To examine the potential for respiratory transmission of rotavirus, we systematically assessed if rotavirus RNA is detectable by real-time quantitative reverse transcription-polymerase chain reaction from nasal and oropharyngeal swab specimens of Bangladeshi children with acute rotavirus gastroenteritis. Forehead swabs were collected to assess skin contamination. Among 399 children aged <2 years hospitalized for gastroenteritis during peak rotavirus season, rotavirus RNA was detected in stool, oral, nasal and forehead swab specimens of 354 (89%). A subset was genotyped; genotype was concordant within a child's specimen set and several different genotypes were detected across children. These findings support possible respiratory transmission of rotavirus and warrant further investigation.
Group A rotavirus is the leading cause of severe acute gastroenteritis (AGE) in young children, especially those in countries without widespread rotavirus vaccination programs, such as Bangladesh [1–3]. While fecal-oral/fecal-fomite-oral are accepted as the main transmission routes from infection in the small intestine, the possibilities of respiratory transmission and respiratory tract infection with rotavirus have long been raised in part because of the observations of the rapidity of disease spread in some situations, the lack of reduced disease incidence with improved sanitation, and the frequency of respiratory symptoms reported in children with rotavirus gastroenteritis in some studies [4–6]. Using predominantly earlier, less sensitive detection methods (eg, enzyme immunoassay [EIA], indirect immunofluorescence), rotavirus was detected in nasopharyngeal aspirates of some children with respiratory symptoms (some of whom had rotavirus detected in stool) in a few studies but was not detected in other studies [7–9]. Using modern molecular methods, the only systematic evaluation in the published literature was designed to assess the performance of polymerase chain reaction (PCR) testing of buccal swab specimens for identifying the causative pathogen in episodes of gastroenteritis [10]. Of 167 patients aged <18 years whose stool tested positive for rotavirus nonstructural protein 3 (NSP3) by real-time quantitative reverse transcription-PCR (qRT-PCR), the buccal swab was NSP3 positive in 15 (8.9%) [10]. The objective of our study was to determine if rotavirus RNA is detectable from nasal and oropharyngeal swab specimens of young children with severe acute rotavirus gastroenteritis, which would support possible respiratory transmission of this pathogen.
During January–April 2018, this study was nested in ongoing rotavirus surveillance conducted [1] in 7 tertiary care hospitals in Bangladesh. In brief, every fourth child aged <5 years with AGE who was hospitalized for rehydration was approached for enrollment. AGE was defined as occurrence of ≥3 watery or looser-than-normal stools, or ≥1 episode of forceful vomiting, within a 24-hour period with symptom duration ≤7 days at presentation. For those children whose parents provided written consent to participate in surveillance, clinical information was obtained and a whole stool sample was collected. Batched stool testing for rotavirus by EIA was performed at the icddr,b laboratory. For this nested project, children aged <2 years who enrolled in rotavirus surveillance during the 2018 seasonal rotavirus gastroenteritis peak were potentially eligible for the respiratory specimen project, and written consent was requested from parents. Children were ineligible for the respiratory project if they had a nasal or palate abnormality or if a parent had observed any blood from the nares on day of surveillance enrollment. After consent was provided, swab specimens from the nares, oropharynx, and skin were obtained from children who had not had any vomiting on that day up to time of specimen collection; if the child had vomited that day, specimens were not collected and child was potentially eligible the following day if vomiting had ceased. Trained nurses or physicians collected the specimens using dry flocked nylon swabs with attention not to allow the respiratory tract swabs to touch any external skin or other surface. Two midturbinate nasal swabs (1 from each nares; any nasal drainage was wiped away before specimen collection) and 1 oropharyngeal swab (collected ≥30 minutes after any food or drink; tonsils were swabbed after the tongue was depressed with a wooden spatula) were collected. A forehead skin swab (swabbing of 2 cm area in middle of forehead, no cleaning of skin before swabbing) was collected to assess contamination of skin surfaces. Each swab was placed in a separate cryovial. Parents were asked additional clinical questions (ie, had child vomited the day before swab collection, did child have cough or rhinorrhea the day of swab collection).
The goal was to enroll approximately 200 children with EIA-confirmed rotavirus gastroenteritis. Based on 2016 surveillance results at these 7 hospitals, approximately 50% of children aged <2 years enrolled in rotavirus surveillance during peak rotavirus season were expected to be rotavirus EIA positive; our goal therefore was to enroll 400 children into the nested respiratory project. Respiratory swab specimen results from enrolled children whose stool specimen ultimately tested rotavirus EIA negative would serve as a control group for comparison to the rotavirus EIA-positive children.
This study protocol was reviewed and approved by the ethical committee of icddr,b, and the Centers for Disease Control and Prevention’s (CDC) Human Research Protection Office approved reliance on icddr,b. Written informed consent was obtained from parents/guardians of the enrolled children.
Stool specimens were tested for rotavirus VP6 antigen by EIA (IDEA Prospect, Oxoid Diagnostics, United Kingdom) at icddr,b. Specimens or aliquots of swab elutions were stored frozen at −80°C until shipped frozen to CDC (Atlanta, Georgia, USA). At CDC, RNA was extracted from all specimen types and tested for rotavirus NSP3 using qRT-PCR (Supplementary material). The 4 specimens from an individual child were considered a specimen set. Rotavirus genotyping by Sanger sequencing (Supplementary material) was performed in a limited number of sets for which rotavirus NSP3 was detected at relatively high viral loads (low cycle threshold [Ct] values) in each specimen to determine if the genotype(s) were concordant or discordant within the set; discordant genotypes within a set could indicate contamination (eg, across children or at a laboratory). For genotyping, we desired to select sets that included ≥1 child from each of 7 hospitals and several for which stool had tested negative for rotavirus by EIA but positive for rotavirus NSP3. The distribution of NSP3 qRT-PCR Ct values were compared using Kolmogorov–Smirnov test. In post hoc analyses, epidemiological characteristics were compared by Wilcoxon rank-sum or χ2 tests.
Of 400 children enrolled, the stool specimen tested rotavirus EIA positive in 301 (75%) and rotavirus EIA negative in 99 (25%). Children testing EIA positive were older than those EIA negative (median, 11.3 months [interquartile range {IQR}, 8.6–14.7] vs 10.2 months [IQR, 7.2–12.8], respectively; P < .05), had higher maximum daily number of diarrheal stools prior to admission (median: EIA positive, 20; EIA negative, 18; P < .01), shorter diarrhea duration from onset to hospitalization (median: EIA positive, 3 days; EIA negative, 4 days; P < .05), and a greater proportion had vomiting prior to admission (EIA positive: 220/301 [73%] vs EIA negative: 57/99 [58%]; P < .01). No child had vomited the day before (or by protocol, the day of) respiratory specimen collection. The number of days from hospital admission to stool collection was similar (EIA positive: 0 days = 16%, 1 day = 75%, 2–4 days = 10%; EIA negative: 0 days = 15%, 1 day = 76%, 2 days = 9%; P = 1.0), as was number of days from stool collection to respiratory specimen collection (EIA positive; 0 days = 96%, 1 day = 4%; EIA negative: 0 days = 99%, 1 day = 1%; P = .15).
Three hundred ninety-nine children had their 4-specimen set tested for rotavirus NSP3; 1 child whose stool tested rotavirus EIA positive had only a 3-specimen set (stool, nasal swab, and forehead swab). Each of the 400 children tested NSP3 positive from ≥1 of their specimens. NSP3 was detected in all of the stool specimens and all the forehead swab specimens of the 301 children whose stool tested EIA positive (Table 1). Among the 99 children whose stool tested EIA negative, NSP3 was detected in 95 (96%) of the stool samples and 97 (98%) of the forehead swab specimens.
Table 1.
EIA Result.
Specimen Type.
No. of Specimens Positive By Rotavirus NSP3 qRT-PCR/Number Tested (% Positive).
Stool rotavirus.
Specimen Type.
No. of Specimens Positive By Rotavirus NSP3 qRT-PCR/Number Tested (% Positive).
Stool rotavirusTable 1.
EIA Result.
Specimen Type.
No. of Specimens Positive By Rotavirus NSP3 qRT-PCR/Number Tested (% Positive).
Stool rotavirus.
Specimen Type.
No. of Specimens Positive By Rotavirus NSP3 qRT-PCR/Number Tested (% Positive).
Stool rotavirusAmong the 300 children whose stool tested EIA positive and who provided a 4-specimen set, NSP3 was detected in each specimen in 286 (95%); the next most common result combination was NSP3-positive stool, oral, and forehead specimens, with NSP3-negative nasal specimen (n = 11 [4%]) (Table 1). Among the 99 stool EIA-negative children, NSP3 was detected in each of their 4 specimens in 68 (69%); the next most common combination was NSP3-positive stool, nasal, and forehead specimens, with NSP3-negative oral specimen (n = 13 [13%]). Within each specimen type, Ct values were lower (ie, higher viral load) for stool EIA-positive children versus stool EIA-negative children (P < .001). Overall, 17% (51/301) of EIA-positive and 19% (19/99; P = .61) of EIA-negative children had cough or rhinorrhea the day of sample collection (Supplementary material). Among the 395 children whose stool tested NSP3 positive and for whom an oral swab was tested, the proportion with oral swab specimen NSP3 positive was similar among those who had vomiting prior to admission (93% [254/272]) versus those without vomiting (95% [117/123]).
Among the 19 specimen sets with genotyping results (12 sets from children whose stool was EIA positive; 7 sets from children whose stool was EIA negative), at least G- or P-type was able to be determined in each specimen with 1 exception—a nasal swab for which G and P were not able to be typed (Table 2). The same G and/or P genotype was detected in each specimen within a child's set, and different G and P genotypes were detected across several children. For the 7 specimen sets in which this was performed, sequencing of a 756-base pair segment within the VP7 (G) coding region and 691-base pair segment within the VP4 (P) coding region supported concordance within sets with a sequence identity ≥99% and differences across children with the same overall G and P genotype (data not shown).
Table 2.
Hospital.
Among Children With Stool Rotavirus EIA Positive (n = 12).
Among Children With.
No. of 4-Specimen Sets With Genotyping Results.
Was the Same Genotype Detected in Each Specimen of the Set?a.
Genotype Detected, by Set.
No. of 4-Specimen Sets With Genotyping Results.
Was the Same Genotype Detected in Each Specimen of the Set?a.
Genotype Detected, by Set.
1 2 Yes G3P[8] 1 Yes G2P[6] G2P[4]b,c 2 2 Yes G3P[8]b 2 Yes G2P[4] G3P[8]b,c G9P[8] 3 2 Yes G2P[4]b,c,d 1 Yes G1P[4] G2P[4]b 4 1 Yes G3P[8]b,e 0 NA NA 5 2 Yes G1P[8]b 1 Yes G1P[8] G1P[8]b 6 1 Yes G3P[8]b 0 NA NA 7 2 Yes G3P[8]b,d 2 Yes G3P[8] G3P[8] G1P[8] Hospital.
Among Children With Stool Rotavirus EIA Positive (n = 12).
Among Children With.
No. of 4-Specimen Sets With Genotyping Results.
Was the Same Genotype Detected in Each Specimen of the Set?a.
Genotype Detected, by Set.
No. of 4-Specimen Sets With Genotyping Results.
Was the Same Genotype Detected in Each Specimen of the Set?a.
Genotype Detected, by Set.
1 2 Yes G3P[8] 1 Yes G2P[6] G2P[4]b,c 2 2 Yes G3P[8]b 2 Yes G2P[4] G3P[8]b,c G9P[8] 3 2 Yes G2P[4]b,c,d 1 Yes G1P[4] G2P[4]b 4 1 Yes G3P[8]b,e 0 NA NA 5 2 Yes G1P[8]b 1 Yes G1P[8] G1P[8]b 6 1 Yes G3P[8]b 0 NA NA 7 2 Yes G3P[8]b,d 2 Yes G3P[8] G3P[8] G1P[8] Open in new tabTable 2.
Hospital.
Among Children With Stool Rotavirus EIA Positive (n = 12).
Among Children With.
No. of 4-Specimen Sets With Genotyping Results.
Was the Same Genotype Detected in Each Specimen of the Set?a.
Genotype Detected, by Set.
No. of 4-Specimen Sets With Genotyping Results.
Was the Same Genotype Detected in Each Specimen of the Set?a.
Genotype Detected, by Set.
1 2 Yes G3P[8] 1 Yes G2P[6] G2P[4]b,c 2 2 Yes G3P[8]b 2 Yes G2P[4] G3P[8]b,c G9P[8] 3 2 Yes G2P[4]b,c,d 1 Yes G1P[4] G2P[4]b 4 1 Yes G3P[8]b,e 0 NA NA 5 2 Yes G1P[8]b 1 Yes G1P[8] G1P[8]b 6 1 Yes G3P[8]b 0 NA NA 7 2 Yes G3P[8]b,d 2 Yes G3P[8] G3P[8] G1P[8] Hospital.
Among Children With Stool Rotavirus EIA Positive (n = 12).
Among Children With.
No. of 4-Specimen Sets With Genotyping Results.
Was the Same Genotype Detected in Each Specimen of the Set?a.
Genotype Detected, by Set.
No. of 4-Specimen Sets With Genotyping Results.
Was the Same Genotype Detected in Each Specimen of the Set?a.
Genotype Detected, by Set.
1 2 Yes G3P[8] 1 Yes G2P[6] G2P[4]b,c 2 2 Yes G3P[8]b 2 Yes G2P[4] G3P[8]b,c G9P[8] 3 2 Yes G2P[4]b,c,d 1 Yes G1P[4] G2P[4]b 4 1 Yes G3P[8]b,e 0 NA NA 5 2 Yes G1P[8]b 1 Yes G1P[8] G1P[8]b 6 1 Yes G3P[8]b 0 NA NA 7 2 Yes G3P[8]b,d 2 Yes G3P[8] G3P[8] G1P[8] Open in new tabAmong children hospitalized for AGE during the peak rotavirus season, rotavirus NSP3 was detectable by qRT-PCR in the stool and in nasal, oropharyngeal, and forehead swab specimens of most, including those whose stool tested rotavirus negative by EIA. Detection of NSP3 on forehead swabs supports contamination of the child's skin with stool, most likely their own stool, that is, a “fecal veneer” identified with such a sensitive assay. While detection of rotavirus NSP3 in oropharyngeal swab specimens, and even nasal specimens, may occur as a result of gastrointestinal reflux during peristalsis from children with rotavirus infection in the small intestine, these findings would also be consistent with rotavirus replication in the upper respiratory tract [11].
As expected given the higher sensitivity of the assay, a greater proportion of children had rotavirus detected in stool by the NSP3 qRT-PCR assay than by VP6 EIA; however, it was surprising that 95 of 99 rotavirus EIA-negative children had NSP3 detected (albeit with overall higher Ct values than in the EIA-positive group). For the EIA-negative children who had low NSP3 Ct values, it is likely that rotavirus is the cause of their current AGE. For some of the EIA-negative children with high NSP3 Ct values, it is probable that they were still shedding rotavirus in the stool from an earlier symptomatic or asymptomatic infection (especially given that this surveillance was performed during peak rotavirus season in the age group with highest incidence of rotavirus infection) and that rotavirus is not the cause of the current AGE. When assessed using a very sensitive assay such as RT-PCR, detection of rotavirus RNA in stool can continue for many weeks after infection [3, 12, 13]. In our surveillance, testing of stool for other pathogens was not systematically performed. For children whose stool tested EIA negative but NSP3 positive (ie, rotavirus replication in the small bowel, including some at low levels), the interpretation of the high detection rate of NSP3 in respiratory and forehead specimens can be the same as for the EIA-positive children.
Given the very high proportion of children NSP3 positive in stools and swab specimens, cross-contamination of samples during collection, laboratory processing or testing, or between patients while in hospital, was a concern. Review of procedures did not identify a likely explanation. Although a relatively small number of sample sets were genotyped, our results that represent each hospital increased confidence that our respiratory sample results are unlikely to be explained by widespread contamination at the laboratory level or across patients. Three children were NSP3 negative in stool but NSP3 positive from the forehead, oral, and/or nasal specimen; it is possible the NSP3 stool result was falsely negative. One child was only NSP3 positive from the forehead specimen, so it is possible the fecal veneer source may sometimes have been another rotavirus-positive child. In addition to other possible interpretations described earlier, it is also possible that some nasal swabs picked up fecal veneer that may have extended to just inside the nostril despite our best attempts at midturbinate swabbing. Finally, while we tested for and detected rotavirus RNA, we did not test for viable virus.
Our high detection rate of rotavirus RNA in the nares and oropharynx bolsters the possibility that rotavirus replicates in the upper respiratory tract of infected children. This paramount question about rotavirus infection and transmission now stands ready to be resolved by capitalizing on the advancements in human respiratory tract organoids [14, 15].
Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
Author contributions. M. M. C. conceived the investigation, analyzed the data, and wrote the initial draft. M. M. C., S. M. S., and U. P. designed the investigation. S. M. S. supervised enrollment and collection of data and specimens. E. K. and M. D. B. designed the laboratory procedure methods. E. K. led the laboratory testing. E. K., M. E. H., M. T., S. L. S., M. D. B., and M. R. managed the clinical specimens and/or performed or supervised laboratory testing. F. F. managed the data at icddr,b. All authors reviewed the manuscript for scientific content and approved the final version.
Acknowledgments. The authors thank Drs Mathew Esona and Rashi Gautam for their assistance with genotyping at the Centers for Disease Control and Prevention (CDC; Atlanta, Georgia, USA). We thank all of the project participants for their time. icddr,b acknowledges with gratitude the commitment of the CDC to its surveillance efforts. icddr,b is also grateful to the governments of Bangladesh, Canada, Sweden, and the United Kingdom as their core donors for providing unrestricted support.
Disclaimer. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the CDC. Mention of a product or company name is for identification purposes only and does not constitute endorsement by the CDC.
Financial support. This work was conducted as part of the cooperative agreement between the CDC and icddr,b (number U01GH001207).
1
Satter
SM
,Aliabadi
N
,Gastañaduy
PA
, et al.An update from hospital-based surveillance for rotavirus gastroenteritis among young children in Bangladesh, July 2012 to June 2017
.Vaccine
2018
;36
:7811
–5
.2
Aliabadi
N
,Antoni
S
,Mwenda
JM
, et al.Global impact of rotavirus vaccine introduction on rotavirus hospitalisations among children under 5 years of age, 2008–16: findings from the Global Rotavirus Surveillance Network
.Lancet Glob Health
2019
;7
:e893
–903
.3
Liu
J
,Platts-Mills
JA
,Juma
J
, et al.Use of quantitative molecular diagnostic methods to identify causes of diarrhoea in children: a reanalysis of the GEMS case-control study
.Lancet
2016
;388
:1291
–301
.4
Foster
SO
,Palmer
EL
,Gary
GW
Jr, et al.Gastroenteritis due to rotavirus in an isolated Pacific island group: an epidemic of 3,439 cases
.J Infect Dis
1980
;141
:32
–9
.5
Santosham
M
,Yolken
RH
,Wyatt
RG
, et al.Epidemiology of rotavirus diarrhea in a prospectively monitored American Indian population
.J Infect Dis
1985
;152
:778
–83
.6
Dennehy
PH
.Transmission of rotavirus and other enteric pathogens in the home
.Pediatr Infect Dis J
2000
;19
:S103
–5
.7
Fragoso
M
,Kumar
A
,Murray
DL
.Rotavirus in nasopharyngeal secretions of children with upper respiratory tract infections
.Diagn Microbiol Infect Dis
1986
;4
:87
–8
.8
Zheng
BJ
,Chang
RX
,Ma
GZ
, et al.Rotavirus infection of the oropharynx and respiratory tract in young children
.J Med Virol
1991
;34
:29
–37
.9
Taboada
B
,Espinoza
MA
,Isa
P
, et al.Is there still room for novel viral pathogens in pediatric respiratory tract infections?
PLoS One
2014
;9
: .: .
10
Zhuo
R
,Parsons
BD
,Lee
BE
, et al.Identification of enteric viruses in oral swabs from children with acute gastroenteritis
.J Mol Diagn
2018
;20
:56
–62
.11
Cotterill
H
,Curry
A
,Riordan
T
.Rotavirus in vomit
.J Infect
1988
;16
:206
–7
.12
Richardson
S
,Grimwood
K
,Gorrell
R
,Palombo
E
,Barnes
G
,Bishop
R
.Extended excretion of rotavirus after severe diarrhoea in young children
.Lancet
1998
;351
:1844
–8
.13
Bennett
A
,Bar-Zeev
N
,Jere
KC
, et al.Determination of a viral load threshold to distinguish symptomatic versus asymptomatic rotavirus infection in a high-disease-burden African population
.J Clin Microbiol
2015
;53
:1951
–4
.14
Bui
CHT
,Chan
RWY
,Ng
MMT
, et al.Tropism of influenza B viruses in human respiratory tract explants and airway organoids
.Eur Respir J
2019
;54
: .: .
15
Ghosh
S
,Kumar
M
,Santiana
M
, et al.Enteric viruses replicate in salivary glands and infect through saliva
.Nature
2022
;607
:345
–50
.Published by Oxford University Press on behalf of Infectious Diseases Society of America 2023.
This work is written by (a) US Government employee(s) and is in the public domain in the US.