A comprehensive analysis of 30-day readmissions after emergency general surgery procedures: Are risk factors modifiable?

BACKGROUND: Modifiable risk factors associated with procedure-related 30-day readmission after emergency general surgery (EGS) have not been comprehensively studied. We set out to determine risk factors associated with EGS procedure-related 30-day unplanned readmissions. METHODS: A retrospective cohort study was conducted using the National Surgical Quality Improvement Project database (2013-2019). It included nine surgical procedures encompassing 80% of the burden of EGS diseases, performed on an urgent/emergent basis. The procedures were classified as low risk (open and laparoscopic appendectomy and laparoscopic cholecystectomy) and high risk (open cholecystectomy, laparoscopic and open colectomy, lysis of adhesions, perforated ulcer repair, small bowel resection, and exploratory laparotomy). Data on patient characteristics, admission status, procedure risk, hospital length of stay, and discharge disposition were analyzed by multivariate logistic regression. RESULTS: A total of 312,862 patients were included (16,306 procedure-related 30-day readmissions [5.2%]). Thirty-day readmission patients were older, had higher American Association of Anesthesiology scores, were more often underweighted or markedly obese, and were more frequently presented with sepsis. Risk factors associated with EGS procedure-related 30-day unplanned readmissions included age older than 40 years (adjusted odds ratio [AOR], 1.15), American Association of Anesthesiology >= 3 (AOR, 1.41), sepsis present at the time of surgery (AOR, 1.84), body mass index <18 kg/m(2) (AOR, 1.16), body mass index >= 40 kg/m(2) (AOR, 1.12), high-risk procedures (AOR, 1.51), LOS >= 4 d (AOR, 2.04), and discharge except to home (AOR, 1.33). Thirty-day readmissions following low-risk procedures occurred at a median of 5 days (interquartile range, 2-11 days) and 6 days (interquartile range, 3-11 days) after high-risk procedures. Surgical site infections, postoperative sepsis, wound disruption, and thromboembolic events were more prevalent in the 30-day readmission group. Mortality rate was fourfold higher in the 30-day readmission group (2.4% vs. 0.6%). CONCLUSION: We identified several unmodifiable patients and EGS disease-related factors associated with 30-day unplanned readmissions. Readmissions could be potentially reduced by the implementation of a postdischarge surveillance systems between hospitals and postdischarge destination facilities, leveraging telehealth and outpatient care.