Adverse Events of Exchange Transfusion in Neonatal Hyperbilirubinemia

Introduction: Jaundice is an important problem during neonatal period. When total serum bilirubin (TSB) level exceeds a critical limit, it crosses the blood brain barrier and results into bilirubin encephalopathy. The main aim of therapy for neonatal hyperbilirubinemia is prevention of bilirubin encephalopathy by phototherapy and/or exchange transfusion. The aims of this study were to evaluate the efficacy of exchange transfusion (ET) and observe the adverse events during and following three days of ET in neonates with hyperbilirubinemia. Materials and Method: Hospital based cross-sectional descriptive study. All neonates admitted to neonatal intensive care unit and /or paediatric wards of a tertiarycare centre between September 2010 to March 2012, requiring ET were enrolled. Results: A total of 139 ETs were performed in 120 neonates. The common causes were ABO incompatibility (30.8%), prematurity (30.8%), idiopathic (27.5%), Rh isoimmunization (6.7%) and cephalhematoma (4.2%). Mean preET total serum bilirubin (TSB) was 24.2 mg% dL. There was 58% reduction in TSB in post ET and 31% net reduction in 6 hr post ET. Term and preterm neonates showed equal percentage of TSB reduction. Respiratory distress (10.8%) and bradycardia (6.7%) were the common adverse events during, and hypocalcemia (98.3%) and thrombocytopenia (34.2%) in 3 days following ET. The sick neonates had significantly higher incidence of thrombocytopenia (p= 0.031), respiratory distress (p=0.009), apnea (p<0.001) and cardiorespiratory arrest (p<0.001). Overall mortality was 4.2%, and non-survivors were mostly low birth weight, born outside the present hospital and had higher incidence of adverse events. Conclusion: Exchange transfusion is an effective intervention in reducing the serum bilirubin level. However, these neonates require monitoring of ionised calcium and thrombocytopenia. Sick neonates had higher incidence of adverse events than healthy and close clinical monitoring is needed to improve the outcome.


Introduction
J aundice is an important problem during neonatal period, especially in the fi rst week of life 1 .The serum bilirubin level varies with birth weight, gesta onal age, chronological age and internal milieu of the body.When total serum bilirubin (TSB) level exceeds a cri cal limit, it crosses the blood brain barrier and results into bilirubin encephalopathy 2 .About 50% of the term and 80% of preterm neonates develop jaundice during early neonatal period 3 .About 0.02-0.16% of neonates develop extreme hyperbilirubinemia 4 .In developed countries, the incidence of severe jaundice and associated bilirubin encephalopathy is low.However, in developing countries, this preventable disease s ll occurs in a signifi cant propor on of neonates 5 .Several clinical factors have been associated in causa on of severe jaundice including breast milk feeding, ABO incompa bly, G6PD defi ciency, East Asian Ethnicity and cephalhematoma 6 .
The main aim of therapy for neonatal hyperbilirubinemia is preven on of bilirubin encephalopathy by phototherapy and/or exchange transfusion (ET).The modality of exchange transfusion in the treatment of hyperbilirubinemia was a proven therapy, probably two decades back, however with the availability of highly eff ec ve double surface and LED phototherapy and administra on of intravenous immunoglobulin G (IVIg) in preven on of isoimmuniza on / blood group incompa bility, the number of exchange transfusions have signifi cantly reduced 7 .However, despite having adverse events, it leads to a rapid decline in serum bilirubin levels in symptoma c neonates suffi cient enough to minimize the brain damage 8 .We report our observa ons on effi cacy, adverse events and factors contribu ng to mortality related with ET performed for the treatment of neonatal hyperbilirubinemia at a ter ary-care centre.

Materials and Methods
This study was conducted at ter ary-care centre of a teaching hospital.All neonates presen ng with indirect hyperbilirubinemia requiring ET during the period of September 2010 to March 2012 were included.The criteria for ET for the management of hyperbilirubinemia were followed as per American Academy of Pediatrics 2004 guidelines 6 .Neonates who had undergone exchange transfusion for reasons other than hyperbilirubinemia and par al ET were excluded from the study.The protocol of the study was approved by Ins tute ethical review board and informed consent was taken from parent of each neonate.
Detailed baseline characteris cs and clinical fi ndings were recorded in a pre-designed proforma.Cases were subjected to inves ga ons which included hematocrit, hemoglobin, total and diff eren al leukocyte counts, platelets count and peripheral blood smear, re culocyte count and direct Coombs test for features of hemolysis.ABO and Rh blood grouping of both baby and mother were done.Biochemical parameters included blood glucose, total and direct serum bilirubin and ionized calcium levels.
Rh disease was defi ned as jaundice in Rh posi ve newborns born to Rh nega ve mothers with elevated an body tres to the Rh an gen and evidence of hemolysis.ABO disease was a ributed as cause of jaundice in newborns with posi ve direct Coombs test against the A or B an gens from type O mothers.An infant was considered healthy if jaundice was the only problem and the baby was ac ve and taking feeds adequately, as opposed to sick neonates who had ongoing illnesses such as birth asphyxia, meconium aspira on syndrome, respiratory distress syndrome that required treatment prior to or at the same me ET.The adverse events which occurred during and within 3 days of the procedure were observed and managed accordingly.
The push-pull method of double volume exchange transfusion (160 ml/kg) with stored whole blood of <5 days dura on having citrate-dextrose phosphate adenosine-1 (CDPA) an coagulant was used.Whole blood collected in CDPA compa ble to mother and baby was used.Removal and infusion of small aliquots of blood was performed in all neonates through umbilical vein catheteriza on as per the standard published guidelines 8 .The catheter was removed a er the procedure.Peripheral vein was not used in any pa ent 9 .
In sick neonates, suppor ve treatment in the form of intravenous fl uids, O 2 inhala on, empiric an bio cs (Ampicillin + Amikacin) were given.The an bio cs were changed a er culture report as per an microbial sensi vity pa ern.Vasopressors (Dopamine and / or Dobutamine) were used wherever required.Biochemical abnormali es such as hypoglycemia and hypocalcemia were corrected before taking up the ET.However, no intermi ent infusions of calcium were given during the procedure as per our NICU protocol.The IVIg was not used in any neonate with Rh isoimmuniza on because of high cost and nonaff ordability in our pa ents.

Monitoring
The neonates were monitored during and 3 days following ET for heart rate, respiratory rate, temperature, SpO 2 and ECG.Phototherapy was uniformly con nued in post ET period ll the level decreased to a safe limit.Serum bilirubin, hematocrit, platelets, blood glucose and ionized calcium were measured post and 6 hr post ET period.ET-related adverse events were defi ned as any complica on which was not present before ET and occurred during and within three days a er the exchange.Complica ons observed were thrombocytopenia (platelet count <100000/mm3), hypocalcemia (plasma ionized calcium <1 mmol/L), hypoglycemia (serum glucose <40 mg/dL), apnea and cardiorespiratory arrest.

Statistical analysis
Data were analyzed using the SPSS version 16.0.Chi-square and Fisher Exact tests were applied to compare the data of propor ons and Student's t-test was used for quan ta ve variables.A p value of less than 0.05 was considered as sta s cally signifi cant.

Results
There were 120 neonates, in whom 139 ETs were performed.Demographic characteris cs of the neonates are shown in Table 1.There were 53 preterm and 67 term neonates.The term neonates had signifi cantly higher mean gesta on, body weight and required longer dura on of ET than preterm (p<0.001).The other parameters such as age at ET, gender, race, birth order and place of delivery were comparable between preterm and term neonates.
The mean serum bilirubin levels in study subjects are presented in Table 2.The bilirubin levels showed signifi cant decline in post ET and 6 hr post ET period in comparison to pre-ET value (p<0.001).The mean reduc ons in serum bilirubin levels were 14.2±3.9mg/ dL in post-ET and 7.7±4.1 mg/dL in 6 hr post-ET period (Fig. 1).The term neonates had signifi cantly higher levels of mean serum bilirubin at all stages (pre-ET, post ET and 6 hr post ET) when compared with preterm (p< 0.001).However, the degree of reduc on in serum bilirubin level was comparable between preterm and term babies.
Of 120 neonates, adverse events occurred in 19 (15.8%) during the ET.The common events were respiratory distress in 13 (10.8%)and bradycardia in 8 (6.7%) cases.Apnea, cardio-respiratory arrest and hypoglycaemia were observed in 2.5%, 1.7%, and 0.8% of cases, respec vely.No signifi cant diff erences in the occurrence of events between preterm and term babies were observed during ET.
The neonates had more than one adverse events and thus 203 adverse events occurred in 120 babies (ra o being 1.69) during three days following ET, as shown in Table 3.The common adverse events were hypocalcemia in 98.3% (symptoma c in 14.2%) and thrombocytopenia in 34.2% (symptoma c in 1.7%) of neonates.Other events such as hypoglycemia (5%), respiratory distress (9.2%) and apnea (5.8%) also occurred following exchange transfusion.Almost equal propor on of preterm and term babies developed adverse events.Table 4 shows adverse events in healthy and sick neonates following 3 days of exchange transfusion.The sick neonates had propor onately higher incidence of hypocalcemia (11), thrombocytopenia (11), apnea (5), respiratory distress (4), and cardio-respiratory arrest (4) in comparison to healthy babies.In sick neonates, there were signifi cant fall in hematocrit in post ET (p= 0.038) and ionised calcium in 6 hr post ET period (p= 0.013) in comparison to healthy neonates.

Discussion
Early detec on and treatment of neonatal hyperbilirubinemia are essen al in the preven on of bilirubin encephalopathy.The ET is an emergency procedure and the mean age at which it was performed was 5.5 days in the present study.This was somewhat late than reported by previous workers 10,11,12 because pa ents repor ng to our centre comes from far off places and there is late referral from peripheral health centres also.
The common causes of hyperbilirubinemia were prematurity and ABO incompa bility (30.8% each).Steiner et al. 7 found that about 28% of cases of ET was due to ABO se ng.However, Behja et al. 13 found ABO (52%) and Rh (12%) incompa bility in higher propor on of their cases.Idiopathic group cons tuted the second most common cause.It may be possible that we may not be able to fi nd any a ributable cause in neonates with indirect hyperbilirubinemia and the range of idiopathic e ology varied from 13.9% to 36.4% in diff erent studies earlier 11,12,14 .Majority of neonates were breast fed, as it is a regular prac ce to have exclusive breast feeding.
The ET was performed once in majority (88.3%) of neonates and more than once in 11.7% of cases.The requirement of more than one ET was also reported by other workers 12,15 .Thirteen cases required ET 2-3 mes and the major e ology was ABO incompa bility while one baby required ET fi ve mes and the e ology was Rh isoimmuniza on.The IVIg could not be used in Rh isoimmunisa on due to non-aff ordability.
As regard to effi cacy of ET, overall there was signifi cant decrease in mean serum bilirubin level in post ET and 6 hr post ET period (p<0.001).The similar trend was observed when comparisons were made in preterm and term babies.Further, we found that overall percentage reduc on in serum bilirubin level was 58% from pre -ET level.Sakha et al. 14 reported 61% reduc on in bilirubin.By contrast, Bhat 16 observed lower percent reduc on in serum bilirubin level (40.3%) in their study.Further, there was a rise in serum bilirubin level in 6 hr post ET period despite con nued phototherapy.The overall net reduc on in serum bilirubin level was 31% in 6 hr post ET period.Thus, ET removes about 60% of bilirubin from plasma, resul ng in a clearance of 30 to 40% of bilirubin 17 .Hypocalcemia (98.3%) and thrombocytopenia (34.2%) were the major adverse events following 3 days of ET, but mostly they were asymptoma c.Other authors 12,14 have observed much lower incidence of hypocalcemia (2.5-38%).Unusually very high incidence of hypocalcemia observed in our study could be because of the fact that we did not use prophylac c calcium administra on during the procedure unless it became symptoma c.Also, hypocalcemia has been associated with the use of CPDA-1 as an an coagulant due to the binding of calcium by citrate.Thus, it appears that possibly these neonates require calcium supplementa on during the procedure to avoid the occurrence of hypocalcemia The need for giving supplemental calcium is controversial.However, if 0.5 to 1.0 ml of 10% intravenous calcium gluconate a er each 100 ml of exchange blood is used, it may reduce the incidence of hypocalcemia 18 .Thrombocytopenia (34.2%) was another major adverse event observed and the exact cause remained uncertain.Besides ET, phototherapy can also cause thrombocytopenia and both together can lead to decrease in platelets count in these babies.However, treatment should be given due considera on especially if the neonates develop bleeding manifesta ons.The other events were hypoglycaemia, respiratory distress, apnea and cardio-respiratory arrest; posing addi onal problems requiring close monitoring during the procedure.
It was further observed that sick neonates had signifi cantly higher incidence of thrombocytopenia, respiratory distress and life threatening events such as apnea and cardiorespiratory arrest in comparison to healthy babies; emphasizing the fact that they had already disturbed physiological status and ET further complicated the course.Therefore adequate a en on should be paid for occurrence of such adverse events in sick neonates.All cases required ET once.However, Jackson 19 reported that sick neonates needed mul ple ETs in their studies.
Seven neonates had features of early bilirubin encephalopathy at presenta on, requiring urgent ET.Rh and ABO incompa bility were present in three cases and no cause could be ascertained in four cases.Neurological features disappeared in majority of cases following ET; indica ng the early performance of ET to avoid brain damage and future neurological sequelae.
Five neonates (4.2%) died within 24 hr of ET.Other studies have observed lower mortality rate (0.5-2.8%) 20,21 .The high mortality observed in our study could be due to that all of them were outborn and mostly low birth weight babies.The two neonates had features of bilirubin encephalopathy at presenta on among non-survivors.They also had higher adverse events such as hypocalcemia, hypoglycemia and life threatening events (apnea and bradycardia).Therefore, adequate precau ons are to be undertaken to correct such metabolic abnormali es in order to prevent lifethreatening complica ons and reduce mortality.

Conclusion
In conclusion, although the rate of adverse events associated with exchange transfusions was high, most events were asymptoma c.Hypocalcemia was found in almost all neonates followed by thrombocytopenia.Therefore intravenous calcium infusion during ET procedure may be required.However, despite these considera ons, ET s ll remains an eff ec ve modality in reducing the neonatal hyperbilirubinemia and preven ng bilirubin encephalopathy in developing countries where sick neonates are referred quite late to a ter ary-care centre.

Table 1 :
Basic demographic of characteris cs pa ents according to gesta onal age (n=120)

Table 4 :
Adverse events following 3 days of ET in sick and healthy neonates