Skeletal effects of GLP-1 receptor agonist therapy in adults with overweight or obesity: a systematic review
- Design
- Systematic review · 493 participants · Mixed outcome
- Match to healthy normal-weight adults aged 55–75
- Different populationINDIRECT: the eight trials enrolled adults aged ≥18 years with study-level mean baseline BMI > 25 kg/m², five of them in adults with type 2 diabetes, and no participant ages are reported.
- Could weight loss explain it?
- PossiblyThe outcomes are bone mineral density and bone turnover rather than weight itself, and the review reports the weight loss in each trial alongside the bone results, noting that the two largest weight losses came with reduced lumbar spine and total hip BMD at closely similar magnitudes. That is a comparison across trials and doses, not a participant-level mediation analysis, a weight-matched comparison or a measurement taken before weight loss, so weight change cannot be separated from drug effect here.
- Study tier
- Not ratedNOT_RATED: the review synthesised its eight trials narratively and performed no meta-analysis, so it contributes no new pooled estimate. Its own certainty assessment was very low for every agent, and the authors conclude that the randomized evidence is too sparse and too heterogeneous to determine skeletal effects.
- Assessment
- Version 3 · Two independent AI reviews agreed · Nov 6, 2026
Study facts come from the paper. Population match, weight-loss explanation and study tier are our judgments, made against the reference group of healthy normal-weight adults aged 55–75.
This systematic review collected eight randomized trials (n = 493) of GLP-1 receptor agonists in adults with overweight or obesity that measured bone mineral density by DXA scan, and reported each drug separately rather than pooling them. The single semaglutide trial (n = 64), which produced the largest weight loss, reported reduced bone density at the lumbar spine and whole body and increased bone resorption, and liraglutide at the weight-management dose of 3.0 mg daily showed similar reductions at the spine and hip; liraglutide at the 1.8 mg diabetes dose changed nothing materially, and the five exenatide trials disagreed with each other. Certainty was very low for every drug, no trial measured fractures, and the authors conclude the evidence is too sparse to determine skeletal effects.
01Findings
What the study reported
- Drugs
- Semaglutide, Dulaglutide, Liraglutide, Exenatide
- Dose
- liraglutide 3.0 mg daily (weight-management dose) and 1.8 mg daily (diabetes dose); doses not stated for semaglutide, dulaglutide or exenatide
- Comparator
- placebo or another antidiabetic therapy; four trials randomized an insulin comparator arm
- Treatment duration
- ≥12 weeks (eligibility requirement)
- Primary outcome
- bone mineral density measured by dual-energy X-ray absorptiometry, and bone turnover markers
- Effect
- Semaglutide (Hansen 2024, n = 64, the largest weight loss at ETD - 6.8 kg vs placebo): lumbar spine BMD ETD - 0.018 g/cm², whole-body BMD MD -0.02 g/cm², C-terminal telopeptide MD 166.50 pg/mL. Liraglutide 3.0 mg (Jensen 2024): lumbar spine MD - 0.019 g/cm², total hip MD - 0.013 g/cm². Liraglutide 1.8 mg (Hygum 2020): lumbar spine MD 0.00 g/cm², no material change. Exenatide trials inconsistent, from 0.03 g/cm² to -0.04 g/cm² at the lumbar spine. Certainty of evidence very low for every agent
- 95% confidence interval
- 95% CI - 0.03 to -0.01 (whole-body BMD MD -0.02 g/cm², semaglutide trial)
- P value
- p = 0.007 (lumbar spine BMD, semaglutide trial)
- Follow-up
- Not stated
- Adverse events
- The skeletal findings are the safety signal: reduced lumbar spine and total hip BMD in the two trials with the largest weight loss and increased bone resorption (CTX) in the semaglutide trial, against no material change at the liraglutide diabetes dose; no fracture outcomes and no other adverse events reported
- Limitations
- The authors state that the randomized evidence is too sparse and too clinically heterogeneous to determine skeletal effects, that each agent has been evaluated in at most five small trials, mostly in adults with type 2 diabetes rather than obesity alone, and that certainty is very low for every agent. No trial reported fracture endpoints, no pooling was possible, and five of the eight trials raised risk-of-bias concerns.
Who was studied
- Age min
- 18
- Bmi min
- 25
- Obesity status
- mixed (study-level mean BMI: five trials ≥30 kg/m², three 25 to <30 kg/m²)
- Diabetes status
- mixed (five of eight trials enrolled adults with type 2 diabetes, three enrolled adults without diabetes)
- Cvd status
- not reported in abstract
- Ckd status
- not reported in abstract
- Baseline condition
- overweight or obesity, with type 2 diabetes in five of the eight trials
- Sample size
- 493
- Inclusion exclusion
- adults ≥18 years, mean baseline BMI > 25 kg/m², GLP-1 RA vs placebo or antidiabetic therapy for ≥12 weeks, DXA bone mineral density
Study quality details
- Study design
- Systematic review
- Sample size
- 493
- Randomization
- yes
- Blinding
- not stated
- Comparator
- placebo or another antidiabetic therapy; four trials randomized an insulin comparator arm
- Follow up duration
- not stated
- Outcome type
- mixed
- Replication
- not assessed
- Consistency with other evidence
- not assessed
- Population applicability
- INDIRECT
- Statistical precision
- per-trial 95% CIs reported; no pooled estimate; eight trials, n = 493; certainty of evidence very low for every agent
- Risk of bias
- not assessed
- Peer review status
- yes
02Funding
Funding and conflicts
- Funding
- not reported in abstract
- Industry funded
- Unclear
- Manufacturer
- not stated in source
- Sponsor role
- not reported in abstract
- Author conflicts
- The four named authors declare no conflict of interest.
- Independent replication
- unknown
Funding is shown on every study and never used to score it.
04Source
The source, as retrieved
Abstract
[UNLABELLED] Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are widely used for weight management, yet their skeletal effects remain unclear. This systematic review of eight randomized controlled trials (n = 493) evaluating dual-energy X-ray absorptiometry (DXA)-derived bone mineral density reports findings separately for each agent, without pooling, because the trials evaluated four pharmacologically distinct agents differing markedly in weight-lowering potency. The evidence is sparse for every agent and is insufficient to establish the skeletal effects of GLP-1 RA therapy in this population. [PURPOSE] Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are widely used for weight management, yet their skeletal effects remain uncertain. We systematically reviewed the randomized evidence on bone mineral density (BMD) and bone turnover markers in adults with overweight or obesity, reporting findings separately for each individual agent. [METHODS] This PRISMA 2020-compliant systematic review (INPLASY202630076) searched PubMed, Embase, Cochrane CENTRAL, and Google Scholar (January 2015 to July 2026). Eligible randomized controlled trials (RCTs) enrolled adults aged ≥18 years with mean baseline body mass index (BMI) > 25 kg/m², compared a GLP-1 RA with placebo or another antidiabetic therapy for ≥12 weeks, and reported BMD assessed by dual-energy X-ray absorptiometry (DXA). Risk of bias was assessed using the Cochrane Risk of Bias 2 (RoB 2) tool and certainty of evidence using GRADE. Because the included trials evaluated pharmacologically distinct agents that differ substantially in weight-lowering potency, results were synthesized narratively and reported separately by agent; no meta-analysis was performed. [RESULTS] Eight RCTs (n = 493) were included. Three trials were at low risk of bias (Eriksson 2019; Hygum 2020; Hansen 2024) and five raised some concerns, principally regarding the randomization process and selective reporting; no agent is supported by more than one trial at low risk of bias. (Cai 2021 contributes two treatment arms assessed separately, giving nine comparisons in the risk-of-bias assessment.) Evidence was unevenly distributed across agents: semaglutide and dulaglutide were each evaluated in one trial, liraglutide in two, and exenatide in five. Five of the eight trials enrolled adults with type 2 diabetes, including all four that randomized an insulin comparator arm; three enrolled adults without diabetes, of which one was defined by obesity alone. By study-level mean baseline BMI, five trials lay in the obese range ( ≥ 30 kg/m²) and three in the overweight range (25 to <30 kg/m²). The single semaglutide trial (Hansen 2024; n = 64; low risk of bias), which produced the largest weight loss of any trial (estimated treatment difference [ETD] -6.8 kg vs placebo), reported reduced lumbar spine BMD (ETD - 0.018 g/cm², p = 0.007), reduced whole-body BMD (mean difference [MD] -0.02 g/cm²; 95% CI - 0.03 to -0.01), and increased C-terminal telopeptide (CTX; MD 166.50 pg/mL; 95% CI 66.60 to 266.40). Liraglutide was evaluated at two doses with divergent results. At the weight-management dose of 3.0 mg daily, in the only trial conducted in adults with obesity and without diabetes (Jensen 2024; ETD - 6.7 kg vs placebo), lumbar spine BMD (MD - 0.019 g/cm²; 95% CI - 0.034 to -0.004) and total hip BMD (MD - 0.013 g/cm²; 95% CI - 0.024 to -0.002) were reduced. At the diabetes dose of 1.8 mg daily (Hygum 2020; -3.8 kg), no site changed materially (lumbar spine MD 0.00 g/cm²; 95% CI - 0.01 to 0.01). Exenatide trials were inconsistent, ranging from 0.03 g/cm² (95% CI 0.02 to 0.03) at the lumbar spine (Renyuan 2015) to -0.04 g/cm² (95% CI - 0.15 to 0.07; Akyay 2023). Certainty of evidence was very low for every agent. [CONCLUSION] The randomized evidence is too sparse and too clinically heterogeneous to determine the skeletal effects of GLP-1 RA therapy in adults with overweight or obesity. Each agent has been evaluated in at most five small trials, most conducted in adults with type 2 diabetes rather than in adults with obesity alone. The two trials producing the largest weight loss, one of semaglutide and one of liraglutide 3.0 mg, both reported reduced lumbar spine and total hip BMD at closely similar magnitudes of weight loss, and the semaglutide trial additionally reported increased bone resorption. This convergence warrants dedicated evaluation rather than extrapolation across the drug class. Adequately powered, agent-specific trials with fracture endpoints are required. [PROTOCOL REGISTRATION] INPLASY202630076.
Where this record came from
| Source | Retrieved | Identifier |
|---|---|---|
| pubmed | Oct 25, 2026 | 42920024 first ingestion |
| pubmed | Oct 25, 2026 | 42920024 duplicate matched on doi |
| pubmed | Nov 1, 2026 | 42920024 duplicate matched on doi |
| pubmed | Nov 1, 2026 | 42920024 duplicate matched on doi |
| owner-review | Nov 6, 2026 | reviewed_by set to Carrie Schluter (2026-11-05) on the owner's own confirmation. The assessment is unchanged and still the auto-classifier's, so its extracted values stay withheld under the 2026-10-18 decision. |
| reextract | Nov 6, 2026 | dose: '3.0 mg daily' -> '3.0 mg'. Q-043: manufacturer and drug_details.dose re-extracted by auto-classifier-v1.2. manufacturer no longer reads companies named only in the authors' conflicts statement (IN-062) and refuses a multi-company roster; dose is tied to the GLP-1 agent by name, never reads a concentration, and records a multi-dose design whole. Every other value is carried forward unchanged. Owner decision IN-064 option 1, 2026-11-06. |