Changes in Left Ventricular Mass During Treatment with Minoxidil and Cilazapril in Hypertensive Patients with Left Ventricular Hypertrophy

Home , Minoxidil

Changes in left ventricular mass during treatment with minoxidil and cilazapril in hypertensive patients with left ventricular hypertrophy

G Poga´tsa-Murray1, L Varga2, A Varga2,GyA´braha´m1, I Nagy1, T Forster2, M Csana´dy2 and S Sonkodi1 11st and 22nd Departments of Medicine, A Szent-Gyo¨rgyi Medical University, Szeged, Hungary

Attainment of the regression of hypertension-associa- kind of therapy decreased the arterial pressures to a ted left ventricular hypertrophy (LVH) seems to be a similar degree. The 1-year treatment with the cilazapril- desirable goal of blood pressure (BP)-reducing therapy. based regimen resulted in a significantly diminished Since antihypertensive drugs of differing types may LVMI (from a mean ± s.d. of 173 ± 38 to 152 ± 22 g/m2; exhibit markedly different abilities to modulate LVH, we P Ͻ 0.05). On the other hand, the minoxidil-based ther- examined the effects of the angiotensin-converting apy led to a significant increase in LVMI (from 148 ± 19 enzyme inhibitor cilazapril, and the potassium channel to 170 ± 35 g/m2; P Ͻ 0.05). There were no significant activator minoxidil, alone or in combination with each LVMI changes in patients receiving the combined, cilaz- other, on the left ventricular mass (LVM) in patients with april + minoxidil-based treatment (172 ± 34 vs the pre- severe essential hypertension who had LVH detected by treatment 183 ± 54 g/m2). The results confirm that long- echocardiography. All patients received the same base term treatment with cilazapril is effective both in reduc- therapy of bopindolol and guanfacine. After a run-in per- ing BP and in reducing LVM. In spite of yielding a satis- iod, they were treated with: (1) cilazapril (n = 10); (2) factory reduction of BP, minoxidil therapy, even in com- minoxidil, combined with a diuretic (n = 10); or (3) both bination with a diuretic and a ␤-blocker, may lead to an cilazapril and monoxidil (n = 6) for 12 months. The LVM aggravation of pre-existing LVH; this effect of minoxidil index (LVMI; LVM per body surface area) was estimated could be prevented by the simultaneous administration every 3 months by means of echocardiography. Each of cilazapril.

Keywords: left ventricular hypertrophy; angiotensin-converting enzyme inhibitor; cilazapril; minoxidil; vasodilator

Introduction in LVM has been found to be accompanied by approximately a 1.5-fold enhanced relative risk of Left ventricular hypertrophy (LVH), which occurs in cardiovascular disease and by a doubled risk of car- approximately 15–20% of adult hypertensive diovascular mortality.5 1 patients, is not simply an incidental compensatory Regression of hypertension-associated LVH can be phenomenon in response to the increased overload, achieved by the long-term administration of antihy- but is recognized as a potent independent risk factor pertensive agents6–8 and there is preliminary evi- 1–5 for cardiovascular morbidity and mortality. The dence that the reversal of LVH might reduce the risk main cardiovascular events related to LVH in hyper- of cardiovascular events.9,10 Clinical and experi- tensive patients are coronary heart disease, cardiac mental studies suggest that antihypertensive drugs failure, stroke and sudden death. In the Framingham may differ in their potencies to reduce LVH; in this Study, electrocardiographically detected LVH was sense, angiotensin-converting enzyme (ACE) inhibi- found to be associated with a relative risk of cardio- tors seem to be the most powerful drug class, fol- vascular disease as high as for hypercholesterole- lowed by calcium antagonists, ␤-blockers and 2 mia, cigarette smoking and diabetes together. How- diuretics, while vasodilators exhibit less or no effi- ever, anatomical LVH is more sensitively detected cacy.6–8,11,12 Moreover, experimental studies showed by echocardiography than by electrocardiography or that, in spite of its excellent blood pressure (BP)- 3 X-ray examination. Recent echocardiographic stud- lowering potency, minoxidil, a direct vasodilator ies confirmed that the left ventricular mass (LVM) acting via activation of the ATP-sensitive potassium predicts a progressively increasing risk of mortal channels, can even induce or aggravate LVH, pre- 4,5 2 and morbid cardiac events; each 50 g/m increase sumably due to reflexive activation of the sympathetic and renin-angiotensin systems and the volume overload.12–15 In clinical studies, minoxidil in com- Correspondence: Dr Sańdor Sonkodi, 1st Department of Medi- bination with other antihypertensive drugs proved cine, A Szent-Gyo¨ rgyi Medical University, Szeged, P.O.B. 469, H- 6701, Hungary effective in the treatment of severe, refractory hyper- Received 6 March 1996; revised 12 December 1996; accepted 2 tension.16,17 There is only scanty experience on the January 1997 effect of minoxidil on LVM in human hypertensives; Minoxidil, cilazapril and cardiac mass G Poga´tsa-Murray et al 150 some early studies18,19 and one involving an echo- Roche, Basel, Switzerland; n = 10), depending on the cardiographic assessment of LVM20 demonstrated an BP response, besides the bopindolol and guanfacine aggravating effect of minoxidil on LVH, whereas they were already taking. All minoxidil-treated another echocardiographic investigation did not.21 patients were given a diuretic (clopamide, 10 mg The primary aim of the present study, therefore, daily) to counteract fluid retention. When necessary, was to investigate the effect of the long-term admin- the dose of clopamide was increased throughout the istration of minoxidil on echocardiographically study. A third group of patients (n = 6) received both detected LVH in patients with severe hypertension minoxidil (15–30 mg/day) and cilazapril (3–5 in comparison with that of cilazapril, a new, long- mg/day) in addition to the bopindolol+guanfacine acting ACE inhibitor,22 the two drugs also being therapy. The application of clopamide in this group applied as combination therapy. was ceased if possible. During the first month of treatment, the doses of cilazapril and minoxidil Subjects and methods were titrated to achieve the target DBP of less than 100 mm Hg. Thereafter, BP, heart rate and clinical Subjects status were controlled every month. Echocardiogra- phy, electrocardiography and routine blood chemis- All patients (n = 26) selected for the study had sev- try analyses were performed before randomization ere essential hypertension and LVH. Their diastolic and every 3 months during the treament for 12 BP (DBP) levels amounted to 115 mm Hg on three months. separate occasions when they were not on antihypertensive therapy (baseline BP; Table 1). The criteria for LVH, detected by echocardiography, were BP measurements a mean wall thickness (the average of the posterior wall thickness [PWT] and the interventricular sep- Systolic BP (SBP) and DBP were measured after tum thickness [IVST]) у11 mm and an LVM index 15 min of rest in the sitting position, using a stan- (LVMI; LVM per body surface area) у110 g/m2 in dard mercury sphygmomanometer. BP was meas- women and у134 g/m2 in men.23,24 Secondary ured five times on the occasion of each ambulatory causes of hypertension were excluded by standard control between 8 am and 4 pm. The BP data clinical and laboratory tests. Patients with evident reported in this study are the average values heart or renal failure were excluded. Informed con- observed before each echocardiographic examin- sent was obtained from all patients before the study. ation. Mean arterial pressure (MAP) was calculated as (SBP + 2DPB)/3. In patients with unstable BP, automated BP monitoring (24–48 h) was performed Study protocol to help dose optimization. The protocol was planned in a randomized, prospective, open-labelled fashion because of the Echocardiography potentially adverse cardiac effects of minoxidil. As a run-in period, all patients were treated for 4 weeks M-mode echocardiography was performed by using with the ␤-receptor blocker bopindolol (1–2 mg/day) an Ultramark 9 HDI echocardiograph (ATL, USA) and with the centrally acting agent guanfacine (1–2 with a 3-2 MHz transducer. All echocardiograms mg/day). This run-in period was intended to make were taken by the same investigator. Recordings the pre-trial antihypertensive therapy comparable, were made at a paper speed of 50 mm/second. All since those patients with severe hypertension could echocardiograms were videorecorded. The internal not be deprived of drug treatment,25 for ethical dimension of the left ventricle (LVIDd), IVST and reasons. PWT at end-diastole were measured according to the After the 4-week period, patients were randomly recommendations of the American Society of Echo- allocated to treatment with either a dose of 15– 40 cardiography.26 All recordings were read by two mg/day minoxidil (Loniten, Upjohn, Crawley, UK; independent observers who were blind to the n = 10) or up to 5 mg/day cilazapril (Inhibace, patient’s treatment status. All measurements were

Table 1 Baseline clinical characteristics in the three patient groups (checked before the run-in period)

Minoxidil Cilazapril Minoxidil+cilazapril (n = 10) (n = 10) (n = 6)

Female/male 4/6 4/6 2/4 Age (years) 52.3 ± 6.3 47.2 ± 8.2 51.3 ± 5.6 Weight (kg) 96.2 ± 14.9 85.6 ± 10.6 96.5 ± 13.7 Body surface area (m2) 2.09 ± 0.22 1.98 ± 0.19 2.10 ± 0.25 Duration of hypertension (years) 13.6 ± 6.6 12.6 ± 5.1 12.0 ± 3.2 SBP (mm Hg) 204 ± 15 199 ± 15 206 ± 17 DBP (mm Hg) 130 ± 14 123 ± 16 127 ± 15 LVMI (g/m2) 149 ± 23 171 ± 34 179 ± 48

Values are means ± s.d. SBP = systolic blood pressure; DBP = diastolic blood pressure; LVMI = left ventricular mass index. For all parameters, there were no significant differences between the groups (␹2-test or Kruskal-Wallis test). Minoxidil, cilazapril and cardiac mass G Poga´tsa-Murray et al 151 carried out over at least three cardiac cycles and throughout the study (Table 2, Figure 1). There were were averaged. LVM was estimated by the formula no significant differences in BP responses between of Devereux and Reicheck:27 the treatment groups. The average daily doses of the drugs required to maintain the antihypertensive LVM = 1.04 ([LVIDd+PWD+IVST]3−[LVIDd]3)−14g effect are listed in Table 3. and was divided by the body surface area to derive The minoxidil-based treatment resulted in a con- LVMI. Since the mean variability in the LVM tinuous significant decrease in TPR and a tendency measurement was approximately 25 g, changes of to increase in cardiac index (Table 2); at 3 months, у50 g were regarded as a statistically significant this latter reached a statistically significant differ- regression or aggravation of LVH in an individual ence (+15%, P Ͻ 0.05). On the other hand, the cilaz- patient.28 Relative wall thickness (RWT) was calcu- april-based regimen led to a significant decrease in lated as: cardiac output throughout the treatment, while a significant fall in TPR was observed only transiently RWT = 2PWT/LVIDd. (at 3 months; −9%, P Ͻ 0.05). In the minoxidil+ Left ventricular geometry was regarded as concen- cilazapril-treated group, both the cardiac index and tric hypertrophy (RWT Ͼ 0.45 and LVMI Ͼ 125 TPR showed a tendency to decrease, but without g/m2), eccentric hypertrophy (RWT Ͻ 0.45 and reaching a statistically significant difference. LVMI Ͼ 125 g/m2) or concentric remodelling (RWT Ͼ 0.45 and LVMI Ͻ 125 g/m2).4 Left ventricu- LVM and morphology lar end-diastolic and end-systolic volumes, determined by the formula of Teichholz et al,29 were used Two-way repeated measures ANOVA revealed that to estimate stroke volume and cardiac output. Total the effects of different types of therapy on LVMI peripheral resistance (TPR) was calculated from depended on the duration of treatment (P Ͻ 0.001 MAP and the cardiac output index. for the interaction between the kind and the length of therapy). Table 2 lists the echocardiographic vari- Clinical status ables of the three groups before and after treatment. The minoxidil-based therapy led to a continuous Every month the patients were interviewed and increase in LVMI (+14.9% at 12 months, P Ͻ 0.05; examined for cardiovascular complaints, such as Figure 1), in consequence of the increasing tend- chest pain, dyspnea, ankle edema, nycturia or other encies in LVIDd, IVST and PWT. The average side effects. monthly rate of increase in LVMI was 1.83 g/m2. There was a significant positive correlation between Statistics the daily dose of minoxidil and the increase in LVMI in this group (r = 0.865, P = 0.0095). The cilazapril- ± Results are expressed as means s.d. unless other- based treatment resulted in a progressive decrease wise stated. For statistical evaluation, Student’s t- in LVMI (−12.1% at 12 months, P Ͻ 0.05) which was test for paired data, analysis of variance (ANOVA) mainly due to a decreasing tendency in LVIDd. In for repeated measures, Friedman’s repeated meas- the minoxidil+cilazapril-treated group, the calcu- ures ANOVA on ranks or the Kruskal–Wallis test lated LVMI showed a 6.0% decrease, which failed was used, as indicated. Relationships between vari- to reach statistical significance. ables were assessed by the Spearman rank corre- Taking a у50 g variation in LVM as an indicator lation procedure or by stepwise regression analysis, of a statistically significant regression or aggravation as appropriate. Categorical variables were evaluated of LVH, the individual data indicated that minoxi- ␹2 by the -test, the McNemar test or the Fisher exact dil-based therapy led to an aggravation of LVH in test, as indicated. Statistical significance was six patients and failed to cause regression in any. Ͻ defined as P 0.05. The cilazapril-treated group showed the opposite result: there was a significant regression in five, and Results aggravation in none of the patients. In the group receiving combined treatment, regression occurred Comparability of study groups in three, aggravation in one, and changes of р50 g There were no significant differences in baseline in two patients. ␹2-analysis revealed a significant clinical characteristics between the treatment (P Ͻ 0.05) correlation between the proportion of groups as regards gender, age, weight or BP these categorical variables (regression, aggravation (Table 1). The baseline LVMI in the group of patients or insignificant change) and the kind of treatment. treated with the minoxidil-based regimen seemed to The RWT values exhibited only minor variations, be somewhat lower than in the other groups, but this but with the value of 0.45 for RWT as a cut-off cri- difference was not statistically significant either in terion, the changes in mean RWT indicate a tend- comparison with the other two groups (by ANOVA ency to shift from eccentric towards concentric LVH on ranks; Table 1) or in comparison with the cilazap- in both the minoxidil- and the cilazapril-treated ril-treated group (by the Student’s t-test). groups.

BP response Correlation between BP and LVMI In each group there was a signiﬁcant decrease in BP At baseline, LVMI correlated signiﬁcantly with DBP due to the treatment, and a satisfactory BP control (r = 0.440, P = 0.028) and MAP (r = 0.410, P = 0.040), Minoxidil, cilazapril and cardiac mass G Poga´tsa-Murray et al 152 Table 2 Effects of a 12-month antihypertensive therapy based on minoxidil, cilazapril or their combination on hemodynamic and echocardiographic variables

Patient groups and parameters Entry 12 months

Minoxidil (n = 10) SBP (mm Hg) 174 ± 16 158 ± 17* DBP (mm Hg) 108 ± 998±10* MAP (mm Hg) 130 ± 10 118 ± 11* HR (beats/min) 84.0 ± 10.1 85.0 ± 13.9 LVIDd (mm) 52.7 ± 4.7 53.1 ± 4.1 IVST (mm) 12.7 ± 1.6 13.9 ± 2.2 PWT (mm) 11.6 ± 0.9 12.4 ± 1.6 RWT 0.44 ± 0.06 0.47 ± 0.06 LVM (g) 307 ± 44 353 ± 66* LVMI (g/m2) 148 ± 19 170 ± 35* EF (%) 57.0 ± 8.5 58.8 ± 4.4 Cardiac index (L/min/m2) 3.53 ± 6.4 3.95 ± 8.4 TPR (arbitrary units) 38.1 ± 8.0 31.2 ± 7.7*

Cilazapril (n = 10) SBP (mm Hg) 174 ± 13 150 ± 15* DBP (mm Hg) 112 ± 12 101 ± 14* MAP (mm Hg) 132 ± 11 117 ± 13* HR (beats/min) 78.9 ± 13.3 70.5 ± 9.2* LVIDd (mm) 54.5 ± 7.9 50.8 ± 6.0 IVST (mm) 13.2 ± 1.6 13.2 ± 1.9 PWT (mm) 11.1 ± 3.7 11.1 ± 3.8 RWT 0.44 ± 0.09 0.48 ± 0.10 LVM (g) 348 ± 108 302 ± 63* LVMI (g/m2) 173 ± 38 152 ± 22* EF (%) 64.9 ± 11.1 67.8 ± 10.1 Cardiac index (L/min/m2) 3.48 ± 0.58 2.94 ± 0.55* TPR (arbitrary units) 39.3 ± 6.3 39.7 ± 7.1

Minoxidil + cilazapril (n = 6) SBP (mm Hg) 170 ± 17 156 ± 13* DBP (mm Hg) 106 ± 11 96 ± 10* MAP (mm Hg) 127 ± 13 116 ± 11* HR (beats/min) 78.0 ± 15.2 72.2 ± 10.8 LVIDd (mm) 55.0 ± 4.2 53.6 ± 4.4 IVST (mm) 14.1 ± 3.2 14.4 ± 2.9 PWT (mm) 12.4 ± 2.0 12.0 ± 1.2 RWT 0.45 ± 0.10 0.45 ± 0.07 LVM (g) 375 ± 88 356 ± 64 LVMI (g/m2) 183 ± 54 172 ± 34 EF (%) 64.4 ± 7.8 67.2 ± 8.1 Cardiac index (L/min/m2) 3.54 ± 0.35 3.30 ± 0.89 TPR (arbitrary units) 37.1 ± 7.3 36.8 ± 10.2

Values are means ± s.d. *P Ͻ 0.05 vs the entry value, by Student’s t-test for paired data. SBP = systolic blood pressure; DBP = diastolic blood pressure; MAP = mean arterial pressure; HR = heart rate; LVIDd = left ventricular internal diastolic diameter; IVST = interventricular septum thickness; PWT = posterior wall thickness; RWT = relative wall thickness; LVM = left ventricular mass; LVMI = left ventricular mass index; EF = ejection fraction; TPR = total peripheral resistance.

but not with SBP (r = 0.047, P = 0.816), as calculated Clinical symptoms and metabolic effects for the data on all the 26 patients by Spearman rank correlation analysis. Such correlations between BP There were no major cardiovascular events and LVMI were found neither during the ‘run-in’ (myocardial infarction, pulmonary edema, stroke, period (under the bopindolol+guanfacine therapy) etc) or death in any cases. Chest pain occurred at nor during the 12-month study period. The univari- rest or during exercise, but dyspnea only during ate analysis of the changes in LVMI in relation to exercise. At the beginning of the study, ankle edema the changes in BP parameters after 12 months of occurred only in the evening, but in some patients treatment revealed no statistically signiﬁcant corre- it also appeared in the daytime during the treatment lations. Stepwise multiple regression analysis with with minoxidil. In three of the minoxidil-treated changes in LVMI as dependent, and the mode of patients, the dose of diuretics had to be increased therapy and changes in MAP, SBP or DBP as inde- during the study in order to control volume reten- pendent variables, showed that the kind of therapy tion. In the minoxidil group, the incidence of ankle was the only signiﬁcant explanatory variable with edema was greater (6/6) among those taking a dose an F value of 5.13 (P Ͻ 0.05) for the change in LVMI. of minoxidil greater than 20 mg/day than in patients Minoxidil, cilazapril and cardiac mass G Poga´tsa-Murray et al 153

Figure 1 Changes in mean arterial pressure (MAP) and left ventricular mass index (LVMI) with time in the three treatment groups. Values are means ± s.e.m. *P Ͻ 0.05 vs the entry (‘0’-month) value (ANOVA for repeated measures, followed by the Bonferroni t-test or Friedman’s repeated measures ANOVA on ranks, followed by Dunn’s test).

Table 3 Average daily doses (mg) of the drugs used to maintain echocardiographic study20 that demonstrated an the antihypertensive effect increase in LVM on chronic administration of + minoxidil to human hypertensives. Some early Minoxidil Cilazapril Minoxidil cilazapril 18,19 (n = 10) (n = 10) (n = 6) human observations relating to minoxidil indicated an aggravating effect of the chronic adminis- Minoxidil 27.5 ± 12.0 — 21.7 ± 6.1 tration of minoxidil on LVH; however, the evalu- Cilazapril — 4.7 ± 1.9 3.7 ± 1.0 ation of cardiac hypertrophy in those studies was Bopindolol 1.5 ± 0.6 1.5 ± 0.5 1.8 ± 0.6 not based on echocardiographic examinations. In Clopamide 14.0 ± 4.7 — 8.3 ± 6.9 ± ± ± only two trials were the effects of minoxidil treat- Guanfacine 1.4 0.6 1.9 0.6 1.7 0.5 ment on LVH in hypertensive patients assessed by means of echocardiography.20,21 Chrysant et al20 Values are means ± s.d. treated 23 hypertensive patients with a minoxidil+furosemide+␤-blocker combination for taking less than 20 mg daily (1/4; P Ͻ 0.05 by the 24 months and found highly significant increases in 21 Fisher exact test). In the minoxidil+cilazapril group, LVM. On the other hand, Julien et al administered only three of the patients required diuretic. The a similar minoxidil-based therapy to 14 patients for McNemar test indicated that there were no signifi- a 6-month period, but observed only insignificant cant differences between the three groups in the changes in LVMI, and concluded that the minoxidil- incidence of clinical symptoms (chest pain, dyspnea based therapy failed to reduce LVH. In our study, or nycturia) during the treatment, but the symptoms LVMI increased continuously during the minoxidil- tended to worsen in the minoxidil-treated group, based therapy, a statistically significant difference in and to diminish in the cilazapril-treated group. No LVMI being attained by the end of the 9th month. significant metabolic side effects such as changes in In fact, as calculated from their data, Julien et al21 serum lipids, glucose, blood urea nitrogen, or creati- observed a similar monthly rate of increase in LVMI nine were noted. Minoxidil did not cause hypertri- (about 1.5 g/m2), but this had not reached statistical chosis with a severity which would have necessi- significance by the end of their 6-month study. As tated the cessation of treatment, either in men or pointed out by Devereux and Dahlo¨f,30 the length of in women. the treatment period for the determination of agent- dependent changes in LVMI should be at least 1 year. Accordingly, we consider that the results of Dicussion Julien et al21 may also indicate an LVH-aggravating The main findings of this study on hypertensive effect of minoxidil. patients with LVH are that the minoxidil-based ther- Additionally, we found a significant positive cor- apy led to an increase, and the cilazapril-based regi- relation between the daily dose of minoxidil and the men to a decrease in LVM, while treatment with a change in LVMI in the minoxidil-treated group, combination of these drugs did not change LVM, in which suggests a dose-dependent LVH-aggravating spite of a comparable BP reduction being achieved effect of minoxidil. A similar dose-dependence was in each group. observed as concerns the development of ankle This study confirms the result of the only earlier edema. Earlier studies reported the occurrence of Minoxidil, cilazapril and cardiac mass G Poga´tsa-Murray et al 154 ankle edema during minoxidil treatment, even in pared to the entry status, the minoxidil-based ther- combination with diuretics.18–21 A worsening of apy led to a more severe concentric LVH. On the edema formation has also been observed in patients other hand, the cilazapril-based regimen resulted in with chronic heart failure during long term minoxi- a less severe concentric LVH, with a shift towards dil treatment, in spite of advantageous changes in concentric remodelling. Previously published some hemodynamic parameters (increased cardiac data4,5 suggest that these changes in cardiac mass index and decreased systemic vascular resistance) and anatomy may predict a possibly improved prog- being achieved by acute application of the drug.31–33 nosis in the cilazapril-treated group, in contrast with Only a few human studies have been performed a possibly worse prognosis for the minoxidil- in which the effects of cilazapril treatment on LVH treated group. were examined in hypertensives. Only three of the Minoxidil, however, may be useful for the treat- human studies lasted for 12 months; in one of ment of patients with resistant hypertension and them,34 nine patients with severe hypertension who renal involvement, eg, nephrosclerosis,16,17 who were on cilazapril therapy displayed an average 43 g often have LVH and in whom ACE inhibitor treat- decrease in LVM, which is comparable with the 48 g ment alone could worsen the renal function.48 We decrease in LVM observed in the present study. The therefore combined minoxidil with cilazapril in other long-term studies35,36 also demonstrated the order to examine whether this ACE inhibitor could potency of this drug in reversing LVH. prevent the LVM-increasing effect of minoxidil. To Elevated arterial BP is usually considered to be a our knowledge, this is the first study which demon- principal stimulus for hypertensive cardiac strates that administration of a long-acting, potent hypertrophy, but in fact only relatively weak corre- ACE-inhibitor may overcome this adverse effect of lations were found to exist between clinical BPs and minoxidil in human hypertensives. LVM in previous studies,37–42 as in the present one. In animal experiments, chronic administration of There are a number of haemodynamic and non- minoxidil leads to an eccentric LVH, while in com- haemodynamic factors which also influence the bination with a diuretic it results in concentric development of LVH, such as the elevated wall LVH,13,41,49 as in the present study. Also during cila- stress caused by a pressure or volume overload, gen- zapril treatment, the left ventricular anatomy der, race, obesity, alcohol and sodium intake, blood showed a change from eccentric to concentric LVH, viscosity, catecholamines, angiotensin II, or growth although no diuretics were used in this group. It hormone.2,43 As concerns the effect of BP reduction could be suggested that the inhibition of the renin- by different types of therapy on changes in LVM, angiotensin-aldosterone system by cilazapril may stepwise regression analysis showed that the type of contribute to the regression of volume overload- therapy and not changes in arterial pressures con- induced LVH by decreasing the intravascular vol- tributed significantly to explaining the change in ume. Furthermore, angiotensin II generated in the cardiac mass, as in a previous study comparing the heart seems to be an important mediator of the effects of enalapril and hydrochlorothiazide on hypertrophic response to a cardiac overload, and LVM.44 thus, the ability of ACE inhibitors to reverse LVH The simplification that LVH in essential hyperten- may also be related to their inhibitory effect on car- sion is a consequence of an elevated BP and diac tissue ACE.50,51 increased afterload may lead to the suggestion that We are aware of the limitations of this study. reduction of the afterload with a direct vasodilator Nevertheless, these results, taken together with drug, eg, minoxidil, could reverse LVH. In contrast those reported by others,20,34–36 empower us to draw with this, as demonstrated by numerous experi- conclusions as follows. The ACE inhibitor cilazapril mental studies, the chronic administration of is effective in reversing LVH in hypertensive minoxidil, by leading to stimulation of the sympath- patients. Minoxidil should be reserved for the treat- etic and renin-angiotensin systems and volume ment of patients with severe, therapy-resistant overload, is able to induce LVH in normotensive hypertension and should not be employed in animals or to aggravate pre-existing LVH, with no patients with overt cardiac hypertrophy or weak- improvement in survival.12–15 Our results showing a ness. In order to prevent the LVH- and volume over- decrease in TPR with increasing tendencies in car- load-inducing effects of minoxidil, it should be diac output and edema formation are also suggestive applied in the smallest effective dose (15–20 mg of the development of a volume overload during daily) and combined not only with the usually minoxidil treatment. Other vasodilator drugs such administered diuretics and ␤-blockers, but also with as hydralazine and prazosine, which are less effec- an effective ACE inhibitor with properties like those tive in decreasing TPR and in increasing renin of cilazapril, under careful control, including an release and the intravascular volume, do not aggra- echocardiographic follow-up. vate LVM to such an extent as minoxidil.28,45,46 Earlier studies demonstrated that the geometric Acknowledgement pattern of the left ventricle could also be predictive This work was supported by the National Scientific of the incidence of cardiovascular complications;4,47 Research Fund of Hungary (Grant No. T 017484). the event rates were highest in patients with concentric hypertrophy, lowest in patients with concentric References remodelling and intermediate in patients with 1 Levy D et al. Echocardiographically detected left ven- eccentric hypertrophy.4 The evaluation of the tricular hypertrophy: prevalence and risk factors: the changes in LVM and geometry showed that, as com- Framingham Study. Ann Intern Med 1988; 108: 7–13. Minoxidil, cilazapril and cardiac mass G Poga´tsa-Murray et al 155 2 Kannel WB. Left ventricular hypertrophy as a risk fac- patients: a 6-month double blind comparison. JAm tor: the Framingham experience. J Hypertens 1991; 9 Coll Cardiol 1990; 16: 137–142. (Suppl 2): S3–S9. 22 Szu¨ cs T. Cilazapril. A review. Drugs 1991; 41 (Suppl 3 Levy D et al. Determinants of sensitivity and speci- 1): 18–24. ficity of ECG criteria for left ventricular hypertrophy. 23 Devereux RB et al. Standardization of M-mode echo- Circulation 1990; 81: 815–820. cardiographic left ventricular anatomic measurements. 4 Koren MJ et al. Relation of left ventricular mass and J Am Coll Cardiol 1984; 4: 1222–1230. geometry to morbidity and mortality in men and 24 Devereux RB et al. Echocardiographic detection of women with essential hypertension. Ann Intern Med pressure-overload left ventricular hypertrophy: effect 1991; 114: 345–352. of criteria and patient population. J Clin Hypertens 5 Levy D et al. Prognostic implications of echocardio- 1987; 3: 66–78. graphically determined left ventricular mass in the 25 Fragerberg B et al. Withdrawal of antihypertensive Framingham Heart Study. N Engl J Med 1990; 322: drug treatment: time course for redevelopment of 1561–1566. hypertension and effects upon left ventricular mass. J 6 Schmieder RE, Martus P, Klingbeil A, Reversal of left Hypertens 1992; 10: 587–593. ventricular hypertrophy in essential hypertension. A 26 Sahn DJ, DeMaria A, Kisslo J, Weyman A. Recommen- meta-analysis of randomized double-blind studies. dations regarding quantitation in M-mode echocardi- JAMA 1996; 275: 1507–1513. ography: results of a survey of echocardiographic 7 Liebson PR. Clinical studies of drug reversal of hyper- measurements. Circulation 1978; 58: 1072–1083. tensive left ventricular hypertrophy. Am J Hypertens 27 Devereux RD, Reicheck N. Echocardiographic determi- 1990; 3: 512–517. nation of left ventricular mass in man: anatomic vali- dation of the method. Circulation 1977; 55: 613–618. 8 Dahlo¨f B, Pennert K, Hansson L. Reversal of left ventricular hypertrophy in hypertensive patients. A meta- 28 Devereux RB, Pickering TG, Cody RJ, Laragh JH. analysis of 109 treatment studies. Am J Hypertens Relation of renin-angiotensin system activity to left ventricular hypertrophy and function in experimental 1992; 7: 95–110. and human hypertension. J Clin Hypertens 1987; 3: 9 Koren MJ, Ulin RJ, Laragh JH, Devereux RB. Reduction 87–103. in left ventricular mass during treatment of essential 29 Teichholz LE, Kreulen T, Herman MV, Gorlin R. Prob- hypertension is associated with improved prognosis. lems in echocardiographic volume determinations: Am J Hypertens 1991; 4: 1A. echocardiographic-angiographic correlations in the 10 Levy D et al. Prognostic implications of baseline elec- presence or absence of asynergy. Am J Cardiol 1976; trocardiographic features and their serial changes in 37: 7–11. subjects with left ventricular hypertrophy. Circulation 30 Devereux RB, Dahlo¨f B. Criteria for an informative trial 1994; 90: 1786–1793. of left ventricular hypertrophy regression. J Hum Hyp- 11 Cruickshank JM, Lewis J, Moore V, Dodd C. Reversi- ertens 1994; 8: 735–739. bility of LVH by differing types of antihypertensive 31 Franciosa JA, Cohn JN. Effects of minoxidil on hemo- therapy. J Hum Hypertens 1992; 6: 85–90. dynamics in patients with congestive heart failure. Cir- 12 Knorr A, Garthoff B, Kazda S, Stasch J. Effects of differ- culation 1981; 63: 652–657. ent antihypertensive drug classes on survival in ani- 32 Franciosa JA, Jordan RA, Wilen MM, Ledy CL. Minoxi- mal models. J Cardiovasc Pharmacol 1991; 17 (Suppl dil in patients with chronic left ventricular failure: 2): S94–S100. contrasting hemodynamic effects in a controlled trial. 13 Tsoporis J, Fields N, Lee RM, Leenen FH. Effects of the Circulation 1984; 70: 63–68. arterial vasodilator minoxidil on cardiovascular struc- 33 Nathan M, Rubin SA, Siemienczuk D, Swan HJ. Effects ture and sympathetic activity in spontaneously hyper- of acute and chronic minoxidil administration on rest tensive rats. J Hypertens 1993; 11: 1337–1345. and exercise hemodynamics and clinical status in 14 Ruzicka M, Yuan B, Leenan FH. Effects of enalapril patients with severe, chronic heart failure. Am J Car- versus losartan of regression on volume overload- diol 1982; 50: 960–966. induced cardiac hypertrophy in rats. Circulation 1994; 34 Sanchez RA et al. Long-term evaluation of cilazapril 90: 484–491. in severe hypertension. Assessment of left ventricular 15 Moravec CS et al. Structural and functional conse- and renal function. Am J Med 1989; 87 (6B): 56S–60S. quences of minoxidil-induced cardiac hypertrophy. J 35 Haberbosch W et al. Improvement of diastolic filling Pharmacol Exp Ther 1994; 269: 290–296. in hypertensive patients treated with cilazapril. J Car- 16 Andreucci VE et al. Long-term follow-up of minoxidil diovasc Pharmacol 1991; 17 (Suppl 2): S159–S162. therapy in refractory hypertension. A prospective trial 36 De Simone R et al. Effect of cilazapril therapy on dias- in patients with various degrees of renal insufficiency. tolic filling and left ventricular hypertrophy in Clin Nephrol 1983; 19: 55–60. patients with arterial hypertension. Minerva Cardioan- 17 Toto RD, Mitchell HC, Pettinger WA. Treatment of giol 1993; 41: 205–209. hypertension in nondiabetic renal disease. Curr Opin 37 Abi-Samra F, Fouad FM, Tarazi RC. Determinants of Nephrol Hypertens 1994; 3: 279–285. left ventricular hypertrophy and function in hyperten- 18 Wilburn RL, Blaufuss A, Bennett CM. Long-term treat- sive patients. An echocardiographic study. Am J Med ment of severe hypertension with minoxidil, propran- 1983; 75: 26–33. olol and furosemide. Circulation 1975; 52: 706–713. 38 Devereux RB et al. Left ventricular hypertrophy in 19 Mitchell HC, Pettinger WA. Long-term treatment of patients with hypertension: importance of blood press- refractory hypertensive patients with minoxidil. JAMA ure response to regularly recurring stress. Circulation 1978; 239: 2131–2138. 1983; 68: 470–476. 20 Chrysant SG, Chrysant C, Sadeghi M, Berlin L. Cardiac 39 Hammond IW, Devereux RB, Alderman MH, Laragh changes from beta-blocker, diuretic and minoxidil JH. Relation of blood pressure and body build to left combination in hypertension. Cardiology 1991; 78: ventricular mass in normotensive and hypertensive 45–52. employed adults. J Am Coll Cardiol 1988; 12: 996– 21 Julien J et al. Effects of captopril and minoxidil on left 1004. ventricular hypertrophy in resistant hypertensive 40 Lauer MS, Anderson KM, Levy D. Influence of contem- Minoxidil, cilazapril and cardiac mass G Poga´tsa-Murray et al 156 porary versus 30-year blood pressure levels on left 46 Leenen FHH et al. Vasodilators and regression of left ventricular mass and geometry: the Framingham Heart ventricular hypertrophy. Hydralazine versus prazosin Study. J Am Coll Cardiol 1991; 18: 1287–1294. in hypertensive humans. Am J Med 1987; 82: 969–978. 41 Prisant LM, Carr AA. Ambulatory blood pressure 47 Verdecchia P et al. Adverse prognostic significance of monitoring and echocardiographic left ventricular concentric remodeling of the left ventricle in hyperten- wall thickness and mass. Am J Hypertens 1990; 3: sive patients with normal left ventricular mass. JAm 81–89. Coll Cardiol 1995; 25: 871–878. 42 Grossman E et al. Left ventricular mass in hyperten- 48 Cooke HM, De Besse A. Angiotensin converting sion: correlation with casual, exercise and ambulatory enzyme inhibitor-induced renal dysfunction: rec- blood pressure. J Hum Hypertens 1994; 10: 741–746. ommendations for prevention. Int J Clin Pharmacol 43 Frohlich ED. Pathophysiological considerations in left Ther 1994; 32: 65–70. ventricular hypertrophy. J Clin Hypertens 1987; 3: 49 Tsoporis J, Fields N, Lee RM, Leenen FH. Arterial 54–65. vasodilation and cardiovascular structural changes in 44 Dahlo¨f B, Hansson L. Regression of left ventricular normotensive rats. Am J Physiol 1991; 260: H1944– hypertrophy in previously untreated essential hyper- H1952. tension: different effects of enalapril and hydrochloro- 50 Dahlo¨f B. Factors involved in the pathogenesis of thiazide. J Hypertens 1992; 10: 1513–1524. hypertensive cardiovascular hypertrophy. A review. 45 Ruskoaho H. Prevention of cardiac hypertrophy by Drugs 1988; 35 (Suppl 5): 6–26. long-term treatment with isosorbide dinitrate and pra- 51 Johnston CI. Tissue angiotensin-converting enzyme in zosin but not by minoxidil in spontaneously hypertensive rats. Acta Pharmacol Toxicol (Copenh) 1984; 54: cardiac and vascular hypertrophy, repair, and 154–157. remodeling. Hypertension 1994; 23: 258–268.