ispa

SOUTHEAST ASIAN J TROP MED PUBLIC HEALTH

1116 Vol 41 No. 5 September 2010

Correspondence: Peter J de Vries, Division ofInfectious Diseases, Tropical Medicine andAIDS, Academic Medical Center, F4-217, POBox 22700, 1100 DE Amsterdam, the Nether-lands.E-mail: [email protected]

VIRAL RESPIRATORY TRACT INFECTIONS AMONGPATIENTS WITH ACUTE UNDIFFERENTIATED FEVER

IN VIETNAM

Hoang Lan Phuong1,2, Tran TT Nga1,3, Gerard J van Doornum4, Jan Groen4,Tran Q Binh2, Phan T Giao2, Le Q Hung2, Nguyen V Nam5, PA Kager1

and Peter J de Vries1

1Division of Infectious Diseases, Tropical Medicine and AIDS, Academic MedicalCenter, Amsterdam, the Netherlands; 2Department of Tropical Diseases,

3Department of Microbiology, Cho Ray Hospital, Ho Chi Minh City, Vietnam;4Department of Clinical Virology, Erasmus Medical Center, Rotterdam,

the Netherlands; 5Binh Thuan Medical College, Binh Thuan Province, Vietnam

Abstract. To investigate the proportion of viral respiratory tract infections amongacute undifferentiated fevers (AUFs) at primary health facilities in southern Viet-nam during 2001-2005, patients with AUF not caused by malaria were enrolled attwelve primary health facilities and a clinic for malaria control program. Serumwas collected on first presentation (t0) and after 3 weeks (t3) for serology. Afterexclusion of acute dengue infection, acute and convalescent serum samples from606 patients were using enzyme-linked immunoassays to detect IgA, as well asIgM and IgG antibodies against common respiratory viruses. Paired sera showedthe following infections: human parainfluenza virus (HPIV, 4.7%), influenza B vi-rus (FLUBV, 2.2%), influenza A virus (FLUAV, 1.9%) and human respiratory syn-cytial virus (HRSV, 0.6%). There was no association between type of infection andage, sex or seasonality; some inter-annual differences were observed for influ-enza. Antibody prevalence, indicative of previous infections, was relatively low:HPV, 56.8%, FLUBV, 12.1%; FLUAV, 5.9% and HRSV, 6.8%.

Key words: viral respiratory tract infection, acute undifferentiated fever, Vietnam

Puzelli et al, 2006). Little is known aboutrespiratory tract infections among patientswith fever who present to primary healthfacilities in developing countries. ARIs con-tribute significantly to mortality, especiallyamong children in developing countries(WHO, 1998). Global threats, including theinfluenza pandemic, or potentially fataldisease, such as avian influenza and severeacquired respiratory syndrome (SARS), canbe hidden among ARIs. It is therefore im-portant to know more about the epidemi-ology, presentation and options for earlydetection of these infections.

INTRODUCTION

Data regarding the etiology of acuterespiratory tract infections (ARIs) in devel-oping tropical countries are sparse. Theavailable data are based on sero-surveys oron diagnostic studies in defined cases ofrespiratory tract infection (Weber et al, 1998;

VIRAL RESPIRATORY TRACT INFECTIONS IN VIETNAM

Vol 41 No. 5 September 2010 1117

Viruses, such as human adenovirus(HAdV), human respiratory syncytial vi-rus (HRSV), influenza A virus (FLUAV),influenza B virus (FLUBV), human parain-fluenza virus (HPIV), human enterovi-ruses and human rhinoviruses, are impor-tant causes of ARI worldwide ( Tumova etal, 1989; Shek and Lee, 2003; Bourgeois etal, 2006). ARIs affect mostly children andthe incidence decreases with increasingage (Yun et al, 1995; Tsai et al, 2001; Shekand Lee, 2003). As respiratory viruses arehighly contagious and easily transmittedby aerosols or direct contact, they maycause epidemics. SARS (caused by SARS -associated human coronavirus) is an ex-ample of a new virus that soon after itsemergence threatened to evolve into a pan-demic with a high case fatality rate (Peiriset al, 2003). In contrast, avian influenza (in-fluenza A virus H5N1), is an example of avirus that is highly pathogenic to humansbut not very infectious (Hayden, 2006;Thomas and Noppenberger, 2007). It hasa large volatile bird reservoir which pre-cludes control and causes repeated out-breaks in the poultry industry worldwide.

In Vietnam the incidence of commu-nicable diseases has fallen in recent de-cades, but respiratory tract infections stillrank high among the ten leading causesof morbidity (WHO-WPRO, 2004, 2005).The morbidity data is mainly based onroutine reporting by health services whichmay not reflect the total burden of disease,similar to what is seen with dengue fever(Phuong et al, 2006c). There are no goodestimates of the incidence of respiratorytract infections in Vietnam. In 1980 an ARIcontrol program was established, whichbecame a national program in 1984, focus-ing on children (WHO, 1992). This pro-gram was set up with the goal to diagnoseand treat 80% of the nationwide occurringARIs by 2000. In 1998, despite high cover-

age (87.5%) by trained health workers anddrug availability, it was estimated thatonly 20% of children with ARI were beingtreated according to guidelines (UN, 1999).For older children, adolescents and adultsthere is no data.

The surveillance of ARI does not in-clude confirmation by laboratory tech-niques. If viral or bacterial cultures, PCRor serological tests are performed, this isusually done in hospitals or medical cen-ters for individual case managementrather than for routine surveillance on abroader scale. Uncomplicated cases whopresent to primary health facilities usuallydo not receive a classifying diagnosis(Phuong et al, 2006a).

A discrepancy between low detectionrates for ARIs and high morbidity associ-ated with ARI is possibly due tounderreporting of early, uncomplicated,and clinically non-specific stages. In thisstudy we investigated how many ARIs arefound within the group of undifferentiatedfevers for which patients seek help at com-munity primary health centers and reportthe sero-prevalence of ARIs in southernVietnam.

MATERIALS AND METHODS

Study sites and populationThe study enrolled patients with acute

undifferentiated fever who sought help atone of twelve non-adjacent communityhealth centers (here after called healthposts), and one clinic at the provincialmalaria station, in Binh Thuan Province.Details of the study sites were describedpreviously (Phuong et al, 2006a). BinhThuan Province has a population of ap-proximately 1.2 million living in an areaof 7,992 km2. The tropical monsoon climatecauses an annual average temperature of26ºC with a hot and dry season alternating



with a rainy season that lasts from May toOctober. The annual rainfall averages 1,400mm. Agro-forestry, fishing and recentlytourism are the major sources of income.

The study started in March 2001. Pa-tients with acute undifferentiated fever(AUF), who presented to the study healthposts, were asked to participate in thestudy. AUF was defined as any febrile ill-ness of less than 14 days duration, con-firmed by an axillary temperature ≥38.0ºC,without evidence of severe systemic ororgan specific disease. Pneumonia, docu-mented by percussion and or auscultationof crackles or rales, was an exclusion cri-terion. X-ray facilities were not available.Malaria was excluded by microscopic ex-amination of a thick blood smear. Studycase record forms were filled in by the at-tending healthcare workers, includingidentifiers (age, sex, occupation, and ad-dress), recent exposure factors (contactwith fresh water or flooded terrains, workin paddy fields, visit to a forest or work ina forest), characteristics of disease (dura-tion, symptoms and signs, including theresults of the tourniquet test), self-treat-ment, a presumptive diagnosis, prescribedtreatment, referral and final outcome. Pre-sumptive diagnoses, such as “acute fever”and “viral infection”, were reclassified as“undifferentiated fever”. Two bloodsamples were collected by venous punc-ture on presentation, at time 0 (t0), and 3weeks later (t3). Sera were stored at -20ºCat the study sites until monthly transfer toCho Ray Hospital, where they were storedat -70ºC.

Serological testing for respiratory vi-rus infections was only done for patientsin whom paired serum samples had beencollected and only after dengue had beenexcluded as a cause of the fever. March2001 through March 2002, patients whocomplained of mild to moderate cough

were selected in order to get a first impres-sion. The t3 sera of 126 patients werescreened for antibodies to influenza A vi-rus (FLUAV), influenza B virus (FLUBV),human respiratory syncytial virus (HRSV)and human adenovirus (HAdV). Later, thesera of 22 patients who did not complainof cough, were also examined.

During the following years, patientswere selected randomly, (two per monthper health post) and those who testednegative were further tested for respira-tory viruses. Cough was not a selectioncriterion any more. During the secondyear, from April 2002 through March 2003,and the third year, from April 2003 throughMarch 2004, t3 sera from 138 patients(equally divided over periods 2 and 3)were screened for FLUAV and HRSV. Dur-ing the final year, January 1 to December31, 2005, paired serum samples from 320patients were tested for FLUAV, FLUBV,HRSV and Human parainfluenza virus 1,2, 3 (HPIV-1; HPIV-2; HPIV-3).

Serology

During Year 1 a monowell combinedcapture ELISA was used for the simulta-neous detection of IgM and IgA antibo-dies to FLUAV, FLUBV, HRSV, and toHAdV (Meddens Diagnostics BV, Vorden,the Netherlands). The tests were per-formed according to the instructions of themanufacturer on convalescent (t3) samplesat the laboratory of the Department of Vi-rology, Erasmus MC, Rotterdam, the Neth-erlands.

During Years 2, 3 and 4, ELISA panelsfrom another manufacturer were used(Virion\Serion GmbH, Würzburg, Ger-many). Separate panels were used to de-tect IgG and IgA antibodies to FLUAV,FLUBV, and HRSV. For HPIV an IgAELISA assay was used that was developedto diagnose acute infection but does not



discriminate between the three HPIVs. ForIgG separate panels were used for HPIV-1and HPIV-2. The tests were performed ac-cording to the instructions of the manu-facturer. For Years 2 and 3 only the t3samples were tested in the laboratory ofvirology of Cho Ray Hospital; for Year 4both t0 and t3 samples were tested at thelaboratory of the Department of Virology,Erasmus MC, Rotterdam, the Netherlands.

Interpretation of the serological resultsDuring the first three years the diag-

nosis was based on the t3 sample resultsonly, based on the concept that the detec-tion of IgA, IgM or IgG antibodies in theconvalescent sample was indicative of anacute infection (Meddens et al, 1990; Voetenet al, 1998; Rothbarth et al, 1999). During thefirst year a positive combined IgM/IgA testkit of Meddens Diagnostics was interpretedas indicative of “acute infection“ (Varsanoet al, 1995). With the Serion ELISA assays,used in the later periods, a positive IgA and/or IgG was classified as acute/past infection.Later, it was acknowledged, for HPIV andHRSV, (recent) past infections could not beclearly discriminated from acute infectionson the basis of a single test result only.Therefore during the last year the tests wereexpanded to testing both the acute and con-valescence samples. These results were clas-sified as acute infection, past infection andno infection. Acute infection was definedas a positive test on the t3 sample and afourfold increase in IgA concentration and/or IgG between t0 and t3. A positive IgA orIgG tests at t3 without seroconversion wasclassified as a past infection. No evidenceof infection was diagnosed when neitherIgA nor IgG was detectable at t3.

Ethical considerations

Medical ethical clearance for the studywas obtained from the Scientific Commit-tees of Cho Ray Hospital in Ho Chi Minh

City and the Binh Thuan Provincial HealthService. The study was explained and dis-cussed in meetings with provincial au-thorities and the staff of the health posts.All patients, (or in children, the parents orguardian) gave written informed consentprior to participation.

Data analysisAll data were stored in Microsoft

Access where patient selection and ran-domization was done. SPSS for Windows(version 14.0; SPSS, Chicago, IL) was usedfor statistical analysis. Descriptive statis-tics were calculated for background vari-ables and a chi-square test was used tocompare categorical factors.

RESULTS

In total, 606 patients were included inthe study. The general characteristics of thestudy population are shown in Table 1.Overall there was no difference in respectto age, sex, occupation and season betweenpatients with “cough” and without“cough”. In 2005 ARIs were more commonamong those who reported cough (chi-square 10.625, df = 1, p-value = 0.001) sug-gesting the initial selection of patients inthe 2001 cohort had more ARI.

In 2001 and 2005, approximately 8% oft3 samples tested positive for HRSV and 1%of test results showed acute infection,whereas in 2002 and 2003 none of the testedsamples were positive for IgA or IgG. Forinfluenza we observed a great inter-annualdifference. Six percent of t3 samples in 2005were positive for FLUAV, whereas in pre-vious years this was 18% or higher. In 2005,approximately 11% of samples tested posi-tive for FLUBV, whereas in 2001 25% ofsamples tested positive for FLUAV andFLUBV. The results of serological screen-ing in Years 1 to 3, and the expanded testresults in Year 4 based on the single t3



test results, are shown in Table 2. Theseroprevalence of respiratory tract virusesfrom 2001 to 2005 in Binh Thuan Province,Vietnam is shown in Fig 1.

In 2005 acute viral respiratory infec-tions caused 10.0% (32) of AUFs. The con-tributions by HPIV, FLUBV, FLUAV andHRSV were 4.7, 2.2, 1.9 and 0.6%, respec-tively. The diagnosis of HPIV was basedon seroconversion of IgA in 5 cases, IgGin 10 cases and both IgA and IgG in 0 cases.

FLUAV was diagnosed by 4, 0,2, for FLUBV by 2, 3, 2 and forHRSV infection by 1, 1, 0 byseroconversion of IgA, IgG andboth IgA and IgG, respectively.Mixed antibody responses werefound in two cases (0.6%). Onewas a 12 year-old girl with agreater than fourfold increase inFLUAV-IgG (9.8 times), FLUBV-IgG (19.4 times), HRSV-IgG (4.7times) and HPIV-2-IgG (4.3times); the other patient was a39 year-old farmer with a 3.9fold increase in FLUAV-IgA anda 3.8 fold increase in HPIV-IgA.In 124 patients (38.5%) no anti-bodies were found in any of thetests.

Characteristics of acute respira-tory viral infections

For further analysis of thedata, only the results from 2005were used. The characteristics ofthe 32 patients with serologicallyconfirmed acute viral respira-tory tract infection were furtheranalyzed. There were no signifi-cant differences in age or season-ality of the virus infections.There were no significant differ-ences between the symptomsand signs of patients with a se-

25

86

23

0

19

0

57

12

7 6

0

10

20

30

40

50

60

FLUAV+FLUBV HRSV HAdV FLUAV HRSV FLUAV HRSV HPIV FLUBV HRSV FLUAV

2001 2002 2003 2005

Se

rop

reva

len

ce

of

resp

ira

tory

tra

ct

viru

se

s (

%)

Fig 1–Seroprevalence of respiratory tract viruses from 2001to 2005 in Binh Thuan Province, southern Vietnam.

Fig 2–Prevalence of antibodies to respiratory tract virusesby age in Binh Thuan, southern Vietnam.

0.0

10.0

20.0

30.0

40.0

50.0

60.0

< 10 10-15 16-25 26-35 > 35

Pre

va

len

ce

of

an

tib

od

ies t

o r

esp

ira

tory

viru

se

s (

%)

HRSV HPIV FLUAV FLUBV

Age groups (years)

rodiagnosis of acute viral respiratory in-fection and those who had negative re-sults.

The responses of the primary healthcare workers were uniform. A presump-tive diagnosis of acute fever was diag-nosed in 290 patients (47.9%) (37.5% vs48.4% in those with and without acute res-piratory infection, respectively); pharyn-gitis was diagnosed in 214 patients (35.3%)(43.8% vs 34.8%); dengue fever was diag-



nosed in 55 patients (9.1%) (15.6 % vs 8.7%)with a p-value of 0.265. For treatment61.3% of patients received at least one an-tibiotic (196/320); this was similar for pa-tients with acute viral respiratory infectionand patients without infection (59.4% and61.5%, respectively, p=0.818). Completerecovery was recorded in 301 patients(94.1%) at t3 and partial recovery in 14patients (4.4%). In 5 patients (1.6%) thisinformation was missing.

The monthly distribution of acute vi-ral infections was different from healthpost to health post, but in general itshowed a scattered pattern in which noseasonality could be discerned. In 4 healthposts no infections were detected through-out the year. The prevalence of IgG anti-

bodies indicated no community was freeof these respiratory tract infections.

Seroprevalence of respiratory virusesThe seroprevalence of respiratory vi-

ral infections was estimated based on thepresence of either IgA or IgG antibodies att3. For HPIV this was 56.8%, FLUBV12.1%;FLUAV 5.9% and HRSV 6.8%. The partici-pants were classified into 5 age groups:<10 years, 10-15, 16-25, 26-35, and > 35. Thedistribution of respiratory viruses by agegroup is shown in Fig 2. There was no dif-ferences between males and females.

DISCUSSION

This study shows the common etio-logic agents of respiratory tract infections

Table 1General characteristics of patients presenting with acute undifferentiated fever at

primary health posts in Binh Thuan Province, Vietnam who were serologically testedfor respiratory tract viruses.

Year

2001 2002 2003 2005

Age groups, no. (%)<10 yrs10-15 yrs16-25 yrs26-35 yrs>35 yrs

11 (7.4 )26 (17.6)34 (23.0)44 (29.7)33 (22.3)

5 (7.2)5 (7.2)

21 (30.4)17 (24.6)21 (30.4)

8 (11.6)19 (27.5)13 (18.8)

9 (13.0)20 (29.0)

40 (12.5)105 (32.8)

63 (19.7)58 (18.1)54 (16.9)

Gender F/M, no. 61/87 27/42 37/32 123/197Season, no. (%)

Rainy seasonDry season

99 (66.9)49 (33.1)

34 (49.3)35 (50.7)

39 (56.5)30 (43.5)

163 (50.9)157 (49.1)

Occupation, no. (%)a

FarmerPupilOthers

aMissing 10 in 2001, 3 in 2002 and 3 in 2005

85 (57.4)33 (22.3)20 (13.5)

42 (60.9)15 (21.7)09 (15.9)

35 (50.7)24 (34.8)10 (14.5)

110 (34.4)157 (49.1)

50 (15.6)

Total number 148 69 69 320Median age (range) 27.0 (4 - 74) 28.3 (7 - 75) 21.3 (4 - 76) 16.8 (5 - 74)



circulate in Binh Thuan Province but theircontribution to undifferentiated feverscausing people to seek medical help issmall. Together, they comprise approxi-mately 10% of undifferentiated fever inpatients older than four years. Humanparainfluenza virus infection was the mostcommon.

The study was health service basedaimed at finding causes of undifferentiatedfever in patients who sought help at publicprimary health facilities. It was not an epi-demiologic surveillance of viral respiratoryinfections, and the results may not reflectthe true epidemiology of viral respira-tory infections in Binh Thuan Province.

The study only addressed the most com-mon viral agents that cause respiratorytract infections with fever and did not in-vestigate the less frequent causes of ARIin Vietnam. However, the results contri-bute to a better understanding of the epi-demiology of viral respiratory tract infec-tions. Another limitation of this study wasthat collecting blood from childrenyounger than four years of age, was almostalways refused. Since respiratory virusescirculate intensively among children un-der age five, this study probably underes-timates the true incidence of acute respi-ratory tract infections.

The seroprevalence found in this

Tested serum samples

Year of study

Number of patients tested n = 126a n = 22b n = 69b n = 69b n = 320b

HAdV Acute 7 (6)Past

FLUAV or FLUBV Acute 31 (25)Past

FLUAV Acute 4 (18) 16 (23) 13 (19) 8 (3)c

Past 11 (3)FLUBV Acute 8 (3)c

Past 30 (9)HPIV- (1+2+3) Acute 17 (5)c

Past 166 (52)HRSV Acute 10 (8) 0 0 0 3 (1)c

Past 19 (6)

Table 2Serological indications of acute respiratory tract infection in patients with undifferen-

tiated fever at primary health posts in Binh Thuan Province, Vietnam.

HAdV, Human adenovirus; FLUAV, Influenza A virus; FLUBV, Influenza B virus; HPIV, Humanparainfluenza virus; HRSV, Human respiratory syncytial virusaTested with a combined IgA and IgM ELISA; bTested with separate IgA and IgG ELISA.cThere were two mixed antibody responses: 1 case with antibodies against FLUAV, FLUBV, HPIVand HRSV and one case with antibodies against FLUAV and HPIV.t0, serum sample taken upon presentation to the primary health facility;t3, convalescent serum sample taken three weeks after t0

No. patients (%) with positive antibodies

Convalescence Paired

2001 2002 2003 2005

}}

} } }

} } } }



study is somewhat lower than what wasfound in other tropical regions (Weber etal, 1998; Yeolekar et al, 2001; Puzelli et al,2006). The differences in seroprevalencemay reflect differences in patient selectionor differences in tests used, but the rela-tively low seroprevalence of IgG indicatesthe true incidence of these viral infectionsis relatively low in Binh Thuan.

Mixed antibody responses occurred intwo cases in 2005. The diagnosis in thesetwo cases is not clear, but the low rate ofsuch non-specific, polyvalent responsesindicates the assays used were specificenough to make a diagnosis in the major-ity of patients. By including the results ofthe t0 sample the specificity of the sero-diagnosis in 2005 was further enhanced.The incidences of the respective respira-tory tract infections in 2005 are conserva-tive estimates.

IgG antibodies remain detectable forat least two to three years after FLUAV andFLUBV, and for one year after infectionswith HRSV and other paramyxoviruses(Couch and Kasel, 1983; Ogra, 2004). Viralrespiratory tract infections are commonthroughout life and humans do not de-velop lifelong (cross protective) immunityto different virus types (Crowe and Will-iams, 2003). HRSV infections, that occurafter the first two years of life, are almostalways reinfections that tend to becomeprogressively milder (Crowcroft et al,1999; Hall and McCarthy, 2005). HPIVreinfections in immunocompetent indi-viduals are more likely to cause mild up-per respiratory tract infections, with spar-ing of the lower respiratory tract after thefirst or second exposure (Welliver et al,1982; Glezen et al, 1984). Infection of influ-enza viruses yields longstanding immu-nity to reinfection with homologous vi-ruses but offers essentially no protectionagainst other (sub)types (Couch and

Kasell, 1983). This, antigenic drift and shiftof influenza viruses, explain why influ-enza continues to cause major epidemics(Treanor and Falsey, 1999). During thestudy period avian influenza was intro-duced into Vietnam, but no avian influ-enza virus A (H5N1) infections were de-tected in Binh Thuan Province (MOH Viet-nam, 2005; OIE, 2007).

It is difficult to distinguish betweenthe different viral respiratory tract infec-tions on clinical grounds only (Nicholsonet al, 1997). This has been addressed inmany studies focusing on children, but hasonly recently received more attention inadults (Dowell et al, 1996; Crowcroft et al,1999; Treanor and Falsey, 1999; Hall, 2001;Mangtani et al, 2006). The manifestationsof viral respiratory tract infections inadults are highly variable, ranging fromasymptomatic infection to lower respira-tory tract involvement and severe diseasein high risk populations (Treanor andFalsey, 1999; Hall, 2001; Zambon et al, 2001;Yang et al, 2003).

Seasonality of the incidence of viralrespiratory tract infections was not ob-served in this study. Documented seasonalpatterns of viral respiratory infections inthe tropics suggest HRSV and influenzavirus infections are associated with rain-fall but this was not confirmed in thisstudy (Chew et al, 1998; Shek et al, 2003).The inter-annual differences in the rates ofFLUAV and FLUBV infections in thisstudy may be partly explained by differ-ences in testing. In 2001 a combinedFLUAV and FLUBV test was used andduring the first years of the study only thet3 samples were tested, whereas in 2005the criterion of increasing antibody con-centrations was added. The number ofFLUAV positive (IgA + IgG) t3 sampleswas still lower in 2005 than in the two pre-vious years, and the number of FLUAV



and FLUBV positive samples was higherthan in 2001. There is not much knownabout inter-annual variation in influenzaincidence in Southeast Asia. In temperatezones this is a well known phenomenonand our results point to similar fluctua-tions in the tropics.

In this population the contribution ofcommon viral respiratory infections to thecauses of AUF was small. Respiratory vi-ruses, especially influenza, may spreadrapidly because it has a short generationtime, even if it has low transmissibility ordoes not spread to many others (Carrat etal, 2008). Approximately one-fourth toone-third of AUFs can be attributed to den-gue fever (Phuong et al, 2006b,c) while thecause of the remaining fevers is unknown.One serosurveillance indicated brucello-sis was unlikely to be a cause of AUF (Ngaet al, 2006). Leptospirosis is of interest sinceIgM and IgG antibodies are found(Wagenaar et al, 2004; Thai et al, 2006) buthow much leptospirosis contributes toAUF still has to be evaluated. No evidenceof histoplasmosis was found in surveil-lance of 180 selected samples from 2002 to2005 and antibodies to Chikungunya vi-rus and rickettsia species were rare (un-published data). There are no epidemio-logic data regarding infectious diseases inthis province to direct further investiga-tions. Less frequent air borne pathogenshave not been studied extensively.

From an operational point of view itis good news the incidence of viral respi-ratory tract infections among these pa-tients with undifferentiated fever was lowand that none progressed to severe respi-ratory tract infections. This means there islittle overlap between AUF and the casedefinitions that are applied in the ARI pro-gram, which are based on the clinical find-ings of respiratory tract infections. Thereis no need to expand the ARI program to

a syndromic approach for undifferentiatedfever. The bad news was antibiotic pre-scription rates were inordinately high forinfections that do not require (empiric)antibiotic therapy.

In conclusion, acute respiratory tractinfections are an infrequent cause of un-differentiated fever in Binh Thuan. Thetransmission of respiratory viruses is rela-tively low in Binh Thuan. A study focus-ing on infections in children is necessaryto complete the assessment of the impactof viral respiratory tract infections in thecommunity.

ACKNOWLEDGEMENTS

The study was supported by theDutch Foundation for the Advancementof Tropical Research (WOTRO). We thankthe patients in Binh Thuan who partici-pated in this study. We gratefully acknow-ledge the participating health care work-ers and the health authorities in BinhThuan, the board and department of mi-crobiology of Cho Ray Hospital, andChantal Burghoorn, Sandra Scherbeyn andBianca Baijens of the Department of Virol-ogy, Eramus Medical Center, Rotterdam,the Netherlands who made this study pos-sible.

REFERENCES

Bourgeois FT, Valim C, Wei JC, McAdam AJ,Mandl KD. Influenza and other respira-tory virus-related emergency departmentvisits among young children. Pediatrics2006; 118: e1-8.

Carrat F, Vergu E, Ferguson NM, et al. Timelines of infection and disease in humaninfluenza: a review of volunteer challengestudies. Am J Epidemiol 2008; 167: 775-85.

Chew FT, Doraisingham S, Ling AE,Kumarasinghe G, Lee BW. Seasonal trendsof viral respiratory tract infections in the



tropics. Epidemiol Infect 1998; 121: 121-8.Couch RB, Kasel JA. Immunity to influenza in

man. Annu Rev Microbiol 1983; 37: 529-49.Crowcroft NS, Cutts F, Zambon MC. Respira-

tory syncytial virus: an underestimatedcause of respiratory infection, with pros-pects for a vaccine. Commun Dis PublicHealth 1999; 2: 234-41.

Crowe JE, Williams JV. Immunology of viralrespiratory tract infection in infancy.Paediatr Respir Rev 2003; 4: 112-9.

Dowell SF, Anderson LJ, Gary HE, et al. Respi-ratory syncytial virus is an importantcause of community-acquired lower res-piratory infection among hospitalizedadults. J Infect Dis 1996; 174: 456-62.

Glezen WP, Frank AL, Taber LH, Kasel JA.Parainfluenza virus type 3: seasonality andrisk of infection and reinfection in youngchildren. J Infect Dis 1984; 150: 851-7.

Hall CB. Respiratory syncytial virus andparainfluenza virus. N Engl J Med 2001;344: 1917-28.

Hall CB, McCarthy CA. Respiratory syncytialvirus. In: Mandell GL,Bennett JE, Dolin R,eds. Principles and practice of infectiousdiseases. 6th ed. Philadelphia: ElsevierChurchill Livingstone, 2005: 2008-26.

Hayden FG. Respiratory viral threats. CurrOpin Infect Dis 2006; 19: 169-78.

John J. Treanor influenza virus. In: Mandell GL,Bennett JE, Dolin R, eds. Principles andpractice of infectious diseases. 6th ed. Phila-delphia: Elsevier Churchill Livingstone,2005: 2060-85.

Mangtani P, Hajat S, Kovats S, Wilkinson P,Armstrong B. The Association of respira-tory syncytial virus infection and influ-enza with emergency admissions for res-piratory disease in London: an analysis ofroutine surveillance data. Clin Infect Dis2006; 42: 640-6.

Meddens MJ, Herbrink P, Lindeman J, van DijkWC. Serodiagnosis of respiratory syncy-tial virus (RSV) infection in children asmeasured by detection of RSV-specificimmunoglobulins G, M, and A with en-zyme-linked immunosorbent assay. J Clin

Microbiol 1990; 28: 152-5.Ministry of Health (MOH). Vietnam avian flu.

Hanoi: MOH, 2005. [Cited 2009 Feb 9].Available from: URL: http://www.cimsi.org.vn/influenza/cumga/ban-do-cum-gia-cam.jpg

Nga TT, de Vries PJ, Abdoel TH, Smits HL.Brucellosis is not a major cause of febrileillness in patients at public health careFacilities in Binh Thuan Province, Viet-nam. J Infect 2006; 53: 12-5.

Nicholson KG, Kent J, Hammersley V, CancioE. Acute viral infections of upper respira-tory tract in elderly people living in thecommunity: comparative, prospective,population based study of disease burden.Br Med J 1997; 315: 1060-4.

Ogra PL. Respiratory syncytial virus: the vi-rus, the disease and the immune response.Paediatr Respir Rev 2004; 5: S119-26.

Peiris JS, Yuen KY, Osterhaus AD, Stohr K. Thesevere acute respiratory syndrome. N EnglJ Med 2003; 349: 2431-41.

Phuong HL, de Vries PJ, Nagelkerke N, et al.Acute undifferentiated fever in Binh ThuanProvince, Vietnam: imprecise clinical diag-nosis and irrational pharmaco-therapy.Trop Med Int Health 2006a; 11: 869-79.

Phuong HL, de Vries PJ, Nga TT, et al. Dengueas a cause of acute undifferentiated feverin Vietnam. BMC Infect Dis 2006b; 6: 123.

Phuong HL, de Vries PJ, Thai KT, et al. Denguevirus infections in Vietnam: Tip of the ice-berg. Dengue Bull 2006c; 30: 15-25.

Puzelli S, Boros S, Affinito C, et al. Prevalenceof antibodies against A and B influenzaviruses in south-western Papua NewGuinea. J Med Virol 2006; 78: 820-4.

Rothbarth PH, Groen J, Bohnen AM, de GrootR, Osterhaus ADME. Influenza virus se-rology–a comparative study. J Virol Meth-ods 1999; 78: 163-9.

Shek LP, Lee BW. Epidemiology and seasonal-ity of respiratory tract virus infections inthe tropics. Paediatr Respir Rev 2003; 4: 105-11.

Thai KT, Binh TQ, Giao PT, et al. Seroepide-



miology of leptospirosis in southern Viet-namese children. Trop Med Int Health 2006;11: 738-45.

Thomas JK, Noppenberger J. Avian influenza:a review. Am J Health Syst Pharm 2007; 64:149-65.

Treanor J, Falsey A. Respiratory viral infectionsin the elderly. Antiviral Res 1999; 44: 79-102.

Tsai HP, Kuo PH, Liu CC, Wang JR. Respira-tory viral infections among pediatric inpa-tients and outpatients in Taiwan from 1997to 1999. J Clin Microbiol 2001; 39: 111-8.

Tumova B, Heinz F, Syrucek L, et al. Occurrenceand aetiology of acute respiratory dis-eases: results of a longterm surveillanceprogramme. Acta Virol 1989; 33: 50-62.

United Nations (UN). Looking ahead: a UnitedNations common country assessment ofVietnam. New York: United Nation. n.d.[Cited 2009 Apr 21]. Available from URL:http://www.undg.org/archive_docs/1721-Vietnam_CCA_-_Vietnam_1999.pdf

Varsano N, Azar R, Ben Bassat M, MendelsonE. Application of ELISA for IgM, IgA andantigen detection for rapid diagnosis ofrespiratory syncytial virus infections: acomparative study. Clin Diagn Virol 1995;3: 17-27.

Voeten JT, Groen J, van Alphen D, et al. Use ofrecombinant nucleoproteins in enzyme-linked immunosorbent assays for detec-tion of virus-specific immunoglobulin A(IgA) and IgG antibodies in influenza vi-rus A- or B-infected patients. J ClinMicrobiol 1998; 36: 3527-31.

Wagenaar JF, Falke TH, Nam NV, et al. Rapidserological assays for leptospirosis are oflimited value in southern Vietnam. AnnTrop Med Parasitol 2004; 98: 843-50.

Weber MW, Mulholland EK, Greenwood BM.Respiratory syncytial virus infection intropical and developing countries. TropMed Int Health 1998; 3: 268-80.

Welliver R, Wong DT, Choi TS, Ogra PL. Natu-ral history of parainfluenza virus infectionin childhood. Pediatrics 1982; 101: 180-7.

World Health Organization (WHO). Pro-gramme for control of acute respiratory

infections - Fifth programme report 1990-1991. Geneva: WHO, 1992. [Cited 2009 Feb9]. Available from: URL: http://whqlibdoc.who.int/hq/1992/WHO_ARI_ 92.22.pdf

World Health Organization (WHO). Child andAdolescent Health and Development.WHO/CHS/CAH/98.1A. 1998. [Cited 2009Apr 19]. Available from: URL: http://www.who.int/child-adolescent-health/New_Publications/IMCI/WHO_CHS_CAH_98.1/Rev99.A.pdf

World Organisation for Animal Health (OIE):Update on highly pathogenic avian influ-enza in animals (Type H5 and H7). Paris:OIE, 2007. [Cited 2009 Feb 9]. Availablefrom: URL: http://www.oie.int/wahis/reports/en_fup_0000006332_20071015_151311.pdf

WHO Regional Office for the Western Pacific(WHO-WPRO). Vietnam: country healthinformation profiles. Manila, thePhilipines: WHO-WPRP, 2004. [Cited 2009Apr 19]. Available from: URL: http://www.wpro.who.int/countries/04vtn/

WHO Regional Office for the Western Pacific(WHO-WPRO). Vietnam: country healthinformation profiles. Manila, the Philip-pines: WHO-WPRO, 2005. [Cited 2009Apr 19]. Available from: URL: http://www.wpro.who.int/countries/05vtn/

Yang TY, Lu CY, Kao CL, et al. Clinical mani-festations of parainfluenza infection inchildren. J Microbiol Immunol Infect 2003;36: 270-4.

Yeolekar LR, Kulkarni PB, Chadha MS, Rao BL.Seroepidemiology of influenza in Pune,India. Indian J Med Res 2001; 114: 121-6.

Yun BY, Kim MR, Park JY, Choi EH, Lee HJ,Yun CK. Viral etiology and epidemiologyof acute lower respiratory tract infectionsin Korean children. Pediatr Infect Dis J1995; 14: 1054-9.

Zambon M, Hays J, Webster A, Newman R,Keene O. Diagnosis of influenza in thecommunity: relationship of clinical diag-nosis to confirmed virological, serologic,or molecular detection of influenza. ArchIntern Med 2001; 161: 2116-22.

ispa

Documents