- Email: [email protected]
Temporal variability in birth prevalence of congenital heart defects as recorded by a general birth defects registry
TEMPORAL VARIABILITY IN BIRTH PREVALENCE OF CONGENITAL HEART DEFECTS AS RECORDED BY A GENERAL BIRTH DEFECTS REGISTRY GIULIANO BOSI, MD, GIAMPAOLO GARANI, MD, MARCO SCORRANO, MD, ELISA C ALZOLARI, MD, AND THE IMER WORKING PARTY*
Objective Our aim was to examine the temporal variability in congenital heart defect (CHD) birth prevalence from 1980 to 2000 in Emilia-Romagna, Italy. Methods The study population consisted of all infants, surveyed by the Emilia-Romagna birth defects registry (Indagine Malformazioni conpenite in Emilia-Romagna [IMER]), who were affected by CHDs. A simplified classification into “simple” and “complex” CHD was adopted. A comparison with another epidemiologic study using different methodology in the same area was performed.
From 1980 to 2000, IMER ascertained 2442 live births with CHD of 480,793 infants born, with an average CHD birth prevalence of 5.1% (Range, 3.1% to 7.5%). A significant increase in prevalence of simple CHD during the second decade of the study was demonstrated because of an increased recognition of “minor” cardiac lesions among the simple CHD. The birth prevalence of complex CHD remained stable.
Results
Conclusions The apparent increase in live births with CHD results mainly from the current widespread availability of color Doppler echocardiography, which allows the early detection of the “minor” cardiac defects. Other differences are the result of the sources of ascertainment, diagnostic criteria, system of classification, and especially the age limit for enrolling infants with suspected CHD. (J Pediatr 2003;142:690-8)
ongenital Heart Defects (CHDs) are among the most common of all congenital malformations (CMs) and have a major impact on pediatric morbidity, mortality, and healthcare costs. Recent population-based epidemiologic studies have indicated a CHD prevalence ranging from 3.5% to 13.7% of live births.1-6 In Italy several general birth defects registries, especially those in the north and center of the country, have monitored CM since the late 1970s. Data suggest a CHD birth prevalence around 4.5% of live births.7 The Italian Multicenter Study on epidemiology of Congenital Heart Disease (IMS-CHD), performed between 1992 and 1993, suggests an average CHD prevalence varying from 3.2% in the South to 6% in Northern Italy.8 The aims of this study were, first, to verify the birth prevalence of CHD in a well-defined population as registered by a general birth defects registry; second, to examine temporal variability in the prevalence of CHD during a 21-year period of registration (1980-2000); and third, to evaluate the association with extracardiac anomalies. Finally, taking advantage of being part of the IMS-CHD, we have the opportunity to evaluate similarities and differences in these two epidemiologic studies, comparing data collected in the same period of time with different methodological approaches.
C
AoCo AoSt AVSD ASD II CAT CHD CM DIV DORV ICD-9-CM
690
Aortic coarctation Aortic stenosis Atrioventricular septal defect Atrial septal defect type ostium secundum Common arterial trunk Congenital heart defect Congenital malformation Double-inlet ventricle Double-outlet right ventricle 9th Revision of the International Classification Diseases
IMER Indagine Malformazioni conpenite in Emilia-Romagna IMS-CHD Italian Multicenter Study on epidemiology of Congenital Heart Disease LHH Left hypoplastic heart syndrome MCA Multiple congenital anomalies PDA Patent arterial ductus PulAtr Pulmonary atresia PulSt Pulmonary valvar stenosis TF Tetralogy of Fallot TGA Transposition of great arteries VSD Ventricular septal defect
From the Pediatric Cardiology and Neonatology Units, Department of Clinical and Experimental Medicine, and Medical Genetics, Department of Experimental and Diagnostic Medicine, University of Ferrara, Italy. *IMER Working Party: Gianni Astolfi, Medical Genetics, University of Ferrara; Guido Cocchi, Institute of Pediatric and Neonatology, University of Bologna; Cinzia Magnani and Pietro Viola, Institute of Puericulture and Neonatal Medicine, University of Parma; and Daniela Prandstraller, Pediatric Cardiology Unit, University of Bologna. Submitted for publication Sept 20, 2002; revision received Jan 28, 2003; accepted Mar 18, 2003. Reprint requests: Giuliano Bosi, MD, Pediatric Cardiology Unit, Department of Clinical and Experimental Medicine, University of Ferrara, 44100 Ferrara, Italy,Via Savonarola 9. E-mail: bsg@unife .it. Copyright © 2003, Mosby, Inc. All rights reserved. 0022-3476/2003/$30.00 + 0
10.1067/mpd.2003.243
Fig 1. Year-by-year birth prevalence of CHD in live births: 21 years of registration by IMER (1980-2000; see text for details).
METHODS Population and Data Sources The study population consisted of all infants born from 1980 to 2000 and surveyed by IMER, a general birth defects registry covering a well-defined area (the Emilia-Romagna region in Northern Italy). The general characteristics of this registry, which participates in the European Registry of Congenital Anomalies and Twins (EUROCAT) and the Clearinghouse for Birth Defects Monitoring System, have been previously described.9,10 In summary, this registry includes all live births and stillbirths of at least 28 weeks’ gestation, whose mother is a resident in the area, with any CMs detected at birth. Data regarding the infant, pregnancy, parents, and other family members are collected through an interview with the mother during the hospital stay. In the IMS-CHD, whose preliminary results have been recently published, the study population consisted of all infants born from 1992 to 1993 in 15 provinces spread all over the country who were examined by 18 centers of pediatric cardiology. The peripheral units concerned with maternal and pediatric care, along with the family pediatricians, constituted the main sources of information. Data were collected using a questionnaire similar to that used by the EUROCAT study.6 In contrast with the IMER registry, any new case of CHD in a newborn or in a child up to 2 years of age, whose mother was resident in the study area, was been registered prospectively.8 Temporal Variability in Birth Prevalence of Congenital Heart Defects as Recorded by a General Birth Defects Registry
Case Definition and Diagnostic Hierarchy All cases of CHD detected in stillbirths and live births were prospectively enrolled by the IMER registry using the 9th revision of the International Classification Diseases (ICD-9-CM).11 In contrast, the ascertainment of live births with CHD, born in 1992 or 1993 in the same region and prospectively enrolled by the IMS-CHD, was based on a simplified classification suggested by Pexieder and others.6,8,12 The comparison between these two epidemiologic studies, limited to two years of observation (1992 and 1993), has been accomplished by evaluating the data from the provinces covered by both registries. Moreover, all cases of CHD from the IMER registry were retrospectively categorized by the pediatric cardiologists operating in the same units, according to the classification criteria used by the IMS-CHD. Consequently, a common nomenclature system, using a classification into “simple” and “complex” CHD, was adopted to harmonize the collected data. Simple CHD included isolated major entities in the condition of visceral and atrial situs solitus, atrioventricular concordance, and levocardia, such as aortic coarctation (AoCo), aortic stenosis (AoSt), atrial septal defect type ostium secundum (ASD II), atrioventricular septal defect (AVSD), left hypoplastic Heart syndrome (LHH), patent arterial ductus (PDA), pulmonary valvar stenosis (PulSt), tetralogy of Fallot (TF), transposition of great arteries (TGA), and ventricular Septal defect (VSD). 691
Fig 2. Comparison of the prevalence of any individual CHD between the first period of the study (total CHDs, n = 972) and the second period (total CHDs, n = 1470). *P < .001; **P < .0001. 99% CL is reported.
Fig 3. Association of any individual CHD with extracardiac anomalies (including either known or unknown conditions) in relationship with isolated CHD (see text for details).
692 Bosi et al
The Journal of Pediatrics • June 2003
Table I. Birth prevalence of any individual CHD in live births (10,000 + 99% CL), as registered by the IMER from 1980 to 2000, using a classification into “simple” and “complex” CHD
CHD
Live births
Simple CHD 2157 VSD 947 ASD II 162 PDA 55 PulSt 104 TF 106 AoSt 35 AoCo 99 LHH 78 AVSD 134 TGA 125 Miscellaneous 211 Not defined 101 Complex CHD 286 DIV 60 CAT 34 DORV 21 PulAtr (with or 59 without VSD) TricAtr 35 Heterotaxia 29 Total anomalous 16 pulmonary venous return Ebstein’s 10 AoArcInt 7
% CHD
Prevalence 10,000 (± 99% CL)
(522) (123) (70) (29) (16) (27) (7) (21) (13) (100) (8) (56) (52) (66) (16) (11) (17) (8)
88.3 38.8 6.6 2.3 4.3 4.3 1.4 4.0 3.2 5.5 5.1 8.6 4.1 11.7 2.5 1.4 0.8 2.4
44.86 ± 2.9 1970 ± 1.90 3.37 ± 0.80 1.14 ± 0.46 2.16 ± 0.64 2.20 ± 0.64 0.73 ± 0.37 2.05 ± 0.64 1.60 ± 0.55 2.79 ± 0.72 2.60 ± 0.70 4.39 ± 0.90 2.10 ± 0.63 6.00 ± 1.10 1.25 ± 0.50 0.71 ± 0.36 0.44 ± 0.30 1.23 ± 0.50
(4) (16) (3)
1.4 1.2 0.6
0.73 ± 0.36 0.60 ± 0.30 0.33 ± 0.25
(0) (1)
0.4 0.3
0.21 ± 0.19 0.15 ± 0.15
The number of live births with associated extracardiac anomalies is in brackets (see text).
Complex CHD included cardiovascular lesions that did not fit into defined categories and required a more detailed description of each lesion, such as common arterial trunk (CAT); double-inlet ventricle (DIV); pulmonary atresia (PulAtr) with or without VSD; double-outlet right ventricle (DORV); Ebstein’s anomaly; hetorotaxia (including dextrocardia and left and right atrial isomerism, usually associated with major cardiac lesions); total anomalous pulmonary venous return; and tricuspid valve atresia (TrAtr). Moreover, we defined as miscellaneous CHD as cardiovascular malformations characterized by an association of more than one defect, the main being usually represented by VSD and/or ASD II, in association with other minor cardiac defects. Premature infants (<37 weeks’ gestation) with PDA were excluded. CHD was considered “isolated” when no major extracardiac malformations were present and “associated” when at least one major additional extracardiac malformation was Temporal Variability in Birth Prevalence of Congenital Heart Defects as Recorded by a General Birth Defects Registry
Fig 4. Comparison between the CHD prevalence as registered by IMER and by IMS-CHD during 2 years of registration in the same geographic areas (see text for details). **P < .0001. 99% CL is calculated.
found. CHD diagnosed within syndromes, sequences, or associations of known or presumed separate etiology were considered among recognized conditions. All remaining cases associated with major extracardiac malformations were considered as multiple congenital anomalies (MCAs). We included only live births in the present study because the proportion of stillbirths was very small. For the same reason, fetal and/or diagnosis at autopsy of CHD in spontaneous and/or induced abortions were excluded. The diagnostic techniques included clinical evaluation, echocardiography, diagnostic and/or interventional catheterization, surgery, and necropsy.
Statistical Analysis The χ2 test with Yates correction was used to assess the statistical differences between (1) the CHD prevalence calculated in the first and second period of the study and (2) the data coming from the 2 epidemiologic studies we compared. Statgraphics V4.0 (STSC, Inc, Rockville, Maryland) was used for data processing.
RESULTS Consecutive live births (n = 480,793) and stillbirths were surveyed by the IMER registry from 1980 to 2000 with a mean of 22,894 per year. In the 21 years of the study, 2456 CHD cases were registered; among these, 2442 cases were detected in live births with a birth prevalence ranging from 3.1% to 7.5%, the average being 5.1% (99% CL= 0.3) (Fig 1). Table I shows the prevalence of any individual cardiovascular malformation using the classification into simple and complex CHD. In the first group, isolated VSD occurred with an average prevalence of 19.7 per 10,000 live births. Isolated ASD II had a prevalence of 3.4 per 10,000 live births. AVSD and TGA formed a cluster with a prevalence around 2.7 per 10,000 live births. A second cluster, including TF, LHH, PulSt, and isolated CoAo, had a prevalence from 1.6 to 2.2 per 10,000 live births. PDA in mature infants had a prevalence of 1.2 and AoSt 0.7 per 10,000 live births. 99% confidence limit for any CHD is reported in Table I. 693
Table II.Year-by-year birth prevalence of any individual “simple” CHD in live births as registered by IMER from 1980 to 2000 (see text) Simple CHD
1980
VSD 10.42 PDA 0.00 ASD II 1.60 PulSt 1.60 TF 0.00 TGA 1.60 AoSt 0.00 AoCo 0.00 AVSD 1.60 LHH 0.80 Miscellaneous 0.80 Not defined 8.82 No. of 12.471 live-born
1981
1982
1983
1984
1985
1986
1987
1988
1989
9.08 0.70 1.40 0.00 0.70 2.79 0.70 0.70 2.79 2.09 2.79 4.19 14.322
13.07 0.00 0.00 1.63 0.00 2.18 0.00 3.27 2.18 0.54 1.09 2.72 18.362
14.70 0.46 0.92 2.30 1.38 4.59 0.46 3.22 0.92 1.38 2.30 3.68 21.767
17.27 0.43 0.43 1.30 2.59 2.16 0.86 2.16 2.59 2.16 1.73 2.59 23.155
13.74 0.83 1.25 1.67 3.33 1.25 0.42 2.08 2.91 2.50 2.50 2.08 24.016
13.12 2.19 1.75 4.37 3.94 3.50 1.31 2.62 1.75 1.75 2.19 2.19 22.868
19.79 1.76 1.32 2.64 1.32 1.32 0.00 1.76 4.40 1.32 2.20 0.88 22.741
20.62 0.88 1.75 1.75 2.63 1.75 1.32 1.32 3.07 1.32 2.19 1.32 22.794
19.67 0.43 1.28 4.28 2.57 2.14 1.71 1.71 2.57 1.71 2.99 0.86 23.386
Miscellaneous CHDs, as previously defined, represent a significant proportion of all cardiovascular malformations with a prevalence approximately 4.4 per 10,000 live births. Among the complex CHDs, which represent 11.7% of all the cardiovascular malformations, DIV, and PulAtr (with and without VSD) had a prevalence about 1.2 per 10,000 live births each. DORV, CAT, and TrAtr had a prevalence from 0.5 to 1.2 per 10,000 live births. TAPVC and Ebstein’s malformation varied from 0.2 to 0.3 per 10,000 live births. In the same table, we reported the number of CHD cases associated with extracardiac anomalies, including chromosomal anomalies, recognized conditions, and/or multiple congenital anomalies. Figure 1 shows the year-by-year prevalence of simple and complex CHD together with the prevalence of the total number of CHDs. There was a significant increasing trend in the number of simple CHD over time (P < .001), whereas no change was evident in the prevalence of complex CHD. Their apparent increase in the last years of the study is almost certainly accounted for by the random variation of small numbers. Table II, which shows the year-by-year variability in the prevalence of any individual simple CHD, demonstrates that the greatest variation was caused by the increased prevalence of isolated VSD, ASD II, and miscellaneous CHD, with the prevalence of any other CHD not being significantly changed. This is clearly confirmed by Figure 2, which demonstrates a significant increase in the diagnosis of VSD (P < .001) and ASD II (P < .0001) especially in the second decade of the study, together with a significant increase in the proportion of the so-called miscellaneous CHD (P < .0001). Of the infants enrolled, 24% had extracardiac anomalies. Chromosomal anomalies were diagnosed in 9.1% of all live births. Among these, 75.7% were trisomy 21; 6.3% Edward syndrome; 4.1% Turner syndrome, 3.1% Patau syn694 Bosi et al
drome; and ,10.8% other syndromes including DiGeorge syndrome. CHDs were also present within the framework of various syndromes and/or sequences in 4.4% of these cases. MCAs were present in about 10% of the infants: among these, 55% had 1 associated extracardiac lesion; 23% had 2, and the remaining (22%) more than 2 extracardiac malformations. Systems of organs were affected with the following frequency: urogenital system (25.8%); musculoskeletal system (22.3%); craniofacial anomalies, including eye and ear anomalies (19.3%); gastrointestinal anomalies (16.9%); central nervous system (10.6%); pulmonary apparatus (2.6%); and integument anomalies (1.7%). Figure 3 shows the association of any individual cardiovascular malformation with extracardiac anomalies, including either known (syndromes, sequences, and chromosomal anomalies) or unknown conditions (MCAs). Figure 4 shows the comparison between the CHD prevalence as collected by IMER and IMS-CHD Registries during 2 years of registration (1992-1993) in the same geographic area represented by three provinces of the same Region (Fig 4, A-C). The average CHD prevalence registered by IMS-CHD was 8.3%, ranging from 6.2 to 10.7% live births. This value is significantly different from that reported by the IMER registry both for simple and complex CHD, as clearly documented in Figure 5, A and B.
DISCUSSION In Western industrialized countries, recent populationbased epidemiologic studies have indicated CHD prevalence ranging from 3.5% to 13.7% live births.1-6 These significant differences are closely related to numerous methodologic difficulties arising in the collection of epidemiologic data.13 In 1990, several factors biasing the ascertainment of CHD prevalence were discussed by Ferencz, who underlined the sigThe Journal of Pediatrics • June 2003
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
21.24 0.85 2.97 4.25 0.85 2.97 0.85 1.27 2.55 2.12 4.25 1.70 23.545
25.07 0.40 3.98 1.99 0.80 3.98 0.40 1.59 3.18 2.79 1.99 1.19 25.129
23.56 1.20 3.59 4.39 1.60 0.80 2.00 2.00 2.40 2.00 3.19 3.19 25.042
22.83 0.00 6.93 2.85 4.08 2.45 0.00 1.22 1.63 1.22 6.11 3.26 24.533
22.41 1.93 6.57 3.09 1.93 1.93 0.39 1.55 3.09 1.55 8.11 3.48 25.886
25.80 2.28 6.07 0.76 2.66 2.66 1.52 2.66 1.52 1.52 6.07 0.76 26.359
26.67 1.04 7.27 0.69 2.77 3.12 0.00 3.46 4.85 1.73 5.54 2.42 28.873
16.29 0.00 4.50 2.08 2.43 1.73 0.69 2.08 3.12 2.43 8.66 0.00 28.859
21.02 1.91 7.17 0.96 4.78 3.82 0.48 0.48 5.73 3.34 8.12 2.87 20.928
26.25 3.75 2.92 1.25 1.25 3.75 0.42 3.33 4.58 2.50 7.08 2.08 24.003
24.36 2.76 2.76 0.46 6.90 4.60 1.38 4.60 4.14 3.22 8.73 1.38 21.754
nificant role represented by population factors, family factors, community medical practice, and cardiology center practice.14 At the same time, the risk of exclusion of CHD in abortions, in stillborns, and in infants dead before detection of “critical” CHD was stressed by Fixler et al together with the risk of exclusion of minor CHDs that are clinically silent in early infancy.15 Consequently, the age limit up to which infants with CHD are registered, together with the selection of a well-defined population and with the use of a common nomenclature system, represents the more significant variable factors in collecting reliable epidemiologic data.5,6,12,14,15 The current study concerns a well-defined population surveyed by the IMER registry since the late 1970s. Moreover, in this region the distribution of maternity units, in proximity to the pediatric cardiology centers, minimizes the problem of different referral practices. Spontaneous abortions were not included because of the absence of an accurate autopsy in all of them. The number of induced abortions (n = 67) diagnosed as critical CHD represents a very small proportion of cases limited to the last years of the study. Similarly, the 14 cases of stillbirths, in which there was evidence of a relatively high prevalence of CHD, have been excluded because a reliable ascertainment was difficult. In this study, the average birth prevalence of CHD, as registered during the period of the study, is 5.1%, with a range from 3.1% to 7.5% (Fig 1). The observed average CHD birth prevalence is similar to that in other epidemiologic studies covering a similar period of time.4,6,13,16-18 Nevertheless, Figure 2 shows a significant difference in CHD prevalence from earlier to later periods of the study, demonstrating a higher prevalence in the second decade of the study. In other words, a significant time-related variation in the birth prevalence of CHD has been confirmed.15,19-22 The main reason for this variation is probably related to an increasing recognition of minor cardiac lesions among the simple Temporal Variability in Birth Prevalence of Congenital Heart Defects as Recorded by a General Birth Defects Registry
CHD. This is also confirmed by the significant increase of the miscellaneous CHD, demonstrating a further improvement in the detection of CMs characterized, as define above, by the association of more than one lesion. Analyzing several epidemiologic studies, a similar temporal trend has been observed by Hoffman et al, who suggested that the widespread availability of color Doppler techniques from the beginning of 1990s has made it possible to diagnose minor asymptomatic cardiac lesions even without murmurs in the first days of life.20 Among the simple CHD, we also noticed a small increase in the prevalence of TGA, TF, and AVSD. In contrast, there was no apparent evidence of variation in birth prevalence of the complex CHD (Figs 1 and 2). The increased care given to the detection of any kind of CM is another factor that has improved the detection of CHD, especially in infants affected by extracardiac MCAs.14 In the literature the reported percentage of associated extracardiac anomalies varies from 7% to 15%1,24-27 and from 34% to 45%.28,29 Our data, which concern only live births, are between these 2 groups, with a percentage of 24%. As expected, chromosomal anomalies are more frequent in infants affected by AVSD, with trisomy 21 being the leading anomaly. Among the simple CHD, ASD II and, to a lesser degree, TF, CoAO, AoSt, and LHH, are seldom associated with either known or unknown extracardiac lesions. Similar findings were observed for complex and miscellaneous CHD (Fig 3).5-8 The comparison with the data collected by the IMSCHD, a study oriented to the enrolment of CHD detected up to 2 years of life, was considered useful to clarify the risk of underestimation of CHD prevalence as registered by a general birth defects registry. As previously described, the differences resulting from different nomenclature systems have been avoided using the same classification system in both the studies. 695
Fig 5. A, Comparison between the prevalence of any individual simple CHD as registered by IMER and by IMS-CHD (see text for details). **P < .0001. 95% CL is reported only for VSD.
Fig 5. B, Comparison between the prevalence of any individual complex CHD as registered by IMER and by IMS-CHD (see text for details). **P < .0001.
696 Bosi et al
The Journal of Pediatrics • June 2003
Figure 4 shows that CHD prevalence, registered by the IMS-CHD in the same geographic areas and during the same period of time, varied from 6.2% to 10.7% live births, the average being 8.4%. These values are in agreement with other studies performed with similar methodological criteria.2-6,13,16,17 The main reason for the significant difference with the data collected by IMER Registry is that in the IMS-CHD, all children as old as 2 years of age with suspected CHD were referred to well-trained groups of pediatric cardiologists. This age limit probably explains the higher figures, rising to 11% of CHD in live births (Fig 4). In fact, among the simple CHD, the increase is mainly related to cardiac lesions, such as VSD, PDA, minimal PulSt, or AoSt, attesting to the care given to the minor cardiac lesions often clinically silent in early infancy (Fig 5, A). In the IMS-CHD, the prevalence of VSD rose to 39 per 10,000 live births, the majority being represented by small muscular VSD (67% of all VSDs).8 This observation, suggested by Carlgren et al in the late 1960s, has been recently confirmed by other epidemiologic studies.13 Figure 5A also shows a small increase in the incidence of TF, AVSD, and CoAo among the simple CHD. A similar trend has been recently reported by Botto et al, who stressed an improved CHD ascertainment resulting from changed diagnostic and reporting practices.22 Although the comparison between the 2 studies shows an increase in complex CHD prevalence registered by IMSCHD (Fig 5, B), these figures should be accounted for by the random variation of small numbers of complex CHD enrolled. However, Kuehl et al, in the Baltimore–Washington Infant Study, have conversely demonstrated the failure to diagnose some complex CHD in early infancy: this was related to the short stay in the nursery, therefore confirming the risk of exclusion of infants with critical CHD.23 Differences in the rates are the result of several variable factors, such as the sources of ascertainment, the system of classification, the diagnostic criteria and, especially, the inclusion of the minor cardiac lesions whose detection is strictly related to the age limit up to which the infants with suspected CHD are enrolled. We suggest that a CHD prevalence around 8% to 10% of live births is probably a reasonable estimate. This value might be higher on the basis of the following observations: the possible loss of an unknown number of CHD cases in stillbirths and especially in induced abortions, whose proportion is probably increasing; the risk of exclusion of infants who died before the detection of critical CHD; and the trend of early discharge of infants from nursery, leading to difficulty in CHD ascertainment. Finally, as suggested by Hoffman in 1995, there are minor cardiac lesions, such as the mitral valve prolapse and the bicuspid aortic valve (recognized in 4% to 6% and 2% of the general population, respectively),30,31 which are not usually identified in the majority of epidemiologic studies. This omission is the result of these cardiac lesions usually being diagnosed in adolescence or early adulthood because significant clinical evidence is lacking in the first years of life.20 This appears to be confirmed by the high frequency of aortic valvar stenosis reported by the Bohemian study, which had a longer follow-up, allowing for better detection of this lesion.4 Temporal Variability in Birth Prevalence of Congenital Heart Defects as Recorded by a General Birth Defects Registry
Hoffmann and Kaplan, in a review of the literature, underline that other minor and/or trivial cardiac defects (eg, silent patent ductus arteriosus, isolated anomalous lobar pulmonary veins, and mild pulmonic stenosis) may be difficult to diagnose in early infancy.32 Consequently, we emphasize the possible underestimation of true CHD prevalence in childhood, confirming that cardiac lesions are among the most common of all the CMs.33,34 We are indebted to Dr Amanda Neville for expert editorial support.
REFERENCES 1. Hoffman JIE. Incidence, mortality and natural history. In: Anderson RH, Macartney FJ, Shineborne, et al, editors. Pediatric cardiology. London: Churchil Livingstone; 2002. p. 111-39. 2. Grabitz RG, Joffres MR, Collins-Nakai RL. Congenital heart disease: incidence in the first year of life. Am J Epidemiol 1988;128:381-88. 3. Pradat P. Epidemiology of major congenital heart defects in Sweden 1981-1986. Epidemiol Community Health 1992;46:211-5. 4. Samanek M, Slavik Z, Zborilova B, Hrobonova V, Voriskova M, Skovranek J. Prevalence, treatment, and outcome of heart disease in live-born children: a prospective analysis of 91.823 live-born children. Pediatr Cardiol 1989;10:205-11. 5. Ferencz C, Correa-Villasenor A. Epidemiology of cardiovascular malformations: the state of the art. Cardiol Young 1991;1:264-84. 6. Pexieder T, Bloch D, EUROCAT Working Party on Congenital Heart Disease. EUROCAT Subproject on epidemiology of congenital heart disease: first analysis of the completed study. In: Clark EB, Markwald RR, Takao A, editors. Developmental mechanisms of heart disease. Armonk, NY: Futura Publishing Co; 1995. p. 655-78. 7. Garani GP, Astolfi G, Cocchi G, Magnani C, Milan M, Calzolari E. Congenital hearts defects in Emilia Romagna, Italy: epidemiological study in 267,127 consecutive births. Med Genetik (A) 1995;2:177. 8. Bosi G, Scorrano M, Tosato G, Forini E, Chakrokh and the Working Party of the Italian Society of Pediatric Cardiology: first step of the analysis. Cardiol Young 1999;9:291-9. 9. Milan M, Astolfi G, Volpato S, Garani GP, Clementi M, Tenconi R, et al. 766 cases of oral cleft in Italy. Eur J Epidemiol 1994;10:317-24. 10. Calzolari E, Bianchi F, Dolk H, Stone D, Milan M, and EUROCAT Working Party. Are omphalocele and neural tube defects related congenital anomalies? Data from 21 registries in Europe (EUROCAT). Am J Med Genet 1997;72:79-84. 11. ICD.9.CM International classification of diseases. 9th revision; 5th ed, 1996. MEDICODE Pub, Salt Lake City, UT. 12. Pexieder T, Bloch D, Ferencz C, Eurocat, BWIS. Standard definitions of cardiovascular malformations (CVM) promote accuracy of regional comparisons. Circulation 1991;84:II-385. 13. Carlgren LE, Ericson A, Kallen B. Monitoring of congenital cardiac defects. Pediatr Cardiol 1987;8:247-56. 14. Ferencz C. On the birth prevalence of congenital heart disease. J Am Coll Cardiol. 1990;16:1701-2. 15. Fixler DE, Pastor P, Chamberlin M, Sigman E, Clayton WE. Trends in congenital heart disease in Dallas county births. Circulation 1990;81:137-42. 16. Ferencz C, Rubin JD, Loffredo CA, Magee CA, editors. Epidemiology of congenital heart disease. The Baltimore-Washington Infant Study: Perspectives in Pediatric Cardiology, vol 4. Mount Kisco: Futura Publishing Co; 1993. 17. Kenna AO, Smithells BW, Fielding DW. Congenital heart disease in Liverpool 1960-69. Quart J Med 1978;43:17-44. 18. Stoll C, Alembik A, Roth MP, Dott B, De Geeter B. Risk factors in congenital heart disease. Eur J Epidemiol 1989;5:382-91. 19. Wilson PD, Correa-Villasenor, Loffredo CA, Ferencz C. Temporal trends in prevalence of cardiovascular malformations in Maryland and the District of Columbia, 1981-1988. The Baltimore-Washington Infant Study Group. Epidemiology 1993;4:259-65.
697
20. Hoffman JIE. Incidence of congenital heart disease: I. Postanatal incidence. Pediatr Cardiol 1995;16:103-13. 21. Wren C, Richmond S, Donaldson L. Temporal variability in birth prevalence of cardiovascular malformations. Heart 2000;83:414-9. 22. Botto LD, Correa A, Erickson JD. Racial and temporal variations in the prevalence of heart defects. Pediatrics 2001;107:E32. 23. Kuehl K, Loffredo CA, Ferencz C. Failure to diagnose congenital heart disease in infancy. Pediatrics 1999;103:743-7. 24. Kramer HH, Majewski F, Trampisch HJ, Rammos S, Bourgeois M. Malformation patterns in children with congenital heart disease. Am J Dis Child 1987;141:789-95. 25. Sancez Cascos A. Malformationes congenitos associadas a los cardiopatias congenitos. Rev Clin Espan 1987;180:253-5. 26. Landtman B. Clinical and morphological studies in congenital heart disease: a review of 777 cases. Acta Paediatr Scand 1971;213:1-27. 27. Khoury MJ, Erickson JD. Improved ascertainment of cardiovascular malformations in infants with Down’s syndrome. Implication for the interpretation of increasing rates of cardiovascular malformations in surveillance systems. Am J Epidemiol 1992;136:1457-64.
28. Ferencz C, Boughman JA, Neil CA, Brenner Ji, Perry LW, and The Baltimore-Washington Infant Study Group. Epidemiology of cardiovascular malformations: questions of inheritance. J Am Coll Cardiol 1989;14:756-63. 29. Ferencz C, Loffredo CA, Correa-Villasenor A, Wilson PD, editors. Genetics and environmental risk factors of major cardiovascular malformations. The Baltimore Washington Infant Study, 1981-89. In: Perspectives in pediatric cardiology series, Vol 5. Mount Kisco, NY: Futura Publishing Co; 1997. 30. Roberts WC. The congenitally bicuspid aortic valve: a study of 85 autopsy cases. Am J Cardiol 1970;26:72-83. 31. Warth DC, King ME, Cohen JM, Tesoriero VL, Marcus E, Weyman AE. Prevalence of mitral valve prolapse in normal children. J Am Coll Cardiol 1985;5:1173-7. 32. Hoffmann JIE, Kaplan S. The incidence of congenital heart disease. J Am Coll Cardiol 2002;39:1890-900. 33. Walker DK. Integrating birth defects surveillance in maternal and child health at the state level. Teratology 2000;61:4-8. 34. Kirby RS. Analytical resources for assessment of clinical genetics in public heath: current status and future prospects. Teratology 2000;61:9-16.
50 Years Ago in The Journal of Pediatrics INFECTIOUS HEPATITIS IN CHILDHOOD, WITH A SPECIAL CONSIDERATION OF “PROGRESSIVE HEPATITIS” Murphy ES, Johns RB. J Pediatr 1953;42:707-14 Icteric hepatitis of infectious origin was reported in 40 children 50 years ago. In the past 30 years, the term of infectious hepatitis has been changed to more specific terms, ie, hepatitis A or hepatitis B, because of the progress of virology and diagnostic tools. Before the clinical application of the tests for Australian antigen (hepatitis B surface antigen), and antibody to hepatitis A virus, it was not easy to differentiate acute hepatitis A from B by clinical manifestations. Similar to 50 years ago, there are three established clinical directions of viral hepatitis in children: (1) benign clinical course with recovery in the majority of cases (acute hepatitis); (2) fulminant hepatitis in a small proportion of cases (<1%); and (3) chronic course, either rapid “progessive hepatitis” leading to subacute hepatic necrosis in a small proportion of cases as described by the authors, or an asymptomatic course into either inactive liver disease or liver cirrhosis. Both hepatitis A and B can run an acute or fulminant course, but only hepatitis B may run a chronic course. In the reported three cases with mortality, either fulminant hepatitis or progressive hepatitis was observed. The definition of fulminant hepatitis 50 years ago “with neurologic manifestations and death in ten days or less, with complete destruction of the parenchyma” is very similar to the definition of hyperacute hepatitis proposed later by O’Grady in 1993. By using specific markers of hepatitis B virus, more accurate observation can be achieved for the long-term natural course of hepatitis B in children. In uncommon occasions, symptoms of bridging hepatitis necrosis may develop suddenly, similar to the description of “progressive hepatitis” 50 years ago, during acute exacerbation of chronic hepatitis B in a previous asymptomatic child. The child may have very high levels of aminotransferases, with jaundice, with or without ascites, followed by subacute hepatic failure and liver cirrhosis. However, the majority of children with chronic hepatitis B virus infection remained asymptomatic throughout the course, with initially high levels of hepatitis B virus DNA and minimal liver inflammation, followed by episodes of acute exacerbation of liver inflammation with various degree of liver injury. The latter asymptomatic course was not described 50 years ago because of the lack of the specific markers. Mei-Hwei Chang, MD Department of Pediatrics National Taiwan University Hospital Taipei, Taiwan YMPD256 10.1067/mpd.2003.256
698 Bosi et al
The Journal of Pediatrics • June 2003
Objective Our aim was to examine the temporal variability in congenital heart defect (CHD) birth prevalence from 1980 to 2000 in Emilia-Romagna, Italy. Methods The study population consisted of all infants, surveyed by the Emilia-Romagna birth defects registry (Indagine Malformazioni conpenite in Emilia-Romagna [IMER]), who were affected by CHDs. A simplified classification into “simple” and “complex” CHD was adopted. A comparison with another epidemiologic study using different methodology in the same area was performed.
From 1980 to 2000, IMER ascertained 2442 live births with CHD of 480,793 infants born, with an average CHD birth prevalence of 5.1% (Range, 3.1% to 7.5%). A significant increase in prevalence of simple CHD during the second decade of the study was demonstrated because of an increased recognition of “minor” cardiac lesions among the simple CHD. The birth prevalence of complex CHD remained stable.
Results
Conclusions The apparent increase in live births with CHD results mainly from the current widespread availability of color Doppler echocardiography, which allows the early detection of the “minor” cardiac defects. Other differences are the result of the sources of ascertainment, diagnostic criteria, system of classification, and especially the age limit for enrolling infants with suspected CHD. (J Pediatr 2003;142:690-8)
ongenital Heart Defects (CHDs) are among the most common of all congenital malformations (CMs) and have a major impact on pediatric morbidity, mortality, and healthcare costs. Recent population-based epidemiologic studies have indicated a CHD prevalence ranging from 3.5% to 13.7% of live births.1-6 In Italy several general birth defects registries, especially those in the north and center of the country, have monitored CM since the late 1970s. Data suggest a CHD birth prevalence around 4.5% of live births.7 The Italian Multicenter Study on epidemiology of Congenital Heart Disease (IMS-CHD), performed between 1992 and 1993, suggests an average CHD prevalence varying from 3.2% in the South to 6% in Northern Italy.8 The aims of this study were, first, to verify the birth prevalence of CHD in a well-defined population as registered by a general birth defects registry; second, to examine temporal variability in the prevalence of CHD during a 21-year period of registration (1980-2000); and third, to evaluate the association with extracardiac anomalies. Finally, taking advantage of being part of the IMS-CHD, we have the opportunity to evaluate similarities and differences in these two epidemiologic studies, comparing data collected in the same period of time with different methodological approaches.
C
AoCo AoSt AVSD ASD II CAT CHD CM DIV DORV ICD-9-CM
690
Aortic coarctation Aortic stenosis Atrioventricular septal defect Atrial septal defect type ostium secundum Common arterial trunk Congenital heart defect Congenital malformation Double-inlet ventricle Double-outlet right ventricle 9th Revision of the International Classification Diseases
IMER Indagine Malformazioni conpenite in Emilia-Romagna IMS-CHD Italian Multicenter Study on epidemiology of Congenital Heart Disease LHH Left hypoplastic heart syndrome MCA Multiple congenital anomalies PDA Patent arterial ductus PulAtr Pulmonary atresia PulSt Pulmonary valvar stenosis TF Tetralogy of Fallot TGA Transposition of great arteries VSD Ventricular septal defect
From the Pediatric Cardiology and Neonatology Units, Department of Clinical and Experimental Medicine, and Medical Genetics, Department of Experimental and Diagnostic Medicine, University of Ferrara, Italy. *IMER Working Party: Gianni Astolfi, Medical Genetics, University of Ferrara; Guido Cocchi, Institute of Pediatric and Neonatology, University of Bologna; Cinzia Magnani and Pietro Viola, Institute of Puericulture and Neonatal Medicine, University of Parma; and Daniela Prandstraller, Pediatric Cardiology Unit, University of Bologna. Submitted for publication Sept 20, 2002; revision received Jan 28, 2003; accepted Mar 18, 2003. Reprint requests: Giuliano Bosi, MD, Pediatric Cardiology Unit, Department of Clinical and Experimental Medicine, University of Ferrara, 44100 Ferrara, Italy,Via Savonarola 9. E-mail: bsg@unife .it. Copyright © 2003, Mosby, Inc. All rights reserved. 0022-3476/2003/$30.00 + 0
10.1067/mpd.2003.243
Fig 1. Year-by-year birth prevalence of CHD in live births: 21 years of registration by IMER (1980-2000; see text for details).
METHODS Population and Data Sources The study population consisted of all infants born from 1980 to 2000 and surveyed by IMER, a general birth defects registry covering a well-defined area (the Emilia-Romagna region in Northern Italy). The general characteristics of this registry, which participates in the European Registry of Congenital Anomalies and Twins (EUROCAT) and the Clearinghouse for Birth Defects Monitoring System, have been previously described.9,10 In summary, this registry includes all live births and stillbirths of at least 28 weeks’ gestation, whose mother is a resident in the area, with any CMs detected at birth. Data regarding the infant, pregnancy, parents, and other family members are collected through an interview with the mother during the hospital stay. In the IMS-CHD, whose preliminary results have been recently published, the study population consisted of all infants born from 1992 to 1993 in 15 provinces spread all over the country who were examined by 18 centers of pediatric cardiology. The peripheral units concerned with maternal and pediatric care, along with the family pediatricians, constituted the main sources of information. Data were collected using a questionnaire similar to that used by the EUROCAT study.6 In contrast with the IMER registry, any new case of CHD in a newborn or in a child up to 2 years of age, whose mother was resident in the study area, was been registered prospectively.8 Temporal Variability in Birth Prevalence of Congenital Heart Defects as Recorded by a General Birth Defects Registry
Case Definition and Diagnostic Hierarchy All cases of CHD detected in stillbirths and live births were prospectively enrolled by the IMER registry using the 9th revision of the International Classification Diseases (ICD-9-CM).11 In contrast, the ascertainment of live births with CHD, born in 1992 or 1993 in the same region and prospectively enrolled by the IMS-CHD, was based on a simplified classification suggested by Pexieder and others.6,8,12 The comparison between these two epidemiologic studies, limited to two years of observation (1992 and 1993), has been accomplished by evaluating the data from the provinces covered by both registries. Moreover, all cases of CHD from the IMER registry were retrospectively categorized by the pediatric cardiologists operating in the same units, according to the classification criteria used by the IMS-CHD. Consequently, a common nomenclature system, using a classification into “simple” and “complex” CHD, was adopted to harmonize the collected data. Simple CHD included isolated major entities in the condition of visceral and atrial situs solitus, atrioventricular concordance, and levocardia, such as aortic coarctation (AoCo), aortic stenosis (AoSt), atrial septal defect type ostium secundum (ASD II), atrioventricular septal defect (AVSD), left hypoplastic Heart syndrome (LHH), patent arterial ductus (PDA), pulmonary valvar stenosis (PulSt), tetralogy of Fallot (TF), transposition of great arteries (TGA), and ventricular Septal defect (VSD). 691
Fig 2. Comparison of the prevalence of any individual CHD between the first period of the study (total CHDs, n = 972) and the second period (total CHDs, n = 1470). *P < .001; **P < .0001. 99% CL is reported.
Fig 3. Association of any individual CHD with extracardiac anomalies (including either known or unknown conditions) in relationship with isolated CHD (see text for details).
692 Bosi et al
The Journal of Pediatrics • June 2003
Table I. Birth prevalence of any individual CHD in live births (10,000 + 99% CL), as registered by the IMER from 1980 to 2000, using a classification into “simple” and “complex” CHD
CHD
Live births
Simple CHD 2157 VSD 947 ASD II 162 PDA 55 PulSt 104 TF 106 AoSt 35 AoCo 99 LHH 78 AVSD 134 TGA 125 Miscellaneous 211 Not defined 101 Complex CHD 286 DIV 60 CAT 34 DORV 21 PulAtr (with or 59 without VSD) TricAtr 35 Heterotaxia 29 Total anomalous 16 pulmonary venous return Ebstein’s 10 AoArcInt 7
% CHD
Prevalence 10,000 (± 99% CL)
(522) (123) (70) (29) (16) (27) (7) (21) (13) (100) (8) (56) (52) (66) (16) (11) (17) (8)
88.3 38.8 6.6 2.3 4.3 4.3 1.4 4.0 3.2 5.5 5.1 8.6 4.1 11.7 2.5 1.4 0.8 2.4
44.86 ± 2.9 1970 ± 1.90 3.37 ± 0.80 1.14 ± 0.46 2.16 ± 0.64 2.20 ± 0.64 0.73 ± 0.37 2.05 ± 0.64 1.60 ± 0.55 2.79 ± 0.72 2.60 ± 0.70 4.39 ± 0.90 2.10 ± 0.63 6.00 ± 1.10 1.25 ± 0.50 0.71 ± 0.36 0.44 ± 0.30 1.23 ± 0.50
(4) (16) (3)
1.4 1.2 0.6
0.73 ± 0.36 0.60 ± 0.30 0.33 ± 0.25
(0) (1)
0.4 0.3
0.21 ± 0.19 0.15 ± 0.15
The number of live births with associated extracardiac anomalies is in brackets (see text).
Complex CHD included cardiovascular lesions that did not fit into defined categories and required a more detailed description of each lesion, such as common arterial trunk (CAT); double-inlet ventricle (DIV); pulmonary atresia (PulAtr) with or without VSD; double-outlet right ventricle (DORV); Ebstein’s anomaly; hetorotaxia (including dextrocardia and left and right atrial isomerism, usually associated with major cardiac lesions); total anomalous pulmonary venous return; and tricuspid valve atresia (TrAtr). Moreover, we defined as miscellaneous CHD as cardiovascular malformations characterized by an association of more than one defect, the main being usually represented by VSD and/or ASD II, in association with other minor cardiac defects. Premature infants (<37 weeks’ gestation) with PDA were excluded. CHD was considered “isolated” when no major extracardiac malformations were present and “associated” when at least one major additional extracardiac malformation was Temporal Variability in Birth Prevalence of Congenital Heart Defects as Recorded by a General Birth Defects Registry
Fig 4. Comparison between the CHD prevalence as registered by IMER and by IMS-CHD during 2 years of registration in the same geographic areas (see text for details). **P < .0001. 99% CL is calculated.
found. CHD diagnosed within syndromes, sequences, or associations of known or presumed separate etiology were considered among recognized conditions. All remaining cases associated with major extracardiac malformations were considered as multiple congenital anomalies (MCAs). We included only live births in the present study because the proportion of stillbirths was very small. For the same reason, fetal and/or diagnosis at autopsy of CHD in spontaneous and/or induced abortions were excluded. The diagnostic techniques included clinical evaluation, echocardiography, diagnostic and/or interventional catheterization, surgery, and necropsy.
Statistical Analysis The χ2 test with Yates correction was used to assess the statistical differences between (1) the CHD prevalence calculated in the first and second period of the study and (2) the data coming from the 2 epidemiologic studies we compared. Statgraphics V4.0 (STSC, Inc, Rockville, Maryland) was used for data processing.
RESULTS Consecutive live births (n = 480,793) and stillbirths were surveyed by the IMER registry from 1980 to 2000 with a mean of 22,894 per year. In the 21 years of the study, 2456 CHD cases were registered; among these, 2442 cases were detected in live births with a birth prevalence ranging from 3.1% to 7.5%, the average being 5.1% (99% CL= 0.3) (Fig 1). Table I shows the prevalence of any individual cardiovascular malformation using the classification into simple and complex CHD. In the first group, isolated VSD occurred with an average prevalence of 19.7 per 10,000 live births. Isolated ASD II had a prevalence of 3.4 per 10,000 live births. AVSD and TGA formed a cluster with a prevalence around 2.7 per 10,000 live births. A second cluster, including TF, LHH, PulSt, and isolated CoAo, had a prevalence from 1.6 to 2.2 per 10,000 live births. PDA in mature infants had a prevalence of 1.2 and AoSt 0.7 per 10,000 live births. 99% confidence limit for any CHD is reported in Table I. 693
Table II.Year-by-year birth prevalence of any individual “simple” CHD in live births as registered by IMER from 1980 to 2000 (see text) Simple CHD
1980
VSD 10.42 PDA 0.00 ASD II 1.60 PulSt 1.60 TF 0.00 TGA 1.60 AoSt 0.00 AoCo 0.00 AVSD 1.60 LHH 0.80 Miscellaneous 0.80 Not defined 8.82 No. of 12.471 live-born
1981
1982
1983
1984
1985
1986
1987
1988
1989
9.08 0.70 1.40 0.00 0.70 2.79 0.70 0.70 2.79 2.09 2.79 4.19 14.322
13.07 0.00 0.00 1.63 0.00 2.18 0.00 3.27 2.18 0.54 1.09 2.72 18.362
14.70 0.46 0.92 2.30 1.38 4.59 0.46 3.22 0.92 1.38 2.30 3.68 21.767
17.27 0.43 0.43 1.30 2.59 2.16 0.86 2.16 2.59 2.16 1.73 2.59 23.155
13.74 0.83 1.25 1.67 3.33 1.25 0.42 2.08 2.91 2.50 2.50 2.08 24.016
13.12 2.19 1.75 4.37 3.94 3.50 1.31 2.62 1.75 1.75 2.19 2.19 22.868
19.79 1.76 1.32 2.64 1.32 1.32 0.00 1.76 4.40 1.32 2.20 0.88 22.741
20.62 0.88 1.75 1.75 2.63 1.75 1.32 1.32 3.07 1.32 2.19 1.32 22.794
19.67 0.43 1.28 4.28 2.57 2.14 1.71 1.71 2.57 1.71 2.99 0.86 23.386
Miscellaneous CHDs, as previously defined, represent a significant proportion of all cardiovascular malformations with a prevalence approximately 4.4 per 10,000 live births. Among the complex CHDs, which represent 11.7% of all the cardiovascular malformations, DIV, and PulAtr (with and without VSD) had a prevalence about 1.2 per 10,000 live births each. DORV, CAT, and TrAtr had a prevalence from 0.5 to 1.2 per 10,000 live births. TAPVC and Ebstein’s malformation varied from 0.2 to 0.3 per 10,000 live births. In the same table, we reported the number of CHD cases associated with extracardiac anomalies, including chromosomal anomalies, recognized conditions, and/or multiple congenital anomalies. Figure 1 shows the year-by-year prevalence of simple and complex CHD together with the prevalence of the total number of CHDs. There was a significant increasing trend in the number of simple CHD over time (P < .001), whereas no change was evident in the prevalence of complex CHD. Their apparent increase in the last years of the study is almost certainly accounted for by the random variation of small numbers. Table II, which shows the year-by-year variability in the prevalence of any individual simple CHD, demonstrates that the greatest variation was caused by the increased prevalence of isolated VSD, ASD II, and miscellaneous CHD, with the prevalence of any other CHD not being significantly changed. This is clearly confirmed by Figure 2, which demonstrates a significant increase in the diagnosis of VSD (P < .001) and ASD II (P < .0001) especially in the second decade of the study, together with a significant increase in the proportion of the so-called miscellaneous CHD (P < .0001). Of the infants enrolled, 24% had extracardiac anomalies. Chromosomal anomalies were diagnosed in 9.1% of all live births. Among these, 75.7% were trisomy 21; 6.3% Edward syndrome; 4.1% Turner syndrome, 3.1% Patau syn694 Bosi et al
drome; and ,10.8% other syndromes including DiGeorge syndrome. CHDs were also present within the framework of various syndromes and/or sequences in 4.4% of these cases. MCAs were present in about 10% of the infants: among these, 55% had 1 associated extracardiac lesion; 23% had 2, and the remaining (22%) more than 2 extracardiac malformations. Systems of organs were affected with the following frequency: urogenital system (25.8%); musculoskeletal system (22.3%); craniofacial anomalies, including eye and ear anomalies (19.3%); gastrointestinal anomalies (16.9%); central nervous system (10.6%); pulmonary apparatus (2.6%); and integument anomalies (1.7%). Figure 3 shows the association of any individual cardiovascular malformation with extracardiac anomalies, including either known (syndromes, sequences, and chromosomal anomalies) or unknown conditions (MCAs). Figure 4 shows the comparison between the CHD prevalence as collected by IMER and IMS-CHD Registries during 2 years of registration (1992-1993) in the same geographic area represented by three provinces of the same Region (Fig 4, A-C). The average CHD prevalence registered by IMS-CHD was 8.3%, ranging from 6.2 to 10.7% live births. This value is significantly different from that reported by the IMER registry both for simple and complex CHD, as clearly documented in Figure 5, A and B.
DISCUSSION In Western industrialized countries, recent populationbased epidemiologic studies have indicated CHD prevalence ranging from 3.5% to 13.7% live births.1-6 These significant differences are closely related to numerous methodologic difficulties arising in the collection of epidemiologic data.13 In 1990, several factors biasing the ascertainment of CHD prevalence were discussed by Ferencz, who underlined the sigThe Journal of Pediatrics • June 2003
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
21.24 0.85 2.97 4.25 0.85 2.97 0.85 1.27 2.55 2.12 4.25 1.70 23.545
25.07 0.40 3.98 1.99 0.80 3.98 0.40 1.59 3.18 2.79 1.99 1.19 25.129
23.56 1.20 3.59 4.39 1.60 0.80 2.00 2.00 2.40 2.00 3.19 3.19 25.042
22.83 0.00 6.93 2.85 4.08 2.45 0.00 1.22 1.63 1.22 6.11 3.26 24.533
22.41 1.93 6.57 3.09 1.93 1.93 0.39 1.55 3.09 1.55 8.11 3.48 25.886
25.80 2.28 6.07 0.76 2.66 2.66 1.52 2.66 1.52 1.52 6.07 0.76 26.359
26.67 1.04 7.27 0.69 2.77 3.12 0.00 3.46 4.85 1.73 5.54 2.42 28.873
16.29 0.00 4.50 2.08 2.43 1.73 0.69 2.08 3.12 2.43 8.66 0.00 28.859
21.02 1.91 7.17 0.96 4.78 3.82 0.48 0.48 5.73 3.34 8.12 2.87 20.928
26.25 3.75 2.92 1.25 1.25 3.75 0.42 3.33 4.58 2.50 7.08 2.08 24.003
24.36 2.76 2.76 0.46 6.90 4.60 1.38 4.60 4.14 3.22 8.73 1.38 21.754
nificant role represented by population factors, family factors, community medical practice, and cardiology center practice.14 At the same time, the risk of exclusion of CHD in abortions, in stillborns, and in infants dead before detection of “critical” CHD was stressed by Fixler et al together with the risk of exclusion of minor CHDs that are clinically silent in early infancy.15 Consequently, the age limit up to which infants with CHD are registered, together with the selection of a well-defined population and with the use of a common nomenclature system, represents the more significant variable factors in collecting reliable epidemiologic data.5,6,12,14,15 The current study concerns a well-defined population surveyed by the IMER registry since the late 1970s. Moreover, in this region the distribution of maternity units, in proximity to the pediatric cardiology centers, minimizes the problem of different referral practices. Spontaneous abortions were not included because of the absence of an accurate autopsy in all of them. The number of induced abortions (n = 67) diagnosed as critical CHD represents a very small proportion of cases limited to the last years of the study. Similarly, the 14 cases of stillbirths, in which there was evidence of a relatively high prevalence of CHD, have been excluded because a reliable ascertainment was difficult. In this study, the average birth prevalence of CHD, as registered during the period of the study, is 5.1%, with a range from 3.1% to 7.5% (Fig 1). The observed average CHD birth prevalence is similar to that in other epidemiologic studies covering a similar period of time.4,6,13,16-18 Nevertheless, Figure 2 shows a significant difference in CHD prevalence from earlier to later periods of the study, demonstrating a higher prevalence in the second decade of the study. In other words, a significant time-related variation in the birth prevalence of CHD has been confirmed.15,19-22 The main reason for this variation is probably related to an increasing recognition of minor cardiac lesions among the simple Temporal Variability in Birth Prevalence of Congenital Heart Defects as Recorded by a General Birth Defects Registry
CHD. This is also confirmed by the significant increase of the miscellaneous CHD, demonstrating a further improvement in the detection of CMs characterized, as define above, by the association of more than one lesion. Analyzing several epidemiologic studies, a similar temporal trend has been observed by Hoffman et al, who suggested that the widespread availability of color Doppler techniques from the beginning of 1990s has made it possible to diagnose minor asymptomatic cardiac lesions even without murmurs in the first days of life.20 Among the simple CHD, we also noticed a small increase in the prevalence of TGA, TF, and AVSD. In contrast, there was no apparent evidence of variation in birth prevalence of the complex CHD (Figs 1 and 2). The increased care given to the detection of any kind of CM is another factor that has improved the detection of CHD, especially in infants affected by extracardiac MCAs.14 In the literature the reported percentage of associated extracardiac anomalies varies from 7% to 15%1,24-27 and from 34% to 45%.28,29 Our data, which concern only live births, are between these 2 groups, with a percentage of 24%. As expected, chromosomal anomalies are more frequent in infants affected by AVSD, with trisomy 21 being the leading anomaly. Among the simple CHD, ASD II and, to a lesser degree, TF, CoAO, AoSt, and LHH, are seldom associated with either known or unknown extracardiac lesions. Similar findings were observed for complex and miscellaneous CHD (Fig 3).5-8 The comparison with the data collected by the IMSCHD, a study oriented to the enrolment of CHD detected up to 2 years of life, was considered useful to clarify the risk of underestimation of CHD prevalence as registered by a general birth defects registry. As previously described, the differences resulting from different nomenclature systems have been avoided using the same classification system in both the studies. 695
Fig 5. A, Comparison between the prevalence of any individual simple CHD as registered by IMER and by IMS-CHD (see text for details). **P < .0001. 95% CL is reported only for VSD.
Fig 5. B, Comparison between the prevalence of any individual complex CHD as registered by IMER and by IMS-CHD (see text for details). **P < .0001.
696 Bosi et al
The Journal of Pediatrics • June 2003
Figure 4 shows that CHD prevalence, registered by the IMS-CHD in the same geographic areas and during the same period of time, varied from 6.2% to 10.7% live births, the average being 8.4%. These values are in agreement with other studies performed with similar methodological criteria.2-6,13,16,17 The main reason for the significant difference with the data collected by IMER Registry is that in the IMS-CHD, all children as old as 2 years of age with suspected CHD were referred to well-trained groups of pediatric cardiologists. This age limit probably explains the higher figures, rising to 11% of CHD in live births (Fig 4). In fact, among the simple CHD, the increase is mainly related to cardiac lesions, such as VSD, PDA, minimal PulSt, or AoSt, attesting to the care given to the minor cardiac lesions often clinically silent in early infancy (Fig 5, A). In the IMS-CHD, the prevalence of VSD rose to 39 per 10,000 live births, the majority being represented by small muscular VSD (67% of all VSDs).8 This observation, suggested by Carlgren et al in the late 1960s, has been recently confirmed by other epidemiologic studies.13 Figure 5A also shows a small increase in the incidence of TF, AVSD, and CoAo among the simple CHD. A similar trend has been recently reported by Botto et al, who stressed an improved CHD ascertainment resulting from changed diagnostic and reporting practices.22 Although the comparison between the 2 studies shows an increase in complex CHD prevalence registered by IMSCHD (Fig 5, B), these figures should be accounted for by the random variation of small numbers of complex CHD enrolled. However, Kuehl et al, in the Baltimore–Washington Infant Study, have conversely demonstrated the failure to diagnose some complex CHD in early infancy: this was related to the short stay in the nursery, therefore confirming the risk of exclusion of infants with critical CHD.23 Differences in the rates are the result of several variable factors, such as the sources of ascertainment, the system of classification, the diagnostic criteria and, especially, the inclusion of the minor cardiac lesions whose detection is strictly related to the age limit up to which the infants with suspected CHD are enrolled. We suggest that a CHD prevalence around 8% to 10% of live births is probably a reasonable estimate. This value might be higher on the basis of the following observations: the possible loss of an unknown number of CHD cases in stillbirths and especially in induced abortions, whose proportion is probably increasing; the risk of exclusion of infants who died before the detection of critical CHD; and the trend of early discharge of infants from nursery, leading to difficulty in CHD ascertainment. Finally, as suggested by Hoffman in 1995, there are minor cardiac lesions, such as the mitral valve prolapse and the bicuspid aortic valve (recognized in 4% to 6% and 2% of the general population, respectively),30,31 which are not usually identified in the majority of epidemiologic studies. This omission is the result of these cardiac lesions usually being diagnosed in adolescence or early adulthood because significant clinical evidence is lacking in the first years of life.20 This appears to be confirmed by the high frequency of aortic valvar stenosis reported by the Bohemian study, which had a longer follow-up, allowing for better detection of this lesion.4 Temporal Variability in Birth Prevalence of Congenital Heart Defects as Recorded by a General Birth Defects Registry
Hoffmann and Kaplan, in a review of the literature, underline that other minor and/or trivial cardiac defects (eg, silent patent ductus arteriosus, isolated anomalous lobar pulmonary veins, and mild pulmonic stenosis) may be difficult to diagnose in early infancy.32 Consequently, we emphasize the possible underestimation of true CHD prevalence in childhood, confirming that cardiac lesions are among the most common of all the CMs.33,34 We are indebted to Dr Amanda Neville for expert editorial support.
REFERENCES 1. Hoffman JIE. Incidence, mortality and natural history. In: Anderson RH, Macartney FJ, Shineborne, et al, editors. Pediatric cardiology. London: Churchil Livingstone; 2002. p. 111-39. 2. Grabitz RG, Joffres MR, Collins-Nakai RL. Congenital heart disease: incidence in the first year of life. Am J Epidemiol 1988;128:381-88. 3. Pradat P. Epidemiology of major congenital heart defects in Sweden 1981-1986. Epidemiol Community Health 1992;46:211-5. 4. Samanek M, Slavik Z, Zborilova B, Hrobonova V, Voriskova M, Skovranek J. Prevalence, treatment, and outcome of heart disease in live-born children: a prospective analysis of 91.823 live-born children. Pediatr Cardiol 1989;10:205-11. 5. Ferencz C, Correa-Villasenor A. Epidemiology of cardiovascular malformations: the state of the art. Cardiol Young 1991;1:264-84. 6. Pexieder T, Bloch D, EUROCAT Working Party on Congenital Heart Disease. EUROCAT Subproject on epidemiology of congenital heart disease: first analysis of the completed study. In: Clark EB, Markwald RR, Takao A, editors. Developmental mechanisms of heart disease. Armonk, NY: Futura Publishing Co; 1995. p. 655-78. 7. Garani GP, Astolfi G, Cocchi G, Magnani C, Milan M, Calzolari E. Congenital hearts defects in Emilia Romagna, Italy: epidemiological study in 267,127 consecutive births. Med Genetik (A) 1995;2:177. 8. Bosi G, Scorrano M, Tosato G, Forini E, Chakrokh and the Working Party of the Italian Society of Pediatric Cardiology: first step of the analysis. Cardiol Young 1999;9:291-9. 9. Milan M, Astolfi G, Volpato S, Garani GP, Clementi M, Tenconi R, et al. 766 cases of oral cleft in Italy. Eur J Epidemiol 1994;10:317-24. 10. Calzolari E, Bianchi F, Dolk H, Stone D, Milan M, and EUROCAT Working Party. Are omphalocele and neural tube defects related congenital anomalies? Data from 21 registries in Europe (EUROCAT). Am J Med Genet 1997;72:79-84. 11. ICD.9.CM International classification of diseases. 9th revision; 5th ed, 1996. MEDICODE Pub, Salt Lake City, UT. 12. Pexieder T, Bloch D, Ferencz C, Eurocat, BWIS. Standard definitions of cardiovascular malformations (CVM) promote accuracy of regional comparisons. Circulation 1991;84:II-385. 13. Carlgren LE, Ericson A, Kallen B. Monitoring of congenital cardiac defects. Pediatr Cardiol 1987;8:247-56. 14. Ferencz C. On the birth prevalence of congenital heart disease. J Am Coll Cardiol. 1990;16:1701-2. 15. Fixler DE, Pastor P, Chamberlin M, Sigman E, Clayton WE. Trends in congenital heart disease in Dallas county births. Circulation 1990;81:137-42. 16. Ferencz C, Rubin JD, Loffredo CA, Magee CA, editors. Epidemiology of congenital heart disease. The Baltimore-Washington Infant Study: Perspectives in Pediatric Cardiology, vol 4. Mount Kisco: Futura Publishing Co; 1993. 17. Kenna AO, Smithells BW, Fielding DW. Congenital heart disease in Liverpool 1960-69. Quart J Med 1978;43:17-44. 18. Stoll C, Alembik A, Roth MP, Dott B, De Geeter B. Risk factors in congenital heart disease. Eur J Epidemiol 1989;5:382-91. 19. Wilson PD, Correa-Villasenor, Loffredo CA, Ferencz C. Temporal trends in prevalence of cardiovascular malformations in Maryland and the District of Columbia, 1981-1988. The Baltimore-Washington Infant Study Group. Epidemiology 1993;4:259-65.
697
20. Hoffman JIE. Incidence of congenital heart disease: I. Postanatal incidence. Pediatr Cardiol 1995;16:103-13. 21. Wren C, Richmond S, Donaldson L. Temporal variability in birth prevalence of cardiovascular malformations. Heart 2000;83:414-9. 22. Botto LD, Correa A, Erickson JD. Racial and temporal variations in the prevalence of heart defects. Pediatrics 2001;107:E32. 23. Kuehl K, Loffredo CA, Ferencz C. Failure to diagnose congenital heart disease in infancy. Pediatrics 1999;103:743-7. 24. Kramer HH, Majewski F, Trampisch HJ, Rammos S, Bourgeois M. Malformation patterns in children with congenital heart disease. Am J Dis Child 1987;141:789-95. 25. Sancez Cascos A. Malformationes congenitos associadas a los cardiopatias congenitos. Rev Clin Espan 1987;180:253-5. 26. Landtman B. Clinical and morphological studies in congenital heart disease: a review of 777 cases. Acta Paediatr Scand 1971;213:1-27. 27. Khoury MJ, Erickson JD. Improved ascertainment of cardiovascular malformations in infants with Down’s syndrome. Implication for the interpretation of increasing rates of cardiovascular malformations in surveillance systems. Am J Epidemiol 1992;136:1457-64.
28. Ferencz C, Boughman JA, Neil CA, Brenner Ji, Perry LW, and The Baltimore-Washington Infant Study Group. Epidemiology of cardiovascular malformations: questions of inheritance. J Am Coll Cardiol 1989;14:756-63. 29. Ferencz C, Loffredo CA, Correa-Villasenor A, Wilson PD, editors. Genetics and environmental risk factors of major cardiovascular malformations. The Baltimore Washington Infant Study, 1981-89. In: Perspectives in pediatric cardiology series, Vol 5. Mount Kisco, NY: Futura Publishing Co; 1997. 30. Roberts WC. The congenitally bicuspid aortic valve: a study of 85 autopsy cases. Am J Cardiol 1970;26:72-83. 31. Warth DC, King ME, Cohen JM, Tesoriero VL, Marcus E, Weyman AE. Prevalence of mitral valve prolapse in normal children. J Am Coll Cardiol 1985;5:1173-7. 32. Hoffmann JIE, Kaplan S. The incidence of congenital heart disease. J Am Coll Cardiol 2002;39:1890-900. 33. Walker DK. Integrating birth defects surveillance in maternal and child health at the state level. Teratology 2000;61:4-8. 34. Kirby RS. Analytical resources for assessment of clinical genetics in public heath: current status and future prospects. Teratology 2000;61:9-16.
50 Years Ago in The Journal of Pediatrics INFECTIOUS HEPATITIS IN CHILDHOOD, WITH A SPECIAL CONSIDERATION OF “PROGRESSIVE HEPATITIS” Murphy ES, Johns RB. J Pediatr 1953;42:707-14 Icteric hepatitis of infectious origin was reported in 40 children 50 years ago. In the past 30 years, the term of infectious hepatitis has been changed to more specific terms, ie, hepatitis A or hepatitis B, because of the progress of virology and diagnostic tools. Before the clinical application of the tests for Australian antigen (hepatitis B surface antigen), and antibody to hepatitis A virus, it was not easy to differentiate acute hepatitis A from B by clinical manifestations. Similar to 50 years ago, there are three established clinical directions of viral hepatitis in children: (1) benign clinical course with recovery in the majority of cases (acute hepatitis); (2) fulminant hepatitis in a small proportion of cases (<1%); and (3) chronic course, either rapid “progessive hepatitis” leading to subacute hepatic necrosis in a small proportion of cases as described by the authors, or an asymptomatic course into either inactive liver disease or liver cirrhosis. Both hepatitis A and B can run an acute or fulminant course, but only hepatitis B may run a chronic course. In the reported three cases with mortality, either fulminant hepatitis or progressive hepatitis was observed. The definition of fulminant hepatitis 50 years ago “with neurologic manifestations and death in ten days or less, with complete destruction of the parenchyma” is very similar to the definition of hyperacute hepatitis proposed later by O’Grady in 1993. By using specific markers of hepatitis B virus, more accurate observation can be achieved for the long-term natural course of hepatitis B in children. In uncommon occasions, symptoms of bridging hepatitis necrosis may develop suddenly, similar to the description of “progressive hepatitis” 50 years ago, during acute exacerbation of chronic hepatitis B in a previous asymptomatic child. The child may have very high levels of aminotransferases, with jaundice, with or without ascites, followed by subacute hepatic failure and liver cirrhosis. However, the majority of children with chronic hepatitis B virus infection remained asymptomatic throughout the course, with initially high levels of hepatitis B virus DNA and minimal liver inflammation, followed by episodes of acute exacerbation of liver inflammation with various degree of liver injury. The latter asymptomatic course was not described 50 years ago because of the lack of the specific markers. Mei-Hwei Chang, MD Department of Pediatrics National Taiwan University Hospital Taipei, Taiwan YMPD256 10.1067/mpd.2003.256
698 Bosi et al
The Journal of Pediatrics • June 2003