Children with Down syndrome -: an epidemiological study with special focus on congenital heart defects

Full text

(1)

(2)

(3)

(4)

(5)

(6)

(7) .

(8) !

(9)

(10)

(11)

(12) "#. $%&'&() %&. )') (&*+%&')'& ,)-&+(& ,,)-) ...

(13) Dissertation for the Degree of Doctor of Philosophy, (Faculty of Medicine) in Pediatrics presented at Uppsala University in 2003 ABSTRACT Frid, C. 2003. Children with Down Syndrome - an epidemiological study with special focus on congenital heart defects. Acta Universitatis Upsaliensis. Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1218. 62 pp. ISBN 91-554-5496-8 To assess the impact of congenital malformations in Down syndrome (DS) on morbidity, mortality and outcome at birth, information on all children with DS born in the northern part of Sweden in 197380 (n=211) and 1995-98 (n=88) was collected. Most common were congenital heart defects (CHD), dominated by atrioventricular septal defects (AVSD). Up to age 10 years, morbidity and mortality were more than 10 times higher in DS children with CHD than in healthy DS children. The DS children seemed more vulnerable at birth than Swedish children in general: they had increased frequencies of Cesarean sections, premature birth, asphyxia, and low birthweight, and higher proportions of children small for gestational age, regardless of the presence of CHD. Infant mortality decreased from 14.2% to 2.3% between the two periods. All children with AVSD with and without DS born in Sweden 1973-1997 (n=801) were followed up retrospectively to 2001. Children with isolated AVSD without complex additional CHDs were studied more closely (n=502). A reduction in age at operation and postoperative mortality (from 28 to 1%) was observed. No significant difference in 5-year postoperative mortality between genders or between DS and non-DS children was found. The 5-year postoperative mortality in DS decreased from 35% in 1973-77 to about 10% in 1993-97. CHD had a major influence on morbidity, infectionrate and mortality in DS, but not on DS birth variables. The formerly high mortality in CHD is now reduced. In isolated AVSD measures seem equally successful in DS and non-DS children. Mortality is still 3 times higher in DS children with isolated AVSD than in healthy DS children. Keywords: Down syndrome, congenital malformations, mortality, morbidity, neonatal data, congenital heart defect, atrioventricular septal defect, operation, gender Christina Frid, Department of Women’s and Children’s Health, Section for Pediatrics, Uppsala University Childrens Hospital, Uppsala University, SE- 751 85 Uppsala ©Christina Frid 2002 ISSN 0282-7476 ISBN 91-554-5496-8 Printed in Sweden by Eklundshofs Grafiska, Uppsala 2002.

(14) Fyrhändigt pianostycke. Jag drömmer om mycket mer: Om en musik för tusen händer, Om en harmoni Där hela världen är med. Ur ”Tusen skäl att leva” av Dom Helder Camara. To the men in my life: My husband Svante My sons Pontus, Jonas, Axel och Gustav And my father Viking Olov.

(15) List of publications. This thesis is based on the following papers, which are referred to in the text by their Roman numerals: I. Frid Christina, Drott Peder, Lundell Bo, Rasmussen Finn, Annerén Göran. (1999). Mortality in Down's syndrome in relation to congenital malformations.. J Intellect Disabil Res 43(( Pt 3):): 234-41 II. Frid Christina, Annerén Göran, Rasmussen Finn, Sundelin Claes, Drott Peder. (2002). Utilization of medical care among children with Down's syndrome.. J Intellect Disabil Res 46((Pt 4):): 310-7 III. Frid Christina, Drott Peder, Otterblad Olausson Petra, Sundelin Claes, Annerén Göran. Children with Down syndrome.. Maternal and neonatal factors and mortality 1973-80 and 1995-98.. (Submitted). IV. Frid Christina, Björkhem Gudrun , Jonzon Anders, Sunnegårdh Jan, Annerén Göran, Lundell Bo. Long-term Survival in Children with Complete Atrioventricular Septal Defect in Sweden. Complete isolated AVSD in Down Syndrome and chromosomally normal children. (Submitted) Reprints were made with the kind permission from the publishers..

(16) Contents. Introduction.....................................................................................................9 General introduction .................................................................................9 History of Down syndrome ......................................................................9 Epidemiology..........................................................................................10 Etiology ..................................................................................................11 Prenatal diagnosis ...................................................................................11 Clinical aspects .......................................................................................12 Congenital heart defects in DS ...............................................................14 The embryology of the heart...................................................................15 Epidemiology and etiology of atrioventricular septal defect..................15 Morphology in AVSD ............................................................................16 Clinical findings in AVSD......................................................................16 The history of thoracic surgery...............................................................17 Aims of the studies .......................................................................................19 National Health Registers .............................................................................20 Material.........................................................................................................21 Population and study design ...................................................................21 Papers I, II and III.............................................................................21 Paper IV..................................................................................................22 Methods ........................................................................................................25 Papers I, II and III...................................................................................25 Paper IV..................................................................................................26 Statistical analysis...................................................................................26 Results...........................................................................................................28 Paper I...............................................................................................28 Paper II .............................................................................................29 Paper III............................................................................................31 Paper IV............................................................................................33 Isolated AVSD..................................................................................33 Isolated AVSD and GIT malformations...........................................37.

(17) General Discussion .......................................................................................38 Methodological considerations...............................................................38 At birth (Papers II and III) ......................................................................39 Congenital malformations (Papers I and IV)..........................................40 Morbidity (Paper II)................................................................................41 Mortality (Papers I, III and IV)...............................................................42 Reflections and future directions ............................................................44 Conclusions...................................................................................................46 Paper I.....................................................................................................46 Paper II ...................................................................................................46 Paper III ..................................................................................................46 Paper IV..................................................................................................47 Swedish summary-Svensk sammanfattning .................................................48 Acknowledgements.......................................................................................50 References.....................................................................................................52.

(18) Abbreviations. ASDI ASD II AVSD CHD CHD-DS CI CS DS GA GIT H-DS IDM IQ OR PDA SD SGA VSD. Atrial septal defect of the primum type Atrial septal defect of the secundum type Atrioventricular septal defect Congenital heart defect Children with DS and a CHD, but no other malformation Confidence interval Cesarean section Down syndrome Gestational age in completed weeks Gastrointestinal tract Healthy children with DS without malformation Mean number of inpatient days per month and child Intelligence quotient Odds ratio Patent ductus arteriosus Standard deviation Small for gestational age Ventricular septal defect.

(19)

(20) Introduction. General introduction Down syndrome (DS) is the most common condition associated with severe mental retardation in the general population.1,2 About 50% of children with DS suffer from congenital heart defects (CHD),3,4 a majority of them from atrioventricular septal defect (AVSD),5 which in former days led to death in early childhood. In the last decades great changes have taken place in the management of children with DS.6 In addition to the advances in neonatal care, new surgical techniques and the development of new medical drugs, a change in attitude toward these children has taken place, and in Sweden they are no longer excluded from medical treatment for curable disorders. The result is an increase in their life expectancy, from 9 years in 19297 to 57 years in 1989,8 today probably even longer.9 Bearing these changes in mind, we considered it important to investigate the situation of children with DS today. The present studies address the distribution of congenital malformations in children with DS and the consequent morbidity and mortality today compared to earlier days, as research points to CHD as the main cause of mortality and morbidity.10-12. History of Down syndrome Down syndrome (trisomy 21) was given its name after John Langdon Down, who gave a description of many of its features in ”Observations on ethnic classification of idiots” in 1866. However, a person with Down syndrome was described for the first time in 1838 by Esquirol. The association of the syndrome with congenital heart defects was first pointed out by Garrod in 1894,13 and the specific association with atrioventricular septal defects was elucidated by Abbot in 1931.14 The development of the science of genetics, and the observation by Tijo och Levan in 195615 that the normal number of chromosomes in humans is 46, constituted the necessary preceding landmarks for the discovery of the genetic background of DS. In 1959. 9.

(21) Jerome Lejeune16 found an extra G chromosome in DS patients, demonstrating that they have 47 instead of the normal 46. The finding by Caspersson, Zech and Johansson17 in 1970 of the specific banding patterns of the chromosomes meant that identification of chromosomes and detection of structural abnormalities were simplified. The entire long arm of human chromosome 21 was sequenced in 200018 and an important research goal now is to determine the DS-specific genes in order to make possible therapeutic interventions.. Epidemiology The incidence of DS varies between 1:500 and 1:2,000 in different countries, depending on the maternal age and the use of prenatal diagnostics.19 The association of DS with increased maternal age was first observed by Penrose20 in 1933 and later confirmed by others.21 In Sweden the prevalence among live births at the present time is 1:700-800, and 100130 children are born each year with this syndrome.. Number per 10 000 liveborn. 60 50. DS , age of mother >35. 20. DS, total % of mothers >35 years. 15. 40. % 30. 10. 20 5 10 0. 0 1976. 1980. 1984. 1988. 1992. 1996. 2000. Year of birth. Figure 1. Newborn children with Down syndrome (DS) reported to the Swedish Registry of Congenital Malformations and/or to the Swedish Cytogenetic Registry in 1978-2001. The percentages of mothers of DS children ≥ 35 years of age in Sweden are also presented in the figure.22. With the prenatal diagnostic tools available today, and offered to women > 35 years of age in Sweden, a decrease in incidence due to legal abortions. 10.

(22) could be expected. However, in Sweden no decrease in incidence has been observed so far (Fig. 1), as it seems to be balanced by the increased maternal age.22. Etiology In DS there is an extra copy of chromosome 21. In 90-95% of the cases this is due to nondisjunction, i.e., a failure of the two chromosomes 21 to separate from each other at one of the two meiotic divisions in gametogenesis. In 4%, DS is the result of a translocation of all or part of chromosome 21, and in 1% of the cases there is mosaicism where only a fraction of the cells contain 47 (trisomy 21) chromosomes. The origin of the extra chromosome is maternal in 90% of the cases.23,24 There is a correlation between increased maternal age and nondisjunction. Before the age of 35 years the risk of having a child with DS is 0.05%, and after that age this risk increases exponentially with increasing age. No such correlation has been found with paternal age. In a study by Lindsten and co-workers in 1981,25 the incidence among mothers aged 40 years was 1%, and this increased to 5% in mothers aged 45. Many reasons for the increased incidence of nondisjunction with advanced maternal age have been discussed, but the underlying mechanism is still unknown.26 The distal part of the long arm of chromosome 21 (21q22.2) is called the Down syndrome specific segment, where many of the genes responsible for the features in DS are situated. In this segment most of the genes of chromosome 21 are located.18. Prenatal diagnosis DS is now diagnosed prenatally with increasing frequency. The screening methods in Sweden have so far been applied to mothers > 35 years of age and in those at increased risk. In Sweden in 1999-2000 about 45% of the pregnant women who were offered prenatal diagnosis because of age (> 35 years) accepted the offer. Several screening methods for DS will become available in the future, apart from the traditional ones used in the last decades, with maternal serum biochemistry screening27-29 at 11-14 weeks of gestation, and invasive procedures to obtain the fetal karyotype - chorionic villus sampling (in the first trimester) and amniocentesis (in the second trimester). New methods are being evaluated for their effectiveness, such as second-trimester triple testing (maternal serum levels of alpha-fetoprotein and unconjugated estriol are. 11.

(23) generally reduced and the level of human chorionic gonadotropin is increased in fetal DS pregnancies), quadruple testing (including InhibinA,30,31 which is increased in fetal DS), non-invasive methods such as evaluation of fetal biometric data; measurement of fetal nuchal translucency thickness (increased in fetal DS and other chromosomal defects),32,33 and ultrasound screening of the fetal nasal bone at 11-14 weeks of gestation (absence of nasal bone in fetal DS at this time).34,35 Analysis of fetal cells in maternal urine or blood is also being investigated as a potential diagnostic method.36. Clinical aspects DS is the most frequent occurring human autosomal chromosomal syndrome compatible with postnatal survival. There are many important clinical features in DS, but none of them alone are diagnostic for DS and all can also be found in the general population. The features in DS have been described extensively.37-39 The presence of 12 or more of the 20 diagnostic criteria at birth set by Hall39 makes the diagnosis likely, but chromosomal analysis will give the definite diagnosis. Some of the features are listed below: • Neurological disturbances In DS mental retardation is a dominating feature. At birth the brain is near normal in size, but the number of brain cells is reduced by 10-50% compared to that in newborn children with normal chromosomes.40 Retardation of brain growth is observed after 3-6 months of age and the relative size drops below normal. The ratio IQ declines with increasing age, from 80 at six months of age to 45 at four years.41 Infantile autism is found in 1%42 to 11.4% 43 of children with DS, and is increased compared to that in the general population. Epilepsy has an age-specific component in DS, and the risk of epilepsy increases with age. In infancy, the prevalence of infantile spasms is 4%. Early dementia is common in DS. An Alzheimer-like condition develops at an early age,44,45 with a reduced number of neuronal cells, and with neurofibrillary tangles and neuritic plaques, consisting of beta-amyloid protein, all of which can be observed in the brain. After the age of 35 years these changes can be found in the central nervous system of all adults with DS.46,47 The amyloid protein precursor gene has been traced to the long arm of chromosome 21,48 and an increase in serum amyloid precursor protein has been found in DS.. 12.

(24) • Congenital malformations of the heart Congenital heart defects are present in more than 50% of children with DS.49 The most common lesion is a severe defect that will need surgical treatment early in life. See separate passage below. • Gastrointestinal (GIT) malformations GIT malformations are present in 10 % of children with DS,49-51 and many of them will require immediate surgery after birth. Duodenal atresia (2%), Hirschsprung’s disease (4%) and annular pancreas are the most common defects, but esophageal atresia (1%) and anal atresia (1%) are also present in increased frequencies. • Immunological problems Autoimmune diseases are increased in DS. Hypothyroidism is present in 2835%,52,53 and 2-3% develop hyperthyroidism. Karlsson et al. observed that 15% acquired the hypothyroidism before the age of 8 years, with no thyroid autoantibodies present, while those that developed hypothyroidism after 8 years of age (15%) had autoantibodies. Celiac disease is also increased in DS, in a frequency varying between 3 and 17 % in different studies,54,55 and juvenile diabetes is present in 2%.56-58 • Oncology Children with DS are known to have a high incidence of leukemia, 20 times higher than in normal children.59,60 However, they constitute a special subgroup of the leukemia population, with a better outcome in cases of acute myeloblastic leukemia (AML), but a tendency to a worse outcome in those with acute lymphatic leukemia (ALL), compared to corresponding children without DS.59,61,62 In contrast, there is a decreased risk of solid tumors.63 • Increased susceptibility to infections Children with DS have an increased susceptibility to infections.64 Many factors contribute to this increase; not only is the thymus reduced in size and effectiveness,65 but other functions in the protection against infections are affected. These include a neutrophil function defect,66 defective chemotaxis,67 shortened granulocyte survival,66,68 reduced T-cell counts, and defective T-cell activation.69 • Growth disturbances Children with DS have a short stature, as their growth is retarded.70 At birth their length is within the normal range, but in the first years of life growth retardation becomes evident, and the growth deficit will come close to -2SD and increase to -4SD in some individuals. Growth hormone treatment in DS. 13.

(25) is only recommended when a deficiency is observed.71 New growth charts for children with DS have recently been published.72 The prognostic final height is 17-20 cm shorter than in the general population. Children with DS are known to have a slightly decreased birthweight in comparison with the general population, even after adjustment for gestational age.51,73,74 In the general population, increased maternal age is associated with risks of having very and moderately preterm small for gestational age (SGA) infants,75 and this might also be true of the DS population, but is not well elucidated. The Apgar score is a standardized system for assessing the status of a newborn, devised by Virginia Apgar in 1953.76 Five different clinical features are evaluated and given a score at 1, 5, and 10 minutes after birth; heart rate, respiration, muscle tone, response to pharyngeal catheter, and color of the skin. As a marker for asphyxia, the Apgar score has been criticized,77-79 but as an indicator of vitality directly after birth it still has its place. In the official statistics this score is available for all children born in Sweden, and it is therefore an important factor for use in comparisons. For further prognosis of the neurological outcome in newborns, complementary information on the Apgar score at 15 and 20 minutes, on neurological symptoms and on blood gas data for detection of acidemia are needed. In DS the Apgar score will be slightly reduced because of hypotonia, but the Apgar score in general in DS is not known. Nor is the impact of a CHD on some neonatal factors such as the birthweight, Apgar score, gestational age (GA) at birth and the frequency of SGA known. The variation in the frequency and severity of the above-described features of newborn infants with DS, and in the nature of any malformation present, will mean that many different health providers will be involved in their management. Recent research dealing with health aspects of the parents has revealed the importance of the first information provided to the family with a newborn child with DS,80 and of the care of the family.. Congenital heart defects in DS The prevalence of CHD in newborn infants in the population at large is 0.81%, while in neonates with DS there is a 50% risk of CHD.3,81 Children with Down syndrome constitute 8% of all children with a CHD. Several studies have addressed the prevalence of different congenital heart defects in DS.4,5,82-89 A predominance of males among children in general with CHDs has been observed previously,90 but among children with DS and CHDs there is a higher proportion of females.91,92. 14.

(26) The most common CHDs in Down syndrome are AVSD (45%) and ventricular septal defects (VSDs 35%), and less frequent are atrial septal defects of the secundum type (ASD II 8%), tetralogy of Fallot (4%) and patent ductus arteriosus (PDA 7%).93 Approximately 30% of the children with DS and CHD have multiple cardiac defects, the most common of which is PDA.4,86,94 Some CHDs are absent or very rare in DS, such as transposition of the great arteries, coarctation of the aorta, and situs inversus, but the reason for this is not known.82 It has been speculated that the presence of three chromosomes 21 somehow interferes with the early formation of the endocardial cushions and also may protect against other lesions of the heart.. The embryology of the heart The development of the heart is a complex process including events on the molecular and cellular levels, influenced by the environment and chance. It involves the transformation and migration of cells at the right time, initiated by special genes and their signal substances. New insights into the understanding of the cardiac development are continuously being achieved.95-97 The cardiovascular system develops from the mesoderm, and is transformed from a muscle-wrapped tube to the mature heart. The primitive heart begins to beat on the 21st day. In the fourth week it undergoes a complicated looping and subsequently the chambers begin to take form in the 5th week by the appearance of septa forming the two atria and ventricles. At the same time the subendocardial cushions grow and fuse to form the final anatomical result, the 4-chamber heart with two separate orifices separating the atria from the ventricles.98 When the development of the septa and the fusion of the endocardial cushions fail, an AVSD will be the result.. Epidemiology and etiology of atrioventricular septal defect In three fourths of the children with AVSD a syndrome is present, and a majority of them have DS.99 In the Baltimore-Washington Infant Study,94 it was found that among all CHDs presenting under the age of 1 year, AVSD constituted 7.7%. The rate of AVSD was 3.5 per 10,000 live births.92 The genetic origin of AVSD is multifactorial, and essentially four subgroups are discussed:. 15.

(27) 1) Chromosome abnormalities, including the most common trisomy 2194 and others, such as del 3p25 and del 8p23; 2) Syndromes, for example Holt Oram (12q2), Noonan (12q22) and Turner; 3) Heterotaxy (isomerisms); 4) AVSD as the only malformation, following a mendelian autosomal dominant inheritance.96 Theories on why the endocardial cushions in DS fail to fuse include increased adhesiveness of the cells, thereby delaying or preventing fusion.100. Morphology in AVSD The atrioventricular defects consist of a primum atrial septal defect (ASDI) with or without a ventricular septal defect (VSD). The defect in complete AVSD has three different components: 1) A VSD , 2) an ASDI, and 3) a common valve orifice instead of the two separate mitral and tricuspid valves, as the fusion of the endocardial cushions has failed.101 The defects at the atrial and ventricular levels can vary in size (Fig. 2).. Figure 2. Normal heart (left) and complete AVSD (right). RA=right atrium, LA= left atrium, RV= right ventricle, LV= left ventricle. Clinical findings in AVSD The neonate with an AVSD may present with symptoms, but more often does not have any symptoms at all. The severity of symptoms depends on. 16.

(28) the presence of associated heart lesions, such as ventricular hypoplasia, outflow tract obstruction and other congenital cardiac malformations, and on whether there is shunting through the septal defects, indicating pulmonary hypertension. A systolic or pansystolic murmur may be heard, and the second heart sound may have an accentuated pulmonary closure sound. The presence of a left axis deviation on the electrocardiogram is the general rule, but in 5% there is a right axis deviation. A chest radiography may show an enlarged heart. Left ventricular angiography will reveal an inlet-outlet disproportion of the left ventricle, called “goose neck” deformity.102 Children with AVSD will later develop irreversible pulmonary hypertension, and subsequently will display the features of Eisenmenger´s syndrome, unless anatomical correction is performed early in infancy.88 Studies indicate that corrective surgery should be undertaken before the age of 6 months to prevent the development of irreversible pulmonary vascular disease.103-105 When pulmonary hypertension develops, there is increased muscularity of the small pulmonary arteries, intimal hyperplasia and thrombosis, with a subsequent increase in pulmonary vascular resistance. Studies indicate that children with DS are at higher risk than other children of developing pulmonary hypertension.106,107 However, Yamaki et al. found that in DS the media of the muscular arteries became thinner as the intimal obstruction progressed.108 Even in the absence of a CHD in children with DS,109 pulmonary hypertension may develop, with congenital alveolar hypoplasia, chronic upper airway obstruction and frequent infections of the respiratory tract as possible contributory factors. Irreversible pulmonary hypertension with pulmonary obstructive disease has been found as early as at 4-6 months of age in DS, but rarely in children with normal chromosomes.. The history of thoracic surgery Since 1950 there have been great advances in pediatric cardiology and thoracic and cardiovascular surgery.110 Heart cathetrization was developed in the 1940s, facilitating the diagnostic procedure, but in many cases today this has been replaced by echocardiography. Improvements in the assessment of pulmonary vascular resistance and better understanding of its development allowed better choice of timing of surgery. Open-heart surgery was made possible by the introduction of the heart-lung oxygenator in the 1950s, and further progress was made when the use of deep hypothermia during intracardiac repair was introduced in the early 1970s. In 1955 the first repair of an atrioventricular defect was reported by Lillehei and co-workers.111 Subsequently the techniques improved, the one- and the two-patch. 17.

(29) techniques were introduced112,113 and the age and body weight at which correction was possible became greatly reduced. The treatment practice for CHD in children with DS has varied over time with advances in medical technology, and with changing attitudes in the society.114 In Sweden the policy of offering surgical treatment to children with DS and AVSD was adopted fairly early, compared to most other countries, and consequently we have a large, unique population of subjects with anatomically corrected AVSD to observe in follow-up studies.6,115. 18.

(30) Aims of the studies. The studies on which this thesis is based had the following aims: •. To investigate different aspects of DS in Sweden, including the prevalence in the population, the spectrum of congenital malformations, especially CHDs, and the mortality rates in relation to these malformations (Paper I). •. To assess the magnitude and examine the reasons for utilization of medical care by DS children during different age-periods in relation to congenital malformations, especially CHDs (Paper II). •. To investigate some perinatal factors in children with DS, analyze the impact of CHDs on these factors, and compare the findings with those in the general population (Paper III). •. To describe the prevalence of AVSD in the Swedish population, and the mortality rate and survival among DS and non-DS children with AVSD, with and without operative treatment, over the last 30 years (Paper IV). 19.

(31) National Health Registers. The National Board of Health and Welfare in Sweden keeps several nationwide registers, all of them relying on the identification system used in Sweden, every Swedish resident being allocated a 10-digit national registration number, unique for the person in question. In addition, different units of the health care system, such as antenatal, obstetric, pediatric, plastic surgery, pediatric cardiology clinics, and genetic laboratories are under obligation to send in standardized reports to the relevant registries. For these purposes a Swedish version of the International Classification of Diseases (ICD) with subsequent revisions is used, currently the tenth revision of 1997 (ICD-10). The Register of Congenital Malformations, established in 1964, contains data on children identified before 6 months of age as having major congenital malformations. There is some underreporting of some diagnoses, especially DS before 1982. The Swedish Medical Birth Register was established in 1973 and contains data on 99% of all births.116 The register stores information on the mother, the delivery, and the child, including diagnoses of the child up to 1 month of age. The diagnoses in children are not always correctly registered, and there is some underreporting. The Cytogenetic Register was established in 1978 and contains information on chromosomal aberrations detected before 1 year of age. The Hospital Discharge Register was initiated in 1964 and includes information on more than 99% of all inpatient care at state- or communityowned hospitals in Sweden from 1987. There are no private pediatric hospitals in Sweden. The information includes up to six diagnoses, any operations, number of inpatient days, and the date of discharge. The Register of Congenital Heart Malformations, which was kept between 1980 and 1996, contains information from the four pediatric cardiology centers in Sweden, including diagnoses of heart disorders up to 1 year of age, together with chromosomal aberrations and other diseases. The Population Database includes all Swedish residents alive, and is frequently updated.. 20.

(32) Material. Population and study design The populations investigated in the different studies and the sources of information are summarized in Table I.. Papers I, II and III Study I (Paper I) comprised all liveborn children with DS born in northern Sweden (Fig. 3) from 1 January 1973 to 31 December 1980. This was made possible by means of the Swedish Register of Congenital Malformations, the Cytogenetic Register, and/or the Medical Birth Register. Children moving into the area were excluded. In a total of 219 children the presence of congenital malformations was studied and in 213 children followed up to 30 June, 1997 the mortality was investigated. The Population Register was checked to verify that the child was still alive. Information on the total number of liveborn children in the area was obtained from the official statistics. Medical records from hospital departments were available for the 219 children and were scrutinized. Figure 3. Map of Sweden with the studied area indicated... In study II (paper II) the same population as in study I was followed up to 31 December 1997. From the Medical Birth Register a second population consisting of 90 children with DS born in the same geographical area from 1 January 1995 to 31 December 1998 was retrieved, and information on the duration of neonatal/post-neonatal care was obtained for both populations. 21.

(33) from this register. By manually scrutinizing medical records in December 1992, information on the number of days of and reason for inpatient care was obtained for the period 1973 to 1992 for 211 children born 1973–80, and by means of the Hospital Discharge Register, this information was updated for the period 1992 to 1997. In study III (paper III) the same populations as in study II born in 1973– 80 and 1995–98 were studied (however, 2 children were excluded from the second population because they proved not to belong to the same geographical area, leaving a total of 88 children in that population). One child born in 1977 moved from Sweden before the age of 1 year and was therefore not included in the calculation of infant mortality. Information on perinatal data in the two studied populations was retrieved from the Medical Birth Register together with summarized perinatal data on children in the general population born in the same region and periods.. Paper IV In study IV (paper IV), information on all children in Sweden born alive between 1 January 1973 and 31 December 1997 with a diagnosis of AVSD (1,150), the total number of live births (2,599,624) and the number of children born with DS (3,355) was retrieved from the Register of Congenital Malformations, the Register of Congenital Heart Malformations, the Medical Birth Register and the local registries at the four pediatric cardiology centers in Sweden. Included in the study were 801 children with complete AVSD (Fig. 4), diagnosed according to the international coding systems (ISC 200– 208, ICD 9; 745G; ICD 10; Q21.2) and complementary information. Children with trisomies other than Down syndrome were excluded, as also were those with isolated ostium primum atrial septal defect. Medical records from hospital departments were available for the 801 children and were scrutinized.. 22.

(34) Number of children born with AVSD. 60. AVSD, DS. 50. AVSD, non-DS 40. 30. c. 20. 10. 0 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. Year Fig. 4. Numbers of children with complete and intermediate atrioventricular septal defect (AVSD) born per year in 1973-97 in Sweden.. 23.

(35) Table 1. Study populations and sources of information. Study populations Children with Paper I Study of congenital DS born 197380 in northern malformations Sweden. Registries used MBR RCM CR. Exclusions. Final study population. 3 not DS (1.3%) 1 stillbirth 5 missed to follow-up (2.2%) 219. Mortality study: Follow-up to June 1997 Paper II Study on morbidity Follow-up to December 1997. MBR RCM CR HDR. Children with DS born 197380 in northern Sweden. Children with DS born 199598 in northern Sweden Paper III Children with Study on maternal DS born 1973and neonatal factors 80 in northern Sweden Study on neonatal/postneonatal period. Paper IV Study of all AVSD Follow-up to December 2001. 213 211. MBR. 90. MBR RCM CR HDR. 219. MBR RCM CR HDR MBR RCM RCHM. Children with DS born 199598 in northern Sweden Children with AVSD born 1973-97 in Sweden. 5 moved abroad 1 lost to follow-up 5 moved abroad 2 incomplete neonatal data. 2 children born outside defined geographical area. 88. 317 not compatible with inclusion criteria (27.6%) 32 insufficient information (2.8%). 801. 47 intermediate AVSD Study of isolated 247 complex AVSD 502 AVSD 5 lost to follow-up Follow-up to (0.7%) December 2001 MBR, Medical Birth Register; RCM, Register of Congenital Malformations; CR, Cytogenetic Register; HDR, Hospital Discharge Register; RCHM, Register of Congenital Heart Malformations. 24.

(36) Methods. Papers I, II and III In study I medical records of the included 219 children were scrutinized and data were collected manually. Data were obtained for all 219 children for the neonatal period and 213 children were followed up to the date of death or to the age of at least 14.5 years. The following data were collected: parental ages at birth, sex, date of birth, karyotype, any congenital malformation and, in deceased cases, age at and cause of death. In study II all medical records of the 211 included children with DS born 1973–80, including records from referral clinics in and outside the region, were scrutinized manually. The following data were collected: sex, gestational age and weight at birth, presence of CHD, GIT or other malformations, number of days in hospital as inpatient and number of admissions from birth to December 1992, the reasons for this medical care, and the ages at which it was provided. Information on dates of hospitalization as inpatient, number of days in hospital as inpatient, and the reasons for this medical care, from 1992 to December 1997, were obtained from the Hospital Discharge Register. Congenital malformations were classified into CHD, GIT, CHD+GIT, and other malformations. Children with DS without any malformations, so-called healthy DS children (H-DS, n=95) were compared with children with DS and CHD but no GIT malformation (CHD-DS, n=91). The reasons for medical care were divided into three categories: infections, cardiopulmonary disorders, and other conditions. The duration of neonatal/post-neonatal care in the 211 DS children born in 1973-1980 was compared with that in 90 children with DS born in 1995-98 in the same region. The neonatal/post-neonatal period was defined as all days from birth in the delivery/neonatal ward to discharge to home/foster home/institution or to another hospital, and the number of days could therefore exceed 30 days. The mean number of inpatient days per month and child (IDM) was calculated for each age-specific group. In study III data were collected in the year 2001 from the Medical Birth Register. The following data were collected from this register: sex, mother’s. 25.

(37) age at the birth of the child, date of birth and death (if relevant) of the child, gestational age based on the last day of the menstrual period (1973-80 and 1995-98) and based on ultrasound measurements (1995-98), Apgar score at 5 minutes, Cesarean section, birthweight, SGA, and presence of CHD. The CHD information was supplemented with data from the Swedish Register of Congenital Malformations, the Swedish Cytogenetic Register, and the Hospital Discharge Register. To adjust for methodological differences in measuring length of gestation between 1973-80 and 1995-98, data based on the last menstrual period were used for comparison. Neonatal mortality (death within 28 days after birth) and infant mortality (less than 1 year after birth) were calculated.. Paper IV In study IV the medical records of the included 801 children with AVSD were scrutinized in the year 2002 to confirm the diagnosis of complete AVSD. This diagnosis included the components ventricular septal defect, a common A-V valve and an atrial septal defect of the primum type. The following data were collected: sex, karyotype, associated CHDs, date of birth, operation (only anatomically corrective surgery, not palliation) and death. Follow-up was pursued until death or the endpoint of the study or the date of the last medical record. All individuals recorded as being alive on December 31, 2001 were confirmed as being so in the National Population Database. Data for a subgroup of 502 children with complete AVSD without significant associated heart malformations, called isolated AVSD, were more thoroughly scrutinized, with comparisons between children with and without DS.. Statistical analysis For the statistical analysis a Stat View program package Ver 5 ( Abacus Concepts, Inc., Berkely, CA, USA) was used. Students´s t-test was employed for comparisons of groups with variables with a normal distribution. Results are presented as means, with the range, median value and 10th and 90th percentiles or 95% confidence intervals (CI) in parentheses. When appropriate, differences between groups were estimated by the Mantel Cox or Mann-Whitney method. The Chi square test was used for comparisons of proportions between groups, and 95% confidence intervals (CI) for the differences are given in parentheses. Survival analyses were performed by the Kaplan-Meier method.117 In study III, a multiple. 26.

(38) logistic regression analysis was performed to detect any relationships between the variables studied (dependent variable: birth period 1973-80 or 1995-98; independent factors: prematurity, SGA, low birthweight, asphyxia, mother’s age, and Cesarean section). Another multiple logistic regression analysis was performed in study III with CHD as dependent variable and with prematurity, SGA, low birthweight, asphyxia, mother’s age and Cesarean section as independent factors. To find statistically significant positive predictors for postoperative mortality and 5-year survival in study IV, a multiple logistic regression analysis was carried out for the following variables: time period of birth, operation/non-operation, age at operation, DS/non-DS, and gender. The dependent variable was survival at 5 years of age. As the period of birth showed significant interaction with both operation/nonoperation and age at operation, correction factors were calculated and introduced in the statistical model. P values less than 0.05 were considered significant. The Swedish standard growth chart was used as reference118 concerning GA and SGA. The studies were approved by the Ethics Committees of the Medical Faculties at Uppsala University, Göteborg University, Lund University and the Karolinska Institute, Stockholm.. 27.

(39) Results. Paper I The prevalence of DS in the region covered in study I was 1.2 per 1,000 liveborn, and the male:female ratio was 1.26:1. No gender difference was detected in the distribution of CHDs. A CHD was reported in 47.5%, GIT malformations were present in 7.3%, other major congenital anomalies were present in 5.5%, and minor anomalies in 5.5%. Among the children with DS and a CHD the two most common CHDs were complete atrioventricular septal defect in 41.3% and VSD in 32.7%, and in addition 10.6% had other CHDs, 7.7% suffered from ASDII , 5.8% from ASDI, and 1.9% from PDA. Additional heart lesions were reported in 28.8% of the children with CHD, and the most common was PDA 7.7%. In the 14.5-year follow-up of 213 children, the survival rate was 75.6%. Mortality rates during the first year and during the first ten years after birth were 14.6% and 23.5% respectively, with no significant differences in gender. However, the 10-year mortality differed significantly between children with (44.1%) and without (4.5%) a congenital heart defect and the relative risk for death in those with a CHD as compared to those without was 12.3 [ P< 0.0001; 95% confidence interval (CI)= 4.9-31.3]. The mortality rates among the children with AVSD and VSD were 56.1% (23/41) and 26.5% (9/34) respectively. The mortality was highest during the first year of life (Fig. 5). In the group of children with CHD, 80% of the deceased had cardiorespiratory symptoms prior to death, making heart failure and respiratory infections, mainly pneumonia, the most frequent causes of death. In children without CHD, there were several causes of death; two died of leukemia, one of sudden infant death syndrome (SIDS), one of complications in connection with surgery for duodenal atresia, and one of an unknown cause. Pulmonary hypertension was diagnosed in 40.2% (41/102) of the children with CHD. During the 10-year follow-up only 25% underwent heart surgery.. 28.

(40) Sixteen children had GIT malformations and 7 died during the 10-year follow-up period. A combination of GIT and CHD malformations was present in 11 children and in this group the mortality was high (6/11).. number of children. 35 30 25 20 15 10 5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 age at death (years) Figure 5. Ten-year mortality (numbers/year) in children with DS born in 1973-80 in northern Sweden,by age at death. Filled bars: 45 children with CHD; unfilled bars: 5 children without CHD.. Leukemia developed in 2.4% (5/213) of the children and 3 of these 5 children died. Malformations other than CHD and those of the gastrointestinal tract did not influence the mortality in our material.. Paper II The mean neonatal/post-neonatal care period in children with DS born in 1995-98 was reduced by more than 50% compared to those born in 1973-80, from 26.7 days to 10.7 days (p<0.05). In the Swedish official statistics, days exceeding 28 are not included in the calculation of mean neonatal care, hence the Swedish infants in general received neonatal care for a mean of 6.2 and 4.25 days in the same time periods. When days exceeding 28 were excluded in the DS groups studied, the mean neonatal care period was 14.3 days in 1973-80 and 9.9 days in 1995-98. The mean number of inpatient days per month decreased as the children grew older and differed significantly between CHD-DS children and H-DS children without any congenital malformation (Fig. 6). IDM differed. 29.

(41) significantly according to which group of congenital malformation the child. p<0.0001 1.6 1.4 1.2. H-DS (n=90). IDM. 1 CHD-DS (n=211). .8 .6 .4. p<0.005 p<0.0001. .2 0. 0-3 years. 4-6 years. 7-10 years. belonged to. Figure 6 . Mean number of inpatient days per month, (IDM) during different age intervals in children with Down syndrome (DS) born between 1973 and 1980 in the northern part of Sweden. Children with DS and a congenital heart defect (CHD-DS) are compared with those without any congenital malformation (H-DS). In the CHDDS group one child with hypospadias, one with a limb reduction defect and one with optic atrophy in addition to the CHD are included.. The CHD-DS children had more than 12 times as many IDM than the children of the H-DS group. The presence of a GIT or other malformation increased IDM four- and twofold respectively compared to healthy children with DS. Children with a combination of CHD and a GIT malformation had the highest IDM, and this group also had the highest mortality rate (derived from study I) (see Table 2). The main reason for admission in the CHD-DS children was a cardiopulmonary disorder. However, this group also had the largest number of admissions due to infections, for which IDM was nine times greater than in the H-DS children. During the first ten years of life the children with DS and a CHD accounted for 71% of all inpatient days due to infections. No gender differences were observed.. 30.

(42) Table 2. Mean numbers of inpatient days per month (IDM) and mortality rates in the first 10 years of life in 211 children with Down syndrome (DS) born in 1973-80 in the northern part of Sweden, including DS children with different categories of congenital malformations and children with DS without such malformations.. Congenital malformation CHD GIT CHD+GIT other healthy total. Number of children n (%) 91 (43.1). 5. (2.4). 11 (5.2). 9. (4.3). 95 (45.0). IDM 0-10 years/ child 1.35. 0.48 3.12. Mortality 0-10 years, number of children* n (%) 39 (42.9%). 1 (20.0%) 6 (54.5%). 0.24. 0. (0.0%). 0.11. 4. (4.2%). 211 (100). 50. * mortality figures from study I. CHD, congenital heart defect; GIT, gastrointestinal tract.. There was a tendency, though not statistically significant, to lower utilization of medical care in children born in the later part of the study period.. Paper III The studied neonatal and maternal factors at birth in the children with DS in study III differed significantly from those of the general population. The maternal age increased in the general population and the same trend was seen in the DS mothers. The proportion of mothers aged 35 years or older in 1995-98 was 40% in the DS group and 13.3% in the general population. The relative number of Cesarean sections increased over time in both the DS group (14.5% to 40.0%) and the population at large (8.8% to 13.6%). The increase was more pronounced in the DS mothers also after adjustment for the effect of the increased maternal age in that group. In 1995-98 the relative risk of Cesarean section in the DS group compared to the general population was 2.0 (95% CI: 1.4-2.8). Cesarean sections were 10% more frequent in mothers aged 35 years or older as compared with younger mothers (<35 years). There were 5% more Cesarean sections in the DS groups as compared with the general population, and these 5% seemed to be due to the diagnosis of DS, regardless of the mother’s age. In the multiple logistic regression model used, no change over time in the proportions of children with asphyxia, prematurity, and low birthweight, and. 31.

(43) SGA children were detected between the two studied populations with DS born in 1973-80 and in 1995-98. In the latter DS population the proportion of children born prematurely was 4 times greater than in the general Swedish population. The proportion of children with asphyxia was 6.7 times greater, the proportion of children with low birthweight was 3 times greater, and the proportion of SGA children was 4.5 times greater (Fig. 7).. 30. p<0.0001. DS. 25. 25. GP. 20. % 15. 10. p<0.0001 p<0.0001. p<0.0001. 13,7. 10,1 8 6,3. 5. 4,2 2,3. 1,2. 0. Apgar <6. GA <36. BW<2500. SGA. Figure 7. Proportions (%) of children born in 1995-98 in northern Sweden with asphyxia (Apgar score at 5 min < 6.), prematurity and low birthweight, and the proportions of SGA children in the study population with Down syndrome (DS)(n=88) and in the Swedish general population (GP). Gestational age is based on ultrasound measurements.. In another multiple logistic regression model, with CHD as a dependent variable, the children with CHD did not differ significantly from those without CHD when adjustments were made for prematurity, SGA, low birthweight, asphyxia, mother’s age and Cesarean section. The neonatal and infant mortality rates in the DS children were significantly higher than those in the general population during both the study periods. The rates decreased between the two periods both in the DS population and in the population at large. In the DS children the reduction in neonatal mortality was near significance (p=0.051), but the reduction in infant mortality from 14.2% to 2.3% was highly significant.. 32.

(44) Paper IV The results of study IV confirmed a female predominance in AVSD, with an over-all female:male ratio of 1.3:1. The gender difference seemed to be especially pronounced in the non-DS group with isolated AVSD, where this ratio was 4.5:1. The prevalence of AVSD in 1973-97 was 0.31 per 1,000 liveborn and the prevalence of AVSD was approximately 16% in children with DS. The mean observation time for the survivors was 13.4 years (4,670 observation years) A total of 801 children with diagnosed AVSD were included in the study, of whom 68.9% had DS. The total mortality in cases of all AVSD (including both the intermediate and complex forms) for the whole study period 19732001 was 48.8%; in the DS group it was 43.4% and in the non-DS group it was 60.8 % (p<0.0001, 95% CI 9.8-24.7).. Isolated AVSD Isolated AVSD was present in 502 children, of whom 86% had DS. The mean age at operation decreased and the rate of operations increased over 100. 97. 90. 1973-77 n=97 1993-97 n=109. 80. 70. 60 51,6 50. 40. 39,9. 39 35,5 28,1. 30. 20 10,1. 10. 9,5. 4,2 1 0. Age at op, months. Op %. Opmort %. 5-year opmort %. Total opmort %. Figure 8. Children with isolated AVSD and Down syndrome (DS) in Sweden born 1973-77 and 1993-97. Mean age at operation in months, and rates (%) of anatomical correction (op), rates of early postoperative mortality (opmort), 5-year cumulative postoperative mortality (5-year opmort, but in children born 1997 only 4-year) and rate of cumulative postoperative mortality to end of study in December 2001 (total opmort).. 33.

(45) time in children with isolated AVSD, and these changes were accompanied by a decrease in postoperative mortality (Fig. 8). Survival improved continuously, but the greatest improvement seemed to be achieved in the last part of the study period, 1993-97 (Fig. 9).. 100. Cumulative survival %. 97. 1993-97 (n=109). 80 80. 1988-92 (n=123) 54. 60. 1983-87 (n=95) 34. 1978-82 (n=75). 40 33. 1973-77 (n=97). 20. 0 0. 50. 100. 150. 200. 250. 300. 350. 400. 450. Months Figure 9. Cumulative survival to endpoint of study in December 2001 (mean observation time 13.5 years), in children born with isolated AVSD 1973-97 in Sweden (n=502), illustrated by a Kaplan-Meier survival curve. Distribution for different periods of birth. The number of survivors and total number of liveborns (n) in each time period are given.. The mortality among children with isolated AVSD decreased significantly over time, both in those with and those without DS (Fig. 10). The cumulative postoperative mortality at the endpoint of the study in December 31, 2001 was 27.5% (28 % in the group of DS children, 22.2% in the non-DS group (ns). In contrast, among the children with isolated AVSD not treated surgically, the mortality was 79.7% (98/123); in the DS group 77.8% (84/108) and in the non-DS group 93.3% (14/15) (ns). The odds ratio (OR) for death among children who did not undergo surgical correction decreased over time (Table 3).. 34.

(46) Table 3. Children with isolated AVSD and Down syndrome born in 1973-97 in Sweden. The risk for death in children treated surgically, compared to those not treated surgically expressed as an odds ratio (OR) for different time periods, where 1 means equal risk of death for both operated and not operated. The mean age´s at operation (months) .. Year of birth. OR. Mean age at operation Months. 1973-77 1978-82 1983-87 1988-92 1993-97. 0.91 0.32 0.15 0.06 0.02. 40 22 10 7 4. 100 1993-97 (n=12). 11 86. DS Non-DS. 1993-97 (n=97). Cumulative Survival %. 80. 60. 7. 1983-87 (n=11). 47. 1983-87 (n=84). 40 28. 1973-77 (n=81). 5 1973-77 (n=16) 20. 0 0. 50. 100. 150. 200. 250. Months. 300. 350. 400. 450. Figure 10. Cumulative survival to endpoint of study in December 2001 (mean observation time 13.5 years) in children with isolated AVSD in Sweden with and without DS, born in 1973-77, 1983-87 and 1993-97.. 35.

(47) Using a multiple logistic regression model, no significant difference in mortality either between genders, between DS and non-DS children, or between children operated on before and after 6 months of age was observed. The 5-year postoperative survival in DS children with isolated AVSD increased from 65% in 1973-77 to about 90% in 1993-97 and the same trend was observed in children without DS.. Birth and death within the same period (%). 100. all AVSD,DS all AVSD, non-DS Isolated AVSD,DS. 90 80. Isolated AVSD,non-DS. 70 60 50 40 30 20 10 0 1973-77. 1978-82. 1983-87. 1988-92. 1993-97. Years Figure 11. Death rate (%) among children with AVSD (n=801) and isolated AVSD (n=502) with and without Down syndrome (DS). Birth and death within the same time period.. When the mortality was described as birth and death within the same time period, 1973-77, 1978-82, 1983-92, 1993-97, an interesting trend was observed, with a continuous decline in mortality among the non-DS children, while in the DS group of children most of the decrease in mortality occurred in the last period studied (Fig. 11).. 36.

(48) Isolated AVSD and GIT malformations The prevalence of GIT malformations in the 502 children with isolated AVSD was 8%; 8.5% in those with DS and 5.9% in those without DS. These figures are in accordance with the rates of GIT malformations derived from study I. In study IV the 1-year mortality in DS children born in 1993-97 was reduced to 2/9 (22%) compared to 3/5 (60%) in corresponding children born in 1973-77.. 37.

(49) General Discussion. The purposes of our studies were to investigate the impact of congenital malformations on morbidity and mortality in children with DS, the outcome at birth in DS as reflected by perinatal data, especially in the presence of CHDs, and the survival in children with AVSD combined with DS compared to that in non-DS children with such defects. The summarized results point to CHD as the major cause of death and morbidity in DS. The studies indicate that the newborn child with DS is more vulnerable than the healthy child, irrespective of the presence or absence of CHD. The advances in the management of children with DS and complete AVSD during the last decades has led to a very low postoperative death rate, and increased survival. Down syndrome does not seem to be a risk factor for a poor outcome after repair of isolated AVSD.. Methodological considerations The studies described in this thesis were made on population-based cohorts of children with DS born in 1973-80 and 1995-98, and a similarly based cohort of children with AVSD born in 1973-97. The region in northern Sweden covered in studies I-III constituted 25% of the Swedish population. As there was good correspondence between compiled data from children in northern Sweden in 1995-98 and data from children in all Sweden, we assume that the region is large enough to be able to generalize the results to Sweden as a whole. Concerning the completeness of our data, studies I-III were based on information from three registers, with a very low frequency of missed cases (1% in the DS cohort of 1973-80 and <0.5% in that of 1995-98). In a study by Lindsten et al.,25 use of the same registers ascertained approximately 99% of all subjects with DS. Furthermore, owing to the stability of the populations, very few subjects moved out of Sweden, and in the follow-up studies I-IV very few children were lost to follow-up. The Swedish system of 10-digit personal identification numbers simplified the tracing of medical records, and children who had moved to other parts of Sweden could easily be found. In study IV, the frequency of missed cases, especially in the. 38.

(50) beginning of the study when the registers were not complete, was difficult to assess. However, the observed distribution of AVSD is consistent with the numbers of children born with DS and the total number of all births in Sweden, making a significant number of undiagnosed cases unlikely. The decline in the number of children with AVSD in the late 1990s follows the decreasing birth rate in Sweden (Fig. 4). Furthermore, the prevalence of AVSD found in the present study is similar to that reported from the Baltimore-Washington Infant Study, i.e., 0.362 per 1,000 livebirths.3 In study II the collection of data was performed in two different ways, as the Hospital Discharge Register only covers data from 1987 onwards. The comparison between the manual recording of admissions and the use of the register for the year 1992 indicated that 5% of the reports were lost with the manual method. In study III neonatal data of 28 children from the 1973-80 cohort could not be found in the Medical Birth Register and were therefore retrieved from the previous study,12 when information was obtained directly from the medical records. Elimination of these 28 children from the study did not alter the results, and they were therefore included. The information on CHD was also retrieved in two different ways. In children born in 1973-80 the information was extracted directly from the medical case records, and in 1995-98 it was derived from the Hospital Discharge Register. This register has information on 99% of all inpatient care and contains information on up to six main diagnoses. The frequency of missed cases with CHD in the cohort of children born in 1995-98 is difficult to assess. A National Health supervision program for the care of children with DS was introduced in 1990 and the recommended policy of an echocardiogram before 3 months of age was rapidly applied all over Sweden.119 Many of the children with DS had this echocardiogram before leaving the neonatal care unit, and according to a study by Amark et al. the median age at confirmation of the diagnosis of AVSD was 10 days in children born 1991-96 in Gothenburg.114 As the rate of CHD was increased in the later cohort, no great loss is likely.. At birth (Papers II and III) The prevalence of DS was 1.2 per 1,000 liveborn in 1973-80 and has remained unchanged to the present time. There was a male predominance, a well known fact in DS, with a male:female ratio of 1.26:1.91 The trend in the general population in Sweden, with an increasing maternal age and rate of Cesarean section, was also true in the population of mothers of DS children. However, the births of the children with DS differed from those of the normal children in many respects.. 39.

(51) The DS mothers were older than the mothers in the general population (by approximately 4 years), and in 1995-98 the proportion of mothers older than 35 years was three times that in the population at large (40% vs 13.3%). In that time period the rate of Cesarean sections was twice as high in DS pregnancies as in the general population. The reasons for this difference remain unclear. One contributory factor was the age of the mothers - older mothers are known to have more complicated deliveries. Another reason may be that measures taken in the antenatal health surveillance program could have detected a child with DS and possibly a CHD prenatally, leading to precautions. The DS diagnosis seemed to raise the rate of Cesarean sections by 5% compared to that in the population at large, regardless of the age of the mother. A similar rate of deliveries by Cesarean section in pregnancies with a DS infant was reported by Nielsen et al. in 1987.120 The DS children seemed to be more fragile at the time of birth, irrespective of the presence of a CHD. At birth they presented prematurely (mean 1 week) with reduced Apgar scores and a six times more frequent rate of asphyxia (8% vs 1.2%), low birthweights (13.7% vs 4.2%) and increased rates SGA ( 13.5% vs 3%). These findings are in accordance with reports by others.73,121 The mean duration of neonatal care was reduced both in the DS and in the general population, but the number of days in the DS children remained on a level twice that of the children in the population at large. The decrease is a result of efforts to reduce hospitalization,122,123 and to support parents and help them to cope with their new role in their home situation.. Congenital malformations (Papers I and IV) Many different congenital malformations can be present in children with DS. In study I we found no significant difference in the distribution of congenital malformations between genders. The main congenital malformations were CHDs, which were present in 50%. The second most frequent were GIT malformations (7%) and the rest were different solitary major malformations (5.5%). Some of the malformations needed immediate surgery in order to save the life of the child. The rate of CHDs in our study was higher than in some contemporary studies,91,124 but consistent with more recently reported prevalence rates.4,87 The rate seemed to increase with time, however, and this must be due mainly to the advances in the detection of CHDs by the echocardiography technique.125 The rate of CHD was not found to be higher in children of older mothers, a finding consistent with an observation by Fixler and Threlkeld in 1998.126. 40.

(52) Several different CHDs may be seen in children with DS. The main congenital heart malformation appeared to be AVSD (41.3%)(study I), a defect well known to be associated with Down syndrome.82,127 Isolated AVSD seemed to be an almost DS-specific CHD, as 86% of the children with isolated AVSD had DS, in accordance with the report from the Baltimore-Washington Infant Study.128 The children with AVSD and DS more seldom had other associated complex CHDs,129 in contrast to the AVSD non-DS group of children (study IV). The predominance of females among children with AVSD was confirmed, with an over-all female:male ratio of 1.3:1.84,92 This gender difference seemed to be especially pronounced in the non-DS group with isolated AVSD and intermediate AVSD, where the ratios were 4.5:1 and 2.1:1 respectively. The reason for this female predominance in AVSD is still unknown. Other relatively frequent CHDs were VSDs (32.7%) and ASD II (7.7%), and PDA was the most common additional heart malformation, in accordance with other studies.4,49,85,87,89. Morbidity (Paper II) Few studies reported in the literature have addressed the use of medical services in relation to congenital malformations and DS. Morbidity is difficult to assess, as it depends on many different factors that change over time. Analysis of the utilization of medical care in terms of IDM is one way of elucidating this, and this variable was used to reflect the most serious episodes of medical care. A CHD was the congenital malformation that caused the greatest increase in morbidity. A covariation between morbidity and mortality was seen in our studies, with both high morbidity and high mortality in children with a CHD. We found that the number of IDM was more than twelve times greater in the CHD-DS children than in the H-DS group. It has been reported that children 4-18 years of age with another serious congenital malformation, myelomeningocele, used hospital services seven times more frequently than the controls.62 Recent research in Sweden indicated that in DS children without a CHD, the frequency of illness in the children in terms of sickness absence of the mother was three times higher than in the healthy controls.80 In a Scottish study in 1984, Murdoch130 found that the number of contacts with the family doctor was doubled in DS children compared to healthy controls. The small group of children with a combination of CHD and GIT malformations were the most seriously ill, with conditions frequently needing urgent surgical operations for survival, and they also had the highest mortality rate.. 41.

(53) The utilization of inpatient care by the DS children decreased with age, as in most other children.62,123,131 The reasons for medical care revealed in study II were mainly connected with CHDs, and DS children with a CHD also accounted for the majority of infections. Whether this is also true after correction of all CHDs remains to be investigated.. Mortality (Papers I, III and IV) The life expectancy in DS has increased during the last decades as a result of progress in surgical and medical care.91,132,133 The congenital malformations present in DS were the main causes of the high mortality observed in study I.12 The frequently present CHDs were not operated on at all, or very late, in the 1970s, and only 25% of the children with CHDs born in 1973-80 had surgery. In children born in the 1970s the diagnosis of an AVSD could be delayed several months in the absence of symptoms, thereby rendering surgical correction impossible. With increasing evidence of an elevated risk of early development of irreversible pulmonary hypertension in DS children with AVSD, early corrective surgery in such children103,108 began to be recommended in the 1990s and was rapidly the general rule. The mortality rates in studies I and II will therefore serve as references of historical mortality figures, but the mortality rates in healthy DS children are of importance to remember in these comparisons. The first year of life in children with DS used to be the most critical,91,134 as could be expected (Fig. 3). The infant mortality decreased significantly over time both in the general population (0.9% to 0.4%) and in the DS cohorts (14.2% to 2.3%). The analyses showed that the mortality in 1973-80 and the observed reduction of mortality in the DS population were mainly attributable to the children with CHDs (studies I and III). The same results were obtained in the population of DS children with isolated AVSD, among whom the infant mortality was reduced from 22% in 1973-77 to 8% in 1993-97 (study IV). One explanation for this decline in infant mortality might be the more frequent use of early surgical correction of CHD in DS infants.135 No gender differences were found in the neonatal or infant mortality in the present studies. Compared to the mortality rate for all Swedish children in 1980 (0.75%), the 10-year mortality in healthy DS children was increased fivefold (4.5%). This latter figure is probably even lower at present with the new leukemia treatments available.61,136 In the DS children with a CHD the 10-year mortality was almost 60 times higher (44.1%) than that in the general. 42.

(54) population in 1980. This mortality rate was very similar to that of the DS children with isolated AVSD in 1973-77 (46%). The major causes of death among the children with CHD were cardiorespiratory disorders, in line with findings in other studies.10,137-139 Among the deceased children without a CHD, leukemia was the cause in a few cases. Whether this spectrum of causes of death has changed remains to be investigated (study I). The mortality in AVSD was analyzed in study IV, and a significant difference in mortality was found between children with and without DS in the cohort of all AVSD. This difference was probably mainly due to morphological dissimilarities, as there were more additional severe CHDs in the non-DS group129 (Fig. 11). In the group of children with isolated AVSD the 30-day and the 5-year cumulative postoperative mortality were both high in the beginning of study IV and were significantly reduced at the end. The results showed that with the competence and policies for treatment of AVSD in Sweden in the 1990s, when children with DS had the same options for surgery and care, the results of surgical correction seemed to be equally successful in DS and non-DS children. It was apparent that in the early years of the study period, DS children with AVSD did not receive the same treatment as the chromosomally normal children with this heart defect, but by the end of the study period almost all children with isolated AVSD underwent corrective surgery at an age below 6 months, with a very small postoperative mortality rate, in accordance with other studies.140,141 Despite indications that children with DS are more prone to develop pulmonary hypertension at an early age, age at operation was not found to be an independent factor for survival in the logistic regression model. The increase in survival to 5 years of age was statistically more an effect of increased use of corrective surgery and other improvements in the care of these children with time, than an effect of lower age at operation per se. The progress in the treatment of other associated malformations may have contributed to the outcome, a trend found in study IV among children with AVSD and GIT malformations. In 1993-97 only a small proportion of children with isolated AVSD were considered inoperable, because of irreversible pulmonary hypertension, prematurity or other additional major malformations. The greatest reduction in mortality, although there had been a continuous decline since the beginning of the study, occurred in the last period, 1993-97 (Figs. 9 and 11). The same observation was made by Yang and co-workers, who found that the median age at death in children with DS and CHD increased substantially after 1992, in a study of death certificates from 198397 in the US.139. 43.

(55) The survival curve for DS children with AVSD is now approaching that for children with DS without CHDs. The 10-year mortality in healthy children with DS was 4.2% in 1973-80, and today is probably even lower. With a longer observation period than in the present study, time will show the long-term outcome of surgical interventions in these children. Some of them may suffer from progressive pulmonary hypertension in spite of early operation, and others may develop atrioventricular valve dysfunction or dysrhythmias. It seems that factors other than heart surgery will dominate as causes of death in the future.. Reflections and future directions Great achievements have been made in the management of children with Down syndrome and a CHD. The postoperative mortality in AVSD has decreased dramatically and the 5-year postoperative mortality is now around 10%, but longer follow-up is needed. The surgical results in the management of children with AVSD and DS need to be followed up continuously to assess the long-term mortality. How the affected and repaired heart will function 30-40 years later is essential to elucidate in studies of adolescents and adults. Another important question to be addressed in future research is whether the occurrence of infectious disorders has decreased since the introduction of the recent treatment strategy of early surgical intervention for heart malformations in children with DS. There are still many factors contributing to the susceptibility to infections. However, as a CHD seemed to have strong influence on the morbidity in terms of inpatient days, and the majority of the children did not undergo corrective surgery, the question of the impact of a CHD on the morbidity needs to be addressed again. The altered care nowadays, with early correction of CHDs, might have lowered the rate of infections. A decrease in morbidity would increase the quality of life for the child and reduce some of the burden on the parents and family. It is evident from the present studies that in earlier years there was a delay in the availability of medical achievements to the population of children with DS compared to the treatments offered to children with the same malformations but without DS. Many parents and doctors were also reluctant to consider surgical correction, in view of the high risk of mortality at surgery at that time, and probably also because of a different attitude toward children with mental retardation. However, views and policies have changed and medical care and surgery have improved. The result of this development is the growing population of children, adolescents and young adults now surviving with a better quality of life. The policy makers and. 44.

(56) administrators of the health care system in Sweden must be made aware of this new situation. More investments are needed in extended health care programs, and more support for this group of children, adults and their families.. 45.

No results found