A Prospective Cohort Study on Spectrum of Community Acquired Late Onset Neonatal Pneumonia: An Eye Opener for Neonatal Care

Abstract

Background

Acute lower respiratory tract infections (ALRTIs) are a leading cause of morbidity and mortality in infants in low- and middle-income countries. We emphasize the differences in presentation and outcome between viral and bacterial origin community acquired ALRTI in late neonatal period.

Methods

All consecutively admitted neonates, aged 7 days or older, with features of ALRTI were included in the study, other than babies referred with prior hospital stay, congenital airway malformations and heart disease. Relevant investigations including nasopharyngeal swab polymerase chain reaction (PCR) and blood culture were done for etiological diagnosis. The predisposing factors, presenting features and short-term outcomes were compared between viral and bacterial etiology cohorts.

Results

Half of the 89 babies had primary viral infections, respiratory syncytial virus (RSV) being predominant (n = 21, 23.6%); 27 (30.3%) had bacterial infections, while no organisms were identified in 18% cases. Ten babies (11.2%) had COVID-19 and 4 (4.5%) had pertussis. Small for date (n = 15, 55.5%, p < .001) and history of past sick neonatal care unit admission were significantly higher (n = 20, 74.1%, p = .009) in babies with bacterial isolates. Median age of presentation was day 10 (interquartile range [IQR] 8-12). Bacterial ALRTI was more severe with significantly high respiratory score (p < .001), need for invasive ventilation (p < .001), systemic complications, and mortality (p = .012), while viral ALRTI group required longer respiratory support and hospital stay (p < .001).

Conclusions

PCR based methods are important in etiological diagnosis of community acquired late onset neonatal pneumonia. Such ALRTI is mostly of viral origin, and this is less likely to be fatal but require longer hospital stay than bacterial pneumonia.

Keywords

Acute lower respiratory infection neonate PCR pneumonia

Introduction

Acute lower respiratory tract infections (ALRTIs) are a leading cause of morbidity and mortality in infants of low to middle income countries.¹ In these countries, LRTI accounts for around 33% of under-five deaths, 20% of deaths in children till 1 year of age, and 6.8% of neonatal deaths. Pneumonia is the most serious presentation and is responsible for almost one-fifth of total mortality during childhood.² According to WHO estimates, 14% of under-5 deaths in India were attributed to ALRTI.³

Neonatal pneumonia occurring prior to 7 days of age is considered as early onset and at or after 7 days as late onset pneumonia.^4–6 Early onset pneumonia usually occurs in the setting of a maternal infection, whereas the epidemiology of late-onset neonatal pneumonia is nosocomial or community acquired.⁵ Global estimates of viral infection in infants of low- and middle-income countries highlights the importance of maternal vaccination as a proposed intervention for policy makers.⁷ Withholding and early stopping of antibiotics in cases of nonbacterial etiology infections, especially in absence of clinical suspicion of bacterial sepsis, strengthens the antibiotic stewardship policy in neonatal care.⁸ Therefore, assessment of etiology of late onset ALRTI is need of the hour against the dogmatic belief that bacteria always cause such pneumonia, especially for <2-month-old children. In the Indian context, there is a paucity of data regarding etiological agents of community acquired ALRTI in neonates.

Our study attempts to fill in the knowledge gaps related to age of presentation, etiology, and clinical spectrum of late onset community acquired pneumonia in neonates, especially in the tropical setting, and to compare the outcome among major etiological cohorts, namely viral versus bacterial.

Materials and Methods

This prospective observational study was conducted in a tertiary care neonatal unit in Kolkata. Institutional ethics committee clearance was obtained beforehand, and babies were recruited only if parental consent was obtained after explaining the study rationale and procedures in local language. All infants meeting selection criteria were included in the prespecified time frame of the study from February to October 2021.

Age for inclusion was 7 days to < 28 days of life at onset. All babies presented with ALRTI⁹ defined as presence of respiratory distress and any 2 features among tachypnoea (respiratory rate > 60/minute), subcostal/intercostal retractions, and expiratory grunt/groaning with or without cough, fever, wheeze, stridor, apnea, or cyanosis, were enrolled if parental consent was obtained. Downe score was used as respiratory distress severity score. All babies underwent chest X-ray (CXR) and screening echocardiography. The CXR was considered abnormal if showing lobar or segmental consolidation, nodular or course patchy infiltrates, diffuse haziness or granularity, air bronchogram, atelectasis, hyperinflation, and radiological findings persisting after 48 hours.¹⁰ Neonates brought from home or out-patient departments were included. Babies admitted to and then referred from other hospitals, babies with respiratory distress attributable to congenital airway/diaphragmatic malformation or congenital heart disease were excluded. The Montreux definition of neonatal ARDS was used to label a case as acute respiratory distress syndrome (ARDS).¹¹ The babies were followed up till discharge or death.

Sepsis screen and blood culture were done at admission. Within 24 hours of admission, nasopharyngeal swab was collected and transported to the laboratory as soon as possible in 5 mL viral transport medium tube.¹² The specimens were held at 4°C in the laboratory. Total nucleic acid extraction kit (Spin Star Viral Nucleic Acid Extraction kit ADT) was used for viral DNA-RNA extraction. The sample were eluted with 50 µL elution buffer provided by the extraction kit. The RespiFinder 2SMART assay used was based on the Smart Finder technology that allows a highly complex analysis of up to 13 targets in a single PCR reaction. The assay contains 23 different 2SMART primer sets combined with 15 fluorescent labeled SMART probes which allows the detection of 22 different pathogens plus an internal control (IC) and 2 amplification controls (AC). Analysis started with pre-amplification which combines a reverse transcription step with a polymerase chain reaction (PCR) step to amplify the target cDNA. Subsequently, a part of the pre-amplification reaction product was transferred to 2 PCR tubes. Two separate SmartFinder reactions were performed. The detection was performed using a melting curve analysis. An IC was included in the assay to discriminate between true and false negative samples due to nucleic acid degradation, PCR inhibition, or test failure. RespiFinder 2SMART additionally contains 2 AC which enables a contingent discrimination between extraction failure and amplification failure in the assay. The multiplex reverse transcriptase PCR (RT-PCR) was performed in 2 steps using the RespiFinder kit for Respiratory Pathogens for the detection of Influenza A & B (Inf A & B); Inf A (H1N1) pdm 09; Respiratory syncytial virus (RSV A/B); Human rhinovirus; Human adenovirus; Human parainfluenza virus 1, 2, 3, 4; Human coronavirus OC43, HKU1, NL63, 229E; Human metapneumovirus (hMPV A/B); Human bocavirus (hBoV A/B); Enterovirus (EV); Chlamydia pneumoniae, Mycoplasma pneumoniae, and Bordetella pertussis as per the manufacturer’s instruction. The total reaction volume for each test of first PCR and second PCR was 20 µL (15 µL reagent mix and 5 µL extracted nucleic acid for first PCR and 15 µL reagent mix with 5 µL first PCR product or cDNA for second PCR). Amplification was performed in the Rotor-Gene Q real time PCR machine. For a correct identification of the smart probes the PCR was followed directly by a melting program which was performed between 40°C and 90°C. The positive control virus was provided by the kit. Interpretation of the melting curve was done manually.

As per unit protocol, endotracheal tube aspirate for culture was sent if we failed to wean and extubate even on day 7 of mechanical ventilation. If complicated with empyema, pleural fluid analysis including culture was done wherever diagnostic tap was possible under ultrasonography guidance. All the babies also underwent cerebrospinal fluid (CSF) analysis by lumbar puncture in view of late onset sepsis.

Viral ALRTI was labeled on basis of PCR based microbiological criteria. Bacterial ALRTI was defined as ALRTI (clinical criteria) with any 2 of the following clinical features of bacterial sepsis other than respiratory distress such as poor feeding and lethargy, isolation of bacteria either in blood, CSF or other body fluids and positive sepsis screen. The remaining neonates with clinical features of ALRTI but without fulfilling viral or bacterial ALRTI criteria were included in unidentified group. CXR changes were not mandatory to label as ALRTI.

Statistical analysis: Data have been summarized by descriptive statistics, namely mean and standard deviation for numerical variables that are normally distributed, median and IQR for skewed numerical variables and counts, and percentages for categorical variables. Kruskal-Wallis ANOVA or Mann-Whitney U test was done for comparison of numerical variables between subgroups. Fisher’s exact test or Pearson’s chi-square was employed for intergroup comparison of categorical variables. All analyses were two-tailed and statistical significance was set at p < .05 for all comparisons.

Results

Total 109 neonates were admitted in our hospital during the study period with late onset respiratory distress, of whom 89 fulfilled inclusion criteria (Figure 1). Around half of them had primary viral infections, one-third bacterial (Table 1). The most common viral infection was RSV infection (n = 21, 23.6%). Ten babies (11.2%) had COVID-19 and 4 (4.5%) had pertussis. Isolates in different cultures and PCR reports have been depicted in Table 1. Demography has been depicted in Table 2.

Table 1.

Etiology of Community Acquired Late Onset Pneumonia in Study Population.

Etiology	Count (N = 89)
Viral (%)
Viral isolates (%)^a	46 (51.7)
RSV	21 (23.6)
Rhino	15 (16.8)
COVID	10 (11.2)
Bacterial (%)
Bacterial isolates(%)^a	27 (30.3)
Only blood culture positive	15 (55.5)
MRSA	6 (22.2)
Acinetobacter	5 (18.5)
Enterococcus	2 (7.4)
Klebsiella	2 (7.4)
Only CSF culture positive	8 (29.6)
MRSA	3 (11.1)
Klebsiella	3 (11.1)
Acinetobacter	2 (7.4)
Nasopharyngeal swab PCR positive for pertussis	4 (14.8)
Both blood + other (CSF/empyema/ET aspirate) culture positive	6 (22.2)
Pus (empyema fluid) culture positive (same as grown in blood)	4 (14.8)
Klebsiella	2 (7.4)
MRSA	2 (7.4)
ET culture positive # (same as grown in blood)	2 (7.4)
Enterococcus	1 (3.7)
Acinetobacter	1 (3.7)
Unidentified (%)	16 (17.9)

Notes: Data in no (%).

^a Bacterial Isolates in % calculated out of total bacterial cases (n = 27).

Abbreviations: RSV, respiratory syncytial virus; COVID, coronavirus disease; MRSA, methicillin resistant Staphylococcus aureus; CSF, cerebrospinal fluid; PCR, polymerase chain reaction; ET, endotracheal tube.

Table 2.

Comparison by Etiology of Demographic Parameters and Risk Factors of Babies with Acute Lower Respiratory Tract Infection.

Parameter	Viral (n = 46)	Bacterial (n = 27)	Unidentified (n = 16)	p Value
Gestational age (week)^a	35 (34, 37)	34 (32, 37)	35 (33, 37)	.274
Birth weight (kg)^a	2 (1.9, 2.5)	1.8 (1.6, 2.1)	1.95 (1.4, 2.4)	.025
Male	16 (34.7)	19 (70.3)	10 (62.5)	.008
Small for gestational age	9 (19.5)	15 (55.5)	10 (62.5)	<.001
Readmission	17 (36.9)	20 (74.0)	8 (50.0)	.009
Formula feeding	12 (26)	11 (40.7)	2 (12.5)	.314

Notes: Data in no (%).

^aMedian (IQR).

Both viral and bacterial pneumonia presented in the second week of life, with median age being day 10 (IQR 8-12) (Table 3). The most common symptom in the study population was tachypnoea except neonatal pertussis where 50% of the babies had recurrent apnea as a predominant feature. In babies with bacterial isolates, respiratory distress was more severe with significantly high Downe score (p < .001) and there was increased need of invasive ventilation (p < .001), although incidence of ARDS was similar in all groups. Viral ALRTI group required longer respiratory support and hospital stay (p < .001). Antibiotic duration was significantly short (median 5 days) in viral ALRTI group without any adverse consequence like increased mortality or emergence of hospital acquired sepsis.

Table 3.

Comparison by Etiology of Course of Disease and Outcome of Babies with Acute Lower Respiratory Tract Infection.

Features	Viral (n = 46)	Bacterial (n = 27)	Unidentified (n = 16)	p Value
Clinical course
Onset of respiratory distress (days)^a	10 (10, 12)	10 (8, 12)	8.5 (8, 12)	0.252
Respiratory distress score^a	5 (5, 6)	7 (7, 7)	5 (4.5-5.5)	<0.001
Fever	17 (36.9)	9 (33.3)	4 (25)	<0.001
Cough	31 (67.3)	9 (33.3)	1 (6.25)	<0.001
Rhinorrhea	36 (78.3)	9 (33.3)	1 (6.25)	<0.001
Apnea	7 (15.2)	5 (18.5)	6 (37.5)	0.155
Wheeze	16 (34)	8 (29)	3 (18)	0.140
ARDS	7 (15.2)	6 (22.2)	2 (12.5)	0.651
Shock	11 (23.9)	22 (81.4)	4 (25)	<0.001
Seizure	3 (6.5)	16 (59.2)	0	<0.001
Acute kidney injury	9 (19.5)	9 (33.3)	0	0.031
MODS	13 (28.2)	24 (88.8)	4 (25)	<0.001
Respiratory complication				0.004
Air leak	12 (26)	4 (14.8)	0
Empyema	0	4 (14.8)	0
Maximum respiratory support				<0.001
MV	11 (23.9)	22 (81.4)	2 (12.5)
NIV	8 (17.4)	4 (14.8)	2 (12.5)
CPAP	17 (36.9)	1 (3.7)	8 (50)
HFNC	10 (21.7)	0	4 (25)
Investigations
Neutropenia	1 (2.2)	16 (59.3)	2 (12.5)	<0.001
CRP (mg/dL)	1.8 (0.8, 2.3)	25.9 (12.3, 46.7)	1.35 (1.05, 2.2)	<0.001
Chest X-ray finding				<0.001
Reticulogranular pattern	37 (80.4)	1 (3.7)	12 (75)
Patchy infiltrate	9 (19.5)	14 (51.8)	4 (25)
Lobar consolidation	0	12 (44.4)	0
Outcome
Death	3 (6.5)	7 (25.9)	0	0.012
Duration respiratory support (days)^a	30 (5, 8)	21 (12, 5)	15 (3.5, 7.5)	<0.001
Duration antibiotic use (days)^a	5 (3, 3)	21 (14, 18)	10 (5, 7)	<0.001
Duration hospital stay (days)^a	35 (5, 10)	23 (15, 20)	16 (6, 10)	<0.001

Notes: Data in no (%).

^aMedian (IQR).

Abbreviations: ARDS, acute respiratory distress syndrome; MODS, multiorgan dysfunction syndrome; MV, mechanical ventilation; NIV, noninvasive ventilation; CPAP, continuous positive airway pressure; HFNC, high flow nasal canula; CRP, C-reactive protein.

Systemic complications and mortality were significantly higher in bacterial group while air leak (12, 26%, p = .004) was found to be more in viral pneumonia. Out of 4 neonatal pertussis cases, 1 died. Two primary viral ALRTI cases and one bacterial ALRTI case developed second blood culture proven staphylococcal sepsis later during hospital stay.

Discussion

The features of ALRTI in young infants from eastern India are not well documented. This prompted use-to-use modern diagnostics in the form of PCR for identification of etiological agents of neonatal ARTI and fill related knowledge gaps.

Viruses are reported worldwide, including in developing countries like India, to account for up to 80% of infections in young children, with RSV and rhinoviruses being the most frequently isolated.^{13, 14} This corroborates with our findings where around 50% cases of ALTRI turned out to be of viral origin with RSV being the most common. This suggests that maternal antibody against RSV is probably not very protective in early age infants. Traditionally, etiological diagnosis of neonatal pneumonia has been based primarily on blood culture positivity, which fails to yield organism other than bacteria in large number of cases.^{4, 10} However in our study, etiologies were detected in 81% cases because of use of PCR based diagnosis. Inclusion of neonates throughout the year (which would have covered the winter seasonal peak) and vulnerable infants such as those with congenital airway malformations would probably have led to a higher viral etiology yield.

Bacterial pneumonias are not uncommon in the neonatal period. Escherichia coli was found to be the most common organism followed by RSV among young infants less than 3 months, hospitalized with pneumonia, in a prospective Indian study decades ago.¹⁵ In infants during first month of life Staphylococcus aureus has been identified as a significant cause of community acquired pneumonia in literature specially in developing countries,⁴ supporting our finding. Neonatal pertussis presenting as late onset pneumonia is a unique finding in this study. Clinical diagnosis of pertussis in the neonatal population is stated to be challenging^{16, 17} as its course often does not follow the classic stages; instead, pertussis pneumonia in neonates often presents with apneic episodes like in our case.

Presenting clinical signs-symptoms of viral and bacterial pneumonias in early infancy are almost similar as per literature although wheeze and apnea more common in viral pneumonia.¹⁸ In contrast we found fever, cough, rhinorrhea to be predominant in viral group, while sepsis and multisystem involvement like acute kidney injury and shock predominantly in bacterial group. However, babies with pertussis presented more with apnea and wheeze, resembling viral etiology. Tachypnoea (100%) was the most common clinical feature followed by chest retractions (88%) in our study like that by Duke et al⁴ and Rahman et al.¹⁹ The CXR findings in our study were quite varied with diffuse alveolar infiltration (56%) being the most common finding. This is similar to findings reported by Mathur et al.³

Mortality in late onset neonatal community acquired pneumonia had been reported as around 15%⁴ which fairly matches our figure of 11%. The mortality figures of bacterial pneumonia in under 5 population²⁰ are similar.

Median age of presentation of our study population was day 10 (range 8-19 days), the earliest age of presentation in recent studies. Median age of ALRTI was reported as 6 months (range 0.5-12 months) during longer follow up of birth cohorts till the end of first year of life.^{21, 22}

Male sex, lower birth weight, small for gestation, and previous sick neonatal care admission were found to be significant associations with bacterial ALRTI like that found in population-based surveillance study for hospitalization due to ALRTI.²³

To the best of our knowledge, this is the first study from eastern India that has investigated etiological profile and clinical spectrum of community acquired pneumonia in neonatal period with the help of an advanced diagnostic tool like multiplex PCR. It is an eye opener for the neonatologists to think of viral respiratory infection, community acquired, even in the second week of life of the neonate. This emphasizes a further reason for rationalizing antibiotic use in neonatal units. PCR based diagnostics can also lead to reduction in the duration of empirical antibiotic use in such context.

Our study has limitations. It reflects a single center experience. The sample size is relatively small, and the study was not conducted throughout the year which means seasonal trend and peak information are lacking. Finally, this is a hospital-based study and therefore cannot provide a true estimate of bacterial and viral burden in ALRTI in the community.

Notwithstanding these limitations, we can conclude that PCR based methods are important in etiological diagnosis of community acquired late onset neonatal pneumonia. Such ALRTI is often of viral origin, and this is less likely to be fatal but require longer hospital stay than bacterial pneumonia.

Footnotes

Acknowledgments

All the nursing staff, resident doctors, and support staff of the unit involved in the care of neonates are gratefully acknowledged for their dedicated service and cooperation with the study team.

Author Contributions

Somosri Ray has substantial contribution in design of the work, interpreting the data, revising it critically for important intellectual content, final approval of the version to be published, and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work.

Dinesh Munian has substantial contribution in acquisition, analysis of data, drafting the work, final approval of the version to be published, and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work.

Oishik Roy has substantial contribution in conception, analysis of data, critical revision of the work, final approval of the version to be published, and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work.

Sumon Poddar has substantial contribution in design of the work, interpreting the data (PCR based), revising it critically for important intellectual content, final approval of the version to be published, and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work.

Avijit Hazra has substantial contribution in analysis of data, critical revision of the work, final approval of the version to be published, and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work.

Data Availability

All authors ensure data transparency.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical Approval

Ethical clearance was obtained from the Institutional Ethics Committee, JNMC Institutional Committee on Human Subjects Research, J.N. Medical College, Belagavi, Ref: MDC/DOME/149. The trial has been registered under Clinical Trial Registry India (CTRI-ICMR NIMS): CTRI/2022/07/043888.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Informed Consent

Informed consent was obtained from one of the parents of enrolled neonates included in the study. Parents also gave consent for publication.

ORCID iD

Somosri Ray

References

Feddema

, van der Geest

, Claassen

, van de Burgwal

LHM

. Lower Respiratory Tract Infection Trends in East and South-East Asia: in the light of economic and health care development. Glob Pediatr Health . 2021 Jan 24;8: 2333794X21989530.

Lozano

, Naghavi

, Foreman

, . Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the global burden of disease study 2010. Lancet . 2012;380: 2095–128.

World Health Organization (WHO). Cause Specific Mortality and Mortality in World Health Statistic . WHO; 2014.

Duke

. Neonatal pneumonia in developing countries. Arch Dis Child Fetal Neonatal Ed . 2005;90: F211–F219.

Nissen

. Congenital and neonatal pneumonia. Paediatr Respir Rev . 2007; 8: 195–203.

Hooven

, Polin

. Pneumonia. Semin Fetal Neonatal Med . 2017; 22: 206–213.

Wang

, Li

, Deloria-Knoll

, . for Respiratory Virus Global Epidemiology Network. Global burden of acute lower respiratory infection associated with human metapneumovirus in children under 5 years in 2018: a systematic review and modelling study. Lancet Glob Health . 2021; 9: e33–e43.

Green

, Zar

, White

, Madhi

. Viral lower respiratory tract infections. In: Green

, ed. Viral Infections in Children . Vol II. Springer International Publishing AG; 2017: 27–56.

Wright

, Cutt

for World Health Organization, Department of Vaccines and Biologicals. Generic protocol to examine the incidence of lower respiratory infection due to respiratory syncytial virus in children less than five years of age [Internet]. 2000. Accessed Feb 14, 2022. https://apps.who.int/iris/bitstream/handle/10665/66276/WHOVandB00.08eng.pdf

10.

Mathur

, Garg

, Kumar

. Respiratory distress in neonates with special reference to pneumonia. Indian Pediatr . 2002; 39: 529–537.

11.

De Luca

, van Kaam

, Tingay

, . The Montreux definition of neonatal ARDS: biological and clinical background behind the description of a new entity. Lancet Respir Med . 2017; 5: 657–666.

12.

Olofsson

, Brittain-Long

, Andersson

, Westin

, Lindh

. PCR for detection of respiratory viruses: seasonal variations of virus infections. Expert Rev Anti Infect Ther . 2011; 9: 615–616.

13.

Sonawane

, Shastri

, Bavdekar

. Respiratory pathogens in infants diagnosed with acute lower respiratory tract infection in a tertiary care hospital of western India using multiplex real time PCR. Indian J Pediatr . 2019; 86: 433–438.

14.

Gilani

, Kwong

, Levine

, . A literature review and survey of childhood pneumonia etiology studies: 2000-2010. Clin Infect Dis . 2012; 54 Suppl 2: S102–S108.

15.

Patwari

, Bisht

, Srinivasan

, Deb

, Chattopadhya

. Aetiology of pneumonia in hospitalized children. J Trop Pediatr . 1996; 42: 15–20.

16.

Swamy

, Wheeler

. Neonatal pertussis, cocooning and maternal immunization. Expert Rev Vaccines . 2014; 13: 1107–1114.

17.

Castagnini

, Munoz

. clinical characteristics and outcomes of neonatal pertussis: a comparative study. J Pediatr . 2010;156: 498–500.

18.

World Health Organization Fact Sheet: Pneumonia [Internet]. https://www.who.int/news-room/fact-sheets/detail/pneumonia. Sep 28, 2022.

19.

Rahman

, Sarkar

, Hye

, Das

, Mostafa

, Ahmed

. Clinical profile of neonatal pneumonia in NICU of a secondary care center. J Med Sci Res . 2019; 7: 75–79.

20.

Bénet

, Picot

, Awasthi

, . for The Gabriel Network. Severity of pneumonia in under 5-year-old children from developing countries: a multicenter, prospective, observational study. Am J Trop Med Hyg . 2017; 97: 68–76.

21.

Regamey

, Kaiser

, Roiha

, . for Swiss Paediatric Respiratory Research Group. Viral etiology of acute respiratory infections with cough in infancy: a community-based birth cohort study. Pediatr Infect Dis J . 2008; 27: 100–105.

22.

Kumar

, Medigeshi

, Mishra

, . Etiology of acute respiratory infections in infants: a prospective birth cohort study. Pediatr Infect Dis J . 2017;36: 25–30.

23.

Iwane

, Edwards

, Szilagyi

, . for New Vaccine Surveillance Network. Population-based surveillance for hospitalizations associated with respiratory syncytial virus, influenza virus, and parainfluenza viruses among young children. Pediatrics . 2004;113: 1758–1764.