Outcomes and risk stratification for late antibody-mediated rejection in recipients of ABO-incompatible kidney transplants: a retrospective study.

The required intensity of monitoring for antibody-mediated rejection (AMR) after of ABO-incompatible (ABOi) kidney transplantation is not clearly formulized. We retrospectively evaluated a single-center cohort of 115 ABO-incompatible (ABOi) kidney transplant recipients, of which 32% were also HLA incompatible (ABOi/HLAi) with their donors. We used an adjusted negative binomial model to evaluate risk factors for late AMR. Using this model, we risk-stratified patients into high- and low-risk groups for the development of late AMR; 26% of patients had at least one AMR episode; 49% of AMR episodes occurred within 30-days after transplant and were considered early AMR. Patients with an early AMR episode had a 5.5-fold greater incidence of developing late AMR [IRR = 5.5, (95% CI: 1.5-19.3), P = 0.01]. ABOi/HLAi recipients trended toward increased late AMR risk [IRR = 1.9, (95% CI: 0.5-6.6), P = 0.3]. High-risk recipients (those with an early AMR or those who were ABOi/HLAi) had a sixfold increased incidence of late AMR [IRR = 6.3, (95% CI: 1.6-24.6), P = 0.008] versus low-risk recipients. The overall incidence of late AMR was 20.8% vs. 1.5% in low-risk recipients. Changes in anti-A/B titer did not correlate with late AMR (IRR = 0.9 per log titer increase, P = 0.7). This risk-stratification scheme uses information available within 30 days of ABOi transplantation to determine risk for late AMR and can help direct longitudinal follow-up for individual patients.