Portal de Periódicos da CAPES

Berlin Heart as a Bridge to Recovery for a Failing Fontan

2009; Elsevier BV; Volume: 87; Issue: 3 Linguagem: Inglês

10.1016/j.athoracsur.2008.07.086

1552-6259

Marcelo Cardarelli, Mubbada Salim, Jon Love, Shari Simone, Jamie Tumulty, Dyana Conway, Bartley P. Griffith,

Cardiac Structural Anomalies and Repair

We report the temporary use of a Berlin Heart ventricular assist device (Berlin Heart AG, Berlin, Germany) for cardiac support of an 18-month-old girl with rapidly progressive ventricular failure after completion of a fenestrated Fontan. After 6 months of cardiac assistance with a single pneumatic pump, catheterization data showed improvement of the ventricular function and the ventricular assist device was successfully removed. A follow-up echocardiogram 6 months after hospital discharge demonstrated marked improvement of ventricular function. We report the temporary use of a Berlin Heart ventricular assist device (Berlin Heart AG, Berlin, Germany) for cardiac support of an 18-month-old girl with rapidly progressive ventricular failure after completion of a fenestrated Fontan. After 6 months of cardiac assistance with a single pneumatic pump, catheterization data showed improvement of the ventricular function and the ventricular assist device was successfully removed. A follow-up echocardiogram 6 months after hospital discharge demonstrated marked improvement of ventricular function. The use of a ventricular assist device (VAD) as a temporary bridge to cardiac remodeling and recovery has been documented in the adult and pediatric population with a two-ventricle system. Unresponsive heart failure in children with single ventricle physiology only has heart transplantation as the commonly accepted, seldom implemented therapeutic choice. This case opens a window into the potential use of VAD as a bridge to recovery for a special group of patients. An 18-month-old woman diagnosed with hypoplastic left heart syndrome was admitted while extubated into the intensive care unit after completion of a lateral tunnel Fontan with a 5-mm fenestration. From her original diagnosis she had been palliated with a Sano procedure followed 6 months later by a bidirectional Glenn. After a rapid surgical recovery with good peripheral saturations (86% to 92%) on postoperative day 10 while still on the ward due to poor feeding, she had bilateral pleural effusions develop and abdominal distension, requiring transfer to the intensive care unit, chest drainage, and placement of a central line (initial central venous pressure, 23 mm Hg). Treatment with gentamicin, cefepime hydrochloride, and Diflucan (fluconazole; Pfizer Inc, New York, NY) was started for a positive urine culture for Enterobacter or Candida Albicans. After 7 days, acute renal failure (BUN, 123 mg/dL and creatinine, 5.7 mg/dL) developed in the patient and she had significant metabolic acidosis and respiratory distress requiring reintubation and mechanical ventilation. A femoral hemodialysis catheter was placed, but due to hemodynamic instability, the original hemodialysis system was transformed into venous-venous hemofiltration. Although on hemofiltration, heparin therapy was found to be ineffective and a workup revealed anti-thrombin III deficiency. A venous duplex and a perfusion–ventilation scan was ordered for a drop in arterial blood saturations from 77.9 to 58.5, which resulted in a diagnosis of a deep venous thrombosis and a right pulmonary embolism. Treatment with tissue plasminogen activator was initiated, which effectively dissolved most of the thrombus. Fibrinolytic therapy was discontinued due to a life-threatening pericardial tamponade that required emergency needle-evacuation and an abnormal neurologic examination due to an intracranial hemorrhage. Lipase level peaked at 1,498 I/U with no attributable surgical cause found. On postoperative day 43, a cardiac catheterization revealed elevated end-diastolic pressures. Therapy was initiated at this time to maximize afterload reduction and improve diuresis. Figure 1 shows a chart with the evolution of hemodynamic, weight, and inotropic score throughout admission. On postoperative day 55, the patient underwent a tracheostomy, and a second cardiac catheterization on postoperative day 56 confirmed previous findings and a previously unknown gradient (30 mm Hg) across the corrected hypoplastic arch and descending aorta, along with newly developed multiple aortopulmonary collaterals. Placement of an aortic stent reduced the gradient to 10 mm Hg and all major collaterals were successfully coiled. Ventilator support was successfully weaned to tracheostomy collar over the ensuing days. A third cardiac catheterization on postoperative day 90 revealed the development of a new set of collaterals, which were appropriately addressed. Inotropic and mechanical ventilation dependence prompted the decision to implant an assist device while entertaining the idea of a heart transplant, but never formally listing the patient. Approval from the insurance, Food and Drug Administration paperwork, and importation of the device delayed the process for more than 3 weeks to postoperative day 136. The VAD inflow cannula was placed in the apex of the single ventricle, the outlet cannula in the ascending aorta, and a single, 25-cc pumping bladder was used. In the immediate postoperative period, vasopressin was added for hypotension, which was weaned during the course of 3 days. Diuretic therapy was resumed with the addition of Coreg (carvedilol; GlaxoSmithKline plc, London, UK). After the first 24 hours, a heparin drip was started and transitioned into sodium warfarin once the enteral feeds were resumed. International normalized ratios for clotting were checked daily, and low molecular heparin was added when nontherapeutic levels were suspected. Aspirin (Bayer, Leverkusen, Germany) and Dipyridamole (Rising Pharmaceuticals, Allendale, NJ) were administered daily. Weekly echocardiograms suggested incomplete ventricular emptying with absence of clots in the inflow cannula. Three weeks into VAD support, the decision was made to exchange the bladder for a larger one (30 mL) at the bedside. The patient's evolution continued to be assessed by weekly cardiac ultrasounds, daily coagulation profile, VAD bladder assessment to rule out clot formation, and overall patient clinical status and weight gain. After 5 months of support, and with the VAD placed on standby for 5 minutes, a cardiac catheterization revealed a lower wedge (16 mm Hg) and pulmonary pressure (20 mm Hg). Confronted with the reality that our patient was never an ideal transplant candidate and after discussion with her parents, the decision was made to optimize her general status and move toward removal of the VAD. After 179 days of cardiac support, she was taken to the operating room and the VAD was removed with the use of cardiopulmonary bypass through femoral cannulation. The outflow cannula site was primarily closed with reinforced sutures. Extended inotropic support was resumed for the following 41 days. A newly formed thrombus extending into the inferior cava was discovered shortly after VAD removal and was treated with heparin and fibrinolytics. Ascites, secondary to portal hypertension, resulted in abdominal wound dehiscence at the skin site of cannula entry and exit. Temporary placement of a peritoneal drain relieved abdominal pressure, while allowing for wound healing until resolution of the thrombus and the ascites. After resolution of the ascites, the patient was rapidly advanced toward tracheostomy decannulation. Failure to decannulate was due to a subglottic stenosis diagnosed by bronchoscopy, with the decision made to discharge her to a chronic care facility for further rehabilitation before going home. The cardiac ultrasound at the time of hospital discharge revealed moderate atrioventricular valve regurgitation and normal ventricular contractility. Medications at the time of discharge were diuretics and beta-blockers. She had spent 1 year and 8 days in the hospital, and 55 days had passed since VAD removal. After a 3-month stay in a rehabilitation facility, the patient was sent home under a regime of tracheostomy collar oxygen, wheelchair bounded due to her original stroke, and on G-tube feeds. A cardiac ultrasound 6 months after discharge demonstrated a resolved valve insufficiency and good systolic function. Shortly after her last cardiology visit, she expired during her sleep. An autopsy revealed a dislodged tracheostomy tube as probable cause of death. The use of a Berlin Heart (Berlin Heart AG, Berlin, Germany) as a bridge to transplant in infants and children with heart failure has proven to be a safe resource. Long-term survival rate is similar to that of patients undergoing elective transplant [1Stiller B. Hetzer R. Weng Y. et al.Heart transplantation in children after mechanical circulatory support with pulsatile pneumatic assist device.J Heart Lung Transplant. 2003; 22: 1201-1208Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar]. In the presence of heart failure in a single ventricle physiology, the Berlin Heart has been used as a bridge to transplant, either as a single or biventricular support device in limited occasions and with mixed results [2Nathan M. Baird C. Fynn-Thompson F. et al.Successful implantation of a Berlin heart biventricular assist device in a failing single ventricle.J Thorac Cardiovasc Surg. 2006; 131: 1407-1408Abstract Full Text Full Text PDF PubMed Scopus (92) Google Scholar, 3Chu M. Sharma K. Tchervenkov C.I. et al.Berlin heart ventricular assist device in a child with hypoplastic left heart syndrome.Ann Thorac Surg. 2007; 83: 1179-1181Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar]. Alternatively, the use of extracorporeal membrane oxygenation as a short-term device in these patients has been reported to have a high mortality with a relatively high incidence of neurologic events in survivors [4Booth K. Roth S.J. Thiagarajan R.R. et al.Extracorporeal membrane oxygenation support of the Fontan and bidirectional Glenn circulations.Ann Thorac Surg. 2004; 77: 1341-1348Abstract Full Text Full Text PDF PubMed Scopus (128) Google Scholar]. Left ventricular remodeling has been the subject of numerous studies describing structural changes in response to chronic alterations in loading or unloading conditions [5Opie L.H. Commerford P.J. Gersh B.J. Pfeffer M.A. Controversies in ventricular remodeling.Lancet. 2006; 367: 356-367Abstract Full Text Full Text PDF PubMed Scopus (705) Google Scholar]. Levin and colleagues [6Levin H.R. Oz M.C. Chen J.M. et al.Reversal of chronic ventricular dilation in patients with end-stage cardiomyopathy by prolonged mechanical unloading.Circulation. 1995; 91: 2717-2720Crossref PubMed Scopus (370) Google Scholar] first described the anatomic and physiologic changes after long-term ventricular mechanical unloading in patients with idiopathic dilated cardiomyopathy and Frazier's group was the first to suggest that chronic mechanical support of failing ventricles had the potential to spare transplantation to a selected group of patients [7Frazier O.H. Radovancevic B. Abou-Awdi N.L. et al.Ventricular remodeling after prolonged ventricular unloading "heart rest" experience with the HeartMate left ventricular assist device.J Heart Lung Transplant. 1994; 13: S51Google Scholar]. High pulmonary vascular resistance also seems to be amenable to significant involution with the help of mechanical support for the failing left ventricle. Adequate reduction of the pulmonary resistance can be expected within 3 to 6 months of implanting an left ventricular assist device in end-stage heart failure patients waiting for a heart transplant [8Martina J. Siegenthalera M.P. Friesewinkela O. et al.Implantable left ventricular assist device for treatment of pulmonary hypertension in candidates for orthotopic heart transplantation—a preliminary study.Eur J Cardiothorac Surg. 2004; 25: 971-977Crossref PubMed Scopus (114) Google Scholar]. The use of the Berlin Heart (Berlin Heart AG) for a prolonged period of time seems to have effectively helped our patient. Although we speculate that chronic unloading led to ventricular remodeling, the true mechanism of recovery remains unknown and we have no further objective data to demonstrate our hypothesis. We believe that there has not been any study that has been published to date on single ventricles remodeling. With a limited supply of hearts for children waiting for a transplant, temporary VAD support may provide the clinician with a dependable alternative for these patients.