Background Generalized joint hypermobility (GJH) is often seen and widely underestimated regarding complexity of diagnosis and treatment. The prevalence of GJH was reported between 10% and 18%, with women more often affected (1). Diagnosis is mainly based on the Beighton score (BS), which is based on increased range of motion (ROM) in specific joints like knee, elbow and fingers (2). However, little is known about the importance of conditional factors like muscle strength or balance in persons with GJH, and their influence on pain and disability.
Objectives Aim of this study was to find factors discriminating between hypermobile women with and without symptoms, and women with normal mobility.
Methods A total of 195 women (mean age 25.5 years) were included in this cross-sectional study, whereby 67 were normomobile, 56 symptomatically hypermobile and 47 asymptomatically hypermobile. Hypermobile women had BS of 6 or higher and classification according to symptoms was based on self-reported pain during 6 months. Measurements comprised passive anterior translation of the tibia, passive ROM of the knees, BS, muscle strength, balance during single-leg standing, as well as muscle activity (MA) and ground reaction forces (GRF) during gait and stair climbing. The main parameters of every measurement were included in a principle component analysis (PCA) and all factors with eigenvalue>1 were extracted (3). Main components of these factors were derived from the variables to describe possible discriminating factors between groups.
Results 17 principle components showed eigenvalue>1 and were included in further analysis. Together they accounted for 80% of variance between groups. The five most important factors (eigenvalue>3) accounted for 46% of variance. These factors were composed as follows: #1 (13.7% of variance) consisted of GRF parameters during gait and on stair, #2 (11.0%) was balance as well as quadriceps and hamstrings MA during gait and on stair, #3 (8.4%) included strength of knee extensors and balance, #4 (6.2%) was composed of MA of lower leg muscles on stair, strength of knee flexors, body weight and BMI and #5 (5.7%) was based on passive ROM of knee, BS and passive tibial translation.
Conclusions PCA could not reduce the variables on few factors. Contrariwise, 17 factors remained and the five most important could explain less than 50% of the variance between groups. Furthermore, yet these five factors consisted of several variables derived from different measurements. In general MA was not very important and passive ROM occurred not before factor #5, as well as BS. The high number of extracted factors might be a sign of the complex nature of hypermobility. BS and passive ROM seemed not to be the primary factors to discriminate between women with hypermobility and with normal mobility. The measurements were probably not enough demanding for young women to clearly discriminate symptomatic hypermobile women from asymptomatic hypermobile women.
Simmonds & Keer, Manual Therapy. 2007:298-309.
Remvig et al. The Journal of Rheumatology. 2007:798-803.
Field: Discovering Statistics using SPSS, London, 2013.
Acknowledgements The project was supported by the Swiss National Science Foundation (# 13DPD6 127285) and approved by the Ethics Committee of Canton Bern, Switzerland (Number 229/2008).
Disclosure of Interest None declared