Faculty Bibliography

OBJECTIVE: The objective of this study was to determine whether the DiaRem, a score that predicts type 2 diabetes (T2D) remission following roux-en-y gastric bariatric surgery (RYGB), also predicts remission following laparoscopic adjustable gastric banding (LAGB) and laparoscopic sleeve gastrectomy (LSG) in white and Hispanic patients. BACKGROUND: While bariatric surgery is highly effective in reversing insulin resistance, there are patients for whom surgery will not lead to remission. To date, there is no score for predicting remission following LAGB or LSG surgery. Additionally, there is little known about how to predict whether Hispanic patients will experience remission. METHODS: We conducted a retrospective cohort study of white and Hispanic patients with T2D who received bariatric surgery. There were 361 white and 130 Hispanic patients among whom 328 had RYGB surgery, 107 had LSG surgery, and 56 had LAGB surgery. We used age, diabetes treatment, and hemoglobin A1c to calculate DiaRem scores. Mann-Whitney U test was used to determine the association between DiaRem scores and remission. Area under the receiver operant curve (AUC) was used to assess the ability of the DiaRem to discriminate between patients who did and did not remit. RESULTS: The DiaRem was associated with partial remission in all surgery types for white and Hispanic patients (Mann-Whitney, p < 0.001). The DiaRem had moderate to high discriminant ability (AUC > 0.70) for all surgical and racial/ethnic groups. CONCLUSIONS: The DiaRem distinguishes between patients likely and unlikely to experience remission, informing expectations of patients making T2D treatment decisions.

OBJECTIVE: High-density lipoprotein cholesterol efflux capacity (CEC) is inversely associated with incident cardiovascular events, independent of high-density lipoprotein cholesterol. Obesity is often characterized by impaired high-density lipoprotein function. However, the effects of different bariatric surgical techniques on CEC have not been compared. This study sought to determine the effects of Roux-en-Y gastric bypass (RYGB) and sleeve gastrectomy (SG) on CEC. APPROACH AND RESULTS: We prospectively studied severely obese, nondiabetic, premenopausal Hispanic women not using lipid medications undergoing RYGB (n=31) or SG (n=36). Subjects were examined before and at 6 and 12 months after surgery. There were no differences in baseline characteristics between surgical groups. Preoperative CEC correlated most strongly with Apo A1 (apolipoprotein A1) concentration but did not correlate with body mass index, waist:hip, high-sensitivity C-reactive protein, or measures of insulin resistance. After 6 months, SG produced superior response in high-density lipoprotein cholesterol and Apo A1 quantity, as well as global and non-ABCA1 (ATP-binding cassette transporter A1)-mediated CEC (P=0.048, P=0.018, respectively) versus RYGB. In multivariable regression models, only procedure type was predictive of changes in CEC (P=0.05). At 12 months after SG, CEC was equivalent to that of normal body mass index control subjects, whereas it remained impaired after RYGB. CONCLUSIONS: SG and RYGB produce similar weight loss, but contrasting effects on CEC. These findings may be relevant in discussions about the type of procedure that is most appropriate for a particular obese patient. Further study of the mechanisms underlying these changes may lead to improved understanding of the factors governing CEC and potential therapeutic interventions to maximally reduce cardiovascular disease risk in both obese and nonobese patients.

BACKGROUND: At our medical center, female patients who have undergone bariatric surgery are advised to defer pregnancy for 2 years after surgery to avoid the following complications and their potential consequences for the fetus: inadequate gestational weight gain, inadequate postsurgical weight loss, hyperemesis gravidarum, nutritional deficiencies, gestational diabetes, and gestational hypertension. OBJECTIVES: To examine the effect of time from surgery to conception on pregnancy course and outcomes in bariatric patients. SETTING: University. METHODS: We identified 73 pregnancies in 54 women who became pregnant after undergoing bariatric surgery. Surgery to conception interval was compared between pregnancies that were carried to delivery and 8 pregnancies that resulted in spontaneous abortion. Of 41 pregnancies that were carried to delivery, 26 occurred in women who had undergone surgery less than 2 years before conception, and 15 occurred in women who had undergone surgery greater than 2 years before conception. Gestational age at delivery, number of neonatal intensive care unit admissions, gestational weight gain, hyperemesis gravidarum, nutritional deficiencies, gestational diabetes, and gestational hypertension during pregnancy were compared for the 2 groups. RESULTS: Eight patients who had spontaneous abortion had a significantly shorter time from surgery to conception. There were no significant differences between our 2 groups in rates of preterm deliveries, neonatal intensive care unit admission, gestational weight gain, hyperemesis, nutritional deficiencies, gestational diabetes, or gestational hypertension. CONCLUSIONS: Becoming pregnant within the first 2 years after bariatric surgery appears to have no effect on pregnancy course and outcomes. Women who miscarried had a significantly lower mean surgery to conception interval. These results fail to show an increased rate of pregnancy complications during the first 2 years after bariatric surgery.

BACKGROUND: Portomesenteric vein thrombosis (PMVT) has been increasingly reported after laparoscopic sleeve gastrectomy (LSG). Factor VIII (FVIII) is a plasma sialoglycoprotein that plays an essential role in hemostasis. There is increasing evidence that FVIII elevation constitutes a clinically important risk factor for venous thrombosis. OBJECTIVES: To report the prevalence of FVIII elevation as well as other clinical characteristics in a multicenter series of patients who developed PMVT after LSG. SETTING: University hospitals. METHODS: A retrospective review was conducted of all patients that developed PMVT after laparoscopic bariatric surgery from 2006 to 2016 at 6 high-volume bariatric surgery centers. RESULTS: Forty patients who developed PMVT postoperatively, all after LSG, were identified. During this timeframe, 25,569 laparoscopic bariatric surgery cases were performed, including 9749 LSG (PMVT incidence after LSG = .4%). Mean age and body mass index were 40 years (18-65) and 43.4 kg/m2 (35-59.7), respectively. Abdominal pain was the most common (98%) presenting symptom. Of patients, 92% had a hematologic abnormality identified, and of these, FVIII elevation was the most common (76%). The vast majority (90%) was successfully managed with therapeutic anticoagulation alone. A smaller number of patients required small bowel resection (n = 2) and surgical thrombectomy (n = 1). There were no mortalities. CONCLUSIONS: A high index of clinical suspicion and prompt diagnosis/treatment of PMVT usually leads to favorable outcomes. FVIII elevation was the most common (76%) hematologic abnormality identified in this patient cohort. Further studies are needed to determine the prevalence of FVIII elevation in patients seeking bariatric surgery.

BACKGROUND: There is debate regarding preoperative endoscopy (EGD) in patients undergoing bariatric surgery. Some centers perform EGD routinely in all patients; others perform EGD selectively. The objective of this study was to perform a systematic review and meta-analysis of the existing literature to estimate how frequently preoperative EGD changes management. METHODS: Our review yielded 28 studies encompassing 6616 patients. Baseline characteristics including age and body mass index (BMI) were included. Patients were grouped based on EGD findings into two groups: Group 1-findings which did not significantly change management (e.g., mild/moderate duodenitis, Grade A/B esophagitis, mild/moderate gastritis, H. pylori infection, hiatal hernia <2 cm); Group 2-findings which delayed, altered, or cancelled surgery (e.g., severe duodenitis, Grade C/D esophagitis, gastric varices, hiatal hernia >2 cm, mass/carcinoma). A general estimating equation (GEE) model accounting for the correlated data within each study was used to calculate confidence intervals around the estimate of how frequently surgery was delayed or altered. RESULTS: Mean age was 41.4 +/- 2.9 years, the majority was women, and mean preoperative BMI was 47 +/- 3.2 kg/m2. Overall 92.4 % (n = 6112) had a normal EGD or findings that did not change clinical management and 7.6 % (n = 504); 95 % CI [4.6, 12.4 %] had findings that delayed/altered surgery. The revised estimate was 20.6 %; 95%CI [14.5, 28.2 %] if all esophagitis (regardless of grade) were categorized into Group 2. The approximate incidence of Barrett's esophagus and carcinoma were 0.1 and 0.08 %, respectively. CONCLUSION: A selective approach to preoperative EGD may be considered, based on the patients' symptoms, risk factors, and type of procedure planned.

BACKGROUND: Patients with type 2 diabetes (T2D) and body mass index (BMI)<35 may benefit from metabolic surgery. The soluble form of the receptor for advanced glycation end products (sRAGE) may identify patients at greater chance for T2D remission. OBJECTIVES: To study long-term outcomes of patients with T2D and BMI 30-35 treated with metabolic surgery or medical weight management (MWM) and search for predictors of T2D remission. SETTING: University METHODS: Retrospective review of the original cohort, including patients who crossed over from MWM to surgery. Repeated-measures linear models were used to model weight loss (%WL), change in glycated hemoglobin (HbA1C) and association with baseline sRAGE. RESULTS: Fifty-seven patients with T2D and BMI 30-35 were originally randomly assigned to metabolic surgery versus MWM. Mean BMI and HbA1C was 32.6% and 7.8%, respectively. A total of 30 patients underwent surgery (19 sleeves, 8 bypasses, 3 bands). Three-year follow-up in the surgery group and MWM group was 75% and 86%, respectively. Surgery resulted in higher T2D remission (63% versus 0%; P<.001) and lower HbA1C (6.9% versus 8.4%; P<.001) for up to 3 years. There was no difference in %WL in those with versus those without T2D remission (21.7% versus 20.6%, P = .771), suggesting that additional mechanisms other than %WL play an important role for the studied outcome. Higher baseline sRAGE was associated with greater change in HbA1C and greater %WL after surgery (P< .001). CONCLUSION: Metabolic surgery was effective in promoting remission of T2D in 63% of patients with BMI 30-35; higher baseline sRAGE predicted T2D remission with surgery. Larger-scale randomly assigned trials are needed in this patient population.

Background: At our medical center, female patients who have undergone bariatric surgery are advised to defer pregnancy for two years after surgery, due to the theoretical risk that becoming pregnant during the period of rapid weight loss after surgery will adversely affect long term weight loss. Methods: We examined our database of bariatric surgery patients from a large, urban, public hospital from March 2011 to January 2014. Out of over 2000 patients, we identified 49 women who became pregnant and carried to term after undergoing bariatric surgery. 29 pregnancies occurred in women who had undergone bariatric surgery less than 2 years prior to conception, and 20 occurred in women who had undergone bariatric surgery greater than 2 years prior to conception. Weight data were recorded and BMI was calculated for each patient from pre-op to post-delivery. Percentage of excess weight loss (%EWL) based on an ideal BMI of 25 at three different time points after delivery (3-12 months, 1-2 years, > 2 years) was compared for the two groups. Results: The two groups were statistically similar with regards to baseline weight, BMI, and BMI loss from surgery to conception. The mothers with <2 years between bariatric surgery and conception had a higher percentage of RYGB and LSG surgeries (p=0.020). %EWL was significantly higher for the <2 year group at 3-12 months post-delivery (57.1% vs. 37.5%, p=0.028), 12-24 months post-delivery (61.1% vs. 37.7%, p=0.012), and >2 years post-delivery (59.9% vs. 38.7%, p=0.044). However, when stratifying by surgery type, there were no significant differences in %EWL between the two groups at any time point (Table). A multiple linear repeated measures model showed that mothers in the <2 year group had significantly greater %EWL after adjusting for age of mother, surgery type, and baseline BMI (+5.6% EWL at 3-12 months post-delivery, +11.1% at 12-24 months, and +16.5% at >2 years; p=0.031). Conclusions: In this study, long term weight loss was significantly greater in women who conceived earlier than 2 years after undergoing bariatric surgery. This result may also have been affected by surgery type, as there were higher percentages of RYGB and LSG in the early pregnancy group. However, after controlling for this difference between the two groups, there remained a significant difference in %EWL at all three time points, suggesting that patients who become pregnant within 2 years after surgery may have greater long term weight loss