Background:
The Immature Platelet Fraction (IPF) reflects bone-marrow thrombopoietic activity. Its prognostic relevance in cardiogenic shock (CS) — and whether it adds information beyond the total platelet count — is unclear.
Methods:
In the prospective multicentre MIRACLE registry (all-cause CS, SCAI B–E), IPF was measured on days 1, 3 and 5; 265 patients with a day-1 value formed the analysis cohort. The primary endpoint was 30-day mortality, the secondary endpoint severe bleeding (SHARC ≥ 3b). Cox and time-varying Cox models (IPF % updated D1→D3→D5; robust SE) were adjusted for age, SCAI stage, first lactate, number of devices and Charlson Comorbidity Index; a sensitivity analysis mutually adjusted IPF and total platelet count. Death as a competing event for bleeding was modelled by Fine-Gray and cause-specific regression.
Results:
Of 265 patients (median age 67.5 years, 28.3 % female, 45.7 % SCAI ≥ D), 92 (34.7 %) died within 30 days and 40 (15.1 %) had severe bleeding; deaths occurred early (median 2.8 days, 76 % within 7 days). Higher IPF was modestly and consistently associated with 30-day mortality across baseline (IPF % adjusted HR 1.22 per SD, 95 % CI 0.99–1.51; p = 0.065) and time-varying models (adjusted HR 1.29, 1.03–1.63; p = 0.027; crude HR 1.35, 1.10–1.66; p = 0.004). This association persisted after adjustment for the total platelet count (IPF absolute HR 1.26 per SD, 0.99–1.60; p = 0.055), whereas the platelet count was not independently prognostic in this cohort (HR 1.02; p = 0.88). IPF was not associated with severe bleeding, including after accounting for the competing risk of death (Fine-Gray subdistribution HR 0.99, 0.67–1.44; p = 0.94). On a day-1 phenotype combining IPF % and platelet count, the "elevated IPF % plus thrombocytopenia" group had the highest mortality (56.5 %; crude HR 2.16, 1.20–3.89; p = 0.010; attenuated to HR 1.70, 0.91–3.17 after adjustment), whereas severe bleeding clustered in the thrombocytopenic groups and was highest with thrombocytopenia but normal IPF % (40.9 %).
Conclusion:
In this prospective multicentre CS cohort, a higher IPF showed a modest association with early 30-day mortality that was independent of the total platelet count and consistent between single and serial measurements, but was unrelated to bleeding. The combination of elevated IPF % and thrombocytopenia marked higher crude — though not independent — risk. As IPF can be measured on automated haematology analysers, it is a readily obtainable candidate marker; its incremental value over established risk scores remains to be demonstrated and warrants external validation.
