NASHVILLE –, according to findings from a meta-analysis presented at the meeting.
Overall, the odds ratio was 1.69 for self-harm or suicide after bariatric surgery (95% confidence interval, 1.62-1.76; P less than .001), “indicating a nearly 70% increase in risk for self-harm or suicide following bariatric surgery,” wrote Dawn Roberts, PhD, of Bradley University, Peoria, Ill., and her coauthor, Nicole Pearl of Washington University, St. Louis, in the poster accompanying the presentation.
Further, as elapsed time from surgery grew, the suicide rate dropped (effect size covariance, r = –0.25). “Thus, the greatest risk for self-harm or suicide appears to emerge in the years immediately following surgery,” Dr. Roberts said at the meeting presented by the Obesity Society and the American Society for Metabolic and Bariatric Surgery.
The investigators had a primary objective of characterizing the association between bariatric surgery and suicide or self-harm. The secondary purpose of the meta-analysis was to find moderators of the association that could explain some of the variability that had previously been seen in studies looking at mental health outcomes after bariatric surgery.
Some of the potential moderators, explained the investigators, included the surgery type. With Roux-en-Y gastric bypass (RYGB), more tissue is removed, potentially causing “more extensive disruption of neural pathways,” the investigators wrote. With greater loss of small-intestine surface area might come more disruption of the gut-brain axis, along with unknown effects on metabolism and pharmacokinetics of psychiatric medication. Additionally, alcohol use disorder might have a more profound effect after gastric bypass surgery.
It had previously been shown that more than two-thirds of suicides happen within the first 3 years after bariatric surgery. With the initial weight loss comes renegotiation of personal relationships, and the potential for more mobility and perhaps expanded career choices; stress accompanies even positive changes in these major life domains. Some patients will also experience weight regain within the first 3 years, after an initial nadir. This also can cause deterioration in quality of life, the investigators explained.
Dr. Roberts and her coinvestigator acknowledge that some of the potential moderators may have been missed in the initial data reporting. For example, the “presurgical sample may be at higher risk for suicide or self-injury but withhold psychiatric history during evaluation for fear of being rejected for surgery.”
From an initial 2,676 studies identified for consideration, investigators in the end extracted data from 28 studies from the United States, Canada, Sweden, and Brazil. The studies had considerable heterogeneity in methods; some included presurgery/postsurgery analyses of the same patients, some had a comparator nonsurgical group, and some used a single postsurgical assessment.
In studies where no nonbariatric comparison sample was available, the investigators assigned interpolated comparison. To arrive at this measure, they drew on the World Health Organization–reported base rate of suicides in the study country at the approximate year of assessment. Suicide was the only interpolated outcome.
Various measures of suicide and self-harm were captured, including completed and probable suicides, suicide attempts, and self-harm events. In some studies, information was drawn from a suicide-specific questionnaire, or from a suicide item on another type of questionnaire.
There was significant variability in the odds ratios for suicide or self-harm across studies, Dr. Roberts said.
The researchers plan to continue analyzing additional measures captured in the meta-analysis, such as gender, age, initial body mass index, surgery type, and the percent of excess weight lost at the time of assessment for suicide or self-harm risk. They reported no outside sources of funding, and no conflicts of interest.
SOURCE: Roberts, D et al. Obesity Week 2018, .