Correlation between Perfusion Index and CRIB Score in Sick Neonates Admitted to a Tertiary Center

Correlation between Perfusion Index and CRIB Score in Sick Neonates Admitted to a Tertiary Center Abstract Objective The study was to determine the correlation of Perfusion Index (PI) and Clinical Risk Index for Babies (CRIB) score, in assessing the severity of illness in sick neonates. Methods This was a cross-sectional study conducted at a tertiary care Neonatal Intensive Care Unit (NICU). All eligible neonates, both term and preterm, admitted to the high-dependency unit of the NICU were included, after parental consent. Relevant details of history and examination were collected with a structured proforma. Severity of illness was assessed using CRIB score within 12 h of admission. PI was recorded within 24 h of admission, and babies were examined for the presence or absence of shock and their outcome was documented. The correlation coefficient between PI and CRIB score was derived. Results A total of 200 eligible newborns were enrolled. The mean gestational age of the neonates was 34 weeks. The median [interquartile range (IQR)] CRIB score was 1.00 (0.00, 3.00), and PI was 1.400 (0.93, 2.30). The Spearman’s rank correlation coefficient between PI and CRIB score was −0.41 with p value <0.05. The median PI of neonates with CRIB score ≤5, 6–10 and >10 was 1.50, 0.74, 0.67, respectively (p value <0.0001). The median (IQR) PI of babies with shock and without shock was 0.63 (0.43, 0.84) and 1.58 (1.19, 2.41), respectively, with p value <0.001. Conclusion PI has a negative correlation with CRIB score and can be used to assess the severity of illness in sick neonates. shock, illness severity, pulse oximeter, monitoring INTRODUCTION Early detection of shock (inadequate tissue perfusion and oxygenation) is important in babies admitted to Neonatal Intensive Care Unit (NICU), so that we can institute early therapy [1]. Shock is clinically assessed by monitoring heart rate, blood pressure, capillary refilling time, acid–base status and urine output [1–3]. However, it has been shown that these parameters are relatively poor indicators of acute blood flow changes in the immediate neonatal period as shown by Doppler ultrasound and near-infrared spectroscopy [4–6]. Perfusion index (PI) is the ratio of the pulsatile blood flow to the nonpulsatile or static blood in peripheral tissue. When peripheral hypoperfusion exists, the pulsatile component decreases, and the ratio drops because the nonpulsatile component does not change [2]. Thus, PI, is a noninvasive numerical measure of real time changes in peripheral perfusion measured continuously and noninvasively, easily obtained from the newer pulse oximeters [7–9]. Various studies have shown that PI provides information about illness severity, early neonatal respiratory outcome, low superior venacaval flow, hemodynamically significant patent ductus arteriosus and subclinical chorioamnionitis [10–12, 13]. The American College of Critical care Medicine update emphasized on early use of age-specific therapies and recommends first hour fluid resuscitation and inotrope therapy targeting threshold heart rates, blood pressure and capillary refill ≤2 s [14]. It is here that we think PI can help in early detection of poor tissue perfusion or shock, which will further translate to early initiation of appropriate management. Assessment of illness severity is important in predicting morbidity and mortality in neonates admitted to the NICU. Clinical Risk Index for Babies (CRIB) score is a score that assesses the initial clinical severity in preterm infants. CRIB score is a reliable, easily applicable and an accurate test for predicting mortality [15, 16]. Also, CRIB is assessed over the first 12 h of life, making it less susceptible to treatment effects than some of the other scores [17]. That is why we chose CRIB score among other illness severity scores to assess the severity of illness. It is known that low PI is suggested to be an objective and accurate measure of acute illness [18]. We wanted to determine the efficacy of PI, as an objective assessment tool of illness in neonates, and compare it with CRIB score and evaluate the correlation between them, if any. METHODS The total number of babies admitted to the NICU during the study period was 250, of which, 234 neonates admitted to the high-dependency unit (HDU) were eligible. Of these, 200 neonates were included in the study as per inclusion criteria as seen in Fig. 1. Fig. 1. View largeDownload slide Study flow chart. Fig. 1. View largeDownload slide Study flow chart. The study period was 7 months from January 2016 to August 2016. Clearance from institutional ethics review board for the study and informed consent from parents were taken. This was a cross-sectional study, and the sample size was calculated based on our pilot study on 20 neonates. Considering a correlation of 0.30 between PI and CRIB score, with 90% power and 5% level of significance, the number required was 106. All neonates, both term and preterm, admitted to the HDU of our NICU were included for the study. In total, 34 babies in whom PI could not be recorded within 24 h of admission because of several reasons, and 16 of those admitted to the low dependency unit were excluded from the study. Criteria for admission to the HDU were preterm and term newborns who need ventilation (invasive or noninvasive), therapeutic hypothermia, had shock requiring fluid boluses and/or inotropes, moderate–severe respiratory distress (RDS score ≥4) and those who required hourly monitoring. Relevant details of history and examination were collected with a structured proforma. The CRIB score of eligible neonates was recorded within 12 h of admission by the principal investigator. Blood gas was done as dictated by the clinical requirements of each infant within few hours of admission. PI was recorded within 24 h of admission by an operator, who was unaware of the infant illness severity score. PI was recorded using Masimo SET Radical-7 Rainbow pulse oximeter, with the probe connected to the right upper limb (preductal). PI values were taken after the pulse wave was verified to be artifact-free, for 20 s every min for 10 mins and the average of 30 readings were taken. This method for recording the values was based on a previous study on PI [12]. Enrolled neonates were also examined for the presence or absence of shock. Shock was defined as decreased perfusion manifested by altered decreased mental status, capillary refill >3 s (cold shock) or flash capillary refill (warm shock), diminished (cold shock) or bounding (warm shock) peripheral pulses, mottled cool extremities (cold shock) or decreased urine output <1 ml/kg/h and/or requirement of fluid bolus and inotropes [14]. The outcome of the study subjects was also documented. STATISTICAL ANALYSIS Descriptive statistics using mean and SD for the normally distributed continuous variables, median with interquartile ranges for nonparametric data and percentages for the categorical variables were used. Correlation between PI and CRIB score was assessed using Spearman’s rank correlation. PI and CRIB scores were compared between shock present and absent groups using Mann–Whitney U test. Chi-square test was used to test the association between study variables. Predictors for PI levels were done using multiple linear regression analysis. In addition, binary logistic regression analysis was done to look for any association of lower PI and mortality, considering mortality as an outcome and adjusted for age and sex. The p value <0.05 was set as statistically significant. All the statistical analyses were done using SPSS version 21.0. RESULTS Among the 200 neonates studied, the mean (SD) gestational age of the babies in our study was 34.7 (3.8) weeks, range of 25–41 weeks, and preterm babies constituted 60%. The birth weight ranged between 737 and 4500 g. Extremely low birth weight, very low birth weight and low birth weight babies were 9, 19.5 and 39.5, respectively. There was a striking male predominance in this study group with a ratio of 1.7:1. The baseline characteristics of the neonates are further summarized in Table 1. Table 1 Baseline characteristics of the study population Characteristics Number of patients % Sex distribution  Male 128 64  Female 72 36 Gestation  Preterm 119 59.5  Term 81 39.5 Mode of delivery  NVD 89 44.5  LSCS 111 55.5 Outcome  Discharge 178 89  Death/DAMA 22 11 Resuscitation required  Yes 83 41.5  No 117 58.5 Shock  Normal saline bolus 1 0.5  Inotropes 19 99.5 Ventilation  CPAP 29 14.5  SIMV 52 26  HFOV 3 1.5  Nil 116 58 Characteristics Number of patients % Sex distribution  Male 128 64  Female 72 36 Gestation  Preterm 119 59.5  Term 81 39.5 Mode of delivery  NVD 89 44.5  LSCS 111 55.5 Outcome  Discharge 178 89  Death/DAMA 22 11 Resuscitation required  Yes 83 41.5  No 117 58.5 Shock  Normal saline bolus 1 0.5  Inotropes 19 99.5 Ventilation  CPAP 29 14.5  SIMV 52 26  HFOV 3 1.5  Nil 116 58 Note: CPAP, continuous positive airway pressure; SIMV, synchronized intermittent mandatory ventilation; HFOV, high-frequency oscillatory ventilation; LSCS, lower segment cesarean section; NVD, normal vaginal delivery. Table 1 Baseline characteristics of the study population Characteristics Number of patients % Sex distribution  Male 128 64  Female 72 36 Gestation  Preterm 119 59.5  Term 81 39.5 Mode of delivery  NVD 89 44.5  LSCS 111 55.5 Outcome  Discharge 178 89  Death/DAMA 22 11 Resuscitation required  Yes 83 41.5  No 117 58.5 Shock  Normal saline bolus 1 0.5  Inotropes 19 99.5 Ventilation  CPAP 29 14.5  SIMV 52 26  HFOV 3 1.5  Nil 116 58 Characteristics Number of patients % Sex distribution  Male 128 64  Female 72 36 Gestation  Preterm 119 59.5  Term 81 39.5 Mode of delivery  NVD 89 44.5  LSCS 111 55.5 Outcome  Discharge 178 89  Death/DAMA 22 11 Resuscitation required  Yes 83 41.5  No 117 58.5 Shock  Normal saline bolus 1 0.5  Inotropes 19 99.5 Ventilation  CPAP 29 14.5  SIMV 52 26  HFOV 3 1.5  Nil 116 58 Note: CPAP, continuous positive airway pressure; SIMV, synchronized intermittent mandatory ventilation; HFOV, high-frequency oscillatory ventilation; LSCS, lower segment cesarean section; NVD, normal vaginal delivery. The median [interquartile range (IQR)] CRIB score was 1.00 (0.00, 3.00), and PI was 1.400 (0.93, 2.30). The median (IQR) PI among the preterm neonates was 1.30 (0.90, 2.20) and term babies was 1.70 (1.05, 2.65) with p = 0.07. There was no significant difference in the PI values with respect to requirement of resuscitation, therapeutic hypothermia and gender. The PI values were compared with CRIB score as seen in Fig. 2. Correlation observed between PI and CRIB scores was R = −0.41, p value < 0.01. PI of neonates with CRIB score ≤ 5, 6–10 and >10 was a mean (SD) of 1.94 (1.36) 0.97 (0.7), 0.84 (0.64), respectively, with p value <0.000. Median values are shown in Table 2. Increasing illness had significantly lower PI values and increasing mortality rates of 7.7, 35.7 and 50% in neonates with Levels 1, 2 and 3 CRIB scores, respectively, which is of statistical significance. Table 2 Comparison of PI values with CRIB scores CRIB category (number of patients) 0–5 (Level 1) [180] 6–10 (Level 2) [14] 11–15 (Level 3) [6] ≥ 16 (Level 4) [0] Median 1.50 0.74 0.67 0 Mortality (%) 7.7 35.7 50 0 CRIB category (number of patients) 0–5 (Level 1) [180] 6–10 (Level 2) [14] 11–15 (Level 3) [6] ≥ 16 (Level 4) [0] Median 1.50 0.74 0.67 0 Mortality (%) 7.7 35.7 50 0 Note: CPAP, continuous positive airway pressure; SIMV, synchronized intermittent mandatory ventilation; HFOV, high-frequency oscillatory ventilation; LSCS, lower segment cesarean section; NVD, normal vaginal delivery. Table 2 Comparison of PI values with CRIB scores CRIB category (number of patients) 0–5 (Level 1) [180] 6–10 (Level 2) [14] 11–15 (Level 3) [6] ≥ 16 (Level 4) [0] Median 1.50 0.74 0.67 0 Mortality (%) 7.7 35.7 50 0 CRIB category (number of patients) 0–5 (Level 1) [180] 6–10 (Level 2) [14] 11–15 (Level 3) [6] ≥ 16 (Level 4) [0] Median 1.50 0.74 0.67 0 Mortality (%) 7.7 35.7 50 0 Note: CPAP, continuous positive airway pressure; SIMV, synchronized intermittent mandatory ventilation; HFOV, high-frequency oscillatory ventilation; LSCS, lower segment cesarean section; NVD, normal vaginal delivery. Fig. 2. View largeDownload slide Scatter plot of CRIB score with PI. Fig. 2. View largeDownload slide Scatter plot of CRIB score with PI. Figure 3 shows the relationship between PI values and the presence of shock in these babies. Fig. 3. View largeDownload slide Box plot diagram of PI values in the absence or presence of shock. Fig. 3. View largeDownload slide Box plot diagram of PI values in the absence or presence of shock. The median (IQR) PI was significantly lower among babies with shock, 0.63 (0.43, 0.84) as compared with those without shock, 1.58 (1.19, 2.41), p value <0.001. Linear regression analysis was performed. After adjusting for age and sex, it was found that PI levels decrease by 1.28 units in the presence of shock. PI among the discharged neonates was 1.47 (1.03, 2.40) and among death or discharged against medical advice (DAMA) neonates was 0.75 (0.55, 2.02) with p value <0.001. In addition, an analysis was done to assess the odds of higher mortality with lower PI. It was observed that the odds of death were 1.7 times [unadjusted odds ratio (OR) 1.73; 95% confidence interval (CI) 0.99—3.00; p = 0.05] higher with lower PI. After adjusting for age and sex, the adjusted OR was 1.60 (95% CI 0.92–2.73; p = 0.09). Multivariate analysis in assessing predictors of confounding factors of PI revealed that lower gestational age and presence of shock were independent factors affecting PI. DISCUSSION The pulse oximeter PI has been linked to illness severity and outcomes in both newborns and adults [2, 18]. We found that PI has a significant negative correlation with illness severity as assessed by CRIB score. The correlation coefficient was −0.41 and was of statistical significance. The PI value was significantly lower with increasing illness severity score, as seen in the Scatter diagram (Fig. 2). In one and the only other study comparing illness severity in newborns using Score for Neonatal Acute Physiology score with PI, by De Felice et al. [18], PI's predictive accuracy was shown to be significant, with 95.5% sensitivity and 93.7% specificity. The high severity group had a PI of 0.86 ± 0.26, and low severity group had a PI of 2.02 ± 0.70. They showed that low PI values <1.24 were an accurate predictor for high illness severity in newborns. Our data also reflect the same, the high severity group had a PI of 0.85 ± 0.64 and low severity group had a PI of 1.94 ± 1.36. In neonates with higher CRIB score (Levels 2 and 3), the PI values were on the lower side. CRIB score of Level 2 and more has been associated with corresponding increase in illness severity and mortality. Sarquis et al. [19] study showed mortality of 6.6% in Level 1, 46.2% in Level 2, 85.7% in Level 3 and 100.0% among neonates with Level 4 CRIB score [20]. Our data showed that increasing illness had significantly lower PI values and increasing mortality rates of 7.7, 35.7 and 50% in neonates with Levels 1, 2 and 3 CRIB scores respectively, which is of statistical significance. PI among the discharged neonates was higher than among Death/DAMA neonates, which is also statistically significant. Reference PI values in newborns have been recently published. Healthy term newborns median PI value in Granelli et al. [9] study was 1.7. The mean PI value among those with a CRIB score ≤5 (Level 1) was 1.94 in our study. Granelli et al. concluded that PI values <0.70 may indicate illness, and Takahashi et al. [11] found that PI < 0.44 was suggestive of low superior vena cava flow. In our study, PI values <0.97 were correlating well with higher illness severity scores. Cresi et al.’s [8] study on 30 clinically and hemodynamically stable preterms showed that the median (IQR) PI values was 0.9 (0.6) on Day 1. A study on term Portuguese neonates by Jardim et al. [21] had preductal median PI of 1.6 (IQR 1.2–2.3). Our results showed a mean (IQR) PI of 1.7 (0.3) on Day 1 among 119 preterms and mean (IQR) PI of 2.04 (1.6) among the 81 term babies included in our study, which is of no statistical significance. However, the difference could be because of different instruments used to monitor the PI between these studies and ours. Multivariate analysis showed that gestational age had a positive correlation with PI; however, resuscitation and therapeutic hypothermia had no significant correlation. On further analysis, we also noted that the PI values were significantly lower in those babies with clinical features of shock, requiring fluid bolus and/or inotropes, which was statistically significant. In Rasmy et al. [22] study, low PI was a good indicator of vasopressor requirement in adults with severe sepsis. The optimum cutoff value of PI for predicting vasopressor in that study was 0.3. This cutoff value had a sensitivity of 100% and a specificity of 93%. The slightly higher reading in our study maybe because of the level of illness not being too grave on Day 1 of admission as demonstrated by our maximum CRIB score of 13 as against 19 and the difference in study population. The importance of the findings in our study is that, PI, which is easily obtained on a pulse oximeter that is now extensively used in intensive care units, can be a good noninvasive tool to assess illness [13]. Strengths of the study are that enrolled neonates were across all gestational ages, had varied underlying illness and even children under therapeutic hypothermia were included. Observer bias was minimal, as the person recording the PI was blinded to the CRIB score. Limitations of this study were that all PI readings of the enrolled neonates were not taken at the same hour after admission, so this could cause physiological variations to the PI values. Our data give no information about the range of variability within the same individual. The other factors, which may affect the PI, have also not been covered in the scope of this study. More studies with regard to factors, which affect PI in neonates, would add more light on this simple clinical tool. CONCLUSION PI has a negative correlation with CRIB score. It is a simple tool and can be used to assess the severity of illness in neonates. It may help in early diagnosis of shock. ACKNOWLEDGEMENTS The authors thank Mrs Sumithra S, our Biostatistician, for all her help with the statistical analysis of the data. REFERENCES 1 Lima A , Bakker J. Noninvasive monitoring of peripheral perfusion . Intensive Care Med 2005 ; 31 : 1316 – 26 . Google Scholar CrossRef Search ADS PubMed 2 Lima AP , Beelen P , Bakker J. Use of a peripheral perfusion index derived from the pulse oximetry signal as a noninvasive indicator of perfusion . Crit Care Med 2002 ; 30 : 1210 – 3 . Google Scholar CrossRef Search ADS PubMed 3 Van Genderen M , van Bommell J , Lima A. Monitoring peripheral perfusion in critically ill patients at the bedside . Curr Opin Crit Care 2012 ; 18 : 273 – 9 . Google Scholar CrossRef Search ADS PubMed 4 Tyszczuk L , Meek J , Elwell C , et al. Cerebral blood flow is independent of mean arterial blood pressure in preterm infants undergoing intensive care . Pediatrics 1998 ; 102 : 337 – 4 . Google Scholar CrossRef Search ADS PubMed 5 Pladys P , Wodey E , Bétrémieux P , et al. Effects of volume expansion on cardiac output in the preterm infant . Acta Paediatr 1997 ; 86 : 1241 – 5 . Google Scholar CrossRef Search ADS PubMed 6 Kluckow M , Evans N. Superior vena cava flow in newborn infants: a novel marker of systemic blood flow . Arch Dis Child Fetal Neonatal Ed 2000 ; 82 : F182 – 7 . Google Scholar CrossRef Search ADS PubMed 7 Sahni R , Schulze KF , Ohira-Kist K , et al. Interactions among peripheral perfusion, cardiac activity, oxygen saturation, thermal profile and body position in growing low birth weight infants . Acta Paediatr 2010 ; 99 : 135 – 9 . Google Scholar PubMed 8 Cresi F , Pelle E , Calabrese R , et al. Perfusion index variations in clinically and hemodynamically stable preterm newborns in the first week of life . Ital J Pediatr 2010 ; 36 : 6. Google Scholar CrossRef Search ADS PubMed 9 Granelli A , Ostman-Smith I. Noninvasive peripheral perfusion index as a possible tool for screening for critical left heart obstruction . Acta Paediatr 2007 ; 96 : 1455 – 9 . Google Scholar CrossRef Search ADS PubMed 10 De Felice C , Del Vecchio A , Criscuolo M , et al. Early postnatal changes in the perfusion index in term newborns with subclinical chorioamnionitis . Arch Dis Child Fetal Neonatal Ed 2005 ; 90 : F411 – 4 . Google Scholar CrossRef Search ADS PubMed 11 Takahashi S , Kakiuchi S , Nanba Y , et al. The perfusion index derived from a pulse oximeter for predicting low superior vena cava flow in very low birth weight infants . J Perinatol 2010 ; 30 : 265 – 9 . Google Scholar CrossRef Search ADS PubMed 12 Balla KC , John V , Rao PNS , et al. Perfusion index—bedside diagnosis of hemodynamically significant patent ductus arteriosus . J Trop Ped 2016 ; 62 : 263 – 8 . Google Scholar CrossRef Search ADS 13 Salyer JW. Neonatal and pediatric pulse oximetry . Respir Care 2003 ; 48 : 386 – 96 . Google Scholar PubMed 14 Brierley J , Carcillo JA , Choong K , et al. Clinical practice parameters for hemodynamic support of pediatric and neonatal septic shock: 2007 update from the American College of Critical Care Medicine . Crit Care Med 2009 ; 37 : 666 – 88 . Google Scholar CrossRef Search ADS PubMed 15 Rautonen J , Makella A. CRIB and SNAP assessing the risk of death for preterm neonates . Lancet 1994 ; 21 : 1272 – 3 . Google Scholar CrossRef Search ADS 16 Bastos G , Gomes A. A comparison of 4 pregnancy assessment scales (CRIB, SNAP, SNAP PE, NTISS) in premature newborns . Acta Med Port 1997 ; 10 : 161 – 5 . Google Scholar PubMed 17 Dorling JS , Field D , Manktelow JB. Neonatal disease severity scoring systems . Arch Dis Child Fetal Neonatal Ed 2005 ; 90 : F11 16 . Google Scholar CrossRef Search ADS PubMed 18 De Felice C , Latini G , Vacca P , et al. The pulse oximeter perfusion index as a predictor for high illness severity in neonates . Eur J Pediatr 2002 ; 161 : 561 – 2 . Google Scholar CrossRef Search ADS PubMed 19 Sarquis AL , Miyaki M , Monica NL. CRIB score for predicting neonatal mortality risk . J Pediatr (Rio J) 2002 ; 78 : 225 – 9 . Google Scholar CrossRef Search ADS PubMed 20 Angela SJ , Tiemi M , Maria RC , et al. CRIB score, birth weight and gestational age in neonatal mortality risk evaluation . Rev Saúde Públic 2003a ; 37 : 597 – 602 . Google Scholar CrossRef Search ADS 21 JardimRocha J , Silva R , Guimarães GH. Peripheral perfusion index-reference range in healthy Portuguese term newborns . J Pediatr Neonatal Indiv Med 2014 ; 3 : 22 Rasmy I , Mohamed H , Nabil H , et al. Evaluation of perfusion index as a predictor of vasopressor requirement in patients with severe sepsis . Shock 2015 ; 44 : 554 – 9 . 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

Correlation between Perfusion Index and CRIB Score in Sick Neonates Admitted to a Tertiary Center

Loading next page...
 
/lp/ou_press/correlation-between-perfusion-index-and-crib-score-in-sick-neonates-nHzesiWPpG
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/fmy016
Publisher site
See Article on Publisher Site

Abstract

Abstract Objective The study was to determine the correlation of Perfusion Index (PI) and Clinical Risk Index for Babies (CRIB) score, in assessing the severity of illness in sick neonates. Methods This was a cross-sectional study conducted at a tertiary care Neonatal Intensive Care Unit (NICU). All eligible neonates, both term and preterm, admitted to the high-dependency unit of the NICU were included, after parental consent. Relevant details of history and examination were collected with a structured proforma. Severity of illness was assessed using CRIB score within 12 h of admission. PI was recorded within 24 h of admission, and babies were examined for the presence or absence of shock and their outcome was documented. The correlation coefficient between PI and CRIB score was derived. Results A total of 200 eligible newborns were enrolled. The mean gestational age of the neonates was 34 weeks. The median [interquartile range (IQR)] CRIB score was 1.00 (0.00, 3.00), and PI was 1.400 (0.93, 2.30). The Spearman’s rank correlation coefficient between PI and CRIB score was −0.41 with p value <0.05. The median PI of neonates with CRIB score ≤5, 6–10 and >10 was 1.50, 0.74, 0.67, respectively (p value <0.0001). The median (IQR) PI of babies with shock and without shock was 0.63 (0.43, 0.84) and 1.58 (1.19, 2.41), respectively, with p value <0.001. Conclusion PI has a negative correlation with CRIB score and can be used to assess the severity of illness in sick neonates. shock, illness severity, pulse oximeter, monitoring INTRODUCTION Early detection of shock (inadequate tissue perfusion and oxygenation) is important in babies admitted to Neonatal Intensive Care Unit (NICU), so that we can institute early therapy [1]. Shock is clinically assessed by monitoring heart rate, blood pressure, capillary refilling time, acid–base status and urine output [1–3]. However, it has been shown that these parameters are relatively poor indicators of acute blood flow changes in the immediate neonatal period as shown by Doppler ultrasound and near-infrared spectroscopy [4–6]. Perfusion index (PI) is the ratio of the pulsatile blood flow to the nonpulsatile or static blood in peripheral tissue. When peripheral hypoperfusion exists, the pulsatile component decreases, and the ratio drops because the nonpulsatile component does not change [2]. Thus, PI, is a noninvasive numerical measure of real time changes in peripheral perfusion measured continuously and noninvasively, easily obtained from the newer pulse oximeters [7–9]. Various studies have shown that PI provides information about illness severity, early neonatal respiratory outcome, low superior venacaval flow, hemodynamically significant patent ductus arteriosus and subclinical chorioamnionitis [10–12, 13]. The American College of Critical care Medicine update emphasized on early use of age-specific therapies and recommends first hour fluid resuscitation and inotrope therapy targeting threshold heart rates, blood pressure and capillary refill ≤2 s [14]. It is here that we think PI can help in early detection of poor tissue perfusion or shock, which will further translate to early initiation of appropriate management. Assessment of illness severity is important in predicting morbidity and mortality in neonates admitted to the NICU. Clinical Risk Index for Babies (CRIB) score is a score that assesses the initial clinical severity in preterm infants. CRIB score is a reliable, easily applicable and an accurate test for predicting mortality [15, 16]. Also, CRIB is assessed over the first 12 h of life, making it less susceptible to treatment effects than some of the other scores [17]. That is why we chose CRIB score among other illness severity scores to assess the severity of illness. It is known that low PI is suggested to be an objective and accurate measure of acute illness [18]. We wanted to determine the efficacy of PI, as an objective assessment tool of illness in neonates, and compare it with CRIB score and evaluate the correlation between them, if any. METHODS The total number of babies admitted to the NICU during the study period was 250, of which, 234 neonates admitted to the high-dependency unit (HDU) were eligible. Of these, 200 neonates were included in the study as per inclusion criteria as seen in Fig. 1. Fig. 1. View largeDownload slide Study flow chart. Fig. 1. View largeDownload slide Study flow chart. The study period was 7 months from January 2016 to August 2016. Clearance from institutional ethics review board for the study and informed consent from parents were taken. This was a cross-sectional study, and the sample size was calculated based on our pilot study on 20 neonates. Considering a correlation of 0.30 between PI and CRIB score, with 90% power and 5% level of significance, the number required was 106. All neonates, both term and preterm, admitted to the HDU of our NICU were included for the study. In total, 34 babies in whom PI could not be recorded within 24 h of admission because of several reasons, and 16 of those admitted to the low dependency unit were excluded from the study. Criteria for admission to the HDU were preterm and term newborns who need ventilation (invasive or noninvasive), therapeutic hypothermia, had shock requiring fluid boluses and/or inotropes, moderate–severe respiratory distress (RDS score ≥4) and those who required hourly monitoring. Relevant details of history and examination were collected with a structured proforma. The CRIB score of eligible neonates was recorded within 12 h of admission by the principal investigator. Blood gas was done as dictated by the clinical requirements of each infant within few hours of admission. PI was recorded within 24 h of admission by an operator, who was unaware of the infant illness severity score. PI was recorded using Masimo SET Radical-7 Rainbow pulse oximeter, with the probe connected to the right upper limb (preductal). PI values were taken after the pulse wave was verified to be artifact-free, for 20 s every min for 10 mins and the average of 30 readings were taken. This method for recording the values was based on a previous study on PI [12]. Enrolled neonates were also examined for the presence or absence of shock. Shock was defined as decreased perfusion manifested by altered decreased mental status, capillary refill >3 s (cold shock) or flash capillary refill (warm shock), diminished (cold shock) or bounding (warm shock) peripheral pulses, mottled cool extremities (cold shock) or decreased urine output <1 ml/kg/h and/or requirement of fluid bolus and inotropes [14]. The outcome of the study subjects was also documented. STATISTICAL ANALYSIS Descriptive statistics using mean and SD for the normally distributed continuous variables, median with interquartile ranges for nonparametric data and percentages for the categorical variables were used. Correlation between PI and CRIB score was assessed using Spearman’s rank correlation. PI and CRIB scores were compared between shock present and absent groups using Mann–Whitney U test. Chi-square test was used to test the association between study variables. Predictors for PI levels were done using multiple linear regression analysis. In addition, binary logistic regression analysis was done to look for any association of lower PI and mortality, considering mortality as an outcome and adjusted for age and sex. The p value <0.05 was set as statistically significant. All the statistical analyses were done using SPSS version 21.0. RESULTS Among the 200 neonates studied, the mean (SD) gestational age of the babies in our study was 34.7 (3.8) weeks, range of 25–41 weeks, and preterm babies constituted 60%. The birth weight ranged between 737 and 4500 g. Extremely low birth weight, very low birth weight and low birth weight babies were 9, 19.5 and 39.5, respectively. There was a striking male predominance in this study group with a ratio of 1.7:1. The baseline characteristics of the neonates are further summarized in Table 1. Table 1 Baseline characteristics of the study population Characteristics Number of patients % Sex distribution  Male 128 64  Female 72 36 Gestation  Preterm 119 59.5  Term 81 39.5 Mode of delivery  NVD 89 44.5  LSCS 111 55.5 Outcome  Discharge 178 89  Death/DAMA 22 11 Resuscitation required  Yes 83 41.5  No 117 58.5 Shock  Normal saline bolus 1 0.5  Inotropes 19 99.5 Ventilation  CPAP 29 14.5  SIMV 52 26  HFOV 3 1.5  Nil 116 58 Characteristics Number of patients % Sex distribution  Male 128 64  Female 72 36 Gestation  Preterm 119 59.5  Term 81 39.5 Mode of delivery  NVD 89 44.5  LSCS 111 55.5 Outcome  Discharge 178 89  Death/DAMA 22 11 Resuscitation required  Yes 83 41.5  No 117 58.5 Shock  Normal saline bolus 1 0.5  Inotropes 19 99.5 Ventilation  CPAP 29 14.5  SIMV 52 26  HFOV 3 1.5  Nil 116 58 Note: CPAP, continuous positive airway pressure; SIMV, synchronized intermittent mandatory ventilation; HFOV, high-frequency oscillatory ventilation; LSCS, lower segment cesarean section; NVD, normal vaginal delivery. Table 1 Baseline characteristics of the study population Characteristics Number of patients % Sex distribution  Male 128 64  Female 72 36 Gestation  Preterm 119 59.5  Term 81 39.5 Mode of delivery  NVD 89 44.5  LSCS 111 55.5 Outcome  Discharge 178 89  Death/DAMA 22 11 Resuscitation required  Yes 83 41.5  No 117 58.5 Shock  Normal saline bolus 1 0.5  Inotropes 19 99.5 Ventilation  CPAP 29 14.5  SIMV 52 26  HFOV 3 1.5  Nil 116 58 Characteristics Number of patients % Sex distribution  Male 128 64  Female 72 36 Gestation  Preterm 119 59.5  Term 81 39.5 Mode of delivery  NVD 89 44.5  LSCS 111 55.5 Outcome  Discharge 178 89  Death/DAMA 22 11 Resuscitation required  Yes 83 41.5  No 117 58.5 Shock  Normal saline bolus 1 0.5  Inotropes 19 99.5 Ventilation  CPAP 29 14.5  SIMV 52 26  HFOV 3 1.5  Nil 116 58 Note: CPAP, continuous positive airway pressure; SIMV, synchronized intermittent mandatory ventilation; HFOV, high-frequency oscillatory ventilation; LSCS, lower segment cesarean section; NVD, normal vaginal delivery. The median [interquartile range (IQR)] CRIB score was 1.00 (0.00, 3.00), and PI was 1.400 (0.93, 2.30). The median (IQR) PI among the preterm neonates was 1.30 (0.90, 2.20) and term babies was 1.70 (1.05, 2.65) with p = 0.07. There was no significant difference in the PI values with respect to requirement of resuscitation, therapeutic hypothermia and gender. The PI values were compared with CRIB score as seen in Fig. 2. Correlation observed between PI and CRIB scores was R = −0.41, p value < 0.01. PI of neonates with CRIB score ≤ 5, 6–10 and >10 was a mean (SD) of 1.94 (1.36) 0.97 (0.7), 0.84 (0.64), respectively, with p value <0.000. Median values are shown in Table 2. Increasing illness had significantly lower PI values and increasing mortality rates of 7.7, 35.7 and 50% in neonates with Levels 1, 2 and 3 CRIB scores, respectively, which is of statistical significance. Table 2 Comparison of PI values with CRIB scores CRIB category (number of patients) 0–5 (Level 1) [180] 6–10 (Level 2) [14] 11–15 (Level 3) [6] ≥ 16 (Level 4) [0] Median 1.50 0.74 0.67 0 Mortality (%) 7.7 35.7 50 0 CRIB category (number of patients) 0–5 (Level 1) [180] 6–10 (Level 2) [14] 11–15 (Level 3) [6] ≥ 16 (Level 4) [0] Median 1.50 0.74 0.67 0 Mortality (%) 7.7 35.7 50 0 Note: CPAP, continuous positive airway pressure; SIMV, synchronized intermittent mandatory ventilation; HFOV, high-frequency oscillatory ventilation; LSCS, lower segment cesarean section; NVD, normal vaginal delivery. Table 2 Comparison of PI values with CRIB scores CRIB category (number of patients) 0–5 (Level 1) [180] 6–10 (Level 2) [14] 11–15 (Level 3) [6] ≥ 16 (Level 4) [0] Median 1.50 0.74 0.67 0 Mortality (%) 7.7 35.7 50 0 CRIB category (number of patients) 0–5 (Level 1) [180] 6–10 (Level 2) [14] 11–15 (Level 3) [6] ≥ 16 (Level 4) [0] Median 1.50 0.74 0.67 0 Mortality (%) 7.7 35.7 50 0 Note: CPAP, continuous positive airway pressure; SIMV, synchronized intermittent mandatory ventilation; HFOV, high-frequency oscillatory ventilation; LSCS, lower segment cesarean section; NVD, normal vaginal delivery. Fig. 2. View largeDownload slide Scatter plot of CRIB score with PI. Fig. 2. View largeDownload slide Scatter plot of CRIB score with PI. Figure 3 shows the relationship between PI values and the presence of shock in these babies. Fig. 3. View largeDownload slide Box plot diagram of PI values in the absence or presence of shock. Fig. 3. View largeDownload slide Box plot diagram of PI values in the absence or presence of shock. The median (IQR) PI was significantly lower among babies with shock, 0.63 (0.43, 0.84) as compared with those without shock, 1.58 (1.19, 2.41), p value <0.001. Linear regression analysis was performed. After adjusting for age and sex, it was found that PI levels decrease by 1.28 units in the presence of shock. PI among the discharged neonates was 1.47 (1.03, 2.40) and among death or discharged against medical advice (DAMA) neonates was 0.75 (0.55, 2.02) with p value <0.001. In addition, an analysis was done to assess the odds of higher mortality with lower PI. It was observed that the odds of death were 1.7 times [unadjusted odds ratio (OR) 1.73; 95% confidence interval (CI) 0.99—3.00; p = 0.05] higher with lower PI. After adjusting for age and sex, the adjusted OR was 1.60 (95% CI 0.92–2.73; p = 0.09). Multivariate analysis in assessing predictors of confounding factors of PI revealed that lower gestational age and presence of shock were independent factors affecting PI. DISCUSSION The pulse oximeter PI has been linked to illness severity and outcomes in both newborns and adults [2, 18]. We found that PI has a significant negative correlation with illness severity as assessed by CRIB score. The correlation coefficient was −0.41 and was of statistical significance. The PI value was significantly lower with increasing illness severity score, as seen in the Scatter diagram (Fig. 2). In one and the only other study comparing illness severity in newborns using Score for Neonatal Acute Physiology score with PI, by De Felice et al. [18], PI's predictive accuracy was shown to be significant, with 95.5% sensitivity and 93.7% specificity. The high severity group had a PI of 0.86 ± 0.26, and low severity group had a PI of 2.02 ± 0.70. They showed that low PI values <1.24 were an accurate predictor for high illness severity in newborns. Our data also reflect the same, the high severity group had a PI of 0.85 ± 0.64 and low severity group had a PI of 1.94 ± 1.36. In neonates with higher CRIB score (Levels 2 and 3), the PI values were on the lower side. CRIB score of Level 2 and more has been associated with corresponding increase in illness severity and mortality. Sarquis et al. [19] study showed mortality of 6.6% in Level 1, 46.2% in Level 2, 85.7% in Level 3 and 100.0% among neonates with Level 4 CRIB score [20]. Our data showed that increasing illness had significantly lower PI values and increasing mortality rates of 7.7, 35.7 and 50% in neonates with Levels 1, 2 and 3 CRIB scores respectively, which is of statistical significance. PI among the discharged neonates was higher than among Death/DAMA neonates, which is also statistically significant. Reference PI values in newborns have been recently published. Healthy term newborns median PI value in Granelli et al. [9] study was 1.7. The mean PI value among those with a CRIB score ≤5 (Level 1) was 1.94 in our study. Granelli et al. concluded that PI values <0.70 may indicate illness, and Takahashi et al. [11] found that PI < 0.44 was suggestive of low superior vena cava flow. In our study, PI values <0.97 were correlating well with higher illness severity scores. Cresi et al.’s [8] study on 30 clinically and hemodynamically stable preterms showed that the median (IQR) PI values was 0.9 (0.6) on Day 1. A study on term Portuguese neonates by Jardim et al. [21] had preductal median PI of 1.6 (IQR 1.2–2.3). Our results showed a mean (IQR) PI of 1.7 (0.3) on Day 1 among 119 preterms and mean (IQR) PI of 2.04 (1.6) among the 81 term babies included in our study, which is of no statistical significance. However, the difference could be because of different instruments used to monitor the PI between these studies and ours. Multivariate analysis showed that gestational age had a positive correlation with PI; however, resuscitation and therapeutic hypothermia had no significant correlation. On further analysis, we also noted that the PI values were significantly lower in those babies with clinical features of shock, requiring fluid bolus and/or inotropes, which was statistically significant. In Rasmy et al. [22] study, low PI was a good indicator of vasopressor requirement in adults with severe sepsis. The optimum cutoff value of PI for predicting vasopressor in that study was 0.3. This cutoff value had a sensitivity of 100% and a specificity of 93%. The slightly higher reading in our study maybe because of the level of illness not being too grave on Day 1 of admission as demonstrated by our maximum CRIB score of 13 as against 19 and the difference in study population. The importance of the findings in our study is that, PI, which is easily obtained on a pulse oximeter that is now extensively used in intensive care units, can be a good noninvasive tool to assess illness [13]. Strengths of the study are that enrolled neonates were across all gestational ages, had varied underlying illness and even children under therapeutic hypothermia were included. Observer bias was minimal, as the person recording the PI was blinded to the CRIB score. Limitations of this study were that all PI readings of the enrolled neonates were not taken at the same hour after admission, so this could cause physiological variations to the PI values. Our data give no information about the range of variability within the same individual. The other factors, which may affect the PI, have also not been covered in the scope of this study. More studies with regard to factors, which affect PI in neonates, would add more light on this simple clinical tool. CONCLUSION PI has a negative correlation with CRIB score. It is a simple tool and can be used to assess the severity of illness in neonates. It may help in early diagnosis of shock. ACKNOWLEDGEMENTS The authors thank Mrs Sumithra S, our Biostatistician, for all her help with the statistical analysis of the data. REFERENCES 1 Lima A , Bakker J. Noninvasive monitoring of peripheral perfusion . Intensive Care Med 2005 ; 31 : 1316 – 26 . Google Scholar CrossRef Search ADS PubMed 2 Lima AP , Beelen P , Bakker J. Use of a peripheral perfusion index derived from the pulse oximetry signal as a noninvasive indicator of perfusion . Crit Care Med 2002 ; 30 : 1210 – 3 . Google Scholar CrossRef Search ADS PubMed 3 Van Genderen M , van Bommell J , Lima A. Monitoring peripheral perfusion in critically ill patients at the bedside . Curr Opin Crit Care 2012 ; 18 : 273 – 9 . Google Scholar CrossRef Search ADS PubMed 4 Tyszczuk L , Meek J , Elwell C , et al. Cerebral blood flow is independent of mean arterial blood pressure in preterm infants undergoing intensive care . Pediatrics 1998 ; 102 : 337 – 4 . Google Scholar CrossRef Search ADS PubMed 5 Pladys P , Wodey E , Bétrémieux P , et al. Effects of volume expansion on cardiac output in the preterm infant . Acta Paediatr 1997 ; 86 : 1241 – 5 . Google Scholar CrossRef Search ADS PubMed 6 Kluckow M , Evans N. Superior vena cava flow in newborn infants: a novel marker of systemic blood flow . Arch Dis Child Fetal Neonatal Ed 2000 ; 82 : F182 – 7 . Google Scholar CrossRef Search ADS PubMed 7 Sahni R , Schulze KF , Ohira-Kist K , et al. Interactions among peripheral perfusion, cardiac activity, oxygen saturation, thermal profile and body position in growing low birth weight infants . Acta Paediatr 2010 ; 99 : 135 – 9 . Google Scholar PubMed 8 Cresi F , Pelle E , Calabrese R , et al. Perfusion index variations in clinically and hemodynamically stable preterm newborns in the first week of life . Ital J Pediatr 2010 ; 36 : 6. Google Scholar CrossRef Search ADS PubMed 9 Granelli A , Ostman-Smith I. Noninvasive peripheral perfusion index as a possible tool for screening for critical left heart obstruction . Acta Paediatr 2007 ; 96 : 1455 – 9 . Google Scholar CrossRef Search ADS PubMed 10 De Felice C , Del Vecchio A , Criscuolo M , et al. Early postnatal changes in the perfusion index in term newborns with subclinical chorioamnionitis . Arch Dis Child Fetal Neonatal Ed 2005 ; 90 : F411 – 4 . Google Scholar CrossRef Search ADS PubMed 11 Takahashi S , Kakiuchi S , Nanba Y , et al. The perfusion index derived from a pulse oximeter for predicting low superior vena cava flow in very low birth weight infants . J Perinatol 2010 ; 30 : 265 – 9 . Google Scholar CrossRef Search ADS PubMed 12 Balla KC , John V , Rao PNS , et al. Perfusion index—bedside diagnosis of hemodynamically significant patent ductus arteriosus . J Trop Ped 2016 ; 62 : 263 – 8 . Google Scholar CrossRef Search ADS 13 Salyer JW. Neonatal and pediatric pulse oximetry . Respir Care 2003 ; 48 : 386 – 96 . Google Scholar PubMed 14 Brierley J , Carcillo JA , Choong K , et al. Clinical practice parameters for hemodynamic support of pediatric and neonatal septic shock: 2007 update from the American College of Critical Care Medicine . Crit Care Med 2009 ; 37 : 666 – 88 . Google Scholar CrossRef Search ADS PubMed 15 Rautonen J , Makella A. CRIB and SNAP assessing the risk of death for preterm neonates . Lancet 1994 ; 21 : 1272 – 3 . Google Scholar CrossRef Search ADS 16 Bastos G , Gomes A. A comparison of 4 pregnancy assessment scales (CRIB, SNAP, SNAP PE, NTISS) in premature newborns . Acta Med Port 1997 ; 10 : 161 – 5 . Google Scholar PubMed 17 Dorling JS , Field D , Manktelow JB. Neonatal disease severity scoring systems . Arch Dis Child Fetal Neonatal Ed 2005 ; 90 : F11 16 . Google Scholar CrossRef Search ADS PubMed 18 De Felice C , Latini G , Vacca P , et al. The pulse oximeter perfusion index as a predictor for high illness severity in neonates . Eur J Pediatr 2002 ; 161 : 561 – 2 . Google Scholar CrossRef Search ADS PubMed 19 Sarquis AL , Miyaki M , Monica NL. CRIB score for predicting neonatal mortality risk . J Pediatr (Rio J) 2002 ; 78 : 225 – 9 . Google Scholar CrossRef Search ADS PubMed 20 Angela SJ , Tiemi M , Maria RC , et al. CRIB score, birth weight and gestational age in neonatal mortality risk evaluation . Rev Saúde Públic 2003a ; 37 : 597 – 602 . Google Scholar CrossRef Search ADS 21 JardimRocha J , Silva R , Guimarães GH. Peripheral perfusion index-reference range in healthy Portuguese term newborns . J Pediatr Neonatal Indiv Med 2014 ; 3 : 22 Rasmy I , Mohamed H , Nabil H , et al. Evaluation of perfusion index as a predictor of vasopressor requirement in patients with severe sepsis . Shock 2015 ; 44 : 554 – 9 . 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: Apr 18, 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