Does generalized joint hypermobility predict joint injury in sport? A review.
AuthorsDonaldson, Peter R
AffiliationMichigan Orthopaedic Institute William Beaumont Hospital Royal Oak, Michigan.
MetadataShow full item record
CitationClin J Sport Med. 2012 Jan;22(1):77-8.
JournalClinical journal of sport medicine : official journal of the Canadian Academy of , Sport Medicine
AbstractOBJECTIVE: To determine whether persons with generalized joint hypermobility have an increased risk of lower limb joint injury during sport. DATA SOURCES: PubMed, CINAHL, EMBASE, and SportDiscus were searched through February 2009, without language restrictions, using terms related to risk; hip, ankle, and knee injuries; and joint instability. Reference lists of included studies and relevant reviews were searched by hand. STUDY SELECTION: Selection criteria were peer-reviewed studies with a prospective design that used an objective scale to measure generalized joint hypermobility; the participants were engaged in sport activity, and the injury data were quantitative and based on diagnosis by a health professional, were self-reported, or resulted in time lost to athletic participation. The studies were screened by 1 researcher and checked by a second. Study methods were independently assessed by 2 investigators using the 6-point scale for prognostic studies developed by Pengel. Disagreements were resolved through discussion. Of 4841 studies identified, 18 met inclusion criteria. Of these, 8 were included in random-effects meta-analyses. DATA EXTRACTION: The data extracted by 2 reviewers included participant and sport characteristics and details of joint hypermobility and injury measurements. More detailed data for 4 investigations were obtained from the study authors. Where possible, hypermobility was defined as >/=4 of 9 points on the British Society of Rheumatology Scale (BSRS). MAIN RESULTS: Lower limb joint injuries (3 studies, 1047 participants) occurred in 14% of participants. Using the BSRS of joint hypermobility, any lower limb injury was not associated with hypermobility [odds ratio (OR), 1.43; 95% confidence interval (CI), 0.56-3.67]. Using the original authors' definitions, hypermobility was associated with risk of knee joint injuries (OR, 2.62; 95% CI, 1.04-6.58) in 5 studies. In 4 studies in which the BSRS could be used (1167 participants; incidence of injury, 8.65%), the association was not significant (OR, 3.98; 95% CI, 0.95-16.55). In 4 studies involving contact sporting activities, BSRS hypermobility was associated with knee joint injury (OR, 4.69; 95% CI, 1.33-16.52). In 5 studies of ankle joint injuries (1361 participants; incidence of injury, 8.74%), generalized joint hypermobility was not associated with risk of injury whether the BSRS was used (OR, 1.28; 95% CI, 0.62-2.63) or the authors' definitions of hypermobility. Hip joint injury data (1 study; 51 male rugby players) showed no increase in risk with increasing BSRS joint mobility (OR, 1.33; 95% CI, 0.12-14.94). The methodologic quality of the studies varied. CONCLUSIONS: Generalized joint hypermobility was associated with knee joint injury, especially among players participating in contact sports. It was not associated with ankle injury or overall with lower limb joint injuries. Findings may have been obscured by the differences between studies in methods of measurement and by the inclusion of a wide range of sports.