TOPLINE:

A score incorporating one angiographic, one procedural, and five clinical variables accurately predicted mortality and facilitated risk stratification in the short-, medium-, and long-term among patients with ST-elevation myocardial infarction (STEMI) who underwent percutaneous coronary intervention (PCI).

METHODOLOGY:

• Researchers derived an externally validated a score (PREDICT-STEMI) to estimate long-term mortality in patients who undergo primary PCI for STEMI.
• For the derivation cohort, they used data on 23,086 patients with STEMI who underwent PCI across 10 Spanish hospitals from January 2010 to December 2020 (mean age, 63 years; 78.1% men). The median follow-up duration was 4.3 years, and the maximum follow-up duration was 10.2 years.
• External validation was performed in one cohort of 1498 patients and another cohort of 1112 patients, with median follow-up durations of 10 years and 2 years, respectively.
• The primary outcome was long‑term all‑cause mortality. Secondary outcomes included short-term (0-30 days) and mid-term (30 days to 5 years) mortality rates.
• Researchers estimated the concordance index (C‑index) to assess how well the score separated patients who died earlier from those who lived longer, with higher values indicating better discrimination. They also compared the performance of PREDICT-STEMI with that of existing prediction models.

TAKEAWAY:

• The rate of long‑term all‑cause mortality in the derivation cohort was 30.2% over the available follow‑up period.
• The PREDICT-STEMI score is made up of seven predictors: age, diabetes, prior MI, prior stroke or transient ischemic attack, hemodynamic instability at presentation, three-vessel disease, and the use of mechanical circulatory support (P < .001 for all).
• The PREDICT-STEMI score achieved a C-index of 0.81 in the derivation cohort and 0.81 and 0.84 in the two external validation cohorts. Results were generally consistent at short and mid term. The score performed similar to or better than many existing models on several measures.
• High‑risk patients (score ≥ 60) had a 6.5‑fold higher risk for long‑term mortality than low‑risk patients (score < 60) in the derivation cohort. The risks in the two validation cohorts were 5.37-fold and 8.59-fold higher for high-risk patients (P < .001 for all).

IN PRACTICE:

“The score is a simple tool that can accurately predict mortality and facilitate risk stratification in the short, medium, and long term,” the researchers of the study wrote.

SOURCE:

The study was led by Luis Ortega-Paz, MD, of the University of Florida College of Medicine in Jacksonville, Florida. It was published online on April 29 in Heart.

LIMITATIONS:

The score was derived from a network dataset focused on system performance and may not have included all relevant variables. The model predicted only mortality and did not assess other outcomes such as major adverse cardiovascular events. Definitions of baseline characteristics may have varied across the derivation and validation datasets.

DISCLOSURES:

The authors did not declare any specific funding for the study. Several authors reported receiving consulting fees, speaker’s honoraria, and institutional research grants from various medical devices and pharmaceutical companies. Some authors also disclosed receiving support for unrelated clinical trials.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.