Titanium Plug Closure after HeartWare Ventricular Assist Device Explantation in a 15-Year-Old Girl: First U.S. Experience

Case Report

A 14-year-old girl (weight, 51 kg; body surface area, 1.4 m²) with a history of acute myeloid leukemia developed severe congestive heart failure 3 months after the termination of chemotherapy. Her left ventricular (LV) ejection fraction (EF) was 0.10, her right ventricular systolic function was moderately depressed, and her brain natriuretic peptide level was 1,900 pg/mL. She could not be weaned from inotropic agents. Support from a VAD was deemed necessary as a bridge to candidacy for transplantation, because she had been in remission for only 3 months.

In January 2013, the patient underwent HeartWare HVAD implantation with use of a modified technique—necessitating a median sternotomy—wherein the VAD was placed in an infradiaphragmatic position to lessen compression of the lower lobe of the left lung.¹⁹ She recovered uneventfully and was discharged from the hospital 11 days later. Her medications included enalapril, nadolol, furosemide twice daily, and a stable warfarin dose.

The patient's heart began to show signs of functional recovery after 3 months of VAD support: the LVEF returned to normal, and the LV diastolic dimension progressively decreased. We placed her on a VAD-weaning protocol developed at our center,²⁰ and she tolerated 2 pump-weaning tests well. Although device explantation was thought to be feasible, uncertainty remained about the sustainability of the patient's cardiac function: LV histologic results had shown moderate myocyte hypertrophy, mild-to-moderate interstitial fibrosis, degenerative myocytolysis, and decreased sarcomere density. In view of possibly recurrent heart failure, cancer, or both, we thought it best to explant as much foreign material as possible from the body while still enabling later reinstitution of VAD support.

The patient's young age and female sex influenced our decision to perform a repeat sternotomy, enabling the removal of the entire device. Thoracotomy would have necessitated making an additional operative incision and leaving the outflow-graft component in place.

In January 2014, 12 months after VAD implantation, the patient (now 15 years old) underwent explantation via repeat median sternotomy with cardiopulmonary bypass support. We removed a small mural thrombus from the LV cavity near the cannulation site. We used the titanium plug (Fig. 1) to cover the LV coring site. The plug—manufactured for this specific use—is made of a titanium alloy with titanium microsphere sintering, similar to the materials used in the HVAD itself. The diameter exactly matches that of the sewing ring. The center of the plug is hollow, to reduce weight. A special plug-holder device is supplied by the manufacturer to help position the plug during surgery.¹⁸

View Figure

• Download as Powerpoint
• Download Figure

The patient was weaned from cardiopulmonary bypass without difficulty. Her biventricular systolic function remained normal, with an LVEF of 0.63 and normal LV dimensions. She recovered uneventfully and was discharged from the hospital on postoperative day 11 with instructions to take enalapril, nadolol, warfarin, and aspirin. She took warfarin and aspirin for 3 months and continued thereafter only with aspirin for anticoagulation. Post-explantation histologic results revealed improvement of the degenerative sarcomeric changes that had been seen previously. As of January 2017, the patient was doing well without recurrence of heart failure.

Discussion

Myocardial recovery during long-term VAD support is infrequently reported in children, perhaps because of their substantially shorter waiting times on the transplant list in comparison with adults.²¹ However, as the waiting time in children increases, myocardial recovery might occur in mechanically supported children.²² In 2015, we reported that the pediatric myocardial response to VAD therapy differs from that in the adult myocardium at a subcellular level.²³ Improved understanding of such differences might help to identify candidates for bridge-to-recovery therapy.

Our patient had chronic anthracycline cardiomyopathy, which is generally thought to be irreversible.²⁴ However, several examples of reversed acute and chronic anthracycline cardiotoxicity have been published.^25–27 Although we were encouraged by our patient's seemingly complete recovery of systolic ventricular function and improved histologic results, the natural history of this disease and her initial histologic profile increase the risk of a future decline in cardiac function.

The most extreme method of VAD explantation involves device removal via sternotomy. Least extreme is simple ligation of the outflow graft, which functionally excludes the device from the circulation but leaves it in place.²⁸ Our removal of everything but the sewing ring obviated potential problems.

In current practice, the coring site can be plicated after removing the ring, but this deforms the LV geometry and makes VAD reinsertion far more technically demanding.²⁹ Another option is to create a felt plug and insert it into the ring. The felt plug obviates the need for LV plication, thus sparing further injury to the myocardium and distortion of the ventricular geometry. We have used the latter technique at our institution.²⁰

The titanium plug's material properties and engineering design make it superior to the felt plug for insertion into the sewing ring. The result is a perfect fit, a hermetically closed hole in the ventricle, and a surface that will not promote thrombus formation.¹⁸ The German investigators who originally used the titanium plug found no visible thrombus 3 years after its placement.³⁰ To our knowledge, our report is the first and, so far, only description of the titanium plug's use in the U.S.