Herpes zoster infection in childhood-onset systemic lupus erythematosus patients: a large multicenter study

Lupus. 2016 Jun;25(7):754-9. doi: 10.1177/0961203315627203. Epub 2016 Jan 27.

Abstract

Objective: The aim of this multicenter study in a large childhood-onset systemic lupus erythematosus (cSLE) population was to assess the herpes zoster infection (HZI) prevalence, demographic data, clinical manifestations, laboratory findings, treatment, and outcome.

Methods: A retrospective multicenter cohort study (Brazilian cSLE group) was performed in ten Pediatric Rheumatology services in São Paulo State, Brazil, and included 852 cSLE patients. HZI was defined according to the presence of acute vesicular-bullous lesions on erythematous/edematous base, in a dermatomal distribution. Post-herpetic neuralgia was defined as persistent pain after one month of resolution of lesions in the same dermatome. Patients were divided in two groups for the assessment of current lupus manifestations, laboratory findings, and treatment: patients with HZI (evaluated at the first HZI) and patients without HZI (evaluated at the last visit).

Results: The frequency of HZI in cSLE patients was 120/852 (14%). Hospitalization occurred in 73 (61%) and overlap bacterial infection in 16 (13%). Intravenous or oral aciclovir was administered in 113/120 (94%) cSLE patients at HZI diagnosis. None of them had ophthalmic complication or death. Post-herpetic neuralgia occurred in 6/120 (5%). After Holm-Bonferroni correction for multiple comparisons, disease duration (1.58 vs 4.41 years, p < 0.0001) was significantly lower in HZI cSLE patients compared to those without HZI. Nephritis (37% vs 18%, p < 0.0001), lymphopenia (32% vs 17%, p < 0.0001) prednisone (97% vs 77%, p < 0.0001), cyclophosphamide (20% vs 5%, p < 0.0001) and SLE Disease Activity Index 2000 (6.0 (0-35) vs 2 (0-45), p < 0.0001) were significantly higher in the former group. The logistic regression model showed that four independent variables were associated with HZI: disease duration < 1 year (OR 2.893 (CI 1.821-4.597), p < 0.0001), lymphopenia <1500/mm(3) (OR 1.931 (CI 1.183-3.153), p = 0.009), prednisone (OR 6.723 (CI 2.072-21.815), p = 0.002), and cyclophosphamide use (OR 4.060 (CI 2.174-7.583), p < 0.0001).

Conclusion: HZI is an early viral infection in cSLE with a typical dermatomal distribution. Lymphopenia and immunosuppressive treatment seem to be major factors underlying this complication in spite of a benign course.

Keywords: Infection; childhood-onset systemic lupus erythematosus; herpes zoster infection; multicenter cohort.

Publication types

  • Multicenter Study

MeSH terms

  • Acyclovir / administration & dosage
  • Adolescent
  • Adult
  • Age of Onset
  • Antiviral Agents / administration & dosage
  • Brazil / epidemiology
  • Child
  • Child, Preschool
  • Cyclophosphamide / adverse effects*
  • Female
  • Herpes Zoster / drug therapy
  • Herpes Zoster / epidemiology*
  • Hospitalization / statistics & numerical data
  • Humans
  • Immunosuppressive Agents / adverse effects*
  • Infant
  • Logistic Models
  • Lupus Erythematosus, Systemic / complications*
  • Lupus Erythematosus, Systemic / drug therapy
  • Lymphopenia / epidemiology
  • Male
  • Nephritis / epidemiology
  • Prednisone / adverse effects*
  • Retrospective Studies
  • Severity of Illness Index
  • Young Adult

Substances

  • Antiviral Agents
  • Immunosuppressive Agents
  • Cyclophosphamide
  • Prednisone
  • Acyclovir