Body mass index percentage and survival in pediatric patients listed for lung transplantation: A modern-era multi-institutional analysis.

ABSTRACT

BACKGROUND: Prior studies suggest that being underweight by body mass index percentiles (BMI%) or thinness grade did not affect post-transplant survival in pediatric lung transplant (LTx) recipients regardless of cystic fibrosis (CF) or non-CF diagnosis. Graft and overall survival from the time of listing was instead evaluated based on listing BMI%, the current standard of practice for BMI definitions in pediatrics, to ascertain the impact of a "severely low" subcategory. METHODS: The UNOS registry was queried for children listed for LTx (aged 2 to <18 years) from January 1986 to March 2020. BMI% at listing and transplant were calculated per CDC guidelines according to age in years, sex, and reported BMI%. Patients were divided by listing BMI% severely low (<3rd), low (3-<5th), normal (5-<85th), overweight (85-<95th), and obese (≥95th). Kaplan-Meier curves were generated to assess differences in overall survival since listing based on BMI% classes. Cox proportional-hazards models were developed to assess risk factors for overall and graft survival, including listing BMI%, transplant listing era (≥2005), and listing age, by reporting hazard ratios (HR). RESULTS: Listing BMI% was calculable for 1,876 patients. The proportion of patients with CF differed significantly between BMI% groups (p < 0.001). Patients listed with a non-CF diagnosis comprised 34% of those in the severely low category, and 88% of those listed with an obese BMI%. Compared to patients with a normal listing BMI%, the cohort with severely low BMI% had worse overall survival regardless of LTx (p = 0.009) and graft survival (p = 0.034). Compared to patients with a low BMI%, those with a severely low BMI% had significantly poorer graft survival as well (p = 0.040). Mean graft survival was not significantly different between groups that remained at listing BMI% vs those that improved in category despite an overall small sample size. Independent predictors of poorer survival from the time of listing include severely low vs low-normal BMI% (HR = 1.20) and listing age (HR = 1.02). CONCLUSION: The proportion of children listed at severely low BMI% has steadily decreased with time, yet pediatric LTx candidates listed with a severely low BMI% had poorer graft and overall survival compared to those of normal BMI%. Severely low listing BMI% was an independent prognostic factor for higher mortality risk from the time of placement on the waitlist. BMI% may be a modifiable target for improving survival regardless of transplantation.