Effects of Valsartan Alone Versus Valsartan/Simvastatin Combination on Ambulatory Blood Pressure, C-Reactive Protein, Lipoproteins, and Monocyte Chemoattractant Protein–1 in Patients With Hyperlipidemia and Hypertension
Article Outline
Angiotensin receptor blockers have been hypothesized to have synergistic effects with statins. We evaluated the effects of valsartan alone or combined with simvastatin on blood pressure (BP) and indexes of inflammation and oxidant stress in hypertensive patients with hyperlipidemia. In this double-blind trial, 404 patients were randomized to 12 weeks valsartan 160 mg (V) or valsartan 160 mg plus simvastatin 20 mg (V/S20) or 80 mg (V/S80). Twenty-four–hour mean ambulatory BP and biochemical marker measurements were recorded at baseline and study end. There were no statistically significant between-treatment differences for least-square mean reductions from baseline in systolic BP (V, −9.22; V/S20, −9.25; V/S80, −9.58 mm Hg; p <0.0001 for all within-treatment changes vs baseline). Plasma high-sensitivity C-reactive protein decreased with the combinations but not with V alone (least-square mean median change from baseline, −0.16, −0.20, −0.70 mg/L; p = 0.0001 for V/S80 vs baseline; p = 0.045 for V/S20 vs baseline; p = 0.0023 for V/S80 vs V/S20; p = 0.0045 for V/S80 vs V). Monocyte chemoattractant protein–1 was reduced by V, with no evidence for additional lowering with V/S combinations. In conclusion, addition of simvastatin to valsartan did not incrementally lower BP. However, V/S80 was superior to V and V/S20 in reducing high-sensitivity C-reactive protein.
It has been postulated that treatment targeting the renin-angiotensin-aldosterone system in combination with a 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor augments the antihypertensive effects of the angiotensin receptor blocker while simultaneously preventing the progression of atherosclerosis.1, 2, 3, 4 We evaluated the effects of valsartan combined with simvastatin on blood pressure (BP), lipid levels, selected markers of inflammation, and oxidative stress in hypertensive patients with hyperlipidemia. The long-term efficacy and tolerability of the combination was further evaluated in an open-label extension of this study.
Methods
This randomized, double-blind study was conducted at 47 centers in the United States, Canada, and France. Eligible patients at visit 0 entered a 2- to 4-week washout period followed by a single-blind placebo run-in, after which they were randomized to double-blind, active therapy (Figure 1). Valsartan was supplied in 160-mg capsules (V) and simvastatin in identical 20 mg (S20) and 80 mg (S80) capsules. The first 4 weeks after randomization were a forced-titration period, during which the V/S80 treatment group received V and S20, and the other 2 groups received the same dose as their respective final doses (i.e., V and V/S20). During weeks 5 through 12, all treatment groups received their final planned doses. In addition to other antihypertensive and lipid-lowering drugs, concomitant administration of medications known to significantly affect lipids and/or BP and agents with known interactions with statins were not permitted. Men and women aged ≥50 years with hypertension and hyperlipidemia were enrolled in the study. Systolic hypertension was defined as a mean sitting systolic BP ≥150 and <180 mm Hg measured by a calibrated sphygmomanometer. Hyperlipidemia was defined as plasma low-density lipoprotein (LDL) level ≥3.36 mmol/L (≥130 mg/dl) and <4.91 mmol/L (<190 mg/dl) and triglyceride level ≤4.52 mmol/L (≤400 mg/dl) despite dietary therapy and off-medication before randomization. Women had to be postmenopausal for 1 year, surgically sterile, or using effective nonhormonal contraceptive methods. Patients unable to safely discontinue all other previous antihypertensive and lipid-lowering drugs for the 16- to 18-week study period were excluded, as were those with any contraindication to valsartan, simvastatin, or other lipid-lowering drugs. Another key exclusion criterion was and/or severe hypertension (systolic BP ≥180 mm Hg or diastolic BP ≥110 mm Hg). After initial assessment at visit 0, patients were evaluated at 5 planned study visits with visits occurring at monthly intervals after randomization. All adverse events were recorded. Ambulatory BP was measured twice using a Spacelabs System (model 90207; Spacelabs Medical, Issaquah, Washington) at the end of the placebo run-in 1 day before randomization with placebo administered and after 12 weeks of randomized treatment. Standard procedures were established for ambulatory BP readings by a core laboratory. For determination of lipid and biochemical markers, fasting blood samples were drawn and processed at designated core laboratories. Serum high-sensitivity C-reactive protein (hs-CRP) was measured using particle-enhanced immunologic agglutination (Tina-quant; Roche Diagnostics, Indianapolis, Indiana), monocyte chemoattractant protein–1 by direct sandwich enzyme-linked immunosorbent assay (R&D Diagnostics, Minneapolis, Minnesota), and plasma F2 isoprostanes by enzyme-linked immunosorbent assay (Cayman Chemical Company, Ann Arbor, Michigan). The primary end point was the mean ambulatory systolic BP over 24 hours. Secondary end points included mean ambulatory diastolic BP (24 hours), mean sitting systolic and diastolic BP (measured with a sphygmomanometer), mean ambulatory systolic and diastolic BP for daytime and nighttime, LDL and high-density lipoprotein (HDL) levels, total cholesterol and triglyceride levels, and biochemical markers. After completion of the main study, eligible patients entered a 52-week open-label extension study. Patients were assessed for BP and lipid levels at weeks 4, 13, 26, 39, and 52.
Figure 1.
Study design and flow of patients through the trial. HCTZ = hydrochlorothiazide.
Sample size was estimated with 80% power at a 5% significance level to detect a difference of 5.5 mm Hg in the primary end point (change from baseline in 24-hour mean ambulatory systolic BP) using an estimate of 12.5 mm Hg for the SD of change from baseline in mean ambulatory systolic BP.1 A total of 300 patients (100 patients per treatment arm) were required, with a total of 354 patients planned to be randomized (15% dropout rate). Primary efficacy analyses were performed on the intent-to-treat population. The intent-to-treat population comprised all randomized patients with a baseline and ≥1 efficacy measurement after baseline. The safety population comprised all randomized patients who received ≥1 dose of double-blind study medication. For the primary end point, 2 pairwise comparisons were performed. The study tested the null hypothesis that the effects of valsartan alone would be the same as those of V/S20 and V/S80 versus the alternative hypotheses that they would not, using an analysis of covariance model. Treatment comparability was examined using Cochran-Mantel-Haenszel chi-square tests with raw mean scores for qualitative variables and using 1-way analysis of variance F-tests for quantitative variables.
Results
Figure 1 illustrates the study design and flow of patients through the trial. The most common reasons for discontinuation in the single-blind, placebo run-in were abnormal laboratory values (n = 99), abnormal test results (n = 71), and condition no longer requiring study drug (n = 37). In the double-blind phase, the most common reasons for discontinuation were unsatisfactory therapeutic effect (3.6%, 1.5%, and 4.5% of patients in the V, V/S20, and V/S80 groups, respectively), adverse events (3.6%, 1.5%, and 2.3%), and abnormal laboratory values (2.9%, 3.0%, and 0.8%). The 3 treatment groups were well matched for baseline demographic characteristics, BP, and lipid levels (Table 1). The groups were also well balanced in terms of concomitant medical conditions and associated treatments. Medications most commonly used during the study were aspirin (24.4% to 26.3% of patients across treatment groups) and anilides (primarily paracetamol; 18.2% to 24.4% of patients). Mean sitting BP levels were consistently higher than 24-hour ambulatory readings.
Table 1. Baseline demographics and characteristics (randomized population except where indicated)
| Parameter | Valsartan | V/S20 | V/S80 | Total |
|---|---|---|---|---|
| Age (yrs) | 62.9 ±8.1 | 63.0±8.2 | 64.6±8.6 | 63.5±8.3 |
| Sex (men/women) | 71/66 | 69/66 | 72/60 | 212/192 |
| Race, n (%) | ||||
| Caucasian | 125(91.2) | 121(89.6) | 120(90.9) | 366(90.6) |
| African-American | 12(8.8) | 12(8.9) | 10(7.6) | 34(8.4) |
| Body mass index (kg/m2) | 30.3±5.2 | 29.0±5.2 | 28.9±4.6 | 29.4±5.1 |
| BP (mm Hg) | ||||
| 24-h mean ambulatory | ||||
| Systolic BP⁎ | 142.1±13.2 | 142.3±12.5 | 143.9±12.7 | 142.8±12.8 |
| 24-h mean ambulatory | ||||
| Diastolic BP⁎ | 83.9±9.0 | 84.8±8.6 | 83.7±9.7 | 84.1±9.1 |
| Mean sitting systolic BP | 157.7±8.3 | 158.0±8.4 | 157.8±8.1 | 157.9±8.2 |
| Mean sitting diastolic BP | 92.0±7.9 | 93.4±8.1 | 91.3±8.1 | 92.2±8.0 |
| Lipids (mmol/L) | ||||
| LDL | 4.01±0.63 | 4.14±0.73 | 4.13±0.60 | 4.09±0.65 |
| HDL | 1.32±0.38 | 1.43±0.43 | 1.37±0.41 | 1.37±0.41 |
| Total cholesterol | 6.24±0.81 | 6.42±0.90 | 6.38±0.72 | 6.35±0.82 |
| Triglycerides | 1.89±0.94 | 1.76±0.84 | 1.82±0.83 | 1.82±0.87 |
| Biomarkers (median) | ||||
| hs-CRP (mg/L) | 3.0 | 2.2 | 3.4 | 2.8 |
| MCP-1 (ng/L) | 138.0 | 146.7 | 135.0 | 140.5 |
| F2 Isoprostanes (μg/L) | 6.3 | 7.2 | 7.2 | 7.0 |
⁎Values for 24-hour mean ambulatory systolic BP and 24-hour mean ambulatory diastolic BP are for the intent-to-treat population and not randomized population. |
Figure 2 depicts the least-square mean change from baseline in 24-hour mean ambulatory systolic BP and mean ambulatory diastolic BP for the intent-to-treat population. Although within-treatment changes were statistically significant versus baseline (p <0.0001), there were no statistically significant between-group differences for least-square mean change from baseline in 24-hour mean ambulatory systolic BP and mean ambulatory diastolic BP at end point (week 12). A similar pattern was seen for within- and between-treatment changes from baseline in daytime and nighttime mean ambulatory systolic BP and mean ambulatory diastolic BP as well as mean sitting systolic BP and mean sitting diastolic BP for all treatment groups for the intent-to-treat population (Table 2). Significant within-treatment decreases from baseline in LDL, total cholesterol, and triglyceride levels were reported in both V/S groups (p <0.0001) but not the V group. These differences were significant compared with V alone (p <0.0001; Table 3). Treatment with V was associated with a small, but significant, increase in mean triglyceride levels (p = 0.0008). The percent reductions in LDL, total cholesterol, and triglycerides in the V/S80 group were significantly greater than that in patients treated with V/S20 (p <0.0001 for LDL and total cholesterol; p = 0.01 for triglycerides), whereas increases in mean HDL from baseline in the V/S80 group were statistically significant compared with baseline (p = 0.0093) and V alone (p = 0.025).
Figure 2.
Between-treatment comparisons for least-square mean changes from baseline in (A) 24-hour mean ambulatory systolic BP and (B) 24-hour mean ambulatory diastolic BP at the end point (week 12; intent-to-treat population). p = NS for between-treatment comparisons of least-square mean changes from baseline; p <0.0001 for within-treatment changes from baseline. n = number of patients with ambulatory BP values obtained over 24 hours at baseline and end point.
Table 2. Least-square mean changes from baseline in BP at end point (week 12; intent-to-treat population)
| BP Parameter | Valsartan (n = 107) | V/S20 (n = 109) | V/S80 (n = 105) |
|---|---|---|---|
| 24-hour | |||
| Mean ambulatory systolic BP (mm Hg) | −9.22 | −9.25 | −9.58 |
| Mean ambulatory diastolic BP (mm Hg) | −5.46 | −5.52 | −5.99 |
| Daytime | |||
| Mean ambulatory systolic BP (mm Hg) | −9.55 | −9.67 | −10.04 |
| Mean ambulatory diastolic BP (mm Hg) | −5.77 | −5.78 | −6.25 |
| Nighttime | |||
| Mean ambulatory systolic BP (mm Hg) | −8.43 | −8.43 | −8.20 |
| Mean ambulatory diastolic BP (mm Hg) | −4.88 | −4.82 | −5.16 |
| Clinic sitting BP | n=135 | n=130 | n=127 |
| Mean sitting systolic BP (mm Hg) | −17.57 | −18.80 | −17.72 |
| Mean sitting diastolic BP (mm Hg) | −7.76 | −8.51 | −8.01 |
Table 3. Least-square mean changes from baseline in lipid levels at end point (week 12; intent-to-treat population)
| Lipid | Valsartan (n = 133) | V/S20 (n = 131) | V/S80 (n = 127) |
|---|---|---|---|
| LDL (mmol/L) | 2.73 | −34.80⁎ | −47.19⁎,† |
| HDL (mmol/L) | −1.49 | 0.44 | 3.14⁎ |
| Total cholesterol (mmol/L) | 1.87 | −24.49⁎ | −33.32⁎† |
| Triglycerides (mmol/L) | 10.20 | −10.69⁎ | −20.43⁎† |
⁎p <0.05 versus V. |
†p <0.05 versus V/S20. |
Both V/S treatment regimens produced significant reductions from baseline in hs-CRP (V/S20, p = 0.045; V/S80, p = 0.0001). The degree of reduction of hs-CRP was greater with V/S80 than with V/S20 (p = 0.0023) or with V alone (p = 0.0045; Figure 3). Treatment on all 3 arms resulted in reductions from baseline in monocyte chemoattractant protein–1 levels (V, p = 0.0329; V/S20, p = 0.001; V/S80, p = 0.0167), but there were no differences between groups (Figure 3). Median changes from baseline in plasma F2-isoprostane levels approached statistical significance for V alone (p = 0.052 vs baseline) but not for the 2 combination groups (changes of 1.0, 0.5, and 0.2 μg/L in the V, V/S20, and V/S80 groups, respectively). The effects of V and the combination of V/S on the other biochemical markers were variable.
Figure 3.
Between-treatment median changes from baseline in (A) hs-CRP and (B) monocyte chemoattractant protein–1 at end point (week 12; intent-to-treat population). ap = 0.0045 versus V; bp = 0.0023 versus V/S20. n = number of patients with hs-CRP and monocyte chemoattractant protein–1 values obtained at baseline and end point.
Relatively few patients in the V, V/S20, and V/S80 groups experienced increases in creatine kinase of >300% from baseline (1.5%, 3.1%, and 3.1%, respectively), increases in aspartate aminotransferase of >150% (0%, 2.3%, and 1.6%), or increases in alanine aminotransferase of >150% (1.5%, 4.6%, and 4.7%). In total, 17 patients had adverse events that resulted in discontinuation of study medication (in the V/S groups, 2 for gastrointestinal symptoms, 2 for elevated BP, and 1 each for elevated LDL, aminotransferases, cerebral tumor, muscle cramps, and myopathy; in the V-alone group, 3 for gastrointestinal symptoms and 1 each for increased LDL, creatine kinase, headache/vertigo, and palpitations). There were no deaths in the trial.
The open-label extension phase included 210 patients and most (n = 195) continued to receive V/S20 for the duration of the study. Reductions in mean sitting systolic and diastolic BP achieved during the core study were maintained during the long-term extension. Patients who received V/S80 exhibited a greater reduction in LDL (0.63 mmol/L [24.2 mg/dl]; 13.1% additional reduction at end point) than patients who received V/S20 (0.32 mmol/L [12.3 mg/dl]; 1.2% additional reduction), perhaps reflecting the fact that patients in this group increased their simvastatin dose to greater than the treatment assignment in the double-blind phase. A similar pattern was observed for total cholesterol. However, triglyceride levels remained stable in the V/S20 and V/S80 groups (additional median reduction of 0 mmol/L from end-of-core baseline at end point in both treatment groups).
Discussion
This trial showed that the V/S combination has no incremental effect on 24-hour ambulatory BP compared with V alone, that V plus high-dose simvastatin lowers hs-CRP more effectively than V alone or a combination of V plus low-dose simvastatin, and that monocyte chemoattractant protein–1 is decreased by V alone, but there is no incremental decrease with a V/S combination. The scientific basis and design of this study was based on previous nonrandomized and single-center trials that demonstrated an independent effect of short-term simvastatin therapy on BP.1, 2, 3, 5, 6 The latter studies employed sitting clinic BP in small cohorts of patients and were neither designed nor sufficiently powered to address the impact of the combination on small changes in systolic BP. To our knowledge, our study is the largest trial that has employed standardized ambulatory BP to resolve this issue. Our population (older and with significantly more hyperlipidemia) was chosen to increase the probability of a significant effect, as previous data seem to suggest a preferential effect of simvastatin on BP in hypertensive patients and a greater effect in the presence of concomitant hyperlipidemia.2 Despite these design considerations, the combination at any dose did not incrementally change BP over a 12-week period. It is possible that higher doses of V may have been required. It is also possible that the combination had important effects on central BP without significant changes in peripheral (i.e., brachial) BP.7 Despite no synergistic impact on BP, V/S does appear to modulate parameters of inflammation.8 The reductions in hs-CRP with V/S are consistent with published data that suggest that simvastatin reduces hs-CRP.9, 10
In conclusion, V combined with high-dose simvastatin has no incremental impact on ambulatory BP but appears to be well tolerated and reduces hs-CRP. The long-term impact of these findings needs to be investigated in larger studies.
Acknowledgments
We thank Eckhard Bauer for assistance with the trial. Editorial support for this manuscript was provided by Ian Woolveridge and Sharon Smalley, medical writers at ACUMED, Cheshire, United Kingdom, with funding from Novartis Pharma AG, East Hanover, New Jersey.
References
- . Effect of the HMG-CoA reductase inhibitors on blood pressure in patients with essential hypertension and primary hypercholesterolemia. Hypertension. 1999;34:1281–1286
- . Statins and blood pressure regulation. Curr Hypertens Rep. 2001;3:281–288
- . Additive beneficial effects of losartan combined with simvastatin in the treatment of hypercholesterolemic, hypertensive patients. Circulation. 2004;110:3687–3692
- . Endothelial Protection, AT1 blockade and Cholesterol-Dependent Oxidative Stress: the EPAS trial. Circulation. 2006;114:I296–I301
- . Effect of fluvastatin on serum lipid levels in essential hypertension. Orv Hetil. 1996;137:1857–1859
- . Intensive cholesterol reduction lowers blood pressure and large artery stiffness in isolated systolic hypertension. J Am Coll Cardiol. 2002;39:1020–1025
- . Large artery stiffness and antihypertensive agents. Curr Pharm Des. 2005;11:3317–3326
- . Antiinflammatory effects of angiotensin II subtype 1 receptor blockade in hypertensive patients with microinflammation. Circulation. 2004;110:1103–1107
- . Inflammation as a cardiovascular risk factor. Circulation. 2004;109:II2–II10
- . High-sensitivity C-reactive protein and cardiovascular risk: rationale for screening and primary prevention. Am J Cardiol. 2003;92:17K–22K
PII: S0002-9149(07)00710-2
doi:10.1016/j.amjcard.2007.02.085
© 2007 Elsevier Inc. All rights reserved.
