Original Investigation 

Association of VA Surgeons

Delayed Diagnosis of Anastomotic Leak and Failure to Rescue After Colon Resection

Samantha L. Savitch, Kiran H. Lagisetty, Pasithorn A. Suwanabol

JAMA Surg Published Online: February 11, 2026

doi: 10.1001/jamasurg.2025.6551

Key Points

Question  Is delay in detection of anastomotic leak after colon resection associated with failure to rescue?

Findings  In this cohort study of the Veterans Affairs Surgical Quality Improvement Program that included 39 175 patients who underwent colon resection, probability of failure to rescue (ie, death after a complication) was significantly higher in patients who had delayed diagnosis of anastomotic leak (that is, leak after the onset of sepsis) compared with those who had a leak identified before sepsis developed.

Meaning  These findings suggest early identification of anastomotic leak and timely and appropriate management may improve the mortality associated with colon resection.

Abstract

Importance  Anastomotic leak remains a leading cause of morbidity and mortality following colon resection. There is increasing evidence to suggest that failure to rescue (FTR), defined as death after a complication, is the culmination of a series of cascading events, which may be exacerbated by delays in diagnosis. Timely identification and management of anastomotic leaks may represent a crucial strategy for reducing FTR after colon resection.

Objective  To determine whether delayed diagnosis of anastomotic leak is associated with FTR following colon resection.

Design, Setting, and Participants  This cohort study used the Veterans Affairs Surgical Quality Improvement Program dataset from 2004 to 2023 to assess the rate of FTR after postoperative organ space surgical site infection (OSSI) among patients who underwent colon resection at a Veteran Affairs hospital. Data were analyzed from September 1, 2024, to December 13, 2025.

Exposure  Colon resection.

Main Outcomes and Measures  FTR rate after diagnosis of OSSI. OSSI was used as a surrogate for anastomotic leak and categorized as delayed (occurring after a sepsis diagnosis) or early (before or without a sepsis diagnosis). FTR rate after delayed or early OSSI diagnosis was compared. Multivariable logistic regression was performed to identify factors associated with FTR after OSSI.

Results  Of 39 175 patients (37 228 males [95.0%] and 1947 females [5.0%]; mean [SD] age, 65.3 [11.1] years) included in the analysis who underwent colon resection, 219 were Asian (0.6%) individuals, 6386 were Black (16.3%) individuals, 1820 were Hispanic (4.7) individuals, 24 612 were White (62.8%) individuals, and 6138 were individuals of other or unknown race and ethnicity (15.7%). The indication for resection was colon cancer in 17 067 patients (43.6%), diverticular disease in 4678 (11.9%), inflammatory bowel disease in 658 (1.7%) and colitis, ischemia, or other indication in 16 772 (42.8%). OSSI was diagnosed in 1227 patients (3.1%); of these diagnoses, 381 (31.1%) were delayed and 846 (68.9%) were early. On multivariable analysis, those with delayed OSSI had a significantly higher mean (95% CI) number of total discrete complications compared with those with early OSSI (3.0 [2.9-3.2] vs 1.7 [1.6-1.8], P < .001), higher probability of reoperation (62.1% vs 40.3%, P < .001), longer mean (95% CI) length of stay (22.6 [20.4-24.8] days vs 17.6 [16.5-18.7] days, P < .001), and higher probability of FTR (7.8% vs 2.2%, P < .001). Probability of FTR was 6.7% higher in patients who developed sepsis (8.1%) compared with those who never developed sepsis (1.4%).

Conclusions and Relevance  Findings of this study suggest that FTR after OSSI, which served as a proxy for anastomotic leak, was associated with delayed diagnosis, not the leak itself. Early identification of leaks and avoidance of progression to sepsis could reduce FTR rates. Quality initiatives aimed at earlier identification and timely and appropriate management of anastomotic leak may improve the mortality associated with colon resection.