SEARCH

SEARCH BY CITATION

Keywords:

  • Fetal death;
  • infection;
  • parvovirus B19;
  • pregnancy;
  • risk factors

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Detection of antibodies against parvovirus B19
  6. Results
  7. Discussion
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Editors' Commentary
  15. Conflict of interest

Objectives  To assess the association between maternal parvovirus B19 infection and fetal death, birthweight and length of gestation.

Design  Case–control study.

Setting  Population based.

Population  Cases were all 281 women with fetal death within a cohort of 35 940 pregnant woxmen in Norway. The control group consisted of a random sample of 957 women with a live born child.

Method  Information on pregnancy outcome was obtained from the Medical Birth Registry of Norway. First trimester serum samples were tested for antibodies against parvovirus B19 (IgM and IgG). In seronegative women, further serum was analysed to detect seroconversion during pregnancy.

Main outcome measures  Fetal death, length of gestation and birthweight.

Results  Two of 281 (0.7%) of the women who experienced fetal death and nine of 957 (0.9%) of the controls had presence of IgM antibodies, crude odds ratio 0.8; 95% CI (0.2–3.5). In initially, seronegative women, 3.1% (2/65) with fetal death and 2.6% (8/307) with a live birth seroconverted, crude odds ratio 1.2; 95% CI (0.2–5.7). Presence of maternal parvovirus-specific IgG or IgM antibodies in the first trimester, or seroconversion during pregnancy were not associated with lower birthweight or reduced length of gestation in live born children, but was associated with low birthweight in stillborn offspring.

Conclusion  Maternal parvovirus B19 infection was not associated with fetal death in our study. Very few cases of fetal death may be attributed to maternal parvovirus B19 infection.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Detection of antibodies against parvovirus B19
  6. Results
  7. Discussion
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Editors' Commentary
  15. Conflict of interest

Studies have revealed a relationship between fetal death and human parvovirus B19. The virus may cause a wide spectrum of clinical symptoms ranging from erythema infectiosum (the fifth disease), arthropathy to triggering of autoimmune disorders and transient aplastic crises in patients with increased red cell turnover.1–3 Special attention has been focused on its fetal impact since 1984, when a link between human parvovirus B19 and adverse pregnancy outcome was first proposed;4,5 namely non-immune hydrops fetalis and fetal death.6

Human parvovirus B19 is a common virus with increasing seroprevalence with age. Antibody studies among pregnant women indicate that 30–57% are susceptible, and that the incidence of seroconversion among these women is 1–2%, but may increase up to 16.7% during an epidemic.7–11 The estimated fetal transmission rate is approximately 30%.12

A few prospective studies on the association between fetal death and parvovirus B19 have been carried out. A study from England, including 190 pregnant women with symptomatic infection, reported a fetal death risk of 9% with a clustering in the second trimester.12 Other studies have found a risk of fetal death of 7–13%, and 3% of hydrops fetalis.7,13–17 In other studies, however, there was no relationship between maternal parvovirus B19 infection and adverse pregnancy outcome.11,18 Approximately, 50% of parvovirus B19 infections in pregnant women are asymptomatic.14 Most of the previous studies of parvovirus B19 infection and fetal death did not include a group of non-infected women for comparison.13–16 However, in one study including all fetal deaths (n = 47) among 14 147 pregnancies, seven (15%) of the dead fetuses were parvovirus B19 DNA positive while none of the 50 live born children tested were positive.19 As previous studies were based on small and selected samples, and have shown conflicting results, the attributable risk of maternal parvovirus B19 infection for fetal death in the general population remains largely unknown.

The aim of our study was to assess the association of fetal death with past and present maternal human parvovirus B19 infection. We also studied the association of maternal parvovirus B19 infection with fetal birthweight and length of gestation.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Detection of antibodies against parvovirus B19
  6. Results
  7. Discussion
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Editors' Commentary
  15. Conflict of interest

Study population

The source population comprised 35 940 pregnant women in Norway. They participated in a prospective study on Toxoplasma gondii infection in pregnancy performed by the Norwegian Institute of Public Health from June 1992 to May 1994.20 The serum samples from this study were stored in a biobank. The study population included almost 100% of the pregnant women in 11 of 19 counties in Norway and approximately 60% of all pregnant women in Norway during the study inclusion period.21

All women who had experienced fetal death and a control group with live born children were identified by linkage between the Toxoplasmosis Study Registry and the Medical Birth Registry of Norway by personal identification numbers.22,23 This registry contains information on all births after 16 weeks of gestation in Norway and is obtained by compulsory notification on standardised forms.

The personal identification numbers of 1851 of the 35 940 women (5%) in the toxoplasmosis study could not be obtained and these women were excluded. Two hundred and eighty-three (0.8%) women with fetal death after 16 weeks of gestation were identified.22,23 A control group of 970 women were randomly selected among all women in the cohort who delivered a live born child (n = 35 657). For 15 of the 1253 women there was insufficient serum available for antibody testing (2 cases and 13 controls). Thus, 1238 women, 281 cases and 957 controls, were included in the study, where the exposure was presence or occurrence of parvovirus B19 antibodies and the main outcome was fetal death. Birthweight and length of gestation was studied in both cases and controls.

The mean maternal age was 28.9 years (range 16–47 years). Forty-one percent of the women had no prior deliveries, 36% had one and 23% had two or more prior deliveries.

Serum sampling

The women were included in the study at their first antenatal visit to the primary healthcare centre (mean 10.2 weeks of gestation). Serum samples were requested at the first antenatal visit; and for the women without antibodies against T. gondii, additional serum was requested at the 22nd and the 38th week of pregnancy. The serum collection was performed at regular visits. If a fetal death was notified to the study administrator an additional serum sample from the woman was requested.20 Serum sampling after delivery was otherwise not performed according to the toxoplasmosis study protocol.

Detection of antibodies against parvovirus B19

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Detection of antibodies against parvovirus B19
  6. Results
  7. Discussion
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Editors' Commentary
  15. Conflict of interest

The serum samples were stored at −20 °C and analysed for human parvovirus B19 antibodies during the autumn of 1996 at the Department of Virology, Norwegian Institute of Public Health. All sera were sampled, stored and tested under identical conditions by one technician without knowledge of pregnancy outcome.

The first serum sample from each woman was tested separately for immunoglobulin G (IgG) and M (IgM) antibodies against parvovirus B19 (IDEIA Parvovirus B-19 IgM and IgG; DAKO A/S, Copenhagen, Denmark). If the first serum sample from the pregnancy period was negative for IgG and IgM antibodies, the last available sample from the woman was also analysed for IgG and IgM antibodies to detect possible seroconversion. Serum that tested positive within the grey zone was retested.

Antibody status against parvovirus B19 was categorised as follows:

  • 1
    No previous infection: antibodies against B19 not detected.
  • 2
    Previous B19 infection: presence of IgG antibodies against parvovirus B19 in the first serum sample, but no IgM antibodies.
  • 3
    Presence of IgM antibodies: presence of IgM with or without concomitant IgG antibodies against parvovirus B19 in the first serum sample.
  • 4
    Seroconversion: occurrence of IgG or IgM antibodies against parvovirus B19 in initially seronegative women.

Fetal outcome variables

Information on fetal death, birthweight and length of gestation was obtained from the Medical Birth Registry. Fetal death was defined as death after 16 weeks of gestation. The distribution of fetal deaths in this study sample according to length of gestation is presented elsewhere.24

Statistical analysis

Differences in maternal parvovirus B19 antibody status between cases and controls were analysed by Fisher’s exact test and chi-square test. The associations between maternal parvovirus B19 infection and fetal death were estimated as crude and adjusted odds ratios (OR) with 95% confidence intervals (95% CI), adjustment was made for parity and maternal age, and for the women followed with regard to seroconversion, adjustment was also made for time between first and last serum sample. Differences in mean birthweight and length of gestation according to maternal parvovirus B19 antibody status in live born and stillborn offspring were tested with Student’s t-test and Mann–Whitney test. All statistical analyses were performed by using the Statistical Package for the Social Sciences (SPSS, version 15.0; SPSS Inc., Chicago, IL, USA).

Ethical aspects

The study was approved by the Norwegian Data Inspectorate, the National Board of Health, the Regional Ethical Committee for Medical Research and the advisory Committee for the Medical Birth Registry of Norway.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Detection of antibodies against parvovirus B19
  6. Results
  7. Discussion
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Editors' Commentary
  15. Conflict of interest

Maternal parvovirus B19 infection and fetal death

In total, 64% (796/1238) of all women had presence of IgG antibodies against parvovirus B19 in the first serum sample and were therefore not susceptible to primary infection. Among the women who experienced fetal death, 68% (190/281) had parvovirus B19-specific IgG, and 63% (606/957) of the women without fetal death were positive (= 0.2, chi square test) OR 1.2 [0.9–1.6, 95% CI] (Table 1).

Table 1.   The risk (odds ratio (OR) with 95% confidence interval [CI]) of fetal death associated with presence or occurrence of antibodies against parvovirus B19, immunoglobulin G (IgG) or immunoglobulin M (IgM), in a case–control study within a cohort of 35 940 pregnant women in Norway
 Fetal death Crude OR [95% CI]*Adjusted OR [95% CI]
YesNo 
  1. *Adjusted for maternal age and parity.

  2. **Additional adjustment for follow-up time (time between first and last serum collected).

  3. ***Only women susceptible to parvovirus B19 infection who had two or more serum samples collected are included (65 women with fetal death and 307 women without fetal death).

Presence of IgG antibodies
Yes190 (67.6%)606 (63.3%)796 (64.3%)1.21 [0.91–1.60]0.84 [0.63–1.12]
No91 (32.4%)351 (36.7%)442 (35.7%)  
Presence of IgM antibodies
Yes2 (0.7%) 9 (0.9%) 11 (0.9%) 0.76 [0.16–3.52]1.45 [0.31–6.78]
No279 (99.3%)948 (99.1%)1227 (99.1%)  
Occurrence of antibodies***
Yes2 (3.1%) 8 (2.6%) 10 (2.7%) 1.18 [0.25–5.70]**0.82 [0.15–4.52]
No63 (96.6%)299 (97.4%)362 (97.3%)  

Eleven women had presence of IgM antibodies against parvovirus B19 in first trimester, representing 0.9% of all women (11/1238) and 2.5% (11/442) of the susceptible women. Eight of these 11 women had presence of IgG in addition to IgM antibodies. A total of 0.7% (2/281) of the women who experienced fetal death and 0.9% (9/957) of the controls (= 1.0, Fisher’s exact test) had presence of IgM antibodies, giving a crude OR of 0.8 (0.2–3.5, 95% CI). Adjustment for parity and maternal age did not change the estimated risk significantly (Table 1).

Ten women had serological evidence of seroconversion during pregnancy, representing 0.9% (10/1238) of all women. Of the 442 women who were parvovirus B19 seronegative in the first serum sample, 372 women had two or more stored serum samples and could be followed with regard to seroconversion. Among these, 3.1% (2/65) with fetal death and 2.6% (8/307) with a live born offspring seroconverted (= 0.7, Fisher’s exact test), crude OR 1.2 (0.2–5.7, 95% CI). Among the 10 women who seroconverted during pregnancy, only two had presence of IgM antibodies against parvovirus B19 without concomitant IgG antibodies. After adjustment for parity, maternal age and time between first and last serum sample (follow-up time) the OR of fetal death was 0.8 (0.2–4.5, 95% CI). The mean follow-up time for cases was 19.0 and 25.5 weeks for controls.

Hence, among the 281 women who experienced fetal death, only four had serological signs of acute parvovirus B19 infection in pregnancy, either presence of IgM antibodies in the first serum sample or seroconversion.

Maternal parvovirus B19 infection, length of gestation and offspring birthweight

Past and present maternal parvovirus B19 infection was not significantly associated with birthweight or length of gestation in live born offspring (Table 2).

Table 2.   Mean [SE] length of gestation and birthweight in 281 stillborn and 957 live born offspring according to maternal parvovirus B19 antibody status
Maternal antibody statusnLength of gestation (weeks)Birthweight (grams)
  1. Presence of antibodies in first trimester was studied in 1238 pregnant women. Occurrence of antibodies was studied in n = 372 women without antibodies in the first trimester who were followed during pregnancy.

Live born
Presence of IgG antibodies
 Yes60639.9 [0.1]3565 [23]
 No35139.7 [0.1]3533 [23]
Presence of IgM antibodies/occurence of antibodies
 Yes 1740.1 [0.6]3726 [137]
 No94039.8 [0.1]3550 [19]
Stillborn
Presence of IgG antibodies
 Yes19028.3 [0.6]1302 [90]
 No9127.9 [0.9]1280 [137]
Presence of IgM antibodies/occurence of antibodies
 Yes 423.0 [1.8]338 [123]
 No27728.3 [0.5]1309 [76]

In women with stillborn offspring, the mean length of gestation was 23.0 weeks for women with acute parvovirus B19 infection and 28.3 weeks in women without acute parvovirus B19 infection (= 0.3, Mann–Whitney test). The mean birthweight in the stillborn offspring was 338 g for women with acute infection, and 1309 g in women without acute parvovirus B19 infection (= 0.1, Mann–Whitney test) (Table 2).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Detection of antibodies against parvovirus B19
  6. Results
  7. Discussion
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Editors' Commentary
  15. Conflict of interest

In this case–control study within a cohort of 35 940 pregnant women, we found no association between maternal parvovirus B19 infection and the risk of fetal death. Of 281 fetal deaths only four were born to mothers with serological signs of acute parvovirus B19 infection in pregnancy. Maternal antibody status had no association with birthweight and length of gestation in live born children.

Since the first case reports, parvovirus B19 infection has been linked to adverse outcome of pregnancy.1–7 To our knowledge, no prior study has explored the association between maternal parvovirus B19 antibody status, length of gestation and birthweight.

Prior estimates of the risk of fetal death associated with parvovirus B19 have varied.4 Such variation may be because of differences in study design, selection of study sample and diagnostic methods.8,9,11–13,15,25 In some prior studies only women with symptomatic parvovirus B19 infection in pregnancy have been included.12,14,15 Yaegashi et al. followed 48 pregnant women either exposed to infected persons or having symptoms of infection. Among these women eight offspring with hydrops fetalis and seven cases (15%) of fetal deaths were reported.9 In the Netherlands, 2567 pregnant women within an unselected population were followed prospectively, with occurrence of antiparvovirus B19 IgG. They reported 18 cases of seroconversion, and no fetal deaths.8 In that study, under detection of acute infection may have occurred, as women with acute infection in first trimester might present with both IgG and IgM antibodies. A Swedish case-control study within 14 147 pregnancies, reported that 15% (7/47) of the cases of intrauterine fetal death after 22nd pregnancy week, and none of the controls (0/53), were parvovirus B19 DNA positive. This study did not describe the selection of the control group nor reported uncertainty of the risk estimates.19 Possible differences between cases and controls in type of tissue or number of tissue samples tested for parvovirus DNA were not described in detail.

Our study included all cases of fetal death (281) after 16th pregnancy week within 35 940 pregnancies. As we included almost all pregnant women in 11 counties in Norway during the inclusion period and have drawn controls at random, we believe no selection bias has occurred.

We used commercially available kits for detection of maternal antibodies against parvovirus B19 (IDEIA Parvovirus B-19 IgM and IgG). These kits have been shown to have the highest specificity (94.8%) compared with five other tests.26 However, as the specificity is not 100%, there is a risk of false positive results. We retested borderline positive (grey zone) sera to increase the specificity. Since parvovirus B19-specific IgM antibodies may persist for 2–3 months after infection,27 some of the women with presence of IgM antibodies in the first trimester may have been infected prior to onset of pregnancy. Also, presence of IgM antibodies without concomitant presence of IgG antibodies may be an unspecific serological reaction. However, 8 of the 11 women with presence of IgM in the first serum sample had concomitant presence of IgG antibodies against parvovirus B19. We do not believe over diagnosis is an important source of error in our study. During 1993–1994 there was an outbreak of parvovirus B19 infection in Norway29 and we found 4.8% (21/442) of the susceptible women to have signs of acute infection during pregnancy. Other studies have reported seroconversion rates as high as 16.7% during epidemics.7,11

Lack of follow up during pregnancy could potentially have caused biased estimates. Sixty-eight of the parvovirus B19 susceptible women (68/442) did not have follow-up serum and could therefore not be studied with regard to seroconversion. Twenty-five of these women experienced fetal death, of whom nine lacked follow-up serum as they had IgG antibodies against T. gondii in the first trimester. For T. gondii IgG positive women, no additional serum was requested according to the study protocol. Hence, 16 women with fetal death were true losses to follow up. Even if all cases lost to follow up had acquired parvovirus B19 in pregnancy, maternal infection could only be linked to 29 cases out of a total of 281 cases of fetal death in this cohort of 35 940 women. It is, however, very unlikely that lack of follow up was associated with increased risk of parvovirus B19 infection, and that such increased infection risk was differential according to offspring vital status.

Follow-up serum from after delivery was not routinely collected for all women in this study, but was requested in women with fetal death. Hence, parvovirus B19 seroconversion before delivery may have been under detected. In total, 10 of 1238 women (8 controls and 2 cases) seroconverted during a mean follow-up time of 24.5 weeks (25.5 weeks for controls and 19.0 weeks for cases). It is unlikely that serum sampling for all women after delivery would significantly increase the estimated impact of parvovirus B19 infection on fetal death. Since the incidence of parvovirus B19 was low in our study, additional 2–3 weeks of follow-up time would probably not have added many seroconverters. The potential lost seroconverters in pregnancy are most likely to be among the controls, as more cases than controls had serum available from after delivery. Among the 65 women with fetal death who were followed with regard to seroconversion, 35 women had serum drawn at or after delivery.

In our study, stillborn offspring of women with signs of acute parvovirus B19 infection in pregnancy had lower mean birthweight than women without acute infection. This observation may be spurious or may suggest growth restriction as a cause of fetal death in these cases. Other studies have reported a higher than expected proportion of small for gestational age offspring among pregnant women with serologically confirmed parvovirus B19 infection.12,25,28 Further research is required to confirm these observations.

In conclusion, we found no association between maternal parvovirus B19 in pregnancy and the risk of fetal death. Nor was exposure to parvovirus B19 during pregnancy associated with shorter length of gestation. Among the 281 women out of a cohort of 35 940 pregnancies, who experienced fetal death, only four had signs of acute parvovirus B19 infection in pregnancy. These observations may be of importance for clinicians dealing with prophylactic strategies and follow-up care of pregnant women with fetal death. In our study, parvovirus B19 was a seldom cause of fetal death and just as many women with a live born child as women experiencing stillbirth had an incident infection during pregnancy. Our findings do not suggest that parvovirus B19 is an important cause of fetal death.

Contribution to authorship

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Detection of antibodies against parvovirus B19
  6. Results
  7. Discussion
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Editors' Commentary
  15. Conflict of interest

Aahshi A. Sarfraz has analysed the data and written the manuscript. Sven O. Samuelsen has contributed to data analysis, the interpretation of the results and writing of the manuscript. Anne-Lise Bruu has been responsible for the serum analysis and interpretation of the results. Pål A. Jenum was the principal investigator of the toxoplasmosis study. Anne Eskild is the principal investigator of the fetal death study and the supervisor for Aahshi A. Sarfraz. All authors have contributed to the manuscript and approved the final version.

Details of ethics approval

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Detection of antibodies against parvovirus B19
  6. Results
  7. Discussion
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Editors' Commentary
  15. Conflict of interest

The study was approved by the Norwegian Data Inspectorate, the National Board of Health, the Regional Ethical Committee for Medical Research, and the advisory Committee for the Medical Birth Registry of Norway.

Funding

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Detection of antibodies against parvovirus B19
  6. Results
  7. Discussion
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Editors' Commentary
  15. Conflict of interest

The study was funded by the Norwegian Foundation for Health and Rehabilitation through Norwegian SIDS and Stillbirth Society and by the Norwegian Institute of Public health.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Detection of antibodies against parvovirus B19
  6. Results
  7. Discussion
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Editors' Commentary
  15. Conflict of interest

Editors' Commentary

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Detection of antibodies against parvovirus B19
  6. Results
  7. Discussion
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Editors' Commentary
  15. Conflict of interest

The incidence of acute primary maternal Parvovirus B19 infection during pregnancy is about 2–4%. It is often due to contact with young infected children, and is mostly asymptomatic in the mother. Maternal infection is usually demonstrated by the finding of positive IgM, or by IgG seroconversion. In mothers with primary infection, about 25–30% of fetuses become infected themselves by vertical transmission. Perinatal complications of fetal infection occur in about 10% of fetuses of infected mothers, and these include fetal anaemia and myocarditis. This can lead to fetal hydrops (in 2–6%) and occasionally fetal death, although this is rare if infection occurs after 20 weeks (Enders M, et al. Prenatal Diagnosis 2004;24:513–8).

Apart from detecting hydropic changes in the fetuses of infected mothers, ultrasound can also screen for the development of fetal anaemia by measurement of the peak systolic velocity (PSV) of the middle cerebral artery (MCA) and using a threshold of >1.5 multiples of the median (MoM). If the MCA PSV values are <1.5 MoM, serial ultrasound scans are recommended for 10–12 weeks after the exposure, because fetal infection and anaemia can occur late. If MCA PSV is >1.5 MoM, or if there is fetal hydrops, fetal transfusion is indicated.

Screening of pregnant women for parvovirus B19 is not recommended, as only about one in 5000 screened women would be at risk for fetal hydrops caused by this virus; the risk of fetal death is even lower. In this article [Sarfraz A, et al. BJOG 2009;116:1492–8], a nested case–control study, it is confirmed that very few cases of fetal death may be attributable to parvovirus B19 infection. This is supported by data from the USA suggesting that fewer than 1% of stillbirths result from this infection (Goldenberg RL, Thompson C. The infectious origins of stillbirth. Am J Obstet Gynecol 2003;189:861–73).

Nevertheless, parvovirus B19 is a known cause of fetal anaemia and hydrops, and by extension fetal death. Several organizations (ACOG: American College of Obstetricians and Gynecologists. Evaluation of stillbirths and neonatal deaths. ACOG Committee Opinion No. 383. Obstet Gynecol 2007;110:963–6) and recent reviews (Silver RM et al. Work-up of stillbirth: a review of the evidence. Am J Obstet Gynecol 2007;5:433–44) recommend testing for parvovirus for unexplained fetal death. Testing amniotic fluid by PCR in such cases increases sensitivity, and in one study of 93 fetal deaths, seven (7.5%) had detectable B19 DNA in freshly-frozen placental tissue (Skioldebrand-Sparre L, et al. BJOG 2000;107:476–80).

The current study found no difference in the incidence of a positive IgM (leading to the diagnosis of primary maternal infection) in the serum of 281 susceptible women who had fetal deaths versus 957 susceptible women who did not. Given that parvovirus is known to cause fetal anaemia and hydrops, which in turn are associated with fetal death, how can this be explained? The most likely reason is that this is due to a type 2 error (i.e. the sample sizes are too small to show a difference). Nevertheless, the findings are important: it appears that parvovirus B19 is rarely a cause of late fetal death, and women with a live born child have incident parvovirus infections during pregnancy as often as those with fetal loss. Based on this and other available data, one logical conclusion could be that testing for Parvovirus is important when the fetal death occurs before 24 weeks, when it is associated with exposure to parvovirus or close contact with children, or when there are fetal findings associated with this infection such as hydrops or anaemia. The value of testing for Parvovirus in the absence of such features is questionable, and merits further research.

Conflict of interest

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Detection of antibodies against parvovirus B19
  6. Results
  7. Discussion
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Editors' Commentary
  15. Conflict of interest

Neither author has a conflict to declare.

V Berghellaa, AT Papageorghioub a Professor, Obstetrics and Gynecology, Thomas Jefferson University, Philadelphia, USA bConsultant in Obstetrics and Fetal Medicine, Fetal Medicine Unit St. George's, University of London, London Email: a.papageorghiou@sgul.ac.uk