Journal of Medical Virology 72:299–303 (2004)

Evidence of Human Metapneumovirus inChildren in Argentina

Monica Galiano, Cristina Videla,* Silvia Sanchez Puch, Alfredo Martınez, Marcela Echavarrıa,and Guadalupe Carballal

Clinical Virology Laboratory, CEMIC University Hospital, Buenos Aires, Argentina

Human metapneumovirus (hMPV) is a virus,which was first associated with acute lowerrespiratory infection in children but is detectedcurrently in all age groups. Clinical symptomsare similar to those described for respiratorysyncytial virus (RSV) infections, ranging frommild respiratory illness to severe bronchiolitisand pneumonia in children. To date, no cases ofhMPV have been reported in Argentina. In thisstudy, 440 respiratory samples obtained duringtheperiod1998–2002 fromchildrenunder 5yearsold with acute respiratory infection were evalu-ated. Routine detection for RSV, adenovirus,influenza, and parainfluenza was undertaken byimmunofluorescent assay. Of the samples nega-tive for these viruses, only 100 were available.All these samples were tested for hMPV by RT-PCR using primers for the L gene. Eleven out of100 (11%) respiratory samples were positive forhMPVbyRT-PCR.Ahigher frequencyofdetectionwas observed in spring. hMPVwas detected in allthe years studied, except in 2001. Ten out of 11children positive for hMPV were hospitalized.Median age was 5 months. Of seven patients,five (71%) required oxygen supplementation.The most frequent diagnosis was bronchiolitis(86%), sometimes accompanied by conjunctivitisand otitis media. The present study showed thathMPV was associated with acute lower respira-tory infections in children in Buenos Aires,Argentina. This evidence strongly suggests thathMPV is a common pathogen with a wide geo-graphical distribution, which should be includedin the routine diagnosis of respiratory viruses inyoung children. J. Med. Virol. 72:299–303,2004. � 2004 Wiley-Liss, Inc.

KEY WORDS: humanmetapneumovirus; viralrespiratory infections in Argen-tina; acute lower respiratoryinfections in children

INTRODUCTION

Human metapneumovirus (hMPV) was identifiedrecently ars a viral respiratory pathogen in 2001 [vanden Hoogen et al., 2001]. Since its genomic organizationwas similar to that of theavianpneumovirus, hMPVwasconsidered as the first human-related member of theMetapneumovirus genus [van den Hoogen et al., 2002].The Metapneumovirus and the Pneumovirus genus aremembers of the Pneumovirinae subfamily within theParamyxoviridae family.

HMPV was first identified in The Netherlands inrespiratory samples from children with clinical syn-dromes resembling those caused by respiratory syncy-tial virus (RSV) [van denHoogen et al., 2001]. This viruswas associated initially with respiratory diseases inchildren from Australia, Canada, and France [Nissenet al., 2002; Pelletier et al., 2002; Peret et al., 2002;Freymouth et al., 2003; Mackay et al., 2003]. Otherstudies showed that hMPVwas also detected in childrenand adults in Canada, England, and USA [Boivin et al.,2002; Stockton et al., 2002; Falsey et al., 2003]. Eventhough this virus can grow in cell culture, it is selective,slow, and produces a mild cytopathic effect (CPE)[Boivin et al., 2002]. To date, at least two differentgenetic groups have been described based on sequenceanalysis of the N, M, and F genes. In addition, someintragroup variability has been observed [Boivin et al.,2002; Peret et al., 2002; Stockton et al., 2002].

The clinical manifestations associated with hMPVrange frommild respiratory symptoms, including cough,rhinorrhea, and influenza-like illness, to lower res-piratory diseases such as bronchiolitis and pneumonia

Grant sponsor: The European Union; Grant number:ERBIC18CT980374; Grant sponsor: Fundacion Rene Baron.

*Correspondence to: Cristina Videla, Laboratorio de VirologıaClınica, CEMIC, Galvan 4102 (C1431FWO) Buenos Aires,Argentina. E-mail: cvidela@cemic.edu.ar

Accepted 30 July 2003

DOI 10.1002/jmv.10536

Published online in Wiley InterScience(www.interscience.wiley.com)

� 2004 WILEY-LISS, INC.

[van den Hoogen et al., 2001; Boivin et al., 2002]. Somechildren with severe bronchiolitis required mechanicalventilation [Boivin et al., 2002; Pelletier et al., 2002].

In Argentina, acute lower respiratory infections arethe third most frequent cause of death among childrenunder 1 year of age [Bossio and Arias, 2001]. The fewstudies on respiratory infections etiology performed inArgentine infants from our country revealed a higherfrequency of viral etiology than of bacterial, accountingfor 30% in children less than 5 years of age. RSV wasthe virus detected most commonly, with a prevalenceranging from 18 to 25%. Nevertheless, no etiologicalcauses were reported in up to 69% of respiratory in-fections in this group [Avila et al., 1990; Weissenbacheret al., 1990; Carballal et al., 2001].

In order to detect additional viral etiological patho-gens causing respiratory diseases and the potentialassociation of hMPV with lower respiratory illness,we evaluated retrospectively the presence of hMPV insamples from infantswith respiratory disease inBuenosAires, Argentina.

MATERIALS AND METHODS

Respiratory Samples

A total of 440 samples from children under 5 years ofage with acute respiratory infection were evaluated(median age¼5 months; range 0–58 months). Ninetypercent of the children were admitted to hospital dueto acute lower respiratory infection in two hospitalsin the city of Buenos Aires, Argentina. In all cases,respiratory samples including nasopharyngeal aspi-rates or nasopharyngeal swabs were obtained at ad-mission. Samples were sent to the CEMIC ClinicalVirology Laboratory for diagnosis. The period studiedwas from 1998 to 2002. Samples were processed aspreviously described [Avila et al., 1990]. Briefly, fornasopharyngeal aspirates, mucus was disrupted with apipette and centrifuged at 1,000 rpm for 10 min at 48C.The pellet was resuspended in PBS and washed twice.Cells were fixed on slides. Direct antigen detectionfor respiratory viruses (RSV, adenovirus, influenza Aand B, parainfluenza 1, 2, 3) was carried out by indirectimmunofluorescent assay (IFA) with the correspondentmonoclonal antibodies (RSV: No. 5006, Light Diagnos-tics; Influenza A: MAB8251; Influenza B: MAB8661;Adenovirus: MAB805; Pan parainfluenza (1, 2, 3): MAB819; Chemicon Int., Temecula, CA) and a fluorescein-labeled anti-mouse IgG (Sigma, St. Louis, MO). Theremaining original samples and supernatants frompelleted cells were stored at �708C for further studies.

An informed consent form signed by the parents wasrequired in all cases.

Reverse-Transcription Polymerase ChainReaction (RT-PCR) Assay for hMPV

A total of 100 respiratory samples, which were ne-gative for RSV, adenovirus, influenza A and B, andparainfluenza 1, 2, 3 by IFA, were evaluated for hMPVby PCR. No sample selection was done since all avail-

able frozen samples that were negative for these viruseswere included in this study. We analyzed an aliquot oforiginal samples or supernatant that were kept at�708C. These included 98 nasopharyngeal aspiratesand two nasopharyngeal swabs, and were similarlydistributed among the 5-year study period.

Viral RNA was extracted from respiratory samplesusing commercial columns (QIAamp viral RNAmini kit,Qiagen, Qiagen Gmbh, Hilden, Germany). RT-PCRwasperformed using AMV reverse transcriptase (AMV,Promega, Madison, WI) and Taq DNA polymerase(Promega) in a one-step procedure. Primer set amplifieda region (170 bp) in the polymerase (L) gene [van denHoogen et al., 2001]. A plasmid containing this 170 bpfragment was used as positive control (primer se-quences, PCR protocol, and plasmid were gentlyprovided byDr. A. Osterhaus, ErasmusMedical Center,Rotterdam, The Netherlands). Amplified products weredetected in agarose gel by electrophoresis. A digitalcamera (Kodak) was used for documentation.

HMPV Isolation

Original respiratory samples positive for hMPV byRT-PCR were inoculated in LLC-MK2 cells. Cultureswere observed during 15 days. After that, two blindpassages were carried out. After inoculation,MEMwithEarle salts without serum and with tripsine 1% wasused for cells maintenance. Cells and supernatant wererecovered and extracted for RT-PCR. RT-PCR for hMPVwas done as described previously.

RESULTS

Of 440 samples, 182 (41%) were positive at least forone of the respiratory viruses tested by IFA. Positivityrate was 144/440 (32.7%) for RSV, 15/440 (3.4%) foradenovirus, 13/440 (2.9%) for influenza A and B, and 10/440 (2.3%) for parainfluenza 1, 2, 3. Of 440, 258 (59%)were negative for these viruses.

Of thenegative samples, the 100availablewere testedretrospectively for hMPV by RT-PCR.

A total of 11/100 (11%) respiratory samples werefound positive for hMPV. Of these, 7/11 (64%) wereoriginal samples and4/11 (36%)were supernatants frompelleted cells obtained from the original sample.

Eight out of 11 hMPV-positive samples by RT- PCRwere confirmed by culture.

hMPV was detected in all the 1998–2002 period,except during 2001. Of 27 negative samples from 2001,only 20 were available for further studies. None werepositive for hMPV. In the remaining years, the fre-quency of hMPV detection in samples negative for otherrespiratory viruses ranged from 7.7 to 18.2% (Table I).

Seasonal Distribution of hMPV

The seasonality of the respiratory viruses (RSV,adenovirus, influenza, and parainfluenza) analyzed inthe initial 440 samples was compared with hMPVdetection in 100 negative samples (Fig. 1). Even thoughsamples were obtained throughout the year in all the

300 Galiano et al.

study period, hMPV was only detected from June toJanuary. A higher frequency of detection was observedin spring (September, October, November).

Clinical Manifestations

Of 11 children with positive hMPV, 10 (91%) werehospitalized due to acute lower respiratory infections.Nine out of 11 (82%) were males, and 2/11 (18%) werefemales.Medianagewas5months (range2–27months),compared with a median age of 4 months (range 0–32months) for a group of 59 childrenhospitalizedwithRSVinfection during the same study period (Mann–WhitneyRank Sum Test, P¼ 0.187). A statistically significantdifference was observed between the median hospitalstay of hMPV-infected children (11.5 days, range 4–16)and that of RSV-infected patients (5 days, range 2–20)(Mann–Whitney Rank Sum Test, P¼0.038).

Of seven patients with clinical data available, five(71%) required oxygen supplementation, but none of

them were intubated; six (86%) were diagnosed bron-chiolitis, and one (14%) pertussis-like syndrome atadmission. Moreover, two children with bronchiolitisalso presented otitis media and conjunctivitis. Themostfrequent symptoms were cough (86%), fever (71%), andbreathing difficulties (57%).

DISCUSSION

Acute lower respiratory infection is one of the maincauses of morbidity and mortality among childrenworldwide. In developing countries, the mortality riskcan be 30 times higher than in developed countries[Organizacion Panamericana de la Salud (OPS), 1995].

In this study, young children with acute respiratorydisease were examined during a 5-year period. Respira-tory viruses including RSV, adenovirus, influenza, andparainfluenzawere detected in 41% of the cases, with noviral etiology shown in the rest of the patients. In anattempt to find another viral agent in patients withrespiratory disease, we have studied retrospectivelysamples negative for the common respiratory viruses, inorder to determine the presence of hMPV. A detectionrate of 11% was observed among these patients.

To our knowledge, this is the first evidence of hMPV inArgentina. This frequency was similar to that foundby other researchers in The Netherlands (10%) andAustralia (9.7%), which also analyzed samples fromchildren negative for other respiratory viruses [van denHoogenet al., 2001;Mackay et al., 2003]. In contrast, thefrequency of hMPV was lower in studies conducted inCanada (2.3%), England (2.2%), and USA (4.5%), whichincluded children, as well as young adults and elderlyadults [Boivin et al., 2002; Stockton et al., 2002; Falseyet al., 2003].

TABLE I. Frequency of hMPV in Negative Samples for OtherRespiratory Viruses From Children Under 5 Years of Age

With ALRI, Buenos Aires, Argentina (1998–2002)

YearNegativesamples

Samples studiedfor hMPV

hMPV positivesamples

n %

1998 86 13 1 7.71999 57 18 3 16.62000 63 38 5 13.22001 27 20 0 0.02002 25 11 2 18.2Total 258 100 11 11.0

Fig. 1. Cocirculation of hMPV with the other respiratory viruses in children <5 years old from BuenosAires (1998–2002).

Human Metapneumovirus in Argentina 301

These variations may be due to the differences in thestudy design. Factors such as the study population,season, type of sample, and PCR method could have aneffect on the detection rates among these studies.

Some of the studies have evaluated respiratory sam-ples during winter. A lower frequency rate (2–4.5%)was reported in studies that examined samples collectedonly in winter [Stockton et al., 2002; Falsey et al., 2003].In our study, samples were obtained throughout theyear, and hMPV was detected in winter and spring.Results were similar to those reported by Boivin et al.[2002], who showed that 87% of hMPV isolates wereisolated in winter and spring. In addition, our studyshowed that the temporal occurrence of hMPV wasconsistent with the seasonal pattern observed forparainfluenza and adenoviruses in Argentina.

Interestingly, no positive samples for hMPV werefound during 2001, even when 20/27 (74%) negativesamples were tested and sampling spanned thewhole year except in summer, when few or no respira-tory samples are received for routine diagnosis. Falseyet al. [2003] also observed a significant difference inrates of hMPV illnesses between two consecutive years(1.5 vs. 7%). These results could suggest a variableintensity of hMPV annual outbreaks, or the absence ofhMPV cases in some years. However, as in the case ofother respiratory viruses, the latter seems unlikely.Further studies with a larger number of samples andlonger periods of time are needed to determine thecirculation pattern of this newly described virus.

Different studies reported variable frequency rates inoriginal samples or isolates evaluated for hMPV. VandenHoogen et al. [2001] reported a detection rate of 10%when testing original samples, while Boivin et al. [2002]described a 2% frequency when carrying out PCR onlyon isolates that showed unidentified CPE.

Different hMPV genes such as L, M, N, and F havebeen used as targets for the various PCR methods.Variability of these genes may affect the sensitivity ofthese approaches, especially for this newly describedvirus onwhich limited sequence information isavailable.In our study, the L gene was amplified [van den Hoogenet al., 2001]. Although the limit of detection for this PCRwas not determined, positive results were obtained notonly with original samples, but also in the supernatantfrom pelleted cells from the original samples.

The clinical features of children with hMPV positivesamples observed in our study were similar to thosereported previously in children under 5 years ofage [Boivin et al., 2002; Nissen et al., 2002]. Thespectrum of symptoms putatively caused by hMPVare indistinguishable clinically from those caused byRSV infections in infants [Greenough et al., 2002].Probably due to the very young age of our patients(median age¼5 months), most of them had severedisease, requiring oxygen supplementation and hospi-talization. In addition, hospital stay of children withhMPV-infection was significantly longer than that ofchildren infected with RSV, despite a similar medianage of these groups.

These results were consistent with those described byBoivin et al. [2002], who reported a median age of15 months and a 91.7% of hospitalization due to severerespiratory conditions. In contrast, the same studyrevealed that immunocompetent adult patients pre-sented milder symptoms such as flu-like syndrome orcommon cold.Other studies on children showedvariableclinical presentations ranging from mild upper respira-tory tract disease to severe bronchiolitis and pneumonia[van den Hoogen et al., 2001; Nissen et al., 2002].

Since we have only analyzed the presence of hMPVin samples negative for the most common respiratoryviruses by IFA and not using PCR techniques, thepossible presence of other viruses as coinfecting patho-gens cannot be excluded. A high frequency of dualinfections with hMPV and RSV has been describedrecently in children with bronchiolitis [Greensill et al.,2003].

The absence of a control group in this retrospectivestudy does not indicate whether hMPV is a colonizing ora pathogenic virus. In addition, a recent study describedthe presence of hMPV in 4% of adults with asympto-matic infections [Falsey et al., 2003]. To date, no asym-ptomatic infections have been reported in children.Therefore, the evidence available currently supportsa high correlation between respiratory illness andhMPV infection [Boivin et al., 2002; Stockton et al.,2002].

In this study, we report on the first evidence of hMPVin children in Buenos Aires, Argentina. The resultsstrongly support the hypothesis that hMPV is a re-latively common pathogen responsible for respiratorydiseases in children in Argentina, who develop clinicalsymptoms that cannot be differentiated from thosecaused by RSV. Therefore, sensitive diagnostic methodsshould be developed in order to include hMPV diagnosisin routine tests for respiratory viruses. Further studiesare necessary to determine the prevalence and season-ality pattern in South America, and to better under-stand the impact of this virus in different populations.

ACKNOWLEDGMENTS

We thank Dr. A. Osterhaus for his kind provisionof hMPV positive control, PCR protocol and primersequences; Carmen Ricarte, Beatriz Ebekian, andCristina Juarez for their technical assistance; pediatri-cians for collecting the samples and Valeria Melia forhelping with English language.

REFERENCES

AvilaM,SalomonH,CarballalG,EbekianB,WoyskovskyN,CerqueiroMC, Weissenbacher M. 1990. Isolation and identification of viralagents in Argentinian children with acute lower respiratory tractinfection. Rev Infect Dis 12:S974–S981.

Boivin G, Abed Y, Pelletier G, Ruel L, Moisan D, Cote S, Peret TCT,Erdman DD, Anderson LJ. 2002. Virological features and clinicalmanifestations associated with human metapneumovirus: A newparamixovirus responsible for acute respiratory tract infections inall age groups. J Infect Dis 186:1330–1334.

Bossio JC, Arias SJ. 2001. Mortalidad infantil en Argentina. ArchArgent Pediatr 99:547–553.

302 Galiano et al.

Carballal G, Videla C, EspinosaM, Savy V,UezO, SequeiraM,KnezV,Requeijo P, Riva Posse C, Miceli I. 2001. Multicentered study ofviral acute lower respiratory infections in children from four citiesof Argentina, 1993–1994. J Med Virol 64:167–174.

Falsey AR, Erdman D, Anderson LJ, Walsh EE. 2003. Humanmetapneumovirus infections in young and elderly adults. J InfectDis 187:785–790.

Freymouth F, Vabret A, Legrand L, Eterradosi N, Lafay-Delaire F,Brouard J, Guillois B. 2003. Presence of the new human metap-neumovirus in French children with bronchiolitis. Pediatr InfectDis J 22:92–94.

Greenough A. 2002. Respiratory syncytial virus infection: Clinicalfeatures, management, and prophylaxis. Curr Opin Pulm Med8:214–217.

Greensill J, McNamara PS, Dove W, Flanagan B, Smyth RL, Hart CA.2003. Human metapneumovirus in severe respiratory syncytialvirus bronchiolitis. Emerg Infect Dis 9:372–375.

Mackay I, Jacob K, Woolhouse D, Waller K, Syrmis M, Whiley D,Siebert D, Nissen M, Sloots T. 2003. Molecular assays for detectionof human metapneumovirus. J Clin Microbiol 41:100–105.

Nissen MD, Siebert DJ, Mackay IM, Sloots TP, Withers SJ. 2002.Evidence of human metapneumovirus in Australian children. MedJ Aust 176:188.

Organizacion Panamericana de la Salud (OPS). 1995. Infeccionesrespiratorias agudas en las Americas. Biol Epidemiol 16:1–5.

Pelletier G, Dery P, Abed Y, Boivin G. 2002. Respiratory tract re-infections by the new humanmetapneumovirus in an immunocom-promised child. Emerg Infect Dis 8:976–978.

Peret TCT, Boivin G, Li Y, Couillard M, Humphrey C, Osterhaus AD,Erdman DD, Anderson LJ. 2002. Characterization of humanmetapneumovirus isolated frompatients inNorthAmerica. J InfectDis 185:1660–1663.

Stockton J, Stephenson I, Fleming D, Zambon M. 2002. Humanmetapneumovirus as a cause of community-acquired respiratoryillness. Emerg Infect Dis 8:897–901.

van den Hoogen BG, de Jong JC, Groen J, Kuiken T, de Groot R,Fouchier RA, Osterhaus AD. 2001. A newly discovered humanpneumovirus isolated from young children with respiratory tractdisease. Nat Med 7:719–724.

van den Hoogen BG, Bestebroer TM, Osterhaus AD, Fouchier RA.2002. Analysis of the genomic sequence of a human metapneumo-virus. Virology 295:119–132.

Weissenbacher M, Carballal G, Avila M, Salomon H, Harisiadi J,CatalanoM,CerqueiroMC,MurtaghP.r 1990.Etiologic andclinicalevaluation of acute lower respiratory tract infections in youngArgentinian children: An overview. Rev Infect Dis 12:S889–S898.

Human Metapneumovirus in Argentina 303