EP120 - Anastomotic Leak in Segmental Colectomy Without Protective Stoma: National Trends From the Premier Healthcare Database

Purpose/Background: Anastomotic leak (AL) is the most feared complication after colectomy, associated with increased morbidity, prolonged hospitalization, higher healthcare costs, and mortality. Reported leak rates vary widely, ranging from 2-19%, depending on anastomotic location, risk factors, indication, and operative approach. This variability is compounded by inconsistent study definitions of AL, which often rely on differing combinations of radiographic, clinical, and procedural criteria. Another issue compounds this inconsistency: timeliness of our information - how often should we review leak rates to assess ourselves and our ability to handle them.

National quality registries offer valuable insight into surgical outcomes but may be limited by participation costs, selective sampling, and variable follow-up. In contrast, administrative claims data like the Premier Healthcare Database (PHD) offer large-scale, real-world data encompassing inpatient and outpatient encounters across diverse U.S. hospitals. This all-payer database includes more than 244 million unique patients, enabling evaluation of outcomes at a national level. Yet few studies have used such datasets to comprehensively assess AL rates after segmental colectomy in the absence of a protective stoma (PS).

To address this gap, we evaluated AL rates in primary segmental colectomies without a PS using standardized ICD-10 codes in the PHD.

Methods/Interventions: This retrospective study used the PHD (January 2018-June 2023) to evaluate AL rates following primary segmental colectomy. Adult patients (≥18 years) undergoing right, transverse, left, sigmoid, or low anterior resection (LAR) were identified using ICD-10 procedure codes. Patients undergoing total colectomy or creation of a PS at index surgery were excluded.

The primary outcome was AL within 30 days of surgery, identified by ICD-10 procedure codes indicating reoperation or procedural intervention for leak, abscess, or related complications. Analyses were stratified by colectomy location, surgical approach, surgical indication, and acuity of admission status. Baseline demographic and clinical characteristics were extracted from claims data. An IRB exemption was granted for this study.

Results/Outcomes: A total of 184,952 segmental colon resections without PS were identified. Of these, 12,625 were complicated by an AL, yielding an overall rate of 6.8% within 30 days. Leak rates varied by colectomy type, ranging from 5.2% following LAR to 13.1% for transverse colectomy (Figure 1). Patients undergoing colectomy for trauma were at particularly high risk, with an AL rate of 34.3%. When elective cases were analyzed, the overall AL rate decreased to 4.1% (Figure 1).

Conclusion/Discussion: This nationally representative cohort highlights meaningful variation in AL risk following segmental colectomy. The overall 6.8% rate observed is consistent with published ranges, yet rates differed by anatomic location and the acuity of the admission setting. Colectomy performed in the context of trauma conferred nearly an 8-fold higher risk of AL and ED admissions a 3-fold risk compared to elective procedures, underscoring the impact of emergent physiology and operative complexity.

Despite advances in surgical techniques, enhanced perioperative care, and novel devices, AL rates are relatively unchanged over the past 15 years. This plateau is concerning and underscores the ongoing need for innovation in operative strategy, perioperative optimization, and technologies designed to reduce the risk of leak.

Figure 1. Anastomotic Leaks By Location and Admission Status