Effects of ranolazine on exercise tolerance and HbA 1c in patients with chronic angina and diabetes

Transcription

1 European Heart Journal (2006) 27, doi: /eurheartj/ehi495 Clinical research Effects of ranolazine on exercise tolerance and HbA 1c in patients with chronic angina and diabetes Adam D. Timmis 1 *, Bernard R. Chaitman 2, and Michael Crager 3 1 Department of Cardiology, London Chest Hospital, Bonner Road, London E2 9JX, UK; 2 St Louis University Health Services Center, St Louis, MO, USA; and 3 CV Therapeutics Inc., Palo Alto, CA, USA Received 31 January 2005; revised 19 July 2005; accepted 19 August 2005; online publish-ahead-of-print 21 September 2005 See page 5 for the editorial comment on this article (doi: /eurheartj/ehi624) KEYWORDS ; Diabetes; CARISA; Chronic stable angina; HbA 1c Aims The anti-anginal efficacy and safety of ranolazine in diabetic and non-diabetic patients included in the Combination Assessment of In Stable Angina (CARISA) trial (JAMA 2004;291:309) were studied. Glycaemic control was also assessed in CARISA and its long-term open-label extension study. Methods and results Patients with chronic angina enrolled in CARISA (189 with diabetes, 634 without diabetes) on background atenolol, diltiazem, or amlodipine therapy were randomized to placebo, ranolazine 750 or 1000 mg twice daily for 12 weeks, during which exercise tolerance, angina frequency, nitroglycerin usage, glucose, HbA 1c, and lipids were measured. Patients completing the randomized study could enroll in an ongoing open-label extension study and were evaluated every 3 months. produced similar improvements in exercise parameters, nitroglycerin use, and angina frequency in diabetic and non-diabetic patients. Adverse events were similar between groups. Fasting glucose and lipids remained unaltered in diabetic patients after 12 weeks of therapy. In a post hoc analysis, ranolazine 750 and 1000 mg reduced HbA 1c vs. placebo by % (P ¼ 0.008) and % (P ¼ ), respectively; the HbA 1c levels appeared to remain unchanged over time during longterm therapy. Conclusion Anti-anginal efficacy and safety of ranolazine for angina were similar between diabetic and non-diabetic patients. significantly improved glycaemic control in diabetic patients. Introduction Type 2 diabetes and the metabolic syndrome have reached epidemic levels in the United States 1 and Europe, and are expected to result in an increased prevalence of cardiovascular disease. 2 Diabetes is a significant predictor of cardiovascular mortality 3 and is a common co-morbidity in angina patients. It has a reported prevalence of 7.9% in North America and 7.8% in Europe. 4 The Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) of the National Cholesterol Education Program classifies diabetes as a coronary heart disease risk equivalent. 5 Cardiovascular death and myocardial infarction rates are significantly increased in diabetic patients. Treatment options are often challenging because of the metabolic abnormalities that affect endothelial function and frequently co-morbid conditions, such as hypertension, dyslipidaemia, autonomic dysfunction, and extracardiac atherosclerosis. 3,6 is a novel haemodynamically neutral antianginal and anti-ischaemic compound under development. Unlike existing anti-anginal agents, ranolazine studies have shown that its anti-anginal effect is achieved with little effect on blood pressure or heart rate *Corresponding author. Tel: þ ; fax: þ address: timmis@1ch.demon.co.uk increases exercise capacity and reduces angina attacks in patients with chronic stable angina. 7,8 was well tolerated with little or no effect on mean blood pressure or heart rate and induced an average increase in the QTc interval of,10 ms at 1000 mg twice daily in humans. 7,8 We previously reported that in the placebo-controlled, multicentre, randomized Combination Assessment of in Stable Angina (CARISA) trial, ranolazine was effective in increasing treadmill exercise time, time to angina onset, and time to onset of ischaemic ST-segment depression in patients with chronic angina who had demand-induced myocardial ischaemia at low-level exercise workloads. The purpose of this analysis of the combination therapy trial was to evaluate the anti-anginal efficacy and adverse event profile of ranolazine in diabetic vs. nondiabetic patients with chronic angina. In addition, the effects of ranolazine on HbA 1c levels, serum glucose, and serum lipid levels during the randomized phase of the CARISA trial and HbA 1c values during its long-term open-label extension study were examined in diabetic patients to determine if ranolazine had specific effects on patients with diabetes. Methods Study design The details of the CARISA trial have been previously described. 7 Briefly, in the CARISA trial, 823 patients with chronic angina on & The European Society of Cardiology All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 in diabetic patients 43 background once daily atenolol (50 mg), diltiazem (180 mg), or amlodipine (5 mg) were randomized to 12 weeks of therapy in a three-arm parallel study to placebo, ranolazine 750 mg twice daily or ranolazine 1000 mg twice daily. As previously described, patients were randomized in 1:1:1 ratio to the three treatment groups in blocks of six, stratifying by background therapy. At the screening and qualifying visits and subsequent visits at Weeks 2, 6, and 12, blood pressure, heart rate, and electrocardiogram (ECG) and exercise tolerance test (ETT) trough measurements, adverse events, and concomitant medication usage were collected as well as angina and nitroglycerin-use diary review. ECG and ETT peak measurements were evaluated at the screening and qualifying visits and Weeks 2 and 12. Physical exam, body weight and inclusion/exclusion criteria assessments, and laboratory tests (including HbA 1c in diabetic patients), were conducted at the screening/qualifying visit and Week 12 (trial endpoint). Upon successful completion of CARISA, patients who met entry criteria were eligible to participate in an open-label extension trial, which is ongoing. The extension trial analyses presented here were based on data availability through August 2003 and represent the results from 2 years of an anticipated 6-year study. Study population This analysis compared the effects of ranolazine in diabetic and nondiabetic patients (taking insulin or not taking insulin) enrolled in CARISA. To be eligible for CARISA, all patients were required to have chronic angina.3 months and documented coronary artery disease. Patients with a resting ECG abnormality that precluded analysis of the exercise ECG were excluded, as were patients who required digoxin therapy, were in New York Heart Association (NYHA) class III or IV failure, or had an acute coronary syndrome or revascularization procedure within 2 months before study entry. Diabetes was diagnosed on the basis of clinical history with no requirement for ongoing treatment with hypoglycaemic agents. Insulin-dependent diabetics were designated as such for analysis if they were receiving insulin treatment at baseline. As previously described, 7 the qualifying criterion for randomization was based on exercise duration with two modified Bruce ETTs, 1 week apart, such that the presence of exercise-induced angina was achieved. Ischaemic ST-segment depression 1mm was validated at St Louis University Core ECG Laboratory. Efficacy assessments The primary efficacy parameter in CARISA was the exercise duration at 12 h post-dose, which is the approximate time of trough ranolazine plasma concentrations. Exercise testing was also performed at the peak time of plasma concentrations, 4 h post-dose. The ETT time required to reach the severity of symptoms that would typically have caused the patient to stop exercise during the course of his or her usual daily activities (and no further) constituted the exercise duration endpoint of the study. Secondary efficacy variables included time to onset of angina, time to 1 mm ST-segment depression, and patient-reported frequency of angina attacks and nitroglycerin consumption. HbA 1c was assessed post hoc as a measure of efficacy in patients with diabetes. Safety analyses Adverse events were recorded and investigators assessed the drug relationship of each event. Laboratory measures The protocol specified that laboratory values were to be collected 12 h (+0.5 h) after patients most recent dose of study medication and following a h fasting period from the previous evening. Key laboratory tests included ECGs, haematology, coagulation, serum chemistry (HbA 1c in diabetic patients only; HbA 1c was not collected in patients who were not diabetic at baseline), urinalysis, and lipid panel. HbA 1c was analysed by Covance Central Laboratory using Bio-Rad Variant analyser that utilizes ion-exchange high performance liquid chromatography. In this analysis, the safety of ranolazine in diabetic vs. non-diabetic patients was compared through evaluation of adverse events, vital signs, and laboratory data. Statistical considerations Within CARISA, the effects of ranolazine on exercise time variables in diabetic vs. non-diabetic patients were assessed in an analysis of covariance (ANCOVA) model with effects for treatment, baseline exercise time, background therapy (atenolol, diltiazem, or amlodipine), centre (pooled by geographic region), diabetes status, and the interaction of treatment with diabetes status, using the last observation carried forward after randomization. The primary analysis data set included all patients who received study drug and had at least one evaluable ETT after randomization, assigning patients to treatment groups as randomized with intention to treat. Angina frequency and nitroglycerin consumption were normalized by time of study participation and expressed as a weekly rate. The weekly rates were ranked and analysed using the ANCOVA model described earlier. For the diabetic patients in CARISA, change from baseline in HbA 1c at 12 weeks of treatment or early termination was analysed post hoc using an ANCOVA model with effects for treatment, baseline laboratory value, pooled site, and background therapy. Treatment effects in diabetic patients who were taking insulin and those who were not were estimated by adding insulin status and the interaction of treatment with insulin status to this ANCOVA model. The proportion of patients with HbA 1c,7% at 12 weeks of treatment or early termination was analysed using a logistic regression model with the same effects as the main ANCOVA model. Early termination was defined as failure of the patient to complete 12 weeks of randomized therapy during the double-blind study. Combining data from the diabetic patients in CARISA and the longterm extension trial who received ranolazine, changes from baseline HbA 1c were summarized as a function of time since start of ranolazine. Patient data for the long-term study were included if patients did not have more than a 7-day interruption in dosing between CARISA and its long-term extension study. For this analysis, the time scale was broken into 100-day intervals. The average value of HbA 1c in each of these intervals was estimated with a repeated measures generalized estimating equations (GEE) model 11 using the normal distribution and with effects for baseline value and time, insulin use, and the interaction of time with insulin use. Assuming that the duration of each patient s study participation is not correlated with HbA 1c, this method provides an unbiased estimate of the changes in HbA 1c over time, appropriately adjusting for missing data. Diagnostic tests using GEE models showed no evidence for correlation between the duration of follow-up and the observed values of HbA 1c (P. 0.53). A sensitivity analysis comparing the results in all patients with those patients who had the longest follow-up was also used to validate the repeated measures analysis. All hypothesis tests were two-sided and conducted at the 0.05 level of significance. The analyses reported are exploratory and as such there was no adjustment for multiple inferences. Results Patients Of the patients enrolled in CARISA, 189 (23%) had a history of diabetes. Baseline characteristics of the diabetic and nondiabetic patients in the CARISA trial are illustrated in Table 1, including the antidiabetic medications taken by the diabetic patients.

3 44 A. Timmis et al. Table 1 Baseline demographics of diabetic vs. non-diabetic patients Characteristic Diabetic Non-diabetic Patients, n (%) 189 (23) 634 (77) Age, years (mean + SE) Men, n (%) 136 (72) 501 (79) Prior MI, n (%) 113 (60) 361 (57) Prior CABG, n (%) 43 (23) 101 (16) Prior PCI, n (%) 51 (27) 101 (16) History of hypertension, n (%) 140 (74) 387 (61) CHF (NYHA class I or II), n (%) 55 (29) 190 (30) Diabetic patients receiving concomitant medications a Medication drug class n (%) (n ¼ 57) Ran 750 mg n (%) (n ¼ 68) Ran 1000 mg n (%) (n ¼ 64) Diabetics receiving any 39 (68.4) 42 (61.8) 46 (71.9) antidiabetic medication a-glucoside inhibitors 2 (3.5) 1 (1.5) 1 (1.6) Secretegogues 19 (33.3) 32 (47.1) 31 (48.4) Thiazolidinediones 1 (1.8) 3 (4.4) 1 (1.6) Biguanides 16 (28.1) 17 (25.0) 19 (29.7) Insulin 11 (19.3) 13 (19.1) 11 (17.2) Calcium dobesilate 0 1 (2.3) 0 CABG, coronary artery bypass graft; CHF, congestive heart failure; MI, myocardial infarction; PCI, percutaneous coronary intervention; SE, standard error; Ran, ranolazine. a Patients could be receiving more than one medication. Efficacy As previously reported in the CARISA study, 7 ranolazine significantly improved exercise duration in the primary analysis of the overall population. Figure 1 shows the effects of ranolazine on exercise duration (primary efficacy parameter), time to onset of angina, and time to 1 mm ST-segment depression at trough plasma concentrations in diabetic and non-diabetic patients in this study. The treatment effects appear to be similar in the two subgroups. There was no significant treatment by subgroup interaction for exercise duration (P ¼ 0.89), time to onset of angina (P ¼ 0.54), or time to 1 mm ST-segment depression (P ¼ 0.44), and therefore no evidence that the treatment effect differed between diabetic and non-diabetic patients. At the time of peak ranolazine plasma concentrations, there was also no evidence that the treatment effect differed between diabetic and non-diabetic patients. At peak (data not shown), there was no significant treatment by subgroup interaction for exercise duration (P ¼ 0.58), time to onset of angina (P ¼ 0.26), or time to 1 mm ST-segment depression (P ¼ 0.088). Over the 12 weeks of the CARISA trial, ranolazine significantly reduced the number of normalized weekly angina episodes and nitroglycerin usage in the overall patient population. 7 The effects of ranolazine in diabetic patients and in non-diabetic patients appear to be similar, as shown in Figure 2. Statistical tests for interaction between diabetes status and treatment effect showed no evidence that the effect of ranolazine differed between diabetic and non-diabetic patients in either the number of weekly angina episodes (P ¼ 0.81) or nitroglycerin usage (P ¼ 0.063). It appears that the effects of ranolazine in diabetic patients are comparable to those in non-diabetic patients. Figure 1 Change from baseline in CARISA in (A ) exercise duration, (B ) time to onset of angina, and (C ) time to 1 mm ST-segment depression at trough.

4 in diabetic patients 45 Figure 2 Normalized (A ) weekly angina episodes and (B ) nitroglycerin use in diabetic vs. non-diabetic patients. Effect of ranolazine on HbA 1c in diabetic patients HbA 1c values were obtained from 160/189 (85%) of diabetic patients at baseline and 140/189 (74%) of diabetic patients at 12 weeks. The analysis of HbA 1c changes from baseline included the 131 (69%) patients who had both baseline and on-treatment HbA 1c values. The analysis adjusted the estimated end-of-treatment and change-from-baseline mean values for baseline differences among the treatment groups. As shown in Table 2, ranolazine 750 and 1000 mg reduced HbA 1c compared with placebo by % (P ¼ 0.008) and % (P ¼ ), respectively. When the diabetic patients in CARISA were stratified by insulin treatment, the reduction in HbA 1c compared with placebo in those receiving insulin was greater, as shown in Table 3. Recent diabetes treatment guidelines emphasize achieving a HbA 1c of,7% as the target for the treatment of diabetes. 12 Table 4 shows the percentage of diabetic patients in CARISA with an HbA 1c of,7% at baseline and at the end of treatment. Using a logistic regression analysis that adjusted for baseline differences, the percentage of diabetic patients with HbA 1c,7% was significantly higher in the ranolazine 1000 mg group (P ¼ 0.004). This change in HbA 1c was not accompanied by an increase in weight Table 2 Effect of ranolazine on HbA 1c (%) in diabetic patients (least-squares mean + SEM) (n ¼ 37) 750 mg BID 1000 mg BID Baseline a Week 12 or early termination b Change from baseline b Difference vs. placebo (P ¼ 0.008) (P ¼ ) a Least-squares mean estimates from analysis of variance with effects for treatment, pooled centre, and background therapy. b Least-squares mean estimates from analysis of covariance with effects for treatment, pooled centre, background therapy, and baseline HbA 1c. Table 3 Effect of ranolazine on HbA 1c (%) in insulin-treated vs. non-insulin-treated diabetic patients (least-squares mean + SE) receiving insulin 750 mg BID receiving insulin 1000 mg BID receiving insulin Yes (n ¼ 10) No (n ¼ 27) Yes (n ¼ 13) No (n ¼ 34) Yes (n ¼ 8) No (n ¼ 39) Baseline a Week 12 or time of early termination b Change from baseline b Difference vs. placebo (P ¼ 0.016) (P ¼ 0.087) (P ¼ 0.008) (P ¼ 0.007) a Least-squares mean estimates an analysis of variance model with effects for treatment, pooled centre, background therapy, insulin treatment, and the interaction between randomized treatment and insulin treatment. b Least-squares mean estimates an analysis of covariance model with effects for treatment, pooled centre, background therapy, insulin treatment, the interaction between randomized treatment and insulin treatment, and baseline HbA 1c.

5 46 A. Timmis et al. Table 4 Patients with HbA 1c, 7% (n ¼ 37) 750 mg BID 1000 mg BID Baseline, n (%) 14 (37.8) 17 (36.2) 12 (25.5) Week 12 or time of early termination, n (%) a 16 (43.2) 24 (51.1) (P ¼ 0.30 vs. placebo) 26 (55.3) (P ¼ vs. placebo) a Logistic regression with effects for treatment, pooled centre, background therapy, and baseline HbA 1c. Figure 3 Change from baseline in HbA 1c levels over time in diabetic patients treated with ranolazine in the double-blind and open-label phase. Data include all patients treated with ranolazine. Table 5 Incidence of adverse events occurring in 5% of any treatment group Adverse event (%) Diabetic patients Non-diabetic patients (n ¼ 57) Ran 750 mg BID (n ¼ 68) Ran 1000 mg BID (n ¼ 64) (n ¼ 212) Ran 750 mg BID (n ¼ 211) Ran 1000 mg BID (n ¼ 211) Patients with adverse events Discontinuations due to adverse events Angina Constipation Dizziness Nausea Asthenia during the 3-month double-blind period. did not appear to significantly lower fasting glucose concentrations during the 12-week treatment phase (Table 5 ). In Figure 3, it appears that levels of HbA 1c remained unchanged beyond 12 weeks of ranolazine treatment in the long-term, open-label extension study (for up to 2 years). The Day-50 point represents an average of all measurements taken at any time,100 days after the start of ranolazine during either CARISA or the long-term extension. For this analysis, data from all ranolazine-treated patients were combined, regardless of dose. A sensitivity analysis using only patients who had values in the latest time interval produced similar results. Adverse events As shown in Table 5, the proportion of patients with adverse events was similar between diabetic patients and nondiabetic patients. There were no notable differences in either laboratory values or vital signs in diabetic vs. nondiabetic patients. Effect of ranolazine on serum lipid levels in diabetic patients As shown in Table 6, ranolazine 1000 mg did not have an effect on lipids during the 12-week double-blind treatment period in CARISA, whereas the 750 mg dose was associated with an increase in low-density lipoprotein and total cholesterol. These disparate values suggest a spurious finding and thus do not appear to be of clinical significance. Discussion We previously reported the results of the CARISA trial whereby ranolazine was effective for treating chronic stable angina during 12 weeks of therapy in combination with atenolol, diltiazem, or amlodipine. 7 In CARISA, ranolazine compared with placebo treatment significantly improved exercise duration, time to onset of angina, and time to 1 mm ST-segment depression. In addition, ranolazine treatment reduced angina frequency and nitroglycerine usage compared with placebo treatment. As patients with diabetes account for 15 25% of the population enrolled in clinical trials of chronic angina therapy, 13 and diabetic patients accounted for 23% of those enrolled in the CARISA trial, we sought to evaluate the effects of ranolazine in this patient population. The adverse event profile and ECG effects (including prolongation of the QT interval) of ranolazine were extensively reviewed in both the CARISA and MARISA study publications. 7,8 This analysis indicated that ranolazine improved exercise parameters similarly in diabetic and non-diabetic patients. Treatment with ranolazine significantly decreased HbA 1c concentrations in diabetic patients. did not appear to be associated with a clinically significant change in serum lipids. The earliest time point that might show the full effect of a therapeutic intervention on HbA 1c is 12 weeks. 14 Therefore, CARISA was the first study of ranolazine in which the duration of double-blind treatment was long enough for HbA 1c to reach a new equilibrium. The reduction in HbA 1c was significant vs. placebo in diabetic patients whether or not they were treated with exogenous insulin. In particular, the decreases in HbA 1c seen in

6 in diabetic patients 47 Table 6 Glucose and serum lipids at week 12 (Mean + SE) Parameter Diabetic patients Non-diabetic patients 1000 mg BID 750 mg BID 1000 mg BID 750 mg BID Glucose, mg/dl baseline Change from baseline, (n ) (49) (63) (57) (186) (189) (189) Total cholesterol, mg/dl baseline Change from baseline, (n ) (51) (65) (58) (199) (198) (197) Triglycerides, mg/dl baseline Change from baseline, (n ) (51) (65) (58) (199) (197) (197) LDL, mg/dl baseline Change from baseline, (n ) (50) (55) (56) (192) (188) (186) HDL, mg/dl baseline Change from baseline, (n ) (51) (65) (58) (199) (197) (197) HDL/LDL ratio baseline Change from baseline, (n ) (50) (55) (56) (192) (188) (186) HDL, high-density lipoprotein; LDL, low-density lipoprotein. insulin-treated patients were greater than those seen in non-insulin-treated patients. In view of this observation, it can be speculated that an increase in insulin sensitivity may at least in part account for this observation. The significant response in HbA 1c that was observed following ranolazine treatment was equivalent to that expected with an oral antiglycaemic agent and thus represents a considerable potential for clinical benefit HbA 1c has become a standard measure of long-term glycaemic control. 18 Raised HbA 1c has been shown to be correlated with an increased incidence of diabetic complications including myocardial infarction. 13 Further, the UK Prospective Diabetes Study (UKPDS) has demonstrated that aggressive reduction of HbA 1c with intensive hypoglycaemic therapy is associated with a reduction in complications of diabetes in Type 2 diabetics. Target levels of HbA 1c of,7% have thus been incorporated into a number of guidelines. 12,19 Cardiologists have an increasing role in the management of diabetic patients because of the substantial co-morbidity of diabetes and ischaemic heart disease 17 and as a consequence, HbA 1c estimation is an integral part of cardiology practice. Although the changes in HbA 1c levels were not accompanied by similar changes in glucose in this study, ranolazine may offer a further therapeutic option in the management of the diabetic patient with ischaemic heart disease. HbA 1c was not a pre-planned efficacy variable in the CARISA trial, which had angina as its primary focus. For this reason, neither the presence of diabetes nor usage of insulin or other antidiabetic medication was a stratification factor in the randomization. Although the analysis of HbA 1c was exploratory, the apparent dose-response is indicative of an effect of therapy rather than an artifact. In addition, hypoglycaemic therapy was not strictly controlled, although few patients changed either the nature or the dose of their therapies during the study. Patient activity was also a possible confounding factor. However, the level of exercise demonstrated by Boule et al. 20 to be required to reduce HbA 1c to the extent seen in this study (three 45 min sessions of moderate aerobic exercise per week) would be unexpected in such an elderly angina population. HbA 1c values from the long-term follow-up study have no concurrent placebo control, and are thus subject to potential biases due to time trends in patients disease states, and the possibility of patients discontinuing for reasons that would correlate with future HbA 1c values. Because the data that would be needed to determine the nature of such a correlation cannot be observed, the magnitude of this potential bias is unknown. Conclusions Improvements in exercise duration and decreases in angina frequency appeared similar between diabetic and nondiabetic patients treated with ranolazine. Adverse events were similar between diabetic and non-diabetic populations. In patients with diabetes, ranolazine significantly decreased HbA 1c. The HbA 1c concentrations appeared to remain consistent over time during long-term treatment. The results of this study suggest that ranolazine may be an effective and well-tolerated anti-anginal medication in diabetic patients with chronic angina. Further, prospectively designed studies are warranted to confirm the findings of this analysis.

7 48 A. Timmis et al. Acknowledgement This study was supported by CV Therapeutics Inc. Conflict of interest: A.D.T. is a member of CV Therapeutics Europe Advisory Faculty. M.C. is an employee of CV Therapeutics Inc., which supported these studies. B.R.C. is a consultant to CV Therapeutics and has received research grant from CV Therapeutics. References 1. Mokdad AH, Bowman BA, Ford ES, Vinicor F, Marks JS, Koplan JP. The continuing epidemics of obesity and diabetes in the United States. JAMA 2001;286: Laakso M. Cardiovascular disease in type 2 diabetes: challenge for treatment and prevention. J Intern Med 2001;249: Uusitupa MI, Niskanen LK, Siitonen O, Voutilainen E, Pyorala K. Ten-year cardiovascular mortality in relation to risk factors and abnormalities in lipoprotein composition in type 2 (non-insulin-dependent) diabetic and non-diabetic subjects. Diabetologia 1993;36: International Diabetes Federation. Diabetes prevalence. idf.org/home/index.cfm?node5264 (2005). 5. Adult Treatment Panel III. Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) Final Report. Circulation 2002;106: Kjaergaard SC, Hansen HH, Fog L, Bulow I, Christensen PD. In-hospital outcome for diabetic patients with acute myocardial infarction in the thrombolytic era. Scand Cardiovasc J 1999;33: Chaitman BR, Pepine CJ, Parker JO, Skopal J, Chumakova G, Kuch J, Wang W, Skettino SL, Wolff AA. Effects of ranolazine with atenolol, amlodipine, or diltiazem on exercise tolerance and angina frequency in patients with severe chronic angina: a randomized controlled trial. JAMA 2004;291: Chaitman BR, Skettino SL, Parker JO, Hanley P, Meluzin J, Kuch J, Pepine CJ, Wang W, Nelson JJ, Hebert DA, Wolff A; MARISA Investigators. Antiischemic effects and long-term survival during ranolazine monotherapy in patients with chronic severe angina. J Am Coll Cardiol 2004; 43: Cocco G, Rousseau MF, Bouvy T, Cheron P, Williams G, Detry JM, Pouleur H. Effects of a new metabolic modulator, ranolazine, on exercise tolerance in angina pectoris patients treated with beta-blocker or diltiazem. J Cardiovasc Pharmacol 1992;20: Pepine CJ, Wolff AA. A controlled trial with a novel anti-ischemic agent, ranolazine, in chronic stable angina pectoris that is responsive to conventional antianginal agents. Study Group. Am J Cardiol 1999;84: Liang K, Zeger S. Longitudinal data analysis using generalized linear models. Biometrika 1986;73: American Diabetes Association. Standards of medical care in diabetes. Diabetes Care 2004;27(Suppl. 1):S15 S Bartnik M, Malmberg K, Hamsten A, Efendic S, Norhammar A, Silveira A, Tenerz A, Ohrvik J, Ryden L. Abnormal glucose tolerance a common risk factor in patients with acute myocardial infarction in comparison with population-based controls. J Intern Med 2004;256: Bunn HF, Haney DN, Kamin S, Gabbay KH, Gallop PM. The biosynthesis of human hemoglobin A1c. Slow glycosylation of hemoglobin in vivo. J Clin Invest 1976;57: UK Prospective Diabetes Study Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 1998;352: UK Prospective Diabetes Study Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet 1998;352: Stratton IM, Adler AI, Neil HA, Matthews DR, Manley SE, Cull CA, Hadden D, Turner RC, Holman RR. Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): prospective observational study. BMJ 2000;321: Derr R, Garrett E, Stacy GA, Saudek CD. Is HbA(1c) affected by glycemic instability? Diabetes Care 2003;26: McIntosh A, Hutchinson A, Home PD, Brown F, Bruce A, Damerell A, Davis R, Field R, Frost G, Marshall S, Roddick J, Tesfaye S, Withers H, Suckling R, Smith S, Griffin S, Kaltenthaler E, Peters J, Feder G. Clinical Guidelines and Evidence Review for Type 2 Diabetes: Management of Blood Glucose. Sheffield: University of Sheffield; Boule NG, Haddad E, Kenny GP, Wells GA, Sigal RJ. Effects of exercise on glycemic control and body mass in type 2 diabetes mellitus: a meta-analysis of controlled clinical trials. JAMA 2001;286: