Materials and methods

Demographic, clinical data, echocardiography, catheterisation, ablation, surgery, and outcome results of patients diagnosed and followed as Ebstein’s anomaly in our centre, between 2000 and 2017, were retrospectively evaluated from the hospital files. The study protocol was approved by the Hacettepe University, School of Medicine Ethics Committee (Ethical consent number: GO 17/762-08).

Inclusion criteria included diagnosis of Ebstein’s anomaly by transthoracic echocardiography. The Mayo Clinic criteria were used for diagnosis and included apical displacement of the septal tricuspid valve leaflet from the insertion of the anterior leaflet of the mitral valve by 0.8 cm/m² body surface area in systole. ^{Reference O’Leary, Eidem, O’Leary and Cetta18} Exclusion criteria included Ebstein’s anomaly patients with missing data and incomplete hospital files.

Valve insufficiencies were graded as mild, moderate, or severe by Doppler and colour imaging according to the recommended guidelines. ^{Reference Lancellotti, Tribouilloy and Hagendorff19}

All patients had electrocardiogram and 24-hour electrocardiogram recordings. Transesophageal and/or invasive electrophysiological studies were performed in all patients with symptomatic tachycardia or documented arrhythmia (Wolff-Parkinson-White pattern or sustained tachycardia) in electrocardiograms or in 24-hour electrocardiogram recordings. Ablation or cryoablation procedures were performed in all patients more than 15 kgs with documented supraventricular tachycardia by electrophysiological studies. Three-dimensional imaging techniques (Ensite NavX navigation display technology) and Coil-Assisted Retrograde Transvenous Obliteration were used in some of the patients during the ablation procedures. Patients less than 15 kgs and patients without successful ablation were treated with anti-arrhythmic drugs.

Cardiac magnetic resonance imaging (MRI) were used in some of the patients as a complementary imaging with echocardiography.

Indications for surgery included severe or progressive cyanosis, congestive heart failure due to severe valvular regurgitation despite medical treatment, decreasing or severe limitation of activity, severe ventricular outflow tract obstruction, and recurrent or intractable life-threatening arrhythmia despite medical treatment.

Discussion

Studies in literature about management and outcome of paediatric patients with Ebstein’s anomaly are limited. ^{Reference Celermajer, Cullen, Sullivan, Spiegelhalter, Wyse and Deanfield4,Reference Geerdink, Delhaas and Helbing8–Reference Holst, Dearani and Said17} In this study, we evaluated a large cohort of children and young adults with Ebstein’s anomaly and analysed factors affecting arrhythmia presence and mortality.

In a retrospective multicentre study by Kapusta et al., which evaluated 93 paediatric patients diagnosed and followed as Ebstein’s anomaly, median diagnosis age was less than 30 days (1 day–162 months). ^{Reference Kapusta, Eveleigh and Poulino9} Median follow-up period was 86 months (maximum, 216 months) and 81% of the patients survived. Younger age (12 months) at presentation, mechanical ventilation, diuretic and prostaglandin requirement at presentation, presence of hepatomegaly, moderate/severe pulmonary stenosis, pulmonary atresia, patent ductus arteriosus, and ventricular septal defect were significantly associated with death. Gender, gestational age, birth weight, degree of tricuspid regurgitation, need for operation, cyanosis, and arrhythmias did not significantly affect mortality. Boston et al. analysed risk factors for mortality in 52 children younger than 12 years who underwent tricuspid valve repair and annuloplasty for Ebstein’s anomaly. ^{Reference Boston, Dearani, O'Leary, Driscoll and Danielson15} Mean age at surgery was 7.1 ± 3.9 years (5 months to 12 years) and follow-up period was 12.2 ± 7.4 years (10 months–24.3 years). Early mortality rate was 5.8%. Risk factors for mortality were found as age younger than 2.5 years and weight less than 10.7 kg. In accordance with these studies, we also observed that younger age at diagnosis (p = 0.012), surgery before age three years (p = 0.037), and pulmonary atresia (p = 0.000014) were significantly associated with mortality. This may be explained by the fact that patients who have mortality have more severe form of disease, become symptomatic earlier, and therefore diagnosed at a younger age. Our median mortality age was 25 days. Higher mortality rates with fetal and neonatal presentations were also reported in other studies. ^{Reference Celermajer, Cullen, Sullivan, Spiegelhalter, Wyse and Deanfield4,Reference Geerdink, Delhaas and Helbing8,Reference Shinkawa, Polimenakos and Gomez-Fifer10–Reference Oxenius, Attenhofer Jost and Prêtre11} . Gender, presence and degree of tricuspid regurgitation, and arrhythmias also did not affect mortality in our study.

Attie et al. analysed risk factors for mortality in 174 Ebstein’s anomaly patients (148 operated and 26 non-operated) with a follow-up from 6 months to 25.3 years. ^{Reference Attie, Casanova and Zabal20} Functional class III or IV and cardiothoracic ratio ≥65% with either cyanosis or arrhythmias were found to be associated with mortality. However, the mortality in patients who had only one of these variables was lower. We also did not observe any significant difference between alive and demised patients in terms of arrhythmia presence and arrhythmia types. We did not analyse functional class, cardiothoracic ratio, cyanosis, and combination of variables as risk factors for mortality in our study.

Geerdink et al. analysed factors associated with death during childhood in a cohort of 176 patients diagnosed with Ebstein’s anomaly. ^{Reference Geerdink, Delhaas and Helbing8} Thirty-one patients (18%) died before the age of 18 years. The authors also observed that the median age at death was less than one month (range, 1 day–49 months). Presentation with Modified Ross Heart Failure Class IV was the most important univariate factor associated with death and others included neonatal diagnosis, severe tricuspid valve regurgitation, severe right ventricular outflow tract obstruction, and patent ductus arteriosus at presentation. Multivariable analysis showed that presentation with Modified Ross Heart Failure Class IV and a ventricular septal defect as additional comorbidity were the strongest predictors of death in childhood and adolescence. The authors included patients with pulmonary atresia in the severe right ventricular outflow tract obstruction group. We also observed that pulmonary atresia was strongly associated with death in our cohort. We did not analyse Modified Ross Heart Failure class and also did not find ventricular septal defect as an additional risk factor. These differences may be because our cohort consisted of a mixed population of children and young adults. Patients diagnosed in young adulthood might have different outcomes from those patients diagnosed during the neonatal period or later in childhood.

Most commonly associated cardiac anomalies reported in Ebstein’s anomaly are atrial septal defect or patent foramen ovale (80–94%). ^{Reference Cetta, Dearani, O’Leary, Shaddy, Penny, Feltes, Cetta and Mital3,Reference Ramcharan, Goff, Greenleaf, Shebani, Salazar and Corno6,Reference Oxenius, Attenhofer Jost and Prêtre11} In Oxenius et al.’s study, atrial septal defect/patent foramen ovale (79%), pulmonary stenosis/atresia (17%), left ventricular noncompaction (12%), and ventricular septal defect (7%) were the most common associated anomalies. ^{Reference Oxenius, Attenhofer Jost and Prêtre11} In our study, atrial septal defect/patent foramen ovale (57%), mitral regurgitation (25.3%), pulmonary stenosis/atresia (17.7%), and ventricular septal defect (16.5%) were also the most common anomalies. Attenhofer et al. stated that Ebstein anomaly should not be regarded as a disease confined to the right side of the heart. ^{Reference Attenhofer Jost, Connolly, O'Leary, Warnes, Tajik and Seward21} They observed left heart abnormalities involving the myocardium or valves in 39% of their patients; 17.9% had left ventricular myocardial changes resembling noncompaction, left ventricular dysfunction was present in 43%, ventricular septal defect in 8%, bicuspid aortic valve in 8%, mitral valve prolapsus in 15%, and mitral valve dysplasia in 4%. We also observed left heart lesions in 58% of our patients, including ventricular septal defect (16.5%), mitral regurgitation (25.3%), mitral valve prolapsus (8.9%), aortic coarctation (3.8%), bicuspid aortic valve (2.5%), and left ventricular noncompaction in one patient.

In Ebstein’s anomaly, chromosomal or Mendelian disorders are detectable in about 20% of the patients, and in 80%, it is an isolated non-syndromic malformation. ^{Reference Digilio, Silvestri, Dallapiccola, Marino, Giamberti and Chessa22} Deletion 1p36, deletion 8p23.1, and terminal deletion 18q are the most frequently reported chromosomal disorders associated with Ebstein’s anomaly. In our series, chromosomal and Mendelian disorders were present in three patients. We also had one patient with deletion 1p36. Apert, Noonan, CHARGE, Holt-Oram, Cornelia de Lange, Kabuki syndromes, and VACTERL association have also been reported monogenic syndromes. ^{Reference Digilio, Silvestri, Dallapiccola, Marino, Giamberti and Chessa22} Congenital anomalies such as low-set ears, micrognathia, cleft lip and/or palate, renal agenesis, megacolon, undescended testis, and bilateral inguinal hernia have also been reported. We observed cleft lip and/or palate in two patients and congenital scoliosis in one patient. In our observation, presence of associated genetic diseases or congenital anomalies did not significantly affect mortality.

Attenhofer et al. compared the use of echocardiography and cardiac MRI in Ebstein’s anomaly patients and concluded that both methods provided complementary data and should be performed together before surgery. ^{Reference Attenhofer Jost, Edmister and Julsrud23} We also used cardiac MRI as a complementary imaging method before surgery.

Accessory pathways are common and present in approximately one-third of the Ebstein’s anomaly patients and may be even multiple. ^{Reference Cetta, Dearani, O’Leary, Shaddy, Penny, Feltes, Cetta and Mital3,Reference Ramcharan, Goff, Greenleaf, Shebani, Salazar and Corno6,Reference Geerdink and Kapusta7,Reference Reich, Alud and Hulse12–Reference Karagöz, Ertuğrul and Aypar14} The prevalence of arrhythmias in children and adolescents with Ebstein’s anomaly was reported lower than in adults. ^{Reference Cetta, Dearani, O’Leary, Shaddy, Penny, Feltes, Cetta and Mital3,Reference Ramcharan, Goff, Greenleaf, Shebani, Salazar and Corno6,Reference Delhaas, du Sarvaas and Rijlaarsdam13} In younger age, atrioventricular re-entrant tachycardia related to accessory pathways or atrioventricular nodal re-entrant tachycardia is common. Later in life, atrial tachycardia, atrial flutter, intra-atrial re-entrant tachycardia, and atrial fibrillation are more common, due to a dilated right atrium and postoperative atrial incisions. ^{Reference Cetta, Dearani, O’Leary, Shaddy, Penny, Feltes, Cetta and Mital3} Studies about arrhythmias and ablation in paediatric Ebstein’s anomaly patients are limited. ^{Reference Reich, Alud and Hulse12–Reference Karagöz, Ertuğrul and Aypar14} Delhaas et al. reported prevalence of arrhythmias as 17% in 93 children with Ebstein’s anomaly, and most were supraventricular tachycardia with a preexcitation prevalence of 15%. ^{Reference Delhaas, du Sarvaas and Rijlaarsdam13} In another study from our centre, in which 89 paediatric Ebstein’s anomaly patients were evaluated, arrhythmias were present in 30.3, 31% required ablation, and most were due to accessory pathways (82%). ^{Reference Karagöz, Ertuğrul and Aypar14} We observed a high incidence of arrhythmias (30.4%) in our study. This finding may be because some of our patients had been referred for ablation, as our centre is a tertiary referral centre. Another reason may be the use of electrophysiological studies in all of our patients with symptomatic tachycardia or documented arrhythmia (Wolff–Parkinson–White pattern or sustained tachycardia) in electrocardiograms or in 24-hour electrocardiogram recordings. Ventricular tachycardia has been reported to be rare and can more commonly be induced during electrophysiological studies. ^{Reference Delhaas, du Sarvaas and Rijlaarsdam13,Reference Karagöz, Ertuğrul and Aypar14} However, following tricuspid valve surgery, there is a higher incidence of ventricular arrhythmias, related to ventriculotomy incisions. ^{Reference Cetta, Dearani, O’Leary, Shaddy, Penny, Feltes, Cetta and Mital3} We had only one patient with ventricular tachycardia diagnosed during catheterisation in the pre-operative period.

It has been reported that although acute success rates for catheter ablation of accessory pathways in Ebstein’s anomaly are high, recurrence rates are also high. ^{Reference El-Assaad, DeWitt and Mah24} In a multicentre study, catheter ablation was successful in 8/9 patients. ^{Reference Delhaas, du Sarvaas and Rijlaarsdam13} In another study, in which 65 paediatric Ebstein’s anomaly patients with 82 accessory pathways (median age: 9.8 ± 5.4 years [4 months–20 years]) were evaluated, acute success rates and recurrence rates for right free wall, right septal, and other mechanisms were 79/32%, 89/29%, and 75/27%, respectively. ^{Reference Reich, Alud and Hulse12} Our acute success rate was higher and recurrence rate was lower compared to this study, the differences may be due to use of three-dimensional imaging techniques during some of our ablation procedures and added procedural experience at our centre. Successful results with surgical ablation in accessory pathway-related tachycardias without recurrence have also been reported. ^{Reference Stulak, Sharma, Cannon, Ammash, Schaff and Dearani25}

Surgical results of Ebstein’s anomaly patients have been reported both in children and adults. ^{Reference Boston, Dearani, O'Leary, Driscoll and Danielson15–Reference Holst, Dearani and Said17,Reference Stulak, Sharma, Cannon, Ammash, Schaff and Dearani25–Reference Davies, Pasquali and Jacobs27} In a multicentre study, in which 63 children with Ebstein’s anomaly with 109 operations were evaluated, biventricular repair was performed in 59%, one-and-half ventricular repair in 5.8%, and univentricular repair in 33%. The median age at first operation was 26 months (range: 0–203 months) and median follow-up period was 121 months. Second surgery was required in 42.8% of the patients. ^{Reference Geerdink, du Sarvaas and Kuipers16} The most common post-operative complication was arrhythmia (21%). Perioperative mortality rate was 14%. In our series, we mostly performed biventricular repair (44%) and palliative surgery (48%). These differences in surgery types may be due to fact that surgery type is primarily determined by the tricuspid anatomy in Ebstein’s anomaly and also by the surgeon and the institution experience. Our lower rate of second surgical interventions compared to this study may be because we only performed 32 operations in 25 patients with a higher median age at first surgery (6 years) and our median follow-up period was shorter. Our post-operative arrhythmia (19.2%) and perioperative mortality (12.1%) rates were similar.

In a study in which 498 Ebstein’s anomaly patients with a total of 595 operations from different centres were analysed: 116 were neonates (19%), 122 were infants (21%), 264 were children (44%), and 93 were adults (16%). ^{Reference Davies, Pasquali and Jacobs27} Neonates mostly had palliative procedures: systemic-to-pulmonary artery shunts, with or without tricuspid valve closure (37.1%), tricuspid valve closure (8.6%), and Ebstein’s repair/tricuspid valvuloplasty (32%). Infants mostly had superior cavopulmonary connections (50.8%) and systemic-to-pulmonary artery shunts (8.2%). In older patients, tricuspid valve surgery (children, 54.5%; adults, 68.8%), arrhythmia procedures (children, 8.7%; adults, 17.3%), and Fontan surgery (children, 14.8%) were applied. Overall mortality rate was 5.9%. Highest in-hospital mortality was in neonatal period (23.4%). We also performed mostly palliative procedures in neonates and infants, and biventricular repair in older children and young adults. We also observed highest in-hospital mortality in the neonatal period.