Journal of Human Hypertension (2006) 20, 787–794 & 2006 Nature Publishing Group All rights reserved 0950-9240/06 $30.00 www.nature.com/jhh ORIGINAL ARTICLE Effects of amlodipine and valsartan on vascular damage and ambulatory blood pressure in untreated hypertensive patients

A Ichihara, Y Kaneshiro, T Takemitsu and M Sakoda Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan

The present study was performed to compare the long- amlodipine but not by valsartan. The reduced variability term effects of 24-h ambulatory blood pressure (BP) of ambulatory systolic BP caused by amlodipine control with amlodipine versus valsartan on vascular significantly contributed to the improvement of PWV, damage in untreated hypertensive patients. Amlodipine although both drugs decreased PWV similarly. Carotid and valsartan have benefits on cardiovascular mortality IMT was unaffected by treatment with either drug. and morbidity in hypertensive patients. Although ambu- Valsartan significantly decreased UAE independently latory BP is associated with severity of target-organ of its depressor effect, but amlodipine had no effect damage in hypertensive patients, beneficial effects of on UAE. These results suggest that the 24-h control ambulatory BP control with amlodipine versus valsartan of ambulatory BP with amlodipine had functionally on vascular damage have not been compared. Pulse improved the stiffened arteries of hypertensive patients wave velocity (PWV), intima–media thickness (IMT) of by the end of 12 months of treatment, in part through the carotid arteries, urinary albumin excretion (UAE) and reducing BP variability, whereas ambulatory BP control 24-h ambulatory BP were determined in 100 untreated with valsartan decreased the arterial stiffness to the hypertensive patients before and 12 months after the same degree as amlodipine without affecting BP varia- start of antihypertensive therapy with amlodipine or bility maybe through some pleiotropic effects. valsartan. Amlodipine and valsartan decreased ambula- Journal of Human Hypertension (2006) 20, 787–794. tory BP similarly, but the variability of 24-h and daytime doi:10.1038/sj.jhh.1002067; published online 29 June 2006 ambulatory systolic BP was significantly reduced by

Keywords: ambulatory blood pressure; intima–media thickness; microalbuminuria; oxidative stress; pulse wave velocity

Introduction throughout a 24-h period by amlodipine than by valsartan.8 Thus, it appeared that the intensive The Valsartan Antihypertensive Long-term Use Eval- control of ambulatory BP by the long-acting nature uation (VALUE) study demonstrated that amlodi- of amlodipine might have a beneficial effect on the pine-based therapy has a greater beneficial effect in vascular damage of untreated hypertensive patients. intensive control of blood pressure (BP) and low- Moreover, a recent study reported a greater reduc- ering the risk of myocardial infarction than valsar- tion in the aortic pulse wave velocity (PWV) of tan-based therapy in hypertensive patients already 1 hypertensive patients, which predicts cardiovascu- on antihypertensive drugs. However, the beneficial lar morbidity and mortality9–12 by long-term treat- effects of amlodipine and valsartan on untreated ment with valsartan than with a long-acting calcium hypertensive patients had never been compared. channel blocker, even though both drugs decreased Studies have demonstrated that 24-h ambulatory the BP similarly.13 Thus, the pleiotropic effects BP predicts cardiovascular morbidity and mortality of valsartan may make an important contribution much more accurately than conventional clinic 2–7 to improving the arterial stiffness of untreated BP, and that ambulatory BP is better controlled hypertensive patients. The aim of this study was to compare the long- term effects of 24-h ambulatory BP control with Correspondence: Dr A Ichihara, Department of Internal Medicine, amlodipine and valsartan on vascular damage in Keio University School of Medicine, 35 Shinanomachi, Shinjuku- hypertensive patients who had never been treated ku, Tokyo 160-8582, Japan. E-mail: [email protected] with any antihypertensive drugs. Renal vascular Received 13 April 2006; revised 29 May 2006; accepted 30 May damage was evaluated by urinary albumin excretion 2006; published online 29 June 2006 (UAE), and extra-renal vascular damage was func- Ambulatory BP and vascular damage A Ichihara et al 788 tionally and structurally assessed on the basis of hours) and every 60 min during the night time PWV and the intima–media thickness (IMT) of the period (between 2200 and 0600 hours). Mean values carotid arteries. and standard deviations of ambulatory BP for each subject were calculated for 24 h and separately for the daytime and night time periods. The standard Materials and methods deviation of the ambulatory BP values was recorded as the variability of ambulatory BP in this study. Study population and design The nocturnal decrease in BP was calculated as The subjects of the present study were 100 con- the average systolic BP during the daytime period secutive untreated hypertensive patients who minus the average systolic BP during the night time attended the outpatient hypertension clinic of Keio period. The morning BP surge was calculated as the University Medical School Hospital and for whom average systolic BP during the first 2 h after waking baseline arterial stiffness measurements by PWV minus the lowest systolic BP during the night time were available. Patients with arrhythmias, previous period. cardiovascular events or peripheral arterial disease were not included in the study. The 24-h ambulatory BP and three times clinic BP were determined Pulse wave velocity during the 3-month observation period before the Clinic BP and PWV were determined with a pulse study. Hypertension was defined as a clinic systolic pressure analyser (model: BP-203RPE, Nihon Colin, BP of 4140 mm Hg at any times and/or a clinic Tokyo, Japan) by the methods described pre- diastolic BP of 490 mm Hg at any times and as 24-h viously.15–18 Briefly, pulse volume waveforms were ambulatory systolic BP of 4135 mm Hg and/or recorded with plethymographic sensors placed over 24-h ambulatory BP of 480 mm Hg. No patients the right brachial artery and both posterior tibial with any form of secondary hypertension based on arteries. The PWV values measured by this method the results of classic laboratory and radiology tests are significantly correlated with carotid–femoral were included. All hypertensive patients were PWV,19 heart–ankle PWV17 and aortic PWV mea- randomly assigned to an amlodipine group or sured by the catheter method,20 with high correla- a valsartan group. The amlodipine group was started tion coefficients of 0.81, 0.89 and 0.87, respectively. at a dose of 2.5 mg/day, and the dose was subse- The best 10 consecutive pulses were analysed in the quently increased by 2.5-mg increments at intervals present study, and the average PWV was recorded. of 4 weeks. If the target BP of o130/85 mm Hg was The PWV was measured when the patients’ heart not achieved at the maximum dose of 10 mg/day, rate was stable at 60–80 beats/min after at least a a b-blocker was added, and then, if necessary, 5-min rest. Two measurements were performed in a diuretic was also added. The valsartan group each leg, and the average values were used in the was started at a dose of 40 mg/day, and the dose analysis. Values are expressed in units of cm/s. was subsequently increased in 40-mg increments at intervals of 4 weeks. If the target BP of o130/85 mm Hg was not achieved at the maximum dose Carotid IMT Ultrasonography B-mode imaging of the carotid of 160 mg/day, a diuretic was added, and then, if artery was performed with a PowerVision 6000 necessary, a b-blocker was also added. Clinical and machine (Toshiba, Tokyo, Japan) at a transducer biological parameters were obtained before and frequency of 7.5 MHz. Each subject was examined 12 months after the start of medication with the in the supine position. Up to 4 cm of the common antihypertensive drugs. The study was approved by carotid artery and the carotid bulb were scanned the review board of Keio University Medical School bilaterally in the longitudinal and the transverse Hospital, and informed consent was obtained from projections. The image was focused on the far wall every subject. of the artery. The IMT was defined as the distance Creatinine, uric acid, total cholesterol, triglyceride, between the leading edge of the lumen–intima high-density lipoprotein, low-density lipoprotein, interface and the leading edge of the media– C-reactive protein, active renin concentrations adventitia interface of the far wall, and the greatest and superoxide dismutase (SOD) activity, which in IMT value in the bilateral longitudinal projections the leucocyte correlates with longevity,14 were mea- was recorded as the maximum IMT. All measure- sured in venous blood samples drawn in the ments were performed under blind conditions. The morning after an overnight fast on the same days mean intraobserver and interobserver coefficients as the PWV measurements were made. of variation for maximum IMT were 4.3 and 4.7%, respectively. Ambulatory BP monitoring The 24-h ambulatory BP monitoring was performed Urinary albumin excretion with an oscillometric-based device (TM-2431; A&D UAE was evaluated on the basis of the mean Co., Tokyo, Japan). BP was measured every 30 min albumin-to-creatinine ratio in three non-consecutive during the daytime period (between 0600 and 2200 overnight urine samples collected 1 week before

Journal of Human Hypertension Ambulatory BP and vascular damage A Ichihara et al 789 ambulatory BP monitoring was started. Urinary Table 1 Patient characteristics at baseline concentrations of albumin and creatinine were determined by turbidimetric immunoassay with a Characteristics Amlodipine Valsartan P-value group group Superior-Microalbumin kit (DPC Co., Tokyo, Japan) (n ¼ 50) (n ¼ 50) and by the Jaffe´ reaction with an autoanalyser. Age (years) 53.971.3 54.371.4 0.58 Male gender (n) 36 38 0.65 Statistical analyses BMI (kg/m2) 24.670.6 24.270.4 0.58 Analyses were performed with StatView 5.0 soft- Diabetes (n) 4 3 0.70 ware (SAS Institute Inc., Cary, NC, USA). Fisher’s Hyperlipidaemia (n) 10 9 0.80 exact test was used to analyse gender and the Renal disease (n) 1 1 1.00 Serum Cr (mg/dl) 0.8479.93 0.8670.03 0.15 frequency of the patients’ characteristics, except Serum UA (mg/dl) 6.871.2 6.270.2 0.64 age and body mass index, which were analysed Serum TC (mg/dl) 21175 21475 0.47 by the Mann–Whitney U-test. Differences in the Serum TG (mg/dl) 155714 147712 0.69 numbers of patients taking diuretics or b-blockers Serum HDL-C (mg/dl) 55715572 0.99 7 7 were tested by Fisher’s exact test. The changes in Serum LDL-C (mg/dl) 130 5 130 4 0.92 Serum CRP (mg/dl) 0.1470.04 0.1770.07 0.33 biological parameters were analysed by a two-way Plasma ARC (pg/ml) 7.170.7 8.671.4 0.43 analysis of variance for repeated measures combined with the Dunnett and Tukey–Kramer post hoc tests. Abbreviations: ARC, active renin concentration; BMI, body mass The contribution of changes in BP to changes in PWV index; Cr, creatinine; CRP, C-reactive protein; HDL-C, high-density was tested by analysis of covariance (ANCOVA). lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TC, total cholesterol; TG, triglyceride; UA, uric acid. Multiple regression analyses were also performed to Data are means7s.e.m. evaluate the association of PWV with SOD, IMT, UAE and 24-h or daytime systolic BP parameters. A P-value 0.05 was considered significant. Data o 4 are reported as means s.e.m. *

3 Results Patient characteristics 2 Clinic BP and 24-h ambulatory BP of 100 patients 7 7 in the present study averaged 154 1/97 1 and 1 15071/9371 mm Hg, respectively. Forty-two

patients had a mild hypertension defined as a clinic Serum SOD activity (U/ml) 0 systolic BP of o160 mm Hg and a clinic diastolic 012 BP of o100 mm Hg. Table 1 shows no significant Month differences in patient characteristics between the amlodipine group and valsartan group in any of Figure 1 Serum activity of SOD at baseline and after 12 months of treatment in the amlodipine group (open bars, n ¼ 50) and these baseline values, and no significant changes valsartan group (closed bars, n ¼ 50). *Po0.05 versus the base- were observed in either group during the 12-month line. observation period. However, as shown in Figure 1, serum SOD activity significantly increased from 2.970.2 to 3.470.2 U/ml in the valsartan group Clinic and ambulatory BP during the 12-month treatment period, whereas the Table 2 shows the changes between clinic BP and serum SOD activity of the amlodipine group 24-h, daytime and night time ambulatory BP in the averaged 3.070.2 U/ml and did not change signifi- amlodipine group and valsartan group at baseline cantly during the 12-month treatment period. The and after 12 months of treatment. Clinic systolic/ average doses of amlodipine and valsartan that the diastolic BP and 24-h, daytime and night time patients were taking at the end of the observation ambulatory systolic/diastolic BP significantly de- period were 7.070.3 and 105.875.5 mg/day, respec- creased during the 12-month treatment period in tively. Prescription of a diuretic was added on for both the amlodipine group and the valsartan group, three patients in the amlodipine group and 14 and the decreases in BP were similar in both groups. patients in the valsartan group, and the proportion However, BP variability assessed on the basis of of patients who received a diuretic in the valsartan the standard deviation of 24-h or daytime systolic group was greater than in the amlodipine group BP significantly decreased in the amlodipine group, (P ¼ 0.003). Prescription of a b-blocker was added on but remained unchanged in the valsartan group. The for eight patients in the amlodipine group and four decrease in BP variability observed in the amlodi- patients in the valsartan group, and the numbers pine group was statistically significant in compar- of patients who received a b-blocker in the amlodi- ison with the change in the valsartan group. The BP pine group and valsartan group were similar variability of night time systolic BP and of random (P ¼ 0.218). diastolic BP in the amlodipine group and valsartan

Journal of Human Hypertension Ambulatory BP and vascular damage A Ichihara et al 790 Table 2 BP before and after treatment 1800 1.0 1700 0.9 Amlodipine Valsartan group Intergroup 0.8 group (n ¼ 50) (n ¼ 50) 1600 0.7 * 1500 IMT (mm) 0.6 Baseline 12-m Baseline 12-m P-value PWV (cm/s) *

1400 Carotid maximum 0.5 012 012 Clinic SBP 15572 13472* 15372 13572* 0.34 Month Month Clinic DBP 96718371* 97718571* 0.85 24-h SBP 15172 13471* 14972 13572* 0.14 Its variability 22711871* 20712171 o0.01 50 24-h DBP 92718371* 94728471* 0.67 40 Its variability 17711571* 16711671 0.07 30 7 7 7 7 * Daytime SBP 156 2 138 2* 153 2 139 2* 0.08 20 Its variability 21711771* 19712071 0.01 o 10 Daytime DBP 94718671* 96728771* 0.93 UAE (mg/gCr) Its variability 1671157115711671 0.07 0 7 7 7 7 012 Night time SBP 132 2 118 2* 134 3 117 3* 0.33 Month Its variability 1571137114711471 0.23 Night time DBP 80727271* 83727272* 0.08 Figure 2 PWV, maximum IMT of the carotid arteries and UAE at Its variability 1071107110711171 0.28 baseline and after 12 months of treatment in the amlodipine group Nocturnal 2471207223721972 0.89 (open circles and bars, n ¼ 50) and valsartan group (closed circles decrease and bars, n ¼ 50). *Po0.05 versus the baseline. Morning surge 5773527356745073 0.74

Abbreviations: BP, blood pressure; DBP, diastolic blood pressure; SBP, systolic blood pressure. factors of clinic and ambulatory BP did not affect Data are means7s.e.m. *Po0.05 versus baseline. the decrease in PWV (data not shown). In addition, an interactive influence by the drugs was observed in the decrease in PWV owing to the decrease in variability of 24-h systolic BP and in the decrease group was similar. The nocturnal BP decrease and in PWV owing to the decrease in variability of day- morning BP surge in both the amlodipine group and time systolic BP. More specifically, the decreases in valsartan group tended to decrease during the 12- variability of 24-h and daytime ambulatory systolic month period, but the differences were not signifi- BP caused by amlodipine significantly contributed cant. The change in nocturnal BP decrease and the to the decrease in PWV of the hypertensive patients. change in morning BP surge were both similar in the Changes in the other clinical and biological para- amlodipine group and valsartan group. meters, including serum SOD activity, did not contribute to the changes in PWV. The ANCOVA analyses also showed that the Target-organ damage decreases in 24-h systolic BP, 24-h diastolic BP, The baseline PWV values in the amlodipine group night time systolic BP and night time diastolic BP and valsartan group were similar, averaging 17237 significantly contributed to the decrease in UAE 52 and 1671755 cm/s, respectively. Figure 2 shows during the 12-month treatment period, but the other that PWV significantly decreased to 1517738 cm/s factors of clinical or ambulatory BP had no effect in the amlodipine group and to 1489738 cm/s in on the decrease in UAE. Although the 12-month the valsartan group, and the decreases in PWV in treatment with valsartan, but not with amlodipine, both groups were also similar. However, 12 months significantly decreased UAE, no interactive influ- of treatment with amlodipine or valsartan did not ence by the drugs was observed in the decrease alter the maximum IMT of the carotid arteries, in UAE owing to the decrease in ambulatory BP. which averaged 0.9270.05 and 0.8870.04 mm, In other words, valsartan treatment decreased UAE respectively, at baseline. UAE in the amlodipine independently of its depressor effects. Changes in group and valsartan group averaged 33.3711.7 and other clinical and biological parameters, including 40.9710.0 mg/g of creatinine, respectively, and serum SOD activity, did not contribute to the significantly decreased to 18.674.9 mg/g of creatinine changes in UAE. with valsartan treatment, but remained unchanged with amlodipine treatment. Multiple regression analysis Multiple regression analysis for five explanatory Factors affecting PWV and UAE variables including 24-h (Table 4, R2 ¼ 0.190, P ¼ Table 3 shows that the decreases in clinic systolic 0.013) or daytime systolic BP parameters (Table 5, BP, clinic diastolic BP, 24-h systolic BP, variability of R2 ¼ 0.168, P ¼ 0.028) indicated significant associa- 24-h systolic BP and variability of daytime systolic tions between changes in PWV and changes in 24-h BP significantly contributed to the decrease in PWV or daytime systolic BP variability, especially in the during the 12-month treatment period, but that other amlodipine group. Changes in PWV were also asso-

Journal of Human Hypertension Ambulatory BP and vascular damage A Ichihara et al 791 Table 3 Effects of drugs and BP changes on PWV changes Table 4 Multiple regression analysis of changes in PWV versus five independent factors including changes in 24-h SBP ANCOVA Coefficient s.e. t-test P-value Independent Coefficient Lower 95% Upper 95% P-value Intercept À41.398 40.380 À1.025 0.3085 factor interval interval Group 24.161 40.380 0.598 0.5514 (amlodipine All patients versus Changes À25.334 À122.445 71.777 0.6043 valsartan) in SOD DClinic SBP 7.759 1.625 4.776 o0.0001 Changes 1.573 0.130 3.016 0.0331 Group  Dclinic 0.210 1.625 0.129 0.8974 in UAE SBP Changes in IMT 89.872 À145.598 325.343 0.4488 Mean of 2.021 À3.629 7.670 0.4778 Intercept À44.020 42.776 À1.029 0.3067 24-h SBP Group 39.123 42.776 0.915 0.3633 Variability 15.428 3.881 26.974 0.0096 (amlodipine of 24-h SBP versus valsartan) Amlodipine DClinic DBP 12.230 2.772 4.411 o0.0001 group Group  Dclinic 3.998 2.772 1.442 0.1534 Changes À14.025 À149.032 120.982 0.8336 DBP in SOD Changes 0.697 À1.760 3.154 0.5669 Intercept À105.378 48.578 À2.169 0.0331 in UAE Group 17.942 48.578 0.369 0.7129 Changes in IMT 98.323 À289.438 486.085 0.6087 (amlodipine Mean of 0.817 À8.861 10.496 0.8644 versus 24-h SBP valsartan) Variability 28.784 7.792 49.776 0.0088 D24-h SBP 5.530 2.466 2.243 0.0278 of 24-h SBP Group  D24-h 1.192 2.466 0.483 0.6302 SBP Valsartan group Changes À23.659 À203.850 156.531 0.7904 Intercept À153.049 30.181 À5.071 0.0001 in SOD Group 35.546 30.181 1.178 0.2425 Changes 2.050 À0.180 4.279 0.0703 (amlodipine in UAE versus Changes in IMT 62.258 À268.939 393.455 0.7038 valsartan) Mean of 0.615 À7.308 8.537 0.8752 DVariability of 16.083 5.045 3.188 0.0021 24-h SBP 24-h SBP Variability 6.829 À8.496 22.155 0.3700 Group  D 13.430 5.045 2.662 0.0095 of 24-h SBP variability of 24-h SBP Abbreviations: IMT, intima–media thickness; PWV, pulse wave velocity; SBP, systolic blood pressure; SOD, superoxide dismutase; Intercept À162.882 29.848 À5.457 0.0001 UAE, urinary albumin excretion. Group 25.666 29.848 0.860 0.3925 (amlodipine versus valsartan) treated with either amlodipine or valsartan, the DVariability of 14.059 4.893 2.874 0.0052 lowering of BP itself may play a more important role daytime SBP in improving arterial stiffness independent of the Group  D 10.508 4.893 2.148 0.0349 variability of class of antihypertensive drug used. However, long- daytime SBP term treatment with amlodipine significantly de- creased the variability of ambulatory BP, and the Abbreviations: ANCOVA, analysis of covariance; BP ¼ blood pressure; reduction contributed to the reduction in PWV. DBP ¼ diastolic blood pressure; PWV ¼ pulse wave velocity; SBP ¼ systolic blood pressure; s.e. ¼ standard error. Long-term treatment with valsartan significantly increased serum SOD activity and decreased UAE. The variability of BP is estimated as the standard ciated with changes in UAE in all patients treated deviation of non-invasively monitored ambulatory with amlodipine or valsartan. BP,21 and is associated with the arterial wall stiffness assessed by PWV22 and the cardiovascular morbidity and mortality of hypertensive patients.3,4,23 These Discussion findings suggested two possible cause–effect rela- tionships between BP variability and arterial stiff- The results of the present study showed that both ness, that is, the increase in BP variability may be amlodipine and valsartan significantly decreased attributable to the increased stiffness and decreased clinic BP, 24-h ambulatory BP and PWV, and that compliance of the large elastic arteries caused by the the decreases in the amlodipine group and valsartan elevated BP,24 or the increased BP variability may group were similar. As the decreases in 24-h ambu- be the cause of the increasing stiffness and decreas- latory systolic BP significantly contributed to the ing compliance of the large elastic arteries. How- decrease in PWV in the hypertensive patients ever, as the present study showed that BP lowering

Journal of Human Hypertension Ambulatory BP and vascular damage A Ichihara et al 792 Table 5 Multiple regression analysis of changes in PWV versus tensive patients,27 and angiotensin II receptor bloc- five independent factors including changes in daytime SBP kers had a beneficial effect on the arterial stiffness of hemodialysis patients with increased oxidative Independent Coefficient Lower 95% Upper 95% P-value 17 factor interval interval stress. As valsartan significantly increased serum SOD activity in the present study, we expected the All patients enhanced ability of valsartan to scavenge serum- Changes À18.986 À117.060 79.088 0.7004 reactive oxygen species to have a beneficial impact in SOD on the arterial wall, but the increase in serum SOD Changes 1.512 0.099 2.926 0.0364 activity was not associated with changes in PWV or in UAE Changes in IMT 86.894 À153.430 327.218 0.4730 UAE. A longer observation period may be required Mean of 1.820 À3.682 7.322 0.5114 to determine the benefit of the antioxidative effects daytime SBP of valsartan. Variability of 13.595 2.293 24.898 0.0191 Previous studies demonstrated that intensive BP daytime SBP lowering in hypertensive patients is associated with 16 Amlodipine group a significant decrease in PWV and a lower rate of Changes À17.609 À158.380 123.163 0.8003 cardiovascular events.28 Prescription of amlodipine in SOD of 47.5 mg/day was useful in successful long-term Changes 0.635 À1.903 3.172 0.6137 BP control with a target BP of o130/85 mm Hg, and in UAE Changes in IMT 59.870 À357.554 477.294 0.7718 the PWV of hypertensive patients receiving amlodi- 16 Mean of 1.407 À8.326 11.141 0.7701 pine decreased significantly. However, the reason daytime SBP why the PWV of hypertensive patients receiving Variability of 24.700 3.222 46.178 0.0256 short-acting calcium channel blockers failed to daytime SBP decrease, despite a significant decrease in clinic 16 Valsartan group BP, remained unclear. In the present study, amlo- Changes À16.717 À194.917 161.484 0.8494 dipine was found to significantly decrease BP in SOD variability and to intensively control BP throughout Changes 2.108 À0.053 4.268 0.0555 a 24-h period. As the decrease in BP variability was in UAE Changes in IMT 61.601 À267.822 391.025 0.7052 significantly associated with the decrease in PWV, Mean of 0.226 À7.294 7.746 0.9514 BP variability may play a key role in the success of daytime SBP calcium channel blockers in decreasing PWV. Variability of 6.942 À7.584 21.468 0.3369 Both 12-month treatment with amlodipine and daytime SBP valsartan did not improve IMT. As previous studies Abbreviations: IMT, intima–media thickness; PWV, pulse wave demonstrated that calcium antagonist lacidipine 29 velocity; SBP, systolic blood pressure; SOD, superoxide dismutase; slows down the progression of IMT, amlodipine UAE, urinary albumin excretion. and valsartan might inhibit the progression of IMT rather not influence IMT. Alternatively, IMT reflects structural changes in the wall of the internal carotid with valsartan significantly decreased PWV in the artery, although PWV reflects both structural absence of any reduction in BP variability, BP changes in the artery wall and its functional flexi- variability may be a causal, but not resultant, factor bility. Recent studies reported that the functional that affects the stiffness and compliance of the large improvement of arterial endothelium occurred in- elastic arteries. itially after 2 months of treatment with amlodipine30 Angiotensin II receptor blockers preferentially and after 6 weeks of treatment with valsartan.31 The dilate efferent arterioles,25 and thereby decrease decrease in PWV after 12 months of treatment with intraglomerular pressure, which leads to a reduction amlodipine and valsartan observed in the present in proteinuria. As a reduction in proteinuria has study may reflect the improved endothelial func- been observed at doses of angiotensin II receptor tion. A longer period of treatment may be needed for blockers that did not affect systemic BP in a rat IMT to improve. model of hypertension,26 the glomerular protection Multiple regression analyses showed significant provided by angiotensin II receptor blockers is correlations between changes in PWV and changes thought to be independent of their effects on in systolic BP variability, although the coefficients systemic BP. In the present study, the angiotensin of ANCOVA models including the variability of II receptor blocker valsartan decreased UAE, but the 24-h ambulatory systolic BP and the variability calcium channel blocker amlodipine, which pro- of daytime systolic BP might not be precise because duced a similar reduction in BP, did not. Thus, the of interactions. These results suggest that changes in decrease in UAE with valsartan may not be owing to systolic BP variability can impact on arterial wall its depressor effect, but owing to pleiotropic effects independently of changes in the mean of 24-h ambu- independent of its depressor effect. In addition, latory systolic BP levels. In addition, there was a valsartan has been reported to reduce the formation significant correlation between changes in PWV of reactive oxygen species to a greater extent than and changes in UAE in all patients treated with amlodipine in monocytes harvested from hyper- amlodipine or valsartan. As microalbuminuria may

Journal of Human Hypertension Ambulatory BP and vascular damage A Ichihara et al 793 represent the early renal manifestation of a wide- dynamic and non-hemodynamic factors can spread vascular dysfunction,32 the changes in UAE contribute to the progression of arterial stiffness. be accountable for the decrease in PWV in the Although amlodipine and valsartan similarly redu- treated hypertensive patients. However, as none of ced BP and PWV, beneficial effects of amlodipine the independent variables in the multiple regression therapy on arterial stiffness were mediated mainly analyses for the valsartan group are significantly through the reduction of BP variability, and those correlated with reductions in PWV, the exact mecha- effects of valsartan therapy were mediated possibly nisms involved in valsartan-dependent reductions through some pleiotropic effects. in PWV remain obscure. The Conduit Artery Function Evaluation study recently showed that amlodipine-based therapy Acknowledgements significantly decreased the central arterial pressure assessed by augmentation index compared to This work was supported in part by grants from the b-blocker-based therapy even if peripheral arterial Ministry of Education, Science and Culture of Japan pressure was similar in both groups.33 Thus, the (16790474 and 17390249). We also appreciate the decrease in central arterial pressure may contribute skilful secretarial work of Ms Yukiko Prendergast to the improvement of cardiovascular risk in and Ms Mayumi Shimada. hypertensive patients. In the present study, how- ever, augmentation index was not determined. References Further study is needed to determine whether there are any differences in the effects of amlodipine and 1 Julius S, Kjeldsen SE, Weber M, Brunner HR, Ekman S, valsartan on central arterial pressure. Hansson L et al. Outocomes in hypertensive patients at Although the numbers of patients who received high cardiovascular risk treated with regimens based a b-blocker in the amlodipine group and valsartan on valsartan or amlodipine: the VALUE randomised group were statistically similar, the proportion of trial. Lancet 2004; 363: 2022–2031. patients who received a diuretic in the valsartan 2 Perloff D, Sokolow M, Cowan R. The prognostic value of ambulatory blood pressures. JAMA 1983; 249: group was significantly greater than in the amlodi- 2792–2798. pine group. In combination with angiotensin II type 3 Parati G, Pomidossi G, Albini F, Malaspina D, Mancia 1 receptor blockers, therefore, the diuretics use was G. Relationship of 24-h blood pressure mean and not only useful in BP reduction but also might variability to severity of target-organ damage in contribute to the decreases in PWV and UAE. hypertension. J Hypertens 1987; 5: 93–98. In conclusion, both amlodipine and valsartan 4 Palatini P, Penzo M, Racioppa A, Zugno E, Guzzardi G, similarly controlled 24-h ambulatory BP in un- Anaclerio M et al. Clinical relevance of nighttime treated hypertensive patients and had improved blood pressure and of daytime blood pressure varia- the function of their stiffened arteries by the end of bility. Arch Intern Med 1992; 152: 1855–1860. 12 months of treatment. However, valsartan therapy 5 Ohkubo T, Imai Y, Tsuji I, Nagai K, Watanabe N, Minami N et al. Prediction of mortality by ambulatory was more effective on microalbuminuria and in- blood pressure monitoring versus screening blood creased serum SOD activity, although none of the pressure measurements: a pilot study in Ohasama. independent variables in the multiple regression J Hypertens 1997; 15: 357–364. analyses for the valsartan group are significantly 6 Staessen JA, Thijs L, Fagard R, O’Brien ET, Clement D, correlated with reductions in PWV. Some hemo- de Leeuw PW et al. Predicting cardiovascular risk using conventional vs ambulatory blood pressure in older patients with systolic hypertension. Systolic What is known about this topic: hypertension in Europe trial investigators. JAMA 1999; K In hypertensive patients already on antihypertensive drugs, 282: 539–546. amlodipine-based therapy has a greater beneficial effect in 7 Bjorklund K, Lind L, Zethelius B, Andren B, Lithell H. intensive control of blood pressure (BP) and lowering the Isolated ambulatory hypertension predicts cardiovas- 1 risk of myocardial infarction than valsartan-based therapy. cular morbidity in elderly men. Circulation 2003; 107: K Ambulatory BP is better-controlled throughout a 24-h 1297–1302. period by amlodipine than by valsartan,8 whereas valsartan exerts a greater reduction in arterial stiffness of 8 Eguchi K, Kario K, Hoshide Y, Hoshide S, Ishikawa J, hypertensive patients than a long-acting calcium channel Morinari M et al. Comparison of valsartan and blocker.13 amlodipine on ambulatory and morning blood pres- sure in hypertensive patients. Am J Hypertens 2004; What this study adds: 17: 112–117. K In untreated hypertensive patients, both amlodipine and 9 Blacher J, Asmar R, Djane S, London GM, Safar ME. valsartan similarly control 24-h ambulatory BP and improve Aortic pulse wave velocity as a marker of cardiovas- the function of their stiffened arteries. cular risk in hypertensive patients. Hypertension 1999; K Reduction in BP variability plays a significant role in the 33: 1111–1117. decrease in arterial stiffness with long-term treatment with amlodipine. 10 Laurent S, Boutouyrie P, Asmar R, Gautier I, Laloux B, K Long-term treatment with valsartan reduces arterial Guize L et al. Aortic stiffness is an independent stiffness through some pleiotropic effects, involving the predictor of all-cause and cardiovascular mortality decrease in UAE and increase in serum SOD activity. in hypertensive patients. Hypertension 2001; 37: 1236–1241.

Journal of Human Hypertension Ambulatory BP and vascular damage A Ichihara et al 794 11 Asmar RG, London GM, O’Rourke ME, Safar ME. 23 Kikuya M, Hozawa A, Ohokubo T, Tsuji I, Michimata Improvement in blood pressure, arterial stiffness and M, Matsubara M et al. Prognostic significance of blood wave reflections with a very-low-dose perindopril/ pressure and heart rate variabilities: the Ohasama indapamide combination in hypertensive patient. study. Hypertension 2000; 36: 901–906. A comparison with atenolol. Hypertension 2001; 38: 24 Imai Y, Aihara A, Ohkubo T, Nagai K, Tsuji I, Minami 922–926. N et al. Factors that affect blood pressure variability. 12 Boutouyrie P, Tropeano AI, Asmar R, Gautier I, Benetos A community-based study in Ohasama, Japan. Am J A, Lacolley P et al. Aortic stiffness is an independent Hypertens 1997; 10: 1281–1289. predictor of primary coronary events in hypertensive 25 Navar LG, Inscho EW, Majid DSA, Imig JD, Harrison- patients. A longitudinal study. Hypertension 2002; 39: Bernard LM, Mitchell KD. Paracrine regulation of the 10–15. renal microcirculation. Physiol Rev 1996; 76: 425–536. 13 Munakata M, Nagasaki A, Nunokawa T, Sakuma T, 26 Pugsley MK. The angiotensin-II (AT-II) receptor block- Kato H, Yoshinaga K et al. Effects of valsartan and er olmesartan reduces renal damage in animal models nifedipine coat-core on systemic arterial stiffness of hypertension and diabetes. Proc West Pharmacol in hypertensive patients. Am J Hypertens 2004; 17: Soc 2005; 48: 35–38. 1050–1055. 27 Yasunari K, Maeda K, Watanabe T, Nakamura M, 14 Niwa Y, Ishimoto K, Kanoh T. Induction of superoxide Yoshikawa J, Asada A. Comparative effects of valsartan dismutase in leukocytes by paraquat: correlation with versus amlodipine on left ventricular mass and reac- age and possible predictor of longevity. Blood 1990; 76: tive oxygen species formation by monocytes in hyper- 835–841. tensive patients with left ventricular hypertrophy. 15 Ichihara A, Hayashi M, Ryuzaki M, Handa M, J Am Coll Cardiol 2004; 43: 2116–2123. Furukawa T, Saruta T. Fluvastatin prevents develop- 28 Hansson L, Zanchetti A, Carruthers G, Dahlof B, ment of arterial stiffness in hemodialysis patients Elmfeldt D, Julius S et al. Effects of intensive blood- with type 2 diabetes mellitus. Nephrol Dial Transplant pressure lowering and low-dose aspirin in patients 2002; 17: 1513–1517. with hypertension: principal results of the Hyper- 16 Ichihara A, Hayashi M, Koura Y, Tada Y, Hirota N, tension Optimal Treatment (HOT) randomised trial. Saruta T. Long-term effects of intensive blood-pressure Lancet 1998; 351: 1755–1762. lowering on arterial wall stiffness in hypertensive 29 Zanchetti A, Bond MG, Hennig M, Neiss A, Mancia G, patients. Am J Hypertens 2003; 16: 959–965. Dal Palu C et al. Calcium antagonist lacidipine slows 17 Ichihara A, Hayashi M, Kaneshiro Y, Takemitsu T, down progression of asymptomatic carotid athero- Homma K, Kanno Y et al. Low doses of losartan and sclerosis: principal results of the European Lacidipine trandolapril improve arterial stiffness in hemodialysis Study on Atherosclerosis (ELSA), a randomized, patients. Am J Kidney Dis 2005; 45: 866–874. double-blind, long-term trial. Circulation 2002; 106: 18 Ichihara A, Hayashi M, Koura Y, Tada Y, Kaneshiro Y, 2422–2427. Saruta T. Long-term effects of statins on arterial 30 On YK, Kim CH, Oh BH, Lee MM, Park YB. Effects of pressure and stiffness of hypertensives. J Hum Hyper- angiotensin converting enzyme inhibitor and calcium tens 2005; 19: 103–109. antagonist on endothelial function in patients 19 Sun K, Daimon M, Watanabe S, Komuro I, Masuda Y. with essential hypertension. Hypertens Res 2002; 25: The relation of pulse wave velocities measured by 365–371. oscillometric and tonometric methods and clinical 31 Klingbeil AU, John S, Schneider MP, Jacobi J, Han- application studies. Jpn J Appl Physiol 2002; 32: drock R, Schmieder RE. Effect of AT1 receptor 81–86. blockade on endothelial function in essential hyper- 20 Yamashina A, Tomiyama H, Takeda K, Tsuda H, Arai T, tension. Am J Hypertens 2003; 16: 123–128. Hirose K et al. Validity, reproducibility, and clinical 32 Mule G, Cottone S, Vadala A, Volpe V, Mezzatesta G, significance of noninvasive brachial–ankle pulse Mongiovi R et al. Relationship between albumin wave velocity measurement. Hypertens Res 2002; 25: excretion rate and aortic stiffness in untreated essen- 359–364. tial hypertensive patients. J Intern Med 2004; 256: 21 Pickering TG, James GDJ. Ambulatory blood pressure 22–29. and prognosis. J Hypertens 1994; 12: S29–S33. 33 Williams B, Lacy PS, Thom SM, Cruickshank K, 22 Ichihara A, Kaneshiro Y, Takemitsu T, Sakoda M, Stanton A, Collier D et al. Differential impact of blood Hayashi M. Ambulatory blood pressure variability and pressure-lowering drugs on central aortic pressure and brachial–ankle pulse wave velocity in untreated clinical outcomes: principal results of the Conduit hypertensive patients. J Hum Hypertens 2006; 20(7): Artery Function Evaluation (CAFE) study. Circulation 529–536. 2006; 113: 1213–1225.

Journal of Human Hypertension