Adrenomedullin system activation and risk of cardiogenic shock and death after acute coronary syndromes

Background Cardiogenic shock (CS) remains a leading cause of mortality in acute coronary syndromes (ACS), with 1-year mortality plateauing at 50%. Disruption of the ADM system leads to vascular leakage and systemic and pulmonary oedema, a core pathological feature of CS. Nonetheless, the predictive value of ADM system components: (i) glycine-extended adrenomedullin (ADM-gly; precursor), (ii) peptidylglycine α-amidating monooxygenase (PAM; converting enzyme) and (iii) biologically active adrenomedullin (Bio-ADM) for the prediction of CS and mortality remains unclear. Purpose The present study aimed (i) to evaluate the association of 3 ADM system components with in-hospital CS and 1-year mortality in ACS patients and (ii) to develop and validate a risk prediction model incorporating ADM biomarkers to improve early identification of patients at high-risk of CS. Methods We analyzed 4,787 ACS patients (SPUM-ACS), assessing ADM-gly, PAM, and Bio-ADM levels at presentation. Patients were classified into four phases based on whether their ADM-gly and Bio-ADM levels were above or below their respective medians: Resting (both below median), Initial (ADM-gly above and Bio-ADM below median), Active (both above median), and Exhaustion (ADM-gly below and Bio-ADM above median).The primary endpoint was in-hospital CS; the secondary endpoint was 1-year mortality. Uni- and multivariable logistic and Cox regression models were fit for CS and mortality prediction, respectively. Model performance was evaluated using receiver operating characteristic (ROC) curves, calibration plots, and reclassification metrics. Results ADM-gly and Bio-ADM were independently associated with in-hospital CS (adjusted OR: 1.65 [95% CI, 1.41–1.93] and 1.64 [95% CI, 1.35–1.98]) and 1-year mortality (adjusted HR: 1.34 [95% CI, 1.11–1.61] and 1.49 [95% CI, 1.20–1.85]). Both biomarkers outperformed NT-proBNP in predicting CS (AUC: 0.67 [95% CI, 0.63–0.72] and 0.65 [95% CI, 0.60–0.69]; p < 0.001), whereas PAM had limited predictive value (AUC: 0.55 [95% CI, 0.51–0.59]; p = 0.764) and was not significantly associated with CS (adjusted OR: 1.51 [95% CI, 0.96–2.40]) or 1-year mortality (adjusted HR: 1.30 [95% CI, 0.78–2.15]) after ORBI adjustment (Figure 1). Patients stratified by ADM activation phases had distinct risks, with the Active phase showing the highest CS (8.1%) and 1-year mortality (5.7%) incidence.The ADM-SHOCK score, integrating ADM-gly, Bio-ADM, and clinical/hemodynamic variables, outperformed ORBI for CS prediction (AUC: 0.81 [95% CI, 0.78–0.84] vs. 0.76 [95% CI, 0.72–0.79], p < 0.001) and accurately predicted 1-year mortality (Figure 2). Conclusion ADM system biomarkers refine risk prediction for in-hospital CS and 1-year mortality in patients with ACS, underscoring the importance of vascular leakage in CS pathogenesis. The ADM-SHOCK score enhances the performance of established prediction tools for the personalized management of patients with ACS.