Comparative evaluation of sensitivity and specificity of immunochromatography kit for the rapid detection of norovirus and rotavirus in Bangladesh

We report a comprehensive analysis of sensitivity and specificity of immunochromatography kit (IC Kit) for the rapid detection of norovirus and rotavirus in Bangladesh. The IC kit (IP-Noro/Rota) provides highest sensitivity (100%) to both viruses compared to the reference method reverse transcriptionpolymerase chain reaction (RTPCR) for diagnosis. Furthermore, the test provides a high specificity of 98.9% and 96.1% to diagnose norovirus and rotavirus, respectively, as well as good agreement with the reference method. We also found high prevalence of rotavirus infection (74%) among Bangladeshi pediatric population, of which most of the patients were less than five years old, suffering from severe dehydration, abdominal pain and vomiting. This study is the first to report the ease and rapid detection of norovirus and rotavirus by IC kits in Bangladesh. Therefore, IP-Noro/Rota kit is recommended for the rapid detection of these viruses in routine diagnosis as well as during outbreaks.


Introduction
Diarrheal diseases represent a major worldwide public health problem, particularly in developing countries 1 . Acute gastroenteritis is a very common disease in young children 2 . It has been reported that about 3-5 billion cases of acute gastroenteritis occur each year in children less than 5 years old and 1.5 to 2.5million children of that group die from severe diarrhoea [3][4][5] . Of that, about half amillion death is caused by rotavirus infections 6 . On the other hand, norovirus is responsible for almost half of the foodborne gastroenteritis outbreaks and 75-90% of non-bacterial gastroenteritis outbreaks 7 .
When outbreaks of gastroenteritis occur in communities, rapid identification of pathogens is essential to ensure the administration of the appropriate treatment and control. Furthermore, definite diagnosis plays an important role to decrease the unnecessary use of antibiotics. In the case of emergency, there is no rapid detection method available in Bangladesh. In this regard, a rapid diagnosis kit with good sensitivity and specificity is essential. Developing such a kit may raise the reliability for rapid diagnosis in developing countries, where the prevalence of norovirus and rotavirus is increased. Herein, we report a comprehensive analysis of the sensitivity and specificity of an immunochromatography kit (IC Kit) for the rapid detection of norovirus and rotavirus in Bangladesh.

Participants
In this study, we evaluate the newly developed IC test kit for norovirus and rotavirus detection (IP-Noro/Rota; ImmunoProbe Co., Ltd., Saitama, Japan) in 100 stool samples collected from pediatric patients with acute gastroenteritis (severe dehydration, abdominal pain and vomiting) in Bangladesh during January to June 2015. The study was ethically approved by the ethical review committee of Jahangirnagar University, Bangladesh.

Test methods
Reverse transcription PCR (RT-PCR) was used as the reference test for both norovirus and rotavirus detection 3 . The PCR is a molecular biology technique that allows for nucleic acid fragment from a complex pool of DNA. Faecal specimens were thawed, diluted with distilled water to 10% suspensions, and centrifuged at 10,000xg for 10 min. Viral RNA was extracted from 140µl of the supernatant using a spin-column technique (QIAamp Viral RNA kit; Qiagen, Hilden, Germany) according to the manufacturer's instructions. For reverse transcription, 3µl of extracted RNA was mixed with a reaction mixture consisting of 1µl of oligo dT primer (Promega, Madison, USA) and 1µl of nuclease free water in microcentrifuge tube, then kept at 70°C for 5 mins and then chill for 5 mins. After that 4µl of 5X reaction buffer (Promega, Madison, USA), 2µl of MgCl 2 , 1µl of PCR Nucleotide Mix (Promega, Madison, USA), 0.5µl of Ribonuclease Inhibitor (Promega, Madison, USA), 1µl of Reverse Transcriptase (Promega, Madison, USA), 6.5µl of nuclease free water were mixed with the same microcentrifuge tube. Then the solution was heated at 25°C for 5 mins, 42°C for 60 min and 70°C for 15 mins. Norovirus and rotavirus were detected by PCR analysis of cDNA with specific primers previously published 8-9 . The amplification was carried out in a thermal cycler (2720 Thermal Cycler, Applied Biosystems, USA). The PCR was performed at 94°C for 3 mins followed by 35 cycles of 94°C for 30 s, 55°C for 30 s, 72°C for 60 s and a final extension at 72°C and then held at 4°C. The PCR products were electrophoresed in a 1.5% agarose gel, followed by staining with ethidium bromide (0.5 g/ml) for 20 min and then visualized under ultraviolet (UV) light. The bands were recorded by photography.
To evaluate the sensitivity and specificity of this IP-Noro/Rota test kits, all the 100 samples were tested for norovirus and rotavirus antigens by this kit following manufacturer's instructions (IP-Noro/Rota; ImmunoProbe Co., Ltd., Saitama, Japan). It took only 10-15 min to obtain the result. A positive result for both pathogens is two lines; the left control line (C) and the right test-positive line (T), whereas, a negative result consisted of a single left control line (C) (Figure 1).

Analysis
The sensitivity and specificity of IP-Rota/Noro test kit were calculated mathematically as described below: Sensitivity for IC kit = Both IC kit and RT-PCR Positive × 100/ RT-PCR positive

Results
The working plan for evaluation of sensitivity and specificity of immunochromatography methods for rapid detection of rotavirus and norovirus associated with paediatric diarrhoea in Bangladesh is described in Figure 2.
By the RT-PCR method, 10 and 74 samples were confirmed as norovirus and rotavirus, respectively. It was found that all the isolated norovirus belongs to the genogroup II (data not shown). On the other hand, G1P 8 rotavirus strain was found the most prevalent among the Bangladeshi pediatric population after characterization of G-types (VP-7) and P-types (VP-4) of rotavirus-positive samples. The youngest patient was 21 days and the oldest 56 months; the average age was 14 months. The most common clinical symptoms of rotavirus and norovirus infected patients were dehydration, vomiting, fever and abdominal pain.
Of the 10 and 74 samples positive for norovirus and rotavirus, respectively, by RT-PCR, IP-Noro/Rota kit recognized all positive samples with 100% sensitivity. However, the kit gave one false positives for norovirus and three false positives for rotavirus detection, resulting in a specificity of 98.9% and 96.1%, respectively (Table 1 and Table 2).

Conclusions
The clinical symptoms of the patients with acute gastroenteritis are generally not indicative of a specific pathogen. In Bangladesh, the outbreak of norovirus and rotavirus diarrhea occurs mainly in the winter season 3 , when the IC kits could be used for rapid screening, as other existing diagnosis methods are time consuming. The rapid IC kit test is easy to perform at a low cost and it takes only 10-15 min to diagnose with a simple procedure and does not require special equipment or a skilled technician.
Our findings clearly indicate that rotavirus and norovirus are the most important enteropathogen responsible for acute viral gastroenteritis among infants and children in Bangladesh, where 74% of the cases were caused by rotavirus only. The IC kit provides a high specificity and sensitivity as well as good agreement with the reference method, RT-PCR, for the detection of rotavirus and norovirus. Therefore, IC-Noro/Rota kit will be easy and useful assay for the rapid detection of these viruses in routine diagnosis as well as during the outbreaks. This is the first report about the rapid detection of rotavirus by IC kits in Bangladesh. Finally, it is strongly recommended to use the IC kit as an alternative method for rapid diagnosis of norovirus and rotavirus infections, especially in developing countries like Bangladesh. The paper reports the performance of an immunochromatography cartridge for the rapid detection of norovirus and rotavirus against an RT-PCR reference method. The specimens selected (n=100) were collected between January and June 2015 in Bangladesh and had been frozen prior to use. The authors also calculated the sensitivity and specificity of the kit and made recommendations regarding the potential use. However, the paper has a few problems:

Data availability
1. The calculations in Table 1 and 2 are incorrect. The sum of the rows and columns do not total 100. Consider correcting the tables an indicated below. The calculations for specificity are, therefore, incorrect. This should be correct in the next version of the paper.

The authors briefly mention the broad typing results for norovirus (genogroup II) and rotavirus (G1P[8]
) but fail to give any indication of the methods used to obtain these results. It makes it difficult for the reviewers and readers to assess the results presented. In addition, no detailed genotyping results were presented for the viral strains detected. Most current papers evaluating diagnostic assays for enteric viruses include genotyping data for the virus strains detected -see references below.
3. The authors failed to report the manufacturers' sensitivity and specificity for the kit or to compare their results to other studies that evaluated the same or similar kits -see articles below. Alternatively, if this is a new formulation of the kit being evaluated the authors should indicate this fact.
4. The authors utilized stool specimens that were frozen for an unspecified period (but approximately three years if the study was completed in 2018). The prolonged storage or repeated freeze/thawing could have contributed to a degree of degradation of the viruses in the stool material and should have been listed as a potential limitation of the study. However, the screening of the frozen stool specimens is a previously utilized approach for evaluating diagnostic kits.
5. If the kit is introduced according to the author's recommendations, it is more likely that fresh stool specimens will be screened. Therefore, the authors should consider evaluating the kit using fresh stool specimens from routine surveillance or during an outbreak which would mimic real world use and provide better estimates of sensitivity and specificity.
6. Based on the detection rate of rotavirus in the stool specimens selected for the evaluation (74%), these specimens were collected during the rotavirus season in Bangladesh. While the results indicate that the kit would be adequate for use during the rotavirus season (barring a few false positive results), it does not give any indication of the kits performance when the rotavirus prevalence declines, outside of the rotavirus season.

If applicable, is the statistical analysis and its interpretation appropriate? No
Are all the source data underlying the results available to ensure full reproducibility? Partly

Are the conclusions drawn adequately supported by the results? Partly
The laboratory analysis was conducted at the time of collection of the samples. After analysis of the samples, the leftover is stored for long time and is still available in the lab.

4.
The evaluation of the kit was tested using the fresh samples. 5.
The kit gave a specificity of 100% as compared to the RT-PCR analysis, though some false positive results may have been given. The statement is updated in the revised manuscript.

6.
Additional corrections--In all the paragraphs, the term "developing countries" is replaced by "low-income countries" in the revised manuscript.
-The nomenclature of the rotavirus genotypes is corrected.
-The statement is updated in the revised manuscript. specificity of the kit for rotavirus detection (in Table 2) should be 89.6%. 4. Concluding remarks: The statement "Our findings clearly indicate that rotavirus and norovirus are the most important enteropathogen responsible for acute viral gastroenteritis among infants and children in Bangladesh, where 74% of the cases were caused by rotavirus only." cannot be made from the data presented in this paper. This study was particularly designed to evaluate a kit; not for describing rotavirus and norovirus prevalence.