BOSTON — Small joint involvement and symptoms that last for more than 2 weeks are among the most important early predictors of the evolution of juvenile idiopathic arthritis in children with acute recent onset of the disease, according to findings from a Norwegian study.
A significant percentage of what appears to be juvenile idiopathic arthritis (JIA) actually is transient arthritis, according to the findings.
Knee joint involvement, a neutrophile white blood cell count within the normal range, and a platelet count above the normal limit are also significant predictors of the chronic inflammatory arthritis in children with early symptoms, Dr. Kai S. Handeland reported at the annual meeting of the American College of Rheumatology.
The inability to distinguish between JIA and other types of recent-onset childhood arthritis at an early stage limits the degree to which aggressive management can be carried forth, said Dr. Handeland of Rikshospitalet-Radiumhospitalet medical center, Oslo.
To determine whether any clinical or laboratory features in children with recent-onset arthritis might predict the diagnosis of JIA, Dr. Handeland and colleagues conducted a multicenter population-based study in three Norwegian counties between May 2004 and June 2005. During this period, the investigators asked all primary care physicians in the region to refer for study inclusion all children aged younger than 16 years with symptoms consistent with recent-onset arthritis or osteomyelitis, confirmed by clinical or radiological examination. In addition to prospectively enrolled patients, children who met inclusion criteria were identified retrospectively by searching electronic medical records for relevant diagnoses.
Of the region's 255,303 children, 504 were referred for possible inclusion in the study. A total of 214 patients were enrolled in the study and followed up for a minimum of 6 months, at which point a final diagnosis was made. The investigators reevaluated the diagnoses by chart review after 2 years.
Of the 214 children with confirmed arthritis symptoms, 40 had JIA, 8 had septic arthritis, 5 had osteomyelitis with coexisting arthritis, 21 had poststreptococcal reactive arthritis, 2 had acute rheumatic fever, 4 had Borrelia-arthritis, 1 had enteropathic arthritis, 61 had transient synovitis of the hip, and 72 had other types of transient arthritis, Dr. Handeland said in a poster presentation.
To identify predictors of JIA, the investigators used multiple logistic regression analyses with JIA diagnosis as the dependent variable. Of the factors identified as statistically significant predictors of JIA, the presence of small joint involvement at presentation was the most common, with an odds ratio of 14.1. The odds ratios for symptom duration of 14 days or more, normal neutrophile count, knee-joint involvement, and elevated platelet count, were, respectively, 13.3, 6.2, 4.1, and 3.4, reported Dr. Handeland. Patient gender, monoarthritis, hip-joint involvement, and elevated temperature were not significantly predictive of a JIA diagnosis, he said.
The investigators also conducted subsequent analyses to determine predictors of JIA versus transient/postinfectious arthritis and predictors of JIA versus infectious arthritis.
With respect to the former, the same determinants that discriminate JIA from all arthritis discriminate between JIA and transient/postinfectious arthritis, according to Dr. Handeland.
The determinants that discriminate JIA from infectious arthritis at the time of symptom presentation are the absence of fever and a low occurrence of hip-joint involvement, he said.
Dr. Handeland reported having no conflicts of interest to disclose relative to this presentation.