拜新同改善胰島素抵抗

1、AJH 2002; 15:927931Original ContributionsImprovement of Insulin Sensitivity By a Long-Acting Nifedipine Preparation (Nifedipine-CRin Patients With Essential HypertensionYuko Koyama, Kenichi Kodama, Masaaki Suzuki, and Yutaka HaranoBackground:Nifedipine has been reported to cause impairment of insuli

2、n sensitivity. But recently a con-trolled-released formulation of nifedipine (nifedipine-GITShas been reported that it could improve insulin sensitivity. Methods:We evaluated insulin sensitivity in two groups of essential hypertensive subjects before and after treatment with either long-acting nifed

3、ipine (nifedipine-CR, Adalat CR tablets; Bayer Yakuhin, Osaka, Japan (n 10 or metoprolol (n 9. Insulin sensitivity was evaluated from the steady-state plasma glucose (SSPGlevel measured at the steady-state insulin level (20to 30U/mLusing a modicationof the SSPG method previ-ously reported.nsulin res

4、istance has been reported in essential hy-pertension, 1but whether this resistance is reversible by antihypertensive treatment remains to be deter-mined. Treatment with vasodilating agents such as capto-pril 2and prazosin 3has been shown to improve insulin resistance. Most -adrenoreceptor antagonist

5、s decrease in-sulin sensitivity, but those with a vasodilating effect can improve insulin sensitivity. 4Regarding the calcium-chan-nel blockers, nifedipine treatment worsens glucose toler-ance in patients with non-insulin-dependent diabetes 5and in nondiabetic subjects, 6,7and also deteriorates both

6、 insu-lin sensitivity and secretion. 8Conversely, diltiazem and verapamil were reported to have no effect on insulin sen-sitivity. 9We previously demonstrated an improvement of insulin sensitivity during treatment with amlodipine 10or benidipine, 11long-acting calcium channel blockers. Sheu et al 12

7、showed that a controlled-released nifedipine (nifed-ipine-GITS could improve insulin sensitivity. To clarify whether another long-acting nifedipine preparation (nifed-ipine-CR, Adalat CR tablets; Bayer Yakuhin, Osaka, Ja-pan could restore insulin sensitivity, we measured insulin sensitivity in hyper

8、tensive subjects without diabetes or obesity who were treated with either nifedipine-CR or aReceived January 10, 2002. First decision February 21, 2002. Accepted June 20, 2002.From the Division of Atherosclerosis, Metabolism, and Clinical Nu-trition (YK,KK, MS, Department of Medicine, National Cardi

9、ovascular ©2002by the American Journal of Hypertension, Ltd. Published by Elsevier Science Inc.Results:The SSPG was initially high, but was signif-icantly reduced by nifedipine-CR treatment (from13314mg/dLto 958mg/dL.However, SSPG was not signicantlyaltered by treatment in the metoprolol group

10、(from10315mg/dLto 11912mg/dL.Conclusions:Our results indicate that the long-acting nifedipine (nifedipine-CRis associated with improved insulin sensitivity. Am J Hypertens 2002;15:927931©2002American Journal of Hypertension, Ltd. Key Words:Calcium channel blockers, nifedipine, hy-pertension, in

11、sulin resistance, human.I-blocker, metoprolol. We also evaluated the changes of free fatty acids (FFAsand ketone bodies during the insulin sensitivity test, as parameters of the action of insulin in liver or adipose tissue.MethodsThis investigation was carried out in 19nonobese outpa-tients with hyp

12、ertension and the results were compared with those obtained in seven healthy control subjects. The subjects had not been treated with any drugs for hyperten-sion, and they had no other endocrine or metabolic disor-ders and no hepatic or renal dysfunction.Using a mercury sphygmomanometer, systolic an

13、d di-astolic blood pressures (BPswere measured by detecting the Korotkoff phase I and V sounds, respectively. The mean of three BP measurements obtained on three differ-ent occasions with the subjects in the sitting position was calculated.Hypertension was denedby a systolic BP exceeding 140mm Hg or

14、 a diastolic BP that exceeded 90mm Hg, whereas normotensive subjects were denedas havingCenter Fujishirodai, Suita, Osaka, and Koshien University College of Nutrition (YH,Hyogo, Japan.Address correspondence and reprint requests to Dr. Yuko Koyama, 5-7-1, Fujishirodai Suita, Osaka 565-8565, Japan; e-

15、mail:yukohsp.ncvc.go.jp0895-7061/02/$22.00PII S0895-7061(0203019-4928IMPROVEMENT OF INSULIN SENSITIVITY BY A LONG-ACTING NIFEDIPINE AJH November 2002VOL. 15, NO. 11systolic and diastolic BP below 140and 90mm Hg, respectively.The control subjects were all nondiabetic, nonobese, and normotensive indiv

16、iduals.Hypertensive subjects were randomly assigned to treat-ment with either long-acting nifedipine (nifedipine-CR,40to 80mg once a day, n 10 or metoprolol (60to 120mg three times a day, n 9 for 12weeks. The target of treatment was to decrease the systolic BP below 140mm Hg or the diastolic BP belo

17、w 90mm Hg. Patients were requested not to change their eating habits and daily ac-tivities during therapy.Before and after treatment, the body weight was re-corded and blood was collected for the measurement of total cholesterol (TC,triglycerides (TG,high-density lipoprotein cholesterol (HDL-C,and a

18、polipoprotein B (ApoB. A standard 75-g oral glucose tolerance test (OGTTand an insulin sensitivity test were performed in all patients before and after treatment in the morning, after an overnight fast, before administration of drugs on that day. The area under the plasma concentration versus time c

19、urve (AUCduring OGTT was calculated for glucose and for insulin. Glucose tolerance was evaluated according to the criteria of the World Health Organization study group. 13Insulin Sensitivity TestWe modi ed the previously reported SSPG method using Sandostatin (octreotideacetate; Sandoz, Basal, Switz

20、er-land. 14Sandostatin was infused through an antecubital vein (125g as a bolus, followed by 100g/hand Novolin R insulin (NovoNordisk S/A,Tokyo, Japan, 7.5mU/kgwas also given as a bolus followed by constant infusion at a rate of 0.385mU/kg/min.Both drugs were infused for 2h. A glucose solution (3mg/

21、kg/minand KCl (0.25Eq/kg/minwere also infused through an antecu-bital vein using an infusion pump. Blood samples were collected at 0, 30, and 120min for the determination of glucose, insulin, TG, FFAs, and ketone bodies. Plasma epinephrine (Eand norepinephrine (NElevels were also determined at 0and

22、120min. The plasma glucose level at 120min (steady-stateplasma glucose SSPGwas used as an indicator of insulin resistance. The infusion rates for glucose, KCl, and insulin were half of those previously reported, 14and the steady-state insulin level reached 20to 30U/mL.Under these conditions, the pla

23、sma glucose level was inversely correlated with insulin sensitivity and a higher SSPG value meant greater insulin resistance. Plasma glucose was determined by the glucose oxidase method, and plasma insulin was measured by a two-antibody radioimmunoassay. 15The plasma concentrations of TC, 16TG, 17HD

24、L-C, 18Apo B, 19ketone bodies, 20FFAs, 21E, 22and NE 22were determined as described elsewhere. Body mass index (BMIwas calculated as theweight in kilograms divided by the height in square meters.The fractional excretion of Na (FENaduring the insu-lin sensitivity test was calculated according to the

25、follow-ing equation:FENa Na clearance/creatinineclearance. Informed consent was obtained from each subject. Statistical AnalysisResults are presented as the mean standard error. Prob-ability values .05(two-tailedwere de ned as indicating statistical signi cance. The signi cance of differences among

26、the control and hypertensive groups before treat-ment was determined by the Student t test. The male/female ratio was compared by the Fisher s exact test. The paired t test was performed to compare the data of hyper-tensive subjects before and after treatment. Responses of various parameters during

27、the insulin sensitivity test were analyzed using ANOVA and Student t test.ResultsBaseline CharacteristicsThe two hypertensive groups had signi cantly lower TG levels and higher HDL-C levels than the control group (Table1. Comparing the nifedipine-CR and metoprolol groups, there were no signi cant di

28、fferences in age, BMI, systolic BP, diastolic BP, heart rate, and the levels of TC, HDL-C, TG, glucose, insulin, and hemoglobin A1c. There were also no signi cant differences in AUC for glucose and insulin calculated from OGTT data.After treatment, systolic and diastolic BP were signif-icantly reduc

29、ed in both the nifedipine-CR and metoprolol groups and the reductions were 35.78.1/14.74.7mm Hg in the nifedipine-CR group and 16.94.9/8.71.8in the metoprolol group. There was no signi cant differ-ence between the two groups in the reduction of BP (systolicBP, P .09; diastolic BP, P .26.Both treatme

30、nts did not cause any signi cant changes of BMI, TC, TG, HDL-C, Apo B, glucose, insulin, HbA1c, AUC of glucose, and AUC of insulin. Effects of Nifedipine-CR orMetoprolol on Insulin SensitivityThe SSPG level was initially elevated in both the nifedi-pine-CR group and the metoprolol group when compare

31、d with the control group (13314mg/dLand 10315mg/dLv 827mg/dL,whereas there was no signi cant difference between the SSPG levels of the nifedipine-CR and metoprolol groups (Fig.1. The SSPG level was signi cantly decreased by treatment with nifedipine-CR (958mg/dL,but did not change after metoprolol t

32、herapy (11912mg/dL.All of the subjects in the nifedipine-CR group showed a decrease of SSPG, indicat-ing an improvement of insulin sensitivity. In contrast, the steady-state plasma insulin levels were not changed by treatment in both groups.AJH November 2002VOL. 15, NO. 11IMPROVEMENT OF INSULIN SENS

33、ITIVITY BY A LONG-ACTING NIFEDIPINE929Table 1. Baseline characteristics of each groupNifedipine-CR BeforeAge (ySex (M/FBody mass index (kg/m2 Systolic blood pressure (mmHgDiastolic blood pressure (mmHgHeart rate (beats/minTotal cholesterol (mg/dLHDL cholesterol (mg/dLTriglyceride (mg/dLFree fatty ac

34、ids (U/mLApoprotein B (mg/dLPlasma insulin (U/mLPlasma glucose (mg/dLHbA1c (%AUC of glucose (mg/dLh AUC of insulin (U/mLh HOMA-R61.84.43/722.40.5168.37.2*96.270.1201.259.387495.387.27.889.85.5267.183.91.87AfterMetoprolol Before 6353/622.81.8147.54.6*83.973.3186.65688.5576.682.96.273.84.7307.779.81.2

35、3AfterControl 60.32.23/422.50.6135.87.722.20.4132.64.1§3.5§3.57.36.29.957.4120.52.90.217.817.60.15241.2130.66.3§4.8*81.53.870.98.6193.33.9*618.87843.1462.38.57634.52.592.20.35.320.827411.487.20.82*0.994.9*75.28.856.816.4176.68.7*51.821124.4135.8394.810.461.217.518.283.40.15.215.7297.8

36、24.865.40.19*1.382.975.24.55*70.23.511.8212.410.56.148.14.133.9137.422.889.7493.362.213.192.312.21.64.11.111.9851.90.15.20.116.2284.424.119.275.611.40.191.090.17HDL high-density lipoprotein; HbA1c hemoglobin A1c; area under curve (AUCof glucose or insulin area under the plasma glucose orplasma insul

37、in versus time curve in the OGTT; HOMA-R homeostasis model assessment for insulin resistance index; OGTT oral glucose tolerance test.Values are the mean SE.*P .05; P .01for before nifedipine-CR or metoprolol treatment versus control by the t test; P .05; §P .01for before versus after nifedipine

38、-CR or metoprolol treatment by the t test.Changes of Serum FFAs and KetoneBodies During the Insulin Sensitivity Test Serum FFAs and ketone bodies showed a signi cant de-crease at 30min during the insulin sensitivity test both before and after treatments in all groups (Fig.2. There was a signi cant d

39、ifference in the FFAs response before and after nifedipine-CR treatment (P .05, ANOVA, with the reduction of FFAs at 30min being signi cantly greater after treatment than before treatment (P .05, t test. The reduction of ketone bodies was also greater after nifedipine-CR treatment than before treatm

40、ent, but the difference did not reach statistical signi cance (P .13.Changes of E, NE, and FENaDuring the Insulin Sensitivity TestThere were no signi cant changes of circulating E and NE levels or FENa after treatment with either drug (Table2. There were no signi cant changes of E, NE, and FENa valu

41、es during insulin sensitivity testing in the control or hypertensive subjects before and after treatment.DiscussionAdministration of long-acting nifedipine (nifedipine-CRfor 12weeks to subjects with hypertension was able to decrease the SSPG value. The observed paired decrease of SSPG in the 10hyper

42、tensive subjects is highly signi cant from our experience in a large number of subjects (500. This improvement of insulin sensitivity after nifedi-pine-CR treatment is valid even in the nonstatistical vari-ation of basal SSPG and BP levels between the two groups with these treatments and those with

43、metoprolol. In con-trast to the previous observation, 5the metoprolol treatment has not further reduced insulin sensitivity with the regular dose of administration. The clinical usefulness of -block-ers in subjects with cardiovascular disease has been well established and our data support this conce

44、pt from the standpoint of insulin sensitivity. Steady-state plasma insu-lin has been quite stable and no signi cant difference has been observed in the two treatment groups before and after the treatment. The half dose of insulin and glucose com-pared with the SSPG method previously reported is used

45、 in the present study to diminish the uid volume of infu-FIG. 1. Effects of nifedipine-CR and metoprolol on insulin sensitiv-ity. Values are the mean SE. P .01, P .05, P not significant (NSby the t test. SSPG steady-state plasma glucose; SSPI steady-state plasma insulin.930IMPROVEMENT OF INSULIN SEN

46、SITIVITY BY A LONG-ACTING NIFEDIPINE AJH November 2002VOL. 15, NO. 11FIG. 2. Changes of serum free fatty acids and ketone bodies in the insulin sensitivity test before and after nifedipine-CR or metoprolol treatment. Black circles before treatment; white circles after treatment. Values are the mean

47、SE. P value or NS for the response before versus after treatment by ANOVA. The bar graphs (black columns before treatment; white columns after treatment shows the reduction of free fatty acids or ketone bodies at 30min during the insulin sensitivity test. P value for before versus after treatment by

48、 the t test. Other abbreviation as in Fig. 1.sion. Although the data are not shown, continued infusion with arginine has been attempted.The reduction of FFAs at 30min during insulin sensi-tivity testing was greater after nifedipine-CR treatment, indicating the improvement of insulin sensitivity for

49、anti-lipolytic action.Homeostasis model assessment-insulin resistance index tended to decrease in the nifedipine-CR group but the change was not signi cant. The AUC of glucose and insulin did not change. Improvement of insulin sensitivity by nifedipine-CR may not have been large enough to in uence t

50、he insulin and glucose levels during OGTT. We have reported an improvement in insulin sensitivity after treatment with amlodipine 10and benidipine, 11which belong to the same class of dihydropyridine derivatives as nifedipine. Amlodipine associates and dissociates more slowly from the calcium channe

51、l than nifedipine, 23whereas benidipine binds to cell membrane binding sites from which it dissociates slowly, 24and both drugs are longer acting than nifedipine.Sheu et al 12reported that a controlled-released type nifedipine (nifedipine-GITScould improve insulin sensi-tivity. Nifedipine-CR is a ne

52、wly developed sustained-release for once-a day administration, which consists of a rapidly dissolving core surrounded by a slowly dissolving coat.Calcium channel blockers may inhibit insulin secretion in vitro. 8There was no change of the AUC for insulin in the OGTT results before and after nifedipi

53、ne-CR treat-ment, indicating that insulin secretion was not impaired in the present study.In this study, there was no elevation of cathecholamine levels or heart rate change during the insulin sensitivity test. The insulin-induced activation of the sympathetic nervous system was not observed in the

54、present study. An increased intracellular calcium level seems to be associated with insulin resistance in essential hyperten-sion. We have previously reported that the platelet free calcium concentration was positively correlated with the SSPG level in hypertension. 11Elevated intracellular cal-Tabl

55、e 2. Changes of epinephrine, norepinephrine, and fractional Na excretion during the insulin sensitivity test after nifedipine-CR or metoprolol treatmentNifedipine-CRMinEpinephrine (pmol/LNorepinephrine (pmol/LFractional excretion of Na012001200120Before 26.423.3463.8433.70.60.8After 6.830.14.828.913

56、2.7371.9106.6373.30.10.60.10.6MetoprololBeforeAfter Control6.848.520.153.510.219.45.625.570.3757.3405.9830.879.1492.4192.8411.50.20.50.20.70.11.40.71.010.729.86.24.920.24.125415.376.2185.8466.382.60.20.50.10.21.20.2AJH November 2002VOL. 15, NO. 11IMPROVEMENT OF INSULIN SENSITIVITY BY A LONG-ACTING N

57、IFEDIPINE931cium level and its suppression effect of insulin were re-ported in essential hypertensive 25and obese subjects. 26Suf cient insulin action corrects the intracellular calcium level due to activation of calcium adenosine triophos-phatase or Na-K adenosine triphosphatase, 27which are insuli

58、n sensitive. 28,29The vasodilative effect of nifedi-pine-CR may also enhance insulin sensitivity through im-provement of peripheral blood ow.Long-acting nifedipine treatment decreased the plasma TG and increased the HDL-C level, 12suggesting that improved insulin sensitivity could improve lipid metabo-lism. However, we found no changes of TC, TG, HDL-C, or Apo B after nifedipine-CR treatment.Increased renal sodium reabsorption was reported un-der supraphysiologic hyperinsulinemic conditions in the