G Model
ARTICLE IN PRESS
JVAC-16269; No. of Pages 8
Vaccine xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
A/H1N1 pandemic influenza vaccination: A retrospective evaluation of adverse maternal, fetal and neonatal outcomes in a cohort of pregnant women in Italy Massimo Fabiani a , Antonino Bella a , Maria C. Rota a , Elena Clagnan b , Tolinda Gallo c , Maurizio D’Amato d , Patrizio Pezzotti d,1 , Lorenza Ferrara e , Vittorio Demicheli e , Domenico Martinelli f , Rosa Prato f , Caterina Rizzo a,∗ a
National Centre for Epidemiology, Surveillance and Health Promotion, Italian National Institute of Health, Viale Regina Elena 299, 00161 Rome, Italy Epidemiological Unit, Regional Central Directorate for Health, Social-Health Integration, Social Policy, and Family, Udine, Italy c Department of Prevention, Local Health Unit 4 Medio Friuli, Udine, Italy d Laziosanità—Agency for Public Health of Lazio Region, Rome, Italy e SeREMI, Local Health Unit of Alessandria—Piemonte Region, Alessandria, Italy f Department of Medical and Surgical Sciences, University of Foggia, Foggia, Italy b
a r t i c l e
i n f o
Article history: Received 4 December 2014 Received in revised form 9 March 2015 Accepted 12 March 2015 Available online xxx Keywords: Pandemic vaccination Influenza A/H1N1pdm09 Pregnancy Pregnancy outcomes Safety
a b s t r a c t Although concerns about safety of influenza vaccination during pregnancy have been raised in the past, vaccination of pregnant women was recommended in many countries during the 2009 A/H1N1 pandemic influenza. A retrospective cohort study was conducted to evaluate the risk of adverse maternal, fetal and neonatal outcomes among pregnant women vaccinated with a MF59-adjuvanted A/H1N1 pandemic influenza vaccine. The study was carried out in four Italian regions (Piemonte, Friuli-Venezia-Giulia, Lazio, and Puglia) among 102,077 pregnant women potentially exposed during the second or third trimester of gestation to the vaccination campaign implemented in 2009/2010. Based on data retrieved from the regional administrative databases, the statistical analysis was performed using the Cox proportional-hazards model, adjusting for the propensity score to account for the potential confounding effect due to the socio-demographic characteristics and the clinical and reproductive history of women. A total of 100,332 pregnant women were eligible for the analysis. Of these, 2003 (2.0%) received the A/H1N1 pandemic influenza vaccination during the second or third trimester of gestation. We did not observe any statistically significant association between the A/H1N1 pandemic influenza vaccination and different maternal outcomes (hospital admissions for influenza, pneumonia, hypertension, eclampsia, diabetes, thyroid disease, and anaemia), fetal outcomes (fetal death after the 22nd gestational week) and neonatal outcomes (pre-term birth, low birth weight, low 5-min Apgar score, and congenital malformations). Pre-existing health-risk conditions (hospital admissions and drug prescriptions for specific diseases before the onset of pregnancy) were observed more frequently among vaccinated women, thus suggesting that concomitant chronic conditions increased vaccination uptake. The results of this study add some evidence on the safety of A/H1N1 pandemic influenza vaccination during pregnancy but, because of the reduced statistical power, meta-analyses and large multi-centres studies are needed in order to obtain more conclusive results, especially for rare outcomes. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction ∗ Corresponding author. Tel.: +39 06 49904277; fax: +39 06 44232444. E-mail address: [email protected] (C. Rizzo). 1 Present address: Department of Infectious, Parasitic and Immune-mediated Diseases, Italian National Institute of Health, Viale Regina Elena 299, 00161 Rome, Italy.
Many studies have shown an increased risk of mortality and influenza severity (hospitalization and admission to intensive care units) among pregnant women during the pandemics occurred in the 20th century [1]. Moreover, it has been reported that, during inter-pandemic periods, pregnant women with at least
http://dx.doi.org/10.1016/j.vaccine.2015.03.041 0264-410X/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Fabiani M, et al. A/H1N1 pandemic influenza vaccination: A retrospective evaluation of adverse maternal, fetal and neonatal outcomes in a cohort of pregnant women in Italy. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.03.041
G Model JVAC-16269; No. of Pages 8 2
ARTICLE IN PRESS M. Fabiani et al. / Vaccine xxx (2015) xxx–xxx
Fig. 1. Identification of pregnant women potentially exposed to the A/H1N1 pandemic influenza vaccination campaign in the 2nd or 3rd trimester of gestation (Italy, 2009–2010).
one pre-existing health risk condition have also an increased risk of mortality and influenza severity compared with healthy pregnant women [2]. The increased susceptibility to the influenza virus infection during pregnancy is caused by the reduction of the immune-response, particularly during the third trimester of gestation [3]. Based on these evidences, both seasonal and pandemic influenza vaccinations are internationally recommended for pregnant women, independently of the presence of pre-existing health risk conditions [4]. Although vaccination of pregnant women has been recommended since 2005 by the World Health Organization (WHO) [5], the utilization of the influenza vaccine during pregnancy is still limited because of concerns about its potential effect on the fetus development and possible teratogenic effects [6]. The seasonal influenza vaccine, available since the sixties’, has shown a good safety profile in the general population. However, most of information on vaccine safety among pregnant women comes from post-marketing pharmacovigilance studies [7], being pregnant women excluded from clinical trials conducted to obtain the authorization to commercialization. Epidemiological studies focusing on safety of seasonal influenza vaccination in pregnant women have not shown an increased risk of either complications or adverse outcomes associated with the administration of inactivated non-adjuvanted vaccines [8]. In Italy, only one type of vaccine (Focetria® , manufactured by Novartis) was distributed for preventing the pandemic influenza A/H1N1pdm09. This vaccine was adjuvanted with MF59, a squalene in-water emulsion. Since 1997, the same adjuvant has been used in some seasonal influenza vaccines (e.g., Fluad® ), which were largely distributed (over 45 millions of doses) in several countries worldwide, including Italy, without showing an increased risk of adverse events [9]. This is consistent with the fact that in Italy, to date, the MF59-adjuvanted pandemic vaccine has shown a tolerability and safety profile very similar to those of the seasonal influenza vaccines [10]. During the 2009 A/H1N1 pandemic, the Italian Ministry of Health recommended and offered the influenza pandemic vaccination free of charge to different risk groups, including pregnant women in the second or third trimester of gestation, independently of the presence of pre-existing health risk conditions [11]. The pandemic influenza vaccine was only available through the National Health Service. This study aims at evaluating the risk of adverse maternal, fetal and neonatal outcomes associated with the administration of the MF59-adjuvanted A/H1N1pdm09 influenza vaccine (hereafter referred to as pandemic influenza vaccine) among Italian pregnant women and their newborns.
2. Methods 2.1. Study population and data sources This retrospective cohort study was conducted in four out of 21 Italian regions accepting to participate and able to link individual data from the pandemic influenza vaccination registry with those from the other data sources (Piemonte, Friuli-Venezia-Giulia, Lazio, and Puglia). It investigated pregnant women potentially exposed during the 2nd and 3rd trimester of gestation to the influenza pandemic vaccination campaign implemented from 15 October 2009 to 31 March 2010. We identified the pregnancies starting in the period 15 January 2009–31 December 2009 and ending with a delivery in the period 15 October 2009–30 September 2010 as those potentially exposed to the vaccination campaign (Fig. 1). We then applied the exclusion criteria listed in Table 1. Different regional databases and registries, linkable through a unique and anonymised personal identifier, were used to retrieve the information analysed in this study. The cohort of women was identified through the regional birth registries, which include information on the date and completed gestational weeks at delivery, fetal and neonatal outcomes, and socio-demographic characteristics and reproductive history of women for all births occurred in the national territory after at least 22 gestational weeks. Using this database, the date of pregnancy onset was estimated by subtracting the gestational age (weeks of amenorrhea) from the date of delivery. The pandemic vaccination databases were used to identify vaccinated women and the corresponding date of vaccination. The hospital discharges databases, including principal and secondary diagnoses coded according to the ICD9-CM system for all admissions occurred to both public and private health structures at all levels, were used to retrieve information on pre-existing health risk conditions and maternal outcomes. From the same data source, we also extracted information on congenital malformations diagnosed within the first six months of life of newborns in all regions except Puglia (data not available). Finally, the drug prescriptions databases were also used to identify pre-existing health risk conditions, based on disease-specific prescriptions issued within the Regional Health Service and coded according to the ATC system. The study was approved by the ethical committee of the Italian National Institute of Health (Prot. CE/11/317). 2.2. Exposure, outcomes and possible confounders We evaluated the association between the administration of the pandemic influenza vaccine in the 2nd and 3rd trimester of gestation (exposure) and several maternal, fetal and neonatal outcomes.
Please cite this article in press as: Fabiani M, et al. A/H1N1 pandemic influenza vaccination: A retrospective evaluation of adverse maternal, fetal and neonatal outcomes in a cohort of pregnant women in Italy. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.03.041
G Model JVAC-16269; No. of Pages 8
ARTICLE IN PRESS M. Fabiani et al. / Vaccine xxx (2015) xxx–xxx
3
Table 1 Women included in the cohort study by exclusion criteria (Italy, 2009–2010).
Pregnant womena Excluded Women giving birth with neither hospital admissions for delivery or pregnancy-related diseases nor vaccination for pregnancy
Unvaccinated N (%)
Vaccinated N (%)
Total N (%)
100,029 (100.0) 1700 (1.7) 226 (0.2)
2048 (100.0) 45 (2.2) 0 (0.0)
102,077 (100.0) 1745 (1.7) 226 (0.2)
Multiple births
1255 (1.3)
41 (2.0)
1296 (1.3)
Age < 13 years or age > 55 years
2 (0.0)
0 (0.0)
2 (0.0)
Gestational age at delivery > 45 weeks
17 (0.0)
1 (0.0)
18 (0.0)
b
Women delivering a livebirth but with an abortive outcome recorded for the same pregnancy
68 (0.1)
0 (0.0)
68 (0.1)
Viral infections during pregancy or congenital viral infections in newbornsc
113 (0.1)
2 (0.1)
115 (0.1)
19 (0.0) 98,329 (98.3)
1 (0.0) 2003 (97.8)
20 (0.0) 100,332 (98.3)
Chromosomal malformations in newbornsd Included in the analysis a
Pregnancies starting in the period 15/1/2009–31/12/2009 and ending with a delivery in the period 15/10/2009–30/9/2010 after at least 22 weeks of gestation. Pregnancies recorded as ending both in an abortion (information retrieved from the hospital discharge database) and a live-birth (information retrieved from the birth registry). c ICD9-CM: 647*, 771*. Women with viral infections during pregnancy or delivering newborns with congenital viral infections (e.g., varicella, rubella and cytomegalovirus) were excluded because possibly confounding the association between the A/H1N1 pandemic influenza vaccination and maternal, fetal and neonatal outcomes. d ICD9-CM: 7580* - 7583*; 7585* - 7589*. b
In particular, we considered as maternal outcomes the occurrence, during the 2nd or 3rd trimester of gestation, of hospital admissions with a principal or secondary diagnosis of influenza, pneumonia, gestational hypertension, eclampsia, gestational diabetes, thyroid disease or anaemia. Fetal and neonatal outcomes included stillbirth (defined as the delivery of a dead fetus at ≥22 weeks of gestation), preterm birth (
ARTICLE IN PRESS
JVAC-16269; No. of Pages 8
Vaccine xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
A/H1N1 pandemic influenza vaccination: A retrospective evaluation of adverse maternal, fetal and neonatal outcomes in a cohort of pregnant women in Italy Massimo Fabiani a , Antonino Bella a , Maria C. Rota a , Elena Clagnan b , Tolinda Gallo c , Maurizio D’Amato d , Patrizio Pezzotti d,1 , Lorenza Ferrara e , Vittorio Demicheli e , Domenico Martinelli f , Rosa Prato f , Caterina Rizzo a,∗ a
National Centre for Epidemiology, Surveillance and Health Promotion, Italian National Institute of Health, Viale Regina Elena 299, 00161 Rome, Italy Epidemiological Unit, Regional Central Directorate for Health, Social-Health Integration, Social Policy, and Family, Udine, Italy c Department of Prevention, Local Health Unit 4 Medio Friuli, Udine, Italy d Laziosanità—Agency for Public Health of Lazio Region, Rome, Italy e SeREMI, Local Health Unit of Alessandria—Piemonte Region, Alessandria, Italy f Department of Medical and Surgical Sciences, University of Foggia, Foggia, Italy b
a r t i c l e
i n f o
Article history: Received 4 December 2014 Received in revised form 9 March 2015 Accepted 12 March 2015 Available online xxx Keywords: Pandemic vaccination Influenza A/H1N1pdm09 Pregnancy Pregnancy outcomes Safety
a b s t r a c t Although concerns about safety of influenza vaccination during pregnancy have been raised in the past, vaccination of pregnant women was recommended in many countries during the 2009 A/H1N1 pandemic influenza. A retrospective cohort study was conducted to evaluate the risk of adverse maternal, fetal and neonatal outcomes among pregnant women vaccinated with a MF59-adjuvanted A/H1N1 pandemic influenza vaccine. The study was carried out in four Italian regions (Piemonte, Friuli-Venezia-Giulia, Lazio, and Puglia) among 102,077 pregnant women potentially exposed during the second or third trimester of gestation to the vaccination campaign implemented in 2009/2010. Based on data retrieved from the regional administrative databases, the statistical analysis was performed using the Cox proportional-hazards model, adjusting for the propensity score to account for the potential confounding effect due to the socio-demographic characteristics and the clinical and reproductive history of women. A total of 100,332 pregnant women were eligible for the analysis. Of these, 2003 (2.0%) received the A/H1N1 pandemic influenza vaccination during the second or third trimester of gestation. We did not observe any statistically significant association between the A/H1N1 pandemic influenza vaccination and different maternal outcomes (hospital admissions for influenza, pneumonia, hypertension, eclampsia, diabetes, thyroid disease, and anaemia), fetal outcomes (fetal death after the 22nd gestational week) and neonatal outcomes (pre-term birth, low birth weight, low 5-min Apgar score, and congenital malformations). Pre-existing health-risk conditions (hospital admissions and drug prescriptions for specific diseases before the onset of pregnancy) were observed more frequently among vaccinated women, thus suggesting that concomitant chronic conditions increased vaccination uptake. The results of this study add some evidence on the safety of A/H1N1 pandemic influenza vaccination during pregnancy but, because of the reduced statistical power, meta-analyses and large multi-centres studies are needed in order to obtain more conclusive results, especially for rare outcomes. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction ∗ Corresponding author. Tel.: +39 06 49904277; fax: +39 06 44232444. E-mail address: [email protected] (C. Rizzo). 1 Present address: Department of Infectious, Parasitic and Immune-mediated Diseases, Italian National Institute of Health, Viale Regina Elena 299, 00161 Rome, Italy.
Many studies have shown an increased risk of mortality and influenza severity (hospitalization and admission to intensive care units) among pregnant women during the pandemics occurred in the 20th century [1]. Moreover, it has been reported that, during inter-pandemic periods, pregnant women with at least
http://dx.doi.org/10.1016/j.vaccine.2015.03.041 0264-410X/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Fabiani M, et al. A/H1N1 pandemic influenza vaccination: A retrospective evaluation of adverse maternal, fetal and neonatal outcomes in a cohort of pregnant women in Italy. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.03.041
G Model JVAC-16269; No. of Pages 8 2
ARTICLE IN PRESS M. Fabiani et al. / Vaccine xxx (2015) xxx–xxx
Fig. 1. Identification of pregnant women potentially exposed to the A/H1N1 pandemic influenza vaccination campaign in the 2nd or 3rd trimester of gestation (Italy, 2009–2010).
one pre-existing health risk condition have also an increased risk of mortality and influenza severity compared with healthy pregnant women [2]. The increased susceptibility to the influenza virus infection during pregnancy is caused by the reduction of the immune-response, particularly during the third trimester of gestation [3]. Based on these evidences, both seasonal and pandemic influenza vaccinations are internationally recommended for pregnant women, independently of the presence of pre-existing health risk conditions [4]. Although vaccination of pregnant women has been recommended since 2005 by the World Health Organization (WHO) [5], the utilization of the influenza vaccine during pregnancy is still limited because of concerns about its potential effect on the fetus development and possible teratogenic effects [6]. The seasonal influenza vaccine, available since the sixties’, has shown a good safety profile in the general population. However, most of information on vaccine safety among pregnant women comes from post-marketing pharmacovigilance studies [7], being pregnant women excluded from clinical trials conducted to obtain the authorization to commercialization. Epidemiological studies focusing on safety of seasonal influenza vaccination in pregnant women have not shown an increased risk of either complications or adverse outcomes associated with the administration of inactivated non-adjuvanted vaccines [8]. In Italy, only one type of vaccine (Focetria® , manufactured by Novartis) was distributed for preventing the pandemic influenza A/H1N1pdm09. This vaccine was adjuvanted with MF59, a squalene in-water emulsion. Since 1997, the same adjuvant has been used in some seasonal influenza vaccines (e.g., Fluad® ), which were largely distributed (over 45 millions of doses) in several countries worldwide, including Italy, without showing an increased risk of adverse events [9]. This is consistent with the fact that in Italy, to date, the MF59-adjuvanted pandemic vaccine has shown a tolerability and safety profile very similar to those of the seasonal influenza vaccines [10]. During the 2009 A/H1N1 pandemic, the Italian Ministry of Health recommended and offered the influenza pandemic vaccination free of charge to different risk groups, including pregnant women in the second or third trimester of gestation, independently of the presence of pre-existing health risk conditions [11]. The pandemic influenza vaccine was only available through the National Health Service. This study aims at evaluating the risk of adverse maternal, fetal and neonatal outcomes associated with the administration of the MF59-adjuvanted A/H1N1pdm09 influenza vaccine (hereafter referred to as pandemic influenza vaccine) among Italian pregnant women and their newborns.
2. Methods 2.1. Study population and data sources This retrospective cohort study was conducted in four out of 21 Italian regions accepting to participate and able to link individual data from the pandemic influenza vaccination registry with those from the other data sources (Piemonte, Friuli-Venezia-Giulia, Lazio, and Puglia). It investigated pregnant women potentially exposed during the 2nd and 3rd trimester of gestation to the influenza pandemic vaccination campaign implemented from 15 October 2009 to 31 March 2010. We identified the pregnancies starting in the period 15 January 2009–31 December 2009 and ending with a delivery in the period 15 October 2009–30 September 2010 as those potentially exposed to the vaccination campaign (Fig. 1). We then applied the exclusion criteria listed in Table 1. Different regional databases and registries, linkable through a unique and anonymised personal identifier, were used to retrieve the information analysed in this study. The cohort of women was identified through the regional birth registries, which include information on the date and completed gestational weeks at delivery, fetal and neonatal outcomes, and socio-demographic characteristics and reproductive history of women for all births occurred in the national territory after at least 22 gestational weeks. Using this database, the date of pregnancy onset was estimated by subtracting the gestational age (weeks of amenorrhea) from the date of delivery. The pandemic vaccination databases were used to identify vaccinated women and the corresponding date of vaccination. The hospital discharges databases, including principal and secondary diagnoses coded according to the ICD9-CM system for all admissions occurred to both public and private health structures at all levels, were used to retrieve information on pre-existing health risk conditions and maternal outcomes. From the same data source, we also extracted information on congenital malformations diagnosed within the first six months of life of newborns in all regions except Puglia (data not available). Finally, the drug prescriptions databases were also used to identify pre-existing health risk conditions, based on disease-specific prescriptions issued within the Regional Health Service and coded according to the ATC system. The study was approved by the ethical committee of the Italian National Institute of Health (Prot. CE/11/317). 2.2. Exposure, outcomes and possible confounders We evaluated the association between the administration of the pandemic influenza vaccine in the 2nd and 3rd trimester of gestation (exposure) and several maternal, fetal and neonatal outcomes.
Please cite this article in press as: Fabiani M, et al. A/H1N1 pandemic influenza vaccination: A retrospective evaluation of adverse maternal, fetal and neonatal outcomes in a cohort of pregnant women in Italy. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.03.041
G Model JVAC-16269; No. of Pages 8
ARTICLE IN PRESS M. Fabiani et al. / Vaccine xxx (2015) xxx–xxx
3
Table 1 Women included in the cohort study by exclusion criteria (Italy, 2009–2010).
Pregnant womena Excluded Women giving birth with neither hospital admissions for delivery or pregnancy-related diseases nor vaccination for pregnancy
Unvaccinated N (%)
Vaccinated N (%)
Total N (%)
100,029 (100.0) 1700 (1.7) 226 (0.2)
2048 (100.0) 45 (2.2) 0 (0.0)
102,077 (100.0) 1745 (1.7) 226 (0.2)
Multiple births
1255 (1.3)
41 (2.0)
1296 (1.3)
Age < 13 years or age > 55 years
2 (0.0)
0 (0.0)
2 (0.0)
Gestational age at delivery > 45 weeks
17 (0.0)
1 (0.0)
18 (0.0)
b
Women delivering a livebirth but with an abortive outcome recorded for the same pregnancy
68 (0.1)
0 (0.0)
68 (0.1)
Viral infections during pregancy or congenital viral infections in newbornsc
113 (0.1)
2 (0.1)
115 (0.1)
19 (0.0) 98,329 (98.3)
1 (0.0) 2003 (97.8)
20 (0.0) 100,332 (98.3)
Chromosomal malformations in newbornsd Included in the analysis a
Pregnancies starting in the period 15/1/2009–31/12/2009 and ending with a delivery in the period 15/10/2009–30/9/2010 after at least 22 weeks of gestation. Pregnancies recorded as ending both in an abortion (information retrieved from the hospital discharge database) and a live-birth (information retrieved from the birth registry). c ICD9-CM: 647*, 771*. Women with viral infections during pregnancy or delivering newborns with congenital viral infections (e.g., varicella, rubella and cytomegalovirus) were excluded because possibly confounding the association between the A/H1N1 pandemic influenza vaccination and maternal, fetal and neonatal outcomes. d ICD9-CM: 7580* - 7583*; 7585* - 7589*. b
In particular, we considered as maternal outcomes the occurrence, during the 2nd or 3rd trimester of gestation, of hospital admissions with a principal or secondary diagnosis of influenza, pneumonia, gestational hypertension, eclampsia, gestational diabetes, thyroid disease or anaemia. Fetal and neonatal outcomes included stillbirth (defined as the delivery of a dead fetus at ≥22 weeks of gestation), preterm birth (
