We treated with nifedipine or lisinopril 38 essential hypertensive patients with left ventricular hypertrophy. The study had a single-blind crossover design; nifedipine or lisinopril was given for the first 24 weeks, and then patients were crossed over to the other antihypertensive agent for another 24 weeks. Both nifedipine and lisinopril significantly decreased mean arterial pressure to the same extent. Although lisinopril decreased left ventricular mass index more rapidly than nifedipine, 48 weeks of antihypertensive treatment with nifedipine or lisinopril reduced the extent of left ventricular hypertrophy to the same level. Stepwise multiple linear regression analysis revealed that the reversal of left ventricular hypertrophy may be mainly due to a reduction in mean arterial pressure during the 24-week nifedipine treatment and due to an improvement of aortic compliance during the lisinopril treatment. Both nifedipine and lisinopril are effective in the reversal of hypertensive left ventricular hypertrophy; however, the agents have disparate actions on hemodynamic factors.
Hypertension can lead to the development of LVH,1 one of the most important risk factors for congestive heart failure, ischemic heart disease, and sudden cardiac death.2 In clinical and experimental studies, antihypertensive agents have shown disparate effects on myocardial mass in LVH despite inducing a similar reduction of blood pressure. ACE inhibitors and calcium antagonists have been shown to reduce left ventricular mass significantly,3 4 whereas other vasodilators such as hydralazine did not cause regression of LVH despite a significant decrease in arterial pressure.5 Although the severity of hypertensive LVH has been shown to be somewhat related to the level of arterial pressure,6 involvement of stimuli other than arterial pressure itself seems more important in hypertensive LVH.7 Recently, after an insertion/deletion polymorphism of the ACE gene was detected, carotid intima-media thickness has been known to be associated with LVH and high cardiac ACE activity.8 It has also been reported that cardiac hypertrophy and hypertension are strongly correlated with PWV and arterial distensibility, which can be ameliorated by calcium antagonists and ACE inhibitors.9 No comparative crossover studies have been made of the effects of these two different drug classes on LVH regression in relation to aortic compliance.
We have previously reported that lisinopril, a new long-acting ACE inhibitor, improved aortic compliance compared with nifedipine although both agents decreased MAP to the same extent.10 Therefore, in this study we treated essential hypertensive patients with either lisinopril or the calcium antagonist nifedipine and compared their effects on cardiac structure and function using an ultrasound technique.
We studied 38 hypertensive patients (16 men and 22 women; mean age, 68.7±6.3 years; World Health Organization stage II) with LVH, excluding individuals with bruits, ischemic attacks, completed strokes, or limb ischemia. Hypertension was defined as blood pressure of at least 160 mm Hg systolic or 90 mm Hg diastolic on three consecutive readings after all medication had been discontinued for 4 weeks or longer. Patients diagnosed as having essential hypertension at least 5 years previously had not been treated with antihypertensive drugs. The cause of hypertension was not known in any of the patients. All patients displayed echocardiographic evidence of LVH, with interventricular septal wall thickness greater than or equal to 12 mm at end diastole, left ventricular end-diastolic posterior wall thickness greater than or equal to 12 mm, or both. None of these patients had evidence of any other heart disease, including coronary artery disease, as assessed by cardiovascular history, physical examination, electrocardiogram, or echocardiogram; all had normal sinus rhythm without signs of heart failure. This study was approved by the institutional review committee of PIA Nakamura Hospital. After the nature and purpose of the study had been explained to them, all patients gave informed consent. The procedures followed were in accordance with institutional guidelines.
The present study had a single-blind, crossover design; lisinopril was given once daily, and sustained-release nifedipine was administered twice daily. The initial dose of nifedipine or lisinopril was 10 mg/d; both of these doses were increased to a maximum of 30 mg/d so that a reduction in MAP of 10 mm Hg or more was achieved. Throughout the study, all patients received a diet containing 50 to 100 mEq/d sodium. After a run-in period of 4 weeks, patients were randomly allocated to four groups: the L-N group took lisinopril as the first active drug for 24 weeks and then were crossed over to nifedipine for another 24 weeks; the N-L group took nifedipine as the first active drug for 24 weeks and then were crossed over to lisinopril for another 24 weeks; the L-L group took lisinopril for 48 weeks; and the N-N group took nifedipine for 48 weeks. All hemodynamic and laboratory studies were performed at the end of the run-in period and both active treatment periods.
All studies were performed while patients were in a supine position and postabsorptive state between 11 am and noon after a 30-minute rest. Room temperature was held constant between 20° and 23.5°C. Three sets of blood pressure measurements were made in both arms simultaneously by the same two observers. MAP was calculated as the sum of diastolic pressure plus one third of pulse pressure.
Ultrasound studies used commercially available phased-array echocardiographic Doppler systems (U-sonic model RT5000 or RT8000, Yokogawa Medical Systems). M-mode echocardiograms were recorded of the left ventricular septum and posterior wall at the level of the chordae tendineae. The end-diastolic (LVDd) and end-systolic (LVDs) dimensions of the left ventricle were determined. Percent fractional shortening of the left ventricle (%FS) was calculated as %FS=100×(LVDd−LVDs)/LVDd.
ESS was determined as follows11 :ESS|<|[|>|g|<|\cdot|>|cm^|<||<|-|>|2|>||<|]|>||<|=|>|\frac|<|Peak Arterial Pressure|<|\times|>|LVDs^|<|2|>||>||<|4|<|\times|>|Wall Thickness|<|\times|>|(LVDs|<|+|>|Wall Thickness)|>|ESS is an expression of the average meridional wall stress, which can be defined as the force per unit area acting at the equatorial plane of the ventricle in the direction of the apex to the base axis. Left ventricular mass was estimated from the formula of Devereux and Reichek12 as follows: Left ventricular mass (g)=1.04×[(IVST+PWT+LVIDd)3−LVIDd3]−13.6, where IVST is interventricular septal thickness, PWT is posterior wall thickness, and LVIDd is internal dimension of the left ventricle in diastole. Left ventricular mass was divided by body surface area to derive LVMI, which was taken as an index of the degree of LVH.
The E velocity was the peak transmitral flow velocity during early diastole, and the A velocity was the peak transmitral flow velocity during atrial systole. The mean percentages of interobserver and intraobserver variabilities for measurement of these velocities are 5.3% and 2.8%, respectively. The ratio of the A velocity to the E velocity (A/E) was calculated. Stroke volume was determined with a mitral inflow method.13
The peak flow velocity of diaphragmatic aortic flow during systole was measured. Flow velocity–time integrals of diaphragmatic aortic flow were determined by planimetric measurement of the flow-velocity curve; one (S) was the integral in systole, and the other (D) was the integral in diastole. The ratio of diastole to systole (D/S) of diaphragmatic aortic flow was calculated. According to the Windkessel model, as the aorta becomes more compliant, more of the stroke volume ejected by the left ventricle is considered to be stored in the elastic aorta, resulting in less forward arterial flow in systole and more forward flow in diastole, that is, an increase in D/S.
PWV was obtained by dividing the aortic distance by the delay time between the simultaneously recorded aortic flow pulses. As previously reported,10 Doppler flow recordings were taken at two sites simultaneously—the aortic valve and the descending thoracic aorta at the diaphragm—and the transit time was obtained from the foot-to-foot delay between the flow waves. The aortic distance between these two sites was measured by computed tomographic reconstruction imaging in the long-axis view or by nuclear magnetic resonance imaging in the long-axis view.
Plasma norepinephrine was measured by an electrochemical method with high-performance liquid chromatography.14 Plasma Ang II was measured by radioimmunoassay with the magnesium silicate (Florisil) absorption method.15
Data are expressed as mean±SD. Intergroup unpaired comparisons during the run-in period were made by one-way ANOVA, and subsequent comparison between two groups was made by the nonparametric Wilcoxon U test.
Intragroup paired absolute data with and without treatment were analyzed by the Wilcoxon matched pairs signed rank test. Changes during the first active treatment were compared with changes during the second active treatment with the Wilcoxon matched pairs signed rank test. When the effects of lisinopril and nifedipine were investigated during the first active treatment, changes in parameters during the first active treatment in the N-N and N-L groups were compared with those in the L-N and L-L groups with the nonparametric Wilcoxon U test. A value of P<.05 was considered statistically significant.
To identify determinants of LVMI, we performed stepwise multiple linear regression computations. The change in LVMI was used as the dependent variable, with changes in systolic pressure, diastolic pressure, MAP, pulse pressure, stroke volume, PWV, peak diaphragmatic aortic flow, S, D, D/S, plasma norepinephrine, and plasma Ang II as potential dependent variables.
Subclassification of Patients
All patients completed the study. The 38 patients were randomly allocated to four groups: N-N (n=9; mean age, 67.7±5.9 years), N-L (n=10, 69.8±6.0 years), L-N (n=9, 68.8±6.9 years), and L-L (n=10, 68.6±6.2 years). Systemic and regional hemodynamic parameters and laboratory data did not differ significantly during the run-in period among the four groups.
Hemodynamic and Humoral Parameters
Both nifedipine and lisinopril significantly reduced systolic pressure, diastolic pressure, and MAP to the same extent during the first active treatment (average dose, 20.2 and 18.4 mg/d, respectively) (Table⇓). The second active treatment elicited no further change in blood pressure in any of the four groups (Fig 1A⇓). Neither nifedipine nor lisinopril had any effect on heart rate.
Although nifedipine and lisinopril significantly decreased LVMI during the first active treatment, the decrease was greater with lisinopril than with nifedipine (P<.01). In patients who took nifedipine as the first active drug, both nifedipine (P<.05) and lisinopril (P<.01) decreased LVMI significantly during the second active treatment. In contrast, those who took lisinopril during the first active treatment showed no further decrease in LVMI by either nifedipine or lisinopril as the second active drug. At the end of the second active treatment, the decreases in LVMI did not differ significantly among the four groups (Fig 1B⇑). These data suggest that lisinopril elicits regression of LVH more rapidly than nifedipine but that whichever agent is used, reversal of LVH eventually may occur to the same extent if patients are treated for a long enough time (1 year).
Both nifedipine and lisinopril decreased PWV significantly during the first active treatment, but the decrease in PWV was greater with lisinopril than with nifedipine (P<.01). In the N-N and L-L groups, the second active treatment had no further effect on PWV. In the N-L group, the decreased PWV by nifedipine was further lowered by lisinopril during the second active treatment (P<.01), whereas the effect of lisinopril on decreasing PWV was reversed by nifedipine in the L-N group (P<.05) (Fig 1C⇑). These data indicate that the effect of antihypertensive agents on PWV is reversible and that lisinopril decreases PWV more than nifedipine irrespective of similar reductions of MAP.
Throughout the first and second active treatments, lisinopril increased the D/S ratio of diaphragmatic aortic flow (P<.01), but nifedipine did not (Fig 1D⇑). Thus, the increase in the D/S ratio of diaphragmatic aortic flow by lisinopril was easily reverted by nifedipine. During the first active treatment, there was a significant correlation between the change in PWV and that in the D/S ratio of diaphragmatic aortic flow in the lisinopril group (r=−.66, P<.01) (Fig 2⇓) but not in the nifedipine group.
Nifedipine decreased ESS during the first active treatment (from 54.4±9.0 to 51.1±9.8 g·cm−2, n=19, P<.01) but lisinopril did not (from 54.2±6.9 to 55.1±9.1 g·cm−2, n=19).
Neither nifedipine nor lisinopril had any effect on plasma norepinephrine or Ang II concentration.
Regression of LVH
Since the effects of both antihypertensive drugs on hemodynamic parameters did not differ significantly during the first active treatment regardless of whether lisinopril or nifedipine was administered first, we combined the results of nifedipine in the N-N and N-L groups and those of lisinopril in the L-L and L-N groups. To identify determinants of decreases in LVMI, we performed stepwise multiple linear regression computations in both nifedipine and lisinopril groups during the first active treatment. We tested the changes in systolic and diastolic pressures, MAP, pulse pressure, stroke volume, PWV, peak diaphragmatic aortic flow, S, D, aortic D/S ratio, plasma norepinephrine, and plasma Ang II as independent variables. In the nifedipine group, stepwise multiple linear regression analysis revealed that only the change in MAP was independently correlated with the dependent variable (the change in LVMI) (Change in LVMI [g·m−2]=2.10×Change in MAP [mm Hg]+9.11; R2=0.56, P<.01, n=19; Fig 3⇓). In the lisinopril group, the change in the D/S ratio of diaphragmatic aortic flow was entered only into the final regression equation (Change in LVMI [g·m−2]=−164×Change in D/S Ratio of Diaphragmatic Aortic Flow−30.3; R2=0.56, P<.01, n=19; Fig 4⇓).
There was a negative correlation between the basal values of ESS before treatment and the change in LVMI by nifedipine (r=−.54, P<.01, n=19), suggesting that nifedipine effectively reduces LVMI in patients with higher ESS values. Meanwhile, there was no significant relationship between these parameters in the lisinopril group.
Cardiac Function and Regression of LVH
Neither lisinopril nor nifedipine showed any significant change in the percentage of fractional shortening throughout the 24- or 48-week treatment. During the first active treatment, the A/E ratio remained unchanged in the nifedipine group but decreased in the lisinopril group (P<.01) (Fig 1E⇑). In the lisinopril group, there was a direct correlation between the change in the A/E ratio and that in LVMI (Change in A/E=0.00645×Change in LVMI [g·m−2]+0.125; r=.67, n=19, P<.01). Thus, cardiac performance remained unchanged in the nifedipine group, whereas lisinopril may have improved early diastolic filling of the left ventricle in the first active treatment. After 48 weeks of treatment, a significant decrease in the A/E ratio was seen in the four groups (all P<.05).
In this study, we assessed the effects of two different kinds of antihypertensive agents, the calcium antagonist nifedipine and the ACE inhibitor lisinopril, on regression of LVH. Both agents decreased MAP to the same extent. Although they both caused significant regression of LVH, lisinopril reduced LVMI more rapidly than nifedipine. The major mechanism by which an antihypertensive agent elicits regression of LVH is through afterload reduction, as in the nifedipine treatment, or an improvement of aortic compliance, as in the lisinopril treatment. Reversal of LVH by either antihypertensive agent resulted in no change or improved early diastolic filling of the left ventricle. The clinical evidence and rationale for this conclusion are discussed below.
Regression of LVH
The mechanisms by which antihypertensive agents induce LVH reversal are still controversial and appear to be multifactorial. Data obtained in vivo and in vitro have shown that mechanical events accompanying hemodynamic overload provide signals for hypertrophy by modifying gene expression and protein synthesis as well as growth of myocardial cells.16 17 In the present study, we have confirmed by stepwise multiple linear regression analysis that when nifedipine was administered as the first active treatment, a reduction in arterial pressure is a leading factor in LVH reversal.
According to Laplace's theorem, ventricular wall stress is directly proportional to intraventricular pressure and ventricular diameter and is inversely related to ventricular wall thickness. Therefore, when hypertension increases ventricular wall stress, LVH actually represents an adaptive myocardial remodeling that serves to normalize this increased wall stress.18 In this study, nifedipine decreased ESS significantly, and more marked reductions in LVMI were seen in hypertensive patients with higher basal values of ESS, which is in line with the major contribution of reduced arterial pressure to the regression of LVH during nifedipine treatment.
In contrast to the situation with nifedipine, reducing arterial pressure is not the sole factor responsible for LVH reversal during lisinopril treatment because the change in LVMI and that in MAP (r=.47) were only weakly correlated. Furthermore, because even a non-antihypertensive dose of ACE inhibitor reversed LVH in thoracic aorta–constricted rats,19 possible mechanisms other than a decrease in arterial pressure must be taken into account in the detection of LVH reversal by lisinopril. In our preliminary study,10 8-week treatment with lisinopril decreased plasma atrial natriuretic peptide (from 42.8±19.5 to 26.4±16.2 pg/mL, n=12, P<.05) but treatment with nifedipine did not (from 40.9±20.5 to 36.8±21.3 pg/mL, n=12). Because this decrease in plasma atrial natriuretic peptide by lisinopril indicates a reduction in cardiac filling pressure, preload reduction may contribute to LVH reversal. Another possibility is that lisinopril may antagonize the growth-promoting influence of Ang II. Recent studies clearly demonstrated that Ang II participates in the development of myocardial hypertrophy independent of changes in arterial pressure and cardiac work.20 21 ACE inhibitors have been known to suppress hypertrophy-associated ventricular remodeling, such as the interstitial fibrosis of myocardium,22 by decreasing the cardiac tissue Ang II.23
In the present study, the increase in the aortic D/S ratio was shown to be the major determinant in the reduction of LVMI by lisinopril. According to the Windkessel model, as the aorta becomes more compliant, more of the stroke volume ejected by the left ventricle is considered to be stored in the elastic aorta, resulting in less forward arterial flow in systole and more forward flow in diastole, ie, an increase in aortic D/S ratio. Thus, lisinopril appears to facilitate regression of LVH via increased aortic compliance. Furthermore, because the change in aortic D/S ratio was closely correlated with that in PWV, the increase in aortic D/S ratio seems to reflect the improvement of aortic compliance that can be estimated from PWV.
Studies of arterial impedance have indicated that LVH is influenced by the state of arterial compliance (buffering function of large arteries), which is reported to be reduced in hypertensive individuals, as well as by the level of arterial pressure and total peripheral resistance.24 25 Recent reports of the existence of a local renin-angiotensin system in blood vessel walls, especially large arteries, have raised the possibility that vascular function may be regulated by local production of Ang II independent of the circulating levels.26 The vascular renin-angiotensin system activated in hypertension27 28 29 30 stimulates vasoconstriction and promotes blood vessel growth directly31 or indirectly via its effect on norepinephrine release from noradrenergic nerve endings.32 Thus, blockade of local Ang II production by lisinopril possibly increases aortic compliance, as demonstrated by other ACE inhibitors,33 leading to LVH reversal.
It is noteworthy that this improvement of aortic compliance by lisinopril is fully reversible, whereas the decreasing effect of lisinopril on LVH is not reversed even if antihypertensive treatment is switched to nifedipine. This discrepancy might be explained by experimental studies that LVH can be reversed with antihypertensive drugs, whereas aortic structural changes remain unmodified, reflecting a lack of parallelism between cardiac and vascular effects.34
In the present study, the decrease in LVMI by lisinopril was greater than that by nifedipine after 24 weeks of treatment. However, after 48 weeks of treatment, the decreases in LVMI among the N-N, N-L, L-N, and L-L groups did not differ significantly. These findings suggest that lisinopril decreases LVMI more rapidly than nifedipine and that whichever antihypertensive agent is used, reversal of LVH will eventually occur if patients are treated for 1 year.
Effects of Antihypertensive Treatments on Cardiac Performance
In hypertensive LVH, left ventricular systolic function is usually normal or preserved.35 Altered diastolic function with impaired diastolic filling, which is characterized by a decrease in the velocity of relaxation and an increase in the time to peak relaxation, has been well demonstrated previously.36 37 Thus, an increase in A/E ratio in transmitral flow is seen when the left atrium functions as a “booster pump” to fill the less-distensible left ventricle. The reason for this impaired diastolic function is not known precisely, but progressive interstitial and perivascular fibrosis of a stiffer and less-distensible left ventricle that accompanies LVH may be at least partly involved.38
Different antihypertensive agents may exert varying effects on cardiac performance with LVH reversal, depending on their ability to affect calcium metabolism, myocardial collagen concentration, or coronary flow reserve. In the present study, neither nifedipine nor lisinopril had any effect on the percentage of fractional shortening despite a reduction in LVMI, suggesting that LVH reversal may not modify systolic cardiac performance. As for diastolic left ventricular function estimated from transmitral flow, its A/E ratio is influenced by structural reversal of LVH and functional afterload reduction. Because nifedipine and lisinopril lowered arterial pressure to the same extent in this study, ESS or arterial impedance rather than arterial pressure seems to be a major determinant of afterload, which modifies the A/E ratio in transmitral flow. During the first 24-week treatment in this study, lisinopril decreased the A/E ratio in parallel with a reduction in LVMI, but nifedipine did not. On the other hand, 48 weeks of treatment brought about similar reductions in LVMI and A/E ratio, regardless of whether nifedipine or lisinopril was used. Moreover, in the L-N group, although there was a significant difference in the change in the D/S ratio of diaphragmatic aortic flow (reflecting compliance) between the first lisinopril and second nifedipine treatments, both LVMI and A/E ratio remained unchanged throughout the first and second active treatments. Therefore, one can conclude that the improved diastolic left ventricular filling may be directly due to structural LVH reversal but not to functionally decreased afterload resulting from a reduction in arterial pressure and improved aortic compliance.
Sphygmomanometric recordings of arterial pressure, as used in this study, do not always correspond to intra-aortic pressure. When lisinopril, as well as other ACE inhibitors, increases arterial distensibility (or aortic compliance) and reduces wave reflection, recordings of intra-aortic pressure are expected to reflect a decrease in augmented systolic pressure caused by inappropriately early wave reflection and a reduction in pulse pressure.39 Therefore, our findings of improved aortic compliance in the lisinopril group may result in a lower intra-aortic pulse pressure with comparable stroke volume than in the nifedipine group. However, in this study, the sphygmomanometric recordings of arterial pressure did not differ significantly between the nifedipine and lisinopril groups. Because the reflected wave constitutes the peak of the pressure wave in the aorta but is usually just an undulation on the downstroke of the wave in the radial artery, sphygmomanometric recordings might underestimate a reduction in intra-aortic systolic pressure and pulse pressure brought about by lisinopril.
Second, in relation to the crossover design of this study, we did not include a washout period between the first and second treatment periods. Therefore, we cannot exclude the possibility that the first active treatment might to some extent have influenced the results of the second treatment, in that the effects of the first 8-week active treatment on hemodynamic parameters might have remained somewhat within 4 weeks of the washout period.10 However, hemodynamic changes during a washout period between both active treatments, including elevation of arterial pressure, might have influenced the reversed LVH because it has been reported that left ventricular mass significantly increased after treatment with calcium antagonists or ACE inhibitors had been interrupted for 3 to 4 weeks.40 41
Selected Abbreviations and Acronyms
|Ang II||=||angiotensin II|
|ESS||=||end-systolic wall stress|
|LVH||=||left ventricular hypertrophy|
|LVMI||=||left ventricular mass index|
|MAP||=||mean arterial pressure|
|PWV||=||pulse wave velocity|
- Received January 16, 1996.
- Revision received February 15, 1996.
- Accepted April 19, 1996.
Pfeffer JM, Pfeffer MA, Mirsky I, Braunwald E. Prevention of the development of heart failure and the regression of cardiac hypertrophy by captopril in the spontaneously hypertensive rats. Eur Heart J. 1983;4:143-148.
Levy D, Anderson KM, Savage DD, Kannel WB, Christiansen JC, Castelli WP. Echocardiographically detected left ventricular hypertrophy: prevalence and risk factors: the Framingham Heart Study. Ann Intern Med. 1988;108:7-13.
Frohlich ED. Overview of hemodynamic and nonhemodynamic factors associated with left ventricular hypertrophy. J Mol Cell Cardiol. 1989;21(suppl V):3-10.
Castellano M, Muiesan ML, Rizzoni D, Beschi M, Pasini G, Cinelli A, Salvetti M, Porteri E, Bettoni G, Kreutz R, Lindpaintner K, Rosei EA. Angiotensin-converting enzyme I/D polymorphism and arterial wall thickness in a general population: the Vobarno study. Circulation. 1995;91:2721-2724.
Dzau VJ, Safar ME. Large conduit arteries in hypertension: role of the vascular renin-angiotensin system. Circulation. 1988;77:947-954.
Shimamoto H, Shimamoto Y. Lisinopril improves aortic compliance and renal flow: comparison with nifedipine. Hypertension. 1995;25:327-334.
Takahashi M, Sasayama S, Kawai C, Kotoura H. Contractile performance of the hypertrophied ventricle in patients with systemic hypertension. Circulation. 1980;62:116-126.
Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man: anatomic validation of the method. Circulation. 1977;55:613-618.
Tsuji U, Nakanishi T, Nakai K, Shiomi J, Funahashi S. Liquid-chromatographic measurement of catecholamines in plasma and urine using HLC-8030. Kiki Shiyaku. 1988;11:635-641.
Morimoto T, Aoyama M, Gotoh E, Shionoiri H. A method of radioimmunoassay of plasma angiotensin II using Florisil. Folia Endocrinol Jpn. 1983;59:215-229.
Swynghedauw B, Moalic JM, Delcayre C. The origins of cardiac hypertrophy. In: Swynghedauw B, ed. Research in Cardiac Hypertrophy and Failure. Paris, France: INSERM; 1990:23-50.
Kent RL, Mann DL, Cooper G IV. Signals for cardiac muscle hypertrophy in hypertension. J Cardiovasc Pharmacol. 1991;17(suppl 2):S7-S13.
Linzbach AJ. Heart failure from the point of view of quantitative anatomy. Am J Cardiol. 1969;56:370-382.
Schölkens BA, Linz W, Martorana PA. Experimental cardiovascular benefits of angiotensin-converting enzyme inhibitors: beyond blood pressure reduction. J Cardiovasc Pharmacol. 1991;18(suppl 2):S26-S30.
Baker KM, Aceto JF. Angiotensin II stimulation of protein synthesis and cell growth in chick heart cells. Am J Physiol. 1990;259:H610-H618.
Morgan HE, Baker KM. Cardiac hypertrophy: mechanical, neural and endocrine dependence. Circulation. 1991;83:13-25.
Brilla CG, Janicki JS, Weber KT. Cardioreparative effects of lisinopril in rats with genetic hypertension and left ventricular hypertrophy. Circulation. 1991;83:1771-1779.
Safar ME, Toto-Moukouo JJ, Bouthier JA, Asmar RE, Levenson JA, Simon AC, London GM. Arterial dynamics, cardiac hypertrophy, and antihypertensive treatment. Circulation. 1987;75(suppl I):I-156-I-161.
Simon AC, Safar ME, Levenson JA, London GM, Levy BI, Chau NP. An evaluation of large arteries compliance in man. Am J Physiol. 1979;237:H550-H554.
Kifer I, Dzau VJ. Endothelial renin-angiotensin pathway: evidence for intracellular synthesis and secretion of angiotensins. Circ Res. 1987;60:422-428.
Assad MM, Antonaccio MJ. Vascular wall renin in spontaneously hypertensive rats: potential relevance to hypertension maintenance and antihypertensive effect of captopril. Hypertension. 1982;4:487-493.
Okamura T, Miyazaki M, Inagami T, Toda N. Vascular renin-angiotensin system in two-kidney, one clip hypertensive rats. Hypertension. 1986;8:560-565.
Geisterfer AAT, Owens GK. Hypertrophic response of cultured vascular smooth muscle cells to angiotensin II. Fed Proc. 1986;45:584.
Loudon M, Bing RF, Thurston H, Swales JD. Arterial wall uptake of renal renin and blood pressure control. Hypertension. 1983;5:629-634.
Safar ME, Laurent SL, Bouthier JD. London GM, Mimran AR. Effect of converting enzyme inhibitors on hypertensive large arteries in humans. J Hypertens. 1986;4(suppl 5):S285-S289.
Rorive GL, Carlier PG, Foidart JM. The structural responses of the vascular wall in experimental hypertension. In: Zanchetti A, Tarazi RC, eds. Handbook of Hypertension. Volume 7: Pathophysiology of Hypertension. Cardiovascular Aspects. Amsterdam, Netherlands: Elsevier Scientific Publishers BV; 1986:427-453.
Gardin JM, Drayer JIM, Weber M, Rohan MK, Knoll M, Shu VWC, Garcia R, Brewer D, Henry WL. Doppler echocardiographic assessment of left ventricular systolic and diastolic function in mild hypertension. Hypertension. 1987;9(suppl II):II-90-II-96.
Capasso JM, Strobeck JE, Sonnenblick EH. Myocardial mechanical alterations during gradual onset of long term hypertension in rats. Am J Physiol. 1981;241:H435-H441.
Weber KT, Brilla CG. Pathological hypertrophy and cardiac interstitium: fibrosis and renin-angiotensin-aldosterone system. Circulation. 1991;83:1849-1865.
O'Rourke M. Systolic blood pressure: arterial compliance and early wave reflection, and their modification by antihypertensive therapy. J Hum Hypertens. 1989;3:47-52.
Asmar RG, Journo HJ, Lacolley PJ, Santoni JP, Billaud E, Levy BI, Safar ME. Treatment for one year with perindopril: effect on cardiac mass and arterial compliance in essential hypertension. J Hypertens. 1988;6(suppl 3):S33-S39.