Share this article on:

Prevalence of De Novo Aortic Valve Insufficiency in Patients After HeartWare VAD Implantation with an Intermittent Low-Speed Algorithm

Saeed, Diyar; Westenfeld, Ralf; Maxhera, Bujar; Keymel, Stefanie; Sherif, Ahmed; Sadat, Najla; Petrov, GeorGI; Albert, Alexander; Lichtenberg, Artur

doi: 10.1097/MAT.0000000000000391
Adult Circulatory Support

De novo aortic valve insufficiency (AI) is a frequent occurrence in patients supported with left ventricular assist device (LVAD). The European version of the HeartWare LVAD has intermittent low-speed software (lavare cycle) to facilitate intermittent aortic valve opening. We examined aortic valve opening status and prevalence of AI in patients supported with HeartWare LVAD and activated lavare cycle. HeartWare LVAD patients were prospectively monitored using serial echocardiograms at different time points after the LVAD implantation. Inclusion criteria were patients with no > mild AI and/or no aortic valve surgery at the time of LVAD implantation and at least 60 days of support. Three of 37 patients had aortic valve surgery and were excluded from the analysis. A total of 34 patients with mean age of 57 ± 12 years met the inclusion criteria. After median support duration of 408 days (77–1250 days), eight patients had trace/mild AI (24%) and one patient developed moderate AI (3%). An average pump flow, speed, and mean arterial pressure of 4.4 ± 0.6 L/min, 2,585 ± 147 rpm, and 88 ± 11 mmHg were documented, respectively. Aortic valve opening was persistently seen in 22 patients (65%). Aortic valve opening is frequent, and the development of > mild AI seems to be rare in patients supported with HeartWare LVAD.

From the *Medical Faculty, Clinic for Cardiovascular Surgery, University of Düsseldorf, Düsseldorf, Germany; and Medical Faculty, Division of Cardiology, Pulmonology and Vascular Medicine, University of Düsseldorf, Düsseldorf, Germany.

Submitted for consideration February 11, 2016; accepted for publication in revised form May 6, 2016.

Disclosures: The authors have no conflicts of interest to report.

Presented in part at the 35th Annual Meeting and Scientific Sessions of the International Society for Heart & Lung Transplantation, Nice, France, April 15–18, 2015, and at the 23rd Congress of the International Society for Rotary Blood Pumps, Dubrovnik, Croatia, September 27–29, 2015.

Correspondence: Diyar Saeed, MD, PhD, Department of Cardiovascular Surgery, University Hospital of Düsseldorf, Heinrich-Heine University, Moorenstraße 5, 40225 Düsseldorf, Germany. Email: diyar.saeed@med.uni-duesseldorf.de.

The field of the mechanical circulatory support systems underwent a steady growth during the last years particularly for indications as bridge to heart transplantation or destination therapy.1,2 The introduction of continuous-flow systems resulted in a significant improvement of the survival rate, quality of life, and functional capacity compared with previous generation pulsatile flow systems.1 The recently introduced devices underwent major improvements, and many centers report up to 80% 1 year survival rate.3 De novo aortic valve insufficiency (AI) is one of the clinical issues being faced by the clinicians nowadays in patients with prolonged ventricular assist device (VAD) support duration.4

De novo AI is a frequent occurrence in patients supported with VAD ranging between 11% and 42%.5 Almost all of the published studies cover the development of AI in patients supported with HeartMate II (Thoratec Corporation, Pleasanton, CA) device.4,6 Furthermore, the persistent closure of the aortic valve (AV) has been reported several times in the past as a significant risk factor for AI development.4–7 The European version of the HeartWare left ventricular assist device (LVAD; HeartWare, Framingham, MA) offers intermittent low-speed software or lavare cycle that may facilitate intermittent opening of the AV; this software is not approved in the United States. The aim of this study was to prospectively investigate the AV opening status and prevalence of AI in patients supported with HeartWare LVAD.

Back to Top | Article Outline

Materials and Methods

Patient Population

The study protocol was approved by the local research ethics board. HeartWare LVAD patients were prospectively monitored using serial transthoracic echocardiograms at different time points after LVAD implantation (2 month intervals). The transthoracic echocardiogram was performed by two experienced echocardiographers. Inclusion criteria were patients with no > mild AI and/or no aortic valve surgery at the time of LVAD implantation and at least 60 days of support. The AI jet width <25% of the left ventricular outflow tract in the parasternal long axis was defined as mild AI, jet width between 25% and 65% was defined as moderate AI, and jet width from >65% was defined as severe AI. AV opening of at least 3 of 10 beats defined AV opening. The AV opening status was coded as open, intermittent, or closed. Intermittent AV included patients whose valves were open and closed at different echocardiographic examinations. At the time of each echocardiographic examination, the pump flow, speed, and the mean arterial pressure (MAP) was monitored and documented. The pump flow is divided by body surface area to calculate the pump index. The MAP was measured using Doppler sonography or digital sphygmomanometer in the presence of arterial pulsatility. The pump speed was adjusted early after VAD implantation based on cardiac index of the patients and catecholamine requirements. Afterward, the pump speed was rarely altered and only if the patient reported clinical signs of heart failure.

Back to Top | Article Outline

Lavare Cycle

The lavare cycle is a controlled variation in speed that takes place periodically when enabled (Figure 1). The lavare cycle consists of a decrease in speed of 200 rotations per minute (rpm) below set speed for 2 seconds followed by an equal increase in speed of 200 rpm above set speed for 1 second and then a return of the speed to the set speed. Lavare cycle is limited by pump speed range of 1,800–4,000 rpm during the low- and high-speed portions of the cycles. This cycle is repeated periodically every 60 seconds. The lavare cycle was activated in all patients roughly 2 weeks after LVAD implantation after weaning off all perioperative inotropic agents.

Figure 1

Figure 1

Back to Top | Article Outline

Statistical Analysis

The statistical analysis was performed using SPSS 16.0 (SPSS, Chicago, IL). Values of continuous data are presented as mean ± standard deviation or as median with range when appropriate. Categorical variables are displayed as frequency distributions (n) and simple percentages (%). Kolmogorov–Smirnov test was used to assess the normal distribution of the data. Student’s t-test was used to evaluate the effect of the pump parameters and blood pressure on the development of AI. Freedom from AI was estimated using Kaplan–Meier model. Log-rank test was used to calculate the survival difference in patients with or without AI.

Back to Top | Article Outline

Results

Three of 37 patients had aortic valve surgery at the time of LVAD implantation and were excluded from the analysis. A total of 34 patients with mean age of 57 ± 12 years met the inclusion criteria. The preoperative characteristics of the patients are described in Table 1. The majority of the patients (88%) were male. The study included a relatively sick patient population with preoperative INTERMACS I profile of 35%. Furthermore, 32% of the patients were on venoarterial extracorporeal membrane oxygenation support before VAD implantation. The VAD indication was a bridge to heart transplantation in 68% of the patients. Seven patients (21%) required postoperative right VAD for right ventricular failure and were all weaned off right VAD after a mean support duration of 14 ± 11 days.

Table 1

Table 1

Regarding the surgical implantation technique, HeartWare LVAD implantation was performed using cardiopulmonary bypass in 21 patients, on venoarterial extracorporeal membrane oxygenation in 10 patients, and off-pump in three patients. Median sternotomy approach was used in all except for 7 patients, who were operated using minimal invasive (left anterolateral thoracotomy and J-sternotomy) approach. The outflow graft was attached anteriorly to the ascending aorta in all patients except for one patient, in whom the outflow graft was anastomosed to the descending aorta.

After a median VAD support duration of 408 days (77–1250 days), 25 patients (73%) had no AI at all, 8 patients had trace/mild AI (24%), and 1 patient developed moderate AI (3%). Figure 2 shows time-related freedom AI in this patient population. Notably, only 2 of 34 patients had trace/mild AI at the time of VAD implantation. The severity of AI did not change over time in these two patients and stayed as mild AI at the follow-up (both patients had intermittent opening of the AV). Considering AV opening status, 22 patients had open AVs, 8 patients had intermittent AV opening, and 4 patients had persistently closed AV. Of the nine patients who developed AI (trace/mild AI = 8, moderate AI = 1), two patients had an open AV, six patients had an intermittently open AV, and one patient had persistently closed AV, which represents 22%, 67%, and 11% of each subgroup, respectively. Figure 3 shows the relationship between the severity of AI and AV opening status. Aortic valve insufficiency graded as trace/mild was more frequently encountered in those with intermittently closed AV.

Figure 2

Figure 2

Figure 3

Figure 3

Considering the rate of AI at different time points of follow-up, Figure 4 shows the percentage of AI in patients supported with HeartWare VAD for various periods of time (3, 6, 12, and >24 months). During the follow-up, an average pump flow, index, speed, and MAP of 4.4 ± 0.6 L/min, 2.3 ± 0.4 L/min/m2, 2,585 ± 147 rpm, and 88 ± 11 mm Hg were documented, respectively. Figure 5, A–D shows plot box of the average MAP, pump speed, flow, and pump index in patients with or without trace/mild AI in the follow-up. The average measured MAP in patients without AI and patients with trace/mild AI was 86 ± 12 and 86 ± 7 mm Hg, respectively (p = 0.96). The mean pump speed in patients without AI and patients with trace/mild AI at follow-up was 2,561 ± 136 and 2,647 ± 161 rpm, respectively (p = 0.17). Meanwhile, an average pump flow (index) in patients with no AI and patients with trace/mild AI of 4.3 ± 0.6 L/min (2.2 ± 0.36 L/min/m2) and 4.6 ± 0.5 L/min (2.3 ± 0.24 L/min/m2), respectively, was measured (p = 0.17 and p = 0.45).

Figure 4

Figure 4

Figure 5

Figure 5

Considering the outcome after VAD implantation in this cohort, five patients were transplanted, six patients died on device because of non AI-related causes, and one patient underwent device inactivation and ligation of the outflow graft after partial recovery of the left ventricle. There were no statistically significant differences in the survival rate between patients who developed AI during the follow-up and patients without AI (p = 0.575).

Back to Top | Article Outline

Discussion

In this study, we present our experience with the prevalence of AI in patients supported with HeartWare LVAD. To our knowledge, this is the first study evaluating specifically the incidence of de novo AI in patients supported with HeartWare LVAD. The main finding of the study is that the incidence of > mild AI appears to be very rare in patients supported with HeartWare LVAD. In this study, after median VAD support duration of 408 days (77–1250 days), eight patients had trace/mild AI (24%) and merely one patient developed moderate AI (3%). None of the patients developed severe AI and/or required surgical or other procedural intervention of AI.

The mechanism involved in VAD-induced AI is multifactorial and related to variations in blood flow and pressure in the aortic root.8 A high velocity in the ascending aorta because of mismatch in the LVAD outflow graft diameter and ascending aorta may potentially cause aortic root weakening, dilation, and AV cusp malcoaptation.9 Furthermore, high-pressure and velocity jets of regurgitant blood volumes contacting the root side of closed AV may result in valvular damage and degeneration. Another contributing mechanism in the development of AI in LVAD patients include changes to the aortic wall because of sheer stress and high diastolic luminal pressures.10 Once AI is developed in a patient with LVAD, it is clear that valvular incompetence decreases pump efficiency and can lead to worsening heart failure. The immediate consequence of any degree AI is the creation of a redundant circulatory loop whereby retrograde blood flow is returned to the left ventricle through the incompetent valve.11

During the last few years, many studies reported on the issues of developing AI in patients supported with continuous-flow LVAD.4–8,10,12,13 Almost all of these studies involved patients supported with HeartMate II.4,6,10 Cowger et al.10 reported a relatively high incidence of AI in a cohort comprising 53 patients with HeartMate II having freedom from moderate and greater AI of 86, 72, and 36%, at 6, 12, and 18 months, respectively. Pak et al.12 from Columbia University analyzed 63 HeartMate II patients and reported freedom from more than mild AI of 83.6% at 6 months and 75.2% at 12 months. The authors concluded that AI occurs frequently in patients who receive continuous-flow support with a HeartMate II and may be associated with aortic root diameter enlargement and AV opening. Jorde et al.4 recently published a comprehensive study from the same institution involving 232 patients (223 HeartMate II und 9 HeartWare patients). In that study freedom from greater than mild de novo AI at 1 year was 78%. However, at least moderate AI was seen in 38% of the patients after 3 years. Therefore, the authors concluded that AI is common in continuous-flow LVAD patients and may lead to clinical deterioration and surgical intervention. Notably, the prevalence of AI was substantially lower in patients whose AV opens. In a similar study from Aggarwal et al.6 involving 79 patients with HeartMate II, a significant AI (mild or greater) was seen in 52% of patients after a median support duration of 761 days. A higher incidence of AI was seen in patients with closed AVs and older age group. Meanwhile, Rajagopal et al.13 described their experience with 148 continuous-flow LVAD patients comprising 139 HeartMate-II, 6 HeartWare LVAD, and 3 VentrAssist devices (Ventracor, Sydney, New South Wales, Australia), with freedom from moderate or greater AI at 1 and 2 years being approximately 50% and 30%, respectively. In another study from Patil et al.7 involving 58 HeartMate II and 35 HeartWare patients, overall freedom from moderate or greater AI was 94.7% at 1 year, 86.9% at 2 years, 82.8% at 3 years, and 31% at 4 years. The most important independent risk factors for of AI development in that study were the duration of support and persistently closed AV.

The aforementioned studies have several points in common: first, a substantial proportion of patients with HeartMate II develop at least moderate AI during the follow-up, and second, the most important risk factor for AI development is the persistently closed AV and longer support duration. In contrast, the finding of our study, which involves only HeartWare patients, does not support these results. Notably, the clinical relevance of AI development in the majority of HeartMate II patients seems to be negligible as these patients are rarely urgently transplanted only because of AI and/or require AV replacement.

Back to Top | Article Outline

Possible Explanations of Lower De Novo AI in HeartWare Patients

There are several explanations of observing only few, clinically not relevant, de novo AI in this cohort of patients. We assume that the activation of the lavare cycle, which involves periodic variation in the pump speed every 60 seconds that allow opening of the AV (Figure 1), is one of the potential reasons beyond having a significantly less de novo AI in our cohort. We observed only eight patients with trace/mild AI (24%) and one patient with moderate AI (3%) after a median VAD support duration of 408 days (77–1250 days) (Figure 2). This AI rate is significantly less than the reported AI incidence in previously reported studies, which mainly involve HeartMate II patients.4,6,10 As lavare cycle may cause intermittent opening of the AV and several studies report persistently closed AV as a risk factor for AI development, it is intuitive to speculate that the activation of lavare cycle may be one of the possible explanations of low AI incidence in this study cohort.

Another possible explanation of having low de novo AI rate in this cohort may be related to postoperative patient management. During the follow-up, an average pump flow, index, and speed of 4.4 ± 0.6 L/min, 2.3 ± 0.4 L/min/m2, and 2585 ± 147 rpm were documented, respectively. We did not see any statistical significant difference between these parameters in patients with no AI versus patients with trace/mild AI (Figure 5, B–D). However, there was a trend toward having higher pump speed and pump index in patients who develop clinically irrelevant trace/mild AI (p = 0.17 and 0.17, respectively).

Further explanation of the low AI rate in this cohort may be related to strict blood pressure monitoring protocols at our institution. All of our patients are advised before discharge to measure the blood pressure daily and keep their MAP below 85 mm Hg. The patients are all supported with blood pressure measurement device (Doppler sonography or digital sphygmomanometer in the presence of arterial pulsatility) shortly after hospital discharge. It is somewhat intuitive to expect higher AI rate in patients with continuously high blood pressure. The average documented MAP in this cohort was 88 ± 11 mmHg. However, no statistically relevant difference was detected in the MAP between patients with or without trace/mild AI (Figure 4A).

Another potential explanation of having less AI in this cohort may be related to opening status of the AV. The negative impact of closed AV on the development of de novo AI in LVAD patients has been previously reported.4,6,7 Therefore, some investigators have proposed to facilitate AV opening actively by running the pump at lower speeds.14 In this study, 22 patients had continuously open AVs, 8 patients had intermittent AV opening and 4 patients had persistently closed AV. Figure 3 shows the relationship between the severity of AI and AV opening status. Similar to the results of the aforementioned studies, AI graded as trace/mild was more frequently encountered in those with intermittently closed AV.

Back to Top | Article Outline

Comparing HeartWare LVAD with HeartMate II

Both HeartWare LVAD and HeartMate II pump are considered as new-generation continuous flow pumps. However, the pump design is completely different. HeartWare is a centrifugal pump, and HeartMate II is an axial pump. The current study was not designed for a rigorous head-to-head comparison of HeartMate II versus HeartWare but does afford an insight into the perspective of postimplantation AI in patients supported with HeartWare LVAD. We observed a significantly lower rate of moderate AI development in this study compared with previous studies with HeartMate II.4,6,10 The limited number of patients supported with HeartMate II at our institution precludes comparing these two pumps with regard to AI development in this study. Patil et al.7 reported in a different study a higher incidence of AI in patients with HeartWare LVAD compared with HeartMate II. However, that study was retrospective in nature and was not a propensity-matched study, and the HeartWare patients in that study were older and comprised more patients with diabetes mellitus. Interestingly, the lavare cycle was deactivated in all of these patients (personal communication with Harefield physicians), which may also explain the higher incidence of de novo AI in that study compared with our study.

Back to Top | Article Outline

Limitations

Limitation of the study includes the limited number of subjects included in this study. On the basis of our data, we can only assume that lavare cycle may explain the lower AI incidence in HeartWare patients. This finding can only be confirmed if these patients are compared with another cohort of HeartWare LVAD patients without activated lavare cycle. This comparison was not feasible in this study as we activate lavare cycle in every HeartWare LVAD patient. Another limitation includes a relatively short follow-up time in our cohort. The median VAD support duration in this study was 408 days (77–1250 days). The number of patients on support after 1 year was few, and more studies are necessary to examine the incidence of de novo AI in patients with HeartWare ventricular assist device and longer durations of support. However, on the basis of the results of the previously reported studies with HeartMate II,4,6,10 we should have seen a much higher moderate AI rate even after this support duration. Another limitation includes the absence of preoperative aortic root diameter measurements in these patients and its correlation with AI development. Finally, no data exist regarding the changes in the right ventricular function and its correlation with AI development in these patients.

Back to Top | Article Outline

Conclusions

It appears that some HeartWare patients develop clinically irrelevant trace/mild AI during the follow-up. However, the development of > mild AI seems to be rare in these patients. The development of trace/mild AI was more frequent in patients with intermittently closed AV during the follow-up. The opening of AV is frequent in HeartWare patients, and this may also explain the low AI development. Further prospective studies are necessary to confirm this finding.

Back to Top | Article Outline

References

1. Pagani FD, Miller LW, Russell SD, et al; HeartMate II Investigators: Extended mechanical circulatory support with a continuous-flow rotary left ventricular assist device. J Am Coll Cardiol 2009.54: 312–321.
2. Slaughter MS, Rogers JG, Milano CA, et al; HeartMate II Investigators: Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med 2009.361: 2241–2251.
3. Milano C, Pagani FD, Slaughter MS, et al; ADVANCE Investigators: Clinical outcomes after implantation of a centrifugal flow left ventricular assist device and concurrent cardiac valve procedures. Circulation 2014.130(11 suppl 1): S3–S11.
4. Jorde UP, Uriel N, Nahumi N, et al: Prevalence, significance, and management of aortic insufficiency in continuous flow left ventricular assist device recipients. Circ Heart Fail 2014.7: 310–319.
5. 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: 183–188.
6. Aggarwal A, Raghuvir R, Eryazici P, et al: The development of aortic insufficiency in continuous-flow left ventricular assist device-supported patients. Ann Thorac Surg 2013.95: 493–498.
7. Patil NP, Sabashnikov A, Mohite PN, et al: De novo aortic regurgitation after continuous-flow left ventricular assist device implantation. Ann Thorac Surg 2014.98: 850–857.
8. Toda K, Fujita T, Domae K, Shimahara Y, Kobayashi J, Nakatani T: Late aortic insufficiency related to poor prognosis during left ventricular assist device support. Ann Thorac Surg 2011.92: 929–934.
9. Gologorsky E, Gologorsky A, Pham SM: Monitoring of aortic valve opening and systolic aortic insufficiency in optimization of continuous-flow left ventricular assist device settings. J Cardiothorac Vasc Anesth 2012.26: 1063–1066.
10. Cowger J, Pagani FD, Haft JW, Romano MA, Aaronson KD, Kolias TJ: The development of aortic insufficiency in left ventricular assist device-supported patients. Circ Heart Fail 2010.3: 668–674.
11. John R, Mantz K, Eckman P, Rose A, May-Newman K: Aortic valve pathophysiology during left ventricular assist device support. J Heart Lung Transplant 2010.29: 1321–1329.
12. Pak SW, Uriel N, Takayama H, et al: Prevalence of de novo aortic insufficiency during long-term support with left ventricular assist devices. J Heart Lung Transplant 2010.29: 1172–1176.
13. Rajagopal K, Daneshmand MA, Patel CB, et al: Natural history and clinical effect of aortic valve regurgitation after left ventricular assist device implantation. J Thorac Cardiovasc Surg 2013.145: 1373–1379.
14. Slaughter MS, Naka Y, John R, et al: Post-operative heparin may not be required for transitioning patients with a HeartMate II left ventricular assist system to long-term warfarin therapy. J Heart Lung Transplant 2010.29: 616–624.
Keywords:

de novo aortic valve insufficiency; aortic regurgitation; circulatory support devices; cardiomyopathy; adult; heart failure

Copyright © 2016 by the American Society for Artificial Internal Organs