Obesity and metabolic health status are determinants for the clinical expression of hypertrophic cardiomyopathy

Aims

We sought to investigate the association of obesity and metabolic health status with the incidence of clinical hypertrophic cardiomyopathy (HCM) diagnosis in the general population. Our goal was to identify modifiable risk factors to attenuate clinical expression of HCM, enabling management evolution from a mostly passive strategy of risk stratification to a proactive strategy of modifying disease expression.

Methods

Using nationwide population-based data from the Korean National Health Insurance Service, 28,679,891 people who were free of prevalent HCM and who underwent health examinations between 2009 and 2015 were followed until 31 December 2016. The primary outcome was clinical HCM that was defined as incident diagnosis of HCM during the follow-up, after a blanking period of 12 months.

Results

Over a median follow-up of 5.2 years, 0.027% (n = 7851) of the study participants were diagnosed as incident HCM. The incidence rate per 1000 person-years was 0.059. A significant association was found between body mass index (BMI) and the incidence of clinical HCM after multivariate adjustment, with a hazard ratio per 1 kg/m² increase in BMI of 1.063 (95% confidence interval 1.051–1.075). Metabolically unhealthy participants had a greater incidence of HCM than metabolically healthy participants, regardless of obesity status. The effect of BMI was more pronounced in several subgroups, including participants with no hypertension, those aged less than 65 years and men.

Conclusion

We found that individuals with obesity and/or metabolic abnormalities had a significantly higher incidence of clinical HCM diagnosis than their counterparts. Efforts to manage obesity and metabolic abnormalities may be important in modifying clinical expression of HCM.

Obesity, metabolic health status, hypertrophic cardiomyopathy

Introduction

Obesity is a major contributor to the risk of most forms of cardiovascular (CV) diseases (CVDs), including left ventricular (LV) hypertrophy (LVH) and sudden cardiac death.^1,2 Furthermore, since obesity is reaching epidemic proportions in the general population, its importance has been particularly emphasized as a common and potentially remedial cause of CVDs. Regarding the role of obesity in hypertrophic cardiomyopathy (HCM), a few animal studies have examined obesity or anti-obesity effects on the development or resolution of HCM. Specifically, a mouse study demonstrated that high-fat diet-induced obesity could result in a cardiac hypertrophic response through inactivation of Foxo3a, which is a crucial transcription factor countering hypertrophic growth by triggering an apoptosis-related gene program.³ It has also been reported that the implementation of voluntary wheel running significantly reduced hypertrophic remodeling and myofiber disarray in a mouse model of HCM,⁴ suggesting that efforts to control obesity might lead to a regression of LVH. Human data from Olivotto et al.⁵ reported that obesity, as assessed by body mass index (BMI), was highly prevalent in 275 HCM patients: 38% were overweight and 37% were obese.⁵ They also found that obesity was independently related to greater LVH and greater symptom burden,⁵ implying that obesity might influence phenotypic variability and clinical course in HCM during the lifespan of the patient. Given the number of HCM patients enrolled in this study, however, further studies with larger sample sizes are clearly required to establish whether there is an association between obesity and the risk of clinical expression of HCM, and whether metabolic health status (i.e., the presence or absence of hypertension, diabetes mellitus, and dyslipidemia) could contribute to or confound the association of obesity with HCM. Such studies will help identify participants at high risk of developing clinical HCM, and also provide data to create a prevention strategy, i.e. risk factor modification, to reduce the incidence and lethality of HCM across the population.

Hence, we sought to assess: (a) the association of obesity, measured by BMI or waist circumference (WC), with the incidence of clinical diagnosis of HCM; and (b) whether the association of obesity with HCM differs depending on metabolic health status.

Methods

Study population

This nationwide population-based cohort study used the database from the Korean National Health Insurance Service (NHIS), which holds anonymized health-related information from approximately 97% of the Korean population.⁶ Briefly, all eligible Korean adults, except 3% of participants with low income who are covered by the Medical Aid program, were recommended to undergo a standardized biennial health checkup. The checkup consisted of detailed surveys of demographics, medical histories and health-related behaviors, vital sign and anthropometric measurements, and laboratory tests. For laboratory tests, quality control of data was conducted in accordance with the procedures of the Korean Association of Laboratory Quality Control. We assessed data from the records of 28,888,616 Korean residents aged at least 20 years who had undergone at least one biennial medical evaluation between 2009 and 2015. Of them, 143,139 were excluded for missing covariates. To minimize the reverse causality bias resulting from changes in obesity or metabolic health status caused by sedentary lifestyle in patients with preexisting HCM, we excluded 7961 participants with a history of HCM before the index year. A priori, we excluded 1528 participants who were diagnosed with HCM during the first year from baseline, because HCM diagnosed shortly after obesity and/or metabolic abnormality are identified is unlikely to be a consequence of obesity and/or metabolic abnormality. We also excluded 56,097 participants in whom diagnosis of HCM was not accepted to be true because the rare intractable diseases (RIDs) program (details given in the following) was rejected after strict review by medical experts and insurance claims professionals. This yielded a final study population of 28,679,891 individuals (Supplemental Figure 1).

Variables and definitions

The primary endpoint was the newly diagnosed clinical HCM, which was defined using the International Classification of Diseases, 10th revision (ICD-10) code (I42.1, I42.2). In Korea, HCM has been categorized as a RID since 2006 for the purpose of subsidizing the medical expenses of patients suffering from these diseases. Since the Korean government covers 90% of the medical expenses for patients diagnosed with RIDs,⁷ this disease category is strictly regulated by the government. Specifically, physicians are mandatorily requested to fill out the application form for RID registration for patients who have been diagnosed with HCM. The application also contains detailed information, including the ICD-10 code, date of diagnosis, diagnostic methods used (echocardiography or cardiac magnetic resonance imaging), and the name and license number of physicians who confirmed the diagnosis. After thorough reviewing and monitoring the application, the NHIS gives the patients a RID code (V127) to certify their diagnoses of HCM. In this regard, data regarding the RID are considered validated and reliable, and thus our study only included patients diagnosed with HCM, the diagnosis of which was clearly certified as a RID in the claims database. The validity of the RID code for HCM was independently assessed by examining the medical records and imaging of a random sample of 1100 subjects, revealing that its sensitivity, specificity, and accuracy were found to be 91.5%, 100%, and 92.5%, respectively.⁸ Analyses were performed using decile categories of BMI and WC, and data on the cut-off points defining deciles are summarized in Supplemental Table 1. Participants were also categorized into nine BMI levels (2 kg/m² interval) and six WC levels (10 cm interval) (see details in supplemental methods). The metabolically healthy obese subgroup was defined as overweight or obese participants with no demonstrable obesity-related metabolic abnormality, such as diabetes, hypertension, or dyslipidemia.⁹ We defined individuals as having diabetes, hypertension, and dyslipidemia by coded diagnoses recorded in the NHIS claim data in which the information accuracy was validated previously.¹⁰

Statistical analysis

Descriptive statistics are presented as means ± standard deviation or median (interquartile ranges) for continuous variables and numbers (percentages) for categorical variables. For the comparison between groups, unpaired Student's t-test was applied for continuous variables and theχ² test or Fisher's exact test was used for categorical variables as appropriate. Cox proportional-hazards models were used to assess the association between BMI and the incidence of clinical HCM. Multivariate regression analyses were performed using the nine BMI categories. Hazard ratios (HRs) and 95% confidence interval (CI) were calculated in an unadjusted model after adjustment for age, sex, and smoking status (model 1) and after adjustment for age, sex, smoking status, systolic blood pressure, and levels of total cholesterol, high-density lipoprotein cholesterol, and glucose (model 2). In sensitivity analyses, we adjusted for age, sex, smoking status, alcohol drinking status, physical activity, and income level (model 3) and also for age, sex, and WC (model 4), which are presented in supplemental information. The interactions between variables were tested. Analyses were performed separately in men and women.

Two-sided p values < 0.05 were considered statistically significant. Statistical tests were performed using SAS version 9.2 (SAS Institute, Cary, NC, USA).

Ethics

This study complied with the Declaration of Helsinki and was exempt from review by the Seoul National University Hospital Institutional Review Board (1804-043-935) owing to the retrospective nature of data collection allowing the maintenance of participant confidentiality.

Results

Baseline characteristics

The mean BMI and WC were 23.7 ± 3.3 kg/m², and 79.9 ± 9.4 cm, respectively, and both were higher in men than women (BMI: 24.2 ± 3.1 versus 23.1 ± 3.5 kg/m², p < 0.0001; WC: 83.7 ± 8.1 versus 76.1 ± 9.2 cm, p < 0.0001). Among a total of 28,679,891 study participants, 31.8% (n = 9,116,181) were obese and 27.1% (n = 7,772,503) had metabolic syndrome, while 49.6% and 50.4% of obese participants were classified as “metabolically healthy” and “metabolically unhealthy,” respectively. Table 1 summarizes baseline characteristics of study participants according to their obesity and metabolic health status. Metabolically healthy obese (BMI≥25 kg/m²) participants were younger and had lower values of BMI and WC than metabolically unhealthy obese ones (all p < 0.0001). Compared with metabolically healthy non-obese (BMI<23 kg/m²) participants, metabolically healthy obese subjects were more likely to be current smokers and heavy drinkers, and had higher blood pressure, cholesterol, and glucose levels (all p < 0.0001).

Table 1.

Baseline characteristics of participants according to BMI and metabolic health status.

	BMI<23		23≤BMI<25		BMI≥25
	Metabolically healthy (n = 9,653,051)	Metabolically unhealthy (n = 3,141,549)	Metabolically healthy (n = 4,016,461)	Metabolically unhealthy (n = 2,752,649)	Metabolically healthy (n = 4,251,023)	Metabolically unhealthy (n = 4,865,158)
Age (years)	40.3 ± 13.2	56.7 ± 13.9	43.5 ± 12.5	56.1 ± 12.5	42.7 ± 12.3	53.9 ± 12.9
Male	3,744,922 (38.8)	1,473,806 (46.9)	2,287,826 (57.0)	1,503,201 (54.6)	2,643,227 (62.2)	2,741,372 (56.4)
Urban residence	4,561,865 (47.3)	1,421,642 (45.3)	1,846,258 (46.0)	1,257,821 (45.7)	1,903,757 (44.8)	2,153,927 (44.3)
BMI (kg/m2)	20.7 ± 1.6	21.2 ± 1.5	24.0 ± 0.6	24.0 ± 0.6	27.2 ± 2.1	27.7 ± 2.4
WC (cm)	72.1 ± 6.5	76.1 ± 6.5	80.2 ± 5.6	82.4 ± 5.6	87.2 ± 7.0	89.6 ± 7.3
SBP (mmHg)	113.9 ± 11.4	128.4 ± 17.1	118.1 ± 10.9	130.3 ± 16.0	120.9 ± 10.4	132.8 ± 15.6
DBP (mmHg)	71.2 ± 1.2	79.4 ± 11.2	73.6 ± 7.8	80.7 ± 10.7	75.4 ± 7.4	82.5 ± 10.7
Hypertension	0 (0)	1,917,446 (61.0)	0 (0)	1,803,463 (65.5)	0 (0)	3,496,473 (71.9)
Diabetes mellitus	0 (0)	706,689 (22.5)	0 (0)	649,277 (23.6)	0 (0)	1,259,935 (25.9)
Dyslipidemia	0 (0)	1,516,268 (48.3)	0 (0)	1,412,217 (51.3)	0 (0)	2,510,958 (51.6)
Atrial fibrillation	4062 (0.04)	9575 (0.3)	2051 (0.05)	7509 (0.27)	2175 (0.05)	13,032 (0.27)
CHF	16,856 (0.17)	56,900 (1.81)	7737 (0.19)	43,507 (1.58)	9213 (0.22)	86,140 (1.77)
Myocardial infarction	2661 (0.03)	7413 (0.24)	1182 (0.03)	5542 (0.2)	1352 (0.03)	9420 (0.19)
Stroke	2585 (0.03)	12,284 (0.39)	1110 (0.03)	8773 (0.32)	1114 (0.03)	13,478 (0.28)
ESRD	2257 (0.02)	11,840 (0.38)	605 (0.02)	4905 (0.18)	455 (0.01)	5536 (0.11)
Physical activity
Walk	4,495,944 (46.6)	1,455,627 (46.4)	1,895,486 (47.3)	1,302,563 (47.4)	1,996,616 (47.0)	2,223,334 (45.8)
Moderate	1,568,894 (16.3)	598,795 (19.1)	779,939 (19.4)	573,030 (20.8)	814,801 (19.2)	953,718 (19.6)
High	1,134,776 (11.8)	452,658 (14.4)	629,463 (15.7)	458,004 (16.7)	661,284 (15.6)	761,189 (15.7)
Smoking
Never	6,618,829 (68.6)	1,996,395 (63.6)	2,352,708 (58.6)	1,652,386 (60.0)	2,291,225 (53.9)	2,832,496 (58.2)
Former	827,497(8.6)	408,841 (13.0)	556,580 (13.9)	474,490 (17.2)	632,651 (14.9)	860,831 (17.7)
Current	2,206,725 (22.9)	736,313 (23.4)	1,107,173 (27.6)	625,773 (22.7)	1,327,147 (31.2)	1,171,831 (24.1)
Alcohol
0 g/day	5,079,706 (52.6)	1,930,831 (61.5)	1,930,599 (48.1)	1,569,688 (57.0)	1,957,063 (46.0)	2,665,547 (54.8)
1–30 g/day	4,125,856 (42.7)	1,007,373 (32.1)	1,812,246 (45.1)	978,106 (35.5)	1,937,730 (45.6)	1,764,778 (36.3)
>30 g/day	447,489 (4.6)	203,345 (6.5)	273,616 (6.8)	204,855 (7.4)	356,230 (8.4)	434,833 (8.9)
Low income	2,863,963 (29.7)	917,149 (29.2)	1,077,584 (26.8)	737,553 (26.8)	1,148,312 (27.0)	1,313,294 (27.0)
Hemoglobin (mg/dl)	13.6 ± 1.6	13.6 ± 1.6	14.1 ± 1.6	14.0 ± 1.6	14.4 ± 1.6	14.2 ± 1.6
Glucose (mg/dl)	89.7 ± 10.3	106.4 ± 34.7	92.1 ± 10.7	107.6 ± 32.7	93.6 ± 11.1	109.7 ± 32.9
TC (mg/dl)	180.2 ± 27.7	208.0 ± 45.1	187.8 ± 27.3	210.2 ± 44.2	191.5 ± 26.8	210.6 ± 43.6
LDL-C (mg/dl)	102.0 ± 25.8	123.5 ± 41.2	109.6 ± 26.1	125.1 ± 40.7	112.0 ± 26.3	124.0 ± 40.3
HDL-C (mg/dl)	59.8 ± 15.6	58.1 ± 18.1	54.4 ± 15.3	54.1 ± 16.9	51.3 ± 15.0	51.9 ± 16.3
Triglyceride* (mg/dl)	82.1 (82.1–82.1)	115.8 (115.8–115.9)	105.1 (105.1–105.2)	136.5 (136.4–136.6)	124.8 (124.8–124.9)	154.9 (154.8–155.0)
Creatinine (mg/dl)	0.91 ± 0.73	0.95 ± 0.69	0.96 ± 0.72	0.97 ± 0.66	0.97 ± 0.66	0.97 ± 0.62
GFR (ml/min/1.73m2)	95.3 ± 43.1	87.1 ± 37.4	92.4 ± 45.3	86.0 ± 39.1	92.2 ± 46.7	86.2 ± 41/0

Values given as number (percentage), mean ± standard deviation, or median (interquartile range) unless otherwise indicated.

BMI, body mass index; CHF, congestive heart failure; DBP, diastolic blood pressure; ESRD, end-stage renal disease; HDL-C, high-density lipoprotein-cholesterol; LDL-C, low-density lipoprotein-cholesterol; SBP, systolic blood pressure; TC, total cholesterol; WC, waist circumference.

*

Geometric mean.

Association of obesity with incidence of clinical HCM

During a median follow-up of 5.2 years (interquartile range: 3.4–6.1 years), only 0.027% (n = 7851) of the total study population were newly diagnosed with clinical HCM with an incidence rate per 1000 person-years of 0.059. The median interval from the baseline examination to HCM diagnosis was 3.0 years (interquartile range: 1.5–4.4 years).

As shown in Figure 1, there was a positive linear relationship between BMI and the incidence of clinical HCM in total population after adjusting for potential confounding factors; the incidence of HCM was approximately 2.6 times greater in patients in the highest decile (BMI ≥ 27.9 kg/m²) than those in the lowest decile (BMI < 19.6 kg/m²) in model 1 (HR = 2.690, 95% CI 2.419–2.992, p < 0.0001), model 2 (HR = 2.609, 95% CI 2.342–2.906, p < 0.0001) (Figure 1(a) and (b)), and model 3 (HR = 2.692, 95% CI 2.420–2.994, p < 0.0001) (Supplemental Figure 2(a)). The association of BMI with the incidence of HCM was attenuated in model 4 (HR = 1.812, 95% CI 1.592–2.063, p < 0.0001) (Supplemental Figure 2(b)). The results remained similar when stratified by sex, with higher HRs in men than women. When the analyses were repeated by using WC, WC had positive linear associations with the incidence of clinical HCM, with a bit weaker but still significant after adjustment for BMI (Figure 1(c) and (d) and Supplemental Figure 2(c) and (d)). These associations were consistent in analyses with nine BMI grades (2 kg/m² interval) and six WC grades (10 cm interval) (Supplemental Figure 3). When BMI or WC was entered into the analysis as a categorical variable based on the World Health Organization classification for Asians or as a continuous variable, they emerged as an independent determinant for the diagnosis of clinical HCM in four multivariate models (Table 2 and Supplemental Table 2).

Figure 1.

Association of BMI and WC with incidence of clinical HCM.

HRs and 95% CIs are shown for the association of BMI (a and b) and WC with the incidence of clinical HCM (c and d). The analyses were adjusted for age, sex, and smoking status (model 1, a and c) and for age, sex, smoking status, systolic blood pressure, and levels of total cholesterol, high-density lipoprotein cholesterol, and glucose (model 2, b and d). BMI=body mass index; CI=confidence interval; HCM=hypertrophic cardiomyopathy; HR=hazard ratio; WC=waist circumference.

Table 2.

Association of BMI and WC with incidence of clinical HCM.

	Events, n	IR	Model 1*, HR (95% CI)	Model 2†, HR (95% CI)
Body mass index‡, kg/m2
<18.5	118	0.02234	0.770 (0.639–0.927)	0.779 (0.647–0.939)
18.5–22.9	1782	0.03398	1 (reference)	1 (reference)
23.0–24.9	2029	0.0635	1.549 (1.454–1.651)	1.530 (1.435–1.631)
25.0–39.9	3435	0.09196	2.209 (2.086–2.34)	2.152 (2.029–2.283)
≥30.0	487	0.09945	3.004 (2.717–3.321)	2.865 (2.585–3.174)
Per 1 kg/m2 increase			1.117 (1.110–1.124)	1.112 (1.105–1.12)
Waist circumference§, cm
<90 in men and <85 in women	4848	0.0460	1 (reference)	1 (reference)
≥90 in men and ≥85 in women	3003	0.1134	1.818 (1.736–1.903)	1.737 (1.657–1.822)
Per 1 cm increase			1.044 (1.041–1.046)	1.042 (1.039–1.045)

CI, confidence interval; HR, hazard ratio; IR, incidence rate per 1000 person-years.

*

HRs have been adjusted for age, sex, and smoking status.

†

HRs have been adjusted for age, sex, smoking status, systolic blood pressure, total cholesterol, high-density lipoprotein cholesterol, and glucose.

‡

The World Health Organization recommendations for Asians was used to categorize participants by their obesity status; underweight (body mass index<18.5 kg/m²), normal (18.5–22.9 kg/m²), overweight (23.0–24.9 kg/m²), obese class I (25.0–29.9 kg/m²), and obese class II (≥30.0 kg/m²).

§

Abdominal obese was defined as waist circumference ≥90 cm in men and ≥85 cm in women, respectively.

Effect of metabolic health status on associations between obesity and clinical HCM diagnosis

Participants with a metabolically unhealthy status had a higher incidence of clinical HCM diagnosis than those with a metabolically healthy status, in all BMI strata (non-obese, overweight, and obese; see Figure 2 and Supplemental Figure 4). Similar results were found in all multivariate models. When study participants were stratified by age (<40, 40–49, 50–59, ≥ 60 years), the incidence of clinical HCM diagnosis increased progressively across the age groups (incidence rates per 1000 person-year, 0.015, 0.037, 0.070, and 0.136, respectively). In each age group, the HCM incidence was lowest in metabolically healthy non-obese participants and highest in metabolically unhealthy obese patients, with intermediate incidences in the other groups (Supplemental Table 3).

Figure 2.

Incidence of HCM according to BMI and metabolic health status.

Forest plot of the graded risk for incident hypertrophic cardiomyopathy across strata of obesity category and metabolic health status in multivariate adjusted models (a for model 1 and b for model 2). Participants were categorized into non-obese (BMI<23 kg/m²), overweight (BMI 23.0–24.9 kg/m²), and obese (BMI≥25.0 kg/m²), and then further sub-stratified by metabolic health status. BMI=body mass index; HCM=hypertrophic cardiomyopathy; HR=hazard ratio; MH=metabolically healthy; MUH=metabolically unhealthy.

Subgroup analyses

A greater BMI was associated with a higher incidence of clinical HCM across the subgroups with similar directional trends in multivariate analyses (Supplemental Figure 5). Significant interactions were present for BMI with age, sex, the presence or absence of hypertension, diabetes, dyslipidemia, WC, smoking status, and alcohol drinking status. The effect of BMI was relatively more pronounced in men, individuals younger than 65 years, those without hypertension, current smokers, heavy drinkers, and those with diabetes or dyslipidemia.

Discussion

The main findings of this study were as follows:

1. during a median follow-up of 5.2 years, only 0.027% (n = 7851) of the total study population were diagnosed as incident cases of clinical HCM;

2. higher values of BMI and WC were significantly associated with a higher incidence of clinical HCM;

3. metabolically unhealthy overweight and obese individuals were significantly and synergistically associated with a higher incidence of clinical HCM versus their metabolically healthy counterparts.

These findings suggest that the occurrence rate of clinically identified patients with HCM is similar to that previously reported in the United States (0.031%),¹¹ and more importantly, that obesity and metabolic abnormalities play a role as potential arbitrators for clinical expression of HCM. This has a clinical implication given there is no intervention that modifies or prevents clinical expression of HCM in the contemporary era.

Considerable evidence demonstrates that obesity is strongly associated with LV concentric remodeling and hypertrophy. Specifically, autopsy studies of morbid obesity (BMI≥40.0 kg/m²) showed that all obese participants had increased heart weight and increased LV wall thickness.¹² Another study corroborated the association between obesity and LVH by demonstrating that myocyte hypertrophy was found in more than two-third of biopsies of obese patients.¹³ In echocardiographic studies, where the degree of obesity varied from class I (BMI 30.0–34.9 kg/m²) to class III (BMI≥40.0 kg/m²), LV volume, LV wall thickness, and LV mass or mass index were significantly greater in obese patients than their leaner counterparts.^14,15 It is likely that the effect of obesity on the myocardium would not be different according to the genetic background of the patient. One might anticipate that the magnitude of the effect could be even more potent in HCM mutation carriers than in non-carriers, given the increased genetic susceptibility toward development of phenotypically overt HCM. However, data describing the effects of obesity on the risk of developing clinical HCM are rare. The present study, the largest one to address this question, showed that the incidence of clinical HCM was significantly higher in overweight or obese participants and suggests that obesity, irrespective of its definition (i.e. BMI or WC), may modulate the clinical expression of HCM.

Intriguingly, the effect of obesity on the incidence of clinical HCM was relatively greater in younger individuals. This result might be indirect evidence that obesity is not a mere consequence of sedentary lifestyle due to the concerns of physical activity-related risks in patients already diagnosed with HCM. More importantly, this finding is in line with the previous study reporting that the association between higher BMI and higher LV mass/height was more prominent in adults with early exposure to obesity than those with late exposure.¹⁶ It can be speculated, based on their observation, that early exposure to obesity in childhood or young adulthood may interact with genetic predisposition, and thus can further augment the risk of developing clinical expression of LVH. Considering the substantial and increasing prevalence of obesity among children, adolescents, and young adults, and its adverse effects on the CV system,^17,18 the prevention or reversal of obesity is likely to have a therapeutic implication in participants with a high-risk of developing clinical HCM, i.e. carriers of HCM-causative mutations. This is especially true, given that there is no proven preventive strategy for minimizing or delaying the clinical expression of HCM and thus, a ‘watchful waiting strategy’ is the only available option for these at-risk individuals. Further studies are mandatory to prove that this therapeutic approach may prevent the clinical expression of HCM and ultimately reduce adverse clinical outcomes.

Another novel and important finding of our study is that obesity and metabolic health status were independently, as well as synergistically, associated with the incidence of clinical HCM. These results suggest that, in the long term, every obese individual should be encouraged to maintain a normal weight to gain the greatest CV health benefits. However, it should be also emphasized that moderate weight loss leading to a transition from metabolically unhealthy to metabolically healthy obesity is a more practical goal leading to reducing the risk of CVD.¹⁹ Our study showed that the incidence of clinical HCM was lower in metabolically healthy overweight and obese participants than their metabolically unhealthy counterparts. Given that a large amount of weight loss to reach a normal weight is not an easily achievable goal, the efforts to improve metabolic health status can be a more practical and effective strategy.

One potential mechanism explaining the association of obesity and metabolic health status with the incidence of clinical HCM is insulin resistance, which has been suggested to be associated with the clinical expression of HCM. Specifically, several studies reported that overexpression of insulin-like growth factor-1 (IGF-1) and transforming growth factor-β (TGF-β), well-known trophic and mitotic factors, may be major contributors to the pathological manifestations of HCM.²⁰ Given that high levels of IGF-1 and TGF-β are related to insulin resistance,^21,22 it can be postulated that clinical expression of HCM can be promoted by a variety of conditions associated with insulin resistance, such as obesity or metabolic syndrome. Indeed, one study demonstrated that HCM patients had a higher homeostasis model assessment of insulin resistance than patients with essential hypertension and normotensive control participants, implying a possible link between insulin resistance and clinical expression of HCM.²³ However, our findings can be also explained by the fact that HCM patients with a high BMI are more likely to be symptomatic and seek medical attention leading to the diagnosis. Further studies are warranted to confirm the postulated association and the therapeutic potential of targeting insulin resistance to prevent or modulate the risk of clinical expression of HCM (Supplemental Figure 6).

Several limitations should be acknowledged. First, we used BMI as a major parameter of obesity, which cannot reliably differentiate body fat from muscle mass. However, BMI was shown to be highly correlated with other parameters of fat mass.¹ Furthermore, our results remained similar when WC, a surrogate measure of abdominal fat,²⁴ was used instead of BMI. Second, since echocardiographic data is not available in the Korean NHIS database, we could not assess the effect of obesity and/or metabolic health status on the phenotypic variability in HCM, which might diminish the impact and novelty of the study findings. Third, information on genetic testing results could have been helpful for further verification of the diagnosis of HCM. In this respect, several patients might have acquired LVH from well-known etiologies rather than the genetic form of HCM. However, a strict review by medical experts and insurance claims professionals was performed, and most other etiologies such as amyloidosis and Fabry disease are routinely excluded in the diagnostic process. In addition, lack of genetic information did not allow us to examine the impact of sarcomere mutations on the association between obesity and the incidence of clinical HCM. Fourth, data on adverse cardiac events after HCM diagnosis were not included in this study since we focused on the incident cases of clinical HCM, as opposed to prevalent cases, resulting in short follow-up duration after HCM diagnosis and thus difficulty in obtaining a sufficient number of events to be statistically significant. Fifth, since the study population was derived from a single country, the findings may not be generalizable to other ethnicities. Finally, causality cannot be inferred from our findings because of the observational study design.

Conclusions

This study suggests that obesity may be associated with an increased incidence of clinical HCM, with a relatively stronger association in men and younger individuals. Metabolic abnormalities can synergistically and independently increase the risk of clinical expression of HCM. These findings provide novel insights into the role of obesity and metabolic abnormalities as arbitrators modifying clinical expression of HCM.

Author contribution

J-BP, SRO, and H-KK contributed to the conception or design of the work. DHK and KH contributed to the acquisition and analysis of data for the work. J-BP, HL, I-CH, YEY, HEP, S-YC, Y-JK, G-YC, SRO, and H-KK contributed to the interpretation of data for the work. J-BP, SRO, and H-KK drafted the manuscript. DHK, HL, I-CH, YEY, HEP, S-YC, Y-JK, G-YC, and KH critically revised the manuscript. All gave final approval and agree to be accountable for all aspects of work ensuring integrity and accuracy. SRO and H-KK contributed equally.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was partly supported by Yuhan 2018–2019 research fund.

Author notes

*

These authors should be considered joint corresponding authors