Heart transplant is in high demand, but the wait list exceeds 6 months in Turkey. Until a donor heart can be procured, venoarterial extracorporeal membrane oxygenation is an important support option to bridge patients on the wait list or as a rescue therapy for patients with right ventricular failure after implant of left ventricular assist device; it is less expensive than other options, provides benefits such as simple percutaneous insertion, and requires neither sternotomy nor biventricular and respiratory support. We present a case of a patient bridged to transplant with 5 months of extracorporeal membrane oxygenation support.
Key words : Mechanical circulatory support, Percutaneous insertion
Heart transplant is the gold standard treatment for end-stage heart failure. Until a donor heart is procured, venoarterial extracorporeal membrane oxygenation (VA-ECMO) is one of several mechanical circulatory support options available to clinicians to bridge patients on the wait list in case of multiorgan failure or as a rescue therapy for patients with right ventricular failure (RVF) after implant of a left ventricular assist device (LVAD). Venoarterial extracorporeal membrane oxygenation, as a less expensive option versus other mechanical circulatory support options, provides benefits such as simple percutaneous insertion without the need for sternotomy and biventricular and respiratory support.1 However, because of the shortage of donor organs, mean wait list time for transplant exceeds 6 months in our country (Turkey), and the support time limitation associated with ECMO remains a point of major concern. We present a case of a patient bridged to transplant with 5 months of ECMO support.
A 38-year-old male patient presented with exertional dyspnea and was diagnosed with dilated cardiomyopathy after a transthoracic echocardiographic examination. Transthoracic echocardiography revealed an ejection fraction of 28%, systolic pulmonary artery pressure of 35 mm Hg, left ventricular end-diastolic diameter of 6.8 cm, severe tricuspid insufficiency, and tricuspid annular plane systolic excursion of 1.4 cm. His pulmonary vascular resistance was 2 Wood units, as documented with catheterization of the right side of the heart. A cardiac resynchronization therapy defibrillator was implanted, and he was placed on the heart transplant elective list and discharged.
During follow-up, the patient’s medical condition deteriorated. The patient was hospitalized, and infusion of dobutamine was initiated; he was accepted as Interagency Registry for Mechanically Assisted Circulatory Support profile 3 (defined as “stable but inotrope dependent”). The heart team decided to implant a HeartWare HVAD durable ventricular assist device (Medtronic) and proceeded with concomitant tricuspid valve ring annuloplasty (Contour 3D annuloplasty ring, Medtronic) via median sternotomy. He was extubated on postoperative day 1.
After an uneventful postoperative course of 17 days, infection with Klebsiella pneumoniae was confirmed in the patient by sputum culture, and pneumonia complicated the course of treatment and RVF developed. He was intubated, and treatment for RVF (inhaled nitric oxide; sildenafil, milrinone, and dobutamine) was initiated. His clinical condition improved by medical treatment, and he was extubated. There were another 30 days of stable medical status; however, severe RVF and hypoxia, necessitating intubation and ECMO support, occurred on postoperative day 73. Percutaneous femoral VA-ECMO (Medos Cardiopulmonary Solutions) was inserted via the right groin vessels, and he was listed as high urgent. The patient was extubated on the following day. The extracorporeal membrane oxygenation circuit was changed after 35 days of support to mitigate ineffective oxygenation and hemolysis.
To wean the patient from ECMO, flow was reduced to 1 L/min, twice; however, both attempts were unsuccessful, and the flow was increased to 2.5 L/min. At day 73 of ECMO support, right groin vessel bleeding at the insertion site of ECMO was encountered, and the cannulas were transferred to the left femoral artery and vein percutaneously, with concomitant right femoral artery repair by interposition graft placement of an expanded polytetrafluoroethylene prothesis. Ten days later, a right femoral incision site abscess was drained. After 28 days of left femoral VA-ECMO, the circuit was exchanged to mitigate clots in the lines leading to hemolysis. Five days later, massive bleeding from the right femoral incision site was stopped with primary sutures by emergent re-exploration. Ten days later, a reverse saphenous vein graft was interposed instead of previous expanded polytetrafluoroethylene graft due to chronic bleeding at the right groin incision.
Left distal pulse was impalpable at day 54 of left VA-ECMO support. The arterial cannula was transferred to the right subclavian artery via an end-to-side 8-mm Dacron graft, and a left femoral embolectomy was performed. Finally, after 139 days of ECMO support, the patient was successfully bridged to heart transplant with an uneventful intraoperative and postoperative course. He was extubated on postoperative day 2 and discharged after 2 endomyocardial biopsies that showed no evidence of rejection (grade 0, according to the 2004 criteria of the International Society for Heart and Lung Transplantation). All events are provided in the timeline (Figure 1).
At the time of this writing, the patient was on triple immunosuppressive therapy that included corticosteroids, mycophenolate mofetil, and cyclosporine, and he was ambulatory with perfect cognitive and physical status.
Severe RVF after LVAD implant is associated with a substantial increase in morbidity and mortality and less successful bridging to heart transplant.2 In cases in which patients are unresponsive to all other options of medical treatment, durable intracorporeal or temporary extracorporeal mechanical support is inevitable. Although conversion to a biventricular assist device is a feasible option, the high costs and reimbursement issues associated with health insurance systems in developing countries may compel transplant centers to choose one of several types of temporary mechanical circulatory support as a bridge to heart transplant; in such situations, VA-ECMO is often the choice.3 While VA-ECMO supports both ventricles indirectly by decompression, it also provides gas exchange. Two major limitations of VA-ECMO are the high rate of device-related complications and the short duration of support. Despite these limitations, the shortage of available donor hearts in Turkey has created a need for longer durations of pretransplant VA-ECMO, generally exceeding 14 days. In this report, we presented the case of a patient who was successfully bridged to heart transplant by long-term temporary VA-ECMO support and had undergone LVAD implant with concomitant tricuspid valve repair complicated with RVF.
There are potential risks to long duration of peripheral VA-ECMO support
First, posttransplant survival rates of these patients are lower than survival rates of transplant recipients who did not require ECMO support. Analysis of data from the French national transplant registry, CRISTAL, revealed a 1-year posttransplant survival rate of 70% in the VA-ECMO group, which was significantly lower than that shown in the comparison group.1 According to the data of the Spanish Heart Transplant Registry, of 129 patients who were on VA-ECMO, 43 of these patients died after transplant (33%).4 In our case, peripheral VA-ECMO was required to mitigate RVF, as well as the hypoxic effects of pneumonia that necessitated respiratory support. The reason for prolonged support was several unsuccessful attempts to wean the patient from ECMO.
Second, with increasing duration of support, complications associated with VA-ECMO may be encountered, such as infection, bleeding, stroke, device dysfunction, vascular access site reinterventions, and pulmonary edema. Anticoagulation is a particularly difficult problem in these patients, leading to bleeding and thrombotic complications. Our anticoagulation strategy for the patient was as follows. After LVAD implant, unfractionated heparin infusion was initiated on postoperative day 1, which was continued until extubation was possible or the targeted international normalized ratio was achieved. When the patient was able to tolerate medications by mouth, a standard anticoagulation regimen with oral warfarin was initiated on postoperative day 1, to keep the international normalized ratio between 2.5 and 3.0. After VA-ECMO insertion, unfractionated heparin infusion to achieve an activated partial thromboplastin time of 60 seconds was administered. If the international normalized ratio was above 2.5, then the unfractionated heparin infusion was stopped. The patient received 100 mg of acetylsalicylic acid as an antiplatelet therapy. In our practice, this strategy generally prevents bleeding and thrombotic complications.
Furthermore, peripheral cannulation of femoral vessels is associated with infection, insertion site bleeding due to chronic injury to the vessels, and limb ischemia. Insertion of a distal perfusion cannula and distal pulse palpation with frequent Doppler ultrasonography are compulsory. In case of such extremity complications, decannulation and insertion of cannulas to the contralateral leg or subclavian artery are necessary with concomitant embolectomy if needed. Insertion of cannulas for VA-ECMO via femoral vessels percutaneously was the preferred method in this case. Urgency of the intervention and consideration of short-term support had compelled the surgical team to use this approach. With the increased number of complications in prolonged support, the arterial cannula was eventually moved to the subclavian artery.
If a patient has both an implanted LVAD and VA-ECMO support, then decompression of the left ventricle can be provided easily; however, because the femoral VA-ECMO steals preload and works against LVAD flow, both flows require adjustment according to the respiratory demand. In addition, total mechanical circulatory support flow requires close monitoring of tissue perfusion by arterial blood gas and lactate level analysis. In cases of donor organ shortage and lack of medical insurance coverage for durable biventricular support, duration of VA-ECMO support could be much longer than planned, as in our case. When duration of VA-ECMO support extends past the standard recommended period, conversion to a biventricular assist device is essential to address the increased rate of complications such as bleeding and vascular problems with peripheral ECMO support. However, the option of conversion to a biventricular assist device was not possible in our case because of delays in the coverage by the national medical insurance institution regarding payment to the device company, and this situation resulted in a prolonged period of ECMO support.
In conclusion, VA-ECMO as a bridge to transplant is a plausible option for clinicians with patients who are waiting for a heart transplant, and outcomes are satisfactory if the duration of support is short. However, in cases of donor shortage and economic obstacles, longer duration VA-ECMO support could be a life-saving option to bridge patients to a successful heart transplant, but meticulous care is essential to facilitate positive outcomes.
DOI : 10.6002/ect.2020.0159
From the 1Department of Cardiovascular Surgery, Ankara City Hospital, Ankara,
Turkey; and the 2Department of Emergency Medicine, Ankara City Hospital, Ankara,
Acknowledgements: The authors have not received any funding or grants in support of the presented research or for the preparation of this work and have no potential declarations of interest.
Corresponding author: Dogan Emre Sert, Ankara City Hospital, Department of Cardiovascular Surgery, 06800 Ankara, Turkey
Phone: +90 312 552 60 00
Figure 1. Timeline for a Case of Long-Duration Venoarterial Extracorporeal Membrane Oxygenation