1. INTRODUCTION

Arrhythmogenic right ventricular cardiomyopathy (ARVC) is an inherited cardiomyopathy that mainly affects the right ventricle (RV) and is characterized by the loss of myocardium and replacement with fibrous and fatty tissue.1,2 It is estimated to affect 0.05% to 0.1% of the general population, but referrals to cardiac magnetic resonance (CMR) imaging centers are high due to the potential for sudden cardiac death (SCD), especially in young athletes.1,3 Symptoms typically include palpitations or syncope caused by ventricular arrhythmias.1,4 Correctly distinguishing ARVC from idiopathic right ventricular outflow tract ventricular arrhythmia (RVOT-VA), the most common cause of ventricular tachycardia in young people without obvious structural heart disease, may be crucial due to differences in management and prognosis.4,5

Diagnosing ARVC is still challenging because no individual approach possesses sufficient specificity to confirm it.6 It is currently based on the revised 2010 task force criteria (TFC), which includes image evaluation, electrocardiographic studies, histological evidence, and family history.1 CMR is the imaging modality of choice because it allows a comprehensive investigation of RV structural and functional abnormalities. The current CMR criteria include RV volume and ejection fraction (EF) quantification and RV regional wall motion abnormalities assessment.3,7 Fatty infiltration of the RV on CMR was deemed to lack specificity and reproducibility due to the thinness of the RV wall and the presence of fat in healthy subjects; hence, it is not included in the TFC.8–10 Late gadolinium enhancement (LGE) was also not incorporated into the TFC due to detection difficulty in the thin-walled RV and challenging differentiation between fibrosis and fat.6,11,12 Prominent focal scalloping of the RVOT or the subtricuspid region of the RV free wall during systole was another CMR feature observed in ARVC patients and described as the “accordion sign.”3,13

Morphological features combined with wall motion abnormalities of the RV can yield high sensitivity and specificity for the diagnosis of ARVC.8,9 On the other hand, the analysis of regional wall motion abnormalities may be limited by the weakness of subjectivity due to RV contraction pattern complexity.7 In recent years, myocardial strain has emerged as a less observer-dependent and quantitative method of evaluating myocardial contraction.14 Previous studies demonstrated the utility of RV strain analysis for assessing patients with suspected ARVC.14,15 Nevertheless, few have directly investigated the feasibility of RV strain analysis of distinguishing ARVC from RVOT-VA patients. In addition, strain analysis requires specialized software, which diminishes its availability. This study aimed to use a quantitative index, the right ventricular scalloping index (RVSI), to measure RV free wall scalloping and aid in the imaging diagnosis of ARVC. The diagnostic accuracies of RVSI and RV myocardial strain analysis were compared to demonstrate their clinical applicability.

2. METHODS 2.1. Study population and design

Among the patients with ventricular arrhythmia who underwent CMR, individuals diagnosed with definite ARVC based on the 2010 TFC were retrospectively identified. Using a 1:3 ratio, patients with RVOT-VA were matched by age and sex. All subjects underwent CMR imaging at our institution between January 2005 and May 2018. Patients with significant coronary artery disease, as well as individuals lacking clinical information, cine imaging, or with poor image quality, were excluded from the study. Available clinical data were obtained from patient medical records, including demographic information, presenting symptoms, family history of ARVC or SCD, echocardiography/Holter monitor findings, and therapies for arrhythmia. This study was approved by our institutional review board, which waived the requirement for informed consent for this retrospective study.

2.2. CMR protocol

CMR imaging is typically performed using a 3T scanner (Discovery MR750; GE Healthcare, Waukesha, WI) or a 1.5T scanner (Optima MR450w; GE Healthcare) with a cardiac phased-array receiver surface coil. Informed consent for the CMR was obtained before imaging. Cine imaging was performed using a steady-state free precession sequence in the breath-hold technique with retrospective electrocardiographic gating. Typical acquisition parameters for short-axis cine images from the cardiac base to apex and long-axis cine images in four-chamber view were as follows: echo time (TE), 1.2 milliseconds; repetition time (TR), 3.3 milliseconds; and flip angle, 45° for the 3T scanner; and TE, 1.6 to 1.7 milliseconds; TR, 4.5 to 4.7 milliseconds; and flip angle, 50° for the 1.5T scanner. The field of view was 300 × 300 mm (typical value, may vary depending on the size of the patient), while the image matrix was 224 × 256. The slice thickness was 8 mm, with a slice gap of 2 mm for the short-axis view and no intersection gap for the long-axis four-chamber view. A total of 20 to 30 phases per cardiac cycle were generated with a temporal resolution of 39.6 to 57 milliseconds.

2.3. CMR imaging analysis

Two radiologists (K.-Y.H. and C.-K.C.) processed the CMR images at a workstation using DICOM viewing software (SmartIris, version 1.3.6.0; SYSPOWER, Taipei, Taiwan, ROC) for the RVSI measurement and a CMR image processing system (CVI42 version 5.12.4; Circle Cardiovascular Imaging Inc., Calgary, Canada) for cardiac functional parameters and RV strain analysis. End-systolic phase images from the four-chamber view of each enrolled patient were identified. The epicardial contour of the RV was outlined from the apex to the atrioventricular groove and measured as the curve length. A straight line was drawn between the two end points of the epicardial contour of the RV and measured as a straight distance. RVSI was defined as the curve length divided by the straight distance and obtained from the end-systolic phase images for further analysis (Fig. 1). To assess intraobserver variability of the RVSI, two consecutive measurements were performed by the first investigator at an interval of longer than 28 days. To evaluate interobserver reproducibility, all cases were measured by the second investigator, who was blinded to the results of the first investigator, and the results were compared. The values from the first measurement by the first investigator were used for the subsequent main analysis.

Fig. 1:

RVSI measurements in a patient with ARVC (A) and another patient with RVOT-VA (B). The RVSI was defined as the curve length divided by the straight distance on end-systolic phase images. The RVSI value was higher in the ARVC patient due to the scalloping of right ventricular free wall. ARVC = arrhythmogenic right ventricular cardiomyopathy; RVOT-VA = right ventricular outflow tract ventricular arrhythmia; RVSI = right ventricular scalloping index.

To measure the functional parameters and perform the RV strain analysis, the endocardial and epicardial contours of the RV and left ventricle (LV) were drawn semi-automatically, followed by manual adjustment of the end-diastolic and -systolic phases of the short- and long-axis four-chamber view cine images. The papillary muscles were included in the volumes and the anterior and inferior insertion points were manually drawn. The functional parameters of the RV and LV were obtained, including EF, end-systolic volume (ESV), and end-diastolic volume (EDV). The ventricular volumes were indexed to the body surface area calculated using the Mosteller formula. The end-diastolic phase was selected as the reference for automatic feature tracking throughout the cardiac cycle. The values of the two-dimensional (2D) longitudinal strain parameters were determined from the long-axis four-chamber view cine images, whereas the values of the 2D radial and circumferential strain parameters were determined from the short-axis cine images (Fig. 2).

Fig. 2:

Two-dimensional right ventricular peak global strain values were analyzed with a CVI42 cardiac magnetic resonance system (A). The global longitudinal strain values were obtained from long-axis four-chamber view cine images (B). The global radial and circumferential strain values were obtained from short-axis cine images (C).

2.4. Statistical analysis

Continuous variables are presented as mean ± SD, while categorical variables are presented as counts and percentages. Intergroup differences were compared using a non-directional two-sample t test for continuous variables and the Chi-squared or Fisher exact test for dichotomous variables. Receiver operating characteristic (ROC) curve analysis was performed to determine the accuracy of RVSI, functional parameters, and strain parameters at distinguishing between ARVC and RVOT-VA patients. A cutoff value of RVSI was identified using Youden index to achieve optimal diagnostic sensitivity and specificity. The Delong test was performed to compare the differences between the areas under the curve (AUC). Intraobserver and interobserver reproducibility and agreement were assessed using the intraclass correlation coefficient (ICC) and Bland-Altman plot, respectively. A linear regression model was performed to detect proportional bias.

Correlations were evaluated using Pearson correlation between RVSI and functional parameters, including RVEF and the RVEDV index (RVEDVI). A subgroup analysis of the ARVC and RVOT-VA groups was also performed. Factors predicting the diagnosis of ARVC over RVOT-VA were analyzed using univariable and multivariable logistic regression. To test the robustness of the model, we performed sensitivity analysis. Statistical significance was defined as a two-sided p < 0.05. The statistical analysis was performed using R software (version 4.2.1; R Foundation for Statistical Computing, Vienna, Austria).

3. RESULTS

CMR images of 15 patients with ARVC (25%) and 45 patients with RVOT-VA (75%) were evaluated. Each of the 15 ARVC patients was diagnosed as definite ARVC by fulfilling a minimum of two major criteria or one major and two minor criteria according to the 2010 TFC. Among the criteria met, seven patients met the major CMR criteria, one patient met the minor CMR criteria, four patients met the major criteria of RV angiogram, and three patients were diagnosed based on criteria involving electrocardiographic studies, histological evidence, and family history. Baseline characteristics and clinical information of the study population are shown in Table 1. The average patient age was 48.2 ± 12 years; 36 (60%) of them were male. Mean age and sex were similar between the two groups as the result of matching. The ARVC and RVOT-VA groups showed significant differences in mean body mass index (BMI) (27.6 ± 5.0 vs 24.7 ± 4.1, p = 0.027), family history of ARVC or SCD (three [20%] vs one [2.2%], p = 0.045), symptoms of syncope (seven [46.7%] vs five [11.1%], p = 0.006), beta-blocker use (14 [93.3%] vs 26 [57.8%], p = 0.011), amiodarone use (12 [80%] vs seven [15.6%], p < 0.001), and implantable cardioverter defibrillator (ICD) implantation (10 [66.7%] vs one [2.2%], p < 0.001). There were no statistically significant intergroup differences in cardiovascular comorbidities, including hypertension, diabetes mellitus, congestive heart failure (CHF), and dyslipidemia.

Table 1 - Baseline characteristics of the study population ARVC (n = 15) RVOT-VA (n = 45) p Clinical data  Age, y 48.3 ± 12.5 48.2 ± 11.9 0.961  Gender (% male) 9 (60%) 27 (60%) >0.99  Weight, kg 76.2 ± 19.3 69.2 ± 13.7 0.133  Height, cm 165.2 ± 10.9 166.9 ± 8.5 0.547  BSA, m2 1.86 ± 0.28 1.78 ± 0.21 0.263  BMI, kg/m2 27.6 ± 5.0 24.7 ± 4.1 0.027  Family historya 3 (20%) 1 (2.2%) 0.045  Sustained VT/VPCs >500/d 12 (80%) 41 (91.1%) 0.351  MACEb 2 (13.3%) 1 (2.2%) 0.151 Comorbidity  Hypertension 6 (40%) 15 (33.3%) 0.639  Diabetes mellitus 2 (13.3%) 5 (11.1%) >0.99  CHF 0 (0%) 3 (6.7%) 0.566  Dyslipidemia 6 (40%) 6 (13.3%) 0.056 Symptom  Asymptomatic 0 (0%) 3 (6.7%) 0.566  Chest pain 8 (53.3%) 15 (33.3%) 0.168  Syncope 7 (46.7%) 5 (11.1%) 0.006  Palpitation 11 (73.3%) 34 (75.6%) >0.99  Dyspnea 5 (33.3%) 6 (13.3%) 0.122  Resuscitated SCD 2 (13.3%) 1 (2.2%) 0.151  Others 5 (33.3%) 13 (28.9%) 0.754 Therapy  Ablation 13 (86.7%) 45 (100%) 0.059  Beta-blocker 14 (93.3%) 26 (57.8%) 0.011  Amiodarone 12 (80%) 7 (15.6%) <0.001  ICD implantation 10 (66.7%) 1 (2.2%)c <0.001

ARVC = arrhythmogenic right ventricular cardiomyopathy; BMI = body mass index; BSA = body surface area; CHF = congestive heart failure; ICD = implantable cardioverter defibrillator; MACE = major adverse cardiovascular events; RVOT-VA = right ventricular outflow tract ventricular arrhythmia; SCD = sudden cardiac death; VPC = ventricular premature complex; VT = ventricular tachycardia.

aFamily history of ARVC or SCD.

bDefined as cardiogenic death, heart transplant, or hospitalization for heart failure.

cICD implantation due to ventricular fibrillation triggered by idiopathic ventricular arrhythmia.

The mean CMR volumetric data, RVSI measurements, and RV global peak strain data are presented in Table 2. The mean RVEF of the ARVC group was lower than that of the RVOT-VA group (39.1 ± 18.5 vs 51.2 ± 12.0, p = 0.03). The mean RVSI value of the ARVC group was higher than that of the RVOT-VA group (1.56 ± 0.23 vs 1.30 ± 0.08, p < 0.001). There was no statistically significant intergroup difference in the RV global peak strain analysis. Moderate linear correlations were found between RVSI and RVEF (r = −0.40, p = 0.002) (Fig. 3A) as well as between RVSI and RVEDVI (r = 0.57, p < 0.001) (Fig. 3B). The subgroup analysis revealed a strong linear correlation between RVSI and RVEDVI (r = 0.77, p < 0.001) (Fig. 3C) in the ARVC subgroup. The diagnostic performances of RVSI, RVEF, RVEDVI, RV peak global longitudinal strain (pGLS), RV peak global circumferential strain (pGCS), and RV peak global radial strain (pGRS) are shown in Fig. 4. RVSI was a strong discriminator between ARVC and RVOT-VA patients (AUC, 0.91; 95% CI, 0.82-0.99), outperforming RV functional parameters and RV global peak strain parameters. Significant differences in AUC between RVSI and the other parameters were validated by pairwise comparisons. A cutoff value of RVSI ≥1.49 provided an accuracy of 90.0%, specificity of 97.8%, sensitivity of 66.7%, positive predictive value (PPV) of 90.9%, and a negative predictive value (NPV) of 89.8%, indicating that patients with an RVSI value ≥1.49 were more likely to have ARVC rather than RVOT-VA, whereas patients with an RVSI value <1.49 were less likely to have ARVC.

Table 2 - CMR volumetric data and the RVSI and RV peak global strain analysis ARVC (n = 15) RVOT-VA (n = 45) p LVEF, % 51.5 ± 16.9 54.4 ± 12.8 0.485 LVEDVI, mL/m2 75.6 ± 23.3 81.6 ± 15.2 0.252 LVESVI, mL/m2 39.2 ± 24.6 38.1 ± 17.4 0.849 RVEF, % 39.1 ± 18.5 51.2 ± 12.0 0.030 RVEDVI, mL/m2 94.0 ± 55.1 80.1 ± 14.9 0.350 RVESVI, mL/m2 64.6 ± 58.6 38.8 ± 10.7 0.112 RVSI 1.56 ± 0.23 1.30 ± 0.08 <0.001 RVpGLS, % −13.9 ± 7.2 −16.3 ± 5.0 0.162 RVpGCS, % −7.6 ± 6.1 −9.2 ± 3.3 0.351 RVpGRS, % 12.2 ± 10.2 14.5 ± 5.6 0.429

ARVC = arrhythmogenic right ventricular cardiomyopathy; CMR= cardiac magnetic resonance imaging; LVEDVI = left ventricular end-diastolic volume index; LVEF = left ventricular ejection fraction; LVESVI = left ventricular end-systolic volume index; RVEDVI = right ventricular end-diastolic volume index; RVEF = right ventricular ejection fraction; RVESVI = right ventricular end-systolic volume index; RVOT-VA = right ventricular peak global circumferential strain; RVpGLS = right ventricular peak global longitudinal strain; RVpGRS = right ventricular peak global longitudinal strain; RVpGRS = right ventricular peak global radial strain; RVSI = right ventricular scalloping index.

Fig. 3:

Pearson correlation demonstrated significant linear relationships of RVSI with RVEF and RVEDVI in all patients (A, B), and RVSI with RVEDVI in the ARVC subgroup (C). ARVC = arrhythmogenic right ventricular cardiomyopathy; RVEDVI = right ventricular end-diastolic volume index; RVEF = right ventricular ejection fraction; RVOT-VA = right ventricular outflow tract ventricular arrhythmia; RVSI = right ventricular scalloping index.

Fig. 4:

ROC curve analysis for parameters distinguishing ARVC patients from RVOT-VA patients was performed. The AUC of the RVSI was significantly higher than that of RVEF, RVEDVI, RVpGLS, RVpGCS, and RVpGRS. aPairwise comparison of AUC between RVSI and RVEF. bPairwise comparison of AUC between RVSI and RVEDVI. cPairwise comparison of AUC between RVSI and RVpGLS. dPairwise comparison of AUC between RVSI and RVpGCS. ePairwise comparison of AUC between RVSI and RVpGRS. ARVC = arrhythmogenic right ventricular cardiomyopathy; AUC = area under the curve; ROC = receiver operating characteristic; RVEDVI = right ventricular end-diastolic volume index; RVEF = right ventricular ejection fraction; RVOT-VA = right ventricular outflow tract ventricular arrhythmia; RVpGCS = right ventricular peak global circumferential strain; RVpGLS = right ventricular peak global longitudinal strain; RVpGRS = right ventricular peak global radial strain; RVSI = right ventricular scalloping index.

As shown in Table 3, the univariable regression analysis suggested that BMI, syncope, family history of ARVC or SCD, an RVEF ≤40%, and an RVSI ≥1.49 were significantly associated with the likelihood of an ARVC diagnosis. In the multivariable regression analysis, a family history of ARVC or SCD and an RVSI ≥1.49 remained significantly associated with the likelihood of disease (odds ratio [95% CI], 38.71 [1.48-1011.05], and 64.72 [4.58-914.63], p = 0.028 and 0.002, respectively). A sensitivity analysis using the rounded highest tercile, quartile, and quintile RVSI values as cutoffs at 1.36, 1.41, and 1.44, respectively, showed a significant association with the likelihood of the disease (odds ratios of 15.43, 22.69, and 24.04, respectively; p values of 0.007, 0.002, and 0.003, respectively).

Table 3 - Factors predicting the diagnosis of ARVC vs RVOT-VA analyzed using univariable and multivariable logistic regression Univariable Multivariable OR (95% CI) p OR (95% CI) p BMI 1.17 (1.01-1.36) 0.035 1.23 (0.95-1.60) 0.117 Syncope 7.00 (1.77-27.71) 0.006 4.16 (0.38-45.36) 0.243 Family historya 11.00 (1.05-115.51) 0.046 38.71 (1.48-1011.05) 0.028 RVEF ≤40% 4.75 (1.30-17.35) 0.018 5.72 (0.67-48.57) 0.110 RVSI ≥1.49 88.00 (9.24-838.05) <0.001 64.72 (4.58-914.63) 0.002

ARVC = arrhythmogenic right ventricular cardiomyopathy; BMI = body mass index; OR = odds ratio; RVEF = right ventricular ejection fraction; RVOT-VA = right ventricular outflow tract ventricular arrhythmia; RVSI = right ventricular scalloping index; SCD = sudden cardiac death.

aFamily history of ARVC or SCD.

The ICC for intraobserver reproducibility of RVSI was 0.94 (95% CI, 0.90-0.96), while that for interobserver reproducibility was 0.96 (95% CI, 0.93-0.97), indicating good reliability. The intraobserver and interobserver agreements are illustrated in the Bland-Altman plots (Fig. 5). No proportional bias was found in the intraobserver (p = 0.072) or interobserver (p = 0.48) RVSI measurement.

Fig. 5:

Bland-Altman plots for interobserver (A) and intraobserver (B) bias for RVSI measurements. The dashed lines indicated the mean difference with the upper and lower 95% limits of agreement. RVSI = right ventricular scalloping index.

4. DISCUSSION

In the current study, patients with ARVC had significantly higher RVSI values than those with RVOT-VA. A cutoff value of RVSI ≥1.49 provided high accuracy for the diagnosis of ARVC. RVSI remained predictive of ARVC after the adjustment for covariates on the multivariable analysis.

While the ratio of path length to distance has been previously referred to as sinuosity or tortuosity for measuring the anatomical features of tubular structures such as the trachea16 and vessels,17 no intuitive index for assessing the anatomical characteristics of cardiac chamber walls has been thoroughly evaluated. RVSI, the index used in this study, demonstrated high diagnostic accuracy and was intuitive to measure on long-axis four-chamber view cine images. The end-systolic phase was chosen as the timing for measurement, as previous studies have shown that scalloping of the RV free wall was most prominent in this period.3 Compared with the RV global strain analysis, higher performance by RVSI in differentiating ARVC patients from RVOT-VA patients was observed, especially for patients with possibly impaired cardiac function due to underlying cardiovascular comorbidities. An RVSI cutoff value of ≥1.49 demonstrated a specificity of 97.8% and a sensitivity of 66.7%, both of which were favorable considering previously reported specificity and sensitivity values of fatty infiltration (79%/84%), RV dilatation (96%/68%), and regional RV wall motion abnormalities (94%/78%).8 A Bland-Altman analysis for intraobserver and interobserver measurements demonstrated relatively small mean differences, and no proportional bias was found. We also observed a linear relationship between RVSI and RVEDVI in ARVC patients, which could be explained by the fact that RV dilatation led to an increase in the RV transverse diameter, thereby partially contributing to the RVSI value. The increased RV transverse diameter in ARVC vs RVOT-VA patients was demonstrated by Saberniak et al18 with echocardiographic assessments. Nonetheless, in the current study, while RVEDI was larger in patients with ARVC, the difference in between patients with ARVC and RVOT-VA was not significant. This may be attributed to the limited number of ARVC and RVOT-VA cases, resulting in constrained statistical power.

Previous studies discussed other imaging features for predicting prognosis or risk stratification for ARVC patients. Aquaro et al19 described that LV involvement (defined as presence of an LVEF <50%, LV wall motion abnormalities, LV fat infiltration, or non-ischemic LGE) was associated with a higher probability of SCD, aborted arrest, and appropriated ICD intervention. In an echocardiographic study by Pinamonti et al,20 the main independent long-term predictors of death or heart transplant included RV dysfunction, significant tricuspid regurgitation, biventricular involvement, and amiodarone treatment. Further studies may be warranted to investigate the association between RVSI and major adverse cardiovascular events in ARVC patients. The disparity of the RV global strain values did not demonstrate a significant difference between ARVC and RVOT-VA patients in our study, which differed from the result as Prati et al7 had reported. The RVpGLS, RVpGCS, and RVpGRS values of our RVOT-VA patients were lower than the normal reference corresponding to the age stratification from healthy subjects provided by Liu et al.21 We postulated that some RVOT-VA patients in our study might have subclinical impaired cardiac function, attributed to underlying cardiovascular comorbidities such as hypertension and diabetes mellitus as previously described.22,23 This could potentially reduce the disparity in myocardial strain observed between ARVC and RVOT-VA patients.

We consider RVSI a useful and practical tool that can be routinely performed in CMR studies to evaluate the possible diagnosis of ARVC. Because ARVC is among the major differential diagnoses to be considered in patients with an initial impression of RVOT-VA, image survey with CMR before ablation is recommended to provide a comprehensive assessment of RV morphology, and RVSI is easy and intuitive to measure.

This study had some limitations that require acknowledgment. First, it was limited by its single-center and retrospective design, which allowed for a potential selection bias. Second, the sample size of the ARVC group was modest due to the rarity of this disease entity. To explore the generalizability of this index and its cutoff value, further external validation or prospective studies with a larger sample size may be necessary. Third, we did not perform a regional strain analysis of the RV, as previous studies reported impairment of longitudinal and circumferential strain parameters in the subtricuspid region or the basal level of the RV.14,15 Nevertheless, some feature tracking-derived strain studies reported low reproducibility of segmental strain measurements.24,25 Finally, we enrolled RVOT-VA patients rather than healthy subjects as the control group. However, in real-world practice, patients who require CMR evaluations for suspected ARVC are usually not healthy. Thus, we consider our study design consistent with actual clinical situations.

In conclusion, RVSI is a quantitative imaging marker with good diagnostic performance that may be routinely performed for evaluating patients with suspected ARVC. The role of this index in prognosis evaluation could be investigated in future studies with larger patient cohorts.

ACKNOWLEDGMENTS

This work was supported by a grant from Taipei Veterans General Hospital (V111C-143).

REFERENCES 1. Corrado D, Link MS, Calkins H. Arrhythmogenic right ventricular cardiomyopathy. N Engl J Med. 2017;376:61–72. 2. Murphy DT, Shine SC, Cradock A, Galvin JM, Keelan ET, Murray JG. Cardiac MRI in arrhythmogenic right ventricular cardiomyopathy. AJR Am J Roentgenol. 2010;194:W299–306. 3. Rastegar N, Burt JR, Corona-Villalobos CP, Riele AST, James CA, Murray B, et al. Cardiac MR findings and potential diagnostic pitfalls in patients evaluated for arrhythmogenic right ventricular cardiomyopathy. Radiographics. 2014;34:1553–70. 4. Bluemke DA, Krupinski EA, Ovitt T, Gear K, Unger E, Axel L, et al. MR imaging of arrhythmogenic right ventricular cardiomyopathy: morphologic findings and interobserver reliability. Cardiology. 2003;99:153–62. 5. Whyte GP, Stephens N, Senior R, Peters N, O’hanlon R, Sharma S. Differentiation of RVOT-VT and ARVC in an elite athlete. Med Sci Sports Exerc. 2008;40:1357–61. 6. Te Riele ASJM, Tandri H, Bluemke DA. Arrhythmogenic right ventricular cardiomyopathy (ARVC): cardiovascular magnetic resonance update. J Cardiovasc Magn Reson. 2014;16:50. 7. Prati G, Vitrella G, Allocca G, Muser D, Buttignoni SC, Piccoli G, et al. Right ventricular strain and dyssynchrony assessment in arrhythmogenic right ventricular cardiomyopathy: cardiac magnetic resonance feature-tracking study. Circ Cardiovasc Imaging. 2015;8:e003647. 8. Tandri H, Castillo E, Ferrari VA, Nasir K, Dalal D, Bomma C, et al. Magnetic resonance imaging of arrhythmogenic right ventricular dysplasia: sensitivity, specificity, and observer variability of fat detection versus functional analysis of the right ventricle. J Am Coll Cardiol. 2006;48:2277–84. 9. Aquaro GD, Barison A, Todiere G, Grigoratos C, Ait Ali L, Di Bella G, et al. Usefulness of combined functional assessment by cardiac magnetic resonance and tissue characterization versus task force criteria for diagnosis of arrhythmogenic right ventricular cardiomyopathy. Am J Cardiol. 2016;118:1730–6. 10. Corrado D, Basso C, Thiene G. Arrhythmogenic right ventricular cardiomyopathy: diagnosis, prognosis, and treatment. Heart. 2000;83:588–95. 11. Oomen A, Semsarian C, Puranik R, Sy RW. Diagnosis of arrhythmogenic right ventricular cardiomyopathy: progress and pitfalls. Heart Lung Circ. 2018;27:1310–7. 12. Perazzolo Marra M, Rizzo S, Bauce B, De Lazzari M, Pilichou K, Corrado D, et al. Arrhythmogenic right ventricular cardiomyopathy. Contribution of cardiac magnetic resonance imaging to the diagnosis. Herz. 2015;40:600–6. 13. Castaños Gutiérrez SL, Kamel IR, Zimmerman SL. Current concepts on diagnosis and prognosis of arrhythmogenic right ventricular cardiomyopathy/dysplasia. J Thorac Imaging. 2016;31:324–35. 14. Heermann P, Hedderich DM, Paul M, Schülke C, Kroeger JR, Baeßler B, et al. Biventricular myocardial strain analysis in patients with arrhythmogenic right ventricular cardiomyopathy (ARVC) using cardiovascular magnetic resonance feature tracking. J Cardiovasc Magn Reson. 2014;16:75. 15. Vigneault DM, Te Riele AS, James CA, Zimmerman SL, Selwaness M, Murray B, et al. Right ventricular strain by MR quantitatively identifies regional dysfunction in patients with arrhythmogenic right ventricular cardiomyopathy. J Magn Reson Imaging. 2016;43:1132–9. 16. Camilo GB, Carvalho ARS, Guimarães ARM, Kasuki L, Gadelha MR, Mogami R, et al. Computed tomography airway lumen volumetry in patients with acromegaly: association with growth hormone levels and lung function. J Med Imaging Radiat Oncol. 2017;61:591–9. 17. Chaikof EL, Fillinger MF, Matsumura JS, Rutherford RB, White GH, Blankensteijn JD, et al. Identifying and grading factors that modify the outcome of endovascular aortic aneurysm repair. J Vasc Surg. 2002;35:1061–6. 18. Saberniak J, Leren IS, Haland TF, Beitnes JO, Hopp E, Borgquist R, et al. Comparison of patients with early-phase arrhythmogenic right ventricular cardiomyopathy and right ventricular outflow tract ventricular tachycardia. Eur Heart J Cardiovasc Imaging. 2017;18:62–9. 19. Aquaro GD, De Luca A, Cappelletto C, Raimondi F, Bianco F, Botto N, et al. Prognostic value of magnetic resonance phenotype in patients with arrhythmogenic right ventricular cardiomyopathy. J Am Coll Cardiol. 2020;75:2753–65. 20. Pinamonti B, Dragos AM, Pyxaras SA, Merlo M, Pivetta A, Barbati G, et al. Prognostic predictors in arrhythmogenic right ventricular cardiomyopathy: results from a 10-year registry. Eur Heart J. 2011;32:1105–13. 21. Liu T, Wang C, Li S, Zhao Y, Li P. Age- and gender-related normal references of right ventricular strain values by tissue tracking cardiac magnetic resonance: results from a Chinese population. Quant Imaging Med Surg. 2019;9:1441–50. 22. Ji M, Wu W, He L, Gao L, Zhang Y, Lin Y, et al. Right ventricular longitudinal strain in patients with heart failure. Diagnostics (Basel). 2022;12:445. 23. Wu T, Li X, Zhang D, Gong LG. Early impairment of right ventricular systolic function in patients with prediabetes and type 2 diabetes mellitus: an analysis of two-dimensional speckle tracking echocardiography. Echocardiogr. 2023;40:831–40. 24. Schuster A, Morton G, Hussain ST, Jogiya R, Kutty S, Asrress KN, et al. The intra-observer