Obstetrics and Gynecology
Influenza in Pregnancy
- 1. What every clinician should know
2. Diagnosis and differential diagnosis
- 3. Management
4. Prognosis and outcome
5. What is the evidence for specific management and treatment recommendations
1. What every clinician should know
Influenza is a common infectious disease caused by a family of RNA viruses (orthomyxoviruses) capable of infecting a wide variety of bird and mammalian species. In humans, characteristic symptoms include fever, sore throat, myalgias, chills, cough, headache, fatigue and malaise.
Influenza viruses are transmitted primarily by aerosols created by coughing or sneezing but can be transmitted by direct contact. The viruses are inactivated by detergents, disinfectants and sunlight.
There are three species of influenza viruses, labeled as influenza A, B and C, with the majority of human disease caused by influenza A. They contain various glycoproteins on their surface, primarily hemagglutinins (H) and neuraminidases (N). The specific combination of H and N glycoproteins are used to sub-type the viruses. There are 16 H and 9 N sub-types currently known, although H 1 through 3 and N 1 and 2 are responsible for the majority of human disease.
Because these viruses can exist in multiple species and because there is constant exchange of genetic material between viruses, viral genetic material from a virus infecting one animal species can be transferred to viruses that infect other species. The most common sources of new genetic material for human infections are birds and pigs. For example, the new influenza strain that was responsible for the 2009-2010 H1N1 influenza A pandemic was a combination of human, bird and pig influenza genes.
Increased susceptibility of pregnancy women
Because of the associated physiologic changes of pregnancy, pregnant women have been recognized for a century to be at increased risk of mortality and severe morbidity from concurrent influenza infection. The increases in minute volume and cardiac output, both of which significantly tax the cardiopulmonary reserve of the third trimester, as well as a shift away from T-cell-mediated immune responsiveness, are in large part responsible for this association.
These changes have largely returned to normal within a couple of weeks after delivery, but it’s important to remember that immediately postpartum pregnant women can be severely affected as well, particularly if they have had multifetal pregnancies or other pregnancy complications.
Pregnant women, particularly third trimester pregnant women, are more likely to be hospitalized, and also admitted to intensive care units when compared to non-pregnant women of similar age. A recent meta-analysis from the 2009-2010 H1N1 influenza A pandemic demonstrated that although approximately 1% of the population is pregnant at any time in the United states, pregnant women accounted for a disproportionate number of hospitalizations (835/13,322, or 6.3%), ICU admissions (236/3,989, or 5.9%) and deaths (188/3,295, or 5.7%).
To be sure, some of these reports focus on hospitalized cases and provide relatively little information about less severe cases managed as outpatients. While some might also argue that the increase in hospitalizations is out of concern for fetal well-being, the substantial increase in the percentage of deaths among pregnant women with influenza speaks strongly to the real significance of this problem.
The absolute best treatment for influenza in pregnancy is prevention via vaccination. In fact, immunization during pregnancy is one of the Healthy People 2020 objectives, aiming for an 80% overall uptake. Increasing influenza vaccine uptake by pregnant women should be a priority for all women’s health providers.
Unfortunately, this is not practiced nearly widely enough, despite the collective recommendations of the Centers for Disease Control and Prevention, American Congress of Obstetricians and Gynecologists, American Academy of Family Practice, American Academy of Pediatrics, American College of Nurse-Midwives, Association of Women’s Health, Obstetric and Neonatal Nurses, Infectious Disease Society for Obstetrics and Gynecology, the March of Dimes and the Society for Maternal-Fetal Medicine.
Despite these recommendations, the uptake of influenza vaccination by pregnant women remains alarmingly low. The recent 2009-2010 H1N1 pandemic did increase the uptake of influenza vaccine in pregnant women in the U.S. from a historic average of less than 15% to approximately 40%. While this is certainly an improvement, it leaves much to be desired, with the majority of pregnant women still not receiving appropriate vaccinations.
A substantial contributor to this public health problem is the collective amnesia and/or reticence of obstetric care providers to recommend (or better yet, advocate on behalf of) immunization. It is thus important that obstetric care providers and their staff educate pregnant women about the potential impact of influenza infection during pregnancy and review the signs and symptoms for which they should promptly report for evaluation.
Nonetheless, several studies in North America, Europe and Australia/New Zealand suggest that a substantial percentage of pregnant women continue to decline influenza vaccinations because of safety concerns. In all these geographic areas approximately 98% of pregnant women have at least one prenatal visit and this should be an opportunity both for education and for vaccination.
In addition, the majority of obstetric care practices now have some form of website (clinic, hospital or both). These portals are widely accessed by pregnant women and should include the aforementioned recommendations.
There are, however, numerous barriers to optimal immunization during pregnancy. Most obstetric offices do not have the infrastructure (e.g. proper storage refrigerator and tracking of administration documentation as required by law) to administer vaccines, and obstetric providers often do not view vaccination as their primary responsibility.
In addition, there is no good central vaccine record keeping system for adult vaccination (e.g. to avoid the problem of repeat vaccination) and there are financial disincentives related to inadequate reimbursements. The ever-present medico-legal concerns also serve as a disincentive for some providers.
Because the vaccines are produced in chicken eggs, women who have known or suspected allergies to egg products should not be vaccinated. Having said all this, however, there are several strategies that have increased vaccination compliance for pregnant women. Among them are standing orders and provider assessment and feedback systems.
It is also important to recognize the reality of viral antigenic drift. Viruses are mutating all the time and what we think will be the optimum strains for immunization may not actually turn out to be the strains that cause the most illness.
Finally, it is critically important that all members of the health care team who have any patient contact be vaccinated as well to ensure that health care workers do not become inadvertent disease vectors.
Pregnant women are probably no more likely to contract influenza than non-pregnant women, but they are at increased risk for more severe manifestations of influenza. Pregnant women need to be vaccinated against current circulating influenza strains!!
Likewise, although we recommend universal influenza vaccine during pregnancy, influenza vaccines are not 100% effective. Thus, a history of influenza vaccination should not preclude clinical suspicion of influenza infection. In addition to antigenic drift, infection either just before, or within 10-14 days after, vaccination can still result in clinical infection.
It is important to remember that, although substantial attention was directed to the 2009-2010 H1N1 influenza A pandemic, seasonal influenza is so named because it predictably occurs during influenza season every year. Influenza season in the Northern hemisphere is October to May (and April to November in the Southern hemisphere). Pregnant women are at increased risk every year whether or not there is an influenza pandemic in progress.
Clinical features and incidence
Influenza symptoms are frequently sudden in onset and characteristically appear within 1-2 days after infections. Fever (frequently 39.0 oC or higher) and chills are generally the first symptoms to appear and infected individuals are frequently quite ill, to the point of being bedridden, within a few hours. This degree of prostration is frequently helpful in distinguishing influenza from the common cold.
Other common symptoms include nasal congestion, sore throat, myalgias, cough, headache, fatigue and malaise. Fever, cough and nasal congestion are the most common symptoms in adults. Gastrointestinal complaints are not common in infected adults but are more common in children. Individuals are capable of shedding the virus beginning the day before symptoms and continuing for 5-7 days thereafter.
Because the effectiveness of antiviral drugs (see neuraminidase inhibitors, below) is dependent on early initiation of treatment, pregnant women during influenza season exhibiting these symptoms should be identified early and considered for initiation of treatment.
Pregnant women who have other concurrent medical problems are at even greater risk for serious illness (and should be high-priority candidates for influenza immunization! - see above). The most commonly reported concurrent medical problems are tobacco abuse, asthma, diabetes, cardiac disease, hypertension and/or obesity. Women with other conditions compromising their immune system, such as transplant patients or women with advanced HIV infections, also are at increased risk for more serious illness. Obviously, women with more than one of these risk factors should be considered at correspondingly increased risk.
Previously healthy young adults have a characteristically more robust immune response than do young children or the elderly. Seasonal influenza is thus characteristically a more serious disease amongst the very young (under 6 months) and people aged 65 and older. However, when triggered by a new antigen, such as the H1N1 Influenza A virus, this robust response in young adults can lead to what has come to be called a “cytokine storm.”
The primary portal of entry of influenza viruses is the respiratory tract and the viruses are only able to invade cells when their hemagglutinin protein is cleaved by host protease enzymes. Because protease enzymes differ between genera and species, the hemagglutinin protein characteristically defines which species the virus can infect. Viruses whose hemagglutinin protein can only be cleaved by proteases found in the throat and upper respiratory tract characteristically result in clinically mild infections whereas those whose hemagglutinins can be cleaved by a wide variety of proteases are able to spread throughout the body.
In addition, hemagglutinins that bind only to the upper respiratory tract tend to be more easily transmissible but milder in clinical symptoms, whereas those that bind to tissues deeper in the lungs tend to be associated with pneumonia but are less contagious. An example of the latter strain is the H5N1 influenza A virus responsible for the 2004 “bird flu.”
2. Diagnosis and differential diagnosis
Pregnant women suspected of having an influenza infection should not be seen during regular clinic hours for fear of transmitting the infection to others. Arrangements should be made for either visits to alternative clinics or during alternative hours.
It is important to remember that the rapid diagnostic influenza tests have very poor sensitivity and should not be the basis for diagnosing or excluding influenza. The real-time reverse transcriptase-polymerase chain reaction test is an ideal test for making the diagnosis. It should be minimally affected by recent antiviral therapy, as it does not require replicating virus for diagnosis.
Women with any suggestion of pneumonia or septic shock should be admitted to a hospital for further evaluation and treatment. Bacterial cultures of blood, sputum and urine should be started and appropriate antibiotic coverage initiated pending culture results. In addition to a chest x-ray (see following section), a complete blood count, electrolytes and serum chemistries should be obtained as a baseline for subsequent comparison.
Pregnant women with any clinical suggestion of pneumonia or septic shock should have a chest x-ray performed. Chest x-ray findings with influenza pneumonia are variable and at least in part reflect the invasive characteristic of the specific viral subtypes (see Pathophysiology, below). Chest x-ray findings may be absent but often include some degree of interstitial prominence, patch consolidation or alveolar hemorrhage (classically as small centribolular nodules).
Cavitation and/or pleural effusions are not characteristic findings of influenza pneumonia and should suggest bacterial super-infection. Historically staphylococcal super-infections have been particularly likely to be associated with these findings.
Remember as well that pregnant women often appear rather healthier than they really are. Previously healthy young adults often think that they can “tough it out” (although fortunately women are not as prone to this as are men). It is particularly important to assess maternal oxygen saturation in the setting of suspected influenza infection and to monitor it serially.
Any pulse oximetry value below 95% (roughly equivalent to a PaO 2 of 70 mm Hg) requires serial in-hospital observation. Previously healthy pregnant women with acute influenza infections can crash and burn in a hurry and you don’t want to miss it!
Particularly during influenza season, the diagnosis of influenza, when accompanied by a clinically consistent exposure and sudden onset of characteristic symptoms, is straightforward. However, bacterial, chlamydial or mycoplasma pneumonia can present similarly, so cultures for these infections should be considered, particularly if there is no response to antiviral medications.
Every study that looked at separate outcomes for pregnant women in the recent H1N1 pandemic who were started on neuraminidase inhibitors within 48 hours of symptom onset found improved maternal outcomes. Even starting within 3-4 days after the onset of symptoms has been shown to be of some value. There have been to date no reports that outcomes are equivalent with or without neuraminidase inhibitors and there are likewise no reports of worsened outcomes.
If you are seeing a pregnant woman with suspected influenza, you should be aggressive in starting neuraminidase inhibitors. I would emphasize that this should be started as soon as the diagnosis is suspected, particularly if she has any additional risk factors. Do not wait for confirmatory testing!
Neuraminidase inhibitors (oseltamivir, or Tamiflu; zanamivir, or Relenza)
In contrast to the adamantanes, essentially all the H1N1 and H3N2 influenza A as well as influenza B reported in recent influenza seasons have been sensitive to the neuraminidase inhibitors. These drugs interfere with replication of influenza virus in the respiratory tract (a process that reaches its peak 24-72 hours after onset of symptoms), so it is important that these drugs be administered as quickly as possible. These medications have very few side effects (usually gastrointestinal) and are much less likely to stimulate drug resistance.
Conflicting data exist on whether or not pregnancy alters the pharmacokinetics of oseltamivir, with some reports suggesting that pregnancy results in increased clearance and others suggesting that clearance is not affected by pregnancy. However, it is known that oseltamivir has a wide margin of safety when administered in substantially higher doses than recommended.
Information on the effects of influenza antiviral medications on the fetus is limited, although there remains no suggestion of harm. Given the clear risks associated with maternal influenza infection during pregnancy and the absence of obvious risks with antiviral medications, the CDC recommends that pregnant women with suspected influenza be given appropriate influenza medications. At the time of this writing (2014) oseltamivir is generally recommended but updated information can be obtained at www.cdc.gov/flu.
Adamantanes (amantadine, rimantadine)
Essentially all the H1N1 and H3N2 influenza A as well as influenza B reported in recent influenza seasons have been resistant to the adamantanes. These medications are used infrequently. They are only effective against influenza A, tend to rapidly induce viral resistance and have some potentially serious side effects (in particular, serious allergic reactions and personality changes).
Maternal fever in early pregnancy has been linked to birth defects and late in pregnancy it can substantially increase fetal oxygen requirements (about 10% per degree Fahrenheit of maternal temperature elevation). For both reasons it is appropriate to treat influenza-associated fever during pregnancy. Acetaminophen, rather than aspirin or non-steroidal anti-inflammatory drugs, is the preferred treatment.
Although the increased oxygen affinity of fetal hemoglobin favors continued adequate fetal oxygenation, significant maternal hypoxemia can still lead to fetal compromise. Maternal PaO 2 levels should be kept above 70 mm Hg, roughly corresponding to an oxygen saturation of 95%.
Adequate hydration and symptomatic rest may assist in recovery and will certainly keep pregnant women more comfortable.
Extracorporeal Membrane Oxygenation (ECMO)
ECMO has been reported to be of value in severely ill pregnant women, although at least a half-dozen of the now several dozen reported cases have died either during pregnancy or afterward.
Will Delivery Improve Outcomes?
Delivery can reduce the metabolic and mechanical burden of severe disease by removing fetal oxygen consumption and decreasing intra-abdominal pressure. A number of reports describe iatrogenic preterm birth, frequently by cesarean delivery, in the setting of severe maternal disease. There are no controlled trial data to guide us on this issue.
Ideally, women with an acute influenza infection should be delivered in a separate area away from other laboring women. During pandemics, information about such arrangements should be widely publicized. Postpartum women should also be cared for in areas removed from other pregnant and puerperal women.
What About Treating Associated Preterm Labor?
Spontaneous preterm birth also is more common in the setting of maternal influenza infection. This raises the common sense question of whether we would, or would not, be smarter than human biology to try to tocolyse such women. I would contend that, except in the most unusual of circumstances, we should not attempt to stop spontaneous preterm labor in pregnant women who are seriously ill.
4. Prognosis and outcome
Although being pregnant does not increase the likelihood of a woman acquiring influenza, it does increase by four- to five-fold her risk of developing a serious infection that could require hospitalization.
Most pregnant women who contract influenza will be clinically ill for 1-2 weeks but can be expected to recover completely. However, both patients and their families should be educated to watch carefully for any suggestion of respiratory de-compensation, particularly for the first several days of the illness.
As noted above, we recommend universal influenza vaccination for all pregnant women, irrespective of trimester. While this can effectively prevent influenza illness in pregnant women (by inducing protective concentrations of hemagglutinin inhibition antibodies), it also provides clear protective benefits to newborns through the first six months of life. This is particularly important because infants through their first six months of extrauterine life are at higher risk for severe morbidity and/or mortality should they acquire an influenza infection and are unable to produce an adequate immune response after influenza vaccinations, and therefore are ineligible for influenza vaccinations in the United States.
If an affected pregnancy results in a stillbirth, the placenta, cord, and fetal tissues should be collected for viral culture and examination by reverse transcriptase-polymerase chain reaction and compared to maternal respiratory specimens.
Vertical transmission may have occurred in a handful of cases but it is exceedingly rare and there is no evidence that congenital or neonatal influenza has ever been acquired from influenza vaccination. The major problems for newborns of influenza-infected women are the sequelae of prematurity. An asymptomatic infant born to a mother who is either symptomatic or who has recovered from an influenza illness during pregnancy does not require special treatment but should be monitored closely for evidence of evolving neonatal influenza.
This generally involves separating the mother (if she is symptomatic) from her baby, although the duration of isolation is uncertain. During the 2009-2010 H1N1 Influenza A pandemic, the CDC recommended separation until the mother had received antiviral treatment for at least 48 hours, was afebrile without antipyretics for at least 24 hours, and was able to control her cough and respiratory secretions. Beyond that, influenza precautions were recommended for a total of 7 days.
Before a puerperal woman and her newborn are discharged home, influenza immunization/vaccination status of all household contacts (at least 6 months old) should be evaluated and anyone who is not immune (particularly during influenza season) should be encouraged to get vaccinated.
Although there is no evidence of harm, there are very few data on antiviral medication levels in breast milk. However, as with vaccination and treatment during pregnancy, the increased risks associated with maternal influenza infection during the first few weeks postpartum and the absence of obvious risks with antiviral medications suggest that women with suspected influenza who are within 2 weeks of delivery and who are breastfeeding should be given appropriate influenza medications.
If a breast feeding woman who is more than 2 weeks postpartum develops clinically apparent influenza, she should also be started on influenza medications and encouraged to continue breast pumping. The baby should be fed the breast milk by an uninfected caregiver until the CDC clinical criteria described above have been fulfilled.
5. What is the evidence for specific management and treatment recommendations
Mosby, LG, Rasmussen, SA, Jamieson, DJ. "2009 pandemic influenza A (H1N1) in pregnancy: a systematic review of the literature". Am J Obstet Gynecol. vol. 205. 2011 Jul/. pp. 10-8.
Beigi, RH, Han, K, Venkataramanan, R. "Pharmacokinetics of oseltamivir among pregnant and nonpregnant women". Am J Obstet Gynecol. vol. 204. 2011. pp. S84-S88.
Greer, LG, Leff, RD, Rogers, VL. "Pharmacokinetics of oseltamivir according to trimester of pregnancy". Am J Obstet Gynecol. vol. 204. 2011. pp. S89-S93.
Fiore, AE, Uyeki, TM, Broder, K. "Prevention and control of influenza with vaccines: recommendations of the Advisory Committee on Immunization Practices (ACIP), 2010". MMWR Recomm Rep. vol. 59. 2010. pp. 1-62.
Haberg, SE, Trogstad, L, Gunnes, N. "Risk of Fetal death after pandemic influenza virus or infection or vaccination". N Engl J Med. vol. 368. 2013. pp. 333-340.(Good recent update on impact of influenza on the fetus as well as the lack of fetal impact from vaccination [outcomes clearly improved].)
Thompson, MG, Li, DK, Shifflett, P. "Effectiveness of seasonal trivalent influenza vaccine for preventing influenza virus illness among pregnant women: a population-based case-control study during the 2010-2011 and 2011-2012 influenza seasons". Clin Infect Dis. vol. 58. 2014. pp. 449-57.(Demonstrates equivalent efficacy of influenza vaccine in pregnant vs non-pregnant adults.)
Copyright © 2017, 2014 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.