Efonidipine Simultaneously Improves Blood Pressure, Endothelial Function, and Metabolic Parameters in Nondiabetic Patients with Hypertension

Emerging Treatments and Technologies BRIEF REPORT

Efonidipine Simultaneously Improves Blood Pressure, Endothelial Function, and Metabolic Parameters in Nondiabetic Patients With Hypertension

1 2 KWANG KON KOH, MD JEONG-A KIM, PHD underwent a 3-week washout period. At 2 1 MICHAEL J. QUON, MD, PHD WOOK-JIN CHUNG, MD the end of the washout period, partici- 1 4 SANG JIN LEE, MD YONGHEE LEE, PHD pants were randomly assigned to either 1 1 SEUNG HWAN HAN, MD EAK KYUN SHIN, MD 3 40–80 mg efonidipine or placebo daily JEONG YEAL AHN, MD during 8 weeks. Patients were then crossed over to the second treatment arm on completion of the first treatment arm ypertension is characterized by en- sure (9). Therefore, we hypothesized (without washout phase). The Green dothelial dysfunction and fre- that efonidipine therapy may simulta- Cross Pharmaceutical company (Yongin, H quently clusters with metabolic neously improve endothelial dysfunc- Korea) provided the identical placebo disorders that are characterized by insulin tion, adipocytokine profiles, and other (purchased by investigators). One patient resistance (1,2). These comorbidities may metabolic parameters in nondiabetic suffered from facial flushing and was be explained, in part, by reciprocal rela- patients with hypertension. withdrawn. Thus, data from 38 patients tionships between endothelial dysfunc- were analyzed. This study was approved tion and insulin resistance (1). By contrast RESEARCH DESIGN AND by the Gil Hospital Institute Review with calcium channel blockers (CCBs), METHODS — We evaluated effects of Board. treatment of hypertension with ␤-block- efonidipine in a randomized, double- Blood samples were obtained at ers and diuretics is associated with a blind, placebo-controlled, crossover 8:00 A.M. following an overnight fast be- higher risk of type 2 diabetes (3). This study. Thirty-nine hypertensive patients fore and after each treatment period. advantage of CCBs may relate to specific (systolic blood pressure [SBP] Ͻ180 Assays for plasma insulin, malondialde- mechanisms that target the vicious syn- mmHg and diastolic blood pressure hyde, adiponectin, leptin, and resistin ergy between endothelial dysfunction and [DBP] Ͻ110 mmHg) were considered el- were performed in duplicate by immu- insulin resistance. CCBs activate nitric ox- igible for this study. We excluded patients noradiometric assay or by enzyme- ide (NO) synthase in vitro and enhance with severe hypertension, unstable an- linked immunosorbent assay as NO production in vivo (4). This may im- gina, acute myocardial infarction, or renal previously described (11–13). Quanti- pact on the roles of adiponectin, leptin, insufficiency. None of our subjects were tative insulin sensitivity check index and resistin to influence metabolic sig- diabetic (based on history or criteria ac- (QUICKI) was calculated as described nals, inflammation, and atherosclerosis cording to the Report of the Expert Com- (14). Imaging studies of the right bra- (5–7). mittee on the Diagnosis and Classification chial artery were performed by ultra- Efonidipine hydrochloride is a 1,4- of Diabetes Mellitus [10]) or smokers. To sound as described (11–13). dihydropyridine–type CCB with long- minimize acute side effects, during an ini- Data are expressed as means Ϯ SEM lasting vasodilator actions and little tial run-in period, study medication was or median (range 25–75%). We used reflex tachycardia (8). Efonidipine im- titrated from 40 to 80 mg efonidipine up- paired Student’s t test or Wilcoxon’s proves endothelial function in patients wards over a 2-week period if no hypo- signed-rank test to compare relative with hypertension when compared with tension (SBP Ͻ100 mmHg) or changes in response to treatment. Pear- doses of nifedipine that result in com- hypertension (SBP Ͼ140 mmHg) was son’s or Spearman’s correlation coeffi- parable decreases in mean blood pres- noted. After the run-in period, all patients cient analysis was used to assess ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● associations between parameters. We cal- From the 1Department of Cardiology, Gil Heart Center, Gachon Medical School, Incheon, Korea; the culated that 30 subjects would provide 2Diabetes Unit, Laboratory of Clinical Investigation, National Center for Complementary and Alternative 80% power for detecting an absolute in- Medicine, National Institutes of Health, Bethesda, Maryland; the 3Department of Cardiology, Laboratory crease of Ն1.5% in flow-mediated dila- 4 Medicine, Gachon Medical School, Incheon, Korea; and the Department of Statistics, Ewha Womans tion of the brachial artery between University, Seoul, Korea. ␣ϭ Address correspondence and reprint requests to Kwang Kon Koh, MD, Professor of Medicine, Department placebo and efonidipine, with 0.05 of Cardiology, Gil Heart Center, Gachon Medical School, 1198 Kuwol-dong, Namdong-gu, Incheon, Korea (15). A value of P Ͻ 0.05 was considered 405-760. E-mail: [email protected]. to represent statistical significance. Received for publication 5 November 2006 and accepted in revised form 18 February 2007. Published ahead of print at http://care.diabetesjournals.org on 10 March 2007. DOI: 10.2337/dc06-2267. Abbreviations: CCB, calcium channel blocker; DBP, diastolic blood pressure; SBP, systolic blood pres- RESULTS — The mean age of our sub- sure; QUICKI, quantitative insulin sensitivity check index. jects was 46 Ϯ 2 years, and the male: A table elsewhere in this issue shows conventional and Syste`me International (SI) units and conversion female proportion was 21:17. Baseline factors for many substances. characteristics are reported in Table 1. No © 2007 by the American Diabetes Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby carryover effects were found (data are not marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. shown).

DIABETES CARE, VOLUME 30, NUMBER 6, JUNE 2007 1605 Vascular and metabolic effects of efonidipine

Table 1—Effects of efonidipine in 38 patients with hypertension

Percentage Variables Baseline Placebo Efonidipine changes BMI (kg/m2) 24.7 Ϯ 0.5 24.7 Ϯ 0.5 24.6 Ϯ 0.5 0.4 Ϯ 0.3 (2.0) Heart rate (bpm) 86 Ϯ 282Ϯ 2 (12) 84 Ϯ 2 (13) 4 Ϯ 3 (16) SBP (mmHg) 155 Ϯ 2 148 Ϯ 2 (15) 134 Ϯ 2 (14)* Ϫ9 Ϯ 1 (8) DBP (mmHg) 95 Ϯ 191Ϯ 1 (9) 83 Ϯ 2 (9)* Ϫ9 Ϯ 1 (8) Lipids (mg/dl) Total cholesterol 186 Ϯ 4 183 Ϯ 5 (31) 187 Ϯ 5 (29) 3 Ϯ 2 (12) Triglycerides 162 Ϯ 19 157 Ϯ 18 (112) 155 Ϯ 16 (101) 15 Ϯ 9 (54) LDL cholesterol 105 Ϯ 4 102 Ϯ 4 (22) 106 Ϯ 4 (25) 5 Ϯ 3 (16) HDL cholesterol 49 Ϯ 353Ϯ 3 (20) 50 Ϯ 2 (14) Ϫ3 Ϯ 3 (21) Vasomotor function Flow-mediated dilation (%) 4.28 Ϯ 0.22 5.42 Ϯ 0.26 (1.58) 6.20 Ϯ 0.25 (1.52)* 21 Ϯ 7 (41) Nitroglycerin dilation (%) 13.87 Ϯ 0.70 14.62 Ϯ 0.68 (4.21) 14.84 Ϯ 0.80 (4.90) 2 Ϯ 4 (23) Malondialdehyde (␮mol/l) 1.20 Ϯ 0.04 1.22 Ϯ 0.04 (0.25) 1.12 Ϯ 0.05 (0.31)† Ϫ8 Ϯ 3 (18) C-reactive protein (mg/l) 0.80 (0.50–1.40) 0.85 (0.40–1.40) 0.65 (0.50–1.30) 25 Ϯ 18 (109) Insulin resistance Adiponectin (␮g/ml) 4.3 Ϯ 0.6 4.2 Ϯ 0.6 (3.7) 4.6 Ϯ 0.6 (3.9)† 15 Ϯ 4 (23) Insulin (␮U/ml) 7.33 (4.89–12.05) 7.63 (4.91–11.25) 6.78 (4.50–9.20) 3 Ϯ 8 (52) Glucose (mg/dl) 103 Ϯ 2 102 Ϯ 2 (12) 100 Ϯ 2 (15) Ϫ2 Ϯ 2 (10) QUICKI 0.360 Ϯ 0.010 0.351 Ϯ 0.006 (0.039) 0.359 Ϯ 0.007 (0.040) 3 Ϯ 2 (13) Leptin (ng/ml) 5.2 Ϯ 0.6 5.2 Ϯ 0.6 (3.6) 4.7 Ϯ 0.5 (3.3)† Ϫ12 Ϯ 4 (26) Resistin (ng/ml) 7.66 (5.81–10.34) 7.76 (5.83–10.41) 8.12 (5.31–9.99)* Ϫ1 Ϯ 6 (39) Data are expressed as means Ϯ SEM (SD) or median (25–75th percentile). *P Ͻ 0.001 vs. placebo. †P Ͻ 0.05 vs. placebo. QUICKI ϭ 1/͓log (insulin) ϩ log (glucose)͔ (ref. 14).

When compared with placebo, efo- adiponectin and HDL cholesterol levels function in patients with hypertension (9) nidipine therapy reduced SBP and DBP by (r ϭ 0.434, P ϭ 0.006) and QUICKI (r ϭ when compared with nifedipine. Finally, 9 Ϯ 1% (P Ͻ 0.001) and 9 Ϯ 1% (P Ͻ 0.379, P ϭ 0.019) following efonidipine urinary excretion of 8-hydroxy-2Ј- 0.001), respectively. When compared therapy. In a multiple regression model, deoxyguanosine and serum malondialde- with placebo, efonidipine improved flow- percentage changes in adiponectin levels hyde–modified LDL are both decreased mediated dilator response to hyperemia following efonidipine therapy persisted as by efonidipine, but not nifedipine, ther- by 21 Ϯ 7% (P Ͻ 0.001) and reduced an independent predictor of percentage apy. Although we did not directly com- plasma malondialdehyde levels by 8 Ϯ changes in HDL cholesterol levels (␤ϭ pare efonidipine with other CCBs in the 3% (P ϭ 0.011). 0.459, P ϭ 0.006) and QUICKI (␤ϭ current study, it will be of interest to do so There were positive correlations be- 0.467, P ϭ 0.067). in future studies. tween baseline adiponectin and HDL cho- Following efonidipine therapy, im- Potential mechanisms for CCBs to in- lesterol levels (r ϭ 0.533, P Ͻ 0.001), as provement in flow-mediated dilation was fluence insulin sensitivity may relate to well as between baseline BMI and baseline correlated with percentage changes in their ability to target the vicious synergy adiponectin (r ϭϪ0.507, P ϭ 0.001) or plasma levels of malondialdehyde (r ϭ between endothelial dysfunction and in- leptin (r ϭ 0.508, P ϭ 0.001) levels. Ϫ0.479, P ϭ 0.002), leptin (r ϭϪ0.424, sulin resistance. Therefore, we assessed When compared with placebo, efonidi- P ϭ 0.008), insulin (r ϭϪ0.354, P ϭ metabolic parameters including plasma pine therapy increased plasma adiponec- 0.029), and QUICKI (r ϭ 0.471, P ϭ levels of lipids, adiponectin, leptin, resistin levels by 15 Ϯ 4% (P ϭ 0.013) and 0.003). Improvement in flow-mediated tin, and QUICKI. Efonidipine had a neu- decreased plasma leptin and resistin lev- dilation persisted as an independent pre- tral metabolic effect with respect to the els by 12 Ϯ 4% (P ϭ 0.030) and 1 Ϯ 6% dictor of percentage changes in malondi- lipid profile and QUICKI. However, im- (P ϭ 0.001), respectively. Insulin sensi- aldehyde (␤ϭϪ0.822, P ϭ 0.017) and provement in flow-mediated dilation per- tivity assessed by QUICKI did not signif- QUICKI (␤ϭ1.032, P ϭ 0.034). Follow- sisted as an independent predictor of icantly change (increase of 3 Ϯ 2%, P ϭ ing efonidipine therapy, there were signif- changes in QUICKI. 0.239). Plasma resistin levels were not icant correlations between percentage Amlodipine has no significant effect correlated with either insulin sensitivity changes in plasma levels of malondialde- on adiponectin levels in patients with hy- or BMI. There were no significant corre- hyde and leptin (r ϭ 0.364, P ϭ 0.025). pertension (17). In our study, efonidipine lations between percentage changes in increased adiponectin levels without a adiponectin and percentage changes in CONCLUSIONS — Efonidipine has corresponding change in BMI. Increasing leptin or resistin levels following efo- distinct properties when compared with adiponectin levels is predicted to improve nidipine therapy (Ϫ0.054 Յ r Յ other CCBs. Efonidipine has higher affin- both insulin sensitivity and endothelial ϩ Ϫ0.030). However, we did observe cor- ity for T-type Ca2 channels (16) and a function by multiple mechanisms (18). relations between percentage changes in larger effect in improving endothelial Regulation of metabolic homeostasis and

1606 DIABETES CARE, VOLUME 30, NUMBER 6, JUNE 2007 Koh and Associates hemodynamic homeostasis may be cou- Shintani Y, Minamino T, Kitamura S, Lee Y, Shin EK: Anti-inflammatory and pled by vascular actions of insulin to stim- Tomoike H, Hori M, Kitakaze M: A cal- metabolic effects of candesartan in hyper- cium channel blocker activates both tensive patients. Int J Cardiol 108:96–100, ulate production of NO (19). In the Ј current study, changes in adiponectin lev- ecto-5( )-nucleotidase and NO synthase 2006 els persisted as an independent predictor in HUVEC. Biochem Biophys Res Com- 13. Koh KK, Quon MJ, Han SH, Chung W-J, mun 311:625–628, 2003 Ahn JY, Kim J, Kang WC, Lee Y, Shin EK: of changes in HDL cholesterol levels and 5. Han SH, Quon MJ, Kim JA, Koh KK: Adi- Additive beneficial effects of fenofibrate QUICKI. Effects of efonidipine to reduce ponectin and cardiovascular disease: re- combined with candesartan in the treat- plasma leptin and malondialdehyde lev- sponse to therapeutic interventions. JAm ment of hypertriglyceridemic, hyperten- els and improve endothelium-dependent Coll Cardiol 49:531–538, 2007 sive patients. Diabetes Care 29:195–201, dilation are significantly correlated. In 6. Correia ML, Haynes WG: Leptin, obesity 2006 summary, efonidipine therapy simulta- and cardiovascular disease. Curr Opin 14. Katz A, Nambi SS, Mather K, Baron AD, neously improves blood pressure, endo- Nephrol Hypertens 13:215–223, 2004 Follmann DA, Sullivan G, Quon MJ: thelial function, and metabolic 7. Verma S, Li SH, Wang CH, Fedak PW, Li Quantitative insulin sensitivity check in- parameters without substantially altering RK, Weisel RD, Mickle DA: Resistin pro- dex: a simple, accurate method for assess- insulin sensitivity in nondiabetic patients motes endothelial cell activation: further ing insulin sensitivity in humans. J Clin evidence of adipokine-endothelial inter- Endocrinol Metab 85:2402–2410, 2000 with hypertension. action. Circulation 108:736–740, 2003 15. Koh KK, Ahn JY, Han SH, Kim DS, Jin DK, 8. Honda M, Hayashi K, Matsuda H, Kubota Kim HS, Shin MS, Ahn TH, Choi IS, Shin E, Tokuyama H, Okubo K, Takamatsu I, EK: Pleiotropic effects of angiotensin II Acknowledgments— This study was sup- Ozawa Y, Saruta T: Divergent renal vaso- receptor blocker in hypertensive patients. ported by grants from an established inves- dilator action of L- and T-type calcium J Am Coll Cardiol 42:905–910, 2003 tigator award (2005-1) from the Gil Medical antagonists in vivo. J Hypertens 19:2031– 16. Lee TS, Kaku T, Takebayashi S, Uchino T, Center, Gachon Medical School, Incheon, 2037, 2001 Miyamoto S, Hadama T, Perez-Reyes E, Korea. 9. Oshima T, Ozono R, Yano Y, Higashi Y, Ono K: Actions of mibefradil, efonidipine Teragawa H, Miho N, Ishida T, Ishida M, and nifedipine block of recombinant T- Yoshizumi M, Kambe M: Beneficial effect and L-type Ca channels with distinct in- References of T-type calcium channel blockers on en- hibitory mechanisms. Pharmacology 78: 1. Kim J, Montagnani M, Koh KK, Quon MJ: dothelial function in patients with essen- 11–20, 2006 Reciprocal relationships between insulin tial hypertension. Hypertens Res 28:889– 17. Watanabe S, Okura T, Kurata M, Irita J, resistance and endothelial dysfunction: 894, 2005 Manabe S, Miyoshi K, Fukuoka T, Mu- molecular and pathophysiological mech- 10. The Expert Committee on the Diagnosis rakami K, Higaki J: The effect of losartan anisms. Circulation 113:1888–1904, and Classification of Diabetes Mellitus: and amlodipine on serum adiponectin in 2006 Report of the Expert Committee on the Japanese adults with essential hyperten- 2. Garcia-Puig J, Ruilope LM, Luque M, Fer- Diagnosis and Classification of Diabetes sion. Clin Ther 28:1677–1685, 2006 nandez J, Ortega R, Dal-Re R, AVANT Mellitus. Diabetes Care 21 (Suppl. 1):S5– 18. Koh KK, Han SH, Quon MJ: Inflammatory Study Group Investigators: Glucose me- S22, 1998 markers and the metabolic syndrome: in- tabolism in patients with essential hyper- 11. Koh KK, Quon MJ, Han SH, Ahn JY, Jin sights from therapeutic interventions. JAm tension. Am J Med 119:318–326, 2006 DK, Kim HS, Kim DS, Shin EK: Vascular Coll Cardiol 46:1978–1985, 2005 3. Taylor EN, Hu FB, Curhan GC: Antihy- and metabolic effects of combined ther- 19. Chen H, Montagnani M, Funahashi T, pertensive medications and the risk of in- apy with ramipril and simvastatin in pa- Shimomura I, Quon MJ: Adiponectin cident type 2 diabetes. Diabetes Care 29: tients with type 2 diabetes. Hypertension stimulates production of nitric oxide in 1065–1070, 2006 45:1088–1093, 2005 vascular endothelial cells. J Biol Chem 278: 4. Asano Y, Kim J, Ogai A, Takashima S, 12. Koh KK, Quon MJ, Han SH, Chung W-J, 45021–45026, 2003

DIABETES CARE, VOLUME 30, NUMBER 6, JUNE 2007 1607