bs_bs_banner
THOUGHTS AND PROGRESS
B-Type Natriuretic Peptide Levels Predict Ventricular Arrhythmia Post Left Ventricular Assist Device Implantation *Yaron Hellman, *Adnan S. Malik, †Hongbo Lin, †Changyu Shen, ‡I-Wen Wang, ‡Thomas C. Wozniak, ‡Zubair A. Hashmi, *Jeanette Pickrell, *Milena Jani, *Marco A. Caccamo, *Irmina Gradus-Pizlo, and *Azam Hadi *Krannert Institute of Cardiology, †Department of Biostatistics, IU School of Medicine; and ‡IU Health Cardiovascular Surgeons, Indianapolis, IN, USA Abstract: B-type natriuretic peptide (BNP) levels have been shown to predict ventricular arrhythmia (VA) and sudden death in patients with heart failure. We sought to determine whether BNP levels before left ventricular assist device (LVAD) implantation can predict VA post LVAD implantation in advanced heart failure patients. We conducted a retrospective study consisting of patients who underwent LVAD implantation in our institution during the period of May 2009–March 2013. The study was limited to patients receiving a HeartMate II or HeartWare LVAD. Acute myocardial infarction patients were excluded. We compared between the patients who developed VA within 15 days post LVAD implantation to the patients without VA. A total of 85 patients underwent LVAD implantation during the study period. Eleven patients were excluded (five acute MI, four without BNP measurements, and two discharged earlier than 13 days post LVAD implantation). The incidence of VA was 31%, with 91% ventricular tachycardia (VT) and 9% ventricular fibrillation. BNP remained the single most powerful predictor of VA even after adjustment for other borderline significant factors in a multivariate logistic regression model (P < 0.05). BNP levels are a strong predictor of VA post LVAD implantation, surpassing previously described risk factors such as age and VT in the past. Key Words: B-type natriuretic peptide—Left ventricular assist device—Ventricular arrhythmia— Ventricular tachycardia—Ventricular fibrillation.
Left ventricular assist devices (LVADs) provide improved long-term survival and quality of life in appropriately selected patients with advanced heart failure (1–3). However, sustained ventricular arrhythmias (VAs) in patients with continuous flow LVADs may result in eventual hemodynamic compromise due to the unsupported right ventricle and lack of preload (4,5).
doi:10.1111/aor.12486 Received October 2014; revised November 2014. Address correspondence and reprint requests to Dr. Azam Hadi, Krannert Institute of Cardiology, IU School of Medicine, IU Health Methodist Professional Center 2, 1801 N. Senate Boulevard, Suite 2000, Indianapolis, IN 46202, USA. E-mail: mazhadi @iu.edu
1051
B-type natriuretic peptide (BNP) has been evaluated as a predictor of sudden cardiac death and VA in the heart failure population, including patients with heart failure post myocardial infarction and implantable cardioverter defibrillator (ICD) recipients (6–8). Elevated BNP levels are specifically associated with an increased rate of ICD appropriate therapy and VA burden (9,10). The majority of these studies did not specifically include patients with advanced heart failure and severe LV systolic dysfunction requiring VAD or transplant. We hypothesized that BNP can aid in the prediction of VA post LVAD implantation. PATIENTS AND METHODS We conducted a retrospective study consisting of patients who underwent LVAD implantation in our institution during the period of May 2009–March 2013. The study was limited to patients receiving a HeartMate II (Thoratec, Pleasanton, CA, USA) or HeartWare LVAD (Framingham, MA, USA). Patients with acute myocardial infarction, active myocarditis, or who did not have BNP measurements were excluded. VA was defined as ventricular fibrillation (VF) or any ventricular tachycardia (VT— including asymptomatic nonsustained VT), which was documented in the electronic medical record within 15 days post LVAD implantation. The timeframe of 15 days was selected as the patients were in hospital and all were continuously monitored with telemetry. Analyses were done to compare the patients who developed VA with the patients without VA. BNP levels were defined as the maximal BNP level measured in the Indiana University Health Methodist hospital laboratory during the 4 months preceding VAD implantation. Demographic and clinical data were collected from the electronic medical records. This study was approved by the institutional review board. Statistical analysis Patient characteristics and outcome variables were compared between the patients with and without VA. The Mann–Whitney–Wilcoxon test was used for the continuous variables, and Fisher’s exact test was used for the discrete variables. A logistic regression model was used to model the relationship between BNP and VA. A multivariate logistic regression model was fitted with multiple predictors including BNP. The predicted probability plot was used to show how probability of VA changes as BNP varies. Cox proportional hazard models and logistic models Artif Organs, Vol. 39, No. 12, 2015
1052
THOUGHTS AND PROGRESS
were performed to study the association of VA and outcomes (length of stay, 90-day readmission and survival). All analyses were performed with SAS 9.3 SAS Institute, Cary, NC27513, USA and R 2.13.1 (SAS Institute, Cary, NC, USA). Free Software Foundation, Boston, MA02110, USA. In all analyses, two-sided P values 30 days post implant) (11). Our study design targeted very early VA (within 15 days) in order to avoid the potential detection bias associated with later VA (some episodes are asymptomatic and unnoticed in patients without an ICD or monitoring device). In contrast to previous studies, we did not use ICD interrogation to detect and define the occurrence of VA. This also limited the time frame where
TABLE 1. Baseline characteristics of patients with and without VA Baseline characteristics Age (years) Gender (% male) Ischemic cardiomyopathy (%) Nonischemic cardiomyopathy (%) Diabetes mellitus (%) Hypertension (%) Coronary artery disease (%) VT in past (%) ICD (%) Amiodarone use pre LVAD (%) Elective INTERMACS 3 or 4 (%) Pre LVAD intra-aortic balloon pump (%) Maximal BNP pre LVAD (pg/mL)
Artif Organs, Vol. 39, No. 12, 2015
Total LVAD cohort (n = 74)
VA (n = 23)
No VA (n = 51)
P value
55.5 66 41 62 35 50 54 50 82 58 23 24 1640
58.6 70 35 70 26 49 56 48 87 65 26 26 2373
54.1 65 43 59 39 51 53 51 80 55 22 24 1309
0.24 0.79 0.61 0.44 0.31 1 0.81 1 0.74 0.45 0.77 1 0.0016
THOUGHTS AND PROGRESS
1053
FIG. 1. Probability of VA as a function of maximal BNP levels preceding LVAD implantation.
it is possible to detect an asymptomatic arrhythmia. Early VA post LVAD is not necessarily associated with increased mortality as it is usually promptly recognized and treated during the postoperative stay. Nevertheless, the prediction of early VA has several important implications. First, patients at high risk for VA post LVAD should be monitored beyond the ICU stay (i.e., the patient should be transferred to a telemetry bed). In patients with an ICD, the device should be activated as soon as possible and programmed with an appropriate algorithm which includes antitachycardia pacing (ATP). As stated earlier, most patients can initially tolerate the VA, and immediate DC shock is not always necessary. In fact, up to 50–74% of VA episodes were terminated by ATP in previous studies (12,13). More importantly and beyond the prediction of VA, this study demonstrates that BNP in patients with advanced heart failure is an important marker of adverse events even after implantation of an LVAD. Interestingly, a study evaluating mortality post-ICD implantation linked preimplantation BNP with an adverse prognosis, especially when BNP was greater than 1500 pg/mL (14). Several mechanisms may explain the relationship between BNP and VA post LVAD implantation. The most intuitive reason is that BNP represents the neurohormonal response to the true hemodynamic burden faced by the failing heart (beyond LV ejection fraction). Patients with very high BNP levels have an extremely diseased myocardium—the exact
substrate needed for VA. Therefore, even after hemodynamic support with the LVAD, the diseased substrate still exists. Perhaps only with prolonged LVAD support and reverse remodeling, the amount of this diseased substrate decreases. This concept of a need for prolonged LV support is reinforced by our finding that intra-aortic balloon pump use and INTERMACS level pre-LVAD implantation were not associated with lower rates of VA. Another possible explanation for the relationship between BNP and VA post LVAD is that elevated BNP levels reflect the presence of other mediators of arrhythmia (catecholamines, cytokines, others). This study has several limitations due to the expected bias and confounding factors inherent to a retrospective design. In order to potentially capture all VA events (including asymptomatic VA), the observation time was limited to 15 days. We have analyzed these data with inclusion of VA up to 60 days post LVAD implant—BNP still remained a powerful predictor of VA. Another significant factor contributing to VA during the first 15 days (and possible confounder) is the pro-arrhythmic effect of inotropes. Indeed, most of the patients with VA were at least on dual inotropes, which may reflect inadequate LVAD support or right ventricular failure. Although we did not differentiate between monomorphic and polymorphic VT, most of the VT episodes observed were monomorphic and the VF episodes were equivalent to polymorphic VT. Lastly, BNP levels in obese patients can be suppresse—in our study the mean body mass index was 28.6. Artif Organs, Vol. 39, No. 12, 2015
1054
THOUGHTS AND PROGRESS
The surgical approach in our institution was standard sternotomy with cardiopulmonary bypass. Perhaps a minimal invasive approach, such as implantation of the LVAD via thoracotomy (avoiding manipulation of the heart), has potential benefits in terms of VA (15). It is important to recognize that in some patients the VA is mediated by the genetic abnormality leading up to the cardiomyopathy or an ischemic substrate. In these patients, BNP may not predict VA as it is not solely attributed to the hemodynamic burden and severity of the cardiomyopathy. Obesity may also alter the correlation between VA and BNP due to the factitiously lower BNP levels observed in these patients. Therefore, BNP should be interpreted in the correct clinical context when assessing risk of VA.
CONCLUSIONS Our study suggests that B-type natriuretic peptide levels pre left ventricular assist device implantation may predict ventricular arrhythmia post LVAD implantation. This has not been evaluated in previous studies of VA in LVADs. The risk of VA is exponentially related to pre LVAD BNP levels with a statistical significance surpassing previously described risk factors for VA. In addition, this study underscores the adverse course associated with extremely high BNP levels in patients with heart failure. BNP levels greater than 1500 pg/mL should prompt evaluation of advanced heart failure options (LVAD and transplant), even if the patient appears “stable.” Acknowledgments: This publication was made possible by the Indiana University Health–Indiana University School of Medicine Strategic Research Initiative. Author contributions: Yaron Hellman—data collection, data analysis/interpretation, drafting article, conception, and design; Adnan S. Malik—drafting article, critical revision of article; Hongbo Lin— data analysis/interpretation; Changyu Shen—data analysis/interpretation; I-Wen Wang—critical revision of article and approval of article; Thomas C. Wozniak—critical revision of article and approval of article; Zubair A. Hashmi—critical revision of article and approval of article; Jeanette Pickrell— data collection and approval of article; Milena Jani— critical revision of article; Marco A. Caccamo— critical revision of article and approval of article; Artif Organs, Vol. 39, No. 12, 2015
Irmina Gradus-Pizlo—critical revision of article and approval of article; Azam Hadi—drafting article, critical revision of article, approval of article, conception, and design. Conflict of Interest: The authors declare no conflict of interest. REFERENCES 1. Rose EA, Gelijns AC, Moskowitz AJ, et al. Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med 2001;345:1435–43. 2. Kirklin JK, Naftel DC, Kormos RL, et al. Fifth INTERMACS annual report: risk factor analysis from more than 6000 mechanical circulatory support patients. J Heart Lung Transplant 2013;32:141–56. 3. Starling RC, Naka Y, Boyle AJ, et al. Results of the post-U.S. Food and Drug Administration-approval study with a continuous flow left ventricular assist device as a bridge to heart transplantation: a prospective study using the INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support). J Am Coll Cardiol 2011;57:1890–8. 4. Oz MC, Rose EA, Slater J, Kuiper JJ, Catanese KA, Levin HR. Malignant ventricular arrhythmias are well tolerated in patients receiving long-term left ventricular assist devices. J Am Coll Cardiol 1994;24:1688–91. 5. Ziv O, Dizon J, Thosani A, Naka Y, Magnano AR, Garan H. Effects of left ventricular assist device therapy on ventricular arrhythmias. J Am Coll Cardiol 2005;45:1428–34. 6. Berger R, Huelsman M, Strecker K, et al. B-type natriuretic peptide predicts sudden death in patients with chronic heart failure. Circulation 2002;105:2392–7. 7. Yu H, Oswald H, Gardiwal A, Lissel C, Klein G. Comparison of N-terminal pro-brain natriuretic peptide vs. electrophysiologic study for predicting future outcomes in patients with an implantable cardioverter defibrillator after myocardial infarction. Am J Cardiol 2007;100:635–9. 8. Scott PA, Barry J, Roberts PR, Morgan JM. Brain natriuretic peptide for the prediction of sudden cardiac death and ventricular arrhythmias: a meta-analysis. Eur J Heart Fail 2009;11: 958–66. 9. Budeus M, Reinsch N, Wieneke H, Sack S, Erbel R. The prediction of ICD therapy in multicenter automatic defibrillator implantation trial (MADIT) II like patients: a retrospective analysis. Indian Pacing Electrophysiol J 2008;8: 80–93. 10. Levine YC, Rosenberg MA, Mittleman M, et al. B-type natriuretic peptide is a major predictor of ventricular tachyarrhythmias. Heart Rhythm 2014;11:1109–16. 11. Shirazi JT, Lopshire JC, Gradus-Pizlo I, Hadi MA, Wozniak TC, Malik AS. Ventricular arrhythmias in patients with implanted ventricular assist devices: a contemporary review. Europace 2013;15:11–7. 12. Andersen M, Videbaek R, Boesgaard S, Sander K, Hansen PB, Gustafsson F. Incidence of ventricular arrhythmias in patients on long-term support with a continuous-flow assist device (HeartMate II). J Heart Lung Transplant 2009;28: 733–5. 13. Oswald H, Schultz-Wildelau C, Gardiwal A, et al. Implantable defibrillator therapy for ventricular tachyarrhythmia in left ventricular assist device patients. Eur J Heart Fail 2010;12: 593–9. 14. Wei S, Loyo-Berríos NI, Haigney MC, et al. Elevated B-type natriuretic peptide is associated with increased in-hospital mortality or cardiac arrest in patients undergoing implantable cardioverter-defibrillator implantation. Circ Cardiovasc Qual Outcomes 2011;4:346–54.
bs_bs_banner
THOUGHTS AND PROGRESS 15. Popov AF, Hosseini MT, Zych B, Simon AR, Bahrami T. HeartWare left ventricular assist device implantation through bilateral anterior thoracotomy. Ann Thorac Surg 2012;93: 674–6.
Banding the Right Ventricular Assist Device Outflow Conduit: Is It Really Necessary With Current Devices? *Casey Lo, †‡Shaun Gregory, †Michael Stevens, §Deirdre Murphy, and *Silvana Marasco *Sir Charles James Officer Brown Department of Cardiothoracic Surgery, §Cardiothoracic Intensive Care Unit, The Alfred Hospital, Prahran, Victoria; †Innovative Cardiovascular Engineering and Technology Laboratory, Critical Care Research Group, The Prince Charles Hospital and ‡School of Medicine, The University of Queensland, Brisbane, Queensland, Australia Abstract: Implantable left ventricular assist devices (LVADs) have been adapted clinically for right-sided mechanical circulatory support (RVAD). Previous studies on RVAD support have established the benefits of outflow cannula restriction and rotational speed reduction, and recent literature has focused on assessing either the degree of outflow cannula restriction required to simulate left-sided afterload, or the limitation of RVAD rotational speeds. Anecdotally, the utility of outflow cannula restriction has been questioned, with suggestion that banding may be unnecessary and may be replaced simply by varying the outflow conduit length. Furthermore, many patients have a high pulmonary vascular resistance (PVR) at the time of ventricular assist device (VAD) insertion that reduces with pulmonary vascular bed remodeling. It is therefore important to assess the potential changes in flow through an RVAD as PVR changes. In this in vitro study, we observed the use of dual HeartWare HVAD devices (HeartWare Inc., Framingham, MA, USA) in biventricular support (BiVAD) configuration. We assessed the pumps’ ability to maintain hemodynamic stability with and without banding; and with varying outflow cannulae length (20, 40, and 60 cm). Increased length of the outflow conduit was found to produce significantly increased afterload to the device, but this was not found to be necessary to maintain the device within the manufacturer’s recommended operational parameters under a simulated normal physiological setting of mild and severe right ventricular (RV) failure. We hypothesize that 40 cm of outflow conduit, laid down along
doi:10.1111/aor.12497 Received June 2014; revised January 2015. Address correspondence and reprint requests to Dr. Casey Lo, Sir Charles James Officer Brown Department of Cardiothoracic Surgery, The Alfred Hospital, Prahran, VIC 3181, Australia. E-mail: [email protected]
1055
the diaphragm and then up over the RV to reach the pulmonary trunk, will generate sufficient resistance to maintain normal pump function. Key Words: Right ventricular assist device—Left ventricular assist device—Biventricular assist device—Right ventricular failure—Mechanical circulatory support—Pulmonary hypertension—Pulmonary vascular resistance—Complications—Heart failure— Heart transplant—Mock circulation loop
Ventricular assist devices (VADs) implanted in the left ventricle (LVAD) have been demonstrated to provide both statistically and clinically significant improvement in the life expectancy and quality of life for end-stage heart failure patients (1). However, with the high incidence of subsequent right ventricular (RV) failure and the associated increased morbidity and mortality (2), and evidence of improved outcomes with early biventricular mechanical support (3,4), the ongoing development of mechanical biventricular assistance techniques remains pertinent. All current commercially available continuousflow centrifugal VADs for long-term implantation are designed for left-sided use (5,6). As such, their design specifications are tailored to a pump against left-sided afterload. For the purpose of RV support, implantable VADs are thus adapted to function in the setting of the lower resistance of the right side. This can either be achieved by decreasing pump speed or modification of the outflow conduit to avoid overpumping. Decreasing right ventricular assist device (RVAD) pump speed to below manufacturer’s specifications moves away from the design point for pump function, and can influence pump washout and thus thrombus formation within the pump (7). Furthermore, with hydrodynamic suspension of the impeller within the device, insufficient pump speed may lead to impeller instability (7). Previous in vitro studies on biventricular support using dual LVADs had already established the benefits of outflow cannula restriction and rotational speed reduction. Recent literature has thus focused on assessing either the degree of outflow cannula restriction required to simulate left-sided afterload, or the limitation of dual VAD rotational speeds (5). However, anecdotally, the utility of outflow cannula restriction has been questioned, with suggestion that banding may be entirely unnecessary in an in vivo setting. Although not described in the literature, there have been anecdotal suggestions that, by Poiseuille’s Law, the same outflow resistance may be achieved by increasing graft length. Furthermore, many patients have a high pulmonary vascular resistance (PVR) at the time of VAD insertion that reduces over time. It is, therefore, important to assess the potential changes in flow through an Artif Organs, Vol. 39, No. 12, 2015
THOUGHTS AND PROGRESS
B-Type Natriuretic Peptide Levels Predict Ventricular Arrhythmia Post Left Ventricular Assist Device Implantation *Yaron Hellman, *Adnan S. Malik, †Hongbo Lin, †Changyu Shen, ‡I-Wen Wang, ‡Thomas C. Wozniak, ‡Zubair A. Hashmi, *Jeanette Pickrell, *Milena Jani, *Marco A. Caccamo, *Irmina Gradus-Pizlo, and *Azam Hadi *Krannert Institute of Cardiology, †Department of Biostatistics, IU School of Medicine; and ‡IU Health Cardiovascular Surgeons, Indianapolis, IN, USA Abstract: B-type natriuretic peptide (BNP) levels have been shown to predict ventricular arrhythmia (VA) and sudden death in patients with heart failure. We sought to determine whether BNP levels before left ventricular assist device (LVAD) implantation can predict VA post LVAD implantation in advanced heart failure patients. We conducted a retrospective study consisting of patients who underwent LVAD implantation in our institution during the period of May 2009–March 2013. The study was limited to patients receiving a HeartMate II or HeartWare LVAD. Acute myocardial infarction patients were excluded. We compared between the patients who developed VA within 15 days post LVAD implantation to the patients without VA. A total of 85 patients underwent LVAD implantation during the study period. Eleven patients were excluded (five acute MI, four without BNP measurements, and two discharged earlier than 13 days post LVAD implantation). The incidence of VA was 31%, with 91% ventricular tachycardia (VT) and 9% ventricular fibrillation. BNP remained the single most powerful predictor of VA even after adjustment for other borderline significant factors in a multivariate logistic regression model (P < 0.05). BNP levels are a strong predictor of VA post LVAD implantation, surpassing previously described risk factors such as age and VT in the past. Key Words: B-type natriuretic peptide—Left ventricular assist device—Ventricular arrhythmia— Ventricular tachycardia—Ventricular fibrillation.
Left ventricular assist devices (LVADs) provide improved long-term survival and quality of life in appropriately selected patients with advanced heart failure (1–3). However, sustained ventricular arrhythmias (VAs) in patients with continuous flow LVADs may result in eventual hemodynamic compromise due to the unsupported right ventricle and lack of preload (4,5).
doi:10.1111/aor.12486 Received October 2014; revised November 2014. Address correspondence and reprint requests to Dr. Azam Hadi, Krannert Institute of Cardiology, IU School of Medicine, IU Health Methodist Professional Center 2, 1801 N. Senate Boulevard, Suite 2000, Indianapolis, IN 46202, USA. E-mail: mazhadi @iu.edu
1051
B-type natriuretic peptide (BNP) has been evaluated as a predictor of sudden cardiac death and VA in the heart failure population, including patients with heart failure post myocardial infarction and implantable cardioverter defibrillator (ICD) recipients (6–8). Elevated BNP levels are specifically associated with an increased rate of ICD appropriate therapy and VA burden (9,10). The majority of these studies did not specifically include patients with advanced heart failure and severe LV systolic dysfunction requiring VAD or transplant. We hypothesized that BNP can aid in the prediction of VA post LVAD implantation. PATIENTS AND METHODS We conducted a retrospective study consisting of patients who underwent LVAD implantation in our institution during the period of May 2009–March 2013. The study was limited to patients receiving a HeartMate II (Thoratec, Pleasanton, CA, USA) or HeartWare LVAD (Framingham, MA, USA). Patients with acute myocardial infarction, active myocarditis, or who did not have BNP measurements were excluded. VA was defined as ventricular fibrillation (VF) or any ventricular tachycardia (VT— including asymptomatic nonsustained VT), which was documented in the electronic medical record within 15 days post LVAD implantation. The timeframe of 15 days was selected as the patients were in hospital and all were continuously monitored with telemetry. Analyses were done to compare the patients who developed VA with the patients without VA. BNP levels were defined as the maximal BNP level measured in the Indiana University Health Methodist hospital laboratory during the 4 months preceding VAD implantation. Demographic and clinical data were collected from the electronic medical records. This study was approved by the institutional review board. Statistical analysis Patient characteristics and outcome variables were compared between the patients with and without VA. The Mann–Whitney–Wilcoxon test was used for the continuous variables, and Fisher’s exact test was used for the discrete variables. A logistic regression model was used to model the relationship between BNP and VA. A multivariate logistic regression model was fitted with multiple predictors including BNP. The predicted probability plot was used to show how probability of VA changes as BNP varies. Cox proportional hazard models and logistic models Artif Organs, Vol. 39, No. 12, 2015
1052
THOUGHTS AND PROGRESS
were performed to study the association of VA and outcomes (length of stay, 90-day readmission and survival). All analyses were performed with SAS 9.3 SAS Institute, Cary, NC27513, USA and R 2.13.1 (SAS Institute, Cary, NC, USA). Free Software Foundation, Boston, MA02110, USA. In all analyses, two-sided P values 30 days post implant) (11). Our study design targeted very early VA (within 15 days) in order to avoid the potential detection bias associated with later VA (some episodes are asymptomatic and unnoticed in patients without an ICD or monitoring device). In contrast to previous studies, we did not use ICD interrogation to detect and define the occurrence of VA. This also limited the time frame where
TABLE 1. Baseline characteristics of patients with and without VA Baseline characteristics Age (years) Gender (% male) Ischemic cardiomyopathy (%) Nonischemic cardiomyopathy (%) Diabetes mellitus (%) Hypertension (%) Coronary artery disease (%) VT in past (%) ICD (%) Amiodarone use pre LVAD (%) Elective INTERMACS 3 or 4 (%) Pre LVAD intra-aortic balloon pump (%) Maximal BNP pre LVAD (pg/mL)
Artif Organs, Vol. 39, No. 12, 2015
Total LVAD cohort (n = 74)
VA (n = 23)
No VA (n = 51)
P value
55.5 66 41 62 35 50 54 50 82 58 23 24 1640
58.6 70 35 70 26 49 56 48 87 65 26 26 2373
54.1 65 43 59 39 51 53 51 80 55 22 24 1309
0.24 0.79 0.61 0.44 0.31 1 0.81 1 0.74 0.45 0.77 1 0.0016
THOUGHTS AND PROGRESS
1053
FIG. 1. Probability of VA as a function of maximal BNP levels preceding LVAD implantation.
it is possible to detect an asymptomatic arrhythmia. Early VA post LVAD is not necessarily associated with increased mortality as it is usually promptly recognized and treated during the postoperative stay. Nevertheless, the prediction of early VA has several important implications. First, patients at high risk for VA post LVAD should be monitored beyond the ICU stay (i.e., the patient should be transferred to a telemetry bed). In patients with an ICD, the device should be activated as soon as possible and programmed with an appropriate algorithm which includes antitachycardia pacing (ATP). As stated earlier, most patients can initially tolerate the VA, and immediate DC shock is not always necessary. In fact, up to 50–74% of VA episodes were terminated by ATP in previous studies (12,13). More importantly and beyond the prediction of VA, this study demonstrates that BNP in patients with advanced heart failure is an important marker of adverse events even after implantation of an LVAD. Interestingly, a study evaluating mortality post-ICD implantation linked preimplantation BNP with an adverse prognosis, especially when BNP was greater than 1500 pg/mL (14). Several mechanisms may explain the relationship between BNP and VA post LVAD implantation. The most intuitive reason is that BNP represents the neurohormonal response to the true hemodynamic burden faced by the failing heart (beyond LV ejection fraction). Patients with very high BNP levels have an extremely diseased myocardium—the exact
substrate needed for VA. Therefore, even after hemodynamic support with the LVAD, the diseased substrate still exists. Perhaps only with prolonged LVAD support and reverse remodeling, the amount of this diseased substrate decreases. This concept of a need for prolonged LV support is reinforced by our finding that intra-aortic balloon pump use and INTERMACS level pre-LVAD implantation were not associated with lower rates of VA. Another possible explanation for the relationship between BNP and VA post LVAD is that elevated BNP levels reflect the presence of other mediators of arrhythmia (catecholamines, cytokines, others). This study has several limitations due to the expected bias and confounding factors inherent to a retrospective design. In order to potentially capture all VA events (including asymptomatic VA), the observation time was limited to 15 days. We have analyzed these data with inclusion of VA up to 60 days post LVAD implant—BNP still remained a powerful predictor of VA. Another significant factor contributing to VA during the first 15 days (and possible confounder) is the pro-arrhythmic effect of inotropes. Indeed, most of the patients with VA were at least on dual inotropes, which may reflect inadequate LVAD support or right ventricular failure. Although we did not differentiate between monomorphic and polymorphic VT, most of the VT episodes observed were monomorphic and the VF episodes were equivalent to polymorphic VT. Lastly, BNP levels in obese patients can be suppresse—in our study the mean body mass index was 28.6. Artif Organs, Vol. 39, No. 12, 2015
1054
THOUGHTS AND PROGRESS
The surgical approach in our institution was standard sternotomy with cardiopulmonary bypass. Perhaps a minimal invasive approach, such as implantation of the LVAD via thoracotomy (avoiding manipulation of the heart), has potential benefits in terms of VA (15). It is important to recognize that in some patients the VA is mediated by the genetic abnormality leading up to the cardiomyopathy or an ischemic substrate. In these patients, BNP may not predict VA as it is not solely attributed to the hemodynamic burden and severity of the cardiomyopathy. Obesity may also alter the correlation between VA and BNP due to the factitiously lower BNP levels observed in these patients. Therefore, BNP should be interpreted in the correct clinical context when assessing risk of VA.
CONCLUSIONS Our study suggests that B-type natriuretic peptide levels pre left ventricular assist device implantation may predict ventricular arrhythmia post LVAD implantation. This has not been evaluated in previous studies of VA in LVADs. The risk of VA is exponentially related to pre LVAD BNP levels with a statistical significance surpassing previously described risk factors for VA. In addition, this study underscores the adverse course associated with extremely high BNP levels in patients with heart failure. BNP levels greater than 1500 pg/mL should prompt evaluation of advanced heart failure options (LVAD and transplant), even if the patient appears “stable.” Acknowledgments: This publication was made possible by the Indiana University Health–Indiana University School of Medicine Strategic Research Initiative. Author contributions: Yaron Hellman—data collection, data analysis/interpretation, drafting article, conception, and design; Adnan S. Malik—drafting article, critical revision of article; Hongbo Lin— data analysis/interpretation; Changyu Shen—data analysis/interpretation; I-Wen Wang—critical revision of article and approval of article; Thomas C. Wozniak—critical revision of article and approval of article; Zubair A. Hashmi—critical revision of article and approval of article; Jeanette Pickrell— data collection and approval of article; Milena Jani— critical revision of article; Marco A. Caccamo— critical revision of article and approval of article; Artif Organs, Vol. 39, No. 12, 2015
Irmina Gradus-Pizlo—critical revision of article and approval of article; Azam Hadi—drafting article, critical revision of article, approval of article, conception, and design. Conflict of Interest: The authors declare no conflict of interest. REFERENCES 1. Rose EA, Gelijns AC, Moskowitz AJ, et al. Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med 2001;345:1435–43. 2. Kirklin JK, Naftel DC, Kormos RL, et al. Fifth INTERMACS annual report: risk factor analysis from more than 6000 mechanical circulatory support patients. J Heart Lung Transplant 2013;32:141–56. 3. Starling RC, Naka Y, Boyle AJ, et al. Results of the post-U.S. Food and Drug Administration-approval study with a continuous flow left ventricular assist device as a bridge to heart transplantation: a prospective study using the INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support). J Am Coll Cardiol 2011;57:1890–8. 4. Oz MC, Rose EA, Slater J, Kuiper JJ, Catanese KA, Levin HR. Malignant ventricular arrhythmias are well tolerated in patients receiving long-term left ventricular assist devices. J Am Coll Cardiol 1994;24:1688–91. 5. Ziv O, Dizon J, Thosani A, Naka Y, Magnano AR, Garan H. Effects of left ventricular assist device therapy on ventricular arrhythmias. J Am Coll Cardiol 2005;45:1428–34. 6. Berger R, Huelsman M, Strecker K, et al. B-type natriuretic peptide predicts sudden death in patients with chronic heart failure. Circulation 2002;105:2392–7. 7. Yu H, Oswald H, Gardiwal A, Lissel C, Klein G. Comparison of N-terminal pro-brain natriuretic peptide vs. electrophysiologic study for predicting future outcomes in patients with an implantable cardioverter defibrillator after myocardial infarction. Am J Cardiol 2007;100:635–9. 8. Scott PA, Barry J, Roberts PR, Morgan JM. Brain natriuretic peptide for the prediction of sudden cardiac death and ventricular arrhythmias: a meta-analysis. Eur J Heart Fail 2009;11: 958–66. 9. Budeus M, Reinsch N, Wieneke H, Sack S, Erbel R. The prediction of ICD therapy in multicenter automatic defibrillator implantation trial (MADIT) II like patients: a retrospective analysis. Indian Pacing Electrophysiol J 2008;8: 80–93. 10. Levine YC, Rosenberg MA, Mittleman M, et al. B-type natriuretic peptide is a major predictor of ventricular tachyarrhythmias. Heart Rhythm 2014;11:1109–16. 11. Shirazi JT, Lopshire JC, Gradus-Pizlo I, Hadi MA, Wozniak TC, Malik AS. Ventricular arrhythmias in patients with implanted ventricular assist devices: a contemporary review. Europace 2013;15:11–7. 12. Andersen M, Videbaek R, Boesgaard S, Sander K, Hansen PB, Gustafsson F. Incidence of ventricular arrhythmias in patients on long-term support with a continuous-flow assist device (HeartMate II). J Heart Lung Transplant 2009;28: 733–5. 13. Oswald H, Schultz-Wildelau C, Gardiwal A, et al. Implantable defibrillator therapy for ventricular tachyarrhythmia in left ventricular assist device patients. Eur J Heart Fail 2010;12: 593–9. 14. Wei S, Loyo-Berríos NI, Haigney MC, et al. Elevated B-type natriuretic peptide is associated with increased in-hospital mortality or cardiac arrest in patients undergoing implantable cardioverter-defibrillator implantation. Circ Cardiovasc Qual Outcomes 2011;4:346–54.
bs_bs_banner
THOUGHTS AND PROGRESS 15. Popov AF, Hosseini MT, Zych B, Simon AR, Bahrami T. HeartWare left ventricular assist device implantation through bilateral anterior thoracotomy. Ann Thorac Surg 2012;93: 674–6.
Banding the Right Ventricular Assist Device Outflow Conduit: Is It Really Necessary With Current Devices? *Casey Lo, †‡Shaun Gregory, †Michael Stevens, §Deirdre Murphy, and *Silvana Marasco *Sir Charles James Officer Brown Department of Cardiothoracic Surgery, §Cardiothoracic Intensive Care Unit, The Alfred Hospital, Prahran, Victoria; †Innovative Cardiovascular Engineering and Technology Laboratory, Critical Care Research Group, The Prince Charles Hospital and ‡School of Medicine, The University of Queensland, Brisbane, Queensland, Australia Abstract: Implantable left ventricular assist devices (LVADs) have been adapted clinically for right-sided mechanical circulatory support (RVAD). Previous studies on RVAD support have established the benefits of outflow cannula restriction and rotational speed reduction, and recent literature has focused on assessing either the degree of outflow cannula restriction required to simulate left-sided afterload, or the limitation of RVAD rotational speeds. Anecdotally, the utility of outflow cannula restriction has been questioned, with suggestion that banding may be unnecessary and may be replaced simply by varying the outflow conduit length. Furthermore, many patients have a high pulmonary vascular resistance (PVR) at the time of ventricular assist device (VAD) insertion that reduces with pulmonary vascular bed remodeling. It is therefore important to assess the potential changes in flow through an RVAD as PVR changes. In this in vitro study, we observed the use of dual HeartWare HVAD devices (HeartWare Inc., Framingham, MA, USA) in biventricular support (BiVAD) configuration. We assessed the pumps’ ability to maintain hemodynamic stability with and without banding; and with varying outflow cannulae length (20, 40, and 60 cm). Increased length of the outflow conduit was found to produce significantly increased afterload to the device, but this was not found to be necessary to maintain the device within the manufacturer’s recommended operational parameters under a simulated normal physiological setting of mild and severe right ventricular (RV) failure. We hypothesize that 40 cm of outflow conduit, laid down along
doi:10.1111/aor.12497 Received June 2014; revised January 2015. Address correspondence and reprint requests to Dr. Casey Lo, Sir Charles James Officer Brown Department of Cardiothoracic Surgery, The Alfred Hospital, Prahran, VIC 3181, Australia. E-mail: [email protected]
1055
the diaphragm and then up over the RV to reach the pulmonary trunk, will generate sufficient resistance to maintain normal pump function. Key Words: Right ventricular assist device—Left ventricular assist device—Biventricular assist device—Right ventricular failure—Mechanical circulatory support—Pulmonary hypertension—Pulmonary vascular resistance—Complications—Heart failure— Heart transplant—Mock circulation loop
Ventricular assist devices (VADs) implanted in the left ventricle (LVAD) have been demonstrated to provide both statistically and clinically significant improvement in the life expectancy and quality of life for end-stage heart failure patients (1). However, with the high incidence of subsequent right ventricular (RV) failure and the associated increased morbidity and mortality (2), and evidence of improved outcomes with early biventricular mechanical support (3,4), the ongoing development of mechanical biventricular assistance techniques remains pertinent. All current commercially available continuousflow centrifugal VADs for long-term implantation are designed for left-sided use (5,6). As such, their design specifications are tailored to a pump against left-sided afterload. For the purpose of RV support, implantable VADs are thus adapted to function in the setting of the lower resistance of the right side. This can either be achieved by decreasing pump speed or modification of the outflow conduit to avoid overpumping. Decreasing right ventricular assist device (RVAD) pump speed to below manufacturer’s specifications moves away from the design point for pump function, and can influence pump washout and thus thrombus formation within the pump (7). Furthermore, with hydrodynamic suspension of the impeller within the device, insufficient pump speed may lead to impeller instability (7). Previous in vitro studies on biventricular support using dual LVADs had already established the benefits of outflow cannula restriction and rotational speed reduction. Recent literature has thus focused on assessing either the degree of outflow cannula restriction required to simulate left-sided afterload, or the limitation of dual VAD rotational speeds (5). However, anecdotally, the utility of outflow cannula restriction has been questioned, with suggestion that banding may be entirely unnecessary in an in vivo setting. Although not described in the literature, there have been anecdotal suggestions that, by Poiseuille’s Law, the same outflow resistance may be achieved by increasing graft length. Furthermore, many patients have a high pulmonary vascular resistance (PVR) at the time of VAD insertion that reduces over time. It is, therefore, important to assess the potential changes in flow through an Artif Organs, Vol. 39, No. 12, 2015
