Patients with ST segment elevation myocardial infarction (STEMI) who undergo percutaneous coronary intervention (PCI) should be triaged for intensive care (ICU) treatment based on risk factors that include reperfusion delay to avoid overuse of ICU facilities. A review of patterns of ICU use found that although >80% of stable patients with STEMI are treated in the ICU after primary PCI, the risk for developing a complication requiring ICU care is 16%.
The study was published in the JACC: Cardiovascular Interventions by Jay Shavadia (Duke Clinical Research Institute, Durham, USA) et al. They recommend “implementing a risk-based triage strategy, inclusive of factors such as degree of reperfusion delay” which, they say, “could optimise ICU utilisation for patients with STEMI”.
The authors aimed to describe variability in ICU use for patients with uncomplicated STEMI, and to evaluate the proportion who developed in-hospital complications requiring ICU care, as well to assess whether there was a relationship between variations in patterns of ICU use and complication rates and first medical contact-to-device times.
They looked at data from the Chest Pain MI Registry, linked to Medicare claims, for haemodynamically stable patients with STEMI aged ≥65 years of age who were treated with uncomplicated primary PCI, and examined patterns of ICU use, stratified by timing of reperfusion—early (first medical contact-to-device time ≤60 minutes), intermediate (61–90 minutes), or late (>90 minutes).
In all, 19,507 patients with STEMI were treated at 707 hospitals. Of these, 82.3% were treated in ICUs. The median ICU stay was one day (interquartile range [IQR] one–two days). The median first medical contact-to-device time was 79 minutes (IQR 63–99 minutes), and 22% of patients had early reperfusion, 44.8% intermediate, and 33.2% late reperfusion. There was no difference in rates of ICU use between patients who had early, intermediate, and late reperfusion times (82%, 83%, and 82%, p for trend=0.44).
Overall, 3,159 patients (16.2%) developed complications that required ICU care during the time they were hospitalised. Of these, 3.7% died, 3.7% had a cardiac arrest, 8.7% shock, 0.9% stroke, 4.1% high-grade atrioventricular block requiring treatment, and 5.7% respiratory failure. Patients with longer first medical contact-to-device times were more likely to develop at least one of these complications (early 13.4%, intermediate 15.7%, late 18.7%, p for trend<0.001, adjusted odds ratio [AOR, using early reperfusion as reference] for intermediate 1.13, 95% confidence interval [CI] 1.01–1.25; AOR for late 1.22, 95% CI 1.08–1.37).
Shavadia et al write: “Following uncomplicated primary PCI for initially stable patients with STEMI, ICU utilisation occurs almost universally, with little variability noted across hospitals in the United States. Yet the incident rate of developing an ICU-requiring complication is 16%, and is correlated with reperfusion delays. Improved prediction aimed at the timely identification of patients at greatest risk for developing complications could help in safely rationalising ICU triage decisions for patients with STEMI.”