- Email: [email protected]
Asymptomatic maternal shedding of herpes simplex virus at the onset of labor: Relationship to preterm labor
OBSTETRICS& GYNECO
Volume
87
Atxil
Number
1996
4
Asymptomatic Maternal Shedding of Herpes Simplex Virus at the Onset of Labor: Relationship to Preterm Labor ZANE A. BROWN, MD, JACQUELINE BENEDETTI, PkD, STACEY SELKE, MA, RHODA ASHLEY, PkD, D. HEATHER WATTS, MD, AND LAWRENCE COREY, MD Objective: To determine if fetal growth restriction and prematurity are observed with subclinical shedding of herpes simplex virus (HSV) at the onset of labor. Methods: Within 48 hours of delivery, cultures were taken from the cervix and external genitalia of 15,923 asymptomatic pregnant women without symptoms or signs of genital HSV infection; results were positive for HSV in 57. Each of these 57 women were compared with a control group composed of the three culture-negative women delivering immediately before and the three delivering immediately after each woman shedding HSV. Results: The median birth weight for infants born to the 57 women with asymptomatic shedding was 3050 g, compared with 3360 g among the 342 women without asymptomatic shedding, a statistically significant difference (P < .002). These differences were due to very low birth weight (LBW) among the five infants of women with subclinical viral shedding secondary to recently acquired primary genital herpes; these five infants had a median gestational age of 33
From
the
Departtnellts
Medicine,
Biostatistics, the Children’s Hospital Sqpwted by grant Allergy and I?+ctious Birth Defects Foundation.
VOL.
87, NO.
4, APRIL
of Obstetrics and Gynecology, Laboratory arid Medicine, University of Washingtorl, and Medical Center, Seattle, Washington. MO. A130732 from the National Institutes of Disease ad u ynrlt from the March of Dimes
weeks, compared with 37 weeks for the 14 infants of mothers with nonprimary, first-episode disease and 39 weeks for the 33 infants of women with reactivation disease, also a significant difference (P = .018). Conclusions: Asymptomatic genital shedding of HSV at the onset of labor because of subclinical primary genital HSV infection is associated with preterm delivery. Women who acquire genital HSV-2 before pregnancy and are shedding subclinically at the onset of labor experience no increase in adverse outcome. Thus, prevention of the prematurity and LBW associated with genital herpes means that acquisition of the infection in late pregnancy must be prevented. (O&et Gynecol 2996;87:483-8)
The relationship between genital herpes simplex virus (HSV) infections during pregnancy and fetal outcome is not understood completely. Many newborns with HSV infections are delivered prematurely and are of low birth weight (LBW).’ In one report of a selected referral population,2 symptomatic first-episode genital herpes during late pregnancy was associated with prematurity and fetal growth restriction. However, this association has not been demonstrated in pregnant women with a
1996 SSDl
OO29-7844/96/$15.00 0029-7844(95)00457-2
483
history of symptomatic recurrent genital herpess7 To explore whether prematurity and LBW are associated with asymptomatic shedding of HSV at the onset of labor because of either reactivation or first episodes of genital herpes, we initiated a case-control study of asymptomatic shedding of HSV at the onset of labor. This report describes the outcome of pregnancy in 57 women with asymptomatic genital HSV shedding at the onset of labor and in 342 women without asymptomatic shedding; all subjects were part of a 5.5-year prospective study of 15,923 women who delivered at two metropolitan Seattle hospitals.
gestational age by the obstetric staff agreed with the pediatric assessment within 2 weeks in all of the women with asymptomatic shedding and in 341 of the 342 women without asymptomatic shedding. There was only one woman in whom disagreement between the various measures of gestational age might have changed the categorization of gestation from before 37 to at least 37 or more completed weeks of pregnancy. Asymptomatic (subclinical) HSV shedding was defined as the isolation of HSV from the genitalia at a time when symptoms were not reported by the patient or signs of disease were not observed by the attending physician.
Materials and Methods HSV Isolation Study
Design
Subjects were identified from an ongoing prospective study of HSV shedding in early labor at two metropolitan Seattle hospitals, the University of Washington Medical Center, which has large indigent and referral populations, and a private community hospital serving primarily white, middle-class patients. The overall prevalence of HSV shedding and the frequency of neonatal herpes among these women have been reported in detail elsewhere.8 Since 1984, all women without symptoms or lesions suggesting genital HSV who were admitted to the labor units of the two study hospitals had HSV cultures collected from the labia majora, perianal and periclitoral areas, and, with a second swab, from the cervix and vaginal fornices. Between January 1984 and June 1989, 18,630 women delivered at the two study hospitals. Genital HSV cultures were collected within 48 hours of delivery on 15,923 (85.5%) of these women. All women undergoing induction of labor received genital HSV cultures. However, if the induction-to-delivery interval exceeded 48 hours, the results were not included in the analysis because they were unlikely to represent the status of the genital tract at delivery. Herpes simplex virus cultures were not collected from women with complications of pregnancy, including advanced labor, active vaginal bleeding, and serious maternal illness that precluded a speculum examination. Cultures were refused by the patient or overlooked by admitting labor room personnel in less than 1% of women who delivered. On admission to the labor units of both hospitals, the obstetric staff assessed gestational age based on available clinical data. Data for these assessments included last menstrual period, size at first obstetric examination, date of pregnancy test, symphysis-to-fundus measurements of uterine size, and fetal ultrasound measurements, when available. Gestational age estimates were also available on all newborns. The assessment of 484
Brown
et al
Asymptomatic
HSV
Shedding
and Serology
Specimens for HSV isolation were inoculated into tissue culture, examined for cytopathic effect, and identified as HSV-1 or HSV-2 by using type-specific mouse monoclonal antibodies.” Antibody status for HSV-1, HSV-2, or HSV-1 and HSV-2 was determined for all women by Western blot on sera that was obtained on admission to the labor unit.” Sera was obtained for 90% of the women in whom genital HSV cultures were obtained and was available for 52 of the 57 women in whom the genital tract culture was positive. We did not have sera for typing in five women. Among women in whom the HSV culture was positive, a second serum sample was obtained, generally at the 6-week postpartum examination. The criteria for defining the presence of antibodies specific to HSV-1 or HSV-2 in the initial serum sample as well as the occurrence of seroconversion to the heterologous type in the subsequent serum sample have been published previously.‘-” All paired serum samples were tested in parallel, and the serologic status of the mother was not known until several months after delivery. Primary genital herpes was defined as the lack of antibodies to HSV-1 and HSV-2 in sera obtained at the onset of labor in patients in whom either HSV-1 or HSV-2 had been recovered from the genital tract. Seropositivity for one of the virus types in a subsequent serum sample confirmed the diagnosis. Nonprimary, first-episode genital herpes was defined as either HSV-1 or HSV-2 antibodies in the initial serum sample with the genital isolation of the heterologous viral type (eg, HSV-2 with HSV-1 antibodies); antibodies to that viral type in a subsequent serum sample confirmed the diagnosis. Reactivation (recurrent) HSV was defined as that found in women in whom the serum HSV antibodies were of the same type as the virus recovered from the genitalia. If the HSV antibody profiles between their initial and subsequent sera remained unchanged and was the same as the virus type isolated from their genital tract, the diagnosis of reactivation (recurrent) HSV was confirmed. Obsfetrics
c5 Gynecology
Table
1. Demographic
Characteristics Asymptomatic
shedding
With (n = 57)
Maternal characteristic
23 12 44 27 31 27 17 7 15 50
Age (y) Years ed. White Welfare Married Smoker Hx herpes Hx prostitution Primigravida >3 prenatal visits
(16-35) (8-19) (77%) (47%) (55%) (47%) (30%) (12%) (26%) (88%)
Without (n = 342) 24 12 250 151 231 101 37 3 102 303
(14-46) (O-20) (73%) (44%) (68%‘) (27%) (11%) (1%) (30%) (89%)
ed. = education; Hx = history of. Data are presented as II (range) or )I (I,).
Case-Control
Methods
and Study End Points
We used a case-control method to evaluate the association of HSV isolation in the genital tract at delivery with obstetric outcome measures. Demographic and clinical data were abstracted on the cases and controls in a blinded fashion using standardized case report forms. Cases were considered to be all 57 women in whom an episode of asymptomatic shedding was detected from a culture obtained within 48 hours of delivery. For controls, we selected the three patients who delivered a singleton infant immediately before and the three immediately after the delivery of the woman with asymptomatic shedding. Although this selection procedure was not strictly random, there is no reason to suspect that it introduced any systematic bias, and it provided a representative sample of the general obstetric population at the two hospitals. This scheme does not provide a true matched comparison because there was no attempt to match according to patient characteristics thought to be related to the outcome of interest. As such, the statistical methods used did not treat the data as matched. We were unable to detect any significant differences in the data from the two hospitals and therefore report the combined data. Statistical comparisons between groups used nonparametric procedures for continuous data (KruskalWallis test) and the 2 or Fisher exact test for categorical data. Because both gestational age and birth weight were not normally distributed, we used nonparametric tests to evaluate the differences between cases and controls. However, we were also interested in assessing the relationship of potential risk factors to birth weight and gestational age. Because nonparametric techniques are more limited in their ability to include multiple concomitant variables, we also explored these relationships among variables using linear models. When appropriate nonparametric analogues were applicable, the results were similar to those obtained using techniques VOL.
87, NO.
4, APRIL
1996
requiring normality. However, we recognize that this does not guarantee that the linear model’s results are uniformly correct, and report these results with a recognition of their more exploratory nature.
Results Patient Population Demographic characteristics, including age, years of education, race, marital or cohabitation status, the number of prenatal visits, and the percentage receiving public assistance were similar in the 57 women with asymptomatic shedding and the 342 women without asymptomatic shedding (Table 1). The frequency of reported cigarette smoking was higher in the women with asymptomatic shedding and is discussed later. Of the 52 women with asymptomatic shedding of HSV at the onset of labor in whom sera was available, 19 had a first episode of genital herpes and 33 had reactivation of prior HSV-2 infection. Five of the 19 women with first episode disease had a primary genital HSV infection with asymptomatic seroconversion occurring during late pregnancy (four to HSV-2 and one to HSV-I). Fourteen women had a nonprimary genital HSV infection; all 14 were HSV-1 seropositive, had asymptomatic seroconversion to HSV-2 during late pregnancy, and were asymptomatically shedding HSV-2 in early labor. Thirty of the 33 women with reactivation disease were HSV-2 seropositive and three were HSV-1 seropositive. The 19 women with firstepisode disease tended to be younger (median 21 years), single (58%), and were more frequently in their
Table
2. Birth
Weight
and
Maternal characteristic
Birth
By presence of shedding Asymptomatic shedding (n = 57) No asymptomatic shedding (11 = 342) By type of disease 01 = 521 Primary 01 = 5) NPFE (,I = 14) Reactivation 01 = 33) By site of shedding (,I = 57) Labia (II = 33) Cervix (11 = 11) Cervix/labia 01 = 13) NPFE = nonprimary Data are presented
Brown
Gestational weight
V50 (500@4560)
Age (g)
P
Gestational age (wk)
P
38 (24-42)
.035
3360 (470-4990)
39 (22-44)
1YYl (1350-3160) 3290 (500-4360) 3147 (802-4320)
,083 33 (30-38) 37 (24-41) 39 (27-42)
,018
3189 (802-4560) 3160 (658-43453 2520 (500-44703
.218 39 (27-42) 38 (25-41) 36 (24 -40)
,006
first episode. as median (range).
et al
Asymptomatic
HSV
Shedding
485
first pregnancies (42%) than the 33 women with reactivation disease (median 24 years, 36 and 21%, respectively). However, the differences were not statistically significant.
Birth Weight and Gestational Infants
Age of HSV-Exposed
The median birth weight for infants born to the 57 women with asymptomatic shedding was 3050 g (range 500-4560), compared with 3360 g (range 470-4990) among infants of women without asymptomatic shedding (P < .002). Similarly, the median gestational age for the infants born to women with asymptomatic shedding was 38 weeks (range 24-42), compared with 39 weeks (range 22-44) for the women without asymptomatic shedding (P = ,035). This latter difference was due to the fact that 14 (25%) of the 57 women with asymptomatic shedding delivered before 37 weeks, compared with 42 (12.3%) of the 342 women without asymptomatic shedding (Table 2). Of the 52 women with asymptomatic shedding who could be classified serologically, the infants of the five women with a primary episode of genital herpes had a lower median gestational age (33 weeks) at birth than either the 14 infants of women with a nonprimary first episode (37 weeks) or the 33 infants of women with reactivation disease (39 weeks) (P = .018 by KruskalWallis test) (Table 2). Three (60%) of the five women with primary disease delivered infants before 37 weeks’ gestation compared with five (37.5%) of the 14 women with nonprimary first-episode disease, four (12.1%) of the 33 women with reactivation disease, and 42 (12.3%) of the 342 women without asymptomatic shedding. Two of the five women who we were unable to classify serologically also delivered infants before 37 weeks. There was a trend toward lower median birth weights among infants of mothers with asymptomatic shedding due to primary disease (1991 g) compared with infants of mothers with nonprimary, first-episode disease (3290 g), reactivation disease (3147 g), and the infants of mothers without asymptomatic shedding (3360 g). These differences were not significant (P = .083), possibly because of the small number of cases in the primarydisease category. Of the 57 women with asymptomatic genital HSV shedding, 33 shed from the labia only, 11 from the cervix only, and 13 from both sites. The median birth weights of their infants were 3189, 3160, and 2520 g, respectively (P = .218). The median gestational ages at delivery were 39, 38, and 36 weeks, respectively (P = .006) (Table 2). Three (9%) of the 33 women with asymptomatic shedding from the labia only delivered before 37 weeks’ gestation, compared with four (36%) of the 11 women with asymptomatic shedding from the 486
Brown
et al
Asymptmznfic
HSV
Shedding
cervix only and seven (54%) of the 13 women with asymptomatic shedding from both cervix and labia (P < .004). All five women with primary genital HSV infection shed from the cervix; three were shedding from both the cervix and labia. Because maternal smoking is a known risk factor for prematurity and LBW, we investigated the effect of smoking in this cohort. Qualitative information regarding smoking was available in 47 (83%) of the 57 women with asymptomatic shedding and in 246 (72%) of the 342 women without asymptomatic shedding. Among these 293 women in whom smoking status was known, the median birth weight of the 128 infants from women who smoked was 3160 g, compared with 3430 g for the 165 infants of women who did not smoke (P < .OOl). The corresponding gestational ages were 39 weeks among smokers and 40 weeks among nonsmokers (P = ,111. Because of the effect of smoking on birth weight, we assessed the relationship of asymptomatic shedding to birth outcome, adjusting for the effects of maternal smoking. Among women who smoked, the median birth weight for newborn infants of women with asymptomatic shedding was 2790 g, compared with 3220 g for women who did not shed asymptomatically (P = .002). Among nonsmokers, the birth weights were 3534 and 3420 g, respectively (a nonsignificant difference, P = .75). Among the smokers, the median gestational age at delivery was 38 weeks for women with asymptomatic shedding, compared with 40 weeks for those without asymptomatic shedding (P = .02), with 30% of the women with asymptomatic shedding delivering before 37 weeks compared with 14% of those without asymptomatic shedding. Among the nonsmokers, the median gestational age at delivery was 40 weeks for women with asymptomatic shedding compared with 39 weeks for those without asymptomatic shedding (P = .83). A regression analysis was performed to further assess the relationship between smoking, culture status, birth weight, and gestational age at delivery. Using this model, with birth weight as the outcome measure, the interaction between smoking and culture status at the onset of labor was not significant even though the individual tests for smoking and culture status were significant (P = .002 for smoking, P = .022 for culture status). The corresponding results using gestational age as the outcome variable were nonsignificant (P = ,098 for smoking, P = .065 for culture status). Thus, these findings suggest that the relationship between a positive HSV culture at delivery and birth outcome cannot be explained solely by the smoking status of the mother. These results should be viewed with caution because of the qualitative nature of the smoking data and the large number of women for whom smoking status was not available. Obsfetrics
0 Gynecology
We also investigated the relationship of other variables that might influence gestational age and birth weight, including maternal age, length of prenatal care, socioeconomic status, bacterial genitourinary tract infections, medical disorders (eg, hypertension and autoimmune disease), obstetric conditions (eg, placenta previa, abruption, incompetent cervix), and fetal malformation. Variables were first assessed using univariate linear regression. All variables indicating some relationship to the two outcomes of interest were then included in a multiple regression model, which was computed both with and without smoking included. The relationship between decreased birth weight and asymptomatic shedding versus no asymptomatic shedding of HSV remained significant even after we accounted for these other variables. The results were similar for gestational age, although in the reduced subset of patients with known smoking status, the decrease in gestational age in offspring of women with asymptomatic shedding compared with those without asymptomatic shedding was no longer statistically significant.
Discussion We used a case-control method to evaluate whether asymptomatic shedding of HSV at the onset of labor was associated with LBW and preterm delivery. Infants exposed to asymptomatic HSV shedding at the onset of labor were compared with infants delivered at the same hospital on the same day. Because the presence of asymptomatic shedding was not known until several days postpartum, it did not influence the obstetric management. The infants of women with asymptomatic HSV shedding at the onset of labor who recently acquired genital HSV infections were more likely to be premature than infants of mothers with asymptomatic shedding due to reactivation infections. Among these women with recently acquired first-episode infections, prematurity and LBW were most marked among women with primary genital HSV. These data extend our previous observations on symptomatic primary infections in that symptomatic and now asymptomatic primary infections increase the risks of preterm delivery.’ It also supports our observations that reactivation disease, whether symptomatic or subclinical, is not associated with an increased risk of preterm delivery.‘-’ Variables that might influence gestational age and birth weight, such as maternal age, length of prenatal care, socioeconomic status, and medical and obstetric complications of pregnancy, did not appear to alter the significance of the relationship between asymptomatic genital HSV shedding and birth outcome. Although there was an interaction between smoking, gestational age, birth weight, and subclinical HSV shedding, it appears that VOL.
87,
NO.
4, APRIL
1996
HSV shedding in association with late acquisition of primary genital herpes still affected pregnancy outcome. Several of the demographic and virologic characteristics of our study support our observations that primary HSV is the clinical form of genital HSV associated with pregnancy morbidity. Cervical shedding of HSV is strongly associated with first-episode infection, and women who shed HSV from the cervix with or without concomitant vulvar shedding had a higher frequency of preterm delivery and LBW. The frequency of cervical shedding among women with primary and firstepisode, nonprimary infection was similar in the current study group, suggesting that it is not just HSV cervicitis but other factors (eg, duration of cervical viral shedding, titer of virus shed at the cervix, the ability of HSV to invade the upper genital tract, and the host’s immune response to cervical HSV infection) that may initiate premature labor.“-‘7 One hypothesis is that primary HSV infection is more likely to ascend from the cervix and cause a viral chorioamnionitis and funisitis than is a nonprimary, first-episode infection.“-*” The viral chorioamnionitis could activate preterm labor either directly or as a result of secondary bacterial infection.‘s These data extend our previous studies on clinically symptomatic genital herpes to those with subclinical infections. Morbidity from the acquisition of genital HSV infection late in pregnancy appears to arise from both the clinically symptomatic and subclinical forms of the infection. Irrespective of the mechanism, our results are reassuring concerning the management of women with known reactivation HSV-2 infection, whether symptomatic or asymptomatic. They suggest that asymptomatic HSV reactivation, which occurs in the presence of known immunity to the infecting virus type, is not associated with adverse pregnancy outcome, such as preterm labor or LBW. The greatest morbidity associated with a history of long-standing genital HSV in the mother is the remarkably increased cesarean delivery rate resulting from the effort to prevent neonatal transmission. Maternal morbidity associated with the excess cesarean rates may eclipse that due to the infrequent neonatal transmission events in this group. Whether asymptomatic primary genital herpes infections (ie, asymptomatic primary seroconversion) during pregnancy, but remote from the onset of labor, restrict fetal growth or cause preterm labor similar to rates observed with symptomatic primary infections cannot be answered by this study design, which examined and cultured only women within 48 hours of the onset of labor. Also, our study was not designed to evaluate what percentage of LBW infants are associated with genital HSV infections. A prospective study design is required for such an approach. Although our data Brown et al
Asymptomatic
WSV
Shedding
487
provide some comfort and reassurance for those women who have already acquired genital herpes before pregnancy, they reaffirm the need to develop strategies to reduce the morbidity associated with the acquisition of genital herpes in late pregnancy. Both symptomatic and subclinical primary genital herpes are associated with frequent adverse pregnancy outcomes. Identification of such women can be achieved by the use of accurate type-specific serologic assays. If HSV serologically discordant couples could then be identified, they could be counseled about HSV seroconversion and the risks associated with oral or genital contact during pregnancy. The commercialization of newly developed type-specific HSV serologic assays is needed along with an analysis of their effectiveness and costbenefit in the detection of HSV serologically discordant couples during pregnancy. Moreover, only a minority of HSV-seronegative pregnant women appear to be in a sexual relationship that places them at risk of acquiring HSV in late pregnancy.21 However, our data indicate that approximately 75% of women with either symptomatic or subclinical primary infections during late pregnancy are delivered before 37 weeks or experience other adverse pregnancy outcomes, such as intrauterine growth restriction and neonatal HSV infection.2 These issues must be considered in the overall use of routine HSV serologic testing in early pregnancy.
Neonatal herpes simplex atic maternal infection 9
Western blot nodot enzyme
RJ, Corey
L, Arvin
A, Lakeman
FD, Sumaya
11
12
S, et al. Effects on infants of a first episode of genital herpes during pregnancy. N Engl J Med 1987;317:1246-51. Vontver LA, Hickok DE, Brown Z, Reid L, Corey L. Recurrent genital
herpes
simplex
virus
infection
come and frequency of asymptomatic Gynecol 1982;143:75-84. Harger JH, Pazin CJ, Armstrong ]A,
in pregnancy:
Breinig
Am
MC,
Ho
teristics and management of pregnancy in women herpes simplex virus infection. Am J Obstet Gynecol 91. Catalano simplex factors. Boehm genital
PM, Meritt AO, Mead PB. incidence virus at the time of delivery in women Am J Obstet Gynecol 1991;164:1303-6. FH, Estes W, Wright PF, Growdon herpes simplex virus infection occurring
Am J Obstet Prober CG,
Gynecol Sullender
1981;141:735-40. WM, Yasukawa
Infant
recurrences.
LL,
et al
1987;18:159-68. Bernstein DI, Lovett episode nonprimary
L. Comparison
MA, Bryson genital herpes.
G-specific to herpes
13
Microbial 1988;26:662-7. analysis for interpretblot. J Virol Methods
YJ. Serologic analysis of firstPresence of type 2 antibody in
acute serum samples. Am J Med 1984;77:1055-60. Corey L, Adams HG, Brown ZA, Holmes KK.
14
15
Prober Bryson genital
17
Genital course,
herpes and com-
CC, Corey L, Brown ZA, Hensleigh PA, Frenkel YJ, et al. The management of pregnancies complicated infections with herpes simplex virus. Clin Infect
1992;15:1031-8. Nahmias AJ, Josey WE, Naib ZM, Wheeler JH. Perinatal risk associated
Freeman MG, with maternal
Fernandez R, genital herpes
infection.
Robb JA, Benirschke herpes simplex virus
I. Spontaneous
abortion.
Hum
K, Mannino F, Voland J. Intrauterine infection. II. Latent neonatal infection.
Pathol 1986;17:1210-17. Hain J, Doshi N, Harger JH. Ascending transcervical simplex infection with intact fetal membranes. Obstet 1980;56:106-8. Mitchell MD,
19
RA, Dudley DJ. Immunologic aspects of preterm labor. I’erinatol 1991;75:210-24. Schwartz DA, Caldwell E. Herpes simplex virus infection
20
placenta. Navarro
DW,
Lundin-Schiller
Arch Pathol Lab Med C, Blanc WA. Chronic
967-73. Kulhanjian Weylman primary nancy.
Branch
JA, Soroush V, Au LE, et al. Identification
S, Romero
1991;115:1141-4. viral funisitis.
Pathol latent Hum herpes Gynecol
18
21
LM, by Dis
simplex virus infection. Am J Obstet Gynecol 1971;110:825-37. Robb JA, Benirschke K, Barmeyer R. Intrauterine latent herpes simplex virus 1986;17:1196-209.
16
of
immusimplex
RJ, Daynes
J Pediatr
Semin of the 1977;91:
DS, Bronzan RN, Yasukawa LL, of women at unsuspected risk of
infection with herpes simplex N Engl J Med 1992;326:916-20.
virus
type
2 during
preg-
out-
M. Charac-
with genital 1983;145:784-
of genital herpes with known risk
Address
reprint
requests
to:
Zane A. Brown, MD Department of Obstetrics and Gynecology RH-20 University of Washington Seattle, WA 98295
JF. Management of during pregnancy. Au
DS, Yeager
AS,
recurrent genital herpes simplex virus infections. N Engl J Med 1987;316:240-4. Brown ZA, Benedetti J, Ashley R, Burchett S, Selke S, Berry S, et al.
Brown
A, Corey
glycoprotein antibodies
J Obstet
Arvin AM. Low risk of herpes simplex virus infection in neonates exposed to the virus at the time of vaginal delivery to mothers with
488
and detecting
simplex virus infections: Clinical manifestations, plications. Ann Intern Med 1983;98:958-72.
CV, Wright
PF, et al. Changing presentation of herpes simplex virus infection in neonates. J Infect Dis 1988;158:109-16, Brown ZA, Vontver L-4, Benedetti J, Critchlow CW, Sells CJ, Berry
J, Lee F, Nahmias
(immunoblot) assay for
virus types 1 and 2 in human sera. J Clin 10 Ashley RL, Militoni J. Use of densitometric ing HSV serologies based on Western
References Whitley
1991;324:1247-52. Ashley R, Militoni
virus infection in relation to asymptomat the time of labor. N Engl J Med
Asymptomatic HSV Shedding
Received May 31, 1995. Receizledirl wvised forrlr Octuber Accepted Novendw 20, 1995.
Copyright Gynecologists.
0
1996
by
Published
The
American
by Elsevier
30, 1995
College Science
of
Obstetricians
and
Inc.
Obstetrics G Gynecology
Volume
87
Atxil
Number
1996
4
Asymptomatic Maternal Shedding of Herpes Simplex Virus at the Onset of Labor: Relationship to Preterm Labor ZANE A. BROWN, MD, JACQUELINE BENEDETTI, PkD, STACEY SELKE, MA, RHODA ASHLEY, PkD, D. HEATHER WATTS, MD, AND LAWRENCE COREY, MD Objective: To determine if fetal growth restriction and prematurity are observed with subclinical shedding of herpes simplex virus (HSV) at the onset of labor. Methods: Within 48 hours of delivery, cultures were taken from the cervix and external genitalia of 15,923 asymptomatic pregnant women without symptoms or signs of genital HSV infection; results were positive for HSV in 57. Each of these 57 women were compared with a control group composed of the three culture-negative women delivering immediately before and the three delivering immediately after each woman shedding HSV. Results: The median birth weight for infants born to the 57 women with asymptomatic shedding was 3050 g, compared with 3360 g among the 342 women without asymptomatic shedding, a statistically significant difference (P < .002). These differences were due to very low birth weight (LBW) among the five infants of women with subclinical viral shedding secondary to recently acquired primary genital herpes; these five infants had a median gestational age of 33
From
the
Departtnellts
Medicine,
Biostatistics, the Children’s Hospital Sqpwted by grant Allergy and I?+ctious Birth Defects Foundation.
VOL.
87, NO.
4, APRIL
of Obstetrics and Gynecology, Laboratory arid Medicine, University of Washingtorl, and Medical Center, Seattle, Washington. MO. A130732 from the National Institutes of Disease ad u ynrlt from the March of Dimes
weeks, compared with 37 weeks for the 14 infants of mothers with nonprimary, first-episode disease and 39 weeks for the 33 infants of women with reactivation disease, also a significant difference (P = .018). Conclusions: Asymptomatic genital shedding of HSV at the onset of labor because of subclinical primary genital HSV infection is associated with preterm delivery. Women who acquire genital HSV-2 before pregnancy and are shedding subclinically at the onset of labor experience no increase in adverse outcome. Thus, prevention of the prematurity and LBW associated with genital herpes means that acquisition of the infection in late pregnancy must be prevented. (O&et Gynecol 2996;87:483-8)
The relationship between genital herpes simplex virus (HSV) infections during pregnancy and fetal outcome is not understood completely. Many newborns with HSV infections are delivered prematurely and are of low birth weight (LBW).’ In one report of a selected referral population,2 symptomatic first-episode genital herpes during late pregnancy was associated with prematurity and fetal growth restriction. However, this association has not been demonstrated in pregnant women with a
1996 SSDl
OO29-7844/96/$15.00 0029-7844(95)00457-2
483
history of symptomatic recurrent genital herpess7 To explore whether prematurity and LBW are associated with asymptomatic shedding of HSV at the onset of labor because of either reactivation or first episodes of genital herpes, we initiated a case-control study of asymptomatic shedding of HSV at the onset of labor. This report describes the outcome of pregnancy in 57 women with asymptomatic genital HSV shedding at the onset of labor and in 342 women without asymptomatic shedding; all subjects were part of a 5.5-year prospective study of 15,923 women who delivered at two metropolitan Seattle hospitals.
gestational age by the obstetric staff agreed with the pediatric assessment within 2 weeks in all of the women with asymptomatic shedding and in 341 of the 342 women without asymptomatic shedding. There was only one woman in whom disagreement between the various measures of gestational age might have changed the categorization of gestation from before 37 to at least 37 or more completed weeks of pregnancy. Asymptomatic (subclinical) HSV shedding was defined as the isolation of HSV from the genitalia at a time when symptoms were not reported by the patient or signs of disease were not observed by the attending physician.
Materials and Methods HSV Isolation Study
Design
Subjects were identified from an ongoing prospective study of HSV shedding in early labor at two metropolitan Seattle hospitals, the University of Washington Medical Center, which has large indigent and referral populations, and a private community hospital serving primarily white, middle-class patients. The overall prevalence of HSV shedding and the frequency of neonatal herpes among these women have been reported in detail elsewhere.8 Since 1984, all women without symptoms or lesions suggesting genital HSV who were admitted to the labor units of the two study hospitals had HSV cultures collected from the labia majora, perianal and periclitoral areas, and, with a second swab, from the cervix and vaginal fornices. Between January 1984 and June 1989, 18,630 women delivered at the two study hospitals. Genital HSV cultures were collected within 48 hours of delivery on 15,923 (85.5%) of these women. All women undergoing induction of labor received genital HSV cultures. However, if the induction-to-delivery interval exceeded 48 hours, the results were not included in the analysis because they were unlikely to represent the status of the genital tract at delivery. Herpes simplex virus cultures were not collected from women with complications of pregnancy, including advanced labor, active vaginal bleeding, and serious maternal illness that precluded a speculum examination. Cultures were refused by the patient or overlooked by admitting labor room personnel in less than 1% of women who delivered. On admission to the labor units of both hospitals, the obstetric staff assessed gestational age based on available clinical data. Data for these assessments included last menstrual period, size at first obstetric examination, date of pregnancy test, symphysis-to-fundus measurements of uterine size, and fetal ultrasound measurements, when available. Gestational age estimates were also available on all newborns. The assessment of 484
Brown
et al
Asymptomatic
HSV
Shedding
and Serology
Specimens for HSV isolation were inoculated into tissue culture, examined for cytopathic effect, and identified as HSV-1 or HSV-2 by using type-specific mouse monoclonal antibodies.” Antibody status for HSV-1, HSV-2, or HSV-1 and HSV-2 was determined for all women by Western blot on sera that was obtained on admission to the labor unit.” Sera was obtained for 90% of the women in whom genital HSV cultures were obtained and was available for 52 of the 57 women in whom the genital tract culture was positive. We did not have sera for typing in five women. Among women in whom the HSV culture was positive, a second serum sample was obtained, generally at the 6-week postpartum examination. The criteria for defining the presence of antibodies specific to HSV-1 or HSV-2 in the initial serum sample as well as the occurrence of seroconversion to the heterologous type in the subsequent serum sample have been published previously.‘-” All paired serum samples were tested in parallel, and the serologic status of the mother was not known until several months after delivery. Primary genital herpes was defined as the lack of antibodies to HSV-1 and HSV-2 in sera obtained at the onset of labor in patients in whom either HSV-1 or HSV-2 had been recovered from the genital tract. Seropositivity for one of the virus types in a subsequent serum sample confirmed the diagnosis. Nonprimary, first-episode genital herpes was defined as either HSV-1 or HSV-2 antibodies in the initial serum sample with the genital isolation of the heterologous viral type (eg, HSV-2 with HSV-1 antibodies); antibodies to that viral type in a subsequent serum sample confirmed the diagnosis. Reactivation (recurrent) HSV was defined as that found in women in whom the serum HSV antibodies were of the same type as the virus recovered from the genitalia. If the HSV antibody profiles between their initial and subsequent sera remained unchanged and was the same as the virus type isolated from their genital tract, the diagnosis of reactivation (recurrent) HSV was confirmed. Obsfetrics
c5 Gynecology
Table
1. Demographic
Characteristics Asymptomatic
shedding
With (n = 57)
Maternal characteristic
23 12 44 27 31 27 17 7 15 50
Age (y) Years ed. White Welfare Married Smoker Hx herpes Hx prostitution Primigravida >3 prenatal visits
(16-35) (8-19) (77%) (47%) (55%) (47%) (30%) (12%) (26%) (88%)
Without (n = 342) 24 12 250 151 231 101 37 3 102 303
(14-46) (O-20) (73%) (44%) (68%‘) (27%) (11%) (1%) (30%) (89%)
ed. = education; Hx = history of. Data are presented as II (range) or )I (I,).
Case-Control
Methods
and Study End Points
We used a case-control method to evaluate the association of HSV isolation in the genital tract at delivery with obstetric outcome measures. Demographic and clinical data were abstracted on the cases and controls in a blinded fashion using standardized case report forms. Cases were considered to be all 57 women in whom an episode of asymptomatic shedding was detected from a culture obtained within 48 hours of delivery. For controls, we selected the three patients who delivered a singleton infant immediately before and the three immediately after the delivery of the woman with asymptomatic shedding. Although this selection procedure was not strictly random, there is no reason to suspect that it introduced any systematic bias, and it provided a representative sample of the general obstetric population at the two hospitals. This scheme does not provide a true matched comparison because there was no attempt to match according to patient characteristics thought to be related to the outcome of interest. As such, the statistical methods used did not treat the data as matched. We were unable to detect any significant differences in the data from the two hospitals and therefore report the combined data. Statistical comparisons between groups used nonparametric procedures for continuous data (KruskalWallis test) and the 2 or Fisher exact test for categorical data. Because both gestational age and birth weight were not normally distributed, we used nonparametric tests to evaluate the differences between cases and controls. However, we were also interested in assessing the relationship of potential risk factors to birth weight and gestational age. Because nonparametric techniques are more limited in their ability to include multiple concomitant variables, we also explored these relationships among variables using linear models. When appropriate nonparametric analogues were applicable, the results were similar to those obtained using techniques VOL.
87, NO.
4, APRIL
1996
requiring normality. However, we recognize that this does not guarantee that the linear model’s results are uniformly correct, and report these results with a recognition of their more exploratory nature.
Results Patient Population Demographic characteristics, including age, years of education, race, marital or cohabitation status, the number of prenatal visits, and the percentage receiving public assistance were similar in the 57 women with asymptomatic shedding and the 342 women without asymptomatic shedding (Table 1). The frequency of reported cigarette smoking was higher in the women with asymptomatic shedding and is discussed later. Of the 52 women with asymptomatic shedding of HSV at the onset of labor in whom sera was available, 19 had a first episode of genital herpes and 33 had reactivation of prior HSV-2 infection. Five of the 19 women with first episode disease had a primary genital HSV infection with asymptomatic seroconversion occurring during late pregnancy (four to HSV-2 and one to HSV-I). Fourteen women had a nonprimary genital HSV infection; all 14 were HSV-1 seropositive, had asymptomatic seroconversion to HSV-2 during late pregnancy, and were asymptomatically shedding HSV-2 in early labor. Thirty of the 33 women with reactivation disease were HSV-2 seropositive and three were HSV-1 seropositive. The 19 women with firstepisode disease tended to be younger (median 21 years), single (58%), and were more frequently in their
Table
2. Birth
Weight
and
Maternal characteristic
Birth
By presence of shedding Asymptomatic shedding (n = 57) No asymptomatic shedding (11 = 342) By type of disease 01 = 521 Primary 01 = 5) NPFE (,I = 14) Reactivation 01 = 33) By site of shedding (,I = 57) Labia (II = 33) Cervix (11 = 11) Cervix/labia 01 = 13) NPFE = nonprimary Data are presented
Brown
Gestational weight
V50 (500@4560)
Age (g)
P
Gestational age (wk)
P
38 (24-42)
.035
3360 (470-4990)
39 (22-44)
1YYl (1350-3160) 3290 (500-4360) 3147 (802-4320)
,083 33 (30-38) 37 (24-41) 39 (27-42)
,018
3189 (802-4560) 3160 (658-43453 2520 (500-44703
.218 39 (27-42) 38 (25-41) 36 (24 -40)
,006
first episode. as median (range).
et al
Asymptomatic
HSV
Shedding
485
first pregnancies (42%) than the 33 women with reactivation disease (median 24 years, 36 and 21%, respectively). However, the differences were not statistically significant.
Birth Weight and Gestational Infants
Age of HSV-Exposed
The median birth weight for infants born to the 57 women with asymptomatic shedding was 3050 g (range 500-4560), compared with 3360 g (range 470-4990) among infants of women without asymptomatic shedding (P < .002). Similarly, the median gestational age for the infants born to women with asymptomatic shedding was 38 weeks (range 24-42), compared with 39 weeks (range 22-44) for the women without asymptomatic shedding (P = ,035). This latter difference was due to the fact that 14 (25%) of the 57 women with asymptomatic shedding delivered before 37 weeks, compared with 42 (12.3%) of the 342 women without asymptomatic shedding (Table 2). Of the 52 women with asymptomatic shedding who could be classified serologically, the infants of the five women with a primary episode of genital herpes had a lower median gestational age (33 weeks) at birth than either the 14 infants of women with a nonprimary first episode (37 weeks) or the 33 infants of women with reactivation disease (39 weeks) (P = .018 by KruskalWallis test) (Table 2). Three (60%) of the five women with primary disease delivered infants before 37 weeks’ gestation compared with five (37.5%) of the 14 women with nonprimary first-episode disease, four (12.1%) of the 33 women with reactivation disease, and 42 (12.3%) of the 342 women without asymptomatic shedding. Two of the five women who we were unable to classify serologically also delivered infants before 37 weeks. There was a trend toward lower median birth weights among infants of mothers with asymptomatic shedding due to primary disease (1991 g) compared with infants of mothers with nonprimary, first-episode disease (3290 g), reactivation disease (3147 g), and the infants of mothers without asymptomatic shedding (3360 g). These differences were not significant (P = .083), possibly because of the small number of cases in the primarydisease category. Of the 57 women with asymptomatic genital HSV shedding, 33 shed from the labia only, 11 from the cervix only, and 13 from both sites. The median birth weights of their infants were 3189, 3160, and 2520 g, respectively (P = .218). The median gestational ages at delivery were 39, 38, and 36 weeks, respectively (P = .006) (Table 2). Three (9%) of the 33 women with asymptomatic shedding from the labia only delivered before 37 weeks’ gestation, compared with four (36%) of the 11 women with asymptomatic shedding from the 486
Brown
et al
Asymptmznfic
HSV
Shedding
cervix only and seven (54%) of the 13 women with asymptomatic shedding from both cervix and labia (P < .004). All five women with primary genital HSV infection shed from the cervix; three were shedding from both the cervix and labia. Because maternal smoking is a known risk factor for prematurity and LBW, we investigated the effect of smoking in this cohort. Qualitative information regarding smoking was available in 47 (83%) of the 57 women with asymptomatic shedding and in 246 (72%) of the 342 women without asymptomatic shedding. Among these 293 women in whom smoking status was known, the median birth weight of the 128 infants from women who smoked was 3160 g, compared with 3430 g for the 165 infants of women who did not smoke (P < .OOl). The corresponding gestational ages were 39 weeks among smokers and 40 weeks among nonsmokers (P = ,111. Because of the effect of smoking on birth weight, we assessed the relationship of asymptomatic shedding to birth outcome, adjusting for the effects of maternal smoking. Among women who smoked, the median birth weight for newborn infants of women with asymptomatic shedding was 2790 g, compared with 3220 g for women who did not shed asymptomatically (P = .002). Among nonsmokers, the birth weights were 3534 and 3420 g, respectively (a nonsignificant difference, P = .75). Among the smokers, the median gestational age at delivery was 38 weeks for women with asymptomatic shedding, compared with 40 weeks for those without asymptomatic shedding (P = .02), with 30% of the women with asymptomatic shedding delivering before 37 weeks compared with 14% of those without asymptomatic shedding. Among the nonsmokers, the median gestational age at delivery was 40 weeks for women with asymptomatic shedding compared with 39 weeks for those without asymptomatic shedding (P = .83). A regression analysis was performed to further assess the relationship between smoking, culture status, birth weight, and gestational age at delivery. Using this model, with birth weight as the outcome measure, the interaction between smoking and culture status at the onset of labor was not significant even though the individual tests for smoking and culture status were significant (P = .002 for smoking, P = .022 for culture status). The corresponding results using gestational age as the outcome variable were nonsignificant (P = ,098 for smoking, P = .065 for culture status). Thus, these findings suggest that the relationship between a positive HSV culture at delivery and birth outcome cannot be explained solely by the smoking status of the mother. These results should be viewed with caution because of the qualitative nature of the smoking data and the large number of women for whom smoking status was not available. Obsfetrics
0 Gynecology
We also investigated the relationship of other variables that might influence gestational age and birth weight, including maternal age, length of prenatal care, socioeconomic status, bacterial genitourinary tract infections, medical disorders (eg, hypertension and autoimmune disease), obstetric conditions (eg, placenta previa, abruption, incompetent cervix), and fetal malformation. Variables were first assessed using univariate linear regression. All variables indicating some relationship to the two outcomes of interest were then included in a multiple regression model, which was computed both with and without smoking included. The relationship between decreased birth weight and asymptomatic shedding versus no asymptomatic shedding of HSV remained significant even after we accounted for these other variables. The results were similar for gestational age, although in the reduced subset of patients with known smoking status, the decrease in gestational age in offspring of women with asymptomatic shedding compared with those without asymptomatic shedding was no longer statistically significant.
Discussion We used a case-control method to evaluate whether asymptomatic shedding of HSV at the onset of labor was associated with LBW and preterm delivery. Infants exposed to asymptomatic HSV shedding at the onset of labor were compared with infants delivered at the same hospital on the same day. Because the presence of asymptomatic shedding was not known until several days postpartum, it did not influence the obstetric management. The infants of women with asymptomatic HSV shedding at the onset of labor who recently acquired genital HSV infections were more likely to be premature than infants of mothers with asymptomatic shedding due to reactivation infections. Among these women with recently acquired first-episode infections, prematurity and LBW were most marked among women with primary genital HSV. These data extend our previous observations on symptomatic primary infections in that symptomatic and now asymptomatic primary infections increase the risks of preterm delivery.’ It also supports our observations that reactivation disease, whether symptomatic or subclinical, is not associated with an increased risk of preterm delivery.‘-’ Variables that might influence gestational age and birth weight, such as maternal age, length of prenatal care, socioeconomic status, and medical and obstetric complications of pregnancy, did not appear to alter the significance of the relationship between asymptomatic genital HSV shedding and birth outcome. Although there was an interaction between smoking, gestational age, birth weight, and subclinical HSV shedding, it appears that VOL.
87,
NO.
4, APRIL
1996
HSV shedding in association with late acquisition of primary genital herpes still affected pregnancy outcome. Several of the demographic and virologic characteristics of our study support our observations that primary HSV is the clinical form of genital HSV associated with pregnancy morbidity. Cervical shedding of HSV is strongly associated with first-episode infection, and women who shed HSV from the cervix with or without concomitant vulvar shedding had a higher frequency of preterm delivery and LBW. The frequency of cervical shedding among women with primary and firstepisode, nonprimary infection was similar in the current study group, suggesting that it is not just HSV cervicitis but other factors (eg, duration of cervical viral shedding, titer of virus shed at the cervix, the ability of HSV to invade the upper genital tract, and the host’s immune response to cervical HSV infection) that may initiate premature labor.“-‘7 One hypothesis is that primary HSV infection is more likely to ascend from the cervix and cause a viral chorioamnionitis and funisitis than is a nonprimary, first-episode infection.“-*” The viral chorioamnionitis could activate preterm labor either directly or as a result of secondary bacterial infection.‘s These data extend our previous studies on clinically symptomatic genital herpes to those with subclinical infections. Morbidity from the acquisition of genital HSV infection late in pregnancy appears to arise from both the clinically symptomatic and subclinical forms of the infection. Irrespective of the mechanism, our results are reassuring concerning the management of women with known reactivation HSV-2 infection, whether symptomatic or asymptomatic. They suggest that asymptomatic HSV reactivation, which occurs in the presence of known immunity to the infecting virus type, is not associated with adverse pregnancy outcome, such as preterm labor or LBW. The greatest morbidity associated with a history of long-standing genital HSV in the mother is the remarkably increased cesarean delivery rate resulting from the effort to prevent neonatal transmission. Maternal morbidity associated with the excess cesarean rates may eclipse that due to the infrequent neonatal transmission events in this group. Whether asymptomatic primary genital herpes infections (ie, asymptomatic primary seroconversion) during pregnancy, but remote from the onset of labor, restrict fetal growth or cause preterm labor similar to rates observed with symptomatic primary infections cannot be answered by this study design, which examined and cultured only women within 48 hours of the onset of labor. Also, our study was not designed to evaluate what percentage of LBW infants are associated with genital HSV infections. A prospective study design is required for such an approach. Although our data Brown et al
Asymptomatic
WSV
Shedding
487
provide some comfort and reassurance for those women who have already acquired genital herpes before pregnancy, they reaffirm the need to develop strategies to reduce the morbidity associated with the acquisition of genital herpes in late pregnancy. Both symptomatic and subclinical primary genital herpes are associated with frequent adverse pregnancy outcomes. Identification of such women can be achieved by the use of accurate type-specific serologic assays. If HSV serologically discordant couples could then be identified, they could be counseled about HSV seroconversion and the risks associated with oral or genital contact during pregnancy. The commercialization of newly developed type-specific HSV serologic assays is needed along with an analysis of their effectiveness and costbenefit in the detection of HSV serologically discordant couples during pregnancy. Moreover, only a minority of HSV-seronegative pregnant women appear to be in a sexual relationship that places them at risk of acquiring HSV in late pregnancy.21 However, our data indicate that approximately 75% of women with either symptomatic or subclinical primary infections during late pregnancy are delivered before 37 weeks or experience other adverse pregnancy outcomes, such as intrauterine growth restriction and neonatal HSV infection.2 These issues must be considered in the overall use of routine HSV serologic testing in early pregnancy.
Neonatal herpes simplex atic maternal infection 9
Western blot nodot enzyme
RJ, Corey
L, Arvin
A, Lakeman
FD, Sumaya
11
12
S, et al. Effects on infants of a first episode of genital herpes during pregnancy. N Engl J Med 1987;317:1246-51. Vontver LA, Hickok DE, Brown Z, Reid L, Corey L. Recurrent genital
herpes
simplex
virus
infection
come and frequency of asymptomatic Gynecol 1982;143:75-84. Harger JH, Pazin CJ, Armstrong ]A,
in pregnancy:
Breinig
Am
MC,
Ho
teristics and management of pregnancy in women herpes simplex virus infection. Am J Obstet Gynecol 91. Catalano simplex factors. Boehm genital
PM, Meritt AO, Mead PB. incidence virus at the time of delivery in women Am J Obstet Gynecol 1991;164:1303-6. FH, Estes W, Wright PF, Growdon herpes simplex virus infection occurring
Am J Obstet Prober CG,
Gynecol Sullender
1981;141:735-40. WM, Yasukawa
Infant
recurrences.
LL,
et al
1987;18:159-68. Bernstein DI, Lovett episode nonprimary
L. Comparison
MA, Bryson genital herpes.
G-specific to herpes
13
Microbial 1988;26:662-7. analysis for interpretblot. J Virol Methods
YJ. Serologic analysis of firstPresence of type 2 antibody in
acute serum samples. Am J Med 1984;77:1055-60. Corey L, Adams HG, Brown ZA, Holmes KK.
14
15
Prober Bryson genital
17
Genital course,
herpes and com-
CC, Corey L, Brown ZA, Hensleigh PA, Frenkel YJ, et al. The management of pregnancies complicated infections with herpes simplex virus. Clin Infect
1992;15:1031-8. Nahmias AJ, Josey WE, Naib ZM, Wheeler JH. Perinatal risk associated
Freeman MG, with maternal
Fernandez R, genital herpes
infection.
Robb JA, Benirschke herpes simplex virus
I. Spontaneous
abortion.
Hum
K, Mannino F, Voland J. Intrauterine infection. II. Latent neonatal infection.
Pathol 1986;17:1210-17. Hain J, Doshi N, Harger JH. Ascending transcervical simplex infection with intact fetal membranes. Obstet 1980;56:106-8. Mitchell MD,
19
RA, Dudley DJ. Immunologic aspects of preterm labor. I’erinatol 1991;75:210-24. Schwartz DA, Caldwell E. Herpes simplex virus infection
20
placenta. Navarro
DW,
Lundin-Schiller
Arch Pathol Lab Med C, Blanc WA. Chronic
967-73. Kulhanjian Weylman primary nancy.
Branch
JA, Soroush V, Au LE, et al. Identification
S, Romero
1991;115:1141-4. viral funisitis.
Pathol latent Hum herpes Gynecol
18
21
LM, by Dis
simplex virus infection. Am J Obstet Gynecol 1971;110:825-37. Robb JA, Benirschke K, Barmeyer R. Intrauterine latent herpes simplex virus 1986;17:1196-209.
16
of
immusimplex
RJ, Daynes
J Pediatr
Semin of the 1977;91:
DS, Bronzan RN, Yasukawa LL, of women at unsuspected risk of
infection with herpes simplex N Engl J Med 1992;326:916-20.
virus
type
2 during
preg-
out-
M. Charac-
with genital 1983;145:784-
of genital herpes with known risk
Address
reprint
requests
to:
Zane A. Brown, MD Department of Obstetrics and Gynecology RH-20 University of Washington Seattle, WA 98295
JF. Management of during pregnancy. Au
DS, Yeager
AS,
recurrent genital herpes simplex virus infections. N Engl J Med 1987;316:240-4. Brown ZA, Benedetti J, Ashley R, Burchett S, Selke S, Berry S, et al.
Brown
A, Corey
glycoprotein antibodies
J Obstet
Arvin AM. Low risk of herpes simplex virus infection in neonates exposed to the virus at the time of vaginal delivery to mothers with
488
and detecting
simplex virus infections: Clinical manifestations, plications. Ann Intern Med 1983;98:958-72.
CV, Wright
PF, et al. Changing presentation of herpes simplex virus infection in neonates. J Infect Dis 1988;158:109-16, Brown ZA, Vontver L-4, Benedetti J, Critchlow CW, Sells CJ, Berry
J, Lee F, Nahmias
(immunoblot) assay for
virus types 1 and 2 in human sera. J Clin 10 Ashley RL, Militoni J. Use of densitometric ing HSV serologies based on Western
References Whitley
1991;324:1247-52. Ashley R, Militoni
virus infection in relation to asymptomat the time of labor. N Engl J Med
Asymptomatic HSV Shedding
Received May 31, 1995. Receizledirl wvised forrlr Octuber Accepted Novendw 20, 1995.
Copyright Gynecologists.
0
1996
by
Published
The
American
by Elsevier
30, 1995
College Science
of
Obstetricians
and
Inc.
Obstetrics G Gynecology