HUMAN PARVOVIRUS B 19 INFECTIONS AMONG EXANTHEMATIC DISEASES NOTIFIED AS MEASLES

finding in the serum of healthy blood donors is now known to cause a spectrum of disease in humans. The most common c l in ica l presentation of B19 infection is the childhood exanthem, erythema infectiosum (fifth disease). This classically begins with a facial rash (slapped cheek) that spreads to the trunk and limbs where it takes on a reticulate or lace-like appearance . However the exanthem is highly variable, and may be misdiagnosed as rubella if laboratory investigation is not performed. In Rio de Janeiro State, rubella and measles notifications are higher during the second half of the year . B19 infections were also reported in Rio de Janeiro for the same period . Since the National Measles Vaccination Campaign in Brazil, during which all children under 14 years old were vaccinated against measles (April to May, 1992), all patients presenting with rash, high fever, and at least one of the following symptoms: cough, coryza or conjunctivitis, should be investigated for evidence of measles infection.

Parvovirus B19, first described as a chance finding in the serum of healthy blood donors 8, is now known to cause a spectrum of disease in humans.The most common clinical presentation of B19 infection is the childhood exanthem, erythema infectiosum (fifth disease).This classically begins with a facial rash (slapped cheek) that spreads to the trunk and limbs where it takes on a reticulate or lace-like appearance 3 20 .However the exanthem is highly variable, and may be misdiagnosed as rubella if laboratory investigation is not performed 22 .In Rio de Janeiro State, rubella and measles notifications are higher during the second half of the year 5 21 .B19 infections were also reported in Rio de Janeiro for the same period 10 11 .
Since the National Measles Vaccination Campaign in Brazil, during which all children under 14 years old were vaccinated against measles (April to May, 1992), all patients presenting with rash, high fever, and at least one of the following symptoms: cough, coryza or conjunctivitis, should be investigated for evidence of measles infection 5 .
In this study all patients had initially been investigated for evidence of measles by detection of measles specific IgM.If this proved negative, their stored sera were tested for the presence of rubella specific IgM.As B19 virus has also been shown to cause exanthematic disease, we used all the negative sera for measles and rubella IgM, to search for evidence of recent human parvovirus B19 and study the occurrence of fifth disease in the states of Espírito Santo and Rio de Janeiro, Brazil.Diagnosis of human parvovirus B19 infection.Detection of specific anti-B19 IgM: All sera negative for measles and rubella IgM were tested for human parvovirus B19 specific IgM by an "in house" IgM capture enzyme immunoassay (MACEIA), as described previously 9 .For this test, serum containing 100, 33, 10, 3.3, 1.0 and 0.33 radioimmunoassay (RIA) arbritary units (a.u.) of B19 IgM and normal human serum were used as positive and negative controls.Sera containing P/N values equal or greater than that one obtained with the 10 RIA a.u.control serum were considered as positive.

P a t i e n t ' s
Detection of specific anti-B19 IgG: All anti-B19 IgM positive cases were tested for the presence of anti-B19 IgG antibodies by an "in house" enzyme immunoassay using recombinant B19 capsids produced in insect cells (Sf9) as antigen 6 .Briefly, flat bottomed microtiter plates (Immulon 2, Dynatech) were coated with 50µl of recombinant B19 capsids (1µg/ml; MedImmune Inc., USA) or 4% of SF9 cells diluted 1:1000 in 0.05M carbonate buffer (pH 9.6) and the plates were incubated for 1.5h at 37 o C. The plates were washed three times with PBS (pH 7.2) containing 0.05% Tween 20 (PBST), then 200µl of blocking buffer (PBS pH 7.2 plus 0.1% bovine serum albumine) was added, and the microtiter plates incubated for 1.5h at 37 o C.After washing the plates, 50µl of patient serum diluted 1:20 in diluent buffer (blocking buffer plus 0.5% Tween 20) was added to wells in duplicate and the plates incubated for 1.5h at 37 o C.After the wells were washed, 50µl of peroxidase-conjugate goat anti-human IgG (Sigma) diluted 1:20,000 in diluent buffer was added, and the plates incubated for 1h at 37 o C. Finally, the plates were washed five times and incubated in the dark at room temperature with 100µl of substrate solution (tetramethylbenzidine in phosphate-citrate buffer).The reaction was stopped by the addition of 100µl of 2M H 2 SO 4 , and the absorbance at 450nm was measured with a Titertek Multiscan Plus spectrophotometer.Positive and negative B19 IgG sera were used as controls.The absorbance values obtained in the wells with the recombinant antigen (P) were compared to that one with the insect cells (N).Sera were considered as positive when P/N values were greater than 2.0 and the P -N values were greater than 0.10.
Detection of B19 DNA: All anti-B19 IgM positive cases were also tested for the p r e s e n c e o f B 1 9 D N A b y " d o t -b l o t " hybridization assay using a biotin-labelled probe 17 .

RESULTS
Sera were obtained from 1095 measles notified cases in RJ and ES States during July 1992 to December 1994, and tested for measles specific IgM.Measles infection was confirmed in only 52 (4.7%) cases (Table 1).About 40% of the cases were due to rubella infection, and relatively few B19 infections (27 cases -2.5%) were identified.However, among the cases received from ES, almost the same number of positives for measles and B19 infection were obtained.For RJ, during the period studied there was an increase in the incidence of B19 infection, while the number of cases due to measles infection was decreasing.
The seasonal distribution of B19 and rubella cases in ES and RJ States was examined (Figures 1 and 2).Rubella and B19 infection had a peak occurrence in the second half of the year, late winter and spring time; with sporadic cases of B19 and rubella during the first half of the year.
A total of 301 sera from 194 cases negative for measles and rubella IgM were obtained from other Brazilian states and investigated for B19 IgM.Twenty six cases were received from Pará (north Brazil) in the year of 1989 and B19 IgM could be detected in 5 (19.2%) cases.Among 47 cases from Paraná (south Brazil) in 1993, B19 infection was confirmed for 5 (10.6%) of these cases while for Distrito Federal we found only one positive (2.6%) in 38 cases in 1992.
From the 38 cases IgM positive for B19, for 19 of them acute and convalescent sera were available (Table 2).From 38 sera collected from   acute phase only two of them were negative for B19 IgM.B19 DNA could be detected in one of these IgM negative sera by dot blot hybridization.For this patient IgM and IgG was present only in the convalescent serum.B19 IgG was detected in 31 of the 35 acute sera tested and all but one convalescent sera was positive for B19 IgG.IgM and IgG antibodies for B19 virus could be detected in sera collected from 1 to 22 days after the beginning of the rash.

DISCUSSION
Since the first association made by Anderson e cols 4 of an outbreak of erythema infectiosum (EI) and B19 infection many other reports have appeared and parvovirus B19 is now known to be the etiological agent of EI 18 20 .
The first description of EI with laboratory confirmation in Brazil was done during 1987 in Belém, PA 12 .Althought anti-B19 IgM detection was possible only for three patients, similar clinical findings were reported for ten other cases during a four -year period, 1984 to 1987 16 .Most of these cases occurred during the first half of the year (January to June).During the following years (1988 to 1989) twenty-two laboratory confirmed cases were reported in Belém 14 .Another report from the same area for 1989 to 1990 found just one positive case out of 42 13 .
We are now reporting 27 (2.5%)anti-B19 IgM positive cases among 1095 cases reported as measles in RJ and ES states during the years 1992 to 1994.Shirley e cols 22 detected B19 IgM positive sera in 6.8% of 627 cases of rubella like-illness in the West Midlands, UK, also during a 2 1/2 -year period (1983 to 1985).The proportion of positive cases found in RJ and ES States is lower and this can be easily explained because EI is clinically more similar to rubella than to measles.In the West Midlands study the percentage varied from 3% to 19% for different years, and probably 1985 was an epidemic year for B19 in the UK 1 .In our present study we probably lose an epidemic period.
It is difficult to discuss the results found for another Brazilian states because only for ES and RJ states a follow-up for a 2 1/2 -year period to investigate the occurrence of measles, rubella and B19 infections was done.
Our results for etiology of exanthematic disease in RJ and ES clearly show the impact of measles mass vaccination over the measles virus circulation in both States.From the 92 cases received from ES in the years of 1992 and 1994, we had the same number of positives for measles and B19.In a previous study of an outbreak of exanthem in Niterói, RJ during 1991 we obtained 80% of laboratory confirmation of 112 measles reported cases 19 .In that study no cases with anti-B19 IgM were found and rubella was diagnosed in only four cases.
Diagnosis of erythema infectiosum is mainly based on serology, because specific antibodies are present at the onset of the symptoms.At this time, in normal individuals viraemia can no more be detected 2 .We were able to detect B19 DNA in a patient with rash.This patient had been splenectomized before and it may have contributed to the longer viraemia.Prolonged viraemia has been described to occur when immunocompromised patients are infected with B19 23 .
In our study no relationship between days of disease and antibody-detection could be observed, since IgM and, in most cases, IgG were present after one day of the onset of the rash.
In about 50% of the cases the etiology of the rash was unknown.It is well known that erythematous rashes occur in many common viral infections, like arboviruses, infectious mononucleosis, enteroviruses and the more recently described human herpesvirus 6 (HHV-6).Preliminary studies of HHV-6 infection in north-east Brazil detected seroprevalence rates of 70% and children were found to possess high level of antibodies 15 .So the etiology of these cases remains to be elucidated.
Amplify this study would be desirable as also to consider B19 infection as an alternative diagnosis to rubella especially during pregnancy in which B19 virus can cross the placenta and s e r a .B l o o d w a s t a k e n b y veinepuncture from cases notified as measles by health workers from the epidemiological surveillance in the States of Espírito Santo (ES) and Rio de Janeiro (RJ) for a 2 1/2-year period.The patient's sera had all been stored at -20 o C until tested.Both acute and convalescent sera

Figure 1 -Figure 2 -
Figure 1 -Seasonal distribution of B19 and rubella cases in ES State.

Table 1 -
Laboratory diagnosis among measles notified cases in Rio de Janeiro (RJ) and Espírito Santo (ES) States.