REF-628 · Silbernagel, Thomeé, Eriksson & Karlsson, 2007 · The American Journal of Sports Medicine, · Paper · PubMed
Continuing to run/jump under a pain-monitoring model during Achilles tendinopathy rehab did NOT compromise recovery
Complete rest is unnecessary
Protocol
Randomized clinical controlled trial: thirty-eight patients with Achilles tendinopathy randomly allocated to 2 treatment groups, n = 19 each (from the abstract).
Exercise training group: allowed to continue Achilles tendon-loading activity such as running and jumping, using a pain-monitoring model; active rest group: had to stop such activities during the first 6 weeks.
All patients followed an identical rehabilitation program that gradually increases the load on the Achilles tendon and calf muscle.
Primary outcomes: VISA-A-S questionnaire and pain level during tendon-loading activity, evaluated through a 12-month follow-up.
Key Findings
n=38 patients with Achilles tendinopathy randomized to continued activity (n=19; allowed to keep running/jumping under a pain-monitoring model) vs active rest (n=19; no Achilles-loading activity for the first 6 weeks); both followed the same tendon-loading rehabilitation program that gradually increases load on the Achilles and calf.
Primary outcomes: VISA-A-S questionnaire and pain during tendon-loading activity.
No significant difference in the rate of improvement between groups; both improved significantly (P < .01) at every evaluation. VISA-A-S: continued-activity group 57 (15.8) at baseline → 85 (12.7) at 12 months; active-rest group 57 (15.7) → 91 (8.2).
Authors' conclusion: no negative effects could be demonstrated from continuing pain-monitored tendon-loading activity during treatment; it "might represent a valuable option".
Pain-monitoring rule used in the trial (full text; not checked against the abstract): pain during activity acceptable up to 5 on a 0-10 scale, must not increase week to week, and must settle by the next morning; otherwise reduce load.
Practical Application
Athletes with Achilles tendinopathy do not need to stop running: in a 38-patient RCT, those who kept running and jumping under a pain-monitoring model improved as much as those who rested from tendon-loading activity for 6 weeks (VISA-A-S 57 → 85 vs 57 → 91 at 12 months, both P < .01, no significant difference in rate of improvement) — keep them training through tendon rehab, with the load governed by pain, not by a rest period.
The governing rule is the pain-monitoring model (full text; not checked against the abstract): pain during and after loading acceptable up to 5/10, not building week to week, back to baseline by next morning. Not from this paper; the base's own practice: many clinicians default to a more conservative ≤3/10, which is reasonable.
Make it a habit: rate tendon pain during the session and check next-morning stiffness — a worse morning = load too high → reduce the last increment, don't stop entirely (the trial's rehab program itself was a gradual load increase).
Inference, not tested: "no pain at all" is the wrong target for tendon rehab and usually means under-loading; this trial governs progression speed, a strengthening protocol (e.g. heavy slow resistance) builds the capacity.
Limitations
Modest sample (n=38); Achilles tendinopathy patients, not a defined athlete tier. The active-rest group's 12-month VISA-A-S was numerically higher (91 vs 85); the trial reports no significant difference in rate of improvement, not superiority of continued activity.
The 5/10 threshold is the trial's pain-monitoring model (full text; not checked against the abstract), not a tested cut-off (hence the base's own conservative ≤3/10); requires an honest self-monitoring athlete — red-flag or non-load pain still needs referral, not "monitoring"; never licenses training through worsening joint/tendon pain (Blocks red line).
Rest arm rested only for the first 6 weeks; the trial does not compare against longer rest.
How this entry was checked and changed
Why this level
regraded 2026-09-27 from A: parallel RCT, n=38
Content correction 2026-09-27
Pain-monitoring thresholds (≤5/10, next morning, week to week) -> labelled full text; the abstract says only "pain-monitoring model". "≤3/10 conservative default" -> labelled the base's own practice.
Added the abstract's numbers: VISA-A-S 57 → 85 (continued activity) vs 57 → 91 (active rest) at 12 months, both P < .01, no significant difference in rate of improvement; noted the rest group's numerically higher score.
"mid-portion Achilles" -> deleted (not in the abstract); Sport "triathlon, court/jumping" -> running (the abstract's population is running and jumping activities).
2026-09-28: added Protocol from the abstract (PubMed PMID 17307888, DOI 10.1177/0363546506298279).