Natural History of Hypertrophic Cardiomyopathy in Korea: A Nationwide Population-Based Retrospective Cohort Study

Hypertrophic cardiomyopathy (HCM) is a primary disease in which a portion of the myocardium is thickened without any obvious cause, creating functional impairment of the cardiac muscle.1, 2, 3 HCM is generally considered to be of genetic origin,4 with a prevalence of 1 in 500 according to previous studies, which were generally performed in western societies.5 HCM has been associated with a high risk of sudden cardiac death in young adults6 and complications such as heart failure, atrial fibrillation (AF), and stroke.7

Few studies have evaluated the long-term epidemiology of HCM in Korea. This study was designed to evaluate the general epidemiology and natural history of HCM in Korea by analyzing its age-adjusted incidence and prevalence, survival rate (SR), cause of death, and death risk from 2006 through 2017 using data from the Korean National Health Insurance Service (KNHIS).

The mean age of individuals with HCM was 51.4 ± 16.0 years; 47.3 ± 14.9 years for males and 57.5 ± 15.4 years for females (P < 0.001). The 58.9% of the participants were male. In those aged ≥ 50 years, HCM prevalence was approximately 46% in males and approximately 78% in females (P < 0.001). Within the subjects, 56.9% had hypertension, 11.7% had DM, 31.7% had dyslipidemia, 2.63% had MI, 7.15% had AF, 3.52% had ischemic stroke, 1.74% had CKD, and 1.84% had malignant neoplasm. The overall proportion of deaths that occurred during follow-up was 20.4% (n = 8,658), with 37.1% of them attributed to diseases of the circulatory system (M:F = 31.5%:42.0%) and 15.8% attributed to malignant neoplasm (M:F = 21.7%:10.7%) (Table 1).

Table 1
The distribution of general characteristics, comorbidities, socioeconomic factors, and causes of death in hypertrophic cardiomyopathy by sex (n = 42,352)

• Click for larger image
• Click for full table
• Download as Excel file

Patients with HCM demonstrated a significantly elevated all-cause mortality. This elevated risk in males remained consistent even after adjusting for income and comorbidities. Based on the 20s age group, the adjusted hazard ratios (HRs) for HCM in the different age groups were as follows: 3.67 (95% confidence interval [CI], 2.65–5.10) for 0–9 years, 1.39 (95% CI, 1.05–1.83) for 30–39 years, 2.01 (95% CI, 1.55–2.60) for 40–49 years, 3.88 (95% CI, 3.02–4.99) for 50–59 years, 9.40 (95% CI, 7.33–12.0) for 60–69 years, 22.2 (95% CI, 17.3–28.5) for 70–79 years, and 48.2 (95% CI, 37.0–62.7) for those older than 80 years. Additionally, the adjusted HR for the lower income level was 1.40 (95% CI, 1.33–1.48), and that for the medium income level was 1.18 (95% CI, 1.12–1.25). DM, MI, AF, stroke, CKD, and malignant neoplasm carried an increased risk of death during follow-up. Conversely, the death risk associated with hypertension and dyslipidemia was significantly lower (Table 2).

Table 2
All-cause death for associated with hypertrophic cardiomyopathy during a decade (2006–2018) (n = 42,352)

• Click for larger image
• Click for full table
• Download as Excel file

Fig. 1 shows the age-standardized incidence and prevalence of HCM. The overall age-standardized incidence of HCM increased from 7.84 to 8.62 persons per 100,000 persons from 2006 to 2017, respectively (Fig. 1A). Although the age-standardized incidence in the 0–9, 10–19, 20–29, 30–39, 40–49, and 50–59 age groups increased from 2006 to 2017, it decreased in the 60–69, 70–79, and 80 and older age groups during the same period (Fig. 1B) (Supplementary Table 1). The overall age-standardized prevalence also rose, from 18.2 to 36.4 persons per 100,000 persons, from 2006 to 2017, respectively (Fig. 1C). Likewise, the age-standardized prevalence within age groups increased during the same period (Fig. 1D, Supplementary Table 2).

Fig. 1
Age-standardized incidence and prevalence of HCM overall, by age group, sex, and year per 100,000 persons between 2006 and 2017. (A) Age-standardized incidence of hypertrophic cardiomyopathy overall (middle), by sex (male: upper; female: lower), and by year per 100,000 persons between 2006 and 2017 and number of HCM patients by year and sex. (B) Age-standardized incidence of hypertrophic cardiomyopathy by year and age group per 100,000 persons between 2006 and 2017. (C) Age-standardized prevalence of hypertrophic cardiomyopathy overall (middle), by sex (male: upper; female: lower), and by year per 100,000 persons between 2006 and 2017 and number of HCM patients by year and sex. (D) Age-standardized prevalence of hypertrophic cardiomyopathy by year and age group per 100,000 persons between 2006 and 2017.
HCM = hypertrophic cardiomyopathy.

• Click for larger image
• Download as PowerPoint slide

Fig. 2 shows the 10-year SR of HCM, which was 70.3% (95% CI, 69.7–70.9). The 10-year SR related to cardiovascular death was 1.10% (95% CI, 0.80–1.50), and the SR for non-cardiovascular death was 4.62% (95% CI, 4.09–5.20) (P < 0.001). The 10-year SR declined with increasing age (P < 0.001) (Fig. 2B). The 10-year SR differed significantly by sex: 75.9% (95% CI, 75.1–76.7) in males and 62.5% (95% CI, 61.5–66.3) in females (P < 0.001) (Fig. 2A) (Supplementary Table 3).

Fig. 2
Survival curve of HCM overall, and by sex, age group, and cause of death from 2006 through 2017. (A) Survival curve of HCM overall, by sex, and by cause of death. (B) Survival curve of HCM by age group.
HCM = hypertrophic cardiomyopathy.

• Click for larger image
• Download as PowerPoint slide

Our study is the first to elucidate the death risk of HCM patients using data from a large-scale national health insurance system. HCM patients exhibited a significantly elevated death risk across all age groups, and the risk was particularly high for those with male sex, lower income levels, DM, MI, AF, stroke, CKD, and malignant neoplasm. We also found that the overall 10-year SR for HCM was approximately 70%. The 10-year SR was lower in females than in males, and it decreased with advancing age. The major cause of death among HCM patients was disease of the circulatory system. During the course of a decade, the overall age-standardized incidence and prevalence of HCM increased notably.

There was a significantly higher death risk across all age groups, except for the 10–19 years. This result is similar to a study of 1,047 HCM patients in Belgium and the Netherlands.14 In addition, the results of this study correspond well with those of an earlier study of 598 Taiwanese HCM patients with AF, which showed that the death risk increased significantly as age increased.15 However, it was difficult to make comparisons with our study due to the scarcity of previous research that divided HCM patients into age groups. Even the study in Belgium and the Netherlands14 did not classify the patients by age group.

The HR for HCM was significantly higher in males than that in females. The crude HR for males was significantly lower, but the age-adjusted HR for males increased. This elevated risk in males remained consistent even after adjusting for income and comorbidities. HCM has consistently shown a male predominance, with men composing approximately 60% of most published HCM cohorts.16, 17, 18, 19 Interestingly, the 10-year SR in HCM was lower in females than in males. This result is in contrast to those of several previous smaller-scale studies. A study of 576 Chinese HCM patients from 2008 to 201620 and 1,007 Dutch HCM patients from 1977 to 2017 showed that sex was not independently associated with the death risk.21 Another study from Korea, which used KNHIS data from 2010–2016 to analyze 9,524 HCM patients, evaluated 1-year follow-up results and found no difference in the HR of all-cause death between males and females.22 However, the former two studies included much fewer subjects than we did, and the results of the latter study are based on a very short follow-up duration. Meanwhile, this may have been due to the fact that the SR is analyzed in a univariate analysis only, or the fixed time period of 10 years resulted in a lower SR in female. This will need further elucidation in a study with a longer time period. Also, the adjusted HR of patients with a low income level was significantly higher than that for patients with up larger incomes in this study, where no previous studies comparable to ours with respect to socioeconomic position could be found.

Stroke was found to increase the risk of death in HCM in this study. This finding is in close agreement with a Taiwanese study on HCM with AF, which showed that the HR of TIA/ischemic stroke was 2.82.15 CKD also increased the risk of all-cause mortality during follow-up in HCM patients. This coincides well with a study of 581 Chinese HCM patients from 2008 to 2016, in which the HR of CKD was 3.42, and it was independently associated with all-cause mortality.23 A study of 491 Japanese HCM patients between 2003 and 2016 also found that those with renal dysfunction had a significantly higher risk of sudden death or potentially lethal arrhythmic events.24 For AF, our results are in agreement with the results of a study of 3,673 adult HCM patients in the United States (US) from 1975 to 2012, which showed that AF with HCM presented a significantly higher HR.16

Interestingly, HCM patients with hypertension or dyslipidemia showed a lower HR for death during follow-up in our study. This may be attributed to the more aggressive control of cardiovascular disease risk factors in those subjects due to their HCM diagnosis. More than 70% of the HCM age group was in their 40s or older. Due to limitations in the KNHIS data, we did not analyze the distribution of anti-hypertensive medication and lipid-lowering agents.10 Meanwhile, in a study on 468 Chinese HCM patients from 2008 to 2018, HCM-related death did not differ significantly between the hypertension and non-hypertension groups, suggesting that hypertension had a negative effect on the clinical prognosis of HCM patients.25 These findings warrant further study in other populations for confirmation.

The 10-year SR of HCM was about 70%, and the 10-year SR in females was lower than that in males. This result is in general agreement with the results of a study of 2,506 adults who received septal myectomy for obstructive HCM from 1961 through 201626; the 10-year SR in obstructive HCM patients at the Mayo Clinic was around 80%, which was higher than in our study. To compare the SR of HCM patients to that in the general population, we calculated the percentage of deaths at each age divided by the total number of deaths per year from 2006 through 2018 using Korean census data and drew the survival curve. The SR for the age groups from 0–4 years to 65–69 years was more than 92%, and the rate for the age groups from 70–74 years to 80–84 years was around 80% (Supplementary Fig. 1), showing that the SR of Korean HCM patients was lower than that of the general Korean population in all age groups.

The two most common causes of death for HCM patients were diseases of the circulatory system and malignant neoplasms. More than one-third of all HCM deaths resulted from cardiac disease. Around one-third of the males and about half of the females died from cardiac disease. The 10-year SR in cardiovascular death vs. non-cardiovascular death was about 1% vs. 5% (P < 0.001) in this study. Additional analysis using Cochran-Mantel-Haenszel test showed significant differences in cardiovascular and non-cardiovascular death by age group (P = 0.001) (Supplementary Table 4). The second most common cause of death was malignant neoplasm, which was understandably consistent with reports showing that the most common cause of death in Korea is malignant neoplasm.9, 27

The proportion of males was about 60% in this study. Interestingly, approximately 70% of males were in their 30s to 50s, and more than 70% of the females were in their 50s to 70s. The results of this study align well with those of an earlier study involving 2,123 US HCM patients (38% female) from 2001 to 2016,28 and they also coincide with a study in the Netherlands involving 1,007 HCM patients from 1997 through 2017 (62% male).21 These results all show that females present with HCM at an older age than males.

The results of this study showing that males had a higher risk of death and a higher SR than females and that females had a late onset are of special interest. The underlying mechanism of sexual dimorphism in HCM remains elusive, but sex hormones might contribute to cardiac remodeling and the lifetime risk of cardiovascular disease.29, 30, 31 Sex hormones improve vascular function and contribute to a favorable lipid profile in women, and pregnancy-related hormonal shifts lead to reversible cardiovascular adaptations. In a study involving 292 adult HCM patients undergoing exercise echocardiography,32 women with HCM were older than men and has a lower incidence of left ventricle outflow tract obstruction, but displayed smaller left ventricles and higher left-sided filling pressures. Also, elevated levels of heat shock proteins typically found in females might contribute to their later disease onset and the reduced protein turnover typically found in men, since heat shock protein have cardio-protective effects and protect tissue from accumulation of damaged proteins that affects proper cellular function.33 One of our study limitations is that we were unable to analyze genetic factors related to HCM, but it is known that the age of HCM diagnosis differs by sex depending on genotype.34 In a study of 3,673 HCM patients, women showed lower exercise capacity and predicted peak oxygen uptake than men,16 and proteomics analyses conducted in the Netherlands suggest that increased levels of tubulin partly underlie the more severe diastolic dysfunction seen in female compared to male,33 but neither studies definitely showed that these factors are associated with poorer clinical outcomes. However, there have been other studies showing contrary results, with higher HCM-related complication and death risks in women,32, 35 which demonstrate that further studies are needed for clarification.

In addition to the aforementioned biological explanations for the differences in HR, SR, and age of diagnosis by sex, an explanation that is more socioeconomical than biological might also be relevant. The economic activity participation rate of individuals in their 50s increased from 54.7% in 2006 to 62.2% in 2014 among women and from 85.4% in 2006 to 89.4% in 2014 among men in Korea.36 Health examinations are conducted every 2 years for white-collar workers and every year for blue-collar workers, in accordance with following occupational safety and health laws in Korea.37, 38 Therefore, men older than 40 are more likely than women to be diagnosed with certain diseases, including HCM, during health checkups at their workplaces. We show that the larger proportion of HCM among women in their 60s and older in our study is due to their increasing opportunities for contact with health services due to the life turning-point health examination at age 66 that has been offered by the KNHIS since 2007. Furthermore, the proportion of elderly patients receiving the life turning-point health examination increased from approximately 51% in 2007 to 76% in 2011.11 However, we could not calculate the risk for sudden cardiac death according to types of HCM (i.e., septal/apical/diffuse, etc.) due to the lack of data on cardiac imaging such as echocardiography in this KNHIS dataset, which is a limitation of this study.

The age-standardized incidence and prevalence of HCM increased during the decade of our evaluation. The results of this study resemble those of an earlier study that examined KNHIS data between 2010 and 2016 and found that HCM increased the prevalence and incidence during the study period.39

Our results have important clinical and public health implications. The age-standardized incidence and prevalence in males were higher than those in females, and the change in age-standardized incidence differed substantially during the 10 years of study between the relatively young groups (age 0–59) and the elderly groups (age 60 and older). The adjusted HR showed relationships with increasing age, male sex, lower income level, and comorbidities (DM, MI, AF, stroke, CKD, and malignant neoplasm). According to these results, more appropriate health resources should be allocated to those factors to improve the diagnoses, treatments, research, and health policies that affect HCM.

This study has several limitations. First, the National Health Insurance benefit records might have missed HCM patients who did not use medical services or paid for their own medical expenses. For this reason, the incidence or prevalence of HCM in this study might be under- or over-estimated. Second, our data included only HCM diagnoses in general and did not differentiate the obstructive and non-obstructive HCM in our analysis. Additionally, we were unable to account for information that can be retrieved from imaging studies such as magnetic resonance imaing,40 and echocardiography,41 or HCM-related genetic42, 43 information due to data limitations. However, in this study, we only sought to find the overall demographic characteristics of HCM in Korea and its evolution during the period, and we think the data have its value in itself in presenting a picture of the past and present status of the disease at this time.

The death risk associated with HCM increased with increasing age, male sex, and a lower income level. The risk was also higher in patients who had comorbid DM, MI, AF, stroke, CKD, or malignant neoplasm. The overall 10-year SR for HCM was 70.3%, which is substantially lower than the SR of the general population. The major cause of death for HCM patients was disease of the circulatory system. These patterns indicate that HCM should be considered a genetic or degenerative disease with effects that increase with age. Our findings should be considered in future research designs and policies addressing healthcare for HCM patients.

Supplementary Table 1

Age-standardized incidence^a and 95% CI of hypertrophic cardiomyopathy overall and by sex (per 100,000 persons)

Supplementary Table 2

Age-standardized prevalence^a and 95% CI of hypertrophic cardiomyopathy overall and by sex (per 100,000 persons)

Supplementary Table 3

Survival rate (%) and 95% confidence interval for hypertrophic cardiomyopathy overall, by sex, by cause of death, by age group and by age group in sex

Supplementary Table 4

Distribution of cardiovascular and non-cardiovascular death by age group

Supplementary Fig. 1

Survival curve by year and age group in Korea from 2006 through 2018. The survival curve was calculated based on the percentage of deaths at each age divided by the total number of deaths per year from 2006 through 2018 according to the census in Korea.