Concurrent prophylactic placement of inferior vena cava filter in gastric bypass and adjustable banding operations in the Bariatric Outcomes Longitudinal Database

Abstract

Introduction: Postoperative pulmonary embolism (PE) is a leading cause of morbidity and mortality after bariatric surgery. However, the concurrent prophylactic placement of an inferior vena cava filter (CPIVCF) in patients undergoing bariatric operations remains controversial. This study used the Bariatric Outcomes Longitudinal Database (BOLD) to establish associated characters and determine outcomes of CPIVCF for patients undergoing Roux-en-Y gastric bypass (GB) and adjustable gastric banding (AB) surgeries.

Methods: We analyzed BOLD, a database of bariatric surgery patient information. GB and AB operations were categorized into open and laparoscopic approaches. Univariate logistic regressions were used to compare between non-CPIVCF and concurrent CPIVCF groups. Significant variables (P < .05) were subsequently input into multivariate regression models: CPIVCF was retained in each model.

Results: A total of 322 CPIVCFs (0.33%) were identified from 97,218 GB and AB operations performed between 2007 and 2010 in this retrospective registry study. Significant differences were identified in male gender (21.1% vs 31.4%; P < .001), preoperative body mass index (BMI; 44.5 ± 6.6 vs 45.3 ± 7; P < .001), and African-American race (10.5% vs 18%; P < .001) between non-CPIVCF and CPIVCF groups. The CPIVCF group had more patients with previous nonbariatric surgery (50% vs 43.6%; P = .02), a history of venous thromboembolism (VTE; 21.4% vs 3.1%; P < .001), impairment of functional status (7.8% vs 3.1%; P < .001), lower extremity edema (47.2% vs 27.1%; P < .001), obesity hypoventilation syndrome (7.1% vs 2.1%; P < .001), obstructive sleep apnea syndrome (58.1% vs 43.3%; P < .001), and pulmonary hypertension (13% vs 4.1%; P < .001). Patients in the CPIVCF group were more likely to receive GB than gastric banding (77% vs 58.1%; P < .001) and an open surgical approach (21.4% vs 4.8%; P < .001). Operative duration was longer in the CPIVCF group (119 ± 67 vs 89 ± 52 minutes; P < .001). The CPIVCF group also had a longer length of hospital stay (3 ± 2 vs 2 ± 6 days; P = .048), was associated with higher incidence of deep venous thrombosis (DVT; 0.93% vs 0.12%; P < .001), and had a higher mortality (0.31% vs 0.03%; P = .003) from PE and indeterminate causes. In multivariate analysis, male gender, African-American race, previous nonbariatric surgery, a high BMI, obesity hypoventilation syndrome, history of VTE, lower extremity edema, and pulmonary hypertension were preoperative factors associated with CPIVCF.

Conclusions: CPIVCF was associated with specific clinical features, increased health care resource utilization, and a higher mortality in patients undergoing bariatric operations. Although selected patient characteristics influence surgeons to perform CPIVCF, this study was unable to establish an outcome benefit for CPIVCF.

Trial registration: ClinicalTrials.gov NCT01002352.