When is an Achilles rupture actually ready to run?

Achilles rupture rehab has a well-worn early phase and a vague later one. Most of us can describe the first twelve weeks confidently. Far fewer can say precisely what a patient must demonstrate before they start running, and the literature has not helped, because until now there have been no evidence-based criteria to point at. The decision has been part protocol, part experience, part the patient asking often enough.

Two papers published this month address different halves of that problem. One tells you what to check. The other tells you what checking will miss.

A modified Delphi study invited 35 international Achilles experts and ran three rounds between January and June 2025, requiring 75% agreement for a criterion to pass. Sixteen candidate criteria went in. Eight reached consensus: seven for inclusion and one that the panel agreed to exclude. It was published in Sports Medicine.

The seven are plain and mostly free:

The criterion they agreed to leave out is calf circumference symmetry. Plenty of clinicians still measure it, and the panel could not agree it earned a place. Girth is a poor proxy for what a calf can actually do, and a patient can regain bulk without regaining power.

Six of the seven are pain, gait quality, balance or readiness. Exactly one is a countable performance measure, and it asks for ten repetitions.

The second paper, published in the Journal of ISAKOS, assessed 22 people approximately twelve months after a unilateral Achilles rupture. Mean age 48. It used sixteen-camera motion capture with EMG from eight muscles bilaterally, comparing the injured limb against the patient's own uninjured side during walking and running at self-selected speeds.

During walking, the affected limb showed lower ankle power in late stance, reduced knee power in early stance, reduced ankle range of motion, and a lower total support moment at both initial contact and toe-off. The compensations were visible in the muscle data. Lateral gastrocnemius activation increased in late stance, which the authors read as greater reliance on biarticular muscle-tendon energy transfer for propulsion. Soleus activation showed lower variability, a more stereotyped pattern they interpret as compensatory control in response to altered tendon compliance.

During running there were fewer differences between limbs, though ankle range of motion and the support moment at toe-off remained reduced.

The authors' conclusion is that deficits at one year were primarily localised to the distal limb, and that rehabilitation should target tendon stiffness and monoarticular plantarflexor strength.

Treat the seven criteria as a floor. They are a reasonable, defensible gate for letting someone start running, and having an agreed list is a genuine improvement on guessing. But they are a threshold for beginning, not evidence of recovery. The people in the gait study were a year out and running, and their calves were still measurably down on the other side.

Keep loading the calf long after running resumes. This is the practical consequence. If ten heel rises and no limp is where your programme stops progressing plantarflexor load, you are stopping at roughly the point the evidence says the deficit persists. The gait paper points specifically at tendon stiffness and the monoarticular plantarflexors, which is an argument for continued heavy, slow, knee-extended calf work well into the running phase.

Do not only assess at speed. Walking exposed more inter-limb difference than running did in this cohort. A patient who looks even at running pace may not be, and a slower, more controlled task can be the more revealing test. If you are only watching them run, you are watching the task that hides it.

The measurement problem is real. Neither paper gives you a clinic protocol. The consensus offers one countable test at ten repetitions. The gait study used equipment nobody has in a treatment room. Taking the heel raise past ten and counting to genuine fatigue, knee straight, tempo controlled, at least converts a pass or fail into a number you can re-test against. The other instinct is a hand-held dynamometer, and there the caveat is one we have covered before: dynamometers hold up well for hip and knee strength, but the evidence is thinner at the ankle, which is exactly the joint you need it for here. Pick one method, keep the protocol identical, and read the number as a trend.

There is now an agreed set of criteria for clearing a patient to start running after an Achilles rupture repair: no pain, no limp, tiptoe walking, ten single-leg heel rises, single-leg balance and psychological readiness, with calf circumference explicitly excluded. Use them as permission to begin, not as proof of recovery, because a year on these patients still walk with measurably less ankle power and support moment on the repaired side.

The gait study is 22 people, cross-sectional, level III, and the same research group has published closely related work on this cohort. The consensus is expert opinion, and the authors are explicit that the criteria require prospective validation before being used as decision rules. Both papers concern Achilles rupture specifically. None of this transfers to midportion Achilles tendinopathy, which is a different condition with a different evidence base.