- Split View
-
Views
-
Cite
Cite
Mark C. Scheper, Lesley L. Nicholson, Roger D. Adams, Louise Tofts, Verity Pacey, The natural history of children with joint hypermobility syndrome and Ehlers–Danlos hypermobility type: a longitudinal cohort study, Rheumatology, Volume 56, Issue 12, December 2017, Pages 2073–2083, https://doi.org/10.1093/rheumatology/kex148
- Share Icon Share
Abstract
The objective of the manuscript was to describe the natural history of complaints and disability in children diagnosed with joint hypermobility syndrome (JHS)/Ehlers–Danlos-hypermobility type (EDS-HT) and to identify the constructs that underlie functional decline.
One hundred and one JHS/EDS-HT children were observed over 3 years and assessed at three time points on the following: functional impairments, quality of life, connective tissue laxity, muscle function, postural control and musculoskeletal and multi-systemic complaints. Cluster analysis was performed to identify subgroups in severity. Clinical profiles were determined for these subgroups, and differences were assessed by multivariate analysis of covariance. Mixed linear regression models were used to determine the subsequent trajectories. Finally, an exploratory factor analysis was used to uncover the underlying constructs of functional impairment.
Three clusters of children were identified in terms of functional impairment: mild, moderately and severely affected. Functional impairment at baseline was predictive of worsening trajectories in terms of reduced walking distance and decreased quality of life (P ⩽ 0.05) over 3 years. Multiple interactions between the secondary outcomes were observed, with four underlying constructs identified. All four constructs (multi-systemic effects, pain, fatigue and loss of postural control) contributed significantly to disability (P ⩽ 0.046).
Children diagnosed with JHS/EDS-HT who have a high incidence of multi-systemic complaints (particularly, orthostatic intolerance, urinary incontinence and diarrhoea) and poor postural control in addition to high levels of pain and fatigue at baseline are most likely to have a deteriorating trajectory of functional impairment and, accordingly, warrant clinical prioritization.
Disability is highly variable in joint hypermobility syndrome/Ehlers–Danlos (hypermobility type), and distinct trajectories are present.
Multi-systemic complaints are strongly linked with several maladaptive processes in joint hypermobility syndrome/Ehlers–Danlos (hypermobility type).
Introduction
Children diagnosed with joint hypermobility syndrome (JHS) have generalized joint hypermobility (GJH) and often have hyper-extensible skin [1]. Although some excel in physical activities, for others GJH is accompanied by chronic musculoskeletal complaints, deconditioning, multi-systemic dysfunction and functional impairment [2]. The mechanisms that contribute to functional decline remain unclear [3]. Similar deficits have been reported in children diagnosed with Ehlers–Danlos (hypermobility type) (EDS-HT), and despite the separate diagnostic criteria [4, 5], it has been suggested that JHS and EDS-HT are clinically indistinguishable [6, 7].
Children with JHS/EDS-HT can experience difficulties with functional mobility, with some dependent on walking aids [8]. Improving function of children with disabling symptoms is often the primary focus of management [1, 3]. Currently, there is neither consensus nor adequate evidence as to how treatment for children with JHS/EDS-HT should be optimized in order to reduce disability [7]. A recent meta-analysis [7] demonstrated that in adolescents with JHS/EDS-HT most interventions lack effectiveness for improving function. There is also emerging evidence of the existence of subtypes of JHS/EDS-HT, which may respond differently to treatment and warrant different treatment approaches [3, 9, 10]. A better understanding of the natural history of the disorder would allow early detection of children at risk for functional decline.
The JHS/EDS-HT phenotype is variable in severity and presentation [10]. Widespread arthralgia, joint dislocations and soft tissue injuries are prevalent, often becoming chronic after extended exposure [1, 11]. Other indicators of deconditioning, muscle weakness and decreased exercise tolerance have also been documented [1, 11]. Within the domain of motor control, proprioceptive deficits [12] and difficulties in motor skills [13] have been demonstrated in children with JHS/EDS-HT. It is assumed that the origin of these musculoskeletal deficits is found in the compromised structural integrity of connective tissue [2, 3, 14]. However, connective tissue comprises not only tendons, muscles and ligaments, but also contributes to the integrity of most systems in the human body, including skin, eyes and internal organs. Multi-systemic involvement, including incontinence, constipation, autonomic and cardiac dysfunction, is prevalent in children with JHS/EDS-HT [15, 16], but because most studies are cross-sectional in nature, the natural course of JHS/EDS-HT in children has not yet been described and, consequently, how symptoms interact and result in functional decline has never been elucidated. Examining the natural history of JHS/EDS-HT will provide insight into the mechanisms influencing the condition and may enable clinicians to identify high-risk subgroups and prioritize treatment for those children most likely to experience further functional loss.
Methods
All participants were recruited from the Sydney hypermobility cohort and met the following inclusion criteria: diagnosed with JHS or EDS-HT according to the Brighton and Villefranche criteria, respectively; Beighton score ⩾4/9; aged 6–18 years; able to adhere to the protocol; and no trauma or surgery unrelated to JHS/EDS-HT that would interfere with functioning. Children were recruited from the orthopaedic, physiotherapy, rheumatology, sports medicine and connective tissue dysplasia clinic of The Children’s Hospital at Westmead, Sydney, NSW, Australia. In addition, private paediatricians and paediatric rheumatologists recruited participants, while siblings or family friends who had not previously sought health-care treatment contacted the researchers to determine their eligibility to participate [15]. Those included were followed for three consecutive years, with measurements made at baseline (0 years), T1 follow-up (1.5 years) and T2 follow-up (3 years). The study was observational, with no restrictions on treatment. Age (years), gender, BMI, medication use (medication %), hospitalization (admission incidence %), co-morbidities and orthotic usage were recorded at each visit. The study had ethics approval from the Children’s Hospital at Westmead and the University of Sydney. Informed consent was obtained from all participants’ parents or guardians.
Primary outcomes
Functional impairment, as defined by the International Classification of Functioning for Child and Youth [17], was operationalized in terms of performance and capacity qualifiers, where capacity refers to which activities a child can perform in a standardized environment and performance to what activities a child does in daily life. Functional impairment capacity was assessed by the 6-min walk test, performed on a 25 m track, with children instructed to cover the largest possible distance in 6 min at a self-selected walking speed [18]. The frequency and duration of physical activity undertaken was assessed by the Adolescent Physical Activity Recall Questionnaire [19]. The average number of minutes per week spent in moderate to vigorous physical activity was calculated. Quality of life, reflecting physical, social and emotional wellbeing, was measured with the Paediatric Quality of Life Generic Scale [10]. Scores range from 0 to 100, where a higher score reflects better quality of life [10].
Secondary outcomes
To assess connective tissue laxity, Beighton scores and hamstring length were assessed. The Beighton score consists of nine dichotomously scored manoeuvres, with a total score (ranging from 0 to 9) derived by summation. GJH was defined when a Beighton score of ⩾4 was obtained [20]. Hamstring length was assessed according to the popliteal angle in the supine position, while controlling for pelvic position [21]. Muscle endurance (in seconds) was quantified according to the leg-lift manoeuvre from the Childhood Myositis Assessment Scale [22]. The Childhood Myositis Assessment Scale is reliable in children, with normative data of healthy children demonstrating that 96–100% of children aged ⩾6 years can obtain the maximal score and maintain the straight leg lift for ⩾2min [22]. The Y-balance test was used as a postural control measure, based on the posteromedial direction item of the Star Excursion Balance Test, which has been found to be reliable and valid in children [23]. The total score from both legs, normalized over leg length, was used for analysis. In terms of complaints, both musculoskeletal and multi-systemic complaints were assessed. The number of painful joints lasting >3 months was recorded on a standard body chart. The number of recurrently unstable joints, as perceived by the participants to have subluxated or dislocated on three or more occasions, was recorded during structured interview. Pain intensity of the last 2 weeks was expressed as a value from 0 to 100, where 0 represents no pain and 100 represents worst pain ever [24]. Fatigue was quantified according to the Paediatric Quality of Life Generic Scale-multi-dimensional fatigue scale questionnaire, which assesses fatigue over the previous month in three domains: general, sleep/rest and cognition. For analysis, a total score was used, ranging from 0 to 100, in which lower scores indicate higher levels of fatigue [25]. The incidence of multi-systemic complaints was obtained by structured interview and physical examination of various systems and expressed as a percentage of the included population: skin, eyes, cardiovascular, gastrointestinal and urinary involvement. The extent of the multi-systemic involvement was expressed as the summation of all involved systems.
Statistics
A hierarchical cluster analysis (at baseline) with between-group linkage (Euclidean distances) was performed, based on the primary outcomes, in order to identify potential clusters of functional impairment with different extents. Thereafter, to create a clinical profile, differences between identified groups were determined by multivariate analysis of covariance, age and gender corrected. Finally, the clinical relevance of the differentiated profiles was determined from longitudinal analysis by mixed linear models for each of the primary outcomes (dependent variables), with the group indicator as the independent variable, corrected for confounders.
In the second phase, the underlying constructs of functional impairment were determined. Owing to the presumed close connectivity of the signs and symptoms associated with JHS/EDS-HT, an exploratory factor analysis was performed to achieve variable reduction and to identify common constructs related to functional impairment. The constructs derived from the exploratory factor analysis represent the interlinkage between defects and their development over time. Factor extraction was achieved by principal component analysis with varimax rotation (orthogonal). Sampling adequacy was assessed by Kaiser–Meyer–Olkin measure (KMO), and adequate sampling was achieved when KMO was >0.5. Factor loadings (FLs) of <0.40 were suppressed. In order to analyse the development of the identified constructs over time, mixed linear models were again constructed, with the group indicator as the independent variable, corrected for confounders. Results were expressed in betas (B) and their 95% CI. A P-value of <0.05 was considered statistically significant.
Results
One hundred and one children (45 boys/56 girls) diagnosed with JHS/EDS-HT, according to the Brighton and Villefranche criteria, were included at baseline (Table 1). The group mean age was 11.4 years (range: 6–16.8). At each assessment, participants were provided with recommendations and referrals to clinical services for ongoing treatment, such as physiotherapy, occupational therapy and psychology. Waiting times to access treatment and the form of treatment provided, if any, were not controlled within this observational study. Four subjects were admitted to hospital for abdominal pain and two subjects for orthopaedic procedures. Medication usage was as follows: asthma medication (n = 11, 9.6%), NSIADs (n = 6, 5.4%), opiates (n = 8, 7.3%), laxatives (n = 14, 12.7%), reflux (n = 6, 5.4%), anti-anxiety/depression (n = 8, 7.3%), CSs (n = 4, 3.6%), sleep (n = 2: 1.8%), urinary (n = 2, 1.8%), orthostatic intolerance (n = 4, 3.6%) and stimulants (n = 2, 1.8%). Neither treatment regimen nor medication usage exerted effects on any of the included outcomes. Statistically significant differences between genders were observed. At baseline, girls were older (Relative difference (ΔD): −13.1%, P = 0.011), had longer hamstring length (ΔD: +6.3%, P = 0.006), a higher number of painful joints (ΔD: +28.6%, P = 0.017), higher pain severity (ΔD: +24.6%, P = 0.033) and less fatigue (ΔD: +13.8%, P = 0.023). All other variables failed to reach significance (P ⩾ 0.05). Therefore, age, BMI and gender were retained as confounders.
Outcomes . | . | Boys . | Girls . | . | ||||
---|---|---|---|---|---|---|---|---|
. | Mean (s.d.) . | Minimum . | Maximum . | Mean (s.d.) . | Minimum . | Maximum . | P-value . | |
Age, years | 10.6 (3.0) | 6.0 | 16.1 | 12.2 (3.0) | 6.6 | 16.8 | 0.011 | |
BMI, kg/cm2 | 18.6 (3.7) | 13.3 | 30.6 | 20.1 (4.7) | 12.5 | 31.2 | 0.085 | |
Gender, n (%) | 45 (44.6) | 56 (55.4) | ||||||
Connective tissue laxity, muscle function and motor control (independent factors) | ||||||||
Beighton, score out of 9 | 7 (1) | 4 | 9 | 7 (2) | 4 | 9 | 0.747 | |
Hamstring length, popliteal angle | 152.1 (19.5) | 120.0 | 180.0 | 162.3 (16.1) | 120.0 | 180.0 | 0.006 | |
FPI, score out of 24 | 14 (5) | 5 | 24 | 13 (6) | 4 | 24 | 0.408 | |
Unstable joints, count | 3 (4) | 0 | 14 | 4 (4) | 0 | 14 | 0.882 | |
Muscle endurance, s | 146.0 (78.3) | 18.2 | 240.0 | 152.5 (75.8) | 17.0 | 240.0 | 0.673 | |
Motor control, Y-balancea | 2.7 (0.4) | 1.8 | 3.4 | 2.5 (0.4) | 1.6 | 3.8 | 0.064 | |
Musculoskeletal complaints (independent factors) | ||||||||
Painful joints, count | 5 (4) | 0 | 14 | 7 (4) | 0 | 15 | 0.017 | |
Pain severity, VAS: 0–100 | 43.5 (33.4) | 0 | 100 | 57.7 (32.4) | 0 | 100 | 0.033 | |
Fatigue, PedsQL MFI 0–100b | 67.8 (15.5) | 40.0 | 93.5 | 55.8 (22.6) | 15.3 | 97.2 | 0.023 | |
Disability (dependent factors) | ||||||||
Walking capacity, 6 MWTa | 7.4 (1.4) | 5.1 | 10.9 | 7.1 (1.7) | 4.1 | 11.2 | 0.335 | |
Habitual physical activity level, METS | 472.8 (332.9) | 0 | 1651.3 | 433.1 (322.8) | 0.0 | 1260.0 | 0.546 | |
QoL, PedsQL: 0–100b | 67.9 (15.5) | 37.0 | 93.4 | 61.1 (19.2) | 19.6 | 95.7 | 0.055 |
Outcomes . | . | Boys . | Girls . | . | ||||
---|---|---|---|---|---|---|---|---|
. | Mean (s.d.) . | Minimum . | Maximum . | Mean (s.d.) . | Minimum . | Maximum . | P-value . | |
Age, years | 10.6 (3.0) | 6.0 | 16.1 | 12.2 (3.0) | 6.6 | 16.8 | 0.011 | |
BMI, kg/cm2 | 18.6 (3.7) | 13.3 | 30.6 | 20.1 (4.7) | 12.5 | 31.2 | 0.085 | |
Gender, n (%) | 45 (44.6) | 56 (55.4) | ||||||
Connective tissue laxity, muscle function and motor control (independent factors) | ||||||||
Beighton, score out of 9 | 7 (1) | 4 | 9 | 7 (2) | 4 | 9 | 0.747 | |
Hamstring length, popliteal angle | 152.1 (19.5) | 120.0 | 180.0 | 162.3 (16.1) | 120.0 | 180.0 | 0.006 | |
FPI, score out of 24 | 14 (5) | 5 | 24 | 13 (6) | 4 | 24 | 0.408 | |
Unstable joints, count | 3 (4) | 0 | 14 | 4 (4) | 0 | 14 | 0.882 | |
Muscle endurance, s | 146.0 (78.3) | 18.2 | 240.0 | 152.5 (75.8) | 17.0 | 240.0 | 0.673 | |
Motor control, Y-balancea | 2.7 (0.4) | 1.8 | 3.4 | 2.5 (0.4) | 1.6 | 3.8 | 0.064 | |
Musculoskeletal complaints (independent factors) | ||||||||
Painful joints, count | 5 (4) | 0 | 14 | 7 (4) | 0 | 15 | 0.017 | |
Pain severity, VAS: 0–100 | 43.5 (33.4) | 0 | 100 | 57.7 (32.4) | 0 | 100 | 0.033 | |
Fatigue, PedsQL MFI 0–100b | 67.8 (15.5) | 40.0 | 93.5 | 55.8 (22.6) | 15.3 | 97.2 | 0.023 | |
Disability (dependent factors) | ||||||||
Walking capacity, 6 MWTa | 7.4 (1.4) | 5.1 | 10.9 | 7.1 (1.7) | 4.1 | 11.2 | 0.335 | |
Habitual physical activity level, METS | 472.8 (332.9) | 0 | 1651.3 | 433.1 (322.8) | 0.0 | 1260.0 | 0.546 | |
QoL, PedsQL: 0–100b | 67.9 (15.5) | 37.0 | 93.4 | 61.1 (19.2) | 19.6 | 95.7 | 0.055 |
Normalized over leg length.
Higher score indicates lower levels of fatigue and higher QoL. FPI: Foot Posture Index; METS: metabolic equivalent tasks; MFI: Multi-dimensional Fatigue Inventory; 6 MWT: 6 min walk test; PEDSQL: Paediatric Quality of Life Inventory; QoL: quality of life; VAS: visual analog scale.
Outcomes . | . | Boys . | Girls . | . | ||||
---|---|---|---|---|---|---|---|---|
. | Mean (s.d.) . | Minimum . | Maximum . | Mean (s.d.) . | Minimum . | Maximum . | P-value . | |
Age, years | 10.6 (3.0) | 6.0 | 16.1 | 12.2 (3.0) | 6.6 | 16.8 | 0.011 | |
BMI, kg/cm2 | 18.6 (3.7) | 13.3 | 30.6 | 20.1 (4.7) | 12.5 | 31.2 | 0.085 | |
Gender, n (%) | 45 (44.6) | 56 (55.4) | ||||||
Connective tissue laxity, muscle function and motor control (independent factors) | ||||||||
Beighton, score out of 9 | 7 (1) | 4 | 9 | 7 (2) | 4 | 9 | 0.747 | |
Hamstring length, popliteal angle | 152.1 (19.5) | 120.0 | 180.0 | 162.3 (16.1) | 120.0 | 180.0 | 0.006 | |
FPI, score out of 24 | 14 (5) | 5 | 24 | 13 (6) | 4 | 24 | 0.408 | |
Unstable joints, count | 3 (4) | 0 | 14 | 4 (4) | 0 | 14 | 0.882 | |
Muscle endurance, s | 146.0 (78.3) | 18.2 | 240.0 | 152.5 (75.8) | 17.0 | 240.0 | 0.673 | |
Motor control, Y-balancea | 2.7 (0.4) | 1.8 | 3.4 | 2.5 (0.4) | 1.6 | 3.8 | 0.064 | |
Musculoskeletal complaints (independent factors) | ||||||||
Painful joints, count | 5 (4) | 0 | 14 | 7 (4) | 0 | 15 | 0.017 | |
Pain severity, VAS: 0–100 | 43.5 (33.4) | 0 | 100 | 57.7 (32.4) | 0 | 100 | 0.033 | |
Fatigue, PedsQL MFI 0–100b | 67.8 (15.5) | 40.0 | 93.5 | 55.8 (22.6) | 15.3 | 97.2 | 0.023 | |
Disability (dependent factors) | ||||||||
Walking capacity, 6 MWTa | 7.4 (1.4) | 5.1 | 10.9 | 7.1 (1.7) | 4.1 | 11.2 | 0.335 | |
Habitual physical activity level, METS | 472.8 (332.9) | 0 | 1651.3 | 433.1 (322.8) | 0.0 | 1260.0 | 0.546 | |
QoL, PedsQL: 0–100b | 67.9 (15.5) | 37.0 | 93.4 | 61.1 (19.2) | 19.6 | 95.7 | 0.055 |
Outcomes . | . | Boys . | Girls . | . | ||||
---|---|---|---|---|---|---|---|---|
. | Mean (s.d.) . | Minimum . | Maximum . | Mean (s.d.) . | Minimum . | Maximum . | P-value . | |
Age, years | 10.6 (3.0) | 6.0 | 16.1 | 12.2 (3.0) | 6.6 | 16.8 | 0.011 | |
BMI, kg/cm2 | 18.6 (3.7) | 13.3 | 30.6 | 20.1 (4.7) | 12.5 | 31.2 | 0.085 | |
Gender, n (%) | 45 (44.6) | 56 (55.4) | ||||||
Connective tissue laxity, muscle function and motor control (independent factors) | ||||||||
Beighton, score out of 9 | 7 (1) | 4 | 9 | 7 (2) | 4 | 9 | 0.747 | |
Hamstring length, popliteal angle | 152.1 (19.5) | 120.0 | 180.0 | 162.3 (16.1) | 120.0 | 180.0 | 0.006 | |
FPI, score out of 24 | 14 (5) | 5 | 24 | 13 (6) | 4 | 24 | 0.408 | |
Unstable joints, count | 3 (4) | 0 | 14 | 4 (4) | 0 | 14 | 0.882 | |
Muscle endurance, s | 146.0 (78.3) | 18.2 | 240.0 | 152.5 (75.8) | 17.0 | 240.0 | 0.673 | |
Motor control, Y-balancea | 2.7 (0.4) | 1.8 | 3.4 | 2.5 (0.4) | 1.6 | 3.8 | 0.064 | |
Musculoskeletal complaints (independent factors) | ||||||||
Painful joints, count | 5 (4) | 0 | 14 | 7 (4) | 0 | 15 | 0.017 | |
Pain severity, VAS: 0–100 | 43.5 (33.4) | 0 | 100 | 57.7 (32.4) | 0 | 100 | 0.033 | |
Fatigue, PedsQL MFI 0–100b | 67.8 (15.5) | 40.0 | 93.5 | 55.8 (22.6) | 15.3 | 97.2 | 0.023 | |
Disability (dependent factors) | ||||||||
Walking capacity, 6 MWTa | 7.4 (1.4) | 5.1 | 10.9 | 7.1 (1.7) | 4.1 | 11.2 | 0.335 | |
Habitual physical activity level, METS | 472.8 (332.9) | 0 | 1651.3 | 433.1 (322.8) | 0.0 | 1260.0 | 0.546 | |
QoL, PedsQL: 0–100b | 67.9 (15.5) | 37.0 | 93.4 | 61.1 (19.2) | 19.6 | 95.7 | 0.055 |
Normalized over leg length.
Higher score indicates lower levels of fatigue and higher QoL. FPI: Foot Posture Index; METS: metabolic equivalent tasks; MFI: Multi-dimensional Fatigue Inventory; 6 MWT: 6 min walk test; PEDSQL: Paediatric Quality of Life Inventory; QoL: quality of life; VAS: visual analog scale.
Cluster analysis identified three groups of children with functional impairment: mildly affected (n = 40), moderately affected (n = 31) and severely affected (n = 30). Statistically significant differences were observed between groups, corrected for age and gender (Table 2). Severely affected children showed shorter hamstring length in comparison with the mildly affected group (ΔD: −5.1%, P = 0.018), but not when compared with the moderately affected group (P = 0.209). In terms of postural control, severely affected children demonstrated a reduced ability to control their body over their base of support in the Y-balance test when compared with the moderately (ΔD: −17.4%, P = 0.002) and mildly affected children (ΔD: −21.7%, P = 0.025), but the remaining variables did not reach statistical significance (P ⩾ 0.05). No statistically significant differences were observed between mildly affected children and moderately affected children for the remaining outcomes. Severely affected children had a higher number of painful joints (ΔD: +44.4%, P = 0.033), higher pain intensity (ΔD: +8.4%, P = 0.002) and worse fatigue (ΔD: +61.3%, P ⩽ 0.0001) in comparison with the mildly affected children. When comparing severely with moderately affected children, higher levels of pain intensity (ΔD: +30.6%, P = 0.042) and fatigue (ΔD: +43.1%, P ⩽ 0.0001) were observed in the severely affected group, but there were no significant differences in the number of painful joints (P = 0.143). Moderately affected children had more fatigue than mildly affected children (ΔD: +35.7%, P ⩽ 0.0001). No other statistically significant differences were found.
Outcomes . | . | . | . | Between-group comparisons . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Mild (M) . | Moderate (Mo) . | Severe (S) . | Mild vs moderate . | Severe vs mild . | Moderate vs severe . | ||||||
. | Mean (s.d.) . | Minimum . | Maximum . | Mean (s.d.) . | Minimum . | Maximum . | Mean (s.d.) . | Minimum . | Maximum . | P-valuea . | P-valuea . | P-valuea . |
Age, years | 10.1 (2.5) | 6.5 | 15.5 | 10.4 (2.7) | 6.0 | 16.4 | 14.0 (2.3) | 8.8 | 16.8 | 0.998 | <0.0001 | 0.999 |
BMI, kg/cm2 | 18.1 (3.7) | 13.3 | 27.7 | 18.2 (3.0) | 14.4 | 24.9 | 22.5 (4.8) | 14.2 | 31.2 | 0.999 | <0.0001 | 0.999 |
Gender, n (boys %/girls %) | 40 (60.0/40.0) | 31 (44.4/55.6) | 30 (20.0/80.0) | 0.011 | – | – | ||||||
Connective tissue laxity, muscle function and motor control (independent factors) | ||||||||||||
Beighton, score out of 9) | 7 (1) | 5 | 9 | 7 (2) | 4 | 9 | 7 (2) | 4 | 9 | 0.959 | 0.999 | 0.390 |
Hamstring length, degree knee extension | 161.9 (16.8) | 120.0 | 180.0 | 156.5 (16.9) | 117.5 | 180.0 | 154.1 (19.4) | 117.5 | 180.0 | 0.564 | 0.018 | 0.209 |
FPI, score out of 24 | 14 (5) | 2 | 24 | 14 (5) | 2 | 24 | 12 (5) | 3 | 23 | 0.912 | 0.978 | 0.999 |
Unstable joints, count | 3 (3) | 0 | 12 | 3 (3) | 0 | 12 | 5 (4) | 0 | 14 | 0.483 | 0.187 | 0.189 |
Muscle endurance, s | 148.8 (77.6) | 30.1 | 240.0 | 163.6 (72.1) | 35.5 | 240.0 | 150.8 (77.5) | 17.0 | 240.0 | 0.997 | 0.200 | 0.496 |
Motor control, Y-balanceb | 2.7 (0.4) | 1.6 | 3.2 | 2.8 (0.4) | 1.8 | 3.8 | 2.3 (.3) | 1.7 | 2.9 | 0.987 | 0.001 | 0.025 |
Musculoskeletal complaints (independent factors) | ||||||||||||
Painful joints, count | 5 (4) | 0 | 15 | 5 (3) | 0 | 12 | 9 (4) | 0 | 15 | 0.596 | 0.033 | 0.143 |
Pain severity, VAS: 0–100 | 43.4 (33.0) | 0 | 100 | 47.3 (32.9) | 0 | 100 | 68.2 (29.0) | 0 | 100 | 0.453 | 0.002 | 0.042 |
Fatigue, PedsQL MFI 0–100c | 84.7 (7.1) | 73.6 | 98.6 | 62.4 (7.3) | 50.0 | 72.2 | 35.7 (11.1) | 15.3 | 62.4 | <0.0001 | <0.0001 | <0.0001 |
Outcomes . | . | . | . | Between-group comparisons . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Mild (M) . | Moderate (Mo) . | Severe (S) . | Mild vs moderate . | Severe vs mild . | Moderate vs severe . | ||||||
. | Mean (s.d.) . | Minimum . | Maximum . | Mean (s.d.) . | Minimum . | Maximum . | Mean (s.d.) . | Minimum . | Maximum . | P-valuea . | P-valuea . | P-valuea . |
Age, years | 10.1 (2.5) | 6.5 | 15.5 | 10.4 (2.7) | 6.0 | 16.4 | 14.0 (2.3) | 8.8 | 16.8 | 0.998 | <0.0001 | 0.999 |
BMI, kg/cm2 | 18.1 (3.7) | 13.3 | 27.7 | 18.2 (3.0) | 14.4 | 24.9 | 22.5 (4.8) | 14.2 | 31.2 | 0.999 | <0.0001 | 0.999 |
Gender, n (boys %/girls %) | 40 (60.0/40.0) | 31 (44.4/55.6) | 30 (20.0/80.0) | 0.011 | – | – | ||||||
Connective tissue laxity, muscle function and motor control (independent factors) | ||||||||||||
Beighton, score out of 9) | 7 (1) | 5 | 9 | 7 (2) | 4 | 9 | 7 (2) | 4 | 9 | 0.959 | 0.999 | 0.390 |
Hamstring length, degree knee extension | 161.9 (16.8) | 120.0 | 180.0 | 156.5 (16.9) | 117.5 | 180.0 | 154.1 (19.4) | 117.5 | 180.0 | 0.564 | 0.018 | 0.209 |
FPI, score out of 24 | 14 (5) | 2 | 24 | 14 (5) | 2 | 24 | 12 (5) | 3 | 23 | 0.912 | 0.978 | 0.999 |
Unstable joints, count | 3 (3) | 0 | 12 | 3 (3) | 0 | 12 | 5 (4) | 0 | 14 | 0.483 | 0.187 | 0.189 |
Muscle endurance, s | 148.8 (77.6) | 30.1 | 240.0 | 163.6 (72.1) | 35.5 | 240.0 | 150.8 (77.5) | 17.0 | 240.0 | 0.997 | 0.200 | 0.496 |
Motor control, Y-balanceb | 2.7 (0.4) | 1.6 | 3.2 | 2.8 (0.4) | 1.8 | 3.8 | 2.3 (.3) | 1.7 | 2.9 | 0.987 | 0.001 | 0.025 |
Musculoskeletal complaints (independent factors) | ||||||||||||
Painful joints, count | 5 (4) | 0 | 15 | 5 (3) | 0 | 12 | 9 (4) | 0 | 15 | 0.596 | 0.033 | 0.143 |
Pain severity, VAS: 0–100 | 43.4 (33.0) | 0 | 100 | 47.3 (32.9) | 0 | 100 | 68.2 (29.0) | 0 | 100 | 0.453 | 0.002 | 0.042 |
Fatigue, PedsQL MFI 0–100c | 84.7 (7.1) | 73.6 | 98.6 | 62.4 (7.3) | 50.0 | 72.2 | 35.7 (11.1) | 15.3 | 62.4 | <0.0001 | <0.0001 | <0.0001 |
Corrected for age and gender.
Normalized over leg length.
Higher score indicates lower levels of fatigue and higher QoL. FPI: Foot Posture Index; METS: metabolic equivalent tasks; MFI: Multi-dimensional Fatigue Inventory; PEDSQL: Pediatric Quality of Life Inventory; QoL: quality of life; VAS: visual analog scale.
Outcomes . | . | . | . | Between-group comparisons . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Mild (M) . | Moderate (Mo) . | Severe (S) . | Mild vs moderate . | Severe vs mild . | Moderate vs severe . | ||||||
. | Mean (s.d.) . | Minimum . | Maximum . | Mean (s.d.) . | Minimum . | Maximum . | Mean (s.d.) . | Minimum . | Maximum . | P-valuea . | P-valuea . | P-valuea . |
Age, years | 10.1 (2.5) | 6.5 | 15.5 | 10.4 (2.7) | 6.0 | 16.4 | 14.0 (2.3) | 8.8 | 16.8 | 0.998 | <0.0001 | 0.999 |
BMI, kg/cm2 | 18.1 (3.7) | 13.3 | 27.7 | 18.2 (3.0) | 14.4 | 24.9 | 22.5 (4.8) | 14.2 | 31.2 | 0.999 | <0.0001 | 0.999 |
Gender, n (boys %/girls %) | 40 (60.0/40.0) | 31 (44.4/55.6) | 30 (20.0/80.0) | 0.011 | – | – | ||||||
Connective tissue laxity, muscle function and motor control (independent factors) | ||||||||||||
Beighton, score out of 9) | 7 (1) | 5 | 9 | 7 (2) | 4 | 9 | 7 (2) | 4 | 9 | 0.959 | 0.999 | 0.390 |
Hamstring length, degree knee extension | 161.9 (16.8) | 120.0 | 180.0 | 156.5 (16.9) | 117.5 | 180.0 | 154.1 (19.4) | 117.5 | 180.0 | 0.564 | 0.018 | 0.209 |
FPI, score out of 24 | 14 (5) | 2 | 24 | 14 (5) | 2 | 24 | 12 (5) | 3 | 23 | 0.912 | 0.978 | 0.999 |
Unstable joints, count | 3 (3) | 0 | 12 | 3 (3) | 0 | 12 | 5 (4) | 0 | 14 | 0.483 | 0.187 | 0.189 |
Muscle endurance, s | 148.8 (77.6) | 30.1 | 240.0 | 163.6 (72.1) | 35.5 | 240.0 | 150.8 (77.5) | 17.0 | 240.0 | 0.997 | 0.200 | 0.496 |
Motor control, Y-balanceb | 2.7 (0.4) | 1.6 | 3.2 | 2.8 (0.4) | 1.8 | 3.8 | 2.3 (.3) | 1.7 | 2.9 | 0.987 | 0.001 | 0.025 |
Musculoskeletal complaints (independent factors) | ||||||||||||
Painful joints, count | 5 (4) | 0 | 15 | 5 (3) | 0 | 12 | 9 (4) | 0 | 15 | 0.596 | 0.033 | 0.143 |
Pain severity, VAS: 0–100 | 43.4 (33.0) | 0 | 100 | 47.3 (32.9) | 0 | 100 | 68.2 (29.0) | 0 | 100 | 0.453 | 0.002 | 0.042 |
Fatigue, PedsQL MFI 0–100c | 84.7 (7.1) | 73.6 | 98.6 | 62.4 (7.3) | 50.0 | 72.2 | 35.7 (11.1) | 15.3 | 62.4 | <0.0001 | <0.0001 | <0.0001 |
Outcomes . | . | . | . | Between-group comparisons . | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
. | Mild (M) . | Moderate (Mo) . | Severe (S) . | Mild vs moderate . | Severe vs mild . | Moderate vs severe . | ||||||
. | Mean (s.d.) . | Minimum . | Maximum . | Mean (s.d.) . | Minimum . | Maximum . | Mean (s.d.) . | Minimum . | Maximum . | P-valuea . | P-valuea . | P-valuea . |
Age, years | 10.1 (2.5) | 6.5 | 15.5 | 10.4 (2.7) | 6.0 | 16.4 | 14.0 (2.3) | 8.8 | 16.8 | 0.998 | <0.0001 | 0.999 |
BMI, kg/cm2 | 18.1 (3.7) | 13.3 | 27.7 | 18.2 (3.0) | 14.4 | 24.9 | 22.5 (4.8) | 14.2 | 31.2 | 0.999 | <0.0001 | 0.999 |
Gender, n (boys %/girls %) | 40 (60.0/40.0) | 31 (44.4/55.6) | 30 (20.0/80.0) | 0.011 | – | – | ||||||
Connective tissue laxity, muscle function and motor control (independent factors) | ||||||||||||
Beighton, score out of 9) | 7 (1) | 5 | 9 | 7 (2) | 4 | 9 | 7 (2) | 4 | 9 | 0.959 | 0.999 | 0.390 |
Hamstring length, degree knee extension | 161.9 (16.8) | 120.0 | 180.0 | 156.5 (16.9) | 117.5 | 180.0 | 154.1 (19.4) | 117.5 | 180.0 | 0.564 | 0.018 | 0.209 |
FPI, score out of 24 | 14 (5) | 2 | 24 | 14 (5) | 2 | 24 | 12 (5) | 3 | 23 | 0.912 | 0.978 | 0.999 |
Unstable joints, count | 3 (3) | 0 | 12 | 3 (3) | 0 | 12 | 5 (4) | 0 | 14 | 0.483 | 0.187 | 0.189 |
Muscle endurance, s | 148.8 (77.6) | 30.1 | 240.0 | 163.6 (72.1) | 35.5 | 240.0 | 150.8 (77.5) | 17.0 | 240.0 | 0.997 | 0.200 | 0.496 |
Motor control, Y-balanceb | 2.7 (0.4) | 1.6 | 3.2 | 2.8 (0.4) | 1.8 | 3.8 | 2.3 (.3) | 1.7 | 2.9 | 0.987 | 0.001 | 0.025 |
Musculoskeletal complaints (independent factors) | ||||||||||||
Painful joints, count | 5 (4) | 0 | 15 | 5 (3) | 0 | 12 | 9 (4) | 0 | 15 | 0.596 | 0.033 | 0.143 |
Pain severity, VAS: 0–100 | 43.4 (33.0) | 0 | 100 | 47.3 (32.9) | 0 | 100 | 68.2 (29.0) | 0 | 100 | 0.453 | 0.002 | 0.042 |
Fatigue, PedsQL MFI 0–100c | 84.7 (7.1) | 73.6 | 98.6 | 62.4 (7.3) | 50.0 | 72.2 | 35.7 (11.1) | 15.3 | 62.4 | <0.0001 | <0.0001 | <0.0001 |
Corrected for age and gender.
Normalized over leg length.
Higher score indicates lower levels of fatigue and higher QoL. FPI: Foot Posture Index; METS: metabolic equivalent tasks; MFI: Multi-dimensional Fatigue Inventory; PEDSQL: Pediatric Quality of Life Inventory; QoL: quality of life; VAS: visual analog scale.
Multi-systemic complaints
The incidence of multi-systemic complaints within groups is presented in Table 3. Skin involvement measured by abnormal scarring (n = 11, 43.5%, P = 0.022) and stretch marks (n = 9, 52.9%, P = 0.039) were seen more frequently in severely affected children. Likewise, orthostatic intolerance (n = 17, 56.7%, P ⩽ 0.0001), diarrhoea (n = 8, 50.0%, P = 0.042) and urinary incontinence (n = 9, 40.9%, P = 0.049) were reported more frequently in the severely affected children. The number of systems involved was also significantly higher in the severely affected group [50th percentile of median (25th percentile-75th percentile): 5 (3–7), P = 0.006].
. | Mildly affected . | Moderately affected . | Severely affected . | . |
---|---|---|---|---|
. | n (%) . | n (%) . | n (%) . | P-value . |
Skin involvement | ||||
Soft skin | ||||
Present | 20 (37.0) | 20 (37.0) | 19 (32.2) | 0.387 |
Absent | 20 (37.0) | 11 (39.3) | 11 (39.3) | |
Stretchy skin | ||||
Present | 7 (30.4) | 8 (34.8) | 8 (34.8) | 0.970 |
Absent | 33 (42.3) | 23 (29.5) | 22 (28.2) | |
Scarring | ||||
Present | 8 (26.7) | 6 (20.0) | 16 (53.3) | 0.022 |
Absent | 32 (45.7) | 25 (34.3) | 14 (20.0) | |
Stretch marks | ||||
Present | 2 (9.1) | 6 (27.3) | 14 (63.6) | 0.039 |
Absent | 38 (38.5) | 25 (36.9) | 16 (24.6) | |
Easy bruising | ||||
Present | 15 (35.7) | 13 (31.0) | 15 (33.3) | 0.555 |
Absent | 25 (43.1) | 18 (31.0) | 15 (25.9) | |
Eye involvement | ||||
Short sighted | ||||
Present | 3 (37.5) | 2 (25.0) | 3 (37.5) | 0.769 |
Absent | 37 (39.8) | 29 (31.2) | 27 (29.0) | |
Long sighted | ||||
Present | 2 (28.6) | 2 (28.6) | 3 (42.9) | 0.755 |
Absent | 38 (40.4) | 29 (30.6) | 27 (28.7) | |
Other | ||||
Present | 6 (33.3) | 5 (27.8) | 7 (38.9) | 0.599 |
Absent | 34 (41.0) | 26 (31.3) | 23 (27.7) | |
Cardiovascular involvement | ||||
Varicose veins | ||||
Present | 1 (100.0) | 0 (0.0) | 0 (0.0) | 0.357 |
Absent | 39 (39.0) | 31 (31.0) | 30 (30.0) | |
Valvular dysfunction (on echocardiogram) | ||||
Present | 1 (33.3) | 1 (33.3) | 1 (33.3) | 0.991 |
Absent | 39 (39.8) | 30 (30.6) | 29 (29.6) | |
Orthostatic intolerance (by self-report) | ||||
Present | 4 (11.4) | 9 (25.7) | 22 (62.9) | <0.0001 |
Absent | 36 (54.5) | 22 (33.3) | 8 (12.1) | |
Other | ||||
Present | 4 (44.4) | 1 (11.1) | 4 (44.4) | 0.241 |
Absent | 36 (39.1) | 30 (32.6) | 26 (28.3) | |
Gastrointestinal and urinary involvement | ||||
Constipation | ||||
Present | 9 (32.1) | 10 (31.2) | 9 (32.1) | 0.973 |
Absent | 31 (42.5) | 21 (28.8) | 21 (28.8) | |
Diarrhoea | ||||
Present | 2 (11.1) | 6 (33.3) | 10 (55.6) | 0.042 |
Absent | 38 (54.7) | 25 (24.1) | 20 (24.1) | |
Urinary incontinence | ||||
Present | 8 (33.3) | 5 (20.8) | 11 (45.8) | 0.049 |
Absent | 32 (41.6) | 26 (33.8) | 19 (24.7) | |
Other | ||||
Present | 7 (33.3) | 7 (33.3) | 7 (33.3) | 0.924 |
Absent | 33 (41.3) | 24 (30.0) | 23 (28.8) | |
Multi-systemic involvement (count) | Median (IQR) | Median (IQR) | Median (IQR) | P-value |
1.5 (1–3) | 2 (1–4) | 5 (3–7) | 0.006 |
. | Mildly affected . | Moderately affected . | Severely affected . | . |
---|---|---|---|---|
. | n (%) . | n (%) . | n (%) . | P-value . |
Skin involvement | ||||
Soft skin | ||||
Present | 20 (37.0) | 20 (37.0) | 19 (32.2) | 0.387 |
Absent | 20 (37.0) | 11 (39.3) | 11 (39.3) | |
Stretchy skin | ||||
Present | 7 (30.4) | 8 (34.8) | 8 (34.8) | 0.970 |
Absent | 33 (42.3) | 23 (29.5) | 22 (28.2) | |
Scarring | ||||
Present | 8 (26.7) | 6 (20.0) | 16 (53.3) | 0.022 |
Absent | 32 (45.7) | 25 (34.3) | 14 (20.0) | |
Stretch marks | ||||
Present | 2 (9.1) | 6 (27.3) | 14 (63.6) | 0.039 |
Absent | 38 (38.5) | 25 (36.9) | 16 (24.6) | |
Easy bruising | ||||
Present | 15 (35.7) | 13 (31.0) | 15 (33.3) | 0.555 |
Absent | 25 (43.1) | 18 (31.0) | 15 (25.9) | |
Eye involvement | ||||
Short sighted | ||||
Present | 3 (37.5) | 2 (25.0) | 3 (37.5) | 0.769 |
Absent | 37 (39.8) | 29 (31.2) | 27 (29.0) | |
Long sighted | ||||
Present | 2 (28.6) | 2 (28.6) | 3 (42.9) | 0.755 |
Absent | 38 (40.4) | 29 (30.6) | 27 (28.7) | |
Other | ||||
Present | 6 (33.3) | 5 (27.8) | 7 (38.9) | 0.599 |
Absent | 34 (41.0) | 26 (31.3) | 23 (27.7) | |
Cardiovascular involvement | ||||
Varicose veins | ||||
Present | 1 (100.0) | 0 (0.0) | 0 (0.0) | 0.357 |
Absent | 39 (39.0) | 31 (31.0) | 30 (30.0) | |
Valvular dysfunction (on echocardiogram) | ||||
Present | 1 (33.3) | 1 (33.3) | 1 (33.3) | 0.991 |
Absent | 39 (39.8) | 30 (30.6) | 29 (29.6) | |
Orthostatic intolerance (by self-report) | ||||
Present | 4 (11.4) | 9 (25.7) | 22 (62.9) | <0.0001 |
Absent | 36 (54.5) | 22 (33.3) | 8 (12.1) | |
Other | ||||
Present | 4 (44.4) | 1 (11.1) | 4 (44.4) | 0.241 |
Absent | 36 (39.1) | 30 (32.6) | 26 (28.3) | |
Gastrointestinal and urinary involvement | ||||
Constipation | ||||
Present | 9 (32.1) | 10 (31.2) | 9 (32.1) | 0.973 |
Absent | 31 (42.5) | 21 (28.8) | 21 (28.8) | |
Diarrhoea | ||||
Present | 2 (11.1) | 6 (33.3) | 10 (55.6) | 0.042 |
Absent | 38 (54.7) | 25 (24.1) | 20 (24.1) | |
Urinary incontinence | ||||
Present | 8 (33.3) | 5 (20.8) | 11 (45.8) | 0.049 |
Absent | 32 (41.6) | 26 (33.8) | 19 (24.7) | |
Other | ||||
Present | 7 (33.3) | 7 (33.3) | 7 (33.3) | 0.924 |
Absent | 33 (41.3) | 24 (30.0) | 23 (28.8) | |
Multi-systemic involvement (count) | Median (IQR) | Median (IQR) | Median (IQR) | P-value |
1.5 (1–3) | 2 (1–4) | 5 (3–7) | 0.006 |
IQR: interquartile range.
. | Mildly affected . | Moderately affected . | Severely affected . | . |
---|---|---|---|---|
. | n (%) . | n (%) . | n (%) . | P-value . |
Skin involvement | ||||
Soft skin | ||||
Present | 20 (37.0) | 20 (37.0) | 19 (32.2) | 0.387 |
Absent | 20 (37.0) | 11 (39.3) | 11 (39.3) | |
Stretchy skin | ||||
Present | 7 (30.4) | 8 (34.8) | 8 (34.8) | 0.970 |
Absent | 33 (42.3) | 23 (29.5) | 22 (28.2) | |
Scarring | ||||
Present | 8 (26.7) | 6 (20.0) | 16 (53.3) | 0.022 |
Absent | 32 (45.7) | 25 (34.3) | 14 (20.0) | |
Stretch marks | ||||
Present | 2 (9.1) | 6 (27.3) | 14 (63.6) | 0.039 |
Absent | 38 (38.5) | 25 (36.9) | 16 (24.6) | |
Easy bruising | ||||
Present | 15 (35.7) | 13 (31.0) | 15 (33.3) | 0.555 |
Absent | 25 (43.1) | 18 (31.0) | 15 (25.9) | |
Eye involvement | ||||
Short sighted | ||||
Present | 3 (37.5) | 2 (25.0) | 3 (37.5) | 0.769 |
Absent | 37 (39.8) | 29 (31.2) | 27 (29.0) | |
Long sighted | ||||
Present | 2 (28.6) | 2 (28.6) | 3 (42.9) | 0.755 |
Absent | 38 (40.4) | 29 (30.6) | 27 (28.7) | |
Other | ||||
Present | 6 (33.3) | 5 (27.8) | 7 (38.9) | 0.599 |
Absent | 34 (41.0) | 26 (31.3) | 23 (27.7) | |
Cardiovascular involvement | ||||
Varicose veins | ||||
Present | 1 (100.0) | 0 (0.0) | 0 (0.0) | 0.357 |
Absent | 39 (39.0) | 31 (31.0) | 30 (30.0) | |
Valvular dysfunction (on echocardiogram) | ||||
Present | 1 (33.3) | 1 (33.3) | 1 (33.3) | 0.991 |
Absent | 39 (39.8) | 30 (30.6) | 29 (29.6) | |
Orthostatic intolerance (by self-report) | ||||
Present | 4 (11.4) | 9 (25.7) | 22 (62.9) | <0.0001 |
Absent | 36 (54.5) | 22 (33.3) | 8 (12.1) | |
Other | ||||
Present | 4 (44.4) | 1 (11.1) | 4 (44.4) | 0.241 |
Absent | 36 (39.1) | 30 (32.6) | 26 (28.3) | |
Gastrointestinal and urinary involvement | ||||
Constipation | ||||
Present | 9 (32.1) | 10 (31.2) | 9 (32.1) | 0.973 |
Absent | 31 (42.5) | 21 (28.8) | 21 (28.8) | |
Diarrhoea | ||||
Present | 2 (11.1) | 6 (33.3) | 10 (55.6) | 0.042 |
Absent | 38 (54.7) | 25 (24.1) | 20 (24.1) | |
Urinary incontinence | ||||
Present | 8 (33.3) | 5 (20.8) | 11 (45.8) | 0.049 |
Absent | 32 (41.6) | 26 (33.8) | 19 (24.7) | |
Other | ||||
Present | 7 (33.3) | 7 (33.3) | 7 (33.3) | 0.924 |
Absent | 33 (41.3) | 24 (30.0) | 23 (28.8) | |
Multi-systemic involvement (count) | Median (IQR) | Median (IQR) | Median (IQR) | P-value |
1.5 (1–3) | 2 (1–4) | 5 (3–7) | 0.006 |
. | Mildly affected . | Moderately affected . | Severely affected . | . |
---|---|---|---|---|
. | n (%) . | n (%) . | n (%) . | P-value . |
Skin involvement | ||||
Soft skin | ||||
Present | 20 (37.0) | 20 (37.0) | 19 (32.2) | 0.387 |
Absent | 20 (37.0) | 11 (39.3) | 11 (39.3) | |
Stretchy skin | ||||
Present | 7 (30.4) | 8 (34.8) | 8 (34.8) | 0.970 |
Absent | 33 (42.3) | 23 (29.5) | 22 (28.2) | |
Scarring | ||||
Present | 8 (26.7) | 6 (20.0) | 16 (53.3) | 0.022 |
Absent | 32 (45.7) | 25 (34.3) | 14 (20.0) | |
Stretch marks | ||||
Present | 2 (9.1) | 6 (27.3) | 14 (63.6) | 0.039 |
Absent | 38 (38.5) | 25 (36.9) | 16 (24.6) | |
Easy bruising | ||||
Present | 15 (35.7) | 13 (31.0) | 15 (33.3) | 0.555 |
Absent | 25 (43.1) | 18 (31.0) | 15 (25.9) | |
Eye involvement | ||||
Short sighted | ||||
Present | 3 (37.5) | 2 (25.0) | 3 (37.5) | 0.769 |
Absent | 37 (39.8) | 29 (31.2) | 27 (29.0) | |
Long sighted | ||||
Present | 2 (28.6) | 2 (28.6) | 3 (42.9) | 0.755 |
Absent | 38 (40.4) | 29 (30.6) | 27 (28.7) | |
Other | ||||
Present | 6 (33.3) | 5 (27.8) | 7 (38.9) | 0.599 |
Absent | 34 (41.0) | 26 (31.3) | 23 (27.7) | |
Cardiovascular involvement | ||||
Varicose veins | ||||
Present | 1 (100.0) | 0 (0.0) | 0 (0.0) | 0.357 |
Absent | 39 (39.0) | 31 (31.0) | 30 (30.0) | |
Valvular dysfunction (on echocardiogram) | ||||
Present | 1 (33.3) | 1 (33.3) | 1 (33.3) | 0.991 |
Absent | 39 (39.8) | 30 (30.6) | 29 (29.6) | |
Orthostatic intolerance (by self-report) | ||||
Present | 4 (11.4) | 9 (25.7) | 22 (62.9) | <0.0001 |
Absent | 36 (54.5) | 22 (33.3) | 8 (12.1) | |
Other | ||||
Present | 4 (44.4) | 1 (11.1) | 4 (44.4) | 0.241 |
Absent | 36 (39.1) | 30 (32.6) | 26 (28.3) | |
Gastrointestinal and urinary involvement | ||||
Constipation | ||||
Present | 9 (32.1) | 10 (31.2) | 9 (32.1) | 0.973 |
Absent | 31 (42.5) | 21 (28.8) | 21 (28.8) | |
Diarrhoea | ||||
Present | 2 (11.1) | 6 (33.3) | 10 (55.6) | 0.042 |
Absent | 38 (54.7) | 25 (24.1) | 20 (24.1) | |
Urinary incontinence | ||||
Present | 8 (33.3) | 5 (20.8) | 11 (45.8) | 0.049 |
Absent | 32 (41.6) | 26 (33.8) | 19 (24.7) | |
Other | ||||
Present | 7 (33.3) | 7 (33.3) | 7 (33.3) | 0.924 |
Absent | 33 (41.3) | 24 (30.0) | 23 (28.8) | |
Multi-systemic involvement (count) | Median (IQR) | Median (IQR) | Median (IQR) | P-value |
1.5 (1–3) | 2 (1–4) | 5 (3–7) | 0.006 |
IQR: interquartile range.
Trajectories of functional impairment
Eighty-one out of 101 children had complete data sets at follow-up and were included for longitudinal analysis. The remaining 20 children did not attend the 3-year follow-up, with 18 unable to be contacted and 2 declining to participate further. No significant differences were found between the included cohort and drop-outs on any measures at baseline (P ⩾ 0.05). The subsequent trajectories of functional impairment are depicted in Fig. 1. Correcting for age, gender and BMI, severely affected children showed decreasing walking distances [B (95% CI): −0.40 (−0.73, −0.08), P = 0.016] and consistently lower quality of life [B (95% CI): −1.19 (−1.41, −0.96), P ⩽ 0.0001] over the course of 3 years. In terms of habitual physical activity, no statistically significant effects were identified [B (95% CI): 0.22 (−0.28, 0.49), P = 0.589].
Principal component analysis factor extraction with a Varimax (orthogonal) rotation of 10 factors collected from 81 participants was conducted (explained variance = 83.5%). An examination of sampling adequacy revealed that the sample was factorable (KMO = 0.714; Fig. 2).
Six variables loaded onto factor 1 (explained variance = 25.7%). Within factor 1, the number of multi-systemic complaints yielded the highest Factor loading (FL) = 0.752, and this was labelled component 1: multi-systemic involvement. This component was characterized by an increasing number of recurrently unstable joints (FL = 0.720), worsening pain intensity (FL= 0.581), increasing fatigue (FL = −0.490), reducing muscle endurance (FL = −0.586) and decreasing postural control (FL = −0.459). Four variables loaded on factor 2 (explained variance = 23.6%). Pain intensity yielded the highest FL = 0.530, hence component 2: pain. This component was characterized by decreasing muscle endurance (FL = −0.528), reducing postural control (FL = −0.500) and increasing fatigue (FL = −0.458). Fatigue loaded independently on factor 3 (FL = −0.688), labelled component 3: fatigue (explained variance = 17.1%). Finally, postural control loaded independently on factor 4 (FL = −0.695), labelled component 4: postural control (explained variance = 17.1%).
The trajectories of the extracted components over time are depicted in Fig. 3. Severely affected children had a distinct pattern of development in comparison with the other groups, when correcting for confounders. Severely affected children showed a statistically significant increase within the multi-systemic component [B (s.e.): 0.41 (0.18), P = 0.023], indicated by increasing multi-systemic complaints, more unstable joints, worsening pain and fatigue and decreasing muscle endurance. Fatigue [B (s.e.): −0.52 (0.14), P ⩽ 0.0001] and postural control [B (s.e.): −2.04 (0.16), P ⩽ 0.0001] worsened consistently over time. With regard to component 2: pain, a more pronounced decline was present in children classified as severely affected [B (s.e.): −0.28 (0.18), P = 0.112], which indicated increasing pain with decreasing muscle endurance and postural control over time. No significantly different trajectories were found when comparing children classified as mildly and moderately affected (P ⩾ 0.05), with the exception of the postural control component, in which moderately affected children showed consistently lower postural control in comparison with mildly affected children [B (s.e.): 2.03 (0.14), P ⩽ 0.0001].
Discussion
This study provides the first description of the natural history of children with JHS/EDS-HT. Four underlying constructs of functional impairment were identified, from which multi-systemic dysfunction was found to be the most important construct for functional impairment and was specific for children who were most likely to deteriorate over time.
Within the multi-systemic component, increased reporting of multi-systemic complaints was related to a cascade of maladaptive processes in terms of increases in the number of unstable joints, pain and fatigue, decreasing muscle endurance and reduced postural control. Although the symptoms that were grouped within this component were diverse in nature, they represented the highest FLs and loaded on the most factors, which would suggest a common construct. The central assumption regarding the pathological mechanism of JHS/EDS-HT is that the structural integrity of connective tissue is compromised [2, 3]. It has been assumed that the expression of GJH, hyperelastic skin and altered muscle–tendon properties [14, 26] are the result of structural modifications in connective tissue. As connective tissue is ubiquitous in the human body, other systems may also be affected, and dysfunction in internal organs [27] and bone [16] has been demonstrated previously. Within the present cohort, the clinical profile of children with JHS/EDS-HT showed stretch marks, orthostatic intolerance, diarrhoea and urinary incontinence to be highly prevalent in the most severely affected children, and this is in line with previous research [28–30]. The high prevalence of these multi-systemic complaints may also be related to increased elasticity of connective tissue in skin, gastrointestinal tracts and vascular smooth muscles, which predominantly consist of type 1 collagen. Although laxity of connective tissue may explain the inability of blood vessels to react to sudden changes in blood flow volumes, thus resulting in orthostatic intolerance, it has recently been suggested that the lack of vasoconstrictive reaction may also be neurologically oriented. De Wandele et al. [31] implicated both neurological factors and connective tissue laxity in cardiovascular and sudomotor dysfunction in adults with EDS-HT. The present study shows that multi-systemic symptoms are specific to children with poor functional ability and that an increase in incidence of these symptoms is not only predictive of functional decline but also the most dominating factor of severely affected children. It would therefore seem essential that clinicians thoroughly assess both the musculoskeletal and multi-systemic complaints, to be able to identify those children most likely to decline functionally. However, the extent to which multi-systemic complaints can be addressed and the pathways by which they develop remain elusive.
The second construct that loaded on the most factors was pain, with an increase in pain intensity being associated with decreases in muscle endurance, postural control and increasing fatigue. Chronic musculoskeletal pain has been shown to induce muscle weakness by means of reflex inhibition [32]. Owing to CNS inhibition, motor unit recruitment within muscles is impaired, and inadequate muscle forces are generated in relationship to activity [33]. The presence of reflex inhibition would explain the clustering of these symptoms, especially as pain thresholds in JHS/EDS-HT are lowered because of generalized hyperalgesia [34, 35]. A close relationship between pain and fatigue has been described frequently in the literature, and this would explain the loading of these variables on the same factor [36].
Finally, loss of postural control was found to be an independent contributor for all three outcomes of functional impairment. It has been postulated that joint instability, pain, muscle weakness and reduced proprioceptive acuity underlie issues regarding postural control [37], which are often expressed as clumsiness, problems with writing and delayed motor development [33, 38]. The present data support the multi-dimensional construct of motor control and also show its importance for functional ability. However, as in fatigue, loss of postural control not only contributed to the multi-systemic component and pain, but was also found to an important independent contributor to functional impairment. Considering the large impact of postural control on the presented models, this finding may also represent a potential pathway for functional recovery [39].
From a clinical perspective, the amount of functional impairment at the first visit is highly predictive for extent of functional decline over time. The clinical profiles presented in this manuscript allow clinicians to determine which children with JHS/EDS-HT are at risk of developing more serious complaints and functional impairment. These profiles are based on low-cost, easy-to-use clinical measures that require little time to perform. Clinicians should therefore not only account for the symptoms as stated by the Brighton/Villefranche criteria, but also develop more extensive clinical profiles to detect children at risk of future decline in function. Although the existence of subgroups within children with JHS/EDS-HT has been suggested, the diagnostic criteria to date do not distinguish different presentations [3], nor do they account for age, gender and ethnicity when classifying GJH [40]. The clinical profiles presented in this paper may be informative for clinicians and aid in early detection of children at risk. Nevertheless, they lack other dimensions, such as psychological co-morbidity, the influence of family function and interpersonal relationships, and this is noted as a limitation of the study.
When considering the clinical utility of the included profiles in this study, a more sophisticated clinical profile could provide more accurate predictions of the course over time, recognizing that the presence of multi-systemic complaints is a dominant feature of children with JHS/EDS-HT who are the most severely affected. In a recent meta-analysis, current physical treatment modalities failed to show significant effectiveness [7], and this may be related to the specific clinical profiles of individual patients. Correct identification of the patient profile and allocation of treatment specific for that profile may be a deciding factor in achieving functional recovery. In many scientific publications regarding JHS/EDS-HT, multi-systemic dysfunction is mentioned and even proposed as a major factor in patient complaints [41], yet little is known about the precise pathways by which these come into effect and how treatment should be adapted when multi-systemic complaints are present. Considering the dominance of the multi-systemic component, it may be necessary to understand this pathway to develop novel interventions. Consequently, clinicians should actively screen for these symptoms, particularly orthostatic intolerance and frequently occurring diarrhoea, when assessing a patient, and this may be an important addition to the diagnostic criteria.
In interpreting these results, the following factors should be considered. Firstly, the included population was followed over time with no standard treatment provided because the study was observational. All children were provided with referrals to allied health practitioners and specialists as part of their clinical routine. The proportion of families who followed through on these referrals, and the waiting times for treatment or the type of treatment provided, were not controlled or recorded. Any treatment received or the lack thereof may have influenced the results; however, the present study was not aimed at determining the effectiveness of assessment or treatment, but set out to describe the natural history of JHS/EDS-HT in children. Secondly, the present paper does suggest mechanisms on the basis of the clustering of symptoms. Although the clustering of these symptoms is plausible and supported by the relevant literature, the present included outcomes do not establish causative mechanisms. Future research should focus on investigating these clusters of symptoms in order to identify the causal pathways. This knowledge may prove vital in enabling the development of more effective diagnostic procedures and tailored care. The included measures do have advantages in terms of low cost, clinical applicability and ease of implementation. In this respect, their usage may serve as a starting point until more exact and consensus-based measurements are available.
Conclusion
Although the phenotype of JHS/EDS-HT has been characterized by expressions of connective tissue laxity of a heterogeneous nature, distinct clinical profiles exist. Children with JHS/EDS-HT who have a high incidence of multi-systemic complaints, in addition to high pain levels, fatigue and poor postural control, are most likely to have a subsequent debilitating trajectory and deserve immediate clinical attention.
Acknowledgements
The authors would like to acknowledge Laurie Wellings, Emma Siegel and Bianca Botha, and the children and their families for their ongoing participation in this study.
Funding: The study was funded by Arthritis Australia 2015 Zimmer Australia Grant and Arthritis Australia and State and Territory Affiliate Grant, the Arthritis Australia 2013 Barbara Cameron Memorial Grant and the Dutch organization for scientific research (NWO: grant number 023.002.094).
Disclosure statement: V.P. is an associate investigator on grants from Arthritis Australia and The Menzies Foundation (administered through university institutions) and a Rheumatology Health Professions Association travel scholarship. L.T. received grant funding for this project from Arthritis Australia. All other authors have declared no conflicts of interest.
Comments