Key words: coronary artery disease; intima-media thickness; occult atherosclerosis; peripheral arterial disease; ultrasonic biopsy

Association of Subclinical Wall Changes of Carotid, Femoral, and Popliteal Arteries With Obstructive Coronary Artery Disease in Patients Undergoing Coronary Angiography* Alexandros Kafetzakis, MD; George Kochiadakis, MD; Aggelos Laliotis, MD; Ioannis Peteinarakis, MD; Emmanouel Touloupakis, MD; Nikos Igoumenidis, MD; and Asterios Katsamouris, MD Study objectives: To examine the association of occult atherosclerosis of carotid, femoral, and popliteal arteries with the presence and severity of obstructive coronary artery disease (CAD) in patients without a history or presence of cerebrovascular or peripheral arterial disease using ultrasound examination of peripheral arteries. Patients/methods: One hundred eighty-four such individuals underwent routine coronary angiography. Obstructive CAD was found in 103 cases, which comprised the patient group. The remaining 81 individuals comprised the control group. All were blindly examined by duplex ultrasonography in order to assess occult atherosclerosis, as indicated by the estimation of intima-media thickness of the carotid artery (IMTC), intima-media thickness of the femoral artery (IMTF), intima-media thickness of the popliteal artery (IMTP), and ultrasonic biopsy (UB) of the carotid and femoral arteries. For the individuals with positive coronary angiography findings, the severity of CAD was estimated by the number of the diseased vessels. Results: IMTC, IMTF, IMTP, and UB showed significant correlation with the presence of obstructive CAD, but only IMTC and IMTF were independent predictive factors, with specificity of 74% and 60% and sensitivity of 76% and 70%, respectively. Additionally, our analysis yielded a regression model that, for a given value of IMTC and IMTF, may estimate the probability of CAD: p (CAD) e ( 4.765 3.36 IMTC 1.91 IMTF) /1 e ( 4.765 13.36 IMTC 1.91 IMTF). Patients with one-vessel disease had significantly lower IMTC (p < 0.001) and UB (p 0.011) and lower IMTF (p 0.057) than those with three-vessel disease. Conclusions: The assessment of occult atherosclerosis by duplex ultrasonography in both the carotid and the femoral arteries is significantly associated with the presence and severity of CAD. (CHEST 2005; 128:2538 2543) Key words: coronary artery disease; intima-media thickness; occult atherosclerosis; peripheral arterial disease; ultrasonic biopsy Abbreviations: CAD coronary artery disease; IMT intima-media thickness; IMTC intima-media thickness of carotid artery; IMTF intima-media thickness of femoral artery; IMTP intima-media thickness of popliteal artery; ROC receiver operator characteristic; UB ultrasonic biopsy *From the Vascular Surgery Department (Drs. Kafetzakis, Laliotis, Touloupakis, and Katsamouris), Department of Cardiology (Drs. Kochiadakis and Igoumenidis), and Department of Radiology (Dr. Peteinarakis), University Hospital of Heraklion, University of Crete Medical School, Crete, Greece. Manuscript received November 5, 2004; revision accepted April 5, 2005. A correlation of coronary artery disease (CAD) with atherosclerosis of peripheral arteries and the determination of noninvasive indexes for its existence and extent have been sought by many researchers. Some studies 1 4 report that the intimamedia thickness (IMT) of peripheral arteries obtained by B-mode ultrasound could play this role, even though more reliable indexes exist only for the carotid artery. Other studies 5,6 have claimed that a morphologic classification of carotid and femoral arterial wall changes detected by high-resolution Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Asterios Katsamouris, MD, Professor of Vascular Surgery, University of Crete Medical School, PO Box 1352, Heraklion, Crete, Greece; e-mail: asterios@med.uoc.gr 2538 Clinical Investigations

ultrasound may be an early and accurate indicator of global atherosclerotic disease. However, the majority of the above studies are fragmentary and involve different patient populations, most of them having clinically manifested peripheral arterial disease. Additionally, the precise relationship between the extent of subclinical atherosclerotic disease of peripheral arteries and the existence and severity of CAD has not been well evaluated. For this purpose, we studied the potential role of subclinical atherosclerotic disease, concurrently detected by ultrasound on three different arterial beds, in identifying the presence and severity of obstructive CAD in a single population of patients at risk for ischemic heart disease with totally asymptomatic peripheral arterial disease. Study Group Materials and Methods One hundred eighty-four consecutive patients (mean age, 61.7 8.3 years [ SD]) undergoing coronary angiography at the University Hospital of Heraklion for suspected CAD were included in this study. The inclusion criteria were as follows: absence of history or presence of peripheral arterial disease, as indicated by an ankle brachial pressure index 0.9, carotid artery surgery, or cerebrovascular event and absence of history of previous angiographically documented CAD, congenital heart disease, cardiomyopathy, coronary artery bypass graft, or percutaneous transluminal coronary angioplasty. Written informed consent was obtained from all patients after they were given a detailed description of the procedure. Coronary Angiography Coronary angiography was performed using a standard Judkins technique. Obstructive CAD was defined as the existence of a stenosis 50% of the lumen diameter of at least one major coronary vessel. Patients with such a stenosis comprised the patient group (n 103), and the remaining individuals (with stenosis 50%) were assigned to the control group (n 81). The severity of the disease referred to the number of identified stenosed vessels (lumen diameter 50%) [one-, two- or threevessel disease] and was evaluated by two experienced cardiologists, blind to the ultrasound evaluation of the carotid, femoral, and popliteal arteries. The IMT of the carotid artery (IMTC), the IMT of the femoral artery (IMTF), and the IMT of the popliteal artery (IMTP) were evaluated at the far wall of each artery 2 cm proximal to their bifurcation. The images were zoomed to standard size. The IMTC was calculated as the mean value of six individual measurements at different points within the region of interest (three for the right and three for the left carotid). The IMTF and IMTP were estimated using the same technique. Ultrasonic Biopsy Estimation Ultrasonic biopsy (UB) assessment was performed using the method originally described by Belcaro et al. 8 The carotid and femoral bifurcations were localized by a transverse scan. The probe was rotated 90 to obtain and record the longitudinal image of both the anterior and posterior wall, with the latter being used for evaluation. Both arteries were evaluated for a length of 3 cm (1.5 cm proximally and distally to the flow divider). The initial classification included five classes corresponding to five scores, ranging from 0 to 8 for each artery (Table 1). The subject s ultrasound score was obtained by dividing the sum of the scores of the four arteries by four. Ultrasound evaluation, including IMT measurement and UB, was performed blind to coronary angiography. We did not perform an assessment of interobserver and intraobserver variability, since our experience in using the ultrasound technique in estimating peripheral vessels is well documented. 9 12 Statistical Analysis All descriptive statistics are presented as mean SD. The differences between patients and control subjects in the IMT of each artery and the UB were assessed using the unpaired t test. Stepwise logistic regression analysis was used to identify the independent prognostic factors among the parameters recognized as being statistically significant. The specificity and sensitivity of our results were evaluated by receiver operator characteristic (ROC) analysis. The UB and IMT of each artery in patients with different severities of CAD were compared using analysis of variance and post hoc Bonferroni adjusted pair-wise comparisons. Probability values 0.05 were considered statistically significant. Detection of CAD Results The mean age of the participants was 61.7 8.3 years. Elective coronary arteriography revealed the presence of obstructive CAD in 103 of 184 individuals (patient group), while the remaining 81 subjects Ultrasound Evaluation IMT Measurement: All scans were performed by one observer following the method described by Geroulakos et al, 7 using a color duplex scanner (Sequoia TM 512; Acuson Corporation; Mountain View, CA) with an 8-MHz linear array scan head. Scan settings (power output, 50%; dynamic range, 60 decibels; gain; gray scale; filters; ramp) were preset at machine startup and remained constant during examination. All subjects were initially examined in a supine position with a slight extension of the neck. The anterior and lateral projections were used in order to image the common carotid and common femoral arteries longitudinally. The popliteal arteries were evaluated in a decubitus position. Class I (score 0) II (score 2) III (score 4) IV (score 6) V (score 8) Table 1 UB Classification Description Normal arterial wall Luminal interface disruption (intervals of 0.5 cm) Intima-media granulation or increased intima-media thickness 1mm Plaque without hemodynamic disturbance Asymptomatic atherosclerotic plaque with hemodynamic disturbance www.chestjournal.org CHEST / 128 / 4/ OCTOBER, 2005 2539

Table 2 Clinical Characteristics of Participants* Characteristics Total Patient Group Control Group p Value Subjects, No. 184 103 81 Age, yr 61.74 8.35 62.92 8.43 60.23 8.04 0.835 Male gender 147 (79.9) 83 (80.6) 64 (79.1) 0.937 Diabetes mellitus 39 (21.2) 25 (24.3) 14 (17.3) 0.332 Hypercholesterolemia 111 (60.3) 60 (58.2) 51 (63) 0.546 Smoking 103 (56) 59 (57.3) 44 (54.3) 0.801 Hypertension 82 (44.5) 49 (47.6) 33 (40.7) 0.437 Elevated cardiac markers 94 (51) 59 (57.3) 35 (43.2) 0.327 Diagnostic ECG changes 132 (71.7) 73 (70.9) 59 (72.8) 0.448 *Data are presented as mean SD or No. (%) unless otherwise indicated. Cardiac troponin I 1.5 ng/ml and/or creatine phosphokinase 5%. New or presumed new ST-segment depression or T-wave abnormalities, or both, in two or more contiguous leads; also, new or presumed new symmetric inversion of T waves 1 mm in at least two contiguous leads. Unpaired t test and the 2 test were used to compare age and other parameters, respectively, between the two groups. (control group) had no angiographically obstructive lesions. The clinical characteristics were similar in both groups (Table 2). Univariate analysis showed that IMTC, IMTF, IMTP, and UB had significantly higher values in patients with obstructive CAD than in control subjects (Table 3). ROC analysis showed that all four indexes yielded a significant area (p 0.05) under the ROC curve (0.81, 0.73, 0.71, and 0.77 for IMTC, IMTF, IMTP, and UB, respectively). However, as Figure 1 shows, IMTC was superior to others almost uniformly. The sensitivity and specificity of IMTC, IMTF, IMTP, and UB for the prognosis of obstructive CAD are shown in Table 4. Stepwise binary logistic regression showed that among IMTC, IMTF, IMTP, and UB, only IMTC (odds ratio, 5.3; 95% confidence interval, 2.6 to 10.6) and IMTF (odds ratio, 2.6; 95% confidence interval, 1.3 to 4.9) were independently associated with obstructive CAD. Specifically, IMTC entered first ( 2 56.3, p 0.001), followed by IMTF ( 2 8.4, p 0.005). Finally, based on the model, the estimated probability that a patient has obstructive CAD as a function of IMTC and IMTF is given by the following equation (Fig 2): p (CAD) e ( 4.765 3.36 IMTC 1.91 IMTF) / 1 e ( 4.765 13.36 IMTC 1.91 IMTF). The model fit did not change when variables such as age, sex, smoking, diabetes mellitus, hypertension, or hypercholesterolemia were forced in the model. Severity of CAD Table 5 shows the values of all the indexes in patients with obstructive CAD according to the number of affected vessels. Analysis of variance showed significant differences for IMTC (p 0.001) and UB (p 0.02), and post hoc Bonferroni-adjusted pairwise comparison revealed that patients with one-vessel disease had significantly lower IMTC (p 0.001), UB (p 0.011), and lower IMTF (p 0.057) than those with three-vessel disease. Discussion Atherosclerosis is a generalized, progressive disease that may simultaneously affect several arterial Table 3 IMT and UB in Patients of Both Groups* Variables Patient Group (n 103) Control Group (n 81) p Value IMTC, mm 1.16 0.31 0.81 0.24 0.001 IMTF, mm 1.06 0.33 0.78 0.21 0.001 IMTP, mm 0.90 0.28 0.71 0.16 0.001 UB 3.94 1.29 2.65 1.07 0.001 *Data are presented as mean SD. See Table 1 for UB classification. Figure 1. ROC curves of IMTC, IMTF, IMTP, and UB for CAD. It can clearly be seen that IMTC is superior to the others. 2540 Clinical Investigations

Table 4 Sensitivity and Specificity of IMT and UB for the Prognosis of Obstructive CAD Variables Cutoff Value for CAD Sensitivity, % Specificity, % IMTC, mm 0.88 76 74 IMTF, mm 0.77 70 60 IMTP, mm 0.70 73 56 UB* 3.25 69 70 *See Table 1 for UB classification. trees of the body. 13 Among other efforts in the direction of management of atherosclerosis, early detection of subclinical (asymptomatic) CAD and subsequent prevention of possible future ischemic events is one of our most important tasks. Since the seminal work of Hertzer et al 14 and Martinez et al 15 in the early 1980s, we have accepted that patients with peripheral arterial occlusive disease frequently have concomitant CAD, which remains the leading cause of both early and late mortality following peripheral vascular reconstructive surgery. 14,15 Along these lines, atherosclerotic changes of peripheral arteries might be considered to mirror the condition of coronary arterial circulation. In this study, we tried to establish an association between subclinical arterial wall changes, as detected by ultrasonography, in three different peripheral arterial territories, including the carotid, femoral, and popliteal arteries, and the presence and severity of obstructive CAD. To our knowledge, such a study evaluating individuals with no clinical signs of peripheral arterial occlusive disease and incorporating those three peripheral arteries concurrently in the same patient population has not been reported yet in the English-language literature. According to our results, IMT changes of both the Figure 2. Estimated probability (prob) of CAD as a function of IMTC and IMTF. Table 5 IMT and UB in Patients With CAD According to the Number of Affected Vessels* Variables One-Vessel Disease (n 29) Two-Vessel Disease (n 33) Three-Vessel Disease (n 41) p Value IMTC, mm 0.99 0.25 1.15 0.24 1.27 0.35 0.001 IMTF, mm 0.94 0.36 1.08 0.32 1.12 0.30 0.057 IMTP, mm 0.82 0.27 0.93 0.27 0.93 0.30 0.269 UB 3.41 1.26 3.95 1.25 4.32 1.25 0.02 *Data are presented as mean SD. See Table 1 for UB classification. carotid and femoral arteries appeared to be independent predictors of both the existence and the severity of obstructive CAD. In addition, our analysis yielded for the first time a regression model that for a given value of IMTC and IMTF may estimate the probability of obstructive CAD. In contrast, the IMT of the popliteal artery and UB do not seem to play a significant predictive role in detecting CAD. Our results for IMTC are in line with those of previous isolated clinical reports. In the Rotterdam study 16 an increased IMTC was found to be associated with future cerebrovascular and cardiovascular events. Similarly, in the Atherosclerosis Risk in Communities study, 2 IMTC was related to clinically manifested cardiovascular disease affecting distant vascular beds. The correlation between carotid and coronary atherosclerosis has also been confirmed when it was assessed by imaging techniques only. In a study by Mack et al, 17 the rate of changes of the IMTC detected by ultrasonography was positively correlated with the changes in coronary artery atherosclerosis determined by angiography. Additionally, Geroulakos et al 7 showed for the first time that the IMTC was associated not only with the presence of atherosclerotic risk factors but also with the presence of angiographic CAD data, and that there was a significant linear trend between increasing IMT and the number of coronary vessels involved. Other relative studies 18,19 used the carotid bulb disease or employed more complex systems to evaluate the IMT at different sites of the carotid artery including the carotid bulb. At this point, it should be pointed out that IMT changes observed in the present study were detected in the absence of focal or diffuse plaque formation and therefore in the absence of hemodynamically or clinical significant atherosclerotic disease. The benefits of estimating the IMT of the common carotid artery are that it is typically constant throughout its length, it can be accurately and easily assessed by ultrasound, and plaques are found in the vessel only during the late stages of atherosclerosis. 7 In contrast, the IMT and plaques of the carotid bulb cannot always be visualized adequately. 20 www.chestjournal.org CHEST / 128 / 4/ OCTOBER, 2005 2541

The available data on the relation of IMTF and CAD are few and conflicting. In this study, we found that IMTF might serve as an indicator and as an independent risk factor for the presence of obstructive CAD. Moreover, we observed that patients with one-vessel disease had relatively, but not significantly, lower IMTF (p 0.057) than those with three-vessel disease. Megnier et al 21 reported that IMTF predicts the existence of coronary calcium as assessed by ultrafast CT and concluded that IMTF could be of clinical value for stratifying CAD risk. Additionally, Held et al 1 found later that IMTF is related to the risk of myocardial infarction and revascularization. Furthermore, Lekakis et al 22 reported that IMTF is a strong predictor of the extent and severity of coronary atherosclerosis; however, this strong correlation was observed by using the Gensini score, a more general marker for the estimation of the severity of CAD. Moreover, Lekakis et al 22 included patients with symptomatic peripheral disease, who were excluded in our study. Patients with already symptomatic peripheral disease can be expected to have more severe arterial disease. However, Hulthe et al, 18 in a relatively small study, did not find a significant relation between IMTF and the severity of CAD as evaluated by angiography. The discrepancy between the results of our study and those reported by Hulthe et al 18 could be explained by the small number of patients included in their study and the different methodology used for evaluating the extent of coronary artery atherosclerosis, since they tried to correlate IMTF with the degree of stenosis and not the number of diseased coronary vessels. Regarding IMTP, there are few studies 2,3 in the literature concerning its association with preexisting CAD; in these studies, CAD was positively associated with increased wall thickness in the popliteal arteries. Our observation that patients with obstructive CAD had generally higher values of IMTP are in line with those findings, even though in our study we found that IMTP was not an independent factor for obstructive CAD. Apart from IMT, Belcaro et al 5,6 found that the UB carotid and femoral classification of subclinical atherosclerotic lesions can separate asymptomatic subjects into groups of different risk for cardiovascular events. They concluded that UB might serve as a simple technique to assess subclinical atherosclerosis and could be useful in selecting subjects prone to have cardiovascular events. In our study, we found that UB does indeed have higher values in patients with obstructive CAD, but it is not an independent diagnostic factor. However, our findings indicate that UB might be useful for determining which CAD patients have significant disease. Certainly, further studies are needed in order to elucidate this matter. The aim of this study was to evaluate whether subclinical atherosclerotic disease occurring concurrently in multiple peripheral arteries (eg, common carotid, common femoral, and popliteal) can mirror the presence and severity of obstructive CAD in an adult patient population with no history or signs of peripheral arterial disease. The size of our study population may be considered small for such a kind of investigation. However, it was enough to demonstrate the strong relationship between the early arterial wall changes and presence and extent of CAD, at the same time yielding a regression model that for a given value of IMTC and IMTF may estimate the probability of CAD. The importance of our study is that the development of obstructive CAD may be reflected by preceded IMT changes in multiple peripheral arteries, which can be easily and accurately detected by ultrasound. This could contribute to a more effective identification of patients at risk of future ischemic heart disease. Certainly, some factors may have limited the apparent strength of the suggested regression model or the relationships we found between the arterial wall changes in different peripheral arteries and CAD. Thus, classification of the severity of CAD based on the number of the diseased vessels may not be as accurate as other specific indexes and coronary angiography frequently underestimates the severity of atherosclerotic disease. 23 Additionally, our control group of patients were referred for coronary angiography for suspected ischemic heart disease. This selection criterion mean that our results cannot necessarily be applied to the general population or extrapolated to other patient populations selected by other criteria. Regarding the regression model, studies of larger samples will be necessary to confirm our results before this model can be used as a diagnostic tool in this specific group of patients. References 1 Held C, Hjemdahl P, Eriksson SV, et al. Prognostic implication of intima-media thickness and plaques in the carotid and femoral arteries in patients with stable angina pectoris. Eur Heart J 2001; 22:62 72 2 Burke GL, Evans GW, Riley WA, et al. Arterial wall thickness is associated with prevalent cardiovascular disease in middleaged adults (The ARIC Study). Stroke 1995; 26:386 391 3 Bucciarelli P, Sramek A, Reiber JH, et al. Arterial intimamedia thickness and its relationship with cardiovascular disease and atherosclerosis: a possible contribution of mediumsized arteries. Thromb Haemost 2002; 88:961 966 4 Smith SC Jr, Greenland P, Grundy SM. Prevention conference V: Beyond secondary prevention; identifying the highrisk patient for primary prevention. Circulation 2000; 101: 111 116 2542 Clinical Investigations

5 Belcaro G, Nicolaides AN, Laurora G, et al. Ultrasound morphology classification of the arterial wall and cardiovascular events in a 6-year follow-up study. Arterioscler Thromb Vasc Biol 1996; 16:851 856 6 Belcaro G, Nicolaides AN, Ramaswami G, et al. Carotid and femoral ultrasound morphology screening and cardiovascular events in low risk subject: a 10-year follow-up study (the CAFES-CAVE study). Atherosclerosis 2001; 156:379 387 7 Geroulakos G, O Gorman DJ, Kalodiki E, et al. The carotid intima-media thickness as a marker of the presence of severe symptomatic coronary artery disease. Eur Heart J 1994; 15:781 785 8 Belcaro G, Laourara G, Cesarone MR, et al. Evaluation of atherosclerosis progression with ultrasonic biopsy and intimamedia thickness measurements. Vasa 1993; 22:15 21 9 Kardoulas DG, Katsamouris AN, Gallis PT, et al. Ultrasonographic and histologic characteristic of symptom- free and symptomatic carotid plaque. Cardiovasc Surg 1996; 4:580 590 10 Katsamouris AN, Giannoukas AD Tsetis D, et al. Can ultrasound replace arteriography in the management of chronic arterial occlusive disease of the lower limb? Eur J Vasc Endovasc Surg 2001; 21:155 159 11 Giannoukas AD, Kostas T, Ioannou C, et al. Perforator reflux and clinical presentation in primary superficial venous insufficiency. Eur J Vasc Endovasc Surg 2003; 25:88 89 12 Kostas T, Ioannou CV, Touloupakis E, et al. Recurrent varicose veins after surgery: a new appraisal of a common and complex problem in vascular surgery. Eur J Vasc Endovasc Surg 2004; 27:275 282 13 Ross R. Atherosclerosis: an inflammatory disease. N Engl J Med 1999; 1340:115 126 14 Hertzer NR, Beven EG, Young JR, et al. Coronary artery disease in peripheral vascular patients. Ann Surg 1984; 199:223 15 Martinez BD, Hertzer NR, Beven EG. Influence of distal arterial occlusive disease on prognosis following aortobifemoral bypass. Surgery 1980; 88:795 805 16 Bots ML, Hoes AW, Koudstaal PJ, et al. Common carotid intima- media thickness and risk of stroke and myocardial infarction (The Rotterdam Study). Circulation 1997; 96:1432 1437 17 Mack WJ, LaBree L, Liu CR, et al. Correlations between measures of atherosclerosis change using carotid ultrasonography and coronary angiography. Atherosclerosis 2000; 150: 371 379 18 Hulthe J, Wikstrand J, Emanuelsson H, et al. Atherosclerotic changes in the carotid artery bulb as measured by B-mode ultrasound are associated with the extent of coronary atherosclerosis. Stroke 1997; 28:1189 1194 19 Ebrahim S, Papakosta O, Whincup P, et al. Carotid plaque, intima media thickness, cardiovascular risk factors, and prevalent cardiovascular disease in men and women (The British Regional Heart Study). Stroke 1999; 30:841 850 20 Craven TE, Ryu JE, Espeland MA, et al. Evaluation of the associations between carotid artery atherosclerosis and coronary artery stenosis. Circulation 1990; 82:1230 1242 21 Megnier JL, Simon A, Gariepy J, et al. Preclinical changes of extracoronary arterial structures as indicators of coronary atherosclerosis in men. J Hypertens 1998; 16:157 163 22 Lekakis JP, Papamichael CM, Cimponeriu AT, et al. Atherosclerotic changes of extracoronary arteries are associated with the extent of coronary atherosclerosis. Am J Cardiol 2000; 85:949 952 23 Kallikazaros I, Tsioufis C, Sideris S, et al. Carotid artery disease as a marker for the presence of severe coronary artery disease in patients evaluated for chest pain. Stroke 1999; 30:1002 1007 www.chestjournal.org CHEST / 128 / 4/ OCTOBER, 2005 2543