Prognostic Usefulness of Cardiac Stress Test Modalities in Patients With Type 2 Diabetes Mellitus Who Underwent Myocardial Perfusion Scintigraphy (from the Basel Asymptomatic High-Risk Diabetics' Outcome Trial)

Our study aimed to assess predictors of the stress test technique used and to evaluate the impact of exercise level achieved on risk stratification in patients with asymptomatic type 2 diabetes without a previous coronary artery disease. Little is known whether co-morbidities of these patients predict the stress technique and whether physical performance provides risk stratification: 400 patients underwent clinical evaluation and myocardial perfusion scintigraphy (MPS) using physical or pharmacological stress. Physical patients were divided into 2 groups: achieving <6 and ≥6 METs, respectively. The mean follow-up time was 2 years. Major cardiac events (MACEs) included myocardial infarction and/or cardiac death. Independent predictors of pharmacological stress were a body mass index of >30 kg/m2 (hazard ratio 1.076, 95% confidence interval 1.027 to 1.127, p = 0.002) and a peripheral arterial disease (hazard ratio 2.888, 95% confidence interval 1.446 to 5.769, p = 0.003). Pharmacological patients had more MACE than physical patients (3.2% vs 1.0%, p = 0.03). Patients achieving <6 METs had a similar MACE rate as pharmacological patients (3.0% vs 3.2%, p = not significant) and more MACE than patients achieving ≥6 METs (3.0% vs 0.4%, p = 0.01). In patients achieving <6 METs and in pharmacological patients, MPS added an incremental prognostic value to pretest information (p values for global chi-square 0.012 and 0.04, respectively). In high-risk asymptomatic diabetic patients, co-morbidities were predictive of the stress technique used. Pharmacological patients had more MACE, similar to those unable to achieve 6 METs. MPS provided an incremental prognostic value in pharmacological patients and in patients with <6METs. In contrast, patients who were able to achieve ≥6 METs were at low risk and do not need further risk stratification.