The role of inflammation in post-STEMI remodeling: findings from the PULSE-MI trial

Left ventricular (LV) remodeling after ST-segment elevation myocardial infarction (STEMI) is a key determinant of heart failure and prognosis. While infarct size predicts prognosis and influence remodeling, it does not fully explain why some patients develop adverse dilation and others do not, suggesting an additional role of inflammation. This study examines the relationships between infarct size, inflammation, and LV remodeling and explores whether glucocorticoids in STEMI affects remodeling and LV dilation.

This was a post-hoc sub-study of the 1:1 investigator-initiated, blinded, placebo-controlled, randomized clinical trial, PULSE-MI, in which patients ≥18 years with <12 hours of acute chest pain and STEMI were randomized to a pulse dose of intravenous glucocorticoid (250 mg methylprednisolone) or placebo in the prehospital setting. Cardiac magnetic resonance (CMR) was conducted during admission and after 3 months. C-reactive protein (CRP) after 24h was used as a marker of acute inflammation. LV remodeling was assessed as the absolute and relative change in LV end-diastolic volume (LVEDV) between acute and 3-month CMR scans. Adverse LV remodeling is defined as a ≥20% relative increase in LVEDV. Associations were explored by linear and logistic regression analysis.

Of the 530 patients included in the primary trial, CMR at both time points was available for 311 (59%). Median age was 63 years (IQR 55,72), 255 (82%) were men, LV function upon arrival was 45% (IQR: 35,55), and the left anterior descending artery was the most common culprit vessel (43%). Baseline characteristics were well-balanced between treatment allocation in this subgroup. Median acute infarct size was 15% (IQR 8, 23) and 12% (IQR 5,22), and median 3-month LVEDV index was 40 ml/m2 (IQR 32,47) and 37 ml/m2 (IQR 30,50), in placebo and glucocorticoid, respectively. Adverse remodeling was seen in 36 patients (12%) (18 in each arm, P = 0.83). Acute infarct size was associated with greater LV dilation at 3 months only in the placebo arm (Figure 1). Although infarct size alone explained only 3 % of the variance in LV remodeling (R² = 0.03), the odds of adverse remodeling increased by 112% (95%CI 35%-233%) in the placebo arm vs. 28% (95%CI -16%-95%) in the glucocorticoid arm per 10 % LV mass infarct, P interaction = 0.10. CRP after 24 hours increased the odds of adverse LV remodeling by 24% (95% CI 1%-54%) per doubling, irrespective of glucocorticoid treatment, but overall CRP did not explain any of the variation in absolute or relative change in LVEDV, with R² < 0.001.

Infarct size was associated with LV remodeling but accounted for only a smaller fraction of the observed variance. CRP was not a predictor of LV dilation. Prehospital glucocorticoid administration did not worsen LV remodeling, suggesting that modulation of inflammation with a pulse-dose of glucocorticoid is safe but may have a limited impact on post-STEMI structural outcomes.

LV Remodelling