OBJECTIVE: To characterize the weight fluctuation and explore its influencing factors after glucagon-like peptide 1 receptor agonist (GLP-1RA) treatment in patients with overweight or obesity.
METHODS: This retrospective cohort study enrolled 679 patients with overweight or obesity who initiated GLP-1RA treatment between November 2022 and October 2024. The weight measurements were collected during in-person clinic visits by uniform weighing scales and documented in the electronic medical record (EMR). The weight fluctuation curves were stratified into three phenotypes (Successful Weight Reduction, Remaining Stable and Weight Regain) and went through intergroup analyses. Subsequently, the association between potential influencing factors and weight fluctuations were estimated by univariate and multivariate logistic regression, where the weight fluctuation phenotypes were dichotomized into a binary variable, classified as "successful weight reduction" or "unsuccessful weight reduction". Subgroup analyses were performed in participants with obesity, prediabetes, liraglutide users and semaglutide users.
RESULTS: Patients with a longer duration of GLP-1RA treatment (OR = 1.014, 95% CI, 1.008-1.019) and higher HOMA-β (Homeostasis Model Assessment of β-Cell Function) levels (OR = 4.912, 95% CI, 1.480 to 16.034) were more likely to achieve successful weight reduction at follow-up of 12 months. Non-diabetic status (OR = 2.176, 95% CI, 1.242-3.812) and using semaglutide (OR = 2.138, 95% CI, 1.162-3.935) were associated with successful weight reduction at the follow-up of 6 months. Additionally, a higher percentage body fat (PBF) in both male (OR = 3.990, 95% CI, 1.118-14.246) and female (OR = 2.266, 95% CI, 1.179-4.354) was associated with successful weight reduction. The weight regain group had a higher baseline eGFR than other two groups at the 3-month follow-up especially in participants with prediabetes (p = 0.009). Moreover, the J-shaped associations of basal metabolic rate, skeletal muscle mass, abdominal and limb muscle mass with the probability of successful weight reduction within 6 months, and a positive non-linear association of serum creatinine with the probability of successful weight reduction within 12 months have been characterized.
CONCLUSION: A longer duration of GLP-1RA treatment, using semaglutide, non-diabetic status and higher PBF might be associated with better weight reduction. Basal metabolic rate, skeletal muscle mass, muscle mass of abdomen and limbs, and serum creatinine were non-linearly associated with the probability of successful weight reduction.
