Progress in Pediatric Cardiology 23 (2007) 17 – 24 www.elsevier.com/locate/ppedcard
Pediatric cardiomyopathy: The Australian experience ☆ Robert G. Weintraub a,⁎,1 , Alan W. Nugent b,2 , Piers E.F. Daubeney c,3 a
Royal Children's Hospital, Melbourne, Australia Texas Children's Hospital, Houston, United States Royal Brompton Hospital, London, United Kingdom
b c
Available online 27 June 2007
Abstract Population-based studies have provided insight into the natural history of adult cardiomyopathy (CM), but comparable information for affected children is lacking. All Australian children diagnosed with primary CM at age 0–10 years between 1.1.1987 and 12.31.1996 were enrolled in a longitudinal cohort study. A single cardiologist and pathologist reviewed all cardiac investigations and histopathological material, respectively. During the study period, the overall incidence of CM was 1.24/105 person–years at risk. Indigenous children had a higher incidence of DCM (relative risk 2.8) and a higher rate of presenting death (16.7% vs. 2.6%). The 5-year freedom from death or transplantation was 63% for children with DCM (n = 184), 83% for those with HCM (n = 80), 53% for those with left ventricular non-compaction (LVNC) (n = 29) and 38% for those with RCM (n = 8). In subjects with DCM, positive viral identification and/or lymphocytic myocarditis were found in 30/44 (68.2%) cases with available early histology and 8/9 cases presenting with sudden death. Risk factors for death/transplantation comprised presenting age N 5 years, familial DCM, lower initial fractional shortening Z score and failure to increase fractional shortening Z score during follow-up. In subjects with HCM, an underlying syndromal, genetic or metabolic condition was identified in 46 (57.5%) of subjects. Risk factors for death/transplantation included concentric LV hypertrophy, presenting age b 1 year, lower initial fractional shortening Z score and increasing LV posterior wall Z score. Lymphocytic myocarditis and LVNC are important causes of childhood CM. This population-based study identifies pediatric subjects at risk of adverse events. © 2007 Elsevier Ireland Ltd. All rights reserved. Keywords: Cardiomyopathy; Myocarditis; Left ventricular non-compaction; Sudden death; Transplantation
1. Introduction Pediatric cardiomyopathies are an uncommon and heterogeneous group of disorders with worse than expected outcomes for children with congenital heart disease. They account for approximately 50% of cardiac transplants performed worldwide ☆
Supported by grants from the Murdoch Children's Research Institute (Grant 98001), the National Heart Foundation of Australia (Grants G98M0159 and G05M2151), and the Australia and New Zealand Children's Heart Research Centre. ⁎ Corresponding author. Department of Cardiology, Royal Children's Hospital, Flemington Road, Parkville, VIC 3052, Australia. Tel.: +61 3 93455718; fax: +61 3 93456001. E-mail address:
[email protected] (R.G. Weintraub). 1 Robert Gideon Weintraub is an Associate Professor in Pediatrics, University of Melbourne. 2 Alan William Nugent is an Assistant Professor Pediatrics, Baylor College of Medicine. 3 Piers Edward Francis Daubeney is a Consultant Paediatric and Fetal Cardiologist. 1058-9813/$ - see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ppedcard.2007.05.006
in children beyond the first year of life [1], they are the commonest inherited form of heart disease [2] and the commonest cause of sudden death in healthy young adults [3]. Multiple medical and surgical therapeutic options exist for children with cardiomyopathy, and successful management strategies depend on identification of individuals who are at risk of sudden death or progressive heart failure. The long-term outcomes and prognostic factors for pediatric cardiomyopathies in the present era remain unclear. Until recently, most available information about pediatric cardiomyopathies has been derived from institutional reviews and relates to short-term outcomes, using patient variables that have been collected at the time of presentation. Outcomes for patients seen in large pediatric cardiac and transplant centres may reflect referral bias. Furthermore institutional reviews may not detect patients for whom sudden death is the first symptom, and those who die shortly after presentation. In adults, population-based studies of hypertrophic cardiomyopathy have corrected misconceptions about disease severity
18
R.G. Weintraub et al. / Progress in Pediatric Cardiology 23 (2007) 17–24
and outcomes [4]. The high rate of urbanization of the Australian population together with the centralization of pediatric cardiac services provides a unique opportunity to undertake population-based studies of uncommon conditions. In the 1990's a national cohort study was established to determine the presenting features and outcomes for children with primary cardiomyopathy. 2. Methods The National Australian Childhood Cardiomyopathy Study is a population-based cohort study of all children within Australia who were diagnosed with cardiomyopathy between January 1st 1987 and December 31st 2006. Cases were recruited during a series of site visits by the same 3 investigators to all 9 pediatric cardiac centers and an additional 12 hospitals where subjects with pediatric cardiomyopathy were identified. At each site, potential study subjects were identified from all available sources including medical records and cardiology databases, using predetermined diagnostic codes, as well as from daily echocardiography logbooks. Additional cases were sought from regional pediatricians, cardiologists caring for adult patients and cardiac transplant centers. The broad objective was to capture children whose outcomes were likely to be determined solely on the basis of their cardiac status. The study exclusion criteria therefore included cardiac dysfunction secondary to congenital heart disease, cardiac arrhythmia, Kawasaki disease, prior anthracycline or corticosteroid exposure, progressive neuromuscular disorders or inborn errors of metabolism with multiple organ dysfunction, and prematurity or maternal diabetes for children presenting during the first 4 weeks of life. A decision about study inclusion was made after review of all the inpatient and outpatient medical records, together with the results of cardiac imaging and investigations relevant to the etiology of the cardiomyopathy. Prospective follow-up was arranged for patients who had not been seen in the past 12 months. Specifically designed data forms were used to record uniform clinical and epidemiological variables, at the time of patient enrolment and during prospective follow-up. Australian law requires an autopsy to be performed in cases of sudden or unexpected death. Centralized records compiled by the Australian Bureau of Statistics using the same diagnostic codes as for hospital medical records, were used to identify potential study subjects. The autopsy records were scrutinized and available cardiac specimens and histology were reviewed centrally. In this way, information was acquired about subjects who had not had cardiomyopathy diagnosed during life or had never been in contact with a physician. Familial cardiomyopathy was considered to be present when a first or second degree relative was identified with the same diagnosis. Cardiomyopathies were classified according to the current WHO classification [5] by a single pediatric cardiologist after direct visualization and interpretation of all available cardiac imaging, together with the results of relevant investigations. The diagnostic criteria for dilated cardiomyopathy included the following: (1) reduced left ventricular function on any form of imaging in subjects with symptoms (usually congestive heart
failure) or a family history of dilated cardiomyopathy, (2) a measured left ventricular fractional shortening of b20% or ejection fraction b 45% in subjects without symptoms of a positive family history, (3) pathological evidence of dilated cardiomyopathy or lymphocytic myocarditis at autopsy. The diagnostic criteria for hypertrophic cardiomyopathy included otherwise unexplained septal or left ventricular free wall hypertrophy (Z score N 2 for either). The right ventricle was considered to be involved if the right ventricular free wall thickness exceeded 4 mm in the absence of pulmonary valve stenosis. The morphology of hypertrophic cardiomyopathy was classified as predominantly either asymmetric septal hypertrophy or concentric left ventricular hypertrophy, depending on the extent of left ventricular free wall involvement. Restrictive cardiomyopathy required impaired diastolic filling with normal left ventricular wall thickness and systolic function. The remaining cases comprised unclassified cardiomyopathy. These included subjects with left ventricular non-compaction. The latter diagnosis required a non-compacted to compacted ratio of N 100% for each affected segment, from any form of cardiac imaging. A single pediatric pathologist, who was unaware of the subject's clinical details, examined all available pathological specimens including cardiac histology. Definite lymphocytic myocarditis was classified according to the Dallas criteria [6]. Serial echocardiographic measurements of left ventricular size, fractional shortening (in those without regional wall motion abnormalities), free wall and septal thickness were expressed as Z scores based on body surface area or age (in the case of fractional shortening) [7,8]. Among subjects with more than one echocardiogram available from the time of presentation, the change in Z score for any given parameter was calculated by subtracting the initial Z score from the subsequent value. The earliest available ECG within 7 days of presentation was read by a single observer, and measurements of QRS duration, QRS amplitude and the corrected QT interval were converted to age-appropriate Z scores [9]. 3. Statistical analysis Incidence rates were calculated using the age-specific population at risk between 1987 and 1997. Ninety-five percent confidence intervals for incidence rates were calculated using the Poisson distribution. Data were summarized using frequencies and percentages for categorical data and median for age at presentation. Pearson's chi-squared statistic or Fisher's exact test (when expected cell values were b 5) were used to assess the association between two categorical variables. The Kruskal– Wallis rank test (or Mann–Whitney test for two groups) was used to compare presenting ages between subgroups. Poisson regression was used to assess trends in the annual incidence of cardiomyopathy. Analyses were performed using Stata 6.0. Age at presentation was summarized using median and interquartile range (IQR) and comparisons between different subgroups were made using the Wilcoxon rank sum test. Lymphocytic myocarditis on endomyocardial biopsy in subjects with dilated cardiomyopathy was predictive of survival on univariable analysis, but was not entered into the multivariable
R.G. Weintraub et al. / Progress in Pediatric Cardiology 23 (2007) 17–24
19
4. Results 4.1. Epidemiological findings
Fig. 1. Cumulative frequency of age at presentation according to type of cardiomyopathy. Solid line indicates unclassified cardiomyopathy, dotted line indicates dilated cardiomyopathy, dashed line indicates hypertrophic cardiomyopathy, dotted and dashed line indicates restrictive cardiomyopathy. Reproduced from Nugent et al. [10]. Copyright © 2003 Massachusetts Medical Society. All rights reserved.
model, as cardiac histology was not available in all subjects. The Wilcoxon rank-sum test was used to compare the distribution of time from diagnosis between subjects with and without lymphocytic myocarditis, in those with available myocardial histology. Standard methods of survival analysis were used for the combined end-point of death or transplantation. Analysis of prognostic factors excluded subjects whose initial manifestation was sudden death. Important prognostic factors from the univariable analysis ( p b 0.10) were included in a multivariable Cox proportional hazards model. Changes in echocardiographic Z scores were entered into the multivariable model as continuous predictors, with the latter a time-dependent covariate defined using the most recent available echocardiographic value. Survival curves were plotted using Kaplan–Meier survival estimates, with accompanying 95% Greenwood confidence bands. All reported p values are two-sided.
The principal epidemiological findings for subjects enrolled in NACCS have previously been described [10]. There were 314 cases of cardiomyopathy diagnosed between children aged 0–10 years during the study period. Dilated cardiomyopathy was the commonest cardiomyopathy type with 184 cases (58.6%), followed by hypertrophic cardiomyopathy with 80 cases (25.5%), unclassified cardiomyopathy with 42 cases (13.4%) and the least common was restrictive cardiomyopathy with 8 cases (2.5%). Fig. 1 shows the cumulative frequency distribution for each cardiomyopathy type [10]. The median age at presentation was less than 12 months for all cardiomyopathy types other than restrictive cardiomyopathy. Altogether, 199 of 314 (63.4%) of cases presented prior to 12 months of age and 243 of 314 (77.4%%) presented prior to 2 years. Table 1 shows the annual incidence of each cardiomyopathy type according to the age at presentation [10]. The overall annual incidence was 1.24/100,000 at-risk children and the highest incidence was that of dilated cardiomyopathy before 12 months of age (annual incidence 4.76/100,000). Indigenous children had an annual incidence of dilated cardiomyopathy that was 2.67 times higher than among non-indigenous children (95% CI 1.42, 4.63). There was a male preponderance among children with both hypertrophic cardiomyopathy (68.8%) and unclassified cardiomyopathy (61.9%), with gender being evenly distributed among children with other cardiomyopathy types. Sudden death was the presenting symptom in 3.5% of cases, and was more common among indigenous compared to nonindigenous children (16.7% vs. 2.6%; p = 0.02) [10]. Familial cardiomyopathy was ascertained in 62 of 314 (19.7%) of cases, varying from 12.5% for children with restrictive cardiomyopathy to 35.7% of those with unclassified
Table 1 Incidence of cardiomyopathy (per 100,000 person–years at risk) by type and by age Diagnosis
DCM
HCM
RCM
Unclassified
Total
Age at presentation (years)
N IR 95% CI N IR 95% CI N IR 95% CI N IR 95% CI N IR 95% CI
0–b1 yrs
1–b2 yrs
2–b5 yrs
5 yrs–10 yrs
Total
121 4.8 (4.0, 5.7) 48 1.9 (1.39, 2.5) 0 0 (0, 0.15) 30 1.2 (0.80, 1.69) 199 7.8 (6.8, 9.0)
29 1.1 (0.77, 1.6) 9 0.36 (0.16, 0.67) 1 0.039 (0.001, 0.22) 5 0.20 (0.064, 0.46) 44 1.7 (1.3, 2.3)
18 0.24 (0.14, 0.37) 10 0.13 (0.063, 0.24) 4 0.053 (0.014, 0.13) 4 0.053 (0.014, 0.13) 36 0.47 (0.33, 0.65)
16 0.13 (0.072, 0.21) 13 0.10 (0.055, 0.18) 3 0.024 (0.005, 0.069) 3 0.024 (0.005, 0.069) 35 0.28 (0.19, 0.39)
184 0.73 (0.63, 0.84) 80 0.32 (0.25, 0.39) 8 0.032 (0.014, 0.062) 42 0.17 (0.12, 0.22) 314 1.24 (1.11, 1.38)
N = number of cardiomyopathy cases; IR = incidence rate per 100,000 person–years at risk; 95% CI = 95% confidence interval for incidence rate (IR); DCM — dilated cardiomyopathy, HCM — hypertrophic cardiomyopathy, RCM — restrictive cardiomyopathy. Reproduced from Nugent et al. [10]. Copyright © 2003 Massachusetts Medical Society. All rights reserved.
20
R.G. Weintraub et al. / Progress in Pediatric Cardiology 23 (2007) 17–24
Table 2 Frequency of lymphocytic myocarditis among subjects with available cardiac histology, according to time between diagnosis and examination of histological findings Time from presentation
0–7 days
N1–4 weeks
N4–8 weeks
N8 weeks
Total
Lymphocytic myocarditis Non-specific histological findings Total
22
2
1
0
25
22
8
7
8
45
44
10
8
8
70
p = 0.009, Wilcoxon rank-sum test. Reproduced from Daubeney et al. [11] with permission from the American Heart Association. Copyright 2006.
cardiomyopathy [10]. Parental consanguinity as a potential marker for a recessively inherited condition was documented in 21 of 314 cases (6.7%) where consanguinity was known, including 8.8% of children with dilated cardiomyopathy, 2.7% of children with hypertrophic cardiomyopathy and 12.5% of those with unclassified cardiomyopathy [10]. A metabolic or mitochondrial disease with predominant cardiac manifestations was documented in 28 (8.9%) of 314 cases, including a respiratory chain enzyme deficiency in 3.2%, Barth syndrome in 3.5% and a carnitine transporter deficiency in 1.3% [10]. 4.2. Dilated cardiomyopathy The presenting features and outcomes for children with dilated cardiomyopathy enrolled in this study have recently been described [11]. Congestive heart failure was the presenting symptom on 165 of 184 (89.7%) cases of dilated cardiomyopathy and 9 children presented with sudden death. Of the 175
cases diagnosed during life, 154 (88%) were hospitalized and 79 (45.1%) were admitted to an intensive care unit. Assisted mechanical ventilation was administered in 62 (35.4%) cases and inotropic support in 70 (40%). A positive family history of dilated cardiomyopathy was found in 27 (14.7%) of subjects with dilated cardiomyopathy. In 41 of 184 subjects, a virus was identified by viral culture or polymerase chain reaction from urine, stools or upper airway secretions at the time of presentation. Lymphocytic myocarditis was present in 25 of 70 (35.7%) cases with available cardiac histology obtained from autopsy, explantation or endomyocardial biopsy. Table 2 shows the inverse relationship between the time from diagnosis to histological findings and the presence of positive histological findings for lymphocytic myocarditis [11]. A positive viral identification and/or lymphocytic myocarditis were present in 30 of 44 (68.2%) of cases with available cardiac histology within 7 days of presentation, and 6 of 9 cases whose initial manifestation was sudden death. For children with dilated cardiomyopathy, freedom from death or transplantation was 72% (95% CI, 65% to 78%) at 1 year after presentation and 63% (95%CI, 55% to 70%) at 5 years after presentation. Table 3 shows risk factors for death or transplantation [11]. By multivariable analysis, these included presenting age N 5 years, the presence of familial cardiomyopathy, a lower fractional Z score at presentation and failure to increase fractional shortening Z score after presentation. Fig. 2 shows actuarial survival for all subjects with dilated cardiomyopathy (panel A), and for subjects categorized according to age at presentation (panel B) and the presence or absence of familial cardiomyopathy (panel C) [11]. Although survival was worse among subjects aged 0–4 weeks among presentation when compared to those aged 1–12 months and those aged 12 months to 5 years, the difference between these three groups did not
Table 3 Survival analysis of predictors of death or transplantation in subjects with dilated cardiomyopathy Univariable survival analysis Variable Age at presentation (4 groups) ≤ 4 weeks N4 weeks and ≤1 year N1 year and ≤ 5 years N5 years Age at presentation (2 groups)N5 years Familial cardiomyopathy Biopsy myocarditis a QRS duration Z score Fractional shortening at presentation b Change in fractional shortening from presentation b Fractional shortening Z score at presentation b Change in fractional shortening Z score from presentation b
Sample size
Hazard ratio
Multivariable survival analysis (N = 150) 95% CI
p value
c
Hazard ratio
95% CI
p value
5.6 2.9
(2.6, 12) (1.5, 5.6)
b0.0001 0.001
0.75 0.68
(0.65, 0.87) (0.58, 0.79)
b0.0001 b0.0001
0.001 34 78 47 16 175 175 39 129 150 150 150 150
1 0.73 0.44 2.26 3.27 2.35 – 1.38 0.92 0.83 0.85 0.66
(0.39, 1.4) (0.20, 0.95) (1.07, 4.8) (1.77, 6.0) (1.35, 4.1) (1.02, 1.9) (0.87, 0.97) (0.78, 0.89) (0.75, 0.96) (0.57, 0.77)
b0.0001 0.003 0.01 0.04 0.002 b0.0001 0.01 b0.0001
Excludes 9 subjects whose first manifestation was sudden death. Measured QRS duration was from the earliest available electrocardiogram. Note: For the three youngest age groups, presenting age was dichotomized at 5 years for entry in the multivariable proportional hazards regression model. Reproduced from Daubeney et al. [11] with permission from the American Heart Association. Copyright 2006. a The finding of myocarditis on biopsy was significantly predictive of survival, but a hazard ratio could not be calculated, as all patients with this characteristic were free from death or transplant; p value calculated using log-rank test. b Per unit (percent fractional shortening or unit Z score). c p values from Wald tests in Cox proportional hazards regression.
R.G. Weintraub et al. / Progress in Pediatric Cardiology 23 (2007) 17–24
21
Fig. 2. Survival to death or transplantation from time of presentation for subject with dilated cardiomyopathy: overall with 95%CI (A); with patients grouped by age at presentation (B); with patients grouped according to presence or absence of familial cardiomyopathy (C); and with patients grouped by presence or absence of lymphocytic myocarditis on endomyocardial biopsy (D). Reproduced from Daubeney et al. [11] with permission from the American Heart Association. Copyright 2006.
reach statistical significance. Lymphocytic myocarditis was associated with a high early mortality, but among 39 subjects who underwent endomyocardial biopsy, those with lymphocytic myocarditis had improved survival compared to those with nonspecific histological findings. However the presence of myocarditis was not entered into the multivariable analysis as not all subjects had available cardiac histology. At latest follow-up, 78 of 109 (71.6%) of surviving subjects are not receiving cardiac medication and 94 of 103 (91.3%) subjects with known symptom class are free of symptoms. 4.3. Hypertrophic cardiomyopathy The presenting symptoms and outcomes for children with hypertrophic cardiomyopathy enrolled in this study have also been described [12]. Congestive heart failure was the presenting symptoms in 6 of 80 (7.5%) subjects and there were no cases that presented with sudden death. The majority of subjects were diagnosed after being screened for a cardiac murmur, a family history of hypertrophic cardiomyopathy or the presence of a syndrome.
A syndromal diagnosis (most commonly Noonan syndrome) was present in 26 of 80 (32.5%) subjects and 17 (21.3%) had a family history of hypertrophic cardiomyopathy. Despite the young age of subjects with familial hypertrophic cardiomyopathy (median age at presentation 20 months), in almost all of these cases the family history included an affected adult. There was a strong association noted between the presence of biventricular involvement and concentric left ventricular hypertrophy. Left ventricular outflow tract obstruction with a resting gradient N15 mm Hg on echocardiography was present in 32 (40%) subjects either at presentation or during follow-up. Freedom from death or transplantation was 83% (95%CI, 73% to 95%) at 5 years after diagnosis and 76% (95%CI, 62% to 86%) at 10 years. The majority of subjects who died had congestive heart failure rather than sudden, unexpected death. Table 4 shows risk factors for death or transplantation [12]. By multivariable analysis these included presenting age b 1 year, the presence of concentric left ventricular hypertrophy, lower initial fractional shortening Z score and increasing left ventricular Z score from presentation. Fig. 3 shows actuarial
22
R.G. Weintraub et al. / Progress in Pediatric Cardiology 23 (2007) 17–24
Table 4 Survival analysis of predictors of death or transplantation in subjects with hypertrophic cardiomyopathy Univariable survival analysis
Multivariable survival analysis (n = 72)
Variable
Sample size (n)
Hazard ratio
95% CI
p value
Hazard ratio
95% CI
p value
Presenting age b 1 year Congestive heart failure at presentation Concentric left ventricular hypertrophy Biventricular involvement Familial hypertrophic cardiomyopathy a Parental consanguinity LV posterior wall Z score at presentation b Change in LV posterior wall Z score from presentation b Fractional shortening Z score at presentation c Change in fractional shortening Z score from presentation c
80
2.94
(0.83, 10)
0.10
6.2
(1.44, 26)
0.01
80 80 80 80 73 77 77 72 72
3.46 5.4 4.4
(0.97, 12) (1.8, 16) (1.57, 12)
8.0
(1.33, 48)
0.02
12.5 1.23 1.19 1.16 1.02
(2.68, 58) (1.03, 1.46) (0.94, 1.50) (0.99, 1.36) (0.89, 1.16)
0.06 0.002 0.005 0.04 0.001 0.02 0.14 0.07 0.79
1.02 1.36 1.31 1.14
(0.76, 1.37) (1.03, 1.81) (1.08, 1.6) (0.99, 1.31)
0.91 0.03 0.01 0.06
p value calculated using log-rank test. For each echocardiographic variable that was included at baseline, the respective time varying variable was also included in the multivariable mode. Reproduced from Nugent et al. [12] with permission from the American Heart Association. Copyright 2005. a Familial cardiomyopathy was predictive of survival, but a hazard ratio could not be calculated, as all patients with this characteristic were free from death or transplant. b Per unit Z score increase. c Per unit (percent fractional shortening or Z score) decrease.
survival for all subjects with hypertrophic cardiomyopathy (panel A), and for subjects categorized according to the presence or absence of familial cardiomyopathy (panel B),
concentric vs. asymmetric septal hypertrophy (panel C) and presenting age b 12 months compared with remaining subjects (panel D) [12].
Fig. 3. Survival to death or transplantation from time of presentation for subjects with hypertrophic cardiomyopathy: overall with 95%CI (A); with patients grouped by presence or absence of familial cardiomyopathy (B); morphology of left ventricular involvement (C); and with patients grouped by age at presentation (D).Reproduced from Nugent et al. [12] with permission from the American Heart Association. Copyright 2005.
R.G. Weintraub et al. / Progress in Pediatric Cardiology 23 (2007) 17–24
At latest follow-up, 46 of 65 (70.8%) of surviving subjects are not receiving any cardiac medication and 54 of 63 (85.7%) with known symptom class were free of any symptoms.
23
childhood cardiomyopathy, from Finland, reported an annual incidence of 0.73/100,000 but did not include children with myocarditis and those with self-limiting cardiomyopathies [15].
4.4. Restrictive cardiomyopathy 5.2. Dilated cardiomyopathy The majority of subjects presented with congestive heart failure or exercise intolerance. Survival without intervention in this group was poor, with 4 of 8 subjects undergoing cardiac transplantation and 3 of the remaining 4 having died. 4.5. Left ventricular non-compaction There were 29 subjects with left ventricular non-compaction identified, comprising 9.2% of the overall study population. Most cases presented with congestive heart failure and had left ventricular systolic dysfunction at presentation. Freedom from death or transplantation in this group was only 53% at 5 years after presentation. 5. Discussion The National Australian Childhood Cardiomyopathy study is the largest population-based study of pediatric cardiomyopathy. Consistency of case classification was ensured by having a single cardiologist review and classify all enrolled subjects, and a single pathologist examine available cardiac histology. In addition to characterizing presentations and outcomes for children with cardiomyopathy, this populationbased study highlights the early mortality among children with dilated cardiomyopathy. 5.1. Epidemiological findings Striking similarities exist between the epidemiological findings from this study and that of the North American Pediatric Cardiomyopathy Registry [10,13]. Both studies have shown the highest incidence of cardiomyopathy to be in the first year of life, with approximately 50% of cases ascertained by this age [14]. Both studies have shown significant differences in the incidence of cardiomyopathy according to ethnic factors [14]. This may be due to either environmental or genetic differences. Indigenous Australian children have one of the highest incidences of rheumatic fever in the world, and appear to be at higher risk of serious complications from this infectious illness. Both studies also found a significant difference in the incidence of different cardiomyopathy types according to gender [14]. Familial cardiomyopathy was found in approximately 9% of subjects in the North American study, 20% of subjects in the Australian study. Despite somewhat differing entrance criteria (the North American study included subjects with neuromuscular and metabolic diseases), there was a remarkable similarity in the overall annual incidence of pediatric cardiomyopathy, with a figure of 1.24/100,000 children in Australia and 1.13/100,000 in two geographical areas of North America [10,13]. The only other national study of
There is considerable variation in reported outcomes for children with dilated cardiomyopathy, with five-year survival rates of between 20% and 84% [16–19]. Proposed risk factors include age N 2 years at presentation [17], left ventricular shape [16], severity of diastolic dysfunction and the presence of right ventricular failure [20]. Some of these variables cannot easily be ascertained in a population-based study. Few studies have reported late outcomes or examined survival in relation to serial assessment of left ventricular function. In the present study risk factors for death or transplantation included age N 5 years at presentation, familial dilated cardiomyopathy, lower initial fractional shortening Z scores and failure to improve fractional shortening Z scores during followup. Similar findings have been reported from the North American Pediatric Cardiomyopathy Registry [21]. Lymphocytic myocarditis was found to be common among subjects with available early cardiac histology. The inverse relation between the finding of lymphocytic myocarditis and duration since time of presentation is consistent with animal studies in which there is disappearance of a lymphocytic myocardial infiltrate within 6 weeks of viral inoculation [22]. It is unclear whether the improved survival noted in subjects with myocarditis diagnosed by endomyocardial biopsy was due in some way to patient selection, the more favorable natural history of this condition or the use of immune modulating therapies. This data contrasts with that of adult studies in which the treatment of myocarditis does not confer a survival advantage [23]. 5.3. Hypertrophic cardiomyopathy The causes of pediatric hypertrophic cardiomyopathy have not been well defined. The presence of a recognizable syndrome in 32.5% of subjects and a positive family history in 21.3% of subjects in our study provide clues to a pre-existing genetic predisposition in a high proportion of cases. Prior institutional reviews have indicated an annual mortality of around 5% in children with hypertrophic cardiomyopathy [24,25], and some studies have indicated mortality in excess of 50% for children presenting prior to 12 months of age [25,26]. In the present study, the average annual mortality was 3.4% for subjects who were diagnosed during the first year of life and only 1.5% for subjects diagnosed after this time. Risk factors for death or transplantation included the absence of a family history, concentric left ventricular hypertrophy, lower fractional shortening Z score at diagnosis and increasing left ventricular wall thickness from the time of presentation. Potentially each of these risk factors reflects an unfavorable spectrum of etiologies such as mitochondrial diseases, characterized by more diffuse hypertrophy, as well as worse systolic and diastolic left ventricular
24
R.G. Weintraub et al. / Progress in Pediatric Cardiology 23 (2007) 17–24
function. The protective effect of familial hypertrophic cardiomyopathy may in some subjects be due to incomplete expressivity of sarcomeric protein mutations during the first decade of life. The majority of subjects died of progressive congestive heart failure rather than cardiac arrhythmias and for all subjects with established risk factors, the highest mortality was in the first year after diagnosis. 5.4. Other outcomes The majority of surviving subjects with dilated and hypertrophic cardiomyopathy were not receiving cardiac medication and did not have symptoms at latest follow-up. However there is a lack of information about psychological and cognitive function in children with cardiomyopathy who have been subjected to various therapies including cardiac transplantation. 6. Summary Pediatric cardiomyopathies are a heterogeneous group of disorders with diverse genetic, infectious, mitochondrial and metabolic etiologies. Lymphocytic myocarditis and left ventricular non-compaction are important causes of cardiomyopathy in Australian children. The timing and severity of presentation vary according to cardiomyopathy type as well as genetic and ethnic factors. The behavior of specific cardiomyopathies can be predicted by morphological and functional attributes, as well as underlying patient characteristics. The diagnosis and management of pediatric cardiomyopathy requires a multidisciplinary approach. References [1] Boucek MM, Edwards LB, Keck BM, Trulock EP, Taylor DO, Hertz MI. Registry of the International Society for Heart and Lung Transplantation: eighth official pediatric report—2005. J Heart Lung Transplant 2005;24: 968–82. [2] Maron BJ. Hypertrophic cardiomyopathy: a systematic review. JAMA 2002;287:1308–20. [3] Maron BJ. Sudden death in young athletes. N Engl J Med 2003;349: 1064–75. [4] Spirito P, Chiarella F, Carratino L, Berisso MZ, Bellotti P, Vecchio C. Clinical course and prognosis of hypertrophic cardiomyopathy in an outpatient population. N Engl J Med 1989;320:749–55. [5] Richardson P, McKenna W, Bristow M, et al. Report of the 1995 World Health Organization/International Society and Federation of Cardiology Task Force on the Definition and Classification of cardiomyopathies. Circulation 1996;93:841–2. [6] Aretz HT, Billingham ME, Edwards WD, et al. A histopathologic definition and classification. Am J Cardiovasc Pathol 1987;1:3–14. [7] Sluysmans T, Colan SD. Theoretical and empirical derivation of cardiovascular allometric relationships in children. J Appl Physiol 2005;99: 445–57.
[8] Colan SD, Parness IA, Spevak PJ, Sanders SP. Developmental modulation of myocardial mechanics: age- and growth-related alterations in afterload and contractility. J Am Coll Cardiol 1992;19:619–29. [9] Nugent AW, Davis AM, Kleinert S, Wilkinson JL, Weintraub RG. Clinical, electrocardiographic, and histologic correlations in children with dilated cardiomyopathy. J Heart Lung Transplant 2001;20:1152–7. [10] Nugent AW, Daubeney PE, Chondros PC, et al. National Australian Childhood Cardiomyopathy S. The epidemiology of childhood cardiomyopathy in Australia. N Engl J Med 2003;348:1639–46. [11] Daubeney PE, Nugent AW, Chondros PC, et al. National Australian Childhood Cardiomyopathy S. Clinical features and outcomes of childhood dilated cardiomyopathy: results from a national populationbased study. Circulation 2006;114:2671–8. [12] Nugent AW, Daubeney PE, Chondros PC, et al. National Australian Childhood Cardiomyopathy S. Clinical features and outcomes of childhood hypertrophic cardiomyopathy: results from a national population-based study. Circulation 2005;112:1332–8. [13] Lipshultz SE, Sleeper LA, Towbin JA, et al. The incidence of pediatric cardiomyopathy in two regions of the United States. N Engl J Med 2003;348:1647–55. [14] Strauss A, Lock JE. Pediatric cardiomyopathy—a long way to go. N Engl J Med 2003;348:1703–5. [15] Arola A, Jokinen E, Ruuskanen O, et al. Epidemiology of idiopathic cardiomyopathies in children and adolescents. A nationwide study in Finland. Am J Epidemiol 1997;146:385–93. [16] Matitiau A, Perez-Atayde A, Sanders SP, et al. Infantile dilated cardiomyopathy. Relation of outcome to left ventricular mechanics, hemodynamics, and histology at the time of presentation. Circulation 1994;90:1310–8. [17] Akagi T, Benson LN, Lightfoot NE, Chin K, Wilson G, Freedom RM. Natural history of dilated cardiomyopathy in children. Am Heart J 1991;121: 1502–6. [18] Friedman RA, Moak JP, Garson Jr A. Clinical course of idiopathic dilated cardiomyopathy in children. J Am Coll Cardiol 1991;18:152–6. [19] Tsirka AE, Trinkaus K, Chen SC, et al. Improved outcomes of pediatric dilated cardiomyopathy with utilization of heart transplantation. J Am Coll Cardiol 2004;44:391–7. [20] Arola A, Tuominen J, Ruuskanen O, Jokinen E. Idiopathic dilated cardiomyopathy in children: prognostic indicators and outcome. Pediatrics 1998;101:369–76. [21] Towbin JA, Lowe AM, Colan SD, et al. Incidence, causes, and outcomes of dilated cardiomyopathy in children. JAMA 2006;296:1867–76. [22] Kawai C. From myocarditis to cardiomyopathy: mechanisms of inflammation and cell death: learning from the past for the future. Circulation 1999;99:1091–100. [23] Grogan M, Redfield MM, Bailey KR, et al. Long-term outcome of patients with biopsy-proved myocarditis: comparison with idiopathic dilated cardiomyopathy. J Am Coll Cardiol 1995;26:80–4. [24] McKenna W, Deanfield J, Faruqui A, England D, Oakley C, Goodwin J. Prognosis in hypertrophic cardiomyopathy: role of age and clinical, electrocardiographic and hemodynamic features. Am J Cardiol 1981;47: 532–8. [25] Maron BJ, Tajik AJ, Ruttenberg HD, et al. Hypertrophic cardiomyopathy in infants: clinical features and natural history. Circulation 1982;65: 7–17. [26] Suda K, Kohl T, Kovalchin JP, Silverman NH. Echocardiographic predictors of poor outcome in infants with hypertrophic cardiomyopathy. Am J Cardiol 1997;80:595–600.
