SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Objective

To explore early changes and predictors of bone mass in children with juvenile idiopathic arthritis (JIA) in order to identify patients who will develop bone mass reductions.

Methods

We conducted a prospective cohort study of 108 children with early JIA (ages 6–18 years; mean disease duration 19.3 months) who were individually matched with 108 healthy children for age, sex, race, and county of residence. Bone mass and changes in total body, spine, femur, and forearm bone mineral density and bone mineral content (BMC), body composition, growth, and biochemical parameters of bone turnover were examined at baseline and at followup a mean of 24 months later. Low bone mass was defined as a Z score >1 SD below the reference population.

Results

Of the 200 children evaluated at followup, the 100 healthy children had greater gains in total body BMC (P = 0.035), distal radius BMC (P < 0.001), and total body lean mass (P < 0.001) than did the 100 JIA patients. Low or very low total body BMC was observed in 24% of the patients and 12% of the healthy children. Bone formation, bone resorption, and weight-bearing activities were reduced in the patients compared with the healthy children. Multiple regression analysis showed that in patients with JIA, serum bone-specific alkaline phosphatase, serum C-telopeptide of type I collagen, and weight-bearing activities were independent predictors of changes in total body BMC. Total body BMC was lower in patients with polyarticular onset than in those with oligoarticular disease onset.

Conclusion

Patients with JIA have moderate reductions in bone mass gains, bone turnover, and total body lean mass early in the disease course.

The foundations for bone fragility in old age are probably partly established during growth, and disorders affecting bone growth, such as childhood illness, have been proposed as an important contributing factor (1). Bone growth and underlying cell biology are complex, little understood processes (2). In active juvenile arthritis, bone mineral accrual may be reduced around affected joints and in skeletal sites distant from the diseased joint. Bone morbidity is influenced by inflammation, medication, nutrition, and physical inactivity (3). Although there are several studies demonstrating reduced bone mass in patients with juvenile idiopathic arthritis (JIA) (4, 5), there are few prospective data on bone mass and bone turnover from an early disease stage in these children.

This study explores early changes in, and predictors of, bone mass in children with JIA in order to identify patients who are likely to develop reduced bone mass. A patient cohort and a matched cohort of healthy children were followed up prospectively. Changes in total body, spine, femur, and forearm bone mineral density (BMD) and bone mineral content (BMC), as well as body composition and fracture rates were examined.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Study participants.

All white children with JIA between the ages of 6 and 18 years who were attending the Department of Rheumatology, Rikshospitalet University Hospital, Oslo, for the first time from May 1995 to February 1999 were invited to participate in the study. Of 127 eligible patients who met the criteria for juvenile rheumatoid arthritis (6), juvenile psoriatic arthritis (7), or juvenile ankylosing spondylitis (8), 108 patients (85%) were included. Their mean ± SD disease duration was 19.3 ± 12.2 months. The patients who were not included were comparable to the participants with regard to age, sex, and disease-onset type (data not shown).

The patients were living in 16 different counties in areas ranging from southern Norway to north of the Arctic Circle (latitudes from 58°N to 68°N). Each patient was individually matched to a healthy child with the same sex, age, race, and county of residence who was randomly selected from the National Population Register. The participants were studied at baseline and at followup, a mean of 24 months later. The Regional Ethics Committee for Medical Research approved the study. Written informed consent was obtained from the parents of the children.

Clinical information was obtained by interviews, physical examination, medical record reviews, and questionnaires (9). Weight was measured on a Seca (Hamburg, Germany) digital scale to the nearest 0.1 kg and height to the nearest 0.1 cm. Pubertal development was determined in the clinic by the adolescent's self-assessment of sexual maturation according to Tanner's standard photographs (10, 11). The youngest children, who according to one of us (GL) as well as the children's parents, had obviously not reached puberty, were given the lowest value (stage 1, preadolescence) and were not shown the Tanner's photographs.

All participants completed a self-report questionnaire that included questions about fractures and leisure-time physical activities outside of school hours (12, 13). Food and nutrition intake were assessed by a standardized quantitative food frequency questionnaire that has been evaluated for reproducibility and validity in adolescents (14). Nutrient calculations were computed in accordance with the Norwegian Food Composition Table (15).

Radiographs of the nondominant hand and wrist of all study subjects were taken for assessment of skeletal maturity (16). Radiographs of the patients' knees and ankles were obtained at the time of admission to the hospital; other joints were radiographed when clinically indicated. Radiographic features were graded according to a classification system for juvenile rheumatoid arthritis (17).

Laboratory measures.

Bone formation was assessed by serum levels of bone-specific alkaline phosphatase and osteocalcin. Bone resorption was assessed by serum levels of C-telopeptide of type I collagen and urinary concentrations of deoxypyridinoline (4, 5, 18). Vitamin D stores were assessed by serum concentrations of 25-hydroxyvitamin D, and the active hormonal metabolite was assessed by serum levels of 1,25-dihydroxyvitamin D3 (19). C-reactive protein levels, platelet counts, and serum levels of albumin, phosphate, calcium, and parathyroid hormone were measured in all study participants. Erythrocyte sedimentation rate and IgM rheumatoid factor were also determined in all patients.

Bone mass measurements.

Measurements of total body, lumbar spine, hip, and nondominant forearm bone mass and density were evaluated with the same dual x-ray absorptiometry (DXA) equipment (Lunar Expert-XL; GE Lunar, Madison, WI). The children wore socks and underwear and were measured using the same scan mode at baseline and followup. All analyses were performed by one investigator using Expert-XL software versions 1.72 and 1.91 and were read by one investigator (GL). Fat and lean composition of soft tissue was measured, and the BMD, BMC, and bone area were calculated for the total body, for L2–L4 of the lumbar spine, the femoral neck, total femur, distal radius, and distal one-third of the radial shaft.

To minimize the effects of bone size on BMD values in the spine, we used a model for correcting BMD values for anteroposterior depth, the bone mineral apparent volumetric density, which was calculated as follows: BMDvol = BMDarea × [4/π width] (20).

Total body BMC was used as the outcome variable in the analyses for predictors of bone mass changes. This variable was chosen because it includes all skeletal regions, and the errors in repositioning subjects are minimized (21).

The BMD and BMC findings were calculated as Z scores in terms of the number of standard deviations above or below the age-specific mean in healthy individuals (22). Norwegian reference values for pediatric DXA measurements were not available, and we chose age- and sex-specific numeric data provided by the manufacturer of the DXA scanner (GE Lunar) (13). Low BMD or BMC was defined as a Z score between −1 SD and −2 SD. Very low BMD or BMC was defined as a Z score >2 SD below the mean.

Statistical analysis.

Differences between patients and healthy children were assessed by paired samples t-test for continuous variables and by McNemar's test for categorical variables. Within the patient cohort, differences were tested by the independent samples t-test or the Mann-Whitney test for continuous variables, the chi-square test or Fisher's exact test for categorical variables, or by one-way analysis of variance, using the Bonferroni correction for multiple comparisons.

Multiple regression analyses were performed to identify predictors of changes in total body BMC for the controls and patients separately. Explanatory variables were included in the model if the P value was less than 0.2 in unadjusted linear regression analyses or if a variable was known to be associated with reductions in bone mass (23). Highly intercorrelated independent variables (r > 0.7) in the multiple model were avoided. To reduce the possibility of body size–related artifacts, the bone area, weight, and height were included in the multiple regression models (24). We also used a variance component model to compare controls and patients while controlling for potential confounders. Forward and backward stepwise regression methods were used.

To determine whether our results were highly influenced by the patients with spondylarthropathy and psoriatic arthritis, we also performed the regression analysis and tests of differences between patients and controls separately for the patients with juvenile rheumatoid arthritis.

For all analyses, P values less than or equal to 0.05 (2-tailed tests) were considered significant. The statistical analysis was performed using SPSS software version 11.0 (SPSS, Chicago, IL).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Demographic and clinical data.

A total of 208 (96%) of the 216 participants who were enrolled at baseline completed the study. Eight healthy children were unavailable at followup. The remaining 100 healthy children and the 100 patients, who were individually matched, were included in the statistical analyses at followup.

Clinical features of the participants at baseline and followup are shown in Table 1. Height, weight, and intake of calcium, vitamin D, energy, and protein were comparable for patients and controls (Table 2) (25). There was no pubertal delay in the patients. Leisure-time weight-bearing activities were less frequent in the patient group.

Table 1. Characteristics of patients with JIA and healthy children at baseline and 2-year followup*
CharacteristicsBaselineFollowup
JIA patients (n = 108)Controls (n = 108)JIA patients (n = 100)Controls (n = 100)
  • *

    Because of rounding, not all percentages total 100. JIA = juvenile idiopathic arthritis; RF = rheumatoid factor.

  • Sixty-nine joints were evaluated.

  • Likert scale (1–5 points), where 1 = best and 5 = worst.

  • §

    Visual analog scale (0–10 cm), where 0 = best and 10 = worst.

  • Scores ranged from 0 = best to 3 = worst.

  • #

    Highest registered score from a range of 0 = best to 5 = worst.

  • **

    Methotrexate (n = 25), hydroxychloroquine (n = 8), etanercept (n = 1), methotrexate plus hydroxychloroquine (n = 4), methotrexate plus sulfasalazine (n = 3), and methotrexate plus etanercept (n = 3).

  • ††

    Prednisolone. None of the patients had taken methylprednisolone.

No. (%) female63 (58)63 (58)58 (58)58 (58)
Age, mean ± SD years10.1 ± 3.210.1 ± 3.212.2 ± 3.212.3 ± 3.2
Disease-onset type, no. (%)    
 Systemic arthritis5 (5)4 (4)
 Oligoarthritis64 (59)59 (59)
 Polyarthritis, RF negative30 (28)28 (28)
 Polyarthritis, RF positive3 (3)3 (3)
 Spondylarthropathy patients3 (3)3 (3)
 Psoriatic arthritis patients3 (3)3 (3)
Disease duration, mean ± SD months19.3 ± 12.244.3 ± 14.2
Disease activity measures, mean ± SD    
 No. of joints with active disease2.1 ± 3.31.3 ± 4.0
 No. of joints with restricted mobility1.7 ± 2.61.4 ± 2.9
 Physician's global assessment of disease severity2.2 ± 1.01.8 ± 1.0
 Parent's global assessment of child's well-being§2.6 ± 2.61.8 ± 2.0
 Childhood Health Assessment Questionnaire0.5 ± 0.60.3 ± 0.5
 Erythrocyte sedimentation rate, mm/hour12.4 ± 11.911.7 ± 13.6
Radiographic score, mean ± SD#1.6 ± 0.71.7 ± 0.7
Disease-modifying antirheumatic drugs, current use, no. (%)57 (53)44 (44)**
Oral corticosteroids††    
 Current use, no. (%)21 (19)12 (12)
  Dosage, mean ± SD mg/day11.0 ± 6.79.6 ± 8.3
 Ever used, no. (%)35 (32)39 (39)
  Cumulative dose, mean ± SD mg1,756 ± 2,3722,507 ± 2,995
  Cumulative dose, mean ± SD mg/kg of body weight61 ± 9868 ± 90
Table 2. Anthropometric measurements, physical activity, nutrient intake, and markers of bone turnover in patients with JIA and healthy children at baseline and 2-year followup*
VariableBaselineFollowup
JIA patients (n = 108)Controls (n = 108)PJIA patients (n = 100)Controls (n = 100)P
  • *

    P values for differences between patients and healthy children were determined by paired samples t-test. Values are the mean ± SD. JIA = juvenile idiopathic arthritis; SDS = standard deviation score (above or below the age-specific mean for weight and height in the Norwegian reference population [see ref. 25]); 25(OH)D = 25-hydroxyvitamin D; 1,25(OH)2D3 = 1,25-dihydroxyvitamin D3; alk. phos. = alkaline phosphatase.

  • Ranging from 1 = preadolescence to 5 = fully mature.

  • Defined as leisure-time physical activities during non-school hours in which the heel touches the ground, excluding activities such as bicycling and swimming. Measured levels reflect the additive sum of activities, reported as the number of times per week.

  • §

    There is no international agreement on cutoff levels for hypovitaminosis D. Insufficiency was defined as a 25(OH)D level <37.5 nmoles/liter and deficiency as <20 nmoles/liter (see ref. 26).

Anthropometric measures      
 Bone age, years9.4 ± 3.49.6 ± 3.50.04511.8 ± 3.411.9 ± 3.40.213
 Height, SDS0.52 ± 1.140.68 ± 0.970.2930.38 ± 1.100.47 ± 0.960.517
 Weight, SDS0.66 ± 1.460.68 ± 1.280.9240.75 ± 1.580.62 ± 1.180.522
 Tanner stage1.7 ± 1.21.6 ± 1.10.1222.3 ± 1.52.1 ± 1.50.013
Weight-bearing physical activity6.4 ± 4.27.7 ± 4.90.0337.5 ± 4.58.9 ± 4.90.040
Dietary intake of nutrients      
 Calcium, mg/day1,042 ± 435947 ± 3720.132948 ± 3861,009 ± 4550.417
 Vitamin D, μg/day4.1 ± 2.14.0 ± 1.80.6094.0 ± 2.94.5 ± 3.60.387
 Energy, kcal/day2,351 ± 7382,276 ± 7510.4892,370 ± 7292,575 ± 1,0620.194
 Protein, gm/day82.2 ± 25.678.8 ± 25.60.36080.5 ± 25.686.7 ± 36.90.266
Serum chemistries      
 Ionized calcium, mmoles/liter1.30 ± 0.041.28 ± 0.040.0011.26 ± 0.051.25 ± 0.040.239
 25(OH)D, nmoles/liter§49.6 ± 16.450.4 ± 20.30.72952.6 ± 24.048.0 ± 20.40.155
 1,25(OH)2D3, pmoles/liter112.0 ± 31.9104.4 ± 27.50.075111.2 ± 37.3104.9 ± 34.40.201
 Parathyroid hormone, pmoles/liter3.15 ± 1.903.27 ± 1.980.6623.57 ± 1.833.89 ± 1.810.243
 Bone-specific alk. phos., units/liter78.7 ± 33.892.0 ± 34.20.00483.4 ± 37.999.6 ± 38.0<0.001
 Osteocalcin, nmoles/liter5.58 ± 2.426.14 ± 2.030.0396.07 ± 2.947.93 ± 3.28<0.001
 C-telopeptide of type I collagen, μg/liter14.7 ± 5.314.3 ± 4.00.48713.1 ± 4.714.5 ± 4.80.013
Urinary deoxypyridinoline, nmoles/liter  per mmoles/liter of creatinine23.8 ± 8.821.0 ± 7.00.00322.2 ± 24.543.1 ± 49.1<0.001

Laboratory evaluation of inflammation and bone metabolism.

Serum levels of C-reactive protein and platelet counts were higher and albumin levels signifi-cantly lower in the patients than in the controls (data not shown). Bone-specific alkaline phosphatase and osteocalcin levels were lower in the patients at baseline and followup, while serum C-telopeptide of type I collagen and urinary deoxypyridinoline levels in the patients were higher at baseline and lower at followup (Table 2). Serum levels of 25-hydroxyvitamin D, 1,25-dihydroxyvitamin D3, calcium, and parathyroid hormone were similar in patients and controls (Table 2). Serum levels of 25-hydroxyvitamin D showed seasonal variations, with a peak during the summer months. At followup, the serum levels of 25-hydroxyvitamin D were insufficient (<37.5 nmoles/liter) (26) in 29% of the patients and in 31% of the controls, and deficient (<20 nmoles/liter) in 1% and 3% of the groups, respectively. Daily dietary intake of vitamin D was insufficient (<5 μg/day) in 75% of both groups (data not shown).

Changes in bone measurements and body composition.

There were no significant differences in bone measurements between patients and controls at baseline. Although measured values were increased for controls and patients from baseline to follow up, the increase was greater in controls than in patients in terms of total body BMC (difference of increase 35 gm [95% CI 2–68]; P = 0.035), total body lean mass (difference of increase 1.2 kg [95% CI 0.5–1.8]; P < 0.001), and distal radius BMC (difference of increase 0.08 gm [95% CI 0.04–0.12]; P < 0.001). The patients had higher gains in the percentage of total body fat (difference of increase 2.8% [95% CI 1.2–4.4]; P = 0.001). There was a trend toward higher gains in femoral neck BMC and total femoral BMC in the controls, but the differences were not statistically significant. Changes in BMC of the spine, apparent volumetric BMD of the spine, and BMC of the distal one-third of the radius were comparable in patients and controls.

Frequencies of low bone mass.

The frequencies of low (i.e., Z score less than −1 SD according to the reference population) and very low (i.e., Z score less than −2 SD) BMD and BMC, and fractures are shown in Table 3. Total body BMC Z scores were low or very low in 24% of the patients who underwent total body imaging and in 12% of the controls at followup (P = 0.045). Low total body BMC was found in 0 of 4 patients with systemic-onset juvenile rheumatoid arthritis, 9 (16%) of 58 with oligoarticular onset, 11 (36%) of 31 with polyarticular onset, 3 of 3 with juvenile ankylosing spondylitis, and in 1 of 3 with juvenile psoriatic arthritis. Twenty-two percent of the patients and 17% of the controls reported fractures, but the differences were not statistically significant. None of the fractures were vertebral.

Table 3. Frequency of normal, low, and very low BMD and BMC values as well as the frequency of fractures in patients with JIA and healthy children at baseline and 2-year followup*
VariableBMD or BMC group 
NormalLowVery lowFracture
JIA patientsControlsJIA patientsControlsJIA patientsControlsJIA patientsControls
  • *

    A total of 216 subjects (108 per group) were evaluated at baseline, and 200 (100 per group) were evaluated at followup. Values are the number of subjects/total number evaluated (%). Because of rounding, not all percentages total 100. BMD = bone mineral density; BMC = bone mineral content; JIA = juvenile idiopathic arthritis.

  • Low BMD and BMC were defined as a Z score between −1 SD and −2 SD. Very low BMD and BMC were defined as a Z score >2 SD below the mean in an age- and sex-matched reference population. Z score = (measurement in subject − mean measurement in the Lunar reference population)/(SD for the reference population) (see refs. 13 and22). Reference data sufficient for calculating Z scores were not available for the femur BMC, the entire femur scan, or the forearm.

  • Number of subjects varies because of missing data.

  • §

    P = 0.045 versus healthy children.

Baseline        
 Total body BMC94/107 (88)98/107 (92)12/107 (11)8/107 (7)1/107 (1)1/107 (1)  
 Total body BMD103/107 (96)105/107 (98)4/107 (4)2/107 (2)00  
 Spine L2–L4 BMC107/108 (99)102/108 (94)1/108 (1)6/108 (6)00  
 Spine L2–L4 BMD99/108 (92)101/108 (94)9/108 (8)7/108 (6)00  
 Femur neck BMD79/108 (73)90/108 (83)22/108 (20)17/108 (16)7/108 (6)1/108 (1)  
 Any fracture      14/107 (13)13/107 (12)
Followup        
 Total body BMC75/99 (76)§87/99 (88)18/99 (18)8/99 (8)6/99 (6)4/99 (4)  
 Total body BMD89/99 (90)93/99 (94)8/99 (8)6/99 (6)2/99 (2)0  
 Spine L2–L4 BMC92/100 (92)91/100 (91)6/100 (6)9/100 (9)2/100 (2)0  
 Spine L2–L4 BMD82/100 (82)90/100 (90)15/100 (15)10/100 (10)3/100 (3)0  
 Femur neck BMD72/99 (73)82/99 (83)20/99 (20)12/99 (12)7/99 (7)5/99 (5)  
 Any fracture      21/94 (22)16/94 (17)

Predictors of changes in bone mass.

Clinical characteristics, disease duration, disease activity variables, radiographic scores, medication use, corticosteroid use, laboratory measurements, dietary intake of nutrients, and physical activity (Tables 1 and 2) assessed at baseline were explored as independent predictors of the change in total body BMC by unadjusted linear regression analysis, and predictors (P < 0.2) were chosen for the multiple regression model. Significant predictors, analyzed separately for patients and controls by forward stepwise multiple regression, were weight-bearing physical activity (P = 0.006 and P = 0.007), serum C-telopeptide of type I collagen (P < 0.001 and P < 0.001), and serum bone-specific alkaline phosphatase (P = 0.029 and P = 0.015) when adjustments were made for baseline BMC and body size (Table 4). Backward stepwise regression analysis gave essentially the same results. The presence of JIA was a predictor of lower gains in total body BMC (P = 0.047) (data not shown) when patients and controls were analyzed together by a variance component model that controls for potential confounders. The model confirmed the difference in total body BMC gains between patients and controls reported above.

Table 4. Independent predictors of changes in total body BMC from baseline to followup in patients with JIA and healthy children, by linear regression analysis*
Explanatory variableJIA patients, multiple regression analysisControls, multiple regresion analysis
Regression coefficient95% CIPRegression coefficient95% CIP
  • *

    Analysis was done separately for the juvenile idiopathic arthritis (JIA) patients and healthy controls. The subjects' characteristics, disease duration, disease activity, radiographic scores, medication use (including corticosteroids), laboratory data, dietary nutrient intake, and physical activity levels (from Tables 1 and 2) assessed at baseline were explored as independent predictors of the change in total-body bone mineral content (BMC) by unadjusted linear regression analysis. Predictors chosen for the multiple regression model (P < 0.2) were baseline BMC, bone area, weight, height, bone age, weight-bearing activity, serum levels of ionized calcium, 25-hydroxyvitamin D, parathyroid hormone, bone-specific alkaline phosphatase, and C-telopeptide of type I collagen, and urinary deoxypyridinoline. Adjusted R2 values indicate the total variance explained in the final model. 95% CI = 95% confidence interval.

  • Final model identifying significant predictors of change, by forward regression analysis.

Weight-bearing activity, no. of times per week9.792.84–16.740.0068.232.35–14.110.007
Serum bone-specific alkaline phosphatase, units/liter1.100.12–2.090.0291.130.22–2.030.015
Serum C-telopeptide of type I collagen, μg/liter15.447.96–22.92<0.00123.9117.02–30.81<0.001
Adjusted R2, %35  46  

In the JIA cohort, the best total body BMC Z score was found in patients with oligoarticular-onset juvenile rheumatoid arthritis (Z = 0.01) followed by systemic-onset disease (Z = −0.16), polyarticular-onset disease (Z = −0.63), juvenile psoriatic arthritis (Z = −0.76), and juvenile ankylosing spondylitis (Z = −2.03). The difference between polyarticular- and oligoarticular-onset disease was statistically significant (P = 0.023) (data not shown).

Patients with low total body BMC at followup had higher disease activity values at baseline than did patients with a normal Z score, but the differences were not statistically significant. The parent's global assessment of the child's well-being (0–10-cm visual analog scale) was 3.8 versus 2.2 (P = 0.054), the number of joints with active disease was 3.3 versus 1.9 (P = 0.119), the number of joints with restricted mobility was 2.5 versus 1.5 (P = 0.163), the physician's global assessment (1–5-point Likert scale) was 2.4 versus 2.2 (P = 0.284), the Childhood Health Assessment Questionnaire score (range 0–3) was 0.52 versus 0.48 (P = 0.835), and the erythrocyte sedimentation rate was 16.7 mm/hour versus 11.5 mm/hour (P = 0.183) (data not shown).

Compared with patients who were not taking prednisolone at the time of followup, the 12 patients who were receiving prednisolone had lower Z scores for total body BMD (−0.21 versus 0.34; P = 0.019) and femoral neck BMD (−1.35 versus −0.16; P = 0.001). The Z scores for total body BMC, spine BMC, and spine BMD were also lower, but the differences were not statistically significant (data not shown). Current use of prednisolone was associated with changes in total body BMC by univariate linear regression analysis (regression coefficient −88.2, P = 0,039), but was not chosen as a significant predictor in the multiple regression model. Neither the cumulative prednisolone dose nor ever use of prednisolone was a statistically significant predictor of changes in total body BMC from baseline to followup, either in the univariate or in the multiple regression model (data not shown).

When we analyzed the patients with juvenile rheumatoid arthritis separately, we found the same significant results as for the total JIA group, in terms of differences between patients and controls in laboratory evaluations of inflammation and bone metabolism, changes in bone measurement values and body composition, and predictors of changes in bone mass (data not shown).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

In this 2-year prospective longitudinal study of early JIA patients and healthy children, we found lower gains in total body BMC, distal radius BMC, and total body lean mass and higher gains in total body fat in the patients compared with the healthy children. Low or very low total body BMC was observed in 24% of the patients and in 12% of the controls at followup, and the patients had lower levels of markers of bone formation and bone resorption. Predictors of the changes in total body BMC were the JIA, bone markers, and weight-bearing physical activity. Patients with a low total body BMC at followup seemed to have more active disease at baseline.

Our patient cohort was taken from a referral-based center for children with pediatric rheumatic diseases. We serve ∼70% of the Norwegian population of 4.5 million inhabitants. Although our patient group is referral-based, it is comparable to the JIA patients in epidemiologic studies with regard to age, sex, and distribution of disease-onset type (27, 28).

Our study has several strengths. Ninety-six percent of all subjects who were enrolled completed the study. The healthy children were randomly chosen from the population and individually matched with the patients according to age, sex, ethnicity, and geography. The geographic match may reduce a bias caused by geographic differences in femoral neck geometry, BMD, and fracture rates related to environmental factors or genetic risk determinants (29, 30).

Published data on bone mass development during the first years of JIA is sparse. To our knowledge, this is the first prospective controlled cohort study of bone mass gains and bone turnover in children early in the disease course.

Markers of bone formation (serum bone-specific alkaline phosphatase) and bone resorption (serum C-telopeptide of type I collagen) were predictors of the changes in total body BMC in the controls as well as in the patients by multiple regression analysis. Bone formation markers had lower values in the patients than in the controls both at baseline and at followup, whereas bone resorption was greater in the patients at baseline and less pronounced at followup. These observations indicate that there is a reduction in bone turnover early in the disease course. The results are consistent with the results from studies of bone turnover in patients with a longer disease duration (31–33). Although corticosteroids are known to reduce bone turnover in children (18), the proportion of our patients who were currently receiving oral corticosteroids fell from 19% at baseline to 12% at followup, and cannot be the only explanation for the reduction in bone turnover during the observation period.

Body composition was altered in the patients; lean mass decreased and fat mass increased. A higher percentage of body fat in patients than in healthy children has previously been reported (32). The altered body composition is probably related to chronic inflammation and reduced physical activity.

Reduced physical activity means that the muscle force applied to bone is diminished, which in turn, may result in low bone mass (34). Studies of healthy children indicate that the growing skeleton is sensitive to, and benefits from, exercise (35–37). Weight-bearing physical activity was a predictor of the increase in total body BMC both in the healthy children and in the patients, although the patients engaged in weight-bearing activities less frequently than the controls. Few studies have analyzed the effects of physical loading in JIA, but weight-bearing physical activity was found to be a positive determinant of bone mineral volumetric density of the femoral neck in a Finnish patient group (38).

The presence of JIA was a predictor of lower gains in total body BMC, but the disease activity variables were not found to be significant in the regression analysis. However, we did find an association between disease activity and low bone mass, which has also been found in several other studies of childhood arthritis (4, 5). It is not known what dosage and duration of oral corticosteroids is needed for the development of osteoporosis in children. Our results showed an association between current corticosteroid use and decreased bone mass Z scores, but the cumulative corticosteroid dose and previous use of corticosteroids were not significant predictors of reductions in bone mass gains. These results are consistent with our findings in a long-term outcome study of adolescent patients with JIA (13). The findings are also consistent with a recent published study of children with nephrotic syndrome who were treated with an average cumulative dose of 23,000 mg of glucocorticoids and without the expected deficits in bone mineral content of the spine or total body (39). Those authors proposed that the finding may reflect the ability of the growing skeleton to sustain glucocorticoid-induced reductions in bone formation and to recover during intervals of remission.

Activated vitamin D is a key regulator of calcium phosphate metabolism, which is important for optimal growth and mineralization. During the winter, limited exposure to sunlight reduces the synthesis of vitamin D in the skin in populations living in countries at high latitudes such as ours, and dietary intake of vitamin D becomes more important. The low dietary intake of vitamin D and the low serum levels of 25-hydroxyvitamin D found in our study subjects may be of clinical importance. Low serum levels of 25-hydroxyvitamin D in healthy children have also been found in other studies of Scandinavian populations (26).

While the identification of pediatric osteoporosis has progressed in the last 10 years, there is still a need for longitudinal studies to investigate the natural history of the various forms of childhood osteoporosis (40). The present prospective study indicates that a process leading to diminished bone mass starts early in the disease course of JIA. The decreased gains in bone mass were both generalized and localized around affected joints, and affected both trabecular and cortical bone. However, the observed reductions in bone mass were moderate. Our knowledge of bone catch-up growth is sparse. To fully define the natural history of osteoporosis in JIA, pediatric cohorts should be followed up with longitudinal studies over a longer period.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

We thank the participants and their parents for their patience, Howard S. Barden (GE Lunar Corporation) for providing us with the normative data for the reference population, Gunn J. Hovland for conducting the DXA scans, Virginia Johnston for grading the radiographs, and Berit Halmrast and Helga V. Bruaseth for help in collecting the data.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  • 1
    Bradney M, Karlsson MK, Duan Y, Stuckey S, Bass S, Seeman E. Heterogeneity in the growth of the axial and appendicular skeleton in boys: implications for the pathogenesis of bone fragility in men. J Bone Miner Res 2000; 15: 18718.
  • 2
    Orwoll ES. Toward an expanded understanding of the role of the periosteum in skeletal health. J Bone Miner Res 2003; 18: 94954.
  • 3
    Cassidy JT. Osteopenia and osteoporosis in children. Clin Exp Rheumatol 1999; 17: 24550.
  • 4
    McDonagh JE. Osteoporosis in juvenile idiopathic arthritis. Curr Opin Rheumatol 2001; 13: 399404.
  • 5
    Rabinovich CE. Bone mineral status in juvenile rheumatoid arthritis. J Rheumatol 2000; 27 Suppl 58: 347.
  • 6
    JRA Criteria Subcommittee of the Diagnostic and Therapeutic Criteria Committee of the American Rheumatism Association. Current proposed revision of the JRA criteria. Arthritis Rheum 1977; 20 Suppl 2: 1959.
  • 7
    Southwood TR, Petty RE, Malleson PN, Delgado EA, Hunt DW, Wood B, et al. Psoriatic arthritis in children. Arthritis Rheum 1989; 32: 100713.
  • 8
    Van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis: a proposal for modification of the New York criteria. Arthritis Rheum 1984; 27: 3618.
  • 9
    Selvaag AM, Flato B, Lien G, Sorskaar D, Vinje O, Forre O. Measuring health status in early juvenile idiopathic arthritis: determinants and responsiveness of the child health questionnaire. J Rheumatol 2003; 30: 160210.
  • 10
    Duke PM, Litt IF, Gross RT. Adolescents' self-assessment of sexual maturation. Pediatrics 1980; 66: 91820.
  • 11
    Tanner JM. Growth at adolescence. 2nd ed. Oxford: Blackwell Scientific Publications; 1962. p. 2839.
  • 12
    Anderssen N, Klepp KI, Aas H, Jakobsen R. Stability in physical activity levels in young adolescents. Eur J Public Health 1994; 4: 17580.
  • 13
    Lien G, Flato B, Haugen M, Vinje O, Sorskaar D, Dale K, et al. Frequency of osteopenia in adolescents with early-onset juvenile idiopathic arthritis: a long-term outcome study of one hundred five patients. Arthritis Rheum 2003; 48: 221423.
  • 14
    Andersen LF, Nes M, Lillegaard IT, Sandstad B, Bjorneboe GE, Drevon CA. Evaluation of a quantitative food frequency questionnaire used in a group of Norwegian adolescents. Eur J Clin Nutr 1995; 49: 54354.
  • 15
    Haga Rimestad A, Blaker B, Flaten AM, Nordbotten A. Food composition table. Oslo: Universitetsforlaget; 1995.
  • 16
    Greulich WS, Pyle SI. Radiographic atlas of skeletal development of the hand and wrist. Stanford: Stanford University Press; 1959.
  • 17
    Dale K, Paus AC, Laires K. A radiographic classification system in juvenile rheumatoid arthritis applied to the knee. Eur Radiol 1994; 4: 2732.
  • 18
    Szulc P, Seeman E, Delmas PD. Biochemical measurements of bone turnover in children and adolescents. Osteoporos Int 2000; 11: 28194.
  • 19
    Mughal Z. Rickets in childhood. Semin Musculoskelet Radiol 2002; 6: 18390.
  • 20
    Kroger H, Kotaniemi A, Vainio P, Alhava E. Bone densitometry of the spine and femur in children by dual-energy x-ray absorptiometry. Bone Miner 1992; 17: 7585.
  • 21
    Nelson DA, Koo WW. Interpretation of absorptiometric bone mass measurements in the growing skeleton: issues and limitations. Calcif Tissue Int 1999; 65: 13.
  • 22
    Leonard MB, Propert KJ, Zemel BS, Stallings VA, Feldman HI. Discrepancies in pediatric bone mineral density reference data: potential for misdiagnosis of osteopenia. J Pediatr 1999; 135: 1828.
  • 23
    Altman DG. Relation between several variables. In: AltmanDG, editor. Practical statistics for medical research. London: Chapman & Hall; 1997. p. 32561.
  • 24
    Prentice A, Parsons TJ, Cole TJ. Uncritical use of bone mineral density in absorptiometry may lead to size-related artifacts in the identification of bone mineral determinants. Am J Clin Nutr 1994; 60: 83742.
  • 25
    Waaler PE. Anthropometric studies in Norwegian children. Acta Paediatr Scand Suppl 1983; 308: 141.
  • 26
    Lehtonen-Veromaa M, Mottonen T, Irjala K, Karkkainen M, Lamberg-Allardt C, Hakola P, et al. Vitamin D intake is low and hypovitaminosis D common in healthy 9- to 15-year-old Finnish girls. Eur J Clin Nutr 1999; 53: 74651.
  • 27
    Gare BA, Fasth A. Epidemiology of juvenile chronic arthritis in southwestern Sweden: a 5-year prospective population study. Pediatrics 1992; 90: 9508.
  • 28
    Moe N, Rygg M. Epidemiology of juvenile chronic arthritis in northern Norway: a ten-year retrospective study. Clin Exp Rheumatol 1998; 16: 99101.
  • 29
    Crabtree N, Lunt M, Holt G, Kroger H, Burger H, Grazio S, et al. Hip geometry, bone mineral distribution, and bone strength in European men and women: the EPOS study. Bone 2000; 27: 1519.
  • 30
    Falch JA, Kaastad TS, Bohler G, Espeland J, Sundsvold OJ. Secular increase and geographical differences in hip fracture incidence in Norway. Bone 1993; 14: 6435.
  • 31
    Hillman L, Cassidy JT, Johnson L, Lee D, Allen SH. Vitamin D metabolism and bone mineralization in children with juvenile rheumatoid arthritis. J Pediatr 1994; 124: 9106.
  • 32
    Pepmueller PH, Cassidy JT, Allen SH, Hillman LS. Bone mineralization and bone mineral metabolism in children with juvenile rheumatoid arthritis. Arthritis Rheum 1996; 39: 74657.
  • 33
    Reed AM, Haugen M, Pachman LM, Langman CB. Repair of osteopenia in children with juvenile rheumatoid arthritis. J Pediatr 1993; 122: 6936.
  • 34
    Frost HM, Schonau E. The “muscle-bone unit” in children and adolescents: a 2000 overview. J Pediatr Endocrinol Metab 2000; 13: 57190.
  • 35
    MacKelvie KJ, Khan KM, Petit MA, Janssen PA, McKay HA. A school-based exercise intervention elicits substantial bone health benefits: a 2-year randomized controlled trial in girls. Pediatrics 2003; 112: 44752.
  • 36
    Slemenda CW, Miller JZ, Hui SL, Reister TK, Johnston CC Jr. Role of physical activity in the development of skeletal mass in children. J Bone Miner Res 1991; 6: 122733.
  • 37
    Zanker CL, Gannon L, Cooke CB, Gee KL, Oldroyd B, Truscott JG. Differences in bone density, body composition, physical activity, and diet between child gymnasts and untrained children 7–8 years of age. J Bone Miner Res 2003; 18: 104350.
  • 38
    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: 1926.
  • 39
    Leonard MB, Feldman HI, Shults J, Zemel BS, Foster BJ, Stallings VA. Long-term, high-dose glucocorticoids and bone mineral content in childhood glucocorticoid-sensitive nephrotic syndrome. N Engl J Med 2004; 351: 86875.
  • 40
    Ward LM, Glorieux FH. The spectrum of pediatric osteoporosis. In: GlorieuxFH, editor. Pediatric bone. San Diego: Academic Press; 2003. p. 40131.