Abstract
Rotavirus and Adenovirus are common causes of gastroenteritis in children younger than 3 years worldwide. Rapid Antigen Testing (RAT) is a quick and easy tool to detect virus antigen in stool samples and is more specific than sensitive (higher specificity and lesser sensitivity). Reverse transcription-polymerase chain reaction (RT-PCR) and PCR are more sensitive and specific than RAT. Sensitive and specific tools are required for true diagnosis. We aim to determine sensitivity and specificity of RAT versus PCR testing of rotavirus and adenovirus. From November 18th 2016 to November 18th 2017, all children up to 3 years of age who presented to Mayo University Hospital with vomiting and diarrhoea had their stool tested for rotavirus and adenovirus by RAT in Galway University Hospital Laboratory (GUHL) and by PCR testing in the National Virus Reference Laboratory (NVRL) in Dublin; 143 stool samples were tested for Adenovirus, 126 (88%) tested negative at NVRL, two false positive at GUHL, specificity (98.5%). Seventeen were adenovirus positive in the NVRL, two false negative in GUHL, sensitivity (88%); 144 samples were tested for rotavirus, 108 (75%) were RV negative in the NVRL, one false positive at GUHL, specificity (99%); 36 samples were rotavirus positive in the NVRL, ten (28%) false negative in GUHL, sensitivity (72%). RAT has higher specificity than sensitivity and may be useful for mass screening at times of rotavirus or adenovirus outbreaks. PCR remains more sensitive and specific than RAT and is still required for true diagnosis.
Keywords: National virus reference laboratory (NVRL), Galway university hospital laboratory (GUHL), Mayo university hospital (MUH), Gastroenteritis (GE), Rotavirus (RV), Adenovirus (AdV)
Introduction
Detection of virus genome using real time reverse transcription-polymerase chain reaction (RT-PCR) is more sensitive and specific than Rapid Antigen Testing (RAT) and comprised of amplification of certain regions of the genome of the virus followed by identification of the genotype by fragment size analysis using electrophoresis [1–3]. RAT—as RV detecting tool—is more specific than sensitive (higher specificity and lesser sensitivity) [4–6]. RAT is a quick and easy tool to detect virus antigen in stool samples using immunochromatography and may be useful for investigating and screening group A rotavirus (RV) during outbreaks of food-borne and person-to-person transmitted gastroenteritis(GE). [5] This rapid diagnostic test is easy to perform at the bedside, as it takes only 20 min to reach a diagnosis with a simple procedure, and does not require special equipments. [5] Of 71 samples that were positive for RV by RT-PCR, 69, 68 and 63 were also recognised by RAT kits, indicating 97.2, 95.8 and 88.7% sensitivity for RAT kits, with only one false positive result in one of the three RAT kits (specificity up to 100%). [5]
When comparing the sensitivity of RAT for group A RV detection, it is clearly demonstrated that among various immunochromatography kits, sensitivity and specificity for group A RV infection were a bit different. In addition, it was observed that several RV genotypes G1, G3 and G9 were reacted with these kits. Therefore, genotype variations of RV may not be a problem for false negative results. [5]
RT-PCR assay was found to be specific to RV and broadly reactive to RV genogroups 1–4, 9, 10 and 12. [3] Specificity testing did not identify any cross-reactivity of the assay with a panel of 36 non-RV enteric virus specimens. [3]
Highly sensitive and specific methods such as one-step RT-PCR are still required for true diagnosis of viral GE following clinical suspicion of GE and GE associated complications and for RV vaccine efficacy trials. [1–3].
Materials and methods
Period of prospective recruitment
November 18th 2016 to November 18th 2017.
Inclusion criteria
All children up to 3 years old that attended the Emergency Department (ED) or were admitted to Mayo University Hospital (MUH), CO Mayo Ireland, with vomiting and diarrhoea (loose stool) or diarrhoeal symptoms.
All children up to 3 years old admitted to the paediatric ward (PW) due to symptoms other than vomiting and diarrhoea, who developed diarrhoea 3 days (72 h) after their admission, fulfilling the definition of possible nosocomial infection.
All children up to 3 years old readmitted to the paediatric ward with vomiting and diarrhoea within 48 h following recent discharge, possible nosocomial infection.
MUH is the main referral hospital in our region of Ireland, a busy regional hospital.
Exclusion criteria
Children presenting with diarrhoea due to other disease e.g., immunodeficiency or inflammatory bowel disease or presenting with the same diagnosis within a 48-h period. In total, four children were excluded.
Parents not willing to participate or those who chose to opt out of the research at any time were allowed to withdraw from the study.
The method of testing used in Galway University Hospital laboratory (GUHL) was by RAT for AdV and RV. The method of testing used in the NVRL to detect viruses was by reverse transcription (RT)-polymerase chain reaction (PCR) of RV RNA genome and by PCR testing of AdV DNA genome. [7–12] All samples received at the NVRL for the investigation of viral GE were extracted on the Roche MagNA Pure 96 as per manufacturer’s protocol. Samples were tested by RT-PCR for the detection of RV, and PCR for AdVF. [7–12] Testing and analysis were performed on ABI Viia7 (RV, AdVF) instrument and analysed using the ABI software version 2.1.1 (RV, AdVF).
During the study period, stool samples were split into two on receipt to the local laboratory at MUH. One half of the stool samples were sent to GUHL and were tested by RAT for adenovirus F and rotavirus and the other half were stored frozen at MUH laboratory.
The results of the samples processed at MUH were followed up and recorded. RV and AdV positive results were identified and their matching samples, already stored frozen at MUH, were sent in quarterly batches to the National Virus Reference Laboratory (NVRL) in Dublin to be tested by (RT)-PCR for rotavirus and by PCR for enteric adenovirus (AdV) F.
Likewise, samples that were tested negative for RV and AdV at MUH were identified. Their other halves, already stored frozen at MUH, were sent in quarterly batches to the NVRL for further testing via RT-PCR for RV and PCR for AdVF. All stool samples that were confirmed positive or negative for AdV and RV were sent to the NVRL at the same time. Samples were also analysed in the NVRL at the same time.
Information leaflets were given to parents of children contributing to the research, explaining the purpose of the research study. Consent was obtained and parents were given the right to choose to withdraw or opt out of the research at any time.
A certificate of ethical approval for the research study was obtained from the Clinical Research Ethical Committee at MUH before study commencement.
Results
The results in this section are based on the objectives to determine sensitivity specificity data analysis defining accuracy of viral detection methods for two most common pathogens (Rotavirus and Adenovirus).
These data assessed Rapid Antigen Testing (RAT) in GUHL Versus PCR testing of Virus in the NVRL), Sensitivity and Specificity of RAT (Tables 1, 2).
Table 1.
Adenovirus detecting method: RAT in GUHL versus PCR detection of virus in the NVRL
| Month | Test | Number of Confirmed AdV Positve Cases | Number of False Negative Cases (GUHL) | Number of Confirmed AdV Negative Cases | Number of False Positive Cases (GUHL) |
|---|---|---|---|---|---|
| Nov 18–30 | RAT(GUH) | 0 | 0 | 2 | 0 |
| PCR(NVRL) | 0 | 2 | |||
| December | RAT(GUH) | 0 | 0 | 18 | 0 |
| PCR(NVRL) | 0 | 18 | |||
| January | RAT(GUH) | 0 | 0 | 12 | 0 |
| PCR(NVRL) | 0 | 12 | |||
| February | RAT(GUH) | 1 | 0 | 11 | 0 |
| PCR(NVRL) | 1 | 11 | |||
| March | RAT(GUH) | 2 | 0 | 13 | 0 |
| PCR(NVRL) | 2 | 13 | |||
| April | RAT(GUH) | 3 | 2 | 14 | 0 |
| PCR(NVRL) | 5 | 12 | |||
| May | RAT(GUH) | 1 | 0 | 18 | 0 |
| PCR(NVRL) | 1 | 18 | |||
| June | RAT(GUH) | 4 | 0 | 11 | 2 |
| PCR(NVRL) | 2 | 13 | |||
| July | RAT(GUH) | 2 | 1 | 11 | 0 |
| PCR(NVRL) | 3 | 10 | |||
| August | RAT(GUH) | 0 | 0 | 10 | 0 |
| PCR(NVRL) | 0 | 10 | |||
| Sept | RAT(GUH) | 1 | 0 | 2 | 0 |
| PCR(NVRL) | 1 | 2 | |||
| October | RAT(GUH) | 1 | 1 | 4 | 0 |
| PCR(NVRL) | 2 | 3 | |||
| Nov 1–18 | RAT(GUH) | 0 | 0 | 2 | 0 |
| PCR(NVRL) | 0 | 2 | |||
| Total stool samples tested | 143 | ||||
Table 2.
Adenovirus detection method: RAT in GUHL versus PCR testing of virus in the NVRL—sensitivity and specificity of RAT
| Total AdV Positive | Total AdV False Negative | Sensitivity of RAT (1-B) B = false negative rate | Total AdV Negative | Total AdV False Positive | Specificity of RAT (1-α) α = false positive rate) | |
|---|---|---|---|---|---|---|
| RAT GUH | 15 | 2 | 88% | 128 | 2 | 98.5% |
| PCR NVRL | 17 | 126 |
Adenovirus detection method (Tables 1, 2) 2016–2017
AdV is detected by RAT in GUHL and by PCR testing of virus in the NVRL. During the study period, 144 stool samples were tested in both GUHL and the NVRL for AdV and RV. One sample was inhibitory to AdV due to interference with enzyme marker and was excluded.
A total of 143 stool samples were valid for analysis, 15 were confirmed AdV positive in GUHL, 17 were confirmed AdV positive in the NVRL; two samples were false negative for AdV in GUHL. Sensitivity of RAT (= 1-B) (B = false negative rate) was (88%).
Whereas 128 samples were confirmed AdV negative in GUHL, 126 were confirmed AdV negative in the NVRL; two samples were false positive for AdV in GUHL. Specificity of (RAT) (1-α) (α = false positive rate) was (98.5%).
Virus detection methods for RV 2016–2017 (Tables 3, 4).
Table 3.
Rotavirus detecting method: RAT in GUHL versus PCR testing of virus in the NVRL
| Month | Test | Number of Confirmed RV Positive Cases | Number of false Negative Cases (GUHL) | Number of Confirmed RV Negative Cases | Number of false Positive Cases (GUHL) |
|---|---|---|---|---|---|
| Nov 18–30 | RAT(GUHL) | 1 | 0 | 1 | 0 |
| PCR(NVRL) | 1 | 1 | |||
| December | RAT(GUHL) | 4 | 0 | 14 | 0 |
| PCR(NVRL) | 4 | 14 | |||
| January | RAT(GUHL) | 1 | 1 | 11 | 0 |
| PCR(NVRL) | 2 | 10 | |||
| February | RAT(GUHL) | 2 | 1 | 10 | 0 |
| PCR(NVRL) | 3 | 9 | |||
| March | RAT(GUHL) | 2 | 2 | 14 | 0 |
| PCR(NVRL) | 4 | 12 | |||
| April | RAT(GUHL) | 2 | 1 | 15 | 0 |
| PCR(NVRL) | 3 | 14 | |||
| May | RAT(GUHL) | 2 | 2 | 17 | 0 |
| PCR(NVRL) | 4 | 15 | |||
| June | RAT(GUHL) | 5 | 0 | 10 | 1 |
| PCR(NVRL) | 4 | 11 | |||
| July | RAT(GUHL) | 4 | 1 | 9 | 0 |
| PCR (NVRL) | 5 | 8 | |||
| August | RAT(GUHL) | 3 | 1 | 7 | 0 |
| PCR(NVRL) | 4 | 6 | |||
| September | RAT(GUHL) | 1 | 0 | 2 | 0 |
| PCR(NVRL) | 1 | 2 | |||
| October | RAT(GUHL) | 0 | 1 | 5 | 0 |
| PCR(NVRL) | 1 | 4 | |||
| Nov 1–18 | RAT(GUHL) | 0 | 0 | 2 | 0 |
| PCR(NVRL) | 0 | 2 | |||
| Total (144) | RAT(GUHl) | 27 | 10 | 117 | 1 |
| PCR(NVRL) | 36 | 108 |
Table 4.
RV detecting method: sensitivity and specificity of RAT
| Total RV positive | Total RV false negative | Sensitivity of RAT (1-B) B = false negative rate | Total RV Negative | Total RV false positive | Specificity of RAT (1-α) α = false positive rate) | |
|---|---|---|---|---|---|---|
| RAT GUHL | 27 | 10 | 72% | 117 | 1 | 99% |
| PCR NVRL | 36 | 108 |
These data assessed RAT in GUHL Versus PCR testing of Virus in the NVRL, Sensitivity and Specificity of RAT. RV is detected by RAT in GUHL and by PCR testing in NVRL. From November 18th 2016 to November 18th 2017, 144 stool samples were tested in both GUHL and NVRL for RV.
Twenty-seven samples were confirmed RV positive in GUHL; 36 samples were confirmed RV positive in the NVRL. One sample was false positive for RV in GUHL. Specificity of RAT (1-α) (α = false positive rate) was 99%.
One hundred and seventeen samples were confirmed RV negative in GUHL; 108 samples were confirmed RV negative in the NVRL. Ten samples were false negative for RV in GUHL. Sensitivity of RAT (1-B) (B = false negative rate) was 72%.
Discussion
Our study demonstrated specificity of RAT was 99% in detecting patients who truly did not contract RV GE and 98.5% in detecting those who truly did not contract AdV GE.
Sensitivity of RAT in detecting those children who truly contracted RV or AdV GE were significantly lower (72%, 88% respectively). RAT is ratified as more specific than sensitive in detecting RV [4]. Our study demonstrated that RAT—as RV and AdV detecting tool- was more specific than sensitive (higher specificity and lesser sensitivity) as shown in other similar studies [4–6]. We conclude that such a high specificity of RAT makes it useful and reliable in mass screening for group A RV and AdV during outbreaks of food-borne and person-to-person transmitted GE where resources can be limited due to increasing health care utilization costs at times of outbreaks. We acknowledge the small sample size in our study. RAT does not require special equipments and can be useful for screening of patients when there is widespread of GE [5].
PCR is more sensitive and specific than RAT [1–3]. Our study has shown similar results.
Although the rapid antigen testing is able to quickly give results, our study found that it has a high false negative rate for both RV (72%) and AdV (88%). Thus, other highly sensitive methods such as (RT)-PCR for RV and PCR testing for AdV are still required for true diagnosis, in support of the findings of other studies worldwide [5, 6].
Future research is recommended to continue to monitor sensitivity and specificity of RAT.
We acknowledge that our original research has not provided novel findings, however, results of research of such type should also be made available to the public in an attempt to avoid “research bias” by focusing only on publishing novel data. At times of pandemics, the public need be reassured that they really do not have a disease by using a rapid, easy and accurate test such as RAT.
Funding
None.
Compliance with ethical standards
Conflict of interest
The author declares that they have no conflict of interest.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Noppornpanth S, Poovorawan Y. Comparison between RT-PCR and rapid agglutination test for diagnosis of human rotavirus infection. Southeast Asian J Trop Med Public Health. 1999;30(4):707–709. [PubMed] [Google Scholar]
- 2.Pang XL, Lee B, Boroumand N, Leblanc B, Preiksaitis JK, Yu Ip CC. Increased detection of rotavirus using a real time reverse transcription-polymerase chain reaction (RT-PCR) assay in stool specimens from children with diarrhea. J Med Virol. 2004;72(3):496–501. doi: 10.1002/jmv.20009. [DOI] [PubMed] [Google Scholar]
- 3.Jothikumar N, Kang G, Hil VR. Broadly reactive TaqMan assay for real-time RT-PCR detection of rotavirus in clinical and environmental samples. J Virol Methods. 2009;155(2):126–131. doi: 10.1016/j.jviromet.2008.09.025. [DOI] [PubMed] [Google Scholar]
- 4.Theron EMC, Nyaga MM, Dewar JB. Ratification of rapid rotavirus diagnostic test strips. South Afr J Infect Dis. 2014;29(2):91–94. [Google Scholar]
- 5.Khamrin P, Tran DN, Chan-it W, Thongprachum A, Okitsu S, Maneekarn N, Ushijima H. Comparison of the rapid methods for screening of group a rotavirus in stool samples. J Trop Pediatr. 2011;57(5):375–377. doi: 10.1093/tropej/fmq101. [DOI] [PubMed] [Google Scholar]
- 6.Chieochansin T, Vutithanachot V, Theamboonlers A, Poovorawan Y. Evaluation of the rapid test for human rotavirus A in Thai children with acute gastroenteritis. Clin Lab. 2014;60(3):511–514. doi: 10.7754/clin.lab.2013.130616. [DOI] [PubMed] [Google Scholar]
- 7.Tiemessen CT, Nel MJ. Detection and typing of subgroup F adenoviruses using the polymerase chain reaction. J Virol Methods. 1996;59(1–2):73–82. doi: 10.1016/0166-0934(96)02015-0. [DOI] [PubMed] [Google Scholar]
- 8.Kageyama T, Kojima S, Shinohara M, et al. Broadly reactive and highly sensitive assay for Norwalk-like viruses based on real time quantitative reverse transcription-PCR. J Clin Microbiol. 2003;41(4):1548–1557. doi: 10.1128/JCM.41.4.1548-1557.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Oka T, Katayama K, Hansman GS, et al. Detection of human sapovirus by real-time reverse transcription-polymerase chain reaction. J Med Virol. 2006;78(10):1347–1353. doi: 10.1002/jmv.20699. [DOI] [PubMed] [Google Scholar]
- 10.Freeman MM, Kerin T, Hull J, et al. Enhancement of detection and quantification of rotavirus in stool using a modified real time RT-PCR assay. J Med Virol. 2008;80(8):1489–1496. doi: 10.1002/jmv.21228. [DOI] [PubMed] [Google Scholar]
- 11.Logan C, O’Leary JJ, OSullivan N. Real-time reverse transcription PCR detection of norovirus, sapovirus and astrovirus as causative agents of acute viral gastroenteritis. J Virol Methods. 2007;146(1–2):36–44. doi: 10.1016/j.jviromet.2007.05.031. [DOI] [PubMed] [Google Scholar]
- 12.Gautam R, et al. One-step multiplex real-time RT-PCR assay for detecting and genotyping wild-type group A rotavirus strains and vaccine strains (Rotarix(R) and RotaTeq(R) in stool samples. PeerJ. 2016;4:e1560. doi: 10.7717/peerj.1560. [DOI] [PMC free article] [PubMed] [Google Scholar]