The Soleus Is Not a Smaller Gastrocnemius: A Clinician’s Case for Rewriting Calf Rehabilitation

A middle-distance runner walks into clinic with three months of mid-calf pain. She has been managed elsewhere as “chronic gastrocnemius strain”, progressed through eccentric heel drops off a step, cleared for a graded return at six weeks, broken down twice. Her single-leg straight-knee calf raise is unrestricted and pain-free. Her bent-knee raise reproduces a familiar deep ache and fails at thirteen reps. Palpation of the mid-posterior calf, two finger-widths medial to the gastrocnemius muscle belly and a fingerbreadth deeper, is exquisitely tender. The MRI request comes back with a low-grade strain of the soleus at the medial musculotendinous junction. None of this is unusual. All of it is preventable.

The soleus and the gastrocnemius are not interchangeable. They are not a single muscle with two heads. They share a tendon, and that is approximately where their similarity ends. The Brukner & Khan tradition has long insisted that the calf be examined as two muscles. The injury data, the architectural data, and the in vivo force data all now point in the same direction — and yet the dominant rehabilitation language remains “calf strengthening” and the dominant exercise prescription remains the straight-knee heel-drop. “Calf raises” is not an exercise prescription. It is a request for clarification.

This piece makes the case that the soleus is the underweighted muscle in calf injury practice. The case rests on four converging lines of evidence: architecture, fibre type, force contribution during gait, and injury epidemiology. It ends with three practice patterns worth changing.

The architectural argument

Human muscle architecture predicts function with uncomfortable precision. Fibre length tells you about velocity. Physiological cross-sectional area tells you about force. Pennation angle tells you how the two are traded off. The soleus sits at one extreme of the human lower limb on all three.

Bolsterlee and colleagues’ three-dimensional in vivo mapping of the whole human soleus puts the muscle volume on the order of 425 cm-cubed, the fascicles at 3–4 cm, and the superficial pennation angles above 50 degrees — the highest PCSA in the lower limb, packaged onto the shortest fascicles, at the steepest pennation (Bolsterlee et al., PeerJ 2018). Read literally, this is a muscle built to produce very high force at very low shortening velocity. Walking is a very low shortening-velocity activity. So is slow running. So is most of standing.

The gastrocnemius is the opposite design. Longer fascicles. Smaller PCSA. Biarticular, so its length is partly hostage to knee position. Its architecture suits ballistic, high-velocity output — sprinting, jumping, the late swing-to-stance transition at speed. When the knee is flexed, the gastrocnemius slackens and its capacity to generate plantarflexion torque drops sharply. The soleus, monoarticular at the ankle, is indifferent to knee angle. This is the basis of the bent-knee plantarflexion test, and it is the basis of the bent-knee calf raise as a soleus-loading exercise.

If you take these architectural facts seriously, the soleus is not a muscle you train with three sets of ten heavy heel raises. It is a muscle you train with sustained submaximal loading at volumes the gastrocnemius would tire of.

The fibre-type argument

The histochemical work converges. Edgerton, Johnson, Smerdu and colleagues put the soleus fibre composition at approximately 70–80% slow-twitch type I — the highest type I dominance of any major lower-limb muscle, with individual variability but a clear and reproducible postural-muscle profile (Edgerton et al., Histochem J 1975; Johnson et al., J Neurol Sci 1973). The gastrocnemius sits closer to 50% type I. This is the fibre-type signature of a muscle that is on, in the background, every minute you spend upright.

The clinical implication is dose. The soleus’s training response is dose-driven. It needs reps. It needs time under tension. It does not respond to a few sets of heavy doubles the way the gastrocnemius does. This matters enormously in rehabilitation, because the load required to drive a hypertrophic and tendon-remodelling stimulus in a predominantly slow-twitch muscle sits closer to 25–30 repetitions of bodyweight calf raise to failure than to 3 x 8 heavy — and most rehab programs prescribe neither. They prescribe three sets of fifteen straight-knee heel drops and call it calf rehab.

The force-contribution argument

Hicks’s 1954 work on the foot and the windlass established the architecture of triceps surae loading during stance. The modern in vivo Achilles literature has since quantified what was always implicit: the soleus is the dominant force generator across the Achilles tendon during walking and slow-to-moderate running, with peak output in late stance as the centre of mass passes over the metatarsal heads and the ankle plantarflexes (Lichtwark & Wilson, J Exp Biol 2007; Lai et al., Sci Rep 2021). The gastrocnemius contributes proportionally more at higher running velocities and in true sprint mechanics, but for the gait pattern that most athletes spend most of their volume in, the soleus is the engine.

The numbers are not subtle. Peak Achilles tendon force during walking sits at 2.5–3.5 times body weight; in moderate running it reaches 6–8 times body weight; and most of that load is generated on the soleus side of the equation (Lai et al., Sci Rep 2021). This is the muscle that bears the running load. It is also the muscle that gets the least targeted rehabilitation.

The injury epidemiology

If you read one paper from the calf-injury literature, read Green et al. 2020. Sixteen AFL clubs, multiple seasons of soft tissue injury registry data, MRI-confirmed calf injuries with site-of-injury detail and return-to-play tracking. The headline finding belongs on the practice wall: 84.6% of MRI-confirmed calf muscle strains in elite Australian Football were soleus injuries. Gastrocnemius injuries were a clear minority. Soleus injuries took longer to resolve — 25.4 ± 16.2 days vs 19.1 ± 14.1 days for gastrocnemius in this elite cohort — and the soleus reinjury rate was substantial enough to drive a separate strand of Green and Pizzari’s subsequent recurrence work.

Pedret’s 2015 cohort of soleus strains examined the site of injury within the muscle and found the central aponeurosis as the worst-prognosis site, averaging 44 days to return to play versus 19 days for lateral aponeurotic lesions (Pedret et al., Orthop J Sports Med 2015). The radiographic anatomy is now well-mapped — the Balius–Pedret cadaver and MRI work (Skeletal Radiol 2013) is the cleanest synthesis — and the central tendon, anterior aponeurosis, and proximal medial musculotendinous junction are the three high-yield sites to interrogate on MRI.

Green & Pizzari’s 2017 systematic review identified age and previous calf strain as the two best-supported risk factors for calf injury, holding up across the literature where other proposed risk factors did not (Green & Pizzari, Br J Sports Med 2017). The implication for the misclassified patient: a “calf strain” that was managed as gastrocnemius and then recurred should be treated as suspected unresolved soleus pathology until proven otherwise.

The picture is consistent. The soleus is the most commonly injured calf muscle. It heals more slowly. It reinjures more often. It is harder to image well on ultrasound because it sits deep to gastrocnemius and is multipennate and highly vascular. It is harder to examine without deliberate bent-knee testing. And it is systematically underloaded by the standard heel-drop protocol that most patients are sent home with.

Examining the soleus, specifically

Two findings belong in every assessment template.

First, the bent-knee single-leg calf raise. Repetition to fatigue. Standardised position. Dominant vs non-dominant comparison. A reduction of more than 20% on the affected side is meaningful. A reproduction of pain in the deep mid-calf with the knee flexed, alongside a pain-free straight-knee raise, is close to pathognomonic for soleus involvement.

Second, deep palpation. The soleus belly lies medial and deep to the medial head of gastrocnemius across the mid-to-distal third of the calf. Effective palpation requires the patient prone with the knee passively flexed to 60–90 degrees to relax the gastrocnemius and expose the deeper muscle. Tenderness here, with a clean superficial gastrocnemius, narrows the diagnosis quickly.

MRI remains the gold standard for grading and is more useful than ultrasound in the deeper soleus tissue. The Pedret site-based classification — proximal medial MTJ, proximal lateral MTJ, distal central tendon, anterior myofascial, posterior myofascial — predicts return-to-play better than a simple Grade 1–3 system. Central aponeurotic injuries are the ones to flag as long-rehabilitation.

Three practice patterns to change

Stop prescribing the straight-knee heel drop as soleus rehab. It is a gastrocnemius-dominant exercise. The straight-knee position elongates the gastrocnemius across the knee and ankle and loads it preferentially; the soleus is recruited but is not the limiting tissue. For a known or suspected soleus injury, the loading exercise is the bent-knee calf raise — single-leg, knee flexed to 60–90 degrees, full ankle range, repetitions in the 20–30 zone to failure with progressive external load. This is the loading position that matches the soleus’s architecture and its fibre-type biology.

Set a longer rehab horizon. Soleus strains commonly take 8–14+ weeks to fully resolve in athletic populations, with central aponeurotic and high-grade musculotendinous junction injuries at the long end. The AFL data and the Pedret cohort both bear this out. A patient with a confirmed soleus strain who is “feeling fine” at four weeks is not ready to run, and the strongest predictor of reinjury is premature return.

Examine the bent-knee position. Build it into the assessment template. The single-leg straight-knee raise is a gastrocnemius test masquerading as a calf test. If you only test one calf raise, test the bent-knee version.

The Sleeve as a soleus loading-dose adjunct

Everything this article has built — the architectural case, the fibre-type biology, the in vivo force data — converges on a single clinical observation about dose. The soleus is a slow-twitch, high-PCSA, short-fascicle postural muscle that generates peak force at low shortening velocity. Walking is exactly that activity. Standing is exactly that activity. The patient who is rehabbing a soleus injury spends three thirty-minute sessions per week with you in the clinic doing bent-knee loading at the dose the tissue needs to remodel. They then spend the other one hundred and sixty-six and a half hours of their week walking, standing, climbing stairs, and otherwise putting low-velocity, high-volume, soleus-dominant load through tissue that has not finished healing. That is the gap. The standard rehab plan loads the soleus in clinic and then sends the patient back out to walk and re-load it all day.

There is now a measurable intervention for that gap. The University of Queensland’s June 2025 biomechanical evaluation of the Sleeve tested the device on healthy gait with EMG instrumentation of the lateral gastrocnemius, medial gastrocnemius and soleus, alongside Achilles tendon force modelling. The findings worth knowing — and worth distinguishing carefully from one another — are these. Across the cohort the device produced a statistically significant 32% reduction in medial gastrocnemius EMG during standing balance (p=0.002) and an 8.1% reduction in peak Achilles tendon force during walking. At the individual level, the maximum reductions observed during the late-stance phase of walking were 47.8% in lateral gastrocnemius EMG, 21.9% in medial gastrocnemius EMG, and 20.4% in soleus EMG (UQ Final Report §3.2.5). The group-level standing balance number and the individual late-stance walking maxima are two different things, and the literature on rehab adjuncts is full of conflations of exactly that kind — flag the distinction when you describe this to colleagues.

The 20.4% soleus figure is the one that matters for the soleus rehab patient. Late stance is when the soleus generates peak output across the Achilles tendon during gait. It is the phase the healing soleus least tolerates re-loading in. A device that demonstrably reduces soleus EMG at that specific phase of walking is the first off-the-shelf option for the daily ambulatory dose problem that the bent-knee calf raise cannot solve, because the bent-knee calf raise is a clinic-based intervention and the problem is a what-the-patient-does-between-clinic-visits problem.

How to position it in practice. The Sleeve is not a substitute for the bent-knee loading programme, the longer rehab timeline, or the graded return-to-running progression. It is the adjunct that takes the daily ambulatory load off the deep calf during the return-to-walking and early return-to-running window, so the in-clinic loading dose can actually translate into tissue remodelling rather than being undone by the next eight thousand steps. The clinical edge it gives you is for the soleus rehab patient who has done the bent-knee work, ticked the return-to-run criteria, and is still plateauing or re-aggravating. That is the patient for whom the daily walking dose is the variable that has not been controlled. The honest scope: temporary, adjunctive, evidence-anchored, and bounded by the biomechanical data. The treatment remains progressive bent-knee loading, in dose, over months.

The soleus is the muscle our profession has spent decades calling “the calf” without quite knowing what we meant. The architecture, the fibre type, the in vivo force data, and the injury registry data all now say the same thing. The soleus is the dominant Achilles load-bearer in the gait pattern most athletes spend most of their time in. It is the most commonly injured calf muscle. It is the one we are most likely to underload in rehabilitation. The fix does not require new technology — it requires bent knees, longer time horizons, and a willingness to stop saying “calf raises” without specifying which.

Key references

1. Pedret C, Rodas G, Balius R, Capdevila L, Bossy M, Vernooij RWM, Alomar X. Return to play after soleus muscle injuries. Orthop J Sports Med. 2015;3(7):2325967115595802.
2. Green B, Lin M, McClelland JA, Semciw AI, Schache AG, Rotstein AH, Cook J, Pizzari T. Return to play and recurrence after calf muscle strain injuries in elite Australian Football Players. Am J Sports Med. 2020;48(13):3306-3315.
3. Green B, Pizzari T. Calf muscle strain injuries in sport: a systematic review of risk factors for injury. Br J Sports Med. 2017;51(16):1189-1194.
4. Edgerton VR, Smith JL, Simpson DR. Muscle fibre type populations of human leg muscles. Histochem J. 1975;7(3):259-266.
5. Bolsterlee B, Finni T, D’Souza A, Eguchi J, Clarke EC, Herbert RD. Three-dimensional architecture of the whole human soleus muscle in vivo. PeerJ. 2018;6:e4610.
6. Lai AKM, Dick TJM, Biewener AA, Wakeling JM. Quantifying mechanical loading and elastic strain energy of the human Achilles tendon during walking and running. Sci Rep. 2021;11:5418.
7. Balius R, Alomar X, Rodas G, Miguel-Pérez M, Pedret C, Dobado MC, Blasi J, Koulouris G. The soleus muscle: MRI, anatomic and histologic findings in cadavers with clinical correlation of strain injury distribution. Skeletal Radiol. 2013;42(4):521-530.