Arthritis & Rheumatism (Arthritis Care & Research) Vol. 49, No. 3, June 15, 2003, pp 435– 443 DOI 10.1002/art.11055 © 2003, American College of Rheumatology

SPECIAL ARTICLE

Exercise and Fitness in Children With Arthritis: Evidence of Benefits for Exercise and Physical Activity SUSAN E. KLEPPER

Introduction A sedentary lifestyle is known to be a significant risk factor for mortality from all causes in adults (1). Increased physical activity (PA) is associated with improved health and delayed mortality from chronic disease (2). The increasing prevalence of risk factors for cardiovascular disease in children (3) and the potential tracking of PA habits into adulthood (4) influenced the Healthy People 2010 objectives that call for all children to increase their PA (5). Children with juvenile arthritis (JA) experience joint swelling, pain, and limited mobility, which contribute to decreased PA, fitness, and function, increasing their risk for disability from long-term disease (6). Research documents significant cardiovascular and musculoskeletal deficits and delayed motor development in children with JA (7–21). Although impairments are most pronounced in children with severe articular disease, poor fitness is also reported in those with mild symptoms, and deficits often persist long after arthritis resolves (12,13). Studies show that children with JA can undergo exercise testing (7–17) and participate in sports and physical conditioning programs (22–27) without disease exacerbation. However, there is a paucity of evidence demonstrating the effectiveness of PA and exercise on physical fitness, function, and general health in children with JA. This article reviews what is known about muscle structure and function, aerobic capacity, and PA in children with JA and their response to exercise programs. Areas of limited knowledge and suggestions for future research are discussed. A search of the literature was conducted using

Presented at the International Conference on Health Promotion and Disability Prevention for Individuals and Populations With Rheumatic Disease: Evidence for Exercise and Physical Activity, St. Louis, MO, March, 2002. Susan E. Klepper, PhD, PT: Columbia University, New York, New York. Address correspondence to Susan E. Klepper, PhD, PT, Program in Physical Therapy, Columbia University, 710 West 168th Street, 8th Floor, New York, NY 10032. Submitted for publication June 20, 2002; accepted in revised form October 19, 2002.

Medline via Ovid for studies examining exercise capacity and the effects of exercise in children with JA. The search was limited to studies involving humans and written in the English language. This was not an exhaustive review, and because many of the studies are published only in abstract form, no attempt was made to assess or rate the quality of the studies cited.

Aerobic capacity in children with arthritis Research documents reduced aerobic fitness in children with JA. A summary of these findings is shown in Table 1. A recent systematic review by Takken and colleagues (21) identified 5 studies (7,8,28 –30) that directly measured peak oxygen consumption (VO2peak; in ml-kg⫺1-min⫺1) in children with JA during progressive graded exercise testing on a treadmill or bicycle ergometer and compared their aerobic fitness with healthy controls or reference values. A metaanalysis using pooled data (to determine effect sizes) from these studies indicated that relative VO2peak was 21.8% lower in children with JA. Giannini and Protas (7), using a progressive graded exercise test on a bicycle ergometer, reported that children with JA (n ⫽ 16) had significantly lower VO2peak and peak workload than controls matched for age, sex, and body size. A later study with 30 subjects confirmed these findings and indicated that children with JA had shorter exercise time and lower peak heart rate (HRpeak), but higher mean submaximal HR and VO2peak than matched controls (8), suggesting children with JA work at a higher percentage of their VO2peak than typical children during routine PA. Similar findings have been reported for young adults diagnosed with a seronegative juvenile spondylarthropathy in childhood. Hebestreit and colleagues (13) measured maximum power (Wmax) and VO2peak in 21 subjects, aged 6 –24 years, with HLA–B27 positive spondylarthropathies, whose disease was inactive or in remission. Peak VO2 and Wmax were significantly lower in subjects older than 18 years compared with matched controls. When VO2peak was normalized to height and weight, mean values for the whole group were also significantly lower than controls. VO2peak values were not significantly associated with disease severity or precautions against PA during previous 435

436

Klepper

Table 1. Comparison of physiologic and performance variables between children with and without juvenile arthritis and relationship to disease duration, activity, and severity*

Variable

Children with JA versus healthy children (reference)

Peak VO2

Lower (7,13,29,30) No difference (28)

Peak heart rate Submaximal heart rate Peak workload Peak anaerobic power

Lower (8) Higher (8) Lower (7) Lower (13) No difference (28) Lower (10,15,16)

Performance on field tests of aerobic or anaerobic fitness Muscle strength Muscle thickness Muscle endurance Daily energy expenditure (PA) Participation in sports

Lower (9,11,12,40) No difference (14) Less (11,12,39) No difference (14) No difference (13,35)

Significantly related to disease duration? (reference) Yes (13)

Significantly related to disease activity (JC)? (reference) No (7)

No (8) No (7)

Significantly related to disease severity (ASS or MD global assessment)? (reference) Yes (MD global assessment) (28) No (7) No (8) No (7) No (28)

No (10,15)

No (9) Yes (39)

No (9) Yes (12)

Yes (12)

Lower (35) No difference (13)

* JA ⫽ juvenile arthritis; JC ⫽ count of joints with active arthritis; ASS ⫽ articular severity score; MD ⫽ physician; Peak VO2 ⫽ peak oxygen consumption; PA ⫽ physical activity.

periods of active disease. However, subjects with active disease lasting longer than 2 years had significantly lower VO2peak and Wmax values than controls. Subjects who reported feeling limited in their aerobic exercise tolerance in the past year had lower VO2peak. This suggests that disease duration and self efficacy for exercise may be more important than disease severity in predicting aerobic fitness. Children with JA also perform poorly on standardized field tests of physical fitness. Klepper and colleagues (10) used the standardized 9 minute walk-run test (9MWRT) to compare aerobic function in 20 children with polyarticular JA, aged 6 –11 years, and matched (age, sex, body mass index) controls. Subjects with JA scored significantly below controls and below the 16th percentile on the norms for the 9MWRT. Although the controls were not randomly selected, their mean score was at the 50th percentile, suggesting this sample represented typical children in this age group. Only one study found no difference in aerobic fitness between children with and without JA. Malleson et al (28) reported that values for VO2peak and peak anaerobic power output of the legs were similar in 31 children with JA and 16 healthy controls, aged 8 –17 years. However, the mean VO2peak values for both groups were below published age- and sex-based reference values, suggesting that the aerobic capacity of both groups was impaired.

Possible factors limiting aerobic capacity in children with arthritis Multiple pathophysiologic factors specific to the disease may limit aerobic capacity and increase the metabolic cost

of exercise in children with JA (31). Anemia, common in children with JA, results in decreased oxygen-carrying capacity of the blood and, in the presence of muscle atrophy, results in poor utilization of oxygen by the exercising muscles, higher mixed-venous oxygen content, and low VO2peak. Jones (32) reported that increased cardiac output during exercise testing in children compensates for hemoglobin levels as low as 8 gm/dl when there are no comorbidities. However, the subjects studied by Klepper et al (10) had hemoglobin levels ⱖ10 gm/dl, suggesting that factors other than anemia may limit exercise capacity in children with JA. There is some evidence that muscle atrophy and weakness contribute to decreased exercise capacity in children with JA. Knook and associates (33) examined lung function, respiratory muscle strength, thorax expansion, spinal mobility, and disability in 31 children with polyarticular and systemic JA. All measures of lung function, including peak expiratory flow (PEF), forced vital capacity (FVC), maximum inspiratory pressure (Pimax), and maximum expiratory pressure (Pemax), were significantly reduced compared with reference values or paired controls. Although thoracic and spinal mobility were normal, and signs of intrapulmonary restrictive or obstructive disease were absent, respiratory muscle strength (Pimax and Pemax) was significantly reduced and positively correlated with FVC and PEF. Disability was inversely correlated with FVC, PEF, and Pemax. These findings suggest respiratory muscle weakness may contribute to impaired lung function, reduced aerobic capacity, and increased disability in children with JA. Generalized mus-

Exercise and Fitness in Children With Arthritis cle weakness and joint stiffness in children with arthritis may also result in poor mechanical efficiency and increased energy expenditure during submaximal exercise tests that involve bicycling, walking, or running. Gait abnormalities, including decreased velocity, cadence, and stride length are common in children with JA and may affect the child’s performance on timed walk or run tests (10,34). Additionally, delayed motor development has been found in young children with JA (18,19). Most studies show that aerobic fitness is not significantly related to disease activity or severity. Neither Giannini and Protas (7,8) nor Klepper et al (10) found a significant association between active joint count (JC) or a cumulative articular severity score (ASS; sum of joint swelling, pain on passive motion, tenderness, and limited motion) and aerobic fitness in children with JA. Malleson et al (28) also failed to find a significant correlation between VO2peak or maximal anaerobic power and JC or physician rated global disease activity. However, VO2peak (controlled for age, sex, and sum of skinfold thicknesses) was negatively correlated with physician rated global disease severity. Because the physician’s assessment was based on the child’s global disease severity, the rating may take into account systemic disease and persistence of symptoms, factors associated with poor aerobic capacity (12,13).

Physical activity and aerobic fitness The relationship between habitual PA and laboratory measures of aerobic capacity in children with JA is unclear. The limited data suggests that overall energy expenditure may not be reduced in young children with JA compared with their healthy peers. Henderson et al (35) examined PA over 3 days in 23 children with JA and controls, aged 5–11 years, using the University of Cincinnati Motion Sensor, Caltrac accelerometer, and a physical activity diary kept by the parent. They found the 2 groups did not differ significantly in the overall energy expenditure. However, compared with controls, children with JA spent more time sleeping and engaged in fewer strenuous PAs and organized sports. Similar findings were reported by Hebestreit and associates (13), who measured energy expenditure with a 7-day recall questionnaire (SDR) in 21 subjects, aged 6 –24 years, with spondylarthropathies whose disease was inactive or in remission. Subjects were asked about their current participation in organized and recreational sports, rated the frequency of their PA, and compared their PA with that of their friends. They also rated the impact of physical impairments on their work capacity and ability to be physically active during the previous year. Data for the group as a whole showed energy expenditure and participation in organized sports were similar in subjects and controls that were normally active, but not involved in highly competitive sports. There were no significant associations between PA and either a history of severe physical limitations or physician restrictions on PA during periods of active disease. In addition, while subjects with disease duration exceeding 2 years had significantly lower VO2peak values than controls, their reported PA was not significantly less than controls.

437 The failure to show a definite association between exercise capacity and recalled PA (SDR) level may be explained in a number of ways. Laboratory measures of aerobic capacity may be poorly related to habitual PA levels in younger children (36). Alternately, it is possible that the control subjects were no more fit or active than the subjects with arthritis. It is also possible the SDR did not accurately reflect the subjects’ actual energy expenditure over the previous week. The validity of self reported PA recall questionnaires is generally poor in children younger than 11 years of age (37). In fact, Hebestreit et al (13) reported a significant relationship between PA (SDR) and VO2peak in subjects older than 18 years, whose aerobic capacity was significantly lower than controls, suggesting that PA decreases in adolescents with JA, as it does in typical children. Further research is needed to understand the PA patterns of children and youth with JA and the relationships among PA, disease activity, general health, and fitness in this population.

Muscle function in children with arthritis The available research indicates that arthritis in childhood may result in significant muscular deficits (Table 1). Vostrejs and colleagues (38) found differences of more than 10% in peak isometric quadriceps torque between the involved and noninvolved legs of children with monarticular (knee) JA. Giannini and Protas (9) also found significantly lower peak isometric knee extensor torque in 28 subjects with polyarticular and pauciarticular JA, aged 7–17 years, compared with controls matched for age, sex, and body surface area. Most recently, Hedengren et al (20) examined lower extremity isometric joint torque in 6 children, aged 5–15 years, with polyarticular JA and 6 age- and sex-matched controls. The JA group had significantly lower torque values for the tibialis anterior and triceps surae muscles. Differences in quadriceps torque approached significance, and the quadriceps/hamstrings ratio was lower in the JA group (0.79/1.00) than in the controls (1.00/1.00), suggesting that the quadriceps may be less capable than the hamstrings in generating muscle force. Several studies examined the relationship between muscle strength and disease symptoms. Giannini and Protas (9) did not find a significant relationship between muscle torque and either JC or the ASS, suggesting that multiple factors influence muscle function in children with arthritis. Two studies from Lindehammar and associates (11,12) compared muscle structure and function in subjects with and without JA and examined the relationship between strength and disease activity and severity longitudinally over a 2-year period. In the first study, Lindehammar and Backman (11) compared isometric strength of the knee extensors, elbow flexors, wrist extensors, and ankle dorsiflexors; isokinetic strength of the ankle dorsiflexors; and thickness of the quadriceps, using ultrasound, in 20 children with JA, aged 8 –17 years, with either matched controls (age and sex) or reference values for typically developing children. Muscle strength, thickness, and power were all significantly less in the children with JA, and deficits were found in both the dominant and nondomi-

438 nant limbs. Although deficits were most pronounced in muscles near inflamed joints, decreased strength was also found in distant muscles, suggesting generalized muscle weakness in children with JA. The coincidence of decreased strength and thickness also indicates this is true muscle weakness. Lindehammar and Sandstedt (12) followed these subjects, repeating the same measures every 3 months over a 2-year period to examine how variations in disease activity (JC) and severity (ASS) influenced muscle strength and bulk over time. Declines in quadriceps muscle strength and bulk were most severe in children with either polyarticular disease or persistent arthritis of the hip or knee. During periods of dramatically increased inflammation, muscle strength and bulk declined significantly and in parallel. While recovery also occurred in parallel once inflammation resolved, neither strength nor bulk returned to premorbid levels, indicating weakness and atrophy persist even after arthritis resolves. Vostrejs and Hollister (39) also reported persistent muscle atrophy long after resolution of arthritis and return to normal function. They examined thigh circumference bilaterally in 32 children with pauciarticular JA and a history of arthritis in 1 knee who were followed through routine clinic visits for an average of 4.5 years. All patients were placed on a home program of range of motion (ROM) and isometric exercise and encouraged to perform activities that promote bilateral leg strength, including riding a bike and climbing stairs. Older children used weights during periods of inactive disease. The sample was divided into 2 groups, based on disease onset before or after the age of 3 years. The results showed significant muscle atrophy in the involved leg long after disease remission. Deficits were significantly greater in subjects with disease onset before the age of 3 years, suggesting arthritis in a very young child may interfere with a critical period of muscle cell division. A study by Dunn supports this premise (40). She compared grip strength (using a modified sphygmomanometer) in 13 children with JA and upper extremity involvement, aged 3–7 years, with normal values (based on 6-month intervals in a cohort of 273 healthy children between the ages of 3 years and 7 years, 11 months). Subjects with JA scored significantly below the age-based reference values. Data for typical children showed that grip strength increased with age and did not vary significantly between the preferred and nonpreferred hand. In contrast, while strength values for the right hand of children with JA followed a relatively even pattern across time, data for the left hand were less consistent. These findings lend support to those of Vostrejs and Hollister (39), suggesting that chronic arthritis in early childhood may alter the normal course of muscle structure and function. Children with arthritis also score poorly on standardized fitness tests of trunk or lower extremity muscle strength and power. Klepper et al (10) reported that children with polyarticular JA, aged 6 –11 years, scored significantly below matched controls and reference values on a 1 minute timed sit-up test of abdominal strength, endurance, and power. Fan et al (15) reported similar findings in their study of 20 girls with JA, aged 6 –16 years. They examined the relationship between isokinetic knee exten-

Klepper sion torque, measured on a Kin-Com dynamometer, and lower limb functional performance, assessed by the Canadian Fitness Award timed 50 meter run. Age and running time were used to classify the child’s fitness according to 1 of 4 categories (excellent, gold, silver, bronze). The mean running time for this group was 11.51 seconds, placing 75% of the subjects below the bronze level. Concentric quadriceps torque was moderately, but significantly related to subjects’ run time. These findings were later confirmed by Wessel et al (16) in a study of 32 girls with JA.

Factors responsible for muscle weakness and atrophy Muscle weakness and atrophy in adults with RA is thought to stem from multiple factors, including alterations in anabolic hormone and cytokine production, increased resting energy metabolism, and derangement in protein metabolism, as well as motor unit inhibition from joint swelling and pain and deconditioning from disuse. Similar changes in body composition and metabolism have been observed in children and adults with arthritis. Knops et al (41) documented significant increases in energy metabolism in 33 children with JA, 6 –18 years of age. Resting energy expenditure (REE) in children with systemic disease was 18% higher per kilogram body weight and 21% higher per kilogram fat-free mass. Smaller increases (8%) in REE were found in children with polyarticular and pauciarticular disease. Alterations in protein metabolism may also contribute to muscle atrophy. Henderson and Lovell (42) found evidence of protein energy malnutrition in 36% of 28 children with JA, aged 5–16 years. Of those, 90% had signs of extreme protein degradation and muscle wasting. Although their skinfold thickness was within the normal range, arm muscle area and arm muscle circumference were decreased, suggesting loss of muscle mass rather than subcutaneous fat. Biopsies of atrophied muscles in adults with RA show a loss of type II fast twitch, anaerobic muscle fibers, which are responsible for force production in brief tests of strength and are particularly sensitive to fatigue (43). Type II muscle atrophy may also occur in children with JA. Oberg and colleagues (14) reported abnormal electromyogram (EMG) responses to electrically induced muscle fatigue in children with JA. The patterns were similar to those seen in adults with work-related pain syndromes. In isokinetic tests of the ankle dorsiflexors in children with JA, Lindehammar and Backman (11) found that strength deficits were greater at higher velocities, suggesting decreased muscle power possibly due to type II fiber atrophy. In summary, these studies indicate that children with JA have generalized weakness that is most pronounced in muscles near inflamed joints. Muscle deficits appear to be similar in children and adults with inflammatory arthritis. These deficits may be greater in children with disease onset at a young age, may persist long after arthritis resolves, and may be associated with decreased function.

Effects of physical activity and exercise Exercise and PA are considered mainstays of the physical management of childhood arthritis. However, there is little

Exercise and Fitness in Children With Arthritis research demonstrating the effectiveness of specific exercise modes or regimens on disease activity, physical status, or function in children with JA. Exercise recommendations for children are often based on the personal philosophy of the physician or therapist, anecdotal reports, or occasionally evidence from studies of adults with arthritis. For example, aquatic programs are frequently recommended over weight-bearing exercise because of the belief that the buoyancy of water places less stress on the joints, allowing the child to increase ROM, strength, and mobility without increased pain or damage to inflamed joints. On the other hand, weight-bearing physical activity is known to be a significant determinant of bone width and density in children with JA (44). The effectiveness of water or land-based exercise in reducing impairments and improving function in children with JA has not been well studied. The literature reveals few efficacy studies, many of which are pilot projects with samples ranging from 6 to 12 subjects. These studies vary considerably in design, measurements, and the mode, structure, and length of the intervention. Exercise programs varied in duration from 30 to 60 minutes, frequency from 1 to 3 times a week, and overall program length from 6 to 20 weeks. Although some studies employed a control group or phase, none used randomized assignment, and few included children with severe systemic disease. The findings of these studies are shown in Table 2.

Effects of physical activity and exercise on disease activity and severity The short-term safety of aquatic and land-based PA and exercise programs for children with JA is documented. Takken and colleagues (23), in an uncontrolled pilot study of 10 children with JA, reported a significant (P ⬍ 0.05) decrease in the general symptoms subscale of the Juvenile Arthritis Quality of Life Questionnaire (JAQQ) following an aerobic aquatic program, 1 hour a week for 15 weeks. Pain, measured on a visual analog scale (VAS), decreased 17%, although the change was not significant. Four studies examined the effect of land-based exercise programs on disease activity and severity. Moncur et al (45) reported significantly lower JC (P ⬍ 0.025) in 7 adolescents and young adults with JA who completed a 12-week stationary cycling program 3 times a week. In another study, Fisher et al (17) studied the effects of an 8-week program of resistance exercise in 6 children with JA, aged 5–12 years, compared with a JA control group that did not exercise. The program was individualized for each child based on strength and endurance measures at baseline. Following the program, the JA exercise group reported a significant reduction in pain (42%) and the number of medications used (25%), whereas the control group had increased pain (74%) and medication use (54%). Lower plasma levels and production of cytokines and inflammatory mediators in these children after training suggest that resistance exercise may play a role in reducing inflammation in JA (26,27). Klepper (22) found similar reductions in disease activity and severity in 25 children with JA, aged 8 –17 years, after an 8-week exercise program. Subjects were assessed for

439 active JC and the ASS at study entry, after an 8-week control (no exercise) period, and after the 8-week exercise program. The 60-minute program consisted of a 10-minute warm up period, 25 minutes of low-impact aerobic exercises, 15 minutes of strengthening exercises, and 10 minutes of cool down and stretching. Subjects participated in a group exercise session with an instructor once a week and performed the exercise twice a week at home using a videotape of the program. Joint count and the ASS increased slightly during the control period, possibly due to normal variations in disease activity. In contrast, significant reductions (P ⬍ 0.05) were found in the ASS (42%) and JC (39%) between the pre- and postexercise tests. Decreases in pain (VAS) of 16% were similar in magnitude to those reported by Takken et al (23) after an aquatic program. Finally, Feldman and colleagues (46) reported that arthritis activity and pain decreased in 6 subjects who completed an 8-week circuit training program that combined exercise in water and on land, although 2 children with severe JA and active hip disease dropped out of the program. The available information thus supports the positive effects of graded exercise in reducing disease symptoms in children with JA. Although the current data suggest that land-based exercise programs may result in greater improvements in functional abilities than aquatic programs, one must keep in mind that the literature is very limited, with few randomized controlled studies. Further research is needed to understand the differential effects of specific modes of exercise in children with JA, as well as the optimal exercise intensity, duration, and frequency.

Effects of physical activity and exercise on musculoskeletal and cardiorespiratory impairments Studies describing the effects of PA and exercise on the musculoskeletal and cardiorespiratory status and function in children with JA also indicate benefits from both waterand land-based exercise. Bacon and colleagues (47) found significant increases in hip mobility, but no other lower extremity motion, in 11 subjects with JA, aged 4 –13 years, who completed a 6-week, 12 session aquatic exercise program. Klepper also found significant improvement in the mean score on the standardized sit and reach test of hamstring flexibility in 20 children with arthritis following an 8-week land-based conditioning program (22). Both aquatic and land-based programs appear to be effective in improving muscle strength in children with JA. Oberg et al (14) reported improved muscle function in children with JA who performed gymnastics and pool training for 40 minutes, twice a week for 3 months. They examined maximum isometric force and endurance of the right trapezius and quadriceps muscles in 10 children, aged 7–15 years, and 10 age- and sex-matched controls. Quadriceps strength increased significantly in both groups after the program, indicating there was no difference in trainability between patients and controls. Also, EMG response to localized muscle fatigue that was abnormal in the JA group before exercise showed a tendency toward normalization following the program, suggesting that

Circuit aerobic exercise

Feldman et al, 2000 (46)

No change in 6MWRT distance No change in VO2peak 1Efficiency of gait –

–

–

1Anaerobic power/legs

–

1Quadriceps and hamstrings strength

1Quadriceps and hamstrings strength

1Quadriceps and hamstrings strength

1Quadriceps strength –

–

–

Muscle strength

1Quadriceps and hamstrings endurance

1Quadriceps and hamstrings endurance

1Quadriceps and hamstrings endurance

1Quadriceps endurance

–

–

Muscle endurance

–

–

1Quadriceps contraction speed

1Quadriceps contraction speed

1Quadriceps contraction speed

–

–

–

–

Muscle contraction

–

–

2Pain 1QOL

2Pain 2Medications Modulates plasma levels of proinflammatory mediators 2Pain 2Medications Modulates plasma levels of proinflammatory mediators 2Pain 2Medications Modulates plasma levels of proinflammatory mediators 2Pain

2JC 2ASS

2JC

Disease signs and symptoms

* ROM ⫽ range of motion; VO2peak ⫽ peak oxygen consumption; JC ⫽ count of joints with active arthritis; HR ⫽ heart rate; RHR ⫽ resting heart rate; 9MWRT ⫽ 9 minute walk-run test; ASS ⫽ articular severity score; 6MWRT ⫽ 6 minute walk-run test; QOL ⫽ quality of life.

Group pool aerobic exercise

Takken et al, 2001 (23)

–

1VO2peak 1HRmax

Group land-based resistance exercise

Venkatraman et al, 1999 (27)

–

1VO2peak 1HRmax

Group land-based resistance exercise

Velazquez et al, 1999 (26)

1VO2peak 1HRmax

Group land-based aerobic exercise

Group land-based resistance exercise

1Sit and reach score (hamstring flexibility) –

19MWRT distance

Group pool exercise

1Hip ROM

1Exercise HR Improved recovery to RHR –

Group pool exercise

–

ROM/flexibility

1VO2peak

Aerobic capacity of performance

Stationary bicycle

Type of program

Fisher et al, 1999, 2001 (17,25)

Oberg et al, 1994 (14) Klepper, 1999 (22)

Moncur et al, 1990 (45) Bacon et al, 1991 (47)

Author, year (reference)

Table 2. Effects of physical activity and exercise in children with juvenile arthritis*

440 Klepper

Exercise and Fitness in Children With Arthritis strengthening exercises may improve the function of type II muscle fibers and possibly reduce muscle fatigue in children with JA. Programs that provide an individualized training regimen within a group exercise format appear to produce the most dramatic change in muscle function. Fisher et al (17) conducted a pilot study in which they measured maximal isometric strength, endurance, and contraction speed of the quadriceps and hamstrings in 11 children with JA. Six of the JA subjects (mean age 10 ⫾ 2.7 years) performed resistance exercise 3 times a week for 8 weeks, while the other 5 subjects (mean age 8.8 ⫾ 1.8 years) served as controls. After training, the JA exercise group showed significant improvements in quadriceps strength (48%) and endurance (32%), hamstrings strength (99%) and endurance (59%), and contraction speed (46 – 69%) compared with decreases of 42– 45% in muscle function in the control group. These subjects also demonstrated significant reductions in JC after training, indicating resistance exercise improves muscle function without disease exacerbation. These findings were confirmed in a subsequent study of 19 subjects with JA (25). Feldman et al (46) also conducted a pilot study to examine the impact of group training programs in which each child’s exercise regimen was individualized and progressed under the supervision of a pediatric physical therapist. Six subjects, who completed the program and all tests, exercised for 8 weeks, once to twice a week. The regimen combined water exercise (45 minutes) and circuit training on land in which subjects exercised for up to 10 minutes each on a treadmill, therapy ball, and stationary bicycle. Peak anaerobic power in the legs, measured with the Wingate protocol, improved significantly. The available evidence also demonstrates that children with JA who engage in moderately vigorous activity for at least 6 weeks can improve their aerobic fitness. The mode of exercise appears to be less important than exercise frequency. Bacon et al (47) reported that subjects who exercised in a pool twice a week for 6 weeks showed significantly higher peak exercise HRs and a fuller recovery to resting HR in the 5 minutes after exercise. On the other hand, Takken et al (21) found no significant improvement on the 6 minute walk test after 15 weeks of aquatics performed only once a week. Feldman et al (46) also found no significant improvement in VO2peak in subjects who participated in 8 weeks of combined water and land exercise once to twice a week, although 5 of the 6 subjects showed decreased energy cost of walking, measured as submaximal oxygen uptake. Studies of land-based PC programs, in which exercise frequency was 2–3 times a week, also demonstrated significant improvements in aerobic fitness. Moncur et al (45) reported significant increases in VO2peak and mean exercise time after 12 weeks, 36 sessions of stationary cycling. Klepper (22) found significant increases in 9MRWT distance in children with JA who completed at least 18 exercise sessions over 8 weeks. Even Fisher and colleagues (17) found significant improvements in VO2peak in their subjects after 8 weeks of resistance exercise, 3 times a week.

441

Effects of physical activity and exercise on function and quality of life Another important consideration in judging the effectiveness of any intervention is the impact on function and quality of life. Pain and impaired joint mobility, muscle weakness, and poor aerobic capacity are often used to explain functional limitations and poor health related quality of life in children with JA. It follows that improvements in these parameters may translate to improved function, self efficacy, and overall satisfaction with life. Of the 3 studies that examined the impact of exercise on function in children with JA, only 1 showed significant improvement. Fisher et al (17) reported significant improvement in functional status (32%), performance on timed tasks (10%), and disability (20%) after lower extremity resistance training. The evidence for aquatic training is less promising. In contrast, Takken et al (23) reported that their subjects showed no significant change in function (Childhood Health Assessment Questionnaire) or health-related quality of life (JAQQ) after 15 weeks of aquatic aerobic training. Bacon and colleagues (47) hypothesized that an aquatic exercise program would result in increased lower extremity ROM and strength and secondary improvements in gait, balance, and performance of timed functional tasks (25 foot run, 100 foot walk, 13 step stair climb, and time to rise from supine to stand). Despite a significant increase in hip mobility, they found no significant change in functional abilities. They did find some improvement in walking velocity, cadence, and stride length, possibly due to increased hip mobility. This warrants further study, because gait abnormalities are common in children with JA (34) and may contribute to their fatigue and pain during routine daily tasks.

Summary and implications for making physical activity or exercise recommendations The literature suggests that compared with healthy children, children with JA have decreased aerobic capacity, perform poorly on standardized physical fitness tests, and exhibit higher energy expenditure during submaximal exercise, often resulting in fatigue with moderate activity. Most studies report that these impairments are not significantly associated with disease activity or severity, but may be related to prolonged disease duration. Other physiologic factors specific to inflammatory arthritis, such as anemia and muscle atrophy, may contribute to poor oxygen delivery and utilization by the exercising tissues, reducing aerobic capacity. Muscle bulk, strength, endurance, and power are also reduced in children with JA, and the deficits appear to be similar to those observed in adults with rheumatoid arthritis. Atrophy and weakness are most pronounced in muscles near inflamed joints, but studies suggest that weakness is generalized. Muscular deficits appear to be greater in children with disease onset at a very young age and to persist long after active disease resolves. Deconditioning secondary to pain, stiffness, and limited joint mobility may contribute to poor aerobic and muscular fitness. Although some studies suggest that overall energy expenditure is not significantly lower in children with JA, their participation in vigorous PA and organized

442 sports is significantly less. Many young children with JA may have delayed gross motor development and lack the higher motor skills normally acquired during vigorous PA and sports. Fear of injury or disease flare or low self efficacy for exercise may prevent children with arthritis from being more active. The importance of exercise in managing JA is no longer disputed, but recommendations by health professionals regarding specific exercise mode, intensity, duration, and frequency vary considerably. Although the research is still limited, the following general observations about the benefits of exercise can be made. 1) Children with JA can participate in either aquatic or low-impact land-based exercise programs without disease exacerbation. 2) Participation in either a water- or land-based exercise program at least twice a week for at least 6 weeks may help to reduce disease symptoms and improve general exercise endurance. 3) The studies reviewed suggest that exercise on land may induce greater improvements in muscle strength, performance on timed tasks, and functional status than aquatic exercise. 4) Weight-bearing exercise is needed to develop optimal bone width and density during childhood. 5) Individualized and supervised resistance exercise appears to be safe and effective for children with JA as young as 8 years old. 6) The choice of venue for exercise may depend on the child’s specific needs and preferences. A child with severe joint and soft tissue limitations will benefit from direct individual therapy, whereas group exercise programs may provide significant physical and social benefits for children with mild to moderate joint disease. 7) Children with mild disease should be able to participate in most sports with proper screening and physical conditioning. However, highly competitive contact sports pose a potential risk for damage to the joint surface and growth plate and should be avoided during periods of active joint disease.

Suggestions for future research The current body of research must be considered preliminary because most of the studies were small pilot projects and many were reported only in abstract form. Where control groups were included, assignment was not randomized, and few children with severe systemic disease were included. Many questions about PA and exercise in children with JA remain, including 1) What are the longterm effects of vigorous weight-bearing exercise on bone health and joint integrity? 2) What are the differential effects of specific exercise modes and regimens (exercise intensity, duration, and frequency) on disease activity and severity, function, and quality of life? 3) What is the most cost-effective way of delivering exercise programs without sacrificing quality? 4) What is the most effective method of increasing lifetime PA in children with JA? Large, controlled, and randomized studies that stratify subjects by age; sex; disease type, duration, activity, and severity; and initial fitness level are needed to determine the effects of exercise in children with JA. Longitudinal studies about the relationships among PA and exercise, aerobic capacity, muscle function, bone density, joint mobility, functional abilities, and quality of life in children

Klepper with JA will allow health professionals to provide appropriate and timely intervention. However, there are significant barriers to conducting randomized controlled exercise trials in children. Children with severe physical impairments and those who live far from a rheumatology center may not volunteer for exercise studies. Multicenter projects increase the potential to study large numbers of children, but they require careful coordination, multidisciplinary collaboration, and significant funding. However, given the prevalence of arthritis in children and the significant and long-term impact of the disease, increased efforts to stimulate research would be prudent.

REFERENCES 1. Blair SN, Kohl HW, Paffenbarger RS, Clark DG, Cooper KH, Gibbons LW. Physical fitness and all-cause mortality: a prospective study of healthy men and women. JAMA 1989;12: 2395–2401. 2. Pate RP, Pratt M, Blair SN, Haskell WL, Macera CA, Bouchard C, et al. Physical activity and public health: a recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA 1995;273: 402–7. 3. Montoye H. Physical fitness, physical activity, and heart disease risk factors in children. In: Effects of physical activity on children: American Academy of Physical Education papers no. 19. Champaign (IL): Human Kinetics; 1985. p. 127–52. 4. Telama R, Yang X, Laasko L, Viikari J. Physical activity in childhood and adolescence as predictor of physical activity in young adulthood. Am J Prev Med 1997;13:317–23. 5. Public Health Service. Healthy people 2010 objectives: draft for public comment. Washington (DC): US Department of Health and Human Services; 1998. 6. Wallace CA, Levinson JE. Juvenile rheumatoid arthritis: outcome and treatment for the 1990s. In: Athreya B, editor. Rheumatic disease clinics of North America. Philadelphia (PA): WB Saunders; 1991. p. 891–905. 7. Giannini MJ, Protas EJ. Aerobic capacity in juvenile rheumatoid arthritis patients and healthy children. Arthritis Care Res 1991;4:131–5. 8. Giannini MJ, Protas EJ. Exercise response in children with and without juvenile rheumatoid arthritis: a case comparison study. Phys Ther 1992;72:365–72. 9. Giannini MJ, Protas EJ. Comparison of peak isometric knee extensor torque in children with and without juvenile arthritis. Arthritis Care Res 1993;6:82– 8. 10. Klepper SE, Darbee J, Effgen SK, Singsen BH. Physical fitness levels in children with polyarticular juvenile rheumatoid arthritis. Arthritis Care Res 1992;5:93–100. 11. Lindehammar H, Backman E. Muscle function in juvenile chronic arthritis. J Rheumatol 1995;22:1159 – 65. 12. Lindehammar H, Sandstedt P. Measurement of quadriceps muscle strength and bulk in juvenile chronic arthritis: a prospective, longitudinal, 2-year survey. J Rheumatol 1998;25: 2240 – 8. 13. Hebestreit H, Muller-Scholden J, Huppertz HI. Aerobic fitness and physical activity in patients with HLA-B27 positive juvenile spondyloarthropathy that is inactive or in remission. J Rheumatol 1998;25:1626 –33. 14. Oberg T, Karsznia B, Gare A, Lagerstrand A. Physical training of children with juvenile chronic arthritis. Scand J Rheumatol 1994;23:92–5. 15. Fan J, Wessel J, Ellsworth J. The relationship between strength and function in females with juvenile rheumatoid arthritis. J Rheumatol 1998;25:1399 – 405. 16. Wessel J, Kaup C, Fan J, Ehalt R, Ellsworth J, Speer C, et al. Isometric strength measurements in children with arthritis: reliability and relation to function. Arthritis Care Res 1999; 12:238 – 46.

Exercise and Fitness in Children With Arthritis 17. Fisher NM, Venkatraman JT, O’Neil KM. Effects of resistance exercise on children with juvenile arthritis [abstract]. Arthritis Rheum 1999;42 Suppl 9:S396. 18. DeNardo B, Miller L, Schaller J, Tucker L. Young children with juvenile rheumatoid arthritis (JRA) are developmentally delayed in gross and fine motor skills [abstract]. Arthritis Rheum 1996;39 Suppl 9:S1715. 19. Morrison CD, Bundy RC, Fisher AG. The contribution of motor skills and playfulness to the play performance of preschoolers. Am J Occup Ther 1991;45:687–94. 20. Hedengren E, Knutson LM, Haglund-Akerlind Y, Hagelberg S. Lower extremity isometric torque in children with juvenile chronic arthritis. Scand J Rheumatol 2001;30:69 –76. 21. Takken T, Hemel A, van der Net JJ, Helders PJ. Aerobic fitness in children with juvenile idiopathic arthritis: a systemic review. J Rheumatol 2002;29:2643–7. 22. Klepper S. Effects of an eight-week physical conditioning program on disease signs and symptoms in children with chronic arthritis. Arthritis Care Res 1999;12:52– 60. 23. Takken T, van der Net JJ, Helders PJ. Do juvenile idiopathic arthritis patients benefit from an exercise program? A pilot study. Arthritis Care Res 2001;45:81–5. 24. 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;12:239 – 42. 25. Fisher NM, Venkatraman JT, O’Neil K. The effects of resistance exercises on muscle function in juvenile arthritis [abstract]. Arthritis Rheum 2001;44 Suppl 9:S276. 26. Velazquez V, Fisher NM, Venkatraman KM, O’Neil KM. Effect of resistance exercise on children with juvenile arthritis [abstract]. Arthritis Rheum 1999;42 Suppl 9:S396. 27. Venkatraman JT, Fisher NM, Lin HJ, O’Neil K. Effects of resistance exercise on plasma levels and production of cytokines and lipid mediators in juvenile arthritis [abstract]. Arthritis Rheum 1999;42 Suppl 9:S396. 28. Malleson PN, Bennett SM, MacKinnon M, Jespersen DK, Coutts KD, Turner SP, et al. Physical fitness and its relationship to other indices of health status in children with chronic arthritis. J Rheumatol 1996;23:1059 – 65. 29. Golebiowska M, Brozik H, Bujnowski T. Aerobic and anaerobic metabolism in children with juvenile chronic arthritis in relation to some clinical data. In: Coudert J, editor. Pediatric work physiology, children and exercise. Vol XVI. Paris: Masson; 1992. p. 175–7. 30. Takken T, van der Net JJ, Helders PJM. Maximum exercise testing in juvenile rheumatoid arthritis (JRA) patients. Clin Exerc Physiol 2002;4:38 – 43. 31. Bar-Or O. Pathophysiological factors which limit the exercise capacity of the sick child. Med Sci Sports Exer 1986;18:276 – 82. 32. Jones NL. Clinical exercise testing, 3rd edition. Philadelphia (PA): WB Saunders; 1998.

443 33. Knook LM, de Kleer IM, van der Ent, van der Net JJ, Prakken BJ, Kuis W. Lung function abnormalities and respiratory muscle weakness in children with juvenile chronic arthritis. Eur Respir J 1999;14:529 –33. 34. Lechner DE, McCarthy CF, Holden MK. Gait deviations in patients with juvenile rheumatoid arthritis. Phys Ther 1987; 67:1335– 41. 35. Henderson CJ, Lovell DJ, Specker BL, Campaigne BN. Physical activity in children with juvenile rheumatoid arthritis: quantification and evaluation. Arthritis Care Res 1995;8: 114 –9. 36. Rowland T. Cardiovascular fitness: exploring a semantic maze. Pediatr Exerc Sci 1996;8:1–5. 37. Wallace JP, McKenzie TL, Nader PR. Observed versus recalled exercise behavior: a validation of a seven day exercise recall for boys 11-13 years old. Res Q Exerc Sport 1985;56:261–5. 38. Vostrejs M, Hollister JR, Bowyer S. Quadriceps muscle function after recovery from monarticular JRA: follow-up with physical therapy intervention [abstract]. Arthritis Care Res 1989;2:S16. 39. Vostrejs M, Hollister JR. Muscle atrophy and leg length discrepancies in pauciarticular juvenile rheumatoid arthritis. Am J Dis Child 1988;142:343–5. 40. Dunn W. Grip strength of children aged 3 to 7 years using a modified sphygmomanometer: comparison of typical children and children with rheumatic disorders. Am J Occup Ther 1993;47:421– 8. 41. Knops K, Wulffraat N, Lodder S, Houwen R, de Meer K. Resting energy expenditure and nutritional status in children with juvenile rheumatoid arthritis. J Rheumatol 1999;26: 2039 – 43. 42. Henderson CJ, Lovell DJ. Assessment of protein-energy malnutrition in children and adolescents with juvenile rheumatoid arthritis. Arthritis Care Res 1989;2:108 –13. 43. Edstrom L, Nordemar R. Differential changes in type I and Type II muscle fibers in rheumatoid arthritis. Scand J Rheumatol 1974;3:155– 60. 44. Kotaniemi A, Savolainen A, Kroger H, Kautiainen H, Isomaki H. Weight-bearing physical activity, calcium intake, systemic glucocorticoids, chronic inflammation, and body constitution as determinants of lumbar and femoral bone mineral in juvenile chronic arthritis. Scand J Rheumatol 1999;28:19 –26. 45. Moncur C, Marcus R, Johnson S. Pilot project of aerobic conditioning of subjects with juvenile arthritis [abstract]. Arthritis Care Res 1990;3:S16. 46. Feldman BM, Wright FV, Bar-Or O, Dent P. Rigorous fitness training and testing for children with polyarticular arthritis: a pilot study. Arthritis Rheum 2000;43(9 suppl):S120. 47. Bacon MC, Nicholson C, Binder H, White P. Juvenile rheumatoid arthritis: aquatic exercise and lower-extremity function. Arthritis Care Res 1991;4:102–5.