Risk factors for sudden infant death syndrome in a developing country

OBJECTIVE: To analyze whether previously identifi ed risk factors for sudden death syndrome have a signifi cant impact in a developing country. METHODS: Retrospective longitudinal case-control study carried out in Porto Alegre, Southern Brazil. Cases (N=39) were infants born between 1996 and 2000 who died suddenly and unexpectedly at home during sleep and were diagnosed with sudden death syndrome. Controls (N=117) were infants matched by age and sex who died in hospitals due to other conditions. Data were collected from postmortem examination records and questionnaires answers. A conditional logistic model was used to identify factors associated with the outcome. RESULTS: Mean age at death of cases was 3.2 months. The frequencies of infants regarding gestational age, breastfeeding and regular medical visits were similar in both groups. Sleeping position for most cases and controls was the lateral one. Supine sleeping position was found for few infants in both groups. Maternal variables, age below 20 years (OR=2, 95% CI: 1.1; 5.1) and smoking of more than 10 cigarettes per day during pregnancy (OR=3, 95% CI: 1.3; 6.4), signifi cantly increased the risk for the syndrome. Socioeconomic characteristics were similar in both groups and did not affect risk. CONCLUSIONS: Infant-maternal and socioeconomic profi les of cases in a developing country closely resembled the profi le described in the literature, and risk factors were similar as well. However, individual characteristics were identifi ed as risks in the population studied, such as smoking during pregnancy and maternal age below 20 years. DESCRIPTORS: Sudden Infant Death. Prone Position. Smoking. Risk Factors. Case-Control Studies. 397 Rev Saúde Pública 2008;42(3):396-401 Sudden infant death syndrome (SIDS) is defi ned as “the sudden death of any infant under one year of age which remains unexplained after a thorough case investigation, including complete postmortem examination, review of clinical history and examination of death scene.” 23 The pathology of SIDS has not been established, but many hypotheses have been raised in the last years.19 Advances in the knowledge about SIDS are associated with the identifi cation of environmental risk factors, the so-called “triggers,” that may favor this lethal outcome in an organically predisposed child.7 Currently, many risk factors are associated with SIDS, some of which are defi nitely correlated with this syndrome. The implications of other factors, however, still need to be further explored. Prone sleeping position and maternal smoking are wellestablished risk factors for SIDS in developed countries. A 50% reduction in SIDS incidence in countries that have adopted the recommendation of supine sleeping position is well documented in the literature.2,4 Pediatricians and health providers in developing countries, such as Brazil, have little knowledge on risk RESUMO OBJETIVO: Analisar se fatores de risco previamente identifi cados para a síndrome da morte súbita do lactente têm impacto signifi cativo em um país em desenvolvimento. MÉTODOS: Estudo retrospectivo, longitudinal, de caso-controle pareado realizado em Porto Alegre, RS. Os casos (N=39) foram lactentes nascidos entre 1996 e 2000 que morreram subitamente e inesperadamente em casa, ao dormir, com diagnóstico fi nal da síndrome; o grupo controle (N=117) foi composto de lactentes pareados por idade e sexo que morreram em hospitais, por outra doença. Os dados foram coletados de registros médicos postmortem e respostas a questionários. Utilizou-se modelo de regressão logística para identifi car os fatores associados ao desfecho. RESULTADOS: A idade média dos casos no momento da morte dos lactentes foi de 3,2 meses. As freqüências de idade gestacional, amamentação e visitas regulares a unidades básicas de saúde foram similares nos dois grupos. A posição de dormir mais comum nos casos e controles foi a lateral. A posição supina foi encontrada em alguns lactentes em ambos grupos. As variáveis maternas idade inferior a 20 anos (OR=2, IC 95%:1,1;5,1) e consumo de mais de 10 cigarros/dia durante a gravidez (OR=3, IC 95%: 1,3;6,4) aumentaram signifi cativamente o risco para a síndrome. As características socioeconômicas foram similares entre os grupos e não afetaram o risco. CONCLUSÕES: Os perfi s lactente-materno e os fatores de risco foram similares aos encontrados em outros países. Todavia, foram identifi cadas características individuais quanto aos riscos de fumar durante a gravidez e da idade materna inferior a 20 anos. DESCRITORES: Morte Súbita do Lactente. Decúbito Ventral. Tabagismo. Fatores de Risco. Estudos de Casos e Controles.

Sudden infant death syndrome (SIDS) is defi ned as "the sudden death of any infant under one year of age which remains unexplained after a thorough case investigation, including complete postmortem examination, review of clinical history and examination of death scene." 23he pathology of SIDS has not been established, but many hypotheses have been raised in the last years. 19dvances in the knowledge about SIDS are associated with the identifi cation of environmental risk factors, the so-called "triggers," that may favor this lethal outcome in an organically predisposed child. 7Currently, many risk factors are associated with SIDS, some of which are definitely correlated with this syndrome.The implications of other factors, however, still need to be further explored.Prone sleeping position and maternal smoking are wellestablished risk factors for SIDS in developed countries.A 50% reduction in SIDS incidence in countries that have adopted the recommendation of supine sleeping position is well documented in the literature. 2,4diatricians and health providers in developing countries, such as Brazil, have little knowledge on risk RESUMO OBJETIVO: Analisar se fatores de risco previamente identifi cados para a síndrome da morte súbita do lactente têm impacto signifi cativo em um país em desenvolvimento.

INTRODUCTION
factors for SIDS since it is not seen as an important cause of death in the fi rst year of life. 15Therefore, the incidence of SIDS is unknown in Brazil, except in the city of Porto Alegre, southern Brazil.A previous study reported that SIDS affects 0.4:1000 live births, and is the probable cause of 6.3% of infant deaths before 12 months of age. 15nce risk factors for SIDS may vary or gain greater importance depending on the population studied, 11,20 developing countries also need to know the impact of each risk factor on its own population so that preventive measures can be implemented.
The present study aimed at investigating whether previously identifi ed risk factors for sudden infant death syndrome have a signifi cant impact in developing regions.It is hypothesized that risk factors for SIDS should be similar to those found for non-SIDS cases in developing countries since they are closely associated with poverty.

METHODS
The local Health Department of the municipality of Porto Alegre (Southern Brazil) keeps a Mortality Surveillance Program named Prá-Viver.The program expert team visits all families of children who died in the fi rst year of life and information is collected on the history of disease and death.Most parents agree to participate in the program, and the refusal rate is 2.5%.The questionnaire consists of questions about well-established SIDS risk factors and general risk factors for infant mortality in Brazil.SIDS diagnosis was based on an extensive review of clinical history, family report on the circumstances of death, and negative postmortem examination, following the US National Institute of Child Health and Human Development (NICHD) defi nition. 7Autopsy reports were provided by a medical examiner of the Forensic Medical Institute of Porto Alegre and included macroscopic examination of all organs.
Records from the program were retrospectively reviewed and all children who died under 12 months of age in the city between January 1996 and December 2000 (N=837) were identifi ed.Data from infants who died during sleep at home (N=142) were selected.Out of these deaths, 39 were identifi ed as SIDS cases and included in the study.The control group of infants (N=117) was selected from the same database (Prá-Viver Program) and matched by sex and age at a 3:1 ratio to SIDS cases.Controls died in the same months as SIDS cases, and causes of death were clearly defi ned during hospital stay.All children with congenital malformations were excluded from the study.
To establish risk factors for SIDS, the study variables were divided into three groups: maternal, infant, and socioeconomic factors.
Maternal factors studied were: age below 20 years (adolescent mother; yes or no), interpregnancy interval (<1 year = short interval; yes or no), prenatal care (no if <6 visits), vaginal delivery (yes or no), smoking during pregnancy (more than 10 cigarette per day = proven risk for SIDS), 6,23 alcohol or marijuana use during pregnancy (yes or no) and urinary infection during pregnancy (yes or no).
Infant factors assessed were gestational age (delivery <37 weeks), Apgar score (risk if <7), breastfeeding (at time of death), regular pediatric care (at least one visit per month), usual sleeping position (lateral, prone or supine), apnea (yes or no), SIDS sibling (yes or no), ethnicity (white, black, Asian).Socioeconomic factors included: weather conditions (death during a cold or hot month), income (low if <1 minimum wage), father and mother schooling (low if <4 years), father living with the family (yes or no).
The measure of association in the univariate analysis was odds ratio (OR) and 95% confi dence interval (95% CI).A multivariate analysis was subsequently carried out to identify the best model to explain risk factors.For that purpose, a conditional logistic model was used for the variables statistically associated with SIDS in the univariate analysis (p≤0.05), as well as for the most important variable reported in the literature -sleeping position.
All tests were carried out using the SPSS 9.0.
The study protocol was approved by the Research Ethics Committees of "Pontifi cia Universidade do Rio Grande do Sul" University Hospital, the "Instituto de Medicina Forense of Porto Alegre", and the "Secretaria de Saúde do Município de São Paulo".

RESULTS
Age at death of SIDS cases ranged from 1 to 9 months (median=2 months), and male sex was slightly predominant at a 1:1.4 ratio.Birthweight in the SIDS group ranged from 1,200 to 3,600 g (median=2,800 g).In the control group, birthweight ranged from 700 to 4,300 g (mean=2,600 ± 800, median=2,650 g).
Gestational age below 37 weeks, breastfeeding and regular medical visits were similar in both groups.Previous episodes of apnea were more frequently seen among controls, but there was no signifi cant difference between the groups.No SIDS siblings were identifi ed in the control group.None of the SIDS cases had 5-min Apgar score below 7 (Table 1).
Most SIDS cases and controls regularly slept in the lateral position.Although lateral position was not identifi ed as a risk factor for SIDS, it was also not protective since 67% of SIDS cases regularly slept in the lateral position.Few infants regularly slept in supine position in both groups.Although not signifi cant, there was a predominance of prone sleeping position among SIDS cases (Table 1).
The analysis of maternal variables showed that age below 20 years and smoking more than 10 cigarettes per day during pregnancy significantly increased SIDS risk (OR=2.3 and 3 respectively, Table 3).Short interpregnancy intervals and alcohol abuse were more frequently seen among SIDS mothers, but differences were not signifi cant between the groups.Prenatal care and route of delivery were similar in both groups.Urinary tract infection and marijuana use were more prevalent among control mothers, but the difference was not statistically signifi cant (Table 2).Socioeconomic characteristics were similar in both groups and did not increase SIDS risk (Table 1).

DISCUSSION
Although SIDS is the main cause of infant mortality in developed countries and more frequently affects low income populations, 1 its characteristics and incidence rates are still not known in developing countries, characterized by very low income population.As infant mortality is still quite high in poor countries and generally associated with infectious diseases and malnutrition, pediatricians and health providers have little knowledge on SIDS.Consequently, SIDS is often misdiagnosed and information about SIDS risk factors is not usually provided to parents. 3,5previous study found that the epidemiological profi le of SIDS cases in Porto Alegre was similar to that described in developed regions.It also reported that none of the infants regularly used to sleeping in supine position, and 71% of mothers smoked during pregnancy.15 Supine sleeping position is not common in Brazil.Parents are generally advised to place their babies to sleep on lateral position.16 Thus, it was not surprising that lateral position was the most prevalent sleeping position in both SIDS and control groups.This position did not increase SIDS risk in the present study.Sleeping in prone position, one of the main risk factor for SIDS, [8][9][10] was similar in both SIDS and non-SIDS cases studied.
The ethnicity variable is a well-known risk factor for SIDS in developed countries where racial miscegenation is modest. 18But, in the present study, it was not confi rmed as a risk factor probably due to the great racial miscegenation in Brazil.
Other well-established risk factors for SIDS, such as gestational age below 37 weeks, low birthweight, low Apgar scores, no breastfeeding, no regular pediatric care, sibling victim of SIDS, and previous episodes of apnea, [12][13][14] were not associated with increased SIDS risk in the present study.In developing countries, these characteristics may perhaps be associated with poverty and increased infant mortality risk regardless of death causes.
Variables of maternal risk factors were associated with SIDS in the population studied.The main risk factor identifi ed was smoking during pregnancy, followed by low maternal age.Cigarette smoking during pregnancy has been repeatedly identifi ed as a major risk factor for SIDS. 6,10,21In the NICHD study, 70% of SIDS mothers smoked during pregnancy. 7Other maternal variables, such as route of delivery, alcohol abuse, use of illicit drugs (marijuana) and urinary tract infection during pregnancy were not associated with increased risk of SIDS. 17,21ne of the socioeconomic characteristics assessed were associated with increased SIDS risk since their prevalence was similar in both groups.This fi nding refl ects the epidemiological pattern of mortality in Brazil.Poor economic, social and cultural conditions in which most people in developing countries live may contribute to increased overall infant mortality but not specifi cally in SIDS. 17,18,24st SIDS studies were conducted in developed countries and reported higher risk in lower socioeconomic populations. 22,25Since most Brazilians have lower socioeconomic conditions, it was hypothesized that SIDS risk factors would be related to poverty and thus would not be specifi cally related to SIDS but rather to overall infant mortality.To test it, it was chosen to study a control group of children deaths.Children who died subsequently to a SIDS case, but the death cause has been clearly determined, were selected.Based on the fi ndings, the study hypothesis was confi rmed since poverty-related SIDS risk factors were similar in both groups.Based on that, it can be assumed that poverty-related SIDS risk factors in developing countries may not be specifi cally related to SIDS but to overall mortality.Further investigation using a control group of live children at 12 months of life may help confi rm these fi ndings.
The results of the present study suggest that, in developing regions, risk factors for SIDS and non-SIDS cases overlap because they are closely associated with poverty.Risk factors for SIDS in Brazil were associated with maternal factors such as low age and smoking during pregnancy.These factors should be targeted in preventive campaigns.

Table 1 .
Frequency of risk factors for sudden infant death syndrome in cases and controls.Porto Alegre, SouthernBrazil, 1996-2000.