HEART2D TRIAL

Department of Family MedicinePerpetual Succour Hospital

JOURNAL CLUB SERIES

Vivian C. Barrera,MD1st year resident

Answerable question…..

Which strategy is better…..?

Is post prandial hyperglycemia an independent risk

factor for cardiovascular

disease in diabetes?

Effects of Prandial Versus Fasting Glycemia on

Cardiovascular Outcomes in Type 2 Diabetes: The HEART2D

TRIAL

Diabetes Care Vol 32, Num 3, March 2009

Itamar Raz, MD; et al..

Cardiovascular disease is the leading cause of M & M in diabetics

65% attributable to Heart disease and stroke

Chronic hyperglycemia increases risk

Postprandial hyperglycemia associated with CVE

↑0xidative stress

inflammation

Endothelial Dysfunction

Decreased fibrinolysis

Plaque instability

Cardiac events

DECODE: IGT Increases Mortality RiskCollaborative Analysis of Diagnostic Criteria in EuropeN = 25,364 aged >30 years

Cum

ula

tive M

ort

alit

y

Haza

rd (

%)

DECODE Study Group. Lancet 1999;354:617-21.

Follow-up (years)0 2 4 6 8 10

Diagnosed diabetes (n = 1275)Undiagnosed diabetes (n = 3071)Impaired glucose tolerance (n = 2766)*Normal glucose tolerance (n = 18,252)*

*2-hour oral glucose tolerance test (OGTT)IGT = impaired glucose tolerance

Acarbosen=662

Placebon=888

Hazard Ratio(95% CI)

P Value

Myocardial infarction 1 12 0.09 (0.01–0.72) 0.02

Angina 5 12 0.45 (0.16–1.28) 0.13

Revascularization procedures

11 20 0.61 (0.29–1.26) 0.18

Cardiovascular death 1 2 0.55 (0.05–6.11) 0.63

Congestive heart failure 0 2

Cerebrovascular stroke 2 4 0.56 (0.10–3.07) 0.51

Any cardiovascular event 15 32 0.51 (0.28–0.95) 0.03

STOP-NIDDM Trial: Acarbose Prevents Type 2 Diabetes Mellitus and Cardiovascular Events in Subjects With Impaired Glucose Tolerance

Study PopulationStudy Population

FavorsFavorsAcarboseAcarbose

FavorsFavorsPlaceboPlacebo

0 0.5 1.0 1.5Hazard Ratio

Chiasson JL, et al. JAMA. 2003;290;486–494. Copyright © 2003American Medical Association. All rights reserved.

CI = confidence interval

Diabetes Mellitus Insulin Glucose Infusion in Acute Myocardial Infarction (DIGAMI): Benefit of Tight Glycemic Control in No Insulin – Low Risk Cohort

Malmberg K, et al. BMJ. 1997;314:1512-1515.

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0

Mort

alit

y

Mort

alit

y

Total Cohort No Insulin – Low Risk

Years in Study Years in Study

Control

Insulin-glucoseInfusion

0 1 2 3 4 5 0 1 2 3 4 5

Insulin-glucoseInfusion

Control

p = .0111 p = .004

n=133

n=139

n=314

n=306

Primary objective Hyperglycemia and its effects

after acute myocardial infarction on cardiovascular outcomes in patients with type 2 diabetes

Demonstrate the difference

vs

RESEARCH DESIGN AND METHODS Prospective, open-label, randomized, two-

arm parallel, clinical trial Conducted at 105 study centers in 17

countries.. Treatment group assignments were blinded to

the sponsor during the trial October 25, 2002 - July 6, 2005

RESEARCH DESIGN AND METHODS

Patients (aged 30–75 years) with type 2 diabetes , duration of ≥3 months

Entered within 18 days of an AMI Within 21 days of hospital admission for the

recent AMI, randomly assigned into one of two treatment groups

Determination of Sample Size and Statistical Methods

To achieved 80% power, 490 patients must

have experienced one of the primary outcomes to detect a difference between groups

assuming the following difference of 2.5mmol/l between groups in postprandial glucose

To achieved 80% power, 490 patients must

have experienced one of the primary outcomes to detect a difference between groups

assuming the following difference of 2.5mmol/l between groups in postprandial glucose

1355 patients were planned for random assignment with 678 patients in each group

1355 patients were planned for random assignment with 678 patients in each group

All comparisons were performed using 2 tailed tests with a

nominal significance level 0.05

All comparisons were performed using 2 tailed tests with a

nominal significance level 0.05

PRANDIAL STRATEGY

3x daily Insulin Lispro

BASAL STRATEGY

target AIC 7.0 %

If >8.0 %

Add NPH at bedtime

NPH insulin BID or Lantus OD

FBG <6.7mmol/L or < 121mg/dl

Human insulin 30/70 BID

Self monitor PPBG <7.5mmol / <135mg/dl

Follow-up 7 years

Results

1227 enrolled in the study , 1,115 randomly assigned

Prandial n 557 ( 338)

Basal n 558 ( 346)

Prandial 51 deaths, 38 discontinuedBsal 51 deaths, 44 discontinued

Table 1Baseline characteristics of the intent to treat study population by treatment group

Variable PRANDIAL BASAL P value n 557 558

Sex 0.680

Female     201 (36.1) 208 (37.3)

Male     356 (63.9) 350 (62.7)

Age (years)

Mean     61.1 ± 9.7 60.9 ± 9.8 0.724Aged ≥65     202 (36.3) 220 (39.4) 0.277Origin 0.302

Caucasian     484 (86.9) 483 (86.6)

Western Asian     61 (11.0) 58 (10.4)

African descent     1 (0.2) 6 (1.1)

Other     11 (2.0) 11 (2.0)

Country >0.999

Duration of diabetes (years) 9.3 ± 7.2 9.0 ± 7.3 0.518Current tobacco use 93 (16.7) 81 (14.5) 0.316Past tobacco use (years) 13.7 ± 16.5 12.3 ± 15.4 0.143Weight (kg) 81.12 ± 15.17 81.86 ± 15.86 0.513BMI (kg/m2) 29.0 ± 4.6 29.2 ± 5.0 0.380Overweight (BMI ≥25 kg/m2) 449 (80.8) 447 (80.3) 0.832Systolic blood pressure (mmHg) 126.88 ±

16.63127.76 ±

17.750.346

Diastolic blood pressure (mmHg) 76.60 ± 9.06 76.87 ± 9.56 0.542Prior myocardial infarction 99 (17.8) 101 (18.1) 0.858Thrombolysis (recent AMI)* 97 (17.4) 98 (17.6) 0.970Intravenous insulin infusion (recent

AMI)160 (28.8) 160 (28.8) 0.807

A1C (%) 8.42 ± 1.40 8.27 ± 1.52 0.089Triglycerides (mmol/l) 1.89 ± 1.15 1.77 ± 0.95 0.074Total cholesterol (mmol/l) 4.45 ± 1.25 4.45 ± 1.25 0.871HDL cholesterol (mmol/l) 0.96 ± 0.25 0.96 ± 0.23 0.607LDL cholesterol (mmol/l) 2.68 ± 1.02 2.71 ± 1.02 0.556Urinary albumin-to-creatinine ratio

(mg/g)115 ± 430 163 ± 610 0.171

QTc interval (ms) 435 ± 33 434 ± 34 0.428LVEF (%) 50.54 ± 10.05 50.97 ± 10.08 0.829

Variable (n, %) PRANDIAL (n=549)

BASAL(n=550)

Diet and exercise 46 ( 8.4%) 54 ( 9.8%)

Biguanides 41 ( 7.5%) 33 ( 6.0%)

Sulfonylureas 138 (25.1%) 147 (26.7%)

Alpha-glucosidase inhibitors 3 (0.5%) 1 (0.2%)

Fast-acting insulin secretagogues 6 (1.1%) 3 (0.5%)

Metformin and sulfonylureas 83 (15.1%) 86 (15.6%)

Metformin and sulfonylurea, + any other oral 16 (2.9%) 11 (2.0%)

Other combination of oral agents 10 (1.8%) 13 (2.4%)

Basal/premixed insulin (once- or twice-daily) 121 (22.0%) 124 (22.5%)

Combination oral + basal/premixed insulin 41 (7.5%) 40 (7.3%)

Multiple daily insulin injection (MDI) [≥3 inj/day] 38 (6.9%) 27 (4.9%)

Any combination of oral agents and MDI 2 (0.4%) 4 (0.7%)

Other 4 (0.7%) 7 (1.3%)

Table 1—Diabetes therapy at baseline (p=0·740 between groups)

Primary and Secondary Cardiovascular Endpoint

Mean # days : 963 ( 2.7 y) ( 1-1678 days)

Individual Outcomes*PRANDIAL

(n=557)BASAL (n=558)

Estimated Hazard Ratio (95% CI)

p-value

All-cause death 51 (9.2%) 51 (9.1%) 1.00 (0.68, 1.48) 0.982

Cardiovascular death 44 (7.9%) 42 (7.5%) 1.05 (0.69, 1.60) 0.816

Fatal stroke 3 (0.5%) 2 (0.4%) 1.50 (0.25, 8.99) 0.654

Nonfatal stroke 19 (3.4%) 16 (2.9%) 1.21 (0.62, 2.35) 0.573

Any stroke 20 (3.6%) 17 (3.0%) 1.20 (0.63, 2.29) 0.581

Fatal myocardial infarction 12 (2.2%) 12 (2.2%) 1.01 (0.45, 2.25) 0.983

Nonfatal clinical myocardial infarction 53 (9.5%) 50 (9.0%) 1.07 (0.73, 1.58) 0.716

Silent myocardial infarction 2 (0.4%) 4 (0.7%) 0.50 (0.09, 2.75) 0.419

Any myocardial infarction 63 (11.3%)63

(11.3%) 1.01 (0.71, 1.43) 0.948

Coronary revascularization planned after randomization 84 (15.1%)

94 (16.8%) 0.91 (0.68, 1.22) 0.525

Hospitalized acute coronary syndromes 58 (10.4%) 54 (9.7%) 1.09 (0.75, 1.58) 0.647

Congestive heart failure 33 (5.9%) 37 (6.6%) 0.90 (0.56, 1.44) 0.662

Revascularization for PVD 11 (2.0%) 12 (2.2%) 0.93 (0.41, 2.11) 0.863

Amputation for PVD 9 (1.6%) 8 (1.4%) 1.13 (0.44, 2.93) 0.800

Coronary angiography planned after randomization 75 (13.5%)

86 (15.4%) 0.89 (0.66, 1.22) 0.471

Table 2—Summary and analysis of individual outcomes for time to first adjudicated event

FIGURE 2— Glycemic measures. A: Mean ± SD A1C at each visit by treatment strategy (PRANDIAL versus BASAL). B: Seven-point mean self-monitored blood glucose profiles at baseline (dotted line) and throughout the study (postrandomization, solid line) by treatment strategy (PRANDIAL versus BASAL

28%

31%

FBG 8.1 7.0 (p0.005)

Premeal BG 7.7 7.3 (0.233)

TGL 2.21 2.18 (0.894)

TCHOL 4.65 4.65 (0.997)

HDL 1.14 1.11 (0.523)

LDL 2.26 2.70 (0.719)

SYSTOLIC 131.8 132.4 (0.782)

DIASTOLIC 77.4 77.5 (0.978)

HEART RATE 71.4 52.39 (0.817)

LVEF 54.3 52.39 (0.257)

QT INT 423.8 424.1 (0.952)

PRIMARY & SECONDARY CARDIOVASCULAR ENDPOINTSprandial basal P value

Cardivascular drugs 95.0 95.9 0.478

Beta- blocker 83.7 78.9 0.046Gain more weight 4.8 3.1 <.001

Received greater insulin dose .60 0.52 <0.00

Regimen intensification,frequency

28 21 0.05

Incidence of hypoglycemia 55. 55.2 0.367

Severe hypoglycemia 12.9 9.5 0.071

Nocturnal hypoglycemia 65.7 61.5 .007

Cardiac failure 2.3 0.7 .030

Congestive HF 2.2 2.3 >.999

sepsis 5 0 0.038

prandial basal P value

CONCLUSIONS No difference in prandial and basal with

respect to risk for the first combined cardiovascular event

Similar overall glycemic control measures by AIC in both treatment groups

AIC values did not reach the goal of 7.0 %

PROactive Trial: Significant Reduction in Secondary Outcome

Even

ts,

%

0 6 18 24 3612 30Time from randomization (months)

*Excluding silent myocardial infarction (MI)

All-cause mortality, nonfatal MI*, stroke

Dormandy JA et al. Lancet 2005;366:1279–1289.

Pioglitazone301 events

Placebo358 events16% RRR

HR 0.84 (0.72–0.98) P = 0.027

16% RRRHR 0.84 (0.72–0.98)

P = 0.027

ACCORD: Deaths in Intensive vs. Standard Glycemic Control Groups: Preliminary Results

In both glycemic control groups, the mortality rate was lower than that seen in similar populations in other studies.1

In the intensive group, a 10% reduction in primary composite outcome (nonfatal myocardial infarction, stroke, or cardiovascular death) was observed.1

Specific cause of increased mortality with intensive glycemic control is not known, but it is not due to specific therapy or hypoglycemia.1,2

What would be the cardiovascular outcome with 10-year follow-up?

1NHLBI Web site. http://www.nhlbi.nih.gov/health/prof/heart/other/accord/remarks.pdf2NHLBI Web site. http://www.nih.gov/news/health/feb2008/nhlbi-06.htm.

Standard Glycemic Control (Hemoglobin A1c 7.0–7.9%)

Intensive Glycemic Control (Hemoglobin A1c, <6.0%) Difference

Deaths, n 203 257 54

Rate per 1,000 patients/year

11 14 3

Average duration of treatment ~4 years. Intensive glycemic control arm stopped at 18 months by Data Safety Monitoring Board.

ADVANCE: Action in Diabetes and Vascular Disease:Preterax and Diamicron MR Controlled EvaluationPrimary objective:Primary objective: To determine the effect of blood pressure treatment (ACE-inhibitor + diuretic vs. placebo) and glycemic intervention (intensive vs. standard) on microvascular and macrovascular complications in patients with type 2 diabetes mellitus (2 2 factorial design)

Patient population:Patient population: Type 2 diabetes (hypertensive + nonhypertensive) (N = 11,140)

Primary endpoint:Primary endpoint: Macrovascular: composite of nonfatal stroke, nonfatal myocardial infarction, cardiovascular death

Microvascular: composite of new or worsening nephropathyor retinopathy

Secondary endpoints:Secondary endpoints: Cerebrovascular disease, cardiovascular disease, peripheral vascular disease, microalbuminuria, visual deterioration, neuropathy, heart failure, cognitive function and dementia, all-cause mortality

Treatment duration:Treatment duration: 4.5 years

ADVANCE Management Committee. Diabetologia. 2001;44:1118–1120.

Glycemic Control Blood Pressure Intervention

Intensive* (HbA1c ≤6.5%) Perindopril + Indapamide Placebo

Standard¶ Perindopril + Indapamide Placebo

*Gliclazide plus oral agents/insulin to achieve hemoglobin A1c (HbA1c) target¶Based on standard guidelines

THE NULL RESULTS…….. Retarding the progression of advanced

atherosclerosis… very difficult

Risk factors were similar between the groups

The difference in glycemic control was modest Greater separation of post prandial blood glucose may be

needed to adequately test the hypothesis Overall glycemic controll were not fully realized

Lee min ho