Purpose: To identify differences in hospital course and hospitalization cost when comparing image-guided percutaneous drainage with surgical repair for gastrointestinal anastomotic leaks.
Materials and methods: A retrospective IRB-approved search using key words "leak" and/or "anastomotic" was performed on all adult CT reports from 2002 to 2011. CT examinations were reviewed for evidence of a postoperative gastrointestinal leak and assigned a confidence score of 1-5 (1 = no leak, 5 = definite leak). Patients with an average confidence score <4 were excluded. Type of surgery, patient data, method of leak management, number of hospital admissions, length of hospital stay, discharge disposition, number of CT examinations, number of drains, and hospitalization costs were collected.
Results: One hundred thirty-nine patients had radiographic evidence of a gastrointestinal anastomotic leak (esophageal, gastric, small bowel or colonic). Nine patients were excluded due to low confidence scores. Twenty-seven patients underwent surgical repair (Group A) and 103 were managed entirely with percutaneous image-guided drainage (Group B). There was no significant difference in patient demographics or number of hospital admissions. Patients in Group A had longer median hospital stays compared to Group B (48 vs. 32 days, p = 0.007). The median total hospitalization cost for Group A was more than twice that for Group B ($99,995 vs. $47,838, p = 0.001). Differences in hospital disposition, number of CT examinations, number of drains, and time between original surgery and first CT examination were statistically significant.
Conclusion: Gastrointestinal anastomotic leaks managed by percutaneous drainage are associated with lower hospital cost and shorter hospital stays compared with surgical management.