The prognostic value of admission D-dimer level in patients with cardiogenic shock after acute myocardial infarction

BACKGROUND: Shock is associated with the activation of the coagulation and fibrinolytic system, and D-dimer is the degradation product of cross-linked fibrin. However, the prognostic value of D-dimer in patients with cardiogenic shock (CS) after acute myocardial infarction (AMI) remains unclear. METHODS: We retrospectively analyzed the data of consecutive patients with CS complicating AMI. The primary endpoint was 30-day mortality and the secondary endpoint was the major adverse cardiovascular events (MACEs) including 30-day all-cause mortality, ventricular tachycardia/ventricular fibrillation, atrioventricular block, gastrointestinal hemorrhage, and non-fatal stroke. Restricted cubic spline (RCS) analyses were performed to assess the association between admission D-dimer and outcomes. A multivariable Cox regression model was performed to identify independent risk factors. The risk predictive potency with D-dimer added to the traditional risk scores was evaluated by C-statistics and the net reclassification index. RESULTS: Among 218 patients with CS complicating AMI, those who died during the 30-day follow-up presented with worse baseline characteristics and laboratory test results, including a higher level of D-dimer. According to the X-tile program result, the continuous plasma D-dimer level was divided into three gradients. The 30-day all-cause mortality in patients with low, medium, and high levels of D-dimer were 22.4, 53.3, and 86.2%, respectively (p < 0.001 for all). The 30-day incidence of MACEs was 46.3, 77.0, and 89.7%, respectively (p < 0.001). In the multivariable Cox regression model, the trilogy of D-dimer level was an independent risk predictor for 30-day mortality (median D-dimer cohort: HR 1.768, 95% CI 0.982–3.183, p = 0.057; high D-dimer cohort: HR 2.602, 95% CI 1.310–5.168, p = 0.006), a similar result was observed in secondary endpoint events (median D-dimer cohort: HR 2.012, 95% CI 1.329–3.044, p = 0.001; high D-dimer cohort: HR 2.543, 95% CI 1.452–4.453, p = 0.001). The RCS analyses suggested non-linear associations of D-dimer with 30-day mortality. The enrollment of D-dimer improved risk discrimination for all-cause death when combined with the traditional CardShock score (C-index: 0.741 vs. 0.756, p(difference) = 0.004) and the IABP-SHOCK II score (C-index: 0.732 vs. 0.754, p(difference) = 0.006), and the GRACE score (C-index: 0.679 vs. 0.715, p(difference) < 0.001). Similar results were acquired after logarithmic transformed D-dimer was included in the risk score. The improvements in reclassification which were calculated as additional net reclassification index were 7.5, 8.6, and 12.8%, respectively. CONCLUSION: Admission D-dimer level was independently associated with the short-term outcome in patients with CS complicating AMI and addition of D-dimer brought incremental risk prediction value to traditional risk prediction scores.