This study is an initial report designed to predict the 10-year probability of stroke in Korean hypertensive patients visiting tertiary hospitals using the Framingham risk score.
We found the average 10-year probability of stroke in hypertensive patients was 26.27% (26.9% women, 25.5% men), which was approximately 2.4 times higher than values found in a previous study of the KCPS cohort . This result may be explained by the fact that 34% of the hypertensive patients did not take antihypertensives (Table 1) and therefore their blood pressure was not well controlled. Another possible explanation is that even among hypertensive patients complying with treatment, 37.2% suffered from uncontrolled HTN (Figure 1-A). The 10-year probability of stroke increased in proportion to blood pressure even among patients being treated for hypertension (Table 2). These findings demonstrate that stricter control of HTN may be an important intervention for prevention of stroke.
In HTN stage 2, the 10-year probability of stroke was higher in women compared to that of men, but in other HTN stages, there was no difference between men and women (Figure 2-B). Female patients in HTN stage 2, were older than their male counterparts, but this difference was not observed in other HTN stages. In addition, the distributions of other risk factors did not segregate by gender at any other stage of HTN (data not shown). Therefore, we believe that the difference in the 10-year probability of stroke between men and women for HTN stage 2 is related to age.
The 10-year probability of stroke in HTN patients increased in proportion to age (Table 2) but the distribution of 10-year probability of stroke between men and women varied according to age. In the 60–69 year age group, the 10-year probability of stroke was higher for men than for women, but in other age groups there was no difference by gender (Figure 2-A). In the 60–69 year age group, frequencies of history of CVD and LVH were higher for men than for women, but not in other age groups. The distribution of other risk factors between men and women did not differ according to age (data not shown). Therefore, we believe that differences in the 10-year probability of stroke between men and women in aged 60–69 are driven by the distribution of CVD and LVH.
The prevalence of diabetes mellitus in hypertensive patients was 16% (17.2% women, 14.7% men; Table 1), which was higher than that seen in the Korean National Health and Nutrition Examination Survey . This study showed that, in the Korean population, the prevalence of diabetes mellitus was 9.1% (7.9% women, 10.2% men). Our enrolled subjects were older and considered high risk patients, which may explain the discrepancy.
The prevalence of dyslipidemia in our sample of hypertensive patients was 12.6% (7.9% women, 16.6% men; Table 1). This prevalence was lower than that observed by the Korean National Health and Nutrition Examination Survey (2005) . That study showed that, in the general Korean population, the prevalence of dyslipidemia was 48.3% (38.2% women, 58.9% men). Our enrolled subjects were older and considered high risk patients, and had been prescribed anti-lipid lowering agents, which may explain the discrepancy.
The prevalence of metabolic syndrome in our sample of hypertensive patients measured according to the ATP-III guideline was 28% (26.1% women, 30.5% men; Figure 1-B). This result was similar to values found by the Korean National Health and Nutrition Examination Survey (2005) . That study showed that, in the general Korean population, the prevalence of metabolic syndrome was 29.5% (26.1% women, 33.1% men). Another report suggests that metabolic syndrome is an important risk factor for ischemic stroke . According to our results, the 10-year risk of stroke is higher for patients suffering from metabolic syndrome (Figure 3).
In Korea, ARB was the most frequently prescribed antihypertensive at 48.6%, with the next being CCB (38.7%). Cardiologists working at general hospitals tend to prefer ARB to other antihypertensives due to higher compliance and fewer side effects., Many clinical papers demonstrating positive effects of ARB for the prevention of stroke and cardiovascular diseases have been published recently . The prescription frequency of ACEI was only 9.9%, probably due to side effects such as cough, and a general preference for ARB. Diuretics were more frequently prescribed for older patients than for younger patients. This may be due to a high prevalence of complications, such as ischemic heart disease and heart failure, in older patients. Among Korean hypertensive patients, 35.9% took more than two antihypertensive drugs (Table 3). This frequency was similar to that in seen in the United States (35.8%, 1999–2002) .
In our study, 76% of hypertensive patients took antihypertensive medication and 63.9% of patients on antihypertensive medication had controlled HTN (Table 1). These results were high compared to those found by the Korean National Health and Nutrition Examination Survey (2005) , but this may be explained by the fact that our study population was derived from a highly selected group of patients who were visiting cardiology clinics for suspected or established cardiac diseases.
This study has some limitations. First, our study population was derived from a highly selected group of patients who were visiting tertiary hospitals for suspected or established cardiac diseases. Although it is appropriate to do a collective study with the cooperation of community based hospitals, public health centers, and general hospitals, our study included only tertiary hospitals. Therefore, cardiovascular risk factors are overrepresented in our sample. Second, this study was a simple cross-sectional study and did not consider information about practical stroke events. Therefore, we could not compare 10-year probability of stroke with real stroke incidence. Third, we did not directly compare the 10-year probability of stroke for our hypertensive sample and the KCPS cohort . Fourth, we assessed smoking status with a single, self-reported questionnaire. Therefore, non-differential misclassification is possible.