Reviews/GLP-1 Receptor Agonists/Reduce peripheral arterial disease and limb events
Cardiovascular · GLP1-CV-05 · v2
Several separate studies all point the same way: patients with type 2 diabetes taking GLP-1 drugs tended to have fewer serious leg and foot problems, such as amputations and peripheral artery disease, than comparison patients [16][17][21]. The strongest single study measured these limb problems directly and found fewer amputations [16], and a very large study in people with severely threatened limbs found a lower amputation rate [21]. But all of these were observational records of real-world prescribing rather than randomised trials, so they cannot fully rule out that the two groups of patients differed in other ways. The one piece of trial-based evidence looked at heart attacks and strokes, not limb events, and its limb-population estimates were not statistically conclusive [18]. The comparisons were also against other active diabetes drugs rather than placebo, and the measured effects sat close to no-difference. The panel found the direction consistent but the foundation not yet firm enough to be sure.
The corpus contains one cohort in which major adverse limb events were the pre-specified primary composite outcome, where GLP1RA beat DPP4i with a subdistribution hazard ratio of 0.63 (95% CI 0.41, 0.96), driven by lower amputation (SHR 0.55, 95% CI 0.30, 0.99) [16]. A very large cohort of 139,173 type 2 diabetes patients with chronic limb-threatening ischaemia found a lower major-amputation risk with GLP-1RA (HR 0.745, 95% CI 0.679-0.816) [21]. A Swedish nationwide cohort reported higher peripheral artery disease risk with SGLT-2i than GLP-1RA (HR 1.68, 95% CI 1.04-2.72) [17]. A pooled analysis of LEADER and SUSTAIN-6 in patients with diabetes and PAD showed consistent MACE benefit (liraglutide HR 0.77, 95% CI 0.58-1.01; semaglutide HR 0.61, 95% CI 0.33-1.13) but measured no limb outcome [18]. The contested point: The evidence divides on whether a consistent direction across observational cohorts, with one direct MALE primary endpoint [16], is enough when no placebo-controlled randomised trial in the corpus is powered for limb events and the only RCT-derived paper [18] measures MACE. The Sceptic stresses that effect estimates hug the null ([16] primary SHR upper bound 0.96; [18] limb-population CIs crossing 1.0), while the Synthesist and Moderator weigh the agreement of four independent datasets.
All 21 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from GLP-1 Receptor Agonists, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
Every paper that actually measures a limb outcome ([16], [17], [21]) is an observational retrospective cohort, vulnerable to residual confounding and prescribing-channel bias; [16]'s authors state 'some unexplored confounding factors may exist in this observation study.' No RCT in the corpus is powered with MALE or PAD as a primary endpoint.
Comparator heterogeneity undermines the unqualified word 'reduce': [16] is versus DPP4i and [17] is versus SGLT-2i. The [17] PAD signal may reflect SGLT-2i increasing PAD risk rather than GLP-1RA lowering it, and neither establishes a placebo-anchored reduction.
Where it remains uncertain. Every limb-outcome paper is observational with an active comparator ([16] vs DPP4i, [17] vs SGLT-2i), and the only RCT-derived paper measures MACE with limb-population confidence intervals that cross 1.0 (semaglutide 0.33-1.13; liraglutide 0.58-1.01) [18].
Aleatory
Primary limb-event incidence is low (2.59 vs 4.22 events per 1000 person-years in [16]), so estimates rest on few events and confidence intervals are wide.
Epistemic
No corpus trial is powered with MALE or PAD as a primary endpoint; the one RCT-derived paper [18] measures MACE, leaving the causal limb effect untested by randomisation.
The load-bearing papers from a corpus of 21, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Direct on-claim evidence: MALE (PAD, critical limb ischaemia, PTA/bypass, amputation) was the pre-specified primary composite in a T2D cohort, significantly reduced with GLP1RA versus DPP4i.
“The incidence of primary composite outcome events was significantly lower in those treated with GLP1RAs compared with those treated with DPP4is (2.59 vs 4.22 events per 1000 person-years; subdistribution HR [SHR] 0.63 [95% CI 0.41, 0.96]), primarily due to lower rates of amputation (1.29 events per 1000 person-years for GLP1RAs vs 2.4 events per 1000 person-years for DPP4is; SHR 0.55 [95% CI 0.30, 0.99]).”
Panel auditInternal validity: medium · scope of inference: Major adverse limb events with GLP-1 RA versus DPP-4 inhibitors in type 2 diabetes.Key limitations: observational · active comparator not placebo · residual confounding acknowledged
Supports a lower PAD event rate with GLP-1RA, but only relative to an active comparator (SGLT-2i), not placebo.
“The risks of stroke and peripheral artery disease were higher in the SGLT-2i group relative to GLP-1RA, with HR 1.44 (95% CI 0.99-2.08) and 1.68 (95% CI 1.04-2.72), respectively.”
Panel auditInternal validity: medium · scope of inference: Relative PAD and cardiorenal outcomes of GLP-1 RA versus SGLT-2 inhibitors in type 2 diabetes.Key limitations: comparator is active drug not placebo · observational · short follow-up
Corroborating context that GLP-1 RAs benefit the T2D-with-PAD population, but the measured outcome is MACE, not limb events, and the PAD-subgroup CIs cross 1.0.
“The effects of both therapies on MACE were consistently beneficial in patients with PAD (liraglutide: HR 0.77, 95% CI 0.58-1.01; semaglutide: 0.61, 0.33-1.13) and without”
Panel auditInternal validity: high · scope of inference: MACE efficacy of liraglutide and semaglutide in type 2 diabetes patients with and without PAD.Key limitations: MACE outcome not limb events · subgroup analysis · wide confidence intervals
3 of 21 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
The direction is consistently supported across independent cohorts with a direct MALE endpoint, but the absence of any placebo-controlled RCT powered for limb events, active-comparator designs, and near-null confidence intervals keep confidence in the Uncertain band.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
No placebo-controlled RCT powered for limb events. Every limb signal is observational [16][17][21] and the sole RCT-derived paper [18] measures MACE, so causation for limb outcomes is untested by randomisation. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A randomised placebo-controlled trial in type 2 diabetes with MALE or PAD events as the pre-specified primary endpoint.
Open question: Do GLP-1 receptor agonists reduce major adverse limb events versus placebo when limb events are the primary endpoint? Why it is still open: The corpus contains no placebo-controlled trial powered for limb events; the only RCT-derived paper [18] measures MACE and its limb-population CIs cross 1.0.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19