495

Genetic Factors Predict Weight Loss After Roux-en-Y Gastric BypassIda Hatoum, Danielle Greenawalt, Daniel M. Kemp, Marc Reitman, Pek Lum, Lee M.Kaplan

Introduction: Although Roux-en-Y gastric bypass (RYBG) leads to substantial weight loss,the magnitude of this effect is highly variable. The determinants of weight loss followingRYGB are poorly understood. Clinical factors have been able to account for only a fractionof the observed variation, and several studies have implicated additional biological determin-ants. In this study, we sought to determine the role of genetic contributions to weight lossafter RYGB. Methods: We collected liver samples from 1018 patients undergoing RYGB.DNA from these samples was genotyped for 660,000 single nucleotide polymorphisms(SNPs). Percent excess weight loss (%EWL) for the nadir weight after surgery was calculatedagainst a reference BMI of 25 kg/m2. We identified 40 genetically related individuals withinthis cohort, based on percentage of their genome shared. Using electronic medical records,we identified 20 additional individuals who were cohabitating and genetically unrelated.We used mixed effects models to determine whether degree of relatedness (1°, 2°, or 3°, orcohabitating) is associated with%EWL. In addition, we performed a genome-wide associationstudy (GWAS) of the 789 non-related Caucasians within the total cohort. Results: Theaverage difference in %EWL was 13.7%, 25.1% and 36.8% for 1°, 2°, and 3° relatives, and25.4% for cohabitating individuals, respectively. There was a significant association of %EWLbetween 1° relatives (p=0.04, Intraclass Correlation Coefficient [ICC]=57.3%), which wasnot seen between 2° relatives (p=0.54, ICC=38.2%), 3° relatives (p=0.36, ICC=0%), orcohabitating individuals (p=0.53, ICC=0). The GWAS identified a single region on chromo-some 6 that was significantly associated with %EWL (p=2.1x10-7; false discovery rate=9.2%for the highest ranked SNP in this region). Conclusions: These data suggest that there aregenetic determinants of weight loss after RYGB. Despite the small number of related indi-viduals, the significant correlation between outcomes of 1° relatives indicates that there isa significant genetic impact beyond any shared environmental influences. In addition, evenin this modest-sized cohort, we identified a chromosomal region significantly associatedwith %EWL. These observations underscore the biological nature of the response to RYGBand suggest that variation in the genetic background of individuals influences weight lossafter this operation. Identification of the genes involved in this response could provideinformation about the method of action of RYGB and facilitate strategies for selecting patientsmost likely to benefit from this effective yet invasive procedure.

496

Quality-Adjusted Life Expectancy Benefits of Laparoscopic Bariatric SurgeryLyndon V. Hernandez, Dominic Klyve

Background: The method of choice for bariatric surgery remains controversial and is depend-ent on the surgeon's expertise and experience. Moreover, there has been no comparativedata on survival benefit among the various types of procedures. Aims: To compare theoutcome of laparoscopic Roux-en-Y gastric bypass (L-RYGB) versus laparoscopic adjustablegastric banding (LAGB) using quality-adjusted life years (QALYs). Methods: For the timeperiod immediately following surgery, we predicted the body mass index (BMI) for eachcohort for each year as a function of choice of surgery using published data. We developeda Markov model of the quality of life and survival of L-RYGB and LAGB in obese patients.Using census data, we estimated the probability of dying and quality of life for each yearof each cohort. Results: For all cohorts, L-RYGB offers the highest advantage in QALYscompared to gastric banding. The youngest cohort showed the greatest discrepancy betweenthe two surgical methods, with 7.8, 6.4, and 4.7 QALYs gained with L-RYGB over LAGBfor the age groups 35, 45, and 55, respectively. Those with the highest pre-surgical BMIacquired the most advantage with L-RYGB, with 2.8, 6.4, and 9.6 QALYs gained with L-RYGB over LAGB for the body mass index (BMI) groups 40, 50, and 60. Males had a slightlyhigher advantage with L-RYGB, with 6.5 QALYs gained with L-RYGB over LAGB comparedto 6.5 QALYs for females. To represent a typical bariatric patient, a 35 year old woman willhave varying degrees of difference in QALYs depending on her pre-surgery BMI and thetype of intervention she chooses (Table 1). Thus, a 35 year old woman with a BMI of 60kg/m2 will gain 11 quality-adjusted life years if she chooses L-RYGB instead of LAGB.Conclusion: Across all age and BMI cohorts, patients can expect a longer quality-adjustedlife expectancy following L-RYGB surgery, showing the early advantage gained by LAGBfrom lower peri-operative morbidity is offset over the long-term due to its limited post-surgical weight loss. The young and extremely obese are core groups who will gain the mostQALYs following L-RYGB. Our findings should provide critical information when helpingpatients decide on the type of bariatric surgery that is most appropriate for them and asbaseline data for funding agencies to assess nascent technologies such as endoluminalbariatric procedures.Quality-Adjusted Life Expectancy (Years) for 35 Year Old Woman

497

Safety and Efficacy of the Trans-Oral Endoscopic Restrictive System (TERIS®)for the Treatment of ObesityKoen de Jong, Elisabeth Mathus-Vliegen, Elisabeth Veldhuyzen, Jan H. Eshuis, PaulFockens

Background and Aims: Medical and dietary treatment of obesity often fail. Surgical treatment,although successful, carries considerable morbidity. Aminimally invasive endoscopic implantprocedure is developed and studied in an animal model with encouraging results. We present

S-71 AGA Abstracts

the 6-months results of this implant device (TERIS, Barosense Inc, Redwood City, USA) forthe treatment of morbid obesity in humans. Aims of this study are to investigate safety andeffectiveness. Primary endpoints were adverse events. Secondary endpoints were weight lossand improvement in co-morbid disease(s). Methods: Patients with a Body Mass Index (BMI)between 40 and 50 kg/m2 or a BMI between 35 and 40 kg/m2 plus obesity related co-morbidities, were invited to participate. Patients with prior esophagogastric surgery wereexcluded. Procedures were performed under general anesthesia. The procedure consistedof placement of a 22mm overtube followed by stapling of 5 transmural plications surroundingthe esophagogastric junction in the stomach. Subsequently 5 silicone anchors were placedthrough the plications and a restrictive silicone device with a 10 mm orifice was attachedto the anchors. Patients were followed for 6 months, after which the device was eitherremoved or kept in place for 6 additional months. Results: Thirteen patients (3 male) witha median BMI of 42.1 kg/m2 (35-49) have been enrolled between 09/08 and 04/09. Medianprocedure time was 142 minutes (93-184). In two patients (#2 and #4) a pneumoperitoneumwas detected which was desufflated percutaneously in one and treated conservatively in theother. In the seventh patient the procedure was abandoned after a gastric perforation dueto stapler malfunctioning. After these complications, the stapling device was redesigned andpressure controlled insufflation with CO2 was used instead of air. No subsequent proceduralcomplications were seen after these adjustments. At six months patients reported no side-effects and obtained a median excess weight loss of 37.6% (9-56). Median BMI decreasedfrom 42.1 to 35.8 kg/m2 (30-47). In 11 patients a gastroscopy was performed after 6 monthsand 9 patients wished to keep the restrictor for an additional 6 months. Conclusions: Thisphase 1 study of a new trans-oral endoscopic restrictive implant system (TERIS) for thetreatment of obesity showed successful placement in 12 out of 13 patients. Proceduralcomplications occurred in 3 patients. Weight loss was excellent and comparable to theresults of laparoscopic gastric band placement. Long-term follow-up and randomized studiesare required to determine its place in treating morbid obesity.

498

Obstructive Sleep Apnea is Prevalent and Persistent Among PatientsUndergoing Bariatric Surgery: A Systematic ReviewJon-Erik C. Holty, Barrett G. Levesque, Jennifer Schneider Chafen, Vaughan Tuohy, JohnR. Kapoor, Ingram Olkin, Dena M. Bravata

Background and Aims: We performed a meta-analysis to estimate obstructive sleep apnea(OSA) prevalence in bariatric candidates and to evaluate OSA outcomes among bariatricrecipients. Methods: We sought bariatric surgery studies (1950-2008) meeting predefinedstandards from 4 databases (Medline, ISI SciSearch, CINAHL, Cochrane Library) and biblio-graphies of retrieved articles.Two authors independently abstracted patient data. Results:Two-hundred and forty-seven reports, describing 143 unique bariatric populations (N=57,632) met inclusion standards. The overall reported prevalence of OSA (123 studiesdescribing 39,111 bariatric candidates) was 31.7% (range: 0.0% to 96.6%). However amongthose 18 studies that performed polysomnographic evaluations on all bariatric candidates (n=2817)the OSA prevalence was 79%. Among bariatric surgery candidates, OSA was relativelycommon compared with other obesity-related comorbidities such as hypertension (44%),diabetes (24%), or asthma (17%). Approximately 50%of bariatric recipients with preoperativeOSA had persistent disease after surgery. Bariatric recipients with OSA experienced higherpostoperative morbidity and mortality than patients without OSA. Change in body massindex was not predictive of greater apnea-hypopnea index (AHI) reduction. Despite main-tained weight loss, initial AHI improvements appeared to decrease over time. Conclusions:OSA is highly prevalent among bariatric surgery candidates, and is associated with greaterrisk for postoperative morbidity and mortality. OSA persists among many patients afterbariatric surgery despite weight loss. Clinicians should be aware of the risk of OSA amongbariatric candidates, and refer them for both pre- and post-operative polysomnographic evalu-ation.

499

Morbidity and Mortality After Bariatric Surgery in Sweden: A NationwidePopulation Based StudyMagdalena Plecka Östlund, Richard Marsk, Finn Rasmussen, Jesper Lagergren, ErikNaslund

It is unclear whether the reduction of comorbid disease and mortality seen after bariatricsurgery approaches that of the general population. We assessed the risk of hospitalizationfor myocardial infarction, angina pectoris, stroke, hypertension, diabetes type 2 and overallmortality after bariatric surgery in relation to the general population. A nationwide Swedishpopulation-based cohort study consisting of 13 273 patients who underwent bariatric surgeryduring the period 1980-2006. From the Register of the Total Population a control cohortconsisting of 10 randomly selected persons (matched for gender and age) was created. Dataon inpatient care and death was obtained from the Swedish Inpatient Care Register and theCause of Death Register respectively. Cox regression was used to calculate hazard ratios(HR) with 95% confi-dence interval (CI). The overall adjusted risk remained elevated afterbariatric surgery: myocardial infarction (HR 1.5; 95% CI 1.3-1.8), angina pectoris (HR 2.0;95% CI 1.8-2.3), stroke (HR 2.1; 95% CI 1.9-2.4), hyper-tension (HR 2.8; 95% CI 2.6-3.0) and diabetes (HR 2.4; 95% CI 2.2- 2.7). Similarly, the adjusted risk of mortalityremained increased after bariatric surgery (HR 1.2; 95% CI 1.1-1.3). In the sub-groupoperated with gastric bypass (4161 patients) there was no statistically significant differencebetween the surgical and control group regarding hospitalization for diabetes (HR 1.2; 95%CI 0.9-1.7) and myocardial infarction (HR 0.8; 95% CI (0.4-1.4)). Mortality remainedelevated after gastric bypass (HR 2.2; 95% CI 1.8-2.7). The most common cause of deathin the surgery cohort was coronary heart disease, followed by cancer, while the oppositewas seen in the general population. No gender differences were identified. In general, bariatricsurgery does not appear to be followed by a decreased risk of inpatient care for obesity-related comorbidities to population levels. However gastric bypass, in particular, seems toreduce the risk for diabetes and myocardial infarction to population levels.

AG

AA

bst

ract

s