Slow maturation of IgG1 avidity and persistence of specific IgM in congenital rubella: implications for diagnosis and immunopathology.

Without appropriately timed specimens, serological confirmation of congenital rubella infection may be a problem. We have compared the persistence of specific IgM and low avidity specific IgG1 in 141 sera from 120 cases of serologically confirmed congenital rubella infection with the known time scales for postnatal primary rubella. The results demonstrate that the maturation of the immune response to the rubella virus is abnormally slow in congenital rubella cases both in terms of the isotype switch and especially the development of high avidity specific IgG1. Thus avidity studies may permit serological confirmation of congenital rubella for longer than is possible with tests currently in use. The pathological implications of prolonged low avidity antibody production are discussed.

Outcome in newborn babies given anti-varicella-zoster immunoglobulin after perinatal maternal infection with varicella-zoster virus.

281 newborn babies whose mothers had chickenpox and 25 whose mothers had herpes zoster during the perinatal period were investigated. IgG antibody was present at birth in all babies born more than 7 days after the onset of maternal chickenpox. When the mother's rash appeared 7-3 days before delivery progressively fewer babies were born with antibody, and no infant born less than 3 days after the onset had antibody at birth. Infants were given 100 or 250 mg of anti-varicella-zoster immunoglobulin (VZIG) shortly after birth or the onset of post-natal maternal chickenpox. 1 infant died, without neonatal varicella. Of the 280 surviving infants 169 (60%) were infected--134 (48%) with chickenpox and 35 (13%) without clinical features. The clinical attack rate was highest (60%) in infants whose mothers had chickenpox between 7 days before and 7 days after delivery. Chickenpox was severe in 19 infants, 16 of whom were exposed to maternal chickenpox between 4 days before and 2 days after delivery. There was no evidence that a severe outcome was associated with transplacentally acquired infection. Perinatal maternal herpes zoster did not cause neonatal infection. VZIG should be given to infants at risk, including those whose mothers have chickenpox during the last 7 days of pregnancy.

Detection of rubella virus in fetal and placental tissues and in the throats of neonates after serologically confirmed rubella in pregnancy.

From 35 therapeutic abortions performed because rubella had occurred at 2-19 weeks of pregnancy, 120 fetal organs, 12 specimens of mixed products of conception, and 15 placentae were tested for rubella virus. Virus was isolated from 10 out of 11 fetuses (91 per cent) from women infected at 2-8 weeks, from 5 out of 8 (63 per cent) infected at 9-10 weeks, and from 2 out of 16 (13 per cent) infected at 11-19 weeks. Hybridization tests for viral RNA on 39 fetal organs from eight cases revealed infection in four additional fetuses. Virus was isolated from only 3 out of 15 aborted placentae, but hybridization tests on six placentae revealed infection in three additional specimens. Hybridization was superior to virus isolation for detecting rubella infection in products of conception and is therefore potentially the better method for examining chorionic villus biopsies. Rubella virus was isolated from the throats of 4 out of 9 infants (44 per cent) infected during the first 12 weeks of gestation, but from none of 13 infected after 17 weeks. Infants in the latter group are unlikely to infect susceptible contacts.

Detection of specific IgM in varicella and herpes zoster by antibody-capture radioimmunoassay.

A simple and sensitive M antibody-capture radioimmunoassay (MACRIA) is described which utilizes crude commercial VZV antigen and a single monoclonal anti-VZV antibody. This was compared to the immunofluorescence (IF) test for IgM antibody and was used to study IgM responses in sera from 261 patients with varicella and 220 patients with herpes zoster. With MACRIA, IgM antibodies were detected in all patients with varicella. The IgM antibodies appeared shortly after onset of rash, reached peak levels between 1 and 4 weeks after onset and then declined to low or undetectable levels in most, though not all, patients after 3 months. IgM antibodies were also detected in 98.2% of patients with herpes zoster, but the levels of IgM were significantly lower than after varicella and there was wider individual variation both in magnitude and duration of the IgM responses, in some cases only lasting 2-3 weeks. Comparison between MACRIA and IF showed good agreement in the detection of IgM antibodies following varicella. Discordant results were obtained with 13% of sera, of which 81% were taken either early or late after onset of rash and contained very low IgM levels. In contrast, 62 (28%) of the 220 sera from patients with zoster gave discordant results in the two tests, all except five being MACRIA-positive but IF-negative. The largest proportion of discordant results were obtained with sera taken more than 3 months after onset of rash, but 18 (29%) contained high IgM levels and were taken during the period of peak IgM responses. The diagnostic applications of the VZV MACRIA are discussed.

Outcome of confirmed periconceptional maternal rubella.

61 pregnant women in whom confirmed rubella occurred from 5 weeks before to 6 weeks after the last menstrual period (LMP) were followed up prospectively. In 39, the pregnancy was terminated and the fetal tissues or mixed products of conception were examined for rubella virus. In 22, the pregnancy continued to term and cord serum was tested for specific IgM antibody. No evidence of intrauterine infection was found in 38 pregnancies in which the mother's rash appeared before, or within 11 days after, the last menstrual period. The shortest interval at which fetal infection occurred was when the rash appeared 12 days after the last menstrual period. All 10 pregnancies in which the rash appeared 3-6 weeks after the last menstrual period resulted in fetal infection: 4 of these pregnancies went to term, and all 4 infants were damaged. The risk to the fetus when rubella occurs before the mother's last menstrual period is probably negligible.

Effect of selective vaccination on rubella susceptibility and infection in pregnancy.

The effect of school and adult vaccination on susceptibility to rubella in women of childbearing age was assessed in the Manchester area, where the population attending antenatal clinics is over 40 000 a year. Between 1979 and 1984 the proportion susceptible fell from 6.4% to 2.7%. In 1984, 4.2% of nulliparous women were susceptible compared with 1.4% of women in their second or subsequent pregnancy. Eighty five per cent of pregnant women screened and found to be non-immune were vaccinated post partum before leaving hospital. Requests for prevaccination screening of non-pregnant women increased in response to a national campaign and at the time of local outbreaks of rubella but only two thirds of those found to be nonimmune were subsequently vaccinated. During 1983 and 1984 infection was confirmed in 57 pregnant women--2% of those non-immune. Selective vaccination has reduced susceptibility to rubella in the childbearing population, but it is suggested that mass vaccination of children of both sexes should be added to the existing policy to control circulation of wild rubella virus and reduce the risk of infection to pregnant women who remain susceptible.

Immunisation of nurses with a live varicella vaccine.

34 nurses who had no previous history of chickenpox and were seronegative to varicella zoster virus (VZV) were immunised with a live attenuated varicella vaccine (OKA-RIT strain) and followed up for up to 36 months. No major vaccine reactions were observed. At 5 months and at 1 year, 94% of the nurses had seroconverted but at 3 years only 64% had detectable antibody. Lymphocyte transformation to VZV antigen tested in seven nurses in whom antibody was no longer detectable gave positive stimulation indices. 2 of 13 vaccinated nurses who looked after children with chickenpox became infected but 1 of them had not seroconverted after vaccination. Chickenpox developed in 6 out of 7 unvaccinated seronegative nurses who looked after children with chickenpox. This study shows that the OKA-RIT strain VZV vaccine is safe and immunogenic in adults and may therefore be considered for use in susceptible health care workers. The investigations also show that both cell-mediated and antibody tests are needed for long-term assessment of immunity to chickenpox after vaccination.

Live varicella immunization in healthy non-immune nurses.

Thirty-four varicella-zoster virus (VZV) seronegative nurses were vaccinated with the live varicella vaccine (Varilrix) and followed for periods of up to 36 months. No major vaccine reactions were observed. At 5 and 12 months, 94% of the nurses had seroconverted but at 3 years, only 64% retained antibody activity. However, lymphocyte transformation to VZV antigen was positive in 7 seronegative nurses, all of whom had previously seroconverted. The one nurse who developed chickenpox had not seroconverted after vaccination. Two out of 11 seroconverted nurses had a subclinical reinfection, as shown by a rise in antibody, upon exposure to varicella. In contrast, 4 out of 5 seronegative nurses who had refused vaccination developed chickenpox. Varilrix is therefore safe, immunogenic, and protective in adults and can be considered for routine use in susceptible health workers. However, it is still uncertain whether lifelong immunity is obtained with this vaccine. Both cell-mediated and antibody tests are needed for long-term assessment of immunity to chickenpox.

Prenatal detection of rubella-specific IgM in fetal sera.

Serum specimens were obtained by fetoscopy at 19-25 weeks' gestation from four fetuses whose mothers had had confirmed rubella earlier in pregnancy. They were tested for rubella-specific IgM by antibody capture radioimmunoassay. No specific IgM was detected in one fetus and a healthy infant was delivered at term. Specific IgM was detected in the other three fetuses. In one case the level was low (1 unit) and this pregnancy went to term resulting in a neonate with clinical and laboratory evidence of congenital rubella infection. The remaining two fetuses had 2.8 and 2.4 units of specific IgM and the pregnancies were terminated. Blood obtained from these two fetuses after abortion showed levels of 5.4 and 2.9 units respectively. No specific IgM was detected in sera from eleven other fetuses aborted because of maternal rubella but five of these cases were terminated before 19 weeks and in five the interval between rash and abortion was three weeks or less. The results show that the human fetus can produce detectable specific IgM antibody by 19-20 weeks' gestation after exposure to rubella several weeks earlier. However, a larger study is required to define the reliability of fetoscopic blood sampling for the diagnosis of intrauterine infection.

A comparison of indirect immunofluorescence and radioimmunoassay for detecting antibody to varicella-zoster virus.

Immunofluorescence (IF) and radioimmunoassay (RIA) were found to be more sensitive methods than complement fixation (CF) for detecting antibody to varicella-zoster (V-Z) virus. RIA yielded titres about 30 times greater than those obtained by IF, but for screening purposes RIA was only about six times more sensitive since the minimum serum dilutions that could be tested were 1/100 and 1/16 respectively. When 539 sera from subjects of different ages were screened for V-Z antibody, IF and RIA gave concordant results with 527 specimens (98%). When 19 patients were tested who had not previously had varicella but were experiencing primary infection with herpes simplex (HS) virus, crossreacting antibodies to V-Z antigens were detected in six patients by IF but in only two of these by RIA. IF and RIA are preferable to CF as tests for immune status because of their greater sensitivity, but weak positive reactions caused by presumptive low titres of homologous antibody or by higher titres of heterologous antibody can occur in one or both tests. Such reactions could cause difficulty in assessing the need for vaccine or for specific immune globulin, and in interpreting the response to vaccination.

Consequences of confirmed maternal rubella at successive stages of pregnancy.

Over a thousand women with confirmed rubella infection at different stages of pregnancy were followed up prospectively. Two-thirds of the women were multiparous. Pregnancy continued in 40%, and the infants were followed up after birth both clinically and serologically. The frequency of congenital infection after maternal rubella with a rash was more than 80% during the first 12 weeks of pregnancy, 54% at 13-14 weeks, and 25% at the end of the second trimester. The infection rate then rose again to reach a high figure in the last month. Follow-up was to 2 years of age--the findings in infected children being compared with those in children who had escaped infection. Rubella defects occurred in all infants infected before the 11th week (principally congenital heart disease and deafness) and in 35% of those infected at 13-16 weeks (deafness alone). No defects attributable to rubella were found in 63 children infected after 16 weeks. Continued surveillance of cases of confirmed rubella during pregnancy is recommended as an additional way of monitoring the effect of rubella vaccination.

A comparison of antibody capture radio- and enzyme immunoassays with immunofluorescence for detecting IgM antibody in infants with congenital rubella.

IgM antibody capture radioimmunoassay (MACRIA) and enzyme immunoassay (MACEIA) were compared with immunofluorescence (IF) for detecting specific IgM antibody in 99 sera from 76 infants with confirmed congenital rubella, and 61 sera from a comparative group of 59 infants who had miscellaneous abnormalities but in whom congenital rubella was not confirmed. All of 35 specimens collected from confirmed cases within 12 weeks of birth were positive by all three methods and all but one of 17 specimens collected after the age of 18 months were uniformly negative. At intermediate ages discrepancies occurred in 18 specimens, of which eight were positive and 10 negative by IF. Three of these 18 specimens were negative by both antibody capture procedures but showed weak fluorescence; the other 15 were negative by MACEIA, but positive by MACRIA which appears to be the more sensitive of the antibody capture methods. Sera from five infants in the comparative group were clearly positive by all three methods. These five infants were probably congenitally infected with rubella. Sera from the other 54 infants were negative, except for one that gave a weakly positive result by MACRIA alone. Antibody capture procedures offer several advantages over previous methods for detecting IgM antibody. Although MACRIA was found to be slightly more sensitive than MACEIA, the greater stability of the enzyme label, together with the possibility of both visual and quantitative assessment, could make MACEIA the method of choice for detecting rubella-specific IgM.

Outcome of asymptomatic infection with rubella virus during pregnancy.

We have tried to detect prenatal infection in 34 infants whose mothers were re-infected with rubella virus during pregnancy and in six infants whose mothers had primary subclinical rubella during pregnancy. Two methods of assessment were used: first, serum obtained soon after birth was tested for IgM antibody; secondly, serum obtained after the age of 8 months was tested for specific IgG. The 34 women with re-infections had increases in IgG antibody titre but no IgM response. No evidence of prenatal infection was found in 33 of their 34 infants. One infant was found to have IgG antibody at the age of 11 months. This infant was IgM-negative at birth and had a rubelliform rash at the age of 5 1/2 months; it therefore probably contracted post- rather than pre-natal infection. Fetal infection from maternal re-infection during pregnancy is probably rare. The six women with primary subclinical rubella produced both IgG and IgM classes of antibody. Three of their six infants showed serological evidence of intrauterine infection. One, infected when its mother was 8 weeks pregnant, had clinical evidence of congenital rubella. Primary subclinical rubella during pregnancy therefore carries a significant risk of fetal infection. Because of the difference in outcome, great care should be taken to distinguish between primary infection and re-infection when investigating symptomless increases in antibody titre after contact with rubella during pregnancy.

Fetal infection resulting from maternal rubella after the first trimester of pregnancy.

We have tried to measure the incidence of prenatal infection in 304 infants whose mothers had had rubella at various times after the first 12 weeks of pregnancy. Two methods of assessment were used: first, serum obtained soon after birth was tested for specific IgM antibody; secondly, serum obtained after the age of eight months was tested for specific IgG. When maternal rubella occurred 12-16 weeks after the last menstrual period specific IgM antibody was detected in 28 out of 50 infants (56%). The proportion fell progressively to 12% after maternal rubella at 24-28 weeks, rose to 19% after rubella at 28-36 weeks and then to 58% when the illness occurred during the last month of pregnancy. In all, IgM antibody was detected in 77 out of 260 infants (29%). The fetus can thus be infected at any time during the second and third trimesters of pregnancy, but the risk varies at different stages.The figures for the prevalence of IgG antibody were greater throughout, because some infants had IgG who had previously lacked specific IgM. After maternal rubella at 12-16 weeks IgG antibody persisted in 22 out of 31 infants (71%). The proportion fell to 28% after rubella at 24-28 weeks and then increased progressively to 94% after rubella during the last month. In all, IgG antibody persisted in 94 out of 190 infants (49%). The true rate of fetal infection probably lies between the rates estimated from the presence of IgM antibody and the subsequent prevalence of IgG.Infants whose mothers had rubella at any time during pregnancy should be examined regularly for possible evidence of damage.

Human anti-chickenpox immunoglobulin in the prevention of chickenpox.

Human anti-chickenpox immunoglobulin (zoster immune globulin, ZIG) was largely ineffective in preventing infection in forty-three high-risk contacts of chickenpox. Twenty-nine of these non-immune infants and children who had been in close contact with cases of varicella became infected, and symptoms developed in twenty-four. Since ZIG may modify chickenpox it should continue to be given to high-risk contacts until the availability of a simple and sensitive test makes it possible to identify those who are susceptible. However, ZIG is not necesary for infants whose mothers have chickenpox or zoster five or more days before delivery, since all such infants (eleven) had varicella-zoster antibody at birth.

Comparison of immunofluorescence and radioimmunoassay for detecting IgM antibody in infants with the congenital rubella syndrome.

Immunofluorescence (IF) and radioimmunoassay (RIA) have been compared as methods for detecting IgM antibody in 124 infants with confirmed or suspected congenital rubella. IF was used to test sucrose density gradient fractions and RIA to test fractions and whole serum. When fractions were tested IF and RIA were equally specific and distinguished clearly between IgM and IgG, but RIA was the more sensitive method. The RIA titre in whole serum was always greater than in the peak IgM fraction and there was no evidence that testing the serum, rather than the fraction, could result in failure to detect IgM. With some sera RIA gave low titres which became negative after absorption with IgG-coated latex beads. The mechanism of this 'false positive' effect, which may have been due to IgM with anti-IgG activity, was not investigated, but if it can be removed by absorption it need not reduce the specificity of the test. During the first 6 months of life IgM antibody was detected by RIA in 30 out of 32 unfractionated sera and by IF in fractions from 28 of these. After the age of 6 months IgM was found progressively less frequently and the greater sensitivity of RIA became a more obvious advantage: 17 out of 60 specimens were positive by RIA and 11 of these were negative by IF. RIA testing of whole serum appears to be an economical, specific and sensitive method for detecting IgM antibody in congenital rubella, of particular value when the titre of antibody is low.

Specific immunoglobulin responses after varicella and herpes zoster.

The indirect immunofluorescence technique has been used to titrate the specific immunoglobulins in 200 sera from 64 patients with varicella, and 195 sera from 67 patients with herpes zoster. IgG and IgM antibodies were detected in all patients with varicella, and IgA in 59 (92%). All three classes of antibody appeared 2--5 days after the onset of the rash, increased virtually simultaneously and reached maximum titres during the second and third weeks. IgG then declined slowly, but never became undetectable and was still present in five subjects who were retested after 2--4 years; it was present in 88 out of 100 healthy young adults and probably persists indefinitely after varicella. IgA and IgM antibodies declined more rapidly and were not detected in specimens taken more than a year after the illness. IgA, however, may possibly persist in some cases since low titres were found in 8 out of 88 young adults who possessed IgG antibody and had presumably had varicella in the past. IgA responses were significantly weaker in children under the age of 6 years than in older children and adults. Six out of 67 patients with zoster were tested at various times before the onset of the rash: IgG antibody was detected in all. IgG was present in all sera taken after the onset of the rash, increased rapidly after 2--5 days, reached maximum titres during the second and third weeks and then declined slowly. IgA antibody was detected in 66 patients (99%) and IgM in 52 (78%); both types of antibody followed transient courses, as in varicella. Maximum titres of IgG and complement-fixing antibodies were greater after zoster than after varicella, but the differences were not significant. IgA and IgM titres in young adults with zoster were significantly lower than in older patients, and also lower than in young adults with varicella. Increases in varicella-zoster antibody in patients with herpes simplex virus infections consisted mainly of IgG, sometimes IgA, but never IgM.