b r a z j i n f e c t d i s .

2 0 1 5;1 9(4):358–362

The Brazilian Journal of

INFECTIOUS DISEASES
www.elsevier.com/locate/bjid

Original article

Parainfluenza virus as a cause of acute respiratory

infection in hospitalized children夽

Rogério Pecchini a , Eitan Naaman Berezin a,∗ , Maria Cândida Souza b ,

Lourdes de Andrade Vaz-de-Lima b , Neuza Sato b , Maristela Salgado b ,
Mirthes Ueda b , Saulo Duarte Passos c , Raphael Rangel d , Ana Catebelota d
a Medical Shool, Santa Casa de São Paulo, São Paulo, SP, Brazil
b Center for Immunology, Instituto Adolfo Lutz, São Paulo, SP, Brazil
c Faculdade de Medicina de Jundiaí, Jundiaí, SP, Brazil
d Irmandade da Santa Casa de Misericórdia de São Paulo, SP, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: Background: Human parainfluenza viruses account for a significant proportion of lower respi-
Received 27 December 2014 ratory tract infections in children.
Accepted 31 March 2015 Objective: To assess the prevalence of Human parainfluenza viruses as a cause of acute respi-
Available online 25 April 2015 ratory infection and to compare clinical data for this infection against those of the human
respiratory syncytial virus.
Keywords: Methods: A prospective study in children younger than five years with acute respiratory
Parainfluenza infection was conducted. Detection of respiratory viruses in nasopharyngeal aspirate sam-
Paramyxoviridae infections ples was performed using the indirect immunofluorescence reaction. Length of hospital
Respiratory tract infections stay, age, clinical history and physical exam, clinical diagnoses, and evolution (admission
Respiratory virus to Intensive Care Unit or general ward, discharge or death) were assessed. Past personal
(premature birth and cardiopathy) as well as family (smoking and atopy) medical factors
were also assessed.
Results: A total of 585 patients were included with a median age of 7.9 months and median
hospital stay of six days. No difference between the HRSV+ and HPIV+ groups was found
in terms of age, gender or length of hospital stay. The HRSV+ group had more fever and
cough. Need for admission to the Intensive Care Unit was similar for both groups but more
deaths were recorded in the HPIV+ group. The occurrence of parainfluenza peaked during
the autumn in the first two years of the study.
Conclusion: Parainfluenza was responsible for significant morbidity, proving to be the second-
most prevalent viral agent in this population after respiratory syncytial virus. No difference
in clinical presentation was found between the two groups, but mortality was higher in the
HPIV+ group.
© 2015 Published by Elsevier Editora Ltda.

夽
Approval Number in Internal Review Board (Research Ethical Committee) – 064/09.
∗
Corresponding author.
E-mail address: eberezin2003@yahoo.com (E.N. Berezin).
http://dx.doi.org/10.1016/j.bjid.2015.03.002
1413-8670/© 2015 Published by Elsevier Editora Ltda.
 b r a z j i n f e c t d i s . 2 0 1 5;1 9(4):358–362 359

recurrent wheezing, while acute crisis was characterized as

Introduction acute wheezing. Cases with crackles on pulmonary auscul-
tation and lung consolidation on chest radiography exams
Acute respiratory tract infections (ARTIs) in childhood are
were considered alveolar pneumonia, whereas cases with dif-
a major cause of morbidity and mortality, with pneumonia
fuse infiltration and with crackling on pulmonary auscultation
being the leading cause of death among children between one
were diagnosed as non-alveolar pneumonia.
and five years old.1,2 Viruses are the predominant etiology of
All patients included in the study had samples of nasopha-
ARTIs in this pediatric group.3–5
ryngeal secretion collected within the first 24 h of admission.
The human parainfluenza viruses (HPIVs) account for a
Samples were then stored on ice and sent to the Adolfo Lutz
significant proportion of lower respiratory tract infections in
institute for Indirect Immunofluorescence (IIF), performed
children, where their rates of detection vary with pathology
using a panel of five monoclonal antibodies specific for HRSV,
(respiratory infection of the upper or lower tract) and with the
Influenza A and B, HPIV 1, 2 and 3, and Adenovirus, for the
investigation setting (outpatient units or hospital wards).6
detection of seven viruses (Kit Light Diagnostic TM, Chemicon
With the attention of parents and medical professionals
International Inc, Temecula, USA).
firmly focused on annual outbreaks of respiratory syncy-
Variables analyzed included age, gender, admission date,
tial virus (HRSV) and influenza, HPIV infections can be
clinical characteristics (cough, fever, shortness of breath, and
overlooked.7 However, HPIVs are the second most common
apnea), data on physical exam (wheezing, dyspnea, cyanosis,
cause of lower and upper respiratory tract infections in chil-
inspiratory nasal wing collapse), clinical diagnoses, personal
dren younger than five years after HRSV.8–11
(cardiopathy, premature birth, and gastroesophagic reflux)
This prospective study was conducted to determine the
and family (smoking and atopy) medical histories, patient hos-
prevalence of HPIVs in children hospitalized for ARTI, assess
pital ward (general or ICU), and outcome (discharge or death).
the clinical characteristics of HPIV+ and HRSV+ patients, and
Patients were categorized into two groups (HPIV+ or HRSV+
to compare the two groups identifying any specific features of
on IIF) and subsequently compared. Medians of quantitative
infections caused by HPIVs. Seasonality was also examined in
variables were compared using the Mann–Whitney test. Qual-
the present study.
itative variables were analyzed by comparing proportions in
each group using the Chi-square or Fisher’s exact tests as
applicable. The level of significance was set at 0.05. All sta-
Methods tistical tests were performed using the Sigma Stat program
version 3.5.
This prospective study included all children with ARTI
younger than five years with prodromes of upper respira-
tory tract infections (URTIs) on clinical history, admitted to Results
the hospital general wards or Intensive Care Unit (ICU) of the
Department of Pediatrics of Santa Casa de São Paulo Hospital A total of 585 patients were included in the study. HPIVs
(São Paulo city), between February 2005 and May 2007. After were detected in 45 patients, corresponding to 19.9% of all
2007, surveillance continued but using other methodology for positive cases, and 7.7% of children overall. The HRSV was
diagnosing viral infections. Immunosuppressed patients like the most frequently detected viral agent (26.5%). IIF was pos-
HIV-infected, and those with neoplasia or transplantions were itive for influenza in 3.42% of patients and adenovirus in
excluded. 1%. The clinical data of the HPIV-positive patients was com-
The Central Hospital at Santa Casa de São Paulo is a refer- pared against data from HRSV-positive subjects (see Table 1).
ence university hospital for the central and northern region of A total of 44.4% of HPIV positive patients were female ver-
the city providing high complexity care, including emergency sus 43.9% in the HRSV-positive group. Median age was similar
care. Approximately 60,000 patients receive medical atten- among HPIV+ and HRSV groups, 6.8 and 7.1 months, respec-
tion annually at urgency and emergency units. Out of 9500 tively. Length of hospital stay (median six days) was also
admissions during the study period, around one-third was due similar in the two groups. Comparing signs and symptoms
to respiratory infections. Data on clinical characteristics and of acute lower respiratory tract infection between HPIV+ and
demographics were collected at patient admission. The study HRSV+ groups revealed greater proportion of cough (96.1%
was approved by the Research Ethics Committee of the Hos- vs 82.2%, p < 0.05) and fever (72.3% vs 48.9%, p < 0.05) in the
pital under the number 064/09. Prior to inclusion in the study, HRSV group. No statistically significant difference was found
parents or legal guardians of the patients signed a free and for the other signs and symptoms examined, including short-
informed consent form. ARTI was defined as the presence of ness of breath, wheezing, dyspnea, and cyanosis. The most
two or more signs of inferior respiratory tract involvement: frequent diagnoses in the HPIV+ group were acute wheezing
respiratory dysfunction, characterized by fatigue, tachyp- (28.9%), non-alveolar pneumonia associated to acute wheez-
nea, wheezing, crackling, and/or fever (axillar temperature ing (26.7%), and bronchiolitis (22.2%). In the HRSV+ group
of 37.8 ◦ C) on physical exam. The prodromes of URTIs were diagnoses were bronchiolitis (35.5%), non-alveolar pneumonia
characterized by the presence of coryza, sneezing, and/or associated with acute wheezing (27.1%), and acute wheezing
nasal obstruction. Disease history was considered in the last (14.5%). Both groups were similar with regard to past personal
five days. Bronchiolitis was characterized by first episodes of medical history of premature birth, congenital cardiopathy,
wheezing in infants with URTI prodromes; children with pre- and gastroesophagic reflux disease, as well as for family his-
vious episodes of pulmonary wheezing were diagnosed as tory of atopy and smoking (Table 1). Of the patients in the
360 b r a z j i n f e c t d i s . 2 0 1 5;1 9(4):358–362

greater sensitivity for detecting HPIVs over culture and IIF

Table 1 – Clinical characteristics associated with
infections by human parainfluenza viruses (HPIVs) and techniques,19,20 but with higher cost.
respiratory syncytial virus (HRSV) (n = 585). Earlier studies have reported higher prevalence of HPIVs
than in the present study but had different study designs and
Clinical characteristics PIVs HRSV
(n = 45) (n = 155) viral detection methods. For instance, the study by Rihkanen
et al., investigating the etiology of croup in patients attend-
Female gender 20 (44.4%) 68 (43.9%) ing emergency services, found HPIVs, particularly type 1, in
Median age (months) 6.8 7.1
29.1% of patients.21 HPIV1 is known to be the main etio-
Length of hospital stay (days) 6 6
logic agent of croup, accounting for 56–74% of such cases. A
Signs and symptoms of ARTI cohort study conducted in Spain, involving patients hospi-
Cough 37 (82.2%) 149 (96.1%)a
talized for ARTI aged up to 14 years using PCR as the viral
Fever 22 (48.9%) 112 (72.3%)b
detection method, showed a 9.9% HPIV prevalence, ranking
Shortness of breath 28 (62.2%) 115 (74.2%)
Wheezing 30 (66.7%) 85 (54.8%) third among the most frequent viruses after the HRSV, Rhi-
Dyspnea 27 (60%) 96 (61.9%) novirus and Adenovirus.6 In a study aiming to determine the
Cyanosis 4 (8.9%) 4 (2.58%) etiology of community-acquired pneumonia in hospitalized
ARTI diagnoses patients aged younger than five in Salvador city in northeast-
Acute wheezing 13 (28.9%) 23 (14.5%) ern Brazil, HPIVs ranked second among the most prevalent
Non-alveolar pneumonia 12 (26.7%) 42 (27.1%) agents, followed by Rhinovirus.22
Bronchiolitis 10 (22.2%) 55 (35.5%) The prevalence of HPIV-3 was substantially higher than
Personal medical history HPIV-1 and 2 in this study, whereas HPIV-4 was not stud-
Premature birth 6 (13.3%) 13 (8.4%) ied. It is known that HPIV-3 is the HPIV type most frequently
Congenital cardiopathy 4 (8.9%) 8 (5.2%) detected in hospitalized patients, particularly among young
Gastroesophagic reflux 5 (11.1%) 11 (7.1%) infants, and associated with the diagnoses of bronchiolitis and
Family medical history pneumonia.12,23
Atopy 20 (44.4%) 56 (36.1%) Other studies have shown HPIV-3 to be the most prevalent
Smoking 6 (13.3%) 20 (12.9%) in hospitalized patients, such as the work by Calvo et al.6 in
ICU admission 7 (15.5%) 17 (10.9%) addition to studies conducted in Brazil by Fe et al.24 assessing
Death 4 (8.9%) 2 (1.3%)c the prevalence of HPIVs in patients up to 16 years old with
ARTI, and the study by Veja-Briceño et al. conducted in Chile.25
a
p = 0.0037.
b
Evidence confirms that HPIV-1 is associated with the diagnosis
p = 0.0058.
c
p = 0.024.
of croup, but not a single case of croup or laryngitis associated
to croup was found in our study, perhaps because our cases
were often less severe, not requiring hospitalization.
HRSV+ group, 10.9% were ICU cases compared with 15.5% in Comparing the clinical data of HPIVs and HRSV infected
the HPIV+ group (Table 1). A greater number of deaths were groups, we found no statistically significant differences in
recorded in the HPIV+ group compared with the HRSV+ group terms of hospital stay or need for ICU admission. With regard
(8.9% vs. 1.29%, p < 0.05) (Table 1). to signs and symptoms of ARTI, there were greater proportions
Of the HPIVs detected, the most frequent type was HPIV-3 of cough and fever in the HRSV-infected group. Despite this
(84.4%) followed by HPIV-2 (8.89%) and HPIV-1 (6.67%). association, because signs and symptoms were very unspe-
The peak detection rate of the HRSV occurred in the cific and general, none of these could be attributed to a specific
autumn season during the study period. In the first two years agent.25,26
of the study, a higher incidence of HPIVs was observed accom- The highest rates of HPIV isolation (detection) occurred
panying the HRSV peaks (Fig. 1). during autumn in the first two years of the study, coinciding
with peak incidence of the HRSV. The classic seasonal pattern
of HPIV infection is described in the United States, with bian-
Discussion nual peaks of HPIV1 during fall and winter, and annual peaks
of HPIV3, particularly at the end of winter or in spring.9,27 Stud-
The HPIVs are important etiologic agents of acute respi- ies in tropical countries are scarce, although one investigation
ratory infection in childhood and rank second among the reported seasonal peaks of HPIV3 at the end of winter and in
most common diseases in this age group by some authors. spring in Rio de Janeiro,28 while other studies carried out in
HPIVs predominantly cause lower respiratory tract infection Northeastern and Southern Brazil both showed peaks during
in infants and young children,12–14 and the significance of the October, coinciding with spring.24,29
virus has been underappreciated.9 Evidence shows that HPIV infections tend to be less severe
In the present investigation, the prevalence of HPIVs was than those caused by HRSV.6 More deaths occurred in children
7.7%, the second most common virus after HRSV, whose preva- hospitalized for HPIVs than among HRSV-infected patients.
lence was 26.5%. This result corroborates the findings of other The deaths occurred in children outside the age group at
studies with similar design.15–17 We adopted IIF as the method higher risk for developing severe disease (less than three
of viral investigation, as it is a reliable technique with 50–75% months old), but all these cases had contributing medical
sensitivity and excellent specificity.12,18 Some authors claim factors, with two having history of premature birth, one recur-
that molecular methods (PCR and Real-Time PCR) offer 10–20% rent wheezing case with gastroesophagic reflux, and another
 b r a z j i n f e c t d i s . 2 0 1 5;1 9(4):358–362 361

Total HPIV VSR

80

70

60

50

40

30

20

10

0
05

06

6
5

7
06

07
5

05

06

06

07

7
5

7
5

6
05

6
/0

/0
/0

/0

/0
/0

/0

/0

/0

/0

/0

/0
r/0

r/0

r/0
t/0

t/0
l/0

l/0
o/

o/
v/

v/
n/

n/

n/

n/
ov

ov
ar

ar

ar
v/

ai

ut

er

ai

ut

er

ai
Ab

Ag

Se

Ab

Ag

Se

Ab
Fe

Fe
Ju

Ju

Ja

Ju

Ju

Ja
fe

M
O

O
N

D
Fig. 1 – Distribution of cases by month and number of patients with human parainfluenza viruses (HPIV) or with human
respiratory syncytial virus (VSR).

with delayed neuro-psychomotor development secondary to 2. Williams BG, Gouws E, Boschi-Pinto C, Bryce J, Dye C.
neonatal anoxia and recurrent wheezing. The prevention Estimates of world-wide distribution of child deaths from
acute respiratory infections. Lancet infect Dis. 2002;2:25–32.
strategy of administering immunoglobulin is well established
3. Lourenção LG, Júnior JBS, Rahal P, Souza FP, Zanetta DMT.
for HRSV, and studies are underway for vaccines against this
Infecção pelo Vírus Sincicial Respiratório em Crianças.
agent. For HPIVs however, viruses with significant impact in Pulmão RJ. 2005;14:59–68.
terms of morbidity and mortality, no established strategy is 4. Pilger DA, Cantarelli VV, Amantea SL, et al. Detection of
currently in place. human bocavirus and human metapneumovirus by real-time
In conclusion, the impact caused by HPIVs infections is sig- PCR from patients with respiratory symptoms in Southern
nificant in hospitalized children, where this agent should be Brazil. Mem Inst Oswaldo Cruz. 2011;106:56–60.
5. Bonzel L, Tenenbaum T, Schroten H, Schildgen O,
considered as a cause of ARTI. Longitudinal studies determin-
Schweitzer-Krantz S, Adams O. Frequent detection of viral
ing the prevalence of respiratory viruses should be encouraged coinfection in children hospitalized with acute respiratory
to help devise prevention and treatment strategies. tract infection using a real-time polymerase chain reaction.
Pediatr Infect Dis J. 2008;27:589–94.
6. Calvo C, Garcia-Garcia ML, Ambrona P, et al. The burden of
Funding infections by parainfluenza virus in hospitalized children in
Spain. Pediatr Infect Dis J. 2011;30:792–4.
7. Chiu SS, Chan KH, Chen H, et al. Virologically confirmed
This study was funded by FAPESP – Fundação de Amparo à population-based burden of hospitalization caused by
Pesquisa do Estado de São Paulo. respiratory syncytial virus, adenovirus, and parainfluenza
viruses in children in Hong Kong. Pediatr Infect Dis J.
2010;29:1088–92.
Conflicts of interest 8. Yang HT, Jiang Q, Zhou X, et al. Identification of a natural
human serotype 3 parainfluenza virus. Virol J. 2011;8:58.
9. Weinberg GA. Parainfluenza viruses: an underappreciated
The authors declare no conflicts of interest.
cause of pediatric respiratory morbidity. Pediatr Infect Dis J.
2006;25:447–8.
10. Lee MS, Mendelman PM, Sangli C, Cho I, Mathie SL, August
references
MJ. Half-life of human parainfluenza virus type 3 (HPIV3)
maternal antibody and cumulative proportion of HPIV3
infection in young infants. J Infect Dis. 2001;183:1281–4.
1. UNICEF. World Health Organization (WHO). Pneumonia: the 11. Wolf DG, Greenberg D, Kalkstein D, et al. Comparison of
forgotten Killer of children. Geneva: UNICEF/WHO; 2006, 40 human metapneumovirus, respiratory syncytial virus and
pp. influenza A virus lower respiratory tract infections in
362 b r a z j i n f e c t d i s . 2 0 1 5;1 9(4):358–362

hospitalized young children. Pediatr Infect Dis J. for the detection of seven common respiratory viruses in
2006;25:320–4. respiratory samples. J Mol Diagn. 2004;6:125–31.
12. Henrickson KJ. Parainfluenza viruses. Clin Microbiol Rev. 21. Peltola V, Heikkinen T, Ruuskanen O. Clinical courses of croup
2003;16:242–64. caused by influenza and parainfluenza viruses. Pediatr Infect
13. Sato M, Wright PF. Current status of vaccines for Dis J. 2002;21:76–8.
parainfluenza virus infections. Pediatr Infect Dis J. 22. Nascimento-Carvalho CM, Ribeiro CT, Cardoso MR, et al. The
2008;27:S123–5. role of respiratory viral infections among children
14. Teo WY, Rajadurai VS, Sriram B. Morbidity of parainfluenza 3 hospitalized for community-acquired pneumonia in a
outbreak in preterm infants in a neonatal unit. Ann Acad Med developing country. Pediatr Infect Dis J. 2008;27:939–41.
Singapore. 2010;39:837–46. 23. Gomez M, Mufson MA, Dubovsky F, Knightly C, Zeng W,
15. Iwane MK, Edwards KM, Szilagyi PG, et al. Population-based Losonsky G. Phase-I study MEDI-534, of a live, attenuated
surveillance for hospitalizations associated with respiratory intranasal vaccine against respiratory syncytial virus and
syncytial virus, influenza virus, and parainfluenza viruses parainfluenza-3 virus in seropositive children. Pediatr Infect
among young children. Pediatrics. 2004;113: Dis J. 2009;28:655–8.
1758–64. 24. Fe MM, Monteiro AJ, Moura FE. Parainfluenza virus infections
16. Noyola DE, Arteaga-Dominguez G. Contribution of respiratory in a tropical city: clinical and epidemiological aspects. Braz J
syncytial virus, influenza and parainfluenza viruses to acute Infect Dis. 2008;12:192–7.
respiratory infections in San Luis Potosi, Mexico. Pediatr 25. Vega-Briceno LE, Pulgar BD, Potin SM, Ferres GM, Sánchez DI.
Infect Dis J. 2005;24:1049–52. Clinical and epidemiological manifestations of parainfluenza
17. Pecchini R, Berezin EN, Felicio MC, et al. Incidence and clinical infection in hospitalized children. Rev Chilena Infectol.
characteristics of the infection by the respiratory syncytial 2007;24:377–83.
virus in children admitted in Santa Casa de Sao Paulo 26. Thomazelli LM, Vieira S, Leal AL, et al. Surveillance of eight
Hospital. Braz J Infect Dis. 2008;12:476–9. respiratory viruses in clinical samples of pediatric patients in
18. Greenhow TL, Weintrub PS. Utility of direct fluorescent southeast Brazil. J Pediatr. 2007;83:422–8.
antibody testing of nasopharyngeal washes in children with 27. Hall CB. Respiratory syncytial virus and parainfluenza virus.
and without respiratory tract illness. Pediatr Infect Dis J. N Engl J Med. 2001;344:1917–28.
2006;25:502–6. 28. Nascimento JP, Siqueira MM, Sutmoller F, et al. Longitudinal
19. Freymuth F, Vabret A, Cuvillon-Nimal D, et al. Comparison of study of acute respiratory diseases in Rio de Janeiro:
multiplex PCR assays and conventional techniques for the occurrence of respiratory viruses during four consecutive
diagnostic of respiratory virus infections in children admitted years. Rev Inst Med Trop Sao Paulo. 1991;33:287–96.
to hospital with an acute respiratory illness. J Med Virol. 29. Straliotto SM, Siqueira MM, Muller RL, Fischer GB, Cunha ML,
2006;78:1498–504. Nestor SM. Viral etiology of acute respiratory infections
20. Syrmis MW, Whiley DM, Thomas M, et al. A sensitive, specific, among children in Porto Alegre, RS, Brazil. Rev Soc Bras Med
and cost-effective multiplex reverse transcriptase-PCR assay Trop. 2002;35:283–91.