Original Article | Published:

Serum cholesterol affects blood pressure regulation


A close relationship between abnormalities of the lipid metabolism and arterial hypertension has been observed in several epidemiological studies. The aim of the present study was to investigate whether serum cholesterol might affect blood pressure (BP) levels at rest, during ambulatory monitoring or during sympathetic stimulation—independently of other variables such as body weight or serum insulin—thus influencing the outcome of hypertensive complications. Seventy-three patients with sustained newly-discovered and never-treated hypertension were divided into tertiles according to their serum cholesterol levels and their resting BP, 24-h BP and BP during isometric exercise (handgrip) were compared. Cardiac mass and carotid wall thickness were measured by echographic technique. The results were that tertiles were similar for body weight, blood glucose and serum insulin, but different for serum cholesterol and triglycerides. BP at rest and during 24-h monitoring was similar in the three groups, whilst a significant difference was detected during sympathetic stimulation by handgrip, with systolic and diastolic BP increasing by 16/12, 28/19 and 30/23 mm Hg (P < 0.01) in lower, medium and higher tertiles, respectively. Intima-media layer of the carotid arteries was also significantly thickened in the groups with higher cholesterol levels (0.54 ± 0.07, 0.67 ± 0.14, 0.68 ± 0.15, P < 0.05). These data support the conclusion that even in patients with recently discovered hypertension, cholesterol levels may influence the BP response to adrenergic stimulation as well as the outcome of target organ disease.


Epidemiological investigations have found that hypertensive patients frequently have a concomitant increase in serum cholesterol levels.1,2 Whether this is only a pure statistical association or it also implies a pathophysiological link is still under discussion. Endothelium, in fact, plays a role in the regulation of systemic blood pressure (BP) and local vascular tone, as suggested by some authors in the last few years.3 The same factors able to affect the endothelium might, in turn, influence BP levels. Thus, lipoproteins, which strongly contribute to atherogenesis, might play a relevant role in the pathophysiology of arterial hypertension:4 this hypothesis is further supported by the observation that hypertension and atherosclerosis share structural and functional changes. High cholesterol has been shown to impair endothelium-dependent dilation,5 which seems to be restored by oral administration of L-arginine in hypercholesterolaemic patients.6

In order to contribute to this interesting topic, the present investigation aimed at evaluating possible differences in BP levels at rest, during 24-h monitoring, and during sympathetic stimulation induced by handgrip isometric exercise, in hypertensive patients according to their fasting serum cholesterol levels.

Patients and methods

Seventy-three patients with recently discovered and never-treated arterial hypertension (high BP levels recently detected with a previous finding of normal values within 1 year) were invited to enter the study.

Exclusion criteria were: (a) white-coat hypertension; (b) biochemical or instrumental evidence of secondary hypertension; (c) total cholesterol >7.76 mmol/L and/or triglycerides >3.39 mmol/L (without nutritional or pharmacological treatment); (d) fasting blood glucose >6.94 mmol/L (without nutritional or pharmacological treatment); (e) evidence of cardiovascular disease (myocardial infarction, angina, stroke or transient ischaemic attacks in the past 6 months or evidence of cardiac valvular disease); (f) evidence of liver cirrhosis or renal failure; and (g) pregnancy or lactation.

After a 1-month run-in during which their BP was measured at rest three times at 2-week intervals, patients were asked to give their informed consent to participate. Thereafter they underwent the following measurements:

  1. 1

    Height and weight, using a platform beam-scale (Seca 760); body mass index (BMI), calculated as weight-to-squared height ratio (w/h2)

  2. 2

    Resting BP, using an automatic sphygmomanometer (Sentron, Bard Biomedical, Lombard, IL, USA).7 Two BP readings were recorded with patients in the sitting position for at least 5 min, between 9.00 and 11.00 in the morning; the average of the two readings was considered as the measurement for that visit. Patients were also asked to complete a questionnaire regarding smoking habits

  3. 3

    Twenty-four hour ambulatory BP monitoring (ABPM), by Spacelabs Inc 90207 (Redmond, WA, USA), during a normal working day. The monitor was fitted on in the morning, between 10.00 and 12.00 am. Subjects were asked to perform their normal daily activities and to consume their habitual diet. The monitor was programmed to record BP and heart rate (HR) at 15-min intervals during the subjects’ waking hours and at 20-min intervals during their usual sleeping hours. Subjects were instructed to keep their arms still during recordings. For the analysis of the data, the 24-h measurements were divided into waking (7.00 am to 10.45 pm) and sleeping (11.00 pm to 6.40 am) periods. The 24-h record was acceptable if more than 80% of the scheduled readings were available. Patients also filled in a 24-h diary to report daily activities and rest, including the sleeping period

  4. 4

    BP was also measured during isometric exercise. Patients refrained from smoking and drinking beverages containing caffeine for at least 12 h before the test. The handgrip was performed, after a 15-min bed rest, by squeezing a dynamometer for 3 min at 30% of previously evaluated maximal stress. BP and HR were measured twice before the test, three times at 1-min intervals during the test and twice, after 2 and 5 min, in the recovery phase. The average of BP and HR measurements at baseline, during the stress, and in the recovery phase were calculated for the statistical analysis of the tests.8

  5. 5

    Laboratory assessments of haematology (haematocrit = Ht) and biochemistry (fasting blood glucose (FBG), serum insulin, total proteins, total serum cholesterol (Chol), triglycerides (Tg), and cholesterol content in high-density lipoproteins (HDL chol)). Blood samples were drawn at 8.30 am, after an overnight fast. Cholesterol content in the low-density lipoproteins (LDL chol) was calculated by the Friedewald formula,9 as follows: LDL chol = total chol − (HDL chol + 1/5 Tg)

  6. 6

    Two-dimensionally-targeted M-mode echocardiograms were performed by an expert sonographer, using a commercially available echocardiograph (AU3 Partner ESAOTE Biomedica, Florence, Italy) connected to a 2.5 to 3.5 MHz annular-array transducer and recorded on videotapes. Strip-chart tracings of the patients were obtained at 50 mm/sec velocity; these were examined by two experienced investigators blind to the knowledge of the status of patients, using a graphic tablet and a pointer device interfaced with a PC and a customised data-acquisition software. In our laboratory, the interobserver variability for measurements of wall thickness, internal dimensions and left ventricular mass was very small with an interclass correlation coefficient of 0.90. End-diastolic (EDD) and end-systolic (ESD) left ventricular internal diameters, posterior wall (PWT) and septal thicknesses (SWT) were measured according to the recommendations of the American Society of Echocardiography.10 A second set of measurements was also taken according to the Penn convention criteria to calculate left ventricular mass (LVM).11 To take into account body size, LVM was normalised for body height to the power of 2.7 (LVMi = g/m2.7), an indexation that has been shown to also detect hypertrophy in obese individuals.12 Relative wall thickness (RWT), an index of LV geometric pattern, was measured at end-diastole, as 2 * PWT/EDD. The aortic root (AR) was measured according to the recommendations of the American Society of Echocardiography. Left atrium (LA) size was measured from the trailing edge to the leading edge, by excluding the posterior wall of the aortic root. LV end diastolic (EDV) and systolic (ESV) volumes were calculated using the Teichholz correction of the cube formula.13 LV chamber and stroke volumes (SV) determined using this approach have been shown to correlate well with invasive and Doppler echocardiographic volume measurements in a variety of populations with symmetric LV wall motion.14 Cardiac output (CO) and ejection fraction (EF) were also calculated

  7. 7

    Carotid ultrasound imaging was performed with a 2000 II SA Biosound ultrasound system, (Bio Dynamics, Indianapolis, IN, USA) according to the methodology already described.15,16 Ultrasonographic images were considered acceptable only if they met the criteria for a high quality examination, that is: (a) presence of the adventitia-media interface and intima-lumen interface in at least two arterial segments; (b) visualisation of anterior, posterior, lateral and medial wall of the common carotid artery, the bifurcation and at least 2 cm of the internal carotid artery. The whole scanning procedure was recorded on a videotape; a hard copy of the distal portion of the common carotid artery, just below the bulb, was thereafter produced. Pictures were scanned (Epson G 2000) and displayed on a computer screen (Macintosh II) and then analysed by a software (Image 1.31) that allows quantitative evaluation of the terminal centimer of the vessel intima-media area. Measurement of mean intima-media thickness (IMT) of the common carotid arteries was derived from the area-to-1 cm length ratio. Lumen diameter was also measured at the same level. Thickness of the common carotid artery was measured only at the level of the far wall (the carotid wall farthest from the probe) as this segment is more constantly visualised with B-mode imaging than the more superficial near wall.15,16 This technique allows a high reproducibility of the measurements, as already shown in a previous study.17


Patients arrived at the Outpatient Hypertension Clinic, after an overnight fast, at 8.00 am. In the first day they underwent the clinical procedures according to the following schedule: (a) measurement of BP at rest; (b) venous sampling for biochemical analysis; (c) two-dimensionally targeted M-mode echocardiographic assessment; (d) carotid ultrasound imaging; (e) measurement of BP during a 3-min isometric exercise; and (f) start of 24-h BP monitoring.

The day after patients returned to the outpatient clinic to undress the ABP monitor. Protocol of the study was approved by the local Institutional Ethical Committee.

Statistical analysis

Data were stored and analysed by SPSS statistical package. Data are expressed as mean ± standard deviation (M ± s.d.) in the text and in the tables, and as M ± SE in the figures. Normally distributed variables were compared by one-way analysis of variance with Tukey's multiple comparisons, and categorical ones by χ2 test. A P value of 0.05 was considered significant.


Seventy-three patients with sustained arterial hypertension, age range 17–64 years, entered the study. The patients were divided, retrospectively, into tertiles according to their serum cholesterol levels. The tertiles comprised (a) patients with cholesterol below 5.172 mmol/L, (b) those with cholesterol between 5.172 and 5.922 mmol/L, and (c) patients with serum cholesterol above 5.922 mmol/L.

As shown in Table 1 these groups were similar for age, sex, BMI and smoking habit; also BP levels, both at rest and during 24-h monitoring, were similar in the tertiles.

Table 1 Age, sex, body mass index, smoking habits, blood pressure (BP), heart rate and metabolic parameters in 73 patients with hypertension recently discovered and never treated, divided according to tertiles of serum cholesterol levels

BP response to sympathetic stimulation by a 3 min handgrip showed the systolic and diastolic BP increase during the test to be significantly more pronounced in hypertensive patients with high cholesterol levels than in the first subgroup (systolic BP (SBP) increase 16 ± 7, 28 ± 12, 30 ± 13 mm Hg, P < 0.01; diastolic BP (DBP) increase 12 ± 4, 19 ± 10, 23 ± 10 mm Hg, P < 0.01, mean BP (MBP) increase 14 ± 4, 22 ± 10, 26 ± 10 mm Hg, P < 0.01) (Figure 1), without difference in heart rate (HR) (7.8 ± 6, 9.4 ± 8, 9.3 ± 6 beats per min).

Figure 1

Systolic, diastolic and mean blood pressure increase during isometric exercise in 73 hypertensive patients divided according to tertiles of serum cholesterol levels (M ± s.e., significance 1st vs other tertiles *P < 0.01).

With regard to biochemistry, all lipid parameters but not HDL-cholesterol progressively increased from the first to the last subgroup whilst no difference was detected in glucose metabolism (Table 1).

Echocardiographic parameters of cardiac structure and systolic function were similar in the three subgroups (Table 2), whereas a significant thickening of the carotid wall was detected in patients with higher cholesterol levels. Figure 2 shows that mean (0.54 ± 0.07, 0.67 ± 0.14, 0.68 ± 0.15 mm, P < 0.05) and peak (0.64 ± 0.12, 0.75 ± 0.13, 0.76 ± 0.16 mm, P < 0.05) thickness of the intima-media layer are significantly increased in the second and third tertiles without differences in the carotid diameter (6.25 ± 1.12, 6.25 ± 0.70, 6.57 ± 0.90 mm).

Table 2 Parameters of cardiac structure and function in 73 patients with hypertension recently discovered and never treated, divided according to tertiles of serum cholesterol levels
Figure 2

Mean and peak thickness of the intima-media layer in 73 hypertensive patients divided according to tertiles of serum cholesterol levels (M ± s.e., significance 1st vs other tertiles *P < 0.05).


Large prospective studies18,19 have pointed out that the individual risk of developing coronary artery disease frequently depends on the synergy of two or more risk factors, each often of only mild degree. This implies that the calculated global risk is a more powerful predictor of the cardiovascular outcome than the individual risks from which it is derived.20,21 Among the combinations of different risk factors, the close relationship between arterial hypertension and abnormalities of lipid metabolism has been shown in epidemiological and genetic studies.1,2,22 Both serum cholesterol and triglyceride seem to be related to BP values and a possible link between these conditions has been sought among different factors such as altered peripheral insulin sensitivity23 or genetically-determined abnormalities of the trans-membrane transport system.24 The plurimetabolic syndrome is the peculiar and, probably, most severe expression of an association that cannot be considered as merely statistical.25

The present study was undertaken in order to investigate whether cholesterol levels may affect BP regulation independently of other well known determinants such as age and body weight. For this reason we have measured BP not only at rest but also during 24-h monitoring and in response to an adrenergic stimulating test, the handgrip isometric exercise. This is an easily performed marker of cardiovascular reactivity,26 which is well correlated with other tests of adrenergic stimulation27 and less closely related to 24-h BP monitoring.8,27 Other investigations, such as those evaluating microneurographic neural activity, do indeed provide better indices of sympathetic function, but the advantage of the handgrip isometric exercise lies in its feasibility28 and intraobserver reproducibility.29 The main results of this clinical investigation indicate that BP levels at rest are not affected by serum cholesterol, contrarily to those in response to sympathetic stimulation, which more markedly increase in patients with higher cholesterol levels, on the opposite of HR response to handgrip, which was not different among tertiles. Accordingly hypertensive patients with higher serum cholesterol also have a thickening of the intima-media layer, which is a marker of a generalised arteriosclerotic disease.30 The mass of the left ventricle, on the other hand, does not seem to be affected by different cholesterol levels.

The hypothesis that lipoproteins might be relevant to hypertension is not new and is pathophysiologically sound. In fact, they have direct stimulatory effects on both contraction-coupled and replication-coupled processes in the vascular smooth muscle cells. These effects have been attributed to the stimulation of the phosphoinositide signal transduction in smooth muscle cells.4 Moreover there is evidence that cholesterol induces endothelial dysfunction in experimental31 and clinical studies, even at normal32 or high-to-normal3 ranges by reducing the bioavailability of endothelium-derived nitric oxide. Endothelium-dependent vasodilation, in fact, correlates inversely with total cholesterol levels.3 Accordingly, lipid-lowering (ie statins) and antihypertensive (ie, angiotensin- converting enzyme inhibitors) drugs improve or normalise endothelial dysfunction33 (ie endothelium-dependent vasodilation). These observations support the hypothesis that endothelium dysfunction may occur in relatively young individuals with mild hypertension even at cholesterol levels considered normal, as in the participants of the present study. This is, probably, the reason of the lack of difference in BP response to handgrip between the second and third tertiles, since cholesterol levels in the range 5.2–5.9 mmol/L appear already able to induce altered endothelial function.

Finally it is possible that higher cholesterol levels may be associated with more atherosclerotic vessels, not only in the carotid circulation, as shown in the present paper by the increase of intima-media thickness, but also in other arteries. Stiffer arteries, therefore, might be less able to counteract the increase in BP due to isometric exercise. Whatever the underlying mechanism, it is conceivable that the effects of cholesterol levels on BP appear more evident in stressing situations, such as that induced by the handgrip-stimulated overactivity of the sympathetic nervous system. This finding is in agreement with a previous observation of the reduction in SBP response to an arithmetic test in patients with high cholesterol levels after lovastatin therapy.34

With this purpose, the dissociation between BP and HR response to handgrip in the tertiles of serum cholesterol further support the hypothesis that abnormal increase may be related to altered endothelium-dependent vasodilation. This discrepancy has already been seen in the above-mentioned study with a different adrenergic system stimulating test—the arithmetic one—where lovastatin treatment was able to reduce SBP response, which depends on endothelial damage, but not HR response, which appears independent of it.34

The other main finding of this study—the increase in the IMT of the carotid artery—appears a natural consequence of the increase in the global risk of arteriosclerosis, determined by the concomitant presence of mild hypertension and mild hypercholesterolaemia.30 To this purpose, it is useful to remember that BP and serum cholesterol levels are positively correlated with IMT in several studies.17,35,36 This finding is also in agreement with the results of another study37 which demonstrated in Chinese patients the role of hypertension, hypercholesterolaemia, and hypertrygliceridaemia in the pathogenesis of carotid atherosclerosis.

The lack of difference in LVM, which is mainly related to BP levels, depends on the similarities of BP at rest and during 24-h measurement in the two subgroups.

It is not unexpected that intima-media thickening does not parallel left ventricular (LV) hypertrophy in hypertensives, as already observed by ourselves in uncomplicated hypertensive patients.17 More recently38 other authors stated that the relationship of BP variability with LV hypertrophy or carotid damage is controversial and this finding may further support the lack of correlation between ventricular hypertrophy and carotid damage.

In conclusion the results of this study strongly support the hypothesis that even mildly increased serum cholesterol levels are able to influence BP, at least during sympathetic stimulation. The relevance of this combination to the increase in global risk and, therefore, to the early evidence of arterio- sclerosis has also been pointed out by the present investigation.


  1. 1

    Laurenzi M et al on behalf of the Gubbio Study Group. Multiple risk factors in hypertension: results from the Gubbio Study J Hypertens 1990; 8 (Suppl 1): S7–S12

  2. 2

    National Heart Foundation of Australia: risk factor prevalence study, report 1 National Heart Foundation of Australia: Canberra 1982

  3. 3

    Vallance P, Collier J, Moncada S . Effects of endothelium-derived nitric oxide on peripheral arteriolar tone in man Lancet 1989; 2: 997–1000

  4. 4

    Resink TJ et al. Interaction between plasma lipoproteins and vascular smooth muscle cells: how relevant is it to arterial hypertension? Nutr Metab Cardiovasc Dis 1994; 4: 163–170

  5. 5

    Creager MA et al. Impaired vasodilation of forearm resistance vessels in hypercholesterolemic humans J Clin Invest 1990; 86: 228–234

  6. 6

    Clarkson P et al. Oral L-arginine improves endothelium-dependent dilation in hypercholesterolemic young adults J Clin Invest 1996; 97: 1989–1994

  7. 7

    Johnson CJ, Kerr JH . Automatic blood pressure monitors. A clinical evaluation of five models in adults Anesthesia 1985; 40: 471–478

  8. 8

    Floras JS, Osman Hassan M, Jones JV, Sleight P . Pressor responses to laboratory stresses and daytime blood pressure variability J Hypertens 1987; 5: 715–719

  9. 9

    Friedewald WT, Lery RI, Fredrickson DS . Estimation of plasma low density lipoprotein cholesterol concentration without use of the preparative ultracentrifuge Clin Chem 1972; 18: 499–502

  10. 10

    Sahn DJ, De Maria A, Kisslo J, Weyman A . The Committee on M-Mode Standardization of the American Society of Echocardiography: recommendations regarding quantitation in M-Mode echocardiography. Results of a survey of echocardiographic measurements Circulation 1978; 58: 1072–1083

  11. 11

    Devereux RB, Reichek N . Echocardiographic determination of left ventricular mass in man. Anatomic validation of the method Circulation 1977; 55: 613–618

  12. 12

    de Simone G et al. Left ventricular mass and body size in normotensive children and adults: assessment of allometric relations and impact of overweight J Am Coll Cardiol 1992; 20: 1251–1260

  13. 13

    Teichholz LE, Kreulen T, Herman MV, Gorlin R . Problems in echocardiographic volume determinations: echocardiographic-angiographic correlations in the presence or absence of asinergy Am J Cardiol 1976; 37: 7–11

  14. 14

    Wallerson DC, Ganau A, Roman MJ, Devereux RB . Measurements of cardiac output by M-mode and two dimensional echocardiography: application topatients with hypertension Eur Heart J 1990; 11 (Suppl 1): 67–78

  15. 15

    Bond MG, Wilmoth SK, Enevold GL, Strickland HL . Detection and monitoring of asymptomatic atherosclerosis in clinical trials Am J Med 1989; 86 (Suppl 4A): 33–36

  16. 16

    Pignoli P et al. Intimal plus medial thickness of the arterial wall: a direct measurement with ultrasound imaging Circulation 1986; 74: 1399–1406

  17. 17

    Ferrara LA et al. Early changes of the arterial carotid wall in uncomplicated primary hypertensivepatients. Study by ultrasound high-resolution B-mode imaging Arterioscler Thromb 1994; 14: 1290–1296

  18. 18

    Levy D et al. Echocardiographically detected left ventricular hypertrophy: prevalence and risk factors. The Framingham Heart Study Ann Int Med 1988; 108: 7–13

  19. 19

    Assmann G, Schulte H . The prospective cardiovascular Münster (Procam) study: prevalence of hyperlipidaemia in persons with hypertension and/or diabetes mellitus and the relationship to coronary heart disease Am Heart J 1988; 116: 1713–1724

  20. 20

    Prevention of coronary heart disease in clinical practice. Recommendations of the Second Joint Task Force of European and other Societes on coronary prevention Eur Heart J 1998; 19: 1434–1503

  21. 21

    The International Task Force for prevention of coronary heart disease in cooperation with the International Atherosclerosis Society. Coronary heart disease: reducing the risk. The scientific background for primary and secondary prevention of coronary heart disease. A worldwide view Nutr Metab Cardiovasc Dis 1998; 8: 205–271

  22. 22

    Williams RR et al. Familial dyslipidemic hypertension. Evidence from 58 Utah families for a syndrome present in approximately 12% ofpatients with essential hypertension JAMA 1988; 259: 3579–3586

  23. 23

    Ferrannini E et al. Insulin resistance in essential hypertension N Engl J Med 1987; 317: 350–357

  24. 24

    Hunt SC, Williams RR, Smith JB, Ash KO . Associations of three erythrocytes cation transport system with plasma lipids in Utah subjects Hypertension 1986; 8: 30–36

  25. 25

    Reaven GR . Role of insulin resistance in human disease Diabetes 1988; 37: 1595–1607

  26. 26

    Musgrave IF, Bachmann AW, Saar N, Gordon RD . A comparison of cardiovascular and catecholamine responses to three stimuli in mild hypertension Metabolism 1984; 33: 718–723

  27. 27

    Guida L et al. Clinic-daytime blood pressure difference and cardiovascular damage J Hypertens 1999; 17: 331–337

  28. 28

    Ferrara LA et al. Olive oil and reduced need for antihypertensive medications Arch Intern Med 2000; 160: 837–842

  29. 29

    Hamada M et al. Enhanced blood pressure response to isometric handgrip exercise with essential hypertension: effects of propanolol and razosin J Hypertens 1987; 5: 305–309

  30. 30

    Salonen JT, Salonen R . Ultrasonographically assessed carotid morphology and the risk of coronary artery disease Arterioscler Thromb 1991; 11: 1245–1249

  31. 31

    Hayakawa H, Raij L . Relationship between hypercholesterolemia, endothelial dysfunction and hypertension J Hypertens 1999; 17: 611–619

  32. 32

    Creager MA, Selwyn A . When ‘normal’ cholesterollevels injure the endothelium Circulation 1997; 96: 3255–3257

  33. 33

    Nazzaro P et al. Distinct and combined vascular effects of ACE blockade and HMG – CoA reductase inhibition in hypertensive subjects Hypertension 1999; 33: 719–725

  34. 34

    Sung BH, Izzo JL, Wilson MF . Effects of cholesterol reduction on BP response to mental stress inpatients with high cholesterol Am J Hypertens 1997; 10: 592–599

  35. 35

    Poli A et al. Ultrasonographic measurement of the common carotid artery wall thickness in hypercholesterolemicpatients Atherosclerosis 1988; 70: 253–261

  36. 36

    Heiss G et al the ARIC Investigators. Carotid atherosclerosis measured by B-mode ultrasound in populations: associations with cardiovascular risk factors in the ARIC Study Am J Epidemiol 1991; 134: 250–256

  37. 37

    Pan WH, Bai CH, Chen JR, Chiu HC . Associations between carotid atherosclerosis and high factor VIII activity, dyslipidemia, and hypertension Stroke 1997; 28: 88–94

  38. 38

    Roman MJ et al. Relation of blood pressure variability to carotid atherosclerosis and carotid artery and left ventricular hypertrophy Arterioscler Thromb Vasc Biol 2001; 21: 1507–1511

Download references


The authors are grateful to Mrs Rosanna Scala for her linguistic revision.

Author information

Correspondence to L A Ferrara.

Rights and permissions

Reprints and Permissions

About this article


  • serum cholesterol
  • blood pressure
  • handgrip
  • endothelium function

Further reading