Dr. Tanaka began with a brief summary of the aforementioned techniques. “Subjective estimation” is the simplest in that the surgeon makes an estimate of the percent of parenchymal mass preserved (PPMP) based on her review of the imaging and personal experience without taking any particular measurements. In “quantitative estimation” the tumor and kidney are segmented on imaging to form 3d models, as is a rim of renal parenchyma located a fixed distance radially from the tumor. The volumes are measured and PMPP is calculated via subtraction. Contact-surface area-based estimation is a similar technique that does not require full segmentation of the tumor on imaging. Finally PMPP and postoperative change in GFR are known to be negatively correlated with nephrometry score, so it can be estimated from linear regression models based on previous data.
In this study, Dr. Tanaka’s group applied each of these estimates to a retrospective cohort of 428 men who had adequate imaging prior to undergoing partial nephrectomy. Actual PMPP was measured using postoperative contrast-enhanced imaging and GFR was estimated from serum creatinine both pre and postoperatively.
Correlation between predicted and observed PMPP was modest at best with each of these 4 models, with the best r value being 0.46. However paradoxically the r for postoperative overall GFR was very good, ~0.9 in each case. The group explained this finding by noting that by far the best predictor of postoperative GFR was simply preoperative GFR. Carrying this further, they demonstrated that a fixed estimate of postoperative GFR being equal to approximately 89% of preoperative GFR was as good as any of the more complex models previously tested (r=0.91). When both predicted postoperative PMPP and preoperative GFR were combined in a multivariate analysis along with several clinical characteristics, both of these values were found to be significant (along with a history of solitary kidney), however the effect size of preoperative GFR largely dominated the other two variables (relative impact 2.68 vs 0.24 and 0.24).
The simplicity of this finding and the fact that it comes from a relatively robust analysis make it instantly appealing for clinical use. I would not be surprised to see this simple 89% rule for estimating postoperative GFR after partial nephrectomy rapidly adopted for patient counseling.
Presented by: Hajime Tanaka, MD, PhD. Cleveland Clinic, Cleveland, OH. Tokyo Medical and Dental University, Tokyo, Japan.
Written by: Marshall Strother, MD, Chief Resident, Division of Urology, University of Pennsylvania, Philadelphia PA at the American Urological Association's 2019 Annual Meeting (AUA 2019), May 3 – 6, 2019 in Chicago, Illinois
1. Tanaka H, Wang Y, Suk-ouichai C, et al. Can We Predict Functional Outcomes after Partial Nephrectomy?. J Urol. 2019;201(4):693-701.