Background: Half of all patients with an end colostomy after sigmoid colectomy (Hartmann's procedure) never undergo Hartmann's reversal, frequently secondary to frailty. This retrospective cohort study evaluates the utility of a five-item modified frailty index (mFI-5) in predicting post-operative outcomes after Hartmann's reversal.
Methods: The National Surgery Quality Improvement Program (NSQIP) database captured patients with elective Hartmann's reversals from 2011 to 2020. Clinical covariates were evaluated with univariate analysis and modified Poisson regression to determine association with overall morbidity, overall mortality, and extended length of stay (eLOS) when categorized by mFI-5 score.
Results: 15,172 patients underwent elective Hartmann's reversal (91.6% open and 8.4% laparoscopic). Patients were grouped by mFI-5 score (0: 48.7%, 1: 38.2%, ≥ 2: 13.1%). Adjusted multivariable analysis showed frail patients (mFI-5≥2) had increased overall mortality (OR 2.23, 95% CI 1.21-4.11), morbidity (OR 1.23, 95% CI 1.12-1.35), and eLOS (OR 1.12, 95% 1.02-1.23). Among frail patients, a laparoscopic approach was associated with decreased overall morbidity (OR .64, 95% CI 0.56-.73) and decreased eLOS (OR .46, 95% CI 0.39-.54) when compared to open approach.
Discussion: An mFI-5 of ≥2 was associated with greater morbidity, mortality, and eLOS following Hartmann's reversal. However, there were no mortality or eLOS differences in patients with an mFI-5 of 1 and only a 14% increase in any morbidity, making these patients potentially good candidates for Hartmann's reversal. Furthermore, laparoscopic surgery was associated with a protective effect for overall morbidity and eLOS, potentially mitigating some of the risk associated with higher frailty scores.
Keywords: Hartmann’s reversal; colorectal; frailty; national surgical quality improvement program; outcomes.