However, the authors of this study distinguish between nephrectomies done for localized disease from those done for metastatic disease (cytoreductive nephrectomy, CN). Per the authors, patients undergoing CN represent a unique subset, as they have a more controversial indication for surgery, greater medical and surgical complexity, and shorter overall survival. Thus, they may stand to benefit more from high volume surgical care. They hypothesize that accounting total nephrectomy volume is not the best assessment of outcomes for patients with metastatic disease requiring nephrectomy; instead, hospital volume for specifically CN might be a better reflection of cancer outcomes.
They identified 11,089 patients who received CN for metastatic renal cell cancer (RCC) in the National Cancer Database from 1998 – 2012. They then ranked facilities based on annual CN volume, rather than total nephrectomy volume, into deciles. They compared outcomes for patients who received surgery in facilities in the top v. bottom deciles for annual volume (mean of 8.27 CN/year v. 1.84 CN/year).
Median follow-up was 60.39 months (IQR 35.12 – 95.18), median overall survival was 17.61 months (IQR 7.16-44). Following propensity score weighting to account for other factors, surgery at high-volume facilities was associated with a small but statistically significant decreased risk of mortality (IPTW-adjusted Cox Proportional Hazard Ratio=0.91; 95% CI: 0.86 – 0.96). On our IPTW-adjusted Kaplan-Meier analysis the median survival was 19.94 months at high-volume institutions vs. 15.97 months at low-volume institutions. As many systemic therapies for mRCC provide just a few months benefit, this change is clinically significant as well.
Looking at secondary outcomes, 30-day mortality, 30-day readmissions, and receipt of subsequent systemic therapy – there were no statistically significant differences.
However, it would have been interesting to compare outcomes if they had just looked total nephrectomy volume. Perhaps the results would have been the same?
It is also important to note that the NCDB is limited by lack of cancer-specific survival data – these are all OS data. This significant difference may also be reflective of smaller volume hospitals being unable to care for the medical comorbidities of disease rather than the oncologic care.
Speaker: S. Berg
Co-Author(s): Cole A., Pucheril D., Fletcher S., Noldus J., Sood A., Abdollah F., Menon M., Trinh Q-D.
Written by: Thenappan Chandrasekar, MD, Clinical Fellow, University of Toronto, Twitter: @tchandra_uromd at the 2018 European Association of Urology Meeting EAU18, 16-20 March, 2018 Copenhagen, Denmark