Respiratory syncytial virus (RSV) is the leading cause of acute lower respiratory tract infection (LRTI) in infants and young children worldwide.1 In 2015, an estimated 33.1 million RSV-associated acute LRTIs occurred in children less than 5 years worldwide, resulting in 3.2 million hospitalizations and 59,600 deaths among hospitalized children.2 In the Middle East and North Africa (MENA) region, RSV continues to be a major health and financial burden, with the highest reported prevalence of RSV in Jordan.3,4
The severity of RSV-associated acute respiratory infection (ARI) depends on the interaction of environmental, host and viral factors; these include age, prematurity, underlying medical conditions (UMCs), smoke exposure, viral inoculum and possibly viral subtype.5,6
RSV encompasses 2 antigenically distinct subtypes, RSV-A and RSV-B, based on the differences in the G surface glycoprotein.7,8 There is conflicting evidence on whether RSV subtype is associated with severity of clinical course.9–14 Previous studies were limited by including small sample sizes from single respiratory seasons. Evaluating these differences across different seasons and among large sample sizes is warranted to further understand this important pathogen and guide treatment and preventive measures.
Our previous ARI surveillance study of 3168 hospitalized children less than 2 years old in Amman, Jordan reported that RSV was associated with increased disease severity compared with other viruses over 3 respiratory seasons.3 However, RSV subtyping was not performed and limited data exist on the seasonality and disease severity among RSV subtypes in the Middle East.4 Therefore, this study aimed to evaluate the clinical characteristics, severity and seasonality of RSV subtypes in hospitalized children less than 2 years of age in Amman, Jordan. We hypothesized that RSV-A would be predominant and associated with more severe disease.
Study Design and Setting
This study used data and specimens from a previous perspective viral surveillance study conducted from March 16, 2010 to March 31, 2013 in Amman, Jordan.3,15 Briefly, children less than 2 years who presented with fever and/or respiratory symptoms were enrolled year-round. Children who had chemotherapy-associated neutropenia and/or were newborns who had never been discharged were excluded from the study. Enrollment was performed 5 days a week (Sunday through Thursday) in Al-Bashir Hospital, a major government referral center located in Amman, Jordan. Further details on the study design specifics, the study population and study setting have been published.3,15
Written informed consent was obtained from parents or guardians before enrollment. The Institutional Review Boards of the University of Jordan, the Jordanian Ministry of Health and Vanderbilt University approved the study.
Data and Specimen Collection
Trained research staff obtained demographic characteristics and medical and social histories using standardized questionnaires. Nasal and throat swabs were collected. Demographic and clinical data were collected systematically from the subjects’ medical charts after discharge.
Intensive care unit (ICU) stays included children who were either admitted directly to the ICU or were transferred during admission. Oxygen requirements included children who required oxygen at any point during the hospital admission. Smoke exposure included both cigarette and/or nargileh (hookah pipe) exposure. UMCs included the following: asthma/reactive airway disease, cerebral palsy, cystic fibrosis, cancer, Down syndrome, heart disease, immunodeficiency and neurologic disease. Prematurity was defined as birth at <37 gestational weeks.
At the time of admission, the treating physician collected data points to calculate a severity score. This score is a modification of a validated severity score by Tal et al, as oxygen saturation on room air was added to the original score variables which were respiratory rate, wheezing, flaring/retractions/accessory muscle use and cyanosis (see Table, Supplemental digital content 1; https://links.lww.com/INF/E394).16 A score of 0–5 represented mild illness, 6–9 represented moderate illness and 10–15 represented severe illness.
Data were entered into a standardized, secured REDCap (Research Electronic Data Capture, Vanderbilt University, Nashville, TN) database.
Nasal and throat swabs were collected and combined in transport medium (M4RT; Remel, Lenexa, KS), aliquoted into MagMAX Lysis/Binding Solution Concentrate (Life Technologies, Carlsbad, CA), snap frozen and stored at −80°C. Original and lysis buffer aliquots were shipped on dry ice and were tested by quantitative reverse-transcription polymerase chain reaction (qRT-PCR) for eleven respiratory viruses [RSV, human metapneumovirus, human rhinovirus, influenza A, B and C, parainfluenza (PIV) virus 1, 2 and 3, adenovirus and Middle East respiratory syndrome coronavirus]. The samples were then stored at Vanderbilt. In 2020, testing for 4 human coronavirus subtypes, PIV 4 and subtyping of RSV-positive samples for RSV-A and RSV-B were performed.17 Cycle threshold (Ct) values were used as an indicator of viral load, with lower Ct values representing higher viral loads.
Vitamin D Testing
Blood was placed directly onto filter paper and air dried for ≥30 min before storage at room temperature and kept in a dry state until shipment to ZRT Laboratory (Beaverton, OR) for vitamin D assay per protocol.3
Demographics and Clinical Characteristics
This analysis only included children who were RSV-positive and further subtyped to either RSV-A or RSV-B. Children with indeterminate samples, missing samples or had simultaneous detection of both RSV-A and RSV-B were excluded. Descriptive statistics were summarized as frequency (percentage) or mean (±standard deviation) where suitable. RSV-A and RSV-B were compared using chi-square test for categorical variables and linear regression with robust standard errors for continuous variables. Significance level was set at 0.05 and all analyses were performed using StataCorp version 16.1 (College Station, TX).
We used multivariable linear regression models with robust standard errors to compare the mean severity score across predictors that were chosen a priori based on what was reported to affect RSV disease severity from the literature and our previous work.3,10 These predictors included RSV subtype, viral coinfection, Ct value, age, sex, UMC, prematurity, vitamin D level, number of people living in the household, smoke exposure and breast-feeding.
To compare the odds of ICU admission and oxygen requirement (which are indicators of severity during the hospitalization course) separately across factors potentially affecting RSV severity, we used multivariable logistic regression models and included a priori selected predictors of interest: RSV subtype, viral coinfection, Ct values, age, sex, UMC, prematurity, vitamin D levels, number of people living in the household, smoke exposure, breast-feeding and severity score.
For all regression models, we used multiple imputation with chained equations with M = 100 imputation iterations to address missing data. Of the variables in the models, vitamin D levels were the only one to have missing data where 204 of 1397 (14.6%) children had missing values.
Demographic and Clinical Characteristics of the RSV-positive Cohort
Of 3168 enrolled children, 1397 (44%) tested positive for RSV. The mean age of the RSV-positive children was 5.3 months, 59.7% were male, 6.4% had UMC, 12.8% were premature, 85.9% had a history of breast-feeding, 77% were exposed to smoke and 2% attended daycare. The 5 most common admission diagnoses among RSV-positive children were bronchopneumonia (34.1%), bronchiolitis (26.8%), suspected sepsis (17.7%), pneumonia (16.1%) and pertussis-like cough (9.0%).
Of the 1397 RSV-positive children, 889 (63.6%) were RSV-A positive, 352 (25.2%) were RSV-B positive, 8 (0.6%) were positive for both RSV-A and RSV-B, 143 (10.2%) had indeterminate PCR results, and 5 (0.4%) were not typed due to unavailability of the sample (Fig. 1).
Demographic and Clinical Characteristics of RSV Subtypes
Table 1 demonstrates that there was no evidence of a difference in the demographics and admission diagnoses between RSV-A and RSV-B. However, children with RSV-A presented with higher frequency of decreased appetite but lower frequency of viral coinfection compared with children with RSV-B (Table 1). Supplemental digital content 2; https://links.lww.com/INF/E395 (Figure; https://links.lww.com/INF/E395) demonstrates the percentage of virus-specific RSV-positive co-detections.
TABLE 1. -
Comparison of Clinical and Demographic Characteristics Between RSV-A Positive and RSV-B Positive Children
||RSV-A Positive, n = 889 (63.6%)
||RSV-B Positive, n = 352 (25.2%)
| Age, mean ± SD, months
||5.1 ± 4.7
||5.3 ± 4.8
| Sex, male
| Daycare attendance
| Number of people living in household, mean ± SD
||5.6 ± 2.2
||5.7 ± 2.4
| Smoke exposure
| Prematurity (<37 weeks)
| Underlying medical conditions
| Suspected sepsis
| Pertussis-like cough
| Shortness of breath
| Runny nose
| Nasal congestion
| Decreased activity
| Respiratory rate, mean ± SD
||42.3 ± 10.8
||42.4 ± 11.6
| O2 sat (%), mean ± SD
||87.6% ± 17.4
||92.6% ± 3.2
| Vitamin D level (ng/ml), mean ± SD
||231.8 ± 148.5
||232.9 ± 138.6
| Length of stay, mean, days
||5.4 ± 3.5
||5.2 ± 3.0
| Oxygen requirements
| ICU admission
| Mechanical ventilation
| Severity Score
| Mild (0–5)
| Moderate (6–9)
| Severe (10–15)
Continuous data are in mean ± SD, categorical data are in n (%). Bold values indicate P < 0.05.
*Simple linear regression with robust standard errors.
†Pearson’s Chi-square test.
ICU indicates intensive care unit; n, number; O2 sat, oxygen saturation; SD, standard deviation.
Seasonality of RSV Subtypes
Throughout the study period, RSV-A and RSV-B subtypes co-circulated, with the highest peaks in January–March of each year. RSV-A predominated in March–April of 2010 and throughout the 2010–2011 and 2011–2012 seasons, while RSV-B was the most frequent circulating subtype in the 2012–2013 season (Fig. 2).
Severity of RSV Subtypes
When comparing severity measurements such as mean length of stay, supplemental oxygen requirement, ICU admission, need for mechanical ventilation, mortality and an objective severity score, we found that RSV-A and RSV-B had similar severity profiles in the univariable analyses (Table 1). Of note, although not statistically significant, RSV-A positive children had higher frequencies of runny nose (2.4% vs. 0.9%, P = 0.08), apnea (0.5% vs. 0%, P = 0.21), mechanical ventilation (4.8% vs. 2.6%, P = 0.08) and death (0.7% vs. 0%, P = 0.08) when compared with RSV-B positive children.
Table 2 illustrates that no evidence of an association between RSV subtype and mean severity score was noted. However, we found that higher Ct values, older age and higher vitamin D levels were all independently associated with lower mean severity score.
TABLE 2. -
Estimates From a Linear Regression Model to Evaluate the Association Between Variables Associated with a Higher Severity Score*
||−0.106 to 0.427
Cycle threshold values
||−0.0701 to −0.0174
||−0.351 to 0.145
||−0.145 to 0.368
||−0.0541 to −0.00238
||−0.350 to 1.12
||−0.551 to 0.203
Vitamin D level (ng/ml × 100)
||−0.204 to −0.0284
|Number of people living in household
||−0.0501 to 0.0699
||−0.262 to 0.351
||−0.363 to 0.368
Bold values indicate P < 0.05.
*Model was adjusted for all the covariates listed in the table.
CI indicates confidence interval; RSV, respiratory syncytial virus; UMC, underlying medical condition.
When using ICU admission and oxygen requirement as surrogates for severity, no association with RSV subtype was found. Nevertheless, UMCs, younger age, prematurity and higher severity score were all independently associated with higher odds of ICU admission and the need for supplemental oxygen. Lower Ct value also was predictive of oxygen requirement (Table 3).
TABLE 3. -
Estimates from logistic regression models to evaluate the association between variable associated with ICU admission and oxygen requirement*
||0.55 to 1.37
||0.84 to 1.43
Cycle threshold values†
||0.94 to 1.04
0.92 to 0.97
||0.81 to 1.86
||0.74 to 1.20
||0.47 to 1.08
||0.65 to 1.06
0.73 to 0.91
0.84 to 0.90
1.31 to 7.23
1.51 to 5.67
1.15 to 3.75
1.06 to 2.23
|Vitamin D level
||0.999 to 1.002
||0.999 to 1.001
|Number of people living in household
||0.95 to 1.14
||0.98 to 1.09
||0.50 to 1.28
||0.62 to 1.11
||0.36 to 1.15
||0.58 to 1.22
1.28 to 1.53
1.04 to 1.15
Bold values indicate P < 0.05.
*Model was adjusted for all the covariates listed in the table.
†Cycle threshold values only significant in the oxygen requirement model.
CI indicates confidence interval; ICU, intensive care unit; RSV, respiratory syncytial virus; UMC, underlying medical condition.
In this large cohort of RSV-positive hospitalized children under 2 years over 3 respiratory seasons, we noted that RSV-A was the most predominant subtype over the study period, representing nearly 2 of 3 of all RSV-positive specimens. This is consistent with results from a systemic review of RSV in the MENA region which also noted that RSV-A subtype was more dominant (63%, OR 2.87) in the MENA region compared with RSV-B, except for 3 regions Iraq, Tunisia and Algeria where RSV-B was more frequent (64%, 74% and 82%, respectively).4
When comparing RSV-A and RSV-B positive children, we did not find any evidence of a difference in demographics, clinical characteristics or severity, apart from RSV-A positive children being more likely to present with decreased appetite but less likely to have co-viral detection. Even though it was not statistically significant, RSV-A children had a higher frequency of mechanical ventilation and represented all RSV-related deaths in our cohort. However, in our multivariable regression model adjusting for host and viral factors thought to affect RSV severity, there were no differences between the 2 RSV subtypes in severity score, ICU admission, or oxygen use. These results are consistent with Fodha et al,18 who showed no clinical distinction between RSV-A and RSV-B in 81 infants with a mean age of 1.9 years, who were hospitalized with a diagnosis of bronchiolitis in the Center Coast of Tunisia in 2005. In a previous study by Nidia et al,19 which randomly selected 93 of 1397 RSV-positive samples from our cohort for whole-genome sequencing, they observed that of the 85 specimens that were successfully sequenced, a higher percentage of RSV-B positive children had bronchopneumonia when compared with RSV-A positive children, but the difference was not significant, which they attributed to the small sample size. Our comprehensive study, including all RSV-positive samples from the same cohort, still found no difference in the clinical characteristics or admission diagnoses between RSV subtypes.
Other studies have shown different outcomes, with the majority reporting that RSV-A is more severe. For example, Walsh et al reported that among 265 infants who were admitted with respiratory tract infections during the winters of 1988–1991 in Rochester, NY, RSV-A positive infants had higher severity indices than RSV-B positive infants.14 In contrast, in a prospective study conducted in Copenhagen which enrolled hospitalized children less than 2 years during 3 RSV seasons (1993–1995), Hornsleth et al11 demonstrated that RSV-B led to more severe infections only among children in the 0- to 5-month age group. Furthermore, in a 3-year multicenter prospective cohort study which included children less than 2 years that were hospitalized for bronchiolitis, Laham et al20 demonstrated no significant association between RSV subtypes and severity outcomes; however, after excluding other viral codetection, RSV-A positive children had higher odds of being admitted to the ICU compared with RSV-B positive children. This suggested that viral coinfection can affect the severity of the RSV subtypes; still, in our multivariable regression analyses, neither RSV subtype nor viral coinfection was associated with RSV disease severity. Potential explanations of these variations include differences in study design (mostly retrospective), disease definition, inclusion criteria, outcome measures and/or the subtyping method.
Another plausible explanation behind the contradictory evidence observed in the literature is the difference in seasonality and disease severity between genotypes within each RSV subtypes, which are frequently not evaluated but have been observed in previous studies.21,22 For example, Martinello et al23 showed that the GA3 clade of RSV-A was associated with more severe RSV illness when compared with GA2 during the 1998–1999 and 1999–2000 winter seasons in New Haven, Connecticut. In the same study by Nidia et al. described above, they demonstrated that 55 of 58 were the GA2 genotype of RSV-A and all the RSV-B subtypes (27/27) were genotyped as GB1 in Amman, Jordan. Additionally, they observed that the emergence of the few RSV-A ON1(3/58) genotype coincided with its surfacing in the USA.19 This indicates that various genotypes of the RSV subtypes may be circulating at any given time. Further research is needed to understand the seasonality and severity of the circulating genotypes in different populations and geographic locations.
Both RSV-A and RSV-B co-circulated and peaked during the winter months in our study. This trend was also observed across the MENA region with Pakistan the only exception, where RSV peaked during the monsoon months (July–September).4 We also found that RSV-A was predominant in the first 2 respiratory seasons, whereas RSV-B became dominant in the final season in 2012–2013. Our yearly seasonal patterns were similar to those of a study completed in Brazil, which enrolled 560 children between 6 and 23 months old who presented to the pediatric emergency department with ARI between September 2009 and October 2013.24 They also observed that both RSV groups co-circulated; however, RSV-A was predominant in 2011 and 2012 while RSV-B took dominance in 2010 and 2013. In contrast to our study, their monthly seasonal patterns of RSV subtypes did not coincide. This might suggest that different climates might affect the circulation of specific subtypes.
Despite the fact that we did not note any differences by subtype in both univariable and multivariable analyses, we found that several host and virus-related factors were independently associated with RSV-disease severity. These included lower Ct value, the presence of UMC, younger age, prematurity status, lower vitamin D level and higher severity score. Our findings are consistent with previous RSV studies noting these same factors associated with severity.3,10,25 For instance, in a prospective viral surveillance study of 898 children under 5 years who presented to the emergency department with fever and/or respiratory symptoms in Nashville, TN, we also noted that lower Ct values, younger age, white race and higher severity score were independently associated with hospital admission in RSV-positive children.10
This study has some limitations. First, we are unable to generalize these findings since only hospitalized children in one hospital setting have been included. Second, we have not evaluated the association between specific RSV genotypes and disease severity. Third, while bacterial coinfection may have played a role in RSV disease severity, it was not evaluated due to the sparsity of reliable bacterial culture data. Nonetheless, the strengths of this study include the use of molecular qRT-PCR testing for RSV and RSV subtypes, and the year-round prospective design over 3 consecutive respiratory seasons. This ensured a large cohort with a myriad of different clinical presentations and characteristics. Additionally, the use of an objective severity score helped mitigate bias when interpreting severity measures affected by provider/parental preferences.
RSV subtypes co-circulated, but RSV-A was most common and peaked in 2 of the 3 respiratory seasons in hospitalized young children in Amman, Jordan. No differences in clinical presentation or disease severity were noted between them. Given that RSV-A and RSV-B co-circulate and have similar severity profiles, future preventive and treatment measures should target both subtypes.
We would like to thank our laboratory personnel for their great efforts in processing and testing specimens, and families who participated in this study. We thank our research recruiters: Hanan Amin, Amani Altaber, Hana’a Khalaf, Isra’a Kharbat, Darin Yasin, Shireen Issa, Nurse Sabah Gharbli and the doctors at Al-Bashir.
1. Nair H, Nokes DJ, Gessner BD, et al. Global burden of acute lower respiratory infections due to respiratory syncytial virus in young children
: a systematic review and meta-analysis. Lancet. 2010;375:1545–1555.
2. Shi T, McAllister DA, O’Brien KL, et al. Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in young children
in 2015: a systematic review and modelling study. Lancet. 2017;390:946–958.
3. Halasa N, Williams J, Faouri S, et al. Natural history and epidemiology of respiratory syncytial virus infection in the Middle East: Hospital surveillance for children
under age two in Jordan. Vaccine. 2015;33:6479–6487.
4. Yassine HM, Sohail MU, Younes N, et al. Systematic review of the respiratory syncytial virus (RSV) prevalence, genotype distribution, and seasonality in children
from the Middle East and North Africa (MENA) region. Microorganisms. 2020;8:E713.
5. Geoghegan S, Erviti A, Caballero MT, et al. Mortality due to respiratory syncytial virus. Burden and risk factors. Am J Respir Crit Care Med. 2017;195:96–103.
6. DeVincenzo JP. Factors predicting childhood respiratory syncytial virus severity: what they indicate about pathogenesis. Pediatr Infect Dis J. 2005;24(11 suppl): S177–S183.
7. Anderson LJ, Hierholzer JC, Tsou C, et al. Antigenic characterization of respiratory syncytial virus strains with monoclonal antibodies. J Infect Dis. 1985;151:626–633.
8. Johnson PR Jr, Olmsted RA, Prince GA, et al. Antigenic relatedness between glycoproteins of human respiratory syncytial virus subgroups A and B: evaluation of the contributions of F and G glycoproteins to immunity. J Virol. 1987;61:3163–3166.
9. Devincenzo JP. Natural infection of infants with respiratory syncytial virus subgroups A and B: a study of frequency, disease severity, and viral load. Pediatr Res. 2004;56:914–917.
10. Haddadin Z, Beveridge S, Fernandez K, et al. Respiratory syncytial virus disease severity in young children
[published online ahead of print October 23, 2020]. Clin Infect Dis. doi: 10.1093/cid/ciaa1612
11. Hornsleth A, Klug B, Nir M, et al. Severity of respiratory syncytial virus disease related to type and genotype of virus and to cytokine values in nasopharyngeal secretions. Pediatr Infect Dis J. 1998;17:1114–1121.
12. Laham FR, Mansbach JM, Piedra PA, et al. Clinical profiles of respiratory syncytial virus subtypes A AND B among children
hospitalized with bronchiolitis. Pediatr Infect Dis J. 2017;36:808–810.
13. Straliotto SM, Roitman B, Lima JB, et al. Respiratory syncytial virus (RSV) bronchiolitis: comparative study of RSV groups A and B infected children
. Rev Soc Bras Med Trop. 1994;27:1–4.
14. Walsh EE, McConnochie KM, Long CE, et al. Severity of respiratory syncytial virus infection is related to virus strain. J Infect Dis. 1997;175:814–820.
15. Khuri-Bulos N, Lawrence L, Piya B, et al. Severe outcomes associated with respiratory viruses in newborns and infants: a prospective viral surveillance study in Jordan. BMJ Open. 2018;8:e021898.
16. Tal A, Bavilski C, Yohai D, et al. Dexamethasone and salbutamol in the treatment of acute wheezing in infants. Pediatrics. 1983;71:13–18.
17. Haddadin Z, Chappell J, McHenry R, et al. Coronavirus surveillance in a pediatric population in Jordan from 2010 to 2013: a prospective viral surveillance study. Pediatr Infect Dis J. 2021;40:e12–e17.
18. Fodha I, Vabret A, Ghedira L, et al. Respiratory syncytial virus infections in hospitalized infants: association between viral load, virus subgroup, and disease severity. J Med Virol. 2007;79:1951–1958.
19. Trovão NS, Khuri-Bulos N, Tan Y, et al. Molecular characterization of respiratory syncytial viruses circulating in a paediatric cohort in Amman, Jordan [published online ahead of print September 18, 2019]. Microb Genom. doi: 10.1099/mgen.0.000292
20. Papadopoulos NG, Gourgiotis D, Javadyan A, et al. Does respiratory syncytial virus subtype influences the severity of acute bronchiolitis in hospitalized infants? Respir Med. 2004;98:879–882.
21. Rodriguez-Fernandez R, Tapia LI, Yang CF, et al. Respiratory syncytial virus genotypes, host immune profiles, and disease severity in young children
hospitalized with bronchiolitis. J Infect Dis. 2017;217:24–34.
22. Yoshihara K, Le MN, Okamoto M, et al. Association of RSV-A ON1 genotype with increased pediatric acute lower respiratory tract infection in Vietnam. Sci Rep. 2016;6:27856.
23. Martinello RA, Chen MD, Weibel C, et al. Correlation between respiratory syncytial virus genotype and severity of illness. J Infect Dis. 2002;186:839–842.
24. Bouzas ML, Oliveira JR, Fukutani KF, et al.; Acute Respiratory Infection, Wheeze Study Group Phase I, II. Respiratory syncytial virus a and b display different temporal patterns in a 4-year prospective cross-sectional study among children
with acute respiratory infection in a tropical city. Medicine (Baltimore). 2016;95:e5142.
25. Banajeh SM. Nutritional rickets and vitamin D deficiency-association with the outcomes of childhood very severe pneumonia: a prospective cohort study. Pediatr Pulmonol. 2009;44:1207–1215.