Improved insulin sensitivity with the angiotensin II-receptor ... - Nature

Journal of Human Hypertension (2006) 20, 860–866 & 2006 Nature Publishing Group All rights reserved 0950-9240/06 $30.00 www.nature.com/jhh

ORIGINAL ARTICLE

Improved insulin sensitivity with the angiotensin II-receptor blocker losartan in patients with hypertension and other cardiovascular risk factors TA Aksnes1, HM Reims2, S Guptha3, A Moan4, I Os5 and SE Kjeldsen6 1

Department of Cardiology, Ullevaal University Hospital, Oslo, Norway; 2Department of Cardiology, Ullevaal University Hospital, Oslo, Norway; 3Merck & Co., Whitehouse Station, New Jersey, USA; 4MSD AS, Drammen, Norway; 5Department of Nephrology, Ullevaal University Hospital, Oslo, Norway and 6 Department of Cardiology, Ullevaal University Hospital, Oslo, Norway

We aimed to compare the effects of two different vasodilating principles, angiotensin II-receptor blockade and calcium channel blockade, on peripheral insulinmediated glucose uptake in patients with hypertension and other cardiovascular risk factors. Twenty-one hypertensive patients (11 women and 10 men) with mean age 58.6 years (range 46–75 years), body mass index 29.271.0 kg/m2 and blood pressure 16073/ 9672 mm Hg entered a 4-week run-in period with open-label amlodipine 5 mg. Thereafter they were randomized double-blindly to additional treatment with amlodipine 5 mg or losartan 100 mg. After 8 weeks of treatment, all patients underwent clinical examination and laboratory testing, and 17 of them underwent a hyperinsulinaemic isoglycaemic glucose clamp. After

a 4-week open-label wash-out phase, the participants crossed over to the opposite treatment regimen and final examinations with hyperinsulinaemic isoglycaemic glucose clamp after another 8 weeks. Blood pressure was lowered to the same level in both treatment periods. The glucose disposal rate was significantly higher after treatment with losartan 100 mg þ amlodipine 5 mg compared to amlodipine 10 mg (4.970.4 vs 4.270.5 mg/kg/ min, P ¼ 0.039). Thus our data suggest that angiotensin II-receptor blockade with losartan improves glucose metabolism at the cellular level beyond what can be expected by the vasodilatation and blood pressure reduction alone. Journal of Human Hypertension (2006) 20, 860–866. doi:10.1038/sj.jhh.1002087; published online 21 September 2006

Keywords: angiotensin II type I receptor blockers; calcium channel blockers; diabetes mellitus; hypertension; insulin

resistance

Introduction Patients with hypertension have an increased prevalence of insulin resistance and an increased risk of developing diabetes mellitus with ageing.1 As high blood pressure is encountered in more than 20% of the adult population2 and its management is a priority in preventing cardiovascular complications,3 antihypertensive strategies which also attenuate the trend towards diabetes mellitus might have considerable clinical significance. Data from cohort and randomized trials suggest that the incidence of type II diabetes mellitus is unchanged or increased by thiazides and b-adrenergic blockers, whereas it appears to be unchanged or decreased by

Correspondence: Dr TA Aksnes, Department of Cardiology, Ullevaal University Hospital, N-0407 Oslo, Norway. E-mail: [email protected] Received 20 March 2006; revised 19 June 2006; accepted 19 July 2006; published online 21 September 2006

angiotensin-converting enzyme inhibitors (ACEIs), angiotensin II-receptor blockers (ARBs) and calcium channel blockers (CCBs).4–6 Furthermore, the Valsartan Antihypertensive Long-Term Use Evaluation (VALUE) trial7 recently observed a 23% reduction in development of type II diabetes mellitus with ARBbased compared with CCB-based treatment in nondiabetic hypertensive subjects. It thus appears that treatment with ARBs/ACEIs may prevent new-onset diabetes mellitus better than treatment with CCBs, which is probably neutral in this aspect. The mechanism for the differential effects of the two different vasodilating agents is, however, unclear, as both ARBs and CCBs lower blood pressure by reducing peripheral resistance. To our knowledge, no double-blind, randomized study has previously compared the effects of ARB and CCB on insulin sensitivity using the hyperinsulinaemic isoglycaemic glucose clamp in patients with essential hypertension. We hypothesized that ARB therapy may have an additive effect and improve insulin

Losartan and insulin sensitivity TA Aksnes et al 861

sensitivity more than CCB therapy alone at a comparable dose with regards to the blood pressure-lowering effect.

Methods Study population

Twenty-five patients with mild-to-moderate essential hypertension (office diastolic blood pressure 95–110 mm Hg and systolic blood pressure o180 mm Hg) were recruited from general practitioners in the city of Oslo. The participants were previously untreated for hypertension or treated with monotherapy, but not with ACEI or ARB. All had impaired glucose tolerance or impaired fasting glucose defined as fasting plasma glucose 6.1– 7.0 mmol/l (110–126 mg/dl). The participants also had to have either microalbuminuria (urine albumin excretion rate X20 mg/min), dyslipidaemia (highdensity lipoprotein (HDL)-cholesterol o0.9 mmol/l (35 mg/dl) or triglycerides 41.7 mmol/l (150 mg/ dl)), body mass index 428 kg/m2 or an increased waist-to-hip ratio (40.9 for men, 40.85 for women). At the first visit a clinical examination and laboratory testing with electrolytes, creatinine and thyroid hormones were taken to screen for secondary hypertension. Owing to side effects of the study medication, mainly ankle oedema, but also headaches, flushing and palpitation, three patients decided to withdraw from the study in the run-in period and are not included in the final analysis. One of the patients who completed the study was excluded from analysis owing to an error at the hospital pharmacy (the patient was given amlodipine 10 mg throughout). Thus, the final study population consisted of 21 subjects (Figure 1), 11 women and 10 men. The mean age was 58.6 years (range 46–75 years).

At inclusion, blood pressure averaged 16073/ 9672 mm Hg and heart rate 6672 beats/min. Body mass index was 29.271.0 kg/m2 in the whole study group and the waist-to-hip ratio was 0.9270.01 in the women and 1.0570.01 in the men. Four patients (19%) were previously untreated for their hypertension and seven (33%) were previously treated with thiazides, four (19%) with b-blockers and six (29%) with a CCB. Five (24%) subjects were smokers. Four of the patients did not complete two glucose clamps owing to technical problems during the clamp procedure.

Study design

The present double-blind, randomized cross-over study was designed to compare the metabolic effects of 10 mg amlodipine and 100 mg losartan þ 5 mg amlodipine (Figure 2). After a 4-week open-label amlodipine 5 mg run-in period, all hypertensive patients were randomized to additional treatment with either amlodipine 5 mg or losartan 100 mg for 8 weeks. At the end of this 8-week treatment period, patients underwent blood pressure measurement, blood sampling and a hyperinsulinaemic isoglycaemic glucose clamp. Following this was a 4-week wash-out phase where they continued open-label 5 mg amlodipine, and then they were crossed-over to the opposite treatment regimen for another 8 weeks before the final examination with blood pressure measurement, blood sampling and hyperinsulinaemic isoglycaemic glucose clamp. The study was approved by the National Committees for Research Ethics in Norway and the Norwegian Medicines Agency, and the patients’ verbal and written informed consent to participate was obtained from each patient before inclusion in the study.

Figure 1 Study population. Journal of Human Hypertension


Figure 2 The double-blind randomized crossover design. All patients used open-label amlodipine 5 mg during the whole trial, and they were randomized double-blindly to additional treatment with either an extra amlodipine 5 mg or losartan 100 mg for the 8-week treatment periods (V1 ¼ visit 1 or at baseline, V2 ¼ visit 2 etc., PE ¼ physical examination, I/E ¼ inclusion/exclusion criteria, BP ¼ blood pressure).

Hyperinsulinaemic isoglycaemic glucose clamp technique

The hyperinsulinaemic isoglycaemic glucose clamp was performed after an overnight fast. Antecubital veins on the right and left arm were cannulated with short Teflon catheters (Optiva 2, 18G; Medex Medical Ltd., Haslingden, UK). Both catheters were kept open with intermittent infusion of isotonic saline, the total amount not exceeding 100 ml. The hyperinsulinaemic isoglycaemic glucose clamp was preformed using a modification of the method described by DeFronzo et al.8 The insulin infusion was prepared in a bag with 100 ml of 0.9% saline. To prevent insulin from adhering to the plastics, 4 ml of saline was exchanged with 4 ml of whole blood. To the 100 ml bag, 30 IE of Insulin Actrapid were added and the bag was shaken well. The mixture was then drawn into a 50 ml syringe and infused at a fixed rate of 0.001 IE/kg body weight/min. The fasting blood glucose level was determined as the average of three measurements. Glucose and insulin were infused through one catheter, whereas the other was used for blood sampling. The insulin infusion was kept unchanged during the clamp. Infusion of glucose 200 mg/ml was started after 5 min at a rate of 20 ml/h, and was adjusted every 5 min according to the blood glucose level. Blood samples were taken every 5 min for the determination of blood glucose concentration, and at baseline and after 30, 60, 90 and 120 min of glucose clamping for the determination of serum insulin. The clamp was performed for 120 min. Insulin sensitivity was expressed as the glucose disposal rate (GDR) (mg/kg/min), calculated from the average glucose infusion rate during the last 20 min. This technique for measuring insulin Journal of Human Hypertension

sensitivity has a coefficient of variation of 5% in our laboratory, as previously detailed.9,10 Blood pressure measurement and analytical methods

Blood pressure was measured with a mercury sphygmomanometer with adequate cuff size and after 5 min rest in sitting position. The pressure was measured at least three times and the values registered were the mean of the two latest measurements. Blood glucose during the hyperinsulinaemic isoglycaemic glucose clamp was measured with an Accu-Chek Sensor (Roche Diagnostics GmbH, Mannheim, Germany). Serum insulin and C-peptide concentrations were measured using enzyme immunoassays (DAKO Insulin and DakoCytomation Cpeptide. DakoCytomation Ltd, UK). For analysing glycated haemoglobin (HbA1c) we used Roche Tinaquant immunoassay. The homeostasis model assessment for insulin resistance (HOMA-IR) was calculated in fasting conditions as serum glucose (mmol/l) multiplied by serum insulin (pmol/l) and divided with 135, as described by Matthews et al.11 Statistical analysis

The sample size needed was calculated based on earlier studies of Moan et al.12 of the effect of losartan on glucose metabolism and insulin sensitivity in hypertensive patients. They had standard errors of mean (s.e.m.) ¼ 0.5 for 10 patients, so standard deviation (s.d.) ¼ 1.6. In a 2 2 crossover design, if assuming correlation ¼ 0.5, then the within-subject mean square error is 2.5/2 ¼ 1.25. Assuming alpha ¼ 0.05 and power ¼ 80%, then the


total sample size required is 22 patients (11 per sequence) if delta ¼ 1. We used SPSS 12.0.1 (SPSS, Chicago, US) software for data management and statistical analysis. Changes over time were analysed using pairedsamples t test. Variables with a skewed distribution were analysed after logarithmical transformation, and were back-transformed to natural units for presentation in text and tables. Non-parametric analysis with Wilcoxon-matched pairs signed rank sum test were used for non-normally distributed data. A two-tailed P-valueo0.05 was considered the limit of statistical significance. All values are presented as mean7s.e.m. unless stated otherwise. Carry-over effects may appear with the crossover design and we therefore used a wash-out period of 4 weeks between the 8 weeks treatment periods in order to minimize carry-over effects and a possible carry-over effect is considered in the results.

min in the patients given the losartan treatment regimen in the second cross-over period. Other assessments of the glucose homeostasis

There was no significant difference in the calculated HOMA-IR between the treatment regimens. However, HOMA-IR was significantly lower during the losartan treatment period and nonsignificantly lower during the amlodipine treatment period, compared to the first visit (Table 1). There was no significant difference between treatment with amlodipine 10 mg and losartan 100 mg þ amlodipine 5 mg, with regards to fasting serum insulin as shown in Table 1. However, fasting serum insulin was significantly lower after treatment with losartan 100 mg þ amlodipine 5 mg compared to the first visit, whereas there was a nonsignificant reduction of serum insulin after treatment with amlodipine 10 mg. Fasting C-peptide was significantly reduced

Results Insulin sensitivity

The GDR was significantly higher during treatment with losartan 100 mg þ amlodipine 5 mg compared to amlodipine 10 mg (4.970.4 vs 4.270.5 mg/kg/min, P ¼ 0.039) (Figure 3). Nine of the 17 patients who successfully completed two glucose clamp examinations were treated with amlodipine in their first cross-over period. GDR on amlodipine 10 mg treatment was the same for the patients randomized to this treatment in their first cross-over period (4.270.8 mg/kg/min, n ¼ 9) and those randomized this in the last cross-over period (4.270.6 mg/kg/min, n ¼ 8). The GDR on losartan 100 mg þ amlodipine 5 mg was 4.570.4 mg/kg/min in the patients given the losartan treatment regimen in the first cross-over period and 5.270.6 mg/kg/

Figure 3 Insulin sensitivity assessed by the hyperinsulinaemic isoglycaemic glucose clamp in 17 hypertensive patients treated with amlodipine 10 mg or losartan 100 mg þ amlodipine 5 mg.

Table 1 Metabolic data at inclusion (visit 1) and after 8 weeks treatment of the two different regimens (n ¼ 21) Visit 1

Amlodipine 10 mg

Losartan 100 mg+Amlodipine 5 mg

P* HOMA-IR Insulin (pmol/l) C-peptide (pmol/l) Glucose (mmol/l) HbA1c (%) Total cholesterol (mmol/l) HDL-cholesterol (mmol/l) LDL-cholesterol (mmol/l) Triglycerides (mmol/l) Uric acid (mmol/l) Urine albumin: creatinine ratio (mg/mmol) BMI (kg/m2)

4.470.8 94.4716.1 1243.87150.2 6.170.3 5.770.1 6.070.3 1.570.1 3.870.3 1.5970.14 364720 1.1470.55 29.271.0

3.170.6 67.5713.3 934.6792.7 6.070.3 5.670.1 5.670.2 1.570.1 3.670.2 1.3070.16 315715 1.2670.66 29.271.0

NS NS 0.032 NS NS 0.041 NS NS 0.036 0.001 NS NS

2.870.5 59.878.1 944.8786.2 5.970.3 5.570.1 5.470.2 1.470.1 3.470.2 1.5970.31 306720 0.8270.30 29.071.0

P*

Pw

0.007 0.009 0.023 NS NS 0.002 0.036 0.002 NS 0.002 NS NS

NS NS NS NS NS NS NS NS NS NS NS NS

Abbreviations: BMI, body mass index; GDR, glucose disposal rate; HbA1c, glycated haemoglobin; HDL, high-density lipoprotein; HOMA-IR, Homeostasis model assessment for insulin resistance; I, insulin; LDL, low-density lipoprotein. P*, P-value for differences between visit 1 and study-treatment. Pw, P-value for differences between treatment with amlodipine 10 mg and losartan 100 mg+amlodipine 5 mg. Journal of Human Hypertension


during both treatment periods compared inclusion into the study, but there was no significant difference between the two treatment regimens. There were no significant differences in fasting glucose and HbA1c between treatment with amlodipine 10 mg and losartan 100 mg þ amlodipine 5 mg (Table 1). Blood pressure and heart rate

There were no significant differences in blood pressure between treatment with amlodipine 10 mg (14172/8871 mm Hg) and losartan 100 mg þ amlodipine 5 mg (14373/8871 mm Hg). But highly significant reductions in systolic and diastolic blood pressure were observed in both treatment regimens from inclusion in the study (16073/9672 mm Hg, Po0.001). There were no significant differences in resting heart rate during the visits. The heart rate was 6672, 6872 and 6972 beats/min at visit 1 and when treated with amlodipine 10 mg and losartan 100 mg þ amlodipine 5 mg, respectively. Serum lipids

There were no significant changes in total serum cholesterol, HDL-cholesterol, low-density lipoprotein (LDL)-cholesterol and triglycerides between the amlodipine and losartan treatment periods (Table 1). However, total serum cholesterol was significantly reduced from the time of inclusion into the study to treatment with both amlodipine 10 mg and losartan 100 mg þ amlodipine 5 mg (Table 1). There was a significant reduction of LDL-cholesterol from inclusion to treatment with losartan 100 mg þ amlodipine 5 mg, but not from inclusion to treatment with amlodipine 10 mg. Conversely, there was a significant reduction of triglycerides from inclusion to treatment with amlodipine 10 mg, but not to the losartan regimen. Uric acid

There were no significant changes in serum uric acid between the amlodipine and losartan treatment groups. However, serum uric acid was significantly reduced from baseline to treatment with both the amlodipine (D 49713 mmol/l) and treatment regimen losartan (D 58716 mmol/l) (Table 1).

Discussion Diabetes mellitus, and in particular type II diabetes, has emerged as a major health problem and tends to cluster with hypertension in individuals at high risk of cardiovascular disease.13 In this study we found that in hypertensive patients with a high risk of developing diabetes mellitus, treatment with the ARB losartan significantly improved insulin sensitivity in skeletal muscle as assessed by the glucose clamp technique. The glucose clamp technique is Journal of Human Hypertension

considered to be the ‘gold standard’ in assessing insulin sensitivity.8,14,15 In this method, a high steady circulating insulin level is maintained by constant insulin infusion and the blood glucose concentration is frequently measured and maintained at the fasting level (isoglycaemic), or at a predetermined (euglycaemic) level, by infusion of glucose at a variable rate. The amount of glucose needed to maintain isoglycaemia per unit of time is an index of sensitivity to the administered insulin. Insulin-mediated glucose uptake during the hyperinsulinaemic isoglycaemic glucose clamp occurs mainly in skeletal muscle,8 and is in part dependent on muscular blood flow. Thus, an increase in blood flow induced by vasodilating drugs would be expected to increase glucose uptake. The observed effect in the present study was beyond the vascular effects and blood pressure reduction achieved by maximal calcium channel blockade with amlodipine, indicating a non-haemodynamic effect of angiotensin II-receptor blocking with losartan on insulin sensitivity. We chose amlodipine as the comparator for losartan, as it is considered to have a neutral effect on insulin sensitivity,4 though studies have shown inconsistent results. In fact in earlier clamp studies16 comparing amlodipine with other antihypertensive drugs in parallel group study designs, insulin sensitivity was significantly improved from baseline after treatment with amlodipine. In our study the patients were hypertensive and placebo was not an option. By using another vasodilating agent as a comparator, we were able to study the metabolic effect of the ARB, beyond the vasodilating effect of a CCB. We chose 5 mg amlodipine as openlabel throughout the study, and added amlodipine 5 mg or losartan 100 mg (blinded) in a randomized cross-over design as shown in Figure 2. The dosage of 100 mg losartan was used in order to achieve an optimal angiotensin II-receptor blocking effect. The additional 5 mg of amlodipine in the other crossover arm compared to losartan 100 mg was given in order to avoid a difference in blood pressure between the regimens. As our aim was to do a direct comparison of the two different vasodilating principles, and in order to minimize the strain on the participants, we chose not to conduct the glucose clamp at baseline. The crossover design has been widely used in similar contexts. One of its advantages is the increased statistical power obtained by within-subject comparison rather than group comparison. A disadvantage of the design is that every patient has to complete both treatment periods and all examinations, and unfortunately we lost four patients owing to technical problems during one of the two clamp procedures. The two treatment periods were separated by a 4-week wash-out period intended to minimize carry-over effects. The order in which the drugs were given apparently had no effect on GDR during treatment with amlodipine 10 mg. However,


GDR on losartan 100 mg þ amlodipine 5 mg was higher in the patients who were given losartan in the last crossover period. The number of patients is too small to prove statistical significance (eight patients got losartan in the first cross-over period, and nine in the last period), but regardless the results do not suggest any carry-over effects. CCBs are considered to be neutral in their effects on glucose homeostasis. A recent meta-analysis17 even suggests that in hypertension, treatment with CCB is associated with lower rates of new-onset diabetes mellitus in comparison to diuretics and badrenergic blockers. Valsartan, another ARB, has been shown to be associated with lower rates of new-onset diabetes mellitus compared to amlodipine in the VALUE trial,7 suggesting that blockade of the renin–angiotensin system affects glucose metabolism separately from the blood pressure lowering effect. Angiotensin II in excess may impair endothelial function and promote vascular disease. Conversely, angiotensin II-receptor blockade may improve endothelial function and structural abnormalities in hypertensive patients.18 Thus, a reasonable explanation for the observed prevention of new-onset diabetes mellitus may lie in the improvement of microcirculatory flow and a better delivery of glucose and insulin to skeletal muscle.19 An alternative, and possible additional, explanation is an improvement of glucose homeostasis at the cellular level through agonistic interaction with peroxisome proliferator-activated receptor-gamma.20,21 In an experimental model, saralasin, a nonselective angiotensin II-antagonist, has been shown to affect pancreatic b-cells through increased pancreatic islet blood flow.22 It has also been proposed that blockade of the renin–angiotensin system may promote the recruitment and differentiation of adipocytes. This would counteract the ectopic deposition of lipids in other tissues like liver, muscle and pancreas, and thereby improve insulin sensitivity and prevent type II diabetes mellitus.23 However, assessment of a potential effect on insulin secretion was beyond the scope of the present study. Insulin sensitivity assessed by the hyperinsulinaemic isoglyceamic glucose clamp was the primary end point of our study, and we did not observe any significant improvement of glycaemic control, as assessed from the fasting glucose and HbA1c concentrations. A modest improvement of insulin sensitivity over a 2-month period may not be reflected by improved fasting glucose levels. Moreover, HbA1c is an integrated measure of glycaemic control over the last 6–8 weeks before measurement, and may not be sensitive enough to detect a recent and gradual change in insulin sensitivity. Oral glucose tolerance test was not included in our study in order to minimize the strain on the recruited nonpaid volunteers. If we had included an oral glucose tolerance test, the patients had to meet at the

hospital on two separate days for each examination and we therefore chose to focus on the ‘gold standard’, the glucose clamp, for examination of insulin sensitivity. The present study confirms the findings and hypothesis in our early studies, which suggested that losartan improves insulin sensitivity,12,24 particularly in patients with severe hypertension24 or insulin resistance.12 In the Insulin Carotids US Scandinavia (ICARUS)-study,25 a Losartan Intervention For Endpoint-Reduction in Hypertension (LIFE) substudy, long-term treatment with losartan for 3 years maintained insulin sensitivity in hypertensive patients with electrocardiographic left ventricular hypertrophy, whereas on treatment with atenolol insulin sensitivity decreased. Atenolol and other bblockers are known to increase the risk of diabetes mellitus.5 However, the 25% relative risk reduction of new-onset diabetes mellitus in the losartan-based regimen in the LIFE study26 may also be explained by preserved insulin sensitivity by losartan. In the present study, treatment with amlodipine plus losartan resulted in an identical blood pressure reduction, but improved insulin sensitivity, compared to a maximal dose of amlodipine. Thus our data suggest that AT-1 receptor blockade with losartan affects glucose metabolism beneficially through mechanisms at the cellular level, beyond what can be expected by the vascular vasodilating effects alone.19

Acknowledgements We thank biochemist Roseli Andreassen for technical assistance and the Department of Clinical Chemistry at Ullevaal University Hospital and Dr Knut Lande for analyzing glucose, HbA1c, lipids and uric acid. We also thank Merck & Co., USA for school grant support. We are indebted to physicians and staff at Bentsebro legesenter, Collosseumklinikken, Ekeberg legesenter, L11 Familiehelsesenter, Manglerud legekontor, Nordseter legekontor, Nordstrand legekontor and Trosterud legekontor in Oslo for their kind help with recruiting patients.

What is known about topic K Different antihypertensive treatment regimens have different effect on glucose homeostasis and development of new-onset diabetes mellitus. K Blockade of the renin–angiotensin system may improve insulin sensitivity, but the mechanism is not clear. What this study adds K There is additive beneficial effects of angiotensin II-receptor blockade with losartan on insulin sensitivity compared with amlodipine alone, despite similar blood pressure reduction. K AT-1 receptor blockade with losartan must have other beneficial effects on the glucose metabolism at the cellular level, beyond what can be expected by the vasodilatation alone.

Journal of Human Hypertension


References 1 Kannel WB, Wilson PW, Zhang TJ. The epidemiology of impaired glucose tolerance and hypertension. Am Heart J 1991; 121(Part 2): 1268–1273. 2 Kearney PM, Whelton M, Reynolds K, Muntner P, Whelton PK, He J. Global burden of hypertension: analysis of worldwide data. Lancet 2005; 365(9455): 217–223. 3 Zimmet P, Alberti KG, Shaw J. Global and societal implications of the diabetes epidemic. Nature 2001; 414(6865): 782–787. 4 Padwal R, Laupacis A. Antihypertensive therapy and incidence of type II diabetes: a systematic review. Diabetes Care 2004; 27(1): 247–255. 5 Gress TW, Nieto FJ, Shahar E, Wofford MR, Brancati FL. Hypertension and antihypertensive therapy as risk factors for type II diabetes mellitus. Atherosclerosis risk in communities study. N Engl J Med 2000; 342(13): 905–912. 6 Aksnes TA, Reims HM, Kjeldsen SE, Mancia G. Antihypertensive treatment and new-onset diabetes mellitus. Curr Hypertens Rep 2005; 7(4): 298–303. 7 Julius S, Kjeldsen SE, Weber M, Brunner HR, Ekman S, Hansson L et al. Outcomes in hypertensive patients at high cardiovascular risk treated with regimens based on valsartan or amlodipine: the VALUE randomised trial. Lancet 2004; 363(9426): 2022–2031. 8 DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 1979; 237(3): E214–E223. 9 Moan A, Nordby G, Os I, Birkeland KI, Kjeldsen SE. Relationship between hemorrheologic factors and insulin sensitivity in healthy young men. Metabolism 1994; 43(4): 423–427. 10 Fossum E, Hoïeggen A, Moan A, Nordby G, Kjeldsen SE. Insulin sensitivity relates to other cardiovascular risk factors in young men: validation of some modifications of the hyperinsulinaemic, isoglycaemic glucose clamp technique. Blood Press Suppl 1997; 2: 113–119. 11 Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28(7): 412–419. 12 Moan A, Hoïeggen A, Seljeflot I, Risanger T, Arnesen H, Kjeldsen SE. The effect of angiotensin II receptor antagonism with losartan on glucose metabolism and insulin sensitivity. J Hypertens 1996; 14(9): 1093–1097. 13 Howard BV, Rodriguez BL, Bennett PH, Harris MI, Hamman R, Kuller LH et al. Prevention conference VI: diabetes and cardiovascular disease: writing group I: epidemiology. Circulation 2002; 105(18): e132–e137.

Journal of Human Hypertension

14 Pacini G, Mari A. Methods for clinical assessment of insulin sensitivity and beta-cell function. Best Pract Res Clin Endocrinol Metab 2003; 17(3): 305–322. 15 Bergman RN, Finegood DT, Ader M. Assessment of insulin sensitivity in vivo. Endocr Rev 1985; 6(1): 45–86. 16 Lender D, rauz-Pacheco C, Breen L, Mora-Mora P, Ramirez LC, Raskin P. A double blind comparison of the effects of amlodipine and enalapril on insulin sensitivity in hypertensive patients. Am J Hypertens 1999; 12(3): 298–303. 17 Opie LH, Schall R. Old antihypertensives and new diabetes. J Hypertens 2004; 22(8): 1453–1458. 18 Schiffrin EL, Park JB, Intengan HD, Touyz RM. Correction of arterial structure and endothelial dysfunction in human essential hypertension by the angiotensin receptor antagonist losartan. Circulation 2000; 101(14): 1653–1659. 19 Julius S, Gudbrandsson T, Jamerson K, Tariq SS, Andersson O. The hemodynamic link between insulin resistance and hypertension. J Hypertens 1991; 9(11): 983–986. 20 Benson SC, Pershadsingh HA, Ho CI, Chittiboyina A, Desai P, Pravenec M et al. Identification of telmisartan as a unique angiotensin II receptor antagonist with selective PPARgamma-modulating activity. Hypertension 2004; 43(5): 993–1002. 21 Schupp M, Janke J, Clasen R, Unger T, Kintscher U. Angiotensin type 1 receptor blockers induce peroxisome proliferator-activated receptor-gamma activity. Circulation 2004; 109(17): 2054–2057. 22 Carlsson PO, Berne C, Jansson L. Angiotensin II and the endocrine pancreas: effects on islet blood flow and insulin secretion in rats. Diabetologia 1998; 41(2): 127–133. 23 Sharma AM, Janke J, Gorzelniak K, Engeli S, Luft FC. Angiotensin blockade prevents type II diabetes by formation of fat cells. Hypertension 2002; 40(5): 609–611. 24 Moan A, Risanger T, Eide I, Kjeldsen SE. The effect of angiotensin II receptor blockade on insulin sensitivity and sympathetic nervous system activity in primary hypertension. Blood Press 1994; 3(3): 185–188. 25 Olsen MH, Fossum E, Hoïeggen A, Wachtell K, Hjerkinn E, Nesbitt SD et al. Long-term treatment with losartan versus atenolol improves insulin sensitivity in hypertension: ICARUS, a LIFE substudy. J Hypertens 2005; 23(4): 891–898. 26 Lindholm LH, Ibsen H, Borch-Johnsen K, Olsen MH, Wachtell K, Dahlo¨f B et al. Risk of new-onset diabetes in the Losartan intervention for endpoint reduction in hypertension study. J Hypertens 2002; 20(9): 1879–1886.