Jornal de Pediatria - Vol. 77, Nº1 , 2001 29
Jornal de Pediatria Copyright
© 2001 by Sociedade Brasileira de Pediatria
Sudden infant death syndrome: clinical aspects of an underdiagnosed disease Magda Lahorgue Nunes,1 Ana Paula Silveira Pinho,2 Denise Aerts,3 Ana Sant’ Anna,3 Maurer Pereira Martins,4 Jaderson Costa da Costa5
Abstract Objectives: to investigate the socio-epidemiologic characteristics of the victims of sudden infant death syndrome and to calculate its prevalence during the 1st year of life. Methods: this is an observational, retrospective, and descriptive study based on the review of autopsy protocols and questionnaires applied to families that had cases of infants death during 1997 and 1998 in the city of Porto Alegre. Results: from 335 infants who died, 21 were fulfilled SIDS criteria, although SIDS was not recorded as a cause of death on the death certificate. The infants had no previous apnea or acute disease that could justify the death. Death occurred predominantly in cold months (winter and autumn). None of the infants who died slept in supine position. Families had a low monthly income (3 minimum wages). Seventy one per cent the mothers were smokers and 42% used alcohol during the pregnancy. SIDS prevalence was estimated as 6.3% from the total number of deaths, and the specific mortality coefficient for SIDS was 4.5:10,000 live births. Conclusions: the profile identified for infants at risk is similar to that reported in the literature. However, since SIDS is often misdiagnosed in our community, it does not appear in the statistics about infant mortality. As a consequence, there have never been campaigns to reduce the risk of death due to this syndrome. J Pediatr (Rio J) 2001; 77(1): 29-34: sudden infant death syndrome, apnea, infant mortality.
Introduction The Sudden Infant Death Syndrome (SIDS) is defined as the unexpected death of children younger than 1 year of age that remains unexplained after extensive investigation including clinical history, complete necropsy, and examination of death scene.1 Sudden infant death occurs while the baby is sleeping (at the bed, baby carriage, and so on) and there is no previous indication that the baby’s life was at risk.
Despite SIDS being a well-known disease, which has been reported even in the Bible itself, its cause is still unknown. Various physiopathological mechanisms have been suggested as the cause of SIDS following studies carried out throughout the years. Currently, the most widely accepted hypothesis is related to abnormal sleep arousal associated with presence of other risk factors.2-7 In developed countries, SIDS constitutes the most prevalent cause of death among infants. The age range that is at higher risk for SIDS is from 2 to 5 months of age. Prevalence of SIDS in developing countries, in turn, is still not well known.8 It has been described that incidence of SIDS has been significantly reduced in regions that carry out educational campaigns regarding SIDS risk factors.9
1. Associate Professor of Neurology, Pontifícia Universidade Católica do Rio Grande do Sul (PUCRS) School of Medicine. 2. Graduate student, Graduate Program in Medicine: Neurosciences, PUCRS. 3. Porto Alegre City Department of Health. 4. Undergraduate intern, Neurology Service, Hospital São Lucas, PUCRS. 5. Professor of Neurology, PUCRS School of Medicine.
30 Jornal de Pediatria - Vol. 77, Nº1, 2001 SIDS is diagnosed after assessment of clinical history of patient and elimination of other possible causes of death, following a thorough postmortem examination. Necropsies should be carried out by a pediatric pathologist following the guidelines recommended by the SIDS International.10,11 The National Institute of Child Health and Human Development SIDS cooperative epidemiological study, carried out during the 1970s in the USA with the objective of determining a risk factor profile for SIDS, presented the following results for maternal and perinatal risk factors: young mothers, multiparas, short interpregnancy interval and no prenatal examinations; and premature babies, low weight at birth, sibling of SIDS victims, and low social and economical status. The study also described a prevalence of SIDS among male babies (60%), among specific ethnic groups (Native-American and African-American populations), in specific geographical regions (in the USA, from 1-3:1,000 livebirths and in Australia from 3-7:1,000 livebirths), and during the months of winter.12 More recently, others have described that prone sleeping increases risk for SIDS.13 Based on results of controlled studies, national risk reduction campaigns were developed in different countries with the objective of reducing SIDS. These campaigns focused on changing from sleeping prone to sleeping supine. An immediate reduction in postnatal mortality rates was described (up to 50% in countries that adopted supine sleeping position).14 Despite the fact that some babies are born at higher risk for SIDS, according to the referred epidemiological risk factors, and that the mechanism of death of SIDS is still not well understood, it is possible that providing more adequate sleeping conditions may eliminate risk factors that can trigger SIDS in vulnerable children and, thus, reduce mortality rates. There are no official statistics regarding incidence of SIDS in Brazil. However, a population-based cohort, which followed-up infants in the city of Pelotas, state of Rio Grande do Sul, and which was aimed at determining causes of death during the first year of life estimated that SIDS was the probable cause of death in 4% of its population. 15 Brazil has not participated effectively in international SIDS risk reduction campaigns. This is demonstrated in preliminary results of an international cooperative study (Maternity Advice Study) related to maternity data collected at Brazilian Teaching Hospitals (study data are still being prepared for publishing). Since risk factors for SIDS affect most of our population, we believe that there are cases of SIDS in Brazil. Moreover, because SIDS is not diagnosed, it is not included in official health statistics and, consequently, it is also not considered an important public healthcare problem. In this sense, prevention campaigns, which have been proved effective and are easy to be carried out, are not put to effect and the population at risk for SIDS remains at risk and uninformed. The objective of our study is to understand social and epidemiological characteristics of infants who died of SIDS
Sudden infant death syndrome... - Nunes ML et alii
in our region and determine the prevalence of SIDS as the cause of death during the first year of life.
Patients and methods We carried out an observational, descriptive, and retrospective study based on the review of questionnaires sent to homes in the city of Porto Alegre, state of Rio Grande do Sul, southern Brazil; and on the review of coroner’s reports from the Department of Forensic Medicine of Porto Alegre (DML). The questionnaire had been sent during a program for surveillance of mortality (Prá Viver) by the City Healthcare Department (SMSPA). We reviewed all questionnaires applied during the Prá Viver program from 1997 to 1998. The program consists of door-to-door visits to the homes of families with children who died younger than 1 (mais tarde aparece como 5 anos) year of age. Based on the data provided, we selected the deaths of infants that had occurred at home and while the infant was sleeping. After an extensive review of clinical history, of the place of death, of the cause of death according to death certificate, and of the coroner’s report, we selected probable cases of SIDS for our study. Characterization of risk factor situations for SIDS were determined according to protocols established for maternal, social and cultural, and infant risk factors.12 The prevalence of SIDS as the cause of death was calculated based on the ratio of deaths that may have been caused by the syndrome to the total postneonatal deaths occurred during the two years of the study (152 deaths in 1997, and 183 deaths in 1998). The SIDS-specific mortality rate was calculated according to the ratio of probable SIDS causes of death to total livebirths in the referred period (23,717 in 1997, and 23,193 in 1998). Our study was approved by the Ethics and Research Committee of the Hospital São Lucas at PUCRS and of the SMSPA. Review of coroner’s reports were authorized by the Department of Forensic Medicine of Porto Alegre.
Results Out of the 335 deaths of children aged 28 to 364 days of life, in Porto Alegre, between 1997 and 1998, there were 57 deaths that occurred at home. Out of these 57 deaths, 21 (36.8%) fulfilled the criteria for diagnosis of SIDS. The prevalence of SIDS was estimated at 6.3% of all deaths of children aged 28 to 365 days of life. The SIDS-specific mortality rate was of 4.5% of deaths for every 10,000 livebirths. Maternal risk factors The risk factor profile for mothers included young mothers; smoking; average third pregnancy; no chronic diseases; routine cases of intercurrence during pregnancy, such as anemia and infection of urinary tract; prenatal exams; and normal delivery (tables 1 and 2).
Sudden infant death syndrome... - Nunes ML et alii
Jornal de Pediatria - Vol. 77, Nº1 , 2001 31
Table 1 -
Maternal risk factors for SIDS - numerical variables
Minimum Maximum Median Average Standard deviation
Number of pregnancies
Number of abortions
* Only 4 mother had previous history of abortion, out of which 1 had 3 abortions and the others had 1 each. † 6 (28%) mothers were primigravidas, 8 (38%) presented interpregnancy interval lower than 15 months, and 7 (33%) presented interval higher than 24 months.
Table 2 -
Maternal risk factors for SIDS - nominal variables
Factors Prenatal exam Intercurrence during pregnancy Chronic disease Medication Smoking Alcohol Illegal drugs Normal delivery
15 (72%) 13 (62%) 0 14 (67%) 15 (72%) 9 (43%) 0 19 (90%)
6 (28%) 8 (38%)* 21 (100%) 7 (33%)† 6 (28%) 12 (57%) 21 (100%) 2 (10%)‡
* Intercurrence reported during pregnancy included 4 cases of anemia, 4 cases of infection of the urinary tract (IUT), 4 cases of anemia + IUT, and one case of congenital lues. † Medication used included 6 cases of use of antibiotic therapy, 4 of ferrous sulfate, 3 of both, and 1 of analgesic. ‡ C-section.
Social and cultural risk factors The social and cultural risk factors included low family income (average of 3 + 1.7 minimum wage income minimum wage ≅ US$ 100/month); less educated parents (incomplete 1st to 8th grade). Only one couple had completed high-school education. Fifty-two percent of mothers had jobs, and all but one, who was a street vendor, were housemaids. There were no information regarding 5 (23%) biological fathers of the infants; the other fathers were reported having different jobs: 4 lower-rank construction workers, 3 drivers, 2 technicians, 2 salesmen, 1 security guard, 1 sanitation worker, 1 gardener, 1 soldier, and 1 unemployed. Houses of people interviewed were either made of woodwork (47%) or of stonework (47%); one house was made of plywood. All houses had electricity, 71% of houses had a ceiling, 90% had floor pavement, and 90% had water and 66% had sewage systems.
Infant risk factors The group of infants selected included 13 boys and 8 girls (1.6:1). As for color of skin, 14 (67%) infants were white and 7 (33%) were black. Age at death varied from 1 to 9 months for an average of 3.2 months (+ 2.4) and a median of 2 months (Figure 1). Most deaths occurred during cooler months (8 in the Winter, 8 in the Fall, and 5 in the Spring). Perinatal history indicated that 18 (86%) infants were born from term pregnancies and all had had Apgar scores > 7 at 5 minutes. Only one infant had presented intercurrence during the delivery, which was described as fetal suffering; as a newborn, however, this infant had had an Apgar score of 10. The average weight at birth was of 2,742 grams (+ 568.8) for a median of 2,800 grams. Four infants had been admitted to the Neonatal ICU, 2 due to premature birth, 1 to jaundice, and 1 to hypoglycemia. Only one baby had been admitted to a hospital after the neonatal period (bronchiolitis). One baby was reported with chronic disease (congenital lues). Three infants were reported with health problems close to the moment of death, out of which there was 1 case of bronchospams the day before death, and 1 of hyperthermia and 1 of vomit on the day of death (Table 3). Table 4 presents the distribution of cases examined according to the basic cause of death indicated on the death certificate prior to investigation at the home of the infant, and to examination of coroner’s reports. Out of the 57 deaths investigated that occurred at home, 21 were listed in this Table for fulfilling the criteria for diagnosis of SIDS. After the review of coroner’s reports, there were no new information that could have definitively determined the cause of death of these infants.
Discussion The new definition for SIDS includes three basic characteristics that determine the diagnosis of this syndrome, which are: sudden death of healthy infant, unsuccessful
Figure 1 - Distribution of age at death
Sudden infant death syndrome... - Nunes ML et alii
32 Jornal de Pediatria - Vol. 77, Nº1, 2001
Table 3 -
Infant risk factors for SIDS
Factors Intercurrence during delivery Admitted to Neonatal ICU Other type of hospital admission Puericulture Chronic disease Acute disease Apneia Sibling of SIDS victim Breastfeeding * † ‡ § ||
1 (5%)* 4 (19%)† 1 (5%)‡ 18 (85%) 1 (5%)§ 3 (15%)|| 2 (10%) 1 (5%) 4 (19%)
20 (95%) 17 (81%) 20 (95%) 3 (15%) 20 (95%) 18(85%) 19 (90%) 20 (%) 17 (81%)
Reported fetal suffering. Two premature babies, one baby with jaundice, and one with hypoglycemia. Bronchiolitis. Congenital lues. At the day before death, one infant had had bronchospams; at the day of death, 1 had presented vomit and 1 hyperthermia.
necropsies for cause of death, and unsuccessful investigation of death scene.1 Information regarding this association of factors is usually not available for coroners at necropsy, especially the result of the investigation of the death scene, which is usually not carried out in Brazil. Consequently, it
Table 4 -
is difficult to establish a final cause of death. In this sense, the literature also describes cases of nonspecific findings in necropsy of cases of SIDS, such as petechyae in the thymus, epicardium, and pleura; signs of inflammation in the upper airways; fibrinous foci and necrosis of larynx; expanded lungs; signs of congestion or pulmonary edema; sanguineous exudate in the cardiac cavity; and prominent lymphoid stroma. These findings can influence the diagnosis of coroners, who are usually not pediatric pathologists experienced with cases of SIDS.16 Moreover, routine necropsy procedures usually do not include examination for alterations of the cerebral trunk, which are considered pathognomonic signs of SIDS.16,17 Since the statistics on causes of death are usually based on information contained in death certificates, there are no officially diagnosed cases of SIDS in Brazil (9th revision of the International Classification of Diseases - ICD 9), despite a percentage of undetermined causes of death. The Program Prá Viver of the City Healthcare Department of Porto Alegre (SMSPA) and its investigation of all deaths of children younger than 5 years of age and questionnaires that allowed for diagnosis of SIDS - in addition to specific review of each postneonatal death at home in the city of Porto Alegre and to analysis of coroner’s reports - allowed for the association of information that led to the diagnosis of SIDS.
Comparison between ICD code recorded in death certificate, coroner’s report, and review of necropsy associated with clinical history
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Cause of death undetermined Cause of death undetermined Cause of death undetermined Cause of death undetermined Cause of death undetermined Bronchopneumonia Cause of death undetermined Acute respiratory failure Acute respiratory failure Cause of death undetermined Cause of death undetermined Cause of death undetermined Acute respiratory failure Cause of death undetermined Pulmonary edema Acute respiratory failure Cause of death undetermined Bronchopneumonia Gastric content aspiration Gastric content aspiration Pulmonary edema
No alterations* No alterations* No alterations* No alterations* No alterations* Lung mucus Tardieu’s spots - epicardial† Tardieu’s spots - pleura† Pulmonary collapse No alterations* No alterations* No alterations* Lung mucus No alterations* Congested lung Tardieu’s spots - pleura† No alterations* No alterations* Lung mucus Congested lung No alterations*
R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95 R.95
* No alterations in organs examined. † Tardieu’s spots were punctiform hemorrhages in the subepicardial and subpleural regions.
Sudden infant death syndrome... - Nunes ML et alii
Figure 2 - Relation between habitual sleeping position and sleeping position at the moment of death of infants (%)
We verified that among other causes of death in children aged 28 to 364 days, SIDS was responsible for 6.3% of deaths. This percentage is higher than that estimated for the city of Pelotas, which is a result of the only study, until this moment, regarding SIDS in our state.15 The prevalence of SIDS in the city of Porto Alegre is, thus, comparable to that of the lowest rates described internationally. 8 This finding is probably related to positive indicators of quality of life in our state. The prevalence of SIDS in other Brazilian states may be different. The risk factor profile for SIDS in the city of Porto Alegre is similar, in some aspects, to that described in the literature,8,12 including prevalence among male infants, age at death from 2 to 3 months, prevalence of deaths during the cooler months (in southern Brazil, the four seasons of the year are well-defined), low breastfeeding rates, absence of perinatal complications or of history of chronic disease, and, occasionally, infectious clinical status on the days prior to sudden death of infant. Prevalence of SIDS among boys has been described in all populations included in the literature and, consequently, some authors have suggested a possible genetic factor related to sex in the etiology of SIDS.19 Reports of history of previous infectious clinical status, despite the following unsuccessful necropsies that do not diagnose the cause of death, have also given basis to the hypothesis of an infectious etiology for SIDS.19 Our population presented a wide variability of weight at birth; however, the average and median weight at birth, respectively 2,742 and 2,800 grams, were higher than those of studies in the United States (< 2,500 grams). This finding may be related to both the low rate of premature births in our study and to the low rate of intrauterine malnutrition in the city of Porto Alegre (about 4.4%).20
Jornal de Pediatria - Vol. 77, Nº1 , 2001 33
Considering established risk factors for SIDS and that all infants in our population were being submitted to puericulture, it is important to note that none of the infants were habitually sleeping supine according to the recommendation of the American Academy of Pediatrics currently being followed in most countries.21,22 We verified that in our population the preferred sleeping position was lateral decubitus. Studies aimed at evaluating sleeping positions as a risk factor for SIDS have demonstrated that sleeping lateral decubitus does not offer the same protective effect of sleeping supine.13,14,21 Maternal risk factors for SIDS found in our study were also very similar to those described in the literature, 8,12 including young mothers, short interpregnancy interval, and multiparas. On an additional note, the high incidence of smoking during pregnancy found in this study is worrisome, especially considering that the general damaging effects of smoking are well known.23 The social and cultural risk factors described in our study overlap with those previously described in the literature.8,12 Since social and cultural risk factors generally affect most of the Brazilian population, it is possible that, in other regions, the prevalence of SIDS is higher and, moreover, not diagnosed. Our results suggest that prematurity is not a risk factor for SIDS in our population. Also, reports of apnea and of family history of siblings of SIDS victims were not significantly high. The absence of previous history of apnea as a marker for SIDS was previously described by others,8,12 and as a result, the term near miss has been banished from the literature and replaced with apparent lifethreatening events (ALTE).24 The finding of only 1 case of family history of SIDS in our population is probably related to the fact that 28% of mothers were primigravidas. For most families, the factor of siblings of SIDS victims is reported in less than 1% of cases; the relative risk calculated for this factor, however, was from 2 to 10.25 The etiology of SIDS is still not well-defined and new hypotheses are being investigated. 26 However, the risk factors for SIDS are well-established and prevention against these factors has been described successful.9 The fact that there are SIDS victims in our population, with a risk factor profile similar to that found in the literature, indicates the need for SIDS awareness of Pediatricians and of the healthcare system in general for future diagnosis of this syndrome.27,28 It is important to underscore the significant effect of educational campaigns for populations at risk for SIDS and who remain exposed to factors that are simple to avoid.
References 1. Willinger M, James LS, Catz C. Defining the sudden infant death syndrome: deliberations of an expert panel convened by the National Institute of Child Health and Human Development. Pediatr Pathol 1991; 11:677.
34 Jornal de Pediatria - Vol. 77, Nº1, 2001 2. Brady JP, Ariagno RL, Watts JL, Goldman SL, Dumpit FM. Apnea, hypoxemia, and aborted sudden infant death syndrome. Pediatrics 1978; 62:686-91. 3. Hunt CE. Abnormal hypercarbic and hypoxic sleep arousal responses in near-miss SIDS. J Appl Physiol 1981; 50:1313-7. 4. Steinschnnneider A, Weinstein SL, Diamond E. The sudden infant death syndrome and apnea/obstruction during neonatal sleep and feeding. Pediatrics 1982; 60:858-63. 5. Glotzbach SF, Ariagno RL, Harper RM. Sleep and the sudden infant death syndrome. In: Ferber R, Kryger M, eds. Principles and Practice of Sleep Medicine in the Child. Philadelphia: WB Saunders; 1995.p.231-44. 6. Kahn A, Franco P, Scaillet S, Groswasser J, Dan B. Development of cardiopulmonary integration and the role of arousability from sleep. Current Opinion in Pulmonary Medicine 1997; 3:440-4. 7. Harper RM, Bandler R. Finding the failure mechanism in suden infant death syndrome. Nature Medicine 1998; 4:157-8. 8. Hoffman HJ, Hillman LS. Epidemiology of the sudden infant death syndrome: maternal, neonatal and postneonatal risk factors. Clinics in Perinatology 1992; 19:717-37. 9. Dwyer T, Ponsoby AL. The decline of SIDS: a success story for epidemiology. Epidemiology 1996; 7:323-5. 10. Guidelines for death scene investigation of sudden, unexplained infant deaths: recommendations of the interagency panel on SIDS. Morbidity and Mortality Weekly Report, Center for Disease Control and Prevention. U.S. Department of Health and Human Services, 1996; 45:1-21. 11. Krous HF. Instruction and reference manual for the international standardized autopsy protocol for sudden unexpected infant death. Journal of Sudden Infant death and Infant Mortality 1996; 1:203-46. 12. Hoffman HJ, Damus K, Hillman L, Kongrad E. Risk factors for SIDS. Results of the National Institute of Child Health and Human Development SIDS cooperative epidemiological study. Ann NY Acad Sci 1988; 533:13. 13. Thach BT. How does prone sleeping increase prevalence of sudden infant death syndrome. Pediatric Pulmonology 1997;16: 115-6. 14. Mitchell EA. The changing epidemiology of SIDS following the national risk reduction campaigns. Pediatric Pulmonology 1997; 16:117-9. 15. Barros FC, Victora CG, Vaughan JP, Teixeira AMB, Ashworth A. Infant mortality in southern Brazil: a population based study of causes of death. Arch Dis Child 1987; 62:487-90. 16. Valdes-Dapena M. The sudden infant death syndrome: pathologic findings. Clinics in Perinatology 1992; 19:701-16.
Sudden infant death syndrome... - Nunes ML et alii
17. Kinney HC, Filiano JJ, Sleeper LA, Mandell F, Valdes-Dapena M, White WF. Decreased muscarine receptor binding in arcuated nucleus in sudden infant death syndrome. Science 1995; 269:1446-59. 18. Mage DT, Donner M. A genetic basis for the sudden infant death syndrome sex ratio. Medical Hypothesis 1997; 48:137-42. 19. Lindsay JA, Blackwell C. Infectious agents and SIDS: an update. Molecular Medicine Today 1996;94-5. 20. Aerts D, Flores R. Porto Alegre em números: dados populacionais, nascidos vivos, mortalidade. Secretaria Municipal de Saúde de Porto Alegre, 1998; 47-48. 21. Willinger M, Hoffman HJ, Hartford RB. Infant sleep position and risk for sudden infant death syndrome: report of meeting held january 13 and 14, 1994, National Institutes of Health, Bethesda, MD. Pediatrics 1994; 93:814-9. 22. American Academy of Pediatrics (Task Force on Infant Positioning and SIDS), Does bed sharing affect the risk of SIDS? Pediatrics 1997;100:272. 23. Anderson HR, Derek GC. Passive smoking and sudden infant death syndrome: review of the epidemiological evidence. Thorax 1997;52:1003-9. 24. Brooks JG. Apparent life-threatening events and apnea of infancy. Clinics in Perinatology 1992; 19:809-38. 25. Beal SM. Siblings of sudden infant death syndrome victims. Clinics in Perinatology 1992; 19:839-48. 26. Mc Namara F, Sullivan CE. Evolution of sleep-disordered breathing and sleep. Journal of Paediatrics & Child Health 1998; 34: 37-43. 27. Nunes ML. Síndrome da morte súbita da infância: este é o momento de iniciarmos um registro nacional. Informativo SBP 1996 (outubro/novembro); 3. 28. Nunes ML. Annual Report from Brazil. SIDS International News, 1999; 3(winter):6.
Correspondence: Dr. Magda Lahorgue Nunes Serviço de Neurologia HSL- PUCRS Av. Ipiranga, 6690 CEP 90610-000 – Porto Alegre, RS - Brazil Phone/fax: +55 51 3339.4936 E-mail: [email protected]