SMART-REACH2 puts a number on years gained from secondary prevention

Design

Risk-model derivation with regional recalibration and external validation

Population

8,708 derivation patients; 2,085,780 validation patients with atherosclerotic disease in 54 countries

Primary outcome

Recurrent myocardial infarction, stroke or cardiovascular death

Effect

Pooled C-statistic 0.68 (0.66–0.69); estimated 2 to 4.4 event-free years gained in a 50-year-old

The 2021 ESC prevention guideline recommends the SMART-REACH model to estimate lifetime risk and treatment benefit in patients with established atherosclerotic disease. SMART-REACH2 updates it. It was derived in 8,708 patients with coronary, cerebrovascular or peripheral artery disease or aortic aneurysm, then recalibrated to European and other global risk regions and externally validated in 2,085,780 patients from 54 countries.

Discrimination was moderate (pooled C-statistic 0.68, range 0.66–0.72 by region), and calibration was adequate across regions, sexes and disease types. For an example 50-year-old, lowering systolic pressure by 15 mmHg and LDL cholesterol by 1.0 mmol/L was estimated to add 2 years free of cardiovascular disease in a low-risk region and 4.4 years in a very-high-risk region.

The practical value is in the conversation. Many patients after an event stop a statin or accept a blood pressure that is close enough. Expressing benefit as years of life without another event is easier to weigh than a relative risk reduction, and the model now adjusts for region and sex.

The gains are model estimates, not observed results, and a C-statistic of 0.68 means the model ranks individuals only moderately well. It supports shared decisions; it does not replace guideline targets.

• In patients with established atherosclerotic disease, consider expressing treatment benefit as years free of another event.
• Use the region-calibrated estimate rather than a European default where your patient's region is covered.
• Combine blood pressure and LDL lowering when discussing benefit; the estimated gains came from both together.
• Younger patients stand to gain the most lifetime years; they are also the ones most likely to stop treatment.
• Record the estimate in the notes so it can be revisited at follow-up.

Why it matters

A guideline-endorsed tool now reflects regional risk, which matters when the default estimates were built in low-risk Europe.

Don't overread it

The years gained are model-based estimates for an example patient, not results observed in treated people.

The statistics, in plain English

A C-statistic of 0.68 means that if you pick one patient who has an event and one who does not, the model ranks them correctly about 68% of the time — useful but far from perfect. The 2 to 4.4 years are model projections for an example patient, not outcomes measured in a trial.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.