Etiology-specific outcomes in refractory cardiogenic shock treated with Impella 2.5/CP: a 8-year real-world cohort
G Chatzis, S Syntila, B Markus, J Kreutz, H Lapp, B Schieffer, F KanoAbstract
Background
Refractory cardiogenic shock (CS) remains highly lethal despite contemporary intensive care and revascularization strategies. Percutaneous left-ventricular (LV) unloading with Impella is widely used, yet real-world outcomes across distinct CS etiologies and their clinical implications remain incompletely characterized.
Purpose
To assess in-hospital and 12-month survival in refractory CS treated with Impella 2.5/CP and compare outcomes across etiological phenotypes.
Methods
This was a retrospective single-center cohort study (2013–2021). Adults with refractory CS treated with Impella 2.5/CP were included. Patients with cardiopulmonary resuscitation on admission and those receiving additional mechanical circulatory support (e.g., VA-ECMO/IABP/other Impella systems) were excluded to obtain an Impella-focused cohort and reduce confounding by combined support strategies. Etiology groups were predefined as: acute coronary syndrome/coronary artery disease (ACS/CAD), dilated cardiomyopathy (DCM)/myocarditis, aortic stenosis, sepsis-associated CS and other causes. The primary endpoint was in-hospital mortality. Secondary endpoints was 12-month survival (estimated with Kaplan–Meier, log-rank), and device-related complications (bleeding, limb ischemia, pseudoaneurysm, stroke). Group comparisons using Fisher’s test for categorical variables with two-sided p<0.05 was considered significant.
Results
A total of 409 patients were analyzed (male 75.6%, mean age 70 years). The etiology distribution was as follows ACS/CAD n=269 (66%), DCM/myocarditis n=53 (13%), aortic stenosis n=32 (8%), sepsis-associated CS n=36 (9%), other causes n=19 (4%). Overall in-hospital mortality was 48.9% (200/409) and differed by etiology (p=0.048): ACS/CAD 46.9%, DCM/myocarditis 49.7%, aortic stenosis 56.3%, sepsis-associated CS 80.6%, other causes 31.6%. Twelve-month Kaplan–Meier curve separated significantly (log-rank p<0.01), driven by early mortality in sepsis-associated CS. Overall complication rates were clinically relevant: transfusion-requiring bleeding 14.9%, limb ischemia 11.7%, pseudoaneurysm 4.9%, and stroke 1.0%.
Conclusion
In refractory CS treated with Impella 2.5/CP, outcomes are strongly etiology-dependent. Sepsis-associated CS has markedly worse early and 12-month survival, suggesting that LV support alone may be insufficient in this phenotype and highlighting the need for phenotype-guided selection and prospective validation of tailored support strategies.