- The Pediatric End-stage Liver Disease (PELD) score underestimates the actual probability of 90-day waitlist mortality in children needing primary liver transplant.
Why this matters
- PELD score accurately ranked risk for children, but direct comparisons with the adult Model for End-stage Liver Disease (MELD) score were not accurate.
- “Children with chronic liver disease who are in need of transplant may be at a disadvantage compared with adults in a similar situation," the authors write.
- Retrospective study of 4298 pediatric patients (mean age, 2.5±4.2 years) on the United Network for Organ Sharing (UNOS) waiting list for primary liver transplant during 2002-2014.
- A reduced cohort excluded patients receiving liver donor transplant (n=464) or PELD exception points (n=1413).
- Follow-up: ≥2 years, until 2016.
- Funding: National Institute of Diabetes and Digestive and Kidney Diseases; NIH; Health Resources and Services Administration.
- PELD scores were concordant with 90-day waitlist mortality in the full cohort (concordance [C]-statistic, 0.8387; 95% CI, 0.8191-0.8584) and reduced cohort (C-statistic, 0.8123; 95% CI, 0.7919-0.8327), suggesting good discrimination.
- However, calibration was low: expected 90-day waitlist mortality rates were underestimated vs observed rates, and the difference rose with PELD score.
- PELD score underestimated the actual probability of death by up to 17%.
- Higher proportion of infants vs original PELD cohorts.