Session: MP47: Kidney Cancer: Localized: Surgical Therapy II
MP47-01: Limitations of Parenchymal Volume Analysis (PVA) for Estimating Split Renal Function (SRF) & New Baseline GFR (NBGFR) after Radical Nephrectomy (RN)
Introduction: Accurate prediction of NBGFR after RN can be important for counseling about RN vs. partial nephrectomy in some patients. SRF-based models have been shown to be most accurate for predicting NBGFR after RN, and PVA appears to be superior to nuclear renal scans (NRS) for this purpose. However, data regarding the potential limitations of PVA remain sparse. Our objective was to identify factors associated with PVA inaccuracy. Methods: 241 RCC patients who underwent RN (2006-2021) with preop. CT/MRI and NRS were included for analysis. Our SRF-based model was: Predicted NBGFR = 1.25(Global GFRPre-RN)(SRFContra), with SRF determined by PVA or NRS. Patients were grouped by PVA accuracy: PVAO (PVA over predicted NBGFR by >20%), PVAU (PVA under predicted NBGFR by >20%) and PVAA (PVA accurately predicted NBGFR). Accuracy of NRS was also studied in a similar manner. Results: Six patients (2.5%) were excluded from PVA due to severely altered renal architecture (e.g. PCKD). PVA accurately predicted NBGFR for 83% (196/235) of patients and over/underpredicted NBGFR in 7% and 9%, respectively. NRS only accurately predicted NBGFR for 66% of patients and over/under predicted NBGFR in 22% and 13% of patients, respectively. PVAO had more patients with active/recent hydronephrosis compared to both PVAU /PVAA (29% vs 14% vs 10%, p=0.05). Prior renal infarct (p=0.01) and infiltrative renal mass (IRM) (p=0.002) were associated with PVA inaccuracy in both directions. In contrast, NRS inaccuracy was random and did not associate with hydronephrosis, prior renal infarct, or IRM (all p>0.3). Overall, 30% of patients fell into cohorts where accuracy of PVA was less reliable, but even in these situations PVA was non-inferior to NRS. Conclusions: PVA accuracy is reduced for IRMs likely due to difficulty distinguishing the true margins of the tumor and, in about 3% of patients, severe disruption of the architecture rendered PVA impossible (e.g. PCKD). Disease processes that distort the volume/function relationship, such as active/recent hydronephrosis, can lead to PVAO, because measured mass may not function optimally. Awareness of cohorts where PVA can be less accurate has clinical utility. However, even in these cohorts, PVA was still non-inferior to NRS, and it remains a cost effective tool for determining SRF/NBGFR. Further study of other processes that can affect the volume/function relationship, such as active or chronic pyelonephritis, or that can affect the degree of renal functional compensation, such as extremes of age, will be required. SOURCE OF Funding: None
