Combined Donor-Recipient Obesity and the Risk of Graft Loss After Kidney Transplantation

Abstract

Background: As the prevalence of obesity increases globally, appreciating the effect of donor and recipient (DR) obesity on graft outcomes is of increasing importance. Methods: In a cohort of adult, kidney transplant recipients (2000-2017) identified using the SRTR, we used Cox proportional hazards models to examine the association between DR obesity pairing (body mass index (BMI) >30 kg/m²), and death-censored graft loss (DCGL) or all-cause graft loss, and logistic regression to examine risk of delayed graft function (DGF) and ≤30 days graft loss. We also explored the association of DR weight mismatch (>30 kg, 10-30 kg (D>R; D<R) and <10 kg (D = R)) with each outcome, stratifying by DR obesity pairing. Results: Relative to non-obese DR, obese DR were highest risk for all outcomes (DCGL: HR 1.26, 95% CI 1.22-1.32; all-cause graft loss: HR 1.09, 95% CI 1.06-1.12; DGF: OR 1.98, 95% CI 1.89-2.08; early graft loss: OR 1.34, 95% CI 1.19-1.51). Donor obesity modified the risk of recipient obesity and DCGL [p = 0.001] and all-cause graft loss [p < 0.001] but not DGF or early graft loss. The known association of DR weight mismatch with DCGL was attenuated when either the donor or recipient was obese. Conclusion: DR obesity status impacts early and late post-transplant outcomes.

Keywords: body mass index; graft loss; kidney transplant outcomes; obesity; obesity pairing; weight mismatch.

FIGURE 1

Final study cohort following exclusions.

FIGURE 2

Temporal changes in donor-recipient obesity pairing over time. The accompanying table displays descriptive statistics for each of the donor-recipient obesity pairings.

FIGURE 3

Kaplan-Meier survival curves for time to death-censored graft loss for each donor-recipient obesity pairing. A number at risk table is included below the figure. *The log-rank p-value is <0.001.

FIGURE 4

Hazard ratio plot for death-censored graft loss for combined donor-recipient weight mismatch, stratified by donor-recipient obesity. Models were adjusted for known literature predictors of graft loss, including donor and recipient age, race, sex, recipient end-stage kidney disease (ESKD) cause, cold ischemia time (CIT), dialysis vintage, pre-emptive status, previous kidney transplant, human leukocyte antigen (HLA) mismatch, peak panel reactive antibody (PRA), and recipient medical comorbidities (coronary artery disease, hypertension, peripheral vascular disease, type 2 diabetes).