 |
|
Brief Reports Indian Pediatrics 2001; 38: 757-762 |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Risk Factors for Kernicterus in Term Babies with Non-Hemolytic Jaundice |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Srinivas Murki, Praveen Kumar, S. Majumdar*, Neelam Marwaha, Anil Narang From
the Departments of Pediatrics, Experimental Medicine* and Hematology+,
Postgraduate Institute of Medical Education and Research, Chandigarh 160
012, India.
Kernicterus, the extreme form of bilirubin induced neurological dysfunction is currently uncommon in term babies but dreaded because of its irreversibility. In many Asian and African countries, we still see a significant number of babies with kernicterus(1,2). Nearly three fourths (77.8 %) of these kernicteric babies do not have any evidence of hemolysis(l). Even in the USA, there have been reports of increased occurrence of kernictrus in otherwise healthy term new-borns(3). The occurrence of bilirubin brain damage in preterm sick babies and in those with hemolysis is understood but why term healthy babies develop this is not clear. It is not known why some babies develop kernicterus at a particular bilirubin level, while others with similar bilirubin levels do not. The American Academy of Pediatrics has recommended 25-29 mg/dl of serum bilirubin as the cut off for exchange transfusion in term babies with non hemolytic jaundice based on the extremely rare occurrence of kernicterus in this group(4). However as many as 9.8% of babies with serum bilirubin between 20-25 mg/dl had kernicterus in a study from north India(1). Are the Asian and African populations particularly vulnerable to bilirubin toxicity? It would be important to find out if we can more precisely define babies who are likely to develop kernicterus in our setting. We, therefore, planned this study to find the risk factors for development of kernicterus in term babies with non hemolytic jaundice. Patients and Methods This prospective study was conducted from July 1998 to June 1999 in term (³37 completed weeks) neonates with severe non hemolytic jaundice referred to the neonatal unit of our hospital. The inclusion criteria were total serum bilirubin ³18 mg/dl, absence of hemolysis and major malformations. The cut off of 18 mg/dl was chosen because in an earlier communication, no case of kernicterus occurred at total serum bilirubin of <18 mg/dl in term babies(l). The diagnosis of hemolysis was based on positive direct Coomb’s test, peripheral blood smear, reticulocyte count, plasma hemoglobin and packed cell volumes. Exchange transfusion was done whenever total serum bilirubin level reached ³20 mg/dl. The diagnosis and classification of kernicterus was based on standard clinical criteria(5). All the babies with kernicterus had stage II bilirubin encephalopathy, i.e., presence of opisthotonus, rigidity and sun-setting of eyeballs. All of them also underwent a cerebrospinal fluid (CSF) examination to rule out meningitis and a cranial ultrasound to rule out intracranial bleed. Assessment of gestation was based on the last menstrual period and was confirmed by Ballards Scoring System. Asphyxia was defined as failure to establish respiration or cry at 1 minute of life. Investigations In addition to total serum bilirubin with conjugated fraction, packed cell volume, reticulocyte count, peripheral blood smear, blood grouping and Rh-typing of baby and mother and G6PD levels, the following investigations were done: (a) blood glucose, (b) pH status, (c) plasma hemoglobin,(d) direct Coombs’ test, (e) serum albumin, (f) free bilirubin (g) free fatty acids. The blood samples for the investigations were taken pre-exchange at peak total serum bilirubin levels or at admission for kernicteric babies. The total serum bilirubin was estimated by spectro-photometry while the free bilirubin was measured by peroxidase method(6). Free fatty acids were measured by spectrophotometric method(7). Ultrasound examination of abdo-men and cranium was done in all babies to look for any internal bleeds. Statistical Analysis The data between the groups of babies with kernicterus and those without kernicterus was compared using Chi-square test and Fishers exact tests for qualitative data. Unpaired Student ‘t’ test was used for quantitative data. The independent association of asphyxia, intrauterine growth retardation and maximum serum bilirubin or free bilirubin with the occurrence of kernicterus was assessed by logistic regression analysis. As the number of events (kerenicterus) were small, only 3 variables at a time were included in the logistic regression analysis. Variables such as birth weight, proportion of SFD babies and proportion of exclusively breast fed babies were not included in the regression analysis as they were statistically insignificant on univariate analysis. Informed parental consent was taken before inclusion into the study and the study was approved by the institute ethics committee. Results During the study period, 92 term babies with a TSB ³18 mg/dl were admitted. Of these, 28 were excluded because 24 babies had evidence of hemolysis (Rh isoimmunization-5, ABO incompatibility-10, G6PD deficiency-9) and 4 babies had major congenital malformations (Downs’ syndrome, duodenal atresia, meningomyelocele, malrotation with volvulus-1 each). There were 64 babies eligible for the study, of whom 14 (21.8%) had kernicterus. The mean age of onset of clinical jaundice as reported by parents was 3.64 ± 1.1 days in babies with kernicterus compared to 3.2 ± 1.1 days in non-kernicteric babies. The duration of jaundice before they reported to our hospital was also similar in both the groups (1.68 ± 0.8 vs 1.98 ± 1.1 days, p >0.05). The average age of onset of kernicterus was 4.64 days and the average age at admission of these babies was 5.5 days. The sex distribution, the mean gestational age, and the mean birth weight of the two groups were comparable (Table I). There were more SFD (both <1 SD and < 2 SD) babies with kernicterus compared to babies without kerrnicterus. However, both of these differ-ences were statistically not significant. Various medical and obstetric problems occurred with similar frequency in mothers of kernicteric and non-kernicteric babies. Ninety three per cent of kernicteric babies had been delivered vaginally as against 74% of babies without kernicterus (p>0.05). Seventy one per cent of kernicteric babies were delivered in the hospital as against 90% of nonkernicteric babies. The use of oxytocin for induction or augmentation of labour was marginally higher in non kernicteric babies (42% vs 28.6%, p >0.05). The two groups were comparable for history of birth asphyxia, pH at admission and postnatal weight loss (Table I). One baby with kernicterus had Klebsiella septicemia without meningitis. In the group without kernicterus one baby each had fracture femur, cephal-hematoma and grade I germinal matrix bleed, respectively. Exclusive breastfeeding was more frequent in non-kernicteric babies, but the differences were insignificant. Table I - Neonatal Risk Factors
Table II - Biochemical Parameters (Mean ± SD) (Range)
Laboratory Parameters The mean maximum total serum bilirubin, free bilirubin and bilirubin/albumin ratio were significantly higher in babies with kernicterus than in those without kernicterus (Table II). The incidence of kernicterus increased with increasing levels of total serum bilirubin, free bilirubin and bilirubin to albumin ratio. The chi-square for linear trend was however significant only for free bilirubin. Fourteen per cent of babies with total bilirubin <25 mg /dl, 18% of babies with total bilirubin 25-29 mg/dl and 43% of babies with total bilirubin >30 mg/dl had kernicterus. Ten per cent of babies with free bilirubin <20 nmol/L had kernicterus while 30% and 46.7% of babies with free bilirubin of 20-23.9 nmol/L and >24 nmol/L, respectively had kernicterus. Seventy one per cent of babies with kernicterus had free bilirubin >20 nmol/L. The FFA were significantly elevated in babies with kernicterus (1.58±2.26 vs 0.88±0.21mmol/L, p = 0.03) though the serum albumin was not different between the two groups (Table II). On multiple logistic regression analysis, asphyxia (OR 8.29, CI 1.17 – 111.8, p = 0.03) and maximum TSB (OR 1.15, CI 1.04-1.28, p = 0.005) or free bilirubin (OR 1.10, CI 0.53-2.24, p = 0.009) retained their independent statistical significance. Discussion Our objective was to compare babies who developed kernicterus and those who did not, in the range of bilirubin known to be definitely toxic. All the babies were from the northern states of India reflecting the factors operating in the population of this region. More of the kernicteric babies were delivered by vaginal route compared to the non kernicteric group. Vaginally delivered babies are likely to be discharged early from the hospital. This may cause delayed diagnosis of jaundice and an increased risk of kernic-terus(8). However, the mean age of onset and the duration of jaundice before reporting to our hospital was similar in both the groups. Oxytocin, especially if administered with hypotonic fluids can aggravate hemolysis in the newborn babies by creating hypo-osmolality of the intravascular fluid(9). This leads to higher bilirubin levels. Though kernicteric babies had higher maximum bili-rubin levels, use of oxytocin was more prevalent in mothers of the non kernicteric babies. Exclusively breast fed babies are reported to have higher bilirubin levels than those on formula feeds(10). In our study, the proportion of babies receiving exclusive breastfeeding before reporting was lesser in babies with kernicterus (57%) compared to those without kernicterus (80%). However, the kernicteric babies had lost 7.7% of their birth weight at admission compared to 4.17% weight loss in non kernicteric babies. This could be due to babies not being able to establish lactation adequately or kernicterus itself resulting in poor feeding. Increased FFA levels in the kernicteric group could have been a result of relative starvation; which in turn could have displaced bilirubin from the binding sites on albumin, resulting in kernicterus. Term babies with non-hemolytic jaundice are generally considered to be at low risk for kernicterus till TSB of 25-29 mg/dl(4). How-ever in our earlier experience, kernicterus was present in 9.8% of babies with TSB 20-25 mg/dl(1). In the current study, kernicterus was seen in 14% of babies with total bilirubin <25 mg/dl and 18% of babies with TSB 25-29 mg/dl. Various hypotheses have been proposed for the increased propensity of Asian and Africans to bilirubin toxicity. Increased prevalence of G6PD deficiency is one major factor(11) but the increased risk is seen even after excluding babies with G6PD defici-ency(1). Measurements of carbonmonoxide production suggest that increased bilirubin production is an important factor contributing to hyperbilirubinemia in these areas(12). In our study babies with kernicterus had higher maximum total serum bilirubin, higher free bilirubin and bilirubin to albumin ratio. Serum albumin was however, similar between the two groups. On logistic regression analysis, asphyxia and maximum total serum bilirubin or free bilirubin are the independent factors associated with the occurrence of kernicterus. The incidence of perinatal asphyxia is significantly higher in developing countries(13). Acidosis, anoxia, hypercarbia and other metabolic altera-tions associated with asphyxia increase levels of free bilirubin, disrupt the blood brain barrier and promote bilirubin deposition in the injured brain and thus increase the risk of kernic-terus(14,15). Study Limitations Events such as onset of jaundice, duration of jaundice, onset and duration of kernicterus before reporting to the hospital and history of asphyxia are historical and are subject to error. Asphyxia is defined as inability to establish cry at 1 min of age. This is the standard definition used in the neonatal data and has obvious limitations. Kernicterus in term babies being a relatively uncommon event , the number of babies included in the study are limited and hence the confidence intervals of most significant variables( asphyxia, small for date) are wide. In conclusion asphyxia and maximum serum bilirubin or free bilirubin levels were the determining factors for the occurrence of kernicterus in term babies with non-hemolytic jaundice. Contributors: SMK and PK designed the study, collected and analyzed the data and prepared the manuscripts; they will act as guarantors for the paper. SM and NM helped in analyzing and interpreting the data. AN helped in designing the study and in preparing the manuscript. Funding:
None.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| References | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
