Etiologies of Rash and Fever Illnesses in ... - Oxford Academic

Etiologies of Rash and Fever Illnesses in Campinas,

Brazil

José Cássio de Moraes, Cristiana M. Toscano, Eliana N. C. de Barros, Brigina

Kemp, Fabio Lievano, Steven Jacobson, Ana Maria S. Afonso, Peter M. Strebel,

K. Lisa Cairns, the VigiFex Group

Dow

nloaded from https://academ

ic.oup.com/jid/article/204/suppl_2/S627/874370 by guest on 07 July 2022

http://akrurl.com/.2zpmo

S U P P L E M E N T A R T I C L E

Etiologies of Rash and Fever Illnesses inCampinas, Brazil

Jose Cassio de Moraes,1 Cristiana M. Toscano,3,5,a Eliana N. C. de Barros,4,a Brigina Kemp,4 Fabio Lievano,5,a

Steven Jacobson,6 Ana Maria S. Afonso,2 Peter M. Strebel,5,a K. Lisa Cairns,5,a and the VigiFex Group1Santa Casa de Sao Paulo Medical Sciences University, and 2Instituto Adolfo Lutz-Sao Paulo, Sao Paulo, 3Pan American Health Organization, Brasilia,and 4Municipal Health Department, Campinas, Brazil; 5National Center for Immunizations and Respiratory Diseases, Centers for Disease Control andPrevention, Atlanta, Georgia; and 6National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, Maryland

Background. Few population-based studies of infectious etiologies of fever-rash illnesses have been conducted.

This study reports on enhanced febrile-rash illness surveillance in Campinas, Brazil, a setting of low measles and

rubella virus transmission.

Methods. Cases of febrile-rash illnesses in individuals aged ,40 years that occurred during the period 1 May

2003–30 May 2004 were reported. Blood samples were collected for laboratory diagnostic confirmation, which

included testing for adenovirus, dengue virus, Epstein-Barr virus (EBV), enterovirus, human herpes virus 6

(HHV6), measles virus, parvovirus-B19, Rickettsia rickettsii, rubella virus, and group A streptococci (GAS)

infections. Notification rates were compared with the prestudy period.

Results. A total of 1248 cases were notified, of which 519 (42%) had laboratory diagnosis. Of these, HHV-6

(312 cases), EBV (66 cases), parvovirus (30 cases), rubella virus (30 cases), and GAS (30 cases) were the most

frequent causes of infection. Only 10 rubella cases met the rubella clinical case definition currently in use.

Notification rates were higher during the study than in the prestudy period (181 vs 52.3 cases per 100,000

population aged ,40 years).

Conclusions. Stimulating a passive surveillance system enhanced its sensitivity and resulted in additional

rubella cases detected. In settings with rubella elimination goals, rubella testing may be considered for all cases of

febrile-rash illness, regardless of suspected clinical diagnosis.

Relatively few comprehensive population-based studies of

the etiology of fever-rash illnesses have been conducted,

and most of these are from the northern hemisphere

[1–6]. Such studies are important to better understand

the age and geographic distribution and the incidence of

these etiologies. In the context of the Americas’ measles

and rubella elimination goals, this information can also

inform a minimum suspected measles and rubella re-

porting rate, which aims to ensure surveillance sensitive

enough to detect circulating measles and rubella virus.

In 1994, the Region of the Americas set the goal of

measles elimination (‘‘elimination’’ is defined as the ab-

sence of endemic measles cases in a defined geographical

area for a period of at least 12 months, in the presence of

a well-perfoming surveillance system) by 2000. This goal

was achieved in November 2002. Since then, all reported

measles cases have been linked to importation from

measles-endemic countries [7]. In Brazil, 2 confirmed

measles cases were reported in 2003 [8] and 0 cases were

reported in 2004 [9]. During 2001–2002, Brazil con-

ducted vaccination campaigns targeting all women of

childbearing age (the target age group varied by state,

with 12 years the lowest and 49 years the upper age limit)

Potential conflicts of interest: F. L. is currently working as Senior Director,Clinical Risk Management and Safety Surveillance, Merck & Co., Inc. All otherauthors: no conflicts.Presented in part: 4th Expo Epi (Brazilian exhibit of experiences on

epidemiology) Conference, Brasilia, Brazil, 2004.aPresent affiliations: Merck & Co, Inc., Whitehouse Station, New Jersey (F. L.);

Federal University of Goias, Goiania, Brazil (C. M. T.); World Health Organization,Beijing, China (K. L. C.); World Health Organization, Geneva, Switzerland (P. M. S.);and National Coordination for Respiratory and Vaccine Preventable DiseasesSurveillance, Brazilian Ministry of Health, Brazil (E. N. C. d. B.).Correspondence: Cristiana M. Toscano, MD, PhD, Av. T-05, 715. Apto. 301. Setor

Bueno - Goiania, GO - Brazil 74230-045 ([email protected]).

The Journal of Infectious Diseases 2011;204:S627–S636� The Author 2011. Published by Oxford University Press on behalf of the InfectiousDiseases Society of America. All rights reserved. For Permissions, please e-mail:[email protected] (print)/1537-6613 (online)/2011/204S2-0010$14.00DOI: 10.1093/infdis/jir490

Rash and Fever Syndromic Surveillance d JID 2011:204 (Suppl 2) d S627

Dow



for accelerated rubella control and congenital rubella syn-

drome (CRS) prevention. Five hundred sixty-three cases of

rubella were reported nationwide in 2003 [8] and 401 were re-

ported in 2004 [10]. In September 2003, the Americas endorsed

the rubella and CRS elimination goal by 2010 [11]

To monitor progress toward measles and rubella elimination,

the Pan American Health Organization (PAHO) recommends

that all countries in the region have integrated measles and rubella

surveillance. In Brazil, all cases of fever and rash meeting clinical

case definitions for measles (defined as fever and maculopapular

rash accompanied by $1 of the 3 following symptoms, cough,

coryza, or conjunctivitis, in any individual, regardless of age or

vaccination status) or rubella (defined as fever and maculopapular

rash accompanied by retroauricular, occipital, or cervical

lymphadenopathy in any individual, regardless of age and vacci-

nation status) should be notified and confirmed or discarded by

laboratory testing [12, 13]. The use of specific indicators to eval-

uate the quality of surveillance has been proposed. The World

Health Organization (WHO) has suggested an annual rate of

reported suspected measles or rubella cases of $2 cases per

100,000 population as an indicator of the sensitivity of the sur-

veillance system. PAHO has recently endorsed this indicator as

a criterion for documentation and verification of measles, rubella,

and CRS elimination in the American region [14].

To better understand the etiologies, distribution, and back-

ground rates of fever-rash illnesses in a community setting with

low rates of measles and rubella virus transmission and to assess

the usefulness and feasibility of a minimum suspected measles or

rubella reporting rate as an indicator of measles-rubella sur-

veillance sensitivity, we conducted enhanced passive surveillance

for fever-rash illnesses in individuals aged ,40 years during the

period from 1 May 2003 through 30 May 2004 in Campinas, Sao

Paulo State, Brazil.

METHODS

Study SettingCampinas, a city of �1 million inhabitants located 120 kilo-

meters northwest of Sao Paulo, Brazil, was selected as a study site

due to its well-functioning disease surveillance system, low rates

of measles and rubella virus transmission, and proximity to

a reference laboratory capable of testing for numerous infectious

agents. Public health care, provided by the National Health

System (NHS; Sistema Unico de Saude), is used exclusively by

�70% of the population; the remaining 30%, although eligible

for NHS services, receive most health care through the private

sector. In the public sector, Campinas has 78 clinics, 3 urgent

care centers, and 1 hospital. In the private sector, there are 5000

practitioners employed mainly by 4 major Health Maintenance

Organizations [15] and 21 hospitals. In addition, there are

2 university hospitals. Before this study, measles, rubella,

and dengue surveillance existed, in accordance with national

guidelines. Reporting came almost exclusively from public

health providers. Surveillance quality was considered good, as

demonstrated by standard performance indicators [16].

Study ImplementationThe study was conducted from 1 May 2003 through 30 May 2004.

Before initiating the study, meetings were held with surveillance

workers and both private and public primary health care pro-

viders in Campinas. An international seminar was convened when

the study was launched, and folders, posters, and CDs were dis-

tributed to health care professionals from both public and private

settings invited to participate. A website through which fever-rash

illnesses could be electronically reported was developed.

Information about the study was made available to the pop-

ulation via the media. Throughout the study, meetings were held

with health care providers from private and university settings

whenever they reported fewer than expected cases. Prior to the

end of the study, a surveillance bulletin was distributed to all

reporting sites, and a seminar targeting surveillance professionals

was held to present the study’s findings.

Implementation of the study was done through the existing

surveillance structure. To provide the additional workforce nec-

essary, the municipal surveillance team was supplemented by

12 staff (1 manager, 8 nurse assistants, 2 laboratory technicians,

and 1 driver). One person oversaw rash and fever surveillance in

each of the 5 health districts of the city. Data were reported

directly to the municipal level responsible for managing the sur-

veillance system.

Data CollectionA standard case reporting form was used to collect demographic,

clinical, and risk factor information for all nonvesicular febrile-

rash illnesses in individuals aged ,40 years. To facilitate private

sector participation, a simplified notification form was developed

for their use. Fever was defined as axillary temperature .37.8�Cand rash as recent presence of generalized red lesions on the skin.

Completed forms were sent daily to the district surveillance

team who identified missing information or specimens and

sought collection of this information through home visits.

Surveillance was passive but was supplemented by weekly or

biweekly active case finding in 3 selected emergency departments

and 3 urgent care clinics that had historically underreported

measles and rubella cases.

Specimen Collection and ProcessingThe specimen collection and testing protocol assumed that some

patients would receive a presumptive clinical diagnosis at the

time of presentation for care, and this diagnosis was factored

into the algorithm. At the time of initial presentation to the

health care provider, 3–5 mL of blood was drawn for laboratory

testing, unless rubella, dengue, or Rocky Mountain spotted fever

(RMSF; ie, Rickettsia rickettsii infection) was the presumptive

S628 d JID 2011:204 (Suppl 2) d de Moraes et al

Dow



clinical diagnosis. If the latter occurred, the first specimen was

drawn, respectively, on the fifth, sixth, or seventh day after rash

onset. Oropharyngeal swabs were collected up to 5 days after

rash onset from all patients with cough and coryza and trans-

ported in viral transport media vials for enterovirus and ade-

novirus testing. Persons with suspected group A streptococci

(GAS) infection had oropharyngeal swabs transported in acti-

vated carbon media. A second convalescent-phase blood sample

was collected from all reported cases 3–4 weeks after rash onset,

when any missing clinical and epidemiologic data were also

collected during home visits.

Whole-blood and oropharyngeal samples were refrigerated

at 4�C and sent every other day to 1 of the 2 municipal refer-

ence laboratories in Campinas, where blood samples were

centrifuged. Samples were shipped 3–5 times weekly from these

reference laboratories to the Instituto Adolfo Lutz Laboratory in

Sao Paulo, where oropharyngeal swabs were centrifuged and

blood and oropharyngeal samples were further processed and

tested.

Laboratory ProceduresSerum specimens were tested in the laboratory for serological

evidence of the following etiologies of febrile-rash illnesses: ad-

enoviruses, dengue virus, Epstein Barr virus (EBV), enteroviruses

(coxsackieviruses B1–B6; echoviruses 4, 6, 9, 11, and 30; entero-

viruses 70 and 71; and polioviruses 1–3), human herpes virus 6

(HHV6), measles virus, parvovirus B19, R. rickettsii, and rubella

virus. Oropharyngeal swabs were tested for GAS bacterial culture

or for enteroviruses and adenoviruses viral culture.

Testing for adenovirus, enterovirus, and RMSF could only be

performed when both acute- and convalescent-phase serum

specimens were available. Testing for GAS was only performed

when an oropharyngeal swab was available (in addition to serum

samples).

The sequence in which specimens were tested followed a flow

diagram (Figure 1) determined by the patient’s clinical findings,

Brazil’s disease control priorities, and disease epidemiology. In

summary, all samples were tested initially for measles, rubella, and

dengue viruses, in that order, except when a different presumptive

clinical diagnosis was suspected by the health care professional. In

those cases, testing for the presumptive condition was done first,

followed by testing for measles, rubella, and dengue viruses. If all

results were negative, samples were further tested for HHV6 if

case patients were aged ,2 years or for parvovirus B19 if case

patients were aged $2 years. If those test results were negative,

case-by-case clinical, epidemiologic, and laboratory data were

assessed by a group of professionals who convened periodically to

determine the sequence of further testing.

Case patients with rash and fever beginning 7–14 days after

vaccination with measles- and rubella-containing vaccine were

tested for both measles and rubella, and cases were classified as

postvaccine rash if measles or rubella immunoglobulin (Ig) M

antibody was detected between the eighth to 30th day (measles) or

eighth to 56th day (rubella) after vaccination, respectively. [17].

When serum from a patient tested positive for 2 different

diseases, it was considered that one of these tests must have

yielded a false-positive result. Each case was therefore in-

dividually evaluated, and demographic and epidemiologic var-

iables were considered for final classification of etiology.

A description of tests performed, commercial kits used, and

their reported sensitivities and specificities for diagnosis of the

various infectious diseases considered is presented in Table 1.

Notification RatesNotification rates for rash and fever illness during the study

period were compared with rates of suspected measles and

rubella cases notified during the prestudy period (January 1999–

December 2002). Notification rates by presumptive diagnosis,

age group, and source of notification (private vs public) were

calculated. Final etiologies by age group are presented.

Costs of Surveillance SystemIncremental costs of VigiFex enhanced surveillance was esti-

mated during the study period. Cost of implementation of

VigiFex in addition to routine surveillance prior to the study

period was estimated.

Figure 1. Sequence for testing specimens collected from cases reportedto VigiFex surveillance study, Campinas, Brazil, 2003–2004.


Dow



Clinical Characterization of Reported CasesClinical characterization of all reported cases of rash and fever

illness was performed, considering clinical information recorded

in the case reporting form. The number of cases that would meet

the Brazilian standard case definitions for measles or rubella was

identified.

Database and Data AnalysisMicrosoft Access databases were developed for collection of

epidemiologic and laboratory data and were updated weekly.

These databases were merged weekly. Data were analyzed using

Epi Info 2002 [32].

RESULTS

Patient CharacteristicsFrom 1 May 2002 through 30 May 2003, 1248 rash and fever

illness cases in individuals aged ,40 years were reported;

657 case patients (52.6%) were male, and 837 (67.1%) were

children aged ,4 years.

The mean interval between symptom onset and presentation

to health care was 4.6 days (range, 0–30 days). At the time of

initial presentation, 1244 case patients (99.7%) had specimens

collected; 4 (0.3%) case patients refused to have blood drawn

but had oropharyngeal swabs collected. Acute-phase serum

specimens were collected for 1244 (99.7%) of 1248 cases; of

these patients, 1132 (90.7%) had convalescent-phase serum

specimens collected. However, acute- and convalescent-phase

serum samples were collected within adequate timing from

915 case patients (73.5%), with the convalescent-phase serum

collected a mean of 21 days (median, 17 days; range, 14–60 days)

after the first specimen. Oropharyngeal specimens were collected

for 244 (41.4%) of 590 case patients with respiratory symptoms.

Vaccination cards were available for 1123 case patients (90%).

Of these, 650 (57.8%) had been vaccinated against rubella and

679 (60.5%) against measles, whereas 443 (39.5%) had not been

previously vaccinated for measles or rubella.

Overall NotificationThe notification rate for all fever-rash illnesses during the study

period was 181 cases per 100 000 population aged ,40 years.

Age-specific notification rates of fever-rash illness ranged from

2644 cases per 100 000 population among those aged ,1 year to

24.6 cases per 100 000 population among those aged 30–39 years

(Table 2). Prestudy period (1999–2002) notification rates of

suspected measles plus suspected rubella in the population aged

,40 years in Campinas varied from 38.1 cases per 100 000

population in 1999 to 82.3 cases per 100 000 population in 2000

Table 1. Differential Diagnosis and Laboratory Testing Conducted, VigiFex Surveillance Study, Campinas, Brazil, 2003–2004

Infectious agent Test description Sample(s) Kit manufacturer Sensitivity/specificity, % Reference(s)

Dengue virus Capture ELISA IgM Blood PanBio 94.7a/100 [18]

Measles virus ELISA IgM, IgG,and capture IgM

Blood Behring 88.6/98.7 (IgM) [19–21]

99/97 (IgG)

Rubella virus ELISA IgM Blood Organon Teknika 100/98 (IgM) [22]

Parvovirus B19 ELISA IgM and IgG Blood Biotrin 78–90/84.5–99 (IgM) [23–25]

86–100/95–100 (IgG)

Adenovirus IgG seroconversionthrough indirectimmunofluorescence(blood) or viralisolation (oropharingeal swab)

Blood andoropharingealswab

ATCC 80–85/100 [26]

Human herpes virus 6 Indirect immunofluorescencereaction (IgM)

Blood Biotrin Not defined

Epstein-Barr virus ELISA IgM (VCA) and IgG(EBNA)

Blood Biomerieux 95/89 (IgM) [27]

Rickettsia rickettsii Indirect immunofluorescencereaction—seroconversion

Blood CDC 99.2/100 (IgG) [28]

Enterovirus Cellular cultureneutralization reactionand viral isolation and indirectimmunofluorescencewith monoclonal antibodiesfor identification

Blood andoropharingealswab

ATCC strains andChemicon monoclonalantibodies

93/99 [29, 30]

Group A Streptococcus Culture Oropharyngealswab

Copan VenturiTransystem

90–95b/100 [31]

NOTE. ATCC, American Type Culture Collection; CCA, viral capsid antigen; EBNA, Epstein-Barr virus nuclear antigen; ELISA, enzyme-linked immunosorbent

assay; Ig, immunoglobulin.a For secondary infections, the sensitivity is 55.7%.b Sensitivity depends on previous antimicrobial use and other conditions of the sample tested.


Dow



(Figure 2), with a mean rate in the period of 52.3 cases per

100 000 population aged ,40 years.

Notification by Initial Presumptive Clinical DiagnosisNotification rates of rash and fever illnesses by presumptive

clinical diagnosis are presented in Table 3. Presumptive clinical

diagnoses before laboratory testing with the highest notification

rates were HHV6 (44.2 cases per 100 000 population aged ,40

years), GAS (43.0 cases per 100 000 population aged,40 years),

rubella (27.1 cases per 100 000 population aged ,40 years), and

dengue (16.9 cases per 100 000 population aged ,40 years). The

notification rate for suspected measles or rubella was 31.6 cases

per 100 000 population aged ,40 years (Table 3 and Figure 2).

Etiology of Rash and Fever IllnessOverall, 519 (41.7%) of 1132 rash-fever cases had an etiologic

diagnosis determined by laboratory, with the highest proportions

with a laboratory-confirmed diagnosis among children aged ,1

year (59.0%) and among persons aged 30–39 years (53.7%)

(Table 2). Children aged ,10 years accounted for 455 (87.7%) of

cases with an identified etiology (Table 2). The most common

laboratory-determined etiologies were HHV6 (312 cases), EBV

(66 cases), parvovirus (30 cases), rubella virus (30 cases), GAS (30

cases), and dengue virus (22 cases) (Table 4).

No cases of measles were confirmed, supporting Brazil’s

achievement of measles elimination. Two hundred two rash

and fever cases were reported as suspected rubella, of which 8

were laboratory-confirmed. An additional 22 rubella cases

were detected through laboratory testing, for a total of 30

cases. Two of these 22 additional cases actually met Brazil’s

clinical case definition for rubella and thus should have been

reported as suspected rubella. These 2 cases were diagnosed as

dengue infection. The majority (n 5 18) of diagnosed rubella

cases (n 5 30) occurred in children aged ,4 years (60.3%).

Table 2. Notification Rate of Rash and Fever Illnesses and Laboratory-Confirmed Diagnosis by Age Group, VigiFex Surveillance Study,Campinas, Brazil, 2003–2004

Age group,

years Population

Number of

notified cases (n)

Notification

rate per 100 000 population

Number (%) of notified

cases with laboratory-confirmed

diagnosis

,1 14 937 395 2644.44 233 (59.0)

1–4 63 028 442 701.28 167 (37.8)

5–9 78 381 213 271.75 55 (25.8)

10–14 85 266 51 59.81 10 (19.6)

15–19 93 913 32 34.07 11 (34.4)

20–29 187 414 74 39.48 21 (28.4)

30–39 166 453 41 24.63 22 (53.7)

Total 689 392 1248 181 519 (41.7)

NOTE. Diagnosis was considered preliminary for cases with positive human herpes virus 6 test results.

Figure 2. Annual notification rates of confirmed rubella/measles cases, all rash and fever illness, and other etiologies meeting the rubella or measles casedefinition in Campinas, Brazil, per 100 000 population aged ,40 years, 1999–2004.


Dow



Of the 4 postvaccine rash cases identified, 1 was rubella and 3

were associated with measles vaccine. Four cases had positive

laboratory results for both HHV6 and another etiology (par-

vovirus and EBV, 2 each) in the first serum sample drawn. For

these patients, final diagnosis was reached by considering the

patient’s age and the test results for the convalescent-phase se-

rum samples. Six cases of rash were classified as allergic reactions

to antibiotics on clinical grounds.

The distribution of fever-rash illness etiologies varied by age

group (Table 4). Among children aged ,5 years, HHV6 was the

most common etiology. Among children aged 5–9 years, EBV

followed by parvovirus, HHV6, and GAS were the most frequent

etiologies. Dengue was the most common etiology among adults

aged .30 years (Table 4). The number of reported cases of

HHV6 and EBV was higher during August–October.

Among young adults aged 20–29 years, rubella was the most

frequent etiology, which is consistent with the rubella virus

circulation pattern in Brazil in that period.

Clinical Characterization of Reported CasesOf 1012 reported cases with presumptive diagnoses other than

measles or rubella, 727 (71.8%) met the suspected measles case

definition, 293 (29%) met the suspected rubella case definition,

and 181 (17.8%) met the suspected case definitions for both

measles and rubella. Considering cases that would have met the

clinical case definition for either measles or rubella, the sus-

pected measles notification rate was 105.2 cases per 100 000

population aged ,40 years, and the suspected rubella notifica-

tion rate was 42.5 cases per 100 000 population aged ,40 years

during the study period (Figure 2).

Reporting by SectorDuring the study period, the proportion of rash and fever ill-

nesses reported by the private sector rose from 1.6% to 6.1% and

the proportion reported by the university sector rose from 5.2%

to 28%, while the proportion reported by the public sector de-

creased from 93.9% to 72.0%, compared with the prestudy

period. However, the absolute number of rash and fever cases

notified by the public sector increased from an annual peak of

445 in 2000 to 821 during the study period.

Costs of Surveillance SystemThe incremental cost of the study VigiFex surveillance was

US$229 895, of which human resources accounted for

US$117 946 (51.3%) and laboratory tests for US$73 720 (32%).

Considering a total of 1248 rash and fever cases reported, this

represents an additional cost of US$184 per case.

DISCUSSION

This article reports the largest study, to our knowledge, of eti-

ologies of fever-rash illness in the published literature. This study

was conducted in a highly measles- and rubella-immunized

population, as were other studies [1, 3, 5]. In general, disease

etiologies were similar to those in previous studies, although

their distribution varied. The proportion of cases with any

laboratory-confirmed diagnosis was similar to that reported

by Ramsay et al [1] in England (48%) and Davidkin et al [3]

in Finland (41%) but lower than that found by Oliveira et al [5]

in another region of Brazil (71%). However, the distribution

of etiologies differed. Our study found that 2.5% of cases were

Table 3. Notification Rate of Rash and Fever Illnesses by Suspected Presumptive Clinical Diagnosis, VigiFex Surveillance Study,Campinas, Brazil, 1 May 2003–30 May 2004

Suspected presumptive

clinical diagnosis

Number of

notified cases

Percentage of

overall notified rash

and illness cases

Notification rate

per 100 000 population aged

,40 years

Adenovirus-related disease 16 1.3 2.1

Brucellosis 1 0.1 0.1

Dengue fever 126 10.1 16.9

Enterovirus-related disease 86 6.9 11.5

Erythema infectiosum/fifthdisease (parvovirus B-19)

34 2.7 4.6

Group A streptococci disease 321 25.7 43

Roseola infantum/exanthem subitum/sixthdisease (human herpes virus type 6)

330 26.4 44.2

Rocky Mountain spotted fever 13 1 1.7

Kawasaki syndrome 3 0.2 0.4

Mononucleosis (Epstein-Barr virus) 39 3.1 5.2

Post-medication febrile rash 22 1.8 2.9

Rubella 202 16.2 27.1

Measles 34 2.7 4.6

Other 21 1.7 2.8

Total 1248 100 181


Dow



Table 4. Notified Cases and Final Diagnosis by Age Group During VigiFex Surveillance Study, Campinas, Brazil, 2003–2004

Age

group,

years Measles virusa Rubella virusbParvo-

virus-B19c HHV-6dEntero-

viruseAdeno-

virusf

Epstein

Barr

virusgPost-

vaccinehKawasaki

syndromeiRickettsia

rickettsii j Leptospirosisk

Allergic

reactionto

anti-bioticslGroup A

Streptococcusm Dengue virusn

,1 0 4 1 216 4 2 2 0 1 0 0 2 1 0

1–4 0 14 9 79 2 4 35 4 0 0 0 1 19 0

5–9 0 1 9 9 1 1 22 0 0 0 0 1 9 2

10–14 0 2 1 2 1 0 3 0 0 0 0 0 1 0

15–19 0 0 1 2 0 0 2 0 0 1 0 2 0 3

20–29 0 7 5 1 1 0 1 0 0 0 1 0 0 5

30–39 0 2 4 3 0 0 1 0 0 0 0 0 0 12

TOTAL 0 30 30 312 9 7 66 4 1 1 1 6 30 22

NOTE. Data are number of cases. ELISA, enzyme-linked immunosorbent assay; Ig, immunoglobulin.a Behring ELISA Ig M, IgG, and capture IgM.b Organon Teknika IgM.c Biotrin ELISA IgM and IgG.d Indirect immunoflurescence reaction. This test cannot distinguish between acute and latent HHV6 infection.e Cellular culture neutralization reaction, viral isolation, indirect immunofluorescence with monoclonal antibodies (ATCC strains with Chemicon monoclonal antibodies).f American Type Culture Collection IgG seroconversion through indirect immunofluorescence or viral isolation.g bioMerieux ELISA IgM viral capsid antigen (VCA) and IgG Epstein-Barr virus nuclear antigen (EBNA).h For measles, defined as measles IgM detected in rash patient 8–30 days after measles vaccination; for rubella, defined as rubella IgM detected in rash patient 8–52 days after rubella vaccination.i Clinical diagnosis by attending physician.j United States Centers for Disease Control (US CDC) indirect immunofluorescence reaction seroconversion.k PanBio ELISA IgM.l Culture.m PanBio Capture ELISA IgM.n Clinical diagnosis. Four cases had used amoxicillin a mean of 1.3 days (range, 1–2 days) prior to rash onset.

Rash

and

Fever

Synd

rom

icSu

rveillance

dJID

2011:204(Suppl2)

dS633

Dow



attributable to parvovirus B19, compared with 9%–19% in

other studies, whereas 25% of our cases were attributed to

HHV-6 in contrast to the 2%–12% reported elsewhere [1, 3–5].

Nonetheless, the age distributions of both HHV-6 and parvo-

virus cases were similar to those previously described [1, 3].

Our findings may reflect regional and temporal variations in

febrile-rash illnesses; however, they may also be influenced by

testing limitations. We diagnosed HHV6 using IgM indirect

immunofluorescence. However, HHV-6 is a ubiquitous in-

fection that may be acute or latent, and establishing a link be-

tween HHV-6 and acute rash and fever illness remains

a challenge because there is no gold-standard diagnostic test that

can distinguish between latent infection and active viral repli-

cation [33, 34]. Viral isolation is the only test that can reliably

distinguish between these 2 states; however, it lacks sensitivity

and is not practical on a population basis [35].

We identified 3 school-based clusters of GAS. Treatment of

GAS can prevent rheumatic heart disease, a documented cause

of economic burden in Brazil [36].

Almost three-quarters of the rubella cases detected through

laboratory testing were not suspected on clinical grounds, in most

part because they did not meet the case definition. Of 30 patients

with confirmed rubella, 20 did not present with lymphadenopa-

thy, which is included in the rubella case definition. These data

demonstrate that clinical case definition for rubella in place in

Brazil [12, 13] and recommended by WHO [37] may lack

sensitivity for settings with rubella and CRS elimination goals.

To our knowledge, this is also the only study designed to

assess the value and feasibility of a recommended minimum

reporting rate for suspected measles and rubella as a surveillance

indicator in settings of measles and rubella elimination. Our

study found a notification rate for all febrile-rash illness of

181 cases per 100 000 population aged ,40 years.

The overall notification rate of suspected measles (4.6) or

rubella (27.1) cases by presumptive clinical diagnosis was 31.6

cases per 100 000 population aged ,40 years. Notification rates

of suspected measles and rubella cases that would meet the

standardized clinical case definitions were 105.2 and 42.5 cases

per 100 000 population aged ,40 years, respectively. All of the

aforementioned reporting rates far exceeded the proposed in-

dicator of $2 suspected measles or rubella cases cases per

100 000 population adequately investigated to monitor surveil-

lance sensitivity [14]. Although the feasibility of meeting this

benchmark has been questioned in diverse settings, our findings

demonstrate that implementing enhanced surveillance was fea-

sible at a reasonable additional cost in a setting with a well-

structured surveillance system of a developing country with an

intermediate per-capita income level.

In fact, our data raise the question of whether a reporting rate

of 2 cases per 100 000 population is sensitive enough to detect

virus circulation in regions where measles elimination has been

achieved and where rubella and CRS elimination is in progress,

such as the Americas. Additional assessments that consider

surveillance data from other countries will be needed to assess

whether this benchmark is adequate.

This study highlights the importance of incorporating the

private sector into surveillance activities. The expansion of no-

tification to all rash and fever illnesses and outreach efforts to

private health care facilities during VigiFex resulted in a marked

increase in notifications, particularly from private and university

facilities. However, the proportion of total notifications coming

from them remained low, and their participation throughout

the study was difficult to maintain.

More than 30% of the incremental costs associated with en-

hanced surveillance implemented in this study were related to

expenses for necessary laboratory tests. Approximately 20% of

all laboratory costs were used to purchase diagnostic material for

enterovirus and HHV-6. It may not be feasible to recommend

routine testing for all fever-rash cases as broadly as was done in

this study because of the high costs and difficulties associated

with this method.

This study had several limitations. Because it was confined to

a 13-month period, the average contribution of specific diseases

with multiyear epidemic cycles, such as dengue, to rash-fever ill-

ness in Campinas may have been either over- or underestimated.

Furthermore, some persons with cases of fever-rash illness

undoubtedly remained at home; because this study was facility

based, these cases would have gone undetected. As previously

mentioned, the proportion of fever and rash that could truly be

attributed to HHV-6 is questionable due to the limitations of the

HHV-6 testing available.

CONCLUSIONS AND RECOMMENDATIONS

Variations in measles and rubella case definitions and the stage

of control will likely influence measles and rubella detection

rates through surveillance. Because the current clinical case

definition for rubella lacks sensitivity, rubella testing may be

considered for all cases of febrile-rash illness, regardless of sus-

pected clinical diagnosis, in settings with rubella elimination

goals. Because enhanced surveillance allowed the identification

of disease clusters that would otherwise remain undetected,

making outbreak-associated GAS a notifiable disease in the

country should be considered.

Reaching the currently recommended febrile-rash illness no-

tification rate is achievable and affordable. Additional evaluation

will be needed after data are made available from countries using

this indicator in the context of measles elimination.

Finally, our findings suggest that stimulating a well-functioning

surveillance system through enhanced rash and fever surveillance

would allow reporting of potential measles cases and additional

rubella cases that may have gone unnoticed considering current

clinical surveillance case definitions and regular rubella and

measles surveillance.


Dow



Funding

This work was supported by the US Centers for Disease Control and

Prevention and the Pan American Health Organization. Laboratory and

surveillance personnel were funded by the municipal and state health de-

partment. The Campinas Medicine and Surgery Society, State University

of Campinas (UNICAMP), and PUC-Campinas University provided funds

for the communication strategies and the symposium.

Acknowledgments

The authors have submitted this manuscript on behalf of the VigiFex

Group, whose participants are listed below:

Claudia Bento Safi, Fabiana Medeiros Lopes de Oliveira, Genoefa

Aparecida Casagrande, Maria Alice Sato, Maria Cristina Siqueira M. Prini,

Mariza Natalina dos Santos, Neuza Teles de Lima Martins, Thais Fernanda

Degan, Municipal Health Department, Campinas; Marcia Regina Pacola,

Regional Direction for Surveillance, DIR XII Campinas Sao Paulo State

Health Department; Telma Regina Marques Pinto Carvalhanas, Neuma

Terezinha Rossetto Hidalgo, Flavia Helena Ciccone—Centers for Surveil-

lance, Sao Paulo State Health Department; Fernando R. de Barros, Tereza

Cristina Segatto—National Coordination for Respiratory Diseases Sur-

veillance, Brazilian Ministry of Health; Vania Martins Fontes Del Guercio,

Patrıcia Poletini—Regional Instituto Adolfo Lutz, Campinas; Luiza T. M.

Souza, Andrea Stangarlin, Suely Pires Curti, Cristina Adelaide Figueiredo,

Marcia Theobaldo, Maria Isabel de Oliveira, Ivani Bisordi Ferreira, Maria

Akiko Ishida, Terezinha Maria de Paiva, Maria do Carmo Sampaio Tavares

Timenetsky, Rita de Cassia Carmona Compagnoli, Denise Hage Russo,

Adriana Luchs, Aurea Silveira da Cruz, Silvia Colombo, Maria Cristina de

Cunto Brandileone, Instituto Adolfo Lutz, Sao Paulo.

This study was carried out in accordance with the Declaration of Hel-

sinki as revised in 2000, and approved by the Ethics Committee of the

Instituto Adolfo Lutz-Sao Paulo. Study participants were not required to

provide informed consent as this study was considered by the Ethics

Committee to be part of routine surveillance activities.

References

1. Ramsay M, Reacher M, O’Flynn C, et al. Causes of morbilliform rash in

a highly immunised English population. Arch Dis Child 2002; 87:202–6.

2. Shirley JA, Revill S, Cohen BJ, Buckley MM. Serological study of

rubella-like illnesses. J Med Virol 1987; 21:369–79.

3. Davidkin I, Valle M, Peltola H, et al. Etiology of measles- and rubella-

like illnesses in measles, mumps, and rubella-vaccinated children.

J Infect Dis 1998; 178:1567–70.

4. Papania M, Bromberg K, Grabowsky M, Bellini W, Stewart J, Erdman D.

Differential diagnosis of febrile rash illness in children, Brooklyn, 1994

[abstract K748]. In: Program and abstracts of the 36th Interscience

Conference on Antimicrobial Agents and Chemotherapy. Washington,

DC: American Society for Microbiology, 1996.

5. Oliveira SA, Siqueira MM, Camacho LA, et al. The aetiology of mac-

ulopapular rash diseases in Niteroi, State of Rio de Janeiro, Brazil: im-

plications for measles surveillance. Epidemiol Infect 2001; 127:509–16.

6. Nordin JD, Harpaz R, Harper P, Rush W. Syndromic surveillance for

measleslike illnesses in a managed care setting. J Infect Dis 2004;

189(Suppl 1):S222–S226.

7. de Quadros CA, Izurieta H, Venczel L, Carrasco P. Measles eradication in

the Americas: progress to date. J Infect Dis 2004; 189(Suppl 1):S227–S235.

8. Pan American Health Organization. Vaccines and Immunizations—

Morbidity Tables. Washington, DC: PAHO, 2006.

9. Pan American Health Organization. Immunization in the Americas—

Summary 2005. Washington, DC: PAHO, 2005.

10. Ministerio da Saude, Brasil. Casos confirmados de rubeola. Brasil e

Grandes Regioes, 1997–2006. Brasılia: Ministerio da Saude, 2007.

11. Pan American Health Organization (PAHO). Sustaining Immunization

Programs—Elimination of rubella and congenital rubella syndrome

(CRS). In: 44th Directing Council CD44/11. Washington, DC: PAHO,

2003.

12. BRASIL.Fundacxao Nacional de Saude. Guia de Vigilancia

Epidemiologica. II, 5th ed. Brasilia: FUNASA, 2002.

13. BRASIL.Ministerio da Saude. Secretaria de Vigilancia em Saude.

Manual de Vigilancia para a Erradicacxao do Sarampo, Controle da

Rubeola e Eliminacxao da Sındrome da Rubeola Congenita (SRC), 3rd

ed. Brasilia: Secretaria de Vigilancia em Saude, 2003.

14. Pan American Health Organization (PAHO). Plan of action for the

documentation and verification of measles, rubella, and congenital rubella

syndrome elimination in the Region of the Americas. Washington, DC:

PAHO, 2009.

15. Kemp B, Toscano C, Barros ENC, Barros FR, Moraes JC. Incorporacxao do

setor privado de saude no sistema de vigilancia epidemiologica de febre e

exantema em Campinas /SP: Licxoes aprendidas. In: Anais do 11� CON-

GRESSO MUNDIAL DE SAUDE PUBLICA E 8� CONGRESSO BRASI-

LEIRO DE SAUDE COLETIVA. Rio de Janeiro, Brazil: ABRASCO, 2006.

16. Barros ENC. Praticas em Vigilancia Epidemiologica: Maneiras de Ver e

Fazer. [Dissertacxao de Mestrado]. Campinas (SP): Universidade Es-

tadual de Campinas, 2004.

17. Pan American Health Organization. Classification of suspect measles/

rubella cases as ‘‘vaccine-related’’: compliance with PAHO recom-

mendations. In: Immunization Newsletter. Vol 28. Washington, DC:

Pan American Health Organization, 2006:5.

18. Innis BL, Nisalak A, Nimmannitya S, et al. An enzyme-linked im-

munosorbent assay to characterize dengue infections where dengue

and Japanese encephalitis co-circulate. Am J Trop Med Hyg 1989; 40:

418–27.

19. Ratnam S, Tipples G, Head C, Fauvel M, Fearon M, Ward BJ. Per-

formance of indirect immunoglobulin M (IgM) serology tests and IgM

capture assays for laboratory diagnosis of measles. J Clin Microbiol

2000; 38:99–104.

20. Tipples GA, Hamkar R, Mohktari-Azad T, et al. Assessment of im-

munoglobulin M enzyme immunoassays for diagnosis of measles. J

Clin Microbiol 2003; 41:4790–2.

21. Hesketh L, Charlett A, Farrington P, Miller E, Forsey T, Morgan-

Capner P. An evaluation of nine commercial EIA kits for the detection

of measles specific IgG. J Virol Methods 1997; 66:51–9.

22. Revello MG, Percivalle E, Zavattoni M, Gerna G. Rubella IgM antibody

determination: comparison of two indirect and two capture com-

mercial enzyme immunoassays. Microbiologica 1987; 10:393–401.

23. Bruu AL, Nordbo SA. Evaluation of five commercial tests for detection

of immunoglobulin M antibodies to human parvovirus B19. J Clin

Microbiol 1995; 33:1363–5.

24. Doyle S, Kerr S, O’Keeffe G, O’Carroll D, Daly P, Kilty C. Detection

of parvovirus B19 IgM by antibody capture enzyme immunoassay:

receiver operating characteristic analysis. J Virol Methods 2000; 90:

143–52.

25. Wildig J, Michon P, Siba P, et al. Parvovirus B19 infection contributes

to severe anemia in young children in Papua New Guinea. J Infect Dis

2006; 194:146–53.

26. Takimoto S, Grandien M, Ishida MA, et al. Comparison of enzyme-

linked immunosorbent assay, indirect immunofluorescence assay, and

virus isolation for detection of respiratory viruses in nasopharyngeal

secretions. J Clin Microbiol 1991; 29:470–4.

27. Bruu AL, Hjetland R, Holter E, et al. Evaluation of 12 commercial tests

for detection of Epstein-Barr virus-specific and heterophile antibodies.

Clin Diagn Lab Immunol 2000; 7:451–6.

28. Philip RN, Casper EA, MacCormack JN, et al. A comparison of sero-

logic methods for diagnosis of Rocky Mountain spotted fever. Am

J Epidemiol 1977; 105:56–67.

29. Bastis D, Simonet S, Patterson MA, Neill S. Identification of enter-

oviruses by indirect immunofluorescence using monoclonal anti-

bodies. Clin Diagn Virol 1995; 3:83–93.

30. Rigonan AS, Mann L, Chonmaitree T. Use of monoclonal antibodies

to identify serotypes of enterovirus isolates. J Clin Microbiol 1998; 36:

1877–81.


Dow



31. Facklam RR, Washington JA II. Streptococcus and related catalase-

negative Gram positive cocci. In Balows A, Hausler WJ Jr, Hermann

KL, Isenberg HD, Shadomy HJ eds. Washington, DC: American Society

for Microbiology, 1991:238–57.

32. Dean AD, Dean JA, Burton JH, Dicker RC. Epi Info, version 6.04:

a word processing, database, and statistics program for epidemiology

on microcomputers. [version 6.04]. Atlanta, GA: Centers for Disease

Control and Prevention, 1998.

33. Caserta MT, McDermott MP, Dewhurst S, et al. Human herpesvirus

6 (HHV6) DNA persistence and reactivation in healthy children.

J Pediatr 2004; 145:478–84.

34. Carrigan DR, Knox KK. Human herpesvirus 6: diagnosis of active

infection. Am Clin Lab 2000; 19:12.

35. Bland RM, Mackie PL, Shorts T, Pate S, Paton JY. The rapid di-

agnosis and clinical features of human herpesvirus 6. J Infect 1998;

36:161–5.

36. Terreri MT, Ferraz MB, Goldenberg J, Len C, Hilario MO. Resource

utilization and cost of rheumatic fever. J Rheumatol 2001; 28:1394–7.

37. World Health Organization (WHO)/Department of Vaccines and Bi-

ologicals/Vaccine Assessment and Monitoring Team. Guidelines for

surveillance of congenital rubella syndrome and rubella. WHO/V&B/

99.22. Geneva, Switzerland: WHO, 1999.


Dow



Etiologies of Rash and Fever Illnesses in ... - Oxford Academic

Documents