| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania; and National Institute of Child Health and Human Development, Bethesda, Maryland.
Address reprint requests to: James H. Harger, MD, University of Pittsburgh School of Medicine, Department of Obstetrics, Gynecology, and Reproductive Sciences, 119 Douglas Drive, Pittsburgh, PA 15213-3180; E-mail: jharger{at}pitt.edu.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: Pregnant women with clinical varicella zoster virus infection were enrolled at ten perinatal centers. Maternal and fetal immunoglobulin (Ig) G and IgM by fluorescent antibody confirmed 74.3% of cases. Specialists examined neonates at 0–6 months, 7–18 months, and 19–30 months after delivery to detect abnormalities of their eyes, hearing, and physical and developmental features. A hierarchical set of criteria was used to define congenital varicella syndrome. A jury of four investigators assigned the classification of all findings.
RESULTS: In 362 women enrolled from 1993 to 1996, 15 had herpes zoster, and 347 had primary varicella zoster virus infection. Varicella zoster virus affected 140 women (38.7%) in the first trimester, 122 (33.7%) in the second trimester, and 100 (27.6%) in the third trimester. Five twin pairs were included. Only one case (0.4%) of definite congenital varicella syndrome was found, a 3360-g female infant having a left retinal macular lesion with typical skin scars after maternal varicella at 24 weeks. The maternal blood sample at birth was negative for IgG antibodies to toxoplasmosis and cytomegalovirus. Two cases involved fetal death at 20 weeks and fetal hydrops at 17 weeks after maternal varicella at 11 and 5 weeks, respectively. We found no cases of limb hypoplasia, microcephalus, or cataract.
CONCLUSION: The frequency of congenital varicella syndrome is very low (0.4%) in a prospectively studied cohort. Eye examinations of exposed infants had a low yield.
In the past two decades, physicians have developed increasing concern about the risk of congenital varicella syndrome, and the incidence of this syndrome has been the subject of many studies. Congenital varicella syndrome was first described by Laforet and Lynch1 in 1947. It includes chorioretinitis, congenital cataracts, cerebral cortical atrophy, variable degrees of limb atrophy, skin scarring, gastroesophageal reflux, and other anomalies. Early studies of the frequency of congenital varicella syndrome consisted of small retrospective case series2–5 with evaluation of infant outcome by incomplete and inconsistent methods. These early studies from 19602 to 19865 yielded frequencies from 0% to 9.1%. The severe morbidity observed in some of the previously reported infants led some pregnant women with varicella to choose voluntary abortion. In an effort to estimate the incidence and severity of congenital varicella syndrome more accurately, later studies were prospective and evaluated larger cohorts, but infant evaluations were still inconsistent and limited in scope and duration.6–9 Although these later studies reported an incidence of only 0–1.5%, they did not examine systematically the eyes of the infants for chorioretinitis and congenital cataracts.
Recent epidemiologic evidence suggests that maternal varicella during pregnancy may become more common in the future. The frequency of varicella in all adults has been estimated to be 1.2–4.5 per 100,000 hospital admissions in Stockholm between 1980–198910 and six of 100,000 in Scotland during 1989–1990.11 Further, Miller et al reported an increasing frequency of varicella in adults in England and Wales over the past two decades,11 and a recent upward shift in the age distribution of varicella has been reported in the United States as well.12 Although the reasons for such changes in age distribution are complex, they could herald an increased frequency of pregnant women with varicella.
There is uncertainty about the risk for congenital varicella syndrome and a need for data derived from complete physical examination of infants after birth. We performed a prospective longitudinal study of pregnant women with varicella zoster virus infection in ten tertiary perinatal centers participating in the National Institute of Child Health and Human Development Network of Maternal-Fetal Medicine Units.
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
After enrollment, the patients remained under the care of their referring physicians until delivery. They were encouraged to have ultrasound scans at 18 weeks’ gestation and once or twice in the third trimester to detect possible effects of varicella zoster virus infection in utero. At delivery, any neonatal abnormalities were recorded in addition to neonatal weight, length, head circumference, and Apgar scores.
The protocol included planned assessment of each infant at three times during the first 2 years of development to detect subtle effects of varicella zoster virus infection. The infants were examined by a pediatrician, a pediatric ophthalmologist, and an audiologist at 0–6 months after delivery, 7–18 months, and 19–30 months. The pediatrician performed a standardized physical examination after specific, detailed study protocols. At the second and third examinations, a Bayley scale or Denver Developmental Index was performed according to the preference of the examiner in each center. We defined an abnormal Bayley score as 80 or less on either the mental or motor scale. Those centers employing the Denver Index recorded the number of delays from age-appropriate landmarks and the nature of the delay, whether gross motor, fine motor, or language development.
Pediatric ophthalmologists completed standard forms to record their evaluations of infant eye anatomy for cataracts, anterior chamber disorders, and retinal lesions at each of the three protocol examinations. Both eyes were examined before and after dilatation of the pupil. Infant hearing was evaluated by audiologists who completed a standard form and employed tests appropriate for the age of the infant. The initial examination used acoustic-stimulated brainstem response or otoacoustic emissions, and later audiograms used visual reinforced audiometry when infant maturity permitted.
As case forms were completed, they were reviewed jointly by the investigators and classified by consensus into one of five categories using a hierarchical system defined from the published literature by the protocol subcommittee before the study began. A jury of four investigators decided the classification of all findings. First, "definite congenital varicella syndrome" meant the presence of typical skin scars as well as one of ten additional abnormalities: microcephalus (defined by head circumference less than the 5th percentile for age), ventriculomegaly by ultrasound (defined by a width of either lateral ventricle of more than 10 mm), cerebral cortical atrophy, chorioretinitis, cataract, limb hypoplasia (defined as one limb more than 10% longer than the other), gastrointestinal anomalies or reflux, genitourinary anomalies, sensorineural hearing loss or deficit, or at least one Bayley or Denver neurologic screening delay. The second category included children with one and only one of the above ten criteria but with no cicatricial skin scarring, termed "possible" cases of congenital varicella syndrome.
The third category, "not congenital varicella syndrome," was defined as adequate follow-up information with none of the above criteria present. Adequate follow-up information required at least one protocol ophthalmologic examination and one protocol physical examination at age 7–18 months or 19–30 months. Designation in the fourth category, "uncertain," meant that the child demonstrated some morphologic malformation or anomaly not included in the ten required for categories one and two. The fifth category, "incomplete," included cases in which the later protocol examinations at age 0–6, 7–18, or 19–30 months either were not performed or did not include at least one ophthalmologic examination.
Because clinical identification of varicella is reliable, we did not require varicella zoster virus isolation, but we did draw maternal blood samples for IgG and IgM testing at enrollment and from the umbilical cord at delivery. At the infant exam performed between 7–18 months, we drew a fingerstick sample from the infant for IgG testing in an effort to determine intrauterine varicella zoster virus infection. All blood samples were allowed to clot, centrifuged, separated into aliquots, frozen at -70C, and sent to Dr. Anne Gershon at Columbia University for testing by the fluorescent antibody to membrane antigen (FAMA) method.13 Blood samples were considered confirmatory of maternal varicella zoster virus infection when antivaricella zoster virus IgM was present at any titer. Maternal samples were considered probable if the IgM was not detected but the IgG titer was 128 or more.13
Infants classified as "definite" or "possible" had an aliquot of cord blood sent for analysis of antitoxoplasmosis IgG and IgM to Palo Alto Medical Foundation (Dr. Jack Remington). Another aliquot was sent to Dr. Stuart Adler in Richmond, Virginia, for anticytomegalovirus IgG and IgM antibody testing.
Statistical analysis was performed by calculating frequencies and exact confidence intervals. The size of the affected cohort in this study limited us to descriptive statistics such as counts, rates, and proportions.
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We enrolled 140 women (38.7%) with rash onset in the first trimester, 122 (33.7%) in the second trimester, and 100 (27.6%) in the third trimester. A total of 190 (54.8%) of the 347 primary varicella cases developed the varicella rash before 20 weeks’ gestation. Mean gestational age at rash onset was 19.0 ± 10 (standard deviation) weeks. Maternal age at rash onset was less than 21 years in 98 (27.1%), 21–25 years in 133 (36.7%), and greater than 25 years in the remaining 131 (36.2%) cases. Predominant ethnic background, classified by the woman herself, was non-Hispanic white in 150 (41.4%), black in 140 (38.7%), Hispanic in 60 (16.6%), Asian in six (1.7%), and "other" in six women (1.7%).
In considering the clinical signs at enrollment, we excluded the 15 women enrolled with secondary varicella zoster virus infection (herpes zoster). These clinical signs in 347 women with varicella included a mean peak temperature of 101 ± 1C (standard deviation, range 99–105). More than 100 vesicles at the peak of the rash were found in 186 (53.5%) of the 347 cases. Although 90 (25.9%) of these 347 women reported dyspnea, only 18 (5.2% of the 347 cases) of those 90 women had chest radiographs supporting a diagnosis of varicella pneumonia.14,15
The results of this study reveal a very low incidence of congenital varicella syndrome, one of 231 (0.4%) infants with satisfactory follow-up data including at least one eye examination, when born to women infected by varicella zoster virus for the first time during pregnancy. Even including one stillborn infant and one voluntary termination performed for fetal hydrops, the incidence was only three of 231 (1.3%). The 95% confidence interval for the rate of congenital varicella syndrome was 0.1, 2.4, based on the one definite case or 0.3, 3.7, if the two possible cases with fetal death and hydrops fetalis are included among the 231 infants with adequate follow-up. The final infant outcome classification was "not congenital varicella syndrome" in 188 (81.4%), "uncertain" in 13 (5.6%), "possible" in 29 (13%), and "definite congenital varicella syndrome" in only one case.
The outcome of 125 infants with insufficient follow-up examinations to assess their physical findings was classified as "incomplete," leaving 226 mothers and the 231 infants. However, these 125 infants with insufficient follow-up evaluation had outcome data that included gender, birth weight, length, Apgar scores, head circumference, and the standard physical examination performed at delivery. In 95 of these 125 infants, no further (protocol-mandated) examinations were performed because mothers formally withdrew their infants or simply failed to return for appointments. In the other 30 cases, protocol physical examinations and/or audiology examinations were obtained, but not even one of the three planned, required eye examinations was performed. The length of the follow-up period varied according to a variety of factors. Table 1
displays the number of infants having each of the three protocol examinations, and that number clearly diminished over time.
|
Two cases of fetal loss may have been attributable to congenital varicella zoster virus infection, though there was no serologic or virus isolation confirmation. One mother developed varicella at 5 weeks’ gestation, and ultrasound at 17 weeks revealed severe hydrops fetalis. The pregnancy was terminated voluntarily by dilation and extraction, but no histologic evidence of varicella zoster virus infection was found. In the second case, a 335-g stillborn male fetus was delivered at 20 weeks’ gestation after maternal rash onset at 11 weeks. Hydrops fetalis was found at autopsy, but no specific histologic evidence of varicella zoster virus infection was detected. Both of these cases were classified as "possible," as was a 2430-g small for gestational age female infant born at 40 weeks after maternal varicella at 30 weeks. The only morbidity in this infant was her head circumference (31.5 cm) at 5 days, which was less than the 5th percentile, but there was no protocol follow-up examination.
A single case of neonatal varicella was observed 20 days after delivery in a 3375-g male infant born at 39 weeks’ gestation. Maternal varicella rash had developed at 37 weeks. At the first protocol examination, no infant morbidity was noted. No further protocol examinations were obtained.
None of the infants displayed any changes on audiology examination that suggested neurologic hearing deficits.
Maternal blood samples were collected at enrollment in 334 cases; no sample was obtained in 13. Of the 334 samples tested, 198 (59.3%) were positive for antivaricella zoster virus IgM with titers ranging from 2 to 8 or more. Of the 136 maternal enrollment samples with an IgM titer less than 2 after rash onset, 20 (14.7%) had an antivaricella zoster virus IgG titer more than 128, another 30 (22.1%) had an antivaricella zoster virus IgG titer of 128, and 86 women (63.2%) had an antivaricella zoster virus IgG titer of 64 or less. Thus, a total of 248 (74.3%) of the 334 women tested had serologic confirmation of their recent varicella zoster virus infection.13
Fetal blood samples were obtained from the umbilical cord at the time of delivery in 261 cases, but only two cases (0.77%) were positive for antivaricella zoster virus IgM. In one case, the initial maternal IgM titer was more than 8, and the cord IgM titer was 2 when measured at delivery 54 days later. The other maternal enrollment IgM titer was less than 2 when drawn 23 days after rash onset, although the cord IgM titer was 8 or more when drawn 60 days later, showing evidence of fetal varicella zoster virus infection in spite of the apparently insensitive and nonconfirmatory maternal antibody test. In the 112 blood samples obtained from the infants at age 7–18 months, 15 (13.4%) were positive for IgG antibodies at titers ranging from 2 to 16 or more.
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The differences in ascertainment and inclusion criteria may account for some of the variations in results between studies. The Brooklyn study in 1957–1964 enrolled only 135 "married women" with singleton pregnancies,3 and all other studies limited inclusion to singleton pregnancies.5,6,9 Only our study and one other included twin pregnancies.7
Another study included 194 women who contracted varicella only from known cases of varicella in children,9 whereas we included clinically identified cases regardless of the source because the source of varicella should not affect the outcome. The largest study enrolled cases in England and Germany,8 but did not define the population further. The other prospective studies all reported on women living in the temperate climates in North America.
The limited information available on the risk of maternal varicella during pregnancy has, in the past, led physicians to counsel women exposed to varicella with alarming risks. The earliest studies and reports gave risks of 0–9.1% but were based on small population sizes ranging from 22 to 113 women with varicella in the first 20 weeks’ gestation.3–5 Since 1992, larger studies have reported that the risk of congenital varicella embryopathy was 0%,6 1.2%,7 1.5%,9 and was 0.4% in the first trimester but was 2.0% with maternal rash onset between 13–20 weeks of pregnancy.8 These more recent rates were based, respectively, upon populations of 35, 86,146, and on 472 first-trimester cases and 351 second-trimester cases.6–9 In contrast, the present study evaluated 140 cases in the first trimester and 122 in the second trimester for a total of 262 at a gestational age comparable with other reports, with an additional 100 cases in the third trimester.
Because none of the previous reports systematically examined at-risk infants for chorioretinitis or for hearing defects, we chose to search for these occult lesions in addition to the more traditional ones associated with congenital varicella syndrome. Some infant examinations in previous reports included an inconsistent combination of physical examinations performed by the investigators, local pediatricians, and general practitioners,5,8,9 and by unspecified examiners.3 Previous reports state only that infant follow-up information was obtained by telephone calls to treating physicians at age 8–12 months of the infant7 or to parents 1 or more years after delivery.6 Those previous follow-up examinations did not discover many subtle, later defects associated with congenital varicella syndrome, so they may be insufficient. The eye examinations in the present study disclosed only the single case of varicella chorioretinitis. Although our intent was to have each infant assessed by trained observers at up to three temporal points during the first 2 years of development, the yield of such extensive infant evaluations may be rather low. Many infants were withdrawn from later protocol examinations because their mothers were reassured by the normal earlier evaluations that the infant had escaped the feared effects of congenital varicella infection.
We conclude that the low risk for congenital varicella syndrome in this and other studies should reassure pregnant women affected by varicella during their pregnancy. This reassurance could lead to a reduced frequency of pregnancy terminations performed to prevent congenital anomalies in these patients.
|
Footnotes |
|---|
Supported by grants #HD 21410, HD 21414, HD 21434, HD 27860, HD 27861, HD 27883, HD 27889, HD 27905, HD 27915, HD 27917, and HD 19897 from the National Institute of Child Health and Human Development, National Institutes of Health.
Received December 6, 2001. Received in revised form February 15, 2002. Accepted March 14, 2002.
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Manson MM, Logan WPD, Loy RM. Rubella and other virus infections in pregnancy. In: Reports on public health and medical subjects (No. 101), Ministry of Health, Her Majesty’s Stationery Office, London, England, 1960.
3. Siegel M. Congenital malformations following chicken-pox, measles, mumps and hepatitis. JAMA 1973;226: 1521–4.[Medline]
4. Enders G. Varicella-zoster virus infection in pregnancy. Prog Med Virol 1984;29:166–96.[Medline]
5. Paryani SG, Arvin AM. Intrauterine infection with varicella-zoster virus after maternal varicella. N Engl J Med 1986;314:1542–6.[Abstract]
6. Balducci J, Rodis JF, Rosenberg S, Vintzileos AM, Spivey G, Vosseller C. Pregnancy outcome following first-trimester varicella infection. Obstet Gynecol 1992;79:5–6.
7. Pastuszak A, Levy M, Schick B, Zuber C, Feldkamp M, Gladstone J, et al. Outcome after maternal varicella infection in the first 20 weeks of pregnancy. N Engl J Med 1994;330:901–5.
8. Enders G, Miller E, Craddock-Watson J, Bolley I, Ridenhalg M. Consequences of varicella and herpes zoster in pregnancy: Prospective study of 1739 cases. Lancet 1994; 343:1547–51.
9. Jones KL, Johnson KA, Chambers CD. Offspring of women infected with varicella during pregnancy: A prospective study. Teratology 1994;49:29–32.[Medline]
10. Nilsson A, Ortqvist A. Severe varicella pneumonia in adults in Stockholm County 1980–1989. Scand J Infect Dis 1996;28:121–3.[Medline]
11. Miller E, Marshall R, Vurdien J. Epidemiology, outcome and control of varicella-zoster virus infection. Rev Med Microbiol 1993;4:222–30.
12. Gray GC, Palinkas LA, Kelly PW. Increasing incidence of varicella hospitalizations in the United States Army and Navy personnel. Pediatrics 1990;86:867–73.
13. Williams V, Gershon AA, Brunell P. Serologic response to varicella-zoster membrane antigens measured by indirect immunofluorescence. J Infect Dis 1974;130: 669–72.[Medline]
14. Weber GM, Pellechia JA. Varicella pneumonia. Study of prevalence in adult men. JAMA 1965;192:572–3.
15. Harger JH, Ernest JM, Thurnau GR, Moawad A, Momirova V, Landon MB, et al. Risk factors and outcome of varicella-zoster virus pneumonia in pregnant women. J Infect Dis 2002;185:422–7.[Medline]
16. Enders G. Management of varicella-zoster contact and infection in pregnancy using a standardized varicella-zoster ELISA test. Postgraduate Med J 1985;61(Suppl. 4):23–30.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  A. E. Dent and B. A. Baetz-Greenwalt Herpes Zoster in an Infant Clinical Pediatrics, September 1, 2007; 46(7): 646 - 649. [PDF]
|
|
|
|
 |
|
 Chickenpox, pregnancy and the newborn: a follow-up DTB, December 1, 2005; 43(12): 94 - 96. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Bohlke, K. Galil, L. A. Jackson, D. S. Schmid, P. Starkovich, V. N. Loparev, and J. F. Seward Postpartum Varicella Vaccination: Is the Vaccine Virus Excreted in Breast Milk? Obstet. Gynecol., November 1, 2003; 102(5): 970 - 977. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Scott 2002 Roy M. Pitkin Award Obstet. Gynecol., May 1, 2003; 101(5): 839 - 839. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
