Uzma Rahman1, Salvatore Docimo2, Aurora D Pryor2, Andrew Bates2, Nabeel R Obeid2, Konstantinos Spaniolas3. 1. Department of Surgery, Stony Brook University, Stony Brook, New York. 2. Division of Foregut, Bariatric and Advanced GI Surgery, Department of Surgery, Stony Brook University, Stony Brook, New York. 3. Division of Foregut, Bariatric and Advanced GI Surgery, Department of Surgery, Stony Brook University, Stony Brook, New York. Electronic address: konstantinos.spaniolas@stonybrookmedicine.edu.
Abstract
BACKGROUND: Although multiple studies demonstrate that routine postoperative contrast studies have a low yield in diagnosing patients with early gastrointestinal (GI) leak after bariatric surgery, the practice pattern is unknown. Additionally, routine imaging may hinder procedural pathways that lead to accelerated postoperative discharge. OBJECTIVES: To report on the nationwide use of routine upper GI studies (UGI) and evaluate the effect on hospital resource utilization. SETTING: Nationwide analysis of accredited centers. METHODS: The Metabolic and Bariatric Surgery Accreditation Quality and Improvement Program public use file for 2015 was used to identify patients who underwent routine UGI after nonrevisional Roux-en-Y gastric bypass or sleeve gastrectomy. Multivariable logistic regression models were developed to identify risk factors for early hospital discharge. RESULTS: Bariatric surgery was performed on 130,686 patients. Routine UGI was performed in 30.9% of Roux-en-Y gastric bypass and 43% of sleeve gastrectomy patients (P<.0001). Patients undergoing routine UGI were less likely to be discharged by postoperative day 1 (odds ratio .7, 95%; confidence interval .69-0.72). There was no difference in postoperative leak rate between the routine UGI versus nonroutine UGI group (.7% versus .8%, P = .208). Among patients who developed a GI leak, there was no significant difference in the rate of reoperation, readmission, and reintervention between the 2 groups. The time interval between index operation and any further management for the leak was longer in the routine UGI group. CONCLUSIONS: Routine UGI evaluation after bariatric surgery remains a common practice in accredited centers. This practice is associated with prolonged hospital length of stay, with no effect on the diagnosis of leak rate.
BACKGROUND: Although multiple studies demonstrate that routine postoperative contrast studies have a low yield in diagnosing patients with early gastrointestinal (GI) leak after bariatric surgery, the practice pattern is unknown. Additionally, routine imaging may hinder procedural pathways that lead to accelerated postoperative discharge. OBJECTIVES: To report on the nationwide use of routine upper GI studies (UGI) and evaluate the effect on hospital resource utilization. SETTING: Nationwide analysis of accredited centers. METHODS: The Metabolic and Bariatric Surgery Accreditation Quality and Improvement Program public use file for 2015 was used to identify patients who underwent routine UGI after nonrevisional Roux-en-Y gastric bypass or sleeve gastrectomy. Multivariable logistic regression models were developed to identify risk factors for early hospital discharge. RESULTS: Bariatric surgery was performed on 130,686 patients. Routine UGI was performed in 30.9% of Roux-en-Y gastric bypass and 43% of sleeve gastrectomy patients (P<.0001). Patients undergoing routine UGI were less likely to be discharged by postoperative day 1 (odds ratio .7, 95%; confidence interval .69-0.72). There was no difference in postoperative leak rate between the routine UGI versus nonroutine UGI group (.7% versus .8%, P = .208). Among patients who developed a GI leak, there was no significant difference in the rate of reoperation, readmission, and reintervention between the 2 groups. The time interval between index operation and any further management for the leak was longer in the routine UGI group. CONCLUSIONS: Routine UGI evaluation after bariatric surgery remains a common practice in accredited centers. This practice is associated with prolonged hospital length of stay, with no effect on the diagnosis of leak rate.