ECPELLA (VA-ECMO + Impella)

Definition

ECPELLA (also called ECMELLA) is the combined use of venoarterial extracorporeal membrane oxygenation (VA-ECMO) and an Impella transvalvular axial flow pump for cardiogenic shock. VA-ECMO provides hemodynamic stabilization and systemic oxygenation; Impella provides left ventricular (LV) unloading — counteracting the increase in LV afterload caused by retrograde VA-ECMO flow.

Key Concepts

Hemodynamic Rationale — PV Loop Perspective

• VA-ECMO retrograde aortic flow → increases LV afterload → shifts PV loop rightward → enlarges PVA (pressure-volume area = myocardial O2 consumption); this worsens myocardial ischemia
• Adding Impella to VA-ECMO drains blood from the LV → reduces PVA while systemic flow and BP are maintained
• Adding a vasodilator (reducing SVR) during ECPELLA further markedly reduces PVA and increases total flow — the "total unloading" condition
• With total unloading: PVA approaches zero (LV no longer ejects); excessive LV decompression risks LV suction → arrhythmias, hemolysis, pump malfunction sources/mcs-jic-2023 (high)

Key Clinical Data

Schrage 2020 propensity-matched cohort (n=255 ECMELLA vs 255 VA-ECMO alone; international multicenter):

• 30-day mortality HR 0.79 (95% CI 0.63–0.98; P=0.03) favoring ECPELLA
• Sub-analysis by timing of Impella addition:
  • Impella within 2h of VA-ECMO initiation: HR 0.76 (95% CI 0.60–0.97; P=0.03) — significant benefit
  • Impella >2h after VA-ECMO: HR 0.77 (95% CI 0.51–1.16; P=0.22) — no significant benefit
• Early LV unloading appears mechanistically critical sources/mcs-jic-2023 (high)

J-PVAD ECPELLA registry (n=300 total; 32.3% out-of-hospital cardiac arrest; lactate 8.7 mmol/L; n=50 with paired LVEF data):

• Mean LVEF improved from 24.9 ± 14.2% → 44.0 ± 16.6% at Impella explant (P<0.001) sources/mcs-jic-2023 (high)

Cappannoli systematic review/meta-analysis (ECPELLA vs VA-ECMO alone):

• Bleeding: RR 1.45 (95% CI 1.20–1.75)
• Hemolysis: RR 1.71 (95% CI 1.41–2.07)
• Limb ischemia: RR 1.43 (95% CI 1.17–1.75)
• Renal replacement therapy: RR 1.54 (95% CI 1.19–1.99)
• Severe infections: RR 1.26 (95% CI 0.84–1.89; NS)
• ECPELLA carries substantially higher complication burden than VA-ECMO alone sources/mcs-jic-2023 (high)

Weaning Protocol (Three-Step)

• General principle: VA-ECMO weaned first, then Impella

1. End-organ adequacy (proceed to step 2 only if met):

   • Elevated BP, normalized lactate, weaned off pressors/inotropes
   • Improved intracardiac pressures (RA pressure and PAWP amelioration)

2. RV assessment → governs VA-ECMO decannulation:

   • RA pressure <15 mmHg AND PAPi ≥1.0
   • Minimum VA-ECMO flow before decannulation: 1–2 L/min (usually 1.5 L/min)
   • Note: PAPi can be influenced by RV preload under VA-ECMO; interpret with context

3. LV assessment → governs Impella weaning:

   • PAWP <20 mmHg AND CPO ≥0.6 W
   • Minimum Impella flow: P-level 2
   • LVOT-VTI increase on echo = LV recovery and Impella weaning criterion

• Pulse pressure must reappear spontaneously before initiating weaning sequence
• If criteria not met at any step: do not proceed; intensify HF therapy or escalate device
• Significant MR at weaning (high PAWP or low CPO): evaluate severity; consider surgical repair or transcatheter edge-to-edge repair before decannulation sources/mcs-jic-2023 (high)

Practical Considerations

• Aortic regurgitation at Impella shaft/aortic valve leaflet gap: impairs LV unloading — ensure appropriate intravascular volume and controlled blood pressure
• Excessive MAP elevation during ECPELLA: risk of intracranial hemorrhage
• Blood pressure control reduces PVA and prevents overloading the pump heads (centrifugal VA-ECMO + axial Impella)
• ECPELLA has no completed RCT; all efficacy data is observational

Controversy — Optimal PVA During Cardiac Recovery

• Mazurek 2023: acute reloading after mechanical LV unloading induces cardiac apoptosis — favors prolonged deep unloading
• Diakos 2016: reloading through LVAD normalizes cardiac metabolism (glycolysis/mitochondrial oxidation mismatch corrected) — favors earlier reloading
• Optimal PVA regulation strategy during cardiac recovery remains unresolved sources/mcs-jic-2023 (high)

Contradictions / Open Questions

• No RCT evidence: Schrage 2020 is propensity-matched observational — confounding by indication (sicker patients may selectively not receive Impella) and selection bias remain
• Optimal LV unloading depth: total unloading risks LV suction/arrhythmias; partial unloading may not sufficiently reduce myocardial O2 demand — target PVA is unknown
• Timing of LV unloading: Schrage sub-analysis suggests early (<2h) Impella addition is critical; whether this is a causal time-window or reflects selection of more salvageable patients remains unclear
• Higher complication burden: Cappannoli meta-analysis — bleeding, hemolysis, limb ischemia, RRT all significantly increased vs VA-ECMO alone; net clinical benefit in broad CS populations requires RCT validation
• Cardiac reloading vs sustained unloading: mechanistic disagreement between Mazurek (apoptosis with reloading) and Diakos (metabolic normalization with reloading) — unresolved

Connections

• Related to concepts/Temporary-Mechanical-Circulatory-Support
• Related to concepts/Cardiogenic-Shock
• Related to concepts/SCAI-Shock-Classification
• Related to concepts/IABP
• Related to concepts/Pulmonary-Artery-Pulsatility-Index

Sources

• sources/mcs-jic-2023