Abstract: FR-PO1026
Histologic Findings on Time-Zero Allograft Biopsies Correlate with Kidney Donor Profile Index (KDPI) and 30-Day Serum Creatinine
Session Information
- Transplantation: Donor-Candidate Assessment and Predictors of Outcome
November 03, 2017 | Location: Hall H, Morial Convention Center
Abstract Time: 10:00 AM - 10:00 AM
Category: Transplantation
- 1702 Transplantation: Clinical and Translational
Authors
- Jen, Kuang-Yu, University of California, Davis, Sacramento, California, United States
- Chen, Ling-Xin, University of California, Davis, Sacramento, California, United States
- Francalacci, Luis C, University of California, Davis, Sacramento, California, United States
- Perez, Richard V., University of California, Davis, Sacramento, California, United States
- De Mattos, Angelo M., University of California, Davis, Sacramento, California, United States
Background
KDPI is used as a numerical measure of deceased donor kidney quality relative to other recovered kidneys. It uses 10 donor factors but does not consider histologic findings. Correlation of KDPI to histologic findings is lacking. In this study, we examined the correlation between KDPI and chronic changes seen in time zero biopsies. We also assess whether such histologic findings add to the predictive ability of KDPI for 30-day serum creatinine and delayed graft function.
Methods
All deceased donor kidney transplants at our institution from 07/01/2016 to 03/15/2017 that had a time-zero biopsy were included. The biopsies were graded according to Banff 2015 guidelines. Distribution of KDPI was compared by Banff scores for chronicity (ci and ct) as well as chronic vascular disease (ah and cv). Linear regression was used to assess: 1) correlation between Banff scores and donor KDPI and 2) the ability of KDPI and Banff scores (either individually or together) to predict 30-day serum creatinine. Logistic regression was used to assess the ability of KDPI and/or Banff scores to predict delayed graft function.
Results
134 recipients had a time-zero biopsy performed. There was correlation between the following Banff scores and KDPI: ci, ct, ah and cv (Table 1). ah score of greater or equal to 2 most closely correlated with KDPI. 30-day serum creatinine was predicted by KDPI, as well as ci, ct and ah scores. Using KDPI with ci + ct scores resulted in the best prediction of 30-day serum creatinine, with a correlation coefficient of 0.27. There was no correlation between Banff scores and the occurrence of delayed graft function. A KDPI score of 20-85% predicted delayed graft function at an OR of 8.18 (95% CI 1.80-37.2, p-value 0.007).
Conclusion
Chronic changes seen in time zero allograft biopsies correlate with KDPI and may be useful for predicting early graft outcomes.
Histologic variables, mean 30-day SCr and DGF by KDPI categories
| KDPI | N | ci + ct ≥ 2 | ah ≥ 2 | cv ≥ 2 | Mean 30-day SCr (mg/dL) | DGF, n (%) |
| 0-20% | 19 | 1 (5%) | 1 (5%) | 4 (21%) | 1.32 | 2 (11%) |
| 21-85% | 104 | 33 (32%) | 40 (38%) | 39 (38%) | 2.25 | 51 (49%) |
| 86-100% | 11 | 4 (36%) | 6 (55%) | 3 (27%) | 2.83 | 3 (27%) |
| p-value* | 0.03 | 0.003 | 0.38 | 0.01 | 0.003 |
*p-values obtained via Fisher’s exact test for categorical variables and ANOVA for continuous variables