Myocardial Oxygen Consumption Is Unchanged but Efficiency Is Reduced in Patients With Essential Hypertension and Left Ventricular Hypertrophy
Background—Patients with hypertension and left ventricular hypertrophy (LVH) are prone to develop heart failure. We tested the hypothesis that compensatory LVH is associated with normalization of myocardial oxygen consumption and that this occurs at the expense of a decrease in the ratio between cardiac work and oxygen consumption (efficiency).
Methods and Results—Nine hypertensive men with LVH (LVH+) (age 42±2 years), left ventricular mass index (LVMI) 161±8 g/m2, blood pressure (BP) 145±16/88±10 mm Hg (mean±SD); 8 hypertensive men without LVH (LVH−) (age 39±5 years, LVMI 107±15 g/m2, BP 140±15/90±11 mm Hg); and 10 normotensive men (CONT) were studied. Myocardial blood flow, oxygen consumption, and glucose uptake were measured during euglycemic hyperinsulinemia using PET techniques. LV dimensions, volumes, and workload were determined by echocardiography, and efficiency was calculated. Myocardial workload (2.5±0.8 versus 3.0±0.6 versus 2.3±0.5 mm Hg · mL · min−1 · g−1 for CONT versus LVH− versus LVH+; P<0.05, LVH− versus LVH+), myocardial blood flow (0.84±0.16 versus 1.06±0.22 versus 0.81±0.09 mL · g−1 · min, respectively; P<0.05, LVH− versus other groups) and oxygen consumption (0.09±0.02 versus 0.14±0.03 versus 0.11±0.01 ml · g−1 · min−1, respectively; P<0.05, LVH− versus other groups) were increased in the LVH− group. Myocardial efficiency was reduced in the LVH+ group (18.1±4.1% versus 15.1±2.3% versus 13.5±1.9%, respectively; P<0.05, LVH+ versus CONT).
Conclusions—Myocardial oxygen consumption per unit weight is increased in hypertensive patients without LVH but is normal in those with LVH. The normalization of oxygen consumption via hypertrophy occurs at the expense of efficiency, which may predispose hypertensive patients with LVH to heart failure.
Patients with hypertension and left ventricular hypertrophy (LVH) have a 10-fold greater likelihood of developing heart failure than those with hypertension and a normal LV mass.1 LVH is a structural adaptive mechanism to increased workload, whereas LV failure is the end stage of hypertensive heart disease, resulting from an inability of the heart to adapt to a chronically increased pressure overload.1 The mechanisms whereby hypertension causes cardiac failure are, however, poorly understood.
In LV failure, myocardial efficiency, ie, the ratio of cardiac work to oxygen consumption, is decreased,2 but there are no data of efficiency during various stages of human hypertension. In hypertensive patients with and without LVH, cardiac workload is increased compared with normotensive subjects.3 Data on cardiac oxygen consumption in patients with essential hypertension are sparse and difficult to interpret. Oxygen consumption measurements have been performed in hypertensive patients in whom angiography has been clinically indicated because of symptoms of coronary heart disease.4 5 6 7 Previous studies have not distinguished between patients with and without LVH, but in symptomatic patients with LVH, oxygen consumption per unit weight has been shown to decrease as the LV mass to volume ratio increases.8 On the other hand, the initial response to an increase in afterload, both in animal models of hypertension9 and in response to cardiac hyperfunction induced by exercise and atrial pacing,10 is an increase in oxygen consumption. Thus, at a stage when blood pressure is elevated but LV size is still normal, the ratio between cardiac work and oxygen consumption (efficiency) may not differ from that in normal subjects. Compensatory hypertrophy could then be expected to normalize oxygen consumption per weight unit, but this could occur at the expense of efficiency. These data thus raise the possibility that the natural course of hypertensive heart disease is a gradual decrease in myocardial efficiency.
PET techniques allow quantification of myocardial perfusion and oxygen consumption in healthy individuals and asymptomatic hypertensive patients without cardiac catheterization. We combined PET and echocardiography to determine cardiac efficiency under standardized metabolic conditions in hypertensive patients with and without LVH, and in a group of matched normotensive subjects.
A total of 27 men were studied: 9 had essential hypertension and left ventricular hypertrophy (LVH+), 8 essential hypertension and normal LV mass (LVH−), and 10 were normotensive with normal LV mass (CONT) (Tables 1⇓ and 2⇓). The hypertensive men were recruited from occupational outpatient clinics and the Research and Development Center in Turku. The inclusion criteria for the hypertensive subjects were: age 18 to 50 years, previously diagnosed hypertension, normoglycemia, no signs or symptoms of other cardiovascular disease, and no regular use of any drugs other than antihypertensive drugs. The hypertensive subjects were divided into LVH− and LVH+ groups using 131 g/m2 as an upper limit of normal LV mass index.11 A complete medical examination was performed to exclude secondary forms of hypertension. The normal subjects were recruited among normotensive and normoglycemic males who used no medications and had no signs or symptoms of any disease. All study subjects had normal exercise echocardiograms and laboratory tests. To minimize the direct effects of high blood pressure on the results, the hypertensive subjects were studied while on metabolically neutral antihypertensive treatment (angiotensin-converting enzyme inhibitors and/or calcium channel blockers). The drugs were taken the morning of the PET study.
The nature, purpose, and potential risks of the study were explained to all subjects before they gave their voluntary consent to participate. The study was approved by the Ethical Committee of the Turku University Central Hospital.
The PET studies were performed after a 10- to 12-hour overnight fast starting between 7:30 and 8 am. Two catheters were inserted, 1 in an antecubital vein for all infusions and another in a contralateral hand vein for sampling of arterialized venous blood.12 Each study consisted of a 20-minute basal and a 140-minute hyperinsulinemic period. At 0 minutes, a primed continuous insulin infusion (1 mU · kg−1 · min−1) was started. Normoglycemia was maintained using infusion of 20% glucose. After 30 minutes of hyperinsulinemia, blood volume was quantified using [15O]CO inhalation. Thereafter, myocardial blood flow and oxygen consumption were measured using [15O]H2O infusion and [15O] inhalation. At 100 minutes, [18F]FDG was injected and dynamic PET imaging was performed to determine the myocardial glucose uptake. Blood pressure and heart rate were monitored with an automatic oscillometric blood pressure monitor (HEM-705C, oscillometric blood pressure monitor, Omron Corp) every 15 minutes during the PET study.
Measurement of Myocardial Blood Flow, Oxygen Consumption, Oxygen Extraction Fraction, and Glucose Uptake
[15O] tracers and [18F]FDG were produced as previously described. 12,13 An 8-ring ECAT 931/08–12 tomograph (Siemens/CTI Inc) was used. [15O]CO, [15O]H2O,13 [15O] steady-state14 and [18F]FDG12 studies were performed as previously described. All data were corrected for dead-time, decay, and measured photon attenuation. The [18F]FDG data were reconstructed into a 128×128 matrix using filtered back projection reconstruction method. The final in-plane resolution in filtered back projection–reconstructed and Hann-filtered (0.3 cycles/s) images was 9.5 mm full-width half-maximum. The data of [15O] studies was reconstructed with a new iterative reconstruction algorithm using median root prior.15
Regions of interest were drawn in the lateral, anterior and septal wall of the left ventricle in 4 representative transaxial slices in each study. Additionally a large horseshoe shaped region of interest was drawn in the same 4 slices and used to calculate the average values of metabolic parameters.
Calculation of Blood Flow, Oxygen Consumption, Oxygen Extraction Fraction, and Glucose Uptake
Values of regional myocardial blood flow (expressed in milliliters per gram of tissue per minute),16 oxygen consumption (expressed in milliliters per gram of tissue per minute), and extraction fraction14 were calculated according to previously published methods. Myocardial glucose uptake was calculated as previously described.12 The lumped constant was assumed to be 1.0.17
Two-dimensional guided M-mode echocardiography (Acuson 128XP/5, Acuson) was performed in all study subjects to determine end-diastolic and end-systolic wall thicknesses, LV dimensions, and volumes. LV mass was calculated according to the Penn convention18 and LV mass index by dividing LV mass by body surface area. Minute work (MINW) was calculated by mean blood pressure×cardiac output, MINW index by dividing MINW by body surface area, and MINW per gram of tissue by dividing MINW by LV mass. Meridional left ventricle wall stress was calculated according to the equation WS=(p · r)/[2Wth · (1+Wth/2r), in which WS is wall stress, p is systolic blood pressure, r is systolic internal radius of left ventricle, and Wth is wall systolic thickness.19
Bicycle exercise echocardiograms were analyzed visually by an experienced physician (M.L.) to rule out silent myocardial ischemia. An upright bicycle-ergometer exercise test was performed by increasing workload by 20 W at 1-minute intervals. The test was continued until extreme fatigue or 90% of the predicted maximum heart rate. The echocardiograms were recorded before and immediately after the exercise. All subjects had a normal exercise capacity, were asymptomatic, had no diagnostic ST-changes in ECG, and no disturbances in wall motion either at rest or immediately after the maximal exercise.
Efficiency was calculated from the following equation20: Efficiency (%) where MAP is mean arterial pressure, SV stroke volume, and HR heart rate. The number 0.0136 represents a constant with units g · m−1 · mL−1 · mm Hg−1 and c is a conversion factor representing energy equivalent per mL of oxygen metabolized equaling 2.059 kg · m−1 · mL−1 oxygen consumed.
Plasma glucose was determined in duplicate by the glucose oxidase method, using an Analox GM7 (Analox Instruments) glucose analyzer. Serum insulin was measured by immunoassay12 and serum free fatty acids (FFA) by fluorometric method.12 Glycosylated hemoglobin was measured with fast-protein liquid chromatography (MonoS, Pharmacia) and serum cholesterol and triglycerides by enzymatic methods (Hibachi 717, Automatic Analyzer, Hibachi).
All results are expressed as mean±SD. The differences between the 3 subject groups were compared using ANOVA and Tukey’s studentized range test. Paired samples were compared by paired comparisons t test. Pearson’s correlation coefficients were calculated when appropriate. The regional values were compared using the Mann-Whitney U test. ANCOVA was used to study the influence of hemodynamic parameters on myocardial blood flow, oxygen extraction fraction, and oxygen consumption. P<0.05 was considered significant. Statistical computations were performed with SAS statistical program package (SAS Institute).
Fasting glucose, FFA, and serum insulin concentrations were comparable between the groups (Table 1⇑). The duration of antihypertensive medication was similar in both hypertensive groups (Table 1⇑). During euglycemic hyperinsulinemia, serum-free insulin (67±9 versus 67±11 versus 61±6 mU/L, CONT versus LVH− versus LVH+, P=NS) and plasma glucose (5.1±0.4 versus 5.2±0.4 versus 5.0±0.4 mmol/L, CONT versus LVH− versus LVH+, P=NS) concentrations were comparable in all groups. Serum FFA were higher in the LVH− group (0.18±0.06 μmol/L) than in the CONT (0.08±0.04 μmol/L) or LVH+ (0.12±0.03 μmol/L) groups (P<0.05 for LVH− versus other groups).
Hemodynamic and Echocardiographic Characteristics
Hemodynamic and echocardiographic characteristics of the groups are shown in Tables 1⇑ and 2⇑. Blood pressures were slightly but not significantly higher in the hypertensive compared with the CONT group but similar in the 2 hypertensive groups (Table 1⇑). LV mass index, LV diameters, and diameters of the septal and posterior walls (Table 2⇑) were significantly higher in the LVH+ group than in the other groups. On the basis of the criteria presented in the Framingham study,21 all hypertensive subjects in the LVH+ group had symmetric concentric LV hypertrophy (data not shown). Wall stress and ejection fraction were not different from each other between the groups (Table 2⇑). MINW, calculated per unit weight, was significantly higher in the LVH− group (Figure 1⇓). When calculated per total LV mass, MINW was significantly higher in both hypertensive groups than in the CONT group (Figure 2⇓).
Myocardial Blood Flow, Oxygen Extraction Fraction, and Oxygen Consumption
Myocardial blood flow per unit weight was higher in the LVH− group than in the other groups (0.84±0.16 versus 1.06±0.22 versus 0.81±0.09 mL · g−1 · min−1, CONT versus LVH− versus LVH+; P<0.05 LVH− versus other groups). When calculated per total LV, mass myocardial blood flow was highest in the LVH+ group (272±41 mL/min), intermediate in the LVH− group (225±31 mL/min) and lowest in the CONT group (180±23 mL/min, P<0.05 versus LVH− and LVH+ groups). Myocardial blood flow correlated significantly with MINW in each group (Figure 3⇓).
The oxygen extraction fraction (fraction of oxygen used from that delivered) was increased in both hypertensive groups compared with the CONT group (0.59±0.02 versus 0.71±0.03 versus 0.73±0.04, CONT versus LVH− versus LVH+; P<0.05 CONT versus hypertensive groups).
Myocardial oxygen consumption calculated per unit weight (Figure 1⇑) or per unit weight and per beat were higher in the LVH− than in the other groups (0.15±0.03 versus 0.19±0.02 versus 0.17±0.02 mL · 100 g−1 · beat−1, CONT versus LVH− versus LVH+; P<0.05 LVH− versus other groups).
When calculated per total LV mass (Figure 2⇑) or per total LV mass and per beat (0.32±0.06 versus 0.42±0.08 versus 0.58±0.09 mL · LV mass−1 · beat−1, CONT versus LVH− versus LVH+; P<0.05 CONT versus other groups, P<0.05 LVH+ versus LVH−), oxygen consumption was increased in both hypertensive groups compared with the controls and it was also significantly higher in the LVH+ than in the LVH− group (Figure 2⇑). Differences in heart rate or blood pressure did not explain the differences in myocardial oxygen consumption, blood flow, or oxygen extraction fraction between the groups.
Myocardial efficiency was lower in the LVH+ group (13.5±1.9%) than in the CONT group (18.1±4.1%, P<0.05). Efficiency in the LVH− group (15.1±2.3%) was between that in the LVH+ and CONT groups but not significantly different from either of the other groups (Figure 4⇓).
Myocardial Glucose Uptake
Myocardial glucose uptake per unit weight was not different between the groups (54±14 versus 62±11 versus 64±17 μmol · 100g−1 · min−1, CONT versus LVH− versus LVH+, P=NS). When calculated per whole heart, glucose uptake was highest in the LVH+ group (118±31 versus 135±24 versus 214±57 μmol/min, P<0.05 LVH+ versus other groups).
In the present study, myocardial oxygen consumption, blood flow, and efficiency were directly measured using [15O]-labeled oxygen, PET, and echocardiography in asymptomatic hypertensive patients with and without LVH and in normal subjects. Myocardial oxygen consumption per unit weight was increased in the LVH− group but normal in the LVH+ group. This apparent normalization of oxygen consumption per unit weight via hypertrophy occurred, however, at the expense of reduced efficiency, ie, a decrease in the ratio of cardiac work to oxygen consumption.
Myocardial oxygen consumption per gram of tissue was significantly higher in the LVH− group than in the other groups of this study. Previous catherization studies have not documented increased myocardial oxygen consumption per gram of tissue in hypertensive patients when compared with normotensive subjects.5 6 7 This could be due to failure of previous studies5 6 7 to subgroup patients with hypertension to those with and without LVH, as oxygen consumption per unit weight has been shown to decrease at increasing ratios of LV mass to volume.4 In these previous studies, the proportion of hypertensive subjects with LVH ranged from 29%7 to 74%,5 and LV mass indices varied widely within the hypertensive group, eg, from 75 to 360 g/m2 in the study of Vogt et al 1992.6 Also, these studies included hypertensive patients with symptoms of angina pectoris and normal coronary angiograms. The present study subjects differed from those in previous studies at least in 2 aspects: our subjects had no symptoms or signs of hypertensive microvascular heart disease and were divided into LVH− and LVH+ groups on the basis of their echocardiographically determined LV masses. Oxygen consumption was found to be significantly lower per unit mass of myocardium in patients with than without LVH. This result is consistent with previous data in rats in which a pressure-induced increase in myocardial mass is accompanied by a decrease in oxygen consumption and a shift in myosin isoenzymes from a faster V1 to a slower V3 enzymatic form.22 Thus, LVH may be viewed as a structural adaptation, which aims at restoring oxygen consumption per unit mass of myocardium toward normal.
In the present study, glucose uptake per gram of myocardial tissue was not different between the groups. Serum FFA were, however, slightly higher in the LVH− than in the CONT and LVH+ groups. Because it is more costly to burn fatty acids than glucose, one could speculate that the decreased efficiency, ie, increased oxygen cost of myocardial work found in the LVH+ group, might be due to increased myocardial fatty acid usage. In this study, FFA uptake was not quantified, but it can be estimated by multiplying the FFA concentrations measured in the present study by previously (under similar metabolic conditions than in our study) quantified FFA uptake rates of normal myocardium.23 Assuming all FFA are oxidized, the percent of oxygen maximally used for FFA oxidation24 can be estimated. These calculations reveal that 6% of total myocardial oxygen uptake could have maximally been used for FFA oxidation in the control group, 8% in the LVH−, and 7% in the LVH+ group. These data imply that decreased efficiency in the LVH+ group cannot be simply attributed by increased use of FFA.
An exaggerated interstitial and perivascular deposition of fibrillar collagen is found in hypertrophied left ventricles of hypertensive subjects,25 and increased amount of myocardial collagen has been held responsible for impaired ventricular pumping capacity.26 In addition to the increment of myocardial collagen, media hypertrophy of intramyocardial coronary arterioles has also been found in myocardial biopsies of hypertensive patients.7 Such structural alterations might partly explain the present finding of reduced myocardial efficiency in patients with LVH. In concentric hypertrophy, an increment of the LV diastolic diameter, which characterized the LVH+ group in the present study, combined with only slightly elevated LV end-diastolic pressure would be consistent with initial heart failure.27 Because LV end-diastolic pressures were not measured in the present study, we cannot exclude this possibility in the LVH+ group.
In the present study all hypertensive subjects used an ACE inhibitor as antihypertensive medication, with the exception of 1 patient in the LVH− group who used a calcium channel blocker. To achieve normotension, a calcium channel blocker was combined with an ACE inhibitor in 1 patient in the LVH− group and in 5 patients in the LVH+ group. Long-term antihypertensive treatment using both ACE inhibitors and calcium channel blockers have been shown to reduce LV hypertrophy and to exert a favorable effect on the matching between myocardial mass and perfusion.28 A recent study in dogs suggested that ACE inhibitors might also play a role in matching myocardial oxygen supply to oxygen demands, because ACE inhibitors were found to reduce oxygen consumption in coronary microvessels and increase cardiac nitric oxide release.29 Consequently, lack of measurements performed in the absence of antihypertensive medication could be viewed as a limitation of the present study. On the other hand, because the LVH patients used more medication than those without LVH, one could argue that differences in efficiency and oxygen consumption between the hypertensive groups would have been even greater if untreated patients had been studied. Because only asymptomatic, middle-aged, hypertensive men with normal stress echocardiographies were studied, our data can obviously not be directly extrapolated to hypertensive patients with symptoms of angina or to women.
In conclusion, the present data demonstrate that myocardial oxygen consumption per unit weight is increased in hypertensive patients without LVH but normal in those with LVH. This apparent normalization in oxygen consumption is accompanied by reduced efficiency, ie, the ratio of myocardial work to oxygen consumption. LV failure is invariably characterized by reduced efficiency. The present data raise the possibility that a decrease in efficiency also predisposes patients with hypertension and LVH to develop heart failure.
This study was supported by grants from the Academy of Finland (to P.N. and H.Y.-J.), the Finnish Foundation for Cardiovascular Research (to J.K.), the Turku University Foundation (to H.L.), and the Maud Kuistila Foundation (to H.L.).
- Received January 21, 1999.
- Revision received July 19, 1999.
- Accepted July 28, 1999.
- Copyright © 1999 by American Heart Association
Vogt M, Motz WH, Schwartzkopf, Strauer BE. Pathophysiology and clinical aspects of hypertensive hypertrophy. Eur Heart J. 1993;14(suppl D):2–7.
Vogt M, Motz W, Schwartzkopff B, Strauer BE. Coronary microangiopathy and cardiac hypertrophy. Eur Heart J. 1990;11(suppl B):133–138.
Vogt M, Motz W, Strauer BE. Coronary hemodynamics in hypertensive heart disease. Eur Heart J. 1992;13(suppl D):44–49.
Schwartzkopff B, Motz B, Frenzel H, Vogt M, Knauer S, Strauer BE. Structural and functional alterations of the intramyocardial coronary arterioles in patients with arterial hypertension. Circulation. 1993;88:993–1003.
Neely JR, Liebermeister H, Battersby EJ. Effect of pressure development on oxygen consumption by isolated heart. Am J Physiol. 1967;212:804–814.
Camici P, Marraccini P, Marzilli M, Lorenzoni R, Buzzigoli G, Puntoni R, Boni C, Bellina CR, Klassen GA, L’Abbate A, Ferrannini E. Coronary hemodynamics and myocardial metabolism during and after pacing stress in normal humans. Am J Physiol. 1989;257:E309–E317.
Nuutila P, Mäki M, Laine H, Knuuti J, Ruotsalainen U, Luotolahti M, Haaparanta M, Solin O, Jula A, Koivisto VA, Voipio-Pulkki L-M, Yki-Järvinen H. Insulin action on heart and skeletal muscle glucose uptake in essential hypertension. J Clin Invest. 1995;96:1003–1009.
Iida H, Rhodes CG, Araujo LI, Yamamoto Y, DeSilva R, Maseri A, Jones T. Noninvasive quantification of regional myocardial metabolic rate for oxygen by use of 15 O2 inhalation and positron emission tomography. Circulation. 1996;94:792–807.
Iida H, Kanno I, Takahashi A, Miura S, Murukami M, Takahashi K, Ono Y, Shishido F, Inugami A, Tomura N, Higano S, Fujita H, Sasaki H, Nakamichi H, Mizusawa S, Kondo Y, Uemura K. Measurement of absolute myocardial blood flow with H215O and dynamic positron-emission tomography: strategy for quantification in relation to the partial-volume effect. Circulation.. 1988;78:104–115.
Ng CK, Soufer R, McNulty PH. Effect of hyperinsulinemia on myocardial fluorine-18-FDG uptake. J Nucl Med. 1998;39:379–383.
Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man. Anatomical validation of the method. Circulation. 1977;55:613–618.
Beanlands BSB, Bach DS, Raylman R, Armstrong WF, Wilson V, Montieth M, Moore CK, Bates E, Schwaiger M. Acute effects of dobutamine on myocardial oxygen consumption and cardiac efficiency measured using carbon-11 acetate kinetics in patients with dilated cardiomyopathy. J Am Coll Cardiol. 1993;22:1389–1398.
Savage DD, Garrison RJ, Kannel WB, Levy D, Anderson SJ, Stokes J, Feinleib M, Castelli WP. The spectrum of left ventricular hypertrophy in a general population sample: the Framingham study. Circulation. 1987;75(suppl I):26–33.
Izumo S, Lompre A-M, Matsuoka R, Koren G, Schwartz K, Nadal-Ginard B, Mahdavi V. Myocsin heavy chain messenger RNA and protein isoform transitions during cardiac hypertrophy. J Clin Invest. 1987;79:970–977.
Mäki M, Haaparanta M, Nuutila P, Oikonen V, Luotolahti M, Solin O, Knuuti J. Free fatty acid uptake in the myocardium and skeletal muscle using (18F)fluoro-6-thia-heptadecanoic acid. J Nucl Med. 1998;39:12:1320–1327.
Frayn KN. Calculation of substrate oxidation rates in vivo from gaseous exchange. J Appl Physiol. 1983;55:628–634.
Capasso JM, Palackal T, Olivetti G, Anversa P. Left ventricular failure induced by long-term hypertension in rats. Circ Res. 1990;66:1400–1412.
Koide M, Nagutso M, Zile MR, Hamawaki M, Swindle MM, Keech G, DeFreyte G, Tagawa H, Cooper G, Carabello BA. Premorbid determinants of left ventricular dysfunction in a novel model of gradually induced pressure overload in the adult canine. Circulation. 1997;95:1601–1610.
Parodi O, Neglia D, Palombo C, Sambuceti G, Giorgietti A, Marabotti C, Gallopin M, Simonetti I, L’Abbate A. Comparative effects of enalapril and verapamil on myocardial blood flow in systemic hypertension. Circulation. 1997;96:864–873.
Zhang X, Xie Y-W, Nasjletti A, Xu X, Wolin MS, Hintze TH. ACE inhibitors promote nitric oxide accumulation to modulate myocardial oxygen consumption. Circulation. 1997;95:176–182.