|Year : 2019 | Volume
| Issue : 2 | Page : 79-84
Adverse events following blood exchange transfusion for neonatal hyperbilirubinemia: A prospective study
Swathi Chacham, Jogender Kumar, Sourabh Dutta, Praveen Kumar
Department of Pediatrics, Division of Neonatology, Postgraduate Institute of Medical Education and Research, Chandigarh, India
|Date of Web Publication||25-Apr-2019|
Prof. Praveen Kumar
Department of Pediatrics, Division of Neonatology, Postgraduate Institute of Medical Education and Research, Chandigarh - 160 012
Source of Support: None, Conflict of Interest: None
Background: Exchange transfusion (ET) for hyperbilirubinemia is associated with many complications. The complications are underreported as most of the published studies are retrospective, used varying definitions of adverse events (AEs) and variable follow-up periods. Aim: We evaluated the incidence of clinical, biochemical, hematological, and radiological AEs, including serious AEs, within 2 weeks of ET for hyperbilirubinemia in neonates, using standard definitions. Materials and Methods: This prospective observational study was conducted in level III newborn unit of north India from February 2008 to February 2009. We enrolled consecutive inborn and outborn neonates admitted with hyperbilirubinemia and required ET. Babies requiring partial exchange for anemia/polycythemia or ET for indications other than hyperbilirubinemia were excluded. They were prospectively monitored for clinical, biochemical, hematological, and radiological AEs up to 2 weeks following the procedure. We calculated the incidence/AE rate (AER) as the rate of events per 100 ET and compared them among various groups using the Chi-square test. The SPSS v20 was used for the analysis, and value of P < 0.05 was considered as statistically significant. Results: A total of 202 neonates were screened and 141 neonates (182 ET) were enrolled. The overall AER was 112/100 ETs. The most common AE was biochemical (45.6/100 ET), followed by hematological (44.5/100 ETs), clinical (15.9/100 ETs), and radiological (8.9/100 ETs). Severe AER was 12.6/100 ETs. The AER was significantly more in lower gestation and birth weight groups. Conclusion: ET is a high-risk procedure and should be performed only when the benefit of the procedure offsets the risks.
Keywords: Adverse events, biochemical adverse events, clinical adverse events, exchange transfusion, hematological adverse events, hyperbilirubinemia, neonate
|How to cite this article:|
Chacham S, Kumar J, Dutta S, Kumar P. Adverse events following blood exchange transfusion for neonatal hyperbilirubinemia: A prospective study. J Clin Neonatol 2019;8:79-84
|How to cite this URL:|
Chacham S, Kumar J, Dutta S, Kumar P. Adverse events following blood exchange transfusion for neonatal hyperbilirubinemia: A prospective study. J Clin Neonatol [serial online] 2019 [cited 2020 Feb 28];8:79-84. Available from: http://www.jcnonweb.com/text.asp?2019/8/2/79/257144
| Introduction|| |
Blood exchange transfusion (BET) was introduced in the late 1940s to decrease mortality and morbidity associated with hemolytic disease of the newborn, but subsequently, it was used for the treatment of severe hyperbilirubinemia due to any cause. In due course, its use extended to the treatment of other conditions such as severe sepsis, drug intoxication, hydrops fetalis, hyperammonemia, and refractory hyperkalemia. However, tremendous progress in prenatal (intrauterine transfusion and anti-D immunoglobulin) and postnatal care of Rh iso-immunized fetuses along with efficient phototherapy has substantially reduced the need for BET over the years. The frequency of BET in developed countries has decreased to such an extent that with a decrease in experience with this age-old procedure, there is a risk of increase in the number of BET-related complications. In developing countries of Asia and Africa, despite advancements in phototherapy, the rates of BET still continue to be high, as several neonates are brought late to the hospital when they already have extreme hyperbilirubinemia and/or signs of acute bilirubin encephalopathy (ABE).
BET; however, can lead to complications such as life-threatening bleeding, sepsis, cardiac arrhythmias and even death, apart from transient hypocalcemia, hyperkalemia, bradycardia, and thrombocytopenia., The adverse events (AEs) associated with BET may be attributable to fluctuations in blood volume and blood pressure,changes in the acid–base status, alterations in platelet count due to use of packed red cells which lack platelets and coagulation factors, electrolyte disturbances and the introduction of infectious agents. Although BET has been considered as a major procedure akin to major surgery, accurate information regarding the incidence of these morbidities is lacking. Most of the published studies are retrospective, have used varying definitions of AEs and have had inconsistent follow-up periods.,, It is vital to have accurate information about AEs, using standard definitions for fair comparisons with any newer modality of treatment and to evaluate any improvement strategies. Hence, we prospectively assessed the incidence of clinical, biochemical, hematological, and radiological AEs, including serious AEs (SAEs), occurring within 14 days of BET for hyperbilirubinemia in neonates, using standard definitions.
| Materials and Methods|| |
This prospective observational study was conducted in a level III neonatal unit in north India. All consecutive inborn, as well as outborn neonates admitted in the newborn unit for a single or double volume ET, were screened. Neonates requiring BET for hyperbilirubinemia were enrolled after getting informed written consent from one of the parents. For phototherapy and BET, we followed the American Academy of Pediatrics 2004 recommendations for the infants ≥35 weeks of gestation and Maisel's charts for preterm infants <35 weeks gestation. We excluded babies who underwent partial ET for anemia or polycythemia, or who underwent BET for indications other than hyperbilirubinemia. We prospectively monitored all enrolled neonates for clinical, biochemical, hematological, and radiological AEs for 14 days following BET. Ethical clearance was obtained from the Institute Ethics Committee.
Fresh (<5 days old) compatible blood, cross-matched with both mother and baby was used. All BETs were double volume. Patient demographics, cause of hyperbilirubinemia, indication for BET, comorbidities which could influence the AEs, details of donor blood (blood group, age, and biochemical parameters), details of the procedure (personnel, route, and duration), clinical, hematological, and biochemical monitoring (before, during and after procedure), neurological status and BET-related complications were recorded in predesigned performa. Neonates were investigated for biochemical parameters (ionized calcium, sodium, potassium, pH, and bicarbonate) before the procedure and at one, two, six, 12, and 24 h after BET. Platelet count was assessed before, at 24 h and 72 h after BET. In neonates, who underwent repeat BET within 24 h of the first exchange, platelet count was repeated at 24 and 72 h after the last BET. In preterm neonates undergoing BET, ultrasound head was performed before and at six and 24 h after BET, to look for intracranial/intraventricular hemorrhage (IVH). Blood glucose was checked 2 h after the procedure.
Neonates were actively monitored for clinical AEs (CAEs) such as sepsis, apnea, bleeding, cardiac arrest, and seizures manifesting immediately after the procedure and until 14 days.
Definitions used in the study
Neonate who did not have any associated morbidity, and who did not require any supplemental oxygen and who had normal hemodynamic and neurologic status at admission.
Neonate who had other comorbidity or who was on supplemental oxygen or other respiratory support or had critical hemodynamic status or abnormal neurological status at admission.
Any abnormal alteration in the physiologic status and/or variation in the laboratory parameters (done in relation to the procedure) during or within 14 days of performing BET, which normally would not have occurred in the absence of the procedure.
AEs were further classified as follows:
Clinical adverse events
Apnea, bradycardia, seizures, hypothermia, new onset/increase in need of respiratory support, hypotension, significant clinical bleeding from any site, sepsis, arrhythmia, and death.
Biochemical adverse events
Hypocalcemia (ionized Ca <1 mmol/L); hyponatremia (Na+<135 meq/L); hyperkalemia (K+ >6.5 meq/L with ECG changes); acidosis (pH < 7.20 or base excess >-10 meq/L).
Hematological adverse events
Thrombocytopenia (platelet count <100,000/mm3); severe thrombocytopenia (platelet count <50,000/mm3). In neonates with preexchange platelet count <100,000/mm3, >2% fall from baseline was considered an HAE, as the coefficient of variation of the platelet analyzer was two percentages.
Serious adverse event
Any AE that resulted in death need for readmission, prolongation of existing hospitalization, a persistent or significant disability or incapacity within 14 days of BET.
Blood exchange transfusion-related mortality
Any death which occurred within 14 days after BET was considered as BET-related mortality. The attribution to BET was further defined as: definite if the AE was clearly related to the BET; possible if the AE was likely related to the BET; probable if the AE may be related to the BET; and unlikely if the AE was doubtfully related to the BET.
Acute bilirubin encephalopathy
Acute clinical manifestations of elevated bilirubin in the central nervous system manifesting as lethargy, poor feeding, hypo/hypertonia, high-pitched cry, torticollis, opisthotonus, setting-sun sign, and seizures, etc. It was further classified into mild, moderate, and severe ABE based on the bilirubin-induced neurologic dysfunction score.
From previously reported studies, the incidence of AEs following blood ET varied from 12% to 44%. To detect 12% AEs, with an allowable error of 5%, (alpha error of 5% and power of 80%), a total of 168 BETs needed to be studied.
We described categorical variables as percentages, normally distributed numerical variables as mean (standard deviation), and numerical variables with skewed distributions as median (1st and 3rd quartile). Skewness of distributions was determined by the Shapiro–Wilk test and Q–Q plot. We calculated the incidence of AE rate (AER) as the number of events per 100 ET and compared them in various groups using the Chi-square test. The Statistical Package for the Social Sciences version 20 (IBM Inc., Armonk, NY, USA) was used for the analysis, and P < 0.05 was taken as statistically significant.
| Results|| |
A total of 202 neonates underwent 257 BETs during the study period and were screened for possible inclusion. Of them, 61 neonates were excluded due to various reasons [Figure 1] and 141 were enrolled. These 141 neonates underwent 182 BETs. Out of 141, 105 (74%) underwent one, 31 (22%) underwent two, and rest five (4%) underwent three BETs. The baseline characteristics of the study population are described in [Table 1]. Most of the neonates were outborn as our center serves as the referral institute for neighboring states. The most common causes for hyperbilirubinemia were G6PD deficiency (n-38,27%), Rh isoimmunization (n-33,23.4%), and ABO incompatibility (n-17,12.1%). In 53 (37.6%) infants, no cause could be determined (labeled idiopathic). Forty (28.4%) neonates had ABE at presentation and most (80%) had mild encephalopathy.
AEs were observed in 96 (68.1%) neonates. There were a total of 204 AEs following 182 BETs, which included 83 (40.6% of total AEs, n-60 neonates) biochemical AEs (BAEs), 81 (39.7% of total AEs, n- 81 neonates) hematological AEs (HAEs), 29 (14.2% of total AEs, n-27 neonates) CAEs, and 11 (5.4% of total AEs, n-11 neonates) radiological AEs (RAEs). Therefore, the AER in the study population was 112/100 BETs. BAEs occurred in 60 (42.5%) neonates (45.6/100 BETs) and the most common event was hypocalcemia (n-41) followed by hypernatremia (n-26), hyponatremia (n-8), hyperkalemia (n-7), and acidosis (n-1). All BAEs except hyperkalemia occurred most commonly at 1 h post-BET and showed a declining trend over the next 24 h [Table 2]. The incidence of hyperkalemia was highest at 2 h post-BET and showed improvement by 6 h post-BET. Thrombocytopenia was observed in 81 (57.4%) neonates (44.5/100 BETs), of which 25 (17.7%) had severe thrombocytopenia. The incidence of thrombocytopenia was the highest at 24 h post-BET and improved by 72 h. None of the infants had clinical bleeding. Twenty-nine CAEs occurred in 27 (19.2%) neonates (15.9/100 BETs). Common benign AEs were jitteriness due to symptomatic hypocalcemia (n-3) followed by symptomatic hyperkalemia (n-1), hypothermia (n-1), and multiple soft-tissue abscesses (n-1). The rest of the 23 CAEs were SAEs that occurred in 23 (16%) neonates (12.6/100 ETs). These included death in nine, sepsis in seven (meningitis-two, culture positive-four, and culture negative-one), seizures in five, and cardiac arrest and intravascular hemolysis in one neonate each. All nine (6.4%) neonates who died (7.5/100 BETs) within 14 days of BET were sick before the procedure. Out of nine, one death was classified as possibly related, five as probably related, and rest three as unlikely to be due to the procedure. Of the infants who died, three died within 6 h of procedure, four between 24 and 48 h, and rest two between 1 and 2 weeks after the procedure.
Only preterm neonates (n-66) were screened for RAEs. Eight (9.1%) babies had IVH (8.9 per 100 BETs). Seven out of these eight were sick neonates, and none of them had severe (grade III/IV) IVH. Two-third of all IVH was observed within 6 h post-BET.
AERs were significantly higher in lower gestation and birth weight groups [Table 3]; and sick neonates [Table 4]. Among the sick neonates, those with ABE at admission had more AEs as compared to those without encephalopathy.
|Table 3: Distribution of rate of adverse events (/100 blood exchange transfusion) according to sickness at admission|
Click here to view
|Table 4: Comparison of adverse event rate across various gestation and weight groups|
Click here to view
We also recorded donor blood and BET procedure-related details. Median age of the donor blood was 2 (1–3) days, mean donor blood pH was 6.89 ± 0.98, HCO3 was 13.89 ± 4.3 meq/L, SBE was −15.39 ± 6 meq/L, potassium was 5.75 ± 1.43 meq/L, sodium was 149.38 ± 6.3 meq/L, and ionized calcium was 0.03 ± 0.05 mmol/L. Mean duration of the procedure was 75.22 ± 25 min (range 60–90 min). Mean aliquot volume (ml/kg) was 3.58 ± 0.98. Majority (53.5%) of BETs were performed by the 3rd year postgraduate residents, followed by 2nd year residents (36.5%). All were supervised by neonatology fellows. Umbilical venous route n = 157 (86.3%) followed by peripheral arterial route n = 24 (13.1%) was used for BET. One neonate (0.6%) had undergone BET through the umbilical artery catheter and peripheral vein. Twenty-two neonates (15.8%) had catheter block during the procedure and required reinsertion.
| Discussion|| |
In this prospective study of 182 BETs in 141 neonates, AEs were assessed systematically for 14 days. Overall AER was 112/100 BETs. Most common AE was biochemical (45.6/100 BETs), followed by hematological (44.5/100 BETs), clinical (15.9/100 BETs), and radiological (8.9/100 BETs). Severe AER was 12.6/100 BETs. The AER was significantly higher in infants with lower gestation, lower birth weight, and sickness. We included only those neonates who required BET primarily for neonatal hyperbilirubinemia as the AEs following BET for other indications, for example, sepsis, hydrops, an inborn error of metabolism, etc., will be quite different because of the underlying conditions and many confounders. We included all gestations and also those neonates with who were sick, but BET was being done for the primary indication of hyperbilirubinemia, to improve generalizability of our results. We stratified the results according to gestation, weight, and sickness at admission for better comparison with other studies., Ours being a large government tertiary care center in north India, the majority of enrolled neonates were outborn.
Previous studies either did not describe any time limit, or used a time frame of 3–7 days for reporting procedure-related AEs.,,, In the present study, we defined the limit of 14 days for reporting of AEs in concordance with standard surgical care guidelines to make it more uniform and comparable. The incidence of AEs was quite high in the present study. This can be explained by the fact that due to multimodal monitoring, >1 AE was detected in the majority of the neonates. Furthermore, due to the prospective nature of the study, meticulous protocolised monitoring, more AEs were identified, which could have been otherwise missed by retrospective chart reviews. Similar rates of AE have been reported from recent prospective studies done in the Southeast Asian region and older studies from the western countries.,, Two prospective studies from Nepal reported AER of 146 and 210/100 BETs, respectively., In a study by Patra et al., 74% of BETs were associated with AEs and AER was 136/100 BETs. Hakan et al. reported the largest prospective series from Turkey involving 306 patients. In this Turkish series, 337 BET-related complications were observed in 306 patients (337 BETs), with an AER of 100/100 BETs. However, compared to the present study, their individuals were healthy and of higher gestation with assessment of AE up to 7 days only. On the contrary, complication rates reported by Steiner et al. were quite low (22.7/100 BETs) as compared to the previous studies. These differences can be due to variation in the definitions used, level of sickness among the participants, health-care system in various countries, and delayed referral in the developing countries. All the studies showed that although BET is a commonly performed procedure in neonates, it is not free of risks and is associated with significant complications.
In the present study, biochemical and hematological complications were the most common AEs. Previous studies have also reported similar findings.,,,, Hypocalcemia was one of the most frequent AEs with the incidence ranging from 22.5% to 98%.,,,, Hypocalcemia following BET is due to the calcium chelating properties of citrate, which is present in a very high concentration in the donor blood as a component of anticoagulant. The citrate in donor blood binds ionic calcium and magnesium in the neonate and produces significant hypocalcemia., Thrombocytopenia was seen in 57.4% neonates undergoing BET, of which almost one-third had severe thrombocytopenia. A similar frequency of thrombocytopenia was reported in other studies.,,,, We also observed that the incidence of thrombocytopenia was highest at 24 h post-ET and showed improvement by 72 h. Jasso-Gutiérrez et al. studied the pattern of recovery of platelet count after BET. Like us, they also observed a decrease in platelet count following BET with a nadir at 24 h and full recovery at or after 72 h. This finding has important implications in management and mandates screening for this, at least in preterm infants. Post-BET thrombocytopenia is due to the replacement of whole blood with packed red blood cells, which are deficient in platelets. Furthermore, the process of BET can activate the coagulation system, leading to vascular thrombosis and consumptive thrombocytopenia. The all-cause mortality rate in the present study was 7.5/100 BETs. However, only one death was directly related to the procedure (0.8/100 BETs). This rate is comparable to other studies,,,, and is lower than previously reported from India., All the neonates who died were sick before BET. As expected, AEs were more frequent in the infants of lower gestation and birth weights and those who were sicker. These results are similar to the other studies, and due to the fact that sicker and smaller infants have multiple comorbidities, and hence, are at risk of more complications.
The strength of the present study is that it was a prospective study of a large number of BETs with a preplanned protocol for monitoring, with clear standard definitions of AEs and a follow-up of 14 days. However, we limited our laboratory testing to not so frequent intervals to minimize the blood sampling. We were able to provide AERs for various groups of neonates undergoing BET for the primary indication of hyperbilirubinemia. These AERs can be used for quality improvement as well as comparison with any other treatment options for critical hyperbilirubinemia.
| Conclusion|| |
Although BET is a commonly performed procedure in the management of severe hyperbilirubinemia, the frequency of associated AEs is quite high. AEs are more common in infants of lower gestation, weight, and those who are sicker. A clear protocol of monitoring of these AEs can help early detection and treatment of these complications.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Steiner LA, Bizzarro MJ, Ehrenkranz RA, Gallagher PG. A decline in the frequency of neonatal exchange transfusions and its effect on exchange-related morbidity and mortality. Pediatrics 2007;120:27-32.
Stockman JA 3rd
. Overview of the state of the art of rh disease: History, current clinical management, and recent progress. J Pediatr Hematol Oncol 2001;23:385-93.
Maisels MJ. Phototherapy – Traditional and nontraditional. J Perinatol 2001;21 Suppl 1:S93-7.
Thayyil S, Milligan DW. Single versus double volume exchange transfusion in jaundiced newborn infants. Cochrane Database Syst Rev 2006;(4):CD004592.
Murki S, Kumar P. Blood exchange transfusion for infants with severe neonatal hyperbilirubinemia. Semin Perinatol 2011;35:175-84.
Hakan N, Zenciroglu A, Aydin M, Okumus N, Dursun A, Dilli D, et al.
Exchange transfusion for neonatal hyperbilirubinemia: An 8-year single center experience at a tertiary neonatal intensive care unit in turkey. J Matern Fetal Neonatal Med 2015;28:1537-41.
American Academy of Pediatrics Subcommittee on Hyperbilirubinemia. Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics 2004;114:297-316.
Maisels MJ, Watchko JF, Bhutani VK, Stevenson DK. An approach to the management of hyperbilirubinemia in the preterm infant less than 35 weeks of gestation. J Perinatol 2012;32:660-4.
Iskander I, Gamaleldin R, El Houchi S, El Shenawy A, Seoud I, El Gharbawi N, et al.
Serum bilirubin and bilirubin/albumin ratio as predictors of bilirubin encephalopathy. Pediatrics 2014;134:e1330-9.
Patra K, Storfer-Isser A, Siner B, Moore J, Hack M. Adverse events associated with neonatal exchange transfusion in the 1990s. J Pediatr 2004;144:626-31.
Jackson JC. Adverse events associated with exchange transfusion in healthy and ill newborns. Pediatrics 1997;99:E7.
Badiee Z. Exchange transfusion in neonatal hyperbilirubinaemia: Experience in Isfahan, Iran. Singapore Med J 2007;48:421-3.
Malla T, Singh S, Poudyal P, Sathian B, Bk G, Malla KK, et al.
Aprospective study on exchange transfusion in neonatal unconjugated hyperbilirubinemia – in a tertiary care hospital, Nepal. Kathmandu Univ Med J (KUMJ) 2015;13:102-8.
Jain A, Puri D, Faridi MM. Biochemical changes during exchange transfusion in hyperbilirubinemia in term newborn babies. Indian J Clin Biochem 1997;12:119-24.
Falciglia HS, Greenwood CS. Double volume exchange transfusion: A review of the “ins and outs.”. NeoReviews 2013;14:e513-20.
Jasso-Gutiérrez L, Rizo-Hernández A, de la Rosa L. Effects of exchange transfusion on platelet counts. Arch Invest Med (Mex) 1981;12:297-306.
Narang A, Gathwala G, Kumar P. Neonatal jaundice: An analysis of 551 cases. Indian Pediatr 1997;34:429-32.
Dikshit SK, Gupta PK. Exchange transfusion in neonatal hyperbilirubinemia. Indian Pediatr 1989;26:1139-45.
Mohan VM, Bhakoo ON. Bacteremia following exchange transfusion in the newborn. Indian Pediatr 1980;17:334-8.
[Table 1], [Table 2], [Table 3], [Table 4]