Berlin Heart EXCOR #3: Implantation—Outflow

Outflow creation is the systemic egress of the EXCOR circuit. The operative objective is to construct an outflow pathway that is hemodynamically efficient, mechanically durable, and anatomically safe—a balance that becomes most demanding in small infants, where the ascending aorta offers limited clamp “real estate” and minor geometric distortion can have outsized consequences. Technical reports emphasize that outflow success is driven less by a single “best” configuration than by disciplined control of geometry, hemostasis, and embolic risk at every junction point. [1, 2]

1) Core concepts (what must be achieved)

1.1 Low-resistance systemic outflow

• Prioritize a smooth flow path and avoid iatrogenic afterload: stenosis at the aorta–graft junction or cannula interface can reduce effective systemic delivery. [1]
• In infants, mediastinal constraints make the outflow uniquely susceptible to kinking and compression, particularly after sternal approximation or with patient growth/edema. [1]

1.2 Secure anastomosis with predictable geometry

• The most reproducible approach is a vascular graft sewn end-to-side to the ascending aorta, then connected to the EXCOR outflow cannula. This provides controllability of trajectory and simplifies “tuning” of the cannula course. [1]

1.3 Avoid coronary compromise (RCA vulnerability as a geometry problem)

• Coronary injury is not only direct trauma. A clinically relevant mechanism is geometric distortion—from clamp traction, graft lie, or anastomosis placement that alters the relationship between the aorta and the coronary origin region. [1]
• Therefore, clamp choice and landing zone selection should be treated as a coronary-protection maneuver, not merely a hemostatic step. [1]

2) Outflow options (what you can do)

Option A — Direct cannulation of the ascending aorta (selected cases)

• Feasible when anatomy and working space are favorable, but technical descriptions in small children tend to favor methods that improve hemostasis and geometric control at the aortic interface. [1]

Option B — End-to-side aorta–graft anastomosis (most common)

A vascular graft is sewn end-to-side to the ascending aorta; the outflow cannula interfaces with the graft.

Advantages

• Trajectory control: easier to keep the course short, straight, and away from compressive vectors. [1]
• Field management: improved access for hemostasis and de-airing at a discrete suture line. [1]
• Re-entry / future procedures: modified strategies (e.g., preserving grafts used for bypass or creating extensions) have been described to facilitate later transplant/explant workflows. [2]

3) Clamp strategy: side-biting clamp vs aortic cross-clamp

3.1 Beating-heart technique with side-biting (tangential) clamp

Preferred when the ascending aorta is long enough to safely accommodate a partial clamp.

Key steps / pearls

• Use a tangential clamp, then perform an aortotomy sized precisely to the graft/cannula interface to avoid purse-string narrowing or ovalization. [1]
• Select an anastomosis site that:
• avoids the coronary origin region, and
• preserves space for other cannulation sites and future re-entry considerations. [1]

Primary risk: coronary distortion/compression

• The beating-heart partial clamp can transmit traction and alter aortic geometry; the hazard is often positional, not lacerative. Treat the clamp as a “geometry determinant.” [1]

3.2 Arrest technique with aortic cross-clamp (small infants / short ascending aorta)

When the ascending aorta is too short for a safe side-biting clamp, cross-clamp with cardioplegic arrest may be required to create a controlled field for a secure, reproducible anastomosis. [1]

Why this matters in small patients

• Limited clamp landing zone
• High consequence of subtle distortion
• Need for meticulous hemostasis + de-airing under direct control (systemic embolic penalty is high). [1]

4) Technical priorities for the aorta–graft anastomosis

4.1 Geometry: “short, straight, and non-compressive”

Aim for a graft course that is:

• Not under tension (prevents traction on the suture line and reduces late bleeding) [1]
• Not redundant (reduces torsion/kinking risk) [1]
• Not positioned for sternal compression after closure or during later operations [1]

4.2 Anastomosis quality: hemostasis + lumen preservation

• Maintain a round, non-stenotic anastomosis; even modest narrowing can create clinically meaningful pump afterload. [1]
• Handle the infant aorta gently; clamp-related injury can become a persistent bleeding problem in a population already at high hemorrhagic risk. Large series report substantial re-exploration rates for bleeding during EXCOR support. [5]

4.3 De-airing: treat it as a systemic embolic-risk step

• Build a deliberate workflow that prevents air entry at:
• the graft,
• the connector/cannula interface,
• and any transition where negative pressure could entrain air. [1]

Given that neurologic injury remains a defining adverse event domain for EXCOR, air management deserves “no-compromise” rigor. [3, 4]

5) Aortic insufficiency: evaluate early, treat deliberately

5.1 Why AI is not a minor echocardiographic detail on LVAD support

Significant AI creates a closed-loop circuit:

• LVAD delivers to aorta → regurgitates into LV → re-enters LVAD inflow

This leads to ineffective systemic output, LV distension, and impaired end-organ perfusion despite apparently adequate pump flows.

5.2 What recent experience emphasizes (infants on durable support)

• Severe AI can develop or progress during durable mechanical support in infants and may require high-risk surgical intervention; recent reports highlight that this is not theoretical and can become a practical determinant of support success. [7]

5.3 Practical implications at implantation (a checklist mindset)

• Define AI severity and mechanism (leaflet pathology vs root geometry vs valve non-opening physiology).
• Anticipate whether planned unloading is likely to keep the aortic valve persistently closed—an environment that may predispose to progressive regurgitation over time. [7]
• If AI is clinically significant, address it proactively according to anatomy and institutional strategy, because untreated AI can negate the intended systemic support.

6) High-yield operative “failure modes” (what to actively prevent)

6.1 Neurologic injury (dominant morbidity domain)

• In the multicenter IDE cohort, 29% experienced ≥1 neurologic event and 21% experienced stroke, with many events occurring early after implantation. [3]
• Across published series, systematic review data show a wide stroke range (5%–47%), underscoring heterogeneity in populations and management practices. [4]
• Contemporary collaborative experience demonstrates meaningful improvement: post-approval data reported reduced stroke incidence and reductions in other adverse events (including major bleeding), coinciding with evolving anticoagulation and care practices. [6]

6.2 Major bleeding and re-exploration

• Large single-center longitudinal experience reported re-exploration for bleeding in 22 of 122 patients, highlighting bleeding as a recurring operational problem in pediatric durable support. [5]

6.3 Outflow-specific technical hazards (prevention targets)

• Coronary compromise from clamp traction or graft lie (geometry-driven risk). [1]
• Outflow graft kinking/compression due to mediastinal constraints, graft length mismatch, or sternal effects. [1]
• Aortic anastomotic stenosis causing pump afterload and reduced systemic delivery. [1]
• Progressive AI producing recirculation and inadequate systemic output during LVAD support. [7]

References

[1] Nguyen K. A technique for implanting outflow cannulas for Berlin Heart EXCOR ventricular assist device in small pediatric patients. J Thorac Cardiovasc Surg. 2011;142(1):223-224.

[2] Botha P, Hasan A, Perri G, Filippelli S, Griselli M. Modified technique for the implantation of berlin heart excor ventricular assist device in children. World J Pediatr Congenit Heart Surg. 2012;3(3):373-377.

[3] Jordan LC, Ichord RN, Reinhartz O, et al. Neurological complications and outcomes in the Berlin Heart EXCOR pediatric investigational device exemption trial. J Am Heart Assoc. 2015;4(1):e001429.

[4] Rohde S, Antonides CFJ, Dalinghaus M, Muslem R, Bogers AJJC. Clinical outcomes of paediatric patients supported by the Berlin Heart EXCOR: a systematic review. Eur J Cardiothorac Surg. 2019;56(5):830-839.

[5] Hetzer R, Kaufmann F, Delmo Walter EM. Paediatric mechanical circulatory support with Berlin Heart EXCOR: development and outcome of a 23-year experience. Eur J Cardiothorac Surg. 2016;50(2):203-210.

[6] Zafar F, Villa CR, Morales DLS, et al. Berlin Heart EXCOR and ACTION post-approval surveillance study report. J Heart Lung Transplant. 2021;40(4):251-259.

[7] Konstantinov IE, Brizard CP, Davies B. Severe aortic valve insufficiency in infants on durable ventricular assist device support. J Thorac Cardiovasc Surg. 2024;168(3):957-960.