Abstract
The NanoChip400 system uses multiplex PCR chemistry and electronic microarray detection of influenza A and B viruses; respiratory syncytial viruses A and B; and human parainfluenza virus types 1, 2, and 3. The results obtained with the NanoChip 400 system were compared with those obtained by direct fluorescent-antibody staining (DFA) and real-time PCR with 122 and 130 specimens, respectively. Concordance between DFA and NanoChip 400 system was obtained for 106 of 122 (86.9%) specimens. On the basis of discrepancy analysis with specimens available for confirmatory real-time PCR testing, the sensitivity and specificity of the NanoChip 400 were 98.6% and 100%, respectively. With respect to specimens previously tested by real-time PCR, the NanoChip 400 system demonstrated a sensitivity of 91.1% and a specificity of 100%. The NanoChip 400 system provides clinical laboratories with a practical, rapid, and sensitive method for the detection of common respiratory viruses.
There has been a considerable effort over the past two decades or more to develop methods for the rapid detection of viral agents. Although enzyme immunoassay and lateral flow-based antigen detection are the most frequently used methods for the rapid detection of respiratory viruses, the most reliable antigen detection method for respiratory viruses is considered to be direct fluorescent-antibody staining (DFA), based on extensive studies conducted over the past 20 years (4, 9). While DFA has generally shown good performance in comparison to viral culture, the actual sensitivity is probably lower than that in published reports of studies that used viral culture as the “gold standard,” given the lability of respiratory viruses.
While viral culture is considered the gold standard, specimen handling is a major factor in the recovery of respiratory viruses; furthermore, even with shell vial cultures, the clinical utility of viral culture has limitations, as the results do not have an immediate effect on clinical as well as infection control decisions (11).
Screening for multiple viral agents by DFA is extremely useful for patient management as well as infection control. While the cost-effectiveness and cost-benefit of a rapid antigen detection infection control program for the reduction of nosocomial respiratory virus transmission have been demonstrated (1, 6, 13), the lack of trained personnel on off-shifts often limits the use of DFA to a single shift in many health care settings. While real-time PCR methods are available as analyte-specific reagents, real-time PCR is limited to the simultaneous detection of three respiratory viruses.
DFA requires highly trained clinical laboratory scientists and is subject to nonspecific staining with specimens containing yeasts, certain bacteria, leukocytes, or mucus as well as because of binding to FC receptors (7, 8, 12). The specimen type and the collection technique are also important factors. Furthermore, it has been reported that approximately 5% of respiratory specimens submitted for immunofluorescence examination are inadequate for interpretation as a result of an inadequate number of cells and the quality of the cells. There is a need for a rapid and sensitive automated method that can replace DFA as a method for screening for respiratory viruses.
The electronic microarray-based NanoChip 400 system (Nanogen, Inc., San Diego, CA) analyzes multiplex PCR amplification reactions on a NanoChip cartridge. Amplicons are bound to specific sites on the microarray by hybridization to complementary sequences on oligonucleotides. The 5′ end biotinylated capture oligonucleotides are attached via streptavidin into the permeation layer that covers the microarray. The detection of amplicons is accomplished through the use of reporter probes.
We report on our efforts to replace the current DFA respiratory screening battery for influenza A and B viruses, respiratory syncytial virus (RSV), and parainfluenza virus types 1 to 3 with a NanoChip assay format for the rapid detection of multiple respiratory viruses in clinical specimens.
MATERIALS AND METHODS
Clinical specimens.
For comparison of DFA and the NanoChip 400 system for the detection of influenza A and B viruses, RSV, and parainfluenza viruses, a total of 122 nasopharyngeal and throat specimens, which were stored at −70°C in M4 transport medium (Remel, Lenexa, KS) and which consisted of 44 specimens negative by DFA (Light Diagnostics SimulFluor respiratory screen; Chemicon Interational, Temecula, CA) and 78 specimens positive by DFA, were analyzed with the NanoChip 400 system. Of the DFA-positive specimens, 23 were influenza A virus positive, 24 were influenza B virus positive, 21 were RSV positive, and 10 were parainfluenza virus positive.
For comparison of the NanoChip 400 system and real-time PCR, a total of 130 nasopharyngeal and throat specimens that were previously tested by real-time PCR (ABI 7500; Applied Biosystems, Foster City, CA) with analyte reagents specific for influenza A virus, influenza B virus, and RSV (Prodesse, Waukesha, WI) and that had been stored at −70°C in Universal viral transport medium (Becton Dickinson, Franklin Lakes, NJ) were analyzed with the NanoChip 400 system. Of the 130 specimens, 40 were real-time PCR negative for influenza A and B viruses and RSV. The remaining 90 specimens consisted of 30 specimens real-time PCR positive for influenza A and B viruses and RSV.
NanoChip 400 system.
Following the extraction of nucleic acid with an EZ1 virus minikit (version 2.0; Qiagen, Valencia, CA), reverse transcription (RT) and PCR amplification were performed according to the NanoChip 400 system protocol. Briefly, RT was performed with a program of 22°C for 10 min (one cycle), 42°C for 60 min (one cycle), 95°C for 5 min (one cycle), and 4°C (hold). A 10-μl volume of the RT reaction mixture was used for PCR amplification with a program of 95°C for 10 min (1 cycle); 95°C for 60 s, 55°C for 30 s, and 72°C for 45 s (2 cycles); 94°C for 60 s, 60°C for 30 s, and 72°C for 30 s (38 cycles); 72°C for 7 min (1 cycle); and 4°C (hold).
Ten microliters of each amplification product was transferred to a 96-well microtiter plate and diluted with buffer. The microtiter plates were sealed with a microplate seal before they were loaded on the NanoChip 400 instrument. The preparation of solutions for use with the NanoChip 400 instrument, any required system maintenance, and configuration of the instrument for the running of the assay were completed during RT and PCR amplification.
When specimens were available for repeat testing, the original specimens or nucleic acid extracts for those samples generating discordant results by DFA and with the NanoChip 400 system were analyzed at Focus Diagnostics (Cypress, CA) by RT-PCR, based on proprietary target sequences and TaqMan chemistry. Likewise, when specimens previously examined by real-time PCR with analytical reagents from Prodesse generated discordant results with the NanoChip 400 system, the samples were retested at Focus Diagnostics by RT-PCR.
Sequence analysis of specimens false negative for RSV with NanoChip 400 system.
Three RSV isolates were available for analysis. PCR primers were designed to hybridize upstream and downstream of the RSV assay primers used in the NanoChip 400 system. Separate sets of primers were used for RSV subgroup A and RSV subgroup B. RT-PCR was performed by using a one-step RT-PCR kit (Qiagen). RT was performed for 30 min at 50°C. The reverse transcriptase was inactivated and the polymerase was activated; both processes were achieved by subjecting the reaction mixture to 95°C for 15 min. PCR was performed by using a program of 40 cycles of 94°C for 1 min, 53°C for 1 min, and 72°C for 1 min and then holding the reaction mixture at 72°C for 10 min. Amplicons consistent with the size expected for RSV subgroup B were produced from two of the discrepant samples, and amplicons consistent with the size expected for RSV subgroup A were produced from a third discrepant sample. The PCR products from the three samples were purified, and automated DNA sequencing in both directions was performed at Sequetech Corporation (Mountain View, CA) with an Applied Biosystems 3730 × L DNA analyzer.
RESULTS
Comparison of DFA respiratory screen and NanoChip 400 system.
We first assessed the performance of the NanoChip 400 system by testing respiratory specimens that were submitted for DFA testing. As shown in Table 1, the concordance between DFA and the NanoChip 400 system was obtained for 106 of 122 (86.9%) specimens. Of the 122 specimens previously tested by DFA, 66 were positive both by DFA and with the NanoChip 400 system, while 40 were negative by both methods. Based on DFA as the gold standard, the sensitivity and the specificity of the NanoChip 400 system were 84.6% and 90.9%, respectively.
TABLE 1.
Comparison of DFA and NanoChip 400 system for detection of influenza A and B viruses, RSV, and parainfluenza virus types 1 to 3a
| NanoChip 400 system result | No. of samples with the following DFA result:
|
||
|---|---|---|---|
| + | − | Total | |
| + | 66 | 4 | 70 |
| − | 12 | 40 | 52 |
| Total | 78 | 44 | 122 |
The numbers of samples positive by use of the NanoChip 400 system/DFA were as follows: influenza A virus, 23/23 (100%); influenza B virus, 15/24 (62.5%); RSV, 18/21 (78.3%); and parainfluenza virus types 1 to 3, 10/10 (100%).
A total of 16 (13.1%) discordant specimens were positive by one method and negative by the other. Of the 12 samples with DFA-positive and NanoChip 400 system-negative discrepancies, 9 were positive for influenza B virus by DFA and negative for influenza B virus with the NanoChip 400 system, while 3 were RSV positive by DFA and RSV negative with the NanoChip 400 system. Three samples negative by DFA were RSV positive with the NanoChip 400 system, and one sample negative for parainfluenza virus by DFA was positive for parainfluenza virus with the NanoChip 400 system. Table 2 summarizes the discordant DFA and NanoChip 400 system results along with the real-time PCR results for 12 samples. Six samples were not available for additional testing.
TABLE 2.
Real-time PCR results for samples with discrepant results by DFA and with the NanoChip 400 system
| No. of specimens | Result
|
||
|---|---|---|---|
| DFA | NanoChip 400 system | Real-time PCR | |
| 3 | Influenza B virus | Negative | Negative |
| 1 | Influenza B virus | Influenza A virus | Influenza A virus |
| 1 | Influenza B virus | Negative | Influenza B virus |
| 4 | Influenza B virus | Negative | —a |
| 1 | RSV | Influenza B virus | Influenza B virus |
| 2 | RSV | Negative | — |
| 3 | Negative | RSV | RSV |
| 1 | Negative | Parainfluenza virus | Parainfluenza virus |
—, of the 16 specimens generating discrepant results by DFA and with the NanoChip 400 system, 6 were unavailable for real-time PCR testing and were excluded from the analysis.
A summary of the resolved data, excluding the data for samples no longer available for additional testing, is shown in Table 3. Following discrepancy analysis, the sensitivity and the specificity of the NanoChip 400 system were found to be 72/73 (98.6%) and 43/43 (100%), respectively. The positive and negative predictive values were 100% and 97.7%, respectively. Not included in the data set were data for two specimens with too few cells for definitive DFA result reporting which were positive for RSV with the NanoChip 400 system and by real-time PCR.
TABLE 3.
NanoChip 400 system versus DFA: resolved NanoChip 400 system resultsa
| NanoChip 400 system result | No. of samples with the following results based on repeat testing of samples with discrepant results by real-time PCR:
|
||
|---|---|---|---|
| True positive | True negative | Total | |
| + | 72 | 0 | 72 |
| − | 1 | 43 | 44 |
| Total | 73 | 43 | 116 |
Of the 16 samples with discrepant results between DFA and NanoChip 400 system testing, real-time PCR testing confirmed that 4 DFA-negative, NanoChip 400 system-positive specimens were true positive. Of the 12 samples with DFA-positive, NanoChip 400 system-negative results, real-time PCR confirmed the results for 5 of the specimens with NanoChip 400 system-negative results. Six specimens were not available for repeat testing by a third method.
Comparison of NanoChip 400 system and Prodesse real-time PCR for detection of influenza A and B viruses and RSV.
Of the 130 specimens analyzed by real-time PCR, the overall concordance of the results with those of the NanoChip 400 system was 122/130 (93.9%). With respect to Prodesse real-time PCR-negative specimens, 40 of 40 specimens were negative with the NanoChip 400 system.
By specific respiratory virus, the NanoChip 400 system detected the virus in 29/30 (96.7%) specimens positive by real-time PCR for influenza A virus, 28/30 (93.3%) specimens positive by real-time PCR for influenza B virus, and 25/30 (83.3%) specimens positive by real-time PCR for RSV. RSV accounted for five of the eight NanoChip 400 system-negative, Prodesse real-time PCR-positive samples.
As shown in Table 4, the overall sensitivity and specificity of the NanoChip 400 system compared to the results of real-time PCR were 91.1% and 100%, respectively.
TABLE 4.
Comparison of NanoChip 400 system and real-time PCR for detection of influenza A and B viruses and RSVa
| NanoChip 400 system result | No. of samples with the following results based on repeat testing of samples with discrepant results by a second real-time PCR method:
|
||
|---|---|---|---|
| + | − | Total | |
| + | 82 | 0 | 82 |
| − | 8 | 40 | 48 |
| Total | 90 | 40 | 130 |
Samples for which the NanoChip 400 system and real-time PCR results did not concur were retested at a reference laboratory by RT-PCR, based on proprietary target sequences and TaqMan chemistry.
DNA sequencing.
The sequencing results for three of the real-time PCR-positive RSV isolates that failed to be detected with the NanoChip 400 system are shown in Fig. 1. Mismatches between the primers and the sequences of the isolates were observed.
FIG. 1.
Sequence alignment of forward and reverse primers used in the assay and isolates that failed to be detected on the NanoChip 400 system. Sequence alignment was performed by the use of the MegAlign program. Any mismatch between the primer and the reference sequences is highlighted in pink. Consensus sequence data were derived from NCBI.
Work flow.
The total labor time per NanoChip 400 system result, excluding any required instrument maintenance, was less than 4 min per sample. The total labor time for required maintenance, preparation, and loading of reagents, as well as configuration of the microtiter tray and instrument, was less than 40 min.
If RT is performed and the PCR master mixture is prepared ahead of time and any required instrument maintenance as well reagent preparation and instrument configuration is performed during the RT and PCR amplifications, the total process flow time up to the loading of the trays onto the NanoChip 400 instrument was less than 4 h; this total time includes the times for nucleic acid extraction, RT and PCR amplification, and the preparation of microtiter trays for loading on the NanoChip 400 instrument. Analysis with the NanoChip 400 system took approximately 3 min per sample.
DISCUSSION
For practical considerations, the rapid detection of respiratory viral infections currently relies on less sensitive lateral flow or enzyme immunoassay antigen detection methods. While the shell vial assay for respiratory viruses can reduce the time for detection from several days to 1 or 2 days after receipt of the specimen in the clinical laboratory, the sensitivity of the shell vial assay is substantially less than that of PCR (3). While the negative predictive value of PCR results within hours makes this assay superior to culture, a screen for a battery of respiratory viruses is useful for infection control, treatment, and patient management. The limitations of the real-time PCR assay for several viral agents limit its usefulness as a screening assay for respiratory viruses.
For the reasons described above, DFA has continued to play a role in the detection a broad spectrum of respiratory viruses. At the same time, the subjectivity of interpretation of the results of DFA and the decreasing availability of qualified licensed clinical laboratory personnel limit the use of DFA to larger laboratories with specialized virology sections. Even in larger laboratories with clinical laboratory personnel trained to interpret DFA results, off-shift testing often relies on rapid lateral flow or enzyme immunoassays. The sensitivities of DFA and indirect immunofluorescence assay (IFA) compared to the results of culture vary from laboratory to laboratory and by the various staining methods used. The sensitivities of DFA and IFA testing for influenza A virus have ranged from 45 to 65%, with a median specificity of 98%, while the sensitivities of DFA and IFA testing for influenza B virus have ranged from 59.8% to 90%, with a median specificity of 73.9% (11). For RSV, the sensitivities range from 80 to 97%, with a median specificity of 97% (10). The sensitivities reported for parainfluenza viruses range from 70 to 83%, with a 94% specificity (5).
The present study is the first study to have compared the performance of the NanoChip 400 system with DFA and real-time PCR for the detection of respiratory viruses. Compared to real-time PCR, the NanoChip 400 system demonstrated excellent sensitivity and specificity and can replace the DFA procedure for multiple respiratory viruses. As the results may be available in less than 4 h, the NanoChip 400 system has the potential to influence the clinical care of individual patients.
Of the six unresolved specimens that tested positive by DFA and negative with the NanoChip 400 system, four were positive for influenza B virus by DFA. As the NanoChip 400 system detected 28 of 30 specimens positive for influenza B virus by real-time PCR, it does not appear that the NanoChip 400 system has a weakness in detecting influenza B virus. In light of the fact that specimens were not available in all cases for confirmatory real-time PCR testing, a number of possibilities may account for the discrepancy in results between DFA and the NanoChip 400 system. It is possible that these particular influenza B virus strains did not amplify with the influenza B virus-specific primers supplied with the NanoChip 400 assay. Degradation of the target during storage may also account for the negative NanoChip 400 system results, which may be reflective of RNA stability differences in M4 viral transport medium and Universal viral transport medium. In an earlier report, Chan et al. reported that storage may adversely affect the performance of influenza B virus detection assays compared to that of influenza A virus detection assays (2).
Of the five specimens with false-negative results for RSV with the NanoChip 400 system and positive results for RSV by real-time PCR, sequencing studies with three of the isolates suggest that there is still the potential for optimization of the sensitivity. A disadvantage of multiplex assays is that the limit of detection for each substrate is often increased, which may lead to false-negative results when small amounts of viral RNA are present. Since the study was performed, additional work has been made in developing primers that will detect the mutations observed in the RSV strains encountered in this study. In addition, the NanoChip 400 system platform is adaptable so that it may detect the different isolates that predominate in different seasons as well as sequence variations between and within RSV subgroups A and B.
In summary, the results obtained with the NanoChip 400 system demonstrated a high overall agreement with those obtained by DFA and real-time PCR for the detection of respiratory viruses. As virus detection with the NanoChip 400 system is not limited to several targets like the real-time PCR is, the system shows considerable promise as an alternative to a DFA battery and real-time PCR for the efficient and timely detection of respiratory viruses. Depending on the volume, consumable costs would be approximately $35 for a screening test for six respiratory viral agents, which is comparable to the cost for testing by other molecular methods. As the $35/test cost for the NanoChip 400 system was negotiated as a reagent rental at a price equivalent to the labor and reagent cost required to perform the DFA for all the viruses covered by the NanoChip 400 system, the overall effect on the reagent budget was cost neutral. The impact on labor cost per test would depend on the number of samples per run but would be approximately $4 per specimen, by use of a labor rate of $60 per hour, including benefits.
Operationally, the declining availability of trained clinical laboratory scientists points to the critical need for an efficient and cost-effective alternative to DFA. The NanoChip 400 system is a promising system for the screening of specimens for multiple viral respiratory agents.
Footnotes
Published ahead of print on 5 March 2008.
REFERENCES
- 1.Barenfanger, J., C. Drake, N. Leon, T. Mueller, and T. Troutt. 2000. Clinical financial benefits of rapid detection of respiratory viruses: an outcomes study. J. Clin. Microbiol. 382824-2828. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Chan, K. H., N. Maldeis, W. Pope, A. Yup, A. Ozinkas, J. Gill, W. H. Seto, K. F. Shortridge, and J. S. M. Peiris. 2002. Evaluation of the Directigen Flu A + B test for rapid diagnosis of influenza virus type A and B infections. J. Clin. Microbiol. 401675-1680. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Espy, M. J., et al. 2006. Real-time PCR in clinical microbiology: applications for routine laboratory testing. Clin. Microbiol. Rev. 19165-256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Landry, M. L., S. Cohen, and D. Ferguson. 2000. Impact of sample type on rapid detection of influenza virus by cytospin-enhanced direct immunofluorescence and enzyme-linked immunosorbent assay. J. Clin. Microbiol. 38429-430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Leland, D. S. 2007. Parainfluenza and mumps virus, p. 1353-1360. In P. R. Murray, E. J. Baron, J. H. Jorgensen, M. L. Landry and M. A. Pfaller (ed.), Manual of clinical microbiology, vol. 2, 9th ed. American Society for Microbiology, Washington, DC. [Google Scholar]
- 6.MacCartney, K. K., M. H. Gorelick, M. L. Manning, R. L. Hodinka, and L. M. Bell. 2000. Nosocomial respiratory virus infections: the cost-effectiveness and cost-benefit of infection control. Pediatrics 106520-526. [DOI] [PubMed] [Google Scholar]
- 7.Michaels, M. G., C. Serdy, K. Barbadora, M. Green, A. Apalsch, and E. R. Wald. 1992. Respiratory syncytial virus: a comparison of diagnostic modalities. Pediatr. Infect. Dis. J. 11613-616. [PubMed] [Google Scholar]
- 8.Minnich, L. L., T. F. Smith, and C. G. Ray. 1988. Cumitech 24, Rapid detection of viruses by immunofluorescence. Coordinating ed., S. C. Spector. ASM Press, Washington, DC.
- 9.Rabalais, G. P., G. G. Stout, K. L. Ladd, and K. M. Cost. 1992. Rapid diagnosis of respiratory viral infections by using a shell vial assay and monoclonal antibody pool. J. Clin. Microbiol. 301505-1508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Tang, Y.-W., and J. E. Crowe, Jr. 2007. Respiratory syncytial virus and human metapneumovirus, p. 1361-1377. In P. R. Murray, E. J. Baron, J. H. Jorgensen, M. L. Landry and M. A. Pfaller (ed.), Manual of clinical microbiology, vol. 2, 9th ed. American Society for Microbiology, Washington, DC. [Google Scholar]
- 11.Uyeki, T. M. 2003. Influenza diagnosis and treatment in children: a review of studies on clinically useful tests and antiviral treatment for influenza. Pediatr. Infect. Dis. J. 22164-167. [DOI] [PubMed] [Google Scholar]
- 12.Whimbey, E., R. E. Champlin, J. A. Englund, N. Q. Mirza, P. A. Piedra, J. M. Goodrich, D. Przepiorka, M. A. Luna, R. C. Morice, J. L. Neuman, et al. 1993. Combination therapy with aerosolized ribavirin and intravenous immunoglobulin for respiratory syncytial virus disease in adult bone marrow transplant recipients. Bone Marrow Transplant. 16393-399. [PubMed] [Google Scholar]
- 13.Woo, P. C., S. S. Chiu, W. H. Seto, and M. Peiris. 1997. Cost-effectiveness of rapid diagnosis of viral respiratory tract infections in pediatric patients. J. Clin. Microbiol. 351579-1581. [DOI] [PMC free article] [PubMed] [Google Scholar]