Introduction

Rotavirus is a leading cause of acute viral gastroenteritis throughout the world, and most children are infected by 5 years of age (1). Many studies conducted in both developing and developed countries report that group A rotaviruses are responsible for 13-50% of all cases of viral gastroenteritis in children under 5 years of age (2). Despite the availability of a rotavirus vaccine, more than 200,000 deaths occur per year under the age of five worldwide (1). In studies conducted in Turkey, rates of rotavirus positivity in children with gastroenteritis ranged from 18.7% to 53% (3-10).

The rapid and accurate diagnostic tests for the detection of the virus in patients with acute gastroenteritis are important not only for the diagnosis of viral gastroenteritis but also to prevent the spread of the disease (11). The specific methods available for detecting the rotavirus in stool specimens include electron microscopy (EM), immuno-EM, cultivation techniques, rapid antigen tests, polyacrylamide gel electrophoresis (PAGE), and reverse transcriptase-polymerase chain reaction (RT-PCR). EM, cultivation, and PAGE are not recommended because they are expensive, time-consuming, and technically difficult (4). Several rapid antigen tests, such as the latex agglutination test (LAT), the immunochromatographic test (ICT), and the enzyme-linked immunosorbent assay (ELISA), are inexpensive, easy-to-perform, and commercially available.

In routine diagnostics, it is important to know the sensitivity and specificity of these rapid tests. Rapid and accurate diagnosis could prevent unnecessary and potentially harmful antibiotic treatment and improve knowledge of the epidemiology of rotavirus infections (12). These rapid tests have good performance for determining rotavirus, and they are frequently used by physicians as an aid to diagnosis. The RT-PCR is the most sensitive molecular method to detect and confirm rotavirus (13,14).

The objective of the presented study is to evaluate the performance, sensitivity, and specificity of three commercially available rotavirus antigen tests: ELISA, ICT, and LAT, compared with a reference test, a multiplex reverse transcription-PCR (mRT-PCR), for detecting rotavirus in fecal specimens from patients with acute gastroenteritis over medical records retrospectively.

Materials and Methods

In this study, the results of three different commercial rotavirus antigen tests, ELISA, ICT, and LAT, which were studied to detect rotavirus in stocked stool samples sent to our laboratory from patients with acute gastroenteritis between January and December 2014, were compared with the results of the RT-PCR, which was used as a reference test. The study was approved by the Aydın Adnan Menderes University Ethics Committee (protocol no: 2022/176, date: 10/11/2022).

Detection of rotavirus by ELISA: The ProSpecTTM® Rotavirus Microplate Assay (Oxoid, Ltd., Basingstoke, Hampshire, UK) is a qualitative sandwich ELISA and was used to detect rotavirus group A antigen in stool samples.

Detection of rotavirus by ICT: The CerTest® Rota-Adeno Card Test (CerTest, Biotec, Spain) was used to detect rotavirus antigen in stool samples. This test is a one-step lateral flow ICT that simultaneously detects group A rotavirus and adenovirus in stool samples.

Detection of rotavirus by LAT: The Virotect Rota (Omega Diagnostics, Scotland, UK) wasbasically a rapid LAT for the detection of rotavirus in fecal samples.

Multiplex RT-PCR: The commercially available Seeplex® Diarrhea ACE Detection multiplex PCR (Seegene, Seoul, Korea) was used to simultaneously detect group A rotaviruses, AdV 40 and 41 (species F), noroviruses GI and GII, and astroviruses. Nucleic acids were extracted from fecal suspensions by using the QIAamp DNA Mini kit (Qiagen, Hilden, Germany) and the QIAcube platform (Qiagen). The nucleic acid was amplified using a PTC-200 thermal cycler (Bio-Rad, Hercules, CA, USA), and the PCR products were visualized after electrophoresis on a 1.5% agarose gel. Under ultraviolet light, the DNA products with 650 base pairs (bp) for Astrovirus, 411 bp for enteric adenovirus, 541 bp for group A rotavirus, 304 bp for norovirus group I, and 214 bp for norovirus group II showed positive results. All test kits were performed according to the manufacturer’s instructions.

Statistical Analysis

Descriptive statistics were presented as percentages, the mean with standard deviation, or the median with minimum and maximum values. Based on the RT-PCR results (15), sensitivity, specificity, and positive and negative predictive values were calculated for the ICT, LAT, and ELISA. Test agreement with the RT-PCR results accepted as the gold standard was assessed using Cohen’s kappa with 95% confidence intervals (CI). The strength of agreement was based on Cohen’s kappa value: <0 poor, 0.0-0.20 slight, 0.21-0.40 fair, 0.41-0.60 moderate, 0.61-0.80 substantial, and 0.81-1.00 almost perfect (16). All data entries were made in SPSS 26.0, and the tables were prepared with it. Calculations were performed and evaluated according to the references mentioned before.

Results

In the study, which included 85 people in total, 25 (29.4%) of them were under 1-year-old and 43 (50.6%) were under 5-years-old. The median age was two years (0-32), and 41 were female and 44 were male.

Out of the 85 samples, 56 (65.9 %) tested positive for rotavirus by RT-PCR. Out of all the samples, 63 (74.1%), 53 (62.4%), and 52 (61.2%) were found to be positive in the ELISA, ICT, and LAT kits, respectively.

In Table 1, it was shown that the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated using the RT-PCR as a gold standard. These values for the ICT test were found to be 78.6%, 69.0%, 83.0%, and 62.5%, respectively. These values were 78.6%, 72.4%, 84.6%, and 63.6% for the LAT and 96.4%, 69.0%, 85.7%, and 90.9% for the ELISA, respectively.

Considering Cohen’s kappa, moderate agreement was found between the RT-PCR and the ICT (k=0.464 (95% CI, 0.268-0.660), p<0.001); moderate agreement was found between the RT-PCR and the LAT (k=0.493 (95% CI, 0.303-0.683), p<0.001); and substantial agreement was found between the RT-PCR and the ELISA (k=0.694 (95% CI, 0.529-0.859), p<0.001) (Table 1).

Discussion

Rapid and accurate rotavirus detection is required to ensure the administration of appropriate treatment plans and infection control. Several rapid test kits, including latex agglutination, ICTs, and enzyme immunoassays for detection of rotavirus infection, are used in routine diagnosis. In our study, we evaluated the performance, sensitivity, and specificity of three commercial rotavirus antigen tests compared with a mRT-PCR assay for detecting rotavirus in fecal specimens from patients with acute diarrhea.

In the presented study, RT-PCR was accepted as the reference method, the ELISA was the antigen test with the highest sensitivity (96.4%), and the LAT was the test with the highest specificity (72.4%) among the three tests compared. When PPV and NPV were examined, it was observed that the ELISA had the highest values (85.7% and 90.9%, respectively). The test with the highest agreement with the RT-PCR was found to be ELISA (k=0.694). Considering our findings, it can be thought that the performances of all three tests are not at the desired level; nevertheless, ELISA is the best among these tests.

The ICT is rapid and easy to perform (17). We found that out of 85 samples, 53 (62.4%) were positive in the ICT. The sensitivity and specificity of ICT were 78% and 69%, respectively. Studies comparing the ICT test with the reference method, the PCR test, showed that the sensitivity of the ICT test was 80-100% and the specificity was 89-100%. In the presented study, the sensitivity and specificity of the ICT test were lower than in previous studies. The latex agglutination assay is faster and simpler but less sensitive than the ELISA (17). In our study, while the LAT test was similar to the ICT, it was less sensitive and specific than the ELISA test. In a previous study, it was reported that the latex agglutination assay had a sensitivity of 63.6% and a specificity of 86.8% when compared with the RT-PCR (18). Xiang et al. (19) found 81.03% and 97.44% for the LAT sensitivity and specificity, respectively. Unlike previous studies, we found 78.6% sensitivity and 72.4% specificity for the LAT.

Although the ELISA is the standard test for detecting rotaviruses, it is time-consuming and not cost-effective. In this study, 63 (74.1%) samples were positive for rotavirus by the ELISA and the sensitivity and specificity of the ELISA were 96.4% and 69%, respectively, when compared with the RT-PCR. Gautam et al. (20) conducted a comparative analysis study of three commercial EIA kits. Using the RT-PCR as the gold standard, the sensitivities were between 75 and 82.1% and all the specificities were found to be 100%. Ibrahim et al. (13) observed that the ELISA had a good performance with 88% sensitivity and 100% specificity when compared with the RT-PCR results. Considering our study, the ELISA test was more sensitive than the the other ELISA tests mentioned but less specific.

In a study in which 95 stool samples were studied for LAT, three ICTs, and ELISA tests and compared with the RT-PCR, it was reported that the sensitivity and specificity were 85.7% and 100% for LAT, 100% and 95% for two ICTs, 86.7% and 87.5% for another ICT, and 98.1% and 97.3% for the ELISA, respectively (21). We found the sensitivity of the ELISA to be higher than LAT and ICT, and the specificity to be lower. In another study comparing rapid tests, three different commercial immunologic tests for rapid detection of group A rotavirus (the ICT method, LAT, and ELISA) were used to evaluate 228 stool specimens obtained from children with acute gastroenteritis. The sensitivity and specificity values of the ELISA, ICT, and LAT methods were 96% and 68%, 99% and 99%, 99%, and 96%, respectively. In the study conducted by Wilhelmi et al. (22), the sensitivity and specificity for the ELISA and ICT were found to be higher than LAT.

Agreement with the RT-PCR for the ELISA was the highest compared with the ICT and LAT. In a previous study (23), the analytical and clinical performance of ICT was investigated, and the test results were compared to the ELISA and RT-PCR, found a high level of agreement (k=0.857). Although there was lower agreement, we also found the highest agreement with the RT-PCR in our study with the ELISA.

The LAT and ICT were rapid and easy to perform but showed lower sensitivity than the ELISA. The ELISA was the best test in terms of sensitivity and specificity but had limitations, such as generating results that were difficult to interpret and being time-consuming. Because of the nature of convenience sampling, the results of this study are not generalizable. However, similar results have been obtained in many studies. In this study, the samples were not fresh; frozen samples were thawed and studied.

Conclusion

In summary, the ELISA test for rotavirus detection showed the highest sensitivity and a high level of agreement with the RT-PCR. Even though it is a less sensitive test, the ICT and LAT may be used alternatively for the rapid screening of group A rotavirus in stool samples, especially during the acute gastroenteritis outbreak seasons. In clinical practice, the possibility of false positive and false negative results with rotavirus should be kept in mind. A false positive antigen test could be caused by cross reactivity with other microorganisms or interference. When testing with rapid kits yields a negative result in suspected patients, testing with RT-PCR should not be delayed.

Ethics

Ethics Committee Approval: The study was approved by the Aydın Adnan Menderes University Ethics Committee (protocol no: 2022/176, date: 10/11/2022).

Informed Consent: Retrospective study.

Peer-review: Externally peer-reviewed.

Authorship Contributions

Concept: S.K., N.E., Design: S.K., N.E., Data Collection or Processing: S.K., N.E., F.K., V.Y., M.A., Analysis or Interpretation: S.K., H.Ö., Literature Search: S.K., N.E., H.Ö., M.A., Writing: S.K., N.E., H.Ö.

Conflict of Interest: No conflict of interest was declared by the authors.

Financial Disclosure: The authors declared that this study received no financial support.