Predictive risk model of 30-day mortality in plastic and reconstructive surgery patients.

BACKGROUND: Death following surgical procedures, although uncommon, represents the most significant and costly postoperative complication. The authors delineate several key independently associated risk factors for 30-day mortality in a large nationwide plastic surgery population and generate a simple risk-assessment tool. METHODS: Patients who underwent a primary plastic surgery procedure (n=24,778) were identified from the 2005 to 2011 American College of Surgeons National Surgical Quality Improvement Program databases. Preoperative patient comorbidities and operative variables were analyzed to determine associations with mortality. Multivariate regression modeling and risk stratification were performed to generate a decision-support tool capable of assessing mortality risk. RESULTS: Seventy-eight patients (0.31 percent) experienced 30-day mortality. Age older than 65 years, inpatient surgery, hepatorenal disease, recent chemotherapy, and partial or dependent functional status were perioperative factors associated with 30-day mortality. A composite mortality risk score was used to stratify patients into groups of low (n=18,063) (30-day mortality, 0.06 percent), intermediate (n=1252) (30-day mortality, 0.96 percent), high (n=314) (30-day mortality, 5.10 percent), or extreme (n=126) (30-day mortality, 17.46 percent) risk. The scoring groups accurately separated risk (p<0.001) with a 291-fold variation between low- and extreme-risk patients (C statistic, 0.93). CONCLUSIONS: This study defines a 0.31 percent risk of perioperative 30-day mortality in a heterogeneous cohort of plastic surgery patients and identifies risk factors independently associated with 30-day mortality. These data can be used to tailor and improve informed consent, to optimize surgical decision-making, and to implement risk-reduction strategies in high-risk patients. CLINICAL QUESTION/LEVEL OF EVIDENCE: Risk, III.