Usefulness of CT scan as part of an institutional protocol for proactive leakage management after low anterior resection for rectal cancer

PURPOSE: Highly selective fecal diversion after low anterior resection (LAR) for rectal cancer requires a strict postoperative protocol for early detection of anastomotic leakage (AL). The purpose of this study was to evaluate C-reactive protein (CRP)–based CT imaging in diagnosis and subsequent management of AL. METHODS: All patients that underwent a CT scan for suspicion of AL after transanal total mesorectal excision for rectal cancer in a university center (2015–2020) were included. Outcome parameters were diagnostic yield of CT and timing of CT and subsequent intervention. RESULTS: Forty-four out of 125 patients underwent CT (35%) with an overall median interval of 5 h (IQR 3–6) from CRP measurement. The anastomosis was diverted in 7/44 (16%). CT was conclusive or highly suspicious for AL in 23, with confirmed AL in all those patients (yield 52%), and was false-negative in one patient (sensitivity 96%). CT initiated subsequent intervention after median 6 h (IQR 3–25). There was no or minor suspicion of AL on imaging in all 20 patients without definitive diagnosis of AL. After CT imaging on day 2, AL was confirmed in 0/1, and these proportions were 6/6 for day 3, 7/10 for day 4, 2/4 for day 5, and 9/23 beyond day 5. CONCLUSION: In the setting of an institutional policy of highly selective fecal diversion and pro-active leakage management, the yield of selective CT imaging using predefined CRP cut-off values was 52% with a sensitivity of 96%, enabling timely and tailored intervention after a median of 6 h from imaging. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00423-022-02652-z.