Supervised exercise ranks top for limb salvage in PAD

Peripheral artery disease narrows the arteries supplying the legs, and it carries a double burden: painful, limiting claudication plus a substantially raised risk of heart attack, stroke and death. Clinicians have three broad tools — endovascular revascularisation to reopen vessels, medical therapy, and supervised exercise therapy — and they are often used in combination. What has been unclear is how these strategies and their combinations compare over the long term across quite different outcomes: cardiovascular and cerebrovascular events, all-cause mortality, amputation and limb perfusion. A network meta-analysis can compare options that have rarely, or never, been tested head to head.

The authors searched Chinese and English databases from inception to 8 April 2026 for randomised controlled trials comparing endovascular reconstruction/revascularisation, supervised exercise therapy, medical therapy or their combinations in adults with peripheral artery disease, with the protocol registered prospectively in PROSPERO. Risk of bias used the Cochrane RoB 2.0 tool and a frequentist network meta-analysis was run in Stata 19.0, with odds ratios for binary outcomes and mean differences for continuous ones. Treatments were ranked using the surface under the cumulative ranking curve (SUCRA), consistency was checked by node-splitting, and certainty was assessed with the CINeMA framework. Fifteen RCTs including 1,461 participants were eligible.

For all-cause mortality, revascularisation plus supervised exercise plus medical therapy had the highest SUCRA value at 99.4%, indicating the greatest probability of benefit within the evidence network. For cardiovascular events and cerebral infarction, revascularisation plus medical therapy ranked highest, at 76.7% and 85.3% respectively. For amputation, supervised exercise plus medical therapy ranked highest at 77.7%, pointing to a limb-salvage advantage, while for ankle-brachial index improvement the triple combination ranked top at 85.2%. Subgroup analysis found the apparent amputation-protective effect of revascularisation with ordinary medical therapy was larger than with best medical therapy, though the direction was consistent. Sensitivity analyses did not materially change the rankings.

This reinforces the growing position that supervised exercise is not a soft add-on in peripheral artery disease but a component that ranks competitively alongside procedures, particularly for keeping limbs. It also suggests that the best strategy depends on which outcome you prioritise, since no single approach topped every ranking. The limitations are substantial and the authors are candid about them: 1,461 participants spread across 15 trials makes for a sparse network, with limited direct comparisons, heterogeneous populations, variable reporting and possible publication bias, and overall certainty mainly moderate to low. Large, long-term head-to-head trials are still needed for different disease phenotypes.

If you or someone close to you has peripheral artery disease, this is a reason to take supervised exercise therapy seriously as part of treatment rather than as an optional extra — in this analysis it ranked highest for amputation prevention when combined with medication, and featured in the top-ranked combination for mortality and limb perfusion. It does not establish that any one strategy is superior; SUCRA rankings describe probability within a limited evidence network. Which combination suits an individual depends on disease severity, symptoms and comorbidity, and is a decision for a vascular team. The broader message is that procedures and structured exercise appear to be complementary, not alternatives.