To summarize the literature regarding 2009 H1N1 influenza A during pregnancy, we conducted a systematic literature review using a PubMed search and other strategies. Studies were included if they reported 2009 H1N1 influenza in pregnant women as original data. In all, 2153 abstracts were reviewed, and a total of 120 studies were included. Data were extracted regarding number of cases, additional risk factors for influenza-associated complications, treatment, and maternal and pregnancy outcomes. Authors were contacted to determine the extent of overlap when it was suspected. Pregnancy was associated with increased risk of hospital and intensive care unit admission and of death. Pregnant women who received delayed treatment with neuraminidase inhibitors or who had additional risk factors were more likely to develop severe disease. Preterm and emergency cesarean deliveries were frequently reported. These results reinforce the importance of early identification and treatment of suspected influenza in this high-risk population.
On April 15 and 17, 2009, a novel 2009 H1N1 influenza A (2009 H1N1) virus was identified in 2 children in the United States by the Centers for Disease Control and Prevention (CDC). The virus spread rapidly throughout the United States and worldwide, leading the World Health Organization to declare a pandemic on June 11, 2009. As predicted based on experience with seasonal influenza and previous pandemics, pregnant women were quickly identified as a high-risk group for severe complications of 2009 H1N1 influenza, including hospital and intensive care unit (ICU) admissions and death.
Since April 2009, a large number of publications have appeared describing the effects of 2009 H1N1 influenza among pregnant women; however, many of these publications describe small case series and often overlap with other published studies. Therefore, a systematic review of the currently available literature, with an attempt to minimize overlap, is needed to better understand the effects of 2009 H1N1 on the pregnant woman and her fetus to guide optimal management. These data will be helpful to plan for future influenza seasons, during which 2009 H1N1 and seasonal influenza strains are expected to circulate, as well as to plan for future influenza pandemics.
Materials and Methods
A PubMed search was performed (L.G.M.) on May 6, 2010, and Aug. 2, 2010, to identify all publications in the medical literature discussing 2009 H1N1 influenza in pregnancy using the following search terms: (“H1N1” [TiAb] OR “swine” [MeSH terms] OR “swine” [tiab]) AND (“influenza, human” [MeSH terms] OR “influenza” [tiab]) AND (“pregnancy” [MeSH terms] OR “pregnancy” [tiab]). A second PubMed search was performed (L.G.M.) on May 6, 2010, using the same terms as above excluding: (“pregnancy” [MeSH terms] OR “pregnancy” [tiab]). We limited all searches to the English language and did not attempt to identify unpublished articles. The abstracts of all identified papers were examined. When the abstract included clinical or surveillance information, or when no abstract was available, the full paper was examined to determine whether cases of pregnant women with 2009 H1N1 infection were described. Papers that included cases reported as original data (ie, without citing another study as the source of data) were included. We included case-control studies, cohort studies, case reports, and letters and editorials in medical journals that described cases. Policy and practice articles were not included. Additionally, we examined the references of all identified articles for further relevant resources, and we hand-searched the personal files of 2 subject matter experts for the CDC (S.A.R. and D.J.J.) who have authored publications regarding 2009 H1N1 and pregnancy issues.
We contacted authors when possible to determine extent of overlap between studies. If no overlap could be established, we assumed nonoverlap. When overlap was established and cases were described in >1 study (eg, as part of a large report and as a detailed case report), both studies were included in the review, but the pregnant woman was counted toward total cases, hospitalizations, ICU admissions, and deaths only once.
After accounting for overlap, the proportion of pregnant women among total hospitalizations was calculated by identifying studies that included numbers of hospitalized pregnant women and of total hospitalizations, and summing the total pregnant women hospitalized divided by the total hospitalized in the general population. Proportions of pregnant women admitted to the ICU and who died were calculated in the same manner. Unless otherwise specified, we assumed pregnant women who died had also been admitted to the ICU. When calculating percentages of affected women with additional risk factors for influenza-related complications identified by the Advisory Committee on Immunization Practices (ACIP), only studies with >10 pregnant patients were included in an attempt to avoid bias of data toward more complex patients.
Results
The PubMed search that included “pregnant OR pregnancy” in the search terms yielded 151 publications, 106 of which were published on or after April 1, 2009, 93 of which were English-language papers. Among these 93, 42 provided clinical information on pregnant women with confirmed, probable, or suspected 2009 H1N1 and were thus included in our review. This search was repeated on Aug. 2, 2010, and yielded an additional 26 papers, of which 14 included cases of pregnant women affected by 2009 H1N1.
The PubMed search that excluded the pregnancy terms yielded 2494 publications, 2032 of which were published in English, on or after April 1, 2009. Of these, 116 provided clinical information on pregnant women with 2009 H1N1 influenza and thus were included in our review. All 42 studies from the initial pregnancy-specific PubMed search were included in this collection of 116. One paper was unavailable through CDC and Emory University library systems as of Aug. 10, 2010, and so was not included in this review.
Review of personal collections of papers of 2 CDC subject matter experts (D.J.J. and S.A.R.) yielded an additional 2 papers to be included in the review. Review of reference lists of included publications yielded no new studies fitting our criteria.
Overall, 132 papers reported clinical information on pregnant women with 2009 H1N1 ( Figure ). Of these, 12 papers were excluded because they did not meet our eligibility criteria: one was published as a policy and practice paper, and other papers reported only cases and information described in previous studies.
All papers were evaluated for overlap if they reported cases from the same geographic location during the same time period. Overlap was suspected and was found to varying degrees in reports from the following regions: Argentina, Australia, Brazil, Canada, Chile, China, France, Hong Kong, Iran, Israel, Mexico, New Zealand, Reunion Island, Saudi Arabia, Singapore, South Africa, United Kingdom, and United States. The 120 papers that were included in the review reported 3110 pregnant women from 29 countries with 2009 H1N1 influenza infection, including 1625 (52.3%) who were hospitalized with 2009 H1N1, of whom 378 (23.3%) were admitted to an ICU and 130 (8%) died.
Hospitalization and severe disease in pregnancy
Several studies demonstrated that pregnant women, when compared to nonpregnant women of similar age, or when compared to the general population, have an increased risk of hospitalization, ICU admission, death, and other severe outcomes due to 2009 H1N1 infection ( Table 1 ). Pregnancy was found to be significantly associated with hospitalization ( P = .015). A study from California identified an association between pregnancy and ICU admission that was not statistically significant ( P = .07). One study of severely ill patients found that pregnant women were more likely to develop acute renal failure ( P < .05).
