Culture Confirmation of Negative Rapid Strep Test Results

CLINICAL QUESTION: Does the use of a high-sensitivity rapid strep test without throat culture result in more complications than a culture-only strategy?

BACKGROUND: Because the sensitivity of currently used rapid strep tests is relatively low, many physicians order a throat culture to confirm the negative results of a test. Recently a high-sensitivity rapid antigen test has become available. However, there are concerns that the use of the rapid test without culture confirmation of negative results can lead to increased clinical complications as a result of missed or delayed diagnosis. The authors report the clinical experience of a health system that changed its diagnostic strategy from culture-only to exclusive use of the rapid strep test.

POPULATION STUDIED: The authors identified 30,036 patient encounters with a diagnosis of pharyngitis during a 4-year period. All patients were seen at one of 2 satellite clinics of the Lahey Clinic, a multispecialty integrated delivery system in Massachusetts.

STUDY DESIGN AND VALIDITY: This is a quasiexperimental study that documented the effect of a systemwide change in diagnostic testing strategy for streptococcal pharyngitis. During the first 2 years of the study, patients with pharyngitis were tested using bacterial culture alone. During the subsequent 2 years, patients were tested using the new high-sensitivity antigen test (STREP A OIA, Biostar, Inc). The authors compared the rates of complications from pharyngitis in these 2 groups of patients. Suppurative complications included peritonsillar or retropharyngeal abscess. Nonsuppurative complications included acute rheumatic fever or poststreptococcal glomerulonephritis. All patients were identified using International Classification of Diseases-ninth revision (ICD-9) diagnostic codes. As the authors correctly point out, retrospective case series are not randomized and do not control for a possible, though unlikely, change in streptococcal virulence during the study period. In addition, ICD-9 codes can be inaccurate, although the authors did attempt to confirm diagnoses through chart review. Of greater concern, however, is the sample size in this study. On the basis of the complication rates that are subsequently reported (approximately 0.25%), each arm of the study would have required more than 500,000 patients to find a 10% difference in complications. With approximately 15,000 patients in each group, the study had the power to detect an increase in the rate of complications from 0.25% to 0.42%. Using only the suppurative complications in patients who received culture or rapid antigen test before the visit, the study would have the power to detect an increase of 0.08% to 0.19%. Finally, the primary author’s financial relationship with the manufacturer of the rapid strep test is clearly and appropriately disclosed.

OUTCOMES MEASURED: The authors reported the number of suppurative and nonsuppurative complications during the study period. Through chart review, the authors identified and excluded patients who initially presented with the complication, because those patients had no antecedent diagnostic testing.

RESULTS: During the initial culture-only period, 15,399 patients were identified with pharyngitis, and 65% received throat culture. Thirty-seven patients (0.24%) suffered a subsequent suppurative complication, and there were no nonsuppurative complications. During the second period 14,637 patients presented with pharyngitis, and 51% received the rapid antigen test. Thirty-six patients (0.25%) had a suppurative complication, and one had a nonsuppurative complication. Of the 71 patients with suppurative complications whose charts were available, 40 had been seen before the complication, and 23 had received a throat culture (12) or rapid antigen test (11). However, only 3 throat cultures and 3 rapid antigen tests were positive. There were no statistically significant differences in the rates of complications between the 2 study periods.