Parvovirus B19 and Onset of Juvenile DermatomyositisRESEARCH LETTER

To the Editor: Juvenile dermatomyositis (JDM) is a systemic autoimmune disease characterized by chronic muscle and cutaneous inflammation of unknown cause; however, some studies have suggested that infectious agents may trigger the onset of JDM and other childhood systemic autoimmune diseases.¹ Several studies have described the detection of parvovirus B19 DNA in skin or muscle of patients with JDM, adult dermatomyositis, and other systemic rheumatic diseases.² Given the similarity of the disease presentation of JDM and parvovirus B19 infection, we performed a case-control study to determine if parvovirus B19 infection is associated with the onset of JDM.

A plasma sample and clinical questionnaire were obtained from 62 patients with JDM (45 girls, mean [SD] age, 9.7 [4.6] years) meeting probable or definite Bohan and Peter criteria³ as part of the Childhood Myositis Heterogeneity Study, a North American registry of patients with juvenile idiopathic inflammatory myopathies. All patients in our study were within 6 months of diagnosis (mean [SD] time after diagnosis, 1.8 [1.6] months), and 86% of the patients were diagnosed between 1995 and 1998. An equal number of healthy control children, matched by age, sex, and race, were enrolled at the National Institutes of Health Clinical Center in Bethesda, Md, from 1995 to 1997 and resided in the area. All patients and parents signed written informed consent for an institutional review board–approved study.

Muscle biopsies were obtained at diagnosis before initiation of therapy from 4 patients with JDM (mean [SD] age, 6.0 [2.9] years), 2 of whom also provided plasma samples, and from a second group of 4 control patients with a history of muscle weakness without a defined myopathy on histological examination (mean [SD] age, 10.0 [3.2] years); these control patients did not provide a plasma sample. Five 5-μm sections from each biopsy were used for DNA extraction. Plasma and biopsy specimens were stored at −80°C before assay in 2004.

Parvovirus B19 IgM and IgG antibodies were detected by commercial enzyme-linked immunosorbent assay (Biotrin International, Dublin, Ireland). Parvovirus B19 DNA was detected by dotblot hybridization (sensitivity >10⁶ genome equivalents [gEq]/mL) and 2 different polymerase chain reaction (PCR) assays (sensitivity <1 gEq/μL).⁴ The PCR results were confirmed by real-time PCR and sequencing. All laboratory studies were performed blinded to clinical information or patient status. Comparisons were made using the McNemar test. The power to detect a difference of 0.17 in the odds ratio, using a 2-tailed test, was estimated to be 0.60. All statistical analyses were performed using nQuery Advisor 5.0 (Statistical Solutions, Cork, Ireland). Statistical significance was set at P<.05.

Only 1 plasma sample from a control patient had detectable parvovirus B19 IgM (Table 1). Parvovirus B19 IgG was detected in the plasma of 25 patients (40.3%) with JDM and 36 age-matched healthy controls (58.1%) (P = .04) (Table 2). All samples were negative for parvovirus B19 DNA by dotblot hybridization, and 2 patients with JDM and 2 control patients had detectable parvovirus B19 DNA in plasma by PCR (Table 1). The titer was low in all (each <1-5 gEq/μL plasma) and sequences were genotype 1.

Parvovirus B19 genotype 1 DNA was detected in low titer (<1 gEq/μL extracted DNA) in the muscle tissue of 1 of 4 patients with JDM and in 0 of 4 control patients. The patient with positive parvovirus B19 genome in the muscle also had parvovirus B19 IgG antibody detected in a plasma sample obtained 31 months after diagnosis, but was parvovirus B19 IgM negative.

In our case-control study, parvovirus B19 infection was not positively associated with the onset of JDM, with no increased prevalence of parvovirus B19 DNA in the plasma or muscle of patients with JDM compared with age-matched controls. The rate of parvovirus B19 seropositivity in patients with JDM is similar to that published for healthy children,⁵ and the higher seroprevalence in controls may be related to geographical differences in patient recruitment (nationwide for patients with JDM vs Washington, DC, metro area for controls). These results indicate that previous anecdotal data suggesting a role for parvovirus B19 in the onset of JDM and other systemic autoimmune diseases may be misleading.