Abstracts

INTRODUCTION

The fetal mortality rate (FMR) is considered one of the best indicators of the quality of health care provided during pregnancy and childbirth.^aa Ministério da Saúde, Secretaria de Vigilância em Saúde, Secretaria de Atenção à Saúde. Manual de vigilância do óbito infantil e fetal e do Comitê de Prevenção do Óbito Infantil e Fetal. 2.ed. Brasília (DF); 2009 [cited 2012 Apr 10]. (Série A. Normas e Manuais Técnicos). Available from: http://bvsms.saude.gov.br/bvs/publicações/manual_obito_infantil_fetal_2ed.pdf FMR is calculated using the number of total fetal deaths in the numerator and the number of total births (live and dead births) in the denominator but by varying the criteria for gestational age and weight to define fetal death.^aa Ministério da Saúde, Secretaria de Vigilância em Saúde, Secretaria de Atenção à Saúde. Manual de vigilância do óbito infantil e fetal e do Comitê de Prevenção do Óbito Infantil e Fetal. 2.ed. Brasília (DF); 2009 [cited 2012 Apr 10]. (Série A. Normas e Manuais Técnicos). Available from: http://bvsms.saude.gov.br/bvs/publicações/manual_obito_infantil_fetal_2ed.pdf

WHO^bb World Health Organization, Department of Reproductive Health and Research. National, regional, and worldwide estimates of stillbirth rates in 2009 with trends since 1995: policy brief. Geneva; 2011 [cited 2013 Dec 19]. Available from: http://www.who.int/reproductivehealth/publications/maternal_perinatal_health/rhr_11_03/en/index.html (2011) estimated that approximately 2.6 million fetal deaths (considering infant weight > 1,000 g or gestational age > 28 weeks) occurred worldwide in 2009. Between 1995 and 2009, FMR showed a 14.0% decrease, from 22.1 per 1,000 total births to 18.9 per 1,000 births.^bb World Health Organization, Department of Reproductive Health and Research. National, regional, and worldwide estimates of stillbirth rates in 2009 with trends since 1995: policy brief. Geneva; 2011 [cited 2013 Dec 19]. Available from: http://www.who.int/reproductivehealth/publications/maternal_perinatal_health/rhr_11_03/en/index.html The number of fetal deaths in the third trimester of pregnancy was slightly smaller than the three million neonatal deaths that occurred worldwide.³⁸38 Stantos C, Lawn C, Lawn JE, Rahaman H, Wilczynska Ketende K, Hill K. Stillbirth rates: delivering estimates in 190 countries.Lancet. 2006;367(9521):1487-94. DOI:10.1016/S0140 6736(06)68586 3 Despite this magnitude, this indicator was not included by the United Nations among the Millennium Development Goals,²³23 Lawn JE, Gravett MG, Nunes TM, Rubens CE, Stanton C, Gapps Review Group. Global report on preterm birth and stillbirth (1of 7): definitions, description of the burden and opportunities to improve data. BMC Pregnancy Childbirth. 2010;10(Suppl 1):S1. DOI:10.1186/1471-2393-10-S1-S1 and fetal births have remained ignored by public health policies and programs, both internationally and domestically.¹⁷17 Frøen JF, Cacciatore J, McClure EM, Kuti O, Jokhio AH, Islam M, et al. Stillbirths: why they matter. Lancet.2011;377(9774):1353-66. DOI:10.1016/S0140-6736(10)62232-5^,cc Lansky S. fetal mortality: invisible and preventable deaths. In: Bittencourt DAS, Dias MAB, Wakimoto MD, organizadores. Vigilância do óbito materno, infantil e fetal e atuação em comitês de mortalidade. Rio de Janeiro (RJ): EAD/ENSP; 2013. FMR is low in developed countries, varying between two per 1,000 births and seven per 1,000 births, whereas in Brazil, FMR was at an intermediate level in 2010 at 10.97 per 1,000 births.²⁴24 Lawn JE, Blencowe H, Pattinson R, Cousens S, Kumar R, Ibiebele I, et al. Stillbirths: Where? When? Why? How to make the data count?Lancet. 2011;377(9775):1448-63. DOI:10.1016/S0140-6736(10)62187-3

According to the latest data available in Brazil,^dd Ministério da Saúde, DATASUS. Sistema de Informações de Mortalidade - SIM. Brasília (DF); c2008 [cited 2013 Apr 10]. Available from: http://www2.datasus.gov.br/DATASUS/index.php?area=060701 the absolute number of fetal deaths (considering infant weight of 500 g or gestational age of 22 weeks) decreased from 38,759 to 31,613 between 2001 and 2011. FMR in Brazil has some limitations and should only be used in the Brazilian states in which this rate is calculated using the direct method. Even so, it can be generally estimated using vital statistics. FMR decreased from 12.3 per 1,000 live births to 10.7 per 1,000 births in the same period.^dd Ministério da Saúde, DATASUS. Sistema de Informações de Mortalidade - SIM. Brasília (DF); c2008 [cited 2013 Apr 10]. Available from: http://www2.datasus.gov.br/DATASUS/index.php?area=060701^,ee Ministério da Saúde, DATASUS. Sistema de Informações de Nascidos Vivos - SINASC. Brasília (DF); c2008 [cited 2013 Apr 10]. Available from: http://www2.datasus.gov.br/DATASUS/index.php?area=060702 In the early 2000s, 30.0% of the data on birth weight were unknown. In 2011, this percentage decreased to 9.0%, and it was reported that approximately 30.0% of fetal deaths in Brazil were associated with infants weighing ≥ 2,500 g.

A review of perinatal deaths occurring in Brazil between 1990 and 2002¹⁴14 Fonseca SC, Coutinho ESF. Pesquisa sobre mortalidade perinatal no Brasil: revisão da metodologia e dos resultados. Cad Saude Publica. 2004;20(Supl 1):S7-19. DOI:10.1590/S0102-311X2004000700002 found few specific studies on this topic, with population and methodological differences such as the definition and calculation of some variables. In the 1990s, FMR was estimated at 14.0 per 1,000 births, and FMR was higher in northeast Brazil and lower in South Brazil. When secondary data were used, the level of data completeness was poor, and most underlying causes were listed as undefined. The studies on prevention identified that most fetal deaths occurred before birth, suggesting deficient prenatal care. The factors associated with fetal death were maternal age > 35 years, low income, low education level, inadequate prenatal care, and previous fetal death. However, these factors have been identified in one or at the most two studies, having little consistency as scientific evidence.

The objective of this study was to review the frequency and factors associated with fetal death in the Brazilian scientific literature.

METHODS

A systematic review of the literature evaluated fetal death as a primary or secondary outcome in Brazil between 2003 and 2013.

Three main factors were used as eligibility criteria: studies on the quality of information (variables: level of completeness, reliability, and validity of death certificates), descriptive studies (variables: description of FMRs and temporal trends, description of underlying causes, and application of a classification of preventable deaths), and studies (transverse and longitudinal) on the factors associated with fetal deaths. The studies involving populations with specific morbidities, basic research (e.g., on chemical mediators and cell biology), diagnostic tests, clinical trials, case reports, dissertations, and theses were excluded. Review studies were also excluded. Studies on perinatal mortality were included when they provided individualized data on fetal death.

The literature search was conducted in the Lilacs database (via the Virtual Health Library) and Medline (via PubMed), without language restriction. In Lilacs, the search strategy was (fetal death OR fetal mortality OR stillbirth) AND Brazil, in the abstract or subject fields. The search for Brazilian states (replacing the keyword “Brazil”) was included to increase the sensitivity of the search. For Medline, the strategy was (fetal death OR fetal mortality OR stillbirth OR natimortality) AND Brazil. An additional manual search was performed in the references of the included studies.

Initially, two examiners evaluated the titles. The abstracts of the studies with titles approved by at least one examiner were read. The studies with an abstracts approved by both examiners were included. In cases of disagreement, a third examiner read the abstract to make the final decision. The reasons for exclusion are listed in the flowchart (Figure).

After selecting the abstracts, the full text was read to confirm the eligibility of the studies and proceed with the collection of relevant information. The studies were read and independently evaluated by at least two examiners, and disagreements were resolved by a third examiner. The data on author, location, population characteristics, outcome, type of study design, criteria of fetal deaths, and main results were collected in accordance with a pre-established worksheet. The authors FMSB and SCF read all the included studies and reviewed the spreadsheet.

The methodological quality of the studies was not a criterion for inclusion because the intention of analysis was to analyze the nature of studies on fetal death in Brazil, revealing aspects related to the methodology used.

Initially, 321 studies were identified in Medline and 105 were identified in Lilacs. The search for Brazilian states retrieved three studies that had not been identified previously. In analysis of titles and abstracts, there was disagreement between the two independent examiners regarding 12 studies retrieved from Medline. After analysis by a third examiner, five of the 12 studies were excluded. In the Lilacs database, the evaluation of three studies was discordant, and one was excluded.

Considering the two databases, duplicates were excluded, leaving 29 studies to be read in full; of these, 25 were selected. Two additional studies were included after the manual search of the references. The Figure shows the flowchart for the search and selection of the 27 studies.

Among the 27 studies selected, 13 solely focused on fetal deaths, seven studies evaluated perinatal deaths, and the others evaluated other outcomes. The scientific production in the Southeast region was predominant (13/27). The Northeast and South regions contributed with five and six studies, respectively, and three studies covered the nation as a whole. No studies conducted in the Midwest and North regions were found. The journals with the most publications on this topic belonged to the area of public health, with 10 studies.

The studies selected and their main results are summarized according to their main approach. Table 1 presents the four studies that addressed the quality of information, Table 2 presents the 12 descriptive studies, and Table 3 shows the 11 studies that evaluated the factors associated with fetal deaths. When a study had more than one approach, the approach that the authors defined as the main objective was selected. The characteristics of the studies are classified according to venue and chronological order.

RESULTS

Quality of information

Four studies were identified; of these, two²2 Almeida MF, Alencar GP, Schoeps D, Minuci EG, Silva ZP, Ortiz LP, et al. Quality of information registered on fetal deaths certificates in São Paulo, Southeastern Brazil. Rev Saude Publica. 2011;45(5):845-53. DOI:10.1590/S0034-89102011005000058^,66 Beringhs EM, Gallo PR, Reis AOA. Declarações de nascidos mortos no município de São Paulo: avaliação descritiva do preenchimento. Rev Bras Saude Mater Infant. 2008;8(3):319-23. DOI:10.1590/S1519-38292008000300011 were conducted in the city of Sao Paulo, SP, Southeastern Brazil, and presented a very poor level of completeness of death certificates (DC) for important variables, particularly sociodemographic variables. The level of completeness of the variable “fetal weight” was approximately 100% in 2008 in comparison with 37.0% in 2001-2003. For the other variables, incompleteness often exceeded 50.0%, and the percentage of underlying causes was not specified.

Almeida et al²2 Almeida MF, Alencar GP, Schoeps D, Minuci EG, Silva ZP, Ortiz LP, et al. Quality of information registered on fetal deaths certificates in São Paulo, Southeastern Brazil. Rev Saude Publica. 2011;45(5):845-53. DOI:10.1590/S0034-89102011005000058 evaluated the difference in quality and the agreement between different data sources (death certificates or medical records), different issuers (death verification service or hospital), and institutions (Brazilian Unified Health System [SUS] or others). The quality was better in hospitals; among these, the quality was better in institutions not affiliated with SUS. Except for sex and weight, the variables had excellent agreement, with values between 2.9% and 82.0%. The agreement was low for the underlying causes, and according to these authors, the autopsies performed by the death verification services did not contribute to elucidating the causes of fetal death.

In Ribeirao Preto, in the state of Sao Paulo,⁵5 Barbuscia DM, Rodrigues-Júnior AL. Completude das informações nas declarações de Nascido Vivo e nas Declarações de Óbito, neonatal precoce e fetal, da região de Ribeirão Preto, São Paulo, Brasil, 2000-2007. Cad Saude Publica. 2011;27(6):1192-9. DOI:10.1590/S0102-311X2011000600016 the completion of nine fields on early neonatal and fetal death certificates during an eight-year period was evaluated. On the fetal death certificates, only the field “sex” was adequately completed, and completion of the field “birth weight” was regular. For the field “mother’s age”, incompleteness changed from regular to poor. For the fields “mother’s education” and “live and dead births”, completeness was inadequate throughout the evaluation period and incompleteness varied between 60.0% and 80.0%.

A study conducted in the Northeast³¹31 Rafael RAA, Ribeiro VS, Cavalcante MCV, Santos AM, Simões VMF. Relacionamento probabilístico: recuperação de informações de óbitos infantis e natimortos em localidade do Maranhão, Brasil. Cad Saude Publica. 2011;27(7):1371-9. DOI:10.1590/S0102-311X2011000700012 on the cross correlation of data from the Hospital Information System (HIS) and Mortality Information System (MIS) identified an additional 165 cases of stillbirth, indicating underreporting of approximately 10.0% cases in MIS. Underreporting was higher in smaller municipalities.

Descriptive studies

Twelve studies were retrieved: five in the Southeast, three in the South, two in the Northeast, and two of a national scope.

In Minas Gerais, in the microregion of Caratinga,²⁹29 Nurdan B, Mattar R, Camargo L. Óbito fetal em microrregião de Minas Gerais: causas e fatores associados. Rev Bras Ginecol Obstet.2003;25(2):103-7. DOI:10.1590/S0100-72032003000200005 a high percentage of full-term fetal deaths was observed (59.79%). In addition, 86.3% deaths occurred before delivery, and for 61.57% cases, the underlying causes of death were unknown.

The FMR in SUS hospitals in Ribeirao Preto¹¹11 Duarte G, Coltro PS, Bedone RV, Nogueira AA, Gelonezzi GM, Franco LJ. Trends in the modes of delivery and their impact on perinatal mortality rates. Rev Saude Publica. 2004;38(3):379-84. DOI:10.1590/S0034-89102004000300007 decreased from 1991 to 2000 regardless of the percentage of cesarean sections, the numbers of which remained unchanged, suggesting that this decrease was correlated with improved obstetric care.

The FMR in Sao Paulo decreased by approximately 20.0% between 2000 and 2006⁷7 Camargo ABM. A natimortalidade e a mortalidade perinatal em São Paulo. Sao Paulo Perspect. 2008;22(1):30-47. and increased along with an increase in maternal age. These studies showed improvement in the completion of the variables “weight”, “gestational age”, and “maternal age” but not other variables. The percentage of non-specific causes and intrauterine hypoxia remained high.

In Campinas, in the state of Sao Paulo,³3 Amaral E, Souza JP, Surita F, Luz AG, Souza MH, Cecatti JG, et al. A population-based surveillance study on severe acute maternal morbidity (near miss) and adverse perinatal outcomes in Campinas, Brazil: the Vigimoma Project.BMC Pregnancy Childbirth. 2011;11:9. DOI:10.1186/1471-2393-11-9 a population-based study conducted in 2005 indicated that intrauterine anoxia was the most frequent cause (46.9%) of fetal death, followed by malformations (25.0%).

The results of a study conducted in a large maternity hospital in the city of Rio de Janeiro, RJ, Southeastern Brazil¹⁵15 Fonseca SC, Coutinho ESF. Características biológicas e evitabilidade de óbitos perinatais em uma localidade na cidade do Rio de Janeiro, 1999 a 2003.Rev Bras Saude Mater Infant. 2008;8(2):171-8. DOI:10.1590/S1519-38292008000200004 indicated that 84.0% of the fetal deaths occurred before hospital admission. The search of medical records and prenatal care cards decreased the ill-defined causes to less than 20.0%. For both pre-term and full-term fetal deaths, maternal complications were the first underlying cause (37.0% and 27.0%, respectively), followed by complications related to the placenta, umbilical cord, and membranes.

In the city of Pelotas, RS, Southern Brazil, three birth cohorts were compared.²⁵25 Matijasevich A, Santos IS, Barros AJD, Menezes AMB, Albernaz EP, Barros FC, et al. Perinatal mortality in three population-based cohorts from southern Brazil: trends and differences. Cad Saude Publica. 2008;24(Supl 3):S399-408. DOI:10.1590/S0102-311X2008001500005 This is one of the few studies that used an old criterion for classifying stillbirths (28^th gestational week or weight > 1,000 g). The decrease in FMR was more pronounced between 1982 and 1993. The decrease occurred in all weight ranges, and intrauterine growth restriction (IGR) also decreased. However, inequalities associated with family income persisted. In 2004, FMR was higher in families with income of less than three minimum wages. One study²⁵25 Matijasevich A, Santos IS, Barros AJD, Menezes AMB, Albernaz EP, Barros FC, et al. Perinatal mortality in three population-based cohorts from southern Brazil: trends and differences. Cad Saude Publica. 2008;24(Supl 3):S399-408. DOI:10.1590/S0102-311X2008001500005 used the classification of Wigglesworth for death prevention⁴²42 Wigglesworth JS. Monitoring perinatal mortality.Lancet. 1980;316(8196):684-6. DOI:10.1016/S0140-6736(80)92717-8 and indicated a 72.0% decrease in intrapartum stillbirths and 36.0% decrease in fetal deaths before delivery.

In the city of Pato Branco, PR, Southern Brazil,³⁹39 Trindade LL, Amestroy SC, Piolo D, Falchetti G, Milbrath VM. Fatores de risco para morte fetal no município de Pato Branco (Brasil). Invest Educ Enferm. 2011;29(3):451-8. the average FMR was 9.3 per 1,000 births between 2000 and 2008. The most frequent underlying cause was placenta previa (38.5%), and the classification of 17.0% cases was not specified.

In Florianópolis, SC, Southern Brazil, a single study used the ReCoDe classification¹⁸18 Gardosi J, Kady SM, McGeown P, Francis A, Tonks A. Classification of stillbirth by relevant condition at death (ReCoDe): population based cohort study. BMJ. 2005;331(7525):1113-7. DOI:10.1136/bmj.38629.587639.7C for death prevention and evaluated cases with complete clinical and pathological data (64.0% of the total) from a university hospital for a period of 10 years.⁴¹41 Vieira MSM, Siebert EC, Ceglio WQGW, Almeira MH, Batista TS, Freitas PF. Dificuldades para a identificação da causa do óbito fetal: como resolver?Rev Bras Ginecol Obstet. 2012;34(9):403-8. DOI:10.1590/S0100-72032012000900003 The clinical evaluation indicated that malformations, maternal hypertension, and placental disorders were the underlying causes of fetal death, while the pathological examination revealed infection, malformations, and placental disorders. The two classifications complemented each other, decreasing the percentage of ill-defined causes. The ReCoDe classification had the best performance, with IGR as the cause of death in 32.5% cases and only 9.9% of non-defined deaths.

In the city of Mombaça, CE, Northeastern Brazil, only 28 hospital-based fetal deaths were reported, with a poor level of completeness of medical records, and the underlying cause of fetal death was not identified in more than 90.0% of cases.³⁵35 Silva Filho FP, Maia MCG. Determinantes de natimortalidade em um serviço municipal de assistência hospitalar. Rev Bras Promoc Saude. 2004;17(4):187-92.

In the city of Salvador, BA, Northeastern Brazil, a time-series study showed that fetal death was the main cause of perinatal death, with a small tendency toward decline. There was a poor level of completeness for data on fetal deaths in MIS, particularly data associated with maternal age and education (40.0% and 17.0%, respectively). The most frequent cause of fetal death was intrauterine hypoxia, represented by 44.7% of the cases.¹⁹19 Jacinto E, Aquino EML, Mota ELA. Mortalidade perinatal no município de Salvador, Bahia: evolução de 2000 a 2009. Rev Saude Publica.2013;47(5):846-53. DOI:10.1590/S0034-8910.2013047004528

The sex ratio of fetal deaths in Brazil was estimated at 1,188, according to maternal and fetal characteristics.⁸8 Chiavegatto Filho ADP, Laurenti R. O sexo masculino vulnerável: razão de masculinidade entre os óbitos fetais brasileiros. Cad Saude Publica. 2012;28(4):720-8. DOI:10.1590/S0102-311X2012000400011 This ratio was higher than that for live births and even higher among mothers at age extremes, among mothers with low education, in case of a short gestational period, and for low-birth-weight fetuses. In addition, the sex ratio was high for most underlying causes, suggesting the vulnerability of male fetuses.

Veloso et al⁴⁰40 Veloso HJF, Silva AAM, Barbieri MA, Goldani MZ, Lamy Filho F, Simões VMF, et al. Secular trends in the rate of low birth weight in Brazilian State Capitals in the period 1996 to 2010. Cad Saude Publica. 2013;29(1):91-101. DOI:10.1590/S0102-311X2013000100011 evaluated the tendency of low birth weight between 1996 and 2010 in the Brazilian capitals and found a negative correlation between the percentage of low weight (which increased until 2004) and FMR. FMR decreased from 14.1 per 1,000 live births to 9.3 per 1,000 births, and in 2011, this indicator was higher in the Northeast (12.9 per 1,000 births) and lower in the South (8.5 per 1,000 births). The annual percentage of decrease in FMR was heterogeneous among the different Brazilian regions: North (-2.68%), Northeast (-1.17%), Midwest (-2.13%), and South (-3.83%). In the Southeast, the annual decrease was 4.5% by 2005, stabilizing thereafter.

Associated factors

Eleven studies were identified: five in the Southeast, two in the Northeast, three in the South, and one covering the entire Brazilian territory. Most studies used hierarchical multivariate analysis, and the factors investigated were maternal age, education, employment (type and employment status), ethnicity, psychosocial factors (presence of a partner, domestic violence), smoking and alcohol abuse, reproductive characteristics (parity, previous outcomes), prenatal care, maternal morbidity (hypertension, diabetes, bleeding), maternal nutritional status, physical activity, and biological variables of the fetus (weight, gestational age, intrauterine growth).

In Juiz de Fora, MG, Southeastern Brazil,²⁸28 Neves LAT, Silva LGP, Luz TP, Reis AF, Alves MJM. Fatores de risco para natimortalidade na cidade de Juiz de Fora. Rev Med Minas Gerais. 2004;14(3):151-7. for stillbirths aged more than 28 weeks, maternal morbidities (hypertension, diabetes, and bleeding in the third trimester) and the presence of fewer than six prenatal consultations were positively correlated with fetal death.

In the city of Sao Paulo,¹1 Almeida MF, Alencar GP, Novaes HMD, França Jr I, Siqueira AAF, Campbell OMR, et al. Risk factors for antepartum fetal death in the city of São Paulo, Brazil. Rev Saude Publica. 2007;41(1):35-43. DOI:10.1590/S0034-89102007000100006 fetal deaths prior to labor were significantly correlated with the following variables: education < 4 years (OR = 2.0), being single (OR = 2.7), previous child with low birth weight (OR = 2.4), gestational bleeding (OR = 6.6), hypertension (OR = 5.9), diabetes (OR = 13.2), inadequate prenatal care (OR = 2.4 ), malformation (OR = 3.7), and IGR (OR = 5.1). The variable “gestational age” was considered to be a consequence and not a determinant of fetal death and was therefore not included. The most attributable fractions were inadequate prenatal care, IGR, maternal hypertension, and marital instability. A correlation between maternal hypertension and prenatal care was observed. Accordingly, the odds ratio among hypertensive women with inadequate prenatal care was 8.7, which was much higher than that for hypertensive women with adequate prenatal care (OR = 5.6).

In addition, in Sao Paulo, the correlation between traffic pollution and perinatal death was evaluated.²⁷27 Medeiros APP, Gouveia N, Machado RPP, Souza MR, Alencar GP, Novaes HMD, et al. Traffic-related air pollution and perinatal mortality: a case-control study. Environ Health Perspect.2009;117(1):127-32. DOI:10.1289/ehp.11679 However, for fetal death in isolation, no correlation was found between these variables.

In a SUS maternity hospital in the city of Rio de Janeiro,¹⁶16 Fonseca SC, Coutinho ESF. Fatores de risco para mortalidade fetal em uma maternidade do Sistema Único de Saúde, Rio de Janeiro, Brasil: estudo de caso-controle. Cad Saude Publica. 2010;26(2):240-52. DOI:10.1590/S0102-311X2010000200004 the following protective factors were identified: marital stability, employment stability, being accompanied upon admission to the maternity hospital, and adequate prenatal care. The risk factors were maternal morbidity, IGR, previous stillbirth, and violence during pregnancy. Weight and gestational age were not included because these variables are not conceptually involved in the genesis of fetal death.

Another study conducted across all the maternity hospitals in the city of Rio de Janeiro³⁰30 Oliveira EFV, Gama SGN, Silva CMFP. Gravidez na adolescência e outros fatores de risco para a mortalidade fetal e infantil no Município do Rio de Janeiro, Brasil. Cad Saude Publica. 2010;26(3):567-78. DOI:10.1590/S0102-311X2010000300014 investigated adolescence and fetal and infant outcomes. However, no correlation was found between fetal deaths and adolescence. Hypertension and pre-existing diabetes, gestational age < 37 weeks, and birth weight < 2,500 g increased the risk of fetal death, whereas adequate prenatal care played a protective role.

The cohorts in Pelotas, Rio de Janeiro, were also investigated with regard to the correlation between adolescence and risk of fetal death.³²32 Restrepo-Méndez MC, Barros AJD, Santos IS, Menezes AMB, Matijasevich A, Barros FC, et al. Childbearing during adolescence and offspring mortality: findings from three population-based cohorts in southern Brazil. BMC Public Health. 2011;11:781. DOI:10.1186/1471-2458-11-781 In the three cohorts, adolescence suggested protection against fetal death, without statistical significance. FMR among adolescent mothers decreased from 9.8 per 1,000 births in 1982 to 7.4 per 1,000 births in 2004.

In a SUS maternity (50.0% of health care coverage) in Caxias do Sul, RS, Southern Brazil,²¹21 Klein CJ, Madi JM, Araújo BF, Zatti H, Dal Bosco DS, Henke CN, et al. Fatores de risco relacionados à mortalidade fetal. Rev AMRIGS. 2012;51(1):11-6. the associated factors were the presence of fewer than six prenatal consultations, previous stillbirths, hypertensive syndrome, and malformations.

In Rio Grande, RS,¹²12 Dumith SC, Domingues MR, Mendoza-Sassi RA, Cesar JA. Atividade física durante a gestação e associação com indicadores de saúde materno-infantil. Rev Saude Publica. 2012;46(2):327-33. DOI:10.1590/S0034-89102012005000012 physical activity during pregnancy was correlated with several maternal and perinatal outcomes. The association was protective against fetal death (OR = 0.43) but reached a borderline significance. This was the only study to address this factor.

In a high-risk maternity hospital in Recife, PE, Northeastern Brazil,⁴4 Andrade LG, Amorim MMR, Cunha ASC, Leite SRF, Vital SA. Fatores associados à natimortalidade em uma maternidade escola em Pernambuco: estudo caso-controle. Rev Bras Ginecol Obstet. 2009;31(6):285-92. DOI:10.1590/S0100-72032009000600004 the following factors were correlated with fetal death: having fewer than six prenatal consultations, gestational hemorrhagic syndrome, medical consultation before admission to the maternity hospital, and congenital malformations. The variable “maternal age” was questionable (OR = 2.2; 95%CI 1.0;4.9). The sex ratio was 1:1, and the percentage of indeterminate underlying causes was high (53.3%).

Two ecological studies evaluated the association of agrochemicals/pesticides with fetal death, with conflicting results. A time-series study conducted in the lower Jaguaribe area of Ceará state³³33 Rigotto RM, Silva AMC, Ferreira MJM, Rosa IF, Aguiar ACP. Tendências de agravos crônicos à saúde associados a agrotóxicos em região de fruticultura no Ceará, Brasil. Rev Bras Epidemiol. 2013;16(3):763-73. DOI:10.1590/S1415-790X2013000300019 between 2000 and 2010 showed a gradual increase in fetal death associated with exposure to pesticides. Another study addressed infant and fetal death in Brazil in 2001³⁷37 Siqueira MT, Braga C, Cabral-Filho JE, Augusto LGS, Figueiroa JN, Souza AI. Correlation between pesticide use in agriculture and adverse birth outcomes in Brazil: an ecological study. Bull Environ Contam Toxicol. 2010;84(6):647-51. DOI:10.1007/s00128-010-0027-8 and found no correlation between these two variables and fetal death.

DISCUSSION

The FMR decreased in almost all the studies evaluated; however, this decrease was heterogeneous. Among SUS users, it became evident that fetal mortality was higher in groups with a lower socioeconomic status. Therefore, it is possible that universal access to health care has not been achieved, indicating inequalities in health. In the studies that described weight ranges, the percentage of stillborns with weight ≥ 2,500 g varied between 24.0% and 40.0%. According to Lansky, these fetuses have a great potential for survival, highlighting the seriousness of the problem in Brazil.^cc Lansky S. fetal mortality: invisible and preventable deaths. In: Bittencourt DAS, Dias MAB, Wakimoto MD, organizadores. Vigilância do óbito materno, infantil e fetal e atuação em comitês de mortalidade. Rio de Janeiro (RJ): EAD/ENSP; 2013.

Among the 27 studies published between 2003 and 2013, only 13 specifically focused on fetal death, supporting the low visibility of this topic despite its great importance.^cc Lansky S. fetal mortality: invisible and preventable deaths. In: Bittencourt DAS, Dias MAB, Wakimoto MD, organizadores. Vigilância do óbito materno, infantil e fetal e atuação em comitês de mortalidade. Rio de Janeiro (RJ): EAD/ENSP; 2013. As with other topics related to maternal and child health,¹⁴14 Fonseca SC, Coutinho ESF. Pesquisa sobre mortalidade perinatal no Brasil: revisão da metodologia e dos resultados. Cad Saude Publica. 2004;20(Supl 1):S7-19. DOI:10.1590/S0102-311X2004000700002^,3636 Silveira MF, Matijasevich A, Horta BL, Bettiol H, Barbieri MA, Silva AA, et al. Prevalência de nascimentos pré-termo por peso ao nascer: revisão sistemática. Rev Saude Publica. 2013;47(5):992-1003. DOI:10.1590/S0034-8910.2013047004997 the studies predominantly addressed the Southeast region, followed by the South region. These regions have lower FMR but are home to the greatest number of research institutions. The evaluated studies were mostly published in public health journals, indicating that other topics are prioritized in periodicals that address obstetrics and maternal and child health.

The quality of information contained in fetal death certificates, despite some improvement, remains deficient. A poor level of completeness of data on sociodemographic variables, including age and maternal education, was observed. Therefore, the use of MIS to evaluate the correlation between social inequalities and fetal deaths is unfeasible, as previously noted by Romero & Cunha³⁴34 Romero DE, Cunha CB. Avaliação da qualidade das variáveis socioeconômicas e demográficas dos óbitos de crianças menores de um ano registrados no Sistema de Informações sobre Mortalidade do Brasil (1996/2001).Cad Saude Publica. 2006;22(3):673-84. DOI:10.1590/S0102-311X2006000300022 in relation to infant deaths in Brazil. Notably, the level of completeness of fetal death certificates was poorer than that reported by these authors for death certificates of infants aged under one year, corroborating the low visibility of fetal deaths.^cc Lansky S. fetal mortality: invisible and preventable deaths. In: Bittencourt DAS, Dias MAB, Wakimoto MD, organizadores. Vigilância do óbito materno, infantil e fetal e atuação em comitês de mortalidade. Rio de Janeiro (RJ): EAD/ENSP; 2013. Although cases of fetal deaths have decreased, underreporting occurs. However, data retrieval using the HIS-SUS system is possible, at least considering births from women using this health care system.³¹31 Rafael RAA, Ribeiro VS, Cavalcante MCV, Santos AM, Simões VMF. Relacionamento probabilístico: recuperação de informações de óbitos infantis e natimortos em localidade do Maranhão, Brasil. Cad Saude Publica. 2011;27(7):1371-9. DOI:10.1590/S0102-311X2011000700012

Evaluation of the underlying causes is challenging, considering that the percentage of ill-defined causes is high and that although it was assessed in a single study, agreement was low. The recurrent diagnosis of intrauterine hypoxia, despite its presence as a constant cause of death in ICD-10, barely contributes to the understanding of the phenomenon. In addition, most studies did not even specify whether hypoxia occurred before or during birth and simply reported the code P20, a limitation that had already been indicated by Lansky.^cc Lansky S. fetal mortality: invisible and preventable deaths. In: Bittencourt DAS, Dias MAB, Wakimoto MD, organizadores. Vigilância do óbito materno, infantil e fetal e atuação em comitês de mortalidade. Rio de Janeiro (RJ): EAD/ENSP; 2013. Studies that investigated the underlying causes by review of either medical records or autopsy records indicated two findings: the contribution of IGR as a proximal factor and the increased allocation of maternal causes, including hypertensive diseases, diabetes, and syphilis, as the cause of fetal death.

In addition to the ongoing training of doctors in relation to the completion of death certificates, fetal and infant death investigation committees are important for the improvement of data quality information systems.^aa Ministério da Saúde, Secretaria de Vigilância em Saúde, Secretaria de Atenção à Saúde. Manual de vigilância do óbito infantil e fetal e do Comitê de Prevenção do Óbito Infantil e Fetal. 2.ed. Brasília (DF); 2009 [cited 2012 Apr 10]. (Série A. Normas e Manuais Técnicos). Available from: http://bvsms.saude.gov.br/bvs/publicações/manual_obito_infantil_fetal_2ed.pdf Although the original death certificates cannot be changed, it is possible to correct the data on MIS after review of the death certificates. In the WHO report^ff World Health Organization, Institutional Repository for Information Sharing - IRIS. Strengthening civil registration and vital statistics for births, Deaths and causes of death: resource kit. Geneva; 2013 [cited 2014 Jan 20]. Available from: http://apps.who.int/iris/bitstream/10665/78917/1/9789241504591_eng.pdf on strengthening civil registration and vital statistics, the use of data on causes of death to evaluate health at the local, regional, and national levels is emphasized.

The classifications of death prevention have hardly been used, and Wigglesworth’s classification,⁴²42 Wigglesworth JS. Monitoring perinatal mortality.Lancet. 1980;316(8196):684-6. DOI:10.1016/S0140-6736(80)92717-8 adapted by Kelling, was predominantly used.²⁰20 Keeling JW, MacGillivray I, Golding J, Wigglesworth J, Berry J, Dunn PM. Classification of perinatal death. Arch Dis Child.1989;64(10 Spec No):1345-51. DOI:10.1136/adc.64.10_Spec_No.1345Analysis of death prevention is important because it can guide managers to important resources for improving prenatal care and childbirth. The most frequent category was antepartum fetal death, suggesting greater deficiency in prenatal care. However, few studies have investigated the history of patients before their admission to the maternity ward where fetal death was confirmed, preventing the confirmation of this finding. This limitation of Wigglesworth’s classification had already been noted; therefore, other studies are necessary to better define the context of fetal death.

With regard to the study of determinants, the case-control design and hierarchical analysis were predominant. A review by Di Mario¹⁰10 Di Mario S, Say L, Lincetto O. Risk factors for stillbirth in developing countries: a systematic review of the literature. Sex Transm Dis. 2007;34(7 Suppl):S11-21. DOI:10.1097/01.olq.0000258130.07476.e3 on fetal death in developing countries between 1996 and 2006 indicated a predominance of case-control studies, whereas in the review of Flenady et al¹³13 Flenady V, Koopmans L, Middleton P, Frøen JF, Smith GC, Gibbons K, et al. Major risk factors for stillbirth in high-income countries: a systematic review and meta-analysis. Lancet. 2011;377(9774):1331-40. DOI:10.1016/S0140-6736(10)62233-7
https://doi.org/10.1016/S0140-6736(10)62... conducted in developed countries, approximately 70.0% of the studies were of the retrospective cohort type, using vital statistics databases.

Among the socioeconomic and demographic determinants of the mother, a low education level and advanced age were associated with fetal death in some studies. In the study by McClure²⁶26 McClure EM, Pasha O, Goudar SS, Chomba E, Garces A, Tshefu A, et al. Epidemiology of stillbirth in low-middle income countries: a Global Network Study. Acta Obstet Gynecol Scand. 2011;90(12):1379-85. DOI:10.1111/j.1600-0412.2011.01275.x and in the aforementioned reviews,¹⁰10 Di Mario S, Say L, Lincetto O. Risk factors for stillbirth in developing countries: a systematic review of the literature. Sex Transm Dis. 2007;34(7 Suppl):S11-21. DOI:10.1097/01.olq.0000258130.07476.e3^,1313 Flenady V, Koopmans L, Middleton P, Frøen JF, Smith GC, Gibbons K, et al. Major risk factors for stillbirth in high-income countries: a systematic review and meta-analysis. Lancet. 2011;377(9774):1331-40. DOI:10.1016/S0140-6736(10)62233-7
https://doi.org/10.1016/S0140-6736(10)62... maternal age was associated with fetal death, with a high attributable fraction in developed countries, where the prevalence of pregnant women aged more than 35 years is high. A low education level was also correlated with fetal death in these reviews.

Ethnicity and working conditions were addressed less frequently, and only one study (Fonseca & Coutinho, 2010)¹⁰10 Di Mario S, Say L, Lincetto O. Risk factors for stillbirth in developing countries: a systematic review of the literature. Sex Transm Dis. 2007;34(7 Suppl):S11-21. DOI:10.1097/01.olq.0000258130.07476.e3indicated that employment had a protective effect. Relationship with a partner was associated with fetal death in three studies. Considering that the prevalence of women with no partner or those in unstable relationships can be high among those with a lower socioeconomic status, the fraction attributable to this factor can be high, as observed in the study conducted in the city of Sao Paulo,¹1 Almeida MF, Alencar GP, Novaes HMD, França Jr I, Siqueira AAF, Campbell OMR, et al. Risk factors for antepartum fetal death in the city of São Paulo, Brazil. Rev Saude Publica. 2007;41(1):35-43. DOI:10.1590/S0034-89102007000100006 where the attributable fraction was 26.0%. In a multicenter study conducted in Latin America, the correlation between lack of a partner and fetal death was borderline (RR = 1.12; 95%CI 1.0;1.25).⁹9 Conde-Agudelo A, Belizán JM, Díaz-Rossello JL. Epidemiology of fetal death in Latin America. Acta Obstet Gynecol Scand. 2000;79(5):371-8. DOI:10.1034/j.1600-0412.2000.079005371.x Domestic violence has not been widely investigated; therefore, the results are inconclusive. Ethnicity did not show any correlation with fetal death in our review, and this topic has not been addressed in recent international reviews or in multicenter studies.⁹9 Conde-Agudelo A, Belizán JM, Díaz-Rossello JL. Epidemiology of fetal death in Latin America. Acta Obstet Gynecol Scand. 2000;79(5):371-8. DOI:10.1034/j.1600-0412.2000.079005371.x^,1010 Di Mario S, Say L, Lincetto O. Risk factors for stillbirth in developing countries: a systematic review of the literature. Sex Transm Dis. 2007;34(7 Suppl):S11-21. DOI:10.1097/01.olq.0000258130.07476.e3^,1313 Flenady V, Koopmans L, Middleton P, Frøen JF, Smith GC, Gibbons K, et al. Major risk factors for stillbirth in high-income countries: a systematic review and meta-analysis. Lancet. 2011;377(9774):1331-40. DOI:10.1016/S0140-6736(10)62233-7
https://doi.org/10.1016/S0140-6736(10)62... ^,2626 McClure EM, Pasha O, Goudar SS, Chomba E, Garces A, Tshefu A, et al. Epidemiology of stillbirth in low-middle income countries: a Global Network Study. Acta Obstet Gynecol Scand. 2011;90(12):1379-85. DOI:10.1111/j.1600-0412.2011.01275.x

Prenatal care evaluated using either quantitative measurements such as the number of visits or adequacy criteria such as the Adequacy of Prenatal Care Utilization Index²²22 Kotelchuck M. An evaluation of the Kessner Adequacy of Prenatal Care Index and a Proposed Adequacy of Prenatal Care Utilization Index. Am J Public Health. 1994;84(9):1414-21. DOI:10.2105/AJPH.84.9.1414 was a strong correlation factor: this variable plays a protective role when the number of prenatal consultations is adequate but increases the risk of fetal death when the number of consultations is low. The correlation with maternal hypertension was shown by Almeida et al,¹1 Almeida MF, Alencar GP, Novaes HMD, França Jr I, Siqueira AAF, Campbell OMR, et al. Risk factors for antepartum fetal death in the city of São Paulo, Brazil. Rev Saude Publica. 2007;41(1):35-43. DOI:10.1590/S0034-89102007000100006 indicating how prenatal care can affect fetal death.

The study by Almeida et al¹1 Almeida MF, Alencar GP, Novaes HMD, França Jr I, Siqueira AAF, Campbell OMR, et al. Risk factors for antepartum fetal death in the city of São Paulo, Brazil. Rev Saude Publica. 2007;41(1):35-43. DOI:10.1590/S0034-89102007000100006estimated that inadequate prenatal care had the highest attributable fraction (40.0%). This result is consistent with the review by Di Mario,¹⁰10 Di Mario S, Say L, Lincetto O. Risk factors for stillbirth in developing countries: a systematic review of the literature. Sex Transm Dis. 2007;34(7 Suppl):S11-21. DOI:10.1097/01.olq.0000258130.07476.e3 who found that the attributable fraction for absent or poor prenatal care varied between 15.0% and 72.0%. In addition, Flenady et al¹³13 Flenady V, Koopmans L, Middleton P, Frøen JF, Smith GC, Gibbons K, et al. Major risk factors for stillbirth in high-income countries: a systematic review and meta-analysis. Lancet. 2011;377(9774):1331-40. DOI:10.1016/S0140-6736(10)62233-7
https://doi.org/10.1016/S0140-6736(10)62... found a strong correlation between lack of prenatal care and fetal death (OR = 3.3). However, the attributable fraction was low, probably because of the high coverage of prenatal care in developed countries. In the study by McClure et al²⁶26 McClure EM, Pasha O, Goudar SS, Chomba E, Garces A, Tshefu A, et al. Epidemiology of stillbirth in low-middle income countries: a Global Network Study. Acta Obstet Gynecol Scand. 2011;90(12):1379-85. DOI:10.1111/j.1600-0412.2011.01275.x conducted in seven low- and middle-income countries, the lack of prenatal care showed the highest correlation among the evaluated factors.

Unfavorable reproductive history of the mother was thoroughly investigated, particularly previous stillbirths, multiparity, and morbidities, and although no consensus was found, this analysis suggests a positive correlation with fetal death, which is corroborated by international reviews.¹⁰10 Di Mario S, Say L, Lincetto O. Risk factors for stillbirth in developing countries: a systematic review of the literature. Sex Transm Dis. 2007;34(7 Suppl):S11-21. DOI:10.1097/01.olq.0000258130.07476.e3^,1313 Flenady V, Koopmans L, Middleton P, Frøen JF, Smith GC, Gibbons K, et al. Major risk factors for stillbirth in high-income countries: a systematic review and meta-analysis. Lancet. 2011;377(9774):1331-40. DOI:10.1016/S0140-6736(10)62233-7
https://doi.org/10.1016/S0140-6736(10)62... With regard to fetal characteristics, the male gender showed vulnerability in the study of Chiavegato et al;⁸8 Chiavegatto Filho ADP, Laurenti R. O sexo masculino vulnerável: razão de masculinidade entre os óbitos fetais brasileiros. Cad Saude Publica. 2012;28(4):720-8. DOI:10.1590/S0102-311X2012000400011 however, this result was not corroborated by other studies. Few studies have addressed weight and gestational age, and most included only IGR, defending the causal relationship of the latter, whereas weight and gestational age would be consequences of the process. The international reviews also identified IGR as an important factor, with the attributable fraction being greater than 20.0%.

Maternal nutritional status, which has not been frequently studied in Brazil,¹1 Almeida MF, Alencar GP, Novaes HMD, França Jr I, Siqueira AAF, Campbell OMR, et al. Risk factors for antepartum fetal death in the city of São Paulo, Brazil. Rev Saude Publica. 2007;41(1):35-43. DOI:10.1590/S0034-89102007000100006^,2727 Medeiros APP, Gouveia N, Machado RPP, Souza MR, Alencar GP, Novaes HMD, et al. Traffic-related air pollution and perinatal mortality: a case-control study. Environ Health Perspect.2009;117(1):127-32. DOI:10.1289/ehp.11679 showed no correlation with fetal death. However, obesity was the modifiable factor with the highest attributable fraction in developed countries.¹³13 Flenady V, Koopmans L, Middleton P, Frøen JF, Smith GC, Gibbons K, et al. Major risk factors for stillbirth in high-income countries: a systematic review and meta-analysis. Lancet. 2011;377(9774):1331-40. DOI:10.1016/S0140-6736(10)62233-7
https://doi.org/10.1016/S0140-6736(10)62...

Smoking, identified as a modifiable factor with a high attributable fraction in populations with a lower socioeconomic status in developed countries,¹³13 Flenady V, Koopmans L, Middleton P, Frøen JF, Smith GC, Gibbons K, et al. Major risk factors for stillbirth in high-income countries: a systematic review and meta-analysis. Lancet. 2011;377(9774):1331-40. DOI:10.1016/S0140-6736(10)62233-7
https://doi.org/10.1016/S0140-6736(10)62... was not widely explored in national studies and was not a determinant of fetal death, probably because of the difficulty in measuring this variable or its absence in vital statistics.

Although our results included a low number of studies, which prevented the realization of a meta-analysis, they provide directions for the investigation of fetal death in Brazil and for investment in public policies.

Factors such as ethnicity, obesity, smoking (and quitting smoking), and intrauterine growth, which have been given little attention, should be addressed in future studies. On the other hand, there is a sufficient amount of scientific evidence on the influence of socioeconomic factors and quality of prenatal care in fetal mortality. Although these two factors may not have a direct biological correlation with fetal death, it is reasonable to assume that providing better prenatal care to vulnerable women and/or to those with adverse reproductive history may decrease FMR in Brazil.

The quality of information on fetal deaths, which allows the realization of epidemiological studies with greater internal validity, and investment in research committees should be prioritized so that fetal mortality can gain visibility as well as to ensure that these problems will be confronted more effectively.

REFERENCES

Publication Dates

History