Journal of Human (2002) 16 (Suppl 1), S34–S37  2002 Nature Publishing Group All rights reserved 0950-9240/02 $25.00 www.nature.com/jhh Endothelial dysfunction, - converting enzyme inhibitors and calcium antagonists

PLo´pez-Jaramillo1,2 and JP Casas1 1Instituto Colombiano de Investigaciones Biome´dicas (ICIB), Bucaramanga, Colombia; 2Universidad Industrial de Santander, Colombia

The plays a crucial role in the pathogen- lar mortality, an effect that could be the consequence esis of . Endothelial function is of an improvement in the endothelial function. Recent attenuated by the presence of different well known car- studies have shown that a calcium antagonist might diovascular risk factors. Evaluation of endothelial vaso- improve the endothelial function, however, there is dilator function serve as an index integrating the overall controversy about this action and also about the poten- stress imposed by cardiovascular risk factors and tial mechanisms for the effect of a calcium antagonist reinforce the suggestion that endothelial dysfunction is in the regulation of endothelial function. an early marker of cardiovascular disease that precedes Journal of Human Hypertension (2002) 16 (Suppl 1), S34– clinical manifestations. Angiotensin-converting enzyme S37. DOI: 10.1038/sj/jhh/1001339 inhibitors have been shown to reduce the cardiovascu-

Keywords: endothelial function; angiotensin-converting enzyme inhibitors; calcium antagonists

Introduction reactive .5,6 Estimates of endothelial function by this test are remarkably stable over time, The endothelium is a complex endocrine and para- but no clear set of normal values has been crine organ that affects , smooth muscle developed.7 Because of the different genetic back- cell proliferation, platelet aggregation, monocyte 1 ground and variability in the distribution of CVD and leukocyte adhesion, and fibrinolysis. The func- risk factors, normal values of FMD may differ among tional integrity of the endothelium is crucial to pro- populations.8 We have found in 253 normotensive portionate to the cardiovascular system vasodilator, healthy volunteers, that the mean %FMD was antiatherosclerotic and antithrombotic effects dependent on the presence of risk factors for CVD. through the secretion of vasoactive substances such Subjects with no risk factors had a mean %FMD of 1–3 as prostacyclin and particularly (NO). 13.74% (95% confidence interval (CI): 13.14, 14.35), while in those with at least one risk factor the mean Endothelial function and cardiovascular was significantly lower: 7.40% (95% CI: 4.33, 9.91).8 disease Obesity, smoking, and hypercholesterolaemia were the modifiable risk factors with the largest Endothelial vasodilator function is attenuated by the independently significant reduction effects on presence of different well known cardiovascular risk %FMD, as evaluated in the multivariate model. factors such as hyperlipidaemia, diabetes mellitus, Obese subjects had a %FMD 4.60% units lower than hypertension and smoking.4 In consequence, attenu- non-obese subjects (95% CI: −6.50, −2.71). Regard- ation of endothelial mediated vasodilatation could less of the effect of other risk factors, smokers and be considered as an early marker of cardiovascular hypercholesterolaemic subjects had a reduction of disease (CVD) and could be an important tool to 4.45 (P Ͻ 0.001) and 3.67 (P Ͻ 0.001) units in investigate cardiovascular pathophysiology. %FMD as compared with non-smokers and non- Recently, a non-invasive method using high-resol- hypercholesterolaemic subjects, respectively.8 ution ultrasound has been developed to assess flow We did a receiver operating characteristic (ROC) mediated dilation (FMD) of the brachial , analysis to identify the best %FMD cut-off point which is modulated by the release of NO during useful to identify subjects with and without cardio- vascular risk factors. The area under the ROC curve Correspondence: P Lo´pez-Jaramillo, Scientific Director, ICIB, PO was 83.28% (95% CI: 77.80, 88.77). The ROC plot Box: 384, Bucaramanga. Colombia suggested a cut-off point of 12 for %FMD. Sensi- E-mail: [email protected] tivity and specificity for this cut-off point were Endothelium and pharmacological interventions PLo´pez-Jaramillo and JP Casas S35 76.7% and 66.1%, respectively. The positive predic- term effects of therapeutic interventions on systemic tive value for this cut-off point was 47.9%, while endothelial function. the negative predictive values reached 87.5%. We also explored the performance of other cut-off Angiotensin-converting enzyme (ACE) points, based on their validity and predictive capacity and selected the point at which the number inhibitors and calcium antagonists as of false-positive began to exceed the number of false- alternatives to improve endothelial negative test results. Based on this criterium, the dysfunction best %FMD cut-off point was 10.4, which resulted A functional ACE system present in the vascular in a sensitivity of 71.2% and a specificity of 77.2%. endothelium contributes to the regulation of Positive and negative predictive values for this cut- 17 off point were 55.9% and 86.9%, respectively. Using adequate vascular biology. ACE inhibitors have been shown to reduce cardio- this cut-off point, endothelial dysfunction was 3.13 times more frequent in subjects with, than in sub- vascular mortality in patients with left ventricular dysfunction, acute myocardial infarction and also in jects without cardiovascular risk factors (95% CI: 18–20 8 patients at high risk of cardiovascular events. 2.30, 4.25). Similar analyses have been realised to 21 evaluate if the %FMD is useful in identifying sub- Vanhoutte et al, showed that ACE inhibition had jects with and without . beneficial effects on endothelial function in animal models. The TREND and BANFF Studies, carry out These analysis reported that %FMD was a sensitive and specific (71.3% and 81% respectively) screen- in humans, evaluated the effect of quinapril, a highly specific tissular ACE inhibitor on endothelial ing test to predict the presence of coronary artery function, in resistance and conduit respect- disease.9 Moreover, our results showed that FMD measure- ively, and confirmed the beneficial effects of ACE inhibitors in reversing the endothelial dysfunc- ments can be made with high accuracy and pre- tion.22,23 cision (Lin’s correlation coefficient of 0.88), and very low bias (mean inter-observer difference of Several studies have suggested that the mech- anism by which the tissular ACE inhibition might −0.30%).8 These results clearly show that no invas- improve the endothelial function involve the NO ive evaluation of endothelial vasodilator function may serve as an index integrating the overall stress pathway. Accumulating bradykinin may lead to increased NO synthesis and activity.24 Another imposed by CVD risk factors and reinforce the potential mechanism involves the inhibition of the suggestion that endothelial dysfunction is an early marker of CVD that precedes clinical manifes- production of angiotensin II and as a consequence decrease formation of superoxide anion (O− ). It is tations.8,10 2 well known that angiotensin II stimulates the An important body of evidence suggests that endothelial dysfunction is a generalised process that NADH/NAD(P)H oxidase of endothelial and smooth muscle cells.25 Figure 1 shows the effects of highly is not necessarily confined to vessels with overt 4–10 specific tissular ACE inhibitors in restoring the bal- atherosclerotic alterations. This proposal is − strengthened by the high correlation (Ͼ0.75) ance on the endothelial production of NO and O 2. Recent evidence has shown that acute adminis- reported between the coronary vasomotor response and the peripheral endothelial function of the brach- tration of calcium antagonists improves the abnor- ial artery evaluated by FMD.11,12 mal coronary vasomotion response in hypertension, and hypercholesterolaemia.26,27 In Recently, three different groups10,13,14 have dem- onstrated that the diagnostic of endothelial dysfunc- addition, amlodipine and nifedipine have been tion realised in coronary and peripheral arteries in recently shown to increase the coronary vaso- patients with different stages of coronary artery dis- ease is related with an increased incidence of an adverse long-term outcome of coronary heart dis- ease. Several studies have demonstrated that treatment of cardiovascular risk factors known to lead to endo- thelial dysfunction is associated with a decrease in cardiac events in both primary and secondary pre- vention studies,15,16 underscoring the concept that the reduction in cardiac events in this patient’s population may be secondary to improvement in coronary endothelial function. In conclusion, FMD is attractive as a non-invasive method to assess the integrity of vascular endo- thelial function and can be recommended as a Figure 1 Action mechanism of ACE inhibitors (ACE-I) to reverse screening test for the detection of patients at risk of endothelial dysfunction, through improving the imbalance − CVD. It also can be used to assess short and long- between nitric oxide (NO) and superoxide (O2).

Journal of Human Hypertension Endothelium and pharmacological interventions PLo´pez-Jaramillo and JP Casas S36 dilation and also the NO production from canine 9 Schroeder S et al. Noninvasive determination of endo- microvessels.28,29 thelium-mediated vasodilation as a screening test for The potential mechanisms for this effect are not coronary artery disease: pilot study to assess the pre- yet clear. Early in the characterisation of the L-argi- dictive value in comparision with angina pectoris, nine:NO pathway the crucial role of physiological exercise electrocardiography, and myocardial per- fusion imaging. Am Heart J 1999; 138: 731–739. calcium concentrations in the production of endo- 10 Schachinger V, Britten MB, Zeiher AM. Prognostic thelial NO and the control of vascular tone was dem- impact of coronary vasodilator dysfunction on adverse 3 onstrated. More recently, it has been proposed that long-term outcome of coronary heart disease. Circu- calcium-antagonists up-regulate the activity of endo- lation 2000; 101: 1899–1906. thelial NO synthase in vitro.30 However, it has been 11 Takase B et al. Endothelium-dependent flow-mediated demonstrated that in the endothelium the principal vasodilation in coronary and brachial arteries in sus- calcium channels are mechano activates.31 Calcium pected coronary artery disease. Am J Cardiol 1998; 82: antagonists act specifically in calcium channel volt- 1535–1539. age operates, so it is unlikely that these drugs per- 12 Anderson TJ et al. Close relation of endothelial func- form any direct effect in the endothelial function. tion in the human coronary and peripheral circulation. So, other mechanisms have been proposed to J Am Coll Cardiol 1995; 26: 1235–1241. 13 Suwaidi JA et al. Long-term follow-up of patients with explain the effect of calcium antagonists in the mild coronary artery disease and endothelial dysfunc- endothelial function, as the antioxidant activity tion. Circulation 2000; 101: 948–954. described for these compounds. Thus, amlodipine 14 Neunteufl T et al. Late prognostic value of flow- has been demonstrated to have antioxidant proper- mediated dilation in the brachial artery of patients ties in atherosclerotic animal models, but its effect with chest pain. Am J Cardiol 2000; 86: 207–210. on endothelial function has not been extensively 15 Shepherd JH et al. For the West of Scotland Coronary assessed in humans.32 Prevention Study Group. Prevention of coronary heart Recently, Anderson et al23 reported that amlodip- disease with pravastatin in men with hypercholestero- ine did not improve reactive hyperaemic blood flow lemia. N Engl J Med 1995; 333: 1301–1307. in peripheral arteries of patients with coronary dis- 16 Scandinavian Simvastatin Survival Study Group. Ran- ease in contrast to the improvement observed with domized trial of cholesterol lowering in 4444 patients quinapril. with coronary heart disease: The Scandinavian Simva- In conclusion, the ACE inhibitors, especially Survival Study (4S). Lancet 1994; 344: 1383– 1389. those with high affinity with the endothelial ACE, 17 Luscher T. Endothelial dysfunction: the role and show the best effects in reversing endothelial dys- impact of the -angiotensin system. Heart 2000; 84 function, an activity that is related to accumulation (Suppl I): i20–i22. 33 of bradykinin. The results with calcium antagon- 18 Pfeffer MA. ACE inhibition in acute myocardial infarc- ists are still controversial. tion. N Engl J Med 1995; 332: 118–120. 19 The Heart Outcomes Prevention Evaluation Study (HOPE) investigators. Effects of an Angiotensin Con- References verting enzyme inhibitor, Ramipril, on cardiovascular events in high risk patients. N Engl J Med 2000; 342: 1 Moncada S et al. The L-arginine: nitric oxide pathway. 145–153. J Cardiovasc Pharmacol 1991; 17:S1–S9. 20 SOLVD investigators. Effect of Enalapril on mortality 2Lo´pez-Jaramillo P, Guarner F, Moncada S. Effects of calcium and parathyroid hormone on prostacyclin and the development of heart failure in asymptomatic synthesis by vascular tissue. Life Sci 1987; 40: 983– patients with reduced left ventricular ejection fraction. 986. N Engl J Med 1992; 327: 685–691. 3Lo´pez-Jaramillo P, Gonza´lez MC, Palmer RMJ, Mon- 21 Vanhoutte P. Endothelium-dependent effects of con- cada S. The crucial role of physiological Ca2+ concen- verting-enzyme inhibitors. J Cardiovasc Pharmacol trations in the production of endothelial nitric oxide 1993; 22: S10–S16. and the control of vascular tone. Br J Pharmacol 1990; 22 Mancini J et al. Angiotensin-converting enzyme inhi- 101: 489–493. bition with quinapril improves endothelial vasomotor 4 Lusher T. The endothelium and cardiovascular disease dysfunction in patients with coronary artery disease. a complex relation. N Engl J Med 1994; 330: 1001– The TREND (Trial on Reversing Endothelial 1003. Dysfunction) Study. Circulation 1996; 94: 258–265. 5 Celermajer DS et al. Non-invasive detection of endo- 23 Anderson TJ et al. Comparative study of ACE-inhi- thelial dysfunction in children and adults at risk of bition, angiotensin II antagonism, and calcium chan- atherosclerosis. Lancet 1992; 340: 1111–1115. nel blockade on flow-mediated vasodilation in 6 Dakak N et al. Contribution of nitric oxide to reactive patients with coronary disease (BANFF Study). JAm hyperemia. Impact of endothelial dysfunction. Hyper- Coll Cardiol 2000; 35:60–66. tension 1998; 32:9–15. 24 Sudhir K et al. Effects of selective angiotensin II recep- 7 Hijmering ML et al. Non-invasive evaluation of vascu- tor antagonism and angiotensin converting enzyme lar function. Eur Heart J 1999; 1 (Suppl): M16–M18. inhibition on the coronary vasculature in vivo: intrava- 8 Accini JL et al. Colombian study to Assess the use of scular two-dimensional and doppler ultrasound stud- Noninvasive Determination of Endothelium-mediated ies. Circulation 1993; 87: 931–938. Vasodilation (CANDEV). Normal values and factors 25 Griendling KK, Mineri CA, Ollerenshaw JD, Alexander associated. Endothelium 2000; 8: 157–166. RW. Angiotensin II stimulates NADH and NADPH oxi-

Journal of Human Hypertension Endothelium and pharmacological interventions PLo´pez-Jaramillo and JP Casas S37 dase activity in cultured vascular smooth muscle cells. 30 Ding Y, Vaziri D. Nifedipine and diltiazem but not ver- Circ Res 1994; 74: 1141–1148. apamil up-regulate endothelial nitric-oxide synthase 26 Frielingsdorf J et al. Normalization of abnormal coron- expression. J Pharmacol Exp Ther 2000; 292: 606–609. ary vasomotion by calcium antagonists in patients 31 Lansman JB, Hallam TJ, Rink TJ. Single stretch- with hypertension. Circulation 1996; 93: 1380–1387. activated ion channels in vascular endothelial cells as 27 Kaufmman P et al. Reversal of abnormal coronary mechanotransducers. Nature 1987; 325: 811–813. vasomotion by calcium antagonist in patients with 32 Chen L et al. Preservation of endogenous antioxidant . Circulation 1998; 97: 1348– activity and inhibition of lipid peroxidation as com- 1354. mon mechanisms of antiatherosclerotic effects of 28 Zhang X, Hintze TH. Amlodipin releases nitric oxide vitamin E, lovastatin and amlodipine. J Am Coll Car- from canine coronary microvessels: an unexpected diol 1997; 30: 569–575. mechanism of action of a calcium channel-blocking 33 Horning B, Kohler C, Drexler H. Role of bradykinin in agent. Circulation 1998; 97: 576–580. mediating vascular effects of angiotensin-converting 29 Kitakaze M et al. Nifedipine-induced coronary enzyme inhibitors in humans. Circulation 1997; 95: vasodilation in ischemic hearts is attributable to bra- 1115–1118. dykinin-and NO-dependent mechanism in dogs. Cir- culation 2000; 101: 311–317.

Journal of Human Hypertension