Design: Systematic review and meta-analysis of randomised trials
Population: 5 trials, 3,315 asymptomatic adults with diabetes and no known coronary disease
Primary outcome: Major adverse cardiovascular events
Effect: Pooled HR/RR 0.76 (95% CI 0.48 to 1.19), four trials pooled

Diabetes can blunt anginal symptoms, so screening for silent myocardial ischaemia has intuitive appeal. This systematic review pooled randomised trials comparing systematic screening with usual care in asymptomatic adults with diabetes and no known coronary disease. Five trials met criteria (3,315 participants), including DIAD, DYNAMIT and FACTOR-64.

In the four trials that could be pooled, screening did not significantly reduce major adverse cardiovascular events (pooled HR/RR 0.76, 95% CI 0.48 to 1.19). Screening did find occult ischaemia and atherosclerosis, but finding it did not translate into fewer events. False positives and procedure-related harms were reported inconsistently, so the downside is likely understated.

For family physicians, the implication is to resist routine stress testing or CT coronary angiography in well people with diabetes. Effort is better spent on what does reduce events: blood pressure, lipids, smoking, and drugs with proven cardiovascular benefit where indicated. Testing is reasonable when symptoms, an abnormal ECG or a planned major change in activity mean the result would change management.

Do not order routine stress testing or coronary imaging in asymptomatic adults with diabetes.
Focus on proven measures: blood pressure, statins, smoking cessation and cardioprotective glucose-lowering drugs where indicated.
Ask about atypical symptoms such as exertional breathlessness or fatigue, which can replace angina in diabetes.
Test when symptoms or an abnormal ECG mean the result would change management.

Why it matters

It gives a clear answer to a test patients and clinicians often feel they should be doing.

The statistics, in plain English

A pooled estimate of 0.76 points towards benefit, but the interval (0.48 to 1.19) includes no effect, so screening was not shown to help. With only about 3,300 participants and modern risk factor treatment in both arms, a small benefit cannot be excluded, but a large one is unlikely.