Losartan, the first specific and orally available angiotensin II receptor antagonist, is a potent antihypertensive drug with a low incidence of side effects in humans. However, the effects of losartan on insulin sensitivity and glucose metabolism have not been investigated in detail. Therefore, we carried out a randomized, double-blind study to compare the effects of losartan (50 mg QD) and metoprolol (95 mg QD) on insulin sensitivity, insulin secretion, glucose tolerance, and lipids and lipoproteins in 20 hyperinsulinemic subjects with essential hypertension. The fall in blood pressure was greater with losartan than with metoprolol. Insulin sensitivity evaluated by the euglycemic clamp technique did not change in either group after 12 weeks of treatment. Similarly, glucose oxidation (losartan: 17.0±0.9 versus 16.9±1.0 μmol/kg per minute [before versus after, P=NS]; metoprolol: 17.9±1.3 versus 16.8±1.6 [P=NS]) and nonoxidation (losartan: 22.3±4.0 versus 23.5±3.4 μmol/kg per minute [P=NS]; metoprolol: 23.3±3.2 versus 25.6±4.7 [P=NS]) remained unchanged during the last 30 minutes of the 3-hour euglycemic clamp. Losartan and metoprolol did not have any significant adverse effects on insulin secretion, glucose tolerance, or lipids and lipoproteins. In conclusion, losartan is metabolically neutral, without any significant adverse effect on glucose and lipid metabolism.
Recent studies have indicated that essential hypertension is an insulin-resistant state. One third or even more of all hypertensive individuals have reduced insulin-mediated glucose uptake.1 Although the association between insulin resistance and hypertension may vary among different populations, it is generally believed that insulin resistance has an important role in the pathogenesis of essential hypertension in a majority of populations.2
Several studies have indicated that thiazide diuretics and β-blocking agents worsen insulin sensitivity, whereas α1-receptor antagonists, calcium channel blockers, and angiotensin-converting enzyme (ACE) inhibitors are neutral or even beneficial in this respect.3 However, the drawback of these studies has been that their design has not differentiated between insulin-sensitive and insulin-resistant hypertensive subjects. For example, it is not known whether the results of all previous trials can be generalized to both insulin-resistant and insulin-sensitive hypertensive subjects. It is possible that antihypertensive drugs may be neutral in insulin-resistant subjects but harmful in insulin-sensitive subjects, or vice versa. To overcome this limitation, the study design with new drugs should concentrate either on insulin-sensitive or insulin-resistant hypertensive individuals.
Losartan, the first specific and orally available angiotensin II (Ang II) receptor antagonist, is a potent antihypertensive drug with low toxicity in animal experiments and low frequency of side effects in human studies.4 Furthermore, this drug blocks and reverses the vasoconstrictor effects of Ang II, leading to the improvement of skeletal muscle blood flow.5 By this mechanism, losartan treatment could be associated with the improvement of insulin sensitivity in insulin-resistant hypertensive subjects. Indeed, one previous open study including five subjects with severe hypertension has indicated that losartan could improve insulin sensitivity.6 Therefore, in the current study, we compared the effects of losartan on insulin sensitivity, insulin secretion, glucose tolerance, and lipids and lipoproteins with those of metoprolol in hyperinsulinemic (insulin-resistant) hypertensive subjects.
The double-blind, parallel, randomized study compared the metabolic effects of 50 mg losartan QD and 95 mg metoprolol QD in 20 hyperinsulinemic hypertensive subjects randomized into two treatment groups. Eligible subjects were men or women, aged 30 to 70 years, with normal or impaired glucose tolerance according to the World Health Organization criteria,7 fasting hyperinsulinemia greater than 60 pmol/L, untreated or treated (monotherapy) hypertension for at least 6 months (diastolic pressure >90 mmHg and/or systolic pressure >160 mmHg after at least 4 to 6 weeks of placebo therapy [washout period]), and stable body weight for at least 6 months. Individuals with diabetes, a known contraindication; sensitivity to losartan or metoprolol; serious disease; myocardial infarction within the preceding 12 months; severe infectious disease during the previous 2 months; decompensated cardiac failure; renal insufficiency; history of alcohol or drug abuse; or pregnancy were excluded. The first visit took place 4 to 6 weeks before the active treatment was started. Subjects eligible to participate were given placebo for the next 4 to 6 weeks before randomization. Subjects were randomized to receive either 50 mg losartan QD and placebo matching metoprolol, or 95 mg metoprolol QD and placebo matching losartan. At weeks 2, 4, and 8, subjects visited the outpatient clinic of the Department of Medicine, Kuopio University Hospital, to have their blood pressure and safety parameters measured. At weeks 0 (randomization) and 12 (the end of the study), the following metabolic studies were performed: a 2-hour oral glucose tolerance test (75 g glucose) with blood glucose, plasma insulin, and plasma C-peptide measurements at 0, 30, 60, 90, and 120 minutes followed by a hyperglycemic clamp, and a hyperinsulinemic euglycemic clamp with indirect calorimetry during the following day. Potential adverse reactions were evaluated on each visit.
The protocol was accepted by the Ethics Committee of the University of Kuopio.
Hyperglycemic Clamp (Day 1)
At 120 minutes immediately followed by a 2-hour oral glucose tolerance test, glucose level was acutely increased to 20 mmol/L by a constant glucose infusion (20% glucose) and clamped at 20 mmol/L from 120 to 180 minutes by application of the hyperglycemic clamp technique. Blood glucose was measured every 5 minutes to keep the glucose level at 20 mmol/L. At 160, 170, and 180 minutes, samples were taken for measurement of plasma insulin and plasma C-peptide.
Euglycemic Hyperinsulinemic Clamp (Day 2)
The degree of insulin resistance was evaluated with the euglycemic clamp technique.8 At 8 am after subjects had fasted 12 hours overnight, an intravenous catheter was placed in an antecubital vein for infusion of insulin and 20% glucose. Another cannula for blood sampling was inserted into a wrist vein surrounded by a heated box (70°C). After baseline blood drawing and measurement of gas exchange (see indirect calorimetry below), a priming dose of insulin (Velosulin Human, Nordisk Insulin) was administered during the initial 10 minutes in a logarithmically decreasing manner to acutely raise plasma insulin to the desired level, where it was maintained by a continuous insulin infusion of 480 pmol/m2 per minute (80 mU/m2 per minute). Blood glucose was clamped at 5.0 mmol/L for the next 180 minutes by infusion of 20% glucose at varying rates according to blood glucose measurements performed at 5-minute intervals. The data were calculated for each 20-minute interval.
Indirect Calorimetry (Day 2)
Indirect calorimetry was performed in connection with the euglycemic clamp study with a computerized flow-through canopy–gas analyzer system (Deltatrac, TM Datex) as previously described.9 This device has a precision of 2.5% for oxygen consumption and 1.0% for carbon dioxide production. On the day of the experiment, gas exchange (oxygen consumption and carbon dioxide production) was measured for 30 minutes after subjects had fasted for 12 hours before the clamp and during the last 30 minutes of the euglycemic clamp. The first 10 minutes of each set of data were discarded, and the mean value of the remaining 20 minutes was used in calculations. Protein, glucose, and lipid oxidation rates were calculated according to Ferrannini.10 Protein oxidation rate was calculated on the basis of the urinary nonprotein nitrogen excretion rate by multiplying this value by 6.25. The rate of carbohydrate nonoxidation during the euglycemic clamp was estimated by subtracting the carbohydrate oxidation rate (determined by indirect calorimetry) from the glucose infusion rate (determined by the euglycemic clamp).
Blood glucose in the fasting state and during glucose clamp studies was measured by the glucose oxidase method (Glucose Auto & Stat HGA-1120 analyzer, Daiichi Co). For determination of plasma insulin, blood was collected in tubes containing EDTA. After centrifugation, the plasma was stored at −20°C until the analysis. Plasma insulin concentrations were determined by radioimmunoassay (Phadeseph Insulin RIA 100, Pharmacia Diagnostics AB). Serum lipids and lipoproteins were determined from fresh serum samples drawn after subjects had fasted 12 hours overnight. Lipoprotein fractionation was performed by ultracentrifugation and selective precipitation11 as previously described.12 Cholesterol and triglyceride levels from whole serum and from lipoprotein fractions as well as plasma lactate were assayed by automated enzymatic methods (Boehringer-Mannheim). Apolipoprotein A1 and B levels were determined by a commercial immunoturbidimetric method (Kone Instruments). Serum free fatty acids were determined by an enzymatic method (Wako Chemicals GmbH). Serum potassium was measured by flame photometry. Nonprotein urinary nitrogen was measured by an automated Kjeldahl method.13
All calculations were performed with SPSS/PC+ programs (SPSS Inc). Data are presented as mean±SE. The nonparametric Mann-Whitney test was used for comparison of the two treatment groups and the Wilcoxon test for comparison of changes within each treatment group.
Ten subjects from the losartan group and 9 from the metoprolol group were followed until the study ended. One subject in the metoprolol group dropped out because she moved outside the study area. As shown in Table 1⇓, the subjects were middle-aged, centrally obese, and had moderately elevated blood pressure. No differences in baseline characteristics were found between the groups.
At baseline, blood pressure did not differ significantly between the groups (losartan: 147±5 and 96±2 mmHg, systolic and diastolic, respectively; metoprolol: 153±5 and 101±3). Blood pressure readings were reduced in both groups compared with baseline measurements at 2 weeks (losartan: −9 and −6 mmHg, P<.05; metoprolol: −4 [P=NS] and −8 [P<.05]), 4 weeks (losartan: −11 [P<.01] and −8 mmHg [P<.05]; metoprolol: 0 [P=NS] and −6 [P<.05]), and 8 weeks (losartan: −11 and −6 mmHg [P<.05]; metoprolol: −8 [P=NS] and −9 [P<.05]). After 12 weeks of treatment, systolic and diastolic pressure readings in the losartan group were 134±3 and 90±2 mmHg (P<.05), respectively, and in the metoprolol group were 149±6 (P=NS) and 93±2 mmHg (P<.05).
Glucose Tolerance and Insulin Levels During the Oral Glucose Tolerance Test
Table 2⇓ shows the results of the 2-hour oral glucose tolerance test before and after treatment in each study group. The results demonstrate that treatment with losartan or metoprolol did not significantly change glucose tolerance or insulin levels after an oral glucose load.
Euglycemic Clamp Study
Plasma insulin levels during the last 60 minutes of the euglycemic clamp were similar in both groups before (1062±52 versus 1033±57 pmol/L, losartan versus metoprolol, P=NS) and after (1054±49 versus 1148±63 pmol/L, P=NS) treatment. Blood glucose levels during the euglycemic clamp were similar at baseline (5.0±0.0 versus 5.0±0.0 mmol/L, losartan versus metoprolol, P=NS) and after 12 weeks of treatment (4.9±0.0 versus 5.0±0.0 mmol/L, P=NS). The coefficient of variation of blood glucose levels was less than 4% during the last 2 hours of the clamp in both study groups. During the last 60 minutes of the euglycemic clamp, the rates of whole-body glucose uptake did not change significantly in either group (losartan: 38.5±4.7 versus 39.3±4.3 μmol/kg per minute [before versus after]; metoprolol: 40.7±4.3 versus 42.2±5.6). As shown in Fig 1⇓, glucose oxidation (losartan: 17.0±0.9 versus 16.9±1.0 μmol/kg per minute [before versus after]; metoprolol: 17.9±1.3 versus 16.8±1.6) and nonoxidation (losartan: 22.3±4.0 versus 23.5±3.4 μmol/kg per minute; metoprolol: 23.3±3.2 versus 25.6±4.7) remained unchanged during the last 30 minutes of the euglycemic clamp. Lipid oxidation also remained unchanged with subjects in the fasting state (losartan: 3.4±0.3 versus 3.5±0.3 μmol/kg per minute [before versus after]; metoprolol: 3.5±0.4 versus 3.4±0.3) and during the euglycemic clamp (losartan: 0.15±0.04 versus 0.23±0.05 μmol/kg per minute [before versus after]; metoprolol: 0.09±0.06 versus 0.17±0.09).
Hyperglycemic Clamp Study
Glucose levels during the last 20 minutes of the hyperglycemic clamp did not differ significantly between the groups before (21.4±0.5 versus 21.0±0.3 mmol/L, losartan versus metoprolol) or after (21.5±0.4 versus 20.5±0.5 mmol/L) treatment. As depicted in Fig 2⇓, insulin secretion assessed by insulin and C-peptide responses during maximal glucose stimulation during the hyperglycemic clamp did not change significantly (losartan: C-peptide, 4.34±0.62 versus 4.60±0.69 nmol/L [before versus after] and insulin, 1011±204 versus 1148±294 pmol/L; metoprolol: C-peptide, 4.58±0.50 versus 4.65±0.59 nmol/L and insulin, 1102±193 versus 1200±207 pmol/L).
Lipids, Lipoproteins, and Apolipoproteins
Total cholesterol (losartan: 6.28±0.37 versus 6.29±0.26 mmol/L [before versus after]; metoprolol: 6.68±0.48 versus 6.54±0.58), high-density lipoprotein cholesterol (losartan: 1.20±0.08 versus 1.17±0.10 mmol/L; metoprolol: 1.31±0.09 versus 1.23±0.08), and total triglycerides (losartan: 1.72±0.11 versus 2.24±0.48 mmol/L; metoprolol: 1.85±0.30 versus 1.70±0.26) did not change significantly during treatment in either group. Similarly, cholesterol and triglyceride lipoprotein fractions (low- and high-density lipoproteins and very-low-density lipoproteins) and apolipoproteins A1 and B remained unchanged (data not shown).
Free Fatty Acids and Lactate
Fasting free fatty acids did not change significantly during the study (losartan: 0.60±0.03 versus 0.64±0.09 mmol/L [before versus after]; metoprolol: 0.58±0.08 versus 0.70±0.13). Similarly, fasting lactate levels remained unchanged (losartan: 1.13±0.06 versus 1.09±0.10 mmol/L [before versus after]; metoprolol: 1.08±0.08 versus 0.99±0.08). The suppression of free fatty acids during the euglycemic hyperinsulinemic clamp did not change significantly after the 12 weeks of treatment with losartan (0.036±0.005 versus 0.044±0.011 mmol/L [baseline versus 12 weeks]) or metoprolol (0.043±0.008 versus 0.056±0.014 mmol/L). Similarly, lactate levels during the euglycemic clamp did not change significantly (losartan: 1.06±0.07 versus 1.06±0.07 mmol/L [before versus after]; metoprolol: 1.09±0.07 versus 1.19±0.09).
Albuminuria and Serum Potassium
Urinary albumin excretion remained unchanged during the trial (losartan: 5.6±0.5 versus 6.5±0.8 mg/d [before versus after]; metoprolol: 8.4±1.3 versus 7.8±1.5).
Serum potassium levels decreased similarly during the euglycemic hyperinsulinemic clamp in the losartan-treated group (baseline: 3.8±0.1 versus 3.4±0.1 mmol/L [P<.001]; 12 weeks: 3.9±0.1 versus 3.4±0.1 [P<.001]) and the metoprolol-treated group (baseline: 3.7±0.1 versus 3.4±0.1 mmol/L [P<.001]; 12 weeks: 3.9±0.1 versus 3.4±0.1 [P<.001]).
Our study demonstrates that losartan is metabolically neutral in hyperinsulinemic (insulin-resistant) hypertensive subjects. Losartan did not have any significant effects on insulin sensitivity; insulin secretion; glucose tolerance; insulin response during an oral glucose tolerance test; or lipids, lipoproteins, and apolipoproteins.
The present report is the first randomized, double-blind study investigating the effects of a new class of antihypertensive drugs, Ang II type 1 receptor antagonists, on insulin sensitivity. Losartan inhibits the renin-angiotensin system specifically and selectively without the bradykinin-potentiating effects of ACE inhibitors. To maximize the potential of losartan for improving insulin sensitivity, we included in our study population only hypertensive subjects who were hyperinsulinemic and therefore insulin resistant. A cutoff point of 60 pmol/L was chosen because our previous study demonstrated that subjects having fasting insulin levels exceeding this limit are insulin resistant.14 Baseline measurements of the rates of whole-body glucose uptake demonstrated that this was indeed the case. The mean whole-body glucose uptake was about 30% reduced compared with that in our previous studies in an unselected sample of middle-aged subjects from the same study area in whom insulin sensitivity was evaluated by a euglycemic clamp protocol similar to the one in the present study.15 Nonetheless, losartan failed to result in any significant improvement in insulin sensitivity. Our results agree with the findings reported by Tomiyama et al,16 who treated 4-week-old spontaneously hypertensive rats with enalapril, doxazosin, and losartan for 3 weeks. Losartan decreased mean blood pressure but did not affect the glucose requirement in the euglycemic clamp study. In contrast, enalapril and doxazosin improved insulin sensitivity, indicating that the improvement in insulin sensitivity produced by an ACE inhibitor does not depend on Ang II antagonism.
When measuring insulin sensitivity by the euglycemic clamp technique, we studied our subjects during a high physiological insulin concentration (1000 to 1100 pmol/L), which stimulates glucose uptake to greater than 80% of the maximal response.17 We did not measure hepatic glucose output because during these study conditions, hepatic glucose production is fully suppressed, according to studies by us18 and others.19 Therefore, the rates of whole-body glucose uptake mainly reflect glucose metabolism in insulin-sensitive peripheral tissues, particularly skeletal muscle. We measured the fate of intracellular glucose by indirect calorimetry and observed no difference in glucose oxidation or nonoxidation (glycogen synthesis, lipid synthesis, and anaerobic glycolysis) after treatment with losartan or metoprolol. Similarly, lipid oxidation in the fasting state and during hyperinsulinemia remained unchanged during the study, further indicating that these drugs did not affect glucose metabolism. We measured insulin secretion using the hyperglycemic clamp technique, which reliably measures insulin secretory capacity under maximal glucose stimulation. Neither losartan nor metoprolol had any significant effects on insulin secretion. Losartan did not cause significant changes in lipids, lipoproteins, or the degree of albuminuria. Previous studies have indicated that losartan could even improve lipid profile and reverse proteinuria.20 The neutral effect of losartan on albuminuria in our study was expected, given the fact that our subjects did not have significant albuminuria or proteinuria.
In our study, metoprolol was neutral with respect to glucose metabolism and lipid and lipoprotein levels, in contrast to previous studies that have reported adverse effects by this drug.3 At least two possibilities might explain our results. First, because of time-consuming metabolic studies, we included only 10 subjects in each treatment arm. Therefore, it is possible that the neutral effects of metoprolol could be due to the small sample size and type II error. Second, and more important, the present study design makes it difficult to demonstrate a significant reduction in insulin sensitivity because we studied only insulin-resistant hypertensive subjects. If we had studied a random sample of hypertensive subjects that included both insulin-sensitive and insulin-resistant subjects, our findings on the metabolic effects of metoprolol might have been different.
Losartan has many favorable properties with respect to glucose metabolism. Activation of the sympathetic nervous system, vascular hypertrophy, and vasoconstriction induced by Ang II may lead to insulin resistance. Losartan blocks all of these actions of Ang II. The peripheral vasodilator effect of losartan has been well documented in several species4 although human studies are still scanty. Because skeletal muscle blood flow correlates closely with the rates of whole-blood glucose uptake,21 22 the vasodilator effect of losartan is likely to be important for its effect on glucose metabolism.
Goals of antihypertensive therapy go beyond the control of blood pressure to the prevention of cardiovascular morbidity and mortality. Essential hypertension is often associated with concomitant conditions such as obesity, diabetes, and hyperlipidemia, and insulin resistance is a common denominator in such conditions. An ideal antihypertensive drug does not induce a worsening of these risk factors. In this respect, ACE inhibitors have been shown to be most beneficial. Several studies have indicated that particularly captopril has a potential to improve insulin sensitivity.3 Losartan, a representative of the new class of antihypertensive drugs, Ang II receptor antagonists, may also offer a good alternative in the treatment of hypertensive subjects because it does not have any adverse effects on insulin sensitivity, glucose tolerance, or lipid metabolism.
- Received December 18, 1995.
- Revision received January 16, 1996.
- Accepted April 2, 1996.
Lithell HOL. Effect of antihypertensive drugs on insulin, glucose, and lipid metabolism. Diabetes Care. 1991;14:203-209.
Timmermans PBMWM, Wong PC, Chiu AT, Herblin WF, Benfield P, Carini DJ, Lee RJ, Wexler RR, Saye JAM, Smith RD. Angiotensin II receptors and angiotensin II receptor antagonists. Pharmacol Rev. 1993;45:206-251.
Fouad-Tarazi FM. Hemodynamic effects of inhibitors of the renin-angiotensin system. J Hypertens. 1994;12(suppl 2):S25-S29.
World Health Organization. Diabetes Mellitus. Report of a WHO Study Group. Geneva, Switzerland: World Health Organization; 1985:10-20. Technical Report Series No. 727.
DeFronzo RA, Tobin JE, Andres R. Glucose clamp techniques: a method for quantifying insulin secretion and resistance. Am J Physiol. 1979;237:E214-E223.
Penttilä IM, Voutilainen E, Laitinen P, Juutilainen P. Comparison of different analytical and precipitation methods for the direct estimation of serum high-density lipoprotein cholesterol. Scand J Clin Lab Invest. 1984;41:353-360.
Laakso M, Sarlund H, Mykkänen L. Insulin resistance is associated with lipid and lipoprotein abnormalities in subjects with varying degrees of glucose tolerance. Arteriosclerosis. 1990;10:223-231.
Hawk PB, Oser BL, Summerson WH. Practical Physiological Chemistry. 12th ed. Toronto, Canada: Blakiston; 1947:814-822.
Laakso M. How good a marker is insulin level for insulin resistance? Am J Epidemiol. 1993;137:959-965.
Laakso M, Sarlund H, Salonen R, Suhonen M, Pyörälä K, Salonen JT, Karhapää P. Asymptomatic atherosclerosis and insulin resistance. Arterioscler Thromb. 1991;11:1068-1076.
Tomiyama H, Kushiro T, Abeta H, Ishii T, Takahashi A, Furukawa L, Asagami T, Hino T, Saito F, Otsuka Y, Kurumatani H, Kobayashi F, Kanmatsuse K, Kajiwara N. Kinins contribute to the improvement of insulin sensitivity during treatment with angiotensin converting enzyme inhibitor. Hypertension. 1994;23:450-455.
Prager R, Wallace P, Olefsky JM. In vivo kinetics of insulin action on peripheral glucose disposal and hepatic glucose output in normal and obese subjects. J Clin Invest. 1986;78:472-481.
de Zeeuw D, Gansevoort RT, Dullaart RPF, de Jong PE. Angiotensin II antagonism improves the lipoprotein profile in patients with nephrotic syndrome. J Hypertens. 1995;13(suppl 1):S53-S58.
Laakso M, Edelman SV, Brectel G, Baron AD. Decreased effect of insulin to stimulate skeletal muscle blood flow in obese man: a novel mechanism for insulin resistance. J Clin Invest. 1990;85:1844-1852.
Anderson EA, Hoffman RP, Balon TW, Sinkey CA, Mark LA. Hyperinsulinemia produces both sympathetic neural activation and vasodilation in normal humans. J Clin Invest. 1991;87:2246-2252.