Heart failure (HF) is the most common diagnosis in patients 65 years old and older admitted to hospitals, and despite therapeutic progress, the prognosis of HF remains worse than that of most cancers.1,2 Population aging and successful multidisciplinary management of cardiovascular diseases will further increase the already high prevalence of HF.3,4 Heart transplantation (HTx) is still the gold standard treatment for end-stage HF, because it provides the best long-term results5; however, because of a limited number of donor organs and growing waiting list for HTx, the use of long-term mechanical circulatory support (MCS) as bridge-to-transplant or destination therapy has become a well-established therapeutic option.6,7 The general acceptance of this treatment has been enabled by the very promising clinical results of magnetically activated centrifugal pumps.8 These pumps are significantly different from the volume displacement pumps that have shown several limitations in size and reliability and are associated with poor clinical results.9 Data from the largest international registry of MCS clearly showed that rotary blood pumps were able to significantly improve long-term clinical results with reduction of pump-related morbidity and mortality.9 These results are probably due to the replacement of mechanical bearings with magnetically levitated rotor systems eliminating mechanical wearing and dramatically reducing blood trauma. The HeartMate 3 uses a sophisticated magnetic bearing that significantly reduces shear and compressive forces on blood10,11 (Figure 1). The potential theoretical benefits from these advances in engineering prompted us to implant this device soon after its Conformité Européene (CE) mark approval, and this report summarizes the results collected from the first consecutive 10 patients treated at the “Lausanne–Geneva Transplantation Network.” The already published data on the HeartMate 3 (Thoratec Corporation, Pleasanton, CA) come from CE mark studies,12 although this is a totally independent report.
The HeartMate 3 system is intended for long-term circulatory support of patients with end-stage HF. It is a left ventricular assist device (LVAD) positioned within the pericardium, with a 20 mm inflow cannula entering in the left ventricular (LV) apex and an outflow 14 mm polyester graft sutured on the ascending aorta. All the components but the battery are implanted (Figure 2). It uses a centrifugal pump to generate up to 10 L/min blood flow.10 The pump rotor is fully supported by magnetic levitation (Figure 1). Internal surfaces are textured with titanium microspheres to stimulate endothelial proliferation, therefore, the blood will permanently be in contact with tissue and not with artificial material. An artificial pulse functioning algorithm promotes washing of the pump to prevent the formation of zone of recirculation and stasis. To simplify the surgical procedure, a thin, mechanical apical cuff lock allows quick and easy pump attachment to the LV apex. The system also incorporates a modular driveline for straightforward replacement of external portion, if needed.
This is a single-arm, prospective, two-center, nonblinded, noncontrolled study to evaluate the safety and efficacy of the HeartMate 3 system in patients with refractory, end-stage HF as a bridge to transplantation or destination therapy. All patients suffered from chronic end-stage HF and met the criteria to be enrolled in the Swiss heart transplant program with the exception of three because of age limitation (destination therapy program). They were all in New York Heart Association class 4 and received optimized medical treatment. Device implantation criteria were persistent low output syndrome despite optimal medical treatment (LV ejection fraction < 25%; cardiac index < 2.2 L/min/m2; inotrope dependent); exclusion criteria were as follows: body surface area less than 1.2 m2 and absence of written consent. Patients were enrolled between November 2015 and June 2016. The primary end-point of the study was survival to heart transplant or survival to 90 days on the device, whatever occurred first. Secondary endpoints were major adverse cardiac events related to pump activity.
All patients were monitored for pump flow, selected laboratory parameters, major adverse events, and device malfunctions. The study received the approval of the competent Swiss Ethics Committee (CER-VD 2016-01370).
Device Implantation and Anticoagulation
Device implantation was performed through a standard median sternotomy using normothermic cardiopulmonary bypass (CPB) with central or peripheral cannulation on beating heart, whenever possible (video available at https://www.youtube.com/watch?v=y3sC1dBJea4&feature=youtu.be or see Video, Supplemental Digital Content 1, http://links.lww.com/ASAIO/A133). The first step was the identification of the optimal LV area for the insertion of the 20 mm inflow cannula using transoesophageal echocardiography (TEE) (Figure 3A). Then, the surgeon cored the myocardium using the dedicate circular knife and sutured the apical cuff using 8–10 2/0 braided polyester sutures with Teflon pledgets (BARD International, Inc., NJ) on epicardial side with U stitches technique (Figure 4). The inflow conduit was inserted into the apical cuff and secured in place by engaging the slide-lock mechanism on the pump (Figure 3B). The outflow graft was anastomosed end-to-side to the ascending aorta (Figure 3C). The driveline was tunneled through the abdominal wall to the target position previously chosen with the patient. The pump was switched on for few minutes deairing through the outflow graft. Cardiopulmonary bypass was progressively weaned while pump flow was progressively increased under TEE control avoiding the shifting of the ventricular septum toward the LV. All patients received appropriate volume and inotropic support to avoid right HF. Postoperative anticoagulation guidelines included starting intravenous heparin to reach a partial thromboplastin time (PTT) of 45–55 seconds (or antifactor Xa [anti-FXa] of 0.2–0.45 U anti-Xa/ml), when bleeding was less than 50 ml/hr for 3 consecutive hours. The PTT was progressively increased in postoperative day 2–3 until it reached 65 seconds. Once the patient was able to take oral medications, aspirin (100 mg daily) and anti-vitamin K were administered during the remainder of support, with a targeted international normalised ratio (INR) of 2.5–3.0. To predict the occurrence of the right ventricular (RV) failure, we calculated the CRITT score (central venous pressure [CVP] > 15 mm Hg, RV echocardiography dysfunction, intubation preoperatively, tricuspid regurgitation, tachycardia > 100, and each variable is assigned a score of 0 or 1).
Continuous data are presented as the number of subject, mean with standard deviation (SD), or median and the range. Categorical data are reported as frequencies and percentages. Adverse event data are presented as the number and percentage of patients with the event. Survival analysis is performed with the Kaplan–Meier method, with censoring for heart transplant or cardiac recovery.
Ten consecutive patients were enrolled and implanted with HeartMate 3 at the “Lausanne–Geneva Transplantation Network,” Switzerland. Patients were all under maximal pharmacologic treatment (angiotensin-converting-enzyme [ACE] inhibitors, β-blockers, angiotensin II antagonists, and diuretics). They were a mix of bridge-to-transplant (70%) and destination therapy (30%). There were nine men and one woman with a mean age of 57.3 ± 12 years (range: 38–71 years). Baseline hemodynamic data are provided in Table 1. Body surface areas varied widely, with a mean of 1.95 m2 (range: 1.73–2.18 m2), and the mean body mass index was 25.7 kg/m2 (range: 19–33.1 kg/m2). One had previous cardiac surgery. Idiopathic dilated cardiomyopathy was the etiology of HF in three patients (30%), whereas 70% had ischemic heart disease. Five (50%) were in Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) profiles 1 and 2, three patients (30%) were in profile 3, and two (20%) were in profile 4. Two patients (20%) were under temporary MCS (peripheral venoarterial extracorporeal membrane oxygenation [VA-ECMO]) at implantation.
Surgical procedures were fast with a mean CPB time of 94 minutes (range: 63–150 minutes), and two patients (20%) received aortic valve surgery: one had mechanical aortic valve replaced with bioprosthesis and the other had aortic valve repair because of moderate aortic regurgitation.13 The core-then-sew apical cuff attachment method was used in all procedures. The slide-lock mechanism was easily activated in all cases. Four patients (40%) required temporary right ventricular support (tRVAD) with CentriMag pump (St. Jude Medical, St. Paul, MN) due to failure to wean from CPB. Venous cannula was inserted in the right atrium through the femoral vein, whereas the arterial cannula was inserted into 8 mm Dacron graft sutured end-to-side to the pulmonary artery. The cannula exited the mediastinum at the epigastric level under the right costal margin. This technique allowed the closure of the sternum during the tRVAD support and the retrieval of the cannulas without reopening the chest. The mean duration of tRVAD support was 8.3 days (range: 6–10 days). Bleeding requiring surgical revision occurred in two patients with tRVAD. According to CRITT score, the RV failure expected rate was 20% (two patients had a score of 5 corresponding to 80% risk of right ventricular function [RVF]).
After 90 days, LVAD support was initiated; the pump flow was 4.4 ± 0.5 L/min at 5,300 ± 250 rpm (Table 1). There were no device failures, thrombosis, or pump exchanges during the follow-up. There were no events of hemolysis, and the lactic dehydrogenase profile over the first 90 days was normal (Figure 5). For the entire cohort, renal and hepatic laboratory parameters improved during support (Figure 5). Mean PTT values in the first 15 days postimplant were 64 ± 18 seconds or anti-FXa 0.39 ± 0.05 U anti-Xa/ml.
At the 90 day follow-up, nine patients (90%) remained alive under MCS (Figure 6). None was transplanted. One patient (10%) who received tRVAD died 74 days after implant because of severe respiratory failure. The mean postoperative intensive care stay was 35.4 ± 28 days (range: 5–74 days). None of the patients had stroke, but three (30%) experienced critical illness polyneuropathy as result of prolonged intensive care treatment.
All reoperations were captured in the study, independently from the indications. Five patients (50%) required nine reoperations. Four reoperations were due to bleeding, three for pericardial fluid collection, and two for respiratory failure (tracheostomy). Four patients experienced continuous fever above 38.5°C during the first 7 postoperative days, and two had fever for more than 15 days without evidence of infection.
Two patients underwent cardiopulmonary exercise test (ergospirometry) with mean peak oxygen consumption (VO2 peak) measured at 26 ± 2 ml/kg/min.
Two patients had driveline infection that appeared 30 and 45 days after the implant that required local treatment associated to oral antibiotics. With this approach, driveline infection was under control, but still active at 90 days postoperatively.
In Switzerland, the average time to heart transplant was 72 days in 2002 and 312 days in 2013, and the mortality rate on the waiting list was 6.5% in 2015.6 The increasing incidence of end-stage HF and limited availability of donors has created the need for an alternative treatment option, and magnetically suspended LVADs with long-lasting durability seem to have all the characteristics to be a valid alternative.
The design objectives of the HeartMate 3 are to significantly reduce the degree of shear forces on blood components, to simplify the surgical approach for implantation and to reduce the surface area of the biomaterial–blood interface. Previous clinical experience with other centrifugal pump LVADs has proven that these features will reduce bleeding complications, thrombogenesis, stroke occurrence, and end-organ dysfunction.12,14 The MOMENTUM 3 trial has recently shown that the fully magnetically levitated centrifugal pump HeartMate 3 has a higher rate of survival free of stroke or reoperation to replace the pump at 6 months after implantation than was implantation of the mechanical-bearing axial continuous flow pump HeartMate II among patients with advanced HF, irrespective of their eligibility for transplantation.15 These results match the results of another centrifugal LVAD, the HeartWare HVAD (HeartWare Inc., Framingham, MA). In a recent report, Schmitto et al.16 shown excellent outcomes for patients on the device with a survival rate of almost 60% at 5 years. These results support the hypothesis that centrifugal pumps are the most “blood friendly” pumps ever implanted.
The role played by pulsatility in preventing von Willebrand factor degradation leads to the hypothesis that gastrointestinal bleeding, hemorrhagic stroke, and epistaxis may be reduced by reducing effects on blood proteins and reducing platelet activation.17,18 Echocardiography performed during the follow-up demonstrated in at least 50% of patients, aortic valve opening approximatively every 2–3 beats with the ventricular septum in midline position. These findings have also been reported for the HeartWare HVAD device16 but not for the axial flow pumps.8,9
The small pump size, the integrated inflow cannula, and the dedicated sewing ring to facilitate LV/device connection have contributed to dramatically simplify the implant technique (Figures 3 and 4). It is no more necessary to create pump pocket in the preperitoneal space, making the retrieval of the pump also easier with respect to HeartMate II (Thoratec Corporation, Pleasanton, CA). From a surgical point of view, the implantation was as complex as the implantation of the HeartWare HVAD device in terms of suturing the apical cuff on the LV and the outflow conduit onto the ascending aorta. However, the slide-lock mechanism is definitely a major improvement with respect to the screw in system of the HVAD.
Four patients required tRVAD. Right ventricular failure after LVAD implantation is a serious complication, leading to an estimated 19–43% increase in operative mortality and decreased survival.19 There is no general consensus on the parameters predicting RV failure, and the CRITT score failed to identify half of the patients who experienced RV failure. These data indicate that even clinical RVF risk prediction models developed in the era of continuous flow pumps have rather limited clinical applicability, probably because the RV function is significantly influenced by the LVAD. The rapid unloading of the left ventricle because of the LVAD can cause left displacement of the interventricular septum eventually worsening RV function. Therefore, we set pump speed to avoid complete LV unloading in the first 2 days after LVAD implant. Our operative management that includes the use of selective pulmonary vasodilators (nitric oxide, prostanoids, or Type 5 phosphodiesterase inhibitors) attenuated the development of early RV failure, as also reported in literature.20,21 In all patients, recovery of the RV function occurred in the first 8–10 days postimplant. The tRVAD did not include a membrane oxygenator. Adding an oxygenator to an RVAD has an additional risk as recently shown by Chen et al.,22 because it induces an oxidative stress injury and inflammatory state with detrimental effects on the patient. The weaning process was performed in a stepwise fashion over 24–48 hours and under inotropic support, keeping the central venous pressure below 15 mm Hg. Pulmonary arterial hypertension with an elevated pulmonary vascular resistance (above 5 Wood units) was once thought to predict RV failure after LVAD, because these factors are associated with poor outcome after HTx. More recent studies suggest that depressed RV myocardial function is more accurately characterized by a low RV stroke work index, low pulmonary arterial pressure, and elevated right atrial pressure. Thus, pulmonary hypertension should not be considered a reliable predictor of post-LVAD RV failure.
Bleeding occurred in four out of five patients in day 6 and day 8 postimplant and was not directly associated to surgical technique as confirmed by surgical exploration. The high rate of postoperative bleeding was probably due to supratherapeutic anticoagulation. Although activated partial thromboplastin time (aPTT) is considered the gold standard to assess anticoagulation level in continuous flow LVADs, it is susceptible to physiologic and nonphysiologic factors that do not reflect intrinsic heparin activity.23 Antifactor Xa has been suggested as an alternative measure, because it may exhibit less variability and be less affected by other biological factors unrelated to heparin.23 We found a high level of discordance between aPTT and anti-FXa, most commonly with supratherapeutic aPTT despite therapeutic anti-FXa values, consistently with previous published reports.22,24 The balance of hemostasis in patients with LVADs is complex with multiple variables influencing aPTT and anti-FXa. Liver disease can lead to antithrombin deficiency, which may increase aPTT but decrease anti-FXa levels. On the opposite, hemolysis and hyperbilirubinemia may cause false lowering of the anti-FXa.24 Our feeling is that following the anticoagulation level with anti-FXa could increase the risk of postoperative bleeding.
The pump showed excellent long-term blood compatibility, as evidenced by low levels of lactate dehydrogenase, and the patient’s conditions improved after pump implantation as also evidenced by normalization of vital laboratory parameters (renal and hepatic functions) starting from day 30 after the implant. These findings have also been reported for the HeartWare HVAD device16 and less frequently when an axial flow pump is implanted.8,9
Full hemodynamic support was achieved with power consumption of 4.6 ± 0.5 Watt which is comparable with other current devices.9,15 The mean pump flow of 4.4 ± 0.5 L/min at mean rotation speed of 5,300 ± 250 rpm reflects positively on battery duration, especially when associated to the small and light external controller.10–12
Because of the possible development of a closed loop of LVAD flow, patients with aortic valve regurgitation need the correction of the regurgitation concomitant to LVAD implantation. Even if the aortic valve works properly at the time of LVAD implant, aortic regurgitation develop in 25–52% of patients having continuous flow pump at 1 year due to the progressive stiffness and fusion of the aortic valve leaflets.23,25 Therefore, we decided to repair the leaking aortic valve in one patient. Another patient had a previous mechanical aortic valve replacement. Because of the high risk of valve thrombosis with cerebral embolism and coronary ostia occlusion during LVAD support,3,4 we decided to replace it with a bioprosthesis.
Fever in the first days after implant without a clear evidence of infection (absence of leukocytosis or positive blood cultures) could be explained by a systemic inflammatory reaction because of the polyester components of the outflow graft. Similar reaction has already been seen in endovascular surgery and known as “postimplantation syndrome” and are more frequent since the introduction of woven polyester. Elevated body temperatures could sometimes last for weeks after the procedure.26
The 3 month survival rate of 90% proves the efficacy of the MCS over time without major adverse events despite all our patients had INTERMACS profiles lower than 4 and had high incidence of post-LVAD RV failure. The ergospirometry, unfortunately performed only in two patients, showed VO2 peak values under LVAD support close to physiologic values, once again confirming the consistency of the hemodynamic support. The only fatal complication was respiratory and not linked to the pump.
The limitations of this study were the small number of patients enrolled, the noncontrolled study design, and the follow-up duration. However, we have the privilege to report the first experience on a very high risk population, substantially different from the population previously described in papers based on the CE mark study.12
In conclusion, the initial experience of the “Lausanne–Geneva Transplantation Network” with the HeartMate 3 demonstrates that the circulatory support was consistent even in low INTERMACS profile patients.
1. Bonacchi M, Harmelin G, Bugetti M, Sani G. Mechanical ventricular assistance as destination therapy for end-stage heart failure
: Has it become a first line therapy? Front Surg 2015.2: 35.
2. Braunwald E. The war against heart failure
: The Lancet lecture. Lancet 2015.385: 812824.
3. Stokes MB, Bergin P, McGiffin D. Role of long-term mechanical circulatory support
in patients with advanced heart failure
. Intern Med J 2016.46: 530540.
4. Mancini D, Colombo PC. Left ventricular assist devices: A rapidly evolving alternative to transplant. J Am Coll Cardiol 2015.65: 25422555.
5. Galeone A, Kirsch M, Barreda E, et al. Clinical outcome and quality of life of patients surviving 20 years or longer after heart transplantation. Transpl Int 2014.27: 576582.
6. Tozzi P, Hullin R. Mechanical circulatory support
for destination therapy. Swiss Med Wkly 2016.146: w14267.
7. Pozzi M, Giraud R, Tozzi P, Bendjelid K, Robin J, Meyer P, et al. Long-term continuous-flow left ventricular assist devices (LVAD) as bridge to heart transplantation. J Thorac Dis 2015.7: 532542.
8. Katz MR, Dickinson MG, Raval NY, Slater JP, Dean DA, Zeevi GR, et al. Outcomes of patients implanted with a left ventricular assist device
at nontransplant mechanical circulatory support
centers. Am J Cardiol 2015.115: 12541259.
9. Kirklin JK, Naftel DC, Pagani FD, Kormos RL, Stevenson LW, Blume ED, et al. Seventh INTERMACS annual report: 15,000 patients and counting. J Heart Lung Transplant 2015.34: 1495504.
10. Bourque K, Cotter C, Dague C, Harjes D, Dur O, Duhamel J, et al. Design rationale and preclinical evaluation of the HeartMate 3 left ventricular assist system for hemocompatibility. ASAIO J 2016.62: 375383.
11. Schibilsky D, Lenglinger M, Avci-Adali M, Haller C, Walker T, Wendel HP, et al. Hemocompatibility of axial versus centrifugal pump technology in mechanical circulatory support
devices. Artif Organs 2015.39: 723728.
12. Zimpfer D, Netuka I, Schmitto JD, Pya Y, Garbade J, Morshuis M, et al. Multicentre clinical trial experience with the HeartMate 3 left ventricular assist device
: 30-day outcomes. Eur J Cardiothorac Surg 2016.50: 548554.
13. Park SJ, Liao KK, Segurola R, Madhu KP, Miller LW. Management of aortic insufficiency in patients with left ventricular assist devices: A simple coaptation stitch method (Park’s stitch). J Thorac Cardiovasc Surg 2004.127: 264266.
14. Esmaeilzadeh F, Wauters A, Wijns W, Argacha JF, van de Borne P. Effects of HeartWare ventricular assist device
on the von Willebrand factor: Results of an academic Belgian center. BMC Cardiovasc Disord 2016.16: 155.
15. Mehra MR, Naka Y, Uriel N, Goldstein DJ, Cleveland JC Jr, Colombo PC, et al.; MOMENTUM 3 Investigators: A fully magnetically levitated circulatory pump for advanced heart failure
. N Engl J Med 2016.376: 440450.
16. Schmitto JD, Zimpfer D, Fiane AE, et al. Long-term support of patients receiving a left ventricular assist device
for advanced heart failure
: A follow-up analysis of the Registry to Evaluate the HeartWare Left Ventricular Assist System. Eur J Cardiothorac Surg 2016.50: 834838.
17. Netuka I, Kvasnička T, Kvasnička J, et al. Evaluation of von Willebrand factor with a fully magnetically levitated centrifugal continuous-flow left ventricular assist device
in advanced heart failure
. J Heart Lung Transplant 2016.35: 860867.
18. Nair PK, Kormos RL, Teuteberg JJ, et al. Pulsatile left ventricular assist device
support as a bridge to decision in patients with end-stage heart failure
complicated by pulmonary hypertension. J Heart Lung Transplant 2010.29: 201208.
19. Kormos RL, Teuteberg JJ, Pagani FD, et al.; HeartMate II Clinical Investigators: Right ventricular failure in patients with the HeartMate II continuous-flow left ventricular assist device
: Incidence, risk factors, and effect on outcomes. J Thorac Cardiovasc Surg 2010.139: 13161324.
20. Potapov E, Meyer D, Swaminathan M, et al. Use of inhaled nitric oxide after left ventricular assist device
placement: A prospective, randomised, double-blind, multicenter, placebo-controlled trial. J Heart Lung Transplant 2011.30: 870878.
21. Noly PE, Kirsch M, Quessard A, et al. Temporary right ventricular support following left ventricle assist device implantation: A comparison of two techniques. Interact Cardiovasc Thorac Surg 2014.19: 4955.
22. Chen Q, Yu W, Shi J, et al. The effect of venovenous extra-corporeal membrane oxygenation (ECMO) therapy on immune inflammatory response of cerebral tissues in porcine model. J Cardiothorac Surg 2013.8: 186.
23. Guervil DJ, Rosenberg AF, Winterstein AG, Harris NS, Johns TE, Zumberg MS. Activated partial thromboplastin time versus antifactor Xa heparin assay in monitoring unfractionated heparin by continuous intravenous infusion. Ann Pharmacother 2011.45: 861868.
24. Adatya S, Uriel N, Yarmohammadi H, et al. Anti-factor Xa and activated partial thromboplastin time measurements for heparin monitoring in mechanical circulatory support
. JACC Heart Fail 2015.3: 314322.
25. Deo SV, Sharma V, Cho YH, Shah IK, Park SJ. De novo aortic insufficiency during long-term support on a left ventricular assist device
: A systematic review and meta-analysis. ASAIO J 2014.60: 183188.
26. Voûte MT, Bastos Gonçalves FM, van de Luijtgaarden KM, et al. Stent graft composition plays a material role in the postimplantation syndrome. J Vasc Surg 2012.56: 15031509.