MCS Weaning & Decannulation — Cardiology Guide

Stability on support is not proof of native-heart recovery. Separate readiness to test, tolerance of reduced support, and readiness for removal. Use a supervised, device-specific trial and integrate perfusion, hemodynamics and LV/RV function. Assess durable options in parallel when recovery is uncertain.

▤

Short-term MCS wean algorithm

three separate decisions

Follow perfusion, congestion and both ventricles together. Agree on stop criteria and rescue before reducing support. The animation shows decision order, not a timetable or a prediction of recovery.

A

Readiness — the three domains

before and during reduction

Domain	What to follow (interpret at documented support and drug doses)
Clinical	Resolution of hypoperfusion without rebound at low support — clearing lactate, adequate urine output, off / low-dose pressors.
Hemodynamic	Adequate MAP and forward flow without worsening congestion: follow PCWP, RA, CI and venous saturation trends. A single threshold does not establish independence.
Echo	Assess both ventricles, forward flow, valve function and distension during the trial. LVEF, LVOT VTI and annular velocities are load-dependent. Published VA-ECMO criteria are supporting observations, not universal pass/fail rules.

B

General principles

how to run a trial

Lower stepwise within the device model and device-specific protocol. Specify the steps, observation period, monitored variables and stop criteria before starting.
Balance thrombosis and bleeding low-flow stasis increases thrombotic risk. Agree on anticoagulation and trial duration with the device team.

Expect repeats multiple trials are normal. Don't wean someone on high pressors / unresolved ischemia / arrhythmia.
Coordinate cardiology + surgery + intensivist + perfusion before any escalation or removal.

C

Partial recovery & bridge-to-decision

the trial clarifies the destination

A failed trial establishes dependence at that time. It does not automatically establish LVAD or transplant candidacy. Identify reversible barriers and reassess the trajectory. Predominant LV failure with adequate RV reserve may favor LVAD evaluation, but organ function, neurological prognosis, operative risk and patient goals also matter. In chronic HF, successful liberation may mean return to a compensated baseline, without a normal EF. See the exit plan.

D

Pitfalls that sink a trial

false reads

Hidden pressors weaning on high-dose inotropes/pressors masks dependence — minimize them first.
Timing read at end-expiration. The respiratory swing fools you.

Forgetting the RV the LV may look fine while the RV fails at low flow — check PAPi, TAPSE, CVP trend.
Low-flow stasis prolonged low flow or a stopped pump can thrombose. Use the device-specific protocol.

A wean trial is a team event — cardiology, surgery, perfusion, and the bedside RN, with echo on hand and a plan for both outcomes (explant vs escalate) agreed before you start.

Match reduction to the device and physiology. The examples below are not standing bedside orders. Device model, cannulation, anticoagulation, vasoactive support and the device-specific protocol determine safe steps and observation times. VA-ECMO tests circulatory reserve. V-V ECMO tests respiratory reserve.

◆

What you turn down, by device

the knob

Device	Wean by	Trial
IABP	Augmentation ratio	Ratio reduction (e.g., 1:1 → 1:2 → 1:3) or volume reduction. Avoid prolonged inactive balloon time
Impella	P-level	Stepwise P-level reduction with hemodynamic and echo reassessment. Model-specific minimum support
VA-ECMO	Pump flow / RPM	Staged flow reduction with echo and perfusion assessment. Minimum flow and trial duration follow the circuit protocol
V-V ECMO	Sweep gas	Reduce sweep while maintaining safe circuit blood flow. Assess gas exchange, work of breathing and ventilator requirements

A

IABP

ratio or volume

Wean use the selected ratio or volume-reduction protocol. Reassess perfusion and congestion without simply escalating inotropes to compensate.
Failed wean worsening perfusion, hypotension, congestion or falling forward flow at reduced augmentation. Use individualized stop criteria.
Who fails iCM and low PAPi were associated with deterioration in a chronic-HF shock cohort. This is a risk association, not a universal weaning cutoff or a device-selection rule.

B

Impella

P-level wean

Reduce P-level in protocol-defined steps. Record the actual device, estimated flow, drugs and loading conditions.
Assess perfusion, LV/RV function, filling pressures, rhythm and device position. Avoid prolonged unsafe low support.
Interpret a P-level does not specify one universal flow. Tolerating a low setting supports further assessment. It does not alone authorize removal.

C

VA-ECMO & V-V ECMO

flow vs sweep

VA-ECMO (cardiac)

Reduce flow in supervised stages. Reassess perfusion, LV/RV performance, filling pressures and gas exchange at each step.
Advanced trial-off maneuvers, including PCRTO, require a trained ECMO team and an explicit circuit-specific protocol.

V-V ECMO (respiratory)

Reduce oxygenator FDO₂ and sweep gas toward off while maintaining safe circuit blood flow. Confirm acceptable gas exchange and effort on lung-protective ventilator support.
Confirm sustained acceptable gas exchange and respiratory effort on an agreed sweep-off observation period with lung-protective settings.

D

Signs the trial is failing → abort

resume, retry later

Hemodynamic ↓ MAP, rising PCWP / RA, falling CI or venous saturation relative to the patient's baseline.
Metabolic rising lactate, falling urine output.

Electrical / echo new arrhythmia, ventricular distension, worsening MR or evidence of stasis. Interpret aortic-valve opening in the device/unloading context.
Action abort, return to support, fix the cause, retry at the lowest tolerated level later.

E

Combined support (ECpella) & venting

order of weaning

On ECMO + Impella ("ECpella") for LV venting, an often-used strategy is to reduce ECMO flow first while maintaining LV unloading, then assess the Impella. The sequence depends on RV, LV and respiratory recovery. As VA-ECMO flow falls, assess LV size, filling pressures, pulmonary edema, root/LV stasis and perfusion. A closed aortic valve during effective Impella unloading is not by itself a failed wean. Interpret opening alongside washout and device flows.

F

RV support & TandemHeart

don't forget the right side

RV support (Impella RP / ProtekDuo)

Wean by reducing flow / P-level. Integrate RV size/function, CVP, perfusion and LV filling at reduced support. TAPSE, RV S′ and PAPi are load-dependent adjuncts, not universal removal cutoffs.

TandemHeart (LA → femoral artery)

Reduce support using the actual pump/cannula protocol. Observe BP, perfusion and filling pressures. Plan arterial closure and management of the transseptal cannula with the implanting team.

Test native reserve: during circulatory-support reduction, assess perfusion and congestion. If inadequate, restore support, reassess the cause and name the next destination. V-V ECMO asks a separate question about respiratory reserve.

Removal is a planned procedure, not an afterthought: manage anticoagulation and choose manual, percutaneous-closure, or surgical repair according to access site, cannula/device, anatomy, and local expertise. Have vascular/surgical backup for large-bore arterial access.

A

Before you pull

anticoag + team

Anticoagulation plan interruption, testing and resumption for the actual drug, access and closure method. There is no universal ACT or hold time for every device.
Team coordinate cardiology + surgery + perfusion. Position the patient. Type & screen.

Hemostasis plan choose manual compression, percutaneous closure or surgical repair using access anatomy, device, bleeding risk and expertise.
Backup vascular surgery on standby for large-bore arterial sites.

B

Identify the device and access before removal

use the actual implant record

Device	What to confirm
IABP	Femoral or upper-body access, catheter/sheath and balloon size. It augments native ejection without a pump-flow lumen.
Impella CP / 5.5	Exact model, access/sheath or surgical graft, insertion record and current IFU. A displayed P-level does not specify one fixed flow or removal technique.
TandemHeart	Transseptal LA drainage and arterial return cannulae, residual atrial communication and the arterial closure plan.
VA-ECMO	Drainage, arterial return and distal-perfusion cannulae. Record actual sizes, sites and any graft. Nominal pump capacity is not native cardiac output.

C

Removal by device

technique

Device	How it comes out
IABP	Manual, percutaneous closure or surgical removal according to access, sheath, vascular disease and local expertise.
Impella 2.5 / CP	Remove using the model-specific instructions and imaging/access plan. Coordinate pump changes and arterial closure with the trained implanting team.
Graft-based Impella	Plan surgical device removal and graft management with the implanting team. Do not apply a femoral CP removal sequence to a 5.5.
VA-ECMO	Planned percutaneous closure or surgical repair according to cannulation and vascular anatomy. Assess bleeding, limb perfusion and venous thrombosis when indicated.

D

Watch after removal

access-site complications

Bleeding site hematoma, retroperitoneal bleed (high stick) — serial Hgb, abdo/flank exam.
Limb check distal pulses / perfusion — limb ischemia, compartment syndrome.

Vascular pseudoaneurysm, AV fistula, dissection — duplex for a new bruit / pulsatile mass.
Reaccumulation watch for recurrent shock — a failed wean can declare late.

E

After the device — recovery pathway

don't stop at explant

Recovered HF optimize/uptitrate GDMT, cardiac rehab, serial echo to confirm sustained recovery.
Treat the cause address the established cause. Revascularization or disease-specific myocarditis therapy requires its own indication.

Still dependent → durable LVAD or transplant workup. If not a candidate → goals-of-care / palliation.
Watch late a fraction re-decompensate — keep follow-up tight after step-down.

E2

Removal-day checklist

at the bedside

Before confirm the agreed anticoagulant interruption and coagulation-testing plan. Surgical/vascular backup + closure devices ready. Ultrasound available.

After confirm hemostasis and frequent distal neurovascular checks at the agreed procedure-specific intervals. Follow Hgb, recurrent shock and access restrictions.

Large-bore arterial decannulation is where the late catastrophes hide — a high femoral puncture with retroperitoneal bleeding, an unrecognized ischemic limb, a pseudoaneurysm. Image a new bruit or pulsatile mass, and keep vascular surgery a phone call away.

Removal is a separate decision: establish sustained tolerance of reduced support, confirm the device-specific procedure and rescue plan, and watch for recurrent shock and access complications afterward. Persistent dependence requires reassessment of recovery and candidacy, not automatic conversion to durable support.

Name the destination while stabilizing the patient. Recovery assessment and LVAD/transplant evaluation can proceed together. Waiting for repeated failed weans before starting candidacy review may allow organ injury or device complications to narrow the options.

A

What today’s support actually tells you

trajectory over a snapshot

Observation	Interpretation and next decision
Normal lactate on support	Reassuring under the current support, but not proof of adequate tissue perfusion or native recovery. Integrate hemodynamics and organ function. Record device settings and drug doses.
Failed reduction	Dependence exists at that level and time. Look for residual ischemia, congestion, arrhythmia, infection, loading abnormalities and RV failure. Restore support and define what must improve before retesting.
Partial recovery	A patient with chronic iCM may return to a compensated baseline with a persistently low EF. Adequate unsupported perfusion and manageable congestion matter more than EF normalization.
Worsening trajectory	Escalating vasoactive needs, recurrent organ injury or device complications require renewed shock-team assessment and an explicit escalation or goals-of-care decision.

B

Choose a feasible destination

clinical synthesis

Destination	What has to be plausible
Recovery / medical therapy	Sustained perfusion and acceptable congestion with support removed. Cause treated. Tolerated HF therapy and follow-up plan.
Revascularization	Treatable coronary disease plausibly contributing to dysfunction, suitable targets and acceptable procedural risk. CABG can occur on support. Complete prior weaning is not a universal prerequisite. Agree on support and fallback plans.
Durable LVAD	Predominant LV failure with sufficient RV reserve or a right-sided support strategy. Reversible organ dysfunction, acceptable operative risk and feasible long-term care.
Transplant	Advanced-HF indication, acceptable neurological/organ prognosis and candidacy, with a feasible bridge while awaiting an organ.
Comfort-focused care	When recovery and durable options are infeasible or inconsistent with informed patient goals, align treatment with those goals and relieve symptoms.

For iCM, review shock phenotype, supported revascularization and durable-exit evidence. No single scar, EF, CI or supported-day threshold allocates CABG versus LVAD/transplant.

C

Keep evidence tied to its population

what remains uncertain

ALTSHOCK-2: randomized HF-CS evidence exists. Routine early IABP did not significantly improve 60-day survival or successful heart-replacement bridging. This does not prohibit individualized rescue use or decide CABG versus LVAD.
Bridge-to-LVAD registries: reported survival begins among patients who reached implantation. It excludes those who died or recovered beforehand and cannot rank the causal benefit of bridge devices.

Daily decision: What is improving? What still requires support? What is the next test or intervention? What is the fallback if recovery fails?