*Purpose: Time-zero biopsies provide important information about baseline histology but their clinical utility compared to other non-invasive predictors is unclear. Our center performs universal time-zero biopsies just after reperfusion of the allograft, which reflect the influence of donor characteristics, ischemia-reperfusion injury, and initial alloimmune responses from the recipient. Here, we tested the hypothesis that time-zero histology is better than donor characteristics and other clinical features for predicting short-term kidney transplant outcomes.

*Methods: We studied all consecutive time-zero kidney biopsies performed at our center between May 2015 and June 2018. The primary exposure was Banff injury scores at time-zero. The primary outcomes were 6-month subclinical rejection (present vs. absent) and 2-year allograft function (estimated GFR). Continuous and categorical outcome variables were compared using student’s t-test and the chi-square test, respectively. We used univariable linear regression to identify determinants of 2-year allograft function. Covariates that were associated with 2-year allograft function at P < 0.20 in the univariable model were force entered into a multivariable linear regression model.

*Results: We analyzed 77 of 272 time-zero biopsies in an interim analysis. Our cohort was clinically diverse compared to previous time-zero studies: 62% had donor age > 35, 57% had donors of black race, and 54% had a Kidney Donor Profile Index (KDPI) score > 35%. Acute Banff lesions were rarely detected at time-zero. Chronic Banff lesions tubular atrophy (ct), vasculopathy (cv), and arteriolar hyalinosis (ah) were each detected in 30% of time-zero biopsies. The presence of cv or ah, but not other Banff lesions, was more common with donor age > 35 years and KDPI > 35% (P < 0.03 for all comparisons). Time-zero histology lesions were not predictive of subclinical inflammation at 6 months. The presence of cv or ah at time-zero was associated with allograft function at 2 years by univariable analysis, but these relationships were attenuated when adjusted for KDPI and other baseline clinical features in multivariable linear regression (Table).

*Conclusions: In this interim analysis, vasculopathy and arteriolar hyalinosis at time-zero were more common in older donors and in those with higher KDPI. Time-zero histology was not predictive of short-term kidney transplant outcomes, but this must be confirmed in the final analysis of our entire cohort.

« Back to 2020 American Transplant Congress