Ongoing substance abuse, poorly-controlled depression, or other major psychiatric illness are considered to be a contraindication to bariatric surgery [6
]. Although mental health screening is recommended—and common—prior to weight loss surgery, there is no consensus on to the type or structure of psychological assessment. Moreover, generally, the correlation between mental health and postoperative outcomes has not been elucidated. More specifically, there is little data examining the relationship between a prior history of substance abuse and bariatric surgery outcomes. In this population, a multidisciplinary approach may be the key to ensure good surgical outcomes. In this study we have shown that patients with SA exhibit equivalent weight loss at 6 and 12 months compared to the patients who have no history of alcohol and/or drug abuse, which may be a reflection of intensive workup and close followup by a dedicated multidisciplinary team. This study also demonstrates low postoperative substance abuse rates in SA patients. More patients with a history of SA developed illicit drug or alcohol abuse after bariatric surgery compared to their NA counterparts, although we did not identify a statistically significant difference. This, however, may reflect a small number of study subjects.
American veterans have been found to have a rate of current alcohol abuse of 4.0% and a rate of illicit drug use of 1.9% [5
]. These values were derived from a self-reported study, including both current use and/or a previously recorded diagnosis of substance abuse, which may have underestimated actual rates of substance abuse. Our study population, which focused on veterans who had undergone bariatric surgery, had a prevalence of drug and alcohol abuse that is ten times higher than that reported by Chwastiak et al. This may reflect that the data in our study are not simply self-reported but rather relied on specific institutional screening methods, in addition to the electronic medical record. Interestingly, multiple studies report a decreased lifetime incidence of substance abuse among obese patients [7
], while others have found no association between BMI and substance abuse [8
]. Although we did not specifically compare the prevalence of substance abuse in the morbidly obese cohort at Palo Alto VA and that of a normal weight cohort, we certainly found high rates of prior SA in the bariatric surgical population. Other studies found that obese persons had significantly increased risk of alcohol use disorder, in addition to other psychiatric disorders [8
This study emphasizes the importance of a thorough mental health screening for bariatric surgery that provides an opportunity for timely intervention and appropriate, individualized postoperative followup. Other psychiatric conditions are also found with higher prevalence in the obese population seeking bariatric surgery. Axis I psychiatric disorders in general are prevalent in the bariatric population [6
], including higher rates of a history of sexual abuse, as well as higher rates of posttraumatic stress disorder (PTSD), which are also more prevalent in the veteran population [9
]. Active substance abuse is readily monitored and screened for using routine toxicology tests that provide quantitative, binary results. Thus, it is possible to insist on preoperative abstinence from illicit drug or alcohol use, and compliance can be closely followed before and after surgery.
Our results support the results of other bariatric surgery programs treating a nonveteran population. Heinberg and Ashton examined excess weight loss in patients who have a history of drug and/or alcohol abuse compared to those with no such history [11
]. They showed that early weight loss is equivalent for those with and without a history of substance abuse. However, at 6, 9, and 12 months, the substance abuse groups lost significantly more weight than the control group. The present study showed greater weight loss in the SA group compared to the NA group (RYGB, LSG only), but this difference was not significant, possibly due to the small sample size. In addition, the rates of prior substance abuse in the Heinberg and Ashton study were much lower than in this study (10.9% versus 36%, resp.), perhaps reflecting the difference between the general and the veteran populations. In another study of 80 bariatric patients followed for two years, Clark et al. reported a history of drug or alcohol abuse in 13% [12
]. They also found that the history of abuse translated favorably into greater weight loss after two years. One potential reason SA patients do just as well postoperatively as the NA counterparts may be due to past treatment received for their substance abuse disorders. Most substance abuse interventions involve learned coping skills and utilizing social support as an alternative to relying on addictive behaviors for managing stress. These skills may be useful in the postoperative bariatric population as well. In addition, the SA patients are required to have a period of sobriety prior to surgery. As a result, they have demonstrated the ability to make positive health behavior changes, which may be indicative of their ability to follow a strict postoperative weight management plan.
The relatively small sample sizes, particularly in the laparoscopic gastric band group, increase the risk of type II error in this study. While we found no difference in percent EWL between the SA and NA groups, perhaps a larger sample size would demonstrate a difference.
Recent attention has been given to the “addictive behavior” of obese patients and the replacement of one addiction (i.e., food) for another (e.g., alcohol) [13
]. However, Suzuki et al. found the prevalence of alcohol use disorders among patients who have undergone bariatric surgery to be similar to the general population. Nonetheless, they found more postoperative alcohol use disorders in those patients who had a prior history of abuse, compared to those with no history of abuse [15
]. This is consistent with our findings of a six-fold increased risk of recurrent abuse in patients with a prior preoperative history of drug or alcohol abuse. Others have found that weight loss surgery, in and of itself, predisposes the patients to substance abuse after bariatric surgery in the period 12 to 24 months postoperatively [16
]. Although it appears that the patients with a history of SA were more likely to develop substance abuse postoperatively in our study, the number of cases was too small to identify a statistically significant difference. In addition, the overall low rates of postoperative substance abuse may reflect the close involvement of behavioral medicine specialists in our clinic, throughout the postoperative course. A larger study with a more prolonged postoperative followup course will be needed to elucidate this point.