Failure-to-rescue in Patients Undergoing Pancreatectomy
Résumé
Objective: To evaluate the influence of hospital volume on failure-to-rescue (FTR) after pancreatectomy in France. Background: There are growing evidences that FTR is an important source of postoperative mortality (POM) after pancreatectomy. However, few studies have analyzed the volume-FTR relationship following pancreatic surgery. Methods: All patients undergoing pancreatectomy between 2012 and 2015 were included. FTR is defined as the 90-day POM rate among patients with major complications. According to the spline model, the critical cutoff was 20 resections per year and hospitals were divided into low (<10 resections/an), intermediate (11–19 resections/yr), and high volume centers (≥20 resections/yr). Results: Overall, 12,333 patients who underwent pancreatectomy were identified. The POM was 6.9% and decreased significantly with increased hospital volume. The rate of FTR was 14.5% and varied significantly with hospital volume (18.3% in low hospital volume vs 11.9% in high hospital volume, P < 0.001), age ( P < 0.001) and ChCl (CCl0-2: 11.5%, ChCl3: 13%, CCl ≥4:18.6%; P < 0.001). FTR for renal failure was the highest of all complications (40.2%), followed by postoperative shock (36.4%) and cardiac complications (35.1%). The FTR was significantly higher in low and intermediate compared with high volume hospitals for shock, digestive, and thromboembolic complications and reoperation. In multivariable analysis, intermediate (OR = 1.265, CI 95% [1.103–1.701], P = 0.045) and low volume centers (OR = 1.536, CI 95% [1.165–2.025], P = 0.002) were independently associated with increased FTR rates. Conclusion: FTR after pancreatectomy is high and directly correlated to hospital volume, highlighting variability in the management of postoperative complications. Measurement of the FTR rate should become a standard for quality improvement programs.