Critical evaluation of a CLIA-waived streptococcal antigen detection test in the emergency department

We compare the performance of a Clinical Laboratory Improvement Amendments (CLIA)-waived antigen detection test (ADT) analyzed in the emergency department and a CLIA moderately complex ADT performed in the hospital microbiology laboratory.

Samples from throat swabs were obtained using a double-headed Culturette II (Becton Dickinson Medical Systems, Sparks, MD) from a consecutive sample of 322 patients 3 years or older who presented to the ED of a university-affiliated pediatric referral hospital with the complaint of sore throat during 1998. One swab was transported to the microbiology laboratory and analyzed using a CLIA moderately complex ADT; negative results were confirmed using sheep blood agar culture. The second swab remained in the ED where a nurse conducted a CLIA-waived ADT. The accepted standard for documentation of group A beta-hemolytic streptococcal (GABHS) infection was either a positive moderately complex ADT or culture result. The time of specimen collection, as well as the time the ED results and microbiology laboratory results were available to treating physicians, were recorded. Main outcome measures were concordance (kappa statistic), sensitivity, and turnaround time (Mann-Whitney U test).

Three hundred twenty-two patients (mean age 7.5 years) had both ADTs performed. One hundred one (31%) patients had documented GABHS in the microbiology laboratory; 83 (82%) had a positive ADT result in the microbiology laboratory, and 18 (18%) had a positive culture result after a negative moderately complex ADT result. In 299 patients or 93% (95% confidence interval [CI] 90.8%, 95.8%) of patients, the waived ADT and the moderately complex ADT results were concordant (kappa 0.82; 95% CI 0.78, 0.86; P <.001). The sensitivity of the waived ADT was 80%; the sensitivity of the moderately complex ADT approximated 82% (difference of 2%; 95%CI -3%, 7%). The median times from swab specimen collection to availability of ADT results were 10 minutes (range 3 to 37 minutes) for the waived ADT and 35 minutes (range 5 to 162 minutes) for the moderately complex ADT (P <.001) with a difference of 25 minutes (95% CI 22.4, 27.6 minutes).

In this study, an ED CLIA-waived rapid streptococcal throat test performed as well as its equivalent CLIA-regulated laboratory test. Further, the ED test provided results more rapidly than the laboratory test. Our results also validate previous work that negative rapid throat test results in pediatric patients in the ED should be confirmed by standard throat culture.