Alcohol and other substance use after bariatric surgery: prospective evidence from a U.S. multicenter cohort study

Abstract

Background: Empirical evidence suggests Roux-en-Y gastric bypass (RYGB) increases risk of developing alcohol use disorder (AUD). However, prospective assessment of substance use disorders (SUD) after bariatric surgery is limited.

Objective: To report SUD-related outcomes after RYGB and laparoscopic adjustable gastric banding (LAGB). To identify factors associated with incident SUD-related outcomes.

Setting: 10 U.S. hospitals METHODS: The Longitudinal Assessment of Bariatric Surgery-2 is a prospective cohort study. Participants self-reported past-year AUD symptoms (determined by the Alcohol Use Disorders Identification Test), illicit drug use (cocaine, hallucinogens, inhalants, phencyclidine, amphetamines, or marijuana), and SUD treatment (counseling or hospitalization for alcohol or drugs) presurgery and annually postsurgery for up to 7 years through January 2015.

Results: Of 2348 participants who underwent RYGB or LAGB, 2003 completed baseline and follow-up assessments (79.2% women, baseline median age: 47 years, median body mass index 45.6). The year-5 cumulative incidence of postsurgery onset AUD symptoms, illicit drug use, and SUD treatment were 20.8% (95% confidence interval (CI): 18.5-23.3), 7.5% (95% CI: 6.1-9.1), and 3.5% (95% CI: 2.6-4.8), respectively, post-RYGB, and 11.3% (95% CI: 8.5-14.9), 4.9% (95% CI: 3.1-7.6), and .9% (95% CI: .4-2.5) post-LAGB. Undergoing RYGB versus LAGB was associated with higher risk of incident AUD symptoms (adjusted hazard ratio or AHR = 2.08 [95% CI: 1.51-2.85]), illicit drug use (AHR = 1.76 [95% CI: 1.07-2.90]) and SUD treatment (AHR = 3.56 [95% CI: 1.26-10.07]).

Conclusions: Undergoing RYGB versus LAGB was associated with twice the risk of incident AUD symptoms. One-fifth of participants reported incident AUD symptoms within 5 years post-RYGB. AUD education, screening, evaluation, and treatment referral should be incorporated in pre- and postoperative care.

Keywords: Abuse; Addiction; Disorder; Gastric band; Obese; Roux-en-Y gastric bypass; Substance use; Treatment.

Figure 1

Modeled Prevalence of Substance Use and Indicators of Related Problems among Adults who underwent RYGB or LAGB. A. Among RYGB participants, there were significant increases over time in prevalence of regular alcohol consumption, AUD and illicit drug use (quadratic trends; P for all <.001) but not of SUD treatment (P =.18). B. Among LAGB participants, there was a significant increase in prevalence of regular alcohol consumption over time (quadratic trend; P =.01). There was not a significant trend in AUD (P =.09), illicit drug use (P =.33), or SUD treatment (P =.40). Abbreviations: AUD, Alcohol Use Disorder; LAGB, laparoscopic adjustable gastric banding; RYGB, Roux-en-Y gastric bypass; SUD, Substance Use Disorder. ^aAnnual assessments occurred within 6 months of the surgery anniversary date. Outcomes were not assessed at year 7. Data are based on observations until January 31, 2015; data collection ended before 432 RYGB and 177 LAGB participants were eligible for a 7 year assessment. ^bModels were adjusted for baseline factors related to missing follow-up data (age, smoking status and site). Observed and modeled data are reported online in eTable 2a and 2b, respectively, supplemental material.

Figure 2

Cumulative Incidence of Alcohol Use Disorder Symptoms, its Subcomponents, Illicit drug use and Substance Use Disorder Treatment among Adults who underwent RYGB or LAGB. The cumulative incidence of post-surgery SUD outcomes, among those without the specified SUD outcome in the year prior to surgery, is shown by surgical procedure, as a function of time since surgery. Abbreviations: AUD, Alcohol Use Disorder; AUDIT, Alcohol Use Disorder Identification Test; LAGB, laparoscopic adjustable gastric banding; RYGB, Roux-en-Y gastric bypass; SUD, Substance Use Disorder. ^aNumbers at risk at each time point are those who had not reported the SUD outcome since surgery and were not censored prior to or at the specified time point. Annual assessments occurred within 6 months of the surgery anniversary date. ^bModeled cumulative incidence with 95% CI of all SUD-related outcomes are reported in eTable 3, supplemental material.