What's Actually Going On
Blood has to leave the leg against gravity. Two things move it: one-way valves that stop backflow, and the calf muscles, which squeeze the deep veins every time you take a step. That squeeze is the calf muscle pump.
When valves leak or veins are blocked, the pressure in the lower leg veins stays high while you stand instead of dropping as you walk. That sustained pressure damages the skin from the inside through inflammation and tissue breakdown, which is what finally opens the ulcer (Raffetto 2018).
The reflux that matters is not the one you can see. Across 18 studies and 11,177 participants, severe disease was significantly linked to deep reflux (OR 2.41, 95% CI 1.53–3.78) and perforator reflux (OR 3.37, 95% CI 2.16–5.27), but not to superficial reflux (OR 2.11, 95% CI 0.87–5.14). Visible varicose veins are the least predictive part of the picture (Tan 2024).
The pump is an independent cause, not just a marker, and that is the whole reason physical therapy has a role here. Christopoulos 1989 measured 205 limbs and found a poor ejection fraction was the primary cause of ulceration in limbs with barely any reflux, while a good ejection fraction significantly reduced ulceration even in limbs with marked reflux (p<0.05). Residual volume fraction tracked ambulatory venous pressure at r=0.81. That paper is 37 years old and nothing since has replaced it. Elbenawi 2025 confirmed the principle at scale in 9,510 limbs: among legs with no other detectable venous abnormality at all, rapid post-exercise refilling still carried 5.7% against 2.7% ulcer prevalence (P=.001).
This is not simply "what happens after a clot." Objective testing of 102 limbs found 79% primary venous disease against 18% secondary, and among ulcerated limbs the split was 43% primary valve failure against 57% post-thrombotic (Kistner 1996). Roughly two in five venous ulcers arrive with no clot history at all.
How to Identify It
The one question that separates it from the dangerous lookalike: does raising the leg help or hurt? Venous symptoms ease with elevation. Arterial symptoms get worse.
This condition has no orthopaedic provocation tests. Its tests are vascular measurements, and the honest reading of them is the most safety-critical thing on this page.
- Ankle-brachial pressure index Sn: 63.5% | Sp: 89.3% — the gate that decides whether compression is safe. Below 0.9 rules arterial disease in (PLR 6.5). Between 0.9 and 1.3 it does not rule it out (NLR 0.31).
- Tibial arterial waveform, triphasic Sn: 82.8% | Sp: 86.8% — the best rule-out retrieved (NLR 0.09–0.28), but from only 2 studies.
- Toe-brachial index above 0.75 Sn: 83.0% | Sp: 66.3% — rules out better than the ankle index (NLR 0.14–0.24), and usable when ankle vessels are calcified.
- Feeling for foot pulses — cannot exclude arterial disease (NLR 0.75). Their absence tells you something; their presence does not.
- Duplex ultrasound — maps deep, superficial and perforator reflux, and decides whether an intervention is possible.
A base-rate warning that is not decorative. Every accuracy figure above comes from populations with diabetes, tested against reference imaging inside vascular or diabetic foot services. No study reports how any of these tests perform in an ordinary first-contact caseload, and Forsythe 2020's own verdict on the whole body of evidence is that its quality is poor and there is insufficient evidence to prefer any one bedside test. Use the numbers to understand that the gate leaks. Do not quote them as this population's performance.
What you look for besides: an ulcer in the gaiter area, usually on the inner ankle, shallow with an irregular border and heavy fluid; brown haemosiderin staining; hard, tight, inverted-champagne-bottle skin; swelling that reduces overnight and returns through the day; a small, stiff calf and reduced ankle range in long-standing disease.
The Debate
Does exercise heal a venous leg ulcer?
Jull 2018 (5 trials, N=190) · Turner 2023 (7 trials) · Pagani 2026 (9 trials)
Yes. An additional 14 healed per 100 (95% CI 1–27, P=.04); RR 1.38 (95% CI 1.11–1.71); RR 1.35.
vs
Smith 2018 (3 trials, N=116), published the same year, on the same literature
Possibly not. RR 1.14 (95% CI 0.71–1.84), described by its authors as "possibly no difference", low-quality evidence.
The disagreement is an inclusion boundary, not new data. Smith pooled only progressive resistance programmes against compression alone; the others pooled every exercise type. The trial set is too small to survive a change of definition. Underneath all four sits the thing that matters most: Turner's own authors state that most of the trials failed to show the exercise arm was any more active than the control arm, with compliance running from 33% to 81%. A pooled benefit is being credited to an exposure nobody verified. Prescribe the exercise, and do not promise a number.
Does treating the veins early stop the ulcer coming back?
EVRA extended phase, recurrence rate (Gohel 2020)
Yes. 0.11 against 0.16 recurrences per person-year, IRR 0.658 (95% CI 0.480–0.898), P=.003.
vs
EVRA extended phase, time to first recurrence — the same trial, the same patients
No clear difference. HR 0.82 (95% CI 0.57–1.17), P=.28.
Two ways of measuring one thing, opposite verdicts, inside a single trial. Time-to-first-event throws away every recurrence after the first; a rate over person-years counts them all. In a relapsing condition where 121 of 426 healed participants (28.4%) recurred at least once, those are genuinely different questions. Quoting one without the other is a choice, not a summary.
Stockings or bandages?
Amsler 2009 (8 trials, 688 legs)
Stockings, clearly. 62.7% healed against 46.6%, P<.00001, and 3 weeks faster.
vs
Patton 2023 meta-review · Mauck 2014
No difference. RR 0.95 (95% CI 0.87–1.03), P=.18, and no overall difference in healing, time to healing or recurrence.
Amsler's own abstract records that baseline patient and ulcer characteristics favoured the stocking arm in 4 of its 8 studies. Two later independent syntheses did not reproduce the advantage. The narrower signal that survives is that stockings beat short-stretch bandages specifically. Decide on who can apply it and who will keep wearing it.
Guideline recency: the most widely cited UK guideline for this pathway, NICE CG168, dates from 2013 and is 13 years old. The current documents are the SCAI Chronic Venous Disease guideline (2025) and the Journal of Wound Care compression consensus (2023), and the field is actively moving on who gets high versus moderate compression, when to intervene, and how to handle mixed arterial and venous disease.
Honest Limitations
The exercise effect rests on an exposure nobody measured
What the research shows: pooled healing benefit of RR 1.38 (95% CI 1.11–1.71) for exercise added to compression.
The real-world gap: the trials could not show their exercise arms were more active than their control arms, and compliance ran 33% to 81%. O'Brien 2017 shows what that does to an estimate: intention-to-treat healing was 77% against 53% and not statistically significant, while looking only at the 19 people who stuck to it 75% of the time gave 95% healed (P=0.01). Analysing only the adherent people in an unblinded behavioural trial selects for those who were going to do well anyway.
What to do about it: prescribe the exercise, expect less than the pooled figure, and treat adherence itself as the intervention.
The evidence covers the patient you see least
What the research shows: everything with a number attached comes from an active, open ulcer in a leg with an ankle index above 0.8.
The real-world gap: the commonest presentation is the aching, heavy, swollen leg with skin changes and no open wound. The Cochrane review covering that group found 5 trials, 146 people, and could not pool a single outcome because the trials all measured different things (Araujo 2023). Mixed arterial and venous disease has no randomised trials at all (Alagha 2025).
What to do about it: say the uncertainty out loud when treating a leg without an ulcer. The evidence in this condition is densest where the caseload is thinnest.
Compression hurts, and that pain blocks the exercise prescribed beside it
What the research shows: compression is first-line and exercise is the adjunct, so patients get both.
The real-world gap: a synthesis of 18 qualitative studies found the barriers were mostly not about motivation. They were not understanding that the condition is chronic or why activity was being asked for, beliefs about one's own capability, pain from the wound and from the compression itself, what the clinician did or did not explain, and fear and embarrassment (Qiu 2022).
What to do about it: explain the chronicity and the reason for the movement explicitly, because failing to is itself one of the documented barriers, and time the exercise away from the most painful part of the compression cycle.
There is no agreed number for "meaningfully better"
What the research shows: a wide range of outcome measures, from proportion healed to ulcer area to disease-specific quality of life.
The real-world gap: zero of the 106 papers retrieved for this review report a minimum clinically important difference for anything. That has a documented cause rather than being an accident: McNally 2023 is a systematic review conducted expressly because no core outcome set exists for this condition, and Launois 2015 found 10 competing disease-specific quality-of-life scales across 103 papers.
What to do about it: track change, and resist importing a threshold from a different condition and calling it validated here.
The Nuance
Three in four ulcers heal without any procedure. In EVRA's compression-only arm, 76.3% had healed at 24 weeks with a median of 82 days. In an independent cohort, 62 of 90 ulcers (69%) healed within 12 weeks on standardised care plus compression.
Adding early vein treatment buys about 26 days at the median, taking 24-week healing from 76.3% to 85.6%, and it was 91.6% likely to be cost-effective at £20,000 per quality-adjusted life year over three years. It is not a cure, because neither path repairs the valves.
The number to sit with is recurrence. 121 of 426 healed participants, 28.4%, had at least one ulcer return during follow-up, and that was in a trial population selected for being early and treatable. Compression after healing is not optional aftercare. It is the treatment continuing.
The direction here is the opposite of most conditions on this site, and it is worth saying plainly. Most pages end by restricting what you may do. Here the prescribed treatment is loaded calf work. Someone who already lifts is, on this evidence, already doing more than the exercise dose that was ever tested, and for them the useful levers are the compression and the arterial check rather than more exercise. For someone sedentary, the exercise is where the whole gap sits.
One gap worth naming honestly. The arterial check that gates every recommendation on this page is a test whose sensitivity is 63.5%. That means roughly one in three cases of the disease it screens for is missed. It is still the right first step, and it is still not sufficient on its own, which is why the toe-brachial index and the tibial waveform belong in the assessment when the ankle index is borderline or the ankles are calcified.
