Are plantarflexor muscle impairments present among individuals with Achilles tendinopathy and do they change with exercise? A systematic review with meta-analysis

Key points

1. 01Moderate evidence supports a small but real reduction in maximal plantarflexor torque on the affected side (9-13% below the unaffected side)
2. 02Evidence for deficits in explosive strength, power, and endurance compared with the unaffected side is conflicting across studies
3. 03When comparing people with Achilles tendinopathy to healthy controls, most plantarflexor measures showed no consistent difference
4. 04Endurance improved with 12+ weeks of resistance training, but torque and power did not improve consistently
5. 05Pain and function improved substantially with exercise (20-95% at 12 weeks) while muscle function gains were more modest (strength 14.9-19.2%, endurance 0.8-33.3%, power 9.4-25.6%)

How it was conducted

Design

Systematic review with meta-analysis (PRISMA)

Databases searched

Cochrane CENTRAL, Ovid (MEDLINE, EMBASE, AMED), EBSCO (CINAHL Plus, SPORTDiscus) to September 2020

Included studies

25 studies (15 addressing muscle impairment, 14 addressing change with exercise, 4 addressing both aims)

Participants

545 total (mean age 40 +/- 7 years, mean VISA-A 60/100, BMI 25 kg/m2, 353 men and 126 women)

Population

Adults with insertional or mid-portion Achilles tendinopathy of any duration

Evidence grading

Modified van Tulder guidelines; JBI critical appraisal tools used for quality assessment

What they found

• Moderate evidence (7 studies) for lower concentric plantarflexor peak torque on affected vs unaffected side at 90 degrees/s: MD = -8.74 Nm (95% CI -13.91 to -3.56, I2 = 0%) and at 225 degrees/s: MD = -4.83 Nm (95% CI -7.59 to -2.08, I2 = 0%)
• Moderate evidence for lower eccentric plantarflexor peak torque at 90 degrees/s on affected side: MD = -12.98 Nm (95% CI -25.75 to -0.22, I2 = 0%)
• Impairments in maximal torque were modest: 9% and 13% when pooled effect was divided by mean unaffected side scores
• Limited evidence (1 study, 39 AT vs 38 controls) for lower isotonic concentric peak torque at 90 degrees/s: MD = -17.30 Nm (95% CI -25.73 to -8.87) and at 225 degrees/s: MD = -8.10 Nm (95% CI -13.54 to -2.66) with knee bent
• Very limited evidence (2 studies, 41 AT vs 68 controls) for no difference in plantarflexor peak torque at 60 degrees/s: MD = -4.51 Nm (95% CI -12.10 to 3.08, I2 = 0%)
• Limited evidence (1 study, 14 participants) for reduced normalised rate of force development 0-30 ms: MD = -60 Ns (95% CI -99.41 to -20.59), 0-50 ms: MD = -69.6 Ns (95% CI -112.92 to -26.28), 0-100 ms: MD = -64.7 Ns (95% CI -99.14 to -30.26) on affected side
• Limited evidence (2 studies, 29 participants) for no change in concentric peak torque at 90 degrees/s after 12 weeks of Alfredson protocol: MD = 8.76 Nm (95% CI -3.45 to 20.97)
• Limited evidence (2 studies, 51 participants) for no change in plantarflexor power after resistance training: MD = 26.00 W (95% CI -0.29 to 0.80) at 6 weeks and MD = 59.32 W (95% CI -3.60 to 122.23) at 1 year
• Limited evidence (2 studies, 49 participants) for improvement in heel raise endurance work after 12 weeks of Silbernagel programme: MD = 616.46 J (95% CI 173.39 to 1059.54)
• Limited evidence (4 studies, 105 participants) that heel raise repetitions increased significantly after 12 weeks of resistance training

Limitations

• A majority of torque studies on affected vs unaffected side came from one research group (Alfredson et al.), limiting independence of evidence
• Most studies focused on mid-portion AT, leaving very limited evidence about plantarflexor impairments in insertional AT
• Exercise adherence and fidelity were incompletely reported; only 1 of 14 interventional studies clearly reported adherence
• Non-randomised studies included for within-group change analysis introduce threats to internal validity such as natural history and regression to the mean

Why it matters

For patients

If you have Achilles tendinopathy, your calf muscle strength may be modestly reduced on the painful side, but targeted exercises are more likely to improve your endurance than your peak strength.

For clinicians

The unaffected limb may not be a reliable strength benchmark for all patients; endurance-focused exercise appears more responsive than peak torque, and strength deficits may reflect fear-avoidance or pre-existing factors rather than purely training effects.

For readers

This review highlights that muscle impairment in Achilles tendinopathy is heterogeneous and complex, and that improvements in pain with exercise do not necessarily parallel improvements in muscle capacity.