CT-measured Cortical Volume Ratio Is an Accurate Alternative to Nuclear Medicine Split Scan Ratio Among Living Kidney Donors. Issue 12 (December 2021)

• Record Type: Journal Article
• Authors: Montgomery, John R.; Brown, Craig S.; Zondlak, Allyse N.; Walsh, Kevin W.; Kozlowski, Julia E.; Pinsky, Alexa M.; Herriman, Emily A.; Sussman, Jeremy; Lu, Yee; Stein, Erica B.; Shankar, Prasad R.; Sung, Randall S.; Woodside, Kenneth J.
• Is Part Of: Transplantation. Volume 105:Issue 12(2021)
• Publication Date: 2021-12
• Subjects: Transplantation of organs, tissues, etc -- Periodicals; Transplantation immunology -- Periodicals; 617.95
• DOI: 10.1097/TP.0000000000003676
• Languages: English
• ISSNs: 0041-1337

Abstract

Abstract : Background: The 125 I-iothalamate clearance and 99m Tc diethylenetriamine-pentaacetic acid ( 99m Tc-DTPA) split scan nuclear medicine studies are used among living kidney donor candidates to determine measured glomerular filtration rate (mGFR) and split scan ratio (SSR). The computerized tomography–derived cortical volume ratio (CVR) is a novel measurement of split kidney function and can be combined with predonation estimated GFR (eGFR) or mGFR to predict postdonation kidney function. Whether predonation SSR predicts postdonation kidney function better than predonation CVR and whether predonation mGFR provides additional information beyond predonation eGFR are unknown. Methods: We performed a single-center retrospective analysis of 204 patients who underwent kidney donation between June 2015 and March 2019. The primary outcome was 1-y postdonation eGFR. Model bases were created from a measure of predonation kidney function (mGFR or eGFR) multiplied by the proportion that each nondonated kidney contributed to predonation kidney function (SSR or CVR). Multivariable elastic net regression with 1000 repetitions was used to determine the mean and 95% confidence interval of R 2, root mean square error (RMSE), and proportion overprediction ≥15 mL/min/1.73 m 2 between models. Results: In validation cohorts, eGFR-CVR models performed best ( R 2, 0.547; RMSE, 9.2 mL/min/1.73 m 2, proportion overprediction 3.1%), whereas mGFR-SSR models performed worst ( R 2, 0.360; RMSE, 10.9 mL/min/1.73 m 2, proportion overprediction 7.2%) ( P < 0.001 for all comparisons). Conclusions: These findings suggest that predonation CVR may serve as an acceptable alternative to SSR during donor evaluation and furthermore, that a model based on CVR and predonation eGFR may be superior to other methods. Abstract : Supplemental Digital Content is available in the text.