SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

Objective

To determine early predictors of 6-month outcomes in a prospective cohort of patients with juvenile idiopathic arthritis (JIA).

Methods

Patients selected were those enrolled in an inception cohort study of JIA, the Research in Arthritis in Canadian Children Emphasizing Outcomes Study, within 6 months after diagnosis. The juvenile rheumatoid arthritis core criteria set and quality of life measures were collected at enrollment and 6 months later. Outcomes evaluated included inactive disease, Juvenile Arthritis Quality of Life Questionnaire (JAQQ) scores, and Childhood Health Assessment Questionnaire (C-HAQ) scores at 6 months.

Results

Thirty-three percent of patients had inactive disease at 6 months. Onset subtype and most baseline core criteria set measures correlated with all 3 outcomes. Relative to oligoarticular JIA, the risks of inactive disease were lower for enthesitis-related arthritis, polyarthritis rheumatoid factor (RF)–negative JIA, and polyarthritis RF-positive JIA, and were similar for psoriatic arthritis. In multiple regression analyses, the baseline JAQQ score was an independent predictor of all 3 outcomes. Other independent baseline predictors included polyarthritis RF-negative and systemic JIA for inactive disease; C-HAQ score and polyarthritis RF-positive JIA for the 6-month C-HAQ score; and active joint count, pain, and time to diagnosis for the 6-month JAQQ score.

Conclusion

Clinical measures soon after diagnosis predict short-term outcomes for patients with JIA. The JAQQ is a predictor of multiple outcomes. Time to diagnosis affects quality of life in the short term.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

Outcomes of children with chronic arthritis have progressively improved over the past 40 years (1, 2). Nevertheless, in recent publications, remissions within 5 to 10 years of disease onset occur in a minority of patients, with the exception of those with pauciarticular juvenile rheumatoid arthritis (JRA) (1–4). Moreover, severe disability, defined as Steinbrocker classes III/IV or Childhood Health Assessment Questionnaire (C-HAQ) indices of >1.5, still occurs in 3–12% of patients, and joint damage is found as early as 1 to 2 years after diagnosis (1–9). Current treatment strategies for children with chronic arthritis include the early use of intraarticular corticosteroids, disease-modifying antirheumatic drugs (DMARDs), and biologic agents (10–12). The goal of this approach is the achievement of inactive disease or remission early on in the hope of avoiding these poor outcomes.

Previous studies concentrated on long-term outcomes, and in most reports patients were classified according to American College of Rheumatology criteria for JRA or European League Against Rheumatism criteria for juvenile chronic arthritis (2–4, 13–19). In these studies, JRA or juvenile chronic arthritis subtypes consistently predicted disease outcome (15, 18, 20–22). However, even within subtypes, variations in disease duration before remission, disability, and radiographic joint damage were evident (5, 18, 23). The more recent juvenile idiopathic arthritis (JIA) classification is not directly comparable with previous classifications; however, it offers the advantage of including all types of chronic arthritis in children and adolescents, allowing comparisons among them (24). This is particularly important for psoriatic arthritis (PsA) and enthesitis-related arthritis (ERA), for which disease course and outcomes are not as well documented (25).

Although more aggressive treatment approaches provide the possibility of improved disease control, determination of which patients warrant these therapies is critical to the optimal balance of efficacy and risk. Past studies of risk factors for poor outcomes may not be applicable today due to differences in treatments and expectations of outcome. On the other hand, data on long-term outcomes of patients managed today are not yet available. Since short-term disease control may in turn favor better long-term outcomes, we examined predictors of short-term outcomes in an inception cohort of patients with JIA managed with current therapies. The specific goals of this study were to determine if patient characteristics, disease activity, and quality of life measures at baseline were predictive of inactive disease, physical function, and quality of life 6 months later.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

Research in Arthritis in Canadian Children emphasizing Outcomes (ReACCh-Out) Study.

ReACCh-Out is an ongoing multicenter prospective inception cohort study of JIA conducted at 16 pediatric rheumatology centers in Canada (14 academic and 2 community centers). The inclusion criterion is a diagnosis of JIA according to International League of Associations for Rheumatology (ILAR) criteria made within 12 months before enrollment (24). Demographic data and rheumatologic, family, and medication histories are prospectively collected on standardized prepared forms at enrollment, and interim and medication histories are collected at subsequent study visits every 6 months for the first 2 years and then yearly. The 6-item JRA core criteria set, the Juvenile Arthritis Quality of Life Questionnaire (JAQQ), and the Quality of My Life Questionnaire are completed at each study visit (26–28). Where developmentally appropriate, children complete their own self-report questionnaires. Otherwise, they are completed by a parent. The study protocol was approved by the research ethics boards at each participating center. Informed consent for participation was obtained from parents and informed consent or assent was obtained from patients as applicable.

Participants.

The subset of participants selected for the present report was enrolled in ReACCh-Out within 6 months after diagnosis between January 1, 2005 and December 31, 2007, had completed baseline and 6-month assessments, and had sufficient data to determine the primary outcome of active or inactive disease (defined below) at 6 months. The requirement of a maximum interval of 6 months between diagnosis and enrollment ensured capturing a subset of patients whose early disease course was minimally altered at baseline, although disease course from onset to diagnosis was not studied.

JIA subtype diagnoses.

ILAR JIA subtype diagnoses were first assigned by site investigators at enrollment. For each participant, subtype assignments and enrollment data were reviewed for compliance with ILAR criteria by one of the authors (KO). Any discrepancies were reviewed once more by 2 additional investigators (DAC) using a computerized program designed by them specifically for the validation of ILAR diagnoses. Diagnoses were reassigned if the 2 reviewers were in agreement or referred to the site investigator if information not captured in the enrollment forms was required. As a result of this review process, JIA subtypes were reassigned for 36 of 356 patients included in the study: 18 to undifferentiated JIA, 17 due to exclusions (first-degree relative family histories of psoriasis [n = 11], ankylosing spondylitis [n = 2], psoriasis and ankylosing spondylitis [n = 2], and acute uveitis [n = 1], and presence of psoriasis [n = 1]), and 1 due to fulfillment of criteria for more than one category. The remaining 18 were reassigned to other subtypes.

Outcomes and determinant variables.

The primary outcome assessed in this report was clinically inactive disease at the 6-month assessment. Secondary outcomes were 6-month C-HAQ and JAQQ scores. Based on the previously published criteria by Wallace et al, clinically inactive disease was defined as an active joint count and physician's global assessment of disease activity of 0; absence of fevers, rash, serositis, splenomegaly, or generalized lymphadenopathy attributable to JIA; and absence of active uveitis, regardless of medication use (29). Because many patients (40%) had neither erythrocyte sedimentation rate (ESR) nor C-reactive protein (CRP) results at the 6-month visit, these measures were not included in the definition.

The possible predictive variables of inactive disease we investigated included demographic data, JIA subtype at enrollment, disease duration from onset to diagnosis and from onset to enrollment, and the JRA core criteria set measures, JAQQ score, and pain scores collected at enrollment. Date of onset of JIA was defined as the date of symptom onset approximated to the nearest month or to the middle month of a season. The core criteria set consist of the following: 1) active joint count (potential range 0–71), 2) number of active joints with limited range of movement (ROM; range 0–71), 3) physician's global assessment of disease activity measured using a 10-cm visual analog scale (VAS), where higher scores indicate greater disease activity, 4) patient or parent global assessment of overall well-being measured on a 10-cm VAS, where higher scores indicate worse well-being, 5) a measure of function, i.e., the C-HAQ in our protocol (potential score 0–3, where higher scores indicate greater physical disability), and 6) ESR or CRP level (26, 30). The JAQQ is a quality of life measure that covers 4 domains: gross motor, fine motor, psychosocial, and general symptoms (27). Potential scores for the JAQQ are 1–7, where higher scores denote worse quality of life (27). Pain was measured by a 0–10-cm VAS.

Statistical analysis.

To determine relationships between continuous baseline measures, Spearman's rank correlation coefficients were calculated. To determine predictors of each outcome, univariate analyses were performed first. Each of the JRA core criteria set measures was analyzed separately. Comparisons of groups were made by chi-square, Fisher's exact, Mann-Whitney U, and Kruskal-Wallis tests as appropriate. Relative risks and 95% confidence intervals (95% CIs) were determined from 2 × 2 tables. An increased or decreased risk was considered to be absent if the 95% CIs included 1.0. Spearman's rank correlation coefficients were used to estimate correlations between continuous variables and 6-month C-HAQ or JAQQ scores. Independent (predictor) variables showing a relationship with any of the outcomes at a level of P < 0.10 were examined using general linear modeling. The primary outcome (clinically inactive disease) was examined using logistic regression. Predictor variables were entered by a backward conditional method and confirmed with a forward conditional method. Logistic regression yielded odds ratios (ORs) for the predictor variables, which are an approximation of relative risk. Multiple linear regression models using a backward elimination method were used for analyses of the secondary outcomes 6-month C-HAQ and JAQQ scores, which were confirmed with forward and stepwise methods. The criterion for variable entry was P < 0.05 and for removal was P = 0.10. P values less than 0.05 were significant. Analyses were performed with SPSS, versions 15 and 16 (SPSS, Chicago, IL).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

Patients.

Of 882 eligible children and adolescents with newly diagnosed JIA, 837 were enrolled in ReACCh-Out between January 1, 2005 and December 31, 2007; 45 refused participation. A total of 437 had completed baseline and 6-month assessments. Of the latter, 81 were excluded from the present analyses, 3 due to incomplete 6-month data and 78 because they were enrolled after an interval longer than 6 months following diagnosis. A total of 356 subjects met the criteria for the present report. They were enrolled at a median of 0.65 months (75th percentile 1.9, range 0–6) after diagnosis and a median of 6.0 months (75th percentile 10.5, range 0.1–140) after disease onset. Patients with undifferentiated arthritis formed the third-largest JIA category (11.5%) after oligoarthritis (41%) and polyarthritis rheumatoid factor (RF)–negative subsets (20%). Other demographic data are shown in Table 1. Excluded patients were similar to included patients at diagnosis in age at onset (median 7.4 years [range 1–15.7] for excluded patients), age at diagnosis (median 7.7 years [range 1.3–15.7]), sex (n = 56 girls, n = 25 boys), and subtype distribution at onset (38% oligoarthritis, 7% systemic, 17% polyarthritis RF-negative JIA, 6% polyarthritis RF-positive JIA, 3% PsA, 17% ERA, and 11% undifferentiated arthritis), but had a shorter interval from symptom onset to diagnosis (median 4.0, range 0.2–95.1 months versus median 4.5, range 0.1–139 months; P = 0.030) and a lower frequency of positive antinuclear antibodies (33% versus 49%; P = 0.011).

Table 1. Characteristics and measures of patients at enrollment*
 Value
  • *

    Values are the median (range) unless otherwise indicated. JIA = juvenile idiopathic arthritis; RF = rheumatoid factor; ANA = antinuclear antibody; ROM = range of motion; C-HAQ = Childhood Health Assessment Questionnaire; ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; JAQQ = Juvenile Arthritis Quality of Life Questionnaire; VAS = visual analog scale.

Age at onset, years (n = 355)7.9 (0.2–16.3)
Age at study enrollment, years (n = 356)9.1 (0.3–16.9)
Age at diagnosis, years (n = 356)9.0 (0.2–16.7)
Time from onset to diagnosis, months (n = 355)4.5 (0.1–139)
Time from onset to enrollment, months (n = 355)6.0 (0.1–140)
Time from diagnosis to enrollment, months (n = 356)0.7 (0–6.0)
No. girls/boys234/122
JIA subtype (n = 356), no. (%) 
 Oligoarticular145 (40.7)
 Systemic26 (7.3)
 Polyarthritis RF negative72 (20.2)
 Polyarthritis RF positive13 (3.7)
 Psoriatic arthritis23 (6.5)
 Enthesitis-related arthritis36 (10.1)
 Undifferentiated41 (11.5)
ANA positive (n = 321), no. (%)157 (48.9)
RF positive (n = 305), no. (%)25 (8.2)
Baseline measures 
 Active joint count (n = 356)2 (0–63)
 Limited ROM joint count (n = 356)1 (0–36)
 Physician's global assessment of  disease activity, cm (n = 353)2.9 (0–10.0)
 Patient or parent global assessment,  cm (n = 322)2 (0–10.0)
 C-HAQ score (n = 322)0.4 (0–3.0)
 ESR, mm/hour (n = 323)22 (0–123)
 CRP level, units/liter (n = 275)5 (0–261)
 JAQQ score (n = 304)2.7 (1–7)
 Pain VAS, cm (n = 321)2.4 (0–9.8)

Medications started before or at baseline included nonsteroidal antiinflammatory drugs (NSAIDs) for 312 (88%), DMARDs for 81 (23%), prednisone for 41 (12%), and intraarticular injections for 37 (10%) patients. Thirty-three (9%) patients were not receiving any medications. Triamcinolone hexacetonide was used almost exclusively for intraarticular injections. DMARDs included methotrexate (n = 73), sulfasalazine (n = 13), hydroxychloroquine (n = 4), and cyclosporine (n = 2). Twelve patients received combinations of these medications. No patients were receiving or were started on biologic therapies at enrollment. By the 6-month visit, an additional 68 (19%) patients had received intraarticular injections, 43 (12%) patients were not receiving any medications, 265 (74%) were receiving NSAIDs, 125 (35%) were receiving DMARDs, 41 (12%) were receiving prednisone, and 2 (0.6%) were receiving biologic therapies.

Baseline measures.

Baseline measures are shown in Table 1. All of the core criteria set measures at baseline were significantly correlated with each other, but there was a wide range in values for correlation coefficients (Table 2). Measures completed by participants (patient or parent global assessment of overall well-being, C-HAQ, JAQQ, and pain VAS) were highly (correlation coefficients 0.61–0.66) or very highly (coefficient 0.74) correlated with each other. In contrast, lower correlation coefficients were found between physician- and patient-completed or parent-completed measures (coefficients 0.23–0.42).

Table 2. Correlations between measures at enrollment*
 Active joint countLimited ROM joint countPhysician globalPatient globalC-HAQ scoreESRCRP levelJAQQ scorePain VAS
  • *

    See Table 1 for abbreviations.

Active joint count         
 Spearman's correlation coefficient1.00.580.650.330.400.230.290.420.38
 P < 0.0001< 0.0001< 0.0001< 0.0001< 0.0001< 0.0001< 0.0001< 0.001
 N356356353322322323275304321
Number of joints with limited ROM         
 Spearman's correlation coefficient0.581.00.410.240.260.110.190.320.230
 P< 0.0001 < 0.0001< 0.0001< 0.00010.050.002< 0.0001< 0.0001
 N356356353322322323275304321
Physician's global assessment of  disease activity         
 Spearman's correlation coefficient0.650.411.00.330.390.260.280.340.28
 P< 0.0001< 0.0001 < 0.0001< 0.0001< 0.00010.001< 0.0001< 0.0001
 N353353353320320320272302319
Patient or parent global assessment         
 Spearman's correlation coefficient0.330.240.331.00.630.230.190.610.74
 P< 0.0001< 0.0001< 0.0001 < 0.0001< 0.00010.002< 0.0001< 0.0001
 N322322320322322292252301321
C-HAQ score         
 Spearman's correlation coefficient0.400.260.390.631.00.230.230.660.65
 P< 0.0001< 0.0001< 0.0001< 0.0001 < 0.0001< 0.0001< 0.0001< 0.0001
 N322322320322322292252301321
ESR         
 Spearman's correlation coefficient0.230.110.260.230.231.00.580.240.17
 P< 0.00010.05< 0.0001< 0.0001< 0.0001 < 0.0001< 0.00010.005
 N323323320292292323264279291
CRP level         
 Spearman's correlation coefficient0.290.190.280.190.230.581.00.180.14
 P< 0.00010.0020.0010.002< 0.0001< 0.0001 0.0040.027
 N275275272252252264275238251
JAQQ score         
 Spearman's correlation coefficient0.420.320.340.610.660.240.1841.00.62
 P< 0.0001< 0.0001< 0.0001< 0.001< 0.001< 0.00010.004 < 0.0001
 N304304302301301279238304300
Pain VAS         
 Spearman's correlation coefficient0.380.230.280.740.650.170.140.621.0
 P< 0.0001< 0.0001< 0.0001< 0.0001< 0.00010.0050.027< 0.0001 
 N321321319321321291251300321

Inactive disease at 6 months after enrollment.

A total of 116 patients (32.6%) achieved the primary outcome of inactive disease status at 6 months (Table 3). Two patients were classified as active due to extraarticular symptoms: fever and rash from systemic JIA in one and asymptomatic uveitis in the other. In order to evaluate predictors of inactive disease, univariate analyses were first performed. Characteristics that correlated with inactive disease at 6 months were a shorter time from symptom onset to diagnosis and younger age at both diagnosis and enrollment (Table 2). Of the 7 JIA subtypes, oligoarticular JIA had the best outcome, with a 46% rate of inactive disease. Relative to oligoarticular JIA, the relative risks of inactive disease were reduced for patients with polyarthritis RF-negative JIA (0.43; 95% CI 0.26, 0.71) and ERA (0.56; 95% CI 0.35, 0.90), but not for those with polyarthritis RF-positive JIA (0.17; 95% CI 0.03, 1.120), systemic JIA (0.59; 95% CI 0.31, 1.14), or PsA (0.76; 95% CI 0.43, 1.38).

Table 3. Univariate comparisons of baseline characteristics and measures in patients with inactive and active disease at 6 months*
 InactiveActiveP
NValueNValue
  • *

    Values are the median (range) unless otherwise indicated. Inactive disease was defined as an active joint count and physician's global assessment of disease of 0, and absence of uveitis, fevers, rash, lymphadenopathy, or hepatosplenomegaly attributable to JIA, regardless of medication use. Comparisons were made by Mann-Whitney U, chi-square, or Fischer's exact tests. See Table 1 for abbreviations.

  • With or without the undifferentiated group.

Age at onset, years1156.8 (0.3–16.2)2408.5 (0.2–16.3)0.10
Age at study enrollment, years1167.7 (0.8–16.4)2409.6 (0.3–16.9)0.026
Age at diagnosis, years1167.6 (0.8–16.4)2409.5 (0.2–16.7)0.023
Time from onset to diagnosis, months1154.8 (0.1–66.6)2405.5 (0.1–139)0.003
Time from onset to enrollment, months1155.1 (0.1–68.8)2406.2 (0.6–140)0.029
Time from diagnosis to enrollment, months1160.9 (0-5.2)2400.5 (0-6.0)0.11
No. girls/boys 82/34 152/880.17
JIA subtype, no. (%) of subtype116 240 0.001
 Oligoarticular 66 (56.9)/45.5 79 (32.9)/54.5 
 Systemic 7 (6.0)/26.9 19 (7.9)/73.1 
 Polyarthritis RF negative 14 (12.1)/19.4 58 (24.2)/80.6 
 Polyarthritis RF positive 1 (0.9)/7.7 12 (5.0)/92.3 
 Psoriatic arthritis 8 (6.9)/34.8 15 (6.2)/65.2 
 Enthesitis-related arthritis 7 (6.0)/19.4 29 (12.1)/80.6 
 Undifferentiated 13 (11.2)/31.7 28 (11.7)/68.3 
ANA positive, no. (%)10451 (49.0)217106 (48.8)0.98
RF positive, no. (%)973 (3.1)20822 (10.6)0.026
Baseline measures     
 Active joint count1161 (0–27)2403 (0–63)< 0.0001
 Limited ROM joint count1161 (0–20)2401 (0–36)0.10
 Physician's global assessment of disease  activity, cm1162.1 (0–8.0)2373.8 (0–10.0)< 0.0001
 Patient or parent global assessment, cm1031.1 (0–9.5)2192.0 (0–10.0)0.013
 C-HAQ score1030.3 (0–3.0)2190.5 (0–2.8)0.001
 ESR, mm/hour9920 (1–123)22422 (0–115)0.56
 CRP level, units/liter814 (0–261)1946 (0–217)0.13
 JAQQ score972.3 (1–5)2072.9 (1–7)< 0.0001
 Pain VAS, cm1031.4 (0–9.0)2183.3 (0–9.8)0.001

Each of the core criteria set measures, apart from number of joints with limited ROM, ESR, and CRP level, were significant predictors of disease status at 6 months (Table 3). Lower numbers of active joints, physician's global assessment of disease activity, patient or parent global assessment of overall well-being, and C-HAQ scores at enrollment all predicted inactive disease at 6 months. Baseline JAQQ and pain VAS scores were also significantly lower in those with inactive disease at 6 months (Table 3).

Variables showing associations with disease status at 6 months in univariate tests at a significance level of P < 0.10 were entered in a regression analysis to determine independent predictors (Table 4). Only patients with data for each of these variables were entered. Among the 3 physician-completed measures, only baseline number of active joints was entered, as this was highly correlated with both number of joints with limited ROM and physician's global assessment of disease activity. Moreover, all 3 were measures of disease activity and severity, so that it was not necessary to control these variables for each other. Baseline JAQQ score, time from symptom onset to diagnosis, and onset subtype remained in the final regression model. The JAQQ had a significant independent predictive effect with higher score, decreasing the odds of inactive disease at 6 months. When active joint count and other variables were controlled, the odds of inactive disease for most JIA subtypes were lower compared with oligoarthritis (OR 0.35–0.69) (Table 4). The difference reached significance only for polyarthritis RF-negative and systemic JIA. The OR for PsA (0.69) was the closest to 1.0. No cases of polyarthritis RF-positive JIA entered in the regression had inactive disease at 6 months, resulting in an OR of 0. The undifferentiated category was not included in the regression because it represented a heterogeneous group of patients, thus obscuring a true JIA subtype effect.

Table 4. Regression analysis for inactive disease at 6 months (n = 263)*
Explanatory variable (measure at enrollment)OR (95% CI)P
  • *

    The Cox and Snell R2 value for the regression analysis was 0.13. Variables correlated with inactive disease in univariate tests at a level of P <0.10 were entered, and those remaining in the final model are shown. Variables eliminated in the regression included number of active joints, patient/parent global assessment, C-HAQ score, pain, age at enrollment, and age at diagnosis. The forward conditional method gave different results, with number of active joints (P = 0.017), baseline JAQQ score (P = 0.013), and time from onset to diagnosis (P = 0.058) remaining in the regression; but the Cox and Snell R2 value was lower (0.10). OR = odds ratio; 95% CI = 95% confidence interval; JAQQ = Juvenile Arthritis Quality of Life Questionnaire; JIA = juvenile idiopathic arthritis; RF = rheumatoid factor; ERA = enthesitis-related arthritis.

JAQQ score0.70 (0.55, 0.89)0.004
Time from onset to diagnosis, months0.97 (0.94, 1.0)0.055
JIA diagnosis 0.075
 Oligoarticular (n = 123)1.0 
 Systemic (n = 23)0.35 (0.13, 0.98)0.045
 Polyarthritis RF negative  (n = 59)0.38 (0.18, 0.80)0.011
 Polyarthritis RF positive  (n = 9)0.00 (0.00) 
 Psoriatic arthritis (n = 17)0.69 (0.24, 2.05)0.51
 ERA (n = 32)0.46 (0.18, 1.17)0.10
Constant2.630.012

Physical function (C-HAQ) at 6 months after enrollment.

A total of 310 patients completed the C-HAQ at 6 months, and the median value was 0.1 (range 0–2.8). The evaluation of predictors was performed as above. With the exception of ESR and CRP level, each of the baseline core criteria set measures, JAQQ scores, pain VAS at enrollment, and time from onset to diagnosis were positively correlated with C-HAQ scores at 6 months in univariate analyses (Table 5). Of these, baseline C-HAQ, JAQQ, and pain VAS scores had the highest correlations (Table 5). There were also significant differences in this outcome among JIA subtypes.

Table 5. Predictors of C-HAQ scores at 6 months (n = 310)*
 Univariate analysesLinear regression
NCoefficientPB (95% CI)βP
  • *

    The median C-HAQ score for all patients at 6 months was 0.1 (range 0–2.8). Spearman's rank correlation coefficients were calculated to assess correlations between baseline measures and C-HAQ scores at 6 months, except as noted. A total of 241 patients were entered in the final model and the R2 for the regression was 0.23. Results with forward and stepwise entry of variables were similar. C-HAQ = Childhood Health Assessment Questionnaire; B = regression coefficient; 95% CI = 95% confidence interval; β = standardized coefficient; ROM = range of motion; ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; JAQQ = Juvenile Arthritis Quality of Life Questionnaire; VAS = visual analog scale; JIA = juvenile idiopathic arthritis; RF = rheumatoid factor; ERA = enthesitis-related arthritis.

  • By Mann-Whitney U test.

  • By Kruskal-Wallis test.

Variable at enrollment      
 Number of active joints3100.28< 0.0001Eliminated  
 Number of joints with limited ROM3100.24< 0.0001Not entered  
 Physician's global assessment of  disease activity3080.190.001Not entered  
 Patient or parent global assessment2910.32< 0.0001Eliminated  
 C-HAQ score2910.43< 0.00010.17 (0.037, 0.30)0.200.012
 ESR, mm/hour280−0.0040.94Not entered  
 CRP level, units/liter2450.0150.82Not entered  
 JAQQ score2760.44< 0.00010.096 (0.037, 0.16)0.250.002
 Pain VAS, cm2900.35< 0.0001Eliminated  
Age at onset, years309−0.020.71   
Time from onset to diagnosis, months3090.110.0440.003 (0, 0.007)0.110.074
Sex310 0.90Not entered  
JIA subtype273 0.001   
 Oligoarticular   Reference  
 Systemic   0.14 (−0.056, 0.34)0.090.16
 Polyarthritis RF negative   0.046 (−0.097, 0.19)0.040.53
 Polyarthritis RF positive   0.39 (0.049, 0.73)0.140.025
 Psoriatic arthritis   0.00003 (−0.23, 0.23)01.0
 ERA   0.017 (−0.16, 0.20)0.010.85
Constant   −0.13 (−0.28, 0.025) 0.10

In the linear regression analysis, baseline C-HAQ and JAQQ scores and polyarthritis RF-positive subtype were significant independent predictors for the C-HAQ at 6 months (Table 5). The regression explained 23% of the variation in this outcome.

Quality of life (JAQQ) at 6 months.

A total of 300 patients completed the JAQQ at 6 months, and the median value was 1.8 (range 1.0–6.3). The same variables were correlated with JAQQ scores as with C-HAQ scores at 6 months in univariate analyses (Table 6). Significant independent predictors for the JAQQ score at 6 months were baseline number of active joints, JAQQ score and pain measures, and time from onset to diagnosis (Table 6). The baseline JAQQ score had the greatest predictive effect (standardized coefficient β = 0.39), followed by time to diagnosis (β = 0.21) and pain VAS (β = 0.16), whereas active joint count had the least effect (β = 0.12). Adjusting for the other variables entered, JIA subtype at enrollment did not have an independent effect on the JAQQ score at 6 months. Altogether, 37% of the variation in the JAQQ score was explained (Table 6).

Table 6. Predictors of JAQQ scores at 6 months (n = 300)*
 Univariate analysesLinear regression
NCoefficientPB (95% CI)βP
  • *

    The median value of the JAQQ score at 6 months was 1.8 (range 1.0–6.3). A total of 236 patients were entered in the final model of the regression analysis and the R2 value was 0.37. Forward and stepwise entry of variables yielded similar results. See Table 5 for definitions.

Variable at enrollment      
 Number of active joints3000.27< 0.00010.015 (0.24, 0.49)0.120.042
 Number of joints with limited ROM3000.21< 0.0001Not entered  
 Physician's global assessment of  disease activity2990.150.009Not entered  
 Patient or parent global assessment2840.38< 0.0001Eliminated  
 C-HAQ score2840.38< 0.0001Eliminated  
 ESR, mm/hour273−0.0020.97Not entered  
 CRP level, units/liter237−0.0380.56Not entered  
 JAQQ score2720.56< 0.00010.37 (0.24, 0.49)0.39< 0.0001
 Pain VAS, cm2830.41< 0.00010.067 (0.012, 0.12)0.160.017
Age at onset, years2990.0120.84Not entered  
Time from onset to diagnosis, months2990.190.0010.015 (0.007, 0.023)0.21< 0.0001
Sex300 0.97Not entered  
JIA subtype263 0.015Eliminated  
Constant   0.70 (0.39, 1.01) < 0.0001

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

In this study, we show that disease control at a median of 6.7 months after diagnosis (6 months after enrollment) is achievable in 33% of patients with JIA managed with currently accepted treatment strategies, and that most of the core criteria set and quality of life measures at baseline are predictive of short-term outcomes. The primary outcome chosen for the present analysis was inactive disease, because complete disease control is an aim of current treatment and is of increasing interest in the literature (10, 29, 31, 32). Treatment strategies used in this cohort reflect current approaches and included intraarticular medications and early use of DMARDs. The modest number of patients receiving biologic treatment at the early end point of this study may reflect the initial use of other therapies. Treatment effects on outcomes were not assessed because this was not an intervention study. Advantages of our study include a prospective design and nationwide patient recruitment. These considerations reduce bias by ensuring a study population representative of the full spectrum of disease. Virtually all patients diagnosed with JIA in Canada were enrolled in REACCH-Out, but an additional inclusion criterion for the present report was a short duration from diagnosis to enrollment. This ensured that the variables assessed were measured as early as possible for a reasonable sample size of patients within the ReACCh-Out program. This strategy allowed an evaluation of disease characteristics and measures early in disease to predict clinical outcomes, although events between symptom onset and diagnosis could not be controlled.

A drawback of our study was the omission of normal acute-phase reactants in the definition of inactive disease. Our protocol was designed to fit into clinical practice and the low number of patients with ESR or CRP results at 6 months reflected the reluctance of many clinicians to initiate tests that were not clinically indicated. However, the variation from the definition proposed in the literature introduces potential difficulties in comparisons with other reports (29, 31, 32). There were also possible sources of unintentional bias, including incomplete baseline data sets for a number of patients and the exclusion of patients with extended intervals between diagnosis and enrollment, who differed from those included in a few disease characteristics. Although the latter differences were of low significance, they were unexpected and the reasons for them are unclear. Other limitations of this study include the relatively small sample sizes for the rarer JIA subsets recruited in this prospective study within the period of this analysis, a relatively short followup period, and assessment of inactive disease at a single time point. Furthermore, despite early enrollment after diagnosis, the median duration of 6 months since symptom onset at baseline was still considerable and resulted in outcome assessment at a median clinical disease duration of 1 year.

Our results show that onset subtype and baseline measures of disease activity, function, and quality of life are predictive of all 3 short-term outcomes assessed. Higher values of most baseline measures were associated with worse 6-month outcomes. However, interpreting independent predictors in the multivariable analyses was not straightforward. For example, of the measures tested, only JAQQ score was an independent (negative) predictor of inactive disease, but the correlations among measures suggest overlap. Furthermore, JIA subtype may also be reflected in disease activity measures. For example, oligoarticular JIA is limited to a lower number of active joints at baseline, whereas other subtypes are not.

In the regression analyses, baseline JAQQ score was an independent predictor not only of the 6-month JAQQ score, but also of disease status and physical function, findings that have not previously been reported. This may be because the JAQQ is a comprehensive instrument reflecting disease activity and disability as well as the impact of disease on a patient's quality of life. The observations made suggest that the JAQQ is a useful instrument for the prediction of outcomes at least in the short term. Further followup of this cohort will determine how the JAQQ performs for long-term predictions.

Time to diagnosis and pain were also important predictors, correlating with physical function and quality of life outcomes, although significant independent effects were only noted for quality of life. The results suggest that these outcomes may be improved by earlier diagnosis, and presumably earlier treatment. Pain may also be a more important immediate determinant of the JAQQ and C-HAQ than disability in early disease, since at baseline pain was highly correlated with these measures but not with the number of joints with limited ROM.

Our analysis includes children and adolescents with PsA and ERA, subtypes that have received less attention in studies of outcomes. The prognosis for disease control in the short term for patients with PsA was close to that for oligoarticular JIA, whereas the prognosis for patients with ERA may be worse. If confirmed, these short-term findings would be consistent with the greater long-term disability in patients with ERA recently reported by Flato et al (25).

The significance of our findings will depend on the relationship between short-term disease control and the subsequent disease course. There is some, although scant, evidence to suggest that early disease suppression leads to improved long-term outcomes. First, the risk of radiographic joint damage and disability increases with duration of active arthritis (7, 15, 22, 33–35). Furthermore, in children with systemic JRA, reduced disease activity at 6 months after onset predicts less disability and joint destruction, and earlier remission (5, 13, 36, 37). However, Wallace et al have shown that inactive disease may be episodic, with only 50% of children who achieve remission remaining in remission for more than 6 months and 25% having sustained remission off medications (32). Followup of our cohort will determine if the outcomes seen in this study are maintained or improved in the long term and whether favorable short-term outcomes are predictive of better long-term prognoses. In the meantime, the predictors identified in our analyses may be useful in guiding treatment choices aimed at early disease control.

Our results are difficult to compare with previous publications, because most have analyzed disease remission in patients followed for 5 years or longer, studied patients who were not treated with currently used approaches, or classified patients using other diagnostic criteria (3, 4, 23, 34, 35). In the study by Bowyer et al, 50% of patients with pauciarticular JRA, 55% with systemic JRA, and 81% with polyarticular JRA were still receiving medications at 1 year after diagnosis (6). Except for patients with systemic JIA, these numbers are similar to the proportions of patients with persistently active disease in the present study, despite a shorter followup. However, in the study by Bowyer et al, the status of disease control was not recorded and those receiving medications may have included both patients with active disease and those with inactive disease.

In conclusion, we have shown that clinical measures of disease and patient- or parent-completed measures shortly after diagnosis can be used to predict clinically important 6-month outcomes in a large prospective inception cohort of children with JIA. Altogether, our results suggest that it may be possible to identify children at risk of poor outcomes early in their illness course, and subsequently intervene to reduce this risk through treatment and monitoring. The ReACCh-Out Study is ongoing and will generate data on larger numbers of children and longer periods of followup. This will ultimately allow us to determine if both early disease measures and short-term outcomes are predictive of long-term outcomes.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Oen had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Oen, Tucker, Miettunen, Feldman, Yeung, Duffy.

Acquisition of data. Oen, Tucker, Huber, Miettunen, Scuccimarri, Campillo, Cabral, Tse, Chédeville, Spiegel, Schneider, Lang, Ellsworth, Ramsey, Dancey, Silverman, Chetaille, Cameron, Johnson, Dorval, Petty, Watanabe Duffy, Boire, Haddad, Houghton, Saint-Cyr, Turvey, Benseler, Yeung, Duffy.

Analysis and interpretation of data. Oen, Huber, Miettunen, Cabral, Feldman, Spiegel, Benseler, Cheang, Yeung.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES

The authors want to thank Jaime Guzman, MD, MSc, Peter N. Malleson, MBBS, Victor Espinosa, MSc: University of British Columbia, Vancouver, British Columbia, Canada; Katherine Gross, MBBS: Penticton Regional Hospital, Penticton, British Columbia, Canada; Ronal Laxer, MD: The Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada; Alan M. Rosenberg, MD: University of Saskatchewan, Saskatoon, Saskatchewan, Canada; and Michelle Gibbon, BA, Garbis Meshefedjian, PhD: McGill University, Montreal, Quebec, Canada, for their substantial contribution to acquisition of data, data analysis, final approval of the article, and/or project coordination. The authors also thank all of the patients in the ReACCh-Out Study for their generous participation and the research assistants at each center.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. Acknowledgements
  9. REFERENCES
  • 1
    Oen K. Long-term outcomes and predictors of outcomes for patients with juvenile idiopathic arthritis. Best Pract Res Clin Rheumatol 2002; 16: 34760.
  • 2
    Adib N, Silman A, Thomson W. Outcome following onset of juvenile idiopathic inflammatory arthritis. I. Frequency of different outcomes. Rheumatology (Oxford) 2005; 44: 9951001.
  • 3
    Oen K, Malleson PN, Cabral DA, Rosenberg AM, Petty RE, Cheang M. Disease course and outcome of juvenile rheumatoid arthritis in a multicenter cohort. J Rheumatol 2002; 29: 198999.
  • 4
    Minden K, Kiessling U, Listing J, Niewerth M, Doring E, Meincke J, et al. Prognosis of patients with juvenile chronic arthritis and juvenile spondyloarthropathy. J Rheumatol 2000; 27: 225663.
  • 5
    Schneider R, Lang BA, Reilly BJ, Laxer RM, Silverman ED, Ibanez D, et al. Prognostic indicators of joint destruction in systemic-onset juvenile rheumatoid arthritis. J Pediatr 1992; 120: 2005.
  • 6
    Bowyer SL, Roettcher PA, Higgins GC, Adams B, Myers LK, Wallace C, et al. Health status of patients with juvenile rheumatoid arthritis at 1 and 5 years after diagnosis. J Rheumatol 2003; 30: 394400.
  • 7
    Oen K, Reed M, Malleson PN, Cabral DA, Petty RE, Rosenberg AM, et al. Radiologic outcome and its relationship to functional disability in juvenile rheumatoid arthritis. J Rheumatol 2003; 30: 83240.
  • 8
    Mason T, Reed AM, Nelson AM, Thomas KB. Radiographic progression in children with polyarticular juvenile rheumatoid arthritis: a pilot study. Ann Rheum Dis 2005; 64: 4913.
  • 9
    Mason T, Reed AM, Nelson AM, Thomas KB, Patton A, Hoffman AD, et al. Frequency of abnormal hand and wrist radiographs at time of diagnosis of polyarticular juvenile rheumatoid arthritis. J Rheumatol 2002; 29: 22148.
  • 10
    Lovell DJ. Update on treatment of arthritis in children: new treatments, new goals. Bull NYU Hosp Jt Dis 2006; 64: 726.
  • 11
    Weiss JE, Ilowite NT. Juvenile idiopathic arthritis. Rheum Dis Clin North Am 2007; 33: 44170.
  • 12
    Beukelman T, Guevara JP, Albert DA, Sherry DD, Burnham JM. Variation in the initial treatment of knee monoarthritis in juvenile idiopathic arthritis: a survey of pediatric rheumatologists in the United States and Canada. J Rheumatol 2007; 34: 191824.
  • 13
    Spiegel LR, Schneider R, Lang BA, Birdi N, Silverman ED, Laxer RM, et al. Early predictors of poor functional outcome in systemic-onset juvenile rheumatoid arthritis: a multicenter cohort study. Arthritis Rheum 2000; 43: 24029.
  • 14
    Ansell BM. Heberden Oration, 1977: chronic arthritis in childhood. Ann Rheum Dis 1978; 37: 10720.
  • 15
    Gare BA, Fasth A. The natural history of juvenile chronic arthritis: a population based cohort study. II. Outcome. J Rheumatol 1995; 22: 30819.
  • 16
    Flato B, Aasland A, Vinje O, Forre O. Outcome and predictive factors in juvenile rheumatoid arthritis and juvenile spondyloarthropathy. J Rheumatol 1998; 25: 36675.
  • 17
    Brewer EJ Jr, Bass J, Baum J, Cassidy JT, Fink C, Jacobs J, et al. Current proposed revision of JRA criteria: JRA Criteria Subcommittee of the Diagnostic and Therapeutic Criteria Committee of the American Rheumatism Section of The Arthritis Foundation. Arthritis Rheum 1977; 20 Suppl: 1959.
  • 18
    Flato B, Lien G, Smerdel A, Vinje O, Dale K, Johnston V, et al. Prognostic factors in juvenile rheumatoid arthritis: a case-control study revealing early predictors and outcome after 14.9 years. J Rheumatol 2003; 30: 38693.
  • 19
    Ruperto N, Ravelli A, Levinson JE, Shear ES, Murray K, Link Tague B, et al. Long-term health outcomes and quality of life in American and Italian inception cohorts of patients with juvenile rheumatoid arthritis. II. Early predictors of outcome. J Rheumatol 1997; 24: 9528.
  • 20
    Selvaag AM, Lien G, Sorskaar D, Vinje O, Forre O, Flato B. Early disease course and predictors of disability in juvenile rheumatoid arthritis and juvenile spondyloarthropathy: a 3 year prospective study. J Rheumatol 2005; 32: 112230.
  • 21
    Oen K, Malleson PN, Cabral DA, Rosenberg AM, Petty RE, Reed M, et al. Early predictors of longterm outcome in patients with juvenile rheumatoid arthritis: subset-specific correlations. J Rheumatol 2003; 30: 58593.
  • 22
    Adib N, Silman A, Thomson W. Outcome following onset of juvenile idiopathic inflammatory arthritis. II. Predictors of outcome in juvenile arthritis. Rheumatology (Oxford) 2005; 44: 10027.
  • 23
    Guillaume S, Prieur AM, Coste J, Job-Deslandre C. Long-term outcome and prognosis in oligoarticular-onset juvenile idiopathic arthritis. Arthritis Rheum 2000; 43: 185865.
  • 24
    Petty RE, Southwood TR, Manners P, Baum J, Glass DN, Goldenberg J, et al. International League of Associations for Rheumatology classification of juvenile idiopathic arthritis: second revision, Edmonton, 2001. J Rheumatol 2004; 31: 3902.
  • 25
    Flato B, Hoffmann-Vold AM, Reiff A, Forre O, Lien G, Vinje O. Long-term outcome and prognostic factors in enthesitis-related arthritis: a case–control study. Arthritis Rheum 2006; 54: 357382.
  • 26
    Giannini EH, Ruperto N, Ravelli A, Lovell DJ, Felson DT, Martini A. Preliminary definition of improvement in juvenile arthritis. Arthritis Rheum 1997; 40: 12029.
  • 27
    Duffy CM, Arsenault L, Duffy KN, Paquin JD, Strawczynski H. The Juvenile Arthritis Quality of Life Questionnaire: development of a new responsive index for juvenile rheumatoid arthritis and juvenile spondyloarthritides. J Rheumatol 1997; 24: 73846.
  • 28
    Feldman BM, Grundland B, McCullough L, Wright V. Distinction of quality of life, health related quality of life, and health status in children referred for rheumatologic care. J Rheumatol 2000; 27: 22633.
  • 29
    Wallace CA, Ruperto N, Giannini E. Preliminary criteria for clinical remission for select categories of juvenile idiopathic arthritis. J Rheumatol 2004; 31: 22904.
  • 30
    Singh G, Athreya BH, Fries JF, Goldsmith DP. Measurement of health status in children with juvenile rheumatoid arthritis. Arthritis Rheum 1994; 37: 17619.
  • 31
    Solari N, Viola S, Pistorio A, Magni-Manzoni S, Vitale R, Ruperto N, et al. Assessing current outcomes of juvenile idiopathic arthritis: a cross-sectional study in a tertiary center sample. Arthritis Rheum 2008; 59: 15719.
  • 32
    Wallace CA, Huang B, Bandeira M, Ravelli A, Giannini EH. Patterns of clinical remission in select categories of juvenile idiopathic arthritis. Arthritis Rheum 2005; 52: 355462.
  • 33
    Lomater C, Gerloni V, Gattinara M, Mazzotti J, Cimaz R, Fantini F. Systemic onset juvenile idiopathic arthritis: a retrospective study of 80 consecutive patients followed for 10 years. J Rheumatol 2000; 27: 4916.
  • 34
    Zak M, Pedersen FK. Juvenile chronic arthritis into adulthood: a long-term follow-up study. Rheumatology (Oxford) 2000; 39: 198204.
  • 35
    Gare BA, Fasth A. The natural history of juvenile chronic arthritis: a population based cohort study. I. Onset and disease process. J Rheumatol 1995; 22: 295307.
  • 36
    Lang BA, Schneider R, Reilly BJ, Silverman ED, Laxer RM. Radiologic features of systemic onset juvenile rheumatoid arthritis. J Rheumatol 1995; 22: 16873.
  • 37
    Singh-Grewal D, Schneider R, Bayer N, Feldman BM. Predictors of disease course and remission in systemic juvenile idiopathic arthritis: significance of early clinical and laboratory features. Arthritis Rheum 2006; 54: 1595601.