Abstract
A recently introduced Sofia respiratory syncytial virus (RSV) fluorescent immunoassay (FIA) was evaluated against the BinaxNOW RSV card and the SD Bioline RSV test using 348 respiratory samples. The Sofia, BinaxNOW, and SD Bioline kits showed sensitivities of 66%, 65%, and 64%, respectively, for detecting RSV-A, and 71%, 63%, and 65% for detecting RSV-B, respectively.
TEXT
Although the reverse transcription-PCR (RT-PCR) method is widely used for respiratory syncytial virus (RSV) detection due to its high sensitivity and specificity, rapid antigen detection tests (RADTs) remain practical tests for the detection of RSV in clinical practice (1, 2). The Sofia RSV fluorescent immunoassay (FIA) (Quidel Corp., San Diego, CA, USA) is a recently introduced U.S. FDA-cleared rapid RSV antigen assay that employs immunofluorescence technology and is interpreted with a portable automatic reader (Sofia). The Sofia RSV FIA eliminates subjective interpretation, thereby greatly reducing the chance of human error. The aim of this study was to evaluate the performance of the Sofia RSV FIA in comparison with that of conventional RSV antigen tests, the BinaxNOW RSV card (Alere Scarborough, Inc., Scarborough, ME, USA) and SD Bioline RSV (Standard Diagnostics, Inc., Kyonggi, South Korea) test, using RSV RT-PCR as the reference standard (3).
Between January and November 2013, a total of 348 respiratory specimens (nasopharyngeal aspirates [NPA] or nasopharyngeal swabs [NS]) were collected from consecutive patients suspected of having a respiratory infection at the Korea University Guro Hospital, Seoul, South Korea. The mean age of the patients was 28.1 years (range, 0 to 98 years), and the male-to-female (M:F) ratio was 1:0.78. The patient samples were transported in a vial containing 3 ml of viral transport medium and were immediately used for viral cell culture and real-time RT-PCR. The leftover specimens were then cryopreserved at −80°C until RSV rapid antigen testing was performed in March 2014. All of the specimens underwent a single freeze-thaw cycle. Among the 348 patient samples, we identified 97 RSV-A-positive and 101 RSV-B-positive samples by RT-PCR and viral cell culture. One hundred fifty samples were negative for RSV. This study was approved by the Institutional Research Ethics Board of the Korea University Guro Hospital (approval no. KUGH13177-001).
For the Sofia RSV FIA, 260 μl of nasopharyngeal specimen in viral transport medium (VTM) was added to the same volume of the provided reagent in a test tube. A total volume of 120 μl of the obtained reaction mixture was then added to the test kit. After waiting approximately 15 min to allow the lateral flow of the reaction mixture, the test cassette was inserted into the portable fluorescence analyzer, called Sofia (Quidel Corp., San Diego, CA, USA), and the results were automatically printed within 1 min as positive, negative, or invalid. For the BinaxNOW RSV card, 50 μl of nasopharyngeal specimen in VTM was mixed with the same volume of reagent solution. A total volume of 100 μl of the reaction mixture was then added to the cassette. For the SD Bioline RSV test, 50 μl of nasopharyngeal specimen in VTM was mixed with the same volume of reagent solution. The test strip was inserted into a tube containing a total volume of 100 μl of the reaction mixture. For both of the kits, the test result was visually examined after approximately 15 min and interpreted as positive, negative, or invalid, according to the manufacturer's instructions.
For a home-brewed RSV RT-PCR, viral RNAs were extracted from 140 μl of respiratory specimen using the QIAamp viral RNA minikit (Qiagen, Hilden, Germany). A one-step real-time RT-PCR method was performed according to van Elden et al. (4), with minor modifications. Briefly, each tube contained a 25-μl reaction mixture that included 2.5 μl of isolated viral RNA, 0.1 μM forward and reverse primer, and 0.1 μM probe. TaqMan amplification and detection were performed with a real-time thermocycler, the CFX96 (Bio-Rad, Hercules, CA, USA). The thermocycling conditions were as follows: reverse transcription at 50°C for 20 min and then initial denaturation at 95°C for 10 min, followed by 45 cycles at 95°C for 15 s and at 60°C for 60 s. Two laboratory technicians who performed the rapid tests and RT-PCR were blinded to the results of the other tests, as well as to the clinical presentations of the patients.
The performance parameters, such as sensitivity, specificity, positive predictive value (PPV), negative predictive value, positive likelihood ratio, and negative likelihood ratio, of the three RADTs were calculated using the RT-PCR results as standards (3) and expressed with a 95% confidence interval. Statistical analysis was performed with the Excel software (Microsoft Corporation, Redmond, WA, USA) using a McNemar test or chi-square test or independent t test, with a P value of <0.05 considered statistically significant.
The performance parameters for the three RADT kits are summarized in Table 1. The differences in the sensitivities between the Sofia RSV FIA and BinaxNOW RSV card kit were 1.0% (95% confidence interval [CI], −5.0 to 7.1%; P = 1.000) for RSV-A and 8.0% (95% CI, 1.4 to 14.5%; P = 0.043) for RSV-B. For the comparison between the Sofia RSV FIA and SD Bioline RSV test, the differences in the sensitivities were 2.0% (95% CI, −3.6 to 7.8%; P = 0.724) for RSV-A and 6.0% (95% CI, −1.2 to 13.1%; P = 0.181) for RSV-B. The Sofia RSV FIA showed a slightly lower specificity than that of the other kits, but the difference (2.0%) was not statistically significant (95% CI, 0.1 to 5.3; P = 0.13). The PPV did not vary substantially for any of the three RADTs (92 to 100%).
TABLE 1.
Performance characteristics of three rapid tests for detecting RSV-A/-B compared to those of RT-PCRa
| RSV type | Rapid test usedb | Sensitivity (% [no./total no., 95% CI])c | Specificity (% [no./total no., 95% CI]) | Positive predictive value (% [no./total no., 95% CI]) | Negative predictive value (% [no./total no., 95% CI]) | Positive likelihood ratio (95% CI)d | Negative likelihood ratio (95% CI) |
|---|---|---|---|---|---|---|---|
| A | SO | 66 (64/97, 56.6–75.4) | 98.0 (246/251, 95.4–99.1) | 92.8 (64/69, 84.1–96.9) | 81.5 (145/178, 75.1–86.5) | 19.8 (8.3–47.4) | 0.35 (0.27–0.47) |
| BN | 65 (63/97, 55.4–74.4) | 100.0 (251/251, 98.5–100.0) | 100.0 (63/63, 94.3–100.0) | 81.5 (150/184, 75.3–86.5) | NA | 0.35 (0.27–0.46) | |
| SD | 64 (62/97, 54.3–73.5) | 100.0 (251/251, 98.5–100.0) | 100.0 (62/62, 94.2–100.0) | 81.1 (150/185, 74.8–86.1) | NA | 0.36 (0.28–0.47) | |
| B | SO | 71 (72/101, 62.5–80.1) | 98.0 (242/247, 95.3–99.1) | 93.5 (72/77, 85.7–97.2) | 83.3 (145/174, 77.1–88.1) | 21.4 (9.0–51.1) | 0.30 (0.22–0.40) |
| BN | 63 (64/101, 54.0–72.8) | 100.0 (247/247, 98.5–100.0) | 100.0 (64/64, 94.3–100.0) | 80.2 (150/187, 73.9–85.3) | NA | 0.37 (0.28–0.47) | |
| SD | 65 (66/101, 56.1–74.6) | 100.0 (247/247, 98.5–100.0) | 100.0 (66/66, 94.5–100.0) | 81.1 (150/185, 74.8–86.1) | NA | 0.35 (0.27–0.45) |
RSV, respiratory syncytial virus; RT-PCR, real-time reverse transcription-PCR.
SO, Sofia RSV FIA kit; BN, BinaxNOW RSV card kit; SD, SD Bioline RSV kit.
CI, confidence interval.
The positive likelihood ratio was not applicable (NA) when the denominator was zero.
The mean (standard deviation [SD]) RT-PCR threshold cycle (CT) values for RSV-A- or RSV-B-positive specimens for each of the three RADTs are presented in Table 2. Regarding samples that were positive with the Sofia RSV FIA but negative with the other kits, the average (SD) RT-PCR CT was 25.0 (±3.8) for RSV-A (n = 5) and 26.3 (±3.3) for RSV-B (n = 9).
TABLE 2.
Threshold cycle levels of real-time PCR for RSV-A/B in rapid RSV test-positive and -negative cases
| Influenza type | Rapid test useda | Positive mean (SD) CT | No. of positive cases | Negative mean (SD) CT | No. of negative cases | P valueb |
|---|---|---|---|---|---|---|
| A | SO | 20.0 (3.1) | 64 | 29.1 (5.1) | 33 | <0.05 |
| BN | 19.7 (2.7) | 63 | 29.3 (4.8) | 34 | <0.05 | |
| SD | 19.6 (2.7) | 62 | 29.1 (4.8) | 35 | <0.05 | |
| SO only positive | 25.0 (3.8) | 5 | ||||
| B | SO | 22.3 (3.2) | 72 | 29.9 (4.2) | 29 | <0.05 |
| BN | 21.7 (2.7) | 64 | 29.2 (4.2) | 37 | <0.05 | |
| SD | 22.1 (3.1) | 66 | 29.0 (4.5) | 35 | <0.05 | |
| SO only positive | 26.3 (3.3) | 9 |
SO, Sofia RSV FIA kit; BN, BinaxNOW RSV card kit; SD, SD Bioline RSV kit.
Independent t test between positive and negative mean CT.
The sensitivity of the Sofia RSV FIA kit was slightly higher than that of the two other RADTs, but a statistically significant difference in sensitivity was noted only for RSV-B detection in comparison with that of the BinaxNOW RSV card kit. Therefore, the Sofia RSV FIA kit demonstrated similar sensitivities for detecting RSV-A and -B compared with those of the BinaxNOW RSV card or SD Bioline RSV test.
The sensitivity values of our study were lower than those reported in previous studies. One reason for this lower sensitivity might be the broad age distribution of the patients who were enrolled in our study. The mean ages of our study population for RSV-A and -B were 17.8 and 41.1 years, respectively, and 25% (24/97) of the RSV-A-positive group was >5 years of age, while 32% (32/101) of the RSV-B-positive group was >5 years of age. Compared to the subjects who were ≤5 years of age, RADT sensitivity in the group of those who were >5 years old was significantly lower in all three RADTs for detecting RSV-A and -B. The sensitivities between the two groups differed by 20.0% (BinaxNOW RSV card) to 22.6% (SD Bioline RSV test) for RSV-A (P < 0.05) and 40.3% (Sofia RSV FIA) to 50.0% (SD Bioline RSV test) for RSV-B (P = 0.00).
In this study, RT-PCR analysis was performed with fresh specimens, but the RADT analysis was performed with stored frozen specimens. However, a previous study showed similar sensitivities with fresh and stored frozen specimens for RSV diagnosis in an elderly population (5). Accordingly, the use of stored specimens was assumed not to have affected the sensitivity values in our study.
In conclusion, the Sofia RSV FIA kits showed comparable performance measures for RSV-A and -B compared to those of the BinaxNOW RSV card and SD Bioline RSV test. The Sofia RSV FIA can be used as a reliable point-of-care (POC) device for the rapid detection of RSV in clinical practice.
ACKNOWLEDGMENTS
This study was supported by a Korea University R&D grant in 2013.
We declare no conflicts of interest.
REFERENCES
- 1.Rabon-Stith KM, McGuiness CB, Saunders B, Edelman L, Kumar VR, Boron ML. 2013. Laboratory testing trends for respiratory syncytial virus, 2007–2011. J Clin Virol 58:575–578. doi: 10.1016/j.jcv.2013.09.012. [DOI] [PubMed] [Google Scholar]
- 2.Popow-Kraupp T, Aberle JH. 2011. Diagnosis of respiratory syncytial virus infection. Open Microbiol J 5:128–134. doi: 10.2174/1874285801105010128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Papenburg J, Buckeridge DL, De Serres G, Boivin G. 2013. Host and viral factors affecting clinical performance of a rapid diagnostic test for respiratory syncytial virus in hospitalized children. J Pediatr 163:911–913. doi: 10.1016/j.jpeds.2013.03.067. [DOI] [PubMed] [Google Scholar]
- 4.van Elden LJ, van Loon AM, van der Beek A, Hendriksen KA, Hoepelman AI, van Kraaij MG, Schipper P, Nijhuis M. 2003. Applicability of a real-time quantitative PCR assay for diagnosis of respiratory syncytial virus infection in immunocompromised adults. J Clin Microbiol 41:4378–4381. doi: 10.1128/JCM.41.9.4378-4381.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Casiano-Colón AE, Hulbert BB, Mayer TK, Walsh EE, Falsey AR. 2003. Lack of sensitivity of rapid antigen tests for the diagnosis of respiratory syncytial virus infection in adults. J Clin Virol 28:169–174. doi: 10.1016/S1386-6532(03)00002-7. [DOI] [PubMed] [Google Scholar]