Association of Rapid Early Weight Loss with One-Year Hepatic Steatosis Improvement After Sleeve Gastrectomy: A Retrospective Cohort Study

Abstract

Background: Metabolic dysfunction-associated steatotic liver disease (MASLD), previously referred to as NAFLD, affects nearly one-third of the global adult population and is a leading cause of chronic liver disease, particularly among individuals with obesity and type 2 diabetes (T2DM). Bariatric surgery, including sleeve gastrectomy (SG), has demonstrated a favorable impact on liver fat reduction. However, the predictive value of early postoperative weight-loss trajectories for long-term hepatic improvement remains uncertain, especially in Asian populations, in which MASLD remains understudied despite its increasing prevalence. Methods: We retrospectively reviewed 198 adults who underwent SG at a tertiary Korean center between January 2019 and April 2024. After excluding 21 who had postoperative complications or missed early follow-up, 177 patients (mean age 37 ± 9 years; 41% male; mean body mass index (BMI) 40 kg/m²) were included in the final analysis. The two-week total weight-loss index (TWL_2W) was calculated, and its association with one-year hepatic steatosis resolution, defined as the normalization of the hepatic steatosis index (HSI < 30) at one year, was assessed using receiver operating characteristic (ROC) analysis to explore the trend toward predictive value. Optimal cut-offs were derived using the Youden Index. Multivariable logistic regression models were adjusted for age, sex, baseline hemoglobin A1c (HbA1c), and BMI. Subgroup analyses were performed according to baseline HSI (35-44, 45-54, ≥55) and type 2 diabetes mellitus (T2DM) status. Results: The mean TWL_2W was 7.9 ± 6.6%. A loss of 7.9% optimally predicted HSI values < 30 at one year (area under the curve [AUC] 0.602; unadjusted odds ratio [OR] 2.34; 95% confidence interval [CI] 1.16-4.73). Predictive accuracy improved in T2DM patients (AUC 0.737, 95% CI 0.54-0.95), in whom TWL_2W ≥ 9.1% conferred an adjusted OR 9.12 (95% CI 1.39-59.82), whereas no association was observed in non-diabetic subjects. Stratified analysis showed a pronounced effect in moderate baseline steatosis (HSI 45-54; OR 3.56), but absolute normalization was rare when the baseline HSI was ≥55. Early weight loss was not significantly linked to one-year HbA1c or triglyceride targets. Conclusions: An 8-9% reduction in body weight within two weeks of SG was independently associated with the resolution of hepatic steatosis at one year, particularly among patients with T2DM or moderate baseline hepatic steatosis. This simple metric may assist in early risk stratification and guide personalized postoperative care.

Keywords: early weight loss; hepatic steatosis index; risk stratification; sleeve gastrectomy; type 2 diabetes mellitus.

Figure 1

Receiver operating characteristic curves (ROCs) for early postoperative weight loss in predicting one-year metabolic outcomes. (A) Predictive performance of %TWL_2W for achieving target levels of HSI < 30, glycemic control (HbA1c < 6.5% for DM or <5.7% for non-DM), LDL-C < 100 mg/dL, and TG < 150 mg/dL. (B) Subgroup analysis of ROC curves for predicting HSI < 30 according to diabetes status.

Figure 2

Bar plot of area under the receiver operating characteristic curve (AUC) for early postoperative weight loss in predicting one-year metabolic outcomes. (A) AUCs for four metabolic outcomes: HSI < 30, HbA1c target, LDL-C < 100 mg/dL, and TG < 150 mg/dL. Error bars represent 95% confidence intervals (CIs). (B) Subgroup comparison of AUC for HSI < 30 according to diabetes status (T2DM vs. non-T2DM). Error bars represent 95% CIs.