Perinatal and Maternal Outcomes in Critically ill Obstetrics Patients With Pandemic H1N1 Influenza A

Transcription

1 OBSTETRICS Perinatal and Maternal Outcomes in Critically ill Obstetrics Patients With Pandemic H1N1 Influenza A Titilayo Oluyomi-Obi, MBBS, FRCSC, 1 Lisa Avery, MD, FRCSC, 1,2,3 Carol Schneider, MD, FRCSC, 1 Anand Kumar, MD, 4 Stephen Lapinsky, MD, 5 Savas Menticoglou, MD, FRCSC, 1 Ryan Zarychanski, MD, 4,6 1 Department of Obstetrics, Gynaecology and Reproductive Sciences, University of Manitoba, Winnipeg MB 2 Centre for Global Pubic Health, Department of Community Health Sciences, University of Manitoba, Winnipeg MB 3 Department of Medical Microbiology and Infectious Diseases, University of Manitoba, Winnipeg MB 4 Department of Internal Medicine, University of Manitoba, Winnipeg MB 5 Department of Internal Medicine, University of Toronto, Toronto ON 6 Department of Haematology and Medical Oncology, Cancer Care Manitoba, Winnipeg MB Abstract Background: During the influenza pandemic of spring 2009, Manitoba had a disproportionate number of pregnant women who became critically ill. Information about these cases will be useful to help us understand the potential impact of future outbreaks and review critical illness in pregnancy. Methods: We describe the clinical details of six critically ill pregnant women with pandemic H1N influenza virus admitted to two ICUs in Manitoba between March 1 and August 31, Results: Thirty adult pregnant women tested positive for pandemic H1N virus in Manitoba. Six women were admitted to the ICU. The time from onset of symptoms to life-threatening deterioration was on average five days. Most patients presented with worsening fever and cough and had H1N1-positive contacts. Five of six patients (83%) were Aboriginal. Four of six cases occurred in the third trimester. These patients frequently required non-conventional ventilatory support such as high frequency ventilation or extracorporeal membrane oxygenation (ECMO). All patients received oseltamivir. Two patients died while in the ICU. Three of six patients had adverse perinatal outcomes, and there was one spontaneous abortion and one early preterm delivery. Conclusion: Pandemic H1N influenza virus has the potential to cause severe illness in pregnant patients. Those patients requiring ICU admission for respiratory support have a high risk for poor fetal and neonatal outcome. The experience of this cohort underscores the appropriateness of public health measures directed at prevention and early treatment of H1N1 infection in pregnancy. Key Words: Pregnancy, pandemic swine origin H1N1 influenza A, pandemic H1N1 influenza Competing Interests: None declared. Received on October 26, 2009 Accepted on December 14, 2009 Résumé Contexte : Pendant la pandémie de grippe du printemps 2009, le Manitoba comptait un nombre disproportionné de femmes enceintes qui ont été gravement atteintes. Les renseignements au sujet de ces cas s avéreront utiles pour nous aider à comprendre les effets potentiels de futures éclosions et à analyser les maladies graves durant la grossesse. Méthodes : Nous décrivons les détails cliniques de six femmes enceintes gravement atteintes par le virus pandémique de la grippe H1N qui ont été admises dans deux unités de soins intensifs au Manitoba entre le 1 er mars et le 31 août Résultats : Trente adultes enceintes se sont avérées séropositives pour ce qui est du virus pandémique de la grippe H1N au Manitoba. Six femmes ont été admises à l USI. Le délai entre l apparition des symptômes et la détérioration constituant un danger de mort était de cinq jours en moyenne. La plupart des patientes ont présenté une exacerbation de fièvre et de toux et comptaient cinq personnes séropositives pour ce qui est du virus H1N1 dans leur entourage. Cinq de ces six patientes (83 %) étaient autochtones. Quatre de ces six cas sont survenus au cours du troisième trimestre. Ces patientes ont souvent nécessité une ventilation assistée non conventionnelle telle qu une ventilation à haute fréquence ou une oxygénation extracorporelle (ECMO). Toutes les patientes ont reçu de l oseltamivir. Deux patientes sont décédées à l USI. Trois des six patientes ont connu des issues périnatales indésirables; il y a également eu un cas d avortement spontané et un cas d accouchement très prématuré. Conclusion : Le virus pandémique de la grippe H1N présente le potentiel de causer une grave maladie chez les patientes enceintes. Les patientes qui nécessitent une admission à l USI aux fins d une ventilation assistée courent un risque élevé de connaître de piètres issues fœtales et néonatales. L expérience de cette cohorte souligne la nécessité de disposer de mesures de santé publique visant la prévention et la prise en charge précoce de l infection au H1N1 pendant la grossesse. J Obstet Gynaecol Can 2010;32(5): MAY JOGC MAI

2 INTRODUCTION Over the past two centuries there have been several influenza pandemics (1847 to 1848, 1889 to 1890, 1918 to 1919, and 1957 to 1958), and the influenza subtype A virus remains one of the most common to infect humans. 1 In 1919, Bland et al. 2 showed that the mortality rate in pregnant women with influenza was 30% to 50%. 2,3 Similarly, Harris noted a mortality rate of more than 50% in pregnant women in the third trimester who developed pneumonia with the influenza A virus during the pandemic of The effect on fetal outcome remained less clear, with some studies demonstrating possible increased morbidity and mortality and others demonstrating negligible effect. 4,5 Since its initial outbreak in Mexico, the novel pandemic H1N virus, a new strain of influenza A virus that originated in swine, has quickly spread to many countries including Canada. As of August 1, 2009 there had been 6904 cases of pandemic H1N1 influenza in Canada, with 0.8% (56) being pregnant women. Rates of hospitalization in Canada were higher among pregnant women (85.7%, 48/56) than in the general population (18.6%, 1282/6904). Of the 56 hospitalized pregnant women in Canada, 10 (17.8%) were admitted to ICU, and of these, two (20%) died. 6 As the world readies itself for future waves of novel pandemic H1N infection, information about the presentation, clinical course, and outcomes of patient groups infected with H1N1 becomes critical for pandemic planning. Jaimeson et al. 7 were the first to publish their experience in managing severe novel H1N1 infection in pregnant women. We describe here our experience during the same time frame in the province of Manitoba, which had the largest pandemic cohort of patients in Canada according to Kumar et al. 8 METHODS This observational study included all cases of pregnant women with confirmed pandemic HINI 2009 influenza ABBREVIATIONS APACHE Acute Physiology and Chronic Health Evaluation ARDS acute respiratory distress syndrome ECMO extracorporeal membrane oxygenation. FiO 2 fraction of inspired oxygen concentration HIE hypoxic ischemic encephalopathy PaO 2 partial pressure of arterial oxygen PCR polymerase chain reaction PEEP positive end-expiratory pressure who were admitted or referred to the two tertiary care centres ICUs (St. Boniface Hospital and Health Sciences Centre). These two ICUs were the only units providing care for pregnant patients in the province, from March 1 to August 31, Charts were identified for review using ICD 10 codes for pregnancy, ICU admission, and influenza. Consent to use patient data was obtained from patients or their next of kin. Pandemic H1N virus was confirmed by PCR. Data were collected both prospectively and retrospectively for selected maternal and fetal variables. Charts were reviewed and data were extracted by one investigator and trained research assistants using a standardized data collection form. These data were then coded, entered into a database, and analyzed using simple descriptive statistics and SAS version 9.2 software (SAS Institute Inc., Cary NC). All data were cross-checked with the original chart by a second investigator. No data directly identifying patients were collected, and anonymity and confidentiality were guaranteed throughout the process. Ethical approval for this study was obtained from the University of Manitoba s Ethics Board. RESULTS During the 22-week study period (March 1 to August 31, 2009), there were 889 confirmed cases of H1N1 influenza, 182 hospitalizations not requiring ICU admission, 43 ICU admissions, and seven deaths in Manitoba (population ). 6 During the same time period there were 7600 pregnant patients in the province, with 1.5% of the population ( of ) having been pregnant during the preceding year. Of the 889 confirmed cases of H1N1 in Manitoba, 30 (3.4%) were pregnant at the time of diagnosis. Of these 30 pregnant women, nine were managed in their home community, 18 were hospitalized without ICU admission, and three were admitted to ICU. The additional members of the pregnant cohort were two non-residents of Manitoba and one pediatric patient (aged 14 years); these patients were also referred to our institutions ICUs for care. In our cohort the women admitted to ICU averaged 22.6 years of age (range 16 29) and five of the six (83%) were Aboriginal. All of the women admitted to ICU presented with cough and shortness of breath, and four of the six presented with influenza-like illness in the third trimester. Five of these women had a BMI > 30 kg/m 2 (mean 37 kg/m 2, range 31 kg/m 2 to 43 kg/m 2 ), one was a known smoker, and four of the six had contact with influenza-like illness within their household (Table 1). None had a history of asthma. Three of the women admitted to ICU were intubated in the peripartum period. At the time of admission, one patient had an arterial ph < 7.2; 444 MAY JOGC MAI 2010

3 Perinatal and Maternal Outcomes in Critically ill Obstetrics Patients With Pandemic H1N1 Influenza A all of the patients admitted had an oxygen saturation < 90% (range 50 88%) and had a temperature > 38 C. All patients received broad spectrum antibiotics and oseltamivir (Tamiflu), starting on the second day in the ICU (mean 6 days from onset of symptoms). The mean PEEP on day three was 15.8 cm H 2 O (range 13 19) in pregnant patients; mean FiO 2 was 0.8, and mean PaO 2 was 80 mm Hg (range ). The mean period of ventilation was 30.5 days (range ) and the mean APACHE score on admission was 17 (range 9 to 24). In addition to intubation, a form of ECMO was used in 33% (2/6) of the pregnant patients. Hemodialysis was also used in two of the six pregnant patients. Two of our cohort also experienced iatrogenic medical complications: one developed a bronchopleural fistula, and the other, bilateral hemothoraces due to anticoagulation for ECMO using Novalung. Two of the six women died during admission. Three of the six women had emergency Caesarean sections (two for fetal indications, one for maternal indication). The perinatal outcome in our cohort was poor, with one stillbirth at 35 weeks gestation, one neonatal death at 39 weeks due to HIE a consequence of perinatal asphyxia, one preterm delivery after preterm premature rupture of membranes at 25 weeks gestation, and one 39-week live-born neonate with severe complications secondary to HIE. There was one spontaneous abortion at six weeks gestation. Three patients were admitted to the ICU for more than 20 days, and median length of stay for all pregnant patients was 20.5 days (range 10 82). All four survivors of our pregnant cohort were eventually discharged. DISCUSSION Our series supports the impression that pregnant women are at increased risk of severe illness from H1N1 in pregnancies complicated by influenza, which has been documented in previous studies. 9,10 In our cohort there appeared to be a predisposition for disease in the third trimester, with rapid deterioration and even mortality in some cases (Tables 1 and 2). This predisposition for the third trimester of pregnancy was noted by Dodds et al., who demonstrated increasing prevalence of morbidity from influenza with each trimester, with odds ratio increasing from 2.9 in the first trimester to 7.9 in the third trimester. 11 While other comorbidities such as pre-existing respiratory or cardiovascular disease have been implicated as risk factors for H1N1, we did not find this to be the case in our series, although this was limited by small numbers (Table 1). Most of the pregnant women in our cohort (83%) had a BMI above 30 kg/m 2, which is associated with increased morbidity and morbidity). However, the significance of this Table 1. Clinical characteristics and risk factors Variable Age, years (50.0) (0.0) 31 Ethnicity Non-Aboriginal Aboriginal 5 (83.3) Gestational age at presentation, weeks Smoker BMI, kg/m 2 < >40 Contacts with influenza-like illness Yes No 2 (33.3) Presenting symptoms Cough Shortness of breath Fever Wheeze 3 (50.0) Duration of admission before ventilation, days 2 5 (83.3) >2 Mean duration of ventilation, days (range) Survivors 36 (14 82) Non-survivors 20.5 (20 21) range is unclear in this population because it is a common comorbidity; approximately one quarter of all parturients in Manitoba have a BMI of > 35 kg/m at term. Rather the mechanical, biochemical, hemodynamic, and immunologic changes in maternal physiology may predispose to adverse outcomes during influenza attacks. 12 Most of our pregnant patients had findings on chest X-ray that were compatible with ARDS and required intubation, high initial ventilator pressure, and additional strategies to maintain oxygenation including NO (nitrous oxide), ECMO, and high PEEP. Although none of these MAY JOGC MAI

4 Table 2. Maternal clinical management and outcome, Need for vasopressors Yes 4 (67) No 2 (33) Use of hemodialysis Yes 2 (33) Tamiflu 6 (100) Maternal outcomes Death 2 (33) Survival 4 (67) Iatrogenic medical complications 2 (33) modalities have yet been demonstrated to reduce mortality in the general ARDS population, they have been associated with improved oxygenation, which is essential in pregnancy. 13 The effects of these strategies on the fetus are not completely clear, but reports on women undergoing cardiopulmonary bypass during pregnancy have shown up to 19% perinatal mortality. 14,15 Our cohort suffered significant mortality, with two of the six pregnant women succumbing to their illness (Table 2). The risk of maternal mortality seems to be greater in pregnant patients who developed ARDS than in pregnant patients admitted to the ICU for other indications. 16 Vasquez et al. 16 showed a mortality of 33% in patients with ARDS, and Catanzarite et al. 17 described a mortality of 39.3% in pregnant patients admitted to the ICU for ARDS compared with 6% for patients admitted for other indications. The reason for the increased morbidity and mortality in women who develop ARDS during pregnancy is not completely understood. The overall mortality of ARDS in pregnancy is still less than that in the general population (40% and 60%, respectively). 16 Younger age and fewer underlying diseases may explain these differences. Numerous studies of non-pandemic influenza have not shown significant teratogenic or adverse effects of influenza on fetal outcome. 5,18 Abromwitz et al. 19 studied the effect of influenza on pregnancy outcome (spontaneous abortion, prematurity, stillbirths, and neonatal deaths) during the influenza epidemic of 1957 in South Africa and found the effect to be negligible. However, Harris s study 4 of the 1918 pandemic showed a 26% rate of pregnancy interruption (abortion or premature birth) in pregnant women with influenza but without pneumonia, and 52% in those with pneumonia. Interestingly, Harris also showed a significant increase in the risk of maternal death if pregnancy interruption occurred (41% if pregnancy was interrupted and 16% if it was not). This association between spontaneous abortion Table 3. Obstetrical and neonatal management and outcome Obstetric status and outcome Gestational age at end of pregnancy (n = 6), weeks (66.6) Antenatal complications (n = 6) Gestational hypertension PPROM Preterm labour Pregnancy outcome (n = 6) SA Live birth 4* (66.6) Stillbirth Mode of delivery (n = 5) SVD 2 (40) CS 3 (60) NICU admissions 3 (60) Length of NICU admission, days, Survivors (2) mean 23.5 Non-survivors (1) 2 Neonatal outcomes (n = 4) Death 1 (25) Survival 3 (75) Without sequelae of HIE 2 (50) With sequelae of HIE 1 (25) Evidence of HIE on investigation 2 *Including 1 preterm birth Including 1 neonatal death PPROM: preterm premature rupture of the membranes; SA: spontaneous abortion; SVD: singleton vaginal delivery; CS: Caesarean section; HIE: hypoxic ischemic encephalopathy. during influenza in pregnancy and adverse maternal outcome was also documented by Robinson et al. 5 in The case is very different for the critically ill patient, where significant hypoxemia may have major implications for fetal outcome. Our cohort demonstrated an excess of perinatal morbidity and mortality that was similar to previous reported neonatal mortality rates of 32% in patients admitted to the ICU (Table 3). 17 Severe hypoxemia in the mother may adversely affect the fetus because oxygen delivery to the fetus is determined by maternal arterial oxygen content and uterine blood flow. 12,20 Cartin-Ceba et al. 21 showed 446 MAY JOGC MAI 2010

5 Perinatal and Maternal Outcomes in Critically ill Obstetrics Patients With Pandemic H1N1 Influenza A adverse effects of shock and early gestational age on the fetus in critically ill patients admitted to the ICU. In our cohort, autopsy after neonatal death confirmed multiorgan hypoxic injury consistent with reduced delivery of oxygen to the fetus. Hyperthermia may also have been a contributing factor, because there is a growing body of evidence linking hyperthermia and inflammation at term to adverse neurologic outcome of the fetus. 22 This case series supports the suggestion that H1N1 can significantly compromise the pregnant patient and her fetus. Although our series is limited by its small size, lack of a well-defined denominator, and retrospective nature, its strengths include the reporting of perinatal outcomes in these cases, providing insight into the ventilatory requirements and possible fetal outcomes of critically ill pregnant patients with H1N1 influenza. CONCLUSION Our data reinforce the suggestion that pregnancy is an important risk factor in H1N1-infected patients. H1N1 infection can significantly compromise the pregnant patient and her fetus and can result in poor maternal and fetal outcomes. Pandemic planning must continue to consider the increased risks associated with pregnancy. ACKNOWLEDGEMENTS We would like to thank Manitoba Health and Healthy Living for their contribution to data collection. REFERENCES 1. Derlet RW, Sandrock CE, Nguyen HH, Lawrence R. Influenza. WebMD Emedicine. Epub December 8, Available at: Accessed March 9, Bland PB. Influenza in its relation to pregnancy and labour. Am J Obstet Dis Women Child 1919;79: Skowronski DM, De Serres G. Is routine influenza immunization warranted in early pregnancy? Vaccine 2009;27: Epub 2009 Apr Harris JW. Influenza occurring in pregnant women. JAMA 1919;72: Buck C. Exposure to virus diseases in early pregnancy and congenital malformation. Can Med Assoc J 1955;72: Manitoba Health and Healthy Living. [Website]. Available at: Accessed March 11, Jamieson DJ, Honein MA, Rasmussen SA, Williams JL, Swerdlow DL, Biggerstaff MS, et al. H1N influenza virus infection during pregnancy in the USA. Lancet 2009;374: Epub 2009 Jul Kumar A, Zarychanski R, Pinto R, Cook DJ, Marshall J, Lacroix J, et al. Critically ill patients with pandemic H1N1 influenza in Canada. JAMA Oct 2009;302: Epub 2009 Oct Cox S, Posner SF, McPheeters M, Jamieson DJ, Kourtis AP, Meikle S. Hospitalizations with respiratory illness among pregnant women during influenza season. Obstet Gynecol 2006;107: Cox S, Posner SF, McPheeters M, Jamieson DJ, Kourtis AP, Meikle S. Influenza and pregnant women: hospitalization burden, United States, J Womens Health (Larchmt) 2006;15: Dodds L, McNeil SA, Fell DB, Allen VM, Coombs A, Scott J, et al. Impact of influenza exposure on rates of hospital admissions and physician visits because of respiratory illness among pregnant women. CMAJ 2007;176: Lapinsky SE. Cardiopulmonary complications of pregnancy. Crit Care Med2005;33: Cole DE, Taylor TL, McCullough DM, Shoff CT, Derdak S. Acute respiratory distress syndrome in pregnancy. Crit Care Med 2005;33(10 Suppl):S269-S King P, Rosalion A, McMillan J, Buist M, Holmes P. Cardiopulmonary bypass during pregnancy. Extracorporeal membrane oxygenation in pregnancy. Lancet 2000;356: Parry AJ, Westaby S. Cardiopulmonary bypass during pregnancy Ann Thorac Surg 1996;61: Vasquez DN, Estenssoro E, Canales HS, Reina R, Saenz MG, Das Neves AV, et al. Clinical characteristics and outcomes of obstetric patients requiring ICU admission. Chest 2007;131: Catanzarite V, Willms D, Wong D, Landers C, Cousins L, Schrimmer D. Acute respiratory distress syndrome in pregnancy and the puerperium: causes, courses, and outcomes. Obstet Gynecol 2001;97(5 Pt 1): Wilson MG, Stein AM. Teratogenic effects of Asian influenza an extended study. JAMA 1969;210: Abramowitz LJ. The effect of Asian flu on pregnancy. S Afr Med J 1958;32: Hallak M, Kupsky WJ, Hotra JW, Irtenkauf SM. Foetal rat brain injury: effect of transient maternal hypoxemia. Fetal Diagn Ther 1997;12:68 71 (DOI: / ). 21. Cartin-Ceba R, Gajic O, Iyer VN, Vlahakis NE. Fetal outcomes of critically ill pregnant women admitted to the intensive care unit for nonobstetric causes. Critical Care Med 2008;36: Goetzl L, Zighelboim I, Badell M, Rivers J, Mastrangèlo MA, Tweardy D, et al. Maternal corticosteroids to prevent intrauterine exposure to hyperthermia and inflammation; a randomized, placebo-controlled trial. Am J Obstet Gynecol 2006;195: MAY JOGC MAI