Palpitation was associated with clinical outcomes in patients with hypertrophic cardiomyopathy

Hypertrophic cardiomyopathy (HCM) is a common genetic heart disease with diversified clinical presentation and it is important to identify new predictors of clinical outcomes and survival in HCM patients. In our study, 206 HCM patients were compared with respect to major adverse cardiovascular and cerebrovascular events. By multivariable logistic analysis, we determined that palpitation, together with chronic heart failure (CHF) > 1 year, was an independent predictor of major adverse cardiovascular and cerebral events (MACCE) in HCM patients (OR 3.24, 95% CI 1.60–6.57, P = 0.001). Specially, palpitation was related to higher prevalence of rehospitalization (OR 3.86, 95% CI 2.08–7.08, P < 0.001), cardiac death (OR 2.96, 95% CI 1.05–8.32, P = 0.04) and heart failure exacerbation (OR 4.07, 95% CI 2.04–8.13, P < 0.001). However, patients presented with palpitation did not show a significantly different cardiac phenotype and function. Finally, palpitation predicted a poor prognosis in HCM patients without atrial fibrillation by utilizing Kaplan–Meier analysis (P = 0.041). In conclusion, palpitation could be a new predictor of clinical outcomes and overall survival in HCM patients.

Multivariable logistic regression analyses to identify independent determinants of MAcce in HcM patients. Multivariable analyses indicated that only palpitation and CHF > 1 year were independent determinants of MACCE in HCM patients (Table 2), whereas gender, NYHA class III or IV, diuretics, QRS duration and LVDd were not. The model indicated that HCM patients presented with palpitation were 3.24 times more likely to develop MACCE than those without palpitation (OR 3.24, 95% CI 1.60-6.57, P = 0.001).

Relation of palpitation to LV mass and function or other clinical variables.
To exclude the impact of palpitation on the LV mass and function, which were closely related to clinical outcomes of HCM patients, we explored the difference in LVM and function between two groups. Average LVM and LVMI in the HCM patient cohort were 323.2 ± 101.9 g/m 2 and 184.2 ± 56.6 g/m 2 respectively. Compared with non-MACCE group, LVM, LVMI and LVEF were decreased in MACCE cohorts, but there was no statistic difference ( Supplementary  Fig. 1). Besides, MACCE patients has a significantly higher E/A ratio that non-MACCE patients (P = 0.002). Besides, we further described the characteristics of patients according to the presence of palpitation. We found that clinical, electrocardiographic and echocardiographic features were comparable in patients with or without palpitation (Supplementary Table 1). These results suggested that palpitation was not associated with LV mass, function or other clinical variables. palpitation and clinical outcome in HcM patients. To specifically assess the influence of palpitation on clinical outcomes in HCM cohorts, we compared the relative risk of rehospitalization, cardiac death, heart failure exacerbation and stroke between palpitation group and non-palpitation group. From Table 3, the HCM patients with palpitation had highly presence of rehospitalization (OR 3.86, 95% CI 2.08-7.08, P < 0.001), cardiac death (OR 2.96, 95% CI 1.05-8.32, P = 0.04) and heart failure exacerbation (OR 4.07, 95% CI 2.04-8.13, P < 0.001). However, no difference was observed for stroke between two groups.
The impact of palpitation on overall survival of HCM patients with or without atrial fibrillation (Af). For all HCM patients, there was no significant difference in overall survival between two groups (Fig. 1).
Palpitation is an important clinical manifestation of AF, which affects the quality of life of HCM patients and predicts poor prognosis 11 . Therefore, we investigated the relationship between palpitation and overall survival rate of HCM patients stratified by AF. Compare with HCM patients without palpitation, palpitation patients had a poor survival among HCM patients without atrial fibrillation (P = 0.041). However, the poor prognosis of palpitation was not statistically significant in HCM patients with atrial fibrillation.

Discussion
We reported that palpitation contributed to MACCE in HCM patients, independent on LV mass and LV function. Palpitation was also associated with increased risk of rehospitalization, cardiac death and heart failure exacerbation. Our novel findings were that HCM patients with palpitation have a more poor survival compared with patients without palpitation, but the prognosis of palpitation is only observed in non-AF populations, with no statistical difference in AF populations.
From Table 1, we found that MACCE patients differed in CHF > 1 year (P < 0.001) and NYHA class III or IV (P < 0.001) from non-MACCE cohorts. For a series of common cardiovascular risk, we did not observe significant difference in age, BMI, family history, blood pressure, as well as presence of CAD and hypertension between two groups. Many electrocardiographic and echocardiographic index were not significantly different other than QRS duration (P = 0.025) and LVDd (P = 0.020). Because CHF > 1 year and NYHA class III or IV reflected cardiac function and were closely related to adverse cardiovascular events 12,13 . So, a multivariable logistic regression model was constructed only with variables that proved significant at univariate analysis. From Table 2, we concluded that only CHF > 1 year, NYHA class III or IV and palpitation were independently associated with MACCE of HCM patients. Strikingly, patients presented with palpitation had a 3.24-time risk to develop MACCE than those without palpitation. To confirm whether worse cardiovascular outcomes caused by palpitation were attributed to a poor cardiac phenotype and function 14,15 , we compared the LV mass, LV mass index, LV systolic and LV diastolic function between palpitation groups and non-palpitation groups, which showed no statistical difference between two groups. These results suggested that palpitation on HCM patients was not related to the increased LV mass and decreased LV function. As shown in Table 3, the HCM patients with palpitation had higher risk of rehospitalization, cardiac death and heart failure exacerbation. Although the occurrence of stroke was relatively higher in patients with palpitation than that without palpitation, there was no significant  16 , we explored the association of palpitation in patients in the absence of AF. Interestingly, palpitation predicted a poor prognosis among HCM patients without AF. Our study showed that palpitation was related to more risk of MACCE in HCM patients, including rehospitalization, cardiac death and heart failure exacerbation, which was independent on LV mass and LV function, and had a poor prognosis on overall survival in HCM patients without AF. These results indicated that palpitation could become a new predictor of MACCE in HCM patients and overall survival of HCM cohorts without AF. Some palpitations may be an indication of cardiac arrhythmias 17 , which influence the prognosis of HCM patients. Besides, some benign palpitation could be psychiatric, causing considerable distress and disability of HCM patients 18 . However, the underlying mechanism between palpitation and MACCE of HCM patients needs to be further investigated.
This study had several limitations. This was a cross-sectional study, which includes a long-term follow-up about all-cause death. Although there was obvious difference in clinical outcomes of HCM patients between two groups, the exact causes of adverse cardiac events could not be analyzed. Besides, palpitation was a self-reported clinical presentation, which may not be definitively objective. In addition, we did not have cardiac magnetic resonance data, which may facilitate accurate diagnosis of HCM compared to echocardiography.
In conclusion, our study demonstrated that palpitation increased the risk of MACCE in HCM patients, which was independent of LV mass and LV function, and had a poor prognosis on overall survival in HCM patients without AF. These results indicated that palpitation could become a new predictor of MACCE in HCM patients and overall survival of HCM cohorts.   People's Hospital from January 2010 to December 2015 were retrospectively recruited in this study. 206 patients with complete follow-up data by June 2018 were enrolled in this study. The criteria of a diagnosis of HCM was in accordance with 2014 European Society of Cardiology guidelines on diagnosis and management of hypertrophic cardiomyopathy 19 . It was defined as unexplained LV hypertrophy with a maximal wall thickness above 15 mm showed by any cardiac imaging, such as echocardiography, cardiac magnetic resonance or computed tomography. 206 patients which were admitted to cardiology department from outpatient were all subject to echocardiography, with some subjected to additional echocardiography. All patients with cardiac hypertrophy caused by physiologic cardiac hypertrophy, rheumatic heart disease, mitral valve disease, amyloidosis and Fabry disease were excluded, and some HCM patients with history of renal disease, cancer disease or acute coronary syndrome with cardiogenic shock were also excluded. Palpitations were described as flip-flopping in the chest, rapid fluttering in the chest, and pounding in the neck, which occurred more than once a month and lasted for at least 10 min 20,21  consent to participate. The patients provided their written informed consent to participate in this study.