Background: Cardiogenic shock (CS) is a critical condition defined by a severe decline in cardiac function, leading to systemic tissue hypoperfusion, which ultimately precipitates progressive multiorgan failure. Despite recent therapeutic advances including a growing use of temporary mechanical circulatory support (tMCS) devices, the short-term mortality rate remains high at around 40-50%. Bleeding events are the most frequent complications of tMCS devices and contribute to morbidity and mortality in CS. Gaining a better understanding of determinants of bleeding risk in these patients is fundamental for future optimization of management strategies.
Methods: MIRACLE is an ongoing, prospective, multicenter cohort study enrolling patients with CS of SCAI stage ≥ B. To complement this cohort, a retrospective sample of comparable patients treated at the University Hospital Freiburg (Bad Krozingen and Freiburg sites) between 2018 and 2023 was included, yielding a total of 627 patients. For the present analysis, we included patients who received tMCS early (6 hours before or after ICU admission (n=93)). Among these, a prospectively enrolled subset (n=38) had available data on immature platelet indices. Statistical analyses were performed using Cox proportional hazards and logistic regression models. Continuous predictors were log-transformed and standardized (z-scored) prior to modeling. Severe bleeding was defined as BARC ≥ 3b.
Results: In patients with CS treated with tMCS, platelet counts decreased markedly in the initial five days after ICU admission (median platelets, 103/µl: 184 on d1, 89.5 on d3, and 82 on d5). When stratified by median platelet count at baseline, age, BMI, the Charleston Comorbidity Index, and the etiology of CS did not differ between groups. However, the incidence of severe bleeding was significantly higher among patients with platelet counts below the median at baseline (66% vs. 37%, p=0.01). In multivariable logistic regression analysis, higher baseline platelets were associated with a lower risk of severe bleeding (adjusted OR 0.60 per 1 SD, 95% CI 0.34–0.97), even after correction for age, two vs one tMCS device at index, and out-of-hospital cardiac arrest (OHCA). Patients with baseline platelet counts above the median had a lower risk of 30d all-cause mortality compared to those below the median (adjusted HR 0.42, 95% CI 0.21–0.81). Immature platelet count (103/µl) demonstrated a decline from a median of 8.45 (IQR 5.6–11.78) on d1 to 5.8 (IQR 4.1–7.4) on d3, with stable values at d5 (median 5.9, IQR 3.6–8.65). The immature platelet fraction (IPF) increased steadily from a median of 4.21% (IQR 2.98–6.26) on d1 to 5.1% (IQR 4.18–6.67) on d3 and 5.9% (IQR 3.75–6.95) on d5, indicating a gradual rise in IPF levels during the observation period. In the cohort without early tMCS initiation (n=534), baseline platelet count did not correlate with bleeding or all-cause mortality.
Conclusion: This study shows an association of lower initial platelet counts with all-cause mortality and severe bleeding in CS patients requiring early tMCS. Further analyses will focus on immature platelet indices and whether maintaining higher platelet counts could be advantageous for this particular patient group.
