Costs associated with anastomotic leak after left-sided colorectal surgery: retrospective real-world study in England

BACKGROUND: Anastomotic leak (AL) is a serious postoperative complication following colorectal surgery (CS). This study compared healthcare resource utilization (HCRU) and costs between patients with and without AL after left-sided CS to highlight the significant burden of AL, and to address the evidence gap regarding its economic impact in England. METHODS: This was a retrospective matched cohort study of patients with and without AL after left-sided CS. Data for adult patients undergoing CS between 2018 and 2021 in England were drawn from the Hospital Episode Statistics database. Patients with and without AL after left-sided CS were exactly matched 1 : 1 on age, sex, surgery date and type, and indication. Comparative analyses were conducted on patients with versus without AL, as well as on patients with minor and major AL versus those without AL. The primary objective was to compare all-cause costs for hospitalization; secondary outcomes were to determine AL incidence rates, baseline characteristics of those with AL (major and minor) versus those without, and the implications of AL in terms of interventions and length of hospital stay (LOS). Multivariate models were adjusted for the Charlson co-morbidity index and geographical region. RESULTS: Of 36 948 adult patients who underwent left-sided CS with anastomosis without stoma formation, after excluding those with bleeding complications, 2059 patients with AL and 27 590 without AL were identified. After exact matching, there were 1982 patients in each group. Among patients with AL, 1116 had major AL and 866 had minor AL. AL was associated with a significantly higher mean all-cause inpatient cost (adjusted difference €11 728; 95% confidence interval (c.i.) €10 313 to €13 209; P < 0.001). Patients with AL also had a significantly higher mean cumulative LOS (adjusted difference 15.93 days; 95% c.i. 14.24 to 17.61 days; P < 0.001). The estimated incidence of AL requiring intervention was 5.6%. Patients with any AL had significantly more mean all-cause hospitalizations (adjusted difference 0.37; 95% c.i. 0.26 to 0.49; P < 0.001) and higher cumulative LOS (adjusted difference 15.93; 95% c.i. 14.24 to 17.61; P < 0.001) than patients without AL. CONCLUSION: AL is associated with poorer clinical outcomes, increased HCRU, and a substantial economic burden.