Pregnancy and Pandemic Flu

For most women, pregnancy is a time of joy, tinged with the excitement and expectation of a new family member. However, the pandemic of H1N1 influenza A 2009 (H1N1/09) (swine flu) changed all that. Fear gripped pregnant women the world over as it became apparent, early after the emergence of the novel H1N1 virus, that some infected women became ill very rapidly. Some developed severe complications and required intensive care. Some of the women and some of their newborns died [1, 2].

But should we have been surprised? Pregnancy causes immunologic and physiologic changes, and both are likely contributors to increased susceptibility to in fection and its complications. It has previously been recognized that pregnant women are more likely to experience complications, especially severe pneumonia, when infected with seasonal influenza. Furthermore, historical data from previous pandemics (in 1918 and 1957) indicate an excess of deaths in pregnancy and an increased risk of premature delivery and stillbirth [1].

The H1N1/09 pandemic hit Australia suddenly in May 2009, brought by travelers to Melbourne from Mexico, and spread quickly to other states. Between the declaration of the pandemic by the World Health Organization on 11 June 2009 and the end of September 2009, there had been >36,000 laboratory-confirmed cases, a likely underestimate because testing was not universal, and 180 deaths in Australia.

The article by Hewagama et al [3], published in this issue of Clinical Infectious Diseases, provides unique, detailed data on the clinical characteristics and outcomes of pregnant Australian women admitted to hospital with pandemic H1N1/09 from 20 May 2009 through 31 July 2009. Although it is a descriptive, partly retrospective case series that covers only part of the influenza season, the data are timely and of interest and relevance internationally because lessons learned from the pandemic in Australia have the opportunity to influence the management and outcomes of influenza in the upcoming winter season in the northern hemisphere. Claimed to be the largest series of pregnant women hospitalized with H1N1 to date, this series includes 43 pregnant women admitted to 6 hospitals in the state of Victoria in southeast Australia with laboratory-confirmed H1N1/09. Although the sample is small, these 6 hospitals account for more than one-third of deliveries in Victoria and include all 3 hospitals that provide tertiary-level maternity and neonatal services for the state, so data are representative and likely to be relevant statewide. It is not clear, however, whether the findings will be applicable nationally, where there are ∼300,000 births each year, and few data are provided about the women to enable the reader to determine how similar (or otherwise) they are to women outside Victoria. Limited historical data are provided to allow comparison of outcomes after the H1N1 and seasonal flu seasons.

Of the women admitted to hospital with H1N1 influenza, almost all were in their second or third trimesters of pregnancy, as has previously been described [3]. Most had typical flulike symptoms, and interestingly 50% had no underlying risk factor apart from pregnancy. However, more than one-third delivered during the hospitalization (40% before term, or 4 times the background rate of preterm labor) and 19% required intensive care, with 4 requiring ventilation support and 2 requiring extracorporeal membrane oxygenation. There was 1 maternal death, 2 deaths in utero, and 1 neonatal death. In their hospital-based population of >6000 pregnant women, the authors calculated a risk of ∼1 per 500 for hospitalization in the second trimester and ∼1 per 200 in the third trimester during the 3-month study period. There is potential for selection bias toward more severe cases and overestimation of risk in this sample, which draws from all 3 of the state's tertiary maternity hospitals, and additional data are required to accurately calculate risk.

So what lessons can clinicians learn from the study by Hewagama et al [3]? The overall rate of hospital admission due to H1N1 is higher in pregnant women than in the general population. Many women who require hospitalization because of influenza will have no underlying risk in addition to pregnancy; however, asthma, obesity, and diabetes mellitus may confer additional risk of complications, and such women require close monitoring in the community. Although the presentation of influenza in pregnancy is usually typical, it is often delayed, as is initiation of antiviral therapy, and women may deteriorate quickly. Therefore, any pregnant woman presenting with a flulike illness should be screened immediately for influenza. However, a clinical suspicion of the diagnosis is sufficient to warrant initiation of antiviral therapy. Clinicians should not delay effective treatment until a definitive diagnosis is confirmed. This is particularly pertinent because some health care facilities do not have access to rapid diagnostic testing (which may also have high false-negative rates), and in others viral subtyping may take days to weeks. Importantly, treatment must be given immediately because it is more likely to prevent complications if given within 24 h of onset of symptoms [1, 3, 4]. In the study reported by Hewagama et al. [3], more than two-thirds of women did not receive treatment until ⩾3 days after symptom onset.

The antivirals oseltamivir and zanamivir are effective against H1N1/09, and both may be used in pregnancy [1, 4, 5]. Some guidelines favor oseltamivir because of its systemic action and availability of more safety data. Conversely, some recommend zanamivir because its topical administration means little is absorbed. In some countries, antivirals are provided free to pregnant women. Although not subjected to formal trials in pregnancy, these antivirals have been widely used in the second and third trimesters without proven adverse effects to the mother or teratogenic effects to the unborn child. Their use is justified on the basis that the potential benefit to the mother exceeds the potential risk to the fetus [1, 4, 5]. The neuraminidase inhibitor peramivir has recently been given emergency use authorization by the US Food and Drug Administration (23 October 2009) [6]. Approval is for intravenous use in selected patients if there is failure of response to oral or inhaled therapy, if drug delivery by these routes is not feasible or likely unreliable, or if intravenous therapy is deemed optimal. Pregnancy is not a contraindication to use of peramivir, but no adequate trials have been conducted and limited safety data, including on nephrotoxicity, are available.

Prevention of infection in pregnancy is the key to our future public health response to H1N1 and is now attainable. Vaccination programs against H1N1/09 are being introduced worldwide, and most countries have prioritized pregnant women. Unfortunately, vaccination rates against seasonal influenza have been unacceptably low in the past and we must do better. For example, despite Australian guidelines recommending vaccination of all women planning pregnancy and of pregnant women who will be in their second and/or third trimester during the influenza season, only 1 of the pregnant women included in the study by Hewagama et al [3] had been vaccinated against seasonal influenza. Good uptake of the H1N1 vaccine will likely require a 2-pronged attack to address both lack of health care professional awareness of the risks of influenza in pregnancy and the need to vaccinate and the reluctance of pregnant women to be vaccinated.

Data on the effect of maternal H1N1 influenza on fetal and newborn outcomes are still lacking. The northern hemisphere winter in 2009–2010 will provide an opportunity for careful collection and analyses of such data in large series of pregnant women. In the meantime, pregnant women should be vaccinated and new mothers should be urged to breastfeed, to pay careful attention to personal hygiene, and to avoid exposing their infant to crowds or symptomatic individuals. Availability of the H1N1 vaccine for children varies considerably worldwide, but in most countries vaccination is recommended for all children aged >6 months.

• Received November 23, 2009.
• Accepted November 24, 2009.

• © 2010 by the Infectious Diseases Society of America