Society: SSAT

Introduction
Laparoscopic fundoplication in patients with a history of lung transplant has an average length of stay (LOS) of 3 days with a 30-day readmission rate of 25%, which is significantly worse compared to the non-lung transplant population. We investigated if lung transplantation was still a risk factor for poor short-term outcomes in a practice that exclusively uses the robot for hiatal hernia/GERD surgeries.

Methods
We performed a single-center retrospective analyses of the Society of Thoracic Surgery (STS) database for patients who underwent elective hiatal hernia/GERD procedures from 1/5/2018 to 2/6/2021. We identified patient and surgical characteristics, morbidity, LOS, 30-day readmission and mortality. Analysis was conducted using Chi-square and Wilcoxon rank-sum tests, univariable linear regression, and bivariate analysis. A p-value below 0.05 was considered significant.

Results
Among 386 patients who underwent barrier creation, 43 patients had previously undergone a lung transplant, either bilateral (n=28) or single (n=14). The hiatal hernia/GERD procedure was performed for the lung transplant population on an average of 2.5 years post-transplant (SD +/- 2.5). All lung transplant patients underwent robotic-assisted laparoscopic hiatal hernia/GERD surgery. There was no significant difference in post-operative complications (9.3% vs. 5.2%, p=0.29, Figure 1A), median hospital LOS (1 vs. 1 day, p=0.27, Figure 1B), 30-day readmission (7.0% vs. 5.0%, p=0.48, Figure 1C), or 30-day all-cause mortality (0% vs. 0.6%, p=1.0) between lung transplant and non-lung transplant patients. Univariate analysis showed older age (p=0.03), opioid dependence (p=0.02), neurocognitive dysfunction (p<0.001), and dependent functional status (p=0.02) were associated with all post-operative complications. However, lung transplantation was not associated with increased risk of postoperative complications (p=0.28).

Discussion
In this cohort of patients who underwent robot-assisted laparoscopic hiatal hernia / GERD surgery, we found no difference in short-term outcomes between lung transplant and non-lung transplant patients. Robot-assisted surgery may provide improved outcomes for high-risk surgical patients.

Introduction: There is an overall tendency toward more restriction of the final sleeve by using a smaller bougie and leaving a shorter antrum. Few observational cohort and randomized controlled trial studies have compared between different resection distances from the pylorus during sleeve gastrectomy. Our aim was to aggregate the evidence available on sleeve gastrectomy with antral preservation (AP) versus antral resection (AR) by a meta-analysis.

Methods: Literature search was done according to the PRISMA guidelines. Observational cohort studies only were included in the analysis (9 studies). Meta-analysis was done using the RevMen 5.4.1 software. Statistical method used was Mantel-Haenszel. Analysis model used was random effects regardless of the heterogeneity (I²).

Results: There was no significant difference in the excess weight loss percentage (EWL%) at 1 and 3 months, mean differences (MDs) were 0.45 (CI -1.44, 2.35) and 2.72 (CI -2.04, 7.48), respectively. However, significant difference was observed in favor of the AR group in EWL% at 6, 12 and 24 months. MDs were 4.61 (CI 2.01, 7.20), 6.02 (CI 2.42, 9.61), and 8.32 (CI 6.45, 10.20), respectively. There was no significant difference in the total body weight loss % (TBWL%) at 1 month, MD was 0.66 (CI -0.06, 1.37). However, significant differences were observed in favor of the AR group regarding the TBWL% at 6, 12 and 24 months. MDs were 2.64 (CI 1.46, 3.82), 4.50 (CI 0.86, 8.13), and 4.10 (CI 3.29, 4.91), respectively. Significant difference was observed in favor of AP in postoperative leaks (OR 1.19, CI 1.00, 1.42). No significant difference was observed in the total number of postoperative complications (OR 1.12, CI 0.71, 1.76), postoperative bleeding (OR 0.88, CI 0.73, 1.06), length of stay (LOS) in days (MD 1.09, CI -2.22, 4.39), LOS in hours (MD 0.41, CI -3.39, 4.20), operative time (MD 0.72, CI -2.68, 4.12), or wound infections (OR 1.01, CI 0.41, 2.47). No significant difference was observed in complete resolution of type 2 diabetes (OR 1.35, CI 0.58, 3.15), hypertension (OR 1.51, CI 0.54, 4.28), and dyslipidemia (OR 1.17, 0.52, 2.64).

Conclusion: Sleeve gastrectomy with AP had significantly less EWL% and TBWL%, but significantly less incidence of postoperative leaks. Although insignificant, sleeve gastrectomy with AP had higher incidence of postoperative bleeding, and lesser overall resolution of all comorbidities (combined), with p values near the significance cutoff, larger studies may be able to prove significant differences. Larger studies with emphasis on incidence of postoperative denovo GERD are warranted to see if there are significant benefits for sleeve gastrectomy with AP that would justify the lesser weigh loss outcomes. Otherwise, sleeve gastrectomy without AP might be a better option.

Introduction
Pancreaticoduodenectomy (PD) is performed for several indications, including pancreatic and biliary malignancies. A common post-operative complication is delayed gastric emptying (DGE), which may occur acutely and/or chronically. Procedural variations have sought to reduce the incidence of DGE and its associated symptoms of nausea, vomiting and fullness, however the underlying pathophysiology is still poorly understood. Emerging evidence suggests that gastric myoelectrical abnormalities may contribute to DGE. A non-invasive medical device for body surface gastric electrical mapping was recently developed to evaluate gastric electrical activity and function. This study aimed to assess the feasibility of the novel device on the stomach following PD, to identify any changes in gastric activity and their correlation with symptoms.

Methods
PD patients from Auckland, New Zealand between 2017-2022 were recruited. Patients with known mechanical obstructions or recurrent malignancies were excluded. The Gastric Alimetry System® (Auckland, New Zealand) was employed, comprising a stretchable array (8x8 electrodes; 196cm2) and cloud-based analytics platform. Following an overnight fast, 30 minutes of baseline recording was performed, followed by a meal challenge and 4 hours of post-prandial recordings. Symptoms were logged on a validated iPad App. Spectral analysis of Gastric Alimetry data was performed, with quantitative analysis including Principle Gastric Frequency, BMI-adjusted amplitude and Gastric Alimetry Rhythm Index (GA-RI, a measure of rhythm stability), compared to reference intervals from 110 healthy volunteers. Adverse events were recorded.

Results
16 patients were recruited; all had a pylorus-resecting PD with 15/16 having a gastroejejunostomy and 1/16 receiving a Roux-en-Y reconstruction. Gastric Alimetry spectral abnormalities were more common in patients with moderate-severe symptom burdens (3/5 patients) vs mild-minimal symptom burdens (1/11); p=0.029. Abnormalities in symptomatic patients encompassed low GA-RI in 2 patients (<0.25); and low amplitude in 1 patient (<22μV) indicating gastric neuromuscular dysfunction. Gastric Alimetry symptom phenotypes in symptomatic patients were variable; sensorimotor (3), post-gastric (2) and continuous (2); (2 having mixed profiles). There were no adverse events.

Conclusion
Gastric Alimetry is a safe and feasible technique to non-invasively assess gastric function following PD. A third of patients had moderate to severe gastric symptoms chronically after PD, and these showed a higher rate of gastric neuromuscular dysfunction. A range of symptom phenotypes were noted, indicating gastric sensory, post-gastric (i.e. dumping) and continuous (likely neuropathic) contributions. These data indicate a role for Gastric Alimetry testing in evaluating the causes of chronic gastric symptoms after PD.

<b>Figure 1</b> A. Array placement on patient’s abdomen. B. Validated iPad App for simultaneous symptom logging. C. Example from a patient with no symptoms and normal gastric slow wave characteristics. D. Example from a symptomatic patient with abnormal gastric rhythm stability.

Figure 1 A. Array placement on patient’s abdomen. B. Validated iPad App for simultaneous symptom logging. C. Example from a patient with no symptoms and normal gastric slow wave characteristics. D. Example from a symptomatic patient with abnormal gastric rhythm stability.

Background: The surgical approach for Esophagogastric Junction Cancers (EJC), Siewert II, has been controversial in terms of margin control, reconstruction, and lymphadenectomy extension. Furthermore, both AJCC and UICC also recognize the biological heterogeneity of EJC coexisting in the same TNM. In the case of Siewert II tumors, predicting the need for either total esophagectomy and proximal gastrectomy (TEPG) or total gastrectomy with distal esophagectomy (TGDE) can be difficult, with each direction usually excluding the other. The pre-operative image workout may not accurately determine the need for each procedure. Complication rates for DETG have historically been higher, affecting systemic treatment and long-term outcomes. In this series, we describe the standardization and outcomes of an intra-operative decision-making strategy that allows both surgical approaches until frozen section guidance, rather than just following preoperative image planning, and offers the oncologic procedure with lower complication rates (TGDE) as much as possible.
Aim: The goal of this study is to describe a surgical strategy for approaching Siewert II EJC, with the intraoperative decision to perform total gastrectomy with lymphadenectomy D2 or esophagectomy with mediastinal and retroperitoneal lymphadenectomy based on intraoperative frozen section.
Methods: All patients underwent surgery, beginning with greater curvature detachment while preserving the right gastroepiploic, right, and left arteries; dissection of the esophageal hiatus for both node harvesting and transection of the distal esophagus and its frozen section. If the margin was free, DETG was preferred; if the margin was positive, TEPG and gastric tube reconstruction were performed, as shown in the figure.
Results: thirty-eight patients with EJC Siewert II with adenocarcinoma who underwent this standardization, 26 (68%) were submitted to TGDE and 12 (32%) for TEPG, and the clinic-pathological data and surgical outcomes are shown in the table. Briefly, the TEPG showed a trend toward higher complication rates, higher positive margins, and shorter overall survival, but it was not statistically significant.
Conclusion: Surgery for EJC requires a careful evaluation of their disease extension, to perform an adequate procedure for each case. Although this study found no significant differences in morbidity between the two procedures, type II errors must be considered a possible cause. This surgical strategy favors a stepwise approach to the cardia tumor, with stomach sparing and vascularization for an occasional gastric tube until free margins in the frozen section can be safely confirmed. Thus, in some cases with Siewert II tumors, unnecessary esophagectomies can be avoided without jeopardizing either surgical or oncologic outcomes, and opting for a procedure with less morbidity, according to the literature.

Figure: Intraoperative image sequence of the stepwise approach for Siewert II tumors; A - exposure of distal esophagus with tumor (dashed ellipsis); B – the esophagus margin preparation for frozen section; C – stomach with preserved right gastroepiploic (red arrows), right and left gastric arterial arcades whereas the frozen section is made; D – cervical esophagus-gastric anastomosis in case of positive margin in the distal esophageal frozen section, and using a gastric tube nourished through right gastroepiploic and right gastric arterial arcades (E).

Figure: Intraoperative image sequence of the stepwise approach for Siewert II tumors; A - exposure of distal esophagus with tumor (dashed ellipsis); B – the esophagus margin preparation for frozen section; C – stomach with preserved right gastroepiploic (red arrows), right and left gastric arterial arcades whereas the frozen section is made; D – cervical esophagus-gastric anastomosis in case of positive margin in the distal esophageal frozen section, and using a gastric tube nourished through right gastroepiploic and right gastric arterial arcades (E).

<b>Table: Clinicopathological distribution and surgical outcomes according to the type of procedure necessary to treat esophagogastric junction cancers (Siewert II)</b>

Table: Clinicopathological distribution and surgical outcomes according to the type of procedure necessary to treat esophagogastric junction cancers (Siewert II)

Background: The surgical treatment of gastroparesis is gastric neurostimulaton (GN), laparoscopic/open pyloromyotomy (PY), and/or per-oral pyloromyotomy (POP). How these treatments affect subsequent health care utilization is poorly understood.
Methods: Medical records of patients undergoing surgical treatment for gastroparesis from 2012-2019 were reviewed. Patients were divided into three groups: GN alone, GN + pyloromyotomy, and GN + delayed pyloromyotomy. Primary outcome: number of readmissions. Secondary outcomes: number of emergency room visits, surgical site occurrences, number of GN setting adjustments, and patient-reported symptom relief.
Results: 78 patients were included: 49 GN alone, 12 GN + pyloromyotomy, and 17 GN + delayed pyloromyotomy. Median readmissions (range): GN alone 0 (0-26), GN + pyloroplasty 0 (0-4), GN + delayed pyloromyotomy 1 (0-9) (p-value= 0.29). However, patients significantly differed in the number of GP setting adjustments (p-value= 0.001), number of repositionings or replacements (p-value= 0.008) and total follow-up months (p-value= 0.001).
Conclusions: Readmissions did not differ among groups. However, the number of office visits needed for GN setting adjustments was lowest in the GP alone group, the number of repositionings/replacements was lowest in the GN + pyloromyotomy group and the number of follow-up months was highest among the GN + delayed pyloromyotomy group.

Introduction: Although sleeve gastrectomy (SG) is relatively a safe surgical option, various possible complications have been reported in the literature. The most feared complications are staple line leak, hemorrhage, and stenosis. Proposed strategies to reduce some of these complications range from starting the resection at least 3 cm distance from the pylorus, using larger bougie sizes, choosing correct staple height, applying staple line reinforcements, and doing omentopexy (OP). However, there is no ideal method or technique to avoid these complications. The aim of this study was to evaluate outcomes of doing sleeve gastrectomy alone (SG group) versus sleeve gastrectomy with omentopexy (SGOP group).

Methods: Literature search was done according to the PRISMA guidelines. 11 observational cohort studies were included in the analysis. Meta-analysis was done using the RevMen 5.4.1 software. Statistical method used was Mantel-Haenszel. Analysis model used was random effects regardless of the heterogeneity (I²).

Results: There was significantly higher incidence of leaks (OR 4.24, CI 1.78, 10.10) and postoperative bleeding (OR 2.85, CI 1.83, 4.44) in the SG group compared to SGOP group. There was no significant difference between SG to SGOP groups in the incidence of readmission (OR 1.62, CI 0.66, 3.97), gastric twist (OR 0.71, CI 0.03, 14.38), denovo GERD (OR 3.14, CI 0.27, 36.92) and esophagitis (OR 3.47, CI 0.45, 26.48). Length of stay was significantly longer in the SG group compared to the SGOP group (mean difference 0.73, CI 0.30, 1.16). Operative time was significantly shorter in the SG group compared to the SGOP group (mean difference -19.24, CI -25.74, -12.75).

Conclusion: SGOP group had lower incidence of postoperative leaks and postoperative bleeding, and shorter length of stay compared with the SG group. However, SGOP group had a mean increase in operative time of about 20 minutes. Studies comparing omentopexy versus other staple line reinforcement techniques are warranted, as other techniques might be equally effective with less operative time.

Background: Roux-en-Y gastric bypass (RYGB) surgery has proven highly effective in improving type 2 diabetes mellitus (T2DM). One of the possible mechanisms is through the exclusion of the foregut from the food digestion and absorption. A duodenal-jejunal bypass liner device or EndoBarrier®, mimicking such exclusion in RYGB, offers a less invasive option compared to the surgery. EndoBarrier has been shown to reduce body weight and improve glycemic control; however, the metabolic and the gut microbial changes have not been explored in a randomized controlled clinical trial. We aimed to investigate EndoBarrier-induced perturbations in both fecal bacterial and global metabolic profiles of urine, serum and feces.

Methods: Participants with obesity and uncontrolled T2DM on oral glucose-lowering medication were recruited into this multicenter, randomized, controlled, and open-label trial (ClinicalTrials.gov Identifier NCT02459561) carried out at Imperial College London and University Hospital Southampton NHS Foundation Trusts. Biofluid samples were collected from a control group receiving conventional T2DM therapy with lifestyle modification, and EndoBarrier group during a 12-month treatment period and a 1-year follow-up after the removal of the device. All samples were analyzed using proton nuclear magnetic resonance (¹H NMR) spectroscopy-based metabolic phenotyping approach and 16S rRNA gene sequencing was used to obtain bacterial profiles.

Results and Discussion: Both EndoBarrier and control groups showed significant reductions in body mass index, HOMA-IR scores, fasting blood glucose and HbA1c levels during the treatment period (within 12 months into the treatment) compared to the baseline. Consistent with previously reported findings from RYGB surgery, EndoBarrier resulted in increased relative abundance of fecal Gammaproteobacteria (e.g., Klebsiella, Escherichia_Shigella) and Bacilli and reduced relative abundance of Clostridia. Significant differences from baseline in metabolic profiles of urine and faeces were observed 6 and 12 months into the treatment. These included higher urinary concentrations of phenylacetylglutamine, indoxyl sulfate, and 4-cresyl sulfate and fecal concentrations of tyramine and lactate, indicating enhanced host-microbial co-metabolism induced by EndoBarrier but not the conventional T2DM therapy. These changes reverted towards baseline levels 1 year after device explantation.

Conclusion: This is the first study to explore the metabolic profiles of patients receiving the EndoBarrier treatment. These metabolic changes highlighted shifts in the gut bacterial functions towards amino acid metabolism. Their association with the clinical outcomes warrants further studies.

Introduction:
Despite pharmacologic advances and recognition of deleterious side effects, opioids remain a cornerstone of pain management in bariatric surgery. Although minimization of post-operative opioids is increasingly employed, the impact of multimodal pain management protocol adherence on cumulative opioid requirements remains poorly understood.

Methods:
Consecutive adult patients who underwent primary laparoscopic Roux-en-Y Gastric Bypass (RYGB) or sleeve gastrectomy (SG) at a single institution between 1/1/2020 and 11/15/2022 were included. A multimodal analgesic pathway was used, including preoperative, intraoperative, and postoperative non-narcotic medications to minimize opioids. Medications were quantified in morphine milligram equivalents (MME). Patients were retrospectively stratified based on adherence to a target aspect of the protocol: Group 1 received preoperative acetaminophen and gabapentin, as well as intraoperative ketorolac and ketamine, while Group 2 was characterized by non-adherence to at least 1 of the 4 medications perioperatively. Values are listed as means ± standard deviation.

Results:
Of 427 patients who met inclusion criteria, mean age was 43.6 and 85.0% were female. 159 (37.2%) patients underwent RYGB and 268 (62.8%) underwent SG. All four perioperative non-narcotic medications were administered in 110 (25.8%) patients (Group 1), while 317 (74.2%) patients demonstrated non-adherence to at least 1 medication (Group 2). Intraoperative ketamine was only administered in 146 patients (34.2%). All patients in Group 1 received intraoperative ketamine (mean 45.0±9.95 mg, range 30-80 mg) and ketorolac (mean 29.2±3.43 mg, range 15-30mg). No difference was found between groups in intraoperative opioids (Group 1: 64.4±28.7 MME, Group 2: 60.2±27.8 MME, p=0.191), recovery unit opioids (Group 1: 78.6±45.8 MME, Group 2: 84.9±129.1 MME, p=0.476), or mean recovery unit pain scores (Group 1: 5.83±1.62, Group 2: 5.81±1.58, p=0.925). Postoperative inpatient use of scheduled acetaminophen (>96%), gabapentin (>91%), and methocarbamol (>38%) were similar between groups, while ketorolac use was increased in Group 1 (93.6% vs 79.2%). Postoperative inpatient pain scores did not differ among groups (Group 1: 4.03±1.42, Group 2: 4.06±1.35, p=0.827). While not statistically significant, postoperative inpatient opioid use (Group 1: 27.5±27.68 MME, Group 2: 66.6±391.43 MME, p=0.142) and cumulative aggregate opioid use (Group 1: 157.8±70.63 MME, Group 2: 188.91±357.47 MME, p=0.143) were decreased in Group 1. Protocol adherence was associated with a decreased mean length of stay (Group 1: 37.2±14.6 hours, 45.4±55.6 hours, p=0.017). The 30-day readmission rate was 3.7%.

Conclusion:
Cumulative perioperative opioid use following elective bariatric surgery can be safely minimized by adherence to a multifaceted non-narcotic analgesic protocol.

Introduction:
Marginal ulcer (MU) is a common complication following bariatric surgery. It mostly presents with abdominal pain and if left untreated, it can be complicated with perforation, bleeding or fistula. Proton pump inhibitors (PPI) are the mainstay of treatment, including primary prophylaxis. In this meta-analysis, we aimed to determine the usefulness of PPI prophylaxis for MU prevention and optimal duration of primary prophylaxis after gastric bypass surgery.
Methods:
We performed a comprehensive literature search of MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials from inception through November 2022. The primary outcome was the overall incidence of MU following PPI prophylaxis. The secondary outcomes were rate of bleeding, perforation, fistula, and rate of medical, surgical management. Standard meta-analysis methods were employed using a random-effect model using Comprehensive Meta-Analysis Software (CMA). Heterogeneity was assessed using the I2 index.
Results:
17 studies (3 prospective, 14 retrospective) involving 10,999 patients were included. Mean age was 43.4 (10.7) years, and 81 % were female. The type of surgery performed included open (0.9%) and laparoscopic (99%) Roux-en-Y gastric bypass. The type, dose, and duration of prophylactic PPI were variable ranging from 1 to 6 months, postoperatively. Pooled incidence of MU after PPI prophylaxis was 5.3% [(95% confidence interval (CI):3.6-7.8, I2=94%)]. The odds ratio of MU in PPI vs no PPI was 0.42 [0.27-0.66, I2=6%, p<0.001] (5 studies, Figure.1). Subgroup analysis, based on duration of PPI prophylaxis, demonstrated incidence of MU to be 7.4% [3.6-14, I2=95%] in <3 months, 3.8% [3.8-6.6, I2=93%] at 3 months, and 5.1% [ 2-12, I2=83%] at 6 months. The pooled rate of bleeding was 16.2% [5.8-37.9, I2=63%], perforation 9.2% [4.3-18.8, I2=42%], and fistula formation 6.6% [3.9-9.1, I2=0%]. The pooled rate of medical management was 84% [71-92, I2=77%], and surgical was 16.2% [8.7-28.3. I2=74%] (Table.1). No evidence of publication bias was found (Egger’s test: P=0.11).
Conclusion:
Based on our meta-analysis, PPI prophylaxis reduced the overall incidence of MU from about 10% to 5.3%. The incidence of MU was higher (7.4%) in the less than 3 months of PPI therapy group. Approximately similar rates of MU were observed in the 3 months and 6 months PPI therapy groups with incidence rates of 3.8 and 5.1%, respectively. Further research is warranted to define the optimal duration of PPI therapy after gastric bypass surgery.

Introduction:

Obesity and several obesity-related co-morbidities are known risk factors for the increased severity of COVID-19. Bariatric surgery (BS) is the most effective treatment for obesity. However, there is a gap in our understanding of whether treatment of obesity with BS will reduce the adverse outcomes of COVID-19. Therefore, our study aimed to evaluate the impact of BS on the severity of COVID-19 in hospitalized patients in the US.

Methods:

National Inpatient Sample (NIS) for 2020 was queried using ICD-10-CM Codes to identify a cohort of inpatient admissions with a primary discharge diagnosis of COVID-19 and a secondary diagnosis of obesity or Bariatric Surgery (BS). A weighted sample was used to get baseline characteristics and resource utilization during the inpatient admissions. Multivariate logistic regression analysis followed by predictive margins was used to obtain adjusted estimates of the utilization of therapeutics (Remdesivir, Tocilizumab, or Convalescent Plasma), mortality, intubation, and Intensive Care Unit (ICU) care.

Results:

Among 36,210 patients admitted with COVID, 280,360 (78.3%) had obesity, and 7850 (21.7%) had prior BS. The mean age was similar in both groups (58.1 years vs. 58.0 years). There was a greater proportion of patients in the BS group who were female, white, and had private insurance compared to those with obesity without BS. In adjusted analysis, the BS group experienced decreased intubation (7.4% vs. 12.4%), ICU Care (7.6% vs. 12.6%), utilization of therapeutics (34.3% vs. 38.2%), and Mortality (6.3% vs. 9.8) which were all statistically significant. Length of stay and total hospital charges were higher for the obesity group compared to BS. A lesser number of patients in the obesity group were discharged home than those with BS (62.1% vs. 67.8%, p-value <0.01).

Discussion: Our study found that patients admitted with COVID-19 with a history of prior BS had better in-hospital outcomes and decreased severity of COVID-19 compared with patients with obesity and no history of BS. These findings confirm the urgency for proper care in this underserved population and demands better access to the full spectrum of effective weight loss therapies.

Introduction
Obesity is an independent risk factor for heart failure (HF). Substantial weight loss has been shown to reverse obesity-related cardiomyopathy. In severe cases of HF, obesity can further limit a patient’s eligibility for heart transplant. The goal of this study was to report our institution’s experience with laparoscopic sleeve gastrectomy (LSG) in morbidly obese end-stage HF patients.

Methods
Between 2018-2022, 26 patients with end-stage heart failure were referred for LSG at a tertiary care center. Of those patients, 16 underwent an operation while 10 did not. A retrospective chart review was conducted, and the following outcomes evaluated: (a) average age at the time of referral (b) number of patients with a ventricular assist device (VAD) (c) preoperative weight (d) total decrease in body mass index (BMI), (e) whether the patient underwent heart transplantation, and (f) mortality. Data analysis was performed using Stata (SE 17.0). Wilcoxon Rank Sum test were used to compare continuous variables between the cohorts, and Pearson Chi Squared test used for binary variables with Bonferroni correction applied.

Results
The LSG and non-LSG cohorts had comparable ages (p=0.088), preoperative BMI (p=0.918), and number of VAD patients (p=0.191). Patients who underwent LSG lost significantly more weight than the patients who did not undergo LSG, with an average decrease in BMI of 8.9 kg/m² (SD±6.13) and 1.1 kg/m² (SD±4.10), respectively (p=0.040). Six (37.5%) LSG patients eventually underwent transplantation, compared to two (20.0%) patients from the matched cohort (p=0.884). Of the 26 total patients, there were 6 deaths: 2 (12.5%) in the LSG cohort, and 4 (40.0%) in the non-LSG cohort (p=0.525).

Conclusion
LSG appears to be safe and effective for weight loss in patients with heart failure. This operation may provide morbidly obese end-stage HF patients the life-saving opportunity to achieve transplant candidacy. Future studies with larger patient cohorts are underway.

Background
It remains unclear to what extent advanced liver disease affects the efficacy of bariatric surgery in weight loss and improving chronic disease. Therefore, we sought to elucidate the relationship between advanced liver disease and patient outcomes following bariatric surgery.

Methods
This was a retrospective cohort study from a single bariatric center of excellence. We collected data on consecutive patients who underwent primary bariatric surgery (either laparoscopic sleeve gastrectomy (LSG) or Roux-en-Y gastric bypass (RYGB) from Jan 2020 to Sep 2022 with concomitant surgical liver biopsy. Primary outcome was % excess weight loss (EWL) at 3, 6, and 12 months. Secondary outcomes include 30-day adverse events and improvement in hypertension, hyperlipidemia, and type 2 diabetes. Multivariate linear and logistical regression models were used to evaluate associations between patient characteristics and outcomes. Statistical analyses were performed using SAS 9.4.

Results
400 patients were included in this study. Summary of patient demographics are listed in Table 1. 20 (5%) patients had advanced fibrosis or cirrhosis on biopsy (i.e. F3 or F4). There was no significant difference in BMI between patients with F0-2 vs F3-4 on biopsy (mean 44.7 vs 46.3, p=0.19). Patients with advanced fibrosis had significant lower % EBMIL at 3 and 12 months post surgery (32.4 vs 38.0%, p=0.0482 at 3 months, 60.6 vs 76.8%, p=0.0066 at 12 months) [Figure 1A]. On multivariate analysis including advanced fibrosis, age, sex, surgery type (RYGB vs LSG), alcohol use disorder (AUD), active smoking history, hypertension, hyperlipidemia, obstructive sleep apnea, and type 2 diabetes, advanced fibrosis was associated with lower % EBMIL at 12 months only (13.5% lower in EWL, p=0.0486) [Figure 1B]. In addition, AUD was also associated with lower % EBMIL at 12 months, and RYGB was associated with higher % EWL at 6 and 12 months. Presence of advanced fibrosis was not associated with significant differences in improvement in hypertension, hyperlipidemia, or type 2 diabetes within 12 months of surgery. FIB-4 and APRI did not perform better than liver biopsy in predicting % EWL at 12 months (p=0.86 and 0.68 respectively when replacing advanced fibrosis on multivariate analysis).

Conclusions
Presence of advanced fibrosis at time of bariatric surgery was associated with decreased weight loss at 12 months. No association was seen with improvement of other factors of metabolic syndrome, although this analysis may be limited by the small number of patients with advanced fibrosis. Liver fibrosis index scores of FIB-4 and APRI were not helpful in predicting amount of weight loss. Larger multicenter studies may be helpful to better elucidate the relationship between advanced fibrosis and improvement in metabolic syndrome after bariatric surgery.

David Lautz

Appearances