Extrapharyngeal group A Streptococcus infection: diagnostic accuracy and utility of rapid antigen testing.

Background: Antigen tests have been well-studied and are widely used in pediatric practice for rapid detection of group A Streptococcus (GAS) infections in the throat, but they have not been examined sufficiently for the detection of infection of skin sites, such as the perineal region or impetiginous lesions.
Methods: During the 3-year period 1999 to 2002, we evaluated 239 patients with suspected GAS skin infection, in 5 pediatric practices, using 3 Dacron swabs for each site. The first swab was tested in the pediatric office laboratory with an antigen detection kit. For the first 91 patients, the Abbott Test Pack Plus antigen detection test (ADT) was used. The Abbott Signify Strep A ADT was used to test subsequent patients. The second swab was tested with BD Directigen 1-2-3 ADT in the hospital laboratory. The third swab was placed in modified Stuart's transport medium for comparison of recovery of GAS from culture in broth or on agar. A positive culture served as the reference standard. Test performance and test accuracy were determined for each ADT.
Results: Of the 247 ADTs and cultures performed on 239 patients, 91 with suspected skin infection were tested with the Test Pack Plus test, 149 with the Signify Strep A test and 247 with the Directigen test. Eighty-six (35%) cultures were positive, 73 from perineal sites (54 rectal, 13 vaginal, 6 penile) and 13 from impetiginous lesions. There was 100% concordance for the 86 cultures positive for GAS in a comparison between dry Dacron swabs and swabs that had been placed in modified Stuart's transport medium. Test Pack Plus and Signify Strep A ADTs had similar performance characteristics for skin infections: sensitivity, 92 and 88%; specificity, 99 and 97%; positive predictive value, 96 and 94%; and negative predictive value, 97 and 93%. Directigen ADT had sensitivity 78%, specificity 100%, positive predictive value 100% and negative predictive value 89%. Accuracy for the tests varied from 92 to 97%.
Conclusion: Tests designed to detect GAS carbohydrate antigen in patients with pharyngitis can be used rapidly and accurately to detect GAS antigen in patients with cutaneous lesions suspected of GAS infection.