Faculty Bibliography

Introduction: Airway dysfunction has been detected by oscillometry in obese subjects despite normal large airway function as assessed by spirometry. This has been attributed to lung/airway compression as reflected by reduced FRC; we previously demonstrated improvement of abnormalities towards normal upon voluntary inflation to predicted FRC (AJRCCM 2010; 181:A2532). However, other causes of airway dysfunction such as inflammation or concomitant intrinsic airway disease may coexist and could not be excluded. The present study re-evaluated these subjects following bariatric surgery induced weight loss to evaluate for residual abnormality. Methods: 22 morbidly obese subjects without history of smoking and/or cardiopulmonary disease, underwent evaluation pre/post bariatric surgery (20% reduction in weight). Spirometry, plethysmography and impulse oscillometry (IOS) were performed. IOS parameters included resistance at 5Hz (R5), resistance at 20Hz (R20) frequency dependence of resistance (R5-20) and reactance at 5Hz (X5).IOS was also performed at an elevated lung volume (~1 liter) targeted to restore FRC to predicted values. All IOS measurements were repeated post bronchodilator. Results: Baseline weight and BMI were 256+/-43 kg and 46+/-7 kg/m², respectively. All subjects lost >20% of body weight, but obesity persisted in all subjects (weight 182 kg, BMI 33 kg/ m2). FEV /FVC was normal at baseline and remained unchanged post weight loss 1 (81+/-3% vs 83+/-4%) indicating normal large airway function. FRC and ERV improved post weight loss but values remained abnormal (FRC from 60+/-12 to 77+/-21% predicted, ERV from 46+/-16 to 75+/-38% predicted, p<0.05). Although IOS parameters improved following weight loss, data remained above the upper limit of normal (R5 from 6.8+/-1.8 to 5.1+/-1.4 cmH₂O/l/s, R₂₀ from 4.7+/-1.1 to 3.9+/-0.9 cmH₂O/l/s, R_5-20 from 2.1+/-1.1 to 1.2+/-0.9 cmH_2O /l/s, X₅ from -3.2+/-1.7 to -1.8+/-0.9 cmH₂O /l/s, p<0.05). Since FRC remained abnormal following weight loss, IOS was repeated following voluntary lung inflation (FRC 142+/-30%). While R20 corrected to normal at the elevated FRC (R₂₀ 3.1+/-1.0 cmH₂ O/l/s), R5, R5-20 and X5 remained abnormal indicating residual distal airway dysfunction (R₅ 4.2+/-1.4 cmH₂O/l/s, , R_5-20 1.1+/-0.7 cmH₂O /l/s, X₅ -2.0+/-0.8 cmH₂O /l/s); these residual oscillometric abnormalities were present in 11/22 subjects. Residual airway dysfunction was demonstrated by low specific conductance (assessed at 5HZ) despite restoration of FRC to supranormal values. Conclusions: Distal airway dysfunction persisted following weight loss and was not attributable to persistent mass loading in a subgroup of patients without clinical evidence of airway disease. These abnormalities may represent either functional abnormalities due to persistent obesity and/or intrinsic airway disease

BACKGROUND: Weight regain after Roux-en-Y gastric bypass (RYGB) is increasingly reported in the bariatric literature. Laparoscopic sleeve reduction of the gastrojejunal complex is a surgical option to revise a dilated gastric pouch. We report our short-term results. METHODS: Sleeve reduction entails serial firing of a linear stapler along the jejunal alimentary limb, across the gastric pouch and towards the left crus, with a bougie in place, thus, creating a new 20-25-cm reduced gastrojejunal complex. Data analyzed included age, body mass index (BMI), excess weight loss (EWL), comorbidity resolution, and any other simultaneous operative procedures. RESULTS: Fourteen patients were identified, all done laparoscopically. Nine underwent gastrojejunal sleeve reduction alone and five underwent additional lengthening of the Roux limb. There were no mortalities. Mean age at revision was 43 years (31-59). Mean BMI and EWL prior to revision were 35.5 +/- 4.0 kg/m(2) and 48.9 +/- 15.8%, respectively. Nine of 14 patients (64%) had obesity-related comorbidities prior to the revision. Average BMI decrease was 2.7 kg/m(2). Post-revision mean BMI and %EWL were 32.9 +/- 4.7 kg/m(2) and 12.0 +/- 13.9%, respectively, with mean follow-up of 12 months. Three of nine patients (33%) experienced improvement and/or resolution of comorbidities. We did not find a significant difference between pre-and post-revision mean BMI and %EWL (p = 0.13) even after separately evaluating those patients who underwent Roux limb lengthening (p = 0.16). CONCLUSION: For RYGB patients who regained weight, laparoscopic gastrojejunal sleeve reduction does not seem to offer a major therapeutic benefit. Additional malabsorptive Roux lengthening also does not provide a significant benefit. Other options should be considered, such as placing a band on the gastric pouch or conversion to duodenal switch.

Introduction: Participation in a medically-supervised weight management (MSWM) program before bariatric surgery is mandated by several insurance payers. However, this requirement is not evidencebased and serves as a barrier to medically necessary treatment. We conducted a pilot randomized trial funded by SAGES to determine the effect of an insurance-mandated MSWM prior to surgery. Our hypotheses are: (1) There is no difference in BMI between patients who have participated in a medically-supervised weight management program and those who don't. (2) MSWM does not change self-reported adherence, physical activity, eating behavior, and health beliefs. Methods & Procedures: 55 patients were scheduled for laparoscopic adjustable gastric banding (LAGB) and consented to enroll in an ongoing prospective pilot study in a large public hospital. of these, 12 patients cancelled surgery and 10 patients have been enrolled for less than 6 months. Thus, 33 patients were included in the analysis, with 17 randomized to MSWM (defined as monthly visits over 6 months directed by a physician or nutritionist) and 16 randomized to usual care. Measures of weight, height, adherence, activity level, health beliefs, and eating behavior were obtained at enrollment (ie, baseline) and 6 months after enrollment (ie, MSWM program completion and 2 weeks prior to surgery). For categorical and continuous data, Fisher's Exact Test and t-test were used to compare groups at the 2 time points before surgery. Results: Mean age was 45.5 (SD = 12.5) years. Majority were female (97%) and non-Caucasian (85%) with an income of less than $20,000 (64%). No significant differences were found between MSWM and usual care for age, gender, ethnicity, education, and income, indicating that the 2 groups were evenly matched for demographic variables. Mean BMI for the MSWM group was 46.3 kg/m2 at baseline and 46.0 kg/m2 at 6 months. Mean BMI for usual care was 44.7 kg/m2 and 44.6 kg/m2 (see Table 1). After 6 months of eitherMSWMor wait and at pre-surgery, no significant differences in BMIor patient behaviors were found between the 2 groups. Preliminary analysis of available 3-month post-operative data (n = 13) showed similar mean BMI when comparing MSWM (41.3 kg/m2) and usual care (41.2 kg/m 2). Conclusion: Our preliminary results indicate that MSWM does not affect BMI changes or patient behaviors prior to or after LAGB surgery

Background: Hiatal hernia repair at the time of bariatric surgery improves patient outcome, decreases GERD symptoms and reduces the need for reoperation. The aim of this report is twofold: first, to compare the sensitivity of esophagram with surgical findings at the time of bariatric surgery, and second, to compare the sensitivities of upright versus right anterior oblique (RAO) recumbent esophagram in predicting the presence of hiatal hernia intraoperatively. Methods: Between 2008 and 2010, 389 patients undergoing bariatric surgery were prospectively evaluated for hiatal hernia by barium esophagram. 70 (18%) were performed only in the upright position and 319 (82%) only in the RAO recumbent position. Esophagram technique was changed from upright to recumbent because we hypothesized that we would be able to better detect hiatal hernia utilizing RAO recumbent technique. Hiatal hernia was assessed intraoperatively by laxity/dimpling of the phrenoesophageal ligament and, when present, was repaired posteriorly with permanent sutures. Results: Compared with the surgical findings, the sensitivity and specificity for upright esophagram was 50% and 97%, respectively. For recumbent esophagram, sensitivity was 70% and specificity was 77%. Recumbent esophagram had a lower percentage of false negatives than upright esophagram (11% vs. 21%). Conclusions: Use of a recumbent technique for preoperative esophagram has a higher sensitivity for diagnosis of hiatal hernia than upright esophagram. Routine use of recumbent esophagram results in increased preoperative detection of hiatal hernia and facilitates planning of crural closure

Background: The purpose of this study was to objectively assess the resolution of obesity-related comorbidities (ORC) after bariatric surgery and to compare the status and resolution of comorbidities following laparoscopic adjustable gastric banding (LAGB), roux-en-Y gastric bypass (RYGB), and sleeve gastrectomy (LSG). Methods: Data was collected from an IRB-approved electronic registry, including patient demographics, weight, BMI, and ORC status. Using the registry, ten ORCs were scored, pre-op and post-op, from 0-5 according to severity using the Assessment of Obesity-Related Comorbidities (AORC) Scale, the basis for the Bariatric Outcomes Longitudinal Database. The ten ORCs were: osteoarthritis (OA), diabetes, hypertension (HTN), obstructive sleep apnea, hyperlipidemia (HLD), gastroesophageal reflux disease, depression, urinary stress incontinence, hernia, and lower extremity edema (LEE). Resolution of disease was defined as having AORC>0 pre-surgery and AORC=0 post-surgery. Change in ORC status was calculated with the following equation: (pre-op AORC score) - (post-op AORC score). Paired t-tests were utilized to determine whether comorbidity change was significant following bariatric surgery. Fisher's exact tests were used to determine if there was a significant difference in ORC resolution between procedures. Results: 264 patients with ORC underwent bariatric surgery between January 2008 and March 2010 at an urban safety-net hospital. Average pre-op age was 42.5, and average pre-op BMI was 44.2. At mean patient follow-up of 17.2 months, the %EWL of RYGB, LSG and LAGB was 43.6%, 37.4% EWL, and 23.3% EWL, respectively (p < .0001). Resolution of 4 comorbidities (OA, HTN, HLD, and LEE) was found to be significantly different between surgery types (p<0.05): The percentage of patients with OA resolution was 71% for RYGB, 63% for LSG, and 51% for LAGB. HTN resolution was 57% for RYGB, 23% for LAGB, and 29% for LSG. HLD resolution was 71% for LSG, 67% for RYGB, and 34% for LAGB. LEE resolution was 100% for LSG (n=6), 94% for RYGB, and 68% for LAGB. RYGB produced an overall mean ORC resolution of 66%, vs 60% and 44% produced by LSG and LAGB, respectively. All bariatric surgery procedures had statistically significant AORC score change for all 10 documented comorbidities (p < .0001). The overall mean change in AORC score for all comorbidities, from pre-op to post-op, was 1.7 for RYGB patients, 1.4 for LSG patients, and 1.2 for LAGB patients. There was no significant association between initial BMI and change in AORC score. The pre-op AORC scores were not significantly different between surgery types. Conclusions: RYGB had the greatest ORC resolution for patients with OA and HTN, as well as the greatest mean ORC status improvement overall. LSG produced the greatest significant ORC resolution for patients with HLD and LEE. RYGB, LSG, and LAGB had statistically significant ORC status improvement for all 10 documented comorbidities

Introduction: The purpose of this study is to correlate gastric emptying after laparoscopic sleeve gastrectomy (LSG) and calculated sleeve size (based on radiographic characteristics and pathologic resection) with post-op weight loss. Methods: Data was collected from an IRB-approved electronic registry, including patient demographics, weight, and body mass index (BMI). All sleeves were done with 40Fr Bougie, starting 5-7 cm proximal to pylorus. Post-op esophagrams were evaluated by 2 attending radiologists who specialized in body-imaging for 1) post-op radiographic sleeve diameter near top of sleeve, mid-sleeve and in antrum and 2) antrum-to-duodenum transit time. Sleeve volume was calculated utilizing the formula for cylinder volume r2h, where r=radius of mid-sleeve and h=height of the sleeve from gastroesophageal junction to distal antrum. Resected gastric volume was calculated utilizing radius and length of resected specimen (based on path report.) Excess weight loss (%EWL) was calculated based on ideal body weight. Pearson's correlation coefficient was used to evaluate the association between: transit time and weight loss, sleeve volume and weight loss, and transit time and sleeve diameter. Results: 62 patients underwent LSG (21% concurrent hiatal hernia repair) between Jan 2009 and Jan 2011 at an urban safety-net hospital. The population was 84% female, average pre-op age and BMI was 42 years and 47.0 kg/m², respectively. The transit time (available in 60 patients) ranged from 0-88 seconds (mean=21.3, SD= 19.8). 99% of the patients demonstrated gastric emptying under 60 seconds. Mean radiographic diameter of mid-sleeve was 4.0 cm and mean radiographic height was 26.4 cm. Based on these dimensions, mean calculated sleeve volume (based on cylindrical volume) was 115 cm3 (+/-81.0). Mean resected gastric volume (based on pathology specimen) was 658 cm3 (+/-945). Mean %EWL at 3, 6, and 12 months was 23.8% (+/-9.8), 37.9% (+/-11.8) and 52.2% (+/-10.8). There was no correlation found between transit time and %EWL at 3, 6 or 12 months. When dichotomizing the data between those with transit time <30 seconds vs. >30 seconds, there was still no significant correlation. There was also no correlation found between calculated sleeve volume or resected gastric volume and %EWL at 3, 6 or 12 months. However, shorter transit times were correlated with smaller mid-sleeve diameter (r=0.295, p-value=0.022) and smaller antrum diameter (r=0.255, p-value=0.049) but were not significantly correlated with upper sleeve diameter (r=0.120, p-value=0.360). Conclusion: We found no correlation between transit time after sleeve gastrectomy and weight loss, between sleeve volume and weight loss, and between resected gastric volume and weight loss. However, shorter transit time was correlated with smaller mid-sleeve and antrum diameter; the clinical significance of this remains to be determined

Laparoscopic sleeve gastrectomy (LSG) techniques vary significantly, including bougie size (32-60Fr), distance from pylorus (2-8 cm), antral preservation, proximity to GE junction, staple height/oversewing, and concurrent hiatal hernia repair. The literature is controversial regarding bougie size and weight loss after LSG, however there is certainly a trend towards the use of more narrow bougies (32Fr) with increasing surgeon experience. We have previously published a study comparing weight loss after LSG between 40Fr and 60Fr bougie and seeing no significant weight loss difference at 1 year; however the 60Fr bougie group was mainly firststage duodenal switch patients who were primarily superobese (mean BMI 63.1 at baseline). If one calculates the volume of the sleeve using the formula for cylinder (r2h) where h=25 cm (length of sleeve), the calculated volume for 32Fr bougie is 20cc, for 40Fr bougie is 32cc and 60Fr bougie is 71cc. At our institution, we have found no correlation between calculated sleeve volume (radiographically) and weight loss up to 1 year postoperatively. We did find a correlation between smaller diameter at midsleeve and more rapid emptying of liquid contrast postoperatively, however the accelerated emptying did not correlate with improved weight loss. What is the downside of using a more narrow bougie? There is conflicting literature regarding the complication rates and bougie size for LSG. There is concern that the tighter the sleeve, the higher the risk of a leak, especially near the GE junction. There are also reports of higher reflux rates with tighter bougies. Other large studies (e.g. Spanish Registry Data) have found no correlation between bougie size and complication rate. It is difficult to compare outcomes based on bougie size due to the variability of other intraoperative factors, including how closely the stapler is applied along the bougie, the amount of posterior fundus mobilized, the distance from the pylorus where the LSG begins, the amount of stretch applied laterally on the fundus, the presence of gastritis that affects thickness and distensibility of the stomach, and the use of buttressing material. These factors combined with the accelerated gastric emptying seen after LSG (suggesting that LSG may not be solely a restrictive procedure) make evidence-based comparisons difficult to perform. Larger studies with standardized techniques are needed to determine optimal bougie size

BACKGROUND: Evidence of the positive effects of gastric banding on patients with diabetes has continued to increase. The long-term follow-up of such patients, however, has been limited. The purpose of the present study was to provide the long-term outcomes of patients with diabetes undergoing laparoscopic adjustable gastric banding at our institution. METHODS: From January 2002 through June 2004, 102 patients with type 2 diabetes mellitus underwent laparoscopic adjustable gastric banding. The study parameters included preoperative age, gender, race, body mass index, duration of diabetes before surgery, fasting glucose level, hemoglobin A1c (HbA1c), and medications used. Preoperative data from all patients were collected prospectively and entered into an institutional review board-approved database. Beginning in 2008, efforts were made to collect the 5-year follow-up data. RESULTS: Of the 102 patients, 7 were excluded because they had not reached the 5-year follow-up point (2 patients had had the band removed early and 5 patients had died; 2 of cancer and 3 of unknown causes), leaving 95 patients for the present study. The mean preoperative age was 49.3 years (range 21.3-68.4). The mean preoperative body mass index was 46.3 kg/m(2) (range 35.1-71.9) and had decreased to 35.0 kg/m(2) (range 21.1-53.7) by 5 years of follow-up, yielding a mean percentage of excess weight loss of 48.3%. The mean duration of the diabetes diagnosis before surgery was 6.5 years. Of 94 patients, 83 (88.3%) were taking medications preoperatively, with 14.9% overall taking insulin. At 5 years postoperatively, 33 (46.5%) of 71 patients were taking medications, with 8.5% taking insulin. The mean fasting preoperative glucose level was 146.0 mg/dL. The glucose level had decreased to 118.5 mg/dL at 5 years postoperatively (P = .004). The mean HbA1c level was 7.53 preoperatively in 72 patients and was 6.58 at 5 years postoperatively in 64 patients (P <.001). Overall, diabetes had resolved (no medication requirement, with HbA1c <6 and/or glucose <100 mg/dL) in 23 (39.7%) of 58 patients and had improved (use of fewer medications and/or fasting glucose levels of 100-125 mg/dL) in 41 (71.9%) of 57 patients. The combined improvement/remission rate was 80% (64 of 80 patients). CONCLUSION: Our data have demonstrated that laparoscopic adjustable gastric banding results in a substantial sustained positive effect on diabetes in morbidly obese patients, with a significant reduction in HbA1c and an 80% overall rate of improvement/remission

OBJECTIVES: The goal of this study was to define the frequency of stent gaps by 64-detector computed tomographic angiography (CTA) and their relation to in-stent restenosis (ISR), stent fracture (SF), and overlap failure (OF). BACKGROUND: SF defined by catheter angiography or intravascular ultrasound has been implicated in ISR. METHODS: A total of 292 consecutive patients, with 613 stents, who underwent CTA were evaluated for stent gaps associated with decreased Hounsfield units. Correlations with catheter coronary angiography (CCA) were available in 143 patients with 384 stents. RESULTS: Stent gaps were noted in 16.9% by CTA and 1.0% by CCA. ISR by CCA was noted in 46.1% of the stent gaps (p < 0.001) as determined by CCA, and stent gaps by CTA accounted for 27.8% of the total ISR (p < 0.001). In univariate analysis, stent diameter > or =3 mm was the only CCA characteristic significantly associated with stent gaps (p = 0.002), but was not a significant predictor by multivariate analysis. Bifurcation stents, underlying calcification, stent type, location, post-dilation, and overlapping stents were not observed to be predisposing factors. Excessive tortuosity and lack of conformability were not associated with stent gaps; however, their frequency was insufficient to permit meaningful analysis. CONCLUSIONS: Stent gap by CTA: 1) is associated with 28% of ISR, and ISR is found in 46% of stent gaps; 2) is associated with > or =3-mm stents by univariate (p = 0.002) but not by multivariate analysis; 3) is infrequently noted on catheter angiography; and 4) most likely represents SF in the setting of a single stent, and may represent SF or OF in overlapping stents.