of tracheal aspirates or pleural fluid. b Viruses detected on polymerase chain reaction testing from nasopharyngeal samples or bronchoalveolar lavages. c No deaths in children who cultured fungal organisms. Table 2 Organisms isolated and associated risk of in-hospital mortality Organism isolated Number Proportion (%) Odds ratio (95% CI) p Value None 59 22.30 Bacteriaa  Single bacterium isolated 25 9.20 4.01 (1.53 to 10.57) 0.005  More than one bacteria isolated 2 1.00 Virusb  Single virus isolated 91 34.20 0.43 (0.15 to 1.21) 0.112   More than one virus isolated 26 10.00 Fungal (Pneumocystis jirovecii)c 4 1.50 Mixed infections  Bacterial/viral 45 17.00 1.07 (0.38 to 3.03) 0.885  Bacterial/fungal 3 1.10  Viral/fungal 7 2.60  Bacterial/viral/fungal 3 1.10 Total 203 100 Organism isolated Number Proportion (%) Odds ratio (95% CI) p Value None 59 22.30 Bacteriaa  Single bacterium isolated 25 9.20 4.01 (1.53 to 10.57) 0.005  More than one bacteria isolated 2 1.00 Virusb  Single virus isolated 91 34.20 0.43 (0.15 to 1.21) 0.112   More than one virus isolated 26 10.00 Fungal (Pneumocystis jirovecii)c 4 1.50 Mixed infections  Bacterial/viral 45 17.00 1.07 (0.38 to 3.03) 0.885  Bacterial/fungal 3 1.10  Viral/fungal 7 2.60  Bacterial/viral/fungal 3 1.10 Total 203 100 a Bacteria were isolated from blood culture, bronchoalveolar lavages, culture of tracheal aspirates or pleural fluid. b Viruses detected on polymerase chain reaction testing from nasopharyngeal samples or bronchoalveolar lavages. c No deaths in children who cultured fungal organisms. Children who had bacterium/bacteria isolated from blood culture, bronchoalveolar lavage, sputum or histology were four times more likely to die (OR = 4.01; 95% = CI 1.53–10.57). Outcomes Of 265 patients admitted to a PICU, 34 (12.8%) died in-hospital—24 (9.1%) patients died in the PICU and 10 in the medical wards following discharge (Table 3). Median duration of hospital stay was 12.5 days (7.9–28.0). Median duration of ICU stay was 4.0 days (2.0–8.0). Ventilatory support was required by 250 (94.3%) patients (mean duration of ventilatory support of 3.5 days [1.5–6.9]). Invasive ventilatory support (conventional mechanical ventilation or high-frequency oscillatory ventilation) was used in 192 (72.5%) patients. Inotropic support for hemodynamic instability was used in 42 (15.7%) patients. Table 3 Details of hospital stay by hospital outcome Characteristic All N (%) Survived N (%) Died N (%) OR (95% CI) p Value Duration in PICU (days) 4.0 (2.0 to 8.0) 4 (2.0 to 7.0) 5.0 (2.0 to 12) 1.00 (1.00 to 1.01) 0.053 Duration in hospital (days) 12.5 (7.9 to 28.0) 12.9 (8.1 to 27.0) 8.35 (2 to 48) 1.01 (1.00 to 1.02) 0.008 Invasive ventilationa 155 (58.4) 126 (81.3) 29 (18.7) 5.75 (2.24 to 14.98) <0.001 N (%) Inotropic support 42 (15.7) 18 (7.8) 23 (67.7) 24.7 (10.4 to 58.7) <0.001 Bacterial isolateb 27 (13.1) 19 (10.5) 8 (32.0) 4.01 (1.53 to 10.54) 0.005 Length of time to start antibiotics (h) 1 (1 to 3) 2 (1 to 3) 1 (1 to 2) 1.29 (0.98 to 1.70) 0.068 Characteristic All N (%) Survived N (%) Died N (%) OR (95% CI) p Value Duration in PICU (days) 4.0 (2.0 to 8.0) 4 (2.0 to 7.0) 5.0 (2.0 to 12) 1.00 (1.00 to 1.01) 0.053 Duration in hospital (days) 12.5 (7.9 to 28.0) 12.9 (8.1 to 27.0) 8.35 (2 to 48) 1.01 (1.00 to 1.02) 0.008 Invasive ventilationa 155 (58.4) 126 (81.3) 29 (18.7) 5.75 (2.24 to 14.98) <0.001 N (%) Inotropic support 42 (15.7) 18 (7.8) 23 (67.7) 24.7 (10.4 to 58.7) <0.001 Bacterial isolateb 27 (13.1) 19 (10.5) 8 (32.0) 4.01 (1.53 to 10.54) 0.005 Length of time to start antibiotics (h) 1 (1 to 3) 2 (1 to 3) 1 (1 to 2) 1.29 (0.98 to 1.70) 0.068 a Invasive ventilation if positive pressure ventilation or high-frequency oscillatory ventilation was used via an endotracheal tube or a tracheostomy. b Bacterium/bacteria isolated from blood culture, bronchoalveolar lavage, sputum or histology. Table 3 Details of hospital stay by hospital outcome Characteristic All N (%) Survived N (%) Died N (%) OR (95% CI) p Value Duration in PICU (days) 4.0 (2.0 to 8.0) 4 (2.0 to 7.0) 5.0 (2.0 to 12) 1.00 (1.00 to 1.01) 0.053 Duration in hospital (days) 12.5 (7.9 to 28.0) 12.9 (8.1 to 27.0) 8.35 (2 to 48) 1.01 (1.00 to 1.02) 0.008 Invasive ventilationa 155 (58.4) 126 (81.3) 29 (18.7) 5.75 (2.24 to 14.98) <0.001 N (%) Inotropic support 42 (15.7) 18 (7.8) 23 (67.7) 24.7 (10.4 to 58.7) <0.001 Bacterial isolateb 27 (13.1) 19 (10.5) 8 (32.0) 4.01 (1.53 to 10.54) 0.005 Length of time to start antibiotics (h) 1 (1 to 3) 2 (1 to 3) 1 (1 to 2) 1.29 (0.98 to 1.70) 0.068 Characteristic All N (%) Survived N (%) Died N (%) OR (95% CI) p Value Duration in PICU (days) 4.0 (2.0 to 8.0) 4 (2.0 to 7.0) 5.0 (2.0 to 12) 1.00 (1.00 to 1.01) 0.053 Duration in hospital (days) 12.5 (7.9 to 28.0) 12.9 (8.1 to 27.0) 8.35 (2 to 48) 1.01 (1.00 to 1.02) 0.008 Invasive ventilationa 155 (58.4) 126 (81.3) 29 (18.7) 5.75 (2.24 to 14.98) <0.001 N (%) Inotropic support 42 (15.7) 18 (7.8) 23 (67.7) 24.7 (10.4 to 58.7) <0.001 Bacterial isolateb 27 (13.1) 19 (10.5) 8 (32.0) 4.01 (1.53 to 10.54) 0.005 Length of time to start antibiotics (h) 1 (1 to 3) 2 (1 to 3) 1 (1 to 2) 1.29 (0.98 to 1.70) 0.068 a Invasive ventilation if positive pressure ventilation or high-frequency oscillatory ventilation was used via an endotracheal tube or a tracheostomy. b Bacterium/bacteria isolated from blood culture, bronchoalveolar lavage, sputum or histology. The 265 patients used 1,544 PICU days, 21% of the annual available bed days. The mean PIM2 score was 0.134 (standard deviation = 0.192), indicating a risk of mortality on admission of 13.4%. In this study, the calculated SMR in the PICU and overall in-hospital following PICU discharge were 0.70 and 0.99, respectively. Causes of mortality in the PICU included multiorgan failure (n = 16), respiratory failure (n = 5) or cardiac failure (n = 3). Of the deaths that occurred in the ward following discharge from the PICU, one of the 10 was unexpected, related to a cardiac arrest caused by electrolyte disturbances. The other nine patients were discharged from the PICU for palliation due to severe hypoxic brain damage or significant irreversible lung disease. Risk factors for death On univariate analysis (Tables 1–3), the following factors were associated with mortality: severe malnutrition (OR = 4.99; 95% CI = 2.90–11.90), informal housing (OR = 2.87; 95% CI = 1.30–6.30), cardiac disease (OR = 2.89; 95% CI = 1.20–6.95), bacteria isolated from a normally sterile site (OR = 4.01; 95% CI = 1.53–10.54), need for inotropic support (OR = 24.7; 95% CI = 10.4–58.7), need for invasive ventilation (OR = 5.75; 95% CI = 2.24–14.98) and duration in hospital (OR = 1.01; 95% CI = 1.00–1.02). Formula feeding (OR = 0.31; 95% CI = 0.10–0.96) and mixed feeding (OR = 0.36; 95% CI = 0.14–0.94) were seen to be protective factors in this study. All of the above variables as well as age, prematurity, HIV status, time to first dose of antibiotics and duration in PICU were included in the multivariate analysis, as they were considered clinically important factors. In the multivariate analysis, severe malnutrition, informal housing, inotropic support and duration of hospital stay were associated with a higher risk of hospital mortality, after adjusting for HIV infection, prematurity, age and method of feeding (Table 4). Table 4 Multivariate analysis of risk factors for children admitted with a diagnosis of ALRTI, by hospital outcome (survived/died) Characteristic Univariate association Multivariate association OR (95% CI) p Value OR (95% CI) p Value Age category  <6 months 0.59 (0.29 to 1.21) 0.15 0.82 (0.19 to 3.56) 0.794 HIV status N (%)  Unexposed reference  Exposed, not infected 1.04 (0.44 to 2.45) 0.935 0.49 (0.03 to 8.05) 0.619  Exposed, infected 0.84 (0.23 to 3.02) 0.792 2.85 (0.55 to 14.86) 0.213 Preterm (<37 weeks) 0.87 (0.36 to 2.11) 0.759 1.62 (0.31 to 8.33) 0.565 Infant feeding N (%)  Exclusive breast-fed 4.38 (0.92 to 20.82) 0.063 Nutritional status  None/mild reference  Moderate 2.47 (0.83 to 7.36) 0.104 0.81 (0.12 to 5.32) 0.822  Severe 4.99 (2.90 to 11.90) <0.001 8.25 (1.47 to 46.2) 0.016 Housing  Formal reference  Informal 2.87 (1.30 to 6.30) 0.009 11.87 (1.89 to 20.8) 0.008 Bacterial isolate 4.01 (1.52 to 10.54) 0.005 3.61 (0.65 to 20.00) 0.141 No comorbid disease reference Cardiac diagnosis 2.89 (1.20 to 6.95) 0.018 1.05 (0.19 to 5.80) 0.957 Other 1.45 (0.59 to 3.59) 0.421 1.97 (0.30 to 12.81) 0.48 Inotropic support 24.7 (10.4 to 58.7) <0.001 44.35 (8.20 to 239.9) <0.001 Length of time to start antibiotics (h) 1.29 (0.98 to 1.70) 0.068 0.71 (0.37 to 1.35) 0.294 Duration in hospital 1.01 (1.00 to 1.02) 0.008 1.04 (1.01 to 1.06) 0.007 Duration in PICU 1.00 (1.00 to 1.01) 0.053 1.00 (0.99 to 1.00) 0.789 Characteristic Univariate association Multivariate association OR (95% CI) p Value OR (95% CI) p Value Age category  <6 months 0.59 (0.29 to 1.21) 0.15 0.82 (0.19 to 3.56) 0.794 HIV status N (%)  Unexposed reference  Exposed, not infected 1.04 (0.44 to 2.45) 0.935 0.49 (0.03 to 8.05) 0.619  Exposed, infected 0.84 (0.23 to 3.02) 0.792 2.85 (0.55 to 14.86) 0.213 Preterm (<37 weeks) 0.87 (0.36 to 2.11) 0.759 1.62 (0.31 to 8.33) 0.565 Infant feeding N (%)  Exclusive breast-fed 4.38 (0.92 to 20.82) 0.063 Nutritional status  None/mild reference  Moderate 2.47 (0.83 to 7.36) 0.104 0.81 (0.12 to 5.32) 0.822  Severe 4.99 (2.90 to 11.90) <0.001 8.25 (1.47 to 46.2) 0.016 Housing  Formal reference  Informal 2.87 (1.30 to 6.30) 0.009 11.87 (1.89 to 20.8) 0.008 Bacterial isolate 4.01 (1.52 to 10.54) 0.005 3.61 (0.65 to 20.00) 0.141 No comorbid disease reference Cardiac diagnosis 2.89 (1.20 to 6.95) 0.018 1.05 (0.19 to 5.80) 0.957 Other 1.45 (0.59 to 3.59) 0.421 1.97 (0.30 to 12.81) 0.48 Inotropic support 24.7 (10.4 to 58.7) <0.001 44.35 (8.20 to 239.9) <0.001 Length of time to start antibiotics (h) 1.29 (0.98 to 1.70) 0.068 0.71 (0.37 to 1.35) 0.294 Duration in hospital 1.01 (1.00 to 1.02) 0.008 1.04 (1.01 to 1.06) 0.007 Duration in PICU 1.00 (1.00 to 1.01) 0.053 1.00 (0.99 to 1.00) 0.789 Table 4 Multivariate analysis of risk factors for children admitted with a diagnosis of ALRTI, by hospital outcome (survived/died) Characteristic Univariate association Multivariate association OR (95% CI) p Value OR (95% CI) p Value Age category  <6 months 0.59 (0.29 to 1.21) 0.15 0.82 (0.19 to 3.56) 0.794 HIV status N (%)  Unexposed reference  Exposed, not infected 1.04 (0.44 to 2.45) 0.935 0.49 (0.03 to 8.05) 0.619  Exposed, infected 0.84 (0.23 to 3.02) 0.792 2.85 (0.55 to 14.86) 0.213 Preterm (<37 weeks) 0.87 (0.36 to 2.11) 0.759 1.62 (0.31 to 8.33) 0.565 Infant feeding N (%)  Exclusive breast-fed 4.38 (0.92 to 20.82) 0.063 Nutritional status  None/mild reference  Moderate 2.47 (0.83 to 7.36) 0.104 0.81 (0.12 to 5.32) 0.822  Severe 4.99 (2.90 to 11.90) <0.001 8.25 (1.47 to 46.2) 0.016 Housing  Formal reference  Informal 2.87 (1.30 to 6.30) 0.009 11.87 (1.89 to 20.8) 0.008 Bacterial isolate 4.01 (1.52 to 10.54) 0.005 3.61 (0.65 to 20.00) 0.141 No comorbid disease reference Cardiac diagnosis 2.89 (1.20 to 6.95) 0.018 1.05 (0.19 to 5.80) 0.957 Other 1.45 (0.59 to 3.59) 0.421 1.97 (0.30 to 12.81) 0.48 Inotropic support 24.7 (10.4 to 58.7) <0.001 44.35 (8.20 to 239.9) <0.001 Length of time to start antibiotics (h) 1.29 (0.98 to 1.70) 0.068 0.71 (0.37 to 1.35) 0.294 Duration in hospital 1.01 (1.00 to 1.02) 0.008 1.04 (1.01 to 1.06) 0.007 Duration in PICU 1.00 (1.00 to 1.01) 0.053 1.00 (0.99 to 1.00) 0.789 Characteristic Univariate association Multivariate association OR (95% CI) p Value OR (95% CI) p Value Age category  <6 months 0.59 (0.29 to 1.21) 0.15 0.82 (0.19 to 3.56) 0.794 HIV status N (%)  Unexposed reference  Exposed, not infected 1.04 (0.44 to 2.45) 0.935 0.49 (0.03 to 8.05) 0.619  Exposed, infected 0.84 (0.23 to 3.02) 0.792 2.85 (0.55 to 14.86) 0.213 Preterm (<37 weeks) 0.87 (0.36 to 2.11) 0.759 1.62 (0.31 to 8.33) 0.565 Infant feeding N (%)  Exclusive breast-fed 4.38 (0.92 to 20.82) 0.063 Nutritional status  None/mild reference  Moderate 2.47 (0.83 to 7.36) 0.104 0.81 (0.12 to 5.32) 0.822  Severe 4.99 (2.90 to 11.90) <0.001 8.25 (1.47 to 46.2) 0.016 Housing  Formal reference  Informal 2.87 (1.30 to 6.30) 0.009 11.87 (1.89 to 20.8) 0.008 Bacterial isolate 4.01 (1.52 to 10.54) 0.005 3.61 (0.65 to 20.00) 0.141 No comorbid disease reference Cardiac diagnosis 2.89 (1.20 to 6.95) 0.018 1.05 (0.19 to 5.80) 0.957 Other 1.45 (0.59 to 3.59) 0.421 1.97 (0.30 to 12.81) 0.48 Inotropic support 24.7 (10.4 to 58.7) <0.001 44.35 (8.20 to 239.9) <0.001 Length of time to start antibiotics (h) 1.29 (0.98 to 1.70) 0.068 0.71 (0.37 to 1.35) 0.294 Duration in hospital 1.01 (1.00 to 1.02) 0.008 1.04 (1.01 to 1.06) 0.007 Duration in PICU 1.00 (1.00 to 1.01) 0.053 1.00 (0.99 to 1.00) 0.789 Impact of HIV exposure and infection on outcomes In all, 87 (32.8%) patients were HIV-exposed and 27 (10.2%) were confirmed to be HIV-infected. Of the HIV-infected group, 22 (88%) patients were newly diagnosed on the current admission with a mean viral load of 2,660,254 (log 6.4), mean CD4 count of 23% and mean WHO clinical stage of 3. Of the five patients known to be HIV-infected, four were on first-line ART, and one was on second-line ART, but only two patients were virally suppressed. HIV exposure or infection did not significantly affect mortality but was associated with a higher PIM2 score on admission and significantly longer duration of PICU and hospital stay (Table 5). Table 5 Impact of HIV exposure/infection on outcomes Characteristic HIV-unexposed HIV-exposed, not infected HIV-exposed and infected p Value Survival rate 0.87 0.87 0.89 0.74 Case fatality rate in-hospital 0.13 (24/178) 0.13 (8/60) 0.11 (3/27) – Duration hospital stay in days (median/IQR) 11.0 (7.2 to 27.0) 11.0 (7.9 to 24.4) 23.0 (16.1 to 43.0) 0.002 Duration ICU stay in days (median/IQR) 4.0 (2.0 to 8.0) 4.0 (2.0 to 7.0) 6.0 (4.0 to 10.0) 0.131 PIM2 score (mean) 0.115 0.116 0.315 <0.001 Characteristic HIV-unexposed HIV-exposed, not infected HIV-exposed and infected p Value Survival rate 0.87 0.87 0.89 0.74 Case fatality rate in-hospital 0.13 (24/178) 0.13 (8/60) 0.11 (3/27) – Duration hospital stay in days (median/IQR) 11.0 (7.2 to 27.0) 11.0 (7.9 to 24.4) 23.0 (16.1 to 43.0) 0.002 Duration ICU stay in days (median/IQR) 4.0 (2.0 to 8.0) 4.0 (2.0 to 7.0) 6.0 (4.0 to 10.0) 0.131 PIM2 score (mean) 0.115 0.116 0.315 <0.001 Table 5 Impact of HIV exposure/infection on outcomes Characteristic HIV-unexposed HIV-exposed, not infected HIV-exposed and infected p Value Survival rate 0.87 0.87 0.89 0.74 Case fatality rate in-hospital 0.13 (24/178) 0.13 (8/60) 0.11 (3/27) – Duration hospital stay in days (median/IQR) 11.0 (7.2 to 27.0) 11.0 (7.9 to 24.4) 23.0 (16.1 to 43.0) 0.002 Duration ICU stay in days (median/IQR) 4.0 (2.0 to 8.0) 4.0 (2.0 to 7.0) 6.0 (4.0 to 10.0) 0.131 PIM2 score (mean) 0.115 0.116 0.315 <0.001 Characteristic HIV-unexposed HIV-exposed, not infected HIV-exposed and infected p Value Survival rate 0.87 0.87 0.89 0.74 Case fatality rate in-hospital 0.13 (24/178) 0.13 (8/60) 0.11 (3/27) – Duration hospital stay in days (median/IQR) 11.0 (7.2 to 27.0) 11.0 (7.9 to 24.4) 23.0 (16.1 to 43.0) 0.002 Duration ICU stay in days (median/IQR) 4.0 (2.0 to 8.0) 4.0 (2.0 to 7.0) 6.0 (4.0 to 10.0) 0.131 PIM2 score (mean) 0.115 0.116 0.315 <0.001 DISCUSSION This study provides data on risk factors for mortality and outcomes of South African children with severe ALRTI admitted to a PICU in the context of a widely available mother-to-child HIV preventive program [1, 6, 26]. The severity of illness in this population was extremely high, as evidenced by the PIM2 scores and requirement for mechanical ventilation. The proportion receiving mechanical ventilation (73%) was much higher than reported in other LMICs such as South America and China, where rates of mechanical ventilation in children admitted to PICU with severe ALRTI were 55% and 21%, respectively [8, 27]. The PIM2 score on admission to PICU predicted a mortality of 13%. The actual PICU mortality rate was 9%. This affirms the high quality of care and can be attributed to strategies including antibiotic stewardship, lung protective ventilation and preventative interventions to decrease the risk of complications. The 10 patients who died following discharge from PICU increased the overall in-hospital mortality to 13%, but all of these deaths, except one, were anticipated and palliative therapy was provided. These outcomes are especially encouraging in comparison with other LMICs such as South America or China who report PICU mortality rates from severe ALTRI of 13% and 5.8%, respectively [8, 27]. Several factors were identified as influencing mortality. Of these, socioeconomic factors were a key determinant. Children from impoverished backgrounds living in informal housing without electricity and running water had an almost four times increase in mortality. This finding is consistent with a recent study in this geographical area that reported that children from informal housing are more severely ill at admission, possibly reflecting higher prevalence of risk factors for illness as well as poorer access to health care, unnecessary delays in diagnosis or referral patterns [26]. This study did not show that exclusive breastfeeding was protective in reducing illness severity or mortality. This is in contrast to other studies. A possible explanation may be the very young age of children (median age 4 months), with almost half the deaths occurring before 6 months of age. Further, the high prevalence of comorbid disease may counteract the positive effect of breastfeeding. The proportion of HIV-exposed infants in this study sample was particularly high. HIV exposure in the infants as well as maternal HIV-related factors such as maternal ART and viral load may be important when assessing disease severity and outcomes. Exclusive breastfeeding may also have been inaccurately reported and/or recorded. HIV infection is recognized as a risk factor for severe disease and worse outcome from childhood ALRTI [7]. However, the mortality rate of HIV-infected children was 11%, substantially lower than that predicted by the PIM2 score (32%) and was much lower than previously reported in a study of ALRTI in PICU at RCH in 1998, which found a mortality of 29% [28]. However, the previous RCH study was done before the availability of ART for children. Other PICUs in South Africa report mortality rates for HIV-infected children of approximately 30% [29]. Early detection of HIV infection with timely initiation of ART, timely presentation to PICU, use of broad-spectrum antibiotics and effective ventilation strategies may explain the lower mortality in the current study. Such outcomes support the decision to accept HIV-infected children into the PICU for supportive treatment, which in the era before ART was perhaps not justified [30, 31]. The duration of PICU and hospital stay of HIV-infected children were double that of HIV-uninfected children, confirming that this is a vulnerable population at substantial risk for severe disease and who constitute a large burden on the health system. Recently, HIV exposure has also been associated with a higher risk of pneumonia. In a South African study, HIV-negative children born to HIV-infected mothers had a higher rate of pneumonia in the first year of life compared with those born to HIV-negative women [32]. As PMTCT programs are strengthened, so increasing attention must be given to HIV-exposed but uninfected children who represent an emerging, important vulnerable group. This study has several limitations. As the study was retrospective, there was limited ability to have complete data records and standardized measures of tests. The ability to attribute etiology was limited, as etiological investigations were highly variable and attributing etiology in childhood ALRTI may be especially difficult. The sample of patients in the PICU represent the sickest children; therefore, these results may not be generalizable to children with lower severity of illness. Further, the sample size was relatively small but representative of a full year of PICU admissions for ALRTI, thus accounting for seasonal variability. Finally, we were unable to assess long-term outcome, as children were only followed until discharge. In conclusion, children admitted for ALRTI to the PICU at RCH in 2012 had severe disease on admission but despite this had fairly good outcomes in terms of survival, which was better than predicted and improved from a decade earlier [7, 28]. Several risk factors for mortality were shown, which require broad interventions to effectively address at a public healthcare level. FUNDING Funding was received from a Department of Pediatrics and Child Health Research Award, Faculty of Health Sciences, University of Cape Town, and the South African Medical Research Council. REFERENCES 1 Sinha A , Kim S , Ginsberg G , et al. Economic burden of acute lower respiratory tract infection in South African children . Paediatr Int Child Health 2012 ; 32 : 65 – 73 . Google Scholar CrossRef Search ADS PubMed 2 Liu L , Oza S , Hogan D , et al. Global, regional, and national causes of child mortality in 2000–13, with projections to inform post-2015 priorities: an updated systematic analysis . Lancet 2015 ; 385 : 430 – 40 . Google Scholar CrossRef Search ADS PubMed 3 WHO . Causes of child mortality for the year 2010 [Internet]. WHO. http://www.who.int/gho/child_health/mortality/mortality_causes_text/en/index.html (17 September 2012, date last accessed). 4 Stats SA Library Cataloguing-in-Publication (CIP) Data . Mortality and causes of death in South Africa, 2014: findings from death notification/statistics South Africa . Pretoria : Statistics South Africa , 2015 . 5 Groenewald P , Msemburi W , Morden E , et al. Western Cape mortality profile 2011 . Cape Town : South African Medical Research Council , 2014 . ISBN: 978-1-920618-23-0. 6 Nair H , Simões EAF , Rudan I , et al. Global and regional burden of hospital admissions for severe acute lower respiratory infections in young children in 2010: a systematic analysis . Lancet 2013 ; 381 : 1380 – 90 . Google Scholar CrossRef Search ADS PubMed 7 Zar HJ. Pneumonia in HIV-infected children admitted to hospital in Cape Town, South Africa. PhD thesis. Paediatrics and Child Health, University of Cape Town, South Africa, 2000. 8 Zhang Q , Guo Z , Bai Z , et al. A 4-year prospective study to determine risk factors for severe community acquired pneumonia in children in Southern China . Pediatr Pulmonol 2013 ; 48 : 390 – 7 . Google Scholar CrossRef Search ADS PubMed 9 Simoes EA. Environmental and demographic risk factors for respiratory syncytial virus lower respiratory tract disease . J Pediatr 2003 ; 143 : 118 – 26 . Google Scholar CrossRef Search ADS 10 Delport SD , Brisley T. Aetiology and outcome of severe community-acquired pneumonia in children admitted to a pediatric intensive care unit . S Afr Med J 2002 ; 92 : 907 – 11 . Google Scholar PubMed 11 Weisman LE. Populations at risk for developing respiratory syncytial virus and risk factors for respiratory syncytial virus severity: infants with predisposing conditions . Pediatr Infect Dis J 2003 ; 22 : S33 – 9 . Google Scholar PubMed 12 Tiewsoh K , Lodha R , Pandey RM , et al. Factors determining the outcome of children hospitalized with severe pneumonia . BMC Pediatr 2009 ; 9 : 15 . Google Scholar CrossRef Search ADS PubMed 13 Figueras-Aloy J , Carbonell-Estrany X , Quero J. Case-control study of the risk factors linked to respiratory syncytial virus infection requiring hospitalization in premature infants born at a gestational age of 33-35 weeks in Spain . Pediatr Infect Dis J 2004 ; 23 : 815 – 20 . Google Scholar CrossRef Search ADS PubMed 14 Chisti MJ , Tebruegge M , La Vincente S , et al. Pneumonia in severely malnourished children in developing countries–mortality risk, aetiology and validity of WHO clinical signs: a systematic review . Trop Med Int Health 2009 ; 14 : 1173 – 89 . Google Scholar CrossRef Search ADS PubMed 15 Mohsin SS , Haque A , Shaikh AS , et al. Outcome of infants with unrepaired heart disease admitted to the Pediatric Intensive Care Unit: single-center developing country perspective . Congenit Heart Dis 2014 ; 9 : 116 – 21 . Google Scholar CrossRef Search ADS PubMed 16 Ghani ASA , Morrow BM , Hardie DR , et al. An investigation into the prevalence and outcome of patients admitted to a pediatric intensive care unit with viral respiratory tract infections in Cape Town, South Africa . Pediatr Crit Care Med 2012 ; 13 : e275 – 81 . Google Scholar CrossRef Search ADS PubMed 17 Mussi-Pinhata MM , Freimanis L , Yamamoto AY , et al. Infectious disease morbidity among young HIV-1-exposed but uninfected infants in Latin American and Caribbean countries: the National Institute of Child Health and Human Development International Site Development Initiative Perinatal Study . Pediatrics 2007 ; 119 : e694 – 704 . Google Scholar CrossRef Search ADS PubMed 18 Mussi-Pinhata MM , Motta F , Freimanis-Hance L , et al. Lower respiratory tract infections among human immunodeficiency virus-exposed, uninfected infants . Int J Infect Dis 2010 ; 14(Suppl 3) : e176 – 82 . Google Scholar CrossRef Search ADS PubMed 19 Onayade AA , Abiona TC , Abayomi IO , et al. The first six month growth and illness of exclusively and non-exclusively breast-fed infants in Nigeria . East Afr Med J 2004 ; 81 : 146 – 53 . Google Scholar CrossRef Search ADS PubMed 20 Khadivzadeh T , Parsai S. Effect of exclusive breastfeeding and complementary feeding on infant growth and morbidity . East Mediterr Health J 2004 ; 10 : 289 – 94 . http://apps.who.int/iris/handle/10665/119410 (3 February 2016, date last accessed). Google Scholar PubMed 21 Henkle E , Steinhoff MC , Omer SB , et al. The effect of exclusive breast-feeding on respiratory illness in young infants in a maternal immunization trial in Bangladesh . Pediatr Infect Dis J 2013 ; 32 : 431 – 5 . Google Scholar CrossRef Search ADS PubMed 22 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 , 312 . 23 Fenton R , Kim JH. A systematic review and meta-analysis to revise the Fenton growth chart for preterm infants . BMC Pediatr 2013 ; 13 : 59 . Google Scholar CrossRef Search ADS PubMed 24 Slater A , Shann F , Pearson G , et al. PIM2: a revised version of the Pediatric Index of Mortality . Intensive Care Med 2003 ; 29 : 278 – 85 . Google Scholar CrossRef Search ADS PubMed 25 Solomon LJ , Morrow BM , Argent AC. Pediatric index of mortality scores: an evaluation of function in the pediatric intensive care unit of the Red Cross War Memorial Children’s Hospital . S Afr J Crit Care 2014 ; 30 : 8 – 13 . Google Scholar CrossRef Search ADS 26 Hodkinson P , Argent A , Wallis L , et al. Pathways to care for critically ill or injured children: a cohort study from first presentation to healthcare services through to admission to intensive care or death . PLoS One 2016 ; 11 : e0145473 . Google Scholar CrossRef Search ADS PubMed 27 Farias JA , Fernández A , Monteverde E , et al. Mechanical ventilation in pediatric intensive care units during the season for acute lower respiratory infection . Pediatr Crit Care Med 2012 ; 13 : 158 – 64 . Google Scholar CrossRef Search ADS PubMed 28 Zar HJ , Apolles P , Argent A , et al. The etiology and outcome of pneumonia in human immunodeficiency virus-infected children admitted to intensive care in a developing country . Pediatr Crit Care Med 2001 ; 2 : 108 – 112 . Google Scholar CrossRef Search ADS PubMed 29 Kitchin OP , Masekela R , Becker P , et al. Outcome of human immunodeficiency virus–exposed and –infected children admitted to a pediatric intensive care unit for respiratory failure . Pediatr Crit Care Med 2012 ; 13 : 516 – 19 . Google Scholar CrossRef Search ADS PubMed 30 Argent AC. Managing HIV in the PICU–the experience at the Red Cross War Memorial Children’s Hospital in Cape Town . Indian J Pediatr 2008 ; 75 : 615 – 20 . Google Scholar CrossRef Search ADS PubMed 31 Jeena PM. Challenges in the provision of ICU services to HIV infected children in resource poor settings: a South African Case Study . J Med Ethics 2005 ; 31 : 226 – 30 . Google Scholar CrossRef Search ADS PubMed 32 Le Roux DM , Myer L , Nicol MP , et al. Incidence and severity of childhood pneumonia in the first year of life in a South African birth cohort: the Drakenstein Child Health Study . Lancet Glob Health 2015 ; 3 : e95 – 103 . Google Scholar CrossRef Search ADS PubMed © 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) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Tropical Pediatrics Oxford University Press

Clinical Features and Outcome of Children with Severe Lower Respiratory Tract Infection Admitted to a Pediatric Intensive Care Unit in South Africa

Loading next page...
 
/lp/ou_press/clinical-features-and-outcome-of-children-with-severe-lower-ptezAkUPqV
Publisher
Oxford University Press
Copyright
© The Author(s) [2018]. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com
ISSN
0142-6338
eISSN
1465-3664
D.O.I.
10.1093/tropej/fmy010
Publisher site
See Article on Publisher Site

Abstract

Abstract Aim Severe acute lower respiratory tract infection (ALRTI) remains an important cause of childhood morbidity and mortality. Methods This is a 12-month retrospective cohort study of children (0–12 years) admitted to a pediatric intensive care unit (PICU) with ALRTI to investigate risk factors, clinical course and in-hospital survival. Results In total, 265 patients (median age = 4 months [2–12 months]) were identified. In all,102 (38.5%) had co-morbid disease. Twenty-seven (10.2%) were HIV-infected and 87 (32.8%) were HIV-exposed. In-hospital mortality was 34 (12.8%)—24 (9.1%) in PICU and 10 in the wards. Median duration of intensive care unit was 4.0 days (2.0–8.0) and hospital stay was 12.5 days (7.9–28.0). In total, 192 (72.5%) children required invasive ventilation and 42 (15.8%) required inotropic support. Risk factors for mortality included severe malnutrition (odds ratio [OR] = 8.25; 95% confidence interval [CI] = 1.47–46.21), informal housing (OR = 11.87; CI = 1.89–20.81) or inotropic support (OR = 44.35; CI = 8.20–239.92). HIV exposure or infection was associated with a longer duration of hospital stay (OR = 4.41; CI = 2.44–6.39). Conclusion Severe ALRTI is associated with a high mortality with several factors impacting on in-hospital survival. respiratory tract infection, intensive care unit, pediatric, survival, risk factor INTRODUCTION Acute lower respiratory tract infection (ALRTI), primarily pneumonia, is a leading cause of death in children under the age of 5 years, placing a considerable burden on health and economic resources particularly in low- and middle-income countries (LMICs) [1]. The World Health Organization (WHO) recognizes that pneumonia alone is responsible for up to 15% of deaths in children [2]. In Africa, pneumonia is the most important single cause of under 5-year mortality outside the neonatal period [3], with a disproportionate number of cases of severe disease relative to the childhood population. In South Africa, pneumonia remains one of the leading causes of death in hospital in children under 5 years [4, 5]. There are limited data on hospital outcome of severe ALRTI in African children. WHO estimated 19.2 million cases of severe ALRTI in 2010, with only a proportion of those being treated in a hospital setting [6]. The in-hospital case fatality rate ranges from 2.3% to 6% depending on severity of disease [6]. A study done in 1998 at Red Cross War Memorial Children’s Hospital (RCH) of HIV-infected and uninfected children admitted to a pediatric intensive care unit (PICU) with ALRTI reported a mortality rate of 29% and 14%, respectively [7]. Importantly this study preceded the roll-out of highly active antiretroviral therapy (ART) for children. The limited African data suggest a much higher case fatality rate than that reported by the WHO [6] and highlights the need for further research into the causes, risk factors and outcome of severe ALRTI in this setting. Several factors may predispose children to ALRTI and to severe disease. Reported risk factors include age less than 1 year [8–10], prematurity, low birth weight [11–13], severe or moderate malnutrition [14], micronutrient deficiency or co-morbid disease (especially congenital cardiac disease [15]) and immunosuppressive disease (especially HIV infection [16]). Increasing evidence suggests that HIV-exposed and uninfected children have a higher risk for ALRTI or opportunistic infections than unexposed children [17, 18]. A protective factor identified in many studies is exclusive breastfeeding for 6 months [19–21]; protection wanes with age, but exclusive breastfeeding may halve the risk of ALRTI in infancy. The aim of this study was to describe the clinical characteristics and outcomes of children with severe ALRTI admitted to a PICU in a tertiary hospital in South Africa and to identify risk factors associated with in-hospital mortality. METHODS Study design and setting A retrospective study of patients admitted to the PICU at RCH, South Africa, with a diagnosis of ALRTI was conducted. RCH is a tertiary hospital with 300 beds offering comprehensive care to pediatric patients in the Western Cape. The PICU is a 20-bedded multidisciplinary unit. Children under 12 years admitted to PICU with a primary diagnosis of ALRTI from January 2012 to December 2012 were included. Children who developed a nosocomial ALRTI (defined as developing after 48 h of hospital admission) or those for whom the medical records were incomplete were excluded from the study. Data source The PICU database (Ethics approval number 139/2011) was used as the primary source of information. This is a database of all patients admitted to the PICU capturing demographic, clinical and socioeconomic details routinely required during a PICU admission. The database was used to identify patients with a primary diagnosis of lower respiratory tract disease using the following subheadings: ‘pneumonia’, ‘lower respiratory tract infection’, ‘bronchopneumonia’, ‘bronchiolitis’ and ‘pulmonary tuberculosis’. Medical records and National Health Laboratory records were used to gain further clinical information regarding risk factors and investigation results. Terminology used in this study Malnutrition was defined according to the WHO classification using z-score cut-offs of <−2 and <−3 for moderate and severe malnutrition, respectively [22]. Prematurity was defined as birth before 37 completed weeks of gestation. Birthweight was categorized as low birthweight (<2.5 kg), very low birth weight (1.00–1.49 kg) and extremely low birthweight (<1.0 kg). Fenton’s correction scale was used to adjust for gestational age in ex-premature patients when assessing birthweight and current weight in terms of z-scores [23]. Feeding data were obtained from maternal self-report. Exclusive breastfeeding, defined by WHO, is the exclusive use of breast milk for nutrition up until age 6 months or until hospital admission if younger than 6 months of age. Mixed feeding was defined as a mixture of breast and formula milk or any other oral substance for nutrition. Ventilatory support was considered noninvasive if nasal cannula, high flow nasal cannula oxygen or continuous positive airway pressure was used or invasive if positive pressure ventilation or high-frequency oscillatory ventilation was used via an endotracheal tube or a tracheostomy. Data on HIV status were confirmed using results accessed via the National Health Laboratory Services. HIV infection was defined as a child who was polymerase chain reaction (PCR)-positive if under 18 months or antibody-positive if older. HIV exposure was defined as a child who tested HIV-negative but whose mother was recorded as being HIV-infected. Data on prevention of mother-to-child transmission (PMTCT) strategies used were obtained from records. Data on microbiological etiology were obtained from culture of blood, induced sputum, tracheal aspirate, bronchoalveolar lavage or pleural fluid for bacteria and fungi. Viral PCR panel testing for seven respiratory viruses was performed on nasopharyngeal aspirates and bronchoalveolar lavage specimens. Fungal PCR testing was performed on bronchoalveolar lavage specimens and tracheal aspirates. Tuberculosis microscopy, culture and PCR were performed on induced sputum and gastric washings. Cytomegalovirus (CMV) was measured in HIV-infected children with a quantitative viral load of log >4 considered positive for disease. Outcomes measured included duration of PICU stay, duration of in-hospital stay and survival to discharge. The pediatric index of mortality score (PIM2) was used to predict mortality, as per standard practice in 2012. PIM2 is a composite score of clinical factors, biochemical markers, ventilatory requirements and underlying risk factors at the time of PICU admission that has been validated for use in LMIC [24, 25]. The standardized mortality rate (SMR) is calculated by dividing the actual mortality by the predicted mortality score. The SMR is a marker of the quality of care or performance in the unit. Statistical analysis Data were analyzed using STATA 13 statistical software (STATA Corporation, TX, USA). Exploratory data analysis of categorical and continuous variables included frequency tables and histograms to determine distribution. Simple descriptive statistics were used to characterize the study population. Statistical tests included the chi-square test and Kruskal–Wallis comparison of medians and Student’s t-test for comparison of means. Risk factors tending toward statistical significance (p < 0.26) or those chosen on clinical reasoning were included in a multivariate Cox regression analysis. Factors considered to be competing risk factors for mortality were excluded (duration of PICU stay and invasive ventilatory requirements). HIV infection and breastfeeding were treated as potential confounding variables. Statistical tests were two-sided at α = 0.05. Ethical considerations Ethical approval: University of Cape Town Faculty of Health Science Human Research Ethics Committee (HREC 435/2013). RESULTS In total, 358 patients were identified; 265 complete patient data sets were included for analysis (Fig. 1). A total of 265 patients (median [interquartile range, IQR] age = 4 months [2–12 months]) were admitted with ALRTI; 157 (59.3%) were male. Co-morbidities were present in 102 (38.5%) cases, cardiac disease in 42 (15.9%) and tuberculosis in 7 (6.4%). Whereas only 27 (10.2%) patients were HIV-infected, 87 (32.8%) were HIV-exposed. Baseline characteristics are described in Table 1. Table 1 Baseline characteristics of children admitted with ALRTI, by hospital outcome including univariate analysis of characteristic on hospital outcome Characteristic All N (%) Survived N (%) Died N (%) Odds ratio (95% CI) p Value Gender  N (%) male 157 (59.3) 133 (57.6) 24 (70.6) 1.76 (0.80 to 3.86) 0.153 Age in months  Median (IQR)a 4 (2 to 12) 4 (2 to 12) 6.5 (2.5 to 12) 1.00 (0.98 to 1.01) 1 Birthweight WAZb  Median (IQR) −0.8 (−1.4 to 0) −0.8 (−1.4 to 0.1) −0.9 (−1.5 to 0.5) 0.88 (0.63 to 1.23) 0.458 Preterm (<37 weeks) 60 (22.6) 53 (22.9) 7 (20.6) 0.87 (0.36 to 2.11) 0.759 HIV status N (%)  Unexposed 178 (67.2) 155 (67.1) 23 (67.7) Reference  Exposed, not infected 60 (22.6) 52 (22.5) 8 (23.5) 1.04 (0.44 to 2.45) 0.935  Exposed, infected 27 (10.2) 24 (10.4) 3 (8.8) 0.84 (0.23 to 3.02) 0.792 PMTCTcN (%)  Antenatal received 44 (50.6) 36 (47.4) 8 (72.7) 1.14 (0.90 to 1.43) 0.281  Perinatal received 50 (57.5) 41 (54.0) 9 (81.8) 1.14 (0.90 to 1.44) 0.285  Postnatal—correct 54 (62.1) 45 (59.2) 9 (81.8) 0.96 (0.73 to 1.27) 0.781  Postnatal—incorrect dose 18 (20.7) 17 (22.4) 1 (9.1) Infant feeding N (%)  Exclusive breast-fedd 121 (46.2) 99 (43.0) 22 (68.8) Reference  Exclusive formula 61 (23.3) 57 (24.8) 4 (12.5) 0.31 (0.10 to 0.96) 0.043  Mixed feeding 80 (30.5) 74 (32.2) 6 (18.8) 0.36 (0.14 to 0.94) 0.038 Nutritional status  None/mild 202 (76.5) 185 (80.1) 17 (51.5) Reference  Moderate 27 (10.2) 22 (9.5) 5 (15.2) 2.47 (0.83 to 7.36) 0.104  Severe 35 (13.3) 35 (13.3) 11 (33.1) 4.99 (2.90 to 11.9) <0.001 Income bracket  ≥R4,000 per month 20 (7.5) 18 (7.8) 2 (5.9) Reference  <R4,000 per month 245 (92.5) 213 (92.2) 32 (94.1) 1.35 (0.30 to 6.10) 0.695 Housing  Formal 136 (52.3) 126 (55.5) 10 (44.5) Reference  Informal 124 (47.7) 101 (30.3) 23 (69.7) 2.87 (1.30 to 6.30) 0.009 Tobacco smoke exposure (self-report) 59 (25.1) 56 (26.8) 3 (11.5) 0.36 (0.10 to 1.23) 0.103 TB disease 17 (6.4) 14 (6.1) 3 (8.8) 1.50 (0.41 to 5.52) 0.542 No comorbid disease 164 (61.9) 148 (64.1) 16 (47.1) Reference Cardiac diagnosis 42 (15.9) 32 (13.9) 10 (29.4) 2.89 (1.20 to 6.95) 0.018 Other 59 (22.3) 51 (22.1) 8 (23.5) 1.45 (0.59 to 3.59) 0.421 At least one dose of pneumococcal conjugate vaccine administered 180 (67.9) 155 (67.1) 25 (73.5) 1.36 (0.61 to 3.06) 0.455 Characteristic All N (%) Survived N (%) Died N (%) Odds ratio (95% CI) p Value Gender  N (%) male 157 (59.3) 133 (57.6) 24 (70.6) 1.76 (0.80 to 3.86) 0.153 Age in months  Median (IQR)a 4 (2 to 12) 4 (2 to 12) 6.5 (2.5 to 12) 1.00 (0.98 to 1.01) 1 Birthweight WAZb  Median (IQR) −0.8 (−1.4 to 0) −0.8 (−1.4 to 0.1) −0.9 (−1.5 to 0.5) 0.88 (0.63 to 1.23) 0.458 Preterm (<37 weeks) 60 (22.6) 53 (22.9) 7 (20.6) 0.87 (0.36 to 2.11) 0.759 HIV status N (%)  Unexposed 178 (67.2) 155 (67.1) 23 (67.7) Reference  Exposed, not infected 60 (22.6) 52 (22.5) 8 (23.5) 1.04 (0.44 to 2.45) 0.935  Exposed, infected 27 (10.2) 24 (10.4) 3 (8.8) 0.84 (0.23 to 3.02) 0.792 PMTCTcN (%)  Antenatal received 44 (50.6) 36 (47.4) 8 (72.7) 1.14 (0.90 to 1.43) 0.281  Perinatal received 50 (57.5) 41 (54.0) 9 (81.8) 1.14 (0.90 to 1.44) 0.285  Postnatal—correct 54 (62.1) 45 (59.2) 9 (81.8) 0.96 (0.73 to 1.27) 0.781  Postnatal—incorrect dose 18 (20.7) 17 (22.4) 1 (9.1) Infant feeding N (%)  Exclusive breast-fedd 121 (46.2) 99 (43.0) 22 (68.8) Reference  Exclusive formula 61 (23.3) 57 (24.8) 4 (12.5) 0.31 (0.10 to 0.96) 0.043  Mixed feeding 80 (30.5) 74 (32.2) 6 (18.8) 0.36 (0.14 to 0.94) 0.038 Nutritional status  None/mild 202 (76.5) 185 (80.1) 17 (51.5) Reference  Moderate 27 (10.2) 22 (9.5) 5 (15.2) 2.47 (0.83 to 7.36) 0.104  Severe 35 (13.3) 35 (13.3) 11 (33.1) 4.99 (2.90 to 11.9) <0.001 Income bracket  ≥R4,000 per month 20 (7.5) 18 (7.8) 2 (5.9) Reference  <R4,000 per month 245 (92.5) 213 (92.2) 32 (94.1) 1.35 (0.30 to 6.10) 0.695 Housing  Formal 136 (52.3) 126 (55.5) 10 (44.5) Reference  Informal 124 (47.7) 101 (30.3) 23 (69.7) 2.87 (1.30 to 6.30) 0.009 Tobacco smoke exposure (self-report) 59 (25.1) 56 (26.8) 3 (11.5) 0.36 (0.10 to 1.23) 0.103 TB disease 17 (6.4) 14 (6.1) 3 (8.8) 1.50 (0.41 to 5.52) 0.542 No comorbid disease 164 (61.9) 148 (64.1) 16 (47.1) Reference Cardiac diagnosis 42 (15.9) 32 (13.9) 10 (29.4) 2.89 (1.20 to 6.95) 0.018 Other 59 (22.3) 51 (22.1) 8 (23.5) 1.45 (0.59 to 3.59) 0.421 At least one dose of pneumococcal conjugate vaccine administered 180 (67.9) 155 (67.1) 25 (73.5) 1.36 (0.61 to 3.06) 0.455 a IQR = interquartile range. b WAZ = weight for age Z score (adjusted for gestational age using Fenton’s correctional scale). c PMTCT = prevention of mother-to-child transmission. d Exclusive breastfeeding from the time of birth until admission to hospital. Table 1 Baseline characteristics of children admitted with ALRTI, by hospital outcome including univariate analysis of characteristic on hospital outcome Characteristic All N (%) Survived N (%) Died N (%) Odds ratio (95% CI) p Value Gender  N (%) male 157 (59.3) 133 (57.6) 24 (70.6) 1.76 (0.80 to 3.86) 0.153 Age in months  Median (IQR)a 4 (2 to 12) 4 (2 to 12) 6.5 (2.5 to 12) 1.00 (0.98 to 1.01) 1 Birthweight WAZb  Median (IQR) −0.8 (−1.4 to 0) −0.8 (−1.4 to 0.1) −0.9 (−1.5 to 0.5) 0.88 (0.63 to 1.23) 0.458 Preterm (<37 weeks) 60 (22.6) 53 (22.9) 7 (20.6) 0.87 (0.36 to 2.11) 0.759 HIV status N (%)  Unexposed 178 (67.2) 155 (67.1) 23 (67.7) Reference  Exposed, not infected 60 (22.6) 52 (22.5) 8 (23.5) 1.04 (0.44 to 2.45) 0.935  Exposed, infected 27 (10.2) 24 (10.4) 3 (8.8) 0.84 (0.23 to 3.02) 0.792 PMTCTcN (%)  Antenatal received 44 (50.6) 36 (47.4) 8 (72.7) 1.14 (0.90 to 1.43) 0.281  Perinatal received 50 (57.5) 41 (54.0) 9 (81.8) 1.14 (0.90 to 1.44) 0.285  Postnatal—correct 54 (62.1) 45 (59.2) 9 (81.8) 0.96 (0.73 to 1.27) 0.781  Postnatal—incorrect dose 18 (20.7) 17 (22.4) 1 (9.1) Infant feeding N (%)  Exclusive breast-fedd 121 (46.2) 99 (43.0) 22 (68.8) Reference  Exclusive formula 61 (23.3) 57 (24.8) 4 (12.5) 0.31 (0.10 to 0.96) 0.043  Mixed feeding 80 (30.5) 74 (32.2) 6 (18.8) 0.36 (0.14 to 0.94) 0.038 Nutritional status  None/mild 202 (76.5) 185 (80.1) 17 (51.5) Reference  Moderate 27 (10.2) 22 (9.5) 5 (15.2) 2.47 (0.83 to 7.36) 0.104  Severe 35 (13.3) 35 (13.3) 11 (33.1) 4.99 (2.90 to 11.9) <0.001 Income bracket  ≥R4,000 per month 20 (7.5) 18 (7.8) 2 (5.9) Reference  <R4,000 per month 245 (92.5) 213 (92.2) 32 (94.1) 1.35 (0.30 to 6.10) 0.695 Housing  Formal 136 (52.3) 126 (55.5) 10 (44.5) Reference  Informal 124 (47.7) 101 (30.3) 23 (69.7) 2.87 (1.30 to 6.30) 0.009 Tobacco smoke exposure (self-report) 59 (25.1) 56 (26.8) 3 (11.5) 0.36 (0.10 to 1.23) 0.103 TB disease 17 (6.4) 14 (6.1) 3 (8.8) 1.50 (0.41 to 5.52) 0.542 No comorbid disease 164 (61.9) 148 (64.1) 16 (47.1) Reference Cardiac diagnosis 42 (15.9) 32 (13.9) 10 (29.4) 2.89 (1.20 to 6.95) 0.018 Other 59 (22.3) 51 (22.1) 8 (23.5) 1.45 (0.59 to 3.59) 0.421 At least one dose of pneumococcal conjugate vaccine administered 180 (67.9) 155 (67.1) 25 (73.5) 1.36 (0.61 to 3.06) 0.455 Characteristic All N (%) Survived N (%) Died N (%) Odds ratio (95% CI) p Value Gender  N (%) male 157 (59.3) 133 (57.6) 24 (70.6) 1.76 (0.80 to 3.86) 0.153 Age in months  Median (IQR)a 4 (2 to 12) 4 (2 to 12) 6.5 (2.5 to 12) 1.00 (0.98 to 1.01) 1 Birthweight WAZb  Median (IQR) −0.8 (−1.4 to 0) −0.8 (−1.4 to 0.1) −0.9 (−1.5 to 0.5) 0.88 (0.63 to 1.23) 0.458 Preterm (<37 weeks) 60 (22.6) 53 (22.9) 7 (20.6) 0.87 (0.36 to 2.11) 0.759 HIV status N (%)  Unexposed 178 (67.2) 155 (67.1) 23 (67.7) Reference  Exposed, not infected 60 (22.6) 52 (22.5) 8 (23.5) 1.04 (0.44 to 2.45) 0.935  Exposed, infected 27 (10.2) 24 (10.4) 3 (8.8) 0.84 (0.23 to 3.02) 0.792 PMTCTcN (%)  Antenatal received 44 (50.6) 36 (47.4) 8 (72.7) 1.14 (0.90 to 1.43) 0.281  Perinatal received 50 (57.5) 41 (54.0) 9 (81.8) 1.14 (0.90 to 1.44) 0.285  Postnatal—correct 54 (62.1) 45 (59.2) 9 (81.8) 0.96 (0.73 to 1.27) 0.781  Postnatal—incorrect dose 18 (20.7) 17 (22.4) 1 (9.1) Infant feeding N (%)  Exclusive breast-fedd 121 (46.2) 99 (43.0) 22 (68.8) Reference  Exclusive formula 61 (23.3) 57 (24.8) 4 (12.5) 0.31 (0.10 to 0.96) 0.043  Mixed feeding 80 (30.5) 74 (32.2) 6 (18.8) 0.36 (0.14 to 0.94) 0.038 Nutritional status  None/mild 202 (76.5) 185 (80.1) 17 (51.5) Reference  Moderate 27 (10.2) 22 (9.5) 5 (15.2) 2.47 (0.83 to 7.36) 0.104  Severe 35 (13.3) 35 (13.3) 11 (33.1) 4.99 (2.90 to 11.9) <0.001 Income bracket  ≥R4,000 per month 20 (7.5) 18 (7.8) 2 (5.9) Reference  <R4,000 per month 245 (92.5) 213 (92.2) 32 (94.1) 1.35 (0.30 to 6.10) 0.695 Housing  Formal 136 (52.3) 126 (55.5) 10 (44.5) Reference  Informal 124 (47.7) 101 (30.3) 23 (69.7) 2.87 (1.30 to 6.30) 0.009 Tobacco smoke exposure (self-report) 59 (25.1) 56 (26.8) 3 (11.5) 0.36 (0.10 to 1.23) 0.103 TB disease 17 (6.4) 14 (6.1) 3 (8.8) 1.50 (0.41 to 5.52) 0.542 No comorbid disease 164 (61.9) 148 (64.1) 16 (47.1) Reference Cardiac diagnosis 42 (15.9) 32 (13.9) 10 (29.4) 2.89 (1.20 to 6.95) 0.018 Other 59 (22.3) 51 (22.1) 8 (23.5) 1.45 (0.59 to 3.59) 0.421 At least one dose of pneumococcal conjugate vaccine administered 180 (67.9) 155 (67.1) 25 (73.5) 1.36 (0.61 to 3.06) 0.455 a IQR = interquartile range. b WAZ = weight for age Z score (adjusted for gestational age using Fenton’s correctional scale). c PMTCT = prevention of mother-to-child transmission. d Exclusive breastfeeding from the time of birth until admission to hospital. Fig. 1. View largeDownload slide Summary of patients included in the study. Fig. 1. View largeDownload slide Summary of patients included in the study. Microbiologic investigations Overall, 960 investigations were performed. In all, 206 (77.7%) patients had at least one organism identified, with 86 (32.5%) having more than one organism (Table 2). Viruses were most commonly identified (117 [44.2%]), especially respiratory syncytial virus (9%), rhinovirus (7%) and adenovirus (5%). Of 27 patients tested, 18 had a CMV log > 4 (66.7%). There was no difference in microbiological findings between HIV-unexposed and HIV-infected groups (odds ratio [OR] = 0.99; 95% confidence interval [CI] = 0.65–1.53). Table 2 Organisms isolated and associated risk of in-hospital mortality Organism isolated Number Proportion (%) Odds ratio (95% CI) p Value None 59 22.30 Bacteriaa  Single bacterium isolated 25 9.20 4.01 (1.53 to 10.57) 0.005  More than one bacteria isolated 2 1.00 Virusb  Single virus isolated 91 34.20 0.43 (0.15 to 1.21) 0.112   More than one virus isolated 26 10.00 Fungal (Pneumocystis jirovecii)c 4 1.50 Mixed infections  Bacterial/viral 45 17.00 1.07 (0.38 to 3.03) 0.885  Bacterial/fungal 3 1.10  Viral/fungal 7 2.60  Bacterial/viral/fungal 3 1.10 Total 203 100 Organism isolated Number Proportion (%) Odds ratio (95% CI) p Value None 59 22.30 Bacteriaa  Single bacterium isolated 25 9.20 4.01 (1.53 to 10.57) 0.005  More than one bacteria isolated 2 1.00 Virusb  Single virus isolated 91 34.20 0.43 (0.15 to 1.21) 0.112   More than one virus isolated 26 10.00 Fungal (Pneumocystis jirovecii)c 4 1.50 Mixed infections  Bacterial/viral 45 17.00 1.07 (0.38 to 3.03) 0.885  Bacterial/fungal 3 1.10  Viral/fungal 7 2.60  Bacterial/viral/fungal 3 1.10 Total 203 100 a Bacteria were isolated from blood culture, bronchoalveolar lavages, culture of tracheal aspirates or pleural fluid. b Viruses detected on polymerase chain reaction testing from nasopharyngeal samples or bronchoalveolar lavages. c No deaths in children who cultured fungal organisms. Table 2 Organisms isolated and associated risk of in-hospital mortality Organism isolated Number Proportion (%) Odds ratio (95% CI) p Value None 59 22.30 Bacteriaa  Single bacterium isolated 25 9.20 4.01 (1.53 to 10.57) 0.005  More than one bacteria isolated 2 1.00 Virusb  Single virus isolated 91 34.20 0.43 (0.15 to 1.21) 0.112   More than one virus isolated 26 10.00 Fungal (Pneumocystis jirovecii)c 4 1.50 Mixed infections  Bacterial/viral 45 17.00 1.07 (0.38 to 3.03) 0.885  Bacterial/fungal 3 1.10  Viral/fungal 7 2.60  Bacterial/viral/fungal 3 1.10 Total 203 100 Organism isolated Number Proportion (%) Odds ratio (95% CI) p Value None 59 22.30 Bacteriaa  Single bacterium isolated 25 9.20 4.01 (1.53 to 10.57) 0.005  More than one bacteria isolated 2 1.00 Virusb  Single virus isolated 91 34.20 0.43 (0.15 to 1.21) 0.112   More than one virus isolated 26 10.00 Fungal (Pneumocystis jirovecii)c 4 1.50 Mixed infections  Bacterial/viral 45 17.00 1.07 (0.38 to 3.03) 0.885  Bacterial/fungal 3 1.10  Viral/fungal 7 2.60  Bacterial/viral/fungal 3 1.10 Total 203 100 a Bacteria were isolated from blood culture, bronchoalveolar lavages, culture of tracheal aspirates or pleural fluid. b Viruses detected on polymerase chain reaction testing from nasopharyngeal samples or bronchoalveolar lavages. c No deaths in children who cultured fungal organisms. Children who had bacterium/bacteria isolated from blood culture, bronchoalveolar lavage, sputum or histology were four times more likely to die (OR = 4.01; 95% = CI 1.53–10.57). Outcomes Of 265 patients admitted to a PICU, 34 (12.8%) died in-hospital—24 (9.1%) patients died in the PICU and 10 in the medical wards following discharge (Table 3). Median duration of hospital stay was 12.5 days (7.9–28.0). Median duration of ICU stay was 4.0 days (2.0–8.0). Ventilatory support was required by 250 (94.3%) patients (mean duration of ventilatory support of 3.5 days [1.5–6.9]). Invasive ventilatory support (conventional mechanical ventilation or high-frequency oscillatory ventilation) was used in 192 (72.5%) patients. Inotropic support for hemodynamic instability was used in 42 (15.7%) patients. Table 3 Details of hospital stay by hospital outcome Characteristic All N (%) Survived N (%) Died N (%) OR (95% CI) p Value Duration in PICU (days) 4.0 (2.0 to 8.0) 4 (2.0 to 7.0) 5.0 (2.0 to 12) 1.00 (1.00 to 1.01) 0.053 Duration in hospital (days) 12.5 (7.9 to 28.0) 12.9 (8.1 to 27.0) 8.35 (2 to 48) 1.01 (1.00 to 1.02) 0.008 Invasive ventilationa 155 (58.4) 126 (81.3) 29 (18.7) 5.75 (2.24 to 14.98) <0.001 N (%) Inotropic support 42 (15.7) 18 (7.8) 23 (67.7) 24.7 (10.4 to 58.7) <0.001 Bacterial isolateb 27 (13.1) 19 (10.5) 8 (32.0) 4.01 (1.53 to 10.54) 0.005 Length of time to start antibiotics (h) 1 (1 to 3) 2 (1 to 3) 1 (1 to 2) 1.29 (0.98 to 1.70) 0.068 Characteristic All N (%) Survived N (%) Died N (%) OR (95% CI) p Value Duration in PICU (days) 4.0 (2.0 to 8.0) 4 (2.0 to 7.0) 5.0 (2.0 to 12) 1.00 (1.00 to 1.01) 0.053 Duration in hospital (days) 12.5 (7.9 to 28.0) 12.9 (8.1 to 27.0) 8.35 (2 to 48) 1.01 (1.00 to 1.02) 0.008 Invasive ventilationa 155 (58.4) 126 (81.3) 29 (18.7) 5.75 (2.24 to 14.98) <0.001 N (%) Inotropic support 42 (15.7) 18 (7.8) 23 (67.7) 24.7 (10.4 to 58.7) <0.001 Bacterial isolateb 27 (13.1) 19 (10.5) 8 (32.0) 4.01 (1.53 to 10.54) 0.005 Length of time to start antibiotics (h) 1 (1 to 3) 2 (1 to 3) 1 (1 to 2) 1.29 (0.98 to 1.70) 0.068 a Invasive ventilation if positive pressure ventilation or high-frequency oscillatory ventilation was used via an endotracheal tube or a tracheostomy. b Bacterium/bacteria isolated from blood culture, bronchoalveolar lavage, sputum or histology. Table 3 Details of hospital stay by hospital outcome Characteristic All N (%) Survived N (%) Died N (%) OR (95% CI) p Value Duration in PICU (days) 4.0 (2.0 to 8.0) 4 (2.0 to 7.0) 5.0 (2.0 to 12) 1.00 (1.00 to 1.01) 0.053 Duration in hospital (days) 12.5 (7.9 to 28.0) 12.9 (8.1 to 27.0) 8.35 (2 to 48) 1.01 (1.00 to 1.02) 0.008 Invasive ventilationa 155 (58.4) 126 (81.3) 29 (18.7) 5.75 (2.24 to 14.98) <0.001 N (%) Inotropic support 42 (15.7) 18 (7.8) 23 (67.7) 24.7 (10.4 to 58.7) <0.001 Bacterial isolateb 27 (13.1) 19 (10.5) 8 (32.0) 4.01 (1.53 to 10.54) 0.005 Length of time to start antibiotics (h) 1 (1 to 3) 2 (1 to 3) 1 (1 to 2) 1.29 (0.98 to 1.70) 0.068 Characteristic All N (%) Survived N (%) Died N (%) OR (95% CI) p Value Duration in PICU (days) 4.0 (2.0 to 8.0) 4 (2.0 to 7.0) 5.0 (2.0 to 12) 1.00 (1.00 to 1.01) 0.053 Duration in hospital (days) 12.5 (7.9 to 28.0) 12.9 (8.1 to 27.0) 8.35 (2 to 48) 1.01 (1.00 to 1.02) 0.008 Invasive ventilationa 155 (58.4) 126 (81.3) 29 (18.7) 5.75 (2.24 to 14.98) <0.001 N (%) Inotropic support 42 (15.7) 18 (7.8) 23 (67.7) 24.7 (10.4 to 58.7) <0.001 Bacterial isolateb 27 (13.1) 19 (10.5) 8 (32.0) 4.01 (1.53 to 10.54) 0.005 Length of time to start antibiotics (h) 1 (1 to 3) 2 (1 to 3) 1 (1 to 2) 1.29 (0.98 to 1.70) 0.068 a Invasive ventilation if positive pressure ventilation or high-frequency oscillatory ventilation was used via an endotracheal tube or a tracheostomy. b Bacterium/bacteria isolated from blood culture, bronchoalveolar lavage, sputum or histology. The 265 patients used 1,544 PICU days, 21% of the annual available bed days. The mean PIM2 score was 0.134 (standard deviation = 0.192), indicating a risk of mortality on admission of 13.4%. In this study, the calculated SMR in the PICU and overall in-hospital following PICU discharge were 0.70 and 0.99, respectively. Causes of mortality in the PICU included multiorgan failure (n = 16), respiratory failure (n = 5) or cardiac failure (n = 3). Of the deaths that occurred in the ward following discharge from the PICU, one of the 10 was unexpected, related to a cardiac arrest caused by electrolyte disturbances. The other nine patients were discharged from the PICU for palliation due to severe hypoxic brain damage or significant irreversible lung disease. Risk factors for death On univariate analysis (Tables 1–3), the following factors were associated with mortality: severe malnutrition (OR = 4.99; 95% CI = 2.90–11.90), informal housing (OR = 2.87; 95% CI = 1.30–6.30), cardiac disease (OR = 2.89; 95% CI = 1.20–6.95), bacteria isolated from a normally sterile site (OR = 4.01; 95% CI = 1.53–10.54), need for inotropic support (OR = 24.7; 95% CI = 10.4–58.7), need for invasive ventilation (OR = 5.75; 95% CI = 2.24–14.98) and duration in hospital (OR = 1.01; 95% CI = 1.00–1.02). Formula feeding (OR = 0.31; 95% CI = 0.10–0.96) and mixed feeding (OR = 0.36; 95% CI = 0.14–0.94) were seen to be protective factors in this study. All of the above variables as well as age, prematurity, HIV status, time to first dose of antibiotics and duration in PICU were included in the multivariate analysis, as they were considered clinically important factors. In the multivariate analysis, severe malnutrition, informal housing, inotropic support and duration of hospital stay were associated with a higher risk of hospital mortality, after adjusting for HIV infection, prematurity, age and method of feeding (Table 4). Table 4 Multivariate analysis of risk factors for children admitted with a diagnosis of ALRTI, by hospital outcome (survived/died) Characteristic Univariate association Multivariate association OR (95% CI) p Value OR (95% CI) p Value Age category  <6 months 0.59 (0.29 to 1.21) 0.15 0.82 (0.19 to 3.56) 0.794 HIV status N (%)  Unexposed reference  Exposed, not infected 1.04 (0.44 to 2.45) 0.935 0.49 (0.03 to 8.05) 0.619  Exposed, infected 0.84 (0.23 to 3.02) 0.792 2.85 (0.55 to 14.86) 0.213 Preterm (<37 weeks) 0.87 (0.36 to 2.11) 0.759 1.62 (0.31 to 8.33) 0.565 Infant feeding N (%)  Exclusive breast-fed 4.38 (0.92 to 20.82) 0.063 Nutritional status  None/mild reference  Moderate 2.47 (0.83 to 7.36) 0.104 0.81 (0.12 to 5.32) 0.822  Severe 4.99 (2.90 to 11.90) <0.001 8.25 (1.47 to 46.2) 0.016 Housing  Formal reference  Informal 2.87 (1.30 to 6.30) 0.009 11.87 (1.89 to 20.8) 0.008 Bacterial isolate 4.01 (1.52 to 10.54) 0.005 3.61 (0.65 to 20.00) 0.141 No comorbid disease reference Cardiac diagnosis 2.89 (1.20 to 6.95) 0.018 1.05 (0.19 to 5.80) 0.957 Other 1.45 (0.59 to 3.59) 0.421 1.97 (0.30 to 12.81) 0.48 Inotropic support 24.7 (10.4 to 58.7) <0.001 44.35 (8.20 to 239.9) <0.001 Length of time to start antibiotics (h) 1.29 (0.98 to 1.70) 0.068 0.71 (0.37 to 1.35) 0.294 Duration in hospital 1.01 (1.00 to 1.02) 0.008 1.04 (1.01 to 1.06) 0.007 Duration in PICU 1.00 (1.00 to 1.01) 0.053 1.00 (0.99 to 1.00) 0.789 Characteristic Univariate association Multivariate association OR (95% CI) p Value OR (95% CI) p Value Age category  <6 months 0.59 (0.29 to 1.21) 0.15 0.82 (0.19 to 3.56) 0.794 HIV status N (%)  Unexposed reference  Exposed, not infected 1.04 (0.44 to 2.45) 0.935 0.49 (0.03 to 8.05) 0.619  Exposed, infected 0.84 (0.23 to 3.02) 0.792 2.85 (0.55 to 14.86) 0.213 Preterm (<37 weeks) 0.87 (0.36 to 2.11) 0.759 1.62 (0.31 to 8.33) 0.565 Infant feeding N (%)  Exclusive breast-fed 4.38 (0.92 to 20.82) 0.063 Nutritional status  None/mild reference  Moderate 2.47 (0.83 to 7.36) 0.104 0.81 (0.12 to 5.32) 0.822  Severe 4.99 (2.90 to 11.90) <0.001 8.25 (1.47 to 46.2) 0.016 Housing  Formal reference  Informal 2.87 (1.30 to 6.30) 0.009 11.87 (1.89 to 20.8) 0.008 Bacterial isolate 4.01 (1.52 to 10.54) 0.005 3.61 (0.65 to 20.00) 0.141 No comorbid disease reference Cardiac diagnosis 2.89 (1.20 to 6.95) 0.018 1.05 (0.19 to 5.80) 0.957 Other 1.45 (0.59 to 3.59) 0.421 1.97 (0.30 to 12.81) 0.48 Inotropic support 24.7 (10.4 to 58.7) <0.001 44.35 (8.20 to 239.9) <0.001 Length of time to start antibiotics (h) 1.29 (0.98 to 1.70) 0.068 0.71 (0.37 to 1.35) 0.294 Duration in hospital 1.01 (1.00 to 1.02) 0.008 1.04 (1.01 to 1.06) 0.007 Duration in PICU 1.00 (1.00 to 1.01) 0.053 1.00 (0.99 to 1.00) 0.789 Table 4 Multivariate analysis of risk factors for children admitted with a diagnosis of ALRTI, by hospital outcome (survived/died) Characteristic Univariate association Multivariate association OR (95% CI) p Value OR (95% CI) p Value Age category  <6 months 0.59 (0.29 to 1.21) 0.15 0.82 (0.19 to 3.56) 0.794 HIV status N (%)  Unexposed reference  Exposed, not infected 1.04 (0.44 to 2.45) 0.935 0.49 (0.03 to 8.05) 0.619  Exposed, infected 0.84 (0.23 to 3.02) 0.792 2.85 (0.55 to 14.86) 0.213 Preterm (<37 weeks) 0.87 (0.36 to 2.11) 0.759 1.62 (0.31 to 8.33) 0.565 Infant feeding N (%)  Exclusive breast-fed 4.38 (0.92 to 20.82) 0.063 Nutritional status  None/mild reference  Moderate 2.47 (0.83 to 7.36) 0.104 0.81 (0.12 to 5.32) 0.822  Severe 4.99 (2.90 to 11.90) <0.001 8.25 (1.47 to 46.2) 0.016 Housing  Formal reference  Informal 2.87 (1.30 to 6.30) 0.009 11.87 (1.89 to 20.8) 0.008 Bacterial isolate 4.01 (1.52 to 10.54) 0.005 3.61 (0.65 to 20.00) 0.141 No comorbid disease reference Cardiac diagnosis 2.89 (1.20 to 6.95) 0.018 1.05 (0.19 to 5.80) 0.957 Other 1.45 (0.59 to 3.59) 0.421 1.97 (0.30 to 12.81) 0.48 Inotropic support 24.7 (10.4 to 58.7) <0.001 44.35 (8.20 to 239.9) <0.001 Length of time to start antibiotics (h) 1.29 (0.98 to 1.70) 0.068 0.71 (0.37 to 1.35) 0.294 Duration in hospital 1.01 (1.00 to 1.02) 0.008 1.04 (1.01 to 1.06) 0.007 Duration in PICU 1.00 (1.00 to 1.01) 0.053 1.00 (0.99 to 1.00) 0.789 Characteristic Univariate association Multivariate association OR (95% CI) p Value OR (95% CI) p Value Age category  <6 months 0.59 (0.29 to 1.21) 0.15 0.82 (0.19 to 3.56) 0.794 HIV status N (%)  Unexposed reference  Exposed, not infected 1.04 (0.44 to 2.45) 0.935 0.49 (0.03 to 8.05) 0.619  Exposed, infected 0.84 (0.23 to 3.02) 0.792 2.85 (0.55 to 14.86) 0.213 Preterm (<37 weeks) 0.87 (0.36 to 2.11) 0.759 1.62 (0.31 to 8.33) 0.565 Infant feeding N (%)  Exclusive breast-fed 4.38 (0.92 to 20.82) 0.063 Nutritional status  None/mild reference  Moderate 2.47 (0.83 to 7.36) 0.104 0.81 (0.12 to 5.32) 0.822  Severe 4.99 (2.90 to 11.90) <0.001 8.25 (1.47 to 46.2) 0.016 Housing  Formal reference  Informal 2.87 (1.30 to 6.30) 0.009 11.87 (1.89 to 20.8) 0.008 Bacterial isolate 4.01 (1.52 to 10.54) 0.005 3.61 (0.65 to 20.00) 0.141 No comorbid disease reference Cardiac diagnosis 2.89 (1.20 to 6.95) 0.018 1.05 (0.19 to 5.80) 0.957 Other 1.45 (0.59 to 3.59) 0.421 1.97 (0.30 to 12.81) 0.48 Inotropic support 24.7 (10.4 to 58.7) <0.001 44.35 (8.20 to 239.9) <0.001 Length of time to start antibiotics (h) 1.29 (0.98 to 1.70) 0.068 0.71 (0.37 to 1.35) 0.294 Duration in hospital 1.01 (1.00 to 1.02) 0.008 1.04 (1.01 to 1.06) 0.007 Duration in PICU 1.00 (1.00 to 1.01) 0.053 1.00 (0.99 to 1.00) 0.789 Impact of HIV exposure and infection on outcomes In all, 87 (32.8%) patients were HIV-exposed and 27 (10.2%) were confirmed to be HIV-infected. Of the HIV-infected group, 22 (88%) patients were newly diagnosed on the current admission with a mean viral load of 2,660,254 (log 6.4), mean CD4 count of 23% and mean WHO clinical stage of 3. Of the five patients known to be HIV-infected, four were on first-line ART, and one was on second-line ART, but only two patients were virally suppressed. HIV exposure or infection did not significantly affect mortality but was associated with a higher PIM2 score on admission and significantly longer duration of PICU and hospital stay (Table 5). Table 5 Impact of HIV exposure/infection on outcomes Characteristic HIV-unexposed HIV-exposed, not infected HIV-exposed and infected p Value Survival rate 0.87 0.87 0.89 0.74 Case fatality rate in-hospital 0.13 (24/178) 0.13 (8/60) 0.11 (3/27) – Duration hospital stay in days (median/IQR) 11.0 (7.2 to 27.0) 11.0 (7.9 to 24.4) 23.0 (16.1 to 43.0) 0.002 Duration ICU stay in days (median/IQR) 4.0 (2.0 to 8.0) 4.0 (2.0 to 7.0) 6.0 (4.0 to 10.0) 0.131 PIM2 score (mean) 0.115 0.116 0.315 <0.001 Characteristic HIV-unexposed HIV-exposed, not infected HIV-exposed and infected p Value Survival rate 0.87 0.87 0.89 0.74 Case fatality rate in-hospital 0.13 (24/178) 0.13 (8/60) 0.11 (3/27) – Duration hospital stay in days (median/IQR) 11.0 (7.2 to 27.0) 11.0 (7.9 to 24.4) 23.0 (16.1 to 43.0) 0.002 Duration ICU stay in days (median/IQR) 4.0 (2.0 to 8.0) 4.0 (2.0 to 7.0) 6.0 (4.0 to 10.0) 0.131 PIM2 score (mean) 0.115 0.116 0.315 <0.001 Table 5 Impact of HIV exposure/infection on outcomes Characteristic HIV-unexposed HIV-exposed, not infected HIV-exposed and infected p Value Survival rate 0.87 0.87 0.89 0.74 Case fatality rate in-hospital 0.13 (24/178) 0.13 (8/60) 0.11 (3/27) – Duration hospital stay in days (median/IQR) 11.0 (7.2 to 27.0) 11.0 (7.9 to 24.4) 23.0 (16.1 to 43.0) 0.002 Duration ICU stay in days (median/IQR) 4.0 (2.0 to 8.0) 4.0 (2.0 to 7.0) 6.0 (4.0 to 10.0) 0.131 PIM2 score (mean) 0.115 0.116 0.315 <0.001 Characteristic HIV-unexposed HIV-exposed, not infected HIV-exposed and infected p Value Survival rate 0.87 0.87 0.89 0.74 Case fatality rate in-hospital 0.13 (24/178) 0.13 (8/60) 0.11 (3/27) – Duration hospital stay in days (median/IQR) 11.0 (7.2 to 27.0) 11.0 (7.9 to 24.4) 23.0 (16.1 to 43.0) 0.002 Duration ICU stay in days (median/IQR) 4.0 (2.0 to 8.0) 4.0 (2.0 to 7.0) 6.0 (4.0 to 10.0) 0.131 PIM2 score (mean) 0.115 0.116 0.315 <0.001 DISCUSSION This study provides data on risk factors for mortality and outcomes of South African children with severe ALRTI admitted to a PICU in the context of a widely available mother-to-child HIV preventive program [1, 6, 26]. The severity of illness in this population was extremely high, as evidenced by the PIM2 scores and requirement for mechanical ventilation. The proportion receiving mechanical ventilation (73%) was much higher than reported in other LMICs such as South America and China, where rates of mechanical ventilation in children admitted to PICU with severe ALRTI were 55% and 21%, respectively [8, 27]. The PIM2 score on admission to PICU predicted a mortality of 13%. The actual PICU mortality rate was 9%. This affirms the high quality of care and can be attributed to strategies including antibiotic stewardship, lung protective ventilation and preventative interventions to decrease the risk of complications. The 10 patients who died following discharge from PICU increased the overall in-hospital mortality to 13%, but all of these deaths, except one, were anticipated and palliative therapy was provided. These outcomes are especially encouraging in comparison with other LMICs such as South America or China who report PICU mortality rates from severe ALTRI of 13% and 5.8%, respectively [8, 27]. Several factors were identified as influencing mortality. Of these, socioeconomic factors were a key determinant. Children from impoverished backgrounds living in informal housing without electricity and running water had an almost four times increase in mortality. This finding is consistent with a recent study in this geographical area that reported that children from informal housing are more severely ill at admission, possibly reflecting higher prevalence of risk factors for illness as well as poorer access to health care, unnecessary delays in diagnosis or referral patterns [26]. This study did not show that exclusive breastfeeding was protective in reducing illness severity or mortality. This is in contrast to other studies. A possible explanation may be the very young age of children (median age 4 months), with almost half the deaths occurring before 6 months of age. Further, the high prevalence of comorbid disease may counteract the positive effect of breastfeeding. The proportion of HIV-exposed infants in this study sample was particularly high. HIV exposure in the infants as well as maternal HIV-related factors such as maternal ART and viral load may be important when assessing disease severity and outcomes. Exclusive breastfeeding may also have been inaccurately reported and/or recorded. HIV infection is recognized as a risk factor for severe disease and worse outcome from childhood ALRTI [7]. However, the mortality rate of HIV-infected children was 11%, substantially lower than that predicted by the PIM2 score (32%) and was much lower than previously reported in a study of ALRTI in PICU at RCH in 1998, which found a mortality of 29% [28]. However, the previous RCH study was done before the availability of ART for children. Other PICUs in South Africa report mortality rates for HIV-infected children of approximately 30% [29]. Early detection of HIV infection with timely initiation of ART, timely presentation to PICU, use of broad-spectrum antibiotics and effective ventilation strategies may explain the lower mortality in the current study. Such outcomes support the decision to accept HIV-infected children into the PICU for supportive treatment, which in the era before ART was perhaps not justified [30, 31]. The duration of PICU and hospital stay of HIV-infected children were double that of HIV-uninfected children, confirming that this is a vulnerable population at substantial risk for severe disease and who constitute a large burden on the health system. Recently, HIV exposure has also been associated with a higher risk of pneumonia. In a South African study, HIV-negative children born to HIV-infected mothers had a higher rate of pneumonia in the first year of life compared with those born to HIV-negative women [32]. As PMTCT programs are strengthened, so increasing attention must be given to HIV-exposed but uninfected children who represent an emerging, important vulnerable group. This study has several limitations. As the study was retrospective, there was limited ability to have complete data records and standardized measures of tests. The ability to attribute etiology was limited, as etiological investigations were highly variable and attributing etiology in childhood ALRTI may be especially difficult. The sample of patients in the PICU represent the sickest children; therefore, these results may not be generalizable to children with lower severity of illness. Further, the sample size was relatively small but representative of a full year of PICU admissions for ALRTI, thus accounting for seasonal variability. Finally, we were unable to assess long-term outcome, as children were only followed until discharge. In conclusion, children admitted for ALRTI to the PICU at RCH in 2012 had severe disease on admission but despite this had fairly good outcomes in terms of survival, which was better than predicted and improved from a decade earlier [7, 28]. Several risk factors for mortality were shown, which require broad interventions to effectively address at a public healthcare level. FUNDING Funding was received from a Department of Pediatrics and Child Health Research Award, Faculty of Health Sciences, University of Cape Town, and the South African Medical Research Council. REFERENCES 1 Sinha A , Kim S , Ginsberg G , et al. Economic burden of acute lower respiratory tract infection in South African children . Paediatr Int Child Health 2012 ; 32 : 65 – 73 . Google Scholar CrossRef Search ADS PubMed 2 Liu L , Oza S , Hogan D , et al. Global, regional, and national causes of child mortality in 2000–13, with projections to inform post-2015 priorities: an updated systematic analysis . Lancet 2015 ; 385 : 430 – 40 . Google Scholar CrossRef Search ADS PubMed 3 WHO . Causes of child mortality for the year 2010 [Internet]. WHO. http://www.who.int/gho/child_health/mortality/mortality_causes_text/en/index.html (17 September 2012, date last accessed). 4 Stats SA Library Cataloguing-in-Publication (CIP) Data . Mortality and causes of death in South Africa, 2014: findings from death notification/statistics South Africa . Pretoria : Statistics South Africa , 2015 . 5 Groenewald P , Msemburi W , Morden E , et al. Western Cape mortality profile 2011 . Cape Town : South African Medical Research Council , 2014 . ISBN: 978-1-920618-23-0. 6 Nair H , Simões EAF , Rudan I , et al. Global and regional burden of hospital admissions for severe acute lower respiratory infections in young children in 2010: a systematic analysis . Lancet 2013 ; 381 : 1380 – 90 . Google Scholar CrossRef Search ADS PubMed 7 Zar HJ. Pneumonia in HIV-infected children admitted to hospital in Cape Town, South Africa. PhD thesis. Paediatrics and Child Health, University of Cape Town, South Africa, 2000. 8 Zhang Q , Guo Z , Bai Z , et al. A 4-year prospective study to determine risk factors for severe community acquired pneumonia in children in Southern China . Pediatr Pulmonol 2013 ; 48 : 390 – 7 . Google Scholar CrossRef Search ADS PubMed 9 Simoes EA. Environmental and demographic risk factors for respiratory syncytial virus lower respiratory tract disease . J Pediatr 2003 ; 143 : 118 – 26 . Google Scholar CrossRef Search ADS 10 Delport SD , Brisley T. Aetiology and outcome of severe community-acquired pneumonia in children admitted to a pediatric intensive care unit . S Afr Med J 2002 ; 92 : 907 – 11 . Google Scholar PubMed 11 Weisman LE. Populations at risk for developing respiratory syncytial virus and risk factors for respiratory syncytial virus severity: infants with predisposing conditions . Pediatr Infect Dis J 2003 ; 22 : S33 – 9 . Google Scholar PubMed 12 Tiewsoh K , Lodha R , Pandey RM , et al. Factors determining the outcome of children hospitalized with severe pneumonia . BMC Pediatr 2009 ; 9 : 15 . Google Scholar CrossRef Search ADS PubMed 13 Figueras-Aloy J , Carbonell-Estrany X , Quero J. Case-control study of the risk factors linked to respiratory syncytial virus infection requiring hospitalization in premature infants born at a gestational age of 33-35 weeks in Spain . Pediatr Infect Dis J 2004 ; 23 : 815 – 20 . Google Scholar CrossRef Search ADS PubMed 14 Chisti MJ , Tebruegge M , La Vincente S , et al. Pneumonia in severely malnourished children in developing countries–mortality risk, aetiology and validity of WHO clinical signs: a systematic review . Trop Med Int Health 2009 ; 14 : 1173 – 89 . Google Scholar CrossRef Search ADS PubMed 15 Mohsin SS , Haque A , Shaikh AS , et al. Outcome of infants with unrepaired heart disease admitted to the Pediatric Intensive Care Unit: single-center developing country perspective . Congenit Heart Dis 2014 ; 9 : 116 – 21 . Google Scholar CrossRef Search ADS PubMed 16 Ghani ASA , Morrow BM , Hardie DR , et al. An investigation into the prevalence and outcome of patients admitted to a pediatric intensive care unit with viral respiratory tract infections in Cape Town, South Africa . Pediatr Crit Care Med 2012 ; 13 : e275 – 81 . Google Scholar CrossRef Search ADS PubMed 17 Mussi-Pinhata MM , Freimanis L , Yamamoto AY , et al. Infectious disease morbidity among young HIV-1-exposed but uninfected infants in Latin American and Caribbean countries: the National Institute of Child Health and Human Development International Site Development Initiative Perinatal Study . Pediatrics 2007 ; 119 : e694 – 704 . Google Scholar CrossRef Search ADS PubMed 18 Mussi-Pinhata MM , Motta F , Freimanis-Hance L , et al. Lower respiratory tract infections among human immunodeficiency virus-exposed, uninfected infants . Int J Infect Dis 2010 ; 14(Suppl 3) : e176 – 82 . Google Scholar CrossRef Search ADS PubMed 19 Onayade AA , Abiona TC , Abayomi IO , et al. The first six month growth and illness of exclusively and non-exclusively breast-fed infants in Nigeria . East Afr Med J 2004 ; 81 : 146 – 53 . Google Scholar CrossRef Search ADS PubMed 20 Khadivzadeh T , Parsai S. Effect of exclusive breastfeeding and complementary feeding on infant growth and morbidity . East Mediterr Health J 2004 ; 10 : 289 – 94 . http://apps.who.int/iris/handle/10665/119410 (3 February 2016, date last accessed). Google Scholar PubMed 21 Henkle E , Steinhoff MC , Omer SB , et al. The effect of exclusive breast-feeding on respiratory illness in young infants in a maternal immunization trial in Bangladesh . Pediatr Infect Dis J 2013 ; 32 : 431 – 5 . Google Scholar CrossRef Search ADS PubMed 22 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 , 312 . 23 Fenton R , Kim JH. A systematic review and meta-analysis to revise the Fenton growth chart for preterm infants . BMC Pediatr 2013 ; 13 : 59 . Google Scholar CrossRef Search ADS PubMed 24 Slater A , Shann F , Pearson G , et al. PIM2: a revised version of the Pediatric Index of Mortality . Intensive Care Med 2003 ; 29 : 278 – 85 . Google Scholar CrossRef Search ADS PubMed 25 Solomon LJ , Morrow BM , Argent AC. Pediatric index of mortality scores: an evaluation of function in the pediatric intensive care unit of the Red Cross War Memorial Children’s Hospital . S Afr J Crit Care 2014 ; 30 : 8 – 13 . Google Scholar CrossRef Search ADS 26 Hodkinson P , Argent A , Wallis L , et al. Pathways to care for critically ill or injured children: a cohort study from first presentation to healthcare services through to admission to intensive care or death . PLoS One 2016 ; 11 : e0145473 . Google Scholar CrossRef Search ADS PubMed 27 Farias JA , Fernández A , Monteverde E , et al. Mechanical ventilation in pediatric intensive care units during the season for acute lower respiratory infection . Pediatr Crit Care Med 2012 ; 13 : 158 – 64 . Google Scholar CrossRef Search ADS PubMed 28 Zar HJ , Apolles P , Argent A , et al. The etiology and outcome of pneumonia in human immunodeficiency virus-infected children admitted to intensive care in a developing country . Pediatr Crit Care Med 2001 ; 2 : 108 – 112 . Google Scholar CrossRef Search ADS PubMed 29 Kitchin OP , Masekela R , Becker P , et al. Outcome of human immunodeficiency virus–exposed and –infected children admitted to a pediatric intensive care unit for respiratory failure . Pediatr Crit Care Med 2012 ; 13 : 516 – 19 . Google Scholar CrossRef Search ADS PubMed 30 Argent AC. Managing HIV in the PICU–the experience at the Red Cross War Memorial Children’s Hospital in Cape Town . Indian J Pediatr 2008 ; 75 : 615 – 20 . Google Scholar CrossRef Search ADS PubMed 31 Jeena PM. Challenges in the provision of ICU services to HIV infected children in resource poor settings: a South African Case Study . J Med Ethics 2005 ; 31 : 226 – 30 . Google Scholar CrossRef Search ADS PubMed 32 Le Roux DM , Myer L , Nicol MP , et al. Incidence and severity of childhood pneumonia in the first year of life in a South African birth cohort: the Drakenstein Child Health Study . Lancet Glob Health 2015 ; 3 : e95 – 103 . Google Scholar CrossRef Search ADS PubMed © 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)

Journal

Journal of Tropical PediatricsOxford University Press

Published: Mar 9, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off