Scapular dyskinesis is not an isolated risk factor for shoulder injury in athletes: a systematic review and meta-analysis

Key points

1. 01Pooled risk ratio of 1.07 (95% CI, 0.85-1.34; P = .59) showed no statistically significant link between scapular dyskinesis and shoulder injury
2. 02Scapular dyskinesis was present in 46% of athletes, who had a 25% injury rate versus 21% in athletes with normal scapular control
3. 03Results from individual studies were mixed - two studies found significant associations, five did not
4. 04Sensitivity analyses using stricter injury definitions and excluding rugby collision injuries did not change the conclusion
5. 05Scapular dyskinesis may only become a meaningful risk factor when combined with other variables such as increased training load

How it was conducted

Design

Systematic review and meta-analysis of prospective cohort studies

Databases searched

MEDLINE, CINAHL Plus, SPORTDiscus, and Embase (search date April 15, 2019)

Participants

923 athletes without shoulder injury at baseline, aged 14 to 37 years, across handball, baseball, rugby, volleyball, badminton, and tennis

Scapular assessment

Dynamic observational scapular dyskinesis tests only (static tests excluded per international consensus)

Primary outcome

Development of shoulder injury (pain, disability, or dysfunction) during follow-up

Statistical method

Mantel-Haenszel method, fixed-effects model, pooled risk ratio with 95% CI

What they found

• Pooled RR for scapular dyskinesis and shoulder injury: 1.07 (95% CI, 0.85-1.34; P = .59) - not statistically significant
• 212 shoulder injuries observed across 923 athletes
• Scapular dyskinesis present in 46% of participants; injury rate 25% vs 21% in those with normal scapular control
• Sensitivity analysis excluding Shitara et al (strict injury definition): RR 1.10 (95% CI, 0.87-1.39; P = .77) - not significant
• Sensitivity analysis using more stringent injury definition (severity score >40/100): RR 1.24 (95% CI, 0.86-1.77; P = .25) - not significant
• Sensitivity analysis excluding rugby study (collision-dominated injuries): RR 0.99 (95% CI, 0.77-1.27; P = .93) - not significant
• Clarsen et al (single included study) found obvious scapular dyskinesis associated with shoulder injury: OR 8.4 (95% CI, 1.5-48.1)
• Kawasaki et al (single included study, rugby) found type III scapular dyskinesis associated with shoulder discomfort: OR 3.6 (95% CI, 1.0-12.5)
• Mean study quality score: 14 out of 20 (range 12-15) on modified Downs and Black checklist

Limitations

• Only scapular dyskinesis was examined in isolation - interactions with other risk factors such as training load, strength, and psychosocial variables were not assessed
• Scapular kinematics were measured at baseline only, so changes during the season that could influence injury risk were not captured
• Studies were primarily male athletes in handball and baseball, limiting generalizability to women and other sports
• Heterogeneity in injury definitions across studies (from any symptom to time-loss criteria) may have diluted the ability to detect true associations

Why it matters

For patients

Athletes should not assume that being told they have abnormal scapular movement means they will definitely sustain a shoulder injury, as the evidence does not support this link on its own.

For clinicians

Clinicians should not rely solely on a dynamic scapular dyskinesis test to screen athletes for shoulder injury risk; a broader multifactorial assessment including load monitoring and strength testing is more appropriate.

For readers

This updated, higher-powered meta-analysis challenges a previously cited 43% increased risk figure and suggests the relationship between scapular movement abnormalities and shoulder injury is more nuanced than previously thought.