Short-term effectiveness of high-load compared with low-load strengthening exercise on self-reported function in patients with hypermobile shoulders: a randomised controlled trial

Key points

1. 01High-load exercise (HEAVY) produced a statistically significant 8.3% greater improvement on the WOSI shoulder function scale compared with low-load exercise (LIGHT) at 16 weeks.
2. 02The mean between-group difference of 174.5 points was statistically significant but below the pre-defined minimal important difference of 252 points; the per-protocol analysis reached 250.7 points, meeting the threshold.
3. 0368% of HEAVY patients versus 54% of LIGHT patients achieved a clinically relevant improvement (number needed to treat = 7).
4. 04High-load exercise was associated with more transient muscle soreness (56% vs 37%) and headaches (40% vs 20%), but there were no serious adverse events.
5. 05Most secondary outcomes (strength, range of motion, quality of life) did not reach statistical significance, suggesting broader benefit remains uncertain.

How it was conducted

Design

Multicentre, two-group, superiority parallel randomised controlled trial with balanced block randomisation and electronic allocation concealment

Participants

100 adults aged 18-65 from primary care in Southern Denmark with hypermobility spectrum disorder (HSD) and shoulder pain and/or instability for at least 3 months

Groups

HEAVY (n=50): supervised high-load full-range progressive shoulder strengthening twice weekly plus one home session, 16 weeks; LIGHT (n=50): low-load neutral/mid-range exercises self-trained three times weekly with three supervised sessions

Primary outcome

Western Ontario Shoulder Instability Index (WOSI) total score (0-2100, 0=best) at 16 weeks, minimal important difference 252 points

Analysis

Intention-to-treat with multiple imputation, multivariable linear regression adjusted for baseline score, age, sex and clinic clustering

What they found

• Primary outcome (ITT): adjusted mean between-group difference in WOSI total -174.5 points (95% CI -341.4 to -7.7), representing 8.3% greater improvement favouring HEAVY (statistically significant, p<0.05).
• Per-protocol analysis: adjusted mean difference -250.7 points (95% CI -323.4 to -178.0), meeting the 12% minimal important difference threshold.
• Proportion achieving clinically relevant WOSI improvement: HEAVY 68% vs LIGHT 54% (ITT); number needed to treat 7 (95% CI 4 to 620).
• Per-protocol responder rates: HEAVY 85% vs LIGHT 55%; number needed to treat 3 (95% CI 2 to 7).
• Positive shoulder rotation test (>180 degrees) at 16 weeks: HEAVY 42% vs LIGHT 62%; adjusted OR 0.32 (95% CI 0.13 to 0.80).
• Global Perceived Effect - important improvement in physical symptoms: HEAVY 64% vs LIGHT 44%; adjusted OR 2.37 (95% CI 1.07 to 5.24).
• Muscle soreness: HEAVY 56% vs LIGHT 37% (crude risk difference 19%, 95% CI -2 to 39).
• Headache: HEAVY 40% vs LIGHT 20% (crude risk difference 20%, 95% CI 2 to 39).
• No serious adverse events in either group; 2 patients in HEAVY withdrew due to adverse events.
• Isometric shoulder strength in scaption at 16 weeks: HEAVY 0.52 Nm/kg vs LIGHT 0.48 Nm/kg, adjusted difference 0.05 (95% CI -0.04 to 0.13) - non-significant.

Limitations

• The mean between-group difference (174.5 WOSI points, 8.3%) fell below the pre-specified minimal important difference of 252 points (12%), leaving the clinical relevance of the primary finding uncertain.
• Patients and treating physiotherapists could not be blinded, and the higher supervision intensity in HEAVY may have introduced attention bias that inflated functional outcomes beyond the effect of load alone.
• Findings may not generalise to patients with hypermobile Ehlers-Danlos syndrome, those in secondary or specialist care, or non-Danish populations.
• Many secondary outcome confidence intervals were wide and inconclusive, and no adjustment for multiple testing was made, so chance findings among secondary results cannot be excluded.

Why it matters

For patients

People with hypermobile shoulders can undertake a supervised high-load exercise programme with a reasonable expectation of greater functional improvement than standard low-load exercises, but should be prepared for transient muscle soreness and occasional headache.

For clinicians

A progressive, individually supervised high-load shoulder strengthening programme is statistically superior to standard low-load care at 16 weeks and appears safe in primary care HSD patients, though the clinical magnitude of benefit just misses the formal importance threshold and long-term data are lacking.

For readers

This is the first RCT to directly compare exercise loads in HSD shoulder patients; results support cautious use of high-load protocols in primary care while highlighting the need for longer-term and cost-effectiveness studies.