Author + information
- Received February 27, 2006
- Revision received June 26, 2006
- Accepted July 3, 2006
- Published online October 17, 2006.
- Elisabeth Villain, MD⁎,⁎ (, )
- Nathalie Coastedoat-Chalumeau, MD†,
- Eloi Marijon, MD⁎,
- Younes Boudjemline, MD⁎,
- Jean-Charles Piette, MD† and
- Damien Bonnet, MD, PhD⁎
- ↵⁎Reprint requests and correspondence:
Dr. Elisabeth Villain, Cardiologie pédiatrique, Hôpital Necker-Enfants Malades, 149, rue de Sèvres, 75015 Paris, France.
Objectives We sought to determine whether the presentation and prognosis of children with complete atrioventricular block (CAVB) were related to maternal antibody status.
Background Comparative studies related to the presence or absence of maternal antibodies anti-SSB/La and anti-SSA/Ro are lacking in children with isolated complete CAVB.
Methods From 1980 to 2004, we screened for maternal antibodies in 111 children <15 years old with CAVB. According to the presence (Ab+) or absence (Ab−) of antibodies, 2 groups of patients were retrospectively compared.
Results The study group included 56 Ab+ and 55 Ab− patients with equal gender distribution. A total of 96% Ab+ patients were diagnosed in utero or within the first month, compared with 24% Ab− patients. Progression from incomplete to complete block was shown in 23 Ab− and 2 Ab+ patients. Echocardiography showed normal heart structures in Ab− patients, but 8 Ab+ patients had ostium secundum or ductus arteriosus. Pacemaker implantation was performed in 105 patients, and age at implantation was younger in the Ab+ group. At follow-up (age 9.7 ± 6 years), all Ab− patients were alive with normal left ventricular function; dilated cardiomyopathy was diagnosed at diagnosis or during follow-up in 16 Ab+ patients, and 6 of 16 have died.
Conclusions Patients with antibody-mediated CAVB were diagnosed and underwent pacing earlier in life and had a more severe prognosis than Ab− patients because of a high risk of dilated cardiomyopathy. The absence of antibody suggests a different pathologic mechanism than autoimmunity, and the term congenital may be not appropriate in these cases.
Complete atrioventricular block (CAVB) is rare in children with normal heart structures (i.e., isolated CAVB). Cardiologists have long stuck to the idea that isolated CAVB diagnosed at a “fairly early age” was congenital, even when diagnosed beyond the neonatal period (1,2). By contrast, congenital CAVB is defined by rheumatologists as CAVB “existing at or dating from birth” (3) and more recently as “CAVB diagnosed in utero, at birth, or in the first month of life” (4,5). Isolated CAVB diagnosed in utero or at birth is often associated with the maternal antibodies anti-SSA and/or anti-SSB, and have a more severe prognosis that those diagnosed later in life (6–9). However, because many studies were retrospective and involved numerous decades, maternal antibodies status is not always known, especially in late onset CAVB (10), and comparative studies related to maternal antibody status are lacking in children (5). In a series of consecutive children with isolated CAVB who had a standard follow-up in our institution since 1980, we compared 2 groups of patients to determine whether the clinical presentation and the prognosis of this condition were related to the presence of maternal antibodies.
The study was performed using the diagnostic database of the Department of Paediatric Cardiology at Necker Enfants-Malades in Paris. All cases of isolated CAVB diagnosed between 1980 and 2004 in children <15 years old were included in the study. Patients born before 1980 were excluded because of the major progress in prenatal diagnosis and neonatal management of CAVB, as well as in pacing technology. Patients with traumatic or postoperative block were excluded from the study, as well as those with myocarditis, neuromuscular disorders, metabolic diseases (such as mitochondrial diseases), and significant congenital heart disease likely to account for CAVB and long QT syndrome. Patients with minor abnormalities (atrial septal defects, patent ductus arteriosus, mild pulmonary stenosis) were included in the study because these defects have already been described in association with maternal antibodies (11).
Complete atrioventricular block was defined by atrioventricular dissociation with a ventricular rate lower than the atrial rate. The time of detection of CAVB was taken as the earliest documentation of a patient’s conduction abnormality. Data collected included demographic information, patient age at presentation, echocardiographic data, pacemaker (PM) intervention procedure, and status at follow-up.
Our indications for pacing followed the guidelines established by the American College of Cardiology/American Heart Association (12). Since 1980, our current approach has been to implant venous leads in children who are >10 kg and to surgically pace children who are <10 kg; epicardial electrodes were initially sutured on the right ventricular apex and on the left ventricular apex in the 5 most recent patients. Double-chamber PMs were programmed with a maximum heart rate of 180 beats/min; ventricular rate-responsive PM systems were programmed with a maximum ventricular rate of 120 to 150 beats/min.
All children had regular follow-up in our institution, including at least 2 visits per year with assessment of clinical status, electrocardiogram (ECG), and echocardiography. The diagnosis of dilated cardiomyopathy (DCM) was defined as left ventricular end diastolic diameter >97th percentile associated with a shortening fraction <25%.
Maternal antibody determination
The screening for antibodies was performed in mothers of children with CAVB and in patients <3 months old. Maternal antibodies to soluble nuclear antigens 48-kd SSB/La, 52-kd SSA/Ro, and 60-kd SSA/Ro were detected using quantitative radioligand assays as previously described (13). Before the year 2000, anti-SSA and anti-SSB autoantibodies were detected and quantified using the classic Ouchterlouny double diffusion technique. For this reason, mothers of patients with CAVB who had been found to be negative before 2000 were re-investigated using radioligand assays. Patients were assigned to 2 different groups according to the presence (group Ab+) or the absence (group Ab−) of maternal antibodies.
Data were expressed as mean value ± standard deviation and median and range. Statistical analysis comparing 2 groups (Ab+) and (Ab−) was performed with unpaired 2-tailed ttesting for the means or chi-square for categorical variables. All p values of 0.05 or less were considered significant.
One hundred thirty-six children with isolated CAVB fulfilled the inclusion criteria. Among them, 25 were excluded from our study because they lived abroad and were lost to follow-up or because the information about maternal antibodies was not available. Therefore, 111 children were included in the study; maternal antibodies were present in 56 cases and absent in 55 cases. Table 1summarizes the data obtained in the 2 groups. No differences were seen between both groups for gender distribution.
Age at diagnosis
Fifty-six cases were diagnosed prenatally. The mean term at diagnosis was 26.2 ± 9 weeks gestation (median 24 weeks, range 18 to 38 weeks). The rate of premature delivery (37%) and the proportion of low-birth-weight infants (66% <25th percentile and 18% <10th percentile) were high in this subgroup. Fifty children belonged to the Ab+ group and 6 to the Ab− group (p < 0.0001). Among 20 children diagnosed between birth and 1 year of age, 11 were <1 month (4 Ab+, 7 Ab−) and 9 between 1 and 12 months (1 Ab+, 8 Ab−) old. In the Ab+ group, 54 of 56 (96%) of the CAVB were diagnosed in utero or within the first month of life versus 13 of 55 (24%) in the Ab− group (p < 0.0001). Thirty-four of 35 patients diagnosed with CAVB after 1 year of age belonged to the Ab− group.
Characteristics of AVB at the time of diagnosis
The mean heart rate at diagnosis of CAVB was 51 ± 11 beats/min (range 35 to 80) in the Ab+ group versus 48 ± 11 beats/min (range 30 to 75) in the Ab− group.
Only 2 patients in the Ab+ group initially had documented incomplete block, both in utero, which progressed to CAVB after birth. Conversely, 23 of 56 patients in the Ab− group had a proven incomplete block before completion (p < 0.0001). In 2 of 6 fetuses, 4 of 7 neonates, 8 of 8 infants, and 9 of 29 children older than 1 year belonging to the Ab− group, 24-h ECG monitoring at the time of diagnosis displayed 1:1 atrioventricular conduction with a long PR interval and episodes of second-degree AVB.
Echocardiographic data at diagnosis
All children in the Ab− group had normal heart structures. In the Ab+ group, 5 patients had an ostium secundum, requiring closure in 4 cases (3 surgeries, 1 percutaneous closure) and no intervention in 1 case; 3 patients had a patent ductus arteriosus that was closed surgically in 2 and percutaneously in 1. One patient with a ductus arteriosus also had a mild pulmonary stenosis.
At the time of diagnosis, left ventricular function was normal in all Ab− patients, whereas 9 Ab+ patients had left ventricular DCM.
Pacemaker implantation was performed in 105 of 111 patients, 52 in the Ab+ group and 53 in the Ab− group (p = NS). Prophylactic pacing was decided in 93 asymptomatic patients according to a low heart rate (n = 76), prolonged pauses (n = 10 patients), and/or frequent premature ventricular beats (n = 7), with no difference in the indications between the 2 groups. Twelve patients had symptoms: cardiac failure in 4 neonates, all with heart rates <45 beats/min, episodes of dizziness and/or asthenia in 6 Ab− children, and syncope in 2 Ab− children.
Age at PM implantation was younger in the Ab+ group: 65% (34 of 52) of the children in the Ab+ group were paced before 1 year of age, versus 23% (12 of 53) in the Ab+ group (p < 0.0001). Consequently, the proportion of epicardial system in the Ab+ group was higher, but there was no significant difference in the proportion of single and double chambers between both groups (Table 1).
After a mean follow-up of 9 ± 6 years (median 9.7 years; range 6 months to 24 years), all patients in the Ab− group (mean age 11.5 ± 5.6 years) were alive, and all had normal left ventricular function on echocardiography.
Conversely, 6 patients died in the Ab+ group. Of note, DCM was observed only in this subgroup of patients, as 16 of 56 had a DCM during follow-up (Table 2).In these 16 patients, the AV block was diagnosed in utero in 14 or at birth in 2. The diagnosis of DCM was done in fetal life or at birth in 10 patients (patients 1 to 10) and DCM developed during follow-up in the remaining 6 (patients 11 to 16), in whom it was diagnosed by systematic echocardiography (4 cases) or according to heart failure (2 cases). In 12 patients, DCM was present before PM implantation and did not worsen after pacing. In the other 4 patients, DCM developed after PM implantation; duration of pacing before diagnosis of DCM was 4 months in 2 cases (patients 11 and 12), 12 years in patient 15, and 16 years in patient 16. We did not find any difference in the PM programming between patients with DCM and patients with normal left ventricular function.
All deaths were related to DCM. One neonate with DCM from fetal life died of low cardiac output shortly after PM implantation (patient 9), and the 5 other patients died of cardiac failure, between 6 months and 12 years of age. Among the 10 survivors with DCM, 2 underwent successful cardiac resynchronization, 1 is awaiting heart transplantation, and 7 received chronic medication including beta-blockers and/or angiotensin-converting enzyme inhibitors.
The term congenital CAVB has been coined by pediatric cardiologists to define atrioventricular block in children with normal heart structures. It has long been supposed that all CAVB diagnosed in childhood were congenital and that the diagnosis had been delayed because of a high ventricular rate or absence of symptoms. However, our study confirms that there are indeed 2 different diseases with regard to the presence of maternal antibodies status. Furthermore, we show here that the outcome of CAVB is closely related to its immune origin.
Age at diagnosis and mechanism of block
When the block is diagnosed in the fetus or in the neonate, it is predominantly associated with maternal anti-Ro/SSA and/or anti-La/SSB antibodies, which enter the fetal circulation and trigger an inflammatory response, leading to fibrosis of the conduction system (14–16). In our series, 54 of 56 AVBs diagnosed in utero, 4 of 11 AVBs diagnosed in neonates, and only 2 of 44 AVBs diagnosed after 1 month of age were associated with maternal antibodies; when CAVB was diagnosed after 5 years of age, no maternal antibodies were found. In recent series and with the most sensitive and advanced methods, maternal autoantibodies were found in nearly 100% of patients diagnosed in utero but only in a minority of children diagnosed after the neonatal period (2,6,7,10,17–21). Because it has been shown that once maternal antibodies have been detected, they remain lifelong in mothers’ sera (21,22), their absence suggests that other mechanisms must be involved in the pathogenesis of the AVB. Thus, the denomination antibody-associated CAVB would be more appropriate than solely congenital CAVB when referring to the subgroup of patients with autoantibody-associated CAVB.
We found that 3 of 56 children in the Ab+ group versus 23 of 55 in the Ab− group had progression from incomplete to complete AVB, mostly after birth. Although there is clear documentation that incomplete block can progress after birth in children with antibody-associated CAVB, despite the clearance of maternal antibodies from the neonatal circulation (18,23), others have found that this progression occurs more frequently in the absence of maternal antibodies (6). These data add further evidence showing that isolated CAVB may be related to a degenerative process involving the conduction tissue and occurring often after birth, making the denomination congenital inappropriate.
Associated structural anomalies
By definition, autoantibody-mediated CAVB occurs in children without cardiac abnormalities that would be causal for the development of the block. However, miscellaneous cardiac lesions, mainly ostium secundum and ductus arteriosus, were found frequently in the Ab+ group (8 cases), whereas none were noticed in the Ab− group. Accordingly, cardiac structural lesions have been reported in 16% to 42% of children with antibody-associated CAVB (10,18), and such lesions have also been noticed in infants without CAVB born to anti-SSA/Ro-positive mothers (24).
Management of children with CAVB
The prognosis of CAVB in children is influenced by age, and numerous series have reported that patients detected in utero or at birth have a poorer outcome than those diagnosed later in life (8,9,18,25–27). In retrospective series covering multiple decades, the high mortality and morbidity of neonates with CAVB were mainly related to prematurity, delayed initiation of pacing therapy, and difficulties in pacing very young children. Improvement in management of these high-risk pregnancies (i.e., development of fetal echocardiography and transfer of pregnant patients with affected fetuses to highly specialized centers) and in the initial management of neonates has changed the prognosis of CABV (8). Accordingly, in our series, the only neonatal death occurred in a premature baby who had been delivered by cesarean section because of a very severe DCM.
In neonates, the mechanism of CAVB did not influence the timing of PM implantation because 9 of 13 Ab− and 28 of 54 Ab+ infants were paced before 1 month of age. There was no death related to PM implantation in neonates (28).
All together, 94.5% of our patients were paced before they reached 15 years of age, and with the development of prophylactic pacing, the majority of patients with CAVB now undergo PM implantation in childhood. There was no sudden death, and 2 patients had syncope; in 1 case it was the first symptom of CAVB in a previously healthy child, and the other child was followed up in another department, where prophylactic pacing had not been proposed despite a heart rate of 45 beats/min. These results confirms that prophylactic pacing has considerably decreased the incidence of previously reported severe complications such as syncope or sudden death in children with isolated CAVB (9,26,27,29,30).
Congenital CAVB and DCM
In the whole group of children, 16 of 111 (14.4%) have developed DCM. This high incidence is probably the consequence of the high number of patients who had an early diagnosis in our population. All 16 patients had been diagnosed with CAVB in utero or at birth, so that the incidence of DCM in this subgroup is 29.5% (16 of 54). In 10 cases, the diagnosis of DCM was made during fetal life or during the neonatal period. In the 6 remaining patients, DCM was diagnosed during follow-up. DCM has already been described with a high incidence in children with early diagnosed CAVB (10,31–33). Among 149 children with CAVB, Udink ten Cate et al. (32) identified 9 patients (6%) who developed DCM at a mean age of 6.5 ± 5 years; however, when considering the group of 34 patients diagnosed in utero, 8 of them developed DCM (23%). In a retrospective series of 91 children with CAVB diagnosed in utero or at birth, Eronen et al. (10) have reported an incidence of 23% of DCM that was diagnosed before 3 years of age in all cases. Moak et al. (31) have reported 16 infants who had normal left ventricular function at the time of diagnosis of congenital CAVB and who developed a DCM between 2 weeks and 30 months of age in 13 cases and later in life (5.8 ± 3 years) in 3 cases. Thus, close echocardiographic follow-up is warranted in children with CAVB, even when left ventricular parameters are initially considered normal.
In our series, none of the Ab− patients and 16 Ab+ patients (28.5%) had DCM. In others series, although DCM was mainly described in children with CAVB diagnosed in utero or at birth, no correlation was established between maternal autoimmunity and DCM because serologic findings were often lacking, especially in mothers of children with late-onset heart block. To our knowledge, our study is the first to compare 2 groups of children with isolated CAVB, according to maternal antibody status, and to show the association between antibody-mediated CAVB and a high risk of DCM. It adds further evidence to suggest that children with antibody-associated CAVB have myocardial damage caused by the inflammatory process, which may progress or be reactivated during follow-up (11,16,31,34–36).
Pacemaker-induced DCM has been offered as a possible etiologic factor because right ventricular pacing induces myocardial dystrophic changes and adverse remodeling caused by ventricular desynchronization. This seems unlikely in our series because the majority of paced children did not develop DCM and the majority of patients with DCM developed it before being paced. Cruz et al. (37) have reported in an adult series that cardiac failure after PM implantation was observed only in patients with heart block associated with maternal antibodies. Moreover, DCM has been reported in unpaced patients with CAVB, which was the case in 12 of our patients, and more recently in fetuses and infants with maternal antibodies, in the absence of any conduction anomaly (11). However, this subgroup of children with maternal antibodies and preexisting myocardial damage may be particularly at risk for pacing-induced cardiomyopathy.
The prognosis of patients in whom DCM developed was very poor; 6 of them died and 1 is awaiting transplantation. This severe prognosis of DCM is reported in all series, despite early diagnosis of DCM and medical treatment. Cardiac resynchronization was performed in 2 patients, with good clinical result, and others have been stabilized with medical therapy.
In conclusion, our study suggests that 2 different forms of isolated CAVB should be differentiated in children. Congenital CAVB is usually associated with maternal antibodies, is increasingly diagnosed in utero, and despite early pacing, still carries significant morbidity and mortality because of a high incidence of DCM. Long-term prognosis is difficult to predict because DCM is not always obvious at diagnosis, and prolonged echocardiographic monitoring of these patients is mandatory. In children whose mothers have no antibodies, the block is often progressive and diagnosed later in life, its etiology remains unknown, and these children have a good prognosis when usual prophylactic recommendations for PM implantation are applied.
Twenty-five patients, representing 20% of potential enrollees, were excluded from the study because of lack of antibody status. Fifteen of these patients did not have the same clinical profile as those included in the study: they lived abroad, were diagnosed later than the patients in our series (neither neonate nor infant), and were sent to Paris only for PM implantation. Twenty other patients, diagnosed before 1988, are now followed up by adult cardiologists or are lost to follow-up; in addition to antibody status, initial echocardiography was often lacking or imprecise in these patients.
The authors thank Dr. Marie-Alice Alyanakian from the Immunology Department of Hospital Necker and Dr. Lucile Musset from the Immunology Department of Hospital La Pitié Salpêtrière for their technical assistance.
- Abbreviations and Acronyms
- complete atrioventricular block
- dilated cardiomyopathy
- Received February 27, 2006.
- Revision received June 26, 2006.
- Accepted July 3, 2006.
- American College of Cardiology Foundation
- Brucato A.,
- Jonzon A.,
- Friedman D.,
- et al.
- Rosenthal E.
- Frohn-Mulder I.M.,
- Meilof J.F.,
- Szatmari A.,
- Stewart P.A.,
- Swaak T.J.,
- Hess J.
- Jaeggi E.T.,
- Hamilton R.M.,
- Silverman E.D.,
- Zamora S.A.,
- Hornberger L.K.
- Eronen M.,
- Siren M.K.,
- Ekblad H.,
- Tikanoja T.,
- Julkunen H.,
- Paavilainen T.
- Gregoratos G.,
- Abrams J.,
- Epstein A.E.,
- et al.
- Yamamoto A.M.,
- Amoura Z.,
- Johannet C.,
- et al.
- Buyon J.P.,
- Hiebert R.,
- Copel J.,
- et al.
- Buyon J.P.,
- Rupel A.,
- Clancy R.M.
- Buyon J.P.,
- Winchester R.J.,
- Slade S.G.,
- et al.
- Derksen R.H.,
- Meilof J.F.
- Tseng C.E.,
- Di Donato F.,
- Buyon J.P.
- Askanase A.D.,
- Friedman D.M.,
- Copel J.,
- et al.
- Costedoat-Chalumeau N.,
- Amoura Z.,
- Lupoglazoff J.M.,
- et al.
- Michaelsson M.,
- Jonzon A.,
- Riesenfeld T.
- Pinsky W.W.G.P.,
- Garson A. Jr..,
- McNamara D.G.
- Reid J.M.,
- Coleman E.N.,
- Doig W.
- Moak J.P.,
- Barron K.S.,
- Hougen T.J.,
- et al.
- Udink ten Cate F.E.,
- Breur J.M.,
- Cohen M.I.,
- et al.
- Taylor-Albert E.,
- Reichlin M.,
- Toews W.H.,
- Overholt E.D.,
- Lee L.A.
- Nield L.E.,
- Silverman E.D.,
- Smallhorn J.F.,
- et al.
- Nield L.E.,
- Silverman E.D.,
- Taylor G.P.,
- et al.