Semiautomated CT-Based Quantification of Donor Kidney Volume Applied to a Predictive Model of Outcomes in Renal Transplantation.

ABSTRACT

OBJECTIVE: The purpose of this study was to assess the use of semiautomated CT-based quantification of renal graft volume as a preoperative predictor of graft function. MATERIALS AND METHODS: All transplants over a 3-year period in which donors underwent CT and for which recipient outcomes were available were included. Two blinded readers used a commercially available reconstruction tool to independently measure donated kidney cortical volume and total parenchymal kidney volume. Transplant characteristics obtained by chart review included subject demographics, recipient pretransplant weight, immunologic matching, and recipient creatinine values at multiple time points. Intraclass correlation of measurements by the two readers was calculated. The ratios between donated kidney cortical volume and recipient pretransplant weight were correlated with graft function over 24 months and used in logistic regression models to calculate the odds of development of diminished renal function. RESULTS: After application of the inclusion and exclusion criteria, 153 transplants were included in the study. Donated kidney cortical and total parenchymal volume measurements had high correlation (R > 0.9) and high reproducibility (intraclass correlation coefficient, 0.93-0.94). Unadjusted correlations existed between estimated glomerular filtration rate (eGFR) and the ratio between donated kidney cortical volume and recipient pretransplant weight 12 months (R = 0.8489) and 24 months (R = 0.6839) after transplant. After adjustment for transplant parameters, recipients in the highest tertile for ratio between donated kidney cortical volume and recipient pretransplant weight (2.7 mL/kg) had higher mean eGFR values at all time points in the 24 months than did recipients in the lower tertiles (1.2 and 1.6 mL/kg). Recipients in the highest tertile had a significantly lower risk of development of diminished renal function 12 and 24 months after transplant (adjusted odds ratios, 0.25 at 12 months [95% CI, 0.09-0.66]; 0.27 at 24 months [95% CI, 0.10-0.71]). CONCLUSION: The CT-derived ratio between donated kidney cortical volume and recipient pretransplant weight is a noninvasively and readily obtained reproducible biomarker that is predictive of 12- and 24-month renal transplant outcomes.