Loss of pulsatility during impella support: incidence, predictors, and impact on short- and long-term outcomes from the IMMERGE registry

Background

Percutaneous microaxial flow pumps such as Impella are increasingly used in cardiogenic shock (CS) and complex high-risk indicated PCI (CHIP). Loss of pulsatility (LoP) during Impella support, defined as a transient reduction in pulse pressure <20 mmHg for >5 seconds, reflects ventriculo-arterial uncoupling and profound ventricular unloading. However, its clinical and prognostic implications remain poorly defined.

Methods

We analyzed data from the international multicentric IMMERGE registry including patients treated with contemporary Impella devices (CP, 5.0, 5.5) between January 2013 and January 2025. Among 1019 enrolled patients, LoP data were available in 593 cases. Patients were stratified according to the occurrence of LoP during support. Multivariable logistic regression assessed predictors of LoP and its association with 30-day mortality. Cox proportional hazards models evaluated long-term mortality (up to 1 year). Hemodynamic and biochemical trajectories were analyzed using linear mixed-effects models.

Results

LoP occurred in 170 of 593 patients (28%). It was more frequent in patients presenting with CS (64.1% vs. 37.8%, p<0.001), cardiac arrest (40.8% vs. 12.8%, p<0.001), and acute vessel occlusion (44.0% vs. 22.9%, p<0.001). Independent predictors of LoP were cardiogenic shock (OR 1.62, p=0.01), cardiac arrest (OR 2.89, p<0.001), and acute vessel occlusion (OR 1.56, p<0.001).

At 30 days, LoP was associated with significantly higher all-cause mortality (43.5% vs. 20.0%, p<0.001), acute kidney injury, major bleeding, and sepsis. After adjustment, LoP independently predicted 30-day mortality (OR 2.2, 95% CI 1.2–3.8, p=0.007), together with CS (OR 9.5, p<0.001). Hemodynamically, LoP patients exhibited lower SvO₂, higher lactate levels, lower systolic blood pressure, higher heart rate, and increased vasoactive support requirements. At a mean follow-up of 499±100 days, crude all-cause mortality remained higher in the LoP group (54.7% vs. 35.6%, p<0.001), but LoP was not independently associated with long-term mortality after multivariable adjustment (HR 1.19, p=0.53). Age, CKD, cardiac arrest, and CS were independent predictors of long-term mortality.

Conclusions

Loss of pulsatility during Impella support is frequent and identifies a subgroup of critically ill patients with markedly increased short-term mortality and end-organ complications. LoP independently predicts 30-day mortality but not long-term mortality after adjustment. Recognition of LoP should prompt immediate hemodynamic reassessment and intensified clinical management. Further studies are warranted to clarify mechanistic pathways and evaluate targeted mitigation strategies.