End-Stage Heart Failure with Multiple Intracardiac Thrombi

Abstract

The use of ventricular assist devices as a bridge to transplantation has become a widely used option for patients with end-stage heart failure. In contrast to total artificial hearts, ventricular assist devices support the failing heart by bypassing one or both ventricles. In certain cases (myocardial tumors, graft failure, transplant rejection, endocarditis, intracardiac thrombus formation), however, it may be advantageous to excise the heart and replace it with an artificial device. Total artificial hearts are intracorporeal devices designed for this purpose. Unfortunately, some patients are too small or are, for other reasons, ineligible for a total artificial heart.

We describe the case of a 55-year-old woman who had ischemic cardiomyopathy and thrombus formation in all 4 cardiac chambers. To reduce the risk of thromboembolic events, we elected to replace her heart completely with 2 extracorporeal ventricular assist devices. The heart was excised via a median sternotomy approach, and the outflow cannulae (from device to patient) were connected to both atrial remnants. The 2 inflow cannulae (from patient to device) were anastomosed end-to-end to the aorta and the pulmonary artery, respectively. After attaining a flow of more than 5 L, the 2 extracorporeal assist devices effectively and efficiently performed the work of the native heart. Thus re-established, organ perfusion was improved by this mechanically driven circulation, as signified by an initial decrease in creatinine and blood urea nitrogen levels. The patient, however, did not recover from postoperative neurological dysfunction and died of respiratory insufficiency and multiple-organ failure on the 26th postoperative day.

End-stage heart failure is associated with a poor prognosis and a high morbidity rate. Many recent advances in pharmacologic treatment have been made, and various surgical approaches have been developed with the aim of improving cardiac function; however, the mortality rates associated with cardiac failure remain high. Intracardiac thrombus formation as a consequence of impaired blood flow within the 4 chambers of the heart is a severe complication in patients with end-stage cardiac failure. Depending on the size and age of the thrombus, it can be removed by means of pharmaceutical thrombolysis or surgical thrombectomy. In certain cases (myocardial tumors, graft failure, transplant rejection, endocarditis, intracardiac thrombus formation), however, it may be advantageous to excise the heart and replace it with an artificial device. Total artificial hearts are intracorporeal devices designed for this purpose.

Several methods have been used successfully to resolve cardiac insufficiency as an underlying disease: mechanical circulatory support via intraaortic balloon counterpulsation; via right, left, and biventricular assist devices (VADs); and, more recently, via total artificial hearts (TAHs). A variety of VADs are currently available, and they provide promising options for patients with end-stage heart failure—especially when implemented in a timely fashion as bridges to transplantation. Total artificial hearts have been used both as temporary and as permanent devices in patients with cardiac failure, acute heart graft rejection, myocardial infarction, and congenital heart disease. Total artificial heart pumps, such as the CardioWest C-70® (SynCardia Systems, Inc.; Tucson, Ariz) and the AbioCor® (ABIOMED, Inc.; Danvers, Mass), are too large to be completely implanted in children or in other patients with body surfaces smaller than 1.7 m². Extracorporeal devices can be used for biventricular assistance in these smaller patients; however, reports of complete replacement of the heart with an external biventricular assist device are rare, and none of those operations were performed as a consequence of thrombosis.

Case Report

In January 2003, a 55-year-old woman presented at our institution with multiple cardiac risk factors, including a 10-year history of hypertension, diabetes mellitus, and obesity. Two years earlier, she had experienced a cerebrovascular accident with no residual deficit. She was admitted to the hospital with fatigue, congestive heart failure, and renal insufficiency. An electrocardiogram showed the patient to be in sinus rhythm. The creatinine was 2.8 mg/dL, and the blood urea nitrogen (BUN) was 80 mg/dL. Heart catheterization, which was performed 2 days after admission, revealed a 50% proximal, 60% mid, and 50% distal left anterior descending coronary artery stenosis, and a 100% mid right coronary artery occlusion with insufficient collateral run-off. Transesophageal echocardiography showed severely depressed left and right ventricular function with an ejection fraction of less than 0.10 and a patent foramen ovale. In addition, it revealed thrombus formation in all 4 cardiac chambers. The patient had no knowledge of any previous thrombotic complications or coagulopathies in her family. Over a period of 2 weeks, she began to experience increasingly severe symptoms of congestive heart failure, with shortness of breath, generalized edema, and hypotensive episodes. The symptoms were not alleviated by aggressive inotropic medical management that included milrinone and dobutamine. Because the patient's condition was rapidly deteriorating, an in-tra-aortic balloon pump was placed. The pump, however, did not improve her condition. The existence of thrombi in all 4 cardiac chambers led us to consider complete replacement of the dilated and ischemic heart with 2 Thoratec® VADs (Thoratec Corporation; Pleasanton, Calif). The underlying rationale for such an operation was that removal of the heart would reduce the likelihood of thromboembolic events, and the patient would remain a candidate for heart transplantation.

Surgical Technique

For surgery, the patient was placed in the supine position. After median sternotomy, the pericardium was opened, and a massively enlarged heart was found. Sutures were placed for separate cannulation of the superior and inferior venae cava and the aorta, heparin was administered, and extracorporeal circulation was initiated. The aorta was cross-clamped and the heart excised in accordance with our heart transplant protocol (Fig. 1). Upon opening the right atrium, we discovered copious amounts of thrombus. The size of the thrombi (13 mm × 21 mm) correlated with our findings on the preoperative transesophageal echocardiogram. There was less thrombotic material in the left atrium. The left ventricle contained a single large thrombus.

Fig. 1 The heart is excised using the standard cardiac transplantation technique.

To extend both atrial remnants, we cut a 32-mm Hemashield Gold™ vascular prosthesis (Boston Scientific Corp.; Natick, Mass) in half and anastomosed it to the right and left atria, respectively, with running sutures (Fig. 2). The use of the Hemashield graft as an adapter has been described previously¹ as a means to overcome the porosity of the VAD outflow conduit. Next, the 2 Thoratec systems were tunneled through the abdominal wall, and the outflow cannulae (from device to patient) were placed in the Hemashield grafts for later tightening. The inflow cannulae (from patient to device) were also tunneled next to the outflow cannula and anastomosed end-to-end with 5-0 running Prolene sutures to the pulmonary artery and the aorta, respectively (Figs. 3 and 4). All sutures were reinforced with BioGlue® surgical adhesive (CyroLife, Inc.; Kennesaw, Ga), and the chest was closed with wires after hemostasis was achieved.

Fig. 2 The left and right atrial remnants are extended by using 2 halves of a 32-mm Hemashield Gold™ vascular prosthesis.

Fig. 3 After the atria are connected to the outflow cannulae (from device to patient), the inflow conduits (from patient to device) of both the right and left Thoratec devices are anastomosed to the pulmonary artery and the aorta, respectively.

Fig. 4 Final positioning of the 2 Thoratec devices is shown for complete replacement of the native heart.

The patient was returned to the surgical intensive care unit in stable cardiovascular condition. During the following weeks, however, her neurological condition did not improve. She could not be weaned from the ventilator and died of multiple-organ and respiratory failure on the 26th postoperative day. Postmortem examination revealed global encephalopathy without evidence of circumscribed stroke.

Discussion

Implantable or paracorporeal cardiac assist devices are currently used for treatment of severe heart failure. For patients with medically refractory heart failure, clinical trials have shown greater benefit and survival with VADs than with optimal medical therapy.² The 2 major purposes for VAD support are to serve as a bridge to transplant or to relieve a failing heart for a limited time in order to allow recovery from certain conditions (for example, myocarditis and postcardiotomy shock). In instances of severe global insufficiency, standard criteria for biventricular device implantation have been identified with regard to cardiac index (>2.0 L/min/m²), central venous pressure (<26 mmHg), and high-dose inotropic treatment (≥2 inotropic agents).³

As stated above, complete replacement of the heart by paracorporeal assist devices has rarely been published. The 1st such case mentioned in the literature was that of a 26-year-old man with aortic valve endocarditis who underwent replacement of the valve with implantation of an aortic conduit. Three months later, he had a relapse of endocarditis, with severe aortic regurgitation, abscesses in the interventricular septum, and end-stage heart failure (left ventricular ejection fraction, <0.20). To eliminate the septic focus, the patient's heart was excised completely, and an extracorporeal TAH (in this case, a Thoratec VAD) was used as a bridge to transplant. Transplantation was successfully accomplished 5 days later, and the patient was eventually discharged from the hospital.⁴ In addition, the same group treated 2 patients for primary graft failure who had been re-transplanted successfully after excision of the graft and bridging with an extracorporeal TAH (again, with a Thoratec VAD).⁴ Recently, that group also published the only available case report describing complete replacement of the heart with an extracorporeal biventricular assist device in a patient with chronic graft rejection, thereby obviating the use of immunosuppressive therapy.⁵

The patient described herein fulfilled the transplantation criteria with respect to decreased cardiac index and resistance to high-dose inotropic medication. The patient had a body surface area above 1.7m². Nonetheless, we decided not to use an intracorporeal TAH for replacement, because the chest diameter was rather small, and the body mass index of 29 suggested that the surface area was most likely increased by obesity. In addition, the patient was experiencing rapidly worsening ischemic and dilated cardiomyopathy, which had already led to thrombi formation within the cardiac chambers. The renal function improved during the 1st weeks after device implantation. Creatinine levels dropped from 2.8 mg/dL to 0.9 mg/dL, and BUN levels dropped from 80 mg/dL to 19 mg/dL.

Little has been published on the treatment of heart failure in combination with intracardiac thrombi. Several patients in the existing reports were diagnosed with antiphospholipid syndrome or other coagulopathies before or after intracardiac clot detection. However, in this patient, no primary coagulation disorders could be ascertained. Pathologic examination of the dissected heart revealed mural thrombi in both ventricles, which were associated with transmural infarct areas. Infarction scars were located on the lower right and upper left free wall and measured 4.5 cm and 6 cm in size, respectively. Furthermore, both ventricles contained several thrombi, some of which were loose in the cavity. The mitral valve showed calcification, degeneration, and fibrous thickening, with foci of plasma cells. Coronary arteries had extensive intimal fibrosis and medial calcification with up to 50% stenosis. The heart tissue itself revealed focal myocyte hypertrophy with mild interstitial inflammation. No organism or neoplastic process was observed.

Most of the reports concerning the adequate treatment of intracardiac thrombi have involved premature neonates.⁶ Thrombolysis with either urokinase or tissue plasminogen activator has been recommended as the 1st-choice treatment for intracardiac thrombosis in infants and adults.⁷ In some cases, surgery was required; however, these patients were without congestive heart failure.⁸

In the present case, we began heparin administration immediately upon admission, with an intrave-nous bolus followed by 1,150 U per hour. The hepa-rinization may have precipitated the development of a retroperitoneal hematoma and prevented any further anticoagulation; therefore, we decided to forgo warfarin therapy, which had been planned originally. Although titers for heparin-dependent antibodies were inconclusive, prophylactic anticoagulation with argatroban was initiated. Subsequent excision of the heart allowed us to restrict antithrombotic medication to argatroban alone after biventricular Thoratec implantation.

As is the case with most survival procedures, replacement of the heart with biventricular Thoratec devices as a bridge to transplant is undoubtedly associated with a high incidence of complications and death. Nevertheless, patients such as the one described here are terminally ill and may not be able to undergo transplantation soon enough, due to the restricted donor supply. Successful biventricular Thoratec implantation for cardiac assistance has been demonstrated⁹ and offers a viable treatment option for patients with end-stage congestive heart failure and compounding complications. Total replacement of the heart with a biventricular mechanical device should be reserved as a last resort for those unique patients who have no other treatment options.