The Systolic Hypertension in the Elderly Program

44 Blood Pressure Treatment Slows the Progression of Carotid Stenosis in Patients With Isolated Systolic Hypertension Kim Sutton-Tyrrell, DrPH; Sidney K. Wolfson, Jr, MD; Lewis H. Kuller, MD Background and Purpose The Systolic Hypertension in the Elderly Program (SHEP) was a randomized trial testing the efficacy of treating systolic hypertension in older adults. A significant reduction in stroke risk was observed among participants assigned to active treatment. Serial carotid duplex scans were performed on 129 participants at the University of Pittsburgh center, and rates of progression and regression of carotid stenosis were observed. Methods Changes in blood flow velocity ratios were used to detect progression because they can be reliably measured and their relation to degree of residual lumen is known. Progression required the development of a 40% to 50% diameter stenosis when stenosis was not initially present or, if already present, further reduction in the lumen diameter. Regression required the absence of a 40% to 50% diameter stenosis when stenosis was initially present or a stenosis significantly less severe than that initially seen. The Systolic Hypertension in the Elderly Program (SHEP) has shown that the treatment of systolic hypertension reduces the risk of both stroke and cardiovascular disease. 1 This may result from a slowing of atherosclerosis progression when blood pressure is controlled. Serial duplex scans of the carotid arteries (two scans separated by 2 years) were obtained in 129 SHEP participants at the University of Pittsburgh field center. These data were evaluated to determine the effect of blood pressure treatment on progression of carotid artery stenosis. Methods SHEP was a randomized, placebo-controlled trial to test the efficacy of treatment of isolated systolic hypertension in adults over age 60. 2 Participants randomized to active treatment received a diuretic and/or /3-blocker for blood pressure control. Those randomized to placebo received matching placebo medication. Results of SHEP demonstrated a 36% reduction in stroke incidence among the active treatment group. 1 As part of an ancillary study of peripheral atherosclerosis, participants at the University of Pittsburgh field center were invited to undergo duplex scanning of the carotid arteries. Two Received August 13, 1993; final revision received September 27, 1993; accepted October 11, 1993. From the Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh (K.S.-T., L.H.K.), and the Department of Surgery, School of Medicine and Montefiore University Hospital, University of Pittsburgh (S.K.W.), Pa. Correspondence to Kim Sutton-Tyrrell, DrPH, Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, 130 DeSoto St, Pittsburgh, PA 151. Results Progression occurred in 22% (28/129) of participants and regression in 16% (8/49). Progression of carotid stenosis occurred more often among participants randomized to placebo as compared with active treatment (31% versus 14%, P=.Q20). All eight patients exhibiting regression were randomized to active treatment. In multivariate analysis, participants assigned to placebo had 4.3 times greater odds of progressing than participants assigned to active treatment. Other factors significantly related to progression were higher degree of plaque at baseline, low high-density lipoprotein-3, high lipoprotein(a), and younger age. Conclusions Treating systolic hypertension appears to slow progression of carotid stenosis. Similar effects occurring in the intracranial vessels may be one reason for the substantial decrease in stroke among SHEP participants assigned to active treatment. (Stroke. 1994;:44-50.) Key Words carotid artery disease hypertension stenosis duplex scans were performed separated by 2 years. For all participants, the baseline scan was performed after they began taking active or placebo medication. The 129 individuals included in this report were consecutive participants who had their baseline scan at least 6 months before the end of the SHEP trial. The average time from SHEP entry to the baseline scan was 42 months. There was an average of months between the two scans with a range of 21 to 34 months. Participants underwent duplex scanning at the Peripheral Vascular Diagnostic Laboratory located in Montefiore University Hospital, Pittsburgh, Pa. Both baseline and follow-up scans were performed using a Diasonics DRF 400 duplex scanner (Milpitas, Calif) with a 10-MHz imaging probe and 4.5-MHz Doppler. B-mode images were used to assess minimal to moderate disease that had not yet caused changes in blood flow. Scans were recorded on videotape for later scoring. A reader assigned a grade from 0 to 3 to each of seven segments in the carotid system based on the number and size of lesions present. Plaque grades from segments in the common carotid artery (CCA), carotid bulb, and internal carotid artery (ICA) from both the right and left carotid systems then were summed to create a measure of extent of atherosclerotic plaque called the plaque index. A complete discussion of this method has been previously published. 3 ' 4 Doppler measures of blood flow velocity were used to determine the presence of a carotid stenosis. Doppler measurements obtained by duplex scanning have shown good agreement with angiography in the identification of carotid stenosis. 58 Visual estimates of stenosis can vary greatly, especially when plaque formation is asymmetric. This is because the apparent size of an asymmetric plaque and the associated residual lumen changes depending on the angle of evaluation. As a vessel lumen narrows, regardless of the symmetry of plaque formation, the velocity of blood flow at

Sutton-Tyrrell et al Stenosis Progression and Blood Pressure 45 TABLE 1. Carotid Disease Progression and Regression in the Pittsburgh SHEP Cohort Carotid Systems (n=8) Participants (n=129) Progression ICA ECA CCA/Bulb Any location Regression ICA ECA CCA/Bulb Any location n 21 15 2 34 5 6 0 9 % 8.1 5.8 0.8 13.2 SHEP indicates Systolic Hypertension in the Elderly Program; ICA, internal carotid artery; ECA, external carotid artery; and CCA, common carotid artery. 1.9 2.3 0.0 3.5 n 19 14 2 28 5 6 0 8 % 14.7 10.9 1.6 21.7 3.9 4.7 0.0 6.2 that site increases. This direct relation between blood velocity and residual vessel lumen makes the quantification of stenosis by Doppler more accurate than visual estimates of stenosis. 9. 10 Assuming that the CCA is relatively free of significant disease, the ratio of internal carotid blood flow velocity to common carotid blood flow velocity (ICA-CCA ratio) is a measure of stenosis that controls for intersubject variation. 5 ICA stenosis was defined as an ICA velocity of 90 mm/s or greater accompanied by an ICA-CCA ratio of 1.4 or above. This definition was based on studies comparing velocity ratios in normal patients with ratios in patients with angiographically documented carotid stenosis. 5 ' 11 This corresponds to a luminal diameter reduction of approximately 40% to 50% or greater. 9 While this definition does not always represent clinically important disease, it represents the lowest level of disease that can be reliably detected by Doppler scan. Velocity ratios were also used to determine carotid stenoses in the external carotid artery (ECA), carotid bulb, and CCA by comparing the elevated velocity in the location of interest with the velocity in a relatively normal portion of the CCA proximal to the area of interest. In these locations, stenosis was defined as a velocity of 100 mm/s or greater at the area of interest accompanied by a velocity ratio of 1.4 or greater. At the beginning of the study, a decision was made that progression of disease would include all areas of the carotid system, not just the ICA. Before analysis of the data, changes in the bloodflowvelocity and velocity ratios were used to ascertain progression. Because Doppler measures are not useful in detecting stenoses of less than 40% to 50% diameter reduction, progression or regression of disease less than this could not be included. Progression was defined as the development of 40% to 50% stenosis when stenosis was not present on the first scan or, if a stenosis was already present, further reduction in the cross-sectional luminal diameter. More specifically, progression in the ICA required an increase in blood flow velocity of at least 20 mm/s to at least 90 mm/s, accompanied by an increase in the ICA-CCA ratio of at least 0.4 to at least 1.4. Criteria for the ECA were more stringent, requiring an increase in blood flow velocity of at least 30 mm/s to at least 100 mm/s and an increase in the ECA-CCA ratio of at least 0.4 to at least 1.4. Criteria for the CCA and carotid bulb were even more stringent, requiring an increase in bloodflowvelocity of at least 40 mm/s and an increase in the velocity ratio of at least 0.4 to at least 1.4. Finally, for any artery, if a velocity of 140 mm/s or greater was present on the baseline scan, the requirement for increase in bloodflowvelocity was raised to 40 mrn/s. A vessel exhibiting a high velocity at baseline and found to be occluded at follow-up was also considered progression. To explore the possibility of regression of carotid stenosis, the algorithm for progression was executed in reverse, comparing the second scan with the first. For example, for the ICA, blood flow velocity had to decrease by 20 mm/s accompanied by a reduction of the ICA-CCA ratio of 0.4. If the ICA velocity remained 140 mm/s or greater on the second scan, then a decrease in velocity of 40 mm/s or greater was required accompanied by a reduction in the ICA-CCA ratio of at least 0.4. Regression could therefore only be identified if a stenosis was present at baseline. The study was conducted under a strict research protocol designed to reduce measurement error. Two sonographers performed 95% (6/8) of all scans. Peak blood flow velocities and velocity ratios obtained by these sonographers were highly reproducible in participants who had duplicate scans on the same day. 4 When progression and regression definitions were applied to these duplicate scans, 3% of patients showed progression and 3% showed regression. These represent the error rates in our measurement of progression and regression. Fasting blood samples were drawn at the time of the baseline carotid scan for determination of lipids and glucose. Standard laboratory methods were used to evaluate triglycerides, total cholesterol, high-density lipoprotein (HDL) cholesterol, HDL-2, HDL-3, apoproteins Al and A2, and apoprotein B. Low-density lipoprotein (LDL) cholesterol was estimated using the Friedwald equation. 12 Lipoprotein(a) (Lp[a]) was measured using a commercially available kit that employs an enzyme-linked immunosorbant assay. A x 2 test was used to determine the significance of differences in proportions, with Fisher's exact test used where appropriate. Differences in means were determined by a t test for normally distributed variables and a Wilcoxon test for variables not normally distributed. Independent predictors of progression were determined by logistic regression. The Wilcoxon signed-rank test for paired data was used to determine differences in the plaque index between baseline and follow-up scans. Finally, multiple linear regression was used to determine the relation between treatment assignment and change in plaque index while controlling for baseline differences between active and placebo groups. A value of/"=.o5 was considered significant. Results risk factors were compared between the 129 individuals in this ancillary study and the full SHEP cohort. The only factor that was significantly different was smoking history. Individuals in the ancillary study were significantly less likely to have ever smoked com-

46 Stroke Vol, No 1 January 1994 TABLE 2. Raw Velocity Data for 38 Arteries (34 Participants) Showing Progression Months Between Scans ICA 116 141 1.47 2.66 58 96 1.14 1.85 27 90 1 1.76 2.36 196 4 4.00 8.50 78 108 1.50 2.12 28 80 112 1.36 1.96 184 2.86 5.11 29 122 366 2.18 6.65 121 172 2.16 3.31 88 114 1.47 1.87 0 Occlusion 3.91 67 129 0.89 1.87 172 1.30 2.02 83 1.28 1.71 221 Occlusion 2.99 133 209 2.46 3.48 61 107 0.95 1.51 56 132 0.92 2.40 209 3 3.87 7.58 52 1 0.91 2.02 29 162 280 3. 5.38 29 ECA 64 62 77 64 187 81 76 63 1 81 135 64 111 116 CCA/Bulb 49 115 174 113 355 127 170 100 163 153 184 103 200 4 197 1 170 0.89 1.02 1.64 0.75 1.75 1.62 1.12 0.80 1.54 1. 1.82 1.05 2.07 1.73 1.90 1.00 1.64 3.84 1.97 2.46 1.60 3. 2.54 2.46 1.43 1.96 2.19 2.45 1.72 4. 3.44 3.58 3.00 2.74 ICA indicates internal carotid artery; ECA, external carotid artery; and CCA, common carotid artery. pared with the full SHEP cohort (37% versus 50%, P<.001). Progression occurred in 22% (28/129) of participants. Regression could only be assessed in the subgroup of 49 patients that had a stenosis (any location) at baseline. Of this subgroup, 8 participants (16%) showed regression. Progression was seen primarily in the ICA (21 arteries), but was also observed in the ECA (15 arteries) and CCA or bulb (1 CCA, 1 bulb) (Table 1). The median increase in blood flow velocity for ICAs that progressed was 52 mm/s compared with 1 mm/s for ICAs not showing progression. Those showing regression had a median decrease in blood flow velocity of 40 mm/s. Similar results were seen for the ECA. Raw blood flow velocity data for arteries showing progression and regression are presented in Tables 2 and 3.

Sutton-Tyrrell et al Stenosis Progression and Blood Pressure 47 TABLE 3. ICA ECA Raw Velocity Data for 11 Arteries (8 Participants) Showing Regression 129 200 97 98 140 103 200 107 119 100 74 134 74 78 78 92 65 75 74 69 ICA indicates internal carotid artery; ECA, external carotid artery. 2.08 1.87 1.73 2.07 1.92 1.41 1.66 4.08 1.70 2.09 1.67 0.97 1.22 1.19 1.20 1.50 0.81 0.86 2.36 1.27 1.14 1.13 Months Between Scans 21 Before analyzing results by treatment assignment, the active and placebo groups were compared with respect to baseline risk factors (Table 4). The two groups were similar, although the active treatment group had a significantly higher median Lp(a) value, a lower proportion of males, a higher mean HDL level, and were less likely to have ever smoked. The mean number of months between scans was.9 for the active treatment group and.0 for the placebo group. In univariate analysis, 31% of those randomized to placebo progressed whereas only 14% of those randomized to active treatment progressed (P=.O2O, Table 5). All 8 patients who showed regression were randomized to active treatment. Among the group with a stenosis at baseline, 32% (8/) of active treatment participants regressed versus 0% (0/) of placebo participants (/>=.004, Fisher's exact test; Table 5). Several other factors were found to be related to progression, the strongest of which was the degree of plaque observed on the baseline scan. Progression rates ranged from 3% among those with a plaque index less than 3 to 54% among those with a plaque index of 9 or greater (Figure). Other baseline variables significantly related to progression were higher Lp(a), apoprotein B or triglycerides, and lower HDL, HDL-2, or HDL-3 (Table 6). Mean reduction in pretreatment blood pressure was less for those who progressed than those who did not (18 mm Hg versus mm Hg), although this relation did not reach statistical significance. Stepwise logistic regression was used to determine whether SHEP treatment assignment was independently associated with progression (Table 7). In this model, placebo treatment assignment was associated with 4.3 times greater odds of progression than an active TABLE 4. Differences in Variables Between Participants Randomized to Active Treatment Versus Placebo Mean age, y Median baseline systolic BP, mm Hg Median baseline diastolic BP, mm Hg Median triglycerides Mean total cholesterol Mean HDL cholesterol Mean HDL-2 Mean HDL-3 Mean LDL Mean apoprotein B Median lipoprotein(a) Median plaque index %Male % Ever smokers Active (n=71) 76 168 79 229 54 12 42 144 105 16 3.0 27 27 Placebo (n=58) 75 168 81 122 2 50 10 40 146 104 13 5.0 52 50 BP indicates blood pressure; HDL, high-density lipoprotein; and LDL, low-density lipoprotein. Laboratory values reported in mg/dl. P.575.570.063.851.614.043.057.135.741.977.0.197.004.007

48 Stroke Vol, No 1 January 1994 TABLE 5. Progression and Regression* of Carotid Stenosis by SHEP Treatment Assignment % Showing Progression 100 N n % 129 28 21.7 Active 71 10 14,1 Placebo 58 18 31.0 Progression Regression* 49 8 16.3 Active 8 32.0 Placebo 0 0.0 Regression could only be assessed in participants who had a stenosis (any location) at baseline. treatment assignment. Additional variables related to progression were greater extent of plaque at baseline, HDL-3 <30 mg/dl, Lp(a) >20 mg/dl, and younger age. Other variables that were significantly associated with progression in univariate analysis were no longer significant after controlling for the variables listed above. Years of smoking, HDL (in place of HDL-3), and diastolic blood pressure contributed little to the model and were therefore removed. While the B-mode information was not the primary focus of this study, the plaque index was analyzed to determine if treatment assignment was also related to an increase in extent of plaque. For the group as a whole, the plaque index was significantly higher for the follow-up scans compared with baseline. The mean plaque index was 4.8 at baseline and 5.7 at follow-up (/><.001). The average change in plaque index was +0.89. The average change was higher for the placebo group ( + 1.) compared with the active treatment group (+0.61). While this difference was not statistically significant in univariate analysis (P=.O9O), it was significant after adjusting for baseline differences between active and placebo groups (/>=.050). Changes in the plaque index for those with and without stenosis progression were also compared. The average change in plaque index was +0.66 for those 3-5 6-8 Plaque Index 9+ Relation between degree of plaque at baseline and rates of carotid stenosis progression. The plaque index was calculated by scoring the number and size of plaques in the common carotid artery, carotid bulb, and internal carotid artery. without a stenosis progression compared with +1.68 for those with a stenosis progression (P=.O27). Discussion This study provides strong evidence that the treatment of isolated systolic hypertension slows the rate of carotid stenosis progression. The primary analysis defined progression in terms of blood flow because we believe that this is the best index of atherosclerosis when dealing with a population that has a high prevalence of advanced disease. Doppler measures were highly reproducible and observed progression rates were much higher than those that would have been due to measurement error alone. The use of blood flow velocity ratios in the progression definition has the advantage of controlling for other factors that may affect blood flow velocity. Such factors would have affected the CCA as well as areas more distal in the carotid system. There is much evidence to suggest that disease in the carotid arteries is a marker for generalized systemic ath- TABLE 6. Differences in Variables Between Participants Who Did and Did Not Show Progression of Carotid Stenosis Progression (n=28) No Progression (n=101) P Mean reduction in systolic BP from baseline, mm Hg 18.117 Mean age, y 73 76.113 Median triglycerides 153.016 Mean total cholesterol 2 1.475 Mean HDL cholesterol 46 54.004 12.2.015 41.8.011 Mean HDL-2 Mean HDL-3 8 5 37.3 Mean LDL 146 145.661 Mean apoprotein B 113 102.0 Median lipoprotein(a) 12.036 % Ever smokers 50 33.093 BP indicates blood pressure; HDL, high-density lipoprotein; and LDL, low-density lipoprotein. Laboratory values reported in mg/dl.

Sutton-Tyrrell et al Stenosis Progression and Blood Pressure 49 TABLE 7. Variable Independent Predictors of Carotid Stenosis Progression Placebo treatment assignment Plaque index (each 3-unit increment) HDL-3 <30 mg/dl Lipoprotein(a) >20 mg/dl Age (each 10-year increment) Male HDL indicates high-density lipoprotein. Odds Ratio 4.3 2.2 9.9 4.3 0.4 1.1 95% Confidence Interval 1.3-14.1 1.3-3.5 1.8-53.4 1.3-14.7 0.14-0.97 0.3-3.4 P.017.002.007.019.044.904 erosclerosis. 1314 Past studies have focused primarily on the ICA because of its clinical relation to stroke. Whereas disease in other areas of the carotid system is not always clinically important, researchers have begun to use such disease as a marker for generalized atherosclerosis. 1516 The progression definition used here thus encompassed all areas of the carotid system, not just the ICA. Other studies of carotid disease have used a variety of techniques to measure progression. Some studies have categorized carotid disease by use of angiography 17 or Doppler spectral analysis, 18 while others have focused specifically on changing characteristics of plaques. 19 More recently, authors have concentrated on change in intima-media thickness over time. 20-22 Intima-media thickness measures are not yet available for this study because these techniques were not well documented at the time data collection was initiated. However, analysis of B-mode data in the form of the plaque index suggests that progression of carotid stenosis was accompanied by a progression of atherosclerotic plaque. It has been shown that treatment of systolic hypertension reduces the risk of stroke,1 - and this effect occurs within a year or two after blood pressure treatment. This rapid response to treatment may involve pathophysiological mechanisms over and above a delay in progression of individual plaques. In addition to lowering blood pressure, antihypertensive therapy may also cause changes in distal small vessel reactivity or changes in arterial wave propagation. Either or both of these may contribute to a reduction in cardiovascular events. Previous studies have shown a direct relation between the degree of carotid stenosis and risk of both stroke and coronary heart disease. 14 We have previously shown that individuals with isolated systolic hypertension have a much higher prevalence of carotid stenosis than age-similar normotensive control subjects. 3 The treatment effect on carotid stenosis presented here may be a demonstration of what is ongoing in the intracranial and coronary vessels. This is likely to be at least part of the reason for the reduction in stroke and myocardial infarction risk among SHEP participants randomized to active blood pressure treatment. One weakness of this study is that the duplex scans were not obtained earlier in the study, before treatment. Unfortunately, the SHEP trial ended before all participants had completed their follow-up scans. Observed treatment effects would probably have been stronger if the scans could have been performed earlier, and more conclusive data on regression might have been obtained. In previous literature, the most consistent factor found to be related to progression has been higher degree of carotid disease at baseline, 17-20 - and this is consistent with the results reported here. Interestingly, older participants were less likely to progress than younger participants. This may have resulted from a selection bias related to the exclusion criteria for the SHEP trial. Older individuals were more likely to have advanced atherosclerosis and therefore be excluded. This may have caused the study population to have an under-representation of older individuals who would have been most likely to progress. Similar results published by Roederer et al 18 reported that patients under 65 years of age were most likely to show progression. Other studies have shown relations between various lipid levels and carotid disease progression including higher total cholesterol, 21 triglycerides 19 and LDL, 20 and lower HDL. 19 This study demonstrated similar results, but in multivariate analysis a high Lp(a) and a low HDL-3 were found to be the most important lipids associated with progression. In conclusion, this study provides strong evidence of a major treatment effect on carotid artery stenosis and progression of plaque among participants in SHEP. With treatment of isolated systolic hypertension, progression of stenosis is slowed and it is possible that regression occurs. Similar effects occurring in the coronary and intracranial vessels are likely to be part of the reason for the reduction in stroke and myocardial infarction risk among SHEP participants randomized to active blood pressure treatment. Acknowledgment This study was supported by National Institutes of Health grant HL-39871. References 1. SHEP Cooperative Research Group. Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension: final results of the Systolic Hypertension in the Elderly Program (SHEP). JAMA. 1991;5:35-34. 2. Black HR, Curb JD, Pressel S, Probstfield JL, Stamler J, eds. Systolic Hypertension in the Elderly Program (SHEP): baseline characteristics of the randomized sample. Hypertension. 1991; 17(suppl II):II-1-II-171. 3. Sutton-Tyrrell K, Alcorn H, Wolfson SK. Predictors of carotid stenosis in older adults with and without isolated systolic hypertension. Circulation. 1993;:355-361. 4. Sutton-Tyrrell K, Wolfson SK, Thompson T, Kelsey SF. Measurement variability in duplex scan assessment of carotid atherosclerosis. Stroke. 1992;:215-220. 5. Blackshear W, Phillips DJ, Chikos PM, Harley JD, Thiele BL, Strandness DE. Carotid artery velocity patterns in normal and stenotic vessels. Stroke. 1980;l 1:67-71.

50 Stroke Vol, No 1 January 1994 6. Glover JL, Bendick PJ, Jackson VP, Becker GJ, Dilley RS, Holden RW. Duplex ultrasonography, digital subtraction angiography and conventional angiography in assessing carotid atherosclerosis. Arch Surg. 1984;119:664-669. 7. Blackshear WM, Lamb SL, Kollipara VSK, Anderson JD, Murtagh FR, Shah CP, Farber MS. Correlation of hemodynamically significant internal carotid stenosis with pulsed Doppler frequency analysis. Ann Surg. 1984;199:475-481. 8. Wetzner SM, Kiser LC, Bezreh JS. Duplex ultrasound imaging: vascular applications. Radiology. 1984;150:507-514. 9. Spencer MP, Reid JM. Quantitation of carotid stenosis with continuous wave (C-W) Doppler ultrasound. Stroke. 1979;10:3-330. 10. Alexandrov AV, Bladin CF, Maggisano R, Norris JW. Measuring carotid stenosis: time for a reappraisal. Stroke. 1993;:1292-1296. 11. Keagy BA, Pharr WF, Thomas D, Bowes DE. A quantitative method for the evaluation of spectral analysis patterns in carotid artery stenosis. Ultrasound Med Biol. 1982;8:6-630. 12. Friedwald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972; 18:499-502. 13. Sutton KC, Wolfson SK, Kuller LH. Carotid and lower extremity arterial disease in elderly adults with isolated systolic hypertension. Stroke. 1987;18:817-822. 14. Norris JW, Zhu CZ, Bornstein NM, Chambers BR. Vascular risks of asymptomatic carotid stenosis. Stroke. 1991;22:1485-1490. 15. Salonen R, Haapanen A, Salonen JT. Measurement of intima-media thickness of common carotid arteries with high-resolution B-mode ultrasonography: inter- and intra-observer variability. Ultrasound MedBiol 1991;17:2-0. 16. O'Leary DH, Polak JF, Wolfson SK, Bond MG, Bommer W, Sheth S, Psaty BM, Sharrett AR, Manolio TA. Use of sonography to evaluate carotid atherosclerosis in the elderly: the Cardiovascular Health Study. Stroke. 1991;22:1155-1163. 17. Kuo PT, Toole JF, Schaaf JA, Jones A, Wilson AC, Kostis JB, Moreyra AE. Exlracranial carotid arterial disease in patients with familial hypercholesterolemia and coronary artery disease treated with colestipol and nicotinic acid. Stroke. 1987;18:716-721. 18. Roederer GO, Langlois YE, Jager KA, Primozich JF, Beach KW, Phillips DJ, Strandness DE. The natural history of carotid arterial disease in asymptomatic patients with cervical bruits. Stroke. 1984; 15:605-613. 19. Hennerici M, Rautenberg W, Trockel U, Kladetzky RG. Spontaneous progression and regression of small carotid atheroma. Lancet. 1985;1:1415-1419. 20. Salonen R, Salonen JT. Progression of carotid atherosclerosis and its determinants: a population-based ultrasonography study. Atherosclerosis. 1990;81:33-40. 21. Tanaka H, Nishino M, Ishida M, Fukunaga R, Sueyoshi K. Progression of carotid atherosclerosis in Japanese patients with coronary artery disease. Stroke. 1992;:946-951. 22. Bond MG, Strickland HL, Wilmoth SK, Safrit A, Phillips R, Szostak L for the MIDAS Research Group. Interventional clinical trials using noninvasive ultrasound endpoints: the multicenter isradipine/diuretic atherosclerosis study. J Cardiovasc Pharmacol. 1990;15(suppl l):s30-s33.. MRC Working Party. Medical research council trial of treatment of hypertension in older adults: principal results. BMJ. 1992;304: 405-412.. Beard K, Bulpitt C, Mascie-Taylor H, O'Malley K, Sever P, Webb S. Management of elderly patients with sustained hypertension. BMJ. 1992;304:412-416.. Senin U, Parnetti L, Mercuri M, Lupattelli G, Susta A, Ciuffetti G. Evolutionary trends in carotid atherosclerotic plaques: results of a two-year follow-up study using an ultrasound imaging system. Angiology. 1988;39:429-436.