대한내과학회지 : 제 75 권제 2 호 2008 고혈압환자의경동맥내중막두께의증가는중막두께의증가에기인한다 건양대학교의과대학심장내과학교실 1, 한국표준과학연구원 2 원희관 1 김원식 2 김기영 1 현대우 1 권택근 1 배장호 1 Increased carotid intima-media thickness in hypertensive patients is caused by increased medial thickness Hee-Kwan Won, M.D. 1, Wuon-Shik Kim, Ph.D. 2, Ki-Young Kim, M.D. 1, Dae-Woo Hyun, M.D. 1, Taek-Geun Kwon, M.D. 1 and Jang-Ho Bae, M.D. 1 Department of Cardiology, College of Medicine, Konyang University, Daejeon, Korea 1 ; Korean Research Institute of Standards and Science, Daejeon, Korea 2 Backgrounds/Aims: It has been suggested that there is a differential response of the vasculature to systemic risk factors for atherosclerosis. We sought to evaluate the impact of hypertension on the carotid arterial wall using new methods that can measure each arterial wall layer. Methods: The study subjects consisted of 163 patients who underwent carotid arterial scanning using high-resolution ultrasound that could measure the left carotid intima-media, intima, and media separately. The individual carotid arterial wall thickness was measured off-line by a new method using the Canny edge-detection algorithm. Results: Hypertensive patients (n=79, mean age 61.8 years) had a higher prevalence of diabetes (31.6% vs 11.9%, p=0.004) and a lower level of HDL-cholesterol than did normotensive patients (41.8±11.0 mg/dl vs 45.7±10.0 mg/dl, p=0.019). Hypertensive patients had higher carotid intima-media thickness (CIMT, 0.81±0.21 mm vs 0.74±0.18 mm, p=0.003) and carotid medial thickness (CMT, 0.46±0.12 mm vs 0.42±0.09 mm, p=0.007) than did normotensive patients, whereas carotid intimal thickness (CIT) was not significantly different (0.34±0.04 mm vs 0.34±0.04 mm, p=0.196). Multivariate analysis revealed that the independent factors of CIMT were CMT (β=0.915, p<0.001), hypertension (β= 0.076, p=0.008), age (β=0.074, p=0.010), and sex (β=-0.079, p=0.005). Pearson correlation coefficient between CIMT and CMT was higher (r=0.932, p<0.001 vs r=0.445, p<0.001) than that between CIMT and CIT. The correlation between CIMT and CMT was higher (r=0.940, p<0.001 vs r=0.910, p<0.001) in hypertensive patients than in normotensive patients, whereas that between CIMT and CIT was lower (r=0.344, p=0.002 vs r=0.583, p<0.001) in hypertensive patients. Conclusions: The increased CIMT is caused by increased CMT in hypertensive patients, and this finding is compatible with the medial hypertrophy seen in hypertension. The carotid medial layer should be the focus of attention in future studies looking at hypertensive patients. (Korean J Med 75:179-185, 2008) Key Words: Hypertension; Tunica media; Carotid arteries Received: 2007. 8. 29 Accepted: 2008. 5. 14 Correspondence to: Jang-Ho Bae, M.D., PhD., Heart Center, Konyang University Hospital, 685 Gasoowon-dong, Seo-gu, Daejeon 302-718, Korea E-mail: janghobae@yahoo.co.kr - 179 -
- The Korean Journal of Medicine: Vol. 75, No. 2, 2008 - 서 론 2. 방법 경동맥내중막두께는관상동맥및전신의다른동맥의죽상경화증과관련이있어동맥경화증의지표로이용될수있으며 1), 고해상도이면성초음파를이용한경동맥내중막두께 (Carotid intima-media thickness, CIMT) 의측정이여러연구에서사용되어지고있다. 이들연구에서경동맥내중막두께가두꺼워질수록관상동맥질환과뇌졸중의위험이증가하며 2), 건강인과비교시관상동맥질환이있는군에서경동맥내중막두께가증가되어있다고보고되었다 3, 4). 경동맥내중막두께는연령이증가할수록두꺼워지고, 특히고혈압환자에서유의하게두꺼워진다고알려져있다 5, 6). 최근의연구에서이면성초음파영상처리에의해경동맥의내막 (carotid intima, CIT), 중막 (carotid media, CMT), 내중막 (CIMT) 의두께를분리측정하여, 죽상경화증의전신적인위험인자들에대한이들의반응이서로다를수있음을시사하였다 7). 그러나고혈압이경동맥내막과중막에미치는영향에대한구체적인생체내연구에대해서는잘알려진바없다. 이에저자들은경동맥의두께를증가시킨다고알려져있는고혈압이경동맥의각층에어떤영향을끼치는지알아보고자본연구를시행하였다. 대상및방법 1. 대상환자저자들은본원에서 437명 ( 평균연령 59세, 남자 243명 ) 의환자들을대상으로고해상도초음파로경동맥초음파를시행하였으며, 그중좌측경동막내막과중막의두께를분리하여측정할수있었던 163명 ( 평균연령 60.7세, 남자 81명 ) 의환자를대상으로본연구를시행하였다. 고혈압환자군은과거에고혈압을진단받고현재본원심장내과에서항고혈압제를복용중인환자혹은 JNC-VII 의기준에의해내원당일과다음방문또는입원다음날 10분이상안정후최소한 5분간격으로 2회이상측정하여수축기혈압이 140 mmhg 이상이거나확장기혈압이 90 mmhg 이상인경우로정의하였다. 전체대상환자군을위의정의에따라고혈압환자군 ( 평균연령 61.8세, 남자 37명 ) 과정상혈압군 ( 평균연령 59.6세, 남자 44명 ) 으로나누어비교분석하였다. 총경동맥은광폭 (1~3L) 선상배열변환기를가진고해상도초음파 (HP Sonos-5500, Phillips, USA) 를통해동일한검사자에의해측정되었다. 환자의두부가신전된상태로누워있는상태에서좌우측각각에서총경동맥에서경동맥팽대부로이행하는경계부로부터근위부 1 cm 구간에서원위벽 (far wall) 에서측정하였다. 현재경동맥내중막측정에있어좌측과우측어느곳이적절하다고밝혀진바는없다. 따라서본연구에서는좌우측모두에서측정을하여비교분석하였다. 영상의질을최상화하기위해깊이조정은 4 cm에고정하였으며, 전체분석이진행되는동안변환기주파수는 0.2 mm의축방향해상도를가지고 11 MHz 에고정하였다. 깨끗한경동맥영상이얻어진뒤에는오프라인분석을위하여디지털방식으로영상을저장하였다. 경동맥내중막두께는고해상도초음파상에서혈관내강과혈관내막의경계부위로부터혈관중막과혈관외막의경계부까지의거리로정의하였다. 경동맥원위부영상은흔히저음영지역에의해분리되는 2개의평행한음영선을보이게된다. 내막은첫번째, 두번째음영선의서로먼쪽끝부분사이의거리를측정함으로써얻어지며중막은가장밝은 2개의음영사이의거리이다 8, 9). 경동맥내중막의두께는수동측정보다컴퓨터를이용한반자동화된방법이더정확하다고알려져있다 10). 이연구에서는내막, 중막, 내중막의두께를분리측정하기위하여이미개발된소프트웨어를이용하여측정하였다 7). 먼저, 측정된경동맥영상을이용하여전자식눈금측정기에의해화소당크기가결정이되고, 그것은 multipurpose phantom (Model 539, ATS laboratory) 의축방향으로 4 cm 깊이에서미리조정된다. 다음으로측정하고자하는부위 (region of interest, ROI) 가경동맥팽대부근위부 1 cm 구간에서최소 1 cm 크기로취해진다. 측정부위 (ROI) 는이부위에동맥경화반 (plaque) 이포함되지않게하기위해수동으로이동될수있다. 측정부위의영상의질평가와잡음제거를위해 filtering algorithm이사용되는데 11), 내막, 중막, 내중막은 Canny edge algorithm을통해경계부영상을얻은후구분된다 12). 각층의경계가 Canny algorithm으로구분되기어려울정도로분명치않은경우는통계적신호처리에기초한 autocorrelation에의해각층을구별한다 13). 각층의두께를측정하기위해각층의두께에해당하는화소수를계산하고, 결국각층의화소수에교정인자 (millimeters per - 180 -
- Hee-Kwan Won, et al: Hypertension and carotid media thickness - Table 1. Clinical characteristics of the study subjects Hypertension (+) Hypertension (-) p value Number 79 84 Age, yrs 61.8±9.9 59.6±10.7 0.165 Male, n (%) 37 (46.8%) 44 (52.4%) 0.582 BMI, kg/ m2 25.3±3.2 24.9±3.1 0.523 SBP, mmhg 123.1±13.9 119.9±13.8 0.145 DBP, mmhg 77.7±9.9 75.2±9.9 0.117 Other risk factors DM, n (%) 25 (31.6%) 10 (11.9%) 0.004 Smoking, n (%) 22 (27.8%) 34 (40.5%) 0.112 Dyslipidemia, n (%) 27 (34.2%) 28 (33.3%) 1.000 Ejection fraction, % 65.6±8.4 65.7±9.4 0.942 Previous MI, n (%) 5 (6.3%) 3 (3.6%) 0.486 CAD, n (%) 59 (74.7%) 57 (67.9%) 0.430 FBS, mg/dl 112.5±33.1 111.5±31.3 0.851 Total cholesterol, mg/dl 176.2±35.0 184.1±44.9 0.222 Triglyceride, mg/dl 168±86.0 156.1±87.8 0.383 HDL-cholesterol, mg/dl 41.8±11.0 45.7±10.0 0.019 LDL-cholesterol, mg/dl 107.4±26.9 109.6±34.0 0.652 Hs-CRP, mg/dl 0.2±0.3 0.4±1.0 0.182 Homocysteine, μmol/l 11.1±3.8 10.4±3.4 0.226 Fibrinogen, g/l 3.3±0.8 3.4±0.8 0.618 yrs, years; BMI, body mass index; CAD, coronary artery disease; DBP, diastolic blood pressure; DM, diabetes mellitus; FBS, fasting blood sugar; MI, myocardial infarction; H(L)DL, high (low)-density lipoprotein; Hs-CRP, high sensitivity c-reactive protein; SBP, systolic blood pressure pixel) 를곱하여내막, 중막, 내중막의두께를구할수있다. 이연구에서경동맥내막, 중막, 내중막두께의평균값은 quality index ( 최소 1 cm 이상의측정부위에서분석이가능한분절의비율 ) 가 0.6 이상인영상에서만구하였다. 이방식에의한경동맥내막, 중막, 내중막측정의타당성검증은이미과거연구에서입증한바있다 7). 이와같은방법으로 163명의환자의내중막, 내막, 중막의두께를측정하였고, 동맥경화반의유무를확인하고이를분석하였다. 3. 통계분석자료의통계분석은 SPSS (Version 13.0, SPSS Inc., Illinois, USA) 을이용하였고, 통계수치는평균 ± 표준편차로표시하였으며, p 값이 0.05 미만일때유의한것으로간주하였다. 고혈압환자군과정상혈압군의변수들의평균치비교는 t-test 를이용하여시행하였다. 경동맥내중막두께와경동맥내 막두께, 경동맥중막두께의상관관계는 Pearson 상관계수를이용하여분석하였고, 심혈관계위험인자와경동맥내중막두께, 중막두께와의독립적인기여정도를파악하기위하여다중회귀분석 (multiple regression analysis) 을시행하였다. 결과 1. 대상환자들의임상적특성전체 163명의대상자중고혈압환자군 (79명) 의연령 (61.8±9.9세 vs 59.6±10.7세, p=0.165), 수축기혈압 (123.1±13.9 mmhg vs 119.9±13.8 mmhg, p=0.145) 과확장기혈압 (77.7± 9.9 mmhg vs 75.2±9.9 mmhg, p=0.117) 은정상혈압군과비교하여유의한차이는없었다. 고혈압환자군에서당뇨병이정상혈압군보다많았으며 (31.6% vs 11.9%, p=0.004), 고밀도지단백콜레스테롤이정상혈압군보다낮았다 (41.8±11.0 vs 45.7±10.0 mg/dl, p=0.019)( 표 1). - 181 -
- 대한내과학회지 : 제 75 권제 2 호통권제 576 호 2008 - Table 2. Ultrasonographic parameters Hypertension (+) Hypertension (-) p value Left CCA, mm IMT 0.81±0.21 0.74±0.18 0.003 Intima thickness 0.34±0.04 0.34±0.04 0.196 Media thickness 0.46±0.12 0.42±0.09 0.007 Plaque (+) n, (%) 16 (20.3%) 21 (25.0%) 0.663 CCA, common carotid artery; IMT, intima media thickness Table 3. Factors affecting left common carotid artery intima-media thickness Beta (β) * 95% confidence interval p value Media thickness 0.915 1.039~1.175 <0.001 Hypertension -0.076-0.034~-0.006 0.008 Diabetes Mellitus 0.051-0.002~0.034 0.079 Age 0.074 0.000~0.001 0.010 Sex -0.079-0.034~-0.006 0.005 All models included age, sex, hypertension, diabetes mellitus, smoking, systolic (diastolic) blood pressure, and left common carotid media thickness as independent variables. Only those variables that remained significant or borderline significant after backward stepwise elimination are shown in this table. *Beta (β), regression coefficient Thickness (mm 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 HTN(+) n=79 HTN(-) n=84 p =0.196 p =0.007 p =0.003 Left CCA IT Left CCA MT Left CCA IMT Figure 1. Ultrasonographic parameters (IMT, IT, and MT of left CCA). Left CCA IMT (0.81±0.21 mm vs 0.74±0.18 mm, p=0.003) and CCA MT (0.46±0.12 mm vs 0.42±0.09 mm, p=0.007) in hypertensive patients were higher than in normotensive patients, but left CCA IT (0.34±0.04 mm vs 0.34±0.04 mm, p=0.196) in hypertensive patients was not significantly different. HTN, hypertension; IMT, intima-media thickness; IT, intimal thickness; MT, medial thickness; CCA, common carotid artery 2. 경동맥내중막, 내막, 중막의두께 163명 ( 고혈압군 79명, 정상혈압군 84명 ) 을대상으로분리측정치의분석을시행하였다. 고혈압환자군의좌측총 경동맥내중막두께 (0.81±0.21 mm vs 0.74±0.18 mm, p= 0.003) 와중막두께 (0.46±0.12 mm vs 0.42±0.09 mm, p=0.007) 는정상혈압군보다유의하게높았으나, 내막두께 (0.34± 0.04 mm vs 0.34±0.04 mm, p=0.196) 는정상혈압군과유의한차이가없었다. 동맥경화반의유무와관련해서는고혈압환자군 79명중 16명 (20.3%) 에서동맥경화반이관찰되었으며, 정상혈압군 84명중 21명 (25.0%) 에서동맥경화반이관찰되었다 (p=0.663) ( 표 2)( 그림 1). 3. 경동맥내중막두께와관련된독립인자경동맥내중막두께와관련된독립인자를알아보기위하여고혈압의유무에따라차이를보인좌측경동맥벽의두께를다변수분석을시행하였다. 다변수분석을통해나타난경동맥내중막두께의독립인자들로는경동맥중막두께 (β=0.915, 95% 신뢰구간, 1.039~1.175, p<0.001), 고혈압 (β=-0.076, 95% 신뢰구간, -0.034~-0.006, p=0.008), 나이 (=0.074, 95% 신뢰구간, 0.000~0.001, p=0.010), 그리고성별 (=-0.079, 95% 신뢰구간, -0.034~-0.006, p=0.005) 이있었다. 당뇨 (=0.051, 95% 신뢰구간, -0.002~0.034, p=0.079) 의경우 - 182 -
- 원희관외 5 인 : 고혈압과경동맥중막두께측정 - 유의한경향을보였다 ( 표 3). 4. 경동맥내중막두께와내막, 중막두께와의상관관계 Figure 2. Correlation between IT and MT of left CCA. As a whole, IT and MT were correlated (n=163, r=0.252, p<0.001). In normotensive patients, this kind of correlation existed (n=84, r=0.476, p<0.001), but not in hypertensive patients (n=79, r=0.073, p=0.521). A dotted line indicates hypertensive patients, and a solid line indicates normotensive patients. HTN, hypertension; IT, intimal thickness; MT, medial thickness; CCA, common carotid artery 각각의경동맥벽두께사이의상관관계분석에도좌측경동맥벽두께측정치를이용하였다. 전체대상자에서보면경동맥내막의두께와중막의두께사이의상관관계는유의한값을보이고 (163명, r=0.252, p<0.001), 정상혈압군에서도유의한상관관계를보이지만 (84명, r=0.476, p<0.001), 고혈압환자군에서는상관관계가없었다 (79명, r=0.073, p=0.521)( 그림 2). 전체대상자에서경동맥내중막두께는경동맥중막두께와경동맥내막두께모두와상관관계를보였으나경동맥중막두께와더좋은상관관계를보였다 (r=0.932, p<0.001 vs r=0.445, p<0.001). 또한고혈압환자군은정상혈압군보다경동맥내중막두께와중막두께의상관관계는높았으나 (r=0.940, p<0.001 vs r=0.910, p<0.001), 경동맥내중막두께와내막두께의상관관계는정상혈압군보다낮았다 (r=0.344, p=0.002 vs r=0.583, p<0.001)( 그림 3). 고찰경동맥은관상동맥및전신의다른동맥의죽상경화증과 Figure 3. Correlation between IMT and IT or MT in the left CCA. As a whole, the Pearson correlation coefficient between IMT and MT was higher (r=0.932, p<0.001 vs r=0.445, p<0.001) than that between IMT and IT. The correlation coefficient between IMT and MT was higher (r=0.940, p<0.001 vs r=0.910, p<0.001) in hypertensive patients than in normotensive patients, whereas that between IMT and IT was lower (r=0.344, p=0.002 vs r=0.583, p<0.001) in hypertensive patients. A dotted line indicates hypertensive patients, and a solid line indicates normotensive patients. HTN, hypertension; IMT, intima-media thickness; IT, intimal thickness; MT, medial thickness; CCA, common carotid artery - 183 -
- The Korean Journal of Medicine: Vol. 75, No. 2, 2008 - 관련이있으며 1) 고해상도이면성초음파를이용하여경동맥내중막두께와순환기질환위험인자들의연관성에관한연구가많이이루어졌다. 고해상도초음파를이용한경동맥내중막두께측정은죽상경화증의조기진단에유용한검사법으로 14), 조기죽상경화증의변화와진행을평가하는데좋은표지자이며 15), 초음파를이용한방법은상대적으로덜침습적이고, 안전하며, 간단하고, 정확성과재현성이뛰어난검사법으로알려져있다 16). 기존의초음파를이용한경동맥내중막두께에대한연구에서는중막의비후와내막의두꺼움을감별하기가어려웠으나최근의연구에서초음파를통해얻은영상을새로운방법의영상처리에의해내막과중막, 내중막의두께를따로측정하였으며, 순환기질환의전신적인위험인자에대한이들각층의반응이다름이보고되었다 7). 본연구에서처음 437명에서고해상도초음파를시행하였으나, 최종좌측경동맥내중막, 내막, 중막의분리측정이가능한 163명만을연구의대상으로하였다. Korean IMT study 연구모임의다기관역학연구 17) 결과에서는우측경동맥의임상적의의가더높을것으로보고를하였으나, 본연구에서는좌측에서만유의한결과를보여이와상반된결과를보이고있다. 이에대해서는아직까지통일된결과를보이고있지않아추후더많은연구가필요하리라생각한다. 또한처음본연구의대상환자였던 437명의환자들중본연구에서사용한측정방법으로좌측경동맥의내막의두께와중막의두께를분리하여측정할수있었던환자의수는총 163명 (37.3%) 으로, 향후경동맥의두께측정방법에있어많은개선이필요할것으로생각한다. 본연구에서경동맥내막과중막사이의관계는전체대상자 (r=0.252, p<0.001) 와정상혈압군 (r=0.476, p<0.001) 에서는내막과중막의두께사이에상관관계가있었으나고혈압환자군에서는이런연관성이없어진다 (r=0.073, p=0.521). 이것으로볼때고혈압은내막과중막에균등한영향을미치지는않는것으로생각된다. 내막과내중막의관계를보면전체대상자에서는내막과내중막사이에상관관계 (r=0.445, p<0.001) 가있으나, 고혈압이있으면 (r=0.344 p=0.002) 고혈압이없을때보다 (r=0.583, p<0.001) 상관관계가떨어짐을보여준다. 반면에중막과내중막은고혈압의유무에상관없이좋은상관관계를보여주고 (r=0.932, p<0.001) 특히고혈압이있을때더좋은상관관계를보여 (r=0.940, p<0.001 vs r=0.910, p<0.001) 고혈압은내막보다는중막에더많은영향을미침을간접적으로알수있다. 병리조직을이용한타 연구에서고혈압환자의동맥벽두께의증가는주로평활근세포의비후및비대와교원질 (collagen) 과탄력소 (elastin) 의밀도증가에기인한중막두께의증가와연관이있다고보고한바있으며 18, 19), 이는본연구의결과와도일치한다. 결론적으로고혈압환자에있어경동맥내중막두께의증가는경동맥중막의증가에의한것으로볼수있으며, 이에따라향후고혈압에대한연구를시행함에있어서는경동맥내중막두께뿐만아니라경동맥중막층에대해서도연구의초점이맞춰져야할것이다. 요약목적 : 죽상경화증의전신적인위험인자에대한혈관의서로구분되는다양한반응이제시되어왔다. 저자들은각각의동맥벽층을측정할수있는새로운방법을사용하여경동맥벽에미치는고혈압의영향을측정하고자하였다. 방법 : 이연구는고해상도의초음파로경동맥초음파를시행하여좌측경동맥내중막, 내막, 중막을분리하여측정할수있었던 163명의환자들을대상으로하였다. 각각의경동맥벽의두께는 Canny edge-detection algorithm을사용한새로운방법으로측정되었다. 결과 : 고혈압환자군 (79명, 평균연령 61.8세 ) 은정상혈압군 (84명, 평균연령 59.6세 ) 에비해더높은당뇨병의유병률을보였으며 (31.6% vs 11.9%, p=0.004), 정상혈압군에비해고밀도지단백콜레스테롤이더낮았다 (41.8±11.0 mg/dl vs 45.7±10.0 mg/dl, p=0.019). 두군에서경동맥내막의두께는유의한차이를보이지않았으나 (0.34±0.04 mm vs 0.34±0.04 mm, p=0.196), 고혈압환자군은정상혈압군에비해경동맥내중막두께가높은것으로나타났으며 (0.81±0.21 mm vs 0.74±0.18 mm, p=0.003), 경동맥중막두께에있어서도더높은값을나타내었다 (0.46±0.12 mm vs 0.42±0.09 mm, p=0.007). 다변수적분석을통해나타난경동맥내중막두께의독립적인자로는경동맥중막두께 (β=0.915, p<0.001), 고혈압 (β=-0.076, p=0.008), 나이 (β=0.074, p=0.010), 그리고성별 (β=-0.079, p=0.005) 가있었으며, 경동맥내중막두께와경동맥중막두께사이의 Pearson 상관관계는경동맥내중막두께와경동맥내막두께사이의것보다더높게나타났다 (r=0.932, p<0.001 vs r=0.445, p<0.001). 또한고혈압환자에서경동맥내중막두께와경동맥내막두께사이의상관관계가낮았던것에비해 (r=0.344, p=0.002 vs r=0.583, p<0.001) 고혈압환자에서경동맥내중막두께와경동맥중막두께사이의상관관계는더높은것으로나타났다 (r=0.940, - 184 -
- Hee-Kwan Won, et al: Hypertension and carotid media thickness - p<0.001 vs r=0.910, p<0.001). 결론 : 고혈압환자에있어서경동맥내중막두께의증가는경동맥중막두께의증가에의해서이며, 이결과는고혈압환자에있어서중막층의비후와일치한다. 향후고혈압환자와관련된연구에있어경동맥중막층에대한연구에더초점이맞춰져야할것이다. 중심단어 : 고혈압 ; 중막 ; 경동맥 REFERENCES 1) Salonen JT, Salonen R. Ultrasound B-mode imaging in observational studies of atherosclerotic progression. Circulation 87(3 Suppl):II56-II65, 1993 2) O'Leary DH, Polak JF, Kronmal RA, Manolio TA, Burke GL, Wolfson SK Jr. Carotid artery intima-media thickness as a risk factor for myocardial infarction and stroke in older adult. N Engl J Med 340:14-22, 1999 3) Terry JG, Tang R, Esplend MA, Davis DH, Vieira JL, Mercuri MF, Crouse JR 3rd. Carotid arterial structure in patients with documented coronary artery disease and disease-free control subjects. Circulation 107:1146-1151, 2003 4) Kim JH, Youn HJ, Hong EJ, Park CS, Lee JM, Lim SH, Oh YS, Chung WS, Seung KB, Kim JH, Choi KB, Hong SJ. Clinical significance of B-mode ultrasound of common carotid artery for prediction of severity of coronary artery disease: important parameters on hand measurement. Korean Circ J 35:467-473, 2005 5) Polak JF, O'Leary DH, Kronmal RA, Wolfson SK, Bond MG, Tracy RP, Gardin JM, Kittner SJ, Price TR, Savage PJ. Sonographic evaluation of carotid artery atherosclerosis in the elderly: relationship of disease severity to stroke and transient ischemic attack. Radiology 188:363-370, 1993 6) Kawamoto R, Abe M. Risk factors related to the wall thickness of the common carotid artery in elderly patients. Nippon Ronen Igakkai Zasshi 33:835-839, 1996 7) Bae JH, Kim WS, Rihal CS, Lerman A. Individual measurement and significance of carotid intima, media, and intima media thickness by B-mode ultrasonographic image processing. Arterioscler Thromb Vasc Biol 26:2380-2385, 2006 8) Pignoli P, Tremoli E, Poli A, Oreste P, Paoletti R. Intimal plus medial thickness of the arterial wall: a direct measurement with ultrasound imaging. Circulation 74:1399 1406, 1986 9) Rodriguez-Macias KA, Naessen T, Bergqvist D. Validation of in vivo noninvasive high-frequency ultrasonography of the arterial wall layers. Ultrasound Med Biol 27:751 756, 2001 10) Baldassarre D, Tremoli E, Amato M, Veglia F, Bondioli A, Sirtori CR. Reproducibility validation study comparing analog and digital imaging technologies for the measurement of intima-media thickness. Stroke 31:1104 1110, 2000 11) Wang Z, Bovik AC. A universal image quality index. IEEE Signal Process Lett 9:81-84, 2002 12) Canny J. A computation approach to edge detection. IEEE Trans Pattern Anal Mach Intell 8:679 698, 1986 13) Kay SM. Fundamentals of statistical signal processing: estimation Theory. p. 197-267, New Jersey, Prentice Hall International Inc, 1993 14) Salonen R, Salonen JT. Progression of carotid atherosclerosis and its determinants: a population-based ultrasonography study. Atherosclerosis 81:33-40, 1990 15) Lone E. Carotid artery intima-media thickness: a new noninvasive gold standard for assessing the anatomic extent of atherosclerosis and cardiovascular risk. Clin Invest Med 22:158-160, 1999 16) Aminbakhsh A, Mancini GB. Carotid intima-media thickness measurements: what defines an abnormality?: a sytemic review. Clin Invest Med 22:149-157, 1999 17) Bae JH, Seung KB, Jung HO, Kim KY, Yoo KD, Kim CM, Cho SW, Cho SK, Kim YK, Rhee MY, Cho MC, Kim KS, Jin SW, Lee JM, Kim KS, Hyun DW, Cho YK, Seong IW, Jeong JO, Park SC, Jeong JY, Woo JT, Koh GP, Lim SW. Analysis of Korean carotid intima-media thickness in Korean healthy subjects and patients with risk factors: Korea multi-center epidemiological study. Korean Circ J 35:513-524, 2005 18) Michel JB, Tedgui A, Salzmann JL, Belmain J, Levy B. Function and structure of the vascular wall in experimental arterial hypertension: effect of treatment. Nephrologie 10:55-57, 1989 19) Chobanian AV, Prescott MF, Haudenschild CC. Recent advances in molecular pathology: the effects of hypertension on the arterial wall. Exp Mol Pathol 41:153-169, 1984-185 -