MBS may improve depression, anxiety and non-normative eating, but slightly increase suicides and substance use disorders

Metabolic and bariatric surgery (MBS) may improve depression, anxiety, non-normative eating, and attention, but slightly increase suicides and substance use disorders, compared to nonsurgical conditions, according to a study led by researchers from McMaster University, Hamilton, Ontario, Canada.

The study’s authors noted that there is ample evidence that obesity negatively affects the brain, manifesting as reduced psychiatric and cognitive functioning and greater prevalence of diagnosed psychiatric disorders. Although this is not thoroughly, there are many proposed mechanisms linking obesity to changes in the brain, such as endothelial dysfunction, elevated inflammation and cortisol, vascular risk factors, weight-based discrimination and pain conditions.

Furthermore, several studies have found a high prevalence of psychiatric disorders in people with obesity, causing significant burden to affected individuals and communities, increasing disability and reducing quality and quantity of life.

Of the existing treatment for obesity, the authors stated that MBS has the strongest evidence for improving health consequences of obesity and there is growing evidence that MBS can result in positive psychiatric and cognitive outcomes.

According to the authors, no meta-analysis of RCTs has compared psychiatric and cognitive outcomes between surgical and nonsurgical groups. Therefore, they carried out a systematic review and meta-analysis to determine what is known about the effect of MBS on psychiatric and cognitive functioning in people with obesity, compared to nonsurgical groups.

Outcomes

In total, they identified 75 studies were (71 non-randomised studies (NRS) and four RCTs). Duration of follow-up ranged from 3 to 240 months (mean = 37 months). Surgical techniques included sleeve gastrectomy, gastric bypass, biliopancreatic diversion with duodenal switch, vertical gastroplasty, DJBL, and gastric banding. A passive control group was used in 70% of studies.

Of 8,402,919 people with obesity, 732,149 underweight MBS and 7,670,770 did not. Mean age of participants was 41 and 63% were female. Of the 38 studies that defined surgical eligibility by BMI cut-off, 23% used BMI > 30 kg/m2, 74% used > 35 kg/m2, and 8% used > 40 kg/m2. Only 25 studies reported psychiatric history at baseline, and of these, 32% of participants had at least 1 disorder or were taking psychiatric treatment.

From a meta-analysis of 93 people from 2 RCTs, MBS may improve depressive symptoms over 12–24 months (SMD = −0.40, 95% CI −1.04, 0.24), but the evidence is very uncertain, they authors stated. In a meta-analysis of 2,072 people from 18 NRS, found that MBS may improve depressive symptoms over 3–58 months (SMD = −0.56, 95% CI −0.87, −0.26; low certainty). Removing studies at critical risk of bias, with passive comparators, and that used a DJBL intervention resulted in similar effect sizes. In 5 NRS ineligible for meta-analysis, evidence of benefit was also found in at least 1 study (p = 0.03). Two additional NRS were consistent with the direction of effect in the meta-analyses.

From a meta-analysis of 6066 people from 15 NRS, MBS may improve non-normative eating symptoms over 6–58 months (SMD = −0.75, 95% CI −0.97, −0.53; low certainty). Symptoms included in this composite included disordered eating (4 studies), loss of control over eating (4 studies), emotional eating (4 studies), binge eating (1 study), weight preoccupation (1 study), and eating concern (1 study).

In 5 NRS ineligible for meta-analysis, there was no evidence of benefit on non-normative eating symptoms (p = 0.17), nor was there in another NRS or RCT conference abstract. However, an RCT found a large evidence of benefit, consistent with the meta-analysis.

An analysis of 271,240 people from ten NRS found MBS may slightly increase substance use disorders (RR = 1.66, 95% CI 1.00–2.78; 2/100 more people, 95% CI from 0 fewer to 5 more); however, heterogeneity was high (I2 = 99%) and there was a significant subgroup effect for duration of follow-up (test for difference p = 0.01). Within 2 years of follow-up, MBS probably does not reduce substance use disorders (RR = 0.97, 95% CI 0.64–1.46; 0/100 fewer people, 95% CI from 13 fewer to 17 more; NNT = 83; moderate certainty), but may slightly increase substance use disorders after 2 years (RR = 2.13, 1.33–3.42; 4/100 more, 95% CI from 1 to 9 more; NNH = 67; low certainty).

Substances included in this composite included alcohol (6 studies), any substance (1 study), and non-alcohol substance (1 study) use disorders. Sensitivity analyses showed consistent results. In 4 NRS ineligible for meta-analysis, there was no difference in substance use disorders (p = 0.80), however, a RCT aligned with the latter subgroup analysis suggesting evidence of harm.

From a meta-analysis of 167,042 people from 5 NRS found MBS may slightly increase suicide deaths over 4–21 years (RR = 1.86, 95% CI 1.07–3.21; 1/1000 more people, 95% CI 0 to 3 more; NNH = 1111), but the evidence is very uncertain. Results were consistent in a sensitivity analysis removing passive comparator studies.

Several meta-analyses on cognitive performance were conducted. In 271 people from 4 NRS, MBS may improve cognitive performance in attention over 3–24 months (SMD = −0.72, 95% CI −1.61, 0.17), but the evidence is very uncertain. No sensitivity analyses were possible. An NRS ineligible for meta-analysis did not align with this finding.

“Evidence suggests that in people with obesity, MBS may improve depression, anxiety, non-normative eating, and cognitive performance in attention, but also slightly increase substance use disorders and suicide death in the longer term. However, there was low to very low certainty in most findings,” the authors concluded. “Despite the reduced certainty, these findings reflect the best available evidence and can be considered by people who are contemplating or have received MBS, their families, healthcare workers, as well as guideline developers in the field of obesity.”

However, they called for more methodologically rigorous and adequately powered studies of MBS compared to best practice medical weight management to strengthen confidence in the effects of MBS on psychiatric and cognitive functioning. They said this could be achieved by incorporating psychiatric and cognitive measures into MBS RCTs as secondary outcomes or embedded sub-studies, and by adjusting for important confounding variables in high-quality NRS.

The findings were reported in the paper, ‘Psychiatric and Cognitive Functioning After Metabolic and Bariatric Surgery: A Systematic Review and Meta-Analysis’, published in Obesity Reviews. To access this paper, please click here