Differential impact of mechanical circulatory support in STEMIversus NSTEMI cardiogenic shock

Background Mechanical circulatory support (MCS) is increasingly used in acute myocardial infarction–related cardiogenic shock (AMI-CS). In STEMI-CS, MCS has been associated with improved outcomes, yet whether similar benefit exists in NSTEMI-CS remains uncertain. Given important differences in comorbidity burden, coronary anatomy, and shock trajectory, treatment effects may differ by AMI phenotype. Purpose To evaluate whether the association between MCS use and in-hospital outcomes differs between STEMI-CS and NSTEMI-CS. Methods Adults with AMI-CS treated from 2016–2022 were stratified by AMI type (STEMI vs NSTEMI). The primary outcome was a composite of in-hospital mortality or in-hospital renal replacement therapy (RRT). The secondary outcome was in-hospital mortality. Multivariable logistic regression models were constructed separately within each AMI type to evaluate associations with the composite outcome. Results Among 1,375 patients (STEMI n=790; NSTEMI n=585), NSTEMI patients were older (71.7±12.0 vs 67.2±13.0 years; p<0.001) and had greater comorbidity burden (CCI 6.3±3.2 vs 5.0±3.0; p<0.001). NSTEMI presented with less severe admission SCAI stage but higher baseline and peak creatinine (both p<0.001). In-hospital mortality was similar (34.6% vs 30.3%; p=0.09), though NSTEMI had longer length of stay (p<0.001). MCS was used less frequently in NSTEMI (54.5% vs 71.6%; p<0.001) and initiated later (median 1.0 vs 0.0 days; p<0.001). Use of IABP, Impella, VA-ECMO, and multi-device support were all lower in NSTEMI (all p<0.05). Despite these differences, rates of primary and secondary outcomes were similar between groups. In stratified multivariable models, MCS use was associated with lower odds of the composite outcome in STEMI (adjusted OR 0.63, 95% CI 0.41–0.97; p=0.04) but not in NSTEMI (adjusted OR 1.28, 95% CI 0.80–2.07; p=0.32). Consistent findings were observed for the secondary outcome of in-hospital mortality, with MCS associated with lower odds of death in STEMI (aOR 0.5, 95% CI 0.3–0.8; p=0.005) but not in NSTEMI (aOR 1.2, 95% CI 0.7–2.2; p=0.4). Mortality models were adjusted for the same covariates as the primary analysis, with haemodialysis treated as a covariate. Revascularisation was strongly protective against the primary outcome in both phenotypes (STEMI aOR 0.26; NSTEMI aOR 0.19; both p<0.001). Conclusion In AMI-CS, the association between MCS and clinical outcomes differs by infarct phenotype. While MCS was associated with improved outcomes in STEMI-CS, no similar association was observed in NSTEMI-CS. These findings suggest potential effect heterogeneity and underscore the need for phenotype-specific evaluation of MCS strategies.