In the office laboratory of our private practice, the sensitivity of a single rapid antigen detection (RAD) test for group A streptococci (compared with backup throat culture) was examined over 3 winter periods. When cultures were held only 24 hours, the sensitivity was 92%; in the second period, when cultures were held to 48 hours, the sensitivity declined significantly to 86%; when the cultures were read without knowledge of the RAD test result, the sensitivity was 85%. As the Red Book Committee has suggested, physician office laboratories should validate the sensitivity of their RAD test against culture onto blood agar before abandoning the backup throat culture.
Approximately 8% percent of office visits in our private group practice are for the presenting complaint of sore throat. Group A β-hemolytic streptococcal (GABHS) pharyngitis is diagnosed in 20% to 30% of these visits. The standard method for establishing the etiologic diagnosis of GABHS pharyngitis is culture of a pharyngeal specimen obtained with a throat swab and inoculated on sheep blood agar. However, the throat culture requires 24 to 48 hours to interpret, thus delaying the diagnosis. Rapid antigen detection (RAD) tests that detect the group A carbohydrate of the bacterial cell wall are used to expedite the diagnosis. In our physician office laboratory, a RAD test is performed on each throat swab. A positive RAD test result is accepted as adequate for the diagnosis of GABHS pharyngitis because of the high specificity of the test.1,2 Negative RAD tests, which account for ∼70% to 80% of tests performed, are not accepted as adequate to exclude the diagnosis of GABHS pharyngitis, because the sensitivity of the RAD test is thought to be too low. Each negative RAD test is confirmed with a backup throat culture by using a standard method.
The necessity for a backup throat culture for negative RAD tests is a controversial issue. The US Food and Drug Administration has not approved any RAD test for use without the backup throat culture (Division of Microbiologic Devices, In Vitro Diagnostic Device Evaluation and Safety, Center for Devices and Radiologic Health, US Food and Drug Administration, written communication, 2003). The 1997 edition of the Red Book3 recommended a backup throat culture for all negative RAD test results. In 2000 and 2003, revised editions of the Red Book1, 4 softened this stance on confirmatory throat cultures and recommended that physicians validate that their RAD tests are as sensitive as culture before abandoning the backup throat culture. Additionally, a practice guideline from the Infectious Disease Society of America (IDSA)2 recommended that a negative RAD test result in a child or adolescent be confirmed with a throat culture unless the physicians have ascertained that the RAD test in their office is comparable to culture. However, neither the Red Book nor the IDSA guidelines have specified how best to determine the sensitivity of a RAD test in an office setting.
This report examines the process that we devised to validate the sensitivity of a RAD test that has been used in our physician office laboratory since 1994. Before 1999, confirmatory throat cultures for negative RAD tests were read at 24 hours and then discarded. During 2 subsequent winter periods, we made minor alterations to the backup culture method to assess what effect these changes would have on the sensitivity of the RAD test. After each winter period was completed, we retrospectively collected data from our laboratory records to determine the prevalence of presumptive GABHS and the sensitivity of the RAD test.
The on-site physician office laboratory at our pediatric group practice in Charlottesville, Virginia, has Clinical Laboratory Improvement Amendment certification for moderate complexity. The laboratory is staffed by 3 medical technologists, each with 5 to 10 years of experience in our office. To establish the diagnosis of presumptive streptococcal pharyngitis, physicians collect throat samples from symptomatic patients by using 2 Dacron swabs (Mainline Confirms Throat Swab, Mainline Technology Inc, Ann Arbor, MI, or Pur-Wraps, Hardwood Products, Guilford, ME) held together for swabbing the tonsils and posterior pharynx. The swabs are returned to their paper wrapper and hand delivered to our centrally located laboratory.
Processing of Throat Swabs
Within 2 to 3 minutes of arrival of the swabs in the laboratory, a technologist (or trained nurse) performs the RAD test on 1 swab by using the Mainline Confirms Strep A test (Mainline Technology Inc). This is a 2-site sandwich immunoassay in which any color change on the test band in the presence of a positive control band is considered a positive test for presumptive GABHS pharyngitis. The RAD test result is reported to the physician to facilitate disposition of the waiting patient.
If the RAD test is positive, the second swab is discarded. If it is negative, the second swab is used to perform a qualitative culture on one half of a blood agar plate (Selective Streptococcal Agar, Becton Dickinson, Cockeysville, MD), and a bacitracin disk (Taxo A Disk, Becton Dickinson) is placed on the streaked agar. Culture plates are incubated at 37°C in air. A technologist (or on-call physician on Sundays) interprets the cultures each day. Growth of β-hemolytic colonies with a zone of inhibition around the bacitracin disk is identified presumptively as GABHS.
Recording of Results
The laboratory maintains a record of each test as required by Clinical Laboratory Improvement Amendment certification. The record includes the patient’s name, date of the test, physician’s name, result of the RAD test, and result of the backup throat culture for negative RAD tests. These records are kept for a minimum of 2 years.
Laboratory records for 3 distinct periods were analyzed. Each period spanned the 5-month winter period, October through February inclusive. In the winter period beginning in 1998, the backup throat cultures on negative RAD tests were read at 24 hours and then discarded. In the 1999 winter period, the protocol for the backup throat culture was the same except that culture plates negative for GABHS at 24 hours were read again at 48 hours. The practice of a second reading at 48 hours before discarding the plate has since been continued in our laboratory.
Before the third winter period, we theorized that the technologist might be underreading the backup culture plates for GABHS because of prior knowledge that the RAD test had been negative. To examine this possibility, 2 changes in the processing of the specimens were made for the 5-month period beginning in the winter of 2001. First, all throat-swab samples were cultured even if the RAD test was positive. Second, a system was devised that allowed the technologist to read the culture plates while being blind to the result of the RAD test. The system worked as follows: when the double-swab sample was received in the laboratory, the patient’s name was entered on a numbered list, which served as the written record in the laboratory. The sample number then was written on a 3/8- by 1/2-inch white label (Avery, Brea, CA) affixed to the culture plate on which that specimen was plated. After plating the culture, a 3/4-inch-round navy-colored coding label (Avery) was placed over the white label on the culture plate to obscure its number. The medical technologist would remove the round label to reveal the number after 24 hours only if the culture plate was positive for GABHS or after 48 hours to allow the recording of the culture result.
Analysis of Data
Results of RAD tests and backup throat cultures were extracted from the laboratory’s record. As is accepted in clinical practice, a positive RAD test or throat culture was considered presumptive evidence of GABHS pharyngitis. The specificity of the RAD test and the throat culture was presumed to be 100%. Prevalence and sensitivity were calculated for each period. Statistical significance and confidence intervals were calculated by using Fisher’s exact test (InStat 3, GraphPad Software, Inc, San Diego, CA).
The number of throat-swab samples for each of the 3 winter periods varied from 1968 to 2148, providing data on 6114 patient samples (Table 1). The prevalence of GABHS pharyngitis ranged from 24.5% in 1998 to 31.6% in 2001. During the first winter period (1998), in which backup throat cultures were held only for 24 hours, the sensitivity of the RAD test was 91.9%. In the second period (1999), during which cultures were held to 48 hours, the sensitivity of the RAD test decreased significantly to 85.7% (P = .002). During the third winter period, in which all samples were cultured so that the RAD test results could be masked from the medical technologists reading the cultures, the sensitivity of the RAD test was 85.4%, similar to the sensitivity when backup cultures were done only on samples with a negative RAD test result (P = .9). During both the second and third periods in which cultures were not discarded until after 48 hours of incubation, >40% of the positive backup cultures became evident only at the second reading.
The detection of GABHS on the throat swab of a patient with a sore throat may occasionally represent a carrier state. However, in a symptomatic patient, a positive RAD test or backup culture is accepted as adequate to establish the diagnosis of streptococcal pharyngitis. This report, which resulted from our review of the physician office laboratory records, assessed the possibility of abandoning the backup throat culture that we have used to confirm negative RAD test results. To our knowledge, this examination of the sensitivity of a RAD test from a single manufacturer represents the largest series reported on data collected from patients in a private practice setting. In the first winter period (1998), the 92% sensitivity of the RAD test compared with a backup throat culture read at 24 hours suggested that the RAD test alone might be sufficient for excluding the diagnosis of GABHS pharyngitis. Between the first and second winter periods, we changed our backup culture procedure so that the cultures were read a second time at 48 hours in response to reports that holding the throat cultures for a second day increased the detection rate of GABHS.5,6 As a consequence of this change, the number of positive backup cultures almost doubled in the second period, resulting in a significant reduction in the sensitivity of the RAD test to ∼85%, which was too low to consider abandoning the backup throat culture. During the third winter period (2001), in which the medical technologists interpreting cultures were unaware of the RAD test result, the sensitivity was again 85%.
The measured sensitivity of a RAD test depends to some degree on the culture method to which it is compared. Currently, no single RAD test has demonstrated consistently that its sensitivity in an office setting is as good as a throat culture performed on a blood agar plate. In 1996, Schlager et al5 demonstrated that the optical immunoassay (OIA) was only 80% sensitive against sheep blood agar; a review of previous studies showed an interesting association between financial support from the manufacturer and a favorable result for the sensitivity of the OIA. In 1 subsequent study7 performed in multiple office settings in 2 different cities, the sensitivities of the OIA compared with blood agar plate were reported as 93% and 96%, but other studies8–10 reported sensitivities of 77% to 82%. Thus, it is not clear that the OIA is sensitive enough to warrant abandoning the backup throat culture. In another recent report11 from an office setting, RAD tests were found to be 93.8% sensitive compared with backup culture. However, little information about culture methods was provided, and RAD tests from 3 different manufacturers were used. Others12,13 have reported findings using an immunochromatographic RAD test, but this RAD test either was not sensitive enough12 or the study suffered from poor design.13
An important consideration when determining the sensitivity of a RAD test is how the reference standard is defined. We elected to use a throat culture on a blood agar plate with a bacitracin disk as the reference standard, because this has been in use in our office for many years. Although a blood agar plate with a bacitracin disk is not 100% sensitive, it is readily available in a physician office laboratory, is relatively inexpensive, and can be performed and interpreted easily. Others have used combinations of inoculation in Todd-Hewitt broth,5,7,8,12 pledget cultures,7,8,12 anaerobic cultures,5, 6 and/or polymerase chain reaction10 as the reference standard, but these alternatives are not available in a physician office laboratory, and their advantages for clinical decision-making over a properly performed culture on blood agar are debatable.2,5
We have answered the call of the Red Book committee and the IDSA to validate the sensitivity of our RAD test as compared with culture in our general pediatric practice. We conclude that a negative RAD test result in our pediatric practice still requires a confirmatory throat culture. We suggest that other practices that wish to examine the sensitivity of their RAD test may compare it to a standard backup throat culture on sheep blood agar that is not discarded until a final reading 48 hours after inoculation.
We thank the physicians, nurses, and staff at Pediatric Associates of Charlottesville, especially our medical technologists, Beth Broderick, Sue Likowski, and Debbie (Coffey) Reynolds; Keith R. Powell, MD, for reviewing our manuscript; and the Pendelton Pediatric Infectious Disease Laboratory (Charlottesville, VA) for its support.
- Received February 24, 2003.
- Accepted August 18, 2003.
- Reprint requests to (C.E.A.) Pediatric Associates of Charlottesville, 1011 E Jefferson St, Charlottesville, VA 22902. E-mail:
- ↵American Academy of Pediatrics. Group A streptococcal infections. In: Pickering LK, ed. 2000 Red Book Report of the Committee on Infectious Diseases. 25th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2000:528–529
- ↵Bisno AL, Gerber MA, Gwaltney JM, Kaplan EL, Schwartz RH. Practice guidelines for the diagnosis and management of group A streptococcal pharyngitis. Clin Infect Dis.2002;35 :113– 125
- ↵American Academy of Pediatrics. Group A streptococcal infections. In: Peter G, ed. 1997 Red Book Report of the Committee on Infectious Diseases. 24th ed. Elk Grove Village, IL: American Academy of Pediatrics; 1997:486
- ↵American Academy of Pediatrics. Group A streptococcal infections. In: Pickering LK, ed. 2003 Red Book Report of the Committee on Infectious Diseases. 26th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2003:576
- ↵Hart AP, Buck LL, Morgan S, Saverio S, McLaughlin JC. A comparison of the Biostar strep A OIA rapid antigen assay, group A Selective Strep Agar, and Todd-Hewitt broth cultures for the detection of group A streptococcus in an outpatient family practice setting. Diagn Microbiol Infect Dis.1997;29 :139– 145
- ↵Kaltwasser G, Diego J, Welby-Sellenriek PL, Ferret R, Caparon M, Storch GA. Polymerase chain reaction for Streptococcus pyogenes used to evaluate an optical immunoassay for the detection of group A streptococci in children with pharyngitis. Pediatr Infect Dis J.1997;16 :748– 753
- ↵Mayes T, Pichichero ME. Are follow-up throat cultures necessary when rapid antigen detection tests are negative for group A streptococci? Clin Pediatr (Phila).2001;40 :191– 195
- ↵Chapin KC, Flintoff MA. Comparison of Genzyme OSOM Ultra Strep A test and Thermo Biostar OIA for detection of group A streptococcus [abstract]. Presented at: Interscience Conference on Antimicrobials Agents and Chemotherapy; September 27–30, 2002; San Diego, CA
- Copyright © 2004 by the American Academy of Pediatrics