Background: Severe neonatal jaundice (SNJ) causes long-term neurocognitive impairment, cerebral palsy, auditory neuropathy, deafness, or death. We developed a mathematical model for allo-hemodialysis, a novel potential treatment for SNJ.
Methods: With allo-hemodialysis, the neonate’s blood flows through hollow fibers of a miniature 0.075m2 hemodialyzer, while the blood of a healthy adult (“buddy”) flows counter-currently through the dialysate compartment. Kinetics of unconjugated bilirubin in allo-hemodialysis were simulated with neonate blood flow rates of 12.5 and 15mL/min (for a 2.5kg and 3.5kg neonate, respectively), and 30mL/min for the buddy. We only simulated unconjugated bilirubin kinetics because the conjugated form is easily excreted and non-toxic. The bilirubin production rates in neonate and buddy were set to 6 and 3mg/kg/day, respectively. Neonatal bilirubin conjugation was set to zero. Buddy bilirubin conjugation rate was calculated to obtain normal steady state bilirubin levels.
Results: Model simulations suggest that a 6-hour allo-hemodialysis session can reduce neonatal bilirubin levels by more than 35% and that this modality is particularly effective in neonates with low serum albumin levels. Also, neonatal extracorporeal blood flow affects the bilirubin kinetics. Due to the high bilirubin conjugation capacity of an adult’s healthy liver and the larger distribution volume, the buddy’s bilirubin level increases only transiently during allo-hemodialysis.
Conclusions: Our modelling suggests that a single allo-hemodialysis session lowers neonatal unconjugated bilirubin levels effectively. If corroborated in ex-vivo, animal, and clinical studies, this bilirubin reduction could lower the risks associated with SNJ, especially kernicterus, and possibly avoiding the morbidity associated with exchange transfusions.