Baseline characteristics of patients with diabetes and coronary artery disease enrolled in the Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) trial

Article Outline

• Abstract
• Methods
  • Trial design
  • Baseline data
  • Definitions
  • Statistical analysis
• Results
  • Patient origin
  • Demographic characteristics and physical examination
  • Clinical history
  • Laboratory evaluations
  • Risk factor status
  • Medications
  • Baseline comparisons
• Discussion
  • Patient recruitment
  • BARI 2D population
  • Comparison with established cohorts
• Conclusion
• Appendix A. BARI 2D Study Group
• References
• Copyright

Background

The Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) trial was undertaken to determine whether early revascularization intervention is superior to deferred intervention in the presence of aggressive medical therapy and whether antidiabetes regimens targeting insulin sensitivity are more or less effective than regimens targeting insulin provision in reducing cardiovascular events among patients with type 2 diabetes mellitus and stable coronary artery disease (CAD).

Methods

The BARI 2D trial is a National Institutes of Health–sponsored randomized clinical trial with a 2 × 2 factorial design. Between 2001 and 2005, 49 clinical sites in North America, South America, and Europe randomized 2,368 patients. At baseline, the trial collected data on clinical history, symptoms, and medications along with centralized evaluations of angiograms, electrocardiograms, and blood and urine specimens.

Results

Most of the BARI 2D patients were referred from the cardiac catheterization laboratory (54%) or cardiology clinic (27%). Of the randomized participants, 30% were women, 34% were minorities, 61% had angina, and 67% had multiregion CAD. Moreover, 29% had been treated with insulin, 58% had hemoglobin A_1c >7.0%, 41% had low-density lipoprotein cholesterol ≥100 mg/dL, 52% had blood pressure >130/80 mm Hg, and 56% had body mass index ≥30 kg/m².

Conclusions

Baseline characteristics in BARI 2D are well balanced between the randomized treatment groups, and the clinical profile of the study cohort is representative of the target population. As a result, the BARI 2D clinical trial is in an excellent position to evaluate alternative treatment approaches for diabetes and CAD.

The Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) trial is a randomized clinical trial initiated to evaluate treatment strategies for diabetes and coronary artery disease (CAD) in patients with both of these conditions. In 1996, post hoc analyses of the original BARI randomized trial demonstrated that multivessel coronary disease patients with type 2 diabetes mellitus had lower survival relative to those without diabetes regardless of the form of revascularization and that coronary artery bypass graft (CABG) was associated with better survival than percutaneous transluminal coronary angioplasty without stents in patients with diabetes. ¹ Because most BARI patients had unstable angina and all required revascularization, the optimal treatment approach was unknown for patients with stable coronary disease whose symptoms did not require urgent revascularization. The BARI 2D trial was designed to answer this question by comparing an initial strategy of coronary revascularization coupled with intensive medical therapy versus an initial strategy of intensive medical therapy alone. At the time the trial was designed, data were emerging that suggested that metformin and thiazolidinedione (TZD) drugs might have beneficial effects on cardiovascular disease independent of their blood glucose–lowering actions. ², ³ Thus, the second aim of BARI 2D was to evaluate whether managing glycemia with insulin-sensitizing medications could improve cardiovascular outcomes compared with insulin-providing medications (insulin or insulin secretagogues).

Patients with type 2 diabetes and established coronary disease are at high risk of cardiovascular events. ⁴ Yet, few studies include sizable cohorts of patients with diabetes presenting at a clinical stage when revascularization is acceptable but not mandatory according to evidence-based American Heart Association/American College of Cardiology class 1 indications. ⁵, ⁶ The BARI 2D trial enrolled 2,368 such patients. Baseline data provide an opportunity to determine how the trial participants relate to broader patient populations with diabetes, coronary disease, or both and to observe associations among risk factors in this population.

Back to Article Outline

Methods 

Trial design 

The BARI 2D trial was designed to compare treatment strategies for diabetes and established CAD in the setting of standardized glycemic control and intensive management of dyslipidemia, hypertension, smoking, and obesity. The trial protocol and rationale have been described in detail. ⁷, ⁸, ⁹, ¹⁰ Using a 2 × 2 factorial design, patients were assigned at random to a diabetes treatment and to a cardiovascular treatment. The diabetes component compares an insulin-sensitizing strategy of glycemic control versus an insulin-providing strategy. The cardiovascular component compares a strategy of intensive medical therapy and initial coronary revascularization versus a strategy of intensive medical therapy alone, with revascularization deferred until mandated by symptoms according to clinical guidelines. Eligible patients underwent angiography to document CAD and to determine suitability for elective revascularization. Randomization was stratified by physician-declared intended revascularization strategy (percutaneous coronary intervention [PCI] or CABG) and by clinical site.

Patients were eligible for BARI 2D if they were at least 25 years old with a diagnosis of type 2 diabetes mellitus, documented ischemia, and angiographically documented CAD with at least 1 significant lesion (≥50% stenosis) suitable for elective revascularization. Those with classic exertional angina but without a preprocedure stress test were eligible if they had a lesion with ≥70% stenosis. Patients were not eligible if they required immediate coronary revascularization or if they had undergone revascularization within 12 months before study entry. Other exclusion criteria include New York Heart Association functional class III or IV congestive heart failure, need for concurrent major vascular surgery, stenosis ≥50% of the left main coronary artery, hemoglobin A_1c (HbA_1c) >13%, serum creatinine >2.0 mg/dL, and hepatic disease. The protocol was approved by Institutional Review Boards at all participating institutions, and patients signed written informed consent.

A total of 2,368 patients were enrolled at 49 clinical centers throughout North America, South America, and Europe between January 1, 2001, and March 31, 2005. One additional site was withdrawn from BARI 2D, and data from this site (n = 14 patients, no deaths reported at the time of site termination) were excluded from the analyses. Under the protocol, risk factors are managed to aim for target levels of HbA_1c <7.0%, blood pressure ≤130/80 mm Hg, and low-density lipoprotein (LDL) cholesterol <100 mg/dL. The BARI 2D patient treatment phase ends November 30, 2008. The primary end point is all-cause mortality; and the principal secondary end point is the composite of death, myocardial infarction, and stroke. Patient safety and treatment efficacy are assessed by an independent Data and Safety Monitoring Board.

The BARI 2D trial is coordinated at the University of Pittsburgh (Pittsburgh, PA). The trial is funded as a cooperative agreement by the National Heart, Lung, and Blood Institute (NHLBI) and receives additional funding from the National Institute of Diabetes and Digestive and Kidney Diseases. Several pharmaceutical companies provide supplemental funding and/or donate medications or supplies. The corporate sponsors played no part in the design, conduct, or analysis of the study.

Baseline data 

The BARI 2D trial collected demographic, clinical history, clinical measurements, and medication data. Core laboratories provided centralized evaluations of angiographic anatomy, electrocardiograms (ECGs), HbA_1c, serum lipid levels, urine albumin and creatinine, and fibrinolytic factors. Missing core HbA_1c and lipid values were imputed based on local site laboratory values.

Definitions 

A significant lesion is defined as a luminal narrowing ≥50% in an anatomically relevant coronary artery segment of size ≥1.5 mm. The number of diseased myocardial regions is the number of territories (anterior, lateral, or posterolateral) with at least 1 significant lesion. Myocardial jeopardy index is the ratio of jeopardized and anatomically relevant segments downstream of significant lesions relative to all viable left ventricular (LV) segments, reflecting the extent of anatomical CAD. ¹¹

The Michigan Neuropathy Screening Instrument (MNSI) clinical score is based on physical examination of ankle reflexes, toe vibration sensation, and designated abnormalities; values >2 provide clinical evidence of neuropathy. The MNSI screening score is a count of patient-identified sensory symptoms; values ≥7 are consistent with neuropathy. ¹²

Statistical analysis 

Descriptive statistics include means ± SDs and proportions; medians are presented for highly skewed data. Variables were compared between the 2 randomized cardiac treatment groups (initial revascularization, n = 1,176 versus medical therapy, n = 1,192) and 2 randomized diabetes treatment groups (insulin providing, n = 1,185 versus insulin sensitizing, n = 1,183). Selected variables were compared among groups defined by age, gender, race/ethnicity, and region. t tests, Wilcoxon nonparametric tests, and χ² tests were used; and P values < .01 were considered statistically significant because of the multiple comparisons in this article.

Back to Article Outline

Results 

Patient origin 

A total of 4,623 patients consented to be screened for the BARI 2D trial. Most were identified in the cardiac catheterization laboratory or cardiology clinic ( Figure 1 ) and generally consented before the qualifying angiogram. Approximately half of screened patients were eligible for enrollment (n = 2,436), and 97% consented to randomization. The reason for ineligibility was not collected before July 15, 2002. Among 1,545 ineligible patients with reason for exclusion available, 51% were excluded for insufficient CAD, 10% for CAD that was inappropriate for revascularization, and 28% for severe CAD where revascularization was considered necessary.

• 
  • View Large Image
  • Download to PowerPoint

• Figure 1. 

  Flowchart of the enrollment process in the BARI 2D clinical trial.

Demographic characteristics and physical examination 

The typical patient in BARI 2D was white, male, and 62 years old; however, 30% of patients were women and 34% were minorities ( Figure 2 ). Most participants (63%) were enrolled from the United States; but substantial numbers were enrolled from Canada (15%) and Brazil (15%), with the remaining 7% from Mexico, the Czech Republic, and Austria. The mean body mass index (BMI) was 31.7 ± 5.9 kg/m²; and the mean systolic and diastolic blood pressures were 131.7 ± 20.0 and 74.5 ± 11.2 mm Hg, respectively, ( Table I ).

• 
  • View Large Image
  • Download to PowerPoint

• Figure 2. 

  The BARI 2D population categorized by race/ethnicity overall and stratified by region.

Table I. Baseline characteristics of BARI 2D population

Demographic and Physical Characteristics	(N = 2368)	Cardiac Status	(N = 2368)	Diabetes Status and Lipids	(N = 2368)
Age, mean, SD	62.4, 8.9	Hypertension, %	82.5	Duration of DM categories, %
Male, %	70.4	Hypercholesterolemia, %	81.9	<5 yrs	33.3
Race, %		Cigarette smoking status, %		5 - <10 yrs	23.5
White	70.4	Current	12.5	10 - <20 yrs	29.2
Black	17.0	Former	54.4	≥ 20 yrs	14.1
Asian	4.2	Never	33.1	History of insulin use, %	29.3
Indian/Native American	4.3	Myocardial infarction, %	32.0	Glycemia measurements, %
Other	4.1	Congestive heart failure, %	6.6	HbA1c ≤ 7.0%	41.7
Hispanic ethnicity, %	12.5	Non-coronary artery disease, %	23.7	7.0% < HbA1c ≤ 8.0%	25.3
Region of World, %		Stroke or TIA, %	9.8	HbA1c >8.0%	33.0
USA	63.3	Any prior revascularization, %	23.6	Albuminuria
Canada	14.9	Prior PCI, %	19.6	Microalbuminuria (30 < ACR ≤ 300)	22.9
Brazil	15.0	Prior stent, %	13.4	Macroalbuminuria (ACR > 300)	9.7
Mexico	3.6	Prior CABG, %	6.4	History of hypoglycemic episode, %	22.8
Czech Republic/Austria	3.2	Angina status, %
BMI categories (kg/m2), %		None	17.9	Laser therapy for diabetic retinopathy or macular edema, %	3.1
Normal or underweight, <25	9.7	Angina equivalents only	21.4	Lower extremity amputation, %	0.8
Overweight, 25 to <30	34.0	Stable CCS 1	14.3	MNSI screening neuropathy score ≥7, %	15.9
Class 1 obese, 30 to <35	32.1	Stable CCS 2	28.8	MNSI clinical neuropathy score >2, %	50.3
Class 2 obese, 35 to <40	15.3	Stable CCS 3	7.5	Total cholesterol ≥200 mg/dL, %	19.0
Class 3 obese, ≥40	9.0	Stable CCS 4	1.2	Triglycerides ≥200 mg/dL, %	31.0
Blood pressure >130/80 mmHg,%	52.4	Unstable angina	9.5	HDL <40 male <50 female mg/dL, %	72.4
Ankle brachial index ≤0.9	20.1			LDL ≥100 mg/dL, %	40.5

TIA, Transient ischemic attack; ACR; albumin to creatinine ratio; HDL, high-density lipoprotein.

Clinical history 

The frequency of cardiac risk factor combinations in BARI 2D are shown in Figure 3 . By design, all participants had type 2 diabetes. More than 80% had a history of hypertension; and 45% had the triad of smoking, hypertension, and hyperlipidemia. Almost a third had prior myocardial infarction, but few had congestive heart failure (7%). In addition, 24% had coronary revascularization before randomization, 61% had classic angina, and 24% had evidence of atherosclerosis in vascular beds beyond coronary disease. The average duration of type 2 diabetes mellitus was 10.4 ± 8.7 years. Clinical manifestations and complications of diabetes were relatively common: 23% of patients reported hypoglycemic episodes, and 50% had clinical evidence of neuropathy.

• 
  • View Large Image
  • Download to PowerPoint

• Figure 3. 

  Observed combinations of cardiac risk factors and clinical history in the BARI 2D population.

Laboratory evaluations 

At baseline, the mean HbA_1c level was 7.7% ± 1.6%; and 33% of patients had microalbuminuria or macroalbuminuria. The median total cholesterol level was 164 mg/dL; median triglycerides, 148 mg/dL; median high-density lipoprotein, 37 mg/dL; and median LDL cholesterol, 92 mg/dL. Notably, 21% of patients had LDL <70 mg/dL. The median plasminogen activator inhibitor 1 concentration and activity levels were 23.0 ng/mL and 16.0 AU/mL, respectively; and the median tissue plasminogen activator antigen was 9.7 ng/mL. ¹³ Fifty-five percent of BARI 2D patients had abnormal ECG findings ( Table II ), and 18% had LV ejection fraction (EF) <50%. Based on angiographic core laboratory data, two thirds of participants had CAD involving multiple myocardial regions.

Table II. Baseline characteristics of BARI 2D population

ECG and angiographic characteristics	(N = 2368)	Current medications	(N = 2368)
Abnormal Q wave, %	18.7	Biguanide (metformin), %	54.1
Major Q wave, %	8.1	Thiazolodinedione (TZD), %	18.9
ST depression >0.5 mm, %	15.6	Sulfonylurea, %	53.3
T wave inversion >1 mm, %	19.5	Insulin, %	27.9
Left bundle branch block, %	2.5	Not taking any diabetes drugs, %	8.7
Abnormal LV function (LVEF < 50%) assessed by clinical site, %	17.5	HMG-CoA reductase (statin),%	74.9
		Fibrate, %	8.6
Number of lesions, mean, SD	4.7, 2.3	Niacin, %	2.2
Number of significant lesions, mean, SD	2.7, 1.8	Omega-3 fatty acids, %	2.3
Diseased myocardial regions, %		Beta blocker, %	73.0
Only non-significant lesions	3.6	Calcium-channel blockers, %	31.4
1	29.8	Long-acting nitrate, %	31.4
2	35.9	ACE inhibitor, %	64.6
3	30.7	Angiotensin receptor blocker, %	14.4
Myocardial jeopardy index, mean, SD	44.5, 24.2	Aspirin, %	88.0
Proximal LAD stenosis ≥50%, %	13.2	Anti-platelet drug, %	19.5
Total occlusion(s)	41.0	Diuretic, %	38.6

HMG-CoA, Hydroxymethylglutaryl–coenzyme A; ACE, angiotensin-converting enzyme.

After coronary angiography and before randomization, 1,605 patients (68%) had PCI designated as the intended revascularization (half of whom were then assigned to initial revascularization and half to initial medical therapy); and 763 (32%) had CABG designated. The intended mode of revascularization correlated to the extent and severity of CAD. In the PCI-intended stratum, the mean number of significant lesions was 2.3 and 10% of the patients had proximal left anterior descending coronary artery (LAD) disease, whereas in the CABG-intended stratum, the mean number of significant lesions was 3.5 and 19% of patients had proximal LAD disease. The myocardial jeopardy index was significantly higher in the CABG-intended stratum ( Figure 4 ).

• 
  • View Large Image
  • Download to PowerPoint

• Figure 4. 

  Myocardial jeopardy index score distribution among patients in the PCI stratum (light bars) and CABG stratum (dark bars).

Risk factor status 

A comparison of risk factor control in BARI 2D and the latest National Health and Nutrition Examination Surveys (NHANES) cohort of patients with diabetes is shown in Figure 5 . ¹⁴ At baseline, 12% of BARI 2D patients were at the prespecified protocol targets for the 4 major risk factors (HbA_1c, LDL, blood pressure, and smoking).

• 
  • View Large Image
  • Download to PowerPoint

• Figure 5. 

  Frequency of total cholesterol ≥200 mg/dL, blood pressure ≥140/90 mm Hg, and current smoking in BARI 2D at baseline and in the NHANES 1999-2000 cohort. (Data from Imperatore et al. ¹⁴ )

Medications 

The BARI 2D participants were taking an average of 1.6 diabetes drugs; 2.2 drugs for angina, hypertension, or heart failure; and 0.9 lipid drugs ( Table II ). Of note, 19% of patients were receiving a TZD; 28%, insulin; and 75%, a statin. However, 9% were not taking any diabetes drugs; and 15% were not taking any antianginal drugs (β-blockers, calcium blockers, and long-acting nitrates). At baseline, 37% of patients were receiving aspirin, angiotensin-converting enzyme inhibitor, statin, and β-blockers, the standard American Heart Association/American College of Cardiology evidence-based recommended medications for diabetes and CAD.

Baseline comparisons 

Comparing all baseline characteristics displayed in Table I, Table II among the randomized treatment groups indicated that the initial revascularization group had worse angina symptoms than the medical therapy group (stable Canadian Cardiovascular Society class 3 or 4 or unstable angina: 22% vs 15%, P = .002). No other imbalances were detected at the P < .01 level among the randomized treatment groups.

In contrast, characteristics varied significantly among subgroups defined by age, gender, race/ethnicity, and region ( Table III ). Younger patients had higher mean HbA_1c and lipid levels. Women had higher mean HbA_1c, systolic blood pressure, and LDL values but fewer diseased myocardial regions than men. Relative to other groups, black patients had significantly higher mean blood pressure and LDL cholesterol. The proportion of women and black participants receiving insulin was higher and that of metformin was lower compared with their counterparts. Patients from the United States had larger body mass, more frequently had abnormal LV function and prior revascularization, and more often received insulin and TZD than non-US patients. Those outside the United States and Canada had worse risk factor control and more extensive myocardial disease.

Table III. Baseline characteristics of BARI 2D by subgroups

Characteristic	Age			Gender			Race/ethnicity					Region
	Age<65 y (n = 1439)	Age ≥65 y (n = 929)	P	Male (n = 1666)	Female (n = 702)	P	White nH (n = 1560)	Black nH (n = 398)	Hispanic (n = 297)	Other (n = 113)	P	United States (n = 1499)	Canada (n = 353)	Other (n = 516)	P
Age, mean, SD	–	–		62.2, 8.7	62.9,9.3	NS	63.2, 8.7	61.1, 9.6	60.6, 8.7	60.4, 9.2	⁎	62.9, 9.1	61.9, 8.5	61.3, 8.5	†
Female, %	28.4	31.6	NS	–	–	–	25.1	48.7	30.0	23.9	⁎	31.8	15.0	33.5	⁎
BMI, mean, SD	32.4, 6.4	30.7, 5.0	⁎	31.3, 5.5	32.8, 6.8	⁎	32.0, 5.7	32.6, 6.9	30.3, 5.3	28.7, 5.6	⁎	32.8, 6.2	31.1, 5.3	29.1, 4.6	⁎
HbA1c, mean, SD	7.9, 1.7	7.3, 1.4	⁎	7.5, 1.6	8.0, 1.7	⁎	7.5, 1.5	8.0, 1.7	8.0, 1.8	7.9, 1.5	⁎	7.6, 1.6	7.5, 1.4	8.0, 1.8	⁎
Systolic blood pressure, mean, SD	129.9, 19.8	134.5, 20.0	⁎	130.4, 18.7	134.8, 22.6	⁎	130.9, 19.3	137.7, 22.2	128.4, 20.2	130.8, 17.6	⁎	130.7, 19.0	131.0, 17.8	135.2, 23.7	⁎
Diastolic blood pressure, mean, SD	76.1, 11.1	72.0, 10.9	⁎	74.8, 10.6	73.8, 12.5	NS	73.7, 11.1	78.0, 12.2	74.5, 10.1	74.0, 9.8	⁎	72.4, 10.4	74.2, 10.3	80.7, 11.8	⁎
LDL cholesterol, mean, SD	98.4, 34.6	93.0, 30.4	⁎	93.3, 32.0	103.2, 34.5	⁎	94.1, 32.0	105.8, 37.4	96.0, 30.0	93.5, 33.4	⁎	94.5, 33.2	92.0, 31.6	104.2, 32.7	⁎
Triglycerides, mean, SD	193.2, 151.2	162.8, 111.7	⁎	183.7, 142.6	175.6, 125.9	NS	192.8, 146.3	133.6, 91.7	187.7, 115.5	173.0, 170.0	⁎	176.0, 136.9	178.1, 144.7	198.9, 134.7	†
Insulin, %	28.9	26.2	NS	24.3	36.4	⁎	25.3	38.8	26.6	28.3	⁎	33.2	17.6	19.4	⁎
Sulfonylurea, %	52.1	55.2	NS	55.8	47.3	⁎	53.1	49.9	58.2	55.4	NS	51.2	55.8	57.6	‡
TZD, %	19.6	17.7	NS	19.8	16.6	NS	18.9	17.1	20.2	21.4	NS	25.6	16.5	1.0	⁎
Metformin, %	56.4	50.5	†	55.7	50.1	‡	55.4	46.9	53.5	62.5	†	51.6	69.8	50.4	⁎
No. of hypertension drugs, mean, SD	2.2, 1.0	2.3, 1.0	‡	2.2, 1.0	2.4, 1.0	⁎	2.2, 1.0	2.5, 1.0	1.9, 1.0	2.3, 1.0	⁎	2.3, 1.0	2.2, 1.0	1.9, 1.0	⁎
Statin, %	74.5	75.6	NS	75.8	72.9	NS	75.2	77.8	70.3	72.6	NS	77.4	77.7	65.8	⁎
Prior CABG, %	6.2	6.8	NS	7.3	4.4	†	6.6	5.3	7.7	4.4	NS	8.9	3.4	1.2	⁎
Prior PCI, %	19.1	20.5	NS	19.4	20.2	NS	20.6	19.1	16.5	15.9	NS	23.2	12.2	14.3	⁎
Diseased myocardial regions, %			NS			⁎					NS				⁎
1 or none	34.8	31.3		30.8	39.5		32.2	38.0	36.0	28.4		38.9	26.9	22.1
2	35.6	36.4		35.6	36.5		35.9	34.2	36.4	40.7		34.3	37.4	39.3
3	29.6	32.4		33.5	24.0		32.0	27.9	27.6	31.0		26.8	35.7	38.6
LVEF <50%, %	17.8	17.0	NS	19.7	12.0	⁎	15.7	22.9	20.0	15.7	†	20.8	19.6	6.6	⁎

nH, Non-Hispanic; NS, not significant (P ≥ .05).

Back to Article Outline

Discussion 

The patients enrolled in the BARI 2D clinical trial represent a large segment of the population of patients with diabetes and coronary disease. Evidence-based guidelines for the optimal treatment of both of these conditions are lacking. The BARI 2D trial has the potential to address the impact of alternative diabetes and CAD treatments on cardiovascular outcomes, the major causes of mortality in this population.

Patient recruitment 

The BARI 2D trial required a close collaboration between cardiologists and endocrinologists to formulate the hypotheses and to determine the characteristics of patients for whom the optimal treatments had not yet been resolved. Eligibility criteria were defined to ensure that study participants were similar to those in the community to facilitate the generalization of trial results and to provide data for future updates of clinical guidelines to practitioners. Several requirements made the implementation of the glycemic trial protocol more difficult. For example, the exclusion criterion of serum creatinine >2.0 mg/dL (rather than >1.5 mg/dL) meant that metformin could not be used in all insulin-sensitizing subjects. Similarly, enrolling patients who already received insulin broadened the trial population but added complexity to the glycemic treatment protocol.

BARI 2D population 

There were few statistically significant and no clinically important baseline differences between the randomized treatment groups. A higher proportion of patients with relatively severe angina symptoms was assigned to initial revascularization. Because the 2 randomization groups have substantial overlap in symptom status, this factor can be addressed with subsequent analyses.

The BARI 2D trial will evaluate early revascularization intervention versus revascularization deferred until mandated by symptoms. It is not a trial comparing CABG and PCI. The revascularization strategy was physician determined and based upon the patients' coronary anatomy rather than random assignment. By design, those with the most extensive coronary disease were more frequently physician selected for the “CABG-intended” stratum. With the stratified randomization, BARI 2D can compare PCI versus medical therapy in a group of PCI-intended patients, and CABG versus medical therapy among CABG-intended patients. Randomized comparisons of PCI versus CABG for patients with diabetes will be performed in other trials such as NHLBI FREEDOM, ¹⁵ VA CARD, ¹⁶ and CARDia. ¹⁷

The BARI 2D trial successfully enrolled a patient group appropriate for the design of the trial. Glycemic and cardiovascular risk factor control was consistent with current guidelines for some participants but inadequate for many others. Establishing a partnership between cardiologists and endocrinologists was an important aspect of BARI 2D. When the final results of BARI 2D are available, the trial will demonstrate the feasibility and potential utility of a combined approach to management of diabetes and CAD.

Comparison with established cohorts 

In the most recent NHANES cohort with diabetes, ¹⁴, ¹⁸ 48% of the participants were female, 53% were <60 years old, 76% were white non-Hispanic, 18% were black non-Hispanic, 7% were Hispanic, and average BMI was 33.0 kg/m². The BARI 2D participants were similar to this cohort regarding race/ethnicity, duration of diabetes, and body mass but were slightly older and more likely to be male. It is somewhat surprising that only 30% of the BARI 2D population is female given that, after menopause, women and men with diabetes have comparable rates of cardiovascular disease. ¹⁹ A positive feature of BARI 2D is the inclusion of 17% blacks and 13% Hispanics because these groups disproportionately have type 2 diabetes. Although hypertension was equally common in BARI 2D and NHANES, high cholesterol was less prevalent in BARI 2D possibly because of the pervasive use of statins among trial participants.

The BARI 2D cohort is also comparable to populations enrolled in clinical trials with related study aims. ²⁰, ²¹, ²² The PROactive ²⁰ randomized pioglitazone versus placebo among patients with diabetes and a history of macrovascular disease. By eligibility requirements, 100% of BARI 2D participants had angiographically documented CAD compared with 48% of PROactive patients; however, a larger proportion of PROactive patients had a history of cardiovascular events including myocardial infarction (47%), stroke (19%), and prior revascularization (31%). The 2 trials are comparable with respect to age, sex, duration of diabetes, and insulin use; but BARI 2D enrolled a greater percentage of minorities (34% vs 1.5%). The COURAGE trial, ²¹ conducted by the Department of Veterans Affairs, randomized intensive medical therapy with PCI versus without PCI; all COURAGE patients had coronary disease, but only 32% had diabetes. As expected, BARI 2D enrolled a larger proportion of women than COURAGE; and BARI 2D patients were less likely to be current smokers or to have classic angina (61% vs 88%). The ADVANCE trial is a factorial-designed trial evaluating blood pressure therapy and intensive glucose control ²² among diabetes patients from 20 countries. Compared with ADVANCE, BARI 2D patients were younger and had higher BMI. Although BARI 2D participants more frequently had history of hypertension (82% vs 69%), they had lower blood pressure (132/75 vs 145/81 mm Hg). Insulin use was more common in BARI 2D (28% vs 1%), whereas metformin and sulfolnylurea use was more common in ADVANCE; HbA_1c was similar.

The NHLBI is concurrently sponsoring the ACCORD. ²³ This large randomized trial tests whether intensive treatment aimed at lowering HbA_1c to <6.0% reduces cardiovascular risk compared with standard treatment aimed at maintaining HbA_1c in the range of 7.0% to 7.9%. The BARI 2D and ACCORD are complementary trials, the former determining the optimum glucose-lowering strategy and the latter the optimal glycemic target for minimizing cardiovascular events and mortality in diabetes. The randomized cohorts are comparable in age, duration of diabetes, minority representation, BMI, HbA_1c, and proportions treated with insulin before entry. Finally, both trials have similar 5-year follow-up, ideal for the long-term evaluation of treatment strategies.

Back to Article Outline

Conclusion 

The baseline data presented in this article confirm that the BARI 2D clinical trial participants comprise an appropriate patient population for assessing the risks and benefits of the selected interventions. With systematic monitoring of glycemic control, blood pressure, and lipids to ensure proper control of cardiovascular risk factors, the BARI 2D study is in an excellent position to evaluate alternative modes of treatment of CAD and to determine whether the mode of diabetes treatment influences cardiovascular outcomes in patients with type 2 diabetes and cardiovascular disease.

Back to Article Outline

Appendix A. BARI 2D Study Group 

Back to Article Outline

References 

1. The BARI Investigators  . Comparison of coronary bypass surgery with angioplasty in patients with multivessel disease. N Engl J Med. 1996;335:217–225
   • View In Article
   • MEDLINE
   • CrossRef
2. 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. 1998;352:854–865
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (582 KB)
   • CrossRef
3. Chiquette E , Ramirez G , DeFronzo R . A meta-analysis comparing the effect of thiazolidinediones on cardiovascular risk factors. Arch Intern Med. 2004;164:2097–2104
   • View In Article
   • MEDLINE
   • CrossRef
4. Gu K , Cowie CC , Harris MI . Diabetes and decline in heart disease mortality in US adults. JAMA. 1999;281:1291–1297
   • View In Article
   • MEDLINE
   • CrossRef
5. Eagle KA , Guyton RA , Davidoff R, et al.   American College of Cardiology; American Heart Association. ACC/AHA 2004 guideline update for coronary artery bypass graft surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1999 Guidelines for Coronary Artery Bypass Graft Surgery). Circulation. 2004;110:e340–e437
   • View In Article
6. Smith SC , Feldman TE , Hirshfeld JW, et al.   ACC/AHA/SCAI 2005 guideline update for percutaneous coronary intervention: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/SCAI Writing Committee to Update the 2001 Guidelines for Percutaneous Coronary Intervention). Circulation. 2006;113:156–175
   • View In Article
   • CrossRef
7. Brooks MM , Frye RL , Genuth S, et al.   Hypotheses, design and methods for the Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) trial. Am J Cardiol. 2006;97(Suppl):9G–19G
   • View In Article
   • MEDLINE
8. Magee MF , Isley WL . Rationale, design and methods for glycemic control in the Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) trial. Am J Cardiol. 2006;97(Suppl):20G–30G
   • View In Article
   • MEDLINE
9. Barsness GW , Gersh BJ , Brooks MM, et al.   Rationale for the revascularization arm of the Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) trial. Am J Cardiol. 2006;97(Suppl):31G–40G
   • View In Article
   • MEDLINE
10. Albu J , Gottlieb SH , August P, et al.   Modifications of coronary risk factors. Am J Cardiol. 2006;97(Suppl):41G–52G
    • View In Article
    • MEDLINE
11. Alderman EL , Stadius M . The angiographic definitions of the Bypass Angioplasty Revascularization Investigation. Coronary Artery Disease. 1992;2:1189–1207
    • View In Article
12. Feldman EL , Stevens MJ , Thomas PK, et al.   A practical two-step quantitative clinical and electrophysiological assessment for the diagnosis and staging of diabetic neuropathy. Diabetes Care. 1994;11:1281–1289
    • View In Article
13. McGill JB , Schneider DJ , Arfken CL, et al.   Factors responsible for impaired fibrinolysis in obese patients. Diabetes. 1994;43:104–109
    • View In Article
    • MEDLINE
14. Imperatore G , Cadwell BL , Geiss L, et al.   Thirty-year trends in cardiovascular risk factor levels among US adults with diabetes: National Health and Nutrition Examination Surveys, 1971-2000. Am J Epidemiol. 2004;160:531–539
    • View In Article
    • MEDLINE
    • CrossRef
15. Clinical trials.gov. Comparison of two treatments for multivessel coronary artery disease in individuals with diabetes (FREEDOM). Available at http://clinicaltrials.gov/ct/show/NCT00086450 Last accessed October 28, 2007
    • View In Article
16. Clinical trials.gov. Coronary Artery Revascularization in Diabetes (VACARD). Available at http://clinicaltrials.gov/show/NCT00326196 Last accessed October 28, 2007
    • View In Article
17. Kapur A , Malik IS , Bagger JP, et al.   The Coronary Artery Revascularisation in Diabetes (CARDia) trial: background, aims, and design. Am Heart J. 2005;149:13–19
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (131 KB)
    • CrossRef
18. Gregg EW , Qiuping G , Cheng YJ, et al.   Mortality trends in men and women with diabetes, 1971 to 2000. Ann Intern Med. 2007;147:149–155
    • View In Article
19. Kanaya A , Grady D , Barrett-Connor E . Explaining the sex difference in coronary heart disease mortality among patients with type 2 diabetes mellitus: a meta-analysis. Arch Intern Med. 2002;162:1737–1745
    • View In Article
    • MEDLINE
    • CrossRef
20. Charbonnel B , Dormandy J , Erdman E, et al.   The Prospective Pioglitazone Clinical Trial in Macrovascular Events (PROactive): can pioglitazone reduce cardiovascular events in diabetes? Study design and baseline characteristics of 5,238 patients. Diabetes Care. 2004;27:1647–1653
    • View In Article
    • MEDLINE
    • CrossRef
21. Boden WE , O'Rourke RA , Teo KK, et al.   Optimal medical therapy with or without PCI for stable coronary disease. N Engl J Med. 2007;356:1503–1516
    • View In Article
    • CrossRef
22. ADVANCE Collaborative Group . ADVANCE—Action in Diabetes and Vascular Disease: patient recruitment and characteristics of the study population at baseline. Diabet Med. 2005;22:882–888
    • View In Article
    • MEDLINE
    • CrossRef
23. Clinical trials.gov. Action to Control Cardiovascular Risk in Diabetes (ACCORD). Available at http://clinicaltrials.gov/ct/show/NCT00000620 Last accessed March 31, 2007
    • View In Article