Trajectories of early mobilisation during Impella 5.5 support and associated outcomes in patients with cardiogenic shock

Background The use of catheter-based micro-axial flow pump, Impella, has progressively increased, and patient prognosis has improved following cardiogenic shock (CS). Notably, Impella 5.5 enables higher flow support and longer duration of use compared with Impella CP, and facilitates patient activity because of the cannulation approaches to the axillary or subclavian artery. However, the feasibility and safety of early mobilisation in this setting have not been well established. This study aimed to investigate mobility level trajectory and its safety and clinical outcomes in patients with CS supported with Impella 5.5 using a large amount of physiotherapy session data. Methods We reviewed consecutive patients with CS who underwent early physiotherapy while supported with Impella 5.5 in the intensive care unit (ICU) of a tertiary university hospital from 2022 to 2025. Daily mobility levels were determined through multidisciplinary discussions involving physiotherapists, cardiologists, intensive care nurses, and clinical engineers. We assessed mobility levels using the ICU Mobility Scale (IMS) as a primary outcome and recorded adverse events that led to physiotherapy termination during each session. Patient status after Impella 5.5 cessation was defined as follows: successful weaning from Impella 5.5; bridging to an implantable left ventricular assist device (iLVAD); death following transition to palliative care due to failure to wean from Impella 5.5; and death after weaning from Impella 5.5. The trajectories of IMS levels and their association with patient status were examined using a linear mixed-effects model, adjusting for age, adverse events, and prior mechanical circulatory support before Impella 5.5 implantation as covariates. Results A total of 876 physiotherapy sessions among 30 patients were included for analysis, with 550 sessions (63.3%) of out-of-bed mobilisation. Overall, IMS significantly increased throughout the clinical course, after adjustment for confounders (P <0.001, Figure 1). IMS trajectories were higher in patients successfully weaned from Impella 5.5 compared with bridging to iLVAD, death following transition to palliative care due to failure to wean, and death after weaning (P for interaction = 0.064, 0.002, and 0.015, respectively). Adverse events leading to termination of physiotherapy were observed in 138 sessions (15.8%), including 43 events related to Impella device alarms (4.9%), 91 due to decreased arterial pressure (< 60 mmHg) (10.4%), and 10 due to increased ventricular arrhythmia (1.1%), 5 of which were non-sustained ventricular tachycardia (0.6%); however, no life-threatening adverse events occurred during physiotherapy (Figure 2). Conclusions Early mobilisation trajectories increased progressively after Impella 5.5 implantation without severe adverse events. Higher mobility trajectories were associated with more favourable clinical status in patients with CS.