BJOG: an International Journal of Obstetrics and Gynaecology

September 2002, Vol. 109, pp. 1030– 1035

Herpes simplex virus type-2 infection in pregnancy: no risk of

fetal death: results from a nested case–control study
within 35,940 women
Anne Eskilda,b,*, Stig Jeanssonc, Babill Stray-Pedersend, Pål A. Jenume
Objective The aim of this study was to assess the association of fetal death with herpes simplex virus type-2
(HSV-2) antibody status during pregnancy: 1. presence of antibodies in first trimester; 2. appearance of
antibodies (incident infection); 3. increase in antibody titre; and 4. loss of antibodies.
Design Prospective study.
Population The source population was a cohort of 35,940 pregnant women in Norway.
Methods Nested case –control study within the cohort. Cases were all women in the study population who
experienced a fetal death after the 16th weeks of gestation (n ¼ 281), and controls were 961 randomly
selected women with a live born child.
Main outcome measures HSV-2 antibody status.
Results Twenty-nine percent (82/281) of women with a fetal death and 27% (256/961) of the controls had
of HSV-2 antibodies present in the first trimester (odds ratio 1.1, 95% CI 0.8– 1.5). HSV-2 antibodies
appeared in 2% (3/136) of initially seronegative cases and 3% (16/623) of the controls during pregnancy
(odds ratio 0.9, 95% CI 0.2– 3.0). An increase in HSV-2 antibodies occurred in 4% (2/55) of initially
seropositive cases and 7% (16/231) of the controls (odds ratio 0.5, 95% CI 0.1 –2.3). Loss of HSV-2
antibodies in initially seropositive women was not associated with fetal death, 42% (23/55) of the
cases and 45% (104/231) of the controls seroreverted (odds ratio 0.8, 95% CI 0.5 – 1.6). Differences in
follow up time, age and parity were controlled and did not influence the comparisons between cases and
controls.
Conclusion This study provides no evidence of an association between HSV-2 infection during pregnancy and
fetal death.

INTRODUCTION few case reports describing congenital fetal infection have

been reported6 – 11.
There is a risk of mother-to-child transmission of neo- To our knowledge, only one large study of HSV infec-
natal herpes simplex virus type-2 (HSV-2) infection if the tion during pregnancy and pregnancy outcomes has been
mother has an active infection during labour1 – 5. Whether reported5. In that prospective study, including more than
maternal HSV-2 infection during pregnancy influences fetal 7000 women, no risk estimate of fetal death according to
development is, however, largely unknown. It has been HSV-2 seroconversion was presented, probably due to an
suggested that a primary HSV-2 infection during pregnancy insufficient number of women included.
may be associated with fetal death and miscarriage and a HSV-2 seroconversion in pregnancy and fetal death are
both rare events. Thus, only studies including a large
number of pregnant women can give reliable estimates of
association. The aim of this nested case control study
a
Section of Epidemiology, National Institute of Public within a cohort of 35,940 women was to study association
Health, Oslo, Norway between fetal death and HSV-2 antibody status in preg-
b
Department of Obstetrics and Gynaecology, Ullevål nancy: (i) presence of HSV-2 antibodies in the first trimes-
University Hospital, Norway ter, (ii) appearance of HSV-2 antibodies (incident
c
Department of Microbiology, Ullevål University Hospital, infection), (iii) increase in HSV-2 antibody titre and (iv)
Norway loss of HSV-2 antibodies during pregnancy.
d
Department of Obstetrics and Gynaecology,
Rikshospitalet, Oslo, Norway
e
Department of Bacteriology, National Institute of Public METHODS
Health, Oslo, Norway
* Correspondence: Dr A. Eskild, Section of Epidemiology, National The source population comprised 35,940 pregnant
Institute of Public Health, PO Box 4404 Nydalen, 0403 Oslo, Norway. women in Norway. They had participated in a prospective
D RCOG 2002 BJOG: an International Journal of Obstetrics and Gynaecology
PII: S 1 4 7 0 - 0 3 2 8 ( 0 2 ) 0 1 9 3 4 - 1 www.bjog-elsevier.com
 HERPES SIMPLEX VIRUS TYPE-2 AND FETAL DEATH 1031

study of Toxoplasma gondii infection in pregnancy per-

formed by the National Institute of Public Health from June
1992 until the last delivery within the study in May 199412.
The source population represented almost 100% of all
pregnant women in 11 of the 19 counties in Norway,
approximately 60% of all pregnant women in the whole
country during the study inclusion period13. Women with
induced abortions were not included.
In order to identify the women who had experienced
fetal death and a control group with live born children,
linkage between the Toxoplasmosis Study Registry and the
Medical Birth Registry of Norway14 by personal identifica-
tion numbers was performed. The Medical Birth Registry
contains information on all births in Norway after 16 weeks
of gestation since 1967. The data in this registry are
obtained by compulsory notification on standardised forms
completed by midwives at the delivery wards within one
week after birth.
Women were included in the study at their first antenatal
visit in primary health care, in the 10th to 12th pregnancy
week (mean 10.2 weeks, SD 3.2). Serum samples were
requested at the first antenatal visit and also in the 22nd and
in the 38th week of pregnancy. The sera were collected at
Fig. 1. The distribution of 281 fetal deaths according to pregnancy week at
regular visits by the doctors or midwives in charge of delivery. The number represents all fetal deaths in a source population of
antenatal care. 35,490 pregnant women in Norway 1992 – 1993 as reported to the Medical
The information on gestation length and vital status at Birth Registry.
delivery was obtained through linkage to the Medical Birth
Registry. If a fetal death was notified to the study admin- antibodies, but negative for HSV-2 were used. Each control
istration during the serum collection period, a serum serum was tested in duplicate. The cut off was calculated as
sample from the women was requested after delivery. the mean absorbance, measured as optical density of the
Fetal death was defined as death in pregnancy after 16 control sera plus three standard deviations. Optical density
weeks of gestation. Figure 1 demonstrates the distribution values equal to or more than cut off was defined as positive.
of gestational age at delivery among cases. Among the 281 The sensitivity of the test has been estimated as 97% and
fetal deaths, 82 (29%) occurred during labour as reported to the specificity as > 94%17. All sera were sampled, stored
the Medical Birth Registry. In the analysis, fetal death, and tested under identical conditions.
regardless of the time of death in relation to labour, was Definitions of the exposure variables:
defined as the outcome variable.
The following main explanatory variables were studied: 1. HSV-2 antibodies were defined as present (seropositive)
(i) presence of HSV-2 antibodies in the first trimester, (ii) if the optical density was equal or more than cut off.
appearance of HSV-2 antibodies (incident infection), (iii) High level of antibodies was defined as having optical
increase in HSV-2 antibody titre and (iv) loss of HSV-2 density z 5 times cut off. This definition was used in a
antibodies during pregnancy. subanalysis.
The serum samples had been stored at the National 2. Appearance of antibodies (incident HSV-2 infection)
Institute of Public Health at 20jC since 1992– 1993 and was defined as more than a twofold increase in optical
were analysed for HSV-2 antibodies during the autumn of density from below cut off in the first serum sample
1996. For determination of HSV-2 specific IgG antibodies, (seronegative) to above cut off in the last available
a protocol described by Ades et al.15 was followed. In sample.
short, the following procedure was used. Microtiter plates 3. Increase in HSV-2 antibody titre was defined as more
were coated with purified gG-2 antigen in an optimal than 1.5-fold increase in optical density from the first to
dilution and used in a conventional enzyme immunoassay the last available serum sample in women initially
system16. Sera were tested at 1 in 200 dilution on the seropositive.
antigen-coated plates. Bound IgG antibodies were detected 4. Loss of HSV-2 antibodies was defined as having
by adding horseradish peroxidase-conjugated; affi Pure detectable antibodies in the first serum sample (sero-
goat anti-human IgG (H&L) (Jackson Immuno Research positive, optical density equal or more than cut off) and
Laboratories, West Baltimore Pike, Pennsylvania). For not detectable antibodies (optical density less than cut
determination of cut off, three control sera with HSV-1 off) in the last available serum sample.
D RCOG 2002 Br J Obstet Gynaecol 109, pp. 1030 – 1035
1032 A. ESKILD ET AL.

The exposure risk time, the time from the first to the last group of 970 women with a live born child were drawn
serum sample, was treated as a confounding variable, at random.
thereby controlling for differences in the time of being at For 11 of the 1253 women, two cases and nine controls,
risk of change in HSV-2 antibody status between cases and sufficient serum amounts for HSV-2 antibody testing were
controls. This information was obtained from the Toxo- not available. Thus, the study sample comprised 281 cases
plasmosis Study Registry. In addition, maternal age and and 961 controls, a total of 1242 women.
parity, as obtained from the Medical Birth Registry, were The mean (SD) age was 28.9 (5.3) years, 29.7 (5.5) years
treated as potential confounders. for the cases and 28.7 (5.2) for the controls. A total of 41%
Crude and adjusted odds ratios with 95% confidence of the women had no prior deliveries reported to the
intervals of fetal death according to HSV-2 antibody status Medical Birth Registry, 36% had one delivery and 23%
were estimated in logistic regression models using the had two or more prior deliveries.
SPSS software version 10. Only one serum sample from the pregnancy period was
The study was approved by the Norwegian Data Inspect- available for antibody analyses for 16% of the women (197/
orate and the National Board of Health and recommended 1242). That was 32% (90/281) of the cases and 11% (107/
by the Regional Committee for Medical Research and the 961) of the controls. The women with only one serum
Advisory Committee for the Medical Birth Registry of sample included 109 women with presence of IgG anti-
Norway. bodies against T. gondii in the first serum sample, for
whom only one serum sample has been requested. The
second most important cause of lack of follow up serum
RESULTS was preterm delivery, including fetal death.
Twenty-seven percent (338/1242) of all women had
For 1851 of the 35,940 women (5%) in the toxoplasmo- antibodies against HSV-2 in the first serum sample from
sis study, the personal identification numbers could not be the pregnancy period. That was 29% (82/281) of the
obtained, neither from the Toxoplasmosis Study Registry women who experienced a fetal death and 27% (256/961)
nor from the Norwegian Central Person Registry, to which of the women with a live born child (odds ratio 1.1, 95% CI
a linkage by name and birth date was performed. Women 0.8– 1.5) (Table 1).
with an incomplete personal identification number were High antibody levels against HSV-2 may be an indica-
excluded from the linkage to the Medical Birth Registry, tion of acute or reactivated infection. An association
and thereby from the study population. between high antibody level (z5 times cut off ) and fetal
When the linkage between the Toxoplasmosis Study death was not seen (5.3% [15/281] vs 5.5% [53/961]; odds
Registry and the Medical Birth Registry had been per- ratio 1.0, 95% CI 0.5– 1.7).
formed in 1996, all women with a fetal death after 16 The 197 women without follow up sera from later stages
weeks of gestation were identified (n ¼ 283), representing in pregnancy were included in a subanalysis. Among these
0.8% (283/34,089) of the source population. A control women, the cases were overrepresented (46%, 90/197).

Table 1. Proportion with antibodies against HSV-2 in the first trimester of pregnancy among women with a fetal death and women with a live born child and
the crude and adjusted odds ratio with 95% confidence intervals (adjusted for age and parity). Values are given as proportion (%) and odds ratios [95% CI].

Fetal death Crude odds ratio Adjusted odds ratio

[95% CI] [95% CI]
Yes No

Total study sample n ¼ 281 n ¼ 961

HSV-2 antibodies presence (optical density > cut off )
Yes 82 (29) 256 (27) 1.1 [0.8 – 1.5] 1.1 [0.8 – 1.5]
No 199 (71) 705 (73) 1.0 1.0
High levels of HSV-2 antibodies (optical density > 5 times cut off )
Yes 15 (5) 53 (6) 1.0 [0.5 – 1.7] 0.9 [0.5 – 1.6]
No 266 (95) 908 (94) 1.0 1.0

Subsample of women not included in follow up n ¼ 90 n ¼ 107

Presence of HSV-2 antibodies (optical density > cut off )
Yes 27 (30) 25 (23) 1.4 [0.7 – 2.6] 1.4 [0.7 – 2.7]
No 63 (70) 82 (73) 1.0 1.0
High levels of HSV-2 antibodies (optical density > 5 times cut off )
Yes 6 (7) 6 (6) 1.2 [0.3 – 3.7] 1.0 [0.3 – 3.5]
No 84 (93) 101 (94) 1.0 1.0

Cut off is defined as a function of the absorbance level (optical density) of the serum, as measured by the enzyme immunoassay16.

D RCOG 2002 Br J Obstet Gynaecol 109, pp. 1030 – 1035

 HERPES SIMPLEX VIRUS TYPE-2 AND FETAL DEATH 1033

Table 2. Proportion with changes in antibodies against HSV-2 during pregnancy in women with a fetal death and in women with a live born child and the
crude and adjusted odds ratio with 95% confidence intervals according to changes in antibody status (adjusted for age, parity and follow up time). Values are
given as proportion (%) and odds ratios [95% CI].

Fetal death Crude odds ratio Adjusted odds ratio

[95% CI] [95% CI]
Yes No

Occurrence of HSV-2 antibodies in 759 HSV-2 seronegative women

Yes 3/136 (2) 16/623 (3) 0.9 [0.2 – 3.0] 0.7 [0.2 – 3.6]
No 133/136 (98) 607/623 (97) 1.0 1.0

Increase in of HSV-2 antibodies in 286 HSV-2 seropositive women

Yes 2/55 (4) 16/231 (7) 0.5 [0.1 – 1.6] 0.6 [0.1 – 2.8]
No 53/55 (96) 215/231 (93) 1.0 1.0

Loss of HSV-2 antibodies in 286 HSV-2 seropositive women

Yes 23/55 (42) 104/231 (45) 0.8 [0.5 – 1.6] 0.9 [0.4 – 1.8]
No 32/55 (58) 127/231 (55) 1.0 1.0

Seventy percent of the cases without follow up serum had a (23/55) of the cases and 45% (104/231) of the controls.
delivery before the 25th week of pregnancy. A non-sig- The adjusted odds ratio of fetal death for the women with
nificant higher proportion of HSV-2 seropositives was seen HSV-2 antibody loss was 0.9 (95% CI 0.4 –1.8) (Table 2).
among cases, 30% (27/90) than among the controls, 23%
(25/107) (odds ratio 1.4, 95% CI 0.7 – 2.6) (Table 1). Also,
no difference between cases and controls in the proportion DISCUSSION
with high levels of HSV-2 antibodies was seen (odds ratio
1.2, 95% CI 0.3 –3.7). In this nested case control study based on a source
When controlling for maternal age at delivery and parity, population of 35,940 pregnant women, there was no
the odds ratio of fetal death for presence of HSV-2 association between fetal death and HSV-2 antibody status
antibodies was not significantly above 1.0 in any of the in pregnancy. Neither the presence of antibodies, as an
subgroups in Table 1. indicator of latent infection, nor the appearance of anti-
A total of 1045 women (84%) had two blood samples bodies, as an indicator of primary infection in pregnancy,
or more and could therefore be included in the follow up was identified as risk factor for fetal death.
(191 cases and 854 controls). The mean (SD) pregnancy Apart from the study by Brown et al.5, which included
duration at delivery was 29.7 (8.0) weeks for cases fetal death as one of many possible adverse pregnancy
and 39.8 (2.3) weeks for controls who were followed outcomes of HSV infection, no other large scale studies on
up. The mean exposure risk time, the period between this topic have, to our knowledge, been performed. In the
first and last serum samples, was 24.5 (6.7) weeks for study by Brown et al., stillbirth occurred in 55 women of
the total study sample, 19.0 (7.6) weeks (range 3– 35) whom one woman had HSV-2 seroconversion and no risk
for the cases and 25.5 (6.1) weeks (range 2 – 34) for estimate of fetal death according to HSV-2 infection was
the controls. For 81 cases, the last serum sample was presented. Our study comprised about five times as many
collected after delivery. women with a fetal death. Still, it may be claimed that
Seven hundred and fifty-nine (73%) of the women insufficient numbers of women were included to give
followed up were HSV-2 seronegative in the first trimester. reliable risk estimates of the impact of HSV-2 infection on
Of these women, 2.5% (19/759) seroconverted. That was fetal death, since both these events are rare. However, based
2.2% of the cases (3/136) and 2.6% of the controls (16/623) on our study and the results in the study by Brown et al.,
(odds ratio 0.9, 95% CI 0.2 – 3.0). When controlling for HSV-2 infection is unlikely to be an important cause of fetal
difference in exposure risk period, age and parity between death.
cases and controls, the adjusted odds ratio was 0.7 (95% Serum samples obtained after delivery were not rou-
CI 0.2– 3.6) (Table 2). tinely collected in our study. For 43% of the cases with
Among the initially seropositive women, an increase in follow up serum (81 out of 188 cases for whom both the
HSV-2 antibody level occurred in 6.3% (18/286). That was date of delivery and the date of last serum sampling were
3.6% of the cases (2/55) and 6.9% of the controls (16/231) available), the serum was collected after delivery. Hence,
(Table 2). The adjusted odds ratio of fetal death in women changes in HSV-2 antibodies shortly before fetal death may
with an increase in HSV-2 antibodies was 0.6 (95% CI not have been detected for some women. There is, how-
0.1 –2.8) (Table 2). ever, little reason to believe that changes in their antibody
Among the initially seropositive women, loss of HSV-2 status occurred more often in women with serum sampling
antibodies occurred in 44% (127/286). That was 42% after than before delivery. Nine of the 81 cases (11%) with
D RCOG 2002 Br J Obstet Gynaecol 109, pp. 1030 – 1035
1034 A. ESKILD ET AL.

serum collected after delivery and 16 of the 107 cases (15%) antibodies in women with or without delivery before the
with the last serum collected before delivery had changes in 25th week of pregnancy or with or without fetal death
antibody status. For one out of the three cases with occur- occurring during labour.
rence of antibodies, the last blood sample was obtained after In our study sample, drawn at random among 35,490
delivery. Hence, we have little reason to believe that pregnant women, 27% of women had antibodies against
changes in antibody status have been under-estimated HSV-2 at study inclusion. Similar prevalence estimates
among cases and thereby under-estimating the association have been reported from other studies. In California, the
between HSV-2 infection and fetal death in our study. prevalence was 32%4. In a Swedish study of young
Sixteen percent (197/1242) of the women could not be females, the HSV-2 seroprevalence was 25% in the age
followed up with regard to HSV-2 antibody changes during group 28– 29 years21. In the study by Brown et al.5 in
pregnancy since follow up serum was lacking. Lack of Seattle, 11% of pregnant women were HSV-2 seropositive.
follow up may have caused biased estimates if the impact Among the initially HSV-2 seronegative women in our
of HSV-2 infection on fetal death among the women not study, 2.5% seroconverted during the follow up period of 25
followed up differed from the remaining study sample. The weeks. Hence, the cumulative incidence of seroconversion
lack of association between high HSV-2 antibody level in during the pregnancy period of 40 weeks can be estimated at
the first trimester, an indication of ongoing infection, and 4%, given the same risk during the whole of pregnancy.
fetal death supports no impact of HSV-2 infection in the Compared with previous studies, the HSV-2 incidence in
subsample lost to follow up. There was no difference in our study was relatively high. In Seattle, Brown et al.5
HSV-2 antibody status at inclusion between women with found that HSV-2 was acquired by 1.4% of susceptible
and without follow up serum, suggesting similarity with women during pregnancy. In the Swedish study of young
regard to HSV-2 infection risk. females, the HSV-2 incidence was estimated at 2.3% per
Twenty-five percent of the fetal deaths in our study year21. In a study of seronegative women in HSV-2 dis-
occurred before the 21st week of pregnancy and 45% cordant sexual relationships, 6% seroconverted during
before the 25th week of pregnancy. For some of the cases pregnancy4. Our definition of incident HSV-2 infection
included, induced abortion may have been misclassified as was the occurrence of HSV-2 antibodies, with twofold
fetal death in the Medical Birth Registry. The extent of increase in optical density from below cut off in the first
such possible misclassification is not known, but assumed serum to above cut off in the last serum sample. Some
to account for less than 20% of fetal deaths before the 25th women may falsely have been defined as having incident
week of pregnancy. In a national study of all induced infection during pregnancy. Differential misclassification of
abortions in 1996 – 1997, the estimated induced abortion incident HSV-2 infection according to vital status at deliv-
rate after the 16th week of pregnancy was 2– 3 per 1000 ery and thereby biased risk estimate is, however, unlikely.
births18,19. Pregnancy termination after the 21st week of Increase in HSV-2 antibodies in initially seropositive
pregnancy is only rarely performed and only in cases of women may be an indicator of reactivation of HSV-2
fetal malformation incompatible with life or severe mater- infection and was detected in 6% of these women. There
nal disease19. The fetal death rate as reported to the was a higher proportion of controls (7%) as compared with
Medical Birth Registry is 9 – 10 per 1000 birth20. Misclas- cases (4%) with an increase in HSV-2 antibodies. This
sification of induced abortion as fetal death would lead to difference was not significant.
an under-estimation of the association between HSV-2 Loss of HSV-2 antibodies was seen in 44% of initially
infection and fetal death. Such misclassification is unlikely seropositive women in our study. An association between
to be an important source of error in the analyses of HSV-2 pregnancy duration and loss of antibodies in this study
antibody changes and fetal death, since the follow up serum population has been presented in a previous report22. This
was collected after the 22nd week of pregnancy and the may be explained by a haemodilution during normal
mean duration of gestation at delivery was 29.7 (8) weeks. pregnancy and thereby a reduction in the concentration of
Misclassification of fetal death may, however, have been a circulating antibodies23,24. Since the risk of antibody loss
source of bias in the subanalyses of women without follow increases by gestational length, adjustment for follow up
up serum of whom 70% (63/90) had delivery before the time should reduce the impact of this potential source of
25th week of pregnancy. confounding in our study.
Among the women with a fetal death and delivery before Haemodilution may not be the only explanation of
the 25th week of pregnancy, 37% (47/127) occurred during antibody loss, since loss of antibodies against HSV-1
labour, whereas 23% (35/154) of the fetal deaths in the during pregnancy has not been seen25. The normal serum
25th week of pregnancy or later occurred during labour, as level of antibodies during the different stages of pregnancy
reported to the Medical Birth Registry. Analyses of the is not well known. Such knowledge would make it easier to
impact of HSV-2 antibody change within subcategories of study whether discrepancies from the normal are associated
fetal death were not performed due to the limited number with adverse pregnancy outcome. The results from our
of women with antibody changes. There was, however, study, however, do not suggest that loss of HSV-2 anti-
no significant difference in the prevalence of HSV-2 bodies is associated with fetal death.
D RCOG 2002 Br J Obstet Gynaecol 109, pp. 1030 – 1035
 HERPES SIMPLEX VIRUS TYPE-2 AND FETAL DEATH 1035

CONCLUSION 10. Young EJ, Chafizadeh E, Oliveira VL, Genta RM. Disseminated her-
pesvirus infection during pregnancy. Clin Infect Dis 1996;22:51 – 58.
11. Robb JA, Benirschke K, Mannino F, Voland J. Intrauterine latent her-
Both fetal death and changes in HSV-2 antibodies are pes simplex virus infection: II. Latent neonatal infection. Human
rare events. Although all fetal deaths in a population of Pathol 1986;17:1210 – 1217.
35,940 pregnant women were included as cases in this 12. Jenum PA, Stray-Pedersen B, Melby KK, et al. Incidence of Toxo-
study, a true association between changes in HSV-2 anti- plasma gondii infection in 35,940 pregnant women in Norway and
body status and fetal death may have remained undetected. pregnancy outcome for infected women. J Clin Microbiol 1998;36:
2900 – 2906.
However, as a conclusion, our study gives no evidence of 13. Statistics in Norway. Statistical Yearbook 1993. 12th Issue. Oslo,
an association between HSV-2 infection and fetal death. Norway: Statistics Norway, 1993.
Resources should be concentrated on preventing HSV-2 14. Lie RT, Wilcox AJ, Skjærven R. A population based study of the risk
transmission to the child during delivery. of recurrence of birth defects. N Engl J Med 1994;331:1 – 4.
15. Ades AE, Peckham CS, Dale GE, Best JM, Jeansson S. Prevalence of
antibodies to herpes simplex virus types 1 and 2 in pregnant women
and estimated rates of infection. J Epidemiol Community Health
Acknowledgements 1989;43:53 – 60.
16. Svennerholm B, Olofsson S, Jeansson S, Vahlne A, Lycke E. Herpes
The authors would like to thank Trine Skjerden, Lise simplex type-selective enzyme-linked immunosorbent assay with
Andresen and Gunilla Løvgården for excellent technical Helix Pomatia lectin-purified antigens. J Clin Microbiol 1984;19:
235 – 239.
assistance. The authors would also like to thank the efforts 17. Ho DWT, Field PR, Sjögren-Jansson E, Jeansson S, Cunningham AL.
of the staff at the Medical Birth Registry. The study was Indirect ELISA for the detection of HSV-2 specific IgG and IgM
supported by a grant from Norske kvinners sanitetsforening. antibodies with glycoprotein G (gG-2). J Virol Methods 1992;36:
249 – 264.
18. Eskild A, Nesheim B-I, Berglund T, Totlandsdal JK, Andresen JF.
Geographical variation in induced abortion after 12th pregnancy
References week in Norway 1996 – 1997. J Nor Med Assoc 2001;121:24 – 27
[in Norwegian].
1. Whitley R, Arvin A, Prober C, et al. Predictors of morbidity and 19. Eskild A, Nesheim B-I, Berglund T, Totlandsdal JK, Andresen JF.
mortality in neonates with herpes simplex infections. N Engl J Med Induced abortion due to fetal malformation in Norway 1996 – 1997.
1991;324:450 – 454. J Nor Med Assoc 2000;120:1000 – 1003 [in Norwegian].
2. Brown ZA, Benedetti J, Ashley R, et al. Neonatal herpes simplex 20. Births in Norway through 30 years. Medical Birth Registry of Norway
virus infection in relation to asymptomatic maternal infection at the 1967 – 1996. Bergen: Medical Birth Registry, 1997 [in Norwegian].
time of labor. N Engl J Med 1991;324:1247 – 1252. 21. Christenson B, Böttiger M, Svensson Å, Jeansson S. A 15-year
3. Prober CG, Hensleigh PA, Boucher FD, Yasukawa LL, Au DS, surveillance study of antibodies to herpes simplex virus types 1 and
Arvin AM. Use of routine viral cultures at delivery to identify neonates 2 in a cohort of young girls. J Infect 1992;25:147 – 154.
exposed to herpes simplex virus. N Engl J Med 1988;318:887 – 891. 22. Eskild A, Jeansson S, Hagen J, Jenum PA, Skrondal A. Herpes
4. Kulhanjian JA, Soroush V, Au DS, et al. Identification of women at simplex virus type-2 antibodies in pregnant women: the impact of
unsuspected risk of primary infection with herpes simplex virus type 2 the stage of pregnancy. Epidemiol Infect 2000;125:685 – 692.
during pregnancy. N Engl J Med 1992;326:916 – 920. 23. Yasuhara M, Tamaki H, Iyama S, Yamaguchi Y, Tachi J, Amino N.
5. Brown ZA, Selke S, Zeh J, et al. The acquisition of herpes simplex Reciprocal changes in serum levels of immunoglobulins (IgG, IgA,
virus during pregnancy. N Engl J Med 1997;337:509 – 515. IgM) and complements (C3, C4) in normal pregnancy and after
6. Hyde SR, Giacoia GP. Congenital herpes infection: placental and delivery. J Clin Lab Immunol 1992;38:137 – 141.
umbilical cord findings. Obstet Gynecol 1993;81:852 – 855. 24. Amino N, Tanizawa O, Miyai K, et al. Changes of serum immuno-
7. Mansour AM, Nichols MM. Congenital diffuse necrotizing retinitis. globulins IgG, IgA, IgM, and IgE during pregnancy. Obstet Gynecol
Graefe’s Arch Clin Exp Ophthalmol 1993;231:95 – 98. 1978;52:415 – 420.
8. Sarkell B, Blaylock WK, Vernon H. Congenital neonatal herpes 25. Arvaja M, Lethinen M, Koskela P, Lappalainen M, Paavonen J,
simplex virus infection. J Am Acad Dermatol 1992;27:817 – 821. Vesikari T. Serological evaluation of herpes simplex virus type 1
9. Nahmias AJ, Josey WE, Naib ZM, Freeman MG, Fernandez RJ, and type 2 in pregnancy. Sex Transm Infect 1999;75:168 – 171.
Wheeler JH. Perinatal risk associated with maternal genital herpes
simplex virus infection. Am J Obstet Gynecol 1971;110:825 – 837. Accepted 11 July 2002

D RCOG 2002 Br J Obstet Gynaecol 109, pp. 1030 – 1035