Clinical, echocardiographic and cardiac MRI predictors of outcomes in patients with apical hypertrophic cardiomyopathy

Recent studies have found that some adverse cardiovascular events could also occur in patients with apical hypertrophic cardiomyopathy (ApHCM), which is different with previous studies suggesting benign nature of this condition. Therefore, the present study aimed to observe the clinical prognosis of ApHCM and to identify the predictors of poor prognosis in clinical, echocardiography and cardiac magnetic resonance (CMR). A total of 126 ApHCM patients with both echocardiography and CMR were identified retrospectively from January 2008 to December 2018. Adverse clinical events were defined as a composite of cardiac death, progressive heart failure, myocardial infarction, thromboembolic stroke, appropriate implantable cardioverter-defibrillator (ICD) interventions for ventricular tachycardia or ventricular fibrillation, and new-onset atrial fibrillation (AF). During a mean follow-up of 96.8 ± 36.0 months, clinical events were observed in 34 (27.0%) patients. As compared with patients without clinical events, patients with clinical events were older and had a higher incidence of heart failure. Moreover, patients with clinical events had a higher incidence of non-sustained ventricular tachycardia (NSVT) and had larger left atrial volume index (LAVI), thicker apical thickness, lower peak systolic mitral annular velocity (S′) than those without clinical events. In addition, late gadolinium enhancement (LGE) in CMR were more frequently observed in patients with clinical events. Five predictors of poor prognosis were identified: age ≥ 55 years, LAVI ≥ 36.7 ml/m2, S′ ≤ 6.7 cm/s, NSVT and LGE. ApHCM was not as benign as expected. Age ≥ 55 years, LAVI ≥ 36.7 ml/m2, S′ ≤ 6.7 cm/s along with NSVT and LGE were independent predictors for poor prognosis of ApHCM.


Introduction
Hypertrophic cardiomyopathy (HCM), with the prevalence of 0.2% in the general population [1], is an autosomal dominant disease of the myocardium, caused by mutations of genes encoding sarcomeric proteins and characterized by the presence of increased left ventricular (LV) wall thickness that is not solely explained by abnormal loading conditions. Apical hypertrophic cardiomyopathy (ApHCM) is a relatively rare subtype of HCM, which mainly involves the left ventricular apex [2]. In the Asian population, ApHCM patients account for about 25% of the total HCM population and 1-10% in the non-Asian population [3]. Sakamoto et al. first described ApHCM in patients with "giant" T-wave negative and end diastolic LV "spade-shaped" configuration measured by electrocardiogram and left ventriculography, respectively [4]. Although it has been reported that ApHCM has a good prognosis in terms of cardiovascular mortality [5,6], it has been found in recent years that about one -third of ApHCM patients may experience adverse clinical events and potentially life-threatening complications, such as malignant arrhythmia, sudden cardiac death (SCD), heart failure, myocardial infarction, atrial fibrillation and stroke [7]. Therefore, the aim of this study was to observe the clinical prognosis of patients with ApHCM and to identify the predictors of poor prognosis including clinical, echocardiographic and cardiac magnetic resonance (CMR) parameters.

Study population
A total of 675 patients were diagnosed with HCM in Nanjing Medical University Affiliated Wuxi No. 2 Hospital from January 2008 to December 2018. Among them, 142 ApHCM patients with both transthoracic echocardiography and CMR were identified.
The criteria for the diagnosis of ApHCM included a demonstration of otherwise unexplained left ventricular hypertrophy, confined predominantly to the left ventricular apex below the papillary muscle level, with an apical thickness > 15 mm or a ratio of maximal apical to posterior wall thickness > 1.5 at the end diastole using standard 2-dimensional transthoracic echocardiography or CMR [2]. 16 patients were excluded for the following reasons: (1) age < 18 years (N = 2); (2) severe heart failure (NYHA, N = 4), liver dysfunction (N = 1), renal failure requiring dialysis treatment (N = 2), severe valvular disease (N = 1) and severe coronary artery disease (N = 3); (3) inadequate clinical data (N = 3). Finally, 126 patients were included in this retrospective study. The baseline clinical characteristics, electrocardiography (ECG), echocardiography and CMR results of these 126 patients were gathered at presentation (Fig. 1).
The study was approved by the institutional review boards and all subjects provided informed consent.

Echocardiography
Transthoracic two-dimensional echocardiography was performed with commercially available instruments (HP5500 color Doppler (USA) with the probe frequency of 2.5 and 2.4 MHz). Standard 2-dimensional measurements were obtained as recommended by the American Society of Echocardiography in the left lateral position [8]. The maximum apical wall thickness was obtained from the standard apical 4-and 2-chamber views at end-diastole. The left atrial (LA) dimensions were measured at end-systole, and the LA volume was calculated using the area-length methods. The LA volume index (LAVI) was calculated as the LA volume divided by the body surface area. Early mitral inflow velocity (E) and late mitral inflow velocity (A) were measured using the pulsed wave Doppler method. Tissue Dopplerderived early diastolic mitral annular velocity (e′), late diastolic mitral annular velocity (a′), and peak systolic mitral annular velocity (S′) were measured from the septal corner of the mitral annulus in the apical 4-chamber view. The ratio of the E velocity to the e′ septal mitral annular velocities were averaged to provide an index of diastolic function. Left ventricular ejection fraction was calculated from LV volumes using Simpson's rule. The echocardiographic data were gathered and analyzed by 2 experienced echocardiographers and were recorded on videotapes or hard disc.

CMR images acquisition and analysis
CMR was performed with a 1.5 T MRI scanner (Magnetom Avanto, Siemens, Germany). Transverse and sagittal dark blood images were obtained by half-Fourier acquisition single-shot turbo spin-echo sequence. Breath-hold cine steadystate free precession images were performed in standard views with full LV coverage, including LV two-chamber, four-chamber, left ventricular outflow tract, and eight slices of short-axis views. Late gadolinium enhancement (LGE) images were acquired 10-15 min after intravenous administration of 0.2 mmol/kg gadolinium-DTPA with breath-held segmented inversion-recovery sequence (inversion time, 240 to 300 ms) acquired in the same views as the cine images.

Follow-up
Clinical follow-ups were initiated from the time of diagnosis of ApHCM. The primary endpoint was a composite of cardiac death, progressive heart failure [with an increase of at least one New York Heart Association (NYHA) functional class], myocardial infarction, thromboembolic stroke, appropriate implantable cardioverter-defibrillator (ICD) interventions for ventricular tachycardia or ventricular fibrillation, and new-onset atrial fibrillation (AF).

Statistical analyses
All analyses were performed with IBM SPSS Statistics 20.0 software. Continuous normally distributed data were presented as mean ± SD, and were compared using independent-samples t-test. Non-normally distributed continuous data were displayed as median and interquartile range (25-75%), and were compared using non-parametric test (Mann-Whitney U test). Categorical data were presented as number and percentage, and were compared using the χ 2 or Fisher's exact test. Cox analysis was used to identify univariable and multivariable predictors of clinical events. Variables exhibiting a p < 0.1 in the univariate analysis were tested in the multivariable analysis. Receiver operating characteristics curves were used to examine the accuracy of variables in predicting the clinical outcomes. The survival analysis was estimated by the Kaplan-Meier method. Differences in survival between groups were assessed using the log-rank test. A two-tailed p value < 0.05 was considered statistically significant.

Prognostic data
The whole study group comprised 126 patients, and their mean age was 58.3 ± 10.4 years (54.0% were male). During a mean follow-up of 96.8 ± 36.0 months, clinical events were observed in 34 (27.0%) patients, representative picture is shown in Fig. 2. The clinical events during the follow-up periods are listed in Table 1. Cardiac death occurred in 6 patients (4.8%) during the monitoring period, 17(13.5%) patients had progressive heart failure, and 5 patients (4.0%) experienced thromboembolic stroke.
A total of 10 patients developed definite ventricular arrhythmias, including 4 patients with ventricular fibrillation (3 patients died, 1 patient underwent successful cardiopulmonary resuscitation, electrical defibrillation, and ICD implantation), and 6 patients developed definite ventricular tachycardia and were implanted with ICD and no patient underwent ablation. Nonfatal myocardial infarction occurred in 3 (2.4%) patients without significant coronary artery disease and was confirmed by coronary angiography. In addition, 8 (6.3%) patients developed new-onset atrial fibrillation.

Baseline characteristics of study patients
The patients were divided into two groups according to whether the clinical event occurred or not. Baseline characteristics of the study population are summarized in Table 2. The patients who experienced events were older (62.6 ± 7.0 vs. 56.8 ± 11.0 years, p = 0.005) and had a higher incidence of heart failure (38.2% vs. 16.3%, p = 0.015) than the patients who experienced no events. No significant differences were found in other clinical characteristics between the 2 groups.

CMR results
CMR results are listed in Table 5. Compared with patients without clinical events, patients with events were more frequently with LGE presence (82.4% vs. 62.0, p = 0.030). There are no significantly difference in other parameters between the two groups.

Predictors of clinical events
The results of the Cox Univariate regression analysis are listed in Tables S1-4 (Data in Supplementary materials). The results of the Cox multivariate regression analysis are listed in Fig. 3 and Table S5 (Data in Supplementary materials). In multivariate analysis, the age, LAVI, S′, presence of NSVT and LGE were independent risk factors for poor prognosis. ROC curve analysis of continuous variables in risk factors showed that age ≥ 55 years, LAVI ≥ 36.7 ml/ m 2 , S′ ≤ 6.7 cm/s were the best cut-off value. Kaplan-Meier curves of event-free survival since the initial presentation according to the risk factors showed that the patients who had an older age, a larger LAVI, a lower S′ along with NSVT and LGE presence experienced significantly worse clinical outcomes during follow-up (Fig. 4).

Discussion
HCM is an autosomal dominant inherited cardiomyopathy with genetic heterogeneity, characterized by asymmetric ventricular hypertrophy, which is one of the main causes of sudden exercise death in adolescents [9]. ApHCM is a phenotypic variation of HCM, with hypertrophy predominantly at the LV apical portion, with or without middle segment involvement and apical aneurysm formation [10]. ApHCM  was initially considered not to increase the risk of mortality, but recent data show that cardiac mortality is 0.5% to 4% per year, close to typical HCM [3]. In our study, we found a mortality rate of 4.8% for ApHCM and our study demonstrated that age ≥ 55 years, LAVI ≥ 36.7 ml/m 2 , S′ ≤ 6.7 cm/s along with NSVT and LGE were independent predictors for poor prognosis of ApHCM.

Prevalence and prognosis of ApHCM
ApHCM was first described in Japan [5,6], where it represents 13% to 25% of the entire HCM population [7,10]. In non-Asians, ApHCM is less common and has been reported in 1-10% of all HCM patients. In our study, the prevalence of ApHCM was 21% (142/675), which was similar with previous studies. ApHCM was originally thought to have a benign prognosis and carry no increased mortality risk, but recent data reported annual cardiac death rates of 0.5% to 4%, approaching rates of classic HCM. In the current study, the total cardiovascular mortality of ApHCM was 4.8% and the annual cardiovascular mortality was 0.6%, which was similar with previous studies and also showed that ApHCM was not as benign as expected. However, the annual cardiovascular mortality was relatively low in our study, which may be related to the small sample size and patient selection.   [3]. However, these studies only considered clinical and echocardiographic parameters. In our study, we found that age ≥ 55 years, LAVI ≥ 36.7 ml/ m 2 , S′ ≤ 6.7 cm/s along with NSVT and LGE were independent risk factors for poor prognosis of ApHCM, which was more comprehensive than previous studies as CMR, echocardiographic parameters, and clinical data were all taken into account. The following is a detailed description of each predictor.

Age and prognosis of ApHCM
Our study found that patients with clinical events were older than those without. Moon et al. and Kim et al. have similar findings, which is related to the fact that older patients may have more cardiovascular risk factors such as hypertension, diabetes, kidney disease or other complications such as cancer [7,11]. Of note, previous studies have found that sudden death in HCM patients is more common in young people [9,12], while in our study, most of the patients with adverse events in ApHCM are elderly people, which may be related to the difference in hypertrophic sites between HCM patients and ApHCM patients and HCM patients are always complicated with left ventricular outflow tract obstruction.

Echocardiographic parameters and prognosis of ApHCM
In our study, patients with adverse clinical events had larger LAVI and lower S′ than those without adverse events. Previous studies reported that left atrial volume was closely related to left ventricular diastolic dysfunction, which reflected the severity degree of left ventricular diastolic dysfunction [13]. Moreover, left atrial volume was considered as an important prognostic factor, not only in the general population, but also in various cardiovascular diseases [14][15][16]. Furthermore, Yang et al. reported that increased LAVI was an independent predictor of cardiovascular events in HCM patients with the best cut-off value of 39 ml/m 2 [17]. Our results suggested that increased LAVI was an independent predictor of adverse clinical outcomes, and the best cut-off value was 36.7 ml/m 2 , which was close to previous research. In addition, decreased S′ velocity, which indicated impaired myocardial contractility, was an independent predictor of poor prognosis. Moon et al. have similar findings [7], and the cut-off value of S′ in their study was 6 cm/s, which was close to ours. Moreover, in their study, E/e′ ratio was also an independent predictor of poor prognosis, which was not in our study for this difference may come from the patient selection and sample size.

NSVT and prognosis of ApHCM
Previous studies have shown that asymptomatic and symptomatic NSVT accounted for 18% and 5% in Holter monitoring of ApHCM patients, and monomorphic ventricular tachycardia often occurs in ApHCM patients with aneurysm, which may be related to the reentry around the aneurysm [2]. In our studies, the prevalence of NSVT was higher, accounted for 38.2% and 17.4% in ApHCM LGE and prognosis of ApHCM CMR can provide high-resolution myocardial images and accurately determine the location and degree of myocardial hypertrophy and presence of apical aneurysms [20], and can accurately quantify and visualize the pattern of dense focal extracellular matrix deposition, such as LGE due to alternative fibrosis, which is related to the clinical severity of ventricular arrhythmia and SCD [21]. In patients with HCM, LGE has been shown to be closely associated with disease progression, small intramural coronary artery dysplasia, adverse outcomes, and cardiac mortality [22], however, in patients with ApHCM, the relationship between LGE and the prognosis of ApHCM is unclear. Our study found that LGE was also associated with poor prognosis and was an independent predictor of poor prognosis. A recent study found that LGE ≥ 15% of LV mass showed that the risk of SCD events increased by 2 times in those patients with low risk, and the probability of SCD events was estimated to be 6% at 5 years [23]. In this cohort of 1293 patients, for every 10% increase in LGE, the relative risk of SCD events increased by 40%. In our study, LGE of left ventricular mass was not measured due to technical limitations, so further studies are needed to determine this point.

Clinical significance
In clinical work, patients with these predictors mean a higher risk of adverse events, indicating that more active treatment should be given in these patients, such as administered the maximum tolerated dose β receptor blockers or calcium channel blockers and anticoagulation therapy for patients with atrial fibrillation. For patients with age ≥ 55 years, especially with other cardiovascular risk factors, such as hypertension, diabetes and ischemic heart disease, they should be closely followed up and actively treated for these comorbidities. Moreover, although the risk of sudden cardiac death is lower in ApHCM than typical phenotype of HCM especially with left ventricular outflow tract obstruction, implantation of an implantable cardioverter-defibrillator (ICD) may also be reasonable, especially in patients with NSVT, extensive LGE or with more predictors. Additionally, a strict followup plan is necessary for all patients with these predictors.

Limitations
First, this study was a single center retrospective study. Therefore, our study subjects might not be representative of the overall patient population with ApHCM. Second, the sample size in this study was small and the follow-up was relatively short. Therefore, a study with a larger sample size and a long-term follow-up is needed. Third, some patients who only underwent echocardiography could have been missed for not diagnosing with ApHCM during the period included in the study, which may lead to selection bias. Fourth, no patients had genetic analysis in the present study, the correlation between genetic cause and poor prognosis remained unresolved. Fifth, selection of patients for device therapy may result in some selection bias with respect to survival analysis. Sixth, previous studies have suggested that LV apical aneurysm (LVAA) was a predictor of poor prognosis [10,24], but in this study, there was no significant difference between the two groups because there were fewer patients with LVAA. Seventh, although we have told all patients that if they had palpitations or syncope attacks, they should go to our hospital or other hospitals in time and have ECG or ambulatory ECG examination, some patients may not see a doctor because of transient tachycardia events, resulting in the loss of some adverse event data. Eighth, although Fabry disease is rare in ApHCM, we have to admit that such patients may exist in the included patients. Finally, due to technical limitations, LGE of LV mass was not available in this study.

Conclusion
ApHCM was not as benign as expected. During a mean follow-up of 96.8 ± 36.0 months, clinical events were observed in 34 (27.0%) ApHCM patients. Age ≥ 55 years, LAVI ≥ 36.7 ml/m 2 , S′ ≤ 6.7 cm/s along with NSVT and LGE were independent predictors for poor prognosis of ApHCM.