OBJECTIVE: To assess the association between the use of gelatin–thrombin matrix and the development of pelvic abscess during hysterectomy as well as factors associated with surgeons' use of this product.
METHODS: Data for patients undergoing hysterectomy for obstetric–gynecologic pathology were abstracted from databases at a tertiary hospital between 2009 and 2012. Open and minimally invasive hysterectomies were included and vaginal hysterectomies were excluded. Blood loss, surgery type, comorbidities, abscess formation, and use of gelatin–thrombin matrix were examined. Abscess was defined as a walled-off fluid collection (documented with computed tomography scan) with fever (greater than 38°C) or leukocytosis (greater than 11,000/microliter). Standard statistical models were used.
RESULTS: Of the 413 patients identified, 213 (51%) underwent surgery for malignancy. Gelatin–thrombin matrix was used in 166 patients (40%). The overall rate of abscess was low (3%). In bivariate analyses, blood loss greater than 500 mL (odds ratio [OR] 3.9, 95% confidence interval [CI] 1.1–12.9, P=.021], ascites (OR 6.5, 95% CI 1.6–26.1, P=.023), drain placement (OR 4.5, 95% CI 1.3–15.1, P=.009), and gelatin–thrombin matrix use (OR 7.0, 95% CI 1.5–32.9, P=.009) were significantly associated with abscess formation. Multivariate logistic regression revealed that only gelatin–thrombin matrix use predicted the development of pelvic abscess (OR 7.0, 95% CI 1.5–32.9, P=.013).
CONCLUSION: We found that gelatin–thrombin matrix use was associated with an increased risk of pelvic abscess. Although these products are important in the setting of bleeding, these data suggest that the liberal use of sealants is not without risk.
LEVEL OF EVIDENCE: III
The use of gelatin&#x2013;thrombin matrix is associated with the development of abscesses in women undergoing surgery for pelvic disease.
Department of Obstetrics and Gynecology, University of Colorado School of Medicine at Denver, Denver, Colorado.
Corresponding author: Saketh R. Guntupalli, MD, Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, University of Colorado School of Medicine, Academic Office 1, Room 4411, 12631 East 17th Avenue, Aurora, CO 80045; e-mail: email@example.com.
Presented at the 45th annual meeting of the Society of Gynecologic Oncology, March 25–29, 2014, Tampa, Florida.
Financial Disclosure The authors did not report any potential conflicts of interest.