4. Determinants Of Long-Term Survival Following Major Surgery and the Adverse Effect Of Postoperative Complications
Shukri F. Khuri, MD1, William G. Henderson, PhD*2, Ralph G. DePalma, MD3, Cecilia Mosca, MSPH*2, Participants In the National Surgical Quality Improvement Program
1VA Boston Healthcare System, West Roxbury, MA; 2University of Colorado Health Outcomes Program, Aurora, CO; 3Department of Veterans Affairs, Washington, DC
OBJECTIVE:
To assess determinants of long-term survival (LTS) following major surgery.
METHODS:
We matched the VA NSQIP database with the VA BIRLS file, which is 95% accurate in depicting the vital status of US veterans. Cox regression analyses identified determinants of LTS in 105,951 patients undergoing one of 8 operations in 4 specialties, followed for an average of 8 years.
RESULTS:
The leading independent predictors of LTS in all 8 operations were: age, preoperative serum albumin, the occurrence of any complication within 30-days postoperatively, and preoperative ASA class, BUN>40, COPD, smoking and diabetes. Complications occurring within 30 days postoperatively independently predicting decreased LTS are shown in table 1. To test the hypothesis that complications alone caused this effect, independent of preoperative risk, we compared survival of two cohorts of patients with equal preoperative risk scores: one with occurrence of one or more complications and a matched cohort in whom complications were absent. Occurrence of any complication in the first cohort independently decreased projected median LTS by 68%.
CONCLUSIONS:
The leading predictors of LTS following major surgery were similar to those of 30-day mortality, except for COPD, smoking, and diabetes. Independent of patients’ preoperative severity of illness, occurrence of any 30-day postoperative complication significantly reduced LTS after major surgery. These novel findings demonstrate the need for process improvements that reduce postoperative complications, thus enhancing long-term survival.
