Effect of Beta-Blockers on Insulin Resistance in Patients with Dilated Cardiomyopathy

Circ J 2003; 67: 701–704

Effect of Beta-Blockers on Insulin Resistance in Patients With Dilated Cardiomyopathy

Yuji Hara, MD; Mareomi Hamada, MD; Yuji Shigematsu, MD; Tomoaki Ohtsuka, MD; Akiyoshi Ogimoto, MD; Jitsuo Higaki, MD

The aim of this study was to evaluate the effect ofβ-blockers on insulin resistance in patients with dilated cardiomyopathy (DCM). A secondary aim was to determine the effect of this treatment on plasma concentrations of tumor necrosis factor-α (TNF-α) and to investigate the relationships between this adipocytokine and insulin resistance. Insulin resistance determined using the Homeostatic Model Assessment (HOMA), echocardiographic measurements and analysis of plasma TNF-αconcentration were carried out in 47 patients with DCM without diabetes mellitus before and after 6 months ofβ-blocker therapy. A reduction in left ventricular dimensions and an associated increase in ejection fraction occurred withβ-blocker. The treatment resulted in a signiﬁcant decrease in insulin resistance (HOMA index: Baseline, 2.73±3.36 vs, Month 6, 1.58±1.33, p=0.0347). Beta-blockade was also associated with a decrease in plasma TNF-αconcentration although no signiﬁcant relationship between this change and the improvement in insulin resistance was observed. Beta-blocker therapy in patients with DCM improved not only cardiac function, but also insulin resistance. The mechanism of the change in insulin function remains unclear, but may be related to improvements in left ventricular function or an attenuation of the inhibitory effect of reduction in TNF-αon insulin signaling. (Circ J 2003; 67: 701–704) Key Words: Beta-blockers; Dilated cardiomyopathy; Insulin resistance; Tumor necrosis factor-α

t is well established that treatment with a β-blocker myopathies.18 All patients underwent coronary angiography, results in improvement in cardiac function and prog- and patients with coronary artery disease were excluded. I nosis of patients with chronic heart failure.1–8 The ma- Patients who had clinical or laboratory evidence of diabetes jority of these patients have increased insulin resistance.9–11 mellitus were also excluded. Of the 47 patients, 12 were Patients with essential hypertension are also insulin resis- classified as New York Heart Association (NYHA) func- tant,12,13 but several studies have shown that treatment of tional class grade III, 30 as Class II and 5 as Class I before essential hypertension with an angiotensin-converting the administration ofβ-blocker. Fifteen subjects (mean age enzyme (ACE) inhibitor improves insulin resistance,14–16 in 53±15, range 23–74 years) who had no evidence of organic contrast toβ-blockade that appears to have adverse effect cardiac disease or diabetes mellitus were retrospectively on insulin function in those patients.14,17 There is evidence selected as a control group for the measurement of insulin that β-blockers have beneficial hemodynamic effects in resistance. There were no significant differences in age, patients with dilated cardiomyopathy (DCM), but we are gender or body mass index between the DCM group and the unaware of any other study that has investigated the effects control group (Table1). The study protocol was approved of these agents on insulin resistance. We conducted a pro- by the Human Investigations Committee of the institution spective 6-month study of the effects of the β-blockers, and all patients gave informed consent before participating metoprolol, bevantolol and carvedilol, on insulin resistance in the study. in patients with DCM. Study Protocol Following admission to hospital, all patients with DCM Methods had baseline biochemical and echocardiographic assess- Subjects We studied 47 patients with DCM (mean age 55±13, Table 1 Characteristics of the Control Subjects and the Patients range 25–78 years) who had been treated with digitalis, With Dilated Cardiomyopathy (DCM) diuretics and an ACE inhibitor for at least 3 months. All Controls DCM patients were diagnosed as having DCM according to the p value criteria of the World Health Organization/International (n=15) (n=47) Society and Federation of Cardiology definition of cardio- Age (years) 53±15 55±13 0.5877 M/F 10/5 39/8 0.1766 (Received March 10, 2003; revised manuscript received May 26, Weight (kg) 62.5±12.6 58.6±9.0 0.2026 2003; accepted June 4, 2003) Height (cm) 164±11 161±7 0.3590 The Second Department of Internal Medicine, Ehime University Body mass index (kg/m2)23.2±3.1 22.6±3.3 0.5168 School of Medicine, Ehime, Japan FBS (mmol/L) 4.86±0.51 5.08±0.47 0.2183 Mailing address: Yuji Hara, MD, The Second Department of Internal Insulin concentration (mU/L) 4.2±1.0 12.2±15.4 0.0787 Medicine, Ehime University School of Medicine, Shigenobu-cho, HOMA index 0.91±0.24 2.73±3.36 0.0347 Onsen-gun, Ehime 791-0295, Japan. E-mail: [email protected]. ac. jp FBS, fasting blood glucose; HOMA, Homeostatic Model Assessment.

Circulation Journal Vol.67, August 2003 702 HARA Y et al.

Table 2 Changes in Hemodynamic, Echocardiographic and Plasma TNF-α Concentration Data Before and After Treatment With a Beta- Blocker in Patients With Dilated Cardiomyopathy

At baseline After treatment p value NYHA functional class 2.15±0.60 1.57±0.50 <0.0001 HR (beats/min) 76±16 65±8 <0.0001 SBP (mmHg) 114±12 116±16 0.4942 LVDd (mm) 63.6±6.7 59.6±6.9 <0.0001 LVDs (mm) 52.6±8.5 46.5±8.8 <0.0001 EF (%) 35.2±12.0 43.8±12.3 <0.0001 TNF-α (pg/ml) 33.9±32.4 15.1±19.1 <0.0001 NYHA, New York Heart Association; HR, heart rate; SBP, systolic blood pressure; LVDd, left ventricular dimension at end-diastole, LVDs, left ventricular dimension at end-systole; EF, ejection fraction; TNF, tumor necrosis factor. Fig1. Insulin resistance (HOMA index) before and after treatment with aβ-blocker in patients with DCM. Statistical Analysis Data were analyzed using StatView (Abacus Concepts, ments carried out. Treatment with a β-blocker was then Inc, Berkeley, CA, USA, 1994) and expressed as mean± begun for a period of 6 months, each under the same stan- SD. The level of significance for comparative analyses was dard therapy, with 14 patients receiving metoprolol, 13 re- set at p<0.05. Baseline characteristics, except for insulin ceiving bevantolol and 20 receiving carvedilol. The initial concentration and HOMA index, were analyzed with an dosage of metoprolol and bevantolol was 2.5 or 5mg/day unpaired t test or chi-square test for non-parametrically dis- and the initial dosage of carvedilol was 1.25 or 2.5mg/day. tributed values. The difference in insulin concentration and The dose was increased every 5–7 days until a maintenance the HOMA index between DCM patients and controls was dose was achieved based on the prior dose being clinically compared using the Mann-Whitney U rank-sum test for tolerated, heart rate at rest remaining between 50 and unpaired data. Changes in the hemodynamic or echocardio- 70 beats/min and systolic blood pressure more than graphic variables after β-blocker therapy were analyzed 90mmHg. Mean maintenance doses were 36±10mg (range using Student’s paired t test, and changes in NYHA func- 20–60mg) for metoprolol, 53±32mg (range 20–150mg) for tional class, HOMA index and TNF-αwere determined by bevantolol and 16±6mg (range 10–30mg) for carvedilol. the Wilcoxon signed rank test. The relationship between HOMA index and hemodynamic or echocardiographic Echocardiographic Study parameters beforeβ-blocker treatment was assessed by lin- M-mode and 2-dimensional echocardiographic studies ear regression. The relationship between changes in insulin were performed with an Aloka SSD 9000 or SSD 5500 resistance and changes in echocardiography parameters or imaging system (Tokyo, Japan) with 2.5- or 3.5-MHz plasma TNF-αconcentration was assessed by linear regres- transducers. The M-mode echocardiogram was recorded on sion. All calculations were performed on a personal compu- a strip-chart recorder at a paper speed of 50 or 100mm/s. ter with the statistical package StatView (Abacus Concepts, The following conventional variables were obtained from Inc). A value of p<0.05 was considered significant. the M-mode measurements according to the criteria of the American Society of Echocardiography:19 left ventricular dimension at end-diastole (LVDd) and end-systole (LVDs), Results left ventricular volume calculated using Teichholtz’s for- Insulin Resistance and Plasma TNF-αConcentration mula, and ejection fraction (EF). Arterial blood pressure Insulin resistance at baseline was significantly greater in was determined in duplicate on the day of the echocardio- patients with DCM than in the control subjects (Table1). graphic studies using a sphygmomanometer. The effect of β-blocker treatment on this variable in the DCM group is shown in Fig1; there was a significant im- Measurement of Insulin Resistance provement in insulin resistance (mean HOMA index 2.73± A blood sample was withdrawn from the antecubital 3.36 vs 1.58±1.33, p=0.04). The plasma TNF-α concen- vein on the morning of the investigations after the patient tration was also significantly decreased after 6 months of had fasted for 12h. Fasting blood glucose (FBS) and in- treatment with theβ-blockers (Table2). sulin concentrations were measured and used to determine the Homeostatic Model Assessment (HOMA) index of Echocardiography insulin resistance, calculated using the equation; HOMA= Table2 summarizes the changes in echocardiographic FBS×insulin/22.5.20 parameters, heart rate and systolic blood pressure from baseline to afterβ-blockade. Although systolic blood pres- Measurement of TNF-α(TNF-α) sure remained unchanged, there was a significant decrease Blood samples for TNF-α analysis were collected into in heart rate. As anticipated, both LVDd and LVDs chilled tubes, immediately centrifuged at 4°C, stored at decreased while EF increased. –80°C and analyzed using an enzyme-linked immuno- sorbent assay (Immunotech Co, Marseille, France). The Relationship Between Insulin Resistance and Hemodynamic, average inter- and intra-assay coefficients of variation were Echocardiographic Parameters or Plasma TNF-αConcen- <10% for all the assays carried out. tration There was no the relationship between HOMA index and

Circulation Journal Vol.67, August 2003 Effect of Beta-Blockers on Insulin Resistance in DCM 703 hemodynamic or echocardiographic parameters before β- Study Limitations blocker treatment. There was also no relationship between The study protocol was not randomized and the research- the improvement in insulin resistance and changes in the ers were not blinded to treatment. However, measurements echocardiographic parameters or plasma TNF-αconcentra- of cardiac function, insulin resistance and TNF-α were tion. performed in a blinded manner with regard to treatment groups and time. Another limitation is the small number of study patients. We are aware that a larger number of sub- Discussion jects would have improved the reliability of our results. The present study confirms that DCM is associated with insulin resistance and shows that treatment with aβ-blocker attenuated this change in insulin function. Conclusion However, the mechanism of this apparent improvement Beta-blocker treatment attenuates insulin resistance in in insulin sensitivity remains unclear. In patients with hy- patients with DCM and we would suggest thatβ-blockers pertension,β-blockers reportedly induce insulin resistance, may be a beneficial adjunct to the treatment regimen of pa- possibly by decreasing the metabolic clearance rate of tients with DCM, especially those individuals with insulin insulin that in turn results in higher insulin concentrations, resistance. impaired glucose disposal and hyperglycemia.14,17 It is pos- sible that the decrease in cardiac output during treatment References with a β-blocker may result in decreased blood flow in 1. Waagstein F, Bristow MR, Swedberg K, Camerini F, Fowler MB, skeletal muscles, thereby reducing the availability of Silver MA, et al. Beneficial effects of metoprolol in idiopathic 17 glucose to the prime target tissue for glucose disposal. dilated cardiomyopathy. Lancet 1993; 342: 1441–1446. However, Malminiemi et al reported that celiprolol, a vaso- 2. Krum H, Sackner-Bernstein JD, Goldsmith RL, Kukin ML, Schwartz dilating β1-adrenoreceptor antagonist, improved insulin B, Penn J, et al. Double-blind, placebo-controlled study of the long- sensitivity in dyslipidemic hypertensive patients; celiprolol term efficacy of carvedilol in patients with severe chronic heart failure. Circulation 1995; 92: 1499–1506. hadβ2-agonist properties, which can cause vasodilation in 3. Packer M, Bristow MR, Cohn JN, Colucci WS, Fowler MB, Gilbert muscle tissue.21 Refsgaard et al recently reported that car- EM, et al. The effect of carvedilol on morbidity and mortality in pa- vedilol is neutral with regard to its influence on insulin tients with chronic heart failure. N Engl J Med 1996; 334: 1349– sensitivity in patients with congestive heart failure.22 Pietila 1355. 4. CIBIS II Investigators and Committees. The cardiac insufficiency et al reported that insulin sensitivity was favorably affected bisoprolol study II (CIBIS-II): A randomised trial. Lancet 1999; 353: by exercise training in chronic heart failure, and that 9–13. celiprolol augmented this effect.23 The findings of the pres- 5. Packer M. Current role of beta-adrenergic blockers in the manage- ent study in patients with DCM indicate that the effects of ment of chronic heart failure. Am J Med 2001; 110(Suppl 7A): 81S– 94S. improved cardiac function associated with β-blockade, 6. Hara Y, Hamada M, Shigematsu Y, Suzuki M, Kodama K, Kuwa- such as enhanced exercise capacity and increase peripheral hara T, et al. Effect of beta-blocker on left ventricular function and blood flow, may indirectly lead to an increase in insulin natriuretic peptides in patients with chronic heart failure treated with sensitivity. Swan et al reached a similar conclusion in a angiotensin-converting enzyme inhibitor. Jpn Circ J 2000; 64: 365– 369. study of patients with heart failure in which it was shown 7. Hirooka K, Yasumura Y, Ishida Y, Hanatani A, Nakatani S, Koma- that insulin resistance was a consequence of reduced peak mura K, et al. Comparative left ventricular functional and neurohor- oxygen consumption rather than a result of abnormal ven- moral effects of chronic treatment with carvedilol versus metoprolol tricular function or changes in norepinephrine activity.10 in patients with dilated cardiomyopathy. Jpn Circ J 2001; 65: 931– However, in the present study we were unable to demon- 936. 8. Hamada M, Hara Y, Ohtsuka T, Suzuki J, Saeki H, Ogimoto A, et al. strate an association between changes in left ventricular Effect of addition of aβ-blocker on left ventricular remodeling and function and improved insulin resistance and therefore prognosis in patients with dilated cardiomyopathy treated with further investigations are warranted in order to better angiotensin-converting enzyme inhibitor. J Cardiovasc Pharmacol understand the etiological role of hemodynamic changes on 2003; 41(Suppl 1): S5–S10. 9. Swan JW, Walton C, Godsland IF, Clark AL, Coats AJ, Oliver MF. insulin activity. Insulin resistance in chronic heart failure. Eur Heart J 1994; 15: Our study results also raise the possibility that reductions 1528–1532. in plasma TNF-αconcentration may be a factor contributing 10. Swan JW, Anker SD, Walton C, Godsland IF, Clark AL, Leyva F, et to the increase in insulin sensitivity induced byβ-blockers al. Insulin resistance in chronic heart failure: Relation to severity and etiology of heart failure. J Am Coll Cardiol 1997; 30: 527–532. in DCM. It is well established that TNF-αin combination 11. Coats AJS, Anker S. Insulin resistance in chronic heart failure. J with norepinephrine has an adverse effect on insulin signal- Cardiovasc Pharmacol 2000; 35(Suppl 4): S9–S14. ing by inhibiting phosphorylation of the insulin recep- 12. Ferrannini E, Buzzigoli G, Bonadonna R, Giorico MA, Oleggini M, tor.24–26 The finding of the present study, and that of an Graziadei L, et al. Insulin resistance in essential hypertension. N earlier investigation,27 that -blocker treatment markedly Engl J Med 1987; 317: 350–357. β 13. Reaven GM. Banting lecture 1988: Role of insulin resistance in reduced the TNF-αconcentration raises the possibility that human disease. Diabetes 1988; 37: 1595–1607. attenuation of the inhibitory effect of this adipocytokine 14. Lind L, Pollare T, Berne C, Lithell H. Long-term metabolic effects of followingβ-blockade results in improved insulin sensitivi- antihypertensive drugs. Am Heart J 1994; 128: 1177–1183. ty. However, as with the hemodynamic results we were 15. Pollare T, Lithell H, Berne C. A comparison of the effects of hydro- chlorothiazide and captopril on glucose and lipid metabolism in unable to demonstrate a significant relationship between patients with hypertension. N Engl J Med 1989; 321: 868–873. changes in cytokine concentration and insulin resistance 16. Kinoshita M, Nakaya Y, Harada N, Takahashi A, Nomura M, Bando and therefore more comprehensive and larger studies are S. Combination therapy of exercise and angiotensin-converting required in order to validate the role of TNF-α in these enzyme inhibitor markedly improves insulin sensitivities in hypertensive patients with insulin resistance. Circ J 2002; 66: 655–658. beneficial changes associated withβ-blockers. 17. Pollare T, Lithell H, Selinus I, Berne C. Sensitivity to insulin during treatment with atenolol and metoprolol: A randomised, double blind study of effects on carbohydrate and lipoprotein metabolism in

Circulation Journal Vol.67, August 2003 704 HARA Y et al.

hypertensive patients. BMJ 1989; 298: 1152–1157. et al. Celiprolol augments the effect of physical exercise on insulin 18. Report of the 1995 World Health Organization/International Society sensitivity and serum lipid levels in chronic heart failure. Eur J Heart and Federation of Cardiology task force on the definition and classifi- Fail 2000; 2: 81–90. cation of cardiomyopathies. Circulation 1996; 93: 841–842. 24. Hotamisligil GS, Arner P, Caro JF, Atkinson RL, Spiegelman BM. 19. Sahn DJ, DeMaria A, Kisslo J, Weyman A. The Committee on M- Increased adipose tissue expression of tumor necrotic factor-alpha in mode Standardization of The American Society of Echocardiogra- human obesity and insulin resistance. J Clin Invest 1995; 95: 2409– phy. Recommendations regarding quantitation in M-mode echocar- 2415. diography: Results of a survey of echocardiographic measurements. 25. Lang CH, Dobrescu C, Bagby GJ. Tumor necrotic factor impairs Circulation 1978; 58: 1072–1083. insulin action on peripheral glucose disposal and hepatic glucose 20. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, output. Endocrinology 1992; 130: 43–52. Turner RC. Homeostasis model assessment: Insulin resistance and 26. Paolisso G, Tagliamonte MR, Rizzo MR, Gambardella A, Gualdiero beta-cell function from fasting plasma glucose and insulin concentra- P, Lama D, et al. Prognostic importance of insulin-mediated glucose tions in man. Diabetologia 1985; 28: 412–419. uptake in aged patients with congestive heart failure secondary to 21. Malminiemi K, Lahtela JT, Huupponen R. Effects of celiprolol on mitral and/or aortic valve disease. Am J Cardiol 1999; 83: 1338– insulin sensitivity and glucose tolerance in dyslipidemic hyperten- 1344. sion. Int J Clin Pharmacol Ther 1995; 33: 156–163. 27. Ohtsuka T, Hamada M, Hiasa G, Sasaki O, Suzuki M, Hara Y, et al. 22. Refsgaard J, Thomsen C, Andreasen F, Gotzsche O. Carvedilol does Effect of beta-blockers on circulating levels of inflammatory and not alter the insulin sensitivity in patients with congestive heart fail- anti-inflammatory cytokines in patients with dilated cardiomyopathy. ure. Eur J Heart Fail 2002; 4: 445–453. J Am Coll Cardiol 2001; 37: 412–417. 23. Pietila M, Malminiemi K, Huupponen R, Rouru J, Pulkki K, Pere E,

Circulation Journal Vol.67, August 2003