Background and Purpose In this study, we investigated intima-media thickness and plaque occurrence in the carotid arteries of men with borderline hypertension compared with that in normotensive control subjects and investigated the relations of these variables to atherosclerotic risk factors.
Methods Using B-mode ultrasonography, we compared carotid artery intima-media thickness and plaque occurrence in men with borderline hypertension (diastolic blood pressure of 85 to 94 mm Hg, n=73) with that in age-matched normotensive control subjects (diastolic blood pressure of 80 mm Hg, n=72). We evaluated the relationships of intima-media thickness and plaque occurrence to atherosclerotic risk factors such as age, smoking, lipoprotein levels, and fasting insulin levels.
Results The borderline hypertensive group exhibited a slight increase in overall intima-media thickness (0.73 versus 0.69 mm, P=.07), which was most evident in the right carotid artery (0.72 versus 0.67 mm, P<.05). There were more borderline hypertensive subjects with plaque (26% versus 16%, NS), again more evident on the right side (18% versus 6%, P<.05). Age and high-density lipoprotein cholesterol were consistently related to intima-media thickness (t=1.94 to 3.24 and t=−2.25 to −2.69, respectively, P<.05), whereas age was the only significant determinant for plaque/nonplaque (F=6.4, P<.05). In addition, there was a significant difference in intima-media thickness between the right and left carotids, irrespective of group (F=4.43, P<.05).
Conclusions Our results indicate that vascular structural changes occur even in borderline hypertension, although this seems more related to general atherosclerotic risk factors than to blood pressure alone. Additionally, a possible difference in the development of atherosclerotic lesions of the left and right carotid arteries is suggested, emphasizing the importance of measuring and reporting values from both sides when studying carotid intima-media thickness and plaque occurrence.
The structural adaptation of vessels in hypertension and the early development of atherosclerotic plaques have attracted major interest over the years. The development of noninvasive techniques such as high-resolution ultrasound imaging has opened up the field for in vivo studies in these two interrelated areas. The technique allows measurements of the combined intima and media (I-M) thickness and detection and evaluation of plaque occurrence.1 2 3 4 There is evidence that an increase in I-M thickness is linked to different risk factors for atherosclerosis such as age, ambulatory pulse and systolic blood pressure, a history of coronary heart disease, diabetes, levels of low-density lipoprotein (LDL) cholesterol, and smoking habits,5 and patients with familial hypercholesterolemia have also been shown to have an increased I-M thickness.6 The ultrasound imaging method does not allow for differentiation between the intimal atherosclerotic process and medial hypertrophy due to pressure effects, and it can be debated whether the influence of blood pressure levels on I-M thickness results from the former or the latter. In our previously reported pilot study of newly detected, untreated hypertensive patients with a minimum of other risk factors, I-M thickness was correlated only to age without any difference between hypertensive and normotensive subjects.7 These findings suggest that I-M thickness is more closely related to an intimal atherosclerotic process rather than to a pressure-induced medial hypertrophy.7 Furthermore, in a study by Salonen and Salonen8 it was shown that the increase in I-M thickness in the carotids over a 2-year period was linked to age, LDL cholesterol, and smoking but not to blood pressure levels.
The areas open to investigation with the ultrasound technique are mainly the carotid and the femoral arteries. In earlier topographical studies of atherosclerosis, it has been shown that atherosclerotic plaques start to develop in the carotids at approximately the same time as in the aorta,9 actually preceeding plaque occurrence in the coronary arteries. It has also been shown that carotid atherosclerosis is significantly correlated with the extent of coronary atherosclerosis.10 11
The aim of the present study was to investigate I-M thickness and plaque occurrence in the right and left carotid arteries of patients with borderline hypertension (BHT) compared with age-matched normotensive control subjects and to investigate the relationship between these variables and blood pressure levels and other atherosclerotic risk factors.
In 1985, a blood pressure screening program was started in Äkersberga, a small community north of Stockholm. All men aged 35 to 55 years were asked by mail to visit the primary healthcare center and have their blood pressure measured. Of a sample of 2694 men, 207 had BHT (diastolic blood pressure of 85 to 94 mm Hg on repeated occasions). Of these men, 14 declined further participation, and 193 were followed up for 3 years with yearly blood pressure measurements. At these follow-up visits, approximately 20% of the subjects were found to be hypertensive and 20% normotensive, with the major change (13% to 15%) already occurring by the 1-year follow-up. In addition, 34 subjects declined further partcipation during the follow-up period. The remaining 81 men thus constituted the group that was still within the range for BHT on the basis of repeated measurements over an extended period.
These 81 individuals with BHT were invited to participate in the present investigation together with 80 age-matched male control subjects from the same population who had a diastolic blood pressure ≤80 mm Hg at the initial measurement. To recruit 80 age-matched control subjects, 105 subjects were asked to participate, of whom 23 declined and 2 had a diastolic blood pressure >80 mm Hg. The blood pressure of the control subjects was measured on two occasions a few weeks apart. To participate in the study, a subject’s diastolic blood pressure had to be ≤80 mm Hg on both occasions. All blood pressure measurements during the entire recruitment procedure were performed by the same specially trained nurse.
The study was approved by the local ethics committee of Karolinska Hospital and conducted in accordance with the Helsinki Declaration. All subjects gave their informed consent before entering the study. Of the 81 men with BHT and the 80 normotensive control subjects who agreed to participate in the study, 2 patients in the BHT group discontinued the program before entering the study. One died of malignancy and 1 moved overseas. In addition, 6 of the BHT patients completed blood sampling only. Data from these patients are not included. One normotensive subject turned out to have insulin-dependent diabetes mellitus and could not complete the study; as in the BHT group, 6 normotensive individuals did not complete the carotid examination. Data for these individuals are not included. The total number of patients who completed the entire program, including carotid examinations, was 73 in the BHT group and 72 in the normotensive group. None of the participants had any other signs of cardiovascular disease, and none were treated with antihypertensive or lipid-lowering drugs.
All subjects were investigated according to the same schedule. Men with BHT and the age-matched control subjects were investigated on the same day when possible and not more than 4 weeks apart. Blood samples for lipoprotein analyses and determination of basal insulin levels were taken between 8 and 9:30 am, after 8 to 12 hours of fasting. Carotid ultrasound was performed on a separate occasion within 3 months from the metabolic investigation.
Plasma Lipoprotein and Insulin Determination
The major plasma lipoproteins were determined by a combination of preparative ultracentrifugation and precipitation of apolipoprotein B–containing lipoproteins followed by lipid analyses in the lipoprotein fractions as previously described.12 Basal plasma insulin was measured with standard radioimmunoassay (Pharmacia).
The right and left carotid arteries were examined with a duplex scanner (Acuson 128XP/5) using a 7.0-MHz linear-array transducer. The subjects were investigated in the supine position with the head slightly turned from the sonographer. All measurements were performed by one trained sonographer who was unaware of the subjects’ blood pressure levels. The carotid arteries were carefully examined with regard to wall changes.
Plaque was defined as a localized I-M thickening with a thickness >1 mm and a 100% increase in thickness compared with normal, adjacent wall segments. Plaque occurrence was scored as present or absent. The cutoff point of 1 mm was based on results from a pilot study in newly diagnosed, untreated hypertensive men and control subjects without other cardiovascular risk factors, recruited from the same population screening as in the present study. In the pilot study, none of the participants had I-M thicknesses >1 mm.7 Plaque was screened for in the common, internal, and external carotid arteries.
The far wall of the common carotid artery (CCA), 0.5 to 1.0 cm proximal to the carotid bulb, was used for measurements of I-M thickness and lumen diameter on both sides. The I-M thickness was defined as the distance between the leading edge of the lumen-intima echo and the leading edge of the media-adventitia echo. The lumen diameter was defined as the distance between the leading edge of the intima-lumen echo of the near wall and the leading edge of the lumen-intima echo of the far wall. The examinations were videotaped for subsequent analysis by a computer system (Macintosh IIvx, QuickImage 24-videoframe grabber card [MASS Microsystems Inc] and Panasonic NV-FS90EB VCR). image software (National Institutes of Health, Research Services Branch, National Institute of Mental Health) was used to trace and measure the distances between the wall echos within a 10-mm-long section of the CCA in late diastole, defined by simultaneous electrocardiographic recording. The mean values of the I-M thickness and lumen diameter over the 10 mm were calculated using an application developed with 4th Dimension (ACI). The differences between repeated measurements of I-M thickness and lumen diameter (healthy subjects, 1 week apart) were on average 9% and 2% (intraobserver), respectively (range, 0.45 to 0.89 mm for I-M thickness and 4.63 to 7.14 mm for lumen diameter).
I-M thickness can be affected not only by a change in tissue mass in the inner layer but also by a simultaneous widening of the vessel. Thus, increased diastolic blood pressure or a compensatory widening in the presence of atherosclerotic wall changes13 could decrease the I-M thickness due to stretching of the vessel wall. To overcome this potential source of error, an estimation of circumferential I-M tissue mass was made by calculation of the I-M area (cI-Ma) in a plaque-free cross-section of the CCA using the following formula: ([lumen diameter+2∗I-M thickness]/2)2∗3.14−(lumen diameter/2)2∗3.14.
Blood Pressure Measurements
All blood pressure measurements were performed with a mercury sphygmomanometer in a standardized fashion. All measurements during the entire recruitment period were performed by the same specially trained nurse. Cuff size was adjusted to the circumference of the arm, and the arm was placed with the cuff at the level of the heart. Blood pressure was recorded as the mean of two measurements taken after 5 minutes of rest in the supine position. Systolic and diastolic blood pressures were defined according to Korotkoff sounds I and V.
All patients were weighed without clothing other than underwear, using the same scale (Delta 707, SECA). Height was measured with a special ruler fixed to the wall. The waist was measured at the level of the umbilicus, hips were measured at the greatest circumference, and the waist-to-hip ratio was calculated. Body mass index (BMI) was subsequently calculated as weight in kilograms/(height in meters)2, and body surface area (BSA) was calculated by the simplified formula suggested by Mosteller.14
Variables were tested for skewness. For skewed variables, nonparametric tests were used for comparisons between the groups (Mann-Whitney U test), whereas Student’s t test was used for normally distributed variables and two sample sign tests for comparison of plaque occurrence. Repeated-measures ANOVA was performed to determine interaction terms and group and side differences. Spearman’s rank correlation coefficients were calculated to estimate interrelations between plasma lipoproteins, insulin, clinical characteristics, I-M thickness, and groups (BHT or normotensive). ANCOVA was performed to control for the confounding effects of differences in BMI and insulin between the groups. Discriminant analysis was performed to determine which variables provided the best discrimination between BHT men and control subjects, and multiple regression analysis was performed using the regression variable selection technique. Correlation, regression, and discriminant analyses were performed in the two groups separately. Skewed variables were log-normalized before they were subjected to covariance, discriminant, and regression analyses. Values in the text are given as mean±SD.
Characteristics of BHT Patients and Control Subjects
Basic characteristics of the two study groups are presented in Table 1⇓. The two groups were well matched for age, and blood pressures were well differentiated. The BHT group had a significantly higher BMI with a somewhat more pronounced central obesity as evidenced by a slightly but significantly higher waist-to-hip ratio. Both groups were, however, below the recommended waist-to-hip ratio of 1.0. There was no difference in BSA between the groups.
Men with BHT had significantly higher basal insulin levels. Plasma cholesterol and LDL cholesterol were similar in the two groups. Plasma triglycerides were elevated in the BHT group, but this did not attain statistical significance (P=.058). The high-density lipoprotein (HDL) cholesterol levels were significantly lower in the BHT group (Table 1⇑). When the group differences in BMI and insulin were taken into account in covariance analyses, the plasma lipoprotein differences between the two groups disappeared (F<1.7, P>.20).
In 4 of the BHT and 4 of the normotensive individuals, ultrasonic I-M values could not be obtained because of a malfunctioning tape recorder. The presence of plaque could, however, be recorded for these patients.
In a repeated-measures ANOVA taking both hypertensive status (ie, group) and right or left carotid measurements (ie, side) into consideration, the BHT group had an increase in I-M thickness compared with the normotensive group (computed overall value, 0.73±0.02 versus 0.69±0.02 mm, respectively; P=.07). Analyzing the two sides separately, there was a significantly increased I-M thickness in the BHT group when the right CCA was measured (P<.05). The difference between the two groups did not reach statistical significance on the left side (Table 2⇓, Fig 1⇓). The results for lumen diameter were similar with an overall trend toward a difference between the BHT and normotensive groups (computed overall value, 6.22±0.07 versus 6.05±0.05 mm, respectively; P=.08), and there was also a significant overall increase in cI-Ma (computed value, 16.0±0.4 versus 14.7±0.4 mm2; P<.05) in the BHT group (Table 2⇓).
In the BHT group, 26% of the subjects had plaque on one or both sides. The corresponding percentage for the normotensive group was 16% (19 versus 11 subjects, NS). Analyzing the two sides separately, there were significantly more subjects in the BHT group when the right CCA was measured (18% versus 6%, P<.05), but there was no difference between the groups as to plaque occurrence on the left side (15% versus 13%, NS) (Fig 2⇓).
When comparing subjects with plaque on one or both sides with subjects without plaques in an ANOVA, the I-M thickness in unafflicted sections was significantly higher overall and on both sides in the group with plaques (0.76±0.15 versus 0.68±0.11 mm on the right side and 0.79±0.16 versus 0.70±0.12 mm on the left side, overall P<.001). In an analysis taking group and plaque occurrence into consideration, subjects with plaque had an increased I-M thickness irrespective of blood pressure status (F=13.63, P<.001).
Right Versus Left Common Carotid Artery
The intima-media was somewhat thicker on the left than on the right side in both groups. In an ANOVA, taking group and side into consideration, there was a significant difference in I-M thickness between the two sides irrespective of group (F=4.43, P<.05). There was a tendency toward a difference between the two groups regarding side differences, but this was not statistically significant (interaction F<2.0; P=.16). There were similar numerical differences between left and right sides in cI-Ma (F<4, interaction F<2, P=.17), whereas the opposite was true for lumen diameter, ie, significantly smaller lumen diameter on the left side (F=10.73, P<.01, interaction F<1, P=.80) (Table 2⇑).
Interrelationships Between I-M Thickness, Plaque Occurrence, Blood Pressure Levels, Anthropometrics, and Metabolic Variables
Correlation analyses were performed in the two groups separately. The single factor significantly correlating with I-M thickness was age (r=.26 to .29, P<.05).
In a regression analysis, age was the most significant factor (t=1.94 to 3.24, P<.05) followed by HDL cholesterol level (t=−2.25 to −2.69, P<.05). However, the above factors predicted the I-M thickness to a small degree only (6% to 15%). The use of BSA or BMI as a forced variable to account for possible differences in stature did not change the results of the regression analysis.
A discriminant analysis was performed taking age, lipoprotein levels, insulin levels, and smoking habits into consideration. The analysis showed the basal insulin and the cI-Ma of the right CCA to be the only significantly discriminating factors between the two groups (F=19.7, P<.05 and F=5.35, P<.05, respectively). Together they could correctly classify the individual in 64% of the cases, implying that other factors not taken into account in this study could be of importance.
Another discriminant analysis to determine the discriminating power of the same factors, with the addition of blood pressure levels, for the occurrence of plaques revealed age as the only significant determinant between the groups with and without plaque (F=6.4, P<.05), with total triglyceride levels coming close to significance (F=3.2, P=.06). The predictive value of age, however, was small (Wilks’ lambda=0.91).
The present study, based on a screened population of men with BHT and normotensive male control subjects, shows that there is a slight increase in I-M thickness and plaque occurrence in borderline hypertensive men compared with normotensive control subjects. Although the differences did not always reach statistical significance, values differed consistently in the same direction. This is well in line with findings in hypertensive patients.15 16
In a previous population-based study in Eastern Finnish men, age, smoking, and levels of LDL cholesterol and systolic blood pressure were among the factors determining the I-M thickness (also including ambulatory pulse pressure, history of coronary heart disease, and diabetes).5 The strongest predictor for the presence of carotid plaques has consistently been age.8 17 18 Our results partly confirm and expand such data.
In the present study, I-M thickness was related to age and HDL cholesterol levels but not consistently to blood pressure levels. This suggests that, in the early phases of hypertension, associated atherosclerotic risk factors already exert effects on vessel walls that are not directly related to blood pressure levels. Previous findings in established hypertension support such a notion.16 19 One should, however, bear in mind that the blood pressure span of the investigated groups in the present study was, by definition, relatively narrow. Age was the single best determinant of presence of plaque. The fact that other risk factors were not associated to atherosclerotic plaque occurrence could be due to the relative “healthiness” of the groups studied.
When comparing the BHT and normotensive groups, an interesting phenomenon occurred. The differences in both I-M thickness and in plaque occurrence were more evident in the right CCA. In a repeated-measures ANOVA, there were significant differences between the right and left CCAs, irrespective of blood pressure status. There could be several explanations for this observation, one being that the observed differences are a result of chance alone. Another possible explanation could be that the observed difference between the left and right sides is real and reflects a time-related difference in the natural course of atherosclerosis in the carotid arteries. In most published morphological studies, the two carotids (left and right) are seldom accounted for separately, or there is a statement that the two sides were “similar.”20 However, in an ultrasonographic study in which data are actually given for both sides,21 mean wall thickness in the left CCA of white men in the same age group is numerically larger as in the present study, and in the study of Solberg and Eggen,20 areas with raised atherosclerotic lesions were consistently larger in the left CCA than in the right CCA. Furthermore, the recently published Rotterdam study, albeit investigating an older population than the present study, shows similar numerical differences between the two sides; the difference between subjects with and without systolic hypertension was significant only in the right carotid artery.15 This raises the possibility that atherosclerotic lesions develop earlier on the left side, perhaps because of the different gross anatomy of the left and right CCA, with the left CCA originating directly from the aortic arch. This could possibly result in the two sides exhibiting different shear stress conditions, a factor that has been shown to influence the localization and rate of development of atherosclerotic lesions.22 The observed difference between left and right CCA could thus conceivably result from a faster development of atherosclerosis on the left side. In the presence of atherosclerotic risk factors such as elevated blood pressure, dyslipidemia, and fasting hyperinsulinemia, as in the BHT group in the present study, the atherosclerotic process accelerates generally, and the difference between the left and right carotid arteries is no longer as clearly evident. It is, of course, impossible to draw any definite conclusions from a study of the present cross-sectional design, but our findings warrant further investigations in this area. Our results also underline the importance of measuring and presenting data on both the left and right CCAs.
In the present study, we introduce a method of calculating the circumferential I-M area, taking into account changes in vessel diameter when evaluating changes in the inner layer of the artery. Parallel to the development of atherosclerosis, a widening of the artery is seen that compensates for the restriction of the vessel lumen.13 Such a widening of the artery at an early stage of the atherosclerotic process may lead to an underestimation of the increase in “I-M mass” due to lipid deposit and fibrosis when studied only by measurement of the I-M thickness. In discriminant analysis, cI-Ma was a more potent discriminator between the different groups, indicating the potential usefulness of this type of measurement. It should, however, be stressed that cI-Ma is a calculated value based on the I-M thickness and lumen dimension in one plane only.
Finally, our results show that there is an increased I-M thickness in unafflicted segments of the carotid arteries of patients with atherosclerotic plaques and that this increase is present irrespective of blood pressure status. This further supports previous suggestions7 8 16 19 that the I-M thickness of the CCA is linked to atherosclerosis, rather than to hypertension per se, and it is also in line with the findings of Craven et al23 showing carotid B-mode ultrasound score to be significantly related to coronary atherosclerosis and the most important classifier for presence of coronary atherosclerosis in men aged >50 years. Ultrasonographic I-M thickness measurements of the carotid arteries could thus be a valuable tool for investigating and evaluating factors related to atherosclerotic development, but they would be less valuable for evaluation of the hypertrophic structural vessel-wall changes of hypertension.
In conclusion, our results show that there is a slight but consistent increase in I-M thickness and plaque occurrence in men with BHT compared with normotensive control subjects. This structural change, present in the early stage of hypertension, could be an indicator of an accelerated atherosclerotic process and could possibly be related to the increased mortality from coronary heart disease observed with BHT.24 The difference in I-M thickness and plaque occurrence between the two groups is explained only to a limited extent by the discretely elevated blood pressure levels, whereas a concomitant increase of other risk factors for atherosclerosis, such as dyslipidemia in the BHT group, seems to be of greater importance.
Another salient finding of this study is the previously unreported difference in I-M thickness and plaque occurrence between the left and right sides of the CCAs. Our results point to the possibility of different rates of early development of atherosclerosis in the left and right carotid arteries, possibly related to the gross anatomic differences between the two sides. These findings warrant further studies and emphasize the importance of evaluating and reporting data from both the left and right sides when performing ultrasonographic measurements of the carotid I-M thickness and plaque occurrence.
This project was funded by grants from the Swedish Heart-Lung Foundation, King Gustaf V 80th Birthday Fund, the Trygg-Hansa Research Fund, the Bank of Sweden Tercentenary Fund, and the Karolinska Institute. We are grateful to Ulla Hellmark Augustsson for her help with blood pressure measurements over the years and for her help with blood sampling, to Margaretha Ekberg for her skill as a sonographer, to Anders Hamsten for help with the lipoprotein fractionation, and to Suad Efendic for help with the insulin analysis.
- Received May 26, 1994.
- Revision received September 27, 1994.
- Accepted September 27, 1994.
- Copyright © 1995 by American Heart Association
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. 1986;74:1399-1406.
Persson J, Stavenow L, Wikstrand K, Israelsson B, Formgren J, Berglund G. Noninvasive quantification of atherosclerotic lesions: reproducibility of ultrasonographic measurement of arterial wall thickness and plaque size. Arterioscler Thromb. 1992;12:261-266.
Dawson DL, Strandness DE. Arterial wall thickness with high-resolution B-mode imaging. In: Bernstein EF, ed. Vascular Diagnosis. 4th ed. St Louis, Mo: CV Mosby Co; 1993:420-431.
Wendelhag I, Wiklund O, Wikstrand J. Arterial wall thickness in familial hypercholesterolemia: ultrasound measurement of intima-media thickness in the common carotid artery. Arterioscler Thromb. 1992;12:70-77.
Holme I, Enger SC, Helgeland A, Hjerman I, Loren P, Lund-Larsen PG, Solberg LA, Strong JP. Risk factors and raised atherosclerotic lesions in coronary and cerebral arteries. Arteriosclerosis. 1981;1:250-256.
Carlsson K. Lipoprotein fractionation. J Clin Pathol. 1973;5(suppl 26):32-37.
Bots ML, Hofman A, de Bruyn AM, de Jong PT, Grobbee DE. Isolated systolic hypertension and vessel wall thickness of the carotid artery: the Rotterdam study. Arterioscler Thromb. 1993;13:64-69.
Giral P, Filitti V, Levenson J, Pithois-Merli I, Plainfosse M-C, Mainardi C, Gold A, Simon A, PCV Metra Group. Relation of risk factors for cardiovascular disease to early atherosclerosis detected by ultrasonography in middle-aged normotensive hypercholesterolemic men. Atherosclerosis. 1990;85:151-159.
Salonen R, Seppänen K, Rauramaa R, Salonen JT. Prevalence of carotid atherosclerosis and serum cholesterol levels in Eastern Finland. Arteriosclerosis. 1988;8:788-794.
Lusiani L, Visonà A, Pagnan A. Noninvasive study of arterial hypertension and carotid atherosclerosis. Stroke. 1990;21:410-414.
Solberg LA, Eggen DA. Localization and sequence of development of atherosclerotic lesions in the carotid and vertebral arteries. Circulation. 1971;43:711-724.
Howard G, Sharrett AR, Heiss G, Evans GW, Chambless LE, Riley WA, Burke GL, for the ARIC investigators. Carotid arterial intimal-medial thickness distribution in general populations as evaluated by B-mode ultrasound. Stroke. 1993;24:1297-1304.
Zarin CK, Giddens DP, Bharadvaj BK, Sottiurai VS, Mabon RF, Glagov S. Carotid bifurcation atherosclerosis: quantitative correlation of plaque localization with flow velocity profiles and wall shear stress. Circ Res. 1983;53:502-514.
Craven TE, Ryu J, Espeland M, Kahl FR, McKinney WM, Toole JF, McMahan MR, Thompson CJ, Heiss G, Crouse JR III. Evaluation of the associations between carotid artery atherosclerosis and coronary artery stenosis. Circulation. 1990;82:1230-1242.