Clinical manifestation of norovirus infection in children aged less than five years old admitted with acute diarrhea in Surabaya, Indonesia: a cross-sectional study

Background: The objective of this study was to investigate the clinical manifestation of norovirus infection between norovirus genogroup and severity of acute diarrhea in pediatric patients at the Dr. Soetomo Hospital, Surabaya, Indonesia. Methods: This cross-sectional study involved 31 participants aged 1-60 months admitted to the hospital with acute diarrhea from April 2012 to March 2013. Norovirus genogroups (GI and II) were identified from patient stool using reverse transcription polymerase chain reaction (RT-PCR). Severity was measured using the Ruuska and Vesikari scoring system. Results: In total, 94 stool samples were obtained, of which 31 (19%) were norovirus positive. Norovirus GI was found in one sample with mild diarrhea. Norovirus GII was found in 30 samples (96.8%); one sample with mild diarrhea (3.3%), 20 samples with moderate diarrhea (66.7%), and nine samples with severe diarrhea (30%). Conclusion: Norovirus GII was the most prevalent cause of acute diarrhea and 30% of the cases manifested as severe diarrhea.


Introduction
Diarrhea is considered the second-leading cause of death in pediatric patients under the age of five with a worldwide annual mortality of 525,000 children. Diarrhea lasting for even a few days causes dehydration 1 . Viruses from the genus Norovirus from the family Caliciviridae are the second-leading cause of acute diarrhea after rotavirus in all age groups of pediatric patients [2][3][4] . Norovirus is responsible for 218,000 pediatric deaths (<15 years old) every year and for 1.1 million pediatric hospitalizations around the world 2 . In Indonesia, previous studies mentioned incidence rate around 17-30% in children 5-8 .
Early identification of norovirus strain and genotype is vital for predicting the development of the disease and selecting the most suitable treatment. Genogroup diversity can be checked using reverse transcriptase polymerase chain reaction (RT-PCR). Norovirus is grouped into 40 viral strains, which are further classified into five different genogroups. Among them, GI and GII possess the most diverse genetic components 9 . From the previous study, Norovirus GII.2 genotypes had been the most prevalent norovirus strain in Indonesia (71.4%) followed by norovirus GII.17 (14.3%), one case was GII.4 and one case was GII.1 10 .
Norovirus commonly causes mild and short-term diarrheal episodes 10 . Nonetheless, this virus can be fatal, particularly in pediatric, geriatric, and immunocompromised patients 11,12 . Norovirus patients showed more severe diarrhea compared to those without norovirus infection in pediatric patients. The type of norovirus strain and genome is thought to be related to diarrhea severity 12 .
This study aimed to examine the clinical manifestation of norovirus infection in pediatric patients aged 1-60 months in the Dr. Soetomo General Hospital, Surabaya, Indonesia.

Ethical statement
The study protocol was approved by the Ethical Research Commission of Dr. Soetomo General Hospital, Surabaya, and conducted in line with the 1964 Helsinki declaration and its later amendments or research code of ethic issued by the Ministry of Research, Technology and Higher education. Written informed consent regarding participation in this study, the right to resign, stool and data collection and confidentiality of patient data was obtained and signed from all individuals' parents. Consent was requested from the patients' parents because the patients were 1-60 months old.

Study population
This cross-sectional study was conducted between April 2012 and March 2013 of all children aged 1-60 months old with acute diarrhea (described as defecation more than three times per day with change of stool consistency to loose or watery) admitted to the pediatric ward. Patients with a gastrointestinalanatomical disorder such as Hirschsprung disease, severe systemic disease including sepsis, central nervous system infection or bronchopneumonia, a malabsorption disorder such as cow's milk allergy, or a compromised immune status were excluded from the study to avoid any bias. On the day of the patients' admission to the pediatric ward, parents were asked to participate in this study, and they agreed by signing the informed consent form. Stool samples were collected within 24 hours of patient admission with a sterile pot; approximately 3g of stool sample was taken from the middle part of the stool and delivered in no longer than three hours to the laboratory institution. Using a total sampling method, all samples collected until the end of March 2013 were studied.

Patient assessment
All subjects underwent physical examination and the participant's parents completed a questionnaire.
The patient assessment was carried out by the physicians. The patient's parents completed questions in the questionnaire regarding characteristic patient data 13 . These data were: patient's identity (age, gender, body weight, and body height); parent's identity (maternal education); history of diarrhea, which were divided into diarrhea duration (≤4 days, 5 days, and ≥6 days) and diarrhea frequency within 24 hours (1-3 times/day, 4-5 times/day, and ≥6 times/day); vomiting history, divided into vomiting duration (no vomiting, 1 day, 2 days, and more than 3 days) and vomiting frequency within 24 hours (no vomiting, 1 time/day, 2-4 times/day, and more than 5 times/day); and history of breastfeeding (not received breastfeeding, breastfeeding <6 months, and breastfeeding ≥6 months). Nutritional status was classified to either normal or malnutrition (underweight, stunted, wasted, and overweight) according to the definition by WHO 14 .
All patients also underwent physical examination of axillary body temperature (°C), arterial pulse measured with a pulseoxymeter (times/minute), respiratory rate (times/minute) and inspection of the signs of dehydration based on WHO classification 15 and all results were written down in the questionnaire form. The questionnaire was then reviewed by the researchers and entered into the research database.

Norovirus diagnosis
Stool samples were delivered to the laboratory institution and kept in a deep freezer at -80°C until they were thawed at room temperature prior to RT-PCR analysis. To prevent laboratory contamination, our laboratory staff wore complete apparatuses, such as mask, coat, and gloves, throughout the process. RNA extraction was conducted in Bio Safety Cabinet. Before conducting PCR, all containers were disinfected using alcohol. A 10% stool suspension was prepared for each sample prior to RNA extraction by mixing 100µl stool sample with 100µl phosphate buffered saline

Amendments from Version 2
We have revised the title because our number of patients is so small that we could not evaluate the correlation between G1 positive and severity of diarrheae.
Any further responses from the reviewers can be found at the end of the article buffer (Sigma-Aldrich, St. Louis, USA) with a vortex mixer (QL System, MX-2500 Vortex Mixer, UK) for 15 seconds and then centrifuging at 13,000-15,000 rpm for 10 minutes (Microfuge 20, Beckman Coulter, Indiana Polis, USA). The supernatant (1µl) from the stool suspension was transferred into a clean testtube and the Viral Nucleic Acid Extraction Kit II (Cat # VR100, Geneaid Biotech Ltd., New Taipei, Taiwan) was used to extract viral RNA, carried out according to the kit manufacturer's instructions. The eluted RNA from the samples was then stored in a deep freezer at -80°C until RT-PCR processing.
Reverse transcription was performed by mixing 75 picomoles of pdN6 random hexamers (Cat # 11034731001, Roche Molecular Biochemicals, Germany), 4U AMV Reverse Transcriptase (Cat # AMS.AMV007-1, AMS Biotechnology, Abingdon, UK) and 5µl of the eluted RNA and incubating at 42°C for 60 minutes. Approximately 10µl of the previous mixture was added to 5µl distilled water, 3µl Ex Taq DNA Polymerase (RR001B, Takara Bio Inc., Kusatsu, Japan) and 2µl of both forward and reverse primer. The primer pair used in this study for G1 were the G1SKF primer with nucleotide chain CTGCCCGAATTYGTAAATGA targeting nucleotide position 5342-5362 and a product size of 329 bp, and the G1SKR primer, with nucleotide chain CCAAC-CCARCCATTRTACA targeting nucleotide position 5652-5671 and a product size of 329 bp. The primer pair used in this study for G2 were the G2SKF primer, with nucleotide chain CNTGGGAGGGCGATCGCAA targeting nucleotide position 5058-5077 and a product size of 344 bp, and G2SKR, with nucleotide chain CCRCCNGCATRHCCRTTRTACAT targeting nucleotide position 5378-5401 and a product size of 344 bp.
PCR reaction was performed as follows. Initial denaturation was done at 94°C for 7 minutes, followed by 40 amplification cycles with Takara PCR Thermal Cycler Dice (TP600, Takara Bio Inc.). Each cycle consisted of denaturation at 94°C for 30 seconds, primer annealing at 50°C for 30 seconds for G1 or 57°C for 30 seconds for G2, extension reaction at 72°C for 45 seconds, followed by a final extension for 2 hours 24 minutes. The PCR product was then separated via gel electrophoresis in a 2% agarose gel and visualized under the UV light after ethidium bromide staining. The gel patterns were captured with Printgraph Fx Series (AE-6933FXN, Atto Corporation, Tokyo, Japan). The RT-PCR method used in this study was the one used by Rasanen et al. in Finland for identifying norovirus 16 , which can reveal the genotype variety via nonstructural proteins within the virus 17 . RT-PCR is considered to have the highest sensitivity for diagnosing norovirus infection compared to other methods 18 .

Vesikari Scoring System
The severity of diarrhea was measured using the Vesikari Scoring System (see Table 1). This severity scale was originally developed to evaluate the effectiveness of rotavirus vaccines based on 20 points 19 . The used parameters have been tested for reliability and validity in a cohort study conducted by Freedman with Cronbach's α > 0.7. Diarrhea severity was assessed by evaluating seven clinical symptoms, including the duration of diarrhea, diarrhea frequency within 24 hours, vomiting duration, vomiting frequency within 24 hours, body temperature, dehydration status, and treatment. From those components, we could use the modified Vesikari score 20 to assess diarrhea severity. Mild diarrhea is equal to a score of < 7; a score of 7-10 is equivalent to moderate manifestation, and severe manifestation scores > 10.

Statistical analysis
Descriptive analysis was used to determine proportions from patients' and parents' identity data (age, gender, nutritional status and maternal education variables) and clinical patient data (diarrhea type, diarrhea duration, diarrhea frequency, vomiting duration and frequency, temperature, dehydration status and  causative pathogen). The results of basic and clinical data are presented in tables and divided based on the PCR norovirus result (positive norovirus group and negative norovirus group). The basic characteristics of all patients participated in this study are presented in Table 2. Most of the participants whose samples were positive for norovirus were male (54.8%), the youngest participant was one month old and the oldest was 24 months old. Twenty-two participants (71%) were between 6-23 months old. Most patients in negative norovirus group were within 6-23 months old (74.6%) and were male (65.1%). Differences were found in the nutritional status of norovirus negative patients, in whom malnutrition was more prevalent (77.8%) than in norovirus positive patients. Breastfeeding for less than six months was also more common in the norovirus negative group (74.6%).

Participant characteristics
Clinical characteristic data are presented in Table 3. On average, the subjects were brought to the hospital after suffering diarrhea for two days with a frequency of diarrhea of five times within 24 hours. Other symptoms experienced by the subjects included vomiting (71% in positive norovirus group and 63% in negative norovirus group) with the most frequent duration of vomiting being one day (14.9% in positive norovirus group and 40.4% in negative norovirus group) and the most commonly observed frequency of vomiting being 2-4 times (9.6% in positive norovirus group and 21.3% in negative norovirus group) per day.
The most frequent body temperature on admission to the hospital was below 37°C for positive norovirus group (48.4%) and sub-febrile (37.1-38.4°C) for the negative norovirus group (50.8%). Dehydration status in this study showed that two of the patients suffered from severe dehydration in both groups, while no dehydration was found only in negative norovirus group (3.2%). Watery stool diarrhea was the most frequent type of diarrhea in both positive norovirus group and negative norovirus group (80.6% and 50.4%, respectively). Bloody or mucoid stool was found only in patients of the negative norovirus group (both 9.5%).

Diarrhea severity
Based on norovirus genogroup identification from gel electrophoresis, GI was found in one sample and GII in 30 samples (96.8%). No products other than norovirus was found.
Distribution between norovirus genogroups and degree of diarrhea severity is presented in Table 4 and shows that norovirus GI was only responsible for one case of diarrhea with moderate severity. From 30 samples that tested positive for norovirus GII, GII was responsible for 20 cases (66.7%) of diarrhea with moderate severity, and nine cases (30%) of diarrhea with severe manifestation. Norovirus infection, in our study, is most prevalent in 6-23 months population. Similar to other studies, this finding might be due to protection from maternal antibodies during breastfeeding for infant of < 6 months old. After 2 years of age, incidence of norovirus infection will decline due to acquired immunity 23-26 . In our study, unfortunately, the degree of diarrheal severity in subjects infected by norovirus GI could not be compared to the degree of diarrheal severity in norovirus GII since norovirus GI was only detected in one sample, which is not enough for comparison.
Although this study achieved its aims, there were unavoidable study limitations. First, our sample size was comparatively small compared to previous norovirus studies in other countries. We did not include neonates below 1 month old due to our limitation to reach the neonatal ward. Secondly, we found no recurrent cases in our study, and therefore we did not analyze the relationship between norovirus genogroup classification and recurrence of diarrhea. Thirdly, since all the patients that participated in this study were all being admitted to the hospital, the treatment criteria are relatively more severe. Nevertheless, our findings largely agree with previous studies in Surabaya, Shanghai, and Rio de Janeiro, as explained in our discussion above 10,20,22 . Fourth, we did not have data about other pathogens, which might be co-infecting. In addition, we could not classify the norovirus GII into genotype, and then use this genotype to infer the severity of the disease. Finally, this study only categorized norovirus genogroups by RT-PCR. We did not perform gene sequencing for norovirus RNA. We also did not include positive controls for each PCR reaction. Future studies will address these limitations.

Conclusion
This study demonstrated that norovirus was responsible for 33.0% of diarrhea cases in the study group, and norovirus GII was significantly dominant compared to norovirus GI. As many as 30% of norovirus cases had severe diarrheal manifestation, all of which were caused by norovirus GII. 1.

Open Peer Review Current Peer Review Status:
Version 3 1.

5.
As I have stated earlier in my previous comments, the sample size is too small to perform correlation analysis because only one GI positive manifested as moderate diarrhea. The authors should then not use term "correlation" in the objective (last paragraph of introduction section), as well as in the title. How can the authors make correlation analysis if they had only one GI positive samples? Thus, the term "correlation" is not supported by their data. Therefore, I highly recommend to change to a more general title, such as "Clinical manifestation of norovirus infection in children aged less than five years old admitted with acute diarrhea in Surabaya, Indonesia: a cross-sectional study" or "Norovirus infection in children aged less than five years old admitted with acute diarrhea in Surabaya, Indonesia: a cross-sectional study".
The authors mentioned, "Genogroup diversity can be checked using immunochromatography and reverse transcriptase polymerase chain reaction (RT-PCR)". Previously I was only concerned about electron microscopy, but I just realized that is it true that genogroup diversity can also be determined by immunochromatography? Please support your statement with solid references.
Please change "severer" to "more severe" (in introduction section). I have mentioned this previously, but the authors did not make any changes.
In the Discussion, the authors mentioned, "We did not perform gene sequencing for norovirus DNA." Comment: Norovirus is an RNA virus, not DNA virus. Please change.
In the Discussion, the authors mentioned, "One of the possible contributing factors is because we did not use positive controls for our PCR; therefore our result might have high false positive result." Comment: high rate of false positive result mostly due to unproper control, not positive negative control or due to high contaminations. In contrast, positive control is employed to reduce false negative results. The best way to confirm the PCR results is by performing sequencing of some or all positive samples. The authors should modify this issue in the next revision.
No competing interests were disclosed.
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above. 1.

11.
prevalence in children <5 years old with acute diarrhea in Indonesia are still limited. Therefore, this paper is highly important to expand the data of norovirus surveillance in Indonesia.

Comments:
However, this paper needs a major revision to improve the way of presenting and discussing the data as I recommend the following: The sample size is too small to perform correlation analysis. Only one GI positively manifested as moderate diarrhea. The authors should then not use "correlation" in the objective (last paragraph of introduction section), as well as in the title, this does not make any sense. The author can change into a more general title, such as "Clinical manifestation of norovirus infection in children aged less than five years old admitted with acute diarrhea in Surabaya, Indonesia: a cross-sectional study" or "Norovirus infection in children aged less than five years old admitted with acute diarrhea in Surabaya, Indonesia: a cross-sectional study." The authors mention in the introduction: "Genogroup diversity can be checked using several methods including immunochromatography, reverse transcriptase polymerase chain reaction (RT-PCR), and electron microscopy." Comments: any supporting evidence or reference of this statement? Norovirus genogroup is based on genetic analysis, so somehow it is weird that it can be differentiated based on electron microscopy.
The authors did not perform comprehensive literature searching of the previous norovirus study in Indonesia. In fact, there are already some "key" publications in this field, but the authors did not include them in this manuscript. Please thoroughly check the referenced papers to improve author statement in both introduction and discussion section .
Please change "severer" to "more severe" (in introduction section).
In the methods: Are there any specific reason to exclude <1 month baby in this study? Figure 2B, lane 1. Is it considered positive or negative? Because it seems a scanty, positive band there. It looks like Figure 2A, lane 2 which is considered as positive for norovirus GI.
Do the authors include positive controls for each PCR reaction? The prevalence of norovirus in this study (about 32%) is higher than that of global prevalence (about 20%) in countries that did not include Rotavirus vaccination in the NIP. So the authors should ensure that a proper and valid PCR assay has been conducted. Please check: Ahmed SM . (2014) . et al Figure 1 is not necessary, so please delete it.

16.
In the discussion, the authors should also discuss associated risk factors of contracting norovirus based on Table 2. For example, why did in your population is the most prevalent age of norovirus positive 6-23 months?
The authors should add discussion about different severity of norovirus infection based on different genotpe of GII norovirus. Although they did not perform genotype identification in this study, the readers of this paper should still be aware that different GII genotypes can cause different severity of clinical manifestations.
In conclusion, it is mentioned: "This study demonstrated that norovirus was responsible for 48.4% of diarrhea cases in the study group". Could the authors clarify about the percentage of 48.4%? 31 out of 94 should be 32.9%. It is also not consistent with the abstract (I mention below).
Reference no. 6, what journal that publishes this paper? The authors did not mention this. If this is not published in a peer-reviewed journal, it would be better to change the reference with one I recommended.
The abstract: "This cross-sectional study involved 31 participants" --> please delete "31". When you design the study, you never know how many patients will participate. "In total, 91 stool samples were obtained, of which 31 (19%) … " --> please clarify the percentage. Also, is it 91 or 94? Please thoroughly check all the numbers and percentages you mentioned in this paper. Table 2, change "Malutrition" to "Malnutrition". Also, please describe in the methods what the criteria to categorize this nutrition status are.

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility? Yes

Are the conclusions drawn adequately supported by the results? No
No competing interests were disclosed.
I confirm that I have read this submission and believe that I have an appropriate level of expertise to state that I do not consider it to be of an acceptable scientific standard, for reasons outlined above.

RESPONSES
We greatly appreciate your complimentary comments and suggestions.
1. We apologize that our sample size is too small, however, from this limited sample still we could explore many things.