Influence of humoral and neurohormonal factors on cardiovascular hypertrophy in intreated essential hypertensives

AJH

1996; 9:207-22.5

Influence of Humoral and Neurohormonal Factors on Cardiovascular Hypertrophy in Untreated Essential Hvpertensives d _

Anne Pauline Schroeder, Inger Sihm, Birgitte Mm-n, Kristian Thygesen, Erliq Bjerregaard Pedersen, and Ole Lederbnlle

In essential hypertension, cardiovascular structure is believed to be influenced by hormonal and by hemodynamic factors. The objective of the present study was, in essential hypertensives, to investigate the relationship between blood pressure (BP) level as well as circulating hormones on the one hanId and cardiovascular structure on the other. Seventy-nine untreated essential hypertensives were examined by 24-h ambulatory BP monitoring, echocardiography, microscopy of subcutaneous resistance vessels and analyzes of plasma for angiotensin II (P-Ang 111, aldosterone, atria1 natriuretic factor and 24-h urinary excretion of catecholamines. Multiple regression analysis showed a statistically significant correlation between P-Ang II and the end diastolic interventricular septal diameter (IVSDdl (R = 0.32, P = .0051 and a weak correlation between P-Ang II and the left ventricular posterior wall diameter (R = 0.22, P = .049). These correlations were closer in the subgroup of patients (N = 541 who had never

received antihypertensive treatment (R = O-42/ 0.32, respectively). A weak, though statistically significant, correlation was found between the catecholamine excretion and systolic BP (R = 0.26, P = .031. A statistically negative correlation existed between catecholamines and end-diastolic left ventricular internal diameter index (R = -0.36, P = .OOl). No significant relationship was found between hormonal levels and the tunica media structure of the resistance arteries. In conclusion, P-Ang II was in this study significantly correlated to IVSD,+, but not to resistance artery structure. In essential hypertension a complex relationship exists between humoral and hemodynamic factors and cardiovascular remodeling. Am J Hypertens 1996;9:207-215

ssential hypertension is a condition characterized by an imbalance of modulating forces of vasodilation and vasoconstriction, by endothelial dysfunction and cardiovascular

structural remodelling.1*2Whether cardiac and vascular hypertrophy are phenomena basically secondary to elevated blood pressure and volume overload, or primarily due to nonhemodymamic vasoconstrictor or growth factors still remains to be established. The renin-angiotensin-aldosterone system (RAAS) has a crucial role as a blood pressure regulator at a circulating as well as at a local tissue level. 3Y4 Angiotensin II (Ang II) favors vasoconstriction and exerts an important pressor function. In experiments with animal vascular cell culture as well as human

Received December 6,1994. Accepted August 21,199s. From the Aarhus Amtssygehus, Department of Cardiology (IS, BM, KT); Viborg Sygehus, Department of Internal Medicine (AI’S, OL); and Aarhus University, Skejby Sygehus, Research Laboratory for Nephrology and Hypertension (EBP); Aarhus, Denmark. Address correspondence and reprint requests to Anne Pauline Schroeder, Tsndergade 90, IIIth, 8000 Aarhus C, Denmark.

8 1996 by iizz American Joumnl ofHypertension, Ltd. Published hy Elsevim Science, Inc.

KEY WORDS: Essential hypertension, left ventricular hypertrophy, resistance vessel structure, hormonal factors.

0895-7061/96/$15.00 SSDI 0895-7061(9SJOO352-5

208

SCHROEDER ET AL

cultured aortic cells, it has been shown to enhance protein synthesis and DNA synthesis, inducing WSCUlar cell hypertrophy’ and proliferation.‘,6 However, others found that Mng II merely maintains cardiac hypertrophy.’ Aldosterone (Aldo) promotes collagen synthesis by stimulating cardiac fibroblasts. It thus has an effect on the development of myocardial interstitial fibrosis.’ The effects of atrial natriuretic factor +WF) are to induce vasorelaxation, enhance permeability of capillaries, and induce natriuresis. ANF inhibits synthesis of Aldo and limits the action of the RAAS. In addition it is an inhibitor of vascular cell growth.’Augmented levels of ANF have been revealed in heart and renal failure and in atrial tachyarrhythmias. A few studies have also shown higher plasma levels in hypertensives,‘,” whereas this has not been confirmed by others.” Norepinephrine favors vasoconstriction. it has been hypothesized to stimulate the development of cardiac hypertrophy?” In some animal studies it has been shown to stimulate protem synthesis, cell growth, and proliferation of cardiac myocytes,‘3 whereas this was not the case in cardiac muscle cell culture.‘” Studies of patients with pheochromocytoma revealed that most of these patients did not have cardiac hypertrophy.‘5,‘6 The purpose of the present study was to evaluate the relations between the circulating levels of Aldo, Ang II, and ANF, the urinary excretion of catecholamines, and the left ventricular dimensions as determined by echocardiography, as well as the structure of small subcutaneous arteries, in untreated essential hypeitensives. METHODS Patier& Essential hypertensive patients were recruited consecutively from the outpatient clinics at two centers. The inclusion criteria in the study were as follows: essential hypertension, office supine arterial diastolic blood pressure (BP) 2 100 mm Hg on three separate occasions during a 2 month period, or mean daytime diastolic ambulatory BP 2 95 mm Hg; no medical treatment or dysregulated hypertension on previous medication; and age 25 to 65 years. The exclusion criteria were: secondary hypertension, diabetes mellitus, advanced c!ir.ical signs of atherosclerosis, angina pectoris, previous myocardisl infarction, severe hypercholesterolaemia, heart failure (NYHA class II to IV), evidence of valvular heart disease, second or third degree atrioventricular block, sick sinus syndrome, tachyarrhythmias, impaired renal function (serum creatinine > 175 pmol/L), or other chronic disease. All patients gave their informed consent to participate in the study, which complied with the Helsinki declaration and was approved by the regional scientific ethical committees. Study Design The following examinations were performed: routine clinical examination and biochemical

AJH-MARCH

1996VOL.

9, NO. 3

screening, resting eiectrocardiogram (ECG 1, chest xray, and renography. Patients who met the inclusion criteria subseque&y underwent the following examinations (patients who were initally on medical treatment after a 6 to 8 week washout period): echocardiography, 24- to 48-h ambulatory blood pressure monitoring, and measurement of 24-h-urinary-excretion of catecholamines (epinephrine + norepinephrine) and sodium. In the morning on the day of the echocardiographic examination, blood samples were taken under standardized conditions (after an overnight fast, with the patient resting for 30 min) for determining plasma Ang II, plasma ANF, and plasma Aldo. Further, on the same day, subcutaneous biopsies were taken and small artery segments were dissected out and examined on a myograph. Hormone Assays Plasma Ang II, plasma Aldo, and plasma ANF were analyzed at the Research Laboratory for Nephrology and IHypertension, Aarhus University Hospital. Ang II was determined by radioimmunoassay by a modification of the method described by Kappelgaard et a1.17Radioimmunoassay was performed after previous extraction from plasma by Sep-pak C-18 cartridges. The antibody was obtained from the Department of Clinical Physiology, Glostrup Hospital, Denmark. Minimal detection level was 2 pm01/ L plasma. The coefficients of variation were 12% (interassay) and 8% (intraassay). Aldo was measured by a slight modification of a previously described method.” Using a rabbit antialdosterone antibody (via International CJS, from Sorin, Milan, Italy), radioimmunoassay was performed on a residue from plasma prepared by extraction with dichloromethane and purification on silica gel columns. Minimal detection level was 42 pmol/ L. The coefficients of variation were 13% (interassay) and 9% (intraassay). ANF wirs determined by radioimmunoassay as previously described.” ANF was extracted from plasma by means of Sep-pak C-18 cartridges. For radioimmunoassay, rabbit anti-ANF antibody was obtained from the Department of Clinical Chemistry, Bispebjerg Hospital, Copenhagen, Denmark. The minimum detection level was 0.5 pmol/ L plasma. The coefficients of variation were 12% (interassay) and 10% (intraassay). The 24-h urinary excretion of norepinephrine and epinephrine were analyzed either by electrochemical (HPLC) or fluorometric assay after chromatographic purification.2”ZZ’ 24h Ambulatory Blood Pressure Monitoring Ambulatory blood pressure (BP) recordings were performed for 24 to 48 h, preferably on working days. A Takeda Medical (A&D TM-2420; Osaka, Japan) noninvasive ambulatory BP recorder was used.2’*23The BP readings were comparable to values obtained by routine sphygmomanonreter and did not deviate

AJH-MARCH 1996-VOL. 9, NO. 3

HORMONAL

more than 5 mm Hg. The recorder registered BP readings once every 30 min between 7 AM and 11 PM and once every 60 min between 11 phf and 7 AM. BP measurements registered while the patient was still in the clinic and BP values deviating more than 3 SD from the mean were discarded. The mnan 24-h, mean daytime (7 AM to 11 PM) and mean nighttime ( 11 PM to 7 AM) systolic BP (SBP), diastolic BP (DBP), and heart rate (HR) were calculated. Echocardiography The patients were examined by transthoracic echocardiography in the left lateral position., using a 3.5 MHz transducer on a Vingmed CFM 750 (Ho&m, Norway) or an Aloka (Tokyo, Japan) SD 650 echocardiograph. Examinations were performed by the same observers at each center. The left ventricular dimensions at end-diastole-the left ventricular internal diameter (LVIDd ) , the interventricular septal diameter ( I.VSDd), and the left ventricular posterior wall diameter (LPWD,) -were assessed by M-mode echocardiography. The measurements were performed according to the Penn convention (excluding the endocardium from the left ventricular walls, including it in the ventricular internal diameter) .24Further, as for the transmitral flow, in 27 patients the ratio of the early diastolic to atrial systolic flow (E/A ratio) was a measured as the mean of three successive heartbeats. The measurements were carried out blindly by two observers who had to be in agreement. LVIDd was indexed for body surface ( LVIDdI) . Left ventricular mass (J,\iM) and mass index (LVMI) were calculated according to the Penn convention.‘* Left ventricular hypertrophy (LVH) was defined as left ventricular mass index > 135 g/m2 for men, > 111 g/m2 for women.25 Examination of Resistance Arteries Measurements on resistance arteries were carried out at the Department of Pharmacology, Aarhus University. From a subcutaneous gluteal biopsy, taken under local anesthesia with lidocaine, two small artery segments, 1 to 2 mm long and 100 to 300 pm in diameter, were dissected

TABLE 1. BLOOD PRESSURB

Office SBP, mm ijg Office DBP, mm Hg 24-h SBP, mm Hg Day SBP, mm Hg Night SBP, mm Hg 24-h DBP, mm Hg Day DBP, mm Hg Night DE?, mm Hg 24-h HR, beats/min

RfCORDINGS

Mean

SD

Range

170.2 109.4 159.79 164 142.5 104.7 107.6 93 74.1

16.1 8 14.9 15.8 16.5 8.7 9.4 9.6 8.4

140-210 100-141 131.1-198.5 132.9-208.2 102.1-178 82-131 85-134.5 71.5-125 57-100

N = 79. SBP, Sysfolic blood pressure; DBP, diastolic blood pressure; HR, heart rate.

FACTORS AND THE NYPERTENSZVE HEART

209

TABLE 2. ECMOCARDIOGRAPHIC MEASUREMENTS

IVSRd, cm LPWD,, cm LVIDd, cm LVIDJ, cm LVM, g LVMI, g/m2 E/A ratio /N = 26)

Mean

SD

Range

1.4 1.3 5 2.5 323.9 164 1.04

0.2 0.2 0.5 0.3 97 43.9 0.33

l-1.3 0.9-1.8 4-6.S 2-3.1 184.2-833.7 101.2-338.1 0.53-1.89

N = 75. iii’sc;d, i,kirud. k.il.v scpla!dxwder i,l ei&finstc!c; LPWD,,, kjf posfrrior ml: thickrms in rtrd-dinsi&;LVID,,,kjt ve’etlfriclrlnr intcn1nl dinmeterin cad-diasfok; LVlDJ, left zwztrimkr ihwinl dinmeterindexin end-diztole; LVM, Iejf uetrfricnfnrnnw; LVMI, left mii~iczdnr tms indm; E/A r&o, ratio of otitml dimtolic e&y to &in1 ,@orocamponer~t.

out, and mounted as ring preparations in a MulvanyHalpern myograph, by threading them onto two 40 pm steel wires. These wires were attached to a force transducer and a micrometer, respectively. The vessels were kept in standard saline solution. The circumference that the vessels would have had in vivo when relaxed and under a transmural pressure of 100 mm Hg was found (according to the law of Laplace, dp = dT /I, where dp is transmural pressure, dT is tension, and r is the radius, and L = 2~ x r, where L is the circumference). The method has been described earlier.‘” The media thickness and the lumen diameter were measured by light microscope by one observer blindly. The ratio between media thickness and lumen diameter (media/ lumen ratio) and the media cross-sectional area were then calculated. The values from the two small arteries from each person were averaged. Statistical Analysis The data are expressed as mean values t standard deviation (SD). The correlations of the echocardiographic parameters to blood pressure levels or to hormonai levels were evaluated by univariate linear regression analysis. The correlations of the echocardiographic parameters to two or more independent variables were evaluated by stepwise multiple regression analysis with analysis of variance. A P < .05 was considered significant. RESULTS Patient Data Seventy-nine persons entered the study, 15 women and 64 men, median age 47 years, mean age 46.9 2 8. Fifty-four patients (68%) had recently diagnosed hypertension or had never received treatment for hypertension; the rest had recognized hypertension, but were insufficiently treated on their previous antihypertensive medication. The patients had mild to moderate hypertension (one patient severe hypertension), The BP recordings are given in Table 1. Echocardiographic Determination of Left Ventricular Mass The echocardiographic measurements are dis-

210

AJH-h4.4RCH

SCHFOEDER ET AL

TABLE 3. RELATION OF BLOOD PRESSURE LEVEL TO LEFT VENTRICULAR

P

R Office Office 24-h 24-h

SBF DBP SBP DBP

MASS LVMI

LVM

-

2996.VOL. 9, NO. 3

--

R

P

02.9

.0004

0.45

.00004

0.38 0.41 0.44

.0006 .0002 .0005

0.31 3.46 0.39

,005 .00002 .0004

N = 79.

The mrasiied hormones were to some extent intercorrelated. Plasma Ang II was positively correlated to plasma AIdo CR = 0.33, P = .003), and weakly to plasma ANF (R = 0.24, P = 04). This was stiU statistically significant when adjusted for urinary sodium excretion, as an approximation of the average sodium ingestion. No statistically significant correlation was found between plasma ANF and plasma Aldo, nor between plasma Ang II and urinary catecholamine excretion. Table 5 displays the correlations of the left ventricular enddiastolic parameters with the biochemical parameters. A r&Teak, though statistically significant, correlation was found between IVSl& and plasma Ang II (Figure 21, which was unchanged when adjusting for average ambulatory SBP, age, gender, and urinary catecholamine excretion by stepwise multiple regression analyMeasurements of Resistance Arteries In 7 patients, sis. There was a weak, though significant, correlation no arteries suitable for microscopy were present in the between plasma Ang II and LPWDd, hardly significant biopsies. In the remaining 72 patients, the microscopic when adjusting for the above-mentioned variables. measurements showed a media cross-sectional area of There was a weak, statistically significant correlation 12.9 t 5.3 X lo3 pm* (6.1 to 30.11, and a media/lumen between LVM and plasma Ang II, still significant when ratio of 10.1% 5 2.6% (5.9% to 19.5%). As reported adjusted for SBP, age, gender, and catecholamine excreelsewhere, the media /lumen ratio, but not the media tion. However, this correlation turned out to be depencross-sectional area, appeared to be weakly, but sigdent on two patients who had LVM considerably above nificantly, correlated to the 24-h blood pressure leve1.27 the mean. No significant correlations between plasma Hormonal Level and Relation to Left Ventricular Ang II and blood pressure level were found. Mass The biochemical test results are listed in Table 4. A statistically significant correlation was found beplayed in Table 2. According to these calculations, 62 patients (78%) had hypertrophy of the left ventricle. Linear regression analysis established a statistically significant correlaticln between the calculated LVM and LVMI and the office as well as the ambulatory BP mea---------** 1 Figure 1 I. The systolic BP was OU‘ NllL‘llUiTrble \. _.-.__-, somewhat closer correlated to LVMI than the diastolic BP. These correlations were still significant when adjusting for the plasma level of Ang II and urinary catecholamine excretion, age, and gender, applying stepwise multiple regression analysis. In 27 patients, the mitral early tc atria1 (E/A) Bow ratio was analyzed (Table 2). Mitral regmgitation with a high left atria1 pressure was present in one patient whose E /A ratio was discarded.

TABLE 4. BIOCHEMICAL ANALYSES Mean

SD

Range

ReferenceRange

PlasmaAngiotensinII 2.9-64.1 1.2-17.8

10-30 2-5

129.5

42-544.9

42-500

342

161.4

97-1218

<470

(N = 49)

309.7

174.3

88-1193

30-350

(N = 47)

33

25.7

2--106

<120

(pmol/L) Plasma AMF (pmol/L) Plasma Aldnsterone

(N = 79) iN = 75)

21.6 5.3

14.6 2.5

(pmol/L) Urinary Catecholamine

(N = 78)

199.3

(nmol/day) Urinary Norepinephrine

(N = 75)

(nmol/day) Urinary Epinephrine (nmol/day) Urinary Sodium brnol/day)

(N = 74)

152.9 --

68.5

60-488

-

AIiTh4ARCH

HORMONAL FACTORS AND THE HWERTENSFVE

1996%VOL. 9, NO. 3

D

/ L1

. -

Is0

170

150

210

24 - houramb. SBP (mm Hg)

FIGURE 1.

The correlation between LVMI (g/m*)

mined by echocardiography, R = 0.46, I’ = .00002.

and 24-h ambulatory

as deferSBP (mm Hg).

tween the average ambulatory SBP and the total catecholamine excretion (R = 0.26, P = .03 j (Figure 3) (but not as regards the ambulatory DBP). No significant correlations between catecholamine excretion and LVM, wall thickness, or HR were found. We fcund significant correlations between IVSD,, LPWDd, and

Ang II (N = 79) ANF (N = 75) Aldo (N = 78) Catechol (N = 75) Norepi (N = 49)

R P R P R

= = = = =

P =

R = P =

R =

0.32 .005” 0.29 .01* 0.09 .42 0.18 ‘13 0.3

P = .04*

LPWD*

211

norepinephrine excretion, but not -when adjusting for SBP level, however. Statistically significant negative correlations were found between both catecholamine excretion and norepinephrine excretion and LVIR,, as well as LVB&I. These correlations were not dependent on heart rate or urinary sodium excretion and were still significant when adjusted for SBP, gender, and age, as well. Linear univariate regression analysis showed st,%itally significant correlations between plasma ANF and LVM, LVMI, IVSD,, respectively (Table 5). These correlations were still significant when making adjustments for ambulatory SBP, age, gender, and urinary sodium excretion, as an approximation to the average sodium ingestion, by multiple regression analysis. However, if discardmg the two above-mentioned patients, who had considerably larger LVM, due to slightly dilated ventricles, these correlations to the left ventricular parameters were not statistically significant. There was a significant correlation between LV?Dd and plasma AXF (R = 0.23, P = .048), though not between LVIDdI and plasma ANF. No significant cor:elations were found between the mitral E/A-ratio, adjusted for age and heart rate, and plasma ANF, or betwcen plasma AJF and BP level. No significant correlations were found between plasma Aldo and the left ventricular variables (Table 5) or BF 1~~1. Further regression analysis was performed in the 54 patients who had never been on antihypertensive medication, as treatment, even months earlier, might obscure the data. These patients did not have a statistically significant higher BP level or higher LVM. This analysis largely yielded the same conclusions. They only deviated in two regards: for the relation to plasma Ang II, the correlations were definitely closer (eg, IVSDd: R = 0.42, P = .002, LPWDd: R = 0.32, P =

TABLE 5. RELATION OF PLASMA AND URINE HORMONAL IVSDd

HEART

LVM

LEVEL TO LEFT VENTRICULAR STRUCTURE LVMI

LVmj

LVIDJ

R = 0.22 P = .049* R = 0.19

R = 0.28 P = .Ol” R = 0.40

R = 0.19

R = 0.06

R = -0.16

P = .G3

P = .6

P = .17

R = 0.36

P = .lO

P = .0004*

P = .ooz*

R = 0.1

R = 0.22

R = 0.14

R = 0.23 P = .048* R = 0.19

P = .37

P = .Oh

P = .21

P = .G9

R = 0.20

R = -0.09 P = .47 R = -0.02

P = .013’

R = 0.29

R = -0.03 P = .77 R = 0.07

R = -0.31

R = 0.14 P = .23 R=O P = .99 R = -0.36 P = .Om* R = -0.44

P = .04’

P = .66

P = .92

T = .03*

P =

P = .G6

R = -0.29

.@022’

Significmzt at P < 0.5.

??

IVSDd: Inferventricular sepfal diameter in end-dinstole; LP&Q, Ieft posterior wall thick~~essin elzd-diastole; LVM, left ventricular mass; LVMI, f@ ventricular massindex;LVIDd, reft ventricular internal diameter in end-dinstole; LVID,J, left ventricrtlar internal diameter index in end-diasfoie; _bg IIT plasm angiotensin II; ANF, plasma atrial nafrirrretic factor; Aldo, plasma nldosterone; Cntechol, 24-h urinary catecholamine e?netion; Norepi, 24-h urina y norepinephrine excretion.

SCHROEDER

AJH-MARCH 1996-VOL. 9, NO. 3

ET AL

DISCUSSION

_-( y II&I 0

.~Ds.,. .... .... IIIIIIIIII 2i 40

!. Mi

80

P-ANGIOTENSIN II (pmol!L)

As demonstrated m other studies, we were able to point to a significant relationship between ambulatory (and office) BP level and LVM.Z5 However, whether this correlation is owing to simple hemodynamic mechanisms, ie, adaptive changes in the heart due to enhanced pressure and wall stress, still remains disputable. It has been suggested that, in a subgroup of patients with hypertensive left ventricular hypertrophy and subnormal end-systolic wall stress, this inappropriate hypertrophy might not be induced by mere hemodynamic factors, as in patients with normal wall stress, but rather by neurohormones.28 On the other hand, as mentioned, studies of patients with pheochromocytoma have not revealed hypertrophic ventricles.“,‘” It is possible, however, that in essential hypertension the enhanced degree of vasoconstriction at rest, the myocardial hypertrophy, and the change in cellular arrangements in arteriolar walls might be regarded as secondary to a disrupted humoral equilibrium. In the present study we demonstrated correlations between BP level and catecholamine excretion and between end diastolic left ventricular wall thickness and norepinephrine excretion. The latter was not statisti-

2. The correlation between IVSDd (cnzi as determilled by echocardiography and plasma Alzg (pmol/i). R = 0.32, P = .oos.

FIGURE

.02, LVPVI:R = 0.39, P = ‘04, LVMI: R = 0.3, P = .03). Secondly, the negative correlations between LVIDd / LVID,I and catecholamine / norepinephrine excretion were weaker and only statistically significant for the correlation of LVIDdI to total catecholamine excretion (R = -0.3, P = .04). The Relation Between Hormones and Architecture of the Resistance Arteries Regression analysis of the small artery structure on the biochemical parameters showed a weak, though statistically significant, positive correlation between media/lumen ratio and plasma ANF (R = 0.29, P = .02), unchanged when adjusting for urinary sodium excretion, SBP, age and gender, by stepwise multiple regression. However, this correlation turned out to depend on one patient with a very high level of plasma ANF (Table 4). No correlation between media cross-sectional area and plasma ANF was found. No statistically significant correlations were established between mefdia/lumen ratio and plasma Ang II, plasma Aldo, or 24-h urinary catecholamine excretion, and no correlations either between media cross-sectional area and these hormonal factors. In the 54 patients who were never on medication, the correlations were similar to those in the group as a whole.

3

6

*

12

U-CATECHOLAMINES (n-moV24h)

FIGURE 3.

The correlatiolz betweclz 24-h ambulatory SBP ~mnz Hg) and 24-h urilzary excretion of catecholamitzes (nmol/ 24 h). R = 0.26, P = .03.

AIH_MARCH 1996VOL.

9, NO. 3

HORMONAL FACTORS AND THE HYPERTENSIVE HEART

tally significant, however, when adjustment for BP level was made. One other study has shown a correlation between IVSDd and plasma norepinephrine level (yet only in patients with hypertrophy of the left ventricle),” while other papers, examining both essential hypertensives and patients with pheochromocytoma, did not report any correlation between urinary or plasma catecholamine and LVM.‘6,3” An interesting finding was a significantly negative correlation between the catecholamine excretion and the internal dimension of the left ventricle in end diastole. A higher degree of sympathetic tone may possibly go hand in hand not only with a higher degree of vasoconstriction, but also with an augmented tonus of the heart muscle. Such a view gains support from the fact that treatment with P-blockers, like certain calcium entry blockers, as a rule tend to increase LVID-presumably due to the negative inotropic properties of these drugs. Again, comparable findings have not been reported in patients with pheochromocytoma, although a smaller left ventricular end-systolic diameter has been found in patients with pheochromocytoma compared with essential hypertensives.‘6 Ang II has been suggested to be a modulator of cardiac adaptation to elevated BP, either as trophic factor on myocyte growth or by stimulating the release of norepinephrine. On the other hand, the increased protein synthesis, when adding Ang II to cell cultures, is an in vitro observation deprived of the mobulating influences of pressure, nerves, and endothelium.’ Nevertheless, in our study, circuiaung ~ng II appeared to be independently correlated to IVSD, , and possibly to LPWDd as well, and thus is conceivably associated with cardiac growth. Another study with a somewhat smaller number of patients has shown a significant correlation between LPWDd, the relative wall thickness, and Ang II level, but could not either establish any relationship to plasma Aldo.3’ Aldo has been regarded as a contributor to hypertensive myocardial fibrosis in many experimental studies.” There was no correlation of piasma Aldo to LVM in our study, but echocardiography does not permit us to ascertain the degree of fibrosis of a hypertrophied myocardium. One other, similar study did show a remarkably close correlation between plasma Aldo and LVMI.‘” This study comprehended a smaller number of patients with significantly lower BP levels and LVMI values, but for some reason with a much higher average plasma Aldo level. Differences in dietary sodium ingestion might account for the difference. The positive correlation between plasma Ang II and plasma Aldo was expected, as Ang II is known to stimulate the synthesis of Aldo. The weak, though significantly positive, correlation between plasma

213

Ang 11 and plasma ANF is less obvious. ANF could be considered a kind of antagonist ta the RAAS, being natriuretic and vasodilating in contrast to the sodium retentive and vasoconstrictive abilities of the BAAS. Elevated levels of plasma ANF have been reported in low renin forms of hypertension,” and ANF infusion has been shown to cause suppression of plasma renin activity.3’ We found a correlation of the plasma ANF level -with the left ventricu!ar mass and mass index, which on the other hand could be shown to depend partly on extreme observations in two individual patients, in whom signs of ventricular dilation were evident. No crrrelation existed with the internal dimension indexed for body surface. As in a similar study we also could not show any relationship between plasma ANF and BP level.33 An inverse correlation between mitral E/A ratio, which characterizes left ventricular diastolic functi-7n, and plasma A-NE has been reported. That study included normotensives.” In this population of hypertensives we could not find any correlation between mitral E/A ratio and plasma ANF (adjusted for age and heart rate). E/A ratio is a difficult parameter to quantitate, however. It decreases with age and in hypertension, as a result of reduced relaxation of the left ventricle. In more advanced cases of hypertension the E/A ratio might increase f “pseudonormalize”) because of increasing atria1 pressure. Further, the diastolic function depends on other factors, such as viscoelastic Troperties, pericardial restraint, left and right ventric:~!;;’ interaction. It ha: 3een hypothesized that whereas the cardiac hypertaophy might mainly be dependent on hemodynamics, the vasci~br structure might rather depend on humoral growth :aztors.*~“’The present study also analyzed the pu&ble relationship between hormonal level and resistance artery anatomy. An altered vessel structure in&tiding an elevated media/lumen ratio has been reported in hypertensives as cornTared to normotensives.35j6 We only found a weakly positive correlation between the media /lumen rat:3 and plasma ANF, depending on a single elevated value of plasma ANF, and no correlation for plasma Ang II or other hormonal factors. T------I: -~-TT Al‘FlCC” kIU l-“C “ s‘_‘J o~~l~n=*inns J..r-s_.“.__- for this apparent lack of a relationship. The representativity of the examined arteries ought to be subject to speculation. Although of comparable size, they might not be equal as regards the level of the arterial tree, also taking the different body size of the patients into account. However, otherwise this method of analyzing the arterial structure has proved to be reliable. The enhanced media /lumen ratio is probably not due to growth of the muscular cells in the media layer, induced by growth factors. If anything, the increment is due to an encroachment of the muscular cells upon

214

A,~-MARCH

SCHROEDERF.T AL

the lumen3’ and thus an a!tercd arrangemen? of the cells without growth. This structural remodeling might not be dependent on single hormones, but rather on an interaction of hormonal factors, c:~simply reflect hemodynamic load. CONCLUSIONS We found statistically significant correlations between IVSDd, LPWDd, and plasma Ang II, which were closer in the subgroup of patients who had never received antihypertensive treatment. We also noted a weak though significant correlation between the average ambulatory SBP and the total catecholamine excretion, and statistically significant negative correlations between both catechoiamine excretion, norepinephrine excretion, and LVIDd as well as LVIDdI. However, we found no significant correlations between left ventricular echocardiographic variables and plasma Aido, no reliable correlations between plasma ANF and left ventricular dimensions or tunica media structure, and no correlations between tunica media structure and plasma Ang II, plasma Aldo, or catecholamine excretion. A statistically ‘highly significant correlation was found between LVh4 and ambulatory BP as well as office BP level. But we were not able to point to any single known humoral factor appearlrg to be strongly correlated to the left ventricular mass as determined by e&cardiography, still less to the arterial structure. We found a statistically significant correlation between plasma Ang II ard left ventricular wall thickness, but no correlation between plasma Ang II and vascular structure. The study thus underlines the complex relationship of hormonal and hemodynamic factors. Probably no simple relations exist between single hormones and the cardiovascular structural o&anges. ACKNOWLEDGMENTS We are indebted to Michael J. Mulvany and Christian Aalkjar, Department of Pharmacology, Aarhus University, who took care of the examination of resistance arteries. REFERENCES 1. Dzau VJ, Gibbons GH: Endothelium aad growth factors in vascular remodeling of hypertension. Hypertension 1991;18(suppl III):III-115-111-121. 2. Heagerty AM: Angiotensin II: Vasoconstrictor or growth factor? J Cardiovasc Charm 1991;18(suppl 2):s14-~19.

1996VOL.

9, NO. 3

sin II-stimulated protein synthesis in cultured vascular smooth muscle cells. Hypertension 1989;13:305-314. 6.

Campbell-Boswell M, Robertson AL: Effects of angiotensin II and vasopressin on human smooth muscle cells in vitro. Exp Mol Path01 198I;35:265-276.

7. Golornb E, Abassi ZA, Cuda G, et al: Anziotensin II maintains,. but does not mediate, isoprzterenol-induced cardiac hypertrophy in rats. Am J Physiol 1994;267:H1496-H1506. 8. Weber KT, Brilla CG: Pathological hypertrophy and cardiac interstitium. Fibrosis and renin-angiotensin-aldosterone system. Circulation 1991;83:1849-1865. 9. Musca A, Cammarella I, Ferri C, et al. Natriuretic hormones in ,~-ounghypertensives and in young normotensives with or without a family history of hypertension. Am J Hypertens 1992;5:592-599. 10. Des&-Fulgheri P, Paiermo R, Di Noto G, et a!: Wgh -1 0-3 natri-retlc factor and impaired left levds of Y.a,,,., ventricular diastolic function in hypertensives without left ventricular hypertrophy. J Hypertens 1992;10:161165. 11. Genest J, Lerochelle P, Cusson JR, Gutkowska J, Cantin M: Th.e atria1 natriuretic factor in hypertension. Hypertension 1988;il(suppI 1):1-3-I-7. 12. Rossi MA, Carillo SV: Does norepinephrine play a central causative role in the process of cardiac hypertrophy? Am Heart J 1985;109:622-624. 13. Pfeffer MA, Lamas GA, Vaughan DE, et al: Effect of captopril on progressive ventricular dilatation after anterior myccardial infarction. N Engl J Med 1988; 31980-86. 14. Mann DL, Kent KL, Parsons J, Cooper G, IV: Adrenergic effects on the biologv of the adult mammalian cardiocyte. Circulation 1992;85:790-804. 15. Shub C, Cueta-Garcia L. Sheus SG. et al: Echocardiographic’ findings in pheochromocytoma. Am J Cardiol 1986;57:971-975. 16. Fouad-Tarazi FM, Imamura M, Bravo EL, et al: Differences in left ventricular structural and functional changes between pheochromocytoma and essential hypertension. Role of elevated circulating catecholamines. Am J Hypertens 1992;5:134-140. 17. Kappelgaard AM, Nielsen MD, Giese J: Measurement of angiotensin II in human plasma: technical modifications and practical experience. Clin Chem Acta 1976;67:299-306. 18. Rask-Madsen J, Bruunsgaard A, Munck 0, et al: The significance of bile acids and aldosterone for the electrical hyperpolarization of human rectum in obese patients with intestinal by-pass operation. Stand J Gastroenterol 1974;9:417-426.

4. Laragh JH: Role of renin secretion and kidney function in hypertension and attendant heart attack and stroke. Clin Exp Hypertens [A] 1992;A14(1&2):285-305.

19. Thomassen AR, Bagger JP, Nielsen ‘IT, Eedersen EB: Atria1 natriuretic peptide during pacing in controls and patients with coronary arterial disease. Int J Cardiol 1987;17:267-276. 20. Davidson DF, Fitzpatrick J: A simple, optimised and rapid assav for urinarv free catecholamines bv HPLC with electrochemical ‘detection. Ann Clin Biochem 1985;22:297-303.

5. Berk BC, Vekshtein V, Gordon HM, Tsuda T: Angioten-

21. Vendsalu A: Studies on adrenaline and noradrenaline

3. Dzau VJ: Circulating versus local renin-angiotensin system in cardiovascular homeostasis. Circulation 1988;77(supplI):I-4-I-13.

AJH-MARCH

1996VOL.

9, NO. 3

HORMONAL

in human plasma. Acta Physiol Stand 1960;49(suppI i73):23-322. 22.

23.

White WB, Pickering TG, Morganroth J, et a!: A multicenter evaluation of the A&D TM-2420 ambulatory blood pressure recorder. Am J Hype:tens 1991;4:890-896. Clark S, Fowlie S, Coats A, et al: Ambulatory blood pressure monitoring: validation of the accuracy and reliability of the TM-2420 according to the AAMI recommendations. J Human Hypertens 1991;5:77-82.

24.

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

25.

Devereux RB. Pickerinrr TG. Alderman MH, et al: Left ventricular hypertrophYy in hypertension. Prevalence and relationship to pathophysiologic variables. Hypertension iV87;9(suppi iij:ii53-1160.

26.

Heagerty AM, Bund SJ, Aalkjaer C: Effects of drug treatment on human resistance arteriole morphology in essential hypertension: Direct evidence for structural remodel ling of resistance vessels. Lancet 1988;1209?2?2.

27.

Sihm I, Schroeder Al’, Aalkjrer C, et ah The relation between peripheral vascular structure, left ventricular hypertrophy, and ambulatory blood pressure in essential hypertension. Am J Hypertens 1995;8:987-995.

28.

Sugishita Y, Iida K, Yukisada K, Ito I: Cardiac determinants of regression of left ventricular hypertrophy in essential hypertension with antihypertensive treatment. J Am Co11 Cardiol1990;15:665-671.

FACTORS

AND THE HYPERTENSIVE

HEART

215

29.

Corea L, Bentivoglio M, Verdecchia P, -Motclese M: Plasma norepinephrine and left ventricular hypertrophy in systemic hypertension. Am J Cardiol 1984; 53:129?-1303.

3G.

Fauwens FR> Drrnrez DA; DrBuvzere MI,. et at: influence of the arter?al blood press’ure and nonhemodynamic factors on left ventricular hyperhophy in moderate essential hypertension. Am J Cardiol 2992;68:925929.

31.

Schmieder RE, Messerli FH, Garavaglia GE, et al: Does the renin-angiotensin-aldosterone system modify cardiac structure and function in essential hypertensioL:? Am J Med 1988;84(suppl3A):136-139.

32.

France-Saenz R, Somani I’, Mulrow PJ: Effect of atria1 natriuretic peptide 1%33-Met ANP) in patients with hypertension. Am J Hypertens 1992;5266-275.

33.

Ekholm D, Edvinsson L, Ekman T, Thulin 7: l?Iasma .,+-:%I n=*ri..r~+ir “Lll”. ..uc ZU._..._ -errtide ~-r._-- !_AJqP) in r&tjfi~ to blood preswre in severp hypertension. J Intern Med 1992;231:281-285.

34.

Dovle Al? Role of aneiotensin-convertinz en”vme inhibitors in preventincor reducing end-oygan damage in hypertension. J Cardiovasc Pharm 1992;191suppl 5) :s21 -s27.

35.

Sivertsson R: The hemodvnamic imoortance of structural vascular changes inessential hkpertension. Acta Physiol Scand 197O;suppl343:1-56.

36.

Aalkjasr C, Eiskj;er H, h?ulvany MJ, et al. Abnormal structure and function of isolated subcutaneous resistance vessels from essentia! hypertensive patients desyr:te antihypertensive treatment. J Hypertension 1989; 7:305-310.