Respiratory viral testing during influenza season presents a challenge for the clinical microbiology laboratory. The Cleveland Clinic Clinical Virology Laboratory performs almost 8000 tests for influenza and respiratory syncytial virus (RSV) annually. There is a balance between clinicians' needs, hospital epidemiology, and the amount of work the laboratory is able to do during a busy influenza season. Clinicians want quick, accurate results from the laboratory to help guide treatment decisions.1 Infection control personnel want to mitigate transmission of respiratory tract infections with appropriate respiratory precautions (droplet for influenza and contact for RSV). Laboratory personnel need to offer tests that are reliable, inexpensive, and can meet the seasonal demand. At our institution, the laboratory staff, infection control practitioners, and clinicians worked together as a team to implement and assess new testing methods for respiratory pathogens during the 2004-2005 and 2005-2006 influenza seasons.
2004-2005 INFLUENZA SEASON
The Cleveland Clinic is a 1200-bed tertiary care flagship hospital of a 10-hospital health care system located in northeast Ohio. The Cleveland Clinic Clinical Virology Laboratory provides services for many surrounding hospitals and clinics. During the 2004-2005 season, we implemented rapid respiratory viral testing for identification of influenza A (Flu A), Flu B, and RSV with a Food and Drug Administration (FDA)-approved enzyme immunoassay (EIA) to provide rapid, accurate results that would satisfy the needs of the clinician, infection control practitioners, and not overwhelm the laboratory staff during an already busy time. Patient samples from the throat, nose, and nasopharynx were submitted to the virology laboratory with 3 possible requests: influenza, influenza and RSV, or RSV testing alone. The EIA tests were performed for RSV, Flu A, and Flu B. Negative EIA tests were confirmed by either viral culture (for Flu A and Flu B) or direct fluorescent antibody (DFA) for RSV (Fig. 1) per FDA recommendations.
Materials and Methods
Binax NOW Flu A and Flu B and BINAX NOW RSV EIA tests (BINAX Inc, Scarborough, Maine) are FDA-approved EIA tests for the rapid diagnosis of Flu A, Flu B, and RSV. The EIAs were used in our central laboratory to increase throughput of testing during influenza season. Because all negative EIAs need to be confirmed by backup testing, clinicians decided to send specimens directly to the Clinical Virology Laboratory to be processed by trained laboratory personnel. Results were placed in the Electronic Health Record (EHR) as soon as available. Negative EIAs for influenza were confirmed by viral culture using R-Mix (Diagnostic Hybrids Inc, Athens, Ohio) with shell vials. Negative EIAs for RSV were confirmed by DFA (Meridian Bioscience Inc, Cincinnati, Ohio) according to standard procedure.
The Clinical Virology Laboratory performed a combined total of 7600 influenza and RSV tests (5900 for influenza and 1700 for RSV) during the 2004-2005 influenza season. The EIA testing accounted for only 10.8% of the respiratory testing volume (696 EIAs for influenza and 128 EIAs for RSV) during the 2004-2005 influenza season. Clinicians went directly to DFA and culture instead of ordering EIA tests midway through the influenza season because EIA was not thought to be sensitive enough, missing patients with influenza and RSV.
Specimens submitted to the Clinical Virology Laboratory were as follows: nasopharyngeal swab (49.6%), nasal wash (0.8%), nasal swab (39.8%), throat swabs (10.1%), and miscellaneous (0.5%). The breakdown by age group of samples submitted for EIA testing is as follows: 0 to 2 years, 20.5%; 2 to 5 years, 2.6%; 5 to 20 years, 6.2%; 20 to 50 years, 26.6%; and older than 50 years, 44.2%.
During the 2004-2005 influenza season, 476 patients had a total of 824 EIA tests performed. Of the 348 patients tested for influenza, 26 had Flu A, 6 had Flu B, and 4 of 128 patients tested positive for RSV. Confirmatory testing with DFA detected another 19 patients with RSV. Viral culture detected another 32 patients with influenza missed by EIA (Figs. 2, 3).
The EIA assays only have proven sensitivities ranging from more than 70%,2 and as a result, require confirmatory testing if negative. We were unable to calculate sensitivity, specificity, or positive predictive value because we assumed our positive, rapid EIA tests to be true positives and could only calculate the negative predictive value (NPV). The NPV of BINAX NOW Flu A and Flu B EIA was 81% (253/306) and 82% (85/104) for the RSV EIA.
Clinicians' ordering patterns for the EIA tests during the 2004-2005 flu season corroborated our suspicions of clinicians' lack of enthusiasm for EIA testing (Fig. 4).
2005-2006 INFLUENZA SEASON
The goal for the 2005-2006 influenza season was to choose a more sensitive and specific test that balanced laboratory, clinician, and infection control needs. We used a molecular platform for the detection of Flu A, Flu B, and RSV by real-time reverse transcriptase (RT)-polymerase chain reaction (PCR) (ProFlu-1 Real Time Assay; Prodesse Inc, Waukesha, Wis). Numerous published studies show that real-time PCR is more sensitive than viral culture and DFA.3-8 The RT-PCR increases diagnostic yield, can be performed quickly, and gives clinicians information in a timely manner because it does not require confirmation.
Materials and Methods
ProFlu-1 Real Time Assay is a one-step RT-PCR assay based on 3 major processes: nucleic acid extraction, reverse transcription to generate complementary DNA from target RNA, and amplification and detection of target complementary DNA by using specific primers and probes.
Specimens from throat, nasal, and nasopharyngeal samples were submitted to the Clinical Virology Laboratory with clinician's ordering through our EHR. Sample extraction was performed by MagNA Pure LC (Roche Diagnostics Corporation, Indianapolis, Ind) with each sample having an internal control. Rotor-Gene (Corbett Inc, San Francisco, Calif) was used for amplification and detection using 4 channels (Cy5, JOE, ROX, FAM). Each sample provided 4 separate results on the internal control, Flu A, Flu B, and RSV. Same-day results were placed in the EHR 7 days a week.
The virology laboratory performed a combined total of 6282 influenza and RSV tests during the 2005-2006 influenza season (October 1, 2005-May 31, 2006). Of the 2094 samples tested, 12.5% (261/2094) were positive for Flu A, 3.9% (82/2094) for Flu B, and 13.3% (277/2094) for RSV (Fig. 5) by RT-PCR.
Several commercial tests are available for the detection of influenza and RSV including EIA, DFA, culture, and real-time PCR (Table 1). The Clinical Laboratory Improvement Amendments-waived, FDA-approved EIAs can be used as point-of-care tests giving results in 15 to 30 minutes. However, the sensitivity of EIA is approximately greater than 70%,2 which means that there can be a high rate of false negatives. Viral culture for influenza and RSV is time consuming and requires several days for identification. The DFA takes 2 to 4 hours to perform because it is a highly complex test and requires well-trained personnel for reading the slides. We chose real-time PCR which is more sensitive and specific than viral culture, DFA, and EIA.3-8 With the availability of automated specimen extraction, real-time amplification and detection, results are available on the same day because no confirmatory tests are required. Hands-on time is significantly decreased because extraction, amplification, and detection are all automated. Less sample handling leads to less cross contamination and risk of exposure for technicians in the clinical laboratory.
We offered this RT-PCR test 7 days a week during the influenza season with 2 possible runs per day (34 samples and 2 controls). After the influenza season ended, we switched to performing RT-PCR twice weekly because it is not cost effective to run so few specimens. Requests for influenza testing after the influenza season were replaced with viral culture or DFA.
Real-time PCR is clinically useful with implications for infection control and bed use. The ease of real-time PCR testing significantly increases throughput, decreases hands-on time and turnaround time, and is a better fit for our busy laboratory.
1. Call SA, Vollenweider MA, Hornung CA, et al. Does this patient have influenza? JAMA
3. Boivin G, Cote S, Dery P, et al. Multiplex real-time PCR assay for detection of influenza and human respiratory syncytial viruses. J Clin Microbiol
4. Oosterheert JJ, van Loon AM, Hoepelman AIM, et al. Impact of rapid detection of viral and atypical bacterial pathogens by real-time polymerase chain reaction for patients with lower respiratory tract infection. Clin Infect Dis
5. Templeton KE, Scheltinga SA, Beersma MFC, et al. Rapid and sensitive methods using multiplex real-time PCR for diagnosis of influenza A and influenza B viruses, respiratory syncytial virus, and parainfluenza viruses 1, 2, 3, and 4. J Clin Microbiol
6. van Elden LJR, Nijhuis M, Schipper P, et al. Simultaneous detection of influenza viruses A and B using real-time quantitative PCR. J Clin Microbiol
7. Syrmis MW, Whiley DM, Thomas M, et al. A sensitive, specific, and cost-effective multiplex reverse transcriptase-PCR assay for the detection of seven common respiratory viruses in respiratory samples. J Mol Diagn
© 2007 Lippincott Williams & Wilkins, Inc.
8. Kuypers J, Wright N, Ferrenberg J, et al. Comparison of real-time PCR assays with fluorescent-antibody assays for diagnosis of respiratory virus infections in children. J Clin Microbiol