Author + information
- Received October 29, 2018
- Revision received January 16, 2019
- Accepted January 31, 2019
- Published online May 20, 2019.
- Chrystalle Katte Carreon, MDa,
- Stephen P. Sanders, MDa,b,c,d,
- Antonio R. Perez-Atayde, MD, PhDa,
- Pedro J. del Nido, MDd,e,
- Edward P. Walsh, MDb,c,f,g,
- Tal Geva, MDb,c,g and
- Mark E. Alexander, MDb,c,f,g,∗ ()
- aDepartment of Pathology, Boston Children’s Hospital and Harvard Medical School, Boston, Massachusetts
- bDepartment of Cardiology, Boston Children’s Hospital, Boston, Massachusetts
- cDepartment of Pediatrics, Harvard Medical School, Boston, Massachusetts
- dDepartment of Cardiac Surgery, Boston Children’s Hospital, Boston, Massachusetts
- eDepartment of Surgery, Harvard Medical School, Boston, Massachusetts
- fArrhythmia Service, Boston Children’s Hospital, Boston, Massachusetts
- gDepartment of Pediatrics, Harvard Medical School, Boston, Massachusetts
- ↵∗Address for correspondence:
Dr. Mark E. Alexander, Department of Cardiology, Boston Children’s Hospital, 300 Longwood Avenue, Boston, Massachusetts 02115.
Objectives This study sought to evaluate for the presence of and characterize the interdigitating and entrapped myocardium within cardiac fibromas (CF) and correlate tissue findings with symptoms and surgical outcomes.
Background The mechanism of ventricular tachycardia (VT) in CF is unclear. The authors hypothesized that CF harbor tongues of interdigitating myocardium, which could be a substrate for episodes of arrhythmia analogous to peri-infarct zones.
Methods A total of 29 patients (14 boys) with CF were identified; all subjects had undergone at least partial tumor resection. A semiquantitative grading system was used to assess the degree of myocardial interdigitation and entrapment, myocyte morphology (hematoxylin and eosin stain and immunohistochemical stain for desmin), tumor collagen density, and cellularity (trichrome stain). The subjects’ ages at presentation, types of arrhythmia, and responses to surgery were correlated with histology.
Results CF consistently demonstrated interdigitating and entrapped myocardium, and the extent correlated negatively with age at surgery, as did cellularity, whereas collagen increased with age. Median age at arrhythmia recognition was 8 months. Sustained VT was present in 18 of 29 patients (62%), and 5 of 6 patients with prenatally diagnosed conditions developed VT before 8 months. All 8 patients who experienced cardiac arrest had clinically significant arrhythmia events. Sustained arrhythmia episodes correlated with more diffuse myocyte interdigitation. Ten patients had abnormal karyotype: chromosomes 9 (n = 3) and 3 (n = 1) deletions; isolated translocations: t(4;13), t(5;11) and t(1;9); and undefined aberrations (n = 3). All patients who underwent complete resection were cured of arrhythmias, whereas 2 of 14 patients who had subtotal resections had recurrence, with resolution following re-resection in 1 patient.
Conclusions Interdigitating myocardium represents a potential histopathologic substrate for VT and cardiac arrest in CF, which may also explain the occasional recurrence of arrhythmia following incomplete resection.
Although cardiac fibromas (CF) are histopathologically benign, they are associated with clinically significant arrhythmia episodes, particularly ventricular tachycardia (VT), in up to 64% of patients (1–5). Tumor excision in these patients may be curative and result in relatively good preservation of ventricular function in the immediate post-operative period and beyond (1,6).
The mechanism of VT in CF is not well understood, although a re-entry mechanism is implied because episodes are usually monomorphic and regular in rate, and electrical cardioversion is often effective in terminating the tachycardia (1). The successful elimination of arrhythmia following total or subtotal resection also supports the idea that these tumors contain arrhythmia substrates that permit re-entry (1,6). The authors consistently observed the absence of a distinct capsule separating the tumor from the adjacent myocardium and frequent intermingling of these 2 components at the interface in surgical specimens (7–9). Given the high frequency of sustained monomorphic VT in CF patients and the lack of a well-formed capsule in these tumors, it was hypothesized that interdigitating myocardium within the CF could be the histologic substrate and may promote re-entry ventricular arrhythmias in a manner analogous to what had been described in a peri-infarct zone.
The Heart Center database was searched from 1968 to 2018 to identify all patients with the diagnosis of CF who had undergone partial or complete resection of the tumor. Patients were excluded if tumor tissue or complete clinical data were not available. Our retrospective record review was approved by the Boston Children’s Hospital Institutional Review Board (IRB-P00026995).
The following data were collected from all patients: demographics, symptoms, rhythm abnormalities from electrophysiology studies, cardiac imaging (echocardiography and cardiac magnetic resonance [CMR]) findings, intraoperative findings from procedure notes, and gross and microscopic pathology findings, as well as tumor cytogenetics when available. All hematoxylin and eosin-stained sections of tumors were retrieved and analyzed microscopically. Representative sections were selected for additional immunohistochemical staining for desmin, to highlight myocardium, and special staining with Masson’s trichrome to emphasize fibrosis.
For the purpose of this study, interdigitating myocardium was defined as tongues of myocardium extending from surrounding myocardium into the edge of the tumor and entrapped myocardium as cardiomyocytes completely surrounded by the fibrous tumor. The extent of interdigitating/entrapped myocardium was graded semiquantitatively on a 4-point scale, using representative sections immunostained for desmin on the basis of the proportion of a cross-sectional area showing interdigitation/entrapment and its distribution or depth into the sections, where: 1) minimal = ≤5% of cross-sectional area of the section, mostly adjacent to the periphery or, rarely, isolated deeper foci; 2) mild = >5% but <30% of cross-sectional area of the section, which may involve deeper location but limited to a few foci; 3) moderate = ≥30% but <75% of cross-sectional area of the section, which may involve multiple deeper foci; and 4) marked = ≥75% of cross-sectional area of the section, usually diffusely distributed throughout the section.
Tumor collagen density was graded semiquantitatively by using a 3-point scale on 10 representative medium-power fields, using the following schemes: 1+ = tumor showing predominantly thin collagen fibers; 2+ = tumor showing thin and thick collagen fibers of relatively equal proportion, admixed or zonal in distribution; and 3+ = tumor showing predominantly thick collagen fibers, usually dense blue in color in Masson’s trichrome-stained sections.
Tumor cellularity was likewise graded semiquantitatively on a 3-point scale on 10 representative high-power fields by using the following schemes: 1+ = relatively few and widely spaced tumor nuclei; 2+ = mostly uniformly spaced tumor nuclei with less crowding than in 3+; and 3+ = tumor nuclei crowding with back-to-back and sometimes overlapping nuclei.
Whether the interdigitating cardiomyocytes appeared compressed or in thicker bundles was also noted; and other distinctive histologic features were noted.
Arrhythmia episodes were classified using standard clinical classifications such as sustained monomorphic VT (SMMVT) lasting >30 s and sustained polymorphic VT (SPMVT), which included cardiac arrest requiring cardiopulmonary resuscitation and defibrillation; as well as hospital notes documenting SPMVT and ventricular ectopy, which included isolated beats or nonsustained VT, or no arrhythmia. Recurrences were similarly classified with all cases of recurrent SMMVT having unambiguous electrocardiographic confirmation.
Information about surgical management was retrieved from the operative notes and included the type of surgical procedure, history of prior biopsy or resection, completeness of tumor excision and reason for subtotal resection, accompanying valve repair, and subsequent re-excision. Clinical features, types of arrhythmia, and responses to surgery were correlated with the tissue findings. Descriptive and exploratory statistics were calculated using Stata software v. 15 (StataCorp, College Station, Texas).
Twenty-nine patients (≤21 years of age) met selection criteria. One patient was excluded because of a lack of detailed clinical data. Portions of this cohort were participants in the present authors’ recent experience (1,6) and were included in the UK experience (10). The cohort consisted of 14 boys (48%) and 15 girls (52%). The median age at diagnosis was 5 months (range prenatal to 120 months). Cardiac fibroma was diagnosed in 15 patients (52%) within the first 6 months of life and diagnosed prenatally in 6 patients (21%). Four patients (14%) carried a genetic diagnosis of Gorlin syndrome, also known as nevoid basal cell carcinoma syndrome. In this subset, odontogenic keratocyst was a consistent additional feature; 1 patient also developed ovarian fibroma. Although most patients were from the United States, 7 (24%) were from other countries. A tabular summary is presented in Table 1.
The patients presented with VT (n = 7 [24%]), cardiac arrest (n = 6 [21%]), tachypnea and other respiratory complaints (n = 4 [14%]), murmur (n = 3 [10%]), and pre-syncope without documented arrhythmia (n = 1 [3%]). Conditions in the remainder (n = 8 [28%]) were incidentally diagnosed (including prenatal diagnosis) or lacked data from referral providers. A total of 8 patients (28%) had an occurrence of cardiac arrest at some point prior to surgery, 6 as the presenting event and 2 following initial diagnosis (1 had a murmur and the other had no symptoms at the time of diagnosis). Seven of the 8 patients who experienced cardiac arrest were <1 year of age at the time of the arrest episode (median age: 5.5 months; range 22 days to 34 months).
Overall, 15 patients had SMMVT at some point, and 6 of those also had cardiac arrest and SPMVT. Nonsustained ventricular arrhythmia episodes were seen in 10 patients, with only 1 patient having no documented arrhythmia.
The median age at which arrhythmia was recognized was 8 months (range 1 day to 10 years). Sustained VT was present in 18 of 29 patients (62%). Of these, 15 patients had SMMVT, 2 had SPMVT, and 1 had both. Furthermore, 5 of the 6 patients who were diagnosed prenatally subsequently developed VT before 8 months of age, and 1 at 60 months of age. All 8 patients who experienced cardiac arrest had documented clinically significant arrhythmias (SMMVT in 6, only SPMVT in 2, and both in 4 patients).
Complete excision of the tumor was performed in 15 patients (52%), whereas the remainder (n = 14 [48%]) underwent subtotal resection. One patient had a biopsy prior to excision, and 1 patient had a repeat resection following subtotal resection. Proximity to crucial structures, particularly the left anterior descending and circumflex coronary arteries, as well as atrioventricular valves, was the most common reason for incomplete tumor resection.
No patient who underwent complete excision of the CF had recurrence of tachycardia (median follow-up interval: 2 years; range 1 week to 14 years). Two of 14 patients (14%) who underwent subtotal resection had recurrence of arrhythmia at 13 months and 7 months after surgery (20 months of age and 3.6 years of age, respectively). One patient underwent repeat tumor resection with no further recurrence at 6.5 years following the second resection of tumor. However, this patient underwent mitral valve surgery. In the second patient with recurrence, a sustained arrhythmia could not be induced during a subsequent electrophysiology study, and a catheter ablation was performed based on pace mapping to match the clinical VT. There was no recurrence of the VT during 2.5 years of additional documented follow-up.
Pathology and tumor cytogenetics
The CF most commonly involved the left ventricular free wall (n = 26 [90%]; 4 of which also extended to the interventricular septum), followed by the right ventricular free wall (n = 2 [7%]; 1 also involved the interventricular septum) and the interventricular septum in isolation (n = 1 [3%]). The mean tumor size was 5 cm (range 3 to 10 cm). All tumors exhibited a whitish-tan, solid, and often whorled and trabeculated cut surface. Thin patches of scanty myocardium were noted along the periphery, which can be seen extending toward the center of the tumor in rare specimens (Figures 1A and 1B). In addition, minute (<0.1 cm) pinkish-tan foci were sometimes noted within the tumor, which in retrospect, represented entrapped or interdigitated strands of myocardium.
Microscopically, all tumors were typical cardiac fibromas characterized by a homogeneous spindle cell proliferation of bland fibroblasts. No nuclear pleomorphism or significant number of mitoses were identified. Interdigitating and entrapped myocardium was uniformly present to various extents and uniformly present and highlighted by an immunostain for desmin in the representative tumor sections from all patients (minimal: 6 [21%]; mild: 8 [28%]; moderate: 4 [14%]; marked: 11 [38%]) (Figure 2). The interdigitating myocytes were either compressed (n = 16 [55%]), in thicker or less-compressed bundles (n = 2 [7%]), or a combination of both (n = 11 [38%]) (Figure 3). Similarly, variations in tumor cellularity (1+: 4 [14%]; 2+: 16 [55%]; 3+: 9 [31%]) and collagen density (1+: 4 [14%]; 2+: 12 [41%]; 3+: 13 [45%]) were noted (Figure 4). Approximately one-third of these tumors showed calcifications and occasional foci of chronic inflammation and necrosis. The 4 specimens with tumor necrosis were all 5 cm or greater in size, possibly secondary to age or tumor size-related changes in vascular supply. Ossification, extramedullary hematopoiesis, and myxoid stroma were rarely seen (Figure 5). No tumor had a well-defined capsule on histology, and the tumor-myocardium interface was poorly delineated. They exhibited a complex architecture consisting of multinodular to plexiform architecture with tongues of interdigitating myocardium extending into and islands of entrapped myocardium scattered in between the fibrous nodular network in at least 80% of the cases (Figure 6). Perimyocyte fibrosis was present in some areas peripheral to the tumor-myocardium junction. It is important to note that the histology observed reflects the tumor at the time of surgery and not at the time of presentation.
Tumor cytogenetics were available for 14 patients, and those for 10 patients were abnormal, revealing deletions in chromosomes 9 (n = 3) and 3 (n = 1); isolated translocations (n = 3): t(4;13), t(5;11), and t(1;9); and undefined aberrations (n = 3) (Table 2). Interestingly, none of these patients had Gorlin syndrome.
More extensive myocardial entrapment (p = 0.023) and increased tumor cellularity and lower tumor collagen scores (p = 0.001) were observed with younger age at the time of surgery (Figure 7). Additionally, a majority of the patients (5 of 8 [62%]) who experienced cardiac arrest had moderate or marked myocardial interdigitation/entrapment. Neither age at surgery, age at presentation, nor arrhythmia type was associated with tumor size. The other histological features (calcification, chronic inflammation, ossification, extramedullary hematopoiesis, and myxoid stroma) showed no specific patterns or trends with age. There were no distinct histological features noted in the subset of tumors from patients known to have Gorlin syndrome.
Although the absence of a well-defined capsule is a common observation in CF (7–9), the presence of interdigitating and entrapped myocardium within these tumors has never been fully explored, along with their potential role as a substrate for ventricular arrhythmias. The incidental finding of “completely entrapped working cardiac muscle cells” with adjacent nerve bundles toward the peripheral region of the tumor had been reported by István et al. (5). Entrapped islands of myocardium have also been described by a few investigators and were assumed to be the result of infiltrating growth of the irregular advancing edge of the tumor (11,12). The present authors suggest that interdigitating tongues of myocardium is the most likely substrate for re-entrant arrhythmias.
Conceptually, a fibroma could serve as an inert central obstacle with compression of myocytes around the periphery, resulting in areas of slow conduction and permitting re-entry. The normal surrounding myocardium and lack of a discrete fibrous capsule speak against that hypothesis. Rather, the interlacing tongues of irregular myocardium with variable density and orientation of collagen and elastic fibers and, rarely, chronic inflammatory cells and dystrophic calcifications mimic the histology demonstrated in peri-infarct zones and seem more likely to be the arrhythmia substrate (13,14). These authors hypothesize that a delay in conduction results from the irregular architecture consisting of alternating areas of nonconducting fibrous tumor with adjacent viable but often morphologically, and likely functionally, altered cardiomyocytes, which altogether promotes re-entry (Central Illustration).
The tumor tissue changes observed with older age at the time of surgery opens a new perspective on this neoplasm and suggests a dynamic process within the tumor as the child grows. This could have important clinical implications. The temporal evolution of collagen density and cellularity observed in the present study is similar to that described by Burke et al. (8) in a cohort of adult and pediatric subjects and to that observed by Miller et al. (15) in a comparison of tumor sections from a 4-week-old, a 7-year-old, and a 27-year-old patient. The mechanism of this gradual “disappearance” of interdigitating/entrapped myocardium with age is unclear. Although fibromas can present at any age (even in the sixth decade) (7), the temporal changes in the tumor that were noted in the present study and by other investigators could explain why most fibromas present in childhood, with one-third presenting in the first year.
All episodes of cardiac arrest in this series occurred in the first 3 years of life, with all except 1 in the first year of life. Moreover, in the 8 patients who experienced cardiac arrest, marked myocardial entrapment was present in 5 patients at the time of surgery. The patient who experienced cardiac arrest with the least amount (minimal) of interdigitating/entrapped myocardium at the time of surgery was initially medically managed and only underwent excision >4 years following the episode of arrest, the longest arrest-to-surgery interval of all 8 patients. Perhaps, if the tumor had been resected closer to the time of arrest, more interdigitating/entrapped myocardium would have been observed. Alternatively, combinations of later presentation or lack of serious arrhythmia may be a marker for a less cellular tumor. More importantly, it suggests that there is a “vulnerable” time period wherein CF patients are more likely to develop a fatal arrhythmia. This period of highest risk appears to be within the first year of life. However, given the small numbers and the referral biases, this observation is not conclusive.
Implantable cardioverter-defibrillators (ICDs) have been reported as a management strategy (16) that the present authors have not adopted. Epicardial ICDs in infants are associated with shorter system longevity and increased incidence of inappropriate therapy compared with ICDs in adult patients (17,18). A sternotomy is required for implantation, and the ICD limits use of CMR. Although there may be cases in which this approach is reasonable, these authors have not adopted it.
The multinodular architecture of CF, more obvious toward the periphery, suggests a multifocal origin of the tumor arising from proliferation of scattered resident stromal fibroblasts/myofibroblasts interspersed between cardiomyocytes from multiple foci, with progressive nodular growth and subsequent formation of a larger and more compact solid tumor mass. The prominent multilobulation observed in some of these tumors is also consistent with this theory. Myocardial interdigitation and entrapment is due to direct infiltrative growth of the tumor from multiple foci dispersed within the myocardium (11,12). The histologic resemblance between CF and desmoid-type fibromatosis and other soft tissue spindle cell proliferations has been previously noted (15,19–21) with some authors suggesting that CF is a cardiac manifestation of or involvement by desmoid fibromatosis (20,22), although Miller et al. (15) found no evidence for abnormal WNT-β-catenin signaling. In fact, the behavior of CF differs from the more aggressive growth behavior of desmoid fibromatosis, which is highly associated with familial adenomatous polyposis syndrome.
The genetics of CF are poorly understood, and only a few published data are available, limited to 3 case reports (23–25) and 1 small series (26). The PTCH1 gene, which is involved in Gorlin syndrome, has been localized to chromosome 9q22.3 (27,28). In the present cohort, tumor cytogenetics revealed isolated deletions and translocations and rare undefined aberrations (Table 2). Prior karyotypic analysis (23,24,26) has shown translocations and deletions on this region of chromosome 9. Fluorescence in situ hybridization analysis has shown homozygous and heterozygous loss of the PTCH1 locus (26), and high-resolution microarray analysis showed losses of somatic copy numbers on chromosome 9q21.33 to q22.33 encompassing the PTCH1 locus as well (25). One of the patients in the present cohort also showed translocation between chromosomes 1 and 9, and 3 patients showed deletions in chromosome 9, similar to the previously reported cases (23,24,26). Given that none of the previously reported cases, nor the patients in the present cohort with these genetic findings, had Gorlin syndrome, one might postulate that this region in chromosome 9 is critical for tumor development regardless of syndromic association. To the best of the present authors’ knowledge, the remainder of the present karyotypic findings have not been reported.
The objective of surgery in CF is to prevent life-threatening arrhythmias and ameliorate hemodynamic abnormalities. In the present cohort, none of the patients experienced clear hemodynamic consequences of these large tumors. Rather, for most patients, the motivation for intervention has been to eliminate the risk of cardiac arrest. The surgical results reported here and elsewhere support this treatment strategy. Complete resection resulted in elimination of the arrhythmia in both the present expanded series and the smaller numbers reported by other groups (29). These results support the present authors’ hypothesis that the critical zone of slow conduction that allows re-entry is within the tumor itself. The associated observation that only 2 of 17 patients with incomplete resection had recurrent tachycardia implies that the number of critical zones for arrhythmia generation within the tumor is limited. There has been no systematic attempt to perform electrophysiology studies in these patients, largely because of patient size and relatively clear pre-operative data regarding the arrhythmia.
Three types of arrhythmia have been observed: SMMVT, which typically has required cardioversion and was seen in 52% of our cohort, is the most clearly re-entrant arrhythmia and the arrhythmia where these findings are most relevant. Nonsustained VT is seen in both the SMMVT group and in isolation (29). This and isolated ectopy may have a mechanical component independent of re-entry. The present authors speculate that, in those with SMMVT and cardiac arrest, the combination of VT and hemodynamic instability can lead to arrest and polymorphic VT. However, there were 2 patients who had had cardiac arrest and no documented SMMVT. Those patients may have had an alternative physiology that led to arrest or infrequent episodes of SMMVT not captured on pre-operative monitoring.
The findings of the current study reinforce and expand previous experience (1,6) that tumor resection is indeed an effective way of managing ventricular arrhythmias in CF patients. The present data do not support the presence of a capsule surrounding these tumors as reported by Jones et al. (10). Rather, the detailed histology presented here supports attributing success of surgery to resection of most if not all of the interdigitating myocardium, the key element in the re-entry circuit. This overall favorable response to surgery has also been reported by other institutions (10,29–31).
The present cohort could have had an over-representation of more symptomatic CF patients given that the authors’ institution is a large tertiary and international referral center. This has also resulted in limited follow-up data for some patients, typical of an international referral cohort. As a retrospective analysis of pathologic specimens, the histology reviewed, particularly the immunohistochemistry, could only be performed on representative sections of the tumor. Although most of these patients had CMR evaluation, this study did not attempt to correlate the histology with CMR findings.
The present data support the hypothesis that interdigitating myocardium is a feasible histopathologic substrate for a re-entry mechanism of VT in patients with CF. The universal resolution of VT with complete resection and the high frequency of VT resolution with incomplete resection support the presence of critical zones of slow conduction, which permit re-entry, within the tumor. Furthermore, the age-related changes in the histology of CF suggest a dynamic nature of the tumor architecture. The mechanism of these changes is not defined. The histologic features particularly toward the tumor periphery together with the presence of interdigitating and entrapped myocardium suggest a multifocal origin, more similar to a hamartomatous process coupled with an infiltrative growth pattern, rather than an expansile growth from a single nidus.
Although the overall size of CF did not appear to change over time, the trend toward decreased interdigitating/entrapped myocardium and tumor cellularity and increased collagen density in older patients suggests dynamic changes within the tumor in growing children, which may influence therapeutic decisions for these patients. The present cohort also revealed that CF harbor complex cytogenetics, which offers insight into a possibly larger role for the PTCH1 gene locus in the development of cardiac fibroma, independent of the patient’s syndromic status.
COMPETENCY IN MEDICAL KNOWLEDGE 1: Cardiac fibromas harbor tongues of interdigitating and entrapped myocardium, which is a potential histopathologic substrate for re-entry arrhythmias, and the favorable response to surgery is likely secondary to the elimination of the arrhythmia substrate.
COMPETENCY IN MEDICAL KNOWLEDGE 2: New, additional cytogenetic abnormalities discovered from the present cohort add to the limited existing information about the genetics of cardiac fibroma in the studies. The identification of interdigitating/entrapped myocardium in cardiac fibromas provides a plausible explanation for the success of surgical resections in the management of these patients. Furthermore, the identification of what seems to be a “vulnerable time frame” within which young patients are more likely to develop significant arrhythmias leading to cardiac arrest may aid in treatment decision making, particularly for surgically resectable tumors.
TRANSLATIONAL OUTLOOK 1: Further investigation of the utility of CMR in determining the degree of interdigitating/entrapped myocardium may help in planning the appropriate treatment method and may facilitate timely surgical resection of tumor.
TRANSLATIONAL OUTLOOK 2: Given that nonsyndromic patients exhibited aberrations involving chromosome 9, where the PTCH1 gene microdeletion, implicated in Gorlin syndrome, is mapped, the likelihood of chromosome 9 being critical for tumor development regardless of syndromic association warrants further investigation.
Supported by the Sean Roy Johnson Fund, Boston Children’s Hospital. Drs. Alexander and Walsh have received royalties from UpToDate. Dr. del Nido is the founder of Nido Surgical, Inc.; and a consultant for Gore, Inc. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- cardiac fibroma
- sustained monomorphic ventricular tachycardia
- sustained polymorphic ventricular tachycardia
- ventricular tachycardia
- Received October 29, 2018.
- Revision received January 16, 2019.
- Accepted January 31, 2019.
- 2019 American College of Cardiology Foundation
- Miyake C.Y.,
- Del Nido P.J.,
- Alexander M.E.,
- et al.
- Bryant V.A.,
- Booth J.,
- Palm L.,
- Ashworth M.,
- Jacques T.S.,
- Sebire N.J.
- Nathan M.,
- Fabozzo A.,
- Geva T.,
- Walsh E.,
- del Nido P.J.
- Burke A.,
- Tavora F.,
- Maleszewski J.,
- Frazier A.
- Travis W.D.,
- Brambilla E.,
- Burke A.P.,
- Marx A.,
- Nicholson A.G.
- Jones J.P.,
- Ramcharan T.,
- Chaudhari M.,
- et al.
- De Bakker J.M.,
- Janes M.J.
- Marchlinski F.E.,
- Waxman H.L.,
- Buxton A.E.,
- et al.
- Park J.K.,
- Pollock M.E.
- Turi G.K.,
- Albala A.,
- Fenoglio J.J.
- Zhang Q.,
- Wang T.,
- Wang D.,
- et al.
- Farndon Peter A.,
- Del Mastro Richard G.,
- Gareth D.,
- et al.