Impact of multidisciplinary steatotic liver disease management on bariatric surgery referral and clinical outcomes: A retrospective cohort study

Abstract

BACKGROUND

Lifestyle modifications aimed at weight loss are key to improving metabolic dysfunction-associated steatotic liver disease (MASLD); however, achieving substantial and sustained weight loss through non-surgical approaches may be difficult for some patients. Bariatric surgery should be considered as a therapeutic option in select patients with MASLD, but the national referral rate for eligible patients is low (< 1%).

AIM

To examine referral rates and one-year outcomes among adults with MASLD in a multidisciplinary clinic integrating hepatology and obesity medicine.

METHODS

We performed a retrospective cohort study of 965 patients seen in a MASLD-specific clinic over a three-year period (2018-2022). Patients were categorized as bariatric surgery eligible or non-eligible based on standard referral criteria. We assessed bariatric surgery referral rates, weight, and liver-related outcomes, including change in nonalcoholic fatty liver disease fibrosis score. Categorical variables were compared with χ² tests, and continuous variables were analyzed with two-tailed t-tests. P value < 0.05 was considered significant.

RESULTS

Among 491 patients eligible for bariatric surgery, 127 patients (26%) were referred for surgical evaluation, with 31 patients (24%) ultimately undergoing bariatric surgery (21 sleeve gastrectomy and 10 Roux-en-Y gastric bypass). The remaining 96 patients continued with medication and/or lifestyle management. Individuals who underwent bariatric surgery achieved greater one-year total body weight loss than those who utilized medication and/or lifestyle management alone (20.6% vs 2.5%, P < 0.001). Average nonalcoholic fatty liver disease fibrosis score at one year was -0.91 in surgery patients vs -0.008 in non-surgery patients (P = 0.005).

CONCLUSION

Integration of weight management with hepatology care in patients with MASLD resulted in bariatric surgery referral rates that substantially exceed the national average, leading to improved weight and liver-related outcomes.

Key Words: Bariatric surgery; Metabolic dysfunction-associated steatotic liver disease; Metabolic dysfunction-associated steatohepatitis; Fatty liver disease; Obesity medicine; Weight loss; Multidisciplinary care; Transient elastography

Core Tip: Bariatric surgery should be considered as a therapeutic option in eligible patients with metabolic dysfunction-associated steatotic liver disease (MASLD). Integration of weight management into a tertiary care center’s MASLD-specific clinic led to substantial improvement in referral rates for bariatric surgery when compared with the national average, with improved weight loss and liver-related outcomes one-year post surgery. Multidisciplinary obesity medicine and hepatology co-management is an effective model for MASLD care and promotes access to bariatric surgery.

INTRODUCTION

Metabolic dysfunction-associated steatotic liver disease (MASLD) represents a broad spectrum of disease ranging from steatosis and steatohepatitis to progressive fibrosis resulting in end-stage liver disease[1]. The prevalence of MASLD has been rising in parallel with the worsening obesity epidemic[2,3], with a predicted 55.7% of the global adult population being affected by 2040[4-6]. Weight loss alone has been consistently shown to improve liver-related outcomes, with 7%-10% total body weight loss (TBWL) resulting in reduction in both steatohepatitis and fibrosis[7,8].

Bariatric surgery is a particularly effective weight management method for individuals with MASLD[9] and should be considered for those with class 3 obesity [body mass index (BMI) ≥ 40], or with class 2 obesity (BMI ≥ 35 to BMI < 40) and any weight-related complication. Overall, less than one percent of the eligible population is referred for bariatric surgery[10,11]. Reasons for low referral rates are multifactorial and include lack of provider knowledge about the role of bariatric surgery in MASLD, patient concerns about surgical complications and safety, and cost of care[12-16].

Our understanding of MASLD has broadened as we better appreciate the systemic impact of metabolic dysfunction on not only the liver but also on a variety of other organs including the heart, kidneys, and muscle[17]. Weight reduction is considered paramount to the management of MASLD. Rather than management solely by hepatology, multidisciplinary care models offer both hepatology and comprehensive obesity care including tailored dietary and lifestyle counseling, anti-obesity pharmacotherapy, and bariatric surgery support. However, the impact of these integrated models on the use of appropriate weight loss strategies, as well as on patient weight loss and liver-related outcomes, has not yet been examined. Therefore, in the current study, we assess rates of referral to bariatric surgery in a multidisciplinary MASLD clinic as well as weight and liver-related outcomes of patients managed surgically vs non-surgically. These findings can help inform evolving standard care practices for patients with MASLD.

MATERIALS AND METHODS

We conducted a retrospective cohort study of adults seen at a tertiary care center’s specialized MASLD clinic with a diagnosis of MASLD based on International Classification of Diseases (ICD-10) diagnostic coding (Supplementary Table 1)[18]. All patients were seen in the clinic at least once during the 3-year period from May 2018 to May 2022. Patient data was extracted from the electronic health record. The study was approved by the Yale University Institutional Review Board (approval No. 2000027433).

Exposure and outcomes of interest

The primary exposure was referral to bariatric surgery, and primary outcomes were percent TBWL and BMI reduction at the one-year follow-up time point. For patients undergoing bariatric surgery, preoperative weight and one-year post-surgical weight were used to calculate outcomes. Predicted weight loss was determined using the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program Bariatric Surgical Risk/Benefit Calculator from the American College of Surgeons (Table 1)[19]. The calculator uses data from hundreds of thousands of metabolic and surgical operations collated from 925 centers to predict surgical outcomes, including one-year TBWL, for a given patient based on several variables including age, BMI, sex, ethnicity, race, and obesity-associated comorbidities. This externally validated calculator was chosen because of its robust dataset of patients who underwent bariatric surgery and inclusion of parameters specific to bariatric surgery[20]. Actual TBWL at one-year post surgery (range 10-14 months) was compared with predicted TBWL. For non-surgical patients, initial weight at first clinic visit and weight at one year were used to calculate TBWL and BMI changes. The one-year time point for the non-surgical group was a window from 10-18 months due to the constraints and challenges of follow-up. Nonalcoholic fatty liver disease fibrosis score (NFS) was calculated at initial visit and at one year using age, BMI, aspartate transaminase, alanine transaminase, platelet count, albumin level, and the presence or absence of impaired fasting glucose/diabetes. Fibrosis-4 (FIB-4) score was calculated at initial visit and at one year using age, aspartate transaminase, alanine transaminase, and platelet count. ΔNFS and ΔFIB-4 were calculated as the difference between baseline and follow-up values at the one-year timepoint. Liver stiffness measure (LSM) as determined by vibration-controlled transient elastography (VCTE) was collected for patients before and after bariatric surgery with a minimum follow-up time of one year, and patients with missing data were excluded from the analysis.

Table 1 Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program weight-loss calculator outcomes.

MBSAQIP weight-loss calculator outcomes	Predicted	Actual	P value
1 year final BMI (kg/m2, mean ± SD)	33.1 ± 4.3	35.6 ± 7.4	0.12
1 year weight loss (kg, mean ± SE)	30.3 ± 1.7	25.4 ± 2.8	0.13
1 year TBWL (%, mean ± SE)	24.5 ± 1.1	21.1 ± 1.8	0.13

MBSAQIP: Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program; BMI: Body mass index; TBWL: Total body weight loss.

Inclusion and exclusion criteria

Patients 18 years and older with a confirmed diagnosis of MASLD by ICD-10 coding (Supplementary Table 1) with at least one documented clinic visit to the MASLD clinic were identified and screened for eligibility. Patients seen at the clinic without a diagnosis of MASLD were excluded. Patients were also excluded if they lacked sufficient longitudinal weight data, defined as lacking follow-up weight data collected ten or more months after bariatric surgery or initial clinic visit.

Covariates

We extracted descriptive, medical intervention, surgical intervention, and weight loss outcome data. Additionally, home address zip codes were referenced with 2020 national census data to determine median household income. Using standard eligibility criteria for bariatric surgery (Table 2), patients were categorized as bariatric surgery eligible (BSE) or bariatric surgery non-eligible (BSNE). Categorization of BSE vs BSNE was based on BMI calculated using height and weight data from their initial visit at the MASLD clinic and the presence of an obesity-related comorbidity by ICD-10 codes (Supplementary Table 1). Hypertension, diabetes, dyslipidemia, and obstructive sleep apnea were considered, as they are among the most frequently used criteria when assessing surgical candidacy and are key cardiometabolic comorbidities linked to obesity. This categorization was retrospectively completed for the current study. Surgical complications were assessed by manual chart review to determine the rate of anastomotic leaks, infections, respiratory failure, and 30-day mortality. Additionally, charts were manually reviewed for evidence of any major adverse cardiac events including myocardial infarction, unstable angina, acute heart failure, or dysrhythmia. Also, use of common anti-obesity medications including semaglutide, liraglutide, phentermine, bupropion/naltrexone, orlistat, and topiramate was assessed.

Table 2 Bariatric surgery eligibility criteria.

Bariatric surgery eligibility criteria
BMI ≥ 35 kg/m2 with at least one obesity-associated comorbidity
Hypertension
Type II diabetes mellitus
Obstructive sleep apnea
BMI ≥ 40 kg/m2

BMI: Body mass index.

Statistical analysis

Descriptive statistics were utilized to characterize BSE vs BSNE patients and patients who underwent surgery vs those who did not. χ² tests were used to compare categorical variables and two-sample t-tests for continuous variables. Difference in mean weight loss between groups was assessed, and a two-sample t-test was used for comparing two groups, or analysis of variance for comparing more than two groups; P < 0.05 was considered significant. In patients who underwent bariatric surgery, actual one-year weight loss was compared to predicted one-year weight loss using an unpaired t-test. A two-sample t-test was used to compare liver-related outcomes between patients who had surgery vs those who did not. The analyses were carried out using Prism 10 (GraphPad Software, La Jolla, CA, United States).

RESULTS

Patient characteristics

A total of 965 patients with a confirmed diagnosis of MASLD were seen in the specialized MASLD clinic during the study period (Figure 1). Among them, 491 patients (51%) were categorized as BSE and 474 (49%) as BSNE (Table 3). When compared with patients in the BSNE group, patients in the BSE group had a higher average BMI (41.6 vs 29.6; P ≤ 0.0001) and were more likely to be female [65% female (318/491) vs 53% (249/474); P ≤ 0.0001], younger (50 ± 13 years vs 56 ± 13 years; P ≤ 0.0001), black [10% (53/491) vs 4% (20/474); P ≤ 0.0001], and of lower socioeconomic status (SES) by median household income (P ≤ 0.0001) (Table 3).

Figure 1 Flowchart representing patient evaluation and characterization in this retrospective cohort study. ¹Standard eligibility criteria are body mass index > 40 kg/m² or > 35 kg/m² plus obesity-associated comorbidity. MASLD: Metabolic dysfunction-associated steatotic liver disease; BSE: Bariatric surgery eligible; BSNE: Bariatric surgery non-eligible; RYGB: Roux-en-Y gastric bypass; SG: Sleeve gastrectomy; MBSAQIP: Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program.

Table 3 Demographic characteristics of patients with metabolic dysfunction-associated steatotic liver disease categorized by bariatric surgery eligibility, n (%).

Patient characteristics		BSE	BSNE	P value
Number of patients	n = 965	491	474
Sex	Male	173 (35)	225 (47)	0.0001
	Female	318 (65)	249 (53)
Age in years, mean ± SD		50.2 ± 13	56.4 ± 12	< 0.0001
Race	White	372 (76)	391 (82)	< 0.0001
	Black	53 (10)	20 (4)
	Asian	3 (1)	16 (3)
	Other/declined	63 (13)	50 (11)
Ethnicity1	Hispanic	71 (14)	57 (12)	0.3
	Non-Hispanic	419 (85)	417 (88)
Insurance status	Uninsured	9 (2)	11 (2)	< 0.0001
	Medicaid	114 (23)	61 (13)
	Medicare	97 (20)	137 (29)
	Private	271 (55)	265 (56)
Median household income by zip code	< $50000	81 (16)	66 (14)	< 0.0001
	$50000-$100000	319 (65)	240 (51)
	> $100000	91 (19)	168 (35)
Trialed anti-obesity medication	Lifetime	255 (52)	234 (49)
	Since FLDP	222 (45)	209 (44)
Body mass index (kg/m2), mean ± SD	At initial visit	41.6 ± 12	29.6 ± 3.7	< 0.0001
Comorbidities	Diabetes	276 (56)	171 (36)
	Chronic liver disease2	57 (12)	91 (19)
	Liver cancer	11 (2.2)	9 (1.9)

¹One subject declined to provide ethnicity information.

²Composite variable for non-neoplastic etiologies, including viral, toxin/drug-induced, genetic, metabolic, and autoimmune.

Continuous variables are presented as mean ± SD and compared using unpaired t-test. Categorical variables are presented as n (%) and compared using the χ² test. BSE: Bariatric surgery eligible; BSNE: Bariatric surgery non-eligible; FLDP: Fatty liver disease program.

Referral rates

Among the 491 patients who were BSE, 127 (26%) were referred for bariatric surgery evaluation, with 31 patients (representing 24% of referred patients and 6.3% of all BSE patients) ultimately receiving bariatric surgery (Table 3). There were no significant differences in the demographic characteristics of patients who were referred for and received surgery vs those who were referred for and did not receive surgery, aside from a difference in insurance status (Table 4). Among patients referred for bariatric surgery, those who did not undergo the procedure frequently missed the initial evaluation or did not complete the required preoperative work-up.

Table 4 Demographic and surgical characteristics of bariatric surgery eligible patients who were referred for surgery, n (%).

Patient characteristics		(+) Surgery	(-) Surgery	P value
Number of patients	n = 127	31	96
Sex	Male	6 (20)	32 (33)	0.14
	Female	25 (80)	64 (67)
Age in years, mean ± SD		49 ± 10.5	52 ± 13.1	0.248
Race	White	21 (68)	76 (79)	0.236
	Black	4 (13)	8 (8)
	Asian	1 (3)	0 (0)
	Other/declined	5 (16)	12 (13)
Ethnicity	Hispanic	6 (19)	20 (21)	0.86
	Non-Hispanic	25 (81)	76 (79)
Insurance status	Uninsured	0 (0)	0 (0)	0.035
	Medicaid	14 (46)	25 (26)
	Medicare	2 (6)	24 (25)
	Private	15 (48)	46 (49)
Median household income by zip code	< $50000	7 (23)	18 (19)	0.74
	$50000-$100000	19 (61)	66 (69)
	> $100000	5 (16)	12 (12)
Interventions and outcomes
Trialed anti-obesity medication	Lifetime	13 (42)	32 (33)	0.59
	Trialed since first FLDP visit	8 (26)	26 (27)
Surgery type	RYGB	10 (32)
	SG	21 (68)
Non-surgical weight loss method	Medication and lifestyle		41 (43)
	Lifestyle only		55 (57)
Nutrition referral	Since first FLDP visit	15 (48)	62 (65)
Mean BMI reduction (kg/m2), mean ± SD	RYGB	8.3 ± 5.4	1.3 ± 3.9	< 0.0001

Categorical variables compared using the χ² test. FLDP: Fatty liver disease program; RYGB: Roux-en-Y gastric bypass; SG: Sleeve gastrectomy; BMI: Body mass index.

In the surgery group, 21 patients received sleeve gastrectomy (SG), while 10 patients received Roux-en-Y gastric bypass (RYGB) (Table 4). Five patients initially had SG and were later converted to RYGB; this subset of patients was included in the RYGB group. Additionally, two patients had adjustable lap-band removal before the incident surgery. There were 96 patients referred for surgery who did not proceed with bariatric surgery and instead continued with lifestyle and/or medical management of their disease (Table 4). Among the 31 patients who underwent bariatric surgery, 13 patients (42%) also used an anti-obesity medication. Among the 96 patients who did not receive bariatric surgery, 38 patients (41%) used an anti-obesity medication and 62 (67%) received a registered dietitian (nutritionist) referral (Table 4).

Weight loss outcomes

When comparing RYGB to SG, BMI reduction was 8.3 ± 5.4 kg/m² vs 9.1 ± 6.3 kg/m², respectively (Table 5, Figure 2A), mean weight reduction was 26.3 ± 10.3 kg vs 27.7 ± 17.9 kg (Table 5), and TBWL was 20% ± 2.7% vs 21% ± 2.1% (Table 5, Figure 2A). BSE patients who were referred for surgery but opted for non-surgical weight management had a mean BMI reduction of 1.3 ± 3.9 kg/m², a TBWL of 2.5% ± 0.8%, and a mean weight reduction of 4.1 ± 10.2 kg after one year (Table 5, Figure 2A). The mean BMI reduction and TBWL achieved in patients who opted for non-surgical weight loss was less than that achieved through bariatric surgery (P < 0.0001) (Table 5, Figure 2A). In patients who received bariatric surgery, predicted weight loss outcomes were calculated using the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program calculator (Table 1). There was no significant difference between actual vs predicted values for BMI, weight loss, or TBWL at one-year post surgery (Figure 2B).

Figure 2 Weight and liver-related outcomes in patients with metabolic dysfunction-associated steatotic liver disease undergoing bariatric surgery vs lifestyle management/medication. A: Comparison between total body weight loss in patients who underwent gastric bypass or sleeve gastrectomy vs non-surgical management (lifestyle/medication). P < 0.0001 for weight loss through either type of bariatric surgery compared to non-surgical management (unpaired t-test, Welch’s); B: Comparison between actual vs predicted body mass index post bariatric surgery (P = 0.12, unpaired t-test, Welch’s). The predicted body mass index was obtained using the Metabolic and Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program Bariatric Surgical Risk/Benefit Calculator; C: Change in fibrosis-4 in patients who underwent bariatric surgery vs lifestyle/medication management (P < 0.048, unpaired t-test, Welch’s); D: Change in nonalcoholic fatty liver disease fibrosis score in patients who underwent bariatric surgery vs lifestyle/medication management (P = 0.005); E: Number of patients with each fibrosis stage pre-surgery. Categories are as follows: F0-1: 2.0-8.4 kPa, F2: 8.5-9.4 kPa, F3: 9.5-13.5 kPa, F4 > 13.5 kPa; F: Change in liver stiffness from pre-surgery to post-surgery for patients undergoing bariatric surgery. TBWL: Total body weight loss; BMI: Body mass index; FIB-4: Fibrosis-4; NFS: Nonalcoholic fatty liver disease fibrosis score; RYGB: Roux-en-Y gastric bypass; SG: Sleeve gastrectomy.

Table 5 Outcomes in surgical vs non-surgical patients who were referred for bariatric surgery (mean ± SD, n/total n).

Patient characteristics		(+) Surgery	(-) Surgery	P value
Mean BMI (kg/m2)	RYGB	8.3 ± 5.4, 10/10	1.3 ± 3.9, 89/96	< 0.0001
	SG	9.1 ± 6.3, 21/21		< 0.0001
Mean weight reduction	RYGB	26.3 ± 10.3, 10/10	4.1 ± 10.2, 89/96	< 0.0001
	SG	27.7 ± 17.9, 21/21		< 0.0001
TBWL (%, mean ± SE, n/total n)	RYGB	20 ± 2.7, 10/10	2.5 ± 0.81, 89/96	< 0.0001
	SG	21 ± 2.1, 21/21		< 0.0001
∆NAFLD fibrosis score ± n/total n		-0.91 ± (22/31)	-0.008 ± (62/96)	0.005
∆FIB-4 score ± n/total n		-0.12 ± (30/31)	-0.055 ± (40/96)	0.58

Comparisons between groups analyzed using Welch’s unpaired t-test. BMI: Body mass index; RYGB: Roux-en-Y gastric bypass; SG: Sleeve gastrectomy; TBWL: Total body weight loss; NAFLD: Non-alcoholic fatty liver disease; FIB-4: Fibrosis-4.

Liver-related outcomes

Patients who underwent surgery had an average ΔFIB-4 score of -0.12, compared to -0.055 for those who did not have surgery (P = 0.58; Table 5, Figure 2C). Patients who had bariatric surgery had an average ΔNFS of -0.91, while those who did not have surgery had an average ΔNFS of -0.008 (P = 0.005; Table 5, Figure 2D). Of the 31 patients who underwent bariatric surgery, 21 had pre-operative VCTE data available. Of those 21 patients, 12 also had post-operative VCTE data obtained between 1.3 years and 3.7 years post-surgery with a median follow-up time of 2.2 years. Patients undergoing surgery had varied pre-operative disease severity, as evidenced by LSM ranging from 3.9 kPa (suggestive of F0-1 fibrosis) through 41.0 kPa (suggestive of F4 fibrosis; Figure 2E). The average LSM for patients who underwent bariatric surgery decreased from 12.4 kPa pre-operatively to 7.3 kPa post-operatively while the median decreased from 8.9 kPa to 7.0 kPa (Figure 2F).

Post-operative complications

Following bariatric surgery, no major complications, including 30-day mortality, anastomotic leaks, surgical site infections, or respiratory failure were reported (Table 5). Furthermore, no major adverse cardiac events were reported in the 30-day post-operative period (Table 5).

DISCUSSION

In this study of MASLD patients receiving care in a program integrating hepatology and weight management expertise, referral for bariatric surgery was substantially higher than the national average, with a referral rate of 27% in the current study vs less than 1% national referral rate based on current literature. As expected, those who underwent surgery had greater weight loss and improved liver outcomes compared to those who did not undergo surgery. Patients eligible for bariatric surgery based on the presence of surgical criteria determined from chart review were more likely to be female, Black, and of lower SES than those who were not eligible. However, in the group referred for surgery, there were no differences in sex, race, or SES between surgical and non-surgical patients. Among the 31 patients who received bariatric surgery, there were no 30-day post-surgical complications or mortality. The study is strengthened by the availability of detailed patient demographic and clinical data which allowed for comprehensive evaluation of patient characteristics and outcomes[21,22]. Nonetheless, its retrospective design and the potential confounding effects of anti-obesity pharmacotherapy must be considered when interpreting the findings.

Patients with MASLD presenting to traditional hepatology or primary care clinics without obesity medicine expertise often fail to meet weight loss goals, as these clinics typically recommend non-targeted lifestyle changes that are ineffective in this population[23]. Treatment of MASLD requires an individualized approach taken by providers equipped to provide personalized weight loss recommendations. For some patients, this may involve offering bariatric surgery. In our cohort, while use of anti-obesity medication was comparable between surgical and non-surgical groups, the surgery group had significantly greater weight loss and improved liver-related outcomes after one year. However, it is important to note that semaglutide was not approved for weight loss until 2021; before then, patients may have been using lower diabetes-approved doses that are less effective for weight loss[24]. With the approval of semaglutide and tirzepatide for weight loss and the more recent approval of semaglutide for metabolic dysfunction-associated steatohepatitis, the landscape of obesity and MASLD pharmacotherapy is rapidly evolving[25]. Further research is needed to determine the relative efficacy of currently employed weight loss strategies in MASLD. It is well established that bariatric surgery achieves the most sustainable weight loss and improvement in fibrosis in MASLD and should remain within a clinician’s armamentarium of weight loss strategies, especially for patients with severe obesity and/or inability to tolerate anti-obesity pharmacotherapy[26,27]. The effectiveness of multidisciplinary clinics has been well-described for management of other complex diseases such as heart failure and lymphoma[28]. Specialized multidisciplinary clinics have been shown in several meta-analyses to reduce rehospitalization and mortality and offer a promising solution to address treatment barriers in patients with MASLD[29]. Although guidelines published by the American Association for the Study of Liver Diseases and the American Diabetes Association underscore the importance of individualized, interdisciplinary care for patients with MASLD, studies reporting patient outcomes from a dedicated MASLD clinic are limited[30,31]. The current study highlights the potential of such a clinic in providing access to appropriate obesity care and improving liver-related outcomes.

We found many demographic differences in patients who were BSE based on standard eligibility criteria and those who were BSNE. BSE patients were more likely to be younger, female, non-White (specifically non-Hispanic Black), and of lower SES compared with BSNE patients. It is well known that individuals with lower SES are more likely to experience obesity when compared with individuals with higher SES, likely as a result of increased psychological and physiologic stress, decreased physical activity due to lack of green space and accessible exercise facilities, and poor access to nutritious food sources[32,33]. The SES disparity seen in our cohort lends credence to the need to address structural inequities as a major contributor to the rising prevalence of metabolic syndrome in the United States.

BSE patients who received surgery were more likely to have Medicaid compared to those who did not receive surgery. However, there was no significant difference in median household income between groups. A large body of evidence has revealed an association between increased median household income with resultant access to private insurance and increased rates of bariatric surgery[34,35]. This is likely because availability of private insurance in those with higher household income enhances access to longitudinal care involving primary care as well as specialty care when compared to those with lower median household incomes. While our cohort stands in contrast to this trend, our sample sizes are small, and more research is needed to determine whether multidisciplinary MASLD clinics provide more equitable access to surgical referrals across various patient demographic groups.

Patients who had surgery achieved greater reductions in NFS and FIB-4 scores compared to those who had only pharmacological and lifestyle management. Although bariatric surgery is the most effective and durable weight loss intervention, weight regain after surgery remains a challenge. The reasons for this may be multifactorial, including difficulties with behavioral and nutritional management after surgery[36]. Studies have shown that patients who had close nutritional and cognitive behavioral follow-up had increased success with maintaining weight loss[37,38]. It is well known that there is no standard diet, medical regimen, or lifestyle program that is universally effective for all patients[39]. At present, optimal obesity management remains a highly individualized process that should involve close follow-up with a multidisciplinary team, with continued dietary, behavioral, and medical management by providers who are familiar with the patient and their needs.

There are several limitations in this study. The study was retrospective and therefore did not have comprehensive baseline and follow up liver fibrosis data and imaging for the cohort of MASLD patients examined. The retrospective nature of the study also limited long-term assessment of outcomes. Given that a large portion of the data was collected during the coronavirus disease pandemic, there was a preponderance of virtual visits, which ultimately affected the availability of weight data. However, virtual visits are unlikely to discontinue altogether, and more work should be done to integrate methods of obtaining vital data when patients are not physically present in the clinic. Also, given that many patients undergoing bariatric surgery also used anti-obesity medications, the outcomes noted in these patients cannot be solely attributed to bariatric surgery. Moreover, the average weight loss in the non-surgical group in this study was only 2.5%, but with the more recent approval and widespread use of single and dual incretin agonists such as semaglutide and tirzepatide, patients pursuing non-surgical methods can achieve a greater degree of weight loss[21]. Another limitation is that the classification of BSE vs BSNE was based solely on BMI calculation and the presence of an obesity related comorbidity by ICD-10 coding; it is possible that the number of patients classified as BSNE would be higher if individual patient characteristics such as severe heart or lung disease or unmanaged psychiatric disorders precluding surgery were considered. Finally, a large proportion of patients who underwent bariatric surgery had non-fibrotic MASLD, which may lead to underestimation of the effect of surgery on FIB-4 and/or NFS scores in this analysis.

CONCLUSION

A multidisciplinary care model for patients with MASLD incorporating obesity medicine care and hepatology care results in substantially higher bariatric surgery referral rates compared to the national average. Bariatric surgery improves both weight loss and liver-related outcomes for eligible patients with MASLD compared with non-surgical interventions. Looking ahead, wider adoption of this multidisciplinary model may transform routine MASLD care, ensuring that effective weight loss interventions including bariatric surgery are systematically incorporated into developing treatment pathways.

ACKNOWLEDGEMENTS

Special thanks to Chehayeb RJ, Valido K, Cotter R, and the members of the Yale joint data analytics team for their assistance in the data extraction.