Author + information
- Received March 12, 2002
- Revision received April 29, 2002
- Accepted May 9, 2002
- Published online August 21, 2002.
- Olakunle O Akinboboye, MD, MPH, FACC*,* (, )
- R.u-Ling Chou, PhD† and
- Steven R Bergmann, MD, PhD, FACC†
- ↵*Reprint requests and correspondence:
Dr. Olakunle O. Akinboboye, Non-Invasive Laboratory, Saint Francis Hospital, 100 Port Washington Boulevard, Roslyn, New York 11576, USA.
Objectives The goal of this study was to compare myocardial perfusion reserve (MPR) before and after long-term treatment with lisinopril and losartan in patients with hypertension and left ventricular hypertrophy (LVH).
Background Studies have suggested that treatment with angiotensin-converting enzyme inhibitors (ACEIs) improves MPR in patients with hypertension by potentiating endogenous bradykinins. Because angiotensin receptor blockers (ARBs) lack a direct effect on bradykinins, we hypothesized that they may not improve MPR.
Methods We measured pre- and post-treatment myocardial blood flow (MBF) by positron emission tomography in 17 patients (lisinopril: 9 patients, losartan: 8 patients) with hypertension and LVH at baseline and after coronary vasodilation with intravenous dipyridamole. In addition, we measured rest and hyperemic blood flow in eight normotensive controls.
Results Post-treatment maximal coronary blood flow and MPR in the lisinopril group increased significantly compared with pretreatment values (3.5 ± 1.2 vs. 2.6 ± 1.1 ml/min/g, p = 0.02; 3.7 ± 1.1 vs. 2.4 ± 1 ml/min/g, respectively, p = 0.002, respectively). Post-treatment hyperemic flow in the patients treated with lisinopril was not significantly different from corresponding measurements in controls (3.5 ± 1.2 vs. 3.9 ± 1 ml/min/g, respectively, p = NS). In the patients treated with losartan, there was no difference between pre- and post-treatment MBF values and MPR.
Conclusions Myocardial perfusion reserve and maximal coronary flow improved in asymptomatic patients with hypertension-induced LVH after long-term treatment with lisinopril but not with losartan. Thus, ACEIs, but not ARBs, might be effective in repairing the coronary microangiopathy associated with hypertension-induced LVH.
Left ventricular hypertrophy (LVH) is a major predictor of cardiovascular mortality and morbidity independent of blood pressure level (1,2). Impaired myocardial perfusion reserve (MPR) is believed to play a role in the cardiovascular mortality and morbidity associated with hypertension-induced LVH (1). Long-term treatment with angiotensin-converting enzyme inhibitors (ACEIs) has been shown to improve MPR in patients with hypertension and clinical signs of ischemia but without angiographic evidence of epicardial coronary artery disease (2). However, it is not clearly known if these findings can be extrapolated to asymptomatic patients with hypertension-induced LVH.
Several studies have shown that the improvement in MPR by ACEIs is mediated by potentiation of endogenous bradykinin (3–6). Because angiotensin-receptor blockers (ARBs) do not directly potentiate endogenous bradykinin, they may not have the same effect on myocardial perfusion as ACEI.
Thus, the aim of this study was to compare MPR before and after long-term treatment with an ACEI, lisinopril and an ARB, losartan, in patients with moderate hypertension and LVH. We also sought to examine if treatment with angiotensin antagonists will normalize MPR by comparing post-treatment measurements in the patients with the MPR in normotensive controls.
The Institutional Review Board of Columbia University approved the study. Informed consent was obtained. Participants with pre-existing hypertension and LVH (left ventricular [LV] mass > 125 g/m2) and normotensive controls were recruited from the patient population at New York Presbyterian Medical Center in New York City and the surrounding community. Pre-existing hypertension was validated by blood pressure measurement on entry. Race/ethnicity was based on self-report.
Exclusion criteria for patients with hypertension and LVH included: echocardiographic evidence of significant cardiac abnormality other than hypertrophy, such as valvular heart disease, LV systolic dysfunction, and wall motion abnormalities; significant cardiovascular disease other than hypertension with LVH; history of chest pain; pregnancy; malignant, accelerated, or secondary hypertension; bronchospastic lung disease; regional differences in MPR on pre-treatment positron emission tomography (PET) studies suggestive of coronary artery disease; diabetes mellitus; smoking; renal failure; secondary hypertension and prior treatment with angiotensin antagonists. Additional exclusion criteria for normotensive controls included hypertension and LVH.
Using these criteria we recruited 20 patients, including 14 African Americans and six Latinos with hypertension-induced LVH and 8 normotensive controls. Blood was tested for occult diabetes mellitus, renal failure, and hyperlipidemia.
Measurement of MPR
Absolute myocardial blood flow (MBF) was measured at rest by PET and after maximal coronary vasodilation with intravenous dipyridamole using 15O-water as the flow tracer. All medications were held for at least five half-lives before the PET studies. Measurement of MBF by PET was performed on an ECAT EXACT-47 PET scanner (Siemens, Iselin, New Jersey), which provides 47 contiguous transaxial slices and has a postprocessing spatial resolution of 9 to 10 mm in plane and 4 to 5 mm in the axial direction.
The subjects were studied after an overnight fast and abstinence from methyl xanthines including caffeine for 24 h. Subjects were placed supine in the PET scanner. The localization of the heart within the axial field of view of the scanner was confirmed by performing a 2-min “positioning” scan using a rotating 68Ge/68Ga-rod source. Patient positioning was marked using laser cross beams and ink marks on the patient’s torso. A 20-min transmission scan with the rotating 68Ge/68Ga-rod source was performed to generate an attenuation correction map for the correction of the emission sinogram.
A bolus administration of 0.20 mCi/kg of 15O-water was given with simultaneous initiation of dynamic data acquisition for 300 s. After completion of the rest perfusion scans, 0.56 mg/kg of dipyridamole was administered over 4 min. An additional 4 min was allowed to achieve peak flow response, and 0.20 mCi/kg of 15O-water was readministered. Dynamic data acquisition was initiated for 300 s. The initial 20 s of the two dynamic datasets were summed to identify the blood pool phase. The emission sinograms were corrected for radioactivity decay, reconstructed into transaxial slices and reoriented into short-axis tomograms. Each of the short-axis tomograms was divided into four equal myocardial sectors and count data in each sector used to generate myocardial tissue time activity curves. The arterial input function was obtained from the time-activity curve generated from a 1.5-cm3 region-of-interest placed in the center of basal short-axis tomograms. Regional MBF was quantified using a one-compartment model (7–9). This parameter-optimization approach incorporates corrections for both partial volume effects and blood to myocardial spillover. As no regional differences were noted, regional MBF values were averaged to yield one global flow value for each patient at rest and after coronary vasodilation. Myocardial perfusion reserve was expressed as the ratio of global maximal to resting absolute flow values.
Measurement of LV mass
An LV mass of >125 g/m2 by M-mode echocardiography was used as an inclusion criterion for the patients and an exclusion criterion for the controls (1). Because of lower measurement variability, we measured baseline and post-treatment LV mass by three-dimensional echocardiography (10,11).
Measurement of LV mass by M-mode echocardiography
Two-dimensional guided M-mode echocardiography was performed to measure LV wall mass. The ultrasound beam was aligned perpendicularly to the interventricular septum and the LV posterior wall at a level slightly below the mitral leaflet tips. Left ventricular mass was calculated from measurements of the septal (SD) and posterior wall thickness (PD) and left ventricular dimensions (LVID) at end-diastole using the cube formula: 1.05 × ([SD + LVID + PD]3 − [LVID]3) (12).
Measurement of LV mass by three-dimensional echocardiography
The three-dimensional echocardiographic system (K3 Systems, Inc., Darien, Connecticut) comprises an acoustic spatial locater (Model GP 8-3D, Science Accessories Corp., Stratford, Connecticut), and personal computer (Model 4DX-33V, Gateway 2000, North Sioux City, South Dakota) (13,14). These components are linked to a conventional two-dimensional echocardiograph interfaced with an acoustic spatial locater. Left ventricular volume is computed from a series of guided six to eight real-time parasternal short-axis images. These images were stored along with their XYZ Cartesian coordinates in the personal computer. End-diastolic video frames from each acquired cine-loop were selected for off-line endocardial and epicardial boundary tracing. Epicardial and endocardial volumes were determined from their corresponding boundaries using a polyhedral surface reconstruction algorithm. Endocardial volume was subtracted from epicardial volume to yield myocardial volume, which was multiplied by myocardial density to yield LV mass.
After completing the baseline studies, the patients were randomized to treatment with either lisinopril or losartan. The starting doses for lisinopril and losartan were 10 mg and 50 mg, respectively. The patients were seen weekly, and their medications were adjusted until either a blood pressure of 140/90 mm Hg or less or the maximal daily dosage of the medication was attained. Hydrochlorothiazide (up to 25 mg daily) and atenolol (up to 100 mg daily) were added as needed for further blood pressure control. After a blood pressure of 140/90 mm Hg or less was attained, patients were followed monthly thereafter for the duration of the study.
Seventeen patients completed the study. Two patients declined the follow-up PET study, and one patient was lost to follow-up. After approximately one year of blood pressure control, resting and hyperemic absolute MBF in the 17 patients was measured by PET, using the same dose of dipyridamole as was used in the pretreatment study. We also used the same dose of 15O-water for the rest and hyperemic flow measurements as was used in the pretreatment studies. We withdrew all antihypertensives for at least five half-lives before the post-treatment studies to minimize the acute affects of the medications on coronary blood flow. The patients had their blood pressures measured daily during this washout phase.
Absolute rest and hyperemic flow and MPR were compared before and after treatment using paired t test analysis. Analysis of variance with Bonferroni correction was used to test for differences in continuous variables between groups. Categorical data were compared using chi-square analysis. A p value of <0.05 was considered significant for all analyses.
The two treatment arms were comparable in terms of race and gender distribution. There were more whites in the control group and more African Americans in the patient groups. The duration of hypertension is not different between the two groups (Table 1). The number of patients who required cotreatment with diuretics and beta-blockers were equal between the two treatment groups (Table 1). Expectedly, pretreatment blood pressure and LV mass were higher in the patients than in the controls (Table 2).
Change in blood pressure
Blood pressure control was achieved for a mean duration of 11 ± 1 months. The change in mean arterial pressure over the treatment period was not significantly different between the two treatment groups (Table 2). The mean doses of lisinopril and losartan that were used were 35 ± 10 mg and 90 ± 10 mg, respectively. Pretreatment systolic and diastolic blood pressures were similar in the two patient groups and higher than corresponding measurements in controls. On-treatment systolic and diastolic blood pressure measurements were also higher than measurements in controls (Table 2).
The baseline blood pressure (after drug wash-out) at the time of the post-treatment PET study was neither different between the two groups nor different from the pretreatment blood pressure in both groups. The frequency of cotreatment with diuretics or beta-blockers and their dosages were not significantly different between the two treatment groups (Table 1).
Hemodynamic response to dipyridamole administration
The response to dipyridamole infusion was not significantly different between the two treatment groups. During the pretreatment PET studies, heart rate increased and systolic blood pressure decreased by 18 ± 4 beats/min and 8 ± 2 mm Hg, respectively, with dipyridamole infusion. The response to dipyridamole during the post-treatment PET study was similar; heart rate increased and blood pressures decreased by 14 ± 3 beats/min and 6 ± 2 mm Hg, respectively.
Change in LV mass
Pre- and post-treatment LV mass were significantly higher in the patients compared with corresponding measurements in controls. Left ventricular mass decreased by 4% in the lisinopril group and by 5% in the losartan group. The change in LV mass was neither significant in either group nor different between the two groups (Table 2). There was no relationship between the magnitude of change in LV mass and the change in systolic or diastolic blood pressure over the treatment period.
Change in absolute MBF in patients versus controls
Pretreatment resting blood flow was not significantly different between patients and controls (1.1 ± 0.4 vs. 1 ± 0.2, respectively). Pretreatment hyperemic flow was significantly lower in the patients than in the controls (2.8 ± 0.9 vs. 3.9 ± 1 ml/g/min, p < 0.01) but not different between the two treatment groups 2.6 ± 1.1 vs. 2.9 ± 0.5, lisinopril vs. losartan, respectively). Pre- and post-treatment resting flow values in the combined patient population were not significantly different (1.1 ± 0.2 vs. 0.9 ± 0.2, respectively). Although post-treatment hyperemic flow in the combined patient population was not significantly different from baseline values (3.2 ± 1.1 vs. 2.8 ± 0.9 ml/min/g, respectively), post treatment flow reserve improved significantly from 2.8 ± 0.8 pretreatment to 3.4 ± 0.8 post-treatment (p = 0.04).
Change in absolute MBF in the lisinopril versus losartan arms
Post treatment resting flow was not different from pretreatment flow values in both patient groups (Figs. 1a and 2a). ⇓⇓ However, in the lisinopril group, post-treatment hyperemic flow increased significantly compared with pretreatment hyperemic flow (3.5 ± 1.2 vs. 2.6 ± 1.1 ml/min/g, p = 0.02) (Fig. 1b). Consequently, MPR in the lisinopril group increased significantly after treatment compared with pretreatment values (3.7 ± 1.1 vs. 2.4 ± 1, respectively, p = 0.002) (Fig. 1c). Post-treatment hyperemic flow in the patients treated with lisinopril was not significantly different from corresponding measurements in controls (Fig. 3). In the patients treated with losartan, posttreatment absolute MBF and MPR did not change compared with pretreatment values (Figs. 2b and 2c), and post-treatment hyperemic flow remained significantly lower than corresponding measurements in controls (Fig. 3).
There was no correlation between the change in absolute MBF and the extent of LV mass regression. Furthermore, the change in absolute MBF did not correlate with the magnitude of the reduction in mean arterial pressure.
Myocardial perfusion reserve increased significantly in the group of patients as a whole after treatment compared with baseline values. However, absolute hyperemic flow did not change significantly from baseline. This would suggest that absolute vasodilatory capacity did not improve in the group as a whole compared with baseline. However, when the lisinopril and losartan treatment arms were analyzed separately, we found an improvement in both MPR and absolute hyperemic flow in the lisinopril group. However, in the losartan group, neither MPR nor absolute hyperemic flow changed significantly from baseline. Furthermore, post-treatment hyperemic flow in the group treated with lisinopril was not different from corresponding measurements in controls (Fig. 3). Thus, treatment with lisinopril resulted in normalization of MPR in this group of patients.
The fact that maximal myocardial perfusion improved in the patients that were treated with a lisinopril-based regimen and was unchanged in those that were treated with a losartan-based regimen suggests that the improvement in maximal MBF might be unrelated to the direct effects of lisinopril on angiotensin II production. Because the angiotensin-converting enzyme plays a major role in the breakdown of kinins, a mechanism for the improvement in myocardial perfusion by lisinopril might be the effect of increased availability of bradykinin (5) and, consequently, vasodilatory prostaglandins, and nitric oxide. Unlike ACEIs, ARBs do not have any direct effect on kinin metabolism.
Because lisinopril was withdrawn for at least five half-lives and blood pressure rose to pretreatment levels before the post-treatment flow measurements were taken, it is unlikely that the observed improvement in hyperemic MBF and MPR could be due to an acute effect of the drug, although the latter cannot be completely excluded. Furthermore, the improvement in MPR despite the absence of a significant reduction in LV mass suggests that the improvement in coronary vasodilatory capacity might not be attributable to a reduction in extravascular compressive forces on the coronary microvasculature.
A potential mechanism for the improvement in maximal myocardial perfusion is an increase in myocardial capillary density (15,16). Several studies have demonstrated an improvement in MBF in hypertensive animals; however, the results of such studies in humans have yielded conflicting results. Canby et al. (15) demonstrated increased capillary density and decreased minimal coronary vascular resistance in captopril-treated spontaneously hypertensive rats compared with untreated controls. Normalization of medial wall thickness and minimal coronary resistance were also found in 14-week old spontaneously hypertensive rats after a 12-week treatment with lisinopril compared with untreated controls.
Goehlke et al. (5) demonstrated an improvement in myocardial capillary density in spontaneously hypertensive rats after treatment with lisinopril. However, in the same study the investigators showed that the improvement in myocardial capillary density was abolished when the rats were pretreated with bradykinin B2 receptor antagonist, icatibant. Motz et al. (4) demonstrated a 43% improvement in maximal coronary blood flow and a 23% reduction in minimal coronary vascular resistance in 15 patients with hypertension, LVH, clinical evidence of myocardial ischemia with normal coronary angiograms after treatment with enalapril for one year. Parodi et al. (17) found an improvement in MPR, measured by PET with 13-NH3 in 10 patients treated with verapamil, while another group of 10 patients that were treated with enalapril did not show any significant change. The majority of the studies on the effects of angiotensin antagonists on myocardial perfusion in patients with hypertension-induced LVH were conducted in patients with clinical symptoms of myocardial ischemia (4,18). To the best of our knowledge, ours is the first study to compare the effects of ARBs and ACEIs in an asymptomatic population of patients with moderate hypertension and LVH.
It is also noteworthy that the majority of our patients are African Americans (70%). Some studies have suggested that African Americans with hypertension or heart failure may not benefit from treatment with ACEIs (19,20). However, we found a significant improvement in MPR and maximal MBF in the African Americans that we studied.
The limitations of this study should be recognized. First, the sample size is not large, and there is significant interpatient variability in the change in absolute MBF after treatment. Thus, further studies in larger populations are needed. Second, despite adequate blood pressure control for a mean duration of 11 months, we did not achieve a significant reduction in LV mass. However, the absolute reduction in LV mass in our study population is close to what was observed in some large-scale clinical trials of LV mass regression with antihypertensive treatment. Furthermore, the study was not powered from the outset to detect statistically significant changes in LV mass. It is possible that absence of LV mass regression might have contributed to the failure of the ARB group to improve their maximal MBF with treatment. However, we observed an improvement in absolute MBF in the ACEI group without a significant change in LV mass.
In addition, the difference in race between patients and controls is unlikely to have influenced our findings because race has not been shown to have an independent effect on coronary blood flow (21).
Our study shows that MPR and maximal coronary flow improved in asymptomatic patients with hypertension-induced LVH after long-term treatment with lisinopril but not with losartan. Thus, ACEIs but not ARBs might be effective in repairing the coronary microangiopathy associated with hypertension-induced LVH.
The authors thank Drs. N. Reichek and J. Gottdiener for reviewing this manuscript, Ms. Karen Ngai for helping with manuscript preparation, and Ms. K. Berekashvili for helping with data collection.
☆ Supported by research grant RR00645, National Center for Research Resources NIH.
Presented, in part, at the AHA 73rd Scientific Sessions, New Orleans, Louisiana, 2000.
- angiotensin-converting enzyme inhibitor
- angiotensin-receptor blocker
- left ventricular
- left ventricular hypertrophy
- left ventricular dimensions
- myocardial blood flow
- myocardial perfusion reserve
- posterior wall thickness
- positron emission tomography
- septal wall thickness
- Received March 12, 2002.
- Revision received April 29, 2002.
- Accepted May 9, 2002.
- American College of Cardiology Foundation
- Houghton J.L.,
- Frank M.J.,
- Carr A.A.,
- von Dohlen T.W.,
- Prisant L.M.
- Ruocco N.A.,
- Bergelson B.A.,
- Yu T.K.,
- Gavras I.,
- Gavras H.
- Gohlke P.,
- Kuwer I.,
- Schnell A.,
- Amam K.,
- Mall G.,
- Unger T.
- Bergmann S.R.,
- Herrero P.,
- Markham J.,
- Weinheimer C.J.,
- Walsh M.N.
- Herrero P.,
- Markham J.,
- Bergmann S.R.
- Bergmann S.R.
- Gopal A.S.,
- Keller A.M.,
- Shen Z.,
- et al.
- Gopal A.S.,
- Schnellbaecher M.J.,
- Shen Z.,
- Akinboboye O.,
- Sapin P.M.,
- King D.L.
- Devereux R.B.,
- Reichek N.
- Canby C.A.,
- Tomanek R.J.
- Brilla C.G.,
- Janicki J.S.,
- Weber K.T.
- Parodi O.,
- Neglia G.,
- Palombo C.,
- et al.
- Faile B.A.,
- Guzzo J.A.,
- Tate D.A.,
- Nichols T.C.,
- Smith S.C.,
- Dehmer G.J.