Predicting perioperative complications for partial versus radical nephrectomy in T1b-T2 renal cell carcinoma.

Abstract

737 Background: The standard of care for larger localized RCC lesions is radical (RN) or partial (PN) nephrectomy. RN is increasingly utilized to maximize oncologic benefit in complex tumors. Although PN is a more technically complex procedure, its nephron-sparing nature confers lasting renal and cardiovascular benefits. We utilized the National Surgical Quality Improvement Program (NSQIP) database to elucidate predictors of perioperative morbidity in T1b-T2 RCC patients. Methods: Using the NSQIP Nephrectomy-Targeted PUF, 2,094 patients undergoing nephrectomy between 2019-2020 for localized T1b-T2 RCC were identified. Variables of interest included surgical procedure and approach, tumor stage, demographics, pre-operative laboratory values, comorbidities, infection and venothromboembolism (VTE) prophylaxis techniques, peri-operative complications, operative time, length of hospital stay, 30-day reoperations, and 30-day readmissions. Chi square test was used to analyze univariate associations between certain comorbidities and complications. Multivariate regression analysis was utilized to compare complication rates between PN and RN after adjusting for baseline characteristics and surgical approach. p<0.05 was considered statistically significant. Results: 816 patients received PN while 1,278 received RN. PN patients had an increase in the following events: 30-day readmissions (7.0% vs. 4.7%, p=0.026), major bleeds (9.19% vs. 5.56%, p=0.001), renal failure requiring dialysis (1.23% vs. 0.31%, p=0.013), and urine leak or ureteric fistulae (1.10% vs. 0.31%, p=0.025). Open surgery was associated with increased VTE, renal failure, bleeds, urine leaks or ureteric fistulae, readmissions, and reoperations. Multivariate analysis revealed that PN remained predictive of all four aforementioned events, although further adjustment for robotic approach led to a loss of significance for renal failure and ureteric fistulae. Additional patient-specific predictors of relevant complications across procedure type included bleeding disorder and dialysis for bleeds, and renal failure, steroid use, and COPD for readmissions. Conclusions: This is the first study to evaluate the new NSQIP Nephrectomy-Targeted PUF. This population-based cohort provides unique insights into nephrectomy for pT1b-T2 localized RCC. We demonstrate significant associations between PN and specific complications, modulated by particular comorbidities, although both PN and RN were exceedingly safe. This analysis supports the development of novel risk stratification tools which account for specific patient comorbidities in predicting near term risk. Improved understanding of case-specific determinants of morbidity following PN or RN may facilitate shared decision making in localized RCC management.