“Pregnant women were urged to get their annual flu jab yesterday as research showed they have a five times greater risk of a stillbirth if they are admitted to hospital with swine flu,” reported The Independent.
This news report is based on a study that followed 256 pregnant women admitted to hospital with the 2009 strain of the swine flu virus, called H1N1. The study compared their pregnancy outcomes to 1,220 healthy pregnant women. Women with swine flu were found to have a significantly higher rate of adverse events than uninfected pregnant women. This included about a fourfold higher rate of stillbirth and fivefold higher rate of neonatal death (when the baby dies within 28 days of life).
The promotion of these findings from a study published in June is designed to raise awareness of the importance of pregnant women having the seasonal flu vaccine, which now also protects against swine flu. Pregnant women are one of the high-risk groups advised to have an annual flu vaccination, and swine flu is predicted to be one of the main flu viruses circulating during the 2011/12 winter.
It is important to bear in mind that infant death is relatively rare in this country, and these figures are extrapolations of quite small numbers (of the 256 women with swine flu, 10 had stillbirth or their babies died between 24 weeks of pregnancy and within one week of delivery (perinatal mortality).
These increases in risk sound alarming but it is important to bear in mind that infant death is relatively rare in this country, and these figures are extrapolations of quite small numbers (of the 256 women with swine flu, five had stillbirth or neonatal death). Due to the small amount of data that was available for this study, and other limitations, these findings should be viewed with some caution.
Despite this need for further research however, influenza is a known risk in pregnancy, and these findings do support the recommendation for pregnant women to be vaccinated against seasonal flu.
Where did the story come from?
The study was carried out by researchers from the National Perinatal Epidemiology Unit at the University of Oxford in the UK. Funding was provided by the National Institute for Health Research Health Technology Assessment Programme. The study was published in the peer reviewed British Medical Journal .
The study was generally reported accurately in the papers. The Independent made a slight error in reporting that women who caught H1N1 were five times more likely to have a stillbirth, when in fact they were only four times more likely (the five times figure was for the perinatal mortality rate, which included stillbirths and neonatal deaths).
The Daily Mail carried a useful additional article explaining more about swine flu.
What kind of research was this?
This was a national cohort study set up to estimate the risk posed by swine flu (H1N1) to unborn and newborn babies. It followed 256 pregnant women who had a confirmed swine flu infection during the second wave of the pandemic between September 2009 and January 2010. The researchers compared this group’s rate of stillbirth, perinatal mortality (stillbirths plus deaths after 24 weeks of pregnancy and up to one week after delivery) and neonatal mortality (deaths up to one month after delivery) with the mortality rates in 1,220 uninfected pregnant women.
The researchers point out that studies of the effects of H1N1 on pregnancy have so far focused on the risk to the mother, rather than the offspring. They also say that it is uncommon to follow women up after their original admission to hospital, so the effect of swine flu infection on pregnancy has not been fully investigated. They add that some evidence from previous pandemics suggests that pregnancies after infection with flu are more likely to end in stillbirth or neonatal death.
What did the research involve?
The researchers identified women admitted to hospital with swine flu between September 2009 and January 2010 through the Obstetric Surveillance System (UKOSS) network. This is a national network of clinicians that is present in every UK hospital with an obstetrician-led maternity unit. For comparison, they used a group of uninfected women who had given birth in the UK before the start of the swine flu pandemic, between February 2005 and February 2006. Data on these women were also collected from the UKOSS network.
The researchers also used data that could be used as a national comparison, such as birth statistics from the ONS, data on births in Scotland, and data on the overall number of perinatal deaths in the UK, for 2008.
The researchers then looked at pregnancy outcomes in both groups. They calculated rates of stillbirths, perinatal mortality and neonatal mortality, for every 1,000 pregnancies. The data was analysed using standard statistical methods and the outcomes adjusted for possible confounders such as socioeconomic status, ethnic group, smoking, age and body mass index (BMI) and multiple pregnancy.
Where some data was missing they used validated statistical methods to reduce potential bias.
What were the basic results?
Of 223 UK hospitals in the UKOSS network, a network that all consultant led maternity units belong to, 221 participated in the study. A total of 272 pregnant women were admitted with H1N1 between September 2009 and January 2010, for whom there was available data on 256 women (94%). These women had 249 live births including five pairs of twins. Five pregnancies were lost or terminated before 24 weeks of pregnancy.
Below is a summary of the results:
- Perinatal mortality was over five times higher in babies born to infected women than in those born to uninfected women. In the swine flu group there were 10 deaths among 256 infants, which equates to a rate of 39 out of 1,000 total births (95% confidence interval [CI] 19 to 71). This compares to a rate of 7 out of 1,000 (95% CI 3 to 13) among babies of uninfected women.
- Mostly, the higher perinatal mortality rate among babies of infected women was due to a higher rate of stillbirth (27 out of 1,000 total births, compared to 6 out of 1,000).
- Infants of infected women were also more likely to be born prematurely than infants of healthy women (adjusted odds ratio 4.0, 95% confidence interval 2.7 to 5.9).
- Infected women who delivered preterm (before 37 weeks of pregnancy) were more likely to have been infected in their third trimester, to have been admitted to an intensive care unit, and to have secondary pneumonia, than were those who delivered at term.
How did the researchers interpret the results?
The researchers say their study suggested that the 2009 H1N1 infection during pregnancy is associated with a great risk of poor perinatal outcomes, a risk that persists after taking into account other characteristics known to be associated with poor outcomes.
The health of pregnant women is an important public health priority in future waves of influenza pandemics, they say.
This well conducted study highlights the possible risk to babies of maternal infection with H1N1. The research has some limitations, which mean the results should be interpreted with some caution.
- The researchers used a historical cohort of pregnant women to act as a comparison, some of whom had given birth during 2005-6. It is possible that pregnancy outcomes have changed between this time and the time the infected women gave birth. The researchers point out that this risk is somewhat mitigated by the fact that the national surveillance of perinatal mortality has not identified any changes that were likely to affect these findings.
- Although the researchers adjusted their analysis for confounders, it is possible that both measured and unmeasured confounders affected the results. There are other possible confounders that were not adjusted for, including whether the mother had previously delivered by caesarean, the quality of obstetric care and time between pregnancies.
- Because perinatal mortality is relatively rare, it was difficult for the researchers to adjust even for known confounders.
This study indicates that there may be a higher risk of perinatal mortality for women who contract swine flu. The strength of this connection will need further research however, as will studies to establish how it might have this effect, and whether or not swine flu risk can be reduced by other means.
Despite this need for further research however, influenza is a known risk in pregnancy, and these findings support the recommendation for pregnant women to be vaccinated. Pregnant women are among the high risk group advised to have an annual flu vaccination, which also protects against H1N1. H1N1 is predicted to be one of the main flu viruses circulating during the 2011/12 winter.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
BBC News, 20 October 2011
Daily Mail, 20 October 2011
The Independent, 20 October 2011
Links to the science
Sands press release, 2011
BMJ 2011; 342: d3214
BMJ 2011; 342: d3237