• Juvenile idiopathic arthritis;
  • Exercise testing;
  • Exercise training;
  • Pilot study;
  • Submaximal oxygen consumption


  1. Top of page
  2. Abstract
  7. Acknowledgements


To 1) assess the safety and feasibility of laboratory-based exercise testing in juvenile idiopathic arthritis (JIA), 2) test the safety and feasibility of a 3-month exercise program in JIA, 3) assess pain during exercise in JIA, 4) compare ratings of perceived effort (RPE) with heart rate (HR) achieved, and 5) estimate the training effect on metabolic efficiency of gait as measured by submaximal exercise testing.


Nine children with JIA were enrolled in a 12-week circuit training program involving pool, stationary bicycle, treadmill, and Fitball. They underwent formal exercise testing before and after the program, underwent a full joint assessment, were administered the Childhood Health Assessment Questionnaire and Juvenile Arthritis Functional Status Index, and were assessed for overall quality of life and health-related quality of life. A visual analog scale was used to assess pain during testing and training, and the Borg scale was used to measure RPE.


Children with JIA were able to participate in exercise testing without any significant problems. Children with severe hip disease dropped out of the exercise program due to pain during the exercise sessions and worsened arthritis symptoms. Target HR was achieved and correlated with RPE in the bicycle and treadmill sessions. Submaximal exercise testing showed an improvement with a small to moderate effect size.


This study suggests that it is safe, feasible, and acceptable for children with arthritis, in the absence of severe hip involvement, to participate in formal exercise testing and structured fitness programs.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Juvenile idiopathic arthritis (JIA) is a chronic illness that affects 1 in 1,000 children (1, 2). Patients with JIA may experience significant disability due to muscular weakness, joint pain, contracture, and physical deconditioning. Children with arthritis have been shown to participate in less physical activities and have a higher number of sleep hours than their peers (3). These factors may result in a spiral of deconditioning and disability resulting in an inactive lifestyle (4).

Supporting this idea of impaired physical conditioning, a systematic review by Takken et al (5) identified 9 studies addressing physical fitness in children with JIA. Five of these studies included peak oxygen consumption (VO2peak) measurements and were included in a meta-analysis that showed an overall 22% lower VO2peak in children with JIA (5). Furthermore, Metin et al (6) showed heterogeneity in VO2peak measures within subgroups of patients with JIA. Those with enthesitis-related arthritis had a higher aerobic capacity, but no difference was seen between patients with active and inactive disease (6).

Traditionally, physical therapy for patients with JIA has been aimed at managing pain and inflammation, preserving range of motion, and maintaining muscle strength through rest and limiting the strain on arthritic joints (7, 8). More active forms of therapy have been instituted recently (9), and guidelines have included recommendations for fitness and strengthening exercise in children with JIA to improve function and promote lifetime physical activity (10). Nevertheless, the best therapeutic regimen for children with arthritis is as yet unknown.

Systematic reviews of trials examining the effects of aerobic training on adult patients with rheumatoid arthritis have found that exercise interventions resulted in improved aerobic capacity, muscle strength, and disease activity, with a possible beneficial effect on pain, function, and quality of life (11, 12). No detrimental effects in terms of disease activity, pain, or function were identified.

Few studies of exercise training in children with arthritis are available. Most are small, not randomized, and rely on field testing such as a timed walk rather than more precise, standardized laboratory measurements of oxygen consumption and power output to assess fitness outcomes. Oberg et al (13) suggested that children with arthritis achieved improvements in muscular strength and endurance after a 3-month training program, and that electromyographic abnormalities observed at the beginning of the study improved with training. In an uncontrolled study of 25 children with polyarticular JIA, Klepper (14) demonstrated improvements in aerobic capacity and flexibility after an 8-week, low-impact aerobics and resistance program performed 3 times a week. The program was well tolerated and all patients improved on all measures of arthritis severity (14). More recently, Takken et al's randomized controlled study of aerobic pool exercise involving 54 children demonstrated an improvement in quality of life, joint status, and submaximal endurance (15); these results were not statistically significant. No large-scale randomized studies of land-based exercise programs are available.

The exact roles of aerobic and anaerobic fitness in the function of children with arthritis are not clear. Takken et al (16) demonstrated an association between anaerobic physical fitness and function, suggesting that in children, activities of daily living may depend on short, intense bursts of anaerobic activity. Prolonged endurance types of activities may be more important in adult life. Although precise roles are unknown, it seems reasonable that there would be long- and short-term benefits to both aerobic and anaerobic conditioning in JIA.

Measurements of maximum oxygen uptake (VO2max) have been used in adults to assess fitness. These tests are rigorous and require a plateau in VO2 despite increasing workload. This has proven difficult to measure accurately in children, and therefore VO2peak has been used instead to assess exercise capacity (17). Maximal and peak VO2 measurements may be inappropriate in persons with limitation or disability, and the use of submaximal VO2 (VO2submax) has been suggested as a predictor of VO2max or to assess performance in standardized physical activities (18).

Disability in children with cerebral palsy is associated with a reduced VO2peak along with reduced muscle power and endurance (19). These children are also known to have an exaggerated energy cost of locomotion as measured by submaximal treadmill exercise testing (15, 20, 21). Giannini and Protas (22) demonstrated that children with juvenile arthritis had higher submaximal heart rates and tended towards higher VO2submax as compared with matched children without arthritis at the same workload during a bicycle task. Therefore, it may be that children with arthritis also have an exaggerated energy cost of locomotion that manifests as early fatigue during normal daily activity. VO2submax may provide a means of measuring disability in JIA and provide an easier and more tolerable way to measure aerobic fitness and response to therapy than maximal exercise testing protocols. Submaximal exercise testing and specifically energy cost of locomotion have not been systematically examined in children with arthritis.

This study was undertaken as the pilot study for a randomized controlled study that will test the hypothesis that fitness exercise benefits the physical function and quality of life of children with JIA. This pilot study had 5 aims. First, to assess the safety and feasibility of laboratory-based exercise testing in children with JIA. Second, to test the safety and feasibility of a 3-month aerobic and strength-training program in JIA. Third, to examine the effect of arthritis pain during exercises in JIA. Fourth, to compare subject ratings of perceived effort (RPE) with actual heart rate (HR) achieved during exercise, indicating whether perceived effort can be used as a guide in exercise programs. Finally, to estimate the training effect of regular exercise on metabolic efficiency of gait as measured by VO2submax.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Nine subjects from the rheumatology program at the Bloorview MacMillan Children's Centre (BMCC), a regional pediatric rehabilitation center in Toronto, Canada, were recruited. Prepubescent children (maximum Tanner stage 2), ages 8–11 years, with JIA and a polyarticular course were eligible. These criteria excluded pubescent children who might experience large increases in lean body mass within the study period and younger children who might not be able to cooperate with training and testing protocols. Comorbidity with cardiac, pulmonary, or metabolic disease was grounds for exclusion. Children already engaged in ≥3 hours of structured physical activity per week (other than a physiotherapy pool program) were excluded because they might not show additional gains from fitness training. No child with active features of systemic arthritis was enrolled. There were no restrictions on medication use in this pilot study.


Laboratory testing was undertaken at The Children's Exercise and Nutrition Centre, McMaster University (Hamilton, Ontario, Canada). A habituation session was undertaken at BMCC to familiarize subjects with procedures.

Subjects underwent testing at baseline prior to the 12-week program (pretesting) and testing after the 12-week program (posttesting). Standing height was measured in stocking feet on a wall mounted Stadiometer (Harpenden, London, UK) with a precision of 0.1 cm, and weight was measured to the nearest 10 gm in a hospital gown or t-shirt and shorts using an electronic scale (Ancaster, Brantford, Ontario, Canada). Subjects were examined prior to and after each testing session by a physiotherapy assessor. A joint was considered active if it was effused or displayed 2 of the following features: heat, limited range of movement, tenderness, or stress pain. Pain was measured using a 10-cm visual analog scale (VAS).

The Childhood Health Assessment Questionnaire (CHAQ) was used to assess physical function. The CHAQ is scored between 0 and 3, with 0 denoting no disability and 3 denoting severe disability; the questionnaire is validated for use in JIA (23). The Juvenile Arthritis Functional Status Index (JASI) was used to give a more detailed assessment of functional ability and priority activities (24). The JASI is scored between 0 and 100, with higher scores denoting better function. Overall quality of life (QOL) and health-related quality of life (HRQOL) were measured on a 10-cm VAS previously validated for children with JIA, with a higher score indicating better QOL or HRQOL (25). A subject/parent satisfaction survey, designed by one of the authors (VW), was administered to the subjects and their parents at the end of the final exercise session.

The exercise testing protocol consisted of 3 main components. Energy cost of locomotion was determined by VO2submax while treadmill walking at 1.5 and 3.0 km/hour for 5 minutes. These relatively slow speeds were chosen to ensure that even those subjects with severe active disease or disability due to chronic disease could complete the protocol. Expired gases were analyzed for O2 uptake and CO2 production (SensorMedics Vmax; SensorMedics, Mississauga, Ontario, Canada). HR was measured every 5 seconds using a telemetric device (Polar Vantage XL; Polar Instuments, Kempele, Finland).

VO2peak was measured through an incremental continuous cycling task on a Fleisch ergometer (Fleisch Metabo, Lausanne, Switzerland). Resistance was increased every 2 minutes with increments selected by the investigator based on the child's HR and overall appearance. The test duration was aimed at 6–10 minutes (4) with HR measured continuously and expired gases analyzed breath by breath for O2 uptake and CO2 production (SensorMedics Vmax). The test was continued until volitional fatigue despite strong verbal encouragement (26, 27); this method has been found reliable in previous studies (22, 28).

Leg anaerobic capacity (peak and mean muscle power) was measured through a 30-second all-out cycling task on a cycle ergometer (Fleisch Metabo) following the protocol of the Wingate Anaerobic Test (26). This test has been shown to yield reliable results in normal children and in those with neuromuscular disease including juvenile dermatomyositis (29, 30). The resistance for this test was determined from each subject's force velocity curve, which was derived from several 5–7-second sprints at different loads (31). The anaerobic performance of subjects' upper limbs was assessed in a similar way using a Fleisch ergometer modified for arm cranking.

Training protocol.

Sessions were held at BMCC twice weekly for 12 weeks. Groups were supervised by a physiotherapist with the assistance of a student physiotherapist.

Sessions began with a 15-minute pool warmup consisting of range of motion and light stretching designed to loosen stiff joints (32), preparing subjects for further exercise. This was followed by a structured pool program of jogging, jumping jacks, marching, running, hurdling, and swimming with the assistance of flotation devices, along with punching, swinging, and rowing of the arms aimed at achieving the desired HR. In week 1, the program lasted 5 minutes, with the duration increased by 1 minute each week until week 6. For weeks 6–12, the structured program lasted for 10 minutes with increasing complexity of the routine each week.

The 3 aerobic gym stations initially lasted 5 minutes each and were graded in duration as described above. Maximum HR was estimated at 220, and HR was increased from 132 (∼60% of maximum HR) at week 6 to 165 (∼75% of maximum HR) at week 12. These relatively low target ranges with slow progression were selected for this pilot study as a safety precaution and to ensure that even children with significant impairment could meet the targets. HR was measured every 60 seconds during the gymnasium sessions using a telemetric device (Polar Vantage XL; Polar Instuments).

RPE were reported by the child at the end of each station in response to the question, “How hard did you work in this exercise?” The child rated their RPE while looking at a printed copy of the Borg scale. This self-reported scale is graded from 6 to 20 and uses descriptive cues for each category of exertion within the scale ranging from “very, very light” to “very, very heavy.” The RPE has been shown to be a reliable measure of physical strain (33–36), but has not previously been examined in children with arthritis. Subjects also were asked to rate their exercise-related pain on a VAS at the same time as rating their RPE.

Details on the routines followed at the 4 stations are as follows. At the cycle station, the child cycled at speeds of 60–100 revolutions per minute. The initial speed was determined at the habituation session as the one that the child believed he/she could sustain for the entire ride. The speed and resistance were adjusted to achieve the targeted HR. Similarly, on the treadmill the initial speed was determined at the habituation session, and the speed and slope were adjusted to achieve the targeted HR.

The Fitball (Ball Dynamics, Longmont, CO) station consisted of leg movements (marching, bouncing, rocking from heel to toe, foot tapping, and leg kicking) and arm movements (clapping, reaching, swinging, and raising). Once mastered, these were combined into more complex routines designed to achieve the target HR range (Klepper SE: personal communication).

Strength training consisted of progressive upper- and lower-limb static (isometric) and dynamic exercises using soft weights, and Thera-Band (Hygenic, Akron, OH) exercises concentrated on the biceps and triceps in the arms; hip abductors, adductors, and quadriceps in the legs; and the abdominal muscles. The intensity of exercise in the arms and legs was graded according to number of repetitions and maximal effort as a percentage of repetition maximum (RM). The RMs for the weights and Thera-Band were determined during habituation testing sessions (33). In weeks 5 and 6, the focus was on a 0.25 RM for 10 repetitions. In weeks 7–12, the number of repetitions was systematically increased to 20, and the RMs from 0.25 to 0.50.

Statistical analysis.

Data were analyzed descriptively with means, medians, and standard deviations for pretest and posttest scores for the exercise and functional measures. Paired Student's t-tests were used to assess the significance of changes in fitness parameters between the pretesting and posttesting sessions. Effect size (ES) and standardized response mean (SRM) were calculated to assess the magnitude of the observed differences (37–41). The ES and SRM are standardized measures that provide an estimate of the magnitude of the effect of the intervention in terms of standard deviations of the pretest (ES) or standard deviations of the change between measures (SRM). An ES or SRM of 0.2 is thought to represent a small effect, 0.5 a moderate effect, and ≥0.8 a large effect (37–41).


  1. Top of page
  2. Abstract
  7. Acknowledgements


Nine children were enrolled (5 girls). The median age was 9.4 years (range 8.0–11.1 years) (Table 1). Four subjects had oligoarticular-onset JIA with mild disease. Three subjects had polyarticular-onset JIA with some functional limitation. Subjects 5 and 6 had systemic arthritis with 10 and 40 active joints, respectively, at study enrollment; both had active hip disease and a CHAQ score of 3 at enrollment (Table 2).

Table 1. Characteristics of patients at commencement of study*
SubjectAge, yearsSexDisease onsetProgram completedSessions attended (max 24)Height, cmWeight, kgNSAIDCSMTXAnti-TNF
  • *

    NSAID = nonsteroidal antiinflammatory drug; CS = corticosteroids; MTX = methotrexate; anti-TNF = anti–tumor necrosis factor therapy (infliximab or etanercept); F= female; M = male; Oligo = oligoarticular; Poly = polyarticular.

Table 2. Disease activity, quality of life, and functional measures at commencement and on completion of training program*
SubjectTotal effused joints (lower limb)Total active joints (lower limb)CHAQJASIQOLHRQOL
  • *

    CHAQ = Childhood Health Assessment Questionnaire (score out of 3); JASI = Juvenile Arthritis Status Index (score out of 100); QOL = quality of life; HRQOL = health-related quality of life (score out of 10); Pre = prior to fitness training; Post = after fitness training; NA = not available.

11 (1)0 (0)1 (1)0 (0)0.25095.7974.
22 (2)2 (2)3 (2)3 (2)0099.595.
33 (1)3 (2)6 (2)6 (2)0.620.8793937.
419 (3)17 (2)40 (14)22 (2)21.7565696.23.446.1
630 (7)NA40 (13)NA3NANANA7.1NA5.8NA
70 (0)0 (0)2 (0)0 (0)0010010099.48.87.7
82 (0)2 (0)2 (0)2 (0)0NA1001007.8NA8.6NA
93 (0)2 (0)5 (1)3 (0)0.370.37100100NA9.9NA4.5

All subjects received nonsteroidal antiinflammatory medications at recruitment. Subject 4 was taking low-dose prednisone, 6 subjects were taking methotrexate, and 2 subjects were taking tumor necrosis factor blockers at the beginning of the study.


All patients completed pretesting. Both subjects with systemic arthritis dropped out and did not undergo posttesting. Subject 8 completed the study but not posttesting due to the inconvenience of traveling to the testing facility.

Fifteen exercise testing sessions were completed and were well tolerated, with only subject 4 requiring modification of the protocol due to limited hip mobility. Subject 3 was noted to have an effusion after testing in a knee joint that was noted to be active but not effused at the commencement of testing. Both subjects with systemic JIA reported pain during the testing sessions, with scores on a 10-point scale of 2.8 and 2.1 in one subject and 3.7 and 2.5 in the other. Subject 4 reported a score of 1.7 in the pretesting session. Pain scores were generally mild, and pain did not necessitate modification of the testing protocol.

Training sessions.

Subjects entered 1 of 2 exercise groups. The first group trained for the planned 12 weeks and the other group trained for 16 weeks due to missed sessions over a holiday period. Of the 7 subjects who completed the program, the median number of sessions attended was 15 of a maximum of 24 (range 7–23; mean ± SD 16.3 ± 5.4). Both subjects with systemic JIA had severe erosive hip disease and dropped out due to symptom exacerbation and prolonged fatigue after the exercise sessions.

Of the 123 exercise sessions, 62 (50.4%) were completed without any pain reported. The pain reported on VAS by each child is shown in Figure 1. Four subjects reported no pain in any sessions, 1 reported pain in 2 sessions, and 4 reported frequent pain. Subjects 5 and 6, with systemic arthritis, reported significant pain with every session. Subject 4, with severe polyarticular JIA, reported pain with every session that required modification of the program. Subject 9 reported frequent pain with a severity of up to 9.2 of 10, although this pain was not reported by instructors to hamper the child's participation.

thumbnail image

Figure 1. Pain scores as measured on a visual analog scale during exercise sessions. Data are shown as box plots. Each box represents the 25th to 75th percentiles. Lines outside the boxes represent the 10th and 90th percentiles. Lines inside the boxes represent the median. Actual pain scores are shown by the superimposed dot plot.

Download figure to PowerPoint

As seen in Table 3, the target HR was achieved with the stationary exercise bicycle and treadmill, but not with the Fitball station. Mean RPE also appeared higher in the stationary exercise bicycle and treadmill sessions than in the Fitball sessions. Correlations between HR and RPE are shown in Table 3. Treadmill sessions showed the strongest correlation, followed by the stationary bicycle. Overall, subjects with oligoarticular JIA appeared to have higher RPE during the gym exercises than those with polyarticular JIA (mean RPE 16.6 of 20 versus 10.1 of 20).

Table 3. Heart rate, rating of perceived exertion (RPE), and their correlation during training exercise*
ActivityHeart rateRPE out of 20Correlation coefficient (95% CI)
  • *

    Values are the mean ± SD unless otherwise indicated. 95% CI = 95% confidence interval; NA = not available.

Fitball131.9 ± 10.210.6 ± 2.90.46 (0.31–0.59)
Exercise bicycle146.2 ± 14.213.1 ± 3.50.55 (0.41–0.66)
Treadmill141.1 ± 12.212.9 ± 4.10.79 (0.71–0.84)
PoolNA11.2 ± 4.8NA

Supervising therapists observed that during the Fitball sessions, children required a significant effort to stay balanced on the apparatus to avoid falling off. This concentration on balance and complexity of the routine distracted subjects from exerting maximal effort. These difficulties were not overcome through familiarity with the apparatus. Fitball also aggravated hip arthritis in the 2 subjects with systemic arthritis, and required modification for subject 4 (who had polyarticular JIA with hip involvement).

No subject that completed the study experienced a definite worsening of arthritis. Three subjects had no change in active joint count and 4 improved. Subject 3's CHAQ score increased by 0.25, subject 4 showed a decrease of 2.8 in QOL, and subject 7 showed a decrease of 1.1 in HRQOL. For those that completed the study, the overall pattern was either an improvement or no change in functional and QOL parameters. Followup parameters were not available for those subjects that dropped out.

Effectiveness of training program.

Complete results from exercise testing at both sessions were available for 6 subjects (Tables 4 and 5). Energy cost of locomotion was reduced in most subjects with a lower VO2submax percentage at both 1.5 km/hour and 3.0 km/hour walking speeds after training, although the P values obtained from paired t-tests were nonsignificant. The ES and SRM for VO2submax percentage suggested moderate and small effect sizes at 1.5 km/hour and 3.0 km/hour, respectively.

Table 4. Results of exercise testing (power and fatigue) before and after training program*
SubjectRelative peak power, watts/kgRelative mean power, watts/kg% fatigue
  • *

    NA = not available; SRM = standardized response mean.

  • Statistics only include those patients for which pre- and posttesting are available.

Mean ± SD      
 Before6.5 ± 2.62.6 ± 1.64.5 ± 2.11.3 ± 0.855.4 ± 23.781.4 ± 17.7
 After7.9 ± 3.62.1 ± 1.74.4 ± 2.21.2 ± 0.967.4 ± 4.876.0 ± 22.3
Mean ± SD difference1.4 ± 1.1−0.5 ± 0.45−0.05 ± 0.7−0.12 ± 0.312.1 ± 23.3−5.4 ± 14.4
Effect size0.530.320.020.140.510.31
Table 5. Results of exercise testing before and after training program*
SubjectsRelative VO2at 1.5 km/hour (ml/kg/minute)VO2% at 1.5 km/hourRelative VO2at 3 km/hour (ml/kg/minute)VO2% at 3 km/hourAbsolute VO2peak(ml/minute)Relative VO2peak(ml/kg/minute)
  • *

    VO2 = oxygen consumption; VO2peak = peak oxygen consumption; NA = not available; SRM = standardized response mean.

  • Statistics only include those patients for which pre- and posttesting are available.

Mean ± SD      
 Before13.6 ± 5.745 ± 1915.8 ± 4.954 ± 221,196 ± 54932.1 ± 11.1
 After11.3 ± 2.536 ± 1415.9 ± 4.649 ± 241,298 ± 46433.7 ± 9.0
Mean ± SD difference−2.3 ± 4.3−9 ± 13−0.1 ± 3.9−6 ± 21102 ± 190−1.6 ± 4.6
Effect size0.400.470.

Absolute VO2peak increased in 4 of 6 subjects; the changes were not significant. The ES and SRM suggested a possible small effect. However, the relative VO2peak measures did not support this finding.

Absolute and relative peak anaerobic leg power were improved in all subjects with a moderate ES and SRM. Leg muscular endurance did not show any overall change.

Absolute and relative peak anaerobic arm power were reduced in all subjects with a significant reduction in relative peak power and a small to moderate ES and SRM. Arm muscular endurance showed no real change.

Satisfaction questionnaires were completed by 8 subjects and 5 parents. Overall, families reported a positive experience with increased fitness and energy levels. The subjects reported that the most enjoyable experiences were related to time spent in the pool, interactions with instructors, and the provision of small gifts as incentive for attendance. Parents reported that duration, frequency, and scheduling of evening and weekend sessions were suitable. Long travel times were highlighted by many as a disincentive to participation. With the exception of 2 subjects who withdrew from the study, most others said they would consider participation in a similar program again.


  1. Top of page
  2. Abstract
  7. Acknowledgements

This study was undertaken as a pilot study for a large randomized controlled study to test the hypothesis that fitness exercise is beneficial in the physical function and quality of life of children with JIA. We have shown that children with JIA are able to participate safely in a 12–16-week exercise program, which is consistent with findings from other studies of aquatic (15, 32) and land-based weight-bearing fitness programs (14, 41). These children can also undertake formal exercise testing protocols without experiencing severe pain or worsening of arthritis.

Some children with severe arthritis required modification of the exercise program, and those with severe hip disease were able to complete exercise testing but not the exercise program. This suggests that intensive weight-bearing training programs may not be feasible in children with JIA and severe hip involvement.

The Fitball program was included in the training schedule because it was hypothesized to cause less stress on the hip joints and provide an enjoyable form of aerobic training. However, subjects failed to achieve target HR ranges on the Fitball, and it aggravated existing hip symptoms. Exercise bicycle and treadmill programs achieved the target HR and did not worsen arthritis activity, suggesting that they would be acceptable forms of training for a larger study.

Published studies have generally used weekly or twice weekly protocols without clearly demonstrating improvements in fitness. It is believed that a minimum of twice weekly training is required for significant improvements in physical fitness (10). Only one uncontrolled study by Klepper (14) has examined a 3 times per week aerobic program in children with arthritis over 8-weeks and showed improved fitness by field testing. Based on the modest changes seen in our study, it would seem that training more often than once or twice weekly is necessary.

Although it is important to study programs with more frequent training sessions, geographic and logistic considerations and family time constraints have an impact on attendance frequency. Training centers closer to subjects' homes were identified from parent surveys as a factor that may improve adherence. Incorporation of a home-based component to exercise programs, possibly with the aid of videotaped or written instructions, may also increase exercise frequency. Future studies might best examine a combination of in-class fitness exercise once weekly with an additional twice weekly exercise session at home guided by an exercise videotape.

During the exercise training sessions, RPE measured by the Borg scale correlated well with measured HR during the treadmill and stationary bicycle sessions but not as well with the Fitball sessions, with correlation coefficients of 0.79, 0.55, and 0.46, respectively. These results suggest that RPE has potential as an indicator of physical strain in children with arthritis, and might be used to guide treadmill and stationary bicycle programs for these children.

A number of potentially beneficial outcomes were observed in this study. A reduced VO2submax percentage at both 1.5 and 3.0 km/hour was noted, suggesting a reduced energy cost of locomotion. The changes did not achieve statistical significance in this pilot study. The ES of 0.51 and 0.26, respectively, were consistent with a small to moderate effect of the intervention. Waters et al estimated that in untrained children between the ages of 6 and 12 years, O2 consumption was ∼28% of VO2peak at an average customary walking speed of 4.2 km/hour (21, 42). At the slower speeds of 1.5 and 3.0 km/hour tested in this study, O2 consumption was higher (45% and 54%, respectively, for the pretraining measures) and may reflect differences in the testing protocol used or may suggest that children with arthritis have an elevated energy cost of locomotion. There is no previous literature on the energy cost of locomotion in JIA or on changes in energy cost of locomotion following training programs in children. Energy cost of locomotion may be an important and modifiable target for future studies of exercise in JIA.

Children <14 years of age without arthritis have been reported to experience an increase in VO2peak between 7% and 26% following training, with most studies reporting an improvement <11% (43). Our study group showed a mean improvement of 7.1% in absolute VO2peak (ml/minute) but a reduction in relative VO2peak (ml/kg/minute) of 5.1%. These nonsignificant and inconsistent findings are likely a result of the small sample size of this pilot study.

Peak anaerobic power was significantly increased in the lower limbs but decreased slightly in the upper limbs of study subjects. This finding may reflect the design of the training program, which concentrated on lower-limb exercise. Anaerobic capacity was not appreciably changed in this study and may reflect the format or frequency of the training protocol. Measures of arthritis activity, functional status, and quality of life were generally stable in those subjects that completed the study, and overall satisfaction with the program was high.

Although it is important to remember that this is a small pilot study without a control group and only fair compliance with the exercise program, our results suggest that it is safe, feasible, and acceptable for children with arthritis who do not have severe hip involvement to participate in formal exercise testing and structured fitness programs. Furthermore, it is possible that these programs will result in an improvement in physical fitness and quality of life of children with arthritis. These findings have been incorporated into the design of a randomized controlled trial examining the role of fitness exercise in improving physical function and quality of life for children with JIA.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Supported by The Arthritis Society of Canada. Dr. Feldman holds a Canada Research Chair in Childhood Arthritis.


  1. Top of page
  2. Abstract
  7. Acknowledgements
  • 1
    Malleson PN, Fung MY, Rosenberg AM. The incidence of pediatric rheumatic diseases: results from the Canadian Pediatric Rheumatology Association Disease Registry. J Rheumatol 1996; 23: 19817.
  • 2
    Manners PJ, Bower C. Worldwide prevalence of juvenile arthritis: why does it vary so much? J Rheumatol 2002; 29; 152030.
  • 3
    Henderson CJ, Lovell DJ, Specker BL, Campaigne BN. Physical activity in children with juvenile rheumatoid arthritis: quantification and evaluation. Arthritis Care Res 1995; 8: 1149.
  • 4
    Bar-Or O. Pediatric sports medicine for the practitioner. In: KatzM, StiehmER, editors. Comprehensive manuals in pediatrics. New York: Springer-Verlag; 1983. p. 68.
  • 5
    Takken T, Hemel A, van der Net J, Helders PJ. Aerobic fitness in children with juvenile idiopathic arthritis: a systematic review. J Rheumatol 2002; 29: 26437.
  • 6
    Metin G, Ozturk L, Kasapcopur O, Apelyan M, Arisoy N. Cardiopulmonary exercise testing in juvenile idiopathic arthritis. J Rheumtol 2004; 31: 18349.
  • 7
    Alexander GJ, Hortas C, Bacon PA. Bed rest, activity and the inflammation of rheumatoid arthritis. Br J Rheumatol 1983; 22: 13440.
  • 8
    Wright V, Smith E. PT management of children with juvenile rheumatoid arthritis. In: WalkerJ, HelewaA, editors. Physical therapy in arthritis. New Jersey: W Saunders Publishers; 1996. p. 21163.
  • 9
    Minor MA. Arthritis and exercise: the times they are a-changin' [editorial]. Arthritis Care Res 1996; 9: 7981.
  • 10
    Minor MA. 2002 Exercise and Physical Activity Conference, St. Louis, Missouri: exercise and arthritis “we know a little bit about a lot of things …” Arthritis Rheum 2003; 49: 12.
  • 11
    Stenstrom CH, Minor MA. Evidence for the benefit of aerobic and strengthening exercise in rheumatoid arthritis. Arthritis Rheum 2003; 49: 42834.
  • 12
    Van den Ende CH, Vliet Vlieland TP, Munneke M, Hazes JM. Dynamic exercise therapy in rheumatoid arthritis: a systematic review. Br J Rheumatol 1998; 37: 67787.
  • 13
    Oberg T, Karsznia A, Gare BA, Lagerstrand A. Physical training of children with juvenile chronic arthritis: effects on force, endurance and EMG response to localized muscle fatigue. Scand J Rheumatol 1994; 23: 925.
  • 14
    Klepper SE. Effects of an eight-week physical conditioning program on disease signs and symptoms in children with chronic arthritis. Arthritis Care Res 1999; 12: 5260.
  • 15
    Takken T, van der Net J, Kuis W, Helders PJ. Aquatic fitness training for children with juvenile idiopathic arthritis. Rheumatology (Oxford) 2003; 42: 140814.
  • 16
    Takken T, van der Net J, Helders PJ. Relationship between functional ability and physical fitness in juvenile idiopathic arthritis patients. Scand J Rheumatol 2003; 32: 1748.
  • 17
    Washington RL, Bricker JT, Alpert BS, Daniels SR, Deckelbaum RJ, Fisher EA, et al. Guidelines for exercise testing in the pediatric age group: from the Committee on Atherosclerosis and Hypertension in Children, Council on Cardiovascular Disease in the Young, the American Heart Association. Circulation 1994; 90: 216679.
  • 18
    Noonan V, Dean E. Submaximal exercise testing: clinical application and interpretation. Phys Ther 2000; 80: 782807.
  • 19
    Hoofwijk M, Unnithan V, Bar-Or O. Maximal treadmill peformance of children with cerebral palsy. Pediatr Exerc Sci 1995; 7: 30313.
  • 20
    Bar-Or O. Role of exercise in the assessment and management of neuromuscular disease in children. Med Sci Sports Exerc 1996; 28: 4217.
  • 21
    Waters RL, Mulroy S. The energy expenditure of normal and pathologic gait. Gait Posture 1999; 9: 20731.
  • 22
    Giannini MJ, Protas EJ. Exercise response in children with and without juvenile rheumatoid arthritis: a case-comparison study. Phys Ther 1992; 72: 36572.
  • 23
    Singh G, Athreya BH, Fries JF, Goldsmith DP. Measurement of health status in children with juvenile rheumatoid arthritis. Arthritis Rheum 1994; 37: 17619.
  • 24
    Wright FV, Kimber JL, Law M, Goldsmith CH, Crombie V, Dent P. The Juvenile Arthritis Functional Status Index (JASI): a validation study. J Rheumatol 1996; 23: 106679.
  • 25
    Feldman BM, Grundland B, McCullough L, Wright V. Distinction of quality of life, health related quality of life, and health status in children referred for rheumatologic care. J Rheumatol 2000; 27: 22633.
  • 26
    Bar-Or O. The Wingate anaerobic test: an update on methodology, reliability and validity. Sports Med 1987; 4: 38194.
  • 27
    Andreacci JL, LeMura LM, Cohen SL, Urbansky EA, Chelland SA, von Duvillard SP. The effects of frequency of encouragement on performance during maximal exercise testing. J Sports Sci 2002; 20: 34552.
  • 28
    Takken T, van der Net J, Helders PJ. Aerobic exercise testing in juvenile rheumatoid arthritis patients. Clin Exerc Phys 2002; 4: 3843.
  • 29
    Tirosh E, Bar-Or O, Rosenbaum P. New muscle power test in neuromuscular disease: feasibility and reliability. Am J Dis Child 1990; 144: 10837.
  • 30
    Takken T, van der Net J, Helders PJ. The reliability of an aerobic and an anaerobic exercise tolerance test in patients with juvenile onset dermatomyositis. J Rheumatol 2005; 32: 7349.
  • 31
    Van Mil E, Schoeber N, Calvert RE, Bar-Or O. Optimization of force in the Wingate Test for children with a neuromuscular disease. Med Sci Sports Exerc 1996; 28: 108792.
  • 32
    Bacon MC, Nicholson C, Binder H, White PH. Juvenile rheumatoid arthritis: aquatic exercise and lower-extremity function. Arthritis Care Res 1991; 4: 1025.
  • 33
    Bar-Or O. Rating of percieved exertion in children and adolescents: clinical aspects. In: LjunggrenG, DornicS, editors. Psychophysics in action. New York: Springer-Verlag; 1989. p. 10513.
  • 34
    Lamb KL. Children's ratings of effort during cycle ergometry: an examination of the validity of two effort rating scales. Pediatr Exerc Sci 1995; 7: 40721.
  • 35
    Ward DS, Bar-Or O. Usefulness of RPE scale for exercise prescription with obese youth [abstract]. Med Sci Sports Exerc 1987; 19: S15.
  • 36
    Pfeiffer KA, Pivarnik JM, Womack CJ, Reeves MJ, Malina RM. Reliability and validity of the Borg and OMNI rating of perceived exertion scales in adolescent girls. Med Sci Sports Exerc 2002; 34: 205761.
  • 37
    Cohen J. Statistical power analysis for the behavioural sciences. 8th ed. New York: Academic Press; 1977.
  • 38
    Liang MH, Fossel AH, Larson MG. Comparisons of five health status instruments for orthopedic evaluation. Med Care 1990; 28: 63242.
  • 39
    Katz JN, Larson MG, Phillips CB, Fossel AH, Liang MH. Comparative measurement sensitivity of short and longer health status instruments. Med Care 1992; 30: 91725.
  • 40
    Beaton DE, Hogg-Johnson S, Bombardier C. Evaluating changes in health status: reliability and responsiveness of five generic health status measures in workers with musculoskeletal disorders. J Clin Epidemiol 1997; 50: 7993.
  • 41
    Kirchheimer JC, Wanivenhaus A, Engel A. Does sport negatively influence joint scores in patients with juvenile rheumatoid arthritis? An 8-year prospective study. Rheumatol Int 1993; 12: 23942.
  • 42
    Waters RL, Hislop HJ, Thomas L, Campbell J. Energy cost of walking in normal children and teenagers. Dev Med Child Neurol 1983; 25: 1848.
  • 43
    Braden DS, Carroll JF. Normative cardiovascular responses to exercise in children. Pediatr Cardiol 1999; 20: 410.