Author + information
- Received May 25, 2012
- Revision received August 13, 2012
- Accepted August 16, 2012
- Published online January 1, 2013.
- Kishan S. Parikh, MD⁎,
- Amit K. Mehrotra, MD, MBA⁎,
- Mark J. Russo, MD, MSC†,
- Roberto M. Lang, MD⁎,
- Allen Anderson, MD⁎,
- Valluvan Jeevanandam, MD†,
- Benjamin H. Freed, MD⁎,
- Jonathan D. Paul, MD⁎,
- Janet Karol, MSN⁎,
- Sandeep Nathan, MD, MS⁎ and
- Atman P. Shah, MD⁎,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Atman P. Shah, Assistant Professor of Medicine, University of Chicago Medical Center, 5841 South Maryland Avenue, MC 6080, Chicago, Illinois 60637
Objectives This study sought to assess the effectiveness of a novel percutaneous method to treat left ventricular assist device (LVAD)–associated severe aortic insufficiency (AI) in a series of patients determined to be poor reoperative candidates.
Background The increased use of continuous-flow LVAD in advanced heart failure has led to marked changes in the management of patients with this condition. However, secondary AI can become a significant complication.
Methods Five patients with continuous-flow LVAD and severe post-LVAD AI underwent percutaneous transcatheter aortic valve closure from September to October 2011 at a single quaternary care academic medical center. All patients had LVAD implanted as destination therapy. LVAD parameters, hemodynamics, and echocardiographic measurements were obtained before and after aortic valve closure.
Results All patients underwent successful closure with the Amplatzer cribriform device (AGA Medical, Plymouth, Minnesota) via a percutaneous transcatheter femoral approach with a significant reduction of AI from severe to trivial. Cardiac hemodynamics improved, and the pulmonary capillary wedge pressure was reduced in all patients. There was no change in mitral or tricuspid regurgitation, LVAD power, or pulsatility index.
Conclusions Percutaneous transcatheter closure of the aortic valve effectively treats LVAD-associated AI and reduces pulmonary capillary wedge pressure. This procedure should be considered to treat LVAD-associated AI in patients who are poor candidates for repeat operation. Further data are needed to assess long-term results.
Left ventricular assist devices (LVAD) improve survival and quality of life in patients with heart failure and are being increasingly used as destination therapy. More than one-third of all listed adult heart transplant candidates and 75% of those initially listed as U.S. United Network for Organ Sharing status 1 have undergone LVAD implantation since 1999 (1). However, development of de novo aortic insufficiency (AI) is a significant complication recognized with long-term LVAD support, especially with increased use of continuous-flow devices. Cowger et al. (2) recently reported a 51% prevalence of moderate or severe AI after 18 months of continuous-flow LVAD therapy with HeartMate II (Thoratec, Pleasanton, California) devices. The development of AI results in blood returning to the left ventricle via a low-resistance circuit and leads to ineffective device support, decreased device durability, and end organ malperfusion. The development of AI in this population is associated with increased morbidity and mortality after 1 year (3).
To avoid the development of LVAD-associated AI, aortic valve (AV) closure during LVAD insertion has been achieved using the Park stitch for suturing the AV leaflets (4) or the sandwich plug technique, which employs felt patches on either side of the AV (5). Unfortunately, surgical closure to treat LVAD-associated AI is also often complicated by AI recurrence resulting in a significant risk and is associated with an operative mortality as high as 7% (6). Many patients with LVAD and significant AI develop severe decompensated heart failure and are not reoperative candidates due to their clinical status. A less-invasive approach using an open surgical transcatheter technique to treat LVAD-associated AI was recently reported by Grohmann et al. (7). We have also previously described a percutaneous transcatheter method to reduce severe AI (8). The aim of this study was to assess the effectiveness of a novel percutaneous technique to treat LVAD-associated severe AI and to assess its effectiveness using echocardiography and hemodynamics in a series of patients determined to be poor reoperative candidates.
Informed, written consent was obtained for data derived from human subjects. After informed consent was obtained, 5 consecutive patients with continuous-flow LVAD (4 HeartMate II, 1 HeartWare device, Thoratec) and post-LVAD severe AI at a single, quaternary care academic medical center underwent the procedure in September and October 2011. Cardiothoracic surgery and interventional cardiology services had previously discussed the care of each patient and jointly determined that their operative risk was excessively high (operative mortality risk >75%) to recommend surgical closure.
After standard placement of femoral arterial and venous sheaths, a 7-F Swan-Ganz pulmonary artery catheter (Edwards Lifesciences, Irvine, California) was used to obtain initial hemodynamic measurements. A 6-F multipurpose catheter was used to cross the aortic valve and then exchanged for an 8-F, 80-cm, 45° TorqVue catheter (AGA Medical, Plymouth, Minnesota). Using transesophageal echocardiography (TEE) guidance, the AV annulus was measured. A 25-, 30-, or 35-mm Amplatzer Multi-Fenestrated Septal Occluder—cribriform—device (AGA Medical, Plymouth, Minnesota) was selected so the size of the device would match as closely as possible the annulus size. The selected device was advanced to the AV. The device was deployed under fluoroscopic and transesophageal guidance across the AV. Closure of the AV and reduction of AI was confirmed by TEE. Patency of the coronary arteries was subsequently established with coronary angiography and removal of the femoral sheaths was managed by manual compression.
Data collection and statistical analysis
Patient characteristics, including demographics, and clinical and laboratory values were obtained from the electronic medical record (Table 1). Determination of anatomically normal AV was made with transthoracic echocardiography (TTE) performed within 1 week of LVAD implantation. LVAD parameters were obtained before AV closure and on days 1 and 3 after AV closure (Table 2). Hemodynamics were measured immediately before and 5 min after AV closure (Table 3). Intraprocedural TEE and TTE acquired 12 days before and 6 days after the procedure were used to assess the severity of aortic insufficiency with standard guideline criteria. The Wilcoxon signed-rank test for matched pairs was performed for pre- and post-AV closure data. A p value <0.05 was considered statistically significant. Statistical analysis was performed using SPSS version 19 (IBM, Armonk, New York). Results are reported as median (range) unless otherwise specified.
All 5 patients underwent successful placement of the Amplatzer device (Figs. 1 and 2). The average age was 52 years, 4 were men, and 40% had an LVAD placed for ischemic cardiomyopathy. The median time of LVAD support before AV closure was 1.0 years (0.4 to 2.1 years). Relevant clinical and laboratory characteristics are reported in Table 1. There was near-complete resolution of the aortic insufficiency (AI) in all patients with TEE (Fig. 1, Online Video 1). There were no immediate complications in any of the procedures. After 4 days, 1 device was found to have embolized to the aortic arch in 1 patient and was successfully retrieved percutaneously. At 30 days, 2 of the patients were alive and TTE demonstrated a static device and no further AI. One patient developed severe right heart failure, and the family withdrew care at post-procedure day 12 after he developed renal failure. The remaining patient died on post-procedure day 18 from sepsis. In both of these patients, TTE (obtained at least 10 days after the index procedure) revealed the device in position and no further AI.
LVAD parameters, including pump flow, speed, pulsatility index, and pump power pre-, 1 day post-, 3 days post-AV closure are shown in Table 2. There was a trend toward increased pump flow (p = 0.083) on day 1 but this trend was not maintained on day 3.
TTE showed improvement of AI from severe to mild or absent (p = 0.038) after AV closure in all patients confirming the results noted with intraprocedural TEE (Fig. 1). Mild-to-moderate mitral regurgitation and severe tricuspid regurgitation were also observed pre-AV closure in all patients. There was no post-procedural change in the severity of either mitral or tricuspid regurgitation in any of the patients.
Right ventricular systolic pressure, mean pulmonary artery pressure, pulmonary capillary wedge pressure (PCWP), and left ventricular end-diastolic pressure all decreased significantly post-Amplatzer device implantation (Table 3, Fig. 3) without a change in the right atrial pressure. Pulmonary artery saturation increased significantly, and there was a trend toward an increase in cardiac output with the procedure, with 1 patient having unchanged cardiac output and the rest having increased values. Notably, 1 patient's cardiac output increased dramatically from 2.4 to 3.3 l/min after AV closure.
We present a series of patients with LVAD-associated severe AI who were effectively treated with a percutaneous transcatheter technique. Our study shows that LVAD-associated severe AI can be successfully and safely treated with our approach to AV closure using an Amplatzer Multi-Fenestrated Septal Occluder—cribriform—device in patients who are poor reoperative candidates. We also observed significant improvement in patients' hemodynamics following the procedure. There was 1 device embolization that occurred in a patient with a prosthetic mitral valve, in whom the prosthetic strut extended into the left ventricular outflow tract may have prevented adequate seating of the device.
The transcatheter technique as a preferred approach to addressing LVAD-associated AI is receiving increased attention because it is less invasive than is open surgical repair. Grohmann et al. (7) used a surgical transcatheter technique with an Amplatzer ventricular septal defect plug, and Santini et al. (9) recently used a CoreValve (Medtronic, Minneapolis, Minnesota) to treat LVAD-associated AI. Additionally, although not in a patient with an LVAD, Riede et al. (10) reported successful closure of the AV with an atrial septal occluder in an infant with hypoplastic left heart syndrome. Our technique is the first percutaneous method to our knowledge reported in a series of patients.
Development of LVAD-associated AI
The increasing use of continuous-flow LVAD for the treatment of advanced heart failure patients and the increased life expectancy of patients who receive LVAD has made the development of severe AI a significant morbidity and mortality risk in this patient population (3). The exact etiology is unknown and likely multifactorial. Mobility and integrity of the valve may be compromised over time due to the LVAD's redirection of blood flow, especially in patients with ventricles too weak to generate enough intrachamber pressure to open the AV (2,11). Intermittent opening of the AV has been demonstrated to result in commissural fusion and development of myxomatous granulation tissue on the coronary cusps, and fused valves may degenerate when exposed to high-velocity flow from the aortic cannula (12). In addition to the ability of pulsatile-flow LVAD to allow physiological opening of the AV, aortic regurgitation has been reported to be increased more in continuous-flow than in pulsatile-flow devices and is likely secondary to differences in aortic blood flow patterns and smaller outflow cannulas (2,13–15).
Benefits and considerations of percutaneous AV closure
Optimizing LVAD Function
The underlying challenge of AI occurrence in patients with LVAD support is that a low-resistance continuous circuit is created by which the regurgitant blood returns from the LVAD outflow cannula directly back into the inflow cannula in the left ventricle. As a result, net systemic flow via the higher resistance arterial circuit will decrease even as device readings indicate higher pump flows. Therefore, despite efforts to improve systemic perfusion by increasing power and flow, effective output will progressively diminish as AI worsens. Closing this low-resistance circuit with the Amplatzer occluder reduces the demands on the LVAD, improves systemic flow, and increases the LVAD's durability and longevity. Although our results do not show a significant change in pump flow before and after AV closure, pre-closure flows are artificially elevated because of LVAD inefficiency in the setting of AI (2,16).
Improvement of Diastolic Filling Pressures
We have shown that by treating LVAD-associated AI, several hemodynamic parameters, including PCWP and left ventricular end-diastolic pressure, are significantly improved. PCWP has been shown to be an independent predictor of survival in patients with an ejection fraction <20% from either ischemic or nonischemic cardiomyopathy (17). There is also evidence supporting that an elevated PCWP contributes to right ventricular dysfunction directly by increasing net right ventricular afterload (18). Elevated PCWP also correlates with symptoms of dyspnea, orthopnea, and paroxysmal nocturnal dyspnea. Though not statistically significant in this study, and likely confounded by significant tricuspid regurgitation and right ventricular failure, the observed decrease in right atrial pressure in all patients suggests that the right ventricle is unloaded in addition to the left ventricle.
The results from this study are limited by its nonrandomized, small sample size. Adjustments for sample bias were made by assuming a nonparametric population in the statistical analysis. Another limitation of our study is the lack of long-term follow-up. The long-term implications of AV closure by any method in patients with LVAD are unclear, and device malfunction in this setting could result in immediate hemodynamic collapse and death.
Percutaneous transcatheter AV closure appears to be a safe and effective method to treat severe aortic insufficiency in patients with LVAD, a population often too sick to undergo reoperation for treatment. Although our results are encouraging, the technique needs to be studied in a larger cohort for longer-term follow-up.
For supplemental videos, please see the online version of this article.
Dr. Lang received a moderate equipment grant from Philips. Dr. Jeevanandam is a scientific adviser to Thoratec, Heartware, and Terumo. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- aortic insufficiency
- aortic valve(s)
- left ventricular assist device(s)
- pulmonary capillary wedge pressure
- transesophageal echocardiography
- transthoracic echocardiography
- Received May 25, 2012.
- Revision received August 13, 2012.
- Accepted August 16, 2012.
- American College of Cardiology Foundation
- Mancini D.,
- Lietz K.
- Cowger J.,
- Pagani F.D.,
- Haft J.W.,
- Romano M.A.,
- Aaronson K.D.,
- Kolias T.J.
- Grohmann J.,
- Blanke P.,
- Benk C.,
- Schlensak C.
- Freed B.H.,
- Paul J.D.,
- Bhave N.M.,
- et al.
- Santini F.,
- Forni A.,
- Dandale R.,
- et al.
- Tedford R.J.,
- Hassoun P.M.,
- Mathai S.C.,
- et al.