OBJECTIVE—Thiazolidinediones (TZDs) and metformin are insulin-sensitizing antihyperglycemic agents with reported benefits on atherosclerosis. Despite extensive use in patients with diabetes and cardiovascular disease, there is a paucity of outcomes data with metformin and none yet with TZDs. We sought to determine the impact of these insulin sensitizers on outcomes in diabetic patients after hospitalization with acute myocardial infarction (AMI).

RESEARCH DESIGN AND METHODS—We conducted a retrospective cohort study of 24,953 Medicare beneficiaries with diabetes discharged after hospitalization with AMI between April 1998 and March 1999 or July 2000 and June 2001. The independent association between discharge prescription for metformin, TZD, or both agents and outcomes at 1 year was assessed in multivariable Cox proportional hazards models, adjusting for patient, physician, and hospital variables. The primary outcome was time to death within 1 year of discharge; secondary outcomes were time to first rehospitalization within 1 year of discharge for AMI, heart failure, and all causes.

RESULTS—There were 8,872 patients discharged on an antihyperglycemic agent, of which 819 were prescribed a TZD, 1,273 metformin, and 139 both drugs. After multivariable analysis, compared with patients prescribed an antihyperglycemic regimen that included no insulin sensitizer, mortality rates were not significantly different in patients treated with either metformin (hazard ratio [HR] 0.92 [95% CI 0.81–1.06]) or a TZD (0.92 [0.80–1.05]) but were lower in those prescribed both drugs (0.52 [0.34–0.82]). The results were similar among patients with heart failure. The prescription of a TZD was associated with a borderline higher risk of all-cause readmission (1.09[1.00–1.20]), predominately due to a higher risk for heart failure readmission (1.17 [1.05–1.30]).

CONCLUSIONS—Individually, prescription of insulin-sensitizing drugs is not associated with a significantly different risk of death in older diabetic patients within 1 year following AMI compared with other antihyperglycemic agents. Combined, however, metformin and TZDs may exert benefit. TZD prescription is associated with a higher risk of readmission for heart failure after myocardial infarction.

Ischemic heart disease (IHD) is the leading cause of mortality in patients with type 2 diabetes (1), an association that may relate, in part, to insulin resistance (2). To date, the benefit of antihyperglycemic therapies that target insulin deficiency (e.g., sulfonylureas or insulin) on IHD outcomes has not been shown (3). In contrast, metformin, which has insulin-sensitizing properties, may reduce myocardial infarction rates in newly diagnosed type 2 diabetic patients (4). Recently, the thiazolidinediones (TZDs), which specifically target insulin resistance and improve metabolic and vascular markers of the metabolic syndrome, have become increasingly popular (5). Although the TZDs are no more effective in lowering glucose than conventional agents, several experts have proposed that they reduce cardiovascular risk due to their pleiotropic effects (59). Despite a large and growing body of evidence demonstrating benefit on cardiovascular disease surrogates, however, there are no data proving a benefit of TZDs on IHD outcomes. Moreover, according to the Food and Drug Administration, both metformin and TZDs are contraindicated in many patients with heart failure, which frequently complicates AMI, due to concerns of lactic acidosis and increased fluid retention, respectively.

Ideally, the effect of metformin or TZDs on IHD outcomes would be addressed by randomized clinical trials. Although some are now underway, full results are not expected for several more years. In addition, many randomized trials of therapy in cardiovascular diseases do not include adequate numbers of elderly patients with significant comorbidity. It is therefore informative to investigate the effect of metformin and TZDs on IHD outcomes in population-based observational studies, which have the advantage of including a broad spectrum of patients typically seen in clinical practice. We therefore evaluated the relationship between the prescription of insulin-sensitizing agents and the outcomes of death and readmission among elderly diabetic patients after hospitalization for AMI.

Data sources

The National Heart Care Project.

The National Heart Care Project is a Centers for Medicare & Medicaid Services initiative designed to assess and improve the quality of care for Medicare beneficiaries with AMI. The data for the project consist of two longitudinal national cohorts of fee-for-service Medicare beneficiaries hospitalized with a principal discharge diagnosis of AMI. The first sample included hospitalizations with discharge dates between April 1998 and March 1999, inclusive, and the second between July 2000 and June 2001, inclusive. All admissions in each of the 50 states, Washington DC, and Puerto Rico identified within these time frames were sorted by age, sex, race, and hospital. Detailed demographic and clinical data were abstracted from these records by trained medical record reviewers. Data quality was enhanced through the use of medical record abstraction software and random record reabstraction.

Additional data sources.

The sample cohort was linked with the American Medical Association Physician Masterfile (10) and with the American Hospital Association Annual Surveys to ascertain the characteristics of the attending physicians and hospitals, respectively (11). Dates of death were assessed through linkage with the Medicare Enrollment Database (12). Linkage with part A Medicare data provided the dates and diagnoses for rehospitalizations occurring within a year of the index hospitalization.

Patients

The cohort consisted of 35,713 records in 1998–1999 and 35,407 in 2000–2001, for a total of 71,120 records. A clinical diagnosis of AMI was confirmed using standard clinical electrocardiogram and laboratory criteria (13).

Candidates for this study were those patients with a diagnosis of diabetes (n = 24,953) established by documentation in the hospital record or the prescription of an antihyperglycemic medication either at hospital admission or discharge. Because this study focused on the effects of diabetes drug therapy, only those patients that received an antihyperglycemic agent (metformin, TZDs, sulfonylurea, nonsulfonylurea insulin secretagogues, α-glucosidase inhibitors, or insulin) upon discharge were included (n = 14,786). Those patients in whom a diagnosis of AMI could not be confirmed were excluded (n = 2,829). Patients receiving long-term hemodialysis and those <65 years of age (n = 2,888) were also excluded because such Medicare beneficiaries may not be representative of older patients with cardiac disease. Finally, those patients who died during their hospitalizations (n = 3,070), were with unknown dates of death (n = 185) or unknown readmission data (n = 1,079), were discharged to hospice, were transferred to another hospital, or left against medical advice (n = 4,250) were also excluded. After the exclusion of those patients meeting one or more of the above criteria, a total sample size of 8,872 records were then analyzed.

Measurements

Variables.

The main independent variable was the prescription of metformin or a TZD at hospital discharge. A complete list of discharge medications was abstracted from the records of all patients surviving hospitalization. The only TZD available in the U.S. during the 1998–1999 sampling period was troglitazone (Rezulin; Parke-Davis), which was withdrawn in March 2000 (between sampling periods). By the beginning of the 2000–2001 sampling period, both rosiglitazone (Avandia; Glaxo-SmithKline; approved May 1999) and pioglitazone (Actos; Takeda; approved July 1999) were available for use in the U.S. Thus, a variable accounting for the sampling frame was included in all models.

Outcomes.

The primary outcome variable was time from hospital discharge to death from any cause censored at 1 year of discharge. Secondary outcomes included time to first readmission for myocardial infarction, first readmission for heart failure, and first readmission for any cause censored at 1 year of discharge.

Statistical analysis

Primary analysis.

Patients were categorized into four mutually exclusive groups by their antihyperglycemic drug prescription: patients who were prescribed an antihyperglycemic regimen that included neither metformin nor a TZD, patients prescribed metformin (but not a TZD), patients prescribed a TZD (but not metformin), and patients discharged on both metformin and a TZD. Differences in the characteristics of patient, physician, and hospital were compared pairwise using χ2 tests for categorical variables and t tests for continuous variables among these four groups, with the group of patients who were prescribed neither insulin-sensitizing drug as the referent group. The relationships between metformin or TZD prescription and outcomes were first compared with χ2 tests pairwise among the four groups, and the independent relationships were evaluated with multivariable Cox proportional hazards models. The multivariable models were adjusted for the following factors: demographics (age, sex, and race); cardiac history (history of heart failure, myocardial infarction, hypertension, and revascularization); noncardiovascular history (admission source, mobility, cerebral vascular accident, chronic pulmonary disease, urinary incontinence, and dementia); clinical characteristics at hospital admission (systolic blood pressure, respiratory rate, heart failure, sodium, glucose, blood urea nitrogen, creatinine, white blood cell count, and hematocrit); hospital course (atrial fibrillation, heart failure/pulmonary edema on admission chest radiograph, cardiac catheterization, percutaneous coronary intervention, coronary artery bypass grafting, and diabetes complications); other discharge prescriptions, including the prescription of other antihyperglycemic agents (sulfonylureas and nonsulfonylurea insulin secretagogues, α-glucosidase inhibitors, and insulin) and agents commonly used to reduce cardiovascular risk in post-AMI patients (aspirin, ACE inhibitors, β-adrenergic receptor blockers, calcium channel blockers, statins, and fibrates); and the sample frame in which the index hospitalization occurred. Covariates reflecting diabetes severity included the admission blood glucose level, prescription of other antihyperglycemic drugs at hospital discharge, and the presence of the ICD-9 code for a diabetes complication in the discharge abstract for the index hospitalization or in any part A record during the year before the admission. Complications included diabetic nephropathy, diabetic retinopathy, diabetic neuropathy, and peripheral vascular disease. Attending physician (specialty and board certification) and the treating hospital (bed size, geographic location, ownership, level of cardiac care facilities, and teaching status) were also included in the model. These factors were chosen based on both clinical significance and statistical significance (P < 0.05) using the stepwise method in the multivariate models. The models also adjusted for clustering of patients by hospitals.

Secondary analyses.

Because both TZDs and metformin are contraindicated in patients with advanced heart failure, analyses were stratified by the presence or absence of heart failure/pulmonary edema by admission chest radiograph and by the degree of left ventricular systolic dysfunction. Because edema occurs more frequently in TZD patients also treated with insulin, results were stratified by concomitant prescription of insulin at discharge.

Statistical analyses were performed with SAS 8.20 (SAS, Cary, NC) and Stata 7.0 (Stata, College Station, TX). The use of the National Heart Care Project database was approved by the Yale University Human Investigation Committee.

The mean age of the cohort was 76.4 ± 7.1 years, and the majority were Caucasian (88.2%) and were women (52.4%). Heart failure/pulmonary edema were present on admission chest radiograph in 30.1% of patients. A total of 1,273 (14.3%) were discharged on metformin, 819 (9.2%) on a TZD, and 139 (1.6%) on both insulin sensitizers (Table 1). A total of 6,641 (74.9%) patients were discharged on an antihyperglycemic drug regimen with neither insulin sensitizer, including 3,515 on a sulfonylurea and 3,509 on insulin. Compared with patients discharged on neither sensitizer, TZD-treated patients were younger but more likely to have a prior history of myocardial infarction, percutaneous coronary intervention, heart failure, or a documented diabetes-related complication. Patients treated with metformin were younger, more likely Caucasian, and less likely female. Proportionately fewer metformin patients had chronic illnesses, including heart failure, and were less likely to have diabetes-associated complications, including renal insufficiency. Metformin-treated patients also more often had preserved left ventricular systolic function and were more likely to have undergone invasive cardiac procedures during the index hospitalization. Finally, they were more often discharged on β-blockers, ACE inhibitors, or sulfonylureas but less often concomitantly treated with insulin. Patients prescribed both insulin sensitizers shared many but not all of the characteristics of metformin-only patients.

Mortality

Overall mortality in the cohort was 6.5% at 30 days, 20.0% at 6 months, and 28.3% at 1 year. In our fully adjusted model, 31 clinical factors were independently associated with mortality (12 with hazard ratio [HR] <1.0 and 19 with >1.0.) The six variables with the greatest positive association with mortality were prior coronary artery bypass grafting (HR 0.47 [95% CI 0.37–0.59]), prior percutaneous transluminal coronary angioplasty (0.72 [0.60–0.86]), cardiac catheterization (0.77 [0.68–0.88]) during the hospitalization, and discharged on a fibrate (0.71 [0.52–0.96]), statin (0.80 [0.73–0.88]), or β-blocker (0.85 [0.78–0.92]). The six variables with the greatest negative association with mortality were documented diabetes complications (1.47 [1.33–1.64]), left ventricular systolic dysfunction (1.46 [1.31–1.62]), age ≥85 years (1.46 [1.28–1.66]), admission from a skilled nursing, intermediate care, or extended care facility (1.44 [1.26–1.64]), a white blood cell count ≥11 mm3 on admission (1.44 [1.31–1.59]), and dementia (1.33 [1.18–1.51]).

Compared with patients not discharged on an insulin sensitizer, patients treated with metformin had significantly lower unadjusted mortality rates at all time points (5.1 vs. 6.8% at 30 days [P = 0.02], 13.0 vs. 21.6% at 6 months [P < 0.0001], and 19.3 vs. 30.3% at 1 year [P < 0.0001]) (Table 2). However, a similar trend was not observed for patients treated with a TZD (6.6, 19.8, and 28.9% [P = NS for all]). After adjustment for patient, provider, and hospital characteristics, sampling period, and differences in other medical treatments at discharge, both metformin-prescribed (HR 0.92 [95% CI 0.81–1.06]) and TZD-prescribed (0.92 [0.80–1.05]) patients (Table 3, Fig. 1A) had a mortality risk that was not significantly different from patients discharged without an insulin sensitizer. Because patients treated with sulfonylureas have different characteristics than patients treated with insulin, we conducted a second analysis in which the patients not treated with an insulin sensitizer were divided into two groups, those discharged on a sulfonylurea and those on insulin. Adjusted analysis showed no difference in mortality or readmissions between these two nonsensitizer groups. Our results were also unchanged by comparison of both groups to either the metformin or TZD patients (data not shown.)

Readmissions

At 1 year, 63.9% of patients had at least one readmission for any cause, with 18.4% being readmitted for myocardial infarction and 42.3% being readmitted for heart failure. Compared with patients not discharged on an insulin sensitizer, those discharged on metformin had significantly lower crude rates of readmission for all causes at 6 and 12 months, for heart failure at all time points, and there was a trend toward lower crude rates of readmission for myocardial infarction at the later time points (Table 2). In contrast, patients treated with a TZD had a higher risk of all-cause readmission of borderline statistical significance at 1 year and higher risks for readmission for heart failure at all time points, while readmission for myocardial infarction was similar.

After multivariable adjustment, the risk for readmission from all causes was no different among patients discharged on metformin than among those not discharged on an insulin sensitizer but was borderline higher among TZD-treated patients (HR 1.09 [95% CI 1.00–1.20]) (Table 3, Fig. 1B), predominately due to a higher risk of heart failure readmission (1.17 [1.05–1.30]). Readmission for myocardial infarction was not significantly different in either group.

Combination therapy and outcomes

After multivariable analysis, adjusted mortality in the small subset of patients treated with both metformin and a TZD (n = 139) was significantly lower (HR 0.52 [95% CI 0.34–0.82]) compared with those patients receiving no insulin sensitizer. The relationship between the two insulin sensitizers and mortality was greater in combination than individually (P = 0.045 for interaction between metformin and TZD) (Table 3, Fig. 1A). There were no differences in readmission rates in the combination treatment group (Table 3, Fig. 1B). The metformin-TZD combination group also had significantly lower mortality compared with either the sulfonylurea-treated (0.51 [0.32–0.81]) or the insulin-treated (0.54 [0.34–0.86]) patients.

Stratified analyses

In those patients with impaired left ventricular systolic function, mortality was not significantly higher in either insulin sensitizer group after multivariable analysis compared with the referent group (metformin [n = 406]: HR 0.92 [95% CI 0.72–1.18]; TZD [n = 255]: 1.04 [0.83–1.31]). Similarly, in those patients with heart failure/pulmonary edema on admission chest radiograph, mortality was also not increased with either insulin sensitizer (metformin [n = 312]: 0.96 [0.78–1.19]; TZD [n = 363]: 0.92 [0.75–1.13]).

The magnitude of the relationship between TZD therapy and heart failure readmissions was not substantively different from the referent group among patients additionally prescribed insulin on discharge (n = 386; HR 1.16 [95% CI 1.01–1.34]) versus those not on insulin (n = 433; 1.14 [0.96–1.36]) nor in those with impaired left ventricular function (n = 255; 1.15 [0.97–1.38]) versus those with preserved ventricular function (n = 354; 1.14 [0.95–1.38]) (P for interaction >0.05 for both comparisons). Also, in those patients who were readmitted for heart failure, mortality was not increased in those treated with metformin (n = 435; 0.87 [0.73–1.04]) or TZDs (n = 402; 0.92 [0.77–1.10]).

Our observational study of a large nationally representative sample of older patients with AMI demonstrates no large apparent short-term mortality benefit nor any mortality risk associated with the prescription of metformin or TZDs. TZD prescription was associated with a borderline relative increased risk of all cause readmission, predominately due to a 17% relative increased risk of readmission for heart failure. One conclusion that may be drawn from our data is that despite a host of prognostic factors that appear worse in TZD-prescribed patients, these drugs are not associated with any higher mortality, even in the unadjusted analyses. Conclusions regarding metformin are less clear. Patients prescribed metformin had significantly better risk profiles, as reflected by their lower crude mortality rates. Adjustments increase the HRs closer to 1. These data appear to indicate that metformin therapy post-AMI may offer little vascular benefit in the short term.

The modestly increased risk of heart failure readmission associated with TZDs is likely the result of the propensity for these drugs to cause fluid retention, usually manifested as peripheral edema alone (14). It should be noted, however, that the accuracy of the readmission diagnoses of heart failure cannot be assessed from these data. Interestingly, heart failure readmission rates were not any greater in those insulin-treated patients also prescribed a TZD (a combination known to increase the risk of edema) versus a noninsulin sensitizer. They were also not greater in TZD-treated patients with left ventricular systolic dysfunction. This might be an indirect reflection of some benefit on cardiac metabolism/ventricular function in patients with preexisting heart failure. The increased heart failure readmission risk was also not associated with any increased mortality. It is interesting to note that, in a different database recently analyzed by our group, Medicare patients discharged with a diagnosis of heart failure and treated with either metformin or a TZD had a 13% lower 1-year mortality, despite a borderline significant 6% increased risk of heart failure readmission in the TZD group (15). Thus, while the prescription of a TZD, compared with non–insulin sensitizer drugs, appears to confer no clear beneficial or adverse effect on mortality over the 1st year following AMI in older diabetic patients, their use is associated with a mildly increased risk of heart failure readmission. Close monitoring for fluid retention in these patients is clearly advisable (16).

When both metformin and TZDs were used in combination, a prominent relationship with lower mortality was demonstrated, with a risk reduction approaching 50%. Although the number of patients in this combination group was small (only 1.5% of the entire cohort), this relationship was consistent at all measured time points and persisted after multivariable analysis that accounted for a large number and variety of patient, physician, and hospital characteristics. There are few trials examining the combined use of these insulin sensitizer classes, with the majority assessing solely glycemic effects (1720). In one study, additive benefits on lipids were demonstrated (21). Conceptually, combined insulin sensitization could have significant cardiovascular benefits, but this finding needs to be confirmed by others and should not affect clinical decision-making at this point.

Insulin resistance, the primary metabolic defect in type 2 diabetes, has been associated with cardiovascular disease (22), although there is yet no compelling evidence from prospective studies to suggest that improving insulin sensitivity decreases cardiovascular event rates. The pharmacological agents that increase insulin sensitivity, metformin and the TZDs, appear to exert their insulin-sensitizing effects by different mechanisms (17), but both are associated with improvements in cardiovascular risk factors (5,6,23,24). Metformin decreases triglyceride and LDL cholesterol concentrations and the antifibrinolytic factor plasminogen activator inhibitor-1 and improves vasoreactivity (24). TZDs increase HDL cholesterol and reduce triglycerides, free fatty acids, plasminogen activator inhibitor-1, and the inflammatory markers C-reactive protein, matrix metalloproteinase-9, and CD40 ligand (5,6,23,25). In addition, TZDs improve endothelial function and markedly increase circulating concentrations of the adipocytokine adiponectin (26), which may have antiatherogenic properties (27). Studies in humans are limited to several investigations (2830) suggesting a benefit on carotid intimal-medial thickness, a surrogate for atherosclerosis, and two small studies (31,32) showing a reduction in restenosis rates after coronary angioplasty. In contrast to these studies that suggest benefit, metformin is known to increase homocysteine levels (33), an emerging cardiovascular disease risk factor, and some TZDs increase LDL cholesterol concentrations (although this may relate to these drugs’ effect on LDL particle size and not LDL particle number) (5,6). Also, both drugs are contraindicated in many patients with heart failure, which frequently accompanies or soon follows the diagnosis of AMI (16).

To date, no beneficial effect has been consistently demonstrated with any insulin sensitizer on IHD outcomes. Metformin monotherapy reduced AMI rates in a small subset of overweight, newly diagnosed type 2 diabetic patients within the UKPDS (U.K. Prospective Diabetes Study) (4). When used in combination with sulfonylureas in the same study, however, an increase in mortality was observed. TZDs have only been available since 1997, and it was not until 2001 that randomized, controlled clinical trials examining cardiovascular outcomes studies were initiated, data from which are not yet available.

Recently, small observational studies have suggested benefit from insulin sensitizers on IHD outcomes in patients with diabetes (3437). Each of these reports have had methodological limitations, including small sample sizes, retrospective design, failure to control for other potential confounders (3437), and relatively low exposure frequency to TZDs (36,37). Current studies underway to test the effect of insulin sensitizer therapy on cardiovascular outcomes in patients with diabetes include BARI-2D (personal communication, Dr. M.M. Brooks, 16 July 2004), PROactive (Prospective Pioglitazone Clinical Trial in Macrovascular Events) (38), and RECORD (Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycemia in Diabetes) (39). Their results are anticipated to shed further light on this important area.

Our study is the first large investigation into outcomes with insulin-sensitizing medications in diabetic patients following a major cardiovascular event. Yet, there are several issues to consider when interpreting our data. Our database is only able to measure outcomes up to 1 year following AMI, a time interval that may not be optimal to assess for the progression of already established, advanced atherosclerosis. Mortality at this time may indeed be driven primarily by the degree of myocardial damage, pump dysfunction, arrhythmic events, and embolic complications. It is possible that the benefits (and/or risks) of treatment with insulin sensitizers would be apparent over a longer time frame, such as after 3–5 years, as is conventionally used in cardiovascular trials. In addition, our database captures medications prescribed to the patients upon discharge, but it was not possible to determine compliance or changes in drug regimens in the follow-up period. Such misclassification would tend to bias the study toward the null. Because of the observational design, unmeasured confounding factors could influence the observed associations (or lack thereof). For example, we were not able to determine metabolic control after hospital discharge, specifically the quality of glucose (i.e., HbA1c), lipid, or blood pressure management, which has been, of course, associated with poorer cardiovascular outcomes by others. On the other hand, other retrospective “real world” studies of this type have replicated the results from subsequent randomized controlled trials (13,40). We were also not able to exclude patients with type 1 diabetes, although in this older Medicare population, it is unlikely that patients with type 1 diabetes represented more than a small minority of the cohort. Finally, since this study only analyzed data from older Medicare beneficiaries discharged from hospitals after AMI, the results may not apply to other patient populations, such as younger individuals or those with chronic, stable coronary artery disease.

In conclusion, this study suggests no clear mortality benefit or risk from the prescription of individual insulin-sensitizing drugs to older type 2 diabetic patients with AMI over the 1st year following hospital discharge compared with patients given other antihyperglycemic agents. However, the prescription of a TZD was associated with a mildly increased risk of rehospitalization for a diagnosis of heart failure. Current antihyperglycemic therapy guidelines, which do not favor one drug over another in patients with type 2 diabetes and cardiovascular disease, therefore appear appropriate until further information becomes available (41). In contrast, the suggestion of a marked mortality benefit from the combined metformin-TZD prescription is provocative in this setting but requires further confirmation. Adequately powered randomized, controlled clinical trials will be instrumental in determining rational antihyperglycemic regimens for the growing population of older patients with both type 2 diabetes and coronary artery disease.

Figure 1—

Adjusted Kaplan-Meier curves for mortality (A) and all-cause readmission (B) for the four treatment groups: patients discharged on an antihyperglycemic regimen that included neither a TZD nor metformin (solid line); patients with a discharge prescription for a TZD (but not metformin) (dotted line); patients with a discharge prescription for metformin (but not a TZD) (dashed line); patients with discharge prescriptions for both a TZD and metformin (dotted and dashed line).

Figure 1—

Adjusted Kaplan-Meier curves for mortality (A) and all-cause readmission (B) for the four treatment groups: patients discharged on an antihyperglycemic regimen that included neither a TZD nor metformin (solid line); patients with a discharge prescription for a TZD (but not metformin) (dotted line); patients with a discharge prescription for metformin (but not a TZD) (dashed line); patients with discharge prescriptions for both a TZD and metformin (dotted and dashed line).

Close modal
Table 1—

The patient population

No insulin sensitizer
Metformin
Thiazolidinedione
Both
n = 6,641n = 1,273P*n = 819P*n = 139P*
Demographics        
    Age (years) 76.8 ± 7.1 75.2 ± 7.0 <0.001 75.8 ± 6.9 <0.001 73.6 ± 6.8 <0.001 
    Female (%)* 53.2%† 48.2%* 0.001 53%* 0.904 46.8%* 0.131 
    Nonwhite 12.5 8.6 <0.001 10.7 0.143 11.5 0.720 
Cardiac history        
    Congestive heart failure 41.6 28.1 <0.001 46.6 0.006 30.9 0.012 
    Myocardial infarction 44.5 39.1 <0.001 52.0 <0.001 50.4 0.172 
    Hypertension 80.3 77.6 0.031 80.7 0.760 76.3 0.242 
    Coronary artery bypass graft 22.7 21.1 0.195 23.9 0.431 31.7 0.013 
    Percutaneous coronary intervention 16.0 17.1 0.328 19.2 0.022 20.9 0.124 
Noncardiac history        
    Cerebrovascular accident 24.1 18.7 <0.001 25.6 0.339 19.4 0.199 
    Chronic obstructive pulmonary disease 22.8 19.2 0.004 21.1 0.276 19.4 0.346 
    Dementia/Alzheimer’s disease 9.2 5.8 <0.001 7.7 0.159 9.4 0.946 
Diabetes complications (12 months prior to admission) 13.7 6.0 <0.001 17.2 0.006 5.0 0.003 
    Diabetic coma (250.1, 250.2, 250.3) 0.6 0.2 0.114 0.6 0.935 0.0 0.365 
    Renal complications (250.4) 4.9 0.5 <0.001 6.0 0.165 0.7 0.023 
    Ophthalmic manifestation (250.5) 2.5 1.0 0.001 3.3 0.166 2.2 0.807 
    Neurological manifestations (250.6) 5.1 3.3 0.006 7.2 0.012 2.9 0.233 
    Peripheral circulatory disorders (250.7) 2.2 1.0 0.007 1.8 0.529 0.0 0.079 
    Other complications/ manifestations (250.8, 250.9) 4.1 1.8 <0.001 4.5 0.525 0.7 0.047 
Presenting features        
    ST elevation on electrocardiogram 23.9 24.7 0.517 20.6 0.038 19.4 0.220 
    Congestive heart failure/pulmonary edema 31.4 22.2 <0.001 33.7 0.175 21.6 0.014 
    Anterior AMI 31.4 31.7 0.837 31.0 0.837 27.3 0.311 
Admission laboratory values        
    Glucose   0.064  0.758  0.529 
        <136 mg/dl 14.1 12.2  14.7  11.5  
        136–185 mg/dl 19.7 21.2  19.5  16.5  
        186–250 mg/dl 28.5 30.9  29.8  29.5  
        >250 mg/dl 37.7 35.7  36.0  42.4  
    Creatinine   <0.001  <0.001  <0.001 
        <1.5 mg/dl 50.2 72.9  41.6  66.9  
        1.5–2.5 mg/dl 31.7 22.5  38.0  29.5  
        >2.5 mg/dl 18.1 4.6  20.4  3.6  
In-hospital events        
    Atrial fibrillation 17.9 15.6 0.040 16.8 0.438 11.5 0.050 
    Ventricular tachycardia 0.6 0.5 0.498 0.9 0.458 0.7 0.898 
    Congestive heart failure/pulmonary edema on chest radiograph 35.5 28.5 <0.001 38.1 0.145 20.9 <0.001 
    Left ventricular systolic function   <0.001  0.290  0.003 
        Preserved (normal or mild) 40.4 46.4  43.2  54.0  
        Impaired (moderate or severe) 32.9 31.9  31.1  21.6  
        Not measured 26.7 21.7  25.6  24.5  
    Cardiac catheterization 41.4 53.8 <0.001 43.5 0.263 55.4 0.001 
    Primary reperfusion        
        Early thrombolytic therapy 2.4 2.7 0.515 1.7 0.194 2.9 0.741 
        Early percutaneous coronary intervention 1.4 1.4 0.931 0.9 0.172 3.6 0.038 
    Percutaneous coronary intervention beyond 12 h of admission 17.2 26.2 <0.001 20.0 0.048 21.6  
    Coronary artery bypass graft 10.0 11.9 0.043 8.4 0.157 10.1 0.181 
    Diabetes complications 24.4 13.5 <0.001 29.1 0.004 20.1 0.248 
Discharge medications        
    ACE inhibitor 54.8 59.9 0.001 54.5 0.848 71.9 <0.001 
    β-Blocker 63.2 68.8 <0.001 66.4 0.074 74.1 0.009 
    Calcium channel blocker 25.8 23.6 0.095 27.6 0.268 18.0 0.037 
    Aspirin 76.0 79.7 0.005 75.7 0.844 77.0 0.792 
    Lipid-lowering agents        
        Statin 32.1 42.3 <0.001 48.0 <0.001 57.6 <0.001 
        Fibrates 2.3 2.9 0.211 3.2 0.132 4.3 0.125 
    Other anticoagulants 0.4 0.5 0.743 1.3 <0.001 2.2 0.002 
    Antihyperglycemic agents        
        Sulfonylureas 52.9 58.9 <0.001 39.3 <0.001 54.0 0.810 
        Nonsulfonylurea secretagogues 1.0 1.3 0.298 1.6 0.129 3.6 0.003 
        α-Glucosidase inhibitors 0.6 0.5 0.731 0.9 0.458 2.2 0.028 
        Insulin 52.8 14.5 <0.001 47.1 0.002 23.7 <0.001 
Attending physican specialty   <0.001  <0.001  <0.001 
    Cardiologist 27.3 33.2  29.5  42.4  
    Noncardiologist 38.2 38.7  45.1  41.0  
    Unknown 34.6 28.0  25.4  16.5  
Hospital characteristics        
    Rural location 25.9 28.4 0.063 27.2 0.398 30.2 0.246 
    Ownership   0.002  0.858  0.022 
        Public 10.7 10.8  10.3  9.4  
        Not for profit 76.9 73.4  76.7  69.8  
        For profit 9.2 10.8  10.0  16.5  
    Teaching status   0.002  0.215  0.745 
        COTH (Council of Teaching Hospitals) member 17.3 14.6  14.8  18.7  
        Residency affiliated 28.5 28.6  28.0  25.2  
        Nonteaching 50.9 51.8  54.2  51.8  
    Cardiac care facilities   <0.001  0.132  0.033 
        Cardiac surgery suite 50.4 54.8  53.1  59.7  
        Cardiac catheterization lab 15.5 13.0  16.7  15.8  
        No invasive facilities 26.7 22.8  23.0  15.8  
        Sampling period   <0.001  <0.001  <0.001 
    1998–1999 52.0 40.5  31.6  14.4  
    2000–2001 48.0 59.5  68.4  85.6  
No insulin sensitizer
Metformin
Thiazolidinedione
Both
n = 6,641n = 1,273P*n = 819P*n = 139P*
Demographics        
    Age (years) 76.8 ± 7.1 75.2 ± 7.0 <0.001 75.8 ± 6.9 <0.001 73.6 ± 6.8 <0.001 
    Female (%)* 53.2%† 48.2%* 0.001 53%* 0.904 46.8%* 0.131 
    Nonwhite 12.5 8.6 <0.001 10.7 0.143 11.5 0.720 
Cardiac history        
    Congestive heart failure 41.6 28.1 <0.001 46.6 0.006 30.9 0.012 
    Myocardial infarction 44.5 39.1 <0.001 52.0 <0.001 50.4 0.172 
    Hypertension 80.3 77.6 0.031 80.7 0.760 76.3 0.242 
    Coronary artery bypass graft 22.7 21.1 0.195 23.9 0.431 31.7 0.013 
    Percutaneous coronary intervention 16.0 17.1 0.328 19.2 0.022 20.9 0.124 
Noncardiac history        
    Cerebrovascular accident 24.1 18.7 <0.001 25.6 0.339 19.4 0.199 
    Chronic obstructive pulmonary disease 22.8 19.2 0.004 21.1 0.276 19.4 0.346 
    Dementia/Alzheimer’s disease 9.2 5.8 <0.001 7.7 0.159 9.4 0.946 
Diabetes complications (12 months prior to admission) 13.7 6.0 <0.001 17.2 0.006 5.0 0.003 
    Diabetic coma (250.1, 250.2, 250.3) 0.6 0.2 0.114 0.6 0.935 0.0 0.365 
    Renal complications (250.4) 4.9 0.5 <0.001 6.0 0.165 0.7 0.023 
    Ophthalmic manifestation (250.5) 2.5 1.0 0.001 3.3 0.166 2.2 0.807 
    Neurological manifestations (250.6) 5.1 3.3 0.006 7.2 0.012 2.9 0.233 
    Peripheral circulatory disorders (250.7) 2.2 1.0 0.007 1.8 0.529 0.0 0.079 
    Other complications/ manifestations (250.8, 250.9) 4.1 1.8 <0.001 4.5 0.525 0.7 0.047 
Presenting features        
    ST elevation on electrocardiogram 23.9 24.7 0.517 20.6 0.038 19.4 0.220 
    Congestive heart failure/pulmonary edema 31.4 22.2 <0.001 33.7 0.175 21.6 0.014 
    Anterior AMI 31.4 31.7 0.837 31.0 0.837 27.3 0.311 
Admission laboratory values        
    Glucose   0.064  0.758  0.529 
        <136 mg/dl 14.1 12.2  14.7  11.5  
        136–185 mg/dl 19.7 21.2  19.5  16.5  
        186–250 mg/dl 28.5 30.9  29.8  29.5  
        >250 mg/dl 37.7 35.7  36.0  42.4  
    Creatinine   <0.001  <0.001  <0.001 
        <1.5 mg/dl 50.2 72.9  41.6  66.9  
        1.5–2.5 mg/dl 31.7 22.5  38.0  29.5  
        >2.5 mg/dl 18.1 4.6  20.4  3.6  
In-hospital events        
    Atrial fibrillation 17.9 15.6 0.040 16.8 0.438 11.5 0.050 
    Ventricular tachycardia 0.6 0.5 0.498 0.9 0.458 0.7 0.898 
    Congestive heart failure/pulmonary edema on chest radiograph 35.5 28.5 <0.001 38.1 0.145 20.9 <0.001 
    Left ventricular systolic function   <0.001  0.290  0.003 
        Preserved (normal or mild) 40.4 46.4  43.2  54.0  
        Impaired (moderate or severe) 32.9 31.9  31.1  21.6  
        Not measured 26.7 21.7  25.6  24.5  
    Cardiac catheterization 41.4 53.8 <0.001 43.5 0.263 55.4 0.001 
    Primary reperfusion        
        Early thrombolytic therapy 2.4 2.7 0.515 1.7 0.194 2.9 0.741 
        Early percutaneous coronary intervention 1.4 1.4 0.931 0.9 0.172 3.6 0.038 
    Percutaneous coronary intervention beyond 12 h of admission 17.2 26.2 <0.001 20.0 0.048 21.6  
    Coronary artery bypass graft 10.0 11.9 0.043 8.4 0.157 10.1 0.181 
    Diabetes complications 24.4 13.5 <0.001 29.1 0.004 20.1 0.248 
Discharge medications        
    ACE inhibitor 54.8 59.9 0.001 54.5 0.848 71.9 <0.001 
    β-Blocker 63.2 68.8 <0.001 66.4 0.074 74.1 0.009 
    Calcium channel blocker 25.8 23.6 0.095 27.6 0.268 18.0 0.037 
    Aspirin 76.0 79.7 0.005 75.7 0.844 77.0 0.792 
    Lipid-lowering agents        
        Statin 32.1 42.3 <0.001 48.0 <0.001 57.6 <0.001 
        Fibrates 2.3 2.9 0.211 3.2 0.132 4.3 0.125 
    Other anticoagulants 0.4 0.5 0.743 1.3 <0.001 2.2 0.002 
    Antihyperglycemic agents        
        Sulfonylureas 52.9 58.9 <0.001 39.3 <0.001 54.0 0.810 
        Nonsulfonylurea secretagogues 1.0 1.3 0.298 1.6 0.129 3.6 0.003 
        α-Glucosidase inhibitors 0.6 0.5 0.731 0.9 0.458 2.2 0.028 
        Insulin 52.8 14.5 <0.001 47.1 0.002 23.7 <0.001 
Attending physican specialty   <0.001  <0.001  <0.001 
    Cardiologist 27.3 33.2  29.5  42.4  
    Noncardiologist 38.2 38.7  45.1  41.0  
    Unknown 34.6 28.0  25.4  16.5  
Hospital characteristics        
    Rural location 25.9 28.4 0.063 27.2 0.398 30.2 0.246 
    Ownership   0.002  0.858  0.022 
        Public 10.7 10.8  10.3  9.4  
        Not for profit 76.9 73.4  76.7  69.8  
        For profit 9.2 10.8  10.0  16.5  
    Teaching status   0.002  0.215  0.745 
        COTH (Council of Teaching Hospitals) member 17.3 14.6  14.8  18.7  
        Residency affiliated 28.5 28.6  28.0  25.2  
        Nonteaching 50.9 51.8  54.2  51.8  
    Cardiac care facilities   <0.001  0.132  0.033 
        Cardiac surgery suite 50.4 54.8  53.1  59.7  
        Cardiac catheterization lab 15.5 13.0  16.7  15.8  
        No invasive facilities 26.7 22.8  23.0  15.8  
        Sampling period   <0.001  <0.001  <0.001 
    1998–1999 52.0 40.5  31.6  14.4  
    2000–2001 48.0 59.5  68.4  85.6  

Data are means ± SE and %.

*

P for comparison with the “no insulin sensitizer” group.

Table 2—

Crude outcomes

OutcomeNo insulin sensitizer
Metformin
Thiazolidinedione
Both
n%n%Pn%Pn%P
Mortality            
    within 30 days of discharge 453 6.8 65 5.1 0.023 54 6.6 0.807 2.2 0.030 
    within 3 months of discharge 954 14.4 113 8.9 0.000 117 14.3 0.951 5.0 0.002 
    within 6 months of discharge 1,431 21.5 165 13.0 <0.001 162 19.8 0.244 12 8.6 <0.001 
    within 1 year of discharge 2,014 30.3 246 19.3 <0.001 237 28.9 0.414 17 12.2 <0.001 
Readmission            
    Myocardial infarction            
        within 30 days of discharge 331 5.0 63 4.9 0.958 46 5.6 0.436 2.9 0.257 
        within 3 months of discharge 664 10.0 118 9.3 0.425 78 9.5 0.668 11 7.9 0.417 
        within 6 months of discharge 953 14.4 155 12.2 0.041 107 13.1 0.320 18 12.9 0.641 
        within 1 year of discharge 1,247 18.8 210 16.5 0.054 154 18.8 0.986 21 15.1 0.272 
    Heart failure            
        within 30 days of discharge 1,046 15.8 162 12.7 0.006 159 19.4 0.007 17 12.2 0.259 
        within 3 months of discharge 1,808 27.2 280 22.0 <0.001 260 31.7 0.006 38 27.3 0.976 
        within 6 months of discharge 2342 35.3 354 27.8 <0.001 313 38.2 0.096 50 36.0 0.863 
        within 1 year of discharge 2859 43.1 435 34.2 <0.001 402 49.1 0.001 54 38.8 0.322 
    All cause            
        within 30 days of discharge 1669 25.1 309 24.3 0.517 235 28.7 0.027 29 20.9 0.250 
        within 3 months of discharge 2829 42.6 489 38.4 0.006 385 47.0 0.016 54 38.8 0.376 
        within 6 months of discharge 3596 54.1 610 47.9 <0.001 465 56.8 0.154 70 50.4 0.375 
        within 1 year of discharge 4268 64.3 759 59.6 0.002 555 67.8 0.048 88 63.3 0.816 
OutcomeNo insulin sensitizer
Metformin
Thiazolidinedione
Both
n%n%Pn%Pn%P
Mortality            
    within 30 days of discharge 453 6.8 65 5.1 0.023 54 6.6 0.807 2.2 0.030 
    within 3 months of discharge 954 14.4 113 8.9 0.000 117 14.3 0.951 5.0 0.002 
    within 6 months of discharge 1,431 21.5 165 13.0 <0.001 162 19.8 0.244 12 8.6 <0.001 
    within 1 year of discharge 2,014 30.3 246 19.3 <0.001 237 28.9 0.414 17 12.2 <0.001 
Readmission            
    Myocardial infarction            
        within 30 days of discharge 331 5.0 63 4.9 0.958 46 5.6 0.436 2.9 0.257 
        within 3 months of discharge 664 10.0 118 9.3 0.425 78 9.5 0.668 11 7.9 0.417 
        within 6 months of discharge 953 14.4 155 12.2 0.041 107 13.1 0.320 18 12.9 0.641 
        within 1 year of discharge 1,247 18.8 210 16.5 0.054 154 18.8 0.986 21 15.1 0.272 
    Heart failure            
        within 30 days of discharge 1,046 15.8 162 12.7 0.006 159 19.4 0.007 17 12.2 0.259 
        within 3 months of discharge 1,808 27.2 280 22.0 <0.001 260 31.7 0.006 38 27.3 0.976 
        within 6 months of discharge 2342 35.3 354 27.8 <0.001 313 38.2 0.096 50 36.0 0.863 
        within 1 year of discharge 2859 43.1 435 34.2 <0.001 402 49.1 0.001 54 38.8 0.322 
    All cause            
        within 30 days of discharge 1669 25.1 309 24.3 0.517 235 28.7 0.027 29 20.9 0.250 
        within 3 months of discharge 2829 42.6 489 38.4 0.006 385 47.0 0.016 54 38.8 0.376 
        within 6 months of discharge 3596 54.1 610 47.9 <0.001 465 56.8 0.154 70 50.4 0.375 
        within 1 year of discharge 4268 64.3 759 59.6 0.002 555 67.8 0.048 88 63.3 0.816 
Table 3—

Adjusted outcomes

MetforminPThiazolidinedionesPBothP
Mortality 0.92 (0.81–1.06) 0.255 0.92 (0.80–1.05) 0.221 0.52 (0.34–0.82) 0.004 
Myocardial infarction readmission 1.02 (0.86–1.20) 0.831 0.92 (0.77–1.10) 0.355 0.88 (0.56–1.37) 0.568 
Heart failure readmission 1.06 (0.95–1.18) 0.334 1.17 (1.05–1.30) 0.004 1.24 (0.94–1.63) 0.122 
All cause Readmission 1.04 (0.96–1.13) 0.368 1.09 (1.00–1.20) 0.052 1.06 (0.87–1.30) 0.543 
MetforminPThiazolidinedionesPBothP
Mortality 0.92 (0.81–1.06) 0.255 0.92 (0.80–1.05) 0.221 0.52 (0.34–0.82) 0.004 
Myocardial infarction readmission 1.02 (0.86–1.20) 0.831 0.92 (0.77–1.10) 0.355 0.88 (0.56–1.37) 0.568 
Heart failure readmission 1.06 (0.95–1.18) 0.334 1.17 (1.05–1.30) 0.004 1.24 (0.94–1.63) 0.122 
All cause Readmission 1.04 (0.96–1.13) 0.368 1.09 (1.00–1.20) 0.052 1.06 (0.87–1.30) 0.543 

Data are HR (95% CI).

The analyses upon which this publication is based were performed under contract no. 500-02-CO01, entitled “Utilization and Quality Control Peer Review Organization for the State of Colorado,” sponsored by the Centers for Medicare & Medicaid Services (formerly Health Care Financing Administration), Department of Health and Human Services. The content of the publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. government. The author assumes full responsibility for the accuracy and completeness of the ideas presented. This article is a direct result of the Health Care Quality Improvement Program initiated by the Centers for Medicare & Medicaid Services, which has encouraged identification of quality improvement projects derived from analysis of patterns of care, and therefore required no special funding on the part of this contractor. Ideas and contributions to the author concerning experience in engaging with issues presented are welcomed.

1.
Hurst RT, Lee RW: Increased incidence of coronary atherosclerosis in type 2 diabetes mellitus: mechanisms and management.
Ann Intern Med
139
:
824
–834,
2003
2.
Haffner SM, D’Agostino R Jr, Mykkanen L, Tracy R, Howard B, Rewers M, Selby J, Savage PJ, Saad MF: Insulin sensitivity in subjects with type 2 diabetes: relationship to cardiovascular risk factors: the Insulin Resistance Atherosclerosis Study.
Diabetes Care
22
:
562
–568,
1999
3.
UK Prospective Diabetes Study (UKPDS) 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
352
:
837
–853,
1998
4.
UK Prospective Diabetes Study (UKPDS) Group: Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34).
Lancet
352
:
854
–865,
1998
5.
Uwaifo GI, Ratner RE: The roles of insulin resistance, hyperinsulinemia, and thiazolidinediones in cardiovascular disease.
Am J Med
115 (Suppl. 8A)
:
12S
–19S,
2003
6.
Parulkar AA, Pendergrass ML, Granda-Ayala R, Lee TR, Fonseca VA: Nonhypoglycemic effects of thiazolidinediones.
Ann Intern Med
134
:
61
–71,
2001
7.
Ginsberg H, Plutzky J, Sobel BE: A review of metabolic and cardiovascular effects of oral antidiabetic agents: beyond glucose-level lowering.
J Cardiovasc Risk
6
:
337
–346,
1999
8.
Fonseca VA: Management of diabetes mellitus and insulin resistance in patients with cardiovascular disease.
Am J Cardiol
92
:
50J
–60J,
2003
9.
Dandona P, Aljada A, Chaudhuri A, Bandyopadhyay A: The potential influence of inflammation and insulin resistance on the pathogenesis and treatment of atherosclerosis-related complications in type 2 diabetes.
J Clin Endocrinol Metab
88
:
2422
–2429,
2003
10.
Kenward K: The scope of the data available in the AMA’s Physician Masterfile.
Am J Public Health
86
:
1481
–1482,
1996
11.
American Hospital Association:
The American Hospital Association annual survey database: fiscal year 1998 and 2000 documentation
. Chicago, IL, Health Forum, 1999 and 2001
12.
Fleming C, Fisher ES, Chang CH, Bubolz TA, Malenka DJ: Studying outcomes and hospital utilization in the elderly: the advantages of a merged data base for Medicare and Veterans Affairs hospitals.
Med Care
30
:
377
–391,
1992
13.
Krumholz HM, Chen YT, Wang Y, Radford MJ: Aspirin and angiotensin-converting enzyme inhibitors among elderly survivors of hospitalization for an acute myocardial infarction.
Arch Intern Med
161
:
538
–544,
2001
14.
Tang WHH, Francis GS, Hoogwerf BJ, Young JB: Fluid retention after initiation of thiazolidinediones therapy in diabetic patients with established chronic heart failure clinic.
J Am Coll Cardiol
41
:
1394
–1398,
2003
15.
Masoudi FA, Inzucchi SE, Wang Y, Havranek EP, Foody JM, Krumholz HM: Thiazolidinediones, metformin, and outcomes in older patients with diabetes and heart failure: an observational study.
Circulation
111
:
583
–590,
2005
16.
Nesto RW, Bell D, Bonow RO, Fonseca V, Grundy SM, Horton ES, Le Winter M, Porte D, Semenkovich CF, Smith S, Young LH, Kahn R: Thiazolidinedione use, fluid retention, and congestive heart failure: a consensus statement from the American Heart Association and American Diabetes Association.
Circulation
108
:
2941
–2948,
2003
17.
Inzucchi SE, Maggs DG, Spollett GR, Page SL, Rife FS, Walton V, Shulman GI: Efficacy and metabolic effects of metformin and troglitazone in type II diabetes mellitus.
N Engl J Med
338
:
867
–872,
1998
18.
Fonseca V, Rosenstock J, Patwardhan R, Salzman A: Effect of metformin and rosiglitazone combination therapy in patients with type 2 diabetes mellitus: a randomized controlled trial.
JAMA
283
:
1695
–1702,
2000
19.
Einhorn D, Rendell M, Rosenzweig J, Egan JW, Mathisen AL, Schneider RL: Pioglitazone hydrochloride in combination with metformin in the treatment of type 2 diabetes mellitus: a randomized, placebo-controlled study: the Pioglitazone 027 Study Group.
Clin Therapeut
22
:
1395
–1409,
2000
20.
Yale JF, Valiquett TR, Ghazzi MN, Owens-Grillo JK, Whitcomb RW, Foyt HL: The effect of a thiazolidinedione drug, troglitazone, on glycemia in patients with type 2 diabetes mellitus poorly controlled with sulfonylurea and metformin: a multicenter, randomized, double blind, placebo-controlled trial.
Ann Intern Med
134
:
737
–745,
2001
21.
Lee I, Maggs D, Katz D, Spollett GR, Page SL, Rife FS, Inzucchi SE: Effects on lipid metabolism of metformin and troglitazone, alone and in combination, in patients with type 2 diabetes mellitus (Abstract).
Diabetes
51 (Suppl. 2)
:
A105
,
2002
22.
Reaven GM: Insulin resistance/compensatory hyperinsulinemia, essential hypertension, and cardiovascular disease.
J Clin Endocrinol Metab
88
:
2399
–2403,
2003
23.
Chu NV, Kong AP, Kim DD, Armstrong D, Baxi S, Deutsch R, Caulfield M, Mudaliar SR, Reitz R, Henry RR, Reaven PD: Differential effects of metformin and troglitazone on cardiovascular risk factors in patients with type 2 diabetes.
Diabetes Care
25
:
542
–549,
2002
24.
Hundal RS, Inzucchi SE: Metformin: new understandings, new uses.
Drugs
63
:
1879
–1894,
2003
25.
Haffner SM, Greenberg AS, Weston WM, Chen H, Williams K, Freed MI: Effect of rosiglitazone treatment on nontraditional markers of cardiovascular disease in patients with type 2 diabetes mellitus.
Circulation
106
:
679
–684,
2002
26.
Phillips SA, Ciaraldi TP, Kong AP, Bandukwala R, Aroda V, Carter L, Baxi S, Mudaliar SR, Henry RR: Modulation of circulating and adipose tissue adiponectin levels by antidiabetic therapy.
Diabetes
52
:
667
–674,
2003
27.
Kumada M, Kihara S, Sumitsuji S, Kawamoto T, Matsumoto S, Ouchi N, Arita Y, Okamoto Y, Shimomura I, Hiraoka H, Nakamura T, Funahashi T, Matsuzawa Y, Osaka CAD Study Group: Coronary artery disease: association of hypoadiponectinemia with coronary artery disease in men.
Arterio Thromb Vasc Biol
23
:
85
–89,
2003
28.
Minamikawa J, Tanaka S, Yamauchi M, Inoue D, Koshiyama H: Potent inhibitory effect of troglitazone on carotid arterial wall thickness in type 2 diabetes.
J Clin Endocrinol Metab
83
:
1818
–1820,
1998
29.
Koshiyama H, Shimono D, Kuwamura N, Minamikawa J, Nakamura Y: Rapid communication: inhibitory effect of pioglitazone on carotid arterial wall thickness in type 2 diabetes.
J Clin Endocrinol Metab
86
:
3452
–3456,
2001
30.
Hodis HN, Buchanan TA, Li Y, Torres M, Mack WJ: PPAR activation and progression of subclinical atherosclerosis in insulin requiring type 2 diabetes mellitus (Abstract).
Circulation
108 (Suppl. IV)
:
741
,
2003
31.
Takagi T, Akasaka T, Yamamuro A, Honda Y, Hozumi T, Morioka S, Yoshida K: Troglitazone reduces neointimal tissue proliferation after coronary stent implantation in patients with non-insulin dependent diabetes mellitus: a serial intravascular ultrasound study.
J Am Coll Cardiol
36
:
1529
–1535,
2000
32.
Takagi T, Yamamuro A, Tamita K, Yamabe K, Katayama M, Mizoguchi S, Ibuki M, Tani T, Tanabe K, Nagai K, Shiratori K, Morioka S, Yoshikawa J: Pioglitazone reduces neointimal tissue proliferation after coronary stent implantation in patients with type 2 diabetes mellitus: an intravascular ultrasound scanning study (Abstract).
Am Heart J
146
:
E5
,
2003
33.
Aarsand AK, Carlsen SM: Folate administration reduces circulating homocysteine levels in NIDDM patients on long-term metformin treatment.
J Intern Med
244
:
169
–174,
1998
34.
Sauer WH, Berlin JA, Kimmel SE: Insulin sensitizing drug use is associated with a reduced risk of myocardial infarction in type 2 diabetics (Abstract).
Circulation
108 (Suppl. IV)
:
767
,
2003
35.
Teirstein PS, Kao JA, Castarella P, Huppe G, Sirkin K, Grise M, Giap H, Koka A, Tripuraneni P: Insulin sensitizers are associated with improved outcomes in diabetic patients undergoing brachytherapy.
J Am Coll Cardiol
43 (Suppl. A)
:
1025
–1041,
2004
36.
Mulestein JB, Pearson RR, Horne BD, Bair TL, Thomas H, Lappe DL, Anderson JL, Jones HU, Lavasani F, Einhorn D, Renlund DG: Use of either metformin or thiazolidinedione is associated with improved survival among paients with type 2 diabetes from a registry of 16,203 diabetic patients.
J Am Coll Cardiol
43 (Suppl. A)
:
810
–815,
2004
37.
Kao J, Tobis J, McClelland RL, Heaton MR, Davis BR, Holmes Dr, Currier JW: Relation of metformin treatment to clinical events indiabetic patients undergoing percutneous intervention.
Am J Cardiol
93
:
1347
–1350,
2004
38.
Charbonnel B, Dormandy J, Erdmann E, Massi-Benedetti M, Skene A, PROactive Study Group: The Prospective Pioglitazone Clinical Trial in Macrovacular Events.
Diabetes Care
27
:
1647
–1653,
2004
39.
Home PD, Pocock S, Beck-Nielsen H, Gomis R, Hanefeld M, Dargie H, Komajda M, Jones NP, Garcia S, Curtis P, the RECORD Study Group:
Rosiglitazone evaluated for cardiac outcomes and regulation of glycaemia in diabetes (RECORD): an interim analysis of glycaemia at 18 months
. Abstract no. 725, presented at 40th Annual Meeting of the European Association for the Study of Diabetes, 5–9 September 2004, Munich, Germany
40.
Chen J, Marciniak TA, Radford MJ, Wang Y, Krumholz HM: Beta-blocker therapy for secondary prevention of myocardial infarction in elderly diabetic patients: results from the National Cooperative Cardiovascular Project.
J Am Coll Cardiol
34
:
1388
–1394,
1999
41.
American Diabetes Association: Standards of medical care for patients with diabetes mellitus (Position Statement).
Diabetes Care
27 (Suppl. 1)
:
S15
–S35,
2003

S.E.I. has served on an advisory board for Takeda and has received honoraria from Takeda and Glaxo-Smith Kline. F.A.M. has served on an advisory board for Takeda and has received honoraria from Pfizer and AstraZeneca. E.P.H. has received honorarium from Takeda.

A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.