Aerobic and anaerobic capacity in Juvenile Idiopathic Arthritis: evaluation of the cardiorespiratory response
K Houghton, J Potts, A Sheel, R Petty, D McKenzie
Keywords
arthritis, exercise, jia, pediatric
Citation
K Houghton, J Potts, A Sheel, R Petty, D McKenzie. Aerobic and anaerobic capacity in Juvenile Idiopathic Arthritis: evaluation of the cardiorespiratory response. The Internet Journal of Rheumatology. 2007 Volume 5 Number 2.
Abstract
Objectives: To compare the cardiac response during maximal aerobic exercise in JIA with healthy children.
Methods: Thirteen children aged 10 to 17 years with JIA and 9 controls (CON) participated. All performed maximal VO2 and Wingate tests on a cycle ergometer. Cardiac output (CO) was measured during exercise with Doppler echocardiography. Arterial – mixed venous oxygen index (A-VO2) and systemic vascular resistance (SVR) were calculated.
Results: JIA subjects had decreased aerobic fitness (VO2 31.3, CON 47.9 ml/kg/min p = 0.013]. There was no significant difference in CO, A-VO2 or SVR but we observed trends towards lower CO and higher SVR in JIA. During anaerobic exercise JIA subjects completed less total work.
Conclusions: Children with JIA have moderate impairment in aerobic fitness. CO and A-VO2 during aerobic exercise did not differ between JIA subjects and CON. Further research with is required to determine factors contributing to limited fitness in JIA.
Contributing institutions
Divisions of Rheumatology and Cardiology, British Columbia Children's Hospital, Department of Pediatrics, University of British Columbia, Vancouver, Canada. School of Human Kinetics, University of British Columbia, Vancouver, Canada.
Introduction
As a group, children with chronic disease or physical disability are less active than their healthy peers and studies show children with juvenile idiopathic arthritis (JIA) have reduced vigorous physical activity levels, sports participation and decreased fitness compared to healthy children. (1,2,3)
Children with JIA have moderate impairments (21.8% in meta-analysis) in aerobic fitness as measured by VO2 peak (4,5,6,7). VO2 peak is the gold standard for aerobic fitness and is equal to the product of cardiac output (CO, maximal heart rate multiplied by maximal stroke volume), and arterial venous oxygen (A-VO2) difference as defined by Fick equation. Suboptimal VO2 peak may be due to central limitations, characterized by suboptimal heart rate response, CO or oxygen saturation; or peripheral limitations, characterized by high mixed venous oxygen content (low A-VO2).
Peripheral limitations due to muscle atrophy and weakness are hypothesized to largely account for diminished aerobic fitness in children with JIA. Children with JIA have generalized muscle weakness and muscle atrophy, most pronounced in muscles surrounding inflamed joints and often persisting even after clinical resolution of inflammation. (8,9,10,11) Muscle atrophy may lead to decreased oxygen extraction from the exercising muscle, resulting in high mixed venous oxygen content and a low VO2 peak.
There is limited data on anaerobic fitness in JIA with recent studies describing significant anaerobic impairment in children and adolescents. (7, 12,13,14,15) Most childhood play and activities of daily living are anaerobic in nature. Impaired anaerobic fitness may make activities difficult or impossible for children to perform. An association between anaerobic fitness and functional capacity has been described. (16)
The purpose of this study is to examine the cardiorespiratory response during aerobic exercise and the anaerobic to aerobic ratio (metabolic index) in children with JIA compared to healthy children.
Methods
Participants: Patients between 8 and 18 years of age followed at British Columbia's Children's Hospital arthritis clinic with a definite diagnosis of JIA and history of lower extremity joint involvement were invited to participate. (17) Patients were excluded if they had active systemic disease manifest by fever, pericariditis or pleuritis; disease remission greater than one year; primary cardio-respiratory disease with the exception of mild asthma; were unable to cycle due to pain or decreased range of motion of lower extremity joints or were unable to tolerate the mouthpiece for VO2peak testing due to temporomandibular joint disease. Controls (CON) were physically healthy age and sex matched peers, friends or relatives of patients with JIA. Study assessments occurred between January and September 2007. All participants and caregivers provided assent and consent prior to participation. Ethical approval was obtained from the University of British Columbia's Clinical Research Ethics Board and the Hospital's Research Review Committee.
Materials and Methods: Study participants with JIA underwent a clinical assessment by a rheumatologist (KH) the same day as the exercise testing. Assessment included a complete physical examination including active joint count, calculation of the articular severity index (ASI), and data collection for age, sex, disease duration, history of previous cardiac or respiratory disease, recent hemoglobin level and current medications. Subjects underwent all testing in the afternoon to negate morning stiffness as a potential confounding factor.
Anthropometric data
Height (Harpenden Stadiometer, London) and body mass (SECA electronics, Hamburg, Germany) were measured to the nearest 0.1cm and 0.1kg respectively. Body mass index (BMI) was calculated (kg/m 2 ). Obesity was defined as BMI or greater than the 95 th percentile according to reference data. (18)
Fitness
A Wingate test was done first followed by a modified stress echocardiogram with VO2 measurement. Both exercise tests were done on an upright cycle ergometer.
Cardiac output (CO)
CO was measured with Doppler echocardiography by a single echocardiographer (AH). Parasternal long-axis view was used to measure the left ventricular outflow tract (LVOT) diameter at the aortic valve hinge-point during systole. The LVOT area was calculated from this diameter. From an apical 4-chamber view with the transducer tilted anteriorly towards the LVOT, a pulsed wave Doppler sample volume was taken from the centre of the LVOT. The velocity-time integral (VTI) was calculated. The frequency of the transducer (7, 5S, M3S MHz) varied depending on the size of the child. Stroke volume (SV) was calculated as the product of the VTI and LVOT area. Both SV and CO were indexed to body surface area (BSA). CO was calculated as the product of SV and HR. Measurements were taken at rest, 150 seconds into each stage of exercise, immediately post exercise and 3 minutes post exercise.
Arterial-mixed venous O difference (A-VO) and systemic vascular resistance (SVR)
A-VO2 was calculated as the absolute oxygen uptake divided by the absolute CO (A-VO2 = VO2peak/CO). A-VO2 is equal to arterial O2 content minus the mixed venous O2 content. SVR was calculated from CO and MAP (SVR= MAP / CO).
Joint Pain
Subjects with JIA completed a visual analogue scale (VAS) for joint pain on the day of testing, over the previous week and immediately after testing using a 10 point validated VAS scale from the Pediatric Pain Questionnaire. (21, 22) Control subjects completed the VAS after completion of the exercise test.
Disease activity assessment
Subjects with JIA were assessed on the day of testing prior to performing any exercise. ASI was calculated as the sum of scores for joint swelling, pain on motion, tenderness, and limitation of motion. (23) Active joint count was determined as the number of joints with either swelling or painful, limited range of motion.
Functional Disability
Subjects with JIA or their proxy (parent or guardian) completed the childhood health assessment questionnaire (CHAQ), a valid and reliable measure of function in children with JIA. (24)
Physical activity
A questionnaire was completed by patients and controls to determine habitual physical activity and general health. (Adapted from the Children's Exercise and Nutrition Centre at McMaster University. Developed by Oded Bar-Or, MD, FACSM.)
Statistical Analysis: Baseline characteristics were compared with Student's T-test. Descriptive statistics were calculated for fitness measures. Median values and ranges are presented for non-normally distributed data. Z-scores are presented for individual subjects. Differences in fitness measures between subjects with JIA and controls were determined using wilcoxin mann-whitney test for non-normally distributed data. The association between continuous variables was assessed using Pearson and Spearman correlation coefficients. Linear regression models were used to explore the relationships between aerobic fitness, anaerobic fitness, disease activity and function in subjects with JIA. Correlation coefficients
Results
Thirteen children and adolescents aged 10 to 17 years with JIA and 9 CON participated in the study. Table 1 shows patient and CON demographics. There was no difference between the two groups. JIA subjects median age was 13.9 (10.5-17.7) and the control subjects was 12.8 (11.3-16.5). Two of the JIA subjects met the operational definition of obesity. (18)
Figure 1
JIA subject disease characteristics are shown in Table 2. Median disease duration was 24 (5-166) months. All JIA subjects were on at least one medication for their arthritis. Eleven (85%) were on a non-steroidal anti-inflammatory medication (NSAID), 9 (69%) on disease modifying anti-rheumatic disease therapy (DMARD), 2 (15%) on corticosteroids and 1 (8%) on biologic anti-cytokine therapy. Disease activity was variable with median active joint count=1 (0-22) and ASI=6 (0-64). Function and pain was also variable with a median CHAQ score=0 (0-1.4); 5 patients had mild-to-moderate disability. (25)
Figure 2
Fitness: Summary fitness measures are shown in Table 3. All subjects were able to complete the exercise tests without any adverse events. There was a large range in aerobic fitness measures for JIA subjects. VO2 peak was 31.3 ml/min/kg (20.2-49.9) corresponding to 67.7% (50.4-101.0) predicted and a Z score of -1.4 (-.06--2.4). Five (38%) JIA subjects had Z scores of -2 or lower. CON subjects VO2 peak was 47.9 ml/min/kg (32.7-54.1) corresponding to 97.0% (77.4-126.6) predicted and a Z score of -.17 (-1.6-.87). The difference between JIA subjects and CON subjects was significant (P = .013 for VO2 peak, P=.012 for % predicted and P=.011 for Z score).
Figure 3
Anaerobic fitness measures were variable for JIA subjects. Peak power was 9.7 W/kg (5.6-13.7) and total work completed 168.5 j/kg (107-252) compared to CON subject values of 11.3 W/kg (9.8-14.5) and 224 j/kg (180-248). There was no significant difference between the two groups for peak power (P=.095) but there was for total work completed (P=.036). The metabolic index did not differ between the two groups. Two subjects with JIA had a metabolic index less than 2.5, suggesting greater impairment of the anaerobic than aerobic system. (26)
Cardiac output: Maximal CO for JIA subjects was 7.4 L/min (5.6-13.1) and for CON subjects 9.0 (6.7-11.5). There was a trend towards lower CO in JIA subjects but no significant difference between the two groups (P=0.11).
A-VO2 and SVR: A-VO2 did not differ between JIA subjects and CON subjects (P=0.29) but there was a trend towards higher SVR in subjects with JIA. (P=0.12)
Physical activity: All CON subjects and 11 (85%) of JIA subjects were physically active.
Correlations: Correlations between aerobic and anaerobic fitness measures for JIA subjects are shown in Table 4. VO2 peak showed moderate positive correlation with CO (.615, P=.025) and A-VO2 (.637, P=.019) consistent with the Fick principle. VO2 peak and CO showed low and high negative correlation, respectively, with SVR which is consistent with basic physiology principles. Aerobic and anaerobic fitness showed moderate postive correlation: VO2 peak and peak power (0.664, P=.019) and VO2 peak and total work (.619, P=.032). In JIA subjects, there was no significant correlation between disease activity, function and fitness measures.
Figure 4
Discussion
Children with JIA have moderate impairments (21.8% in meta-analysis) in aerobic fitness as measured by VO2 peak (5,6,7, 27). Our patients' aerobic fitness is similar to previous reports with VO2 peak 67.7% (50.4-101.0) predicted and a Z score of -1.4 (-.06--2.4). VO2 peak is the gold standard for aerobic fitness and is equal to the product of cardiac output (maximal heart rate multiplied by maximal stroke volume), and A-VO2 (arterial oxygen content – mixed venous oxygen content) as defined by the Fick equation. Suboptimal VO2 peak may be due to central limitations, characterized by suboptimal heart rate response, CO or arterial oxygen content; or peripheral limitations, characterized by high mixed venous oxygen content (low A-VO2).
Maximal CO in healthy children reaches three to four times the resting value. Most of the increase is due to HR with only 20-25% due to an increase in stroke volume. (28) VO2 peak in children with JIA may be limited centrally by low stroke volume (deconditioning) or low maximal heart rate (cessation of exercise due to fatigue or pain prior to reaching peak heart rate). (26) Children with JIA also have high sub-maximal energy expenditures suggesting increased metabolic demands for routine physical activity. (6) Doppler echocardiography measures of CO at a given VO2 did not differ between our JIA subjects and controls. This finding supports our original hypothesis that children with JIA have normal cardiac response to exercise and suboptimal VO2 peak is due to peripheral limitations.
VO2 peak may be limited peripherally by low arterial oxygen content or high mixed venous oxygen content. Low red cell mass leads to decreased arterial oxygen content, a reduction in maximal A-VO2 and limitation of VO2 peak. A moderate positive relationship between VO2 peak and total body hemoglobin is well described. (29) Anemia is common in children with poorly controlled polyarticular disease and systemic JIA. Only one of our JIA subjects had anemia and this subject had a VO2 peak of 91.5% predicted, making reduced arterial oxygen carrying capacity an unlikely contributor to suboptimal VO2 peak in our study population. Therefore, in our population high mixed venous oxygen content likely accounts for low A-VO2 and low VO2 peak. High mixed venous oxygen content is present when there is suboptimal blood flow to exercising muscles or deficient oxygen extraction from exercising muscles. During aerobic exercise, arterial systolic blood pressure increases in proportion to exercise intensity. Changes (increase or decrease) in diastolic pressure are smaller and reflect changes in peripheral vascular resistance. Mean arterial pressure (MAP), defined as the average arterial pressure during a single cardiac cycle, is considered the perfusion pressure seen by organs in the body. MAP and CO both increase during exercise. Lower SVR results in greater blood flow to exercising muscles.
Children with JIA have generalized muscle weakness and muscle atrophy, most pronounced in muscles surrounding inflamed joints and often persisting even after clinical resolution of inflammation. (8,9,10,11, 30, 31) Children with JIA likely have a combination of peripheral limitations to their aerobic capacity including: increased SVR which may limit blood flow to exercising muscles; muscle atrophy may lead to decreased oxygen extraction from the exercising muscle; and low muscle endurance may lead to decreased oxygen extraction at some stage during an exercise task. We found trends towards higher SVR and lower muscle endurance (lower total work completed during the anaerobic Wingate Test) in our subjects with JIA. We did not specifically measure muscle atrophy or muscle strength. We attempted to non-invasively measure of tissue oxygenation using near infrared spectrophotometer (NIRS) technology but were unable to obtain reliable data in our subjects (data not published). (32, 33) There are no data on A-VO2 in children with JIA or other chronic musculoskeletal conditions.
Anaerobic exercise is high intensity exercise that can't be sustained for more than 30 to 60 seconds. Anaerobic fitness depends predominantly on non-oxidative energy turnover and is related to local characteristics of working muscle groups. There is recognition that children's play activities are anaerobic in nature with short, intense bursts of high energy activity. Anaerobic fitness improves during childhood and adolescence concomitant with increased muscle mass, increased glycolytic capability and improved neuromuscular coordination. Until very recently, there was limited data on anaerobic fitness in JIA. (7, 12, 13) Two recent Dutch studies describe significant anaerobic impairment in children and adolescents with JIA. (14, 15) They performed Wingate exercise tests on 62 children with JIA and found impairment of mean power (66.7% predicted) and peak power (65.5% predicted) compared to healthy children. (14) A similar study of 22 adolescents with JIA found lower mean power for adolescent girls (74%) and boys (88%) and lower peak power for girls (67%). (15)
Anaerobic fitness may be lower in children with JIA due to muscle atrophy. In adults with rheumatoid arthritis, neuromuscular complications are not uncommon. Both neuropathy and a selective reduction in type II muscle fibres have been described. (34-36) A single study of muscle biopsies in children with JIA demonstrated the presence of inflammatory changes in the muscle but no evidence of type II muscle fibre hypotrophy or neuropathy. (37)
In our study, we found children with JIA had mild impairments in anaerobic fitness with less total work completed and a trend towards lower peak power. There are several possible reasons that our subjects with JIA did not have significant impairments. Our clinic's philosophy is to encourage activity and most of our patients participate in regular physical activity. There was also a study selection bias; the patients who volunteered to participate were generally quite fit and interested in sport. Impaired total work during maximal anaerobic exercise suggests poor muscle endurance. Poor muscle endurance may translate to muscular fatigue in daily activities. Given the anaerobic nature of children's activities they may be limited by low muscular endurance rather than suboptimal oxygen consumption. Children with significant anaerobic impairment may be unable to perform all their activities of daily living. In support of this hypothesis, a recent study described a positive relationship between functional ability and anaerobic fitness in children with JIA. (16)
The anaerobic: aerobic ratio or metabolic index is usually greater than 2.5 in healthy children. A lower ratio implies that anaerobic power is compromised more than aerobic power. Children with advanced neuromuscular disease have a low metabolic index. (26) The metabolic index in our patients did not differ from our control population or referenced normative data.
Most studies show aerobic fitness is not significantly related to disease severity or activity but may be related to disease duration. (5,6,7, 16, 27) Increased physical activity levels and self efficacy for exercise correlate with improved aerobic capacity but a causal effect has not been established. (7) There is suggestion that anaerobic fitness is positively related to function. (16) We did not find any significant correlation between fitness measures, disease activity and function in this study.
This study has several limitations. Small subject numbers, a heterogeneous JIA population, and enrollment bias of fitter patients and controls limits generalization of our results. Age and gender differences for VO2 peak are captured by normative data and calculation of Z scores. There are no gender differences in cardiac output and A-VO2 in children so measures amongst JIA subjects and controls were grouped. (38, 39) We did not subdivide participants based on pubertal status due to small numbers. This may limit generalization of our findings as prepubertal children have lower maximal CO and higher A-VO2 at a given VO2 peak. (40) Future studies with larger numbers, subgroups of pubertal status and disease subtype may allow greater power to determine differences in aerobic and anaerobic fitness for children and teenagers with JIA.
In summary, we found moderate
Acknowledgements
This research was conducted as part of Kristin Houghton's Masters of Science (Human Kinetics) degree from the University of British Columbia. Thanks to AJ Hollinger for performing the echocardiograms and calculating cardiac output, and Natalie Taha BSc HKinetics and Karen Slater for helping with the exercise testing procedures.