Study of Serious Infections in Children (GAS-NF)

Objectives

To evaluate whether use of nonsteroidal anti-inflammatory drugs (NSAID) is a risk factor for the development of severe group A streptococcal (GAS) infections, particularly necrotizing fasciitis (NF), in children with varicella infection.

Methods

Because NF is a rare complication of varicella, a multicenter case-control study was conducted. Children were enrolled prospectively. Detailed information on illness, medication use, and other potential risk factors for infection in the two weeks before the clinical onset of NF or other invasive GAS infection was collected from cases and controls, the latter were selected to have indications and opportunities for NSAID use similar to cases. Case-patients include children with NF or other invasive GAS infection and a history of varicella in the preceding two weeks. Children living in the same communities as the cases with a history of a recent primary varicella infection but with no evidence of NF or other invasive GAS disease served as controls. Up to four times as many controls as cases were enrolled. The primary care physician for each case was asked to identify a colleague with a comparable practice in the community. Children, similar in age to the cases, with uncomplicated varicella were enrolled from these practices.

After notification of a potential study participant and verification of eligibility, a nurse interviewer contacted the child’s parent or guardian and conducted a telephone interview. The interview obtained detailed information on the timing and severity of symptoms of both the primary varicella infection and subsequent complications (case or control illness), and on the timing of prescription and non-prescription medication use, including the number of doses of medication given. To facilitate accurate recall, we provided the parents with a booklet displaying color images of the most commonly used pediatric non-prescription medications. A copy of the hospital medical record, primary care physician records, emergency room visits and any other relevant medical care contact was obtained and relevant data abstracted. Interviews with practitioners involved in the child’s management were conducted to supplement information obtained from medical records.

GAS isolates from cases were saved and forwarded to the Centers for Disease Control for laboratory evaluation. All isolates were characterized by M-type and other phenotypic characteristics or potential virulence factors such as proteases and exotoxins. The characteristics of the isolates from NF cases were compared with those of non-NF invasive GAS cases.