Rapid improvement of hepatic steatosis and liver stiffness after metabolic/bariatric surgery: a prospective study

Metabolic dysfunction-associated steatotic liver disease (MASLD) and related steatohepatitis (MASH) are common among obese patients and may improve after metabolic/bariatric surgery (MBS). 93 Patients undergoing MBS in 2021–2022 were prospectively enrolled. Liver stiffness measurement (LSM; via vibration-controlled transient elastography [VCTE], point [pSWE] and 2D [2DSWE] shear wave elastography) and non-invasive steatosis assessment (via controlled attenuation parameter [CAP]) were performed before (baseline [BL]) and three months (M3) after surgery. 93 patients (median age 40.9 years, 68.8% female, median BL-BMI: 46.0 kg/m2) were included. BL-liver biopsy showed MASLD in 82.8% and MASH in 34.4% of patients. At M3 the median relative total weight loss (%TWL) was 20.1% and the median BMI was 36.1 kg/m2. LSM assessed by VCTE and 2DSWE, as well as median CAP all decreased significantly from BL to M3 both in the overall cohort and among patients with MASH. There was a decrease from BL to M3 in median levels of ALT (34.0 U/L to 31 U/L; p = 0.025), gamma glutamyl transferase (BL: 30.0 to 21.0 U/L; p < 0.001) and MASLD fibrosis score (BL: − 0.97 to − 1.74; p < 0.001). Decreasing LSM and CAP, as well as liver injury markers suggest an improvement of MASLD/MASH as early as 3 months after MBS.

Metabolic dysfunction-associated steatotic liver disease (MASLD), which may progress to metabolic dysfunctionassociated steatohepatitis (MASH), has become a global healthcare burden and its prevalence is constantly increasing 1,2 .MASLD is strongly associated with obesity, diabetes and hyperlipidemia and is hence considered the hepatic manifestation of the metabolic syndrome 3 .To this day there are no pharmacological treatments specifically approved for MASLD/MASH 4 .However, in obese patients, weight loss substantially improves liver health by reducing insulin resistance, decreasing hepatic steatosis, and improving liver enzymes 5,6 .Lifestyle interventions that promote weight loss, such as physical activity and dietary changes are thus, effective in improving liver health in individuals with MASLD/MASH 7,8 .Unfortunately, long-term weight loss is often not sufficiently achieved by lifestyle interventions 9 .Metabolic/Bariatric surgery (MBS) is now increasingly recognized as effective treatment for MASLD/MASH 10 , with particularly strong evidence for its effectiveness generated from studies involving patients with obesity 11,12 .Recently, a multicenter randomized trial was conducted, demonstrating the superiority of MBS to lifestyle intervention plus best medical care in the therapy of obese patients with MASH 13 .The underlying mechanisms include weight loss, metabolic improvements, beneficial effects on the gut microbiota and their metabolic function, gut hormones and systemic inflammation, and amelioration of adipokine levels [12][13][14] .However, we have also observed new-onset of hepatic dysfunction after MBS in some patients 15 .Thus, additional data on the course of MASLD/ MASH after MBS is needed.
While liver biopsy still represents the gold-standard for liver disease evaluation, it is an invasive procedure associated with morbidity and even mortality 16 .Consequently, non-invasive liver stiffness measurement (LSM) is now widely used as a surrogate marker for liver fibrosis 16 .LSM can be performed via vibration-controlled transient elastography (VCTE), point-shear wave elastography (pSWE) or 2D shear wave elastography (2DSWE) 16 .While VCTE is well-validated and widely available 17,18 , pSWE and 2DSWE can be integrated in ultrasound devices and therefore are promising technologies combining LSM with liver imaging 16 .The dedicated VCTE device (FibroScan®, Echosens, France) allows for simultaneous assessment of liver steatosis by controlled attenuation parameter (CAP) 19 .
The aim of this prospective, observational study was to investigate the early trajectories of (i) LSM as assessed by VCTE, pSWE and 2DSWE, (ii) CAP and (iii) lab-based surrogates of MASLD/MASH severity before and 3 months after MBS.

Study design
Patients who underwent MBS at the Vienna General Hospital between January 2021 and December 2022 were prospectively enrolled and characterized.Examinations were performed both preoperatively (BL) and during the first postoperative follow-up visit, scheduled at 3 months (M3).Patients lost to follow-up prior to M3 after MBS were excluded from further analysis.Weight loss outcomes were quantified in terms of excess weight loss percentage (EWL%), utilizing a reference standard of 25 kg/m 2 BMI and percentage of total weight loss (%TWL) relative to initial preoperative body weight.
The study included patients who underwent one of the following metabolic/bariatric procedures: Single Anastomosis Duodeno-Ileal Bypass with Sleeve Gastrectomy (SADI-S), One Anastomosis Gastric Bypass (OAGB), Roux-en-Y Gastric Bypass (RYGB) or Sleeve Gastrectomy (SG).The surgical method was decided based on BMI, comorbidities, the occurrence of reflux, patient preference and other factors.

Surgical method
All surgical procedures followed standardized protocols and techniques were previously described [20][21][22][23] .Briefly, each surgery involved the use of 5 trocars, except in cases of SADI-S where a 6th trocar above the symphysis was occasionally necessary.
www.nature.com/scientificreports/For SG, a sleeve was created using a 12 mm bougie starting approximately 2-4 cm orally from the pylorus, using 2 black cartridges at the beginning for the thicker tissue near the pylorus, followed by purple and beige cartridges.After resection, the gastrosplenic and gastrocolic ligament were refixed at the staple line.
For SADI-S, a 16 mm bougie was used to create the pouch.After this, 300 cm of small bowel were measured from the ileocecal transition for the common limb, followed by creation of a hand-sewn duodenoileostomy.
In OAGB and RYGB, the gastric pouch was formed long and narrow using the same bougie size (12 mm) as in SG.The pouch in OAGB was slightly longer than that in RYGB.The biliopancreatic limb was measured 150 cm long in both procedures.For RYGB, the jejunojejunostomy was set at 70 cm of the alimentary limb.The Peterson space was closed in both procedures, additionally the mesenteric gap was closed in RYGB.

Surgical complications
A detailed description of the surgical procedures is provided in the Supplementary material.Postoperative complications were classified according to the Clavien Dindo Classification (CDC) 24 .

Liver stiffness measurement and liver steatosis assessment
Patients underwent VCTE (FibroScan®, Echosens, France) for LSM and CAP evaluations, as well as point shearwave elastography (pSWE) and 2D shear wave elastography (2DSWE) (Siemens Medical Solutions USA, Inc., Issaquah, WA) using the 5C1 probe on the Siemens ACUSON Sequoia system (Software Revision VA30) using a standardized protocol.Prior to examinations, patients fasted for at least 4 h.
Adhering to the 2020 SRU Guidelines 25 and local expertise 26,27 measurements were performed in a right intercostal space, with patients in a supine position and breath held at mid-respiration.For pSWE and 2DSWE, the probe was positioned 1.5 cm beneath the liver capsule, as directed by the ultrasound machine's built-in guide, while avoiding larger structures or vessels in the region of interest.For VCTE the medium (M) or extra-large (XL) probe was chosen according to the built-in probe selection tool.Reliable VCTE measurements were defined as LSM < 7.1 kPa or interquartile range/median (IQR/med) ≤ 0.3 for LSM ≥ 7.1 kPa and as IQR/med ≤ 0.3 dB/m for CAP corresponding to previously established criteria 18,19 .

Liver biopsy
A protocol liver biopsy was conducted during bariatric surgery.The biopsies were analyzed in a standardized manner at the Department of Pathology at the Medical University of Vienna following previously established methods 32 .MASLD was defined as hepatic steatosis in the presence of at least one metabolic risk factor 2 and MASH was diagnosed according to the algorithm proposed by Bedossa et al. 33 .Moreover, the non-alcoholic fatty liver disease activity score (NAS) was assessed.

Laboratory parameters
Laboratory parameters were analyzed at the ISO-certified Department of Laboratory Medicine at the Vienna General Hospital.Standard laboratory methods were used for the measurements of routine laboratory parameters that were assessed before (BL) and M3 after MBS.
Standard laboratory thresholds for men and women were used as upper limit of normal (ULN) for parameters of hepatocellular (aspartate aminotransferase [AST], alanine aminotransferase [ALT]) and cholestatic liver injury (gamma glutamyl transferase [GGT], bilirubin).Hepatocellular liver injury was defined as increase of AST or ALT > 3xULN, while cholestatic liver injury was defined as increase of GGT or bilirubin > 2xULN as modified from the definition of the American College of Gastroenterology Clinical Guideline 34,35 .

Statistical analysis
Continuous data are presented as median and interquartile range (IQR), while categorical variables are reported as the number (n) and proportion (%) of patients exhibiting the parameter of interest.Wilcoxon's signed-rank test was employed for comparing metric variables in dependent samples, and Pearson's Chi-squared test was used for comparing categorical variables across multiple groups.Statistical analyses were performed using IBM SPSS 27.0 (IBM, Armonk, NY, USA) and GraphPad Prism 8 (GraphPad Software, La Jolla, CA, USA).A two-sided p-value of < 0.05 was deemed statistically significant.

Ethics
This study received approval from the Medical University of Vienna's ethics committee (No. 1509/2020) and adhered to the current version of the Helsinki Declaration and the institution's Guidelines on Good Scientific Practice.All participating patients provided written informed consent prior to study inclusion.

Patient characteristics
The cohort building process is depicted in Fig.

Surgical outcomes and complications
After M3 median TWL% was 20.1% and median EWL% was 46.8% (Table 1).The overall complication rate was 15.1%; 53.3% of complications were early complications (within 30 days) and 46.7% were late complications (> 30 days after MBS).Of the eight early postoperative complications, 75% were classified as CDC grade IIIb and 25% as grade I. Out of the patients who required revision surgery within 30 days, three being attributed to bleeding, two to dysphagia, and one to leakage at the gastrojejunostomy site.Additionally, seven cases of late complications were observed, occurring after the first postoperative month.Of those, 57% classified as CDC Grade II, requiring intravenous substitutional therapy due to iron deficiency and 14% as CDC grade I, necessitating elongated treatment with proton pump inhibitors due to an anastomotic ulcer at 9 moths.28% were classified

Success rate and reliability of VCTE and CAP
In total, longitudinal LSM was available in 56 patients.Among these patients, baseline or follow-up VCTE was unsuccessful or unreliable in 23.2% (n = 13), which is comparable to the success rate of 77.4% in the entire cohort.Baseline or follow-up CAP was unsuccessful or unreliable in 16.1% (n = 9) patients.Again, this corresponded well with the success rate of 82.5% in the entire cohort.In comparison, pSWE and 2DSWE, which were longitudinally performed in 54 patients, were successful in 94.4% (n = 51) of patients.
There was a marked decrease in MASLD fibrosis score 3 months after surgery (BL: − 0.97 points vs. M3: − 1.74 points; p < 0.001).At the same time, AST to platelet ratio index (APRI) remained unchanged (p = 0.974).Comparable results were obtained in the subgroup of patients with MASH (Table S3, Fig. S1).
While there were 1 patient with hepatocellular and 2 patients with cholestatic liver injury at baseline, at M3 after MBS laboratory signs of hepatocellular or cholestatic liver injury were found in 2 and 3 patients, respectively (Table S4).Notably, one patient with laboratory signs of hepatocellular liver injury 3 months after surgery also exhibited a bilirubin of 2.48 mg/dL (as compared to 0.49 mg/dL at BL). Importantly these findings were transient without deterioration of hepatic synthetic function or signs of liver failure and normalized without specific medical/therapeutic interventions.

Discussion
Obesity and obesity-related comorbidities including MASLD/MASH 2 represent significant challenges for health care on a global scale, with dramatically increasing prevalence 36 .While lifestyle modification and medical treatment for obesity and MASLD is challenging, MBS offers an effective option 10 .In this study, we investigate the short-term changes in LSM, CAP and laboratory-based surrogates of liver injury and metabolic parameters after MBS.
The overall composition of the cohort reflects a representative sample of patients undergoing MBS.After 3 months, the cohort exhibited an average excess weight loss (EWL) of 46.8%, with an average BMI reduction www.nature.com/scientificreports/ of 10 kg/m 2 which is comparable to the existing literature concerning short term results after MBS 37 .Typically, the nadir weight is reached 12 to 18 months after MBS.Around 25% of all patients suffered from type II diabetes preoperatively, with 32% of them requiring insulin therapy for treatment.After 3 months 28% of these patients showed an improvement regarding their medication.Also, the HbA1c levels of all patients decreased significantly.The development of MASLD is closely related to insulin resistance and the development of type II diabetes 38 .MBS and especially RYGB surgery showed an excellent performance in improvement or even remission of diabetes mellitus 39 .
The complication pattern observed in this study is common for a patient collective after MBS 40 .Compared to existing literature, the complication rate depicted here is relatively low 40,41 , but there were some severe complications requiring surgical correction.www.nature.com/scientificreports/Non-invasive tests (NIT) and particularly LSM has become increasingly relevant for the diagnosis of compensated advanced chronic liver disease (cACLD) and risk assessment for adverse clinical events such as hepatic decompensation or death 42 .In this study, we used VCTE as a well-established method for the non-invasive assessment of LSM 16,18,42 .Obesity is one of the most important factors associated with failure of VCTE-LSM 43 .Indeed, in our study in a cohort of severely obese individuals there was a high rate of failure or unreliable result of VCTE-LSM (i.e.23.2%), which was comparable in patients undergoing different types of bariatric procedures.
Importantly, a significant decrease in LSM assessed by VCTE and 2DSWE was visible already 3 months after MBS.All but one patient with available longitudinal and reliable VCTE measurements exhibited VCTE-F0/F1 at M3.
Congruently, CAP as a parameter of liver steatosis significantly decreased 3 months after MBS.While noninvasive staging of liver steatosis using CAP is difficult, particularly in (still) obese subjects, and different cutoffs have been proposed in patients with MASLD [29][30][31] , in our cohort there was a significant decrease in non-invasively assessed liver steatosis stage, no matter which system of CAP cutoffs was applied in our cohort.Of note, there was also a significant decrease in LSM and CAP at M3 when only analyzing patients with MASH, underscoring the potential benefit of MBS particularly in these patients by early improvement of MASH already early after MBS.These findings are in line with previous studies showing regression of MASLD/MASH in patients with obesity after MBS 10,44 .
The importance of MBS in the treatment of MASLD/MASH in patients with obesity is increasing in light of escalating incidence 1,2 .Recently published findings of a multicenter randomized trial suggest the superiority of MBS compared to lifestyle intervention and optimal medical care after a 1-year observational period 13 .Additionally, evidence indicates that the resolution of MASLD/MASH following MBS is enduring, as demonstrated by other authors who reported substantial percentages of 84% of MASLD/MASH resolution after a 5-year period 10 .Our findings supplement the existing knowledge regarding the speed at which these changes manifest.In this brief 3-month observational period, significant improvements in MASLD/MASH were already evident.This signifies that MBS may be capable of rapidly improving hepatic steatosis/inflammation.
Furthermore, importantly, this study provides evidence that development of liver dysfunction or even liver cirrhosis after MBS is uncommon and that, on the contrary, liver chemistries including ALT and GGT improved further even within the normal range.Since many patients with MASLD/MASH exhibit transaminases within the normal range 45 , this decrease of ALT might indicate a rapprochement to a "true normal range" in these patients.
On the other hand, there was a slight increase of bilirubin in patients after bariatric procedures.Moreover, there was one patient developing hepatocellular and cholestatic liver injury 3 months after MBS, respectively.While our data does not suggest that liver injury or liver dysfunction is particularly prevalent after MBS, liver chemistries should be controlled regularly in all patients undergoing metabolic/bariatric procedures, as liver dysfunction and chronic liver injury has been reported in some individuals 15 .
This study has some limitations.Firstly, due to organizational reasons during the SARS-CoV-2 (COVID-19) pandemic 46 , not every parameter was available for all patients.This was especially relevant for LSM at follow-up.Apart from the missing values, this reduced the sample size for these parameters, which represents the major limitation of our study.Secondly, it is important to note that this study was conducted as an observational study.The selection of the surgical method was based on factors such as BMI, patient preference, comorbidities (excluding liver health), and the presence of reflux.The consideration of MASLD/MASH status did not play a role in determining the choice between surgical methods.Moreover, this longitudinal study investigated the noninvasive parameters of liver stiffness and liver steatosis over a course of three months after MBS.As such, we did not directly investigate changes in liver fibrosis and liver fat content and cannot derive from our data whether the short-term changes in non-invasive parameters of liver fibrosis and steatosis truly reflect a regression of fibrosis and steatosis in MASLD or reflect the dynamics of other confounding factors such as inflammation, liver congestion, blood flow or cholestasis.Notably, while MASLD and MASH were prevalent in the cohort included in this study, no patient exhibited advanced fibrosis or liver cirrhosis.Moreover, our study only examined short-term changes after MBS.Further studies are required to analyze the trajectory of liver stiffness, liver steatosis and liver chemistries after MBS over a longer period of time.Finally, we did not examine the associations between liver histology features and non-invasive parameters of liver fibrosis and liver stiffness, as this was already done by a previous study 47 .
In conclusion, in this prospective observational study, we thoroughly investigated the short-term trajectory of various non-invasive markers of liver stiffness and liver steatosis in a cohort of obese patients undergoing MBS with a high prevalence of MASLD and MASH.Importantly, LSM-as assessed by both VCTE and SWE-and fibrosis stage decreased significantly 3 months after MBS.Furthermore, liver steatosis as assessed by CAP declined with not only a significant decrease in median CAP, but also a regression of steatosis stage.Importantly, comparable findings were obtained in the subgroup of patients with MASH.This denotes that MBS may be a viable treatment for obese patients with MASLD/MASH patients facilitating rapid improvement of hepatic steatosis and inflammation, although further research is required to confirm our findings.Finally, assessing the trajectory of liver chemistries, we showed that liver injury does occur, but is infrequent after MBS procedures.Importantly, no patient in our study developed hepatic dysfunction after MBS.Still, regular postoperative laboratory check-up of liver chemistries is warranted to detect liver injury early and be able to enact countermeasures in a timely fashion.
29 Overall, 93 patients (SADI-S: n = 28; OAGB: n = 26; RYGB: n = 27; SG: n = 12) with a median age of 40.9 years and female predominance (68.8%) were included in the final cohort.Detailed patient characteristics are provided in Table-1.The median BMI was 46.0 kg/m 2 .Among patients with reliable CAP at BL, 15.0% (n = 12) of patients had steatosis stage S0 in liver histology, while 33.7% (n = 27) had S1, 28.8% (n = 23) had S2 and 22.5% (n = 18) had S3.In comparison, Table S1 lists the prevalence of non-invasively-assessed liver steatosis stages staged by different previously proposed CAP cut-offs that have been previously proposed for patients with MASLD.The CAP steatosis stage cutoffs proposed by Karlas et al.29had the highest correlation with liver steatosis in histology (ρ = 0.312; p = 0.005), but classified 82.5% (n = 66) of patients with reliable CAP at BL as steatosis stage ≥ S2, which was markedly higher than the proportion of patients with ≥ S2 in liver histology (51.7%).In contrast, the cutoffs proposed by Naveau et al.
CDC grade IIIb requiring revision surgery.One needed endoscopic esophageal stent treatment for two weeks due to dysphagia after SG, while the other one was converted to RYGB due to biliary reflux after SADI-S.

Table 1 .
Patient characteristics and weight change after metabolic/bariatric surgery.3M 3 months, BL baseline, BMI body mass index, CAP controlled attenuation parameter, COPD chronic obstructive pulmonary disease, LSM liver stiffness measurement, MASH metabolic dysfunction-associated steatohepatitis, MASLD metabolic dysfunction-associated steatotic liver disease.

Table 2 .
2rajectory of LSM and CAP after metabolic/bariatric surgery.CAP controlled attenuation parameter, LSM liver stiffness measurement.1Indicatedasmedian of reliable longitudinal VCTE measurements.Available in n = 43 patients.2Indicatedas median of reliable longitudinal CAP measurements.Available in n = 47 patients.

Table 3 .
Short-term changes of VCTE-based fibrosis risk and CAP-based steatosis assessment after bariatric/ metabolic surgery.CAP controlled attenuation parameter, LSM liver stiffness measurement, VCTE vibrationcontrolled transient elastography. 1Indicated as fibrosis stage of reliable longitudinal VCTE measurements.Available in n = 43 patients. 2Indicated as median of reliable longitudinal CAP measurements.Available in n = 47 patients.