Frequency and Clinical Associating Factors of Valvular Heart Disease in Asymptomatic Korean Adults

Limited information is available on the prevalence and clinical determinants of valvular heart disease (VHD) in apparently healthy people. This study sought to assess the frequency and clinical associating factors of aortic stenosis (AS), aortic regurgitation (AR), mitral stenosis (MS), mitral regurgitation (MR), tricuspid regurgitation (TR) in asymptomatic individuals with health check-up examination. We included 23,254 subjects ≥50 years of age who underwent a health check-up examination with transthoracic echocardiography (TTE) between 2012 and 2016 in a single tertiary-care hospital in Korea. Among a total of 23,254 subjects, 15,358 men (66.0%) and 7,896 women (34.0%) underwent TTE. Newly identified (predominantly mild) VHD was detected in 9.4% of subjects. The most common VHD were TR (4.6%), AR (3.0%) and MR (2.4%). Clinically significant (more than moderate) VHD was identified in 176 subjects (0.8%). Age ≥75 years was associated with all clinically significant VHD, and female gender was associated with AR, MS and TR. Korea has been very active in the health check-up examination including echocardiography. We find that VHD in apparently healthy people is not uncommon than believed; all VHD except MS were more frequent in elderly over 75 years of age in a large population-based study.

transthoracic echocardiography (tte). Standard two-dimensional TTE and Doppler echocardiography using multiple windows were conducted as part of a health promotion program using commercially available equipment (Vivid7; GE Medical Systems, Horten, Norway and SC2000; Siemens Medical Solution, Mountain view, USA). Image acquisition and measurements were performed according to the 2003 American Society of Echocardiography (ASE) guidelines and 2009 European Association of Echocardiography/ASE guidelines 6,7 . The severity of VHD was determined using an integrated approach and defined as none, mild, mild to moderate, moderate, moderate to severe, and severe (Supplementary Table S1). Clinically significant VHD was defined as moderate or more severe VHD. The primary outcome was a presence of VHD, indicated by mild or significant AS, AR, MS, MR, or TR.
The mean transaortic pressure gradient and peak transaortic velocity were measured in all views possible (that is, apical five-or three chamber and right parasternal views) and the highest values were used for analysis. The time-velocity integral at the aortic valve and the left ventricular outflow tract level were obtained by using continuous wave and pulse wave Doppler echocardiography, respectively, and the aortic valve area was calculated using the continuity equation with the parameters mentioned previously. The mean of three consecutive Doppler signals was used.
Severity of AR was determined using an integrated approach that included the size of the regurgitant jet in the left ventricular cavity, size of the vena contracta, jet deceleration rate, magnitude of the diastolic flow reversal in descending aorta, and when available, regurgitant volume (RV) and effective regurgitant orifice area (ERO).
We assessed mitral valve area by direct planimetry or, when planimetry was not feasible, the Doppler pressure half-time method. The mean diastolic transmitral pressure gradient was calculated via continuous wave Doppler echocardiography.
Severity of MR was rated based on the presence of systolic flow reversal in the pulmonary vein, size of the vena contracta, RV and ERO. One or more available measures were used for MR grading.
We conducted TR quantification by the proximal flow convergence (PISA) method. Color Doppler images of TR PISA were obtained from apical or para-apical views, zooming in on the region of interest. Size of the vena contracta and presence of systolic flow reversal in the hepatic vein were also evaluated. Statistical analysis. Continuous data are reported as the mean and standard deviation, and categorical data as the total sample size (n) and relative (%) frequency. For each type of VHD, Kruskal-Wallis test or Fisher's exact test explored relationships between the 5 severity groups of the disease and quantitative or categorical variables, respectively. Then, we defined a binary level event for the severity by combining moderate to severe conditions as the primary outcome for each disease. As a preliminary investigation to find relevant variables, we performed univariable logistic regression analyses to associate VHD with demographic and clinical variables. Sparse outcomes may limit full multivariable analyses. However, perhaps because we mostly have healthy adult data, most predictors were not significantly related to the event. Finally, we analyzed the dichotomized outcome of diseases in the multivariable logistic regression to adjust for age and sex, along with a few significantly relevant predictors. All analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC) and the R 3.5.1 statistical software (Vienna, Austria). P values < 0.05 were considered statistically significant but a large sample effect was considered in our discussion of the results.
Clinically significant (more than moderate) undiagnosed VHD was identified in 176 subjects (0.8%). There were 25 subjects (0.1%) with clinically significant valvular stenosis, 140 subjects (0.6%) with clinically significant valvular regurgitation and 11 subjects diagnosed with both clinically significant valvular stenosis and regurgitation (Supplementary Table S3). Among cases of clinically significant VHD, TR (n = 71) was the most common followed by AR (n = 56) (Fig. 1). However, among in 18 subjects who diagnosed asymptomatic severe VHD, eight were diagnosed with severe AS, 6 with severe AR, 5 with severe MR, and only one with severe TR. Two subjects had severe AS and severe AR simultaneously. On the other hand, there were no cases of severe MS. The clinical characteristics of subjects in each type of VHD are summarized in Tables 2-4. In Tables 2-4, subjects with multi-valvular heart disease were duplicated, but they were classified according to involved valve to show clinical characteristics of each valve disease. Age was positively related to the severity of VHD (p < 0.001), except in cases of MS.
Univariable logistic analyses were performed for clinically significant VHD (more than moderate) (Supplementary Table S4). Previous prospective studies of subjects aged 65 years and older have demonstrated that the proportion of patients with clinically significant VHD was increased in subjects aged 75 years or older 8 . To explore the effects on aging, we dichotomized the subjects based on their age less than or at least 75 years old. Age ≥75 years was associated with all clinically significant VHD, and female gender was associated with AR, MS and TR. The results were also consistent in the multivariable analysis. Details on the associating factors are provided in Table 5 and Supplementary Table S4. When the frequency of clinically significant VHD was examined in subjects by age groups by 75 years, TR was the most frequent form of VHD in both groups (Fig. 2). There were 1034 subjects who were older than 75 years old. While the frequency of newly diagnosed VHD in all study population was 9.4%, the frequency was higher in those aged ≥75 years (29.3%) than those aged <75 years, which was 8.4%. A total of 34 subjects (3.3%) were diagnosed clinically significant VHD in subjects who were older than 75  www.nature.com/scientificreports www.nature.com/scientificreports/ years old. Among them, two clinically significant VHD were simultaneously present in 5 subjects; AS with AR, AS with TR, AR with MR, AR with TR, and MR with TR. None of the subjects were diagnosed with more than 3 types of clinically significant VHD.

Discussion
In this study, based on a large study population of asymptomatic Korean adults aged >50 years, we found that the frequency of VHD was 9.4%, which was higher than that observed in Western studies. Previous studies in the West have varied widely in their estimates of the prevalence of VHD, as the overall prevalence of VHD in the United States is 2.5% with wide age-related variation 1,9 . On the other hand, some studies detected previously undetected VHD in just over half of 2500 participants aged >65 years 8 . Consistent with these results, we observed a significantly higher rate of VHD in subjects aged ≥75 years. While the frequency of newly identified VHD in subjects aged <75 years was 8.4%, the frequency was higher in those aged ≥75 years (29.3%). The use of a study population over 50 years of age is expected to result in a higher apparent frequency of VHD compared to previous studies.
The frequency and severity of all types of VHD except MS were higher in the elderly. Our findings that age significantly affects the prevalence of VHD are consistent with those of previous studies 8, 10,11 . A previous study based on data derived from the Korean National Health Insurance Service between 2006 and 2011, Jang et al. reported that the overall age-standardized cumulative prevalence of non-rheumatic valve disease was increased in patients over 65 years of age 12 . Although the present study could not provide information regarding the prevalence of VHD because we obtained health screening data from a single center, our results demonstrate the possibility of identifying asymptomatic VHD and its associating factors in apparently healthy Korean adults, with applications to real world practice.
Regarding the distribution of clinically significant VHD (Fig. 2), the frequency of AS was relatively lower in subjects ≥75 years than in those <75 years. However, based on the distribution of all VHD, AS and AR were more frequent in subjects ≥75 years (Supplementary Fig. S2). There are several possible explanations for this  Table 4. Clinical characteristics and severity of tricuspid regurgitation. Numbers are presented as number (%) or mean ± standard deviation. BMI -body mass index, HTN -hypertension, DM -diabetes mellitus, HbA1c -hemoglobin A1c, TG -triglyceride, HDL-C -high density lipoprotein-cholesterol, LDL-C -low density lipoprotein-cholesterol. Please see the legend for Table 2.  www.nature.com/scientificreports www.nature.com/scientificreports/ Although the association between hypertension and AR was significant (OR 1.78(1.04-3.06)), there was a strong association between AR and age (OR 4.71(2.37-9.36)) (Supplementary Table S4). Earlier reports suggested that hypertension predisposes subjects to aortic root enlargement and AR. However, other previous studies demonstrated a strong association between age and AR, which may explain the lack of association between AR and hypertension after adjusting for age in the multivariable model 13,14 . In addition, obesity was associated with a lower frequency of TR. Lower BMI has been associated with valvular regurgitation, perhaps due the increased ease of echocardiographic imaging in thin individuals 9,15 .
The frequency of clinically significant (more than moderate) VHD was also considerably higher than expected in the present study. Severe MS was not found in anyone, but a total of 18 subjects were diagnosed with asymptomatic severe VHD. Although further explorations of the contemporary prevalence and natural course of VHD is required, VHD often has a long latent period and the optimum timing of intervention requires initial identification by screening programs.
This study has several limitations. First, it was a retrospective study performed in a sample of subjects who voluntarily visited our center for health examinations and our study might reflect some selection bias, because individuals in the study population were self-referred for health examinations. It is not yet known what the difference may exist among healthy subjects who did not bother with health examinations, as the national program is not mandatory. Although we do not intend to generalize to general populations, we believe that any differences will be small and our data and results are uniquely based on health promotion initiative, including echocardiography.  Table 5. Clinical associating factors by types of more than moderate valvular heart diseases based on multivariable logistic regressions. Age and sex were retained in all models, and other covariates considered before eliminating for statistical insignificance include: hypertension, diabetes mellitus, BMI, glucose, cholesterol, triglyceride, high density lipoprotein-cholesterol, low density lipoprotein-cholesterol, and hemoglobin A1c (See Supplementary Table S4). CI -confidential interval, BMI -body mass index. www.nature.com/scientificreports www.nature.com/scientificreports/ Second, our study subjects were all ethnically Korean, and our findings may not apply to other populations or different ethnicities. Although this study was conducted at a single center among asymptomatic subjects, we believe our large sample accurately represent healthy, asymptomatic, and older Korean adults with regard to the prevalence and severity of VHD. conclusion Korea has actively promoted health check-up examinations, including echocardiography. We found that VHD in apparently healthy people is not less uncommon than hypothesized. These data demonstrate that all forms of VHD except MS were more frequent in elderly subjects over 75 years of age in a large population-based study. Valve diseases thus represent important public-health problems.

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.