Baseline Potassium and 1 Year Myocardial Recovery After Primary PCI in STEMI: Mediating Role of Microvascular Obstruction

Background Serum potassium levels and microvascular obstruction (MVO) are both associated with myocardial recovery after ST‐segment elevation myocardial infarction (STEMI), but whether MVO mediates the relationship between potassium and long‐term cardiac function remains unknown. This study investigated whether baseline MVO mediates the association between baseline serum potassium levels and 1 year global longitudinal strain (GLS) recovery in STEMI patients undergoing primary percutaneous coronary intervention (PCI). Methods This retrospective cohort study enrolled 258 STEMI patients (mean age 51.3 ± 11.3 years, 93.8% male) at Affiliated Jiangyin Hospital of Nantong University, China, from January 2020 to December 2023. All patients underwent primary PCI after hospital admission (median time from symptom onset to reperfusion: 15.0 [IQR: 4.0–35.5] hours) and completed cardiovascular magnetic resonance (CMR) imaging at baseline (5–9 days post‐PCI) and at 12‐month follow‐up (10–14 months post‐PCI). Baseline serum potassium averaged 4.0 ± 0.6 mmol/L (range: 3.0–5.2 mmol/L). The primary outcome was 1 year GLS recovery (change in absolute GLS from baseline to 1‐year). Causal mediation analysis with bootstrap resampling (5000 iterations) was employed to decompose total effects into natural indirect effects (NIE) and natural direct effects (NDE), adjusting for 8 confounders selected via variance inflation factor analysis. Results Higher baseline potassium was associated with reduced baseline MVO (path a: β = −3.638, 95% CI: −3.98 to −3.29, p < 0.001) and greater 1 year GLS recovery (total effect: β = 3.621, 95% CI: 3.09 to 4.15, p < 0.001). The NIE through MVO was statistically significant ( β = 2.232, 95% CI: 1.474 to 3.036, p < 0.001), with a mediation proportion of 61.6% (95% CI: 39.8%–86.1%), indicating that approximately 61.6% of potassium’s protective association with GLS recovery was mediated through reduction in MVO. Drug‐adjusted sensitivity analyses showed robust results (mediation proportion range: 61.0%–62.2%). Reverse mediation analysis confirmed the primary causal direction (NIE for reverse pathway: −0.246). The direct effect remained significant ( β = 1.389, 95% CI: 0.463 to 2.319, p = 0.002). Conclusions In STEMI patients, baseline MVO substantially mediates the beneficial association between baseline serum potassium and 1 year GLS recovery, accounting for approximately 61.6% of the total effect. Sensitivity analyses adjusting for cardiovascular medications (diuretics, potassium supplementation, RAS inhibitors, beta‐blockers, insulin, and MRA) confirmed robust results. Reverse mediation analysis supported the primary causal direction (K ⟶ M ⟶ GLS) versus alternative pathways. These findings suggest that optimizing potassium levels may support myocardial recovery partly by mitigating MVO, though the observational design precludes definitive causal inference. Future randomized trials are warranted to evaluate potassium optimization as a therapeutic strategy for improving cardiac function post‐STEMI. Consistent findings were observed for left ventricular ejection fraction recovery and across ischemia time subgroups.