Abstract
Exchange transfusion remains the definitive treatment for severe neonatal hyperbilirubinaemia; yet its aetiology and outcomes in rural tropical settings are poorly characterised. This five-year retrospective analysis of 58 neonates at a North Indian rural tertiary centre reveals that Rh incompatibility accounted for 76% of cases – a striking divergence from high income-country patterns where ABO incompatibility predominates – indicating critical gaps in antenatal Rh immuno-prophylaxis. The procedure achieved a mean bilirubin reduction of 50% (424 ± 106 to 212 ± 70&mu/L, p < 0.001) with no procedure-related mortality, though clinically significant adverse events occurred in 19% of neonates. Notably, 10.2% presented with acute bilirubin encephalopathy at the time of intervention, representing potentially preventable neurological injury. These findings make a compelling case for urgent, systematic improvements in antenatal screening, Rh immuno-prophylaxis access, and early jaundice recognition in tropical resource-limited settings.
Keywords
Introduction
Neonatal hyperbilirubinaemia affects approximately 60% of newborns worldwide, with severe cases potentially causing irreversible neurological damage if managed inadequately.1,2 In tropical and low-income countries, delayed recognition combined with limited access to intensive phototherapy contributes to higher rates of extreme hyperbilirubinaemia requiring exchange transfusion. 3 This invasive procedure remains the definitive treatment when total serum bilirubin approaches neurotoxic levels or when conventional therapy proves insufficient. 4
Haemolytic disease of the foetus and newborn, primarily due to blood group incompatibilities, constitutes a major indication for exchange transfusion in resource-limited settings. 5 The procedure aims to remove circulating bilirubin and maternal antibodies rapidly whilst correcting anaemia and preventing acute bilirubin encephalopathy. 6 Despite advances in phototherapy technology, exchange transfusion remains indispensable for managing severe cases, particularly where intensive phototherapy may be unavailable or inadequate. 7
Contemporary literature from Indian tertiary centres demonstrates variable exchange transfusion rates and outcomes, influenced by referral patterns, antenatal care quality, and institutional protocols.8,9 Understanding local practice patterns is crucial for optimising neonatal care and identifying preventable causes of severe hyperbilirubinaemia in tropical settings.
Our study characterises the clinical profile, aetiological spectrum, and immediate outcomes of neonates undergoing exchange transfusion at our tertiary facility, providing insights into current practices and identifying areas for quality improvement in resource-constrained environments.
Materials and methods
Our retrospective observational study was conducted at a tertiary care teaching hospital in rural North India managing approximately 1200 neonatal admissions annually, serving a 200-kilometre referral radius. All neonates undergoing exchange transfusion between January 2019 and December 2023 were included. Of 63 neonates who underwent the procedure, five (7.9%) were excluded owing to incomplete documentation of bilirubin values, demographic data, or follow-up; available data from excluded cases suggested random distribution across the study period with no systematic bias.
Structured data extraction captured demographics, gestational age, birth weight, blood group compatibility, Direct Coombs Test results, pre- and post-procedural laboratory values, procedure details, and complications within 24 h. Preterm birth was defined as gestational age below 37 completed weeks; low birth weight as below 2500 g. Immediate adverse events were defined as clinically significant complications within 24 h, including thrombocytopenia (moderate: 50–100 × 109/L; severe: < 50 × 109/L), hypocalcaemia, electrolyte disturbances, haemodynamic instability, or catheter-related complications. 10
Exchange transfusion indications followed American Academy of Pediatrics guidelines adjusted for gestational age and risk factors.
During the study period, 63 neonates underwent exchange transfusion. Five neonates (7.9%) were excluded due to incomplete documentation: three with missing pre-procedure or post-procedure bilirubin values, one with incomplete demographic data (gestational age and birth weight not documented), and one transferred to another facility within 12 h with no follow-up data. From limited available information, 3/5 excluded cases had Rh incompatibility, one had ABO incompatibility of unknown aetiology. The low exclusion rate (7.9%) and similarity in available aetiological data suggest minimal selection bias, with excluded cases appearing randomly distributed across the study period rather than clustered.
Structured data extraction captured demographic characteristics, gestational age, birth weight, delivery details, maternal and neonatal blood groups, direct Coombs Test results, exchange transfusion indications, pre- and post-procedural laboratory values, number of procedures per patient, and documented complications within 24 h.
All procedures utilised reconstituted whole blood with exchange volumes of 170 ± 8 mL/kg over 2.5 ± 0.4 h. Central venous access was established in all cases. Concurrent interventions included intensive phototherapy (100%), intravenous immunoglobulin when indicated (30.5%), and prophylactic calcium supplementation (69.5%).
Continuous variables are expressed as mean ± SD or median (IQR); categorical variables as frequencies and percentages. Pre- and post-procedure values were compared using paired t-tests; categorical comparisons used chi-square or Fisher's exact tests. Statistical significance was p < 0.05 (SPSS v26.0). The study received Institutional Ethics Committee approval with waiver of informed consent given its retrospective, de-identified nature.
Results
During the 5-year study period, 58 neonates underwent 64 exchange transfusion procedures. Demographic characteristics are presented in Table 1.
Demographic and clinical characteristics (n = 58).
IQR: interquartile range; SD: standard deviation.
Of 58 neonates, 56.9% were preterm and 62.7% were low birth weight; the majority were delivered by Caesarean section and referred from peripheral facilities (Table 1). Rh incompatibility was the dominant aetiology (76%), followed by sepsis-related hyperbilirubinaemia (19%) and ABO incompatibility (5%); no cases of G6PD deficiency were identified (Table 2). Direct Coombs Test was positive in 47.5%. Among the 11 neonates with sepsis-related hyperbilirubinaemia, all failed intensive phototherapy and demonstrated progressive anaemia and clinical deterioration, with 54.5% showing signs of acute bilirubin encephalopathy prior to the procedure.
Aetiological distribution and laboratory parameters (n = 58).
*p < 0.001 compared to pre-ET values. ET: exchange transfusion; G6PD: glucose-6-phosphate dehydrogenase.
The decision to perform exchange transfusion in sepsis-associated hyperbilirubinaemia reflected: (1) concern for haemolysis contribution (particularly in Gram-negative sepsis), (2) extremely elevated bilirubin levels approaching neurotoxic range, (3) failed conventional therapy (intensive phototherapy + treatment of underlying infection), and (4) clinical deterioration suggesting imminent risk of kernicterus.
Exchange transfusion achieved a mean bilirubin reduction of 50%, with concurrent improvement in haemoglobin; platelet counts fell modestly post-procedure (Table 2). A single procedure was sufficient in 93%. The most common indication was total serum bilirubin exceeding the exchange threshold; six neonates (10.2%) presented with clinical signs of acute bilirubin encephalopathy at the time of intervention (Table 3).
Procedural characteristics and efficacy (n = 58).
TSB: total serum bilirubin; SD: standard deviation.
Clinically significant immediate adverse events occurred in 11 neonates (18.97%), with hypocalcaemia and thrombocytopenia requiring intervention being most frequent; no procedure-related deaths occurred (Table 4). Bilirubin reduction was consistent across gestational age groups (p = 0.87).
Immediate adverse events and clinical management (n = 58).
IV: intravenous.
Bold value is for analysing clinically significant thrombocytopenia.
Discussion
This five-year analysis confirms that exchange transfusion remains effective for severe neonatal hyperbilirubinaemia in rural tropical settings, achieving the 50% bilirubin reduction consistently reported in the literature, 11 with no procedure-related mortality. The principal finding, however, is not procedural but aetiological. The overwhelming predominance of Rh incompatibility (76%) contrasts sharply with series from high-income countries, where ABO incompatibility typically leads, 12 and with several Indian urban centre reports where the aetiological mix is more varied.
This pattern is not a reflection of biological difference but of a preventable programmatic failure: inadequate access to antenatal blood group screening and Rh immunoprophylaxis across our predominantly rural catchment area.
The demographic profile – predominantly preterm, low birth weight, outborn neonates referred from distances up to 200 kilometres – underscores the vulnerability of this population. That 10.2% of neonates arrived with clinical signs of acute bilirubin encephalopathy indicates that severe jaundice was not being identified or acted upon early enough at the peripheral level, whether due to limited screening, restricted phototherapy availability, or delayed care-seeking. These are not isolated failures but interconnected gaps in a system where the tertiary centre absorbs the consequences of deficiencies upstream.
Detailed analysis revealed critical failures at multiple system levels:
Recognition: mean age at presentation was 5.2 days (range: 3–8 days), indicating delayed identification of severe jaundice. Mean total serum bilirubin at presentation was 554 ± 72µmol/L (range: 478–667), well above exchange thresholds.
Access to phototherapy: all six (100%) were born at peripheral facilities. Only one received any form of phototherapy before referral. The median time from birth to first phototherapy was 72 h (IQR: 48–96 h), representing missed opportunities for prevention.
Knowledge and awareness: Five (83.3%) had documented home visits by healthcare workers who failed to recognise severe jaundice. Four (66.7%) had family members who reported concerns about deepening jaundice but were advised to wait. None received transcutaneous or serum bilirubin screening before symptom onset.
Access and referral: Mean time from the family's decision to seek care to arrival at tertiary centre was 18 h (range: 12–36), delayed by transportation challenges (200 km mean distance), financial barriers (arranging funds for transport and admission), and referral system inefficiencies.
These findings emphasise need for multi-level interventions: universal pre-discharge bilirubin screening, structured early postnatal follow-up within 48–72 h, clear referral protocols with defined bilirubin thresholds, enhanced phototherapy availability at district hospitals, community education about jaundice warning signs, and subsidised emergency neonatal transport systems.
Cases described highlight the complex interaction between infection, haemolysis, and bilirubin metabolism in critically ill neonates in tropical settings. The sepsis-related cases (19%) merit contextual interpretation. Exchange transfusion is not standard treatment for sepsis-associated hyperbilirubinaemia alone; however, the combination of extreme bilirubin levels, failure of conventional therapy, concurrent haemolysis, and impending encephalopathy justified intervention in these critically ill neonates. The predominance of Gram-negative organisms is consistent with tropical healthcare settings.
The complication rate of 19%, calculated using clinically meaningful thresholds, falls within the 15–25% range reported in contemporary literature.13,14 Mild dilutional thrombocytopenia – a nearly universal and self-resolving consequence of double-volume exchange – was appropriately excluded from this count. The absence of mortality supports the safety of the procedure when performed by experienced teams,14,15 though the higher overall complication rate probably reflects the severity of illness at presentation.
The core message of our study is that a procedure that carries inherent risk is being performed predominantly for a condition that is preventable. Strengthening antenatal Rh screening and immunoprophylaxis in rural North India would reduce exchange transfusion rates, complication burden, and the tragic toll of kernicterus in this region.
Limitations
Our study has inherent retrospective design limitations, including potential selection bias. The single-centre experience may limit generalisability. The higher complication rate observed warrants further investigation to identify modifiable procedural or patient factors. Long-term neurodevelopmental outcomes were not assessed, preventing evaluation of ultimate clinical effectiveness.
Conclusions
The burden of exchange transfusion at this rural North Indian centre is driven overwhelmingly by a preventable cause. That three in four procedures were performed for Rh haemolytic disease – and that one in ten neonates arrived with established bilirubin encephalopathy – reflects systemic gaps in antenatal care rather than failures at the bedside. The procedure itself performs reliably and safely in experienced hands. The imperative, therefore, is not to refine exchange transfusion technique but to reduce the need for it: through universal antenatal blood group screening, consistent access to Rh immunoprophylaxis, and strengthened early jaundice detection at the primary care level. Future research should quantify the neurodevelopmental impact on survivors and evaluate the cost-effectiveness of upstream preventive strategies in comparable resource-limited settings.
Footnotes
Acknowledgements
We extend our gratitude to dedicated SNCU nursing staff and paediatric residents whose meticulous patient care and data documentation and also to MRU (multi-disciplinary unit) at our institute UPUMS Saifai that made this investigation possible.
Author contribution
Muniba Alim and Jyoti Kala Bharati: conceptualisation, methodology, data curation, formal analysis, writing–original draft, and review and editing. Dinesh Kumar and Durgesh Kumar: supervision, validation, and writing–review and editing. All authors reviewed the results and approved the final version of the article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data availability statement
The dataset supporting the conclusions of this article is available upon reasonable request from the corresponding author.
Ethics committee approval
The study was approved by Institutional Ethics Committee. Written informed consent was waived due to the retrospective nature and use of de-identified data.