Immediate postoperative and mid-term survival in nonagenarians undergoing non-traumatic emergency surgery.

BACKGROUND

Nonagenarian people are increasingly using the emergency services, however, few studies have addressed this population. The goal of this prospective observational study is to determine relevant factors that independently reduce their immediate postoperative and mid-term survival.

METHODS

We included all the nonagenarians who underwent non-traumatic emergency surgery between 2006-2011. The dependent variable, days of postoperative survival, was studied at 2 intervals: 0-30 days and 31 days to 1 year after surgery (immediate and mid-term respectively). Firstly, we calculated the Kaplan-Meier survival curve (KMsc) of the whole population. To analyse the bivariate relationship between each of the pre-, intra-, and postoperative variables and mortality we used c2 and Fischer's test. Finally, we performed a stepwise Cox regression analysis and developed two models -one for each interval.

RESULTS

The study included 142 patients. The overall one-year mortality rate was 47.2%, and the 30-day mortality rate was 35.9%. The KMsc showed a steep drop during the first month, and then a more gradual one for the rest of year. The inflection point was at day 26. The variables associated with a reduced immediate postoperative survival were neoplasms and 4 complications: heart failure, pulmonary aspiration, renal impairment and stroke. The factors associated with a reduced mid-term survival were: postoperative arrhythmia, surgical reintervention, and perioperative red blood cell transfusion.

CONCLUSION

Postoperative complications are an important predictor of reduced survival both immediately, and in the mid-term. Furthermore, most postoperative mortality occurs within the first 26 days postsurgery; hence the need for an aggressive treatment of such complications during this period.