Faculty Bibliography

Aims: Gatstro-gastric fistula (GGF), a complication of roux-en-y gastric bypass (RYGB), has an occurrence rate of 1-2% and can result in weight regain, relapse of comorbidities, and marginal ulceration. Surgical management varies-revision of the bypass with concurrent remnant gastrectomy is often avoided for fear of complexity, while division of the fistula, remnant gastrectomy, and endoscopic repairs can result in recurrence of the GGF or marginal ulceration. Institutional data on GGF after RYGB and the step-by-step revision to an esophagojejunostomy are herein presented. Methods: This is a retrospective review of prospectively collected data at a single institution and video presentation. Results: Between 2005 and 2014, 13 patients presented with GGF after RYGB. The mean time to presentation was 4.4 years [range: 7 mos-16.4 years]. Surgical treatments included resection of GGF, remnant gastrectomy, and subtotal gastrectomy with either revision of gastrojejunostomy or esophagojejunostomy. Mean OR time was 191 minutes. Four patients required an additional reoperation for recurrent marginal ulcer (2), persistent GGF (1), and enterocutaneous fistula (1). Patients who underwent full resection of the gastric remnant and conversion to esophagojejunostomy did not develop postoperative complications. Video Presentation: A 47-year-old male s/p laparoscopic RYGB one year prior presented with 1 day of abdominal pain and was diagnosed with a GGF on CT scan. His weight loss was adequate and his BMI was 28.8. Marginal ulceration was presumed to be the source of his acute pain and the patient was taken to the OR for revisional surgery. Intraoperatively, the roux-limb and remnant stomach were divided. The GE junction was marked and divided proximally to the GGF and an endoscopic stapler created an end-to-side esophagojejunal anastomosis. The gastric remnant, gastrojejunostomy, and GGF were removed through the umbilical incision. Incidentally found internal and hiatal hernias were also repaired. Total operative time was 162 minutes. On 6 months follow-up the patient has no complications. Conclusion: Surgical treatment of GGF after RYGB should include revision to esophagojejunostomy with complete excision of the gastric remnant. In experienced hands, this operation is safe and effective and has a lower likelihood of recurrent GGF and marginal ulceration compared to other procedures

Introduction: The impact of resident involvement on outcomes after bariatric procedures is not well understood. Prior studies have demonstrated increased complication rates with resident involvement in Roux-en-y gastric bypass (RYGB). These studies did not include data for laparoscopic sleeve gastrectomy (LSG) and laparoscopic adjustable gastric band (LAGB). At our institution, attending surgeons operate with both residents and physician assistants (PA) interchangeably, thus controlling for surgeon variability. Our objective was to demonstrate that resident involvement in complex bariatric surgeries does not increase complication rates when residents and PA's work with the same attending surgeons. Methods and Procedures: Patients undergoing bariatric procedures at our institution between 3/2012 and 3/2015 were identified using the Metabolic and Bariatric Surgery Quality Improvement Program (MBSAQIP) database. All patients under 18 years of age were excluded. Cases were stratified into four different categories- RYGB, LSG, LAGB, and LAGB revision (replacement, removal, or port revision). Data included patient demographics, comorbidities, length of stay, and 30-day complications. The primary outcomes of the study were operative time and 30-day overall complication rate. Secondary outcomes included length of stay, major complications and reoperation rates. Results: There were 2741 bariatric surgeries performed from 2012-2015. Of those, 2067 had resident involvement and 674 had PA involvement. 30-day complication rates for all surgery types, with and without residents, were 5.3 % and 6.1 %, respectively (p = .45). Complication rates for LSG (p = .716), LAGB (p = .694), LAGB revision (p = .493), and RYGB (p = .126) were also not significant. Operative duration for all surgery types was longer with residents (77.0 vs 60.6 min, p<.0001). Operative duration was longer with resident involvement for LSG (101.1 vs 76.6 min, p<.0001), LAGB (51.6 vs 42.4 min, p<.0001), and LAGB revision (63.0 vs 51.6 min, p = .007). After risk adjustment, OR time for RYGB was significantly longer as well (134.3 min vs 119.3 min, p = .038). Length of stay was also found to be significantly increased in the resident group (1.23 vs 1.09 days, p = .0007). Conclusion: Resident involvement as first assistant in the OR does not increase complication rates in bariatric surgery. Operative times and length of stay are prolonged, as shown in other studies, but no difference was seen in complication rates for cases involving a resident and cases involving a dedicated operative PA. Teaching advanced bariatric surgery techniques to residents is both safe and essential to their education

BACKGROUND: Thirty-day readmission post-bariatric surgery is used as a metric for surgical quality and patient care. We sought to examine factors driving 30-day readmissions after laparoscopic sleeve gastrectomy (LSG). METHODS: We reviewed 1257 LSG performed between March 2012 and June 2014. Readmitted and nonreadmitted patients were compared in their demographics, medical histories, and index hospitalizations. Multivariable regression was used to identify risk factors for readmission. RESULTS: Forty-five (3.6 %) patients required 30-day readmissions. Forty-seven percent were readmitted with malaise (emesis, dehydration, abdominal pain) and 42 % with technical complications (leak, bleed, mesenteric vein thrombosis). Factors independently associated with 30-day readmission include index admission length of stay (LOS) >/=3 days (OR 2.54, CI = [1.19, 5.40]), intraoperative drain placement (OR 3.11, CI = [1.58, 6.13]), postoperative complications (OR 8.21, CI = [2.33, 28.97]), and pain at discharge (OR 8.49, CI = [2.37, 30.44]). Patients requiring 30-day readmissions were 72 times more likely to have additional readmissions by 6 months (OR 72.4, CI = [15.8, 330.5]). CONCLUSIONS: The 30-day readmission rate after LSG is 3.6 %, with near equal contributions from malaise and technical complications. LOS, postoperative complications, drain placement, and pain score can aid in identifying patients at increased risk for 30-day readmissions. Patients should be educated on postoperative hydration and pain management, so readmissions can be limited to technical complications requiring acute inpatient management.

BACKGROUND: Short- and mid-term data on Roux-en-Y gastric bypass (RYGB) indicate sustained weight loss and improvement in co-morbidities. Few long-term studies exist, some of which are outdated, based on open procedures or different techniques. OBJECTIVES: To investigate long-term weight loss, co-morbidity remission, nutritional status, and complication rates among patients undergoing RYGB. SETTING: An academic, university hospital in the United States. METHODS: Between October 2000 and January 2004, patients who underwent RYGB>/=10 years before study onset were eligible for chart review, office visits, and telephone interviews. Revisional surgery was an endpoint, ceasing eligibility for study follow-up. Outcomes included weight loss measures and rates of co-morbidity remission, complications, and nutritional deficiencies. RESULTS: RYGB was performed in 328 patients with a mean preoperative body mass index of 47.5 kg/m2. Of 294 eligible patients, 134 (46%) were contacted for follow-up at>/=10 years (10+Year follow-up). Mean percentage excess weight loss (%EWL) was 58.9% at 10+Year. Higher %EWL was achieved by non-super-obese versus super-obese (61.3% versus 52.9%, P = .034). Blood pressure, lipid panel, and hemoglobin A1c improved significantly. At 10 years, remission of co-morbidities was 46% for hypertension and hyperlipidemia and 58% for diabetes mellitus. Thirty patients (9%) had revisional surgery for weight regain. Sixty-four patients (19.5%) had long-term complications requiring surgery. All-cause mortality was 2.7%. Nutritional deficiencies were seen in 87% of patients. CONCLUSIONS: Weight loss after RYGB appears to be significant and sustainable, especially in the non-super-obese. Co-morbidities are improved, with a substantial number in remission a decade later. Nutritional deficiencies are almost universal.

Objectives: Over the last decade there has been a sharp increase in the use of laparoscopic sleeve gastrectomy (LSG) relative to Roux-en-Y gastric bypass (RYGB) and laparoscopic adjustable gastric band (LAGB). We sought to further examine national trends in bariatric surgery and determine whether the changes in bariatric procedure use over time, particularly the growth in LSG, were uniformly represented among clinical subgroups in the US. Methods: Patients with a body mass index (BMI) >35kg/m² who underwent bariatric surgery from 2005-2012 in the American College of Surgeons-National Surgical Quality Improvement Program (ACS-NSQIP) database were identified. Procedure use, year of surgery, and patient population were evaluated. Logistic regression was used to analyze yearly trends in LSG versus RYGB use among clinically relevant subgroups. Results: We identified 100,304 patients who underwent bariatric procedures between 2005 and 2012 in the NSQIP database. Laparoscopic RYGB was the most common bariatric surgery type performed each year, but had a decreasing trend (75% of cases in 2005 vs. 53.2% in 2012) [Figure 1]. From 2010 to 2012, LSG increased substantially from 9.5% to 34.4% of bariatric procedures performed per year, while LAGB decreased from 28.8% to 9.7%. Over this period, minority patients represented an increasing proportion of those who underwent LSG compared to Caucasian patients. The proportion of LSG patients who were elderly (>65 years), superobese (BMI >50kg/m²), diabetic, high-risk, and ASA class 4-5 also rose, but to a significantly lesser degree than patients <65 years, BMI <50kg/m², non-diabetics, low-risk patients, and ASA class 1-3, respectively [Table 1]. In multiple logistic regression, factors independently associated with having LSG over RYGB in 2012 were age <65 (OR=1.58, 95% CI=[1.32-1.89], p<0.0001]), minority race (OR=1.30, 95% CI=[1.21-1.40], p<0.0001]), non-diabetic (OR=1.51, 95% CI=[1.39-1.63], p<0.0001), low-risk (OR=1.38, 95% CI=[1.27-1.50], p<0.0001]), ASA class 1-3 (OR=1.25, 95% CI= [1.01-1.55], p<0.0037), and BMI <50kg/m2 (OR=1.25, 95% CI=[1.16-1.35], p<0.0001]). Conclusions: LSG use has grown substantially over the last decade, but to a greater extent among certain patient subgroups. Although LSG was developed as the first stage of a two-stage biliopancreatic diversion-duodenal switch for high-risk patients, LSG is being performed on patients who are younger with less risk, whereas RYGB is targeted to those with more severe cases of obesity, surgical risk, and diabetes. Further studies will be necessary to clarify whether this approach reflects surgical preference or has been influenced by patient preferences

Intro: Laparoscopic adjustable gastric banding is well-known for its safety profile. However, band removal, especially for a device-related complication, may be more complex due to the scar tissue created by the band. The objective of this study is to review perioperative outcomes of patients requiring band removal for device-related complications. Methods: A retrospective review was conducted of all band removals over a 13 year period (2001-2014) for a device-related complication (e.g. slippage, erosion, gastric necrosis). Bands removed for weight loss failure or intolerance were excluded from this review. Perioperative complication, readmission and reoperation/re-intervention was defined according to the Metabolicand Bariatric Surgery Accreditation and Quality Improvement Program standards. Results: A total of 104 patients required band removal for a device-related complication. In the same time frame 7633 bands were implanted. The average age at band removal was 44 years old and the average BMI was 35.6. The most common reason was slip (42%) and erosion (28%). The 30-day complication rate from the removal was 26% (27/104) - most commonly pneumonia and perigastric abscess. The 30-day readmission rate and reoperation/ re-intervention rate were 15% and 10%, respectively. There was one mortality (1%) from septic shock secondary to erosion. There were no statistically significant differences in age (p = 0.452) or BMI (p = 0.523) between those who had a 30-daycomplication and those who did not. Conclusions: Band-related complications are rare. Band removal for device-related complication may be associated with significant morbidity

Introduction: Several studies have shown that obesity is more common among patients with psoriasis and psoriatic arthritis, and this correlation may be related to the systemic inflammation associated with obesity. Although bariatric surgery has been shown to improve several obesity-related comorbidities, the effects of surgical weight loss on psoriasis and psoriatic arthritis have not been adequately studied. Our objective was to investigate the effects of weight loss from bariatric surgery on psoriasis and psoriatic arthritis. Methods: A retrospective database of 9,073 bariatric surgeries performed at a single center between 2002 and 2013 was queried. Patients with a diagnosis of psoriasis prior to bariatric surgery were identified. Preoperative demographic, anthropometric, and comorbidity data were collected. Patients were contacted about their history of psoriasis, changes in symptoms after surgery, diagnosis of psoriatic arthritis, and treatment modalities for psoriasis and psoriatic arthritis pre- and postoperatively. The primary outcome was the percentage of patients who reported improvement in psoriasis after surgery. Secondary analyses were performed to define factors associated with improvement in psoriasis. Results: We identified 128 patients with a preoperative diagnosis of psoriasis. Seventy-four (58%) patients completed the study. Baseline patient characteristics are listed in Table 1. The mean time from surgery was 6.2 years, with a mean excess weight loss (EWL) of 46.5%. At the time of contact, forty-one (55%) patients reported improvement in their psoriasis, 24.3% reported improvement with subsequent relapse, 6.8% had no change, and 12.6% reported that their psoriasis progressively worsened. Sixteen (22%) patients also had a preoperative diagnosis of psoriatic arthritis; 62.5% reported improvement in their psoriatic arthritis, whereas 19% had no change and 19% worsened. In secondary analyses, lower preoperative BMI (43.7kg/m² vs. 48.4 kg/m², p=0.004) was found to be independently associated with postoperative improvement in psoriasis. Patients with severe psoriasis at the time of surgery and significant postoperative improvement, excluding those whose improvement may have been due to escalation in medication class, demonstrated greater weight loss (101.4 lb vs. 66.0 lb, p=0.025) and EWL (63.7% vs. 44.7% EWL, p=0.028). Similarly, improvement in psoriatic arthritis was associated with greater EWL, but this did not reach statistical significance (51.4 vs. 48.3, p=0.815). Conclusion: Although the natural history of psoriasis and psoriatic arthritis is typically chronic, a majority of patients experience improvement after bariatric surgery. Based on our results, there is an association between excess weight loss and symptomatic improvement in severe cases of psoriasis. Factors such as lower preoperative BMI may be used to identify those patients with a greater likelihood of remission. Additionally, ours is the first study to show an improvement in psoriatic arthritis after bariatric surgery and a possible association between surgical EWL and improvement in psoriatic arthritis. Larger prospective studies are needed to further define the true effect of surgical weight loss on psoriasis and psoriatic arthritis

BACKGROUND: Marginal ulcers (MUs) are potentially complex complications after Roux-en-Y gastric bypass. Although most resolve with medical management, some require surgical intervention. Many surgical options exist, but there is no standardized approach, and few reports of outcomes have been documented in the literature. The objective of this study was to determine the outcomes of surgical management of marginal ulcers. METHODS: Data from all patients who underwent surgical intervention between 2004 and 2012 for treatment of MU after previous Roux-en-Y gastric bypass were reviewed. RESULTS: Twelve patients with MUs underwent reoperation. Nine patients had associated gastrogastric fistulae (75%). The median time to reoperation was 43 months. Ten patients underwent subtotal gastrectomy, of which 9 had a revision of the gastrojejunal anastomosis and 1 did not. One underwent total gastrectomy with esophagojejunal anastomosis for ulcer after previous revisional partial gastrectomy, and 1 patient underwent video-assisted thoracoscopic truncal vagotomy for persistent ulcer-related bleeding in the early postoperative period. Three patients (25%) experienced postoperative complications associated with revisional surgery requiring reoperation. At median follow-up time of 35 months, 7 patients (58%) had chronic abdominal pain, and 4 patients (33%) had intermittent diarrhea. Three patients (25%) were lost to recent follow-up. None had recurrence of MU. CONCLUSION: Patients can undergo one of several available surgical interventions, including laparoscopic subtotal gastrectomy with gastrojejunostomy revision. Though this appears to offer definitive treatment of MU, its benefits must be weighed against the increased risk of significant postoperative complications and chronic symptoms related to revisional surgery.