The prognostic role of metabolic syndrome after myocardial infarction

Diabetes mellitus can be viewed as the last step in a prolonged process, during which impaired glucose metabolism with insulin resistance (ie, progressive resistance to the action of insulin) has been present for a long time despite normal fasting glucose levels, placing individuals at high risk for cardiovascular events.^1-3

The complexity of the diagnosis of insulin resistance in clinical practice favors the use of surrogates like metabolic syndrome (METS), whose diagnosis is based on simple clinical criteria.⁴ Metabolic syndrome, which is differentiated by a group of insulin-resistance—related risk factors, is believed to be an early sign of impaired glucose metabolism and a prognostic indicator of vascular risk in patients with no overt coronary artery disease.

In patients with previous myocardial infarction (MI), we evaluated the database from the Gruppo Italiano por lo Studio della Sopravvivenza nell’Infarto miocardico (GISSI-Prevenzione) study to determine the prevalence and prognostic role of metabolic syndrome and diabetes in cardiovascular events following MI.⁵

Patients and methods

The GISSI-Prevenzione study was an open-label clinical trial performed at multiple institutions and designed to evaluate the cardiovascular benefits of 850 mg of polyunsaturated fatty acids and 300 mg of vitamin E compared with no treatment in 11,323 patients with prior MI. Assessment of events was performed by blinded evaluators, and patients were followed up for a period of 3.5 years.

For this analysis, all-cause mortality, cumulative rates of cardiovascular events (defined as the occurrence of cardiovascular death, nonfatal MI, or nonfatal stroke), development of congestive heart failure (CHF; defined as hospitalization for CHF during follow-up in patients with no heart failure at baseline), or diabetes development (defined as the appearance of a fasting glucose level >= 126 mg/dL or the beginning of antidiabetic treatment in patients who were nondiabetic at baseline) were the outcome measures. Patients with missing data for METS diagnosis at baseline were ex­cluded from the study. A total of 10,384 patients were analyzed for overall mortality and cardiovascular events, 8417 patients were analyzed for CHF development, and 7468 pa­tients were analyzed for diabetes development.

We also evaluated the effects of change in weight between baseline and the first follow-up visit on the risk of cardiovascular events and diabetes development in 7027 patients (those who, at the first follow-up visit, had weight measurements and did not have diabetes or cardiovascular events). The same analysis was repeated among patients with a body mass index (BMI) of >= 25 kg/m² (n = 4422). The group of patients with a decrease in weight of less than 5% was considered the reference class. A decrease in body weight of more than 10% was considered substantial, and a decrease of 6% to 10% was considered moderate. A weight increase of more than 10% was considered substantial, an inncrease of 6% to 10% was considered moderate, and an increase of 0% to 5% was considered slight.

Nondiabetic patients with 3 of 5 National Cholesterol Education Pro­gram Adult Treatment Panel III criteria were diagnosed with METS, modified by using a BMI of >= 26 kg/m² above instead of waist circumference values. If patients were reported to have diabetes by a physician, were taking anti­diabetic medication, or had a glucose level of >= 126 mg/dL, they were diagnosed as having diabetes.

Final predictive model adjusted for potential confounders of total mortality, cardiovascular events (CVE), hospitalization for CHF, and development of diabetes in normal patients, patients with METS, and diabetic patients. (Reprinted with permission from J Am Coll Cardiol. 2005;46(2):277-283.)

Table.

Total mortality

Cardiovascular Events

Hospitalization for CHF

Development of diabetes

Normal patients

5198

5198

4306

4737

Events, n (%)

384 (7.4)

456 (8.8)

71 (1.6)

454 (9.6)

RR (95% CI)

1.00

1.00

1.00

1.00

METS patients

3047

3047

2522

2731

Events, n (%)

258 (8.5)

316 (10.4)

50 (2.0)

486 (17.8)

RR (95% CI)

1.29 (1.10-1.51)

1.23 (1.06-1.42)

1.24 (0.86-1.79)

1.93 (1.69-2.19)

= .002

P

= .005

P

< .001

P

Diabetic patients

2139

2139

1589

—

Events, n (%)

310 (14.5)

319 (14.9)

62 (3.9)

—

RR (95% CI)

1.68 (1.44-1.95)

1.47 (1.27-1.70)

1.89 (1.34-2.67)

—

< .001

P

< .001

P

= .003

P

—

New diagnosis of diabetes during follow-up, late-onset CHF, cardiovascular events, and all-cause mortality were analyzed as outcome measures using Cox proportional models. Peri&shy;pheral vascular disease; the use of beta blocking agents, HMG-CoA re&shy;ductase inhibitors (statins), angio&shy;ten&shy;sin-converting enzyme inhibitors, anti&shy;platelet agents, and experimental treatments; residual ischemia; ejection fraction; smoking; total cholesterol level; electrical instability; sex; and age were analyzed using multivariable analysis. BMI levels taken at the start of the study were included in the multivariate model to evaluate the effect of weight reduction.

Results

At the start of the study, 3047 pa&shy;tients (29.3%) had METS, and 2139 patients (20.6%) had diabetes mellitus. Both diabetic and METS patients had a higher probability of death and cardiovascular events compared with control patients (Table). Hospital&shy;iza&shy;tion for CHF was more likely to occur in diabetic patients, but not in METS patients (Table). The risk of developing diabetes was almost two times greater for METS patients compared with normal patients, and this risk increased dramatically depending on how many diagnostic criteria for METS a patient had (Table and Figure 1).

Compared with the reference class, the risk of developing diabetes was increased 113% for those with a substantial weight gain, 57% for those with a moderate weight gain, and 48% for those with a slight weight gain (Figure 2). The risk of developing diabetes decreased 41% for those pa&shy;tients who lost a substantial amount of weight and 18% for those who lost a moderate amount of weight. When a subgroup of patients with a BMI >25 kg/m² was compared with patients who had no weight increase or a slight decrease in weight, the risk of diabetes increased 121% for those with a substantial weight gain, 63% for those with a moderate weight gain, and 50% for those with a slight weight gain. Pa&shy;tients who achieved a moderate and substantial reduction of weight, however, had a 15% and 45% decreased risk of diabetes development, respectively. Weight change did not show any significant effect on the risk of cardiovascular events in all patients and in patients with BMI > 25 kg/m².

Discussion

In this analysis, abnormalities of glucose metabolism occurred in half of patients with a recent MI; 29% of participants had METS, and 21% had type 2 diabetes mellitus, both of which correlate with a higher risk of all-cause mortality and cardiovascular events. The risk of cardiovascular events in METS patients is between that of normal patients and those with diabetes, and the risk of death is mainly associated with the development of diabetes. There was no difference in the risk of death between METS patients who did not become diabetic (5.0%) and normal patients (4.7%) in the last 2.5 years of follow-up. For the approximately 10% of patients with METS who later developed diabetes, the risk of death increased significantly (6.5% of patients), which was in between normal patients (4.7%) and diabetic patients (9.2%) at baseline.

As in patients without coronary heart disease,⁶ for those with a previous MI, METS was a powerful prognosticator for developing diabetes, with a greater number of METS diagnostic criteria correlating with a greater risk of developing diabetes. This latter aspect concurs with previous studies⁷ and suggests that the number of METS criteria may be related to higher levels of insulin resistance and consequently a progressive increase in risk.

From a pathophysiological standpoint, these findings confirm the stronger association of hyperinsulinemia with macrovascular problems, which is present several years before the development of manifest diabetes, compared with hyperglycemia.^8-11 Be&shy;cause METS could be considered a marker of insulin resistance and hy&shy;perinsulinemia,^1-3 it represents an early indicator of increased vascular risk.

From a clinical standpoint, the prognostic role of METS in patients without coronary artery disease shown in other studies can now be extended to post-MI patients based on our results.^6,12-14 These results highlight the need to identify patients at high risk for developing diabetes as early as possible so that proper lifestyle habits can be instituted and more aggressive drug treatment can be initiated to decrease cardiovascular risk. We found that favorable weight changes during the first 6 months of follow-up were associated with a reduced risk of diabetes mellitus, corroborating earlier studies indicating that reduced weight, exercise, and improved diet can de&shy;crease diabetes risk by 50%.^15-18

Conclusion

We found that 50% of post-MI patients had 2 conditions related to insulin resistance, METS and diabetes. These 2 conditions portend an increased risk of cardiovascular events and death. In patients with METS, the risk of developing diabetes decreased with weight loss. The results of this study indicate that a more aggressive approach to the treatment of these patients, particularly with regard to changes in life&shy;style, is needed.