Ratification of rapid rotavirus diagnostic test strips

Opion Papers: Ratification of rapid rotavirus diagnostic test strips

Ratification of rapid rotavirus diagnostic test strips EMC Theron, MM Nyaga, JB Dewar

Elizabeth Theron, BSc(Hons), MSc, Junior Lecturer Department of Life and Consumer Sciences, UNISA, Florida Campus Martin Nyaga, BSc(Hons), MSc, MRC, Research Student Diarrhoeal Pathogens Research Unit, University of Limpopo, Medunsa Campus; WHO Rotavirus Regional Reference Laboratory John Dewar, BSc(Hons), MSc, PhD, Professor Department of Life and Consumer Sciences, UNISA, Florida Campus E-mail: [email protected] Keywords: rotavirus, enzyme immunoassay, immunochromographic test, pathology practice, South Africa

The World Health Organization and Centers for Disease Control and Prevention estimate that rotavirus-associated gastroenteritis is the third most common cause of death in young children, with approximately 453 000 deaths annually, and that more than 90% of these deaths occur in developing countries. An accurate diagnosis of rotavirus in stool samples is important to determine the rotavirus burden of disease, possible rotavirus vaccine failure, and to limit the inappropriate use of antibiotics. While enzyme immunoassay (EIA) is one of the most sensitive and specific testing methods to achieve this, it is relatively time consuming and expensive for laboratories that require short turnaround times. This study evaluated the sensitivity and specificity of the simpler and cheaper Combi® immunochromatographic test (ICT) strip (Coris BioConcepts, Gembloux, Belgium) routinely used by pathology practices to screen stool samples for rotaviruses. Of 6 050 stool samples collected in 2010 and 2011 from a private pathology practice in Pretoria and screened using the Combi® ICT strip, 752 (12%) were tested using a commercial, large-scale EIA detection method, e.g. ProSpecT® Rotavirus EIA kit (Oxoid Diagnostics, Basingstoke, UK). Results showed the sensitivity of the Combi® ICT strip to be 93.7% and the specificity of this assay to be 99.8%, when compared to the gold standard EIA. The study results support the use of the Combi® ICT strip as an appropriate detection assay, showing that it has a highly sensitive and specific reaction to rotavirus antigens. South Afr J Infect Dis 2014;29(2):91-94

Peer reviewed. (Submitted: 2014-00-00. Accepted: 2014-00-00.) © SAJID

Introduction

associated with rotavirus disease, will further reduce these estimated figures.4 Between 140 000 and 230 000 annual deaths in sub-Saharan Africa are due to rotavirus infections.5-7 Against this background, it is critically important that rotavirus infection is diagnosed rapidly so as to be able to contribute to correct statistics on rotavirus-associated diarrhoea per country, and worldwide to raise awareness of the impact of this disease on healthcare professionals and parents/caregivers, and finally, to evaluate the impact of vaccinations and identify potential vaccine failure.

Estimates by the World Health Organization (WHO) and the Centers for Disease Control and Prevention rate rotavirus gastroenteritis to be the third most common cause of death in young children,1 accounting for more than 70% of all serious diarrhoeal cases.2 Prior to the introduction of rotavirus vaccines, the WHO estimated that rotavirus infections were responsible for an average of 527 000 (range of 475 000–580 000) deaths annually, with more than 90% of these annual deaths occurring in developing countries.3 More recent estimates by the WHO and The United

When selecting a diagnostic assay, the most important parameters include the specificity and sensitivity of these tests. Different methods used to detect rotavirus in stool samples or rectal swabs have been described by various authors.8-10 These include electron microscopic analysis of negatively stained virions, antigen detection by enzyme immunoassays (EIA), passive particle agglutination tests or lateral flow immunoassays, polyacrylamide gel electrophoresis (PAGE) with silver nitrate staining, deoxyribonucleic acid-agarose gel electrophoresis dotblot hybridisation assays, nucleic acid sequence-based amplification (NASBA), reverse transcription polymerase chain reaction (RT-PCR) and virus isolation in cell cultures. The use of the cell culture of rotaviruses is considered to be unreliable for the diagnostic analysis of infections as the culturing of rotaviruses is often difficult and time consuming,11 and it has been suggested that RT-PCR is the preferred choice for strain typing.9

Nations Children’s Fund indicate an overall decline in total mortality due to diarrhoeal-associated disease in children younger than five years of age from 1.8 million deaths in 2003 to 1.3 million deaths in 2008. Owing to this decline, there was a need to re-evaluate the estimated number of deaths associated with rotavirus disease, especially when taking into consideration the introduction of the rotavirus vaccines, Rotarix® and RotaTeq®, and current improvements in sanitation and hygiene worldwide. Currently, it is estimated that rotavirus-associated diarrhoea is responsible for 453 000 deaths annually in children younger than five years of age. This reduction is attributed to an overall decline in diarrhoeal-associated mortality, as well as increases in the detection rates of rotavirus-associated diarrhoea. The future introduction of rotavirus vaccines, particularly in countries with a high mortality rate

South Afr J Infect Dis

91

2014;29(2)


When levels of virus shedding are very high, electron microscopy, EIA and PAGE are suitable for diagnostic purposes, and when high-titre monoclonal antibodies are used, antigen detection assays can demonstrate high sensitivity and specificity.10,12 When samples are collected in a late phase of an illness, or collected from alternative clinical sites, including throat swabs, cerebrospinal fluid or respiratory secretions, or when the number of viral particles is expected to be low, then more sensitive methods, such as RT-PCR, should be used.12 Electron microscopic detection is the most rapid detection method used to identify rotaviruses in samples, but is less sensitive than immunoassays, as the analysed sample requires at least 106–108 virions/ml to be detectable.8,10,12,13

Other nucleic acid-based rotavirus diagnostic assays routinely involve RT-PCR analysis of the rotavirus RNA. This requires converting rotavirus RNA to complementary deoxyribonucleic acid, which is then amplified to detectable levels by PCR. The PCR is a method in which DNA fragments are amplified to detect the presence of a specific gene sequence.16 As a detection method, RT-PCR improves group A rotavirus detection more than EIA or electron microscopy.13 The real-time RT-PCR of rotavirus genes allows for amplification and fluorescent detection phases to be performed in one test tube by a single instrument. The product for real-time RT-PCR is plotted in the form of an amplification curve, consisting of an early background phase, log phase indicating the presence of the tested organism, and the end-point.3 Real-time RT-PCR is considered to be a more rapid and more sensitive method of detection and quantitation of rotavirus than the conventional RT-PCR or EIA detection methods.23 To date, several sensitive real-time RT-PCR methods, based on primers specific to several different rotavirus genes, have been developed.3 A study conducted by Logan, O’Leary and O’Sullivan showed a 111% increase in the rate of detection by realtime RT-PCR, compared to latex agglutination, and a 186% increase rate of detection compared to electron microscopy.13 The need for a sophisticated laboratory containing expensive equipment and reagents used by highly trained laboratory staff is an important consideration in PCR-based rotavirus detection.

Latex agglutination assays are based on the agglutination of virusspecific, immunoglobulin-coated latex beads by antigens that are present in the tested samples.11 These assays are less sensitive than rotavirus EIA.10,12 False negative results are noted in samples with a low number of virions, resulting in the low sensitivity and low specificity of these assays.8 Latex assays are an insensitive rotavirus diagnostic tool. A sensitivity of only 57% and a specificity of 93% was indicated in a previous study when a latex agglutination assay was compared against EIA analysis of the same samples.14 As per the WHO,3 the most common method of rotavirus detection involves EIA, and this method has been used in large-scale rotavirus surveillance studies. The assay is based on the detection of the rotavirus group A antigen in stool samples by labelled antibody molecules.15,16A sandwich EIA consists of a plate coated with a capture antibody, to which the diluted stool sample solution and an enzyme-linked detector antibody are added, in order to determine the presence of a specific antigen. This technique has also been used to identify different G serotypes. However 20–30% of all samples tested failed to provide serotyping results, leading to the development of alternative characterisation techniques, including RT-PCR.17 The overall sensitivity of EIA requires a range of 105–106 viral particles/ml.12 Authors from a study conducted in Argentina indicated that the EIA resulted in a specificity of 100% and sensitivity of 98.4%, when compared to RT-PCR analysis. The ProSpecT® Rotavirus EIA kit (Oxoid Diagnostics, Basingstoke, UK) was used during the research study. This rotavirus test uses a polyclonal antibody that is in a solid-phase sandwich EIA to detect the specific antigen present in group A rotaviruses. The principle of the EIA method is based on a colour change observed for positive samples, whereas negative samples remain colourless. Positive samples can be identified when the liquid in the well changes to a yellow colour, thereby showing the presence or absence of the specific antigen in the tested samples.

A relatively new development in diagnostic assays involves rapid immunochromatographic test kits, such as the Combi® ICT strips. The specificity of these rapid test kits is based on the interaction between a rotavirus-specific monoclonal antibody and the specific rotavirus protein. In addition to the monoclonal antibody, such kits use a homogeneous nitrocellulose membrane system impregnated with colloidal gold particles. Samples to be tested are diluted in a dilution buffer that is included in the kit. The diluted sample is allowed to interact with the test strip, and rotaviruses then bind to the monoclonal antibody, and the interaction is visualised following adsorption of colloidal gold particles within the immune complex. A previous study by Dewar et al indicated that the Combi® ICT strip resulted in a sensitivity of 88% and a specificity of 100% in detecting rotavirus when compared to EIA analysis of stool samples.14 In comparison to the more sensitive PCR-analysis, EIA, and particularly, rapid ICT, the analysis of rotaviruses is relatively inexpensive, simple and straightforward using the Combi® ICT strip. The objective of this study was to evaluate the sensitivity and specificity of the Combi® ICT strips that are used by pathology practices during routine testing of stool samples for rotaviruses, by comparing these results against the results obtained after retesting the same stool specimens using a commercial, large-scale EIA detection method, e.g. ProsPectT® Rotavirus EIA kit.

In the molecular characterisation of rotavirus, the isolation of rotavirus double-stranded RNA (dsRNA), followed by PAGE, allows these negatively charged dsRNA molecules to separate through the gel, depending on molecular weight. Resulting migration patterns can then be visualised by staining the gel with silver nitrate.3,18-20 Electropherotypes provide some information on the genetic diversity of rotaviruses circulating in a population.9,21,22 The overall sensitivity of PAGE is relatively low, and requires a range of 108–109 viral particles/ml.12 Samples that test PAGEnegative, after testing EIA-positive, could indicate degradation of the virus particles.9


Method Seven hundred and fifty-two (12%) stool samples were selected from a collection of 6 050 stool samples. The tested samples included those from Trichard, Middelburg and Rustenburg, and 327 randomly selected ones from Gauteng and selected areas within North West province. The majority of samples from Gauteng and North West province were selected

92

2014;29(2)


Results

from the expected peak periods for rotavirus infections (May to August). The selection was also based on the age of the children. The majority included diarrhoeal samples collected from children aged 6-24 months. In addition, the tested samples included 4 (0.5%) samples which were not initially tested by the private pathology practice for the presence of rotavirus. Thus, 748 samples that were screened using the Combi® ICT strip by the private pathology practice were included in this study. For the purposes of the latter, this subset of stool samples was retested using the EIA method, according to the manufacturer’s instructions. Of the samples analysed, 538 (71.5%) were recorded as Combi® ICT stripnegative and 210 (28%) were recorded as Combi® ICT strip-positive.

The EIA analysis was used to evaluate the effectiveness, sensitivity and specificity of the rapid test strips used by private pathology practices during routine testing for rotaviruses. The results of this analysis are shown in Table I. The results from this study included 223 positive and 525 negative samples. The high number of retested and confirmed negative samples reflects the high specificity of the Combi® ICT strip kits. While the vast majority of the EIA results corresponded to the Combi® ICT strip results, an additional 14 (5.9%) rotavirus-positive samples were detected using the EIA method.

The EIA procedure involves preparing a 10% suspension for each faecal sample, by mixing approximately 0.1 g or 100 µl of sample with 1 ml of distilled water (dH2O). Each sample applied per well was identified by labelling the corresponding well on an EIA worksheet. Two drops (equal to 100 µl) of the positive control, supplied with the ProSpecT® Rotavirus EIA kit, was added into the first micro-well. An aliquot of 100 µl of dH2O was added to the second micro-well as a negative control. An aliquot of 100 µl of each of the prepared stool suspensions was added to each of the remaining 94 micro-wells. Two drops of conjugate (an enzymelinked detector antibody) were then added to each micro-well, and the plate was covered and sealed it in its original packet, and incubated for 60 minutes in a 37°C incubator. After incubation, the contents of the micro-wells were discarded into a waste dish, and freshly prepared wash buffer, supplied with the kit, was poured into the micro-wells. Thereafter, all fluid from the wells was discarded into the waste dish, and the plate was inverted and tapped on absorbent paper to remove all traces of the wash buffer. This washing step was repeated an additional four times. Two drops of the enzyme substrate were added to each micro-well, and the plate was covered and sealed in its original packet, and incubated for 10 minutes in a 37°C incubator. Any observed colour changes were noted on the EIA worksheet. After incubation, two drops of stop solution were added to each micro-well to stop the substrate reaction. The plate was read spectrophotometrically at a setting of 450 nm, and the spectrometer printout compared to the visually observed results.

In addition, one sample tested positive by Combi® ICT strip, but negative using the EIA method. This appears to be a false positive as only the faintest Combi® ICT strip positive result was obtained, and subsequent PAGE and RT-PCR assays remained negative for that sample. From the results of this study, the sensitivity of the rapid Combi® ICT strip was calculated to be 93.7% (94%) when compared to the gold standard ProSpecT® Rotavirus EIA kit, while the specificity of the Combi® ICT strip was 99.8% (100%) when compared to the EIA. The chi-square test statistical comparison between the EIA and Combi® ICT strip tests was 0.4556, with one degree of freedom (p-value 0.50). This indicates that there was no significant statistical difference between the two testing methods. Nonetheless, it should be mentioned that on a global scale, when testing for the presence of rotavirus in numerous infected children, it is essential to ensure that the most accurate diagnostic assay is used so as to contribute to correct global statistics, raise disease awareness and evaluate the effectiveness of rotavirus vaccines.

Discussion An accurate diagnosis of rotavirus in stool samples is important in the global evaluation of rotavirus-associated disease, and EIA is one of the most sensitive and specific testing methods to achieve this. However, as diagnostic laboratories receive a large number of samples to test, and as EIA is a relatively costly and time-consuming procedure for use in diagnostic laboratories, such as private pathology practices that utilise large volumes of samples and which operate within short turnaround times when reporting results, EIA analysis may not always be a viable option. Alternative tests should be evaluated. The rotavirus latex agglutination test is insufficiently sensitive, while the Combi® ICT strip is a viable, alternative, rapid rotavirus diagnostic assay. Previous test data have demonstrated the high sensitivity of the Combi® ICT strip.14 The results of the current study confirmed that the sensitivity of the currently used Combi® ICT strip in detecting rotavirus in stool samples was slightly higher than that reported in the study published in 2005, with similar, consistent specificity. Bearing in mind the increased mutability of RNA viral genomes, these results underscore the need for continuous monitoring of detection rates in antibody-based detection assays, so as to ensure the continued validity and reliability of these products, particularly when they are used for diagnostic purposes.

Positive and negative results from the EIA analysis were then compared with the diagnostic results initially obtained from testing the same specimens for rotavirus using the rapid Combi® ICT strip. The results of this comparison allowed an evaluation of the sensitivity and specificity, i.e. the effectiveness, of the rapid Combi® ICT strip used for the diagnosis of rotavirus in stool specimens by private pathology practices. To validate the results statistically, the chi-square test was used, as this test assesses if results observed on two variables, as expressed in a contingency table (Table I), are independent of each other. Table I: Sensitivity and specificity of the test strips against enzyme immunoassay EIA Combi® ICT test strips

Positive

Negative

Total

Positive

209

1

210

Negative

14

524

538

Total

223

525

748

EIA: enzyme immunoassay, ICT: immunochromatographic test


93

2014;29(2)


Conclusion

6. Aminu M, Page NA, Ahmad AA, et al. Diversity of rotavirus VP7 and VP4 genotypes in northwestern Nigeria. J Infect Dis. 2010;202(Suppl 1): S198-S204.

This study, as a follow-up to a previous study conducted by Dewar et al, serves to support the use of the Combi® ICT strip by commercial diagnostic laboratories as an appropriate detection assay, as it demonstrated a highly sensitive and specific reaction to rotavirus antigens.

7. Madhi SA, Cunliffe NA, Steele D, et al. Effect of human rotavirus vaccine on severe diarrheoa in African infants. N Engl J Med. 2010;362(4):289-298. 8. White DO, Fenner FJ. Medical virology. 4th ed. San Diego: Academic Press, 1994; p. 138-139, 524-528. 9. Fischer TK, Gentsch JR. Rotavirus typing methods and algorithms. Rev Med Virol. 2004;14(2):71-82. 10. Farkas T, Jiang X. Rotaviruses, calciviruses, astroviruses, enteric adenoviruses and other diarrheic viruses. In: Murray PR, Baron EJ, Jorgensen JH, et al. Manual of clinical microbiology. 9th ed. Washington: ASM Press, 2007; p. 1453-1464. 11. Murray PR, Rosenthal KS, Pfaller MA. Medical microbiology. 5th ed. Pennsylvania: Elsevier Mosby, 2005; p. 627-633.

Acknowledgements

12. Arguelles MH, Villegas GA, Castello A, et al. VP7 and VP4 genotyping of human group A rotavirus in Buenos Aires, Argentina. J Clin Microbiol. 2000;38(1):252-259.

The authors would like to thank the Medical Research Council (MRC) of South Africa for providing funding for the research project, the MRC Diarrhoeal Pathogens Research Unit for use of equipment, and the pathologist in charge of the microbiology department of the private pathology practice in Pretoria for the samples, as well as the data required to conduct the research.

13. Logan C, O’Leary JJ, O’Sullivan N. Real-time reverse transcription-PCR for detection of rotavirus and adenovirus as causative agents of acute viral gastroenteritis in children. J Clin Microbiol. 2006;44(9):3189-3195. 14. Dewar J, de Beer M, Elliott E, et al. Rapid detection of rotaviruses: are laboratories underestimating infection in infants? S Afr Med J. 2005;95(7):494-495. 15. Beards GM, Campbell AD, Cottrell NR, et al. Enzyme-linked immunosorbent assays based polyclonal and monoclonal antibodies for rotavirus detection. J Clin Microbiol. 1984;19(2):248-254. 16. Atlas RM. Principle of microbiology. Student ed. Sydney: WCB Publishers, 1997; p. 713. 17. Page NA, De Beer MC, Seheri LM, et al. The detection and molecular characterization of human G12 genotypes in South Africa. J Med Virol. 2009;81(1):106-113.

References

18. Herring AJ, Inglis NF, Ojeh CK, et al. Rapid diagnosis of rotavirus infection by direct detection of viral nucleic acid in silver-stained polyacrylamide gels. J Clin Microbiol. 1982;16(3):473-477.

1. Ciarlet M, Schödel F. Development of a rotavirus vaccine: clinical safety, immunogenicity, and efficacy of the pentavalent rotavirus vaccine, RotaTeq®. Vaccine. 2009;27(6):G72-G81.

19. Laird AR, Gentsch JR, Nakagomi T, et al. Characterization of serotype G9 rotavirus strains isolated in the United States and India from 1993 to 2001. J Clin Microbiol. 2003;41(7):3100-3111.

2. Adler M, Schulz S, Fischer R, Niemeyer CM. Detection of rotavirus from stool samples using a standardized immune-PCR (“Imperacer”) method with end-point and real-time detection. Biochem Biophys Res Commun. 2005;333(4):1289-1294.

20. Desselberger U, Gray J, Estes MK. Rotaviruses. In: Mahy BWJ, Vermeulen V, editors. Topley and Wilson’s microbiology and microbial infections virology. 10th ed. London: Edward Arnold Publishers, 2005; p. 946-955.

3. World Health Organization. Manual of rotavirus detection and characterization methods. Geneva: WHO; 2010 [homepage on the Internet].. c2010. Available from: http://whqlibdoc.who.int/hq/ 2008/WHO_IVB_08.17_eng.pdf

21. Uchida R, Pandey BD, Sherchand JB, et al. Molecular epidemiology of rotavirus diarrhea among children and adults in Nepal: detection of G12 strains with P[6] or P[8] and a G11P[25] strain. J Clin Microbiol. 2006;44(10):3499-3505.

4. Tate JE, Burton AH, Boschi-Pinto C, et al. 2008 estimate of worldwide rotavirus-associated mortality in children younger than 5 years before the introcution of universal rotavirus vaccination programmes: a systematic review and meta-analysis. Lancet Infect Dis. 2012;12(12):136-141.

22. Dzikwi AA, Umoh JU, Kwaga JKP, et al. Electropherotypes and subgroups of Group A rotaviruses circulating among diarrhoeic children in Kano, Nigeria. Ann Afr Med. 2008;7(4):163-167. 23. Zeng SQ, Halkosalo A, Salminen M, et al. One-step quantitative RT-PCR for the detection of rotavirus in acute gastroenteritis. J Virol Methods. 2008;153(2):238-240.

5. Steele AD, de Vos B, Tumbo J, et al. Co-administration study in South African infants of a live-attenuated oral human rotavirus vaccine (RIX4414) and poliovirus vaccines. Vaccine. 2010;28(39):6542-6548.


94

2014;29(2)