Hospital costs of different treatment strategies for anastomotic leakage after total mesorectal excision: a multicentre cost analysis.

BACKGROUND

Limited data exist on hospital costs incurred by anastomotic leakage (AL), particularly in relation to specific treatment approaches. This study aimed to analyse the incremental hospital costs of AL after total mesorectal excision (TME), stratified by treatment strategy, over a 1-year time horizon.

METHODS

Patients undergoing total mesorectal excision (TME) for rectal cancer (2020-2023), included in the control cohort of the IMARI-study at 15 Dutch centres, were analysed. A cost analysis was conducted according to Dutch National Healthcare Institute guidelines. The primary outcome was the incremental hospital costs incurred by patients with AL detected within 30 days postoperatively, stratified by treatment strategy.

RESULTS

The analysis compared treatment costs in 32 patients with AL and 82 patients without AL. The average hospital costs per patient in the first postoperative year were €15.312. In patients with AL, the mean incremental costs were €24.333. Major cost drivers in the AL group were prolonged hospitalization (+€13.150) and (re)interventions (+€8.910). The treatment costs differed significantly between strategies: no faecal diversion (€10.062), faecal diversion with passive drainage (€23.903), faecal diversion with active drainage (€35.552), and salvage surgery (€38.793).

CONCLUSIONS

AL after TME resulted in a nearly fourfold increase in hospital costs compared with patients without AL. Salvage surgery was the most expensive treatment strategy, followed by faecal diversion with active drainage. Future studies should evaluate how these treatment costs relate to clinical success rates, including rates of chronic pelvic sepsis and permanent stomas.

TRIAL REGISTRATION: This study used data from the IMARI-study. The IMARI-study is registered with the Dutch Central Committee on Research Involving Human Subjects (NL67600.018.18) and is submitted to the http://www.onderzoekmetmensen.nl/en database (NL-OMON26456 and NL-OMON55903).