AGA Abstracts Mo1949 ASSOCIATIONS AMONG ADIPOSITY, MEDIATING BIOMARKERS AND COLORECTAL ADENOMATOUS POLYPS RISK IN A JAPANESE POPULATION Keisuke Nakai, Jiro Watari, Mayu Takimoto, Kumiko Nakamura, Akio Tamura, Ken Hara, Takahisa Yamasaki, Masaya Kobayashi, Takashi Kondo, Katsuyuki Tozawa, Yukako Taniguchi, Toshihiko Tomita, Yoshio Ohda, Tadayuki Oshima, Hirokazu Fukui, Jun Sakurai, Yongmin Kim, Yuji Hayakawa, Takashi Fujisawa, Hiroto Miwa Background: Obesity has been shown to be an increased-risk factor for colorectal adenoma- tous polyps (APs). There is a link between obesity-related adipokines (including adiponectin and leptin) and APs, and between homeostatic model assessment of insulin resistance (HOMA-IR) values and APs. However, the effects of obesity and adipose tissue-related biomarkers on the development of APs have been unclear. Aim: We sought to clarify the associations among obesity (i.e., body mass index [BMI] and visceral obesity), adipokines, HOMA-IR, and APs in a multicenter cross-sectional study of a Japanese population. Methods: From April 2014 to October 2016, a total of 402 consecutive individuals (215 men and 187 women) underwent a first colonoscopy at one of three hospitals in Nishinomiya and Himeji cities, Japan. Obesity was defined as BMI (≥25 kg/m 2 ), waist circumference (WC) (males, ≥85 cm; females, ≥90 cm) and waist-to-hip ratio (WHR) (≥0.9 for males and ≥0.85 for females). Metabolic syndrome (MetS) was defined based on the Japanese criteria. The serum levels of adiponectin, leptin, and HOMA-IR (glucose [mg/dL] ×insulin [mU/L] / 405) were also evaluated. We compared obesity, MetS, and adipose tissue-related biomarkers in the patients with APs (n=206) and without APs (controls; n=196). Results: (1) The prevalence of APs was 59.1% in the men and 42.2% in the women (p<0.001). (2) In a univariate analysis, elderly (≥65 yrs) status, BMI, WC, WHR, MetS, and high serum level of HbA1c were each significantly associated with APs in both genders. Alcohol drinking in the men and high serum levels of triglycerides and low-density lipoprotein cholesterol in the women were associated with APs. In the multivariate logistic regression analysis, elderly status (relative risk [RR]=4.93, 95%CI 2.52-9.64, p<0.0001) and WC (RR=2.60, 95%CI 1.12- 5.69, p=0.02) were independent risk factors for AP in the men, and elderly status (RR= 2.78, 95%CI 1.36-5.66, p=0.005) and BMI (RR=3.75, 95%CI 1.17-12.0, p=0.03) were independent risk factors for AP in the women, but MetS was not a risk factor in either gender. (3) The number of APs was significantly associated with WC and WHR (indicators of visceral obesity) in the men and with all obesity indices including BMI in the women, but a positive association between AP number and obesity was not seen. (4) No significant associations between adipokines and APs or between HOMA-IR and APs were observed, S-830 AGA Abstracts although all of these biomarkers showed strong positive correlations with the three anthropo- metric indices (BMI, WC, and WHR; p<0.001 each). Conclusions: The risk factors for APs in a Japanese population were different between men and women; visceral obesity as characterized by WC was a risk factor in the men, and BMI was a risk factor in the women. In contrast, adiposity-related adipokines and HOMA-IR may not be involved in the development of APs. Mo1950 RISK OF SUICIDE AND SELF-HARM IS INCREASED AFTER BARIATRIC SURGERY: A SYSTEMATIC REVIEW AND META-ANALYSIS Praneet Wander, Daniel Castaneda, Violeta Popov, Christopher C. Thompson Introduction: Annual rates of bariatric surgery have risen dramatically over the past decade and are expected to increase further, as bariatric surgery has been endorsed by multiple societies for treatment of obesity. However, the long term psychiatric effects of surgery in this population have been a concern. Primary Aim: To assess the impact of bariatric surgery on the occurrence of suicide attempt/self-harm and suicide rates. Methods: MEDLINE, Embase, and Web of Science were searched from inception through September 2016 with MeSH terms "suicide", "mortality", "outcomes", "self-harm" and "bariatric surgery". Dual extraction and quality assessment of studies using the Newcastle-Ottawa scale was performed independently by two authors. Inclusion criteria were retrospective and prospective studies reporting long-term outcomes (greater than 12 months) after bariatric procedures, with greater than 100 patients in a cohort, and reporting rates of self-harm and suicides rates. Primary outcomes included the pooled event rate and 95% confidence interval (95% CI) for suicides and self-harm admissions after bariatric surgery, and risk ratio comparing rates of self-harm before and after surgery. Results: 121 citations were identified; ten prospective cohort studies and six retrospective or cross-sectional cohort studies including 89,595 subjects were analyzed. The average follow-up was 3.59 years, range (0.85-35 years), the ages of participants ranged from 26.1 to 44.6 years. There were 75.89% female patients. The majority of the patients had undergone Roux-en-Y gastric bypass (54.8%), followed by laparoscopic gastric banding (11.9%), and other restrictive procedures. In 32.1% of the cases, no specific procedure could be identified. After bariatric surgery, 2 per 1000 patients committed suicide, event rate 0.002 (95% CI 0.0013, 0.0032), I square=76%, Tau square= 0.47, 14 studies, 80,172 patients (Figure 1); and 12 per 1000 patients were admitted for suicidal or self-harm attempt, event rate 0.012 (95% CI 0.004, 0.036). The risk of self- harm/suicide attempt admission was also increased after bariatric surgery within the same patient population: risk ratio of 1.9 (95% CI 1.23, 2.95), I square=99%, Tau square=0.15, 3 studies (43 406 patients) (Figure 2). Conclusion: Bariatric surgery is associated with twice the risk of self-harm/suicide attempt compared to before surgery in the same patient cohorts. Also, higher risk of suicide (2/1000) compared to the general population (1.2/1000) was noted after weight loss surgery. Further studies are needed to identify modifiable factors, and to improve patient selection and follow-up protocols. FIGURE 1: Event rate for self-harm /suicide attempt after bariatric surgery (I sq=99%, Tau sq=1.7, 5 studies, 48 368 patients) FIGURE 2: Rate ratio for self-harm is twice as high after bariatric surgery than before surgery. (I square=99%, Tau sq=0.15, 3 studies, 43 406 patients) Mo1951 PSYCHIATRIC DISEASE DOES NOT NEGATIVELY IMPACT WEIGHT LOSS OUTCOMES FOLLOWING ENDOSCOPIC REVISION FOR FAILED ROUX- EN-Y GASTRIC BYPASS Austin L. Chiang, Pichamol Jirapinyo, Christopher C. Thompson Background: Mental health issues have been shown to impact quality of life and an individu- al's ability to adhere to treatment recommendations and engage in physical activity. The impact of psychiatric disease (PD) on weight loss after bariatric surgery remains unclear, as