Heart failure is a major public health problem affecting approximately 5.7 million Americans, and despite medical treatment advances, the prognoses at 1 year and 5 years remain poor.1 It is estimated that approximately 40,000 to 200,000 patients could benefit from heart replacement therapy annually,2 but because of the limited availability of donor hearts, transplantations are limited to approximately 2,000 patients annually in the United States.3 Left ventricular assist devices (LVADs) have been proven to increase survival in patients with end-stage systolic heart failure. Recently, continuous-flow LVADs (cfLVADs) have been shown to provide improved survival with better durability and fewer device-related complications when compared with pulsatile pumps (XVE, Thoratec paracorporeal ventricular assist device or implantable ventricular assist device).4,5 As a result, the number of patients with heart failure being treated with cfLVADs as bridge-to-transplantation (BTT) or destination therapy (DT) is increasing rapidly.6,7
Unlike pulsatile LVADs, in which pump filling and ejection are determined, in part, by patient physiology, cfLVADs are limited by a fixed speed and their inability to increase flow through the pump during exercise, producing a differential pressure as a function of pump speed whereas flow is dependent upon loading conditions. Accordingly, current cfLVADs do not adjust to changes in physiologic demand. Therefore, increases in oxygen delivery for exercise can only be achieved by either expanding the arteriovenous oxygen difference or increasing the cardiac output via intrinsic left ventricular work (ventricular ejection), which may be limited in the setting of end-stage heart failure. This is of little concern in the BTT application, where survival and preservation of organ function are the ultimate therapeutic goals. However, in DT, as patients are leaving the hospital and returning to their daily activities, functional capacity and quality of life are important. It has been observed that despite similar quality of life and health satisfaction scores, patients with HeartMate II support have diminished peak oxygen consumption (peak VO2) compared with patients who had heart transplant.8
There are few reports of exercise capacity in patients with cfLVADs.9–14 These reports are from a relatively young patient population that is typically implanted with cfLVADs as BTT. It is not clear whether recovery of heart function during support, residual heart function, or peripheral mechanisms allows better exercise capacity in a younger, BTT cohort, than would be seen with a population that is older and included patients receiving DT. The objective of this study was to determine peak aerobic capacity in heart failure patients with a cfLVAD before and after device implantation.
Data were collected from all patients with cfLVADs at Penn State Hershey Medical that had cardiopulmonary exercise testing measured at any time from December 2002 to December 2011, which represented 60 consecutive patients. We performed a retrospective analysis of exercise in patients with cfLVADs using a cycle ergometer (Ergoline ViaSprint 150P [CareFusion, San Diego, CA] or General Electric Case V6.51 EKG [GE Healthcare, Waukesha, WI]). All patients who underwent exercise testing did so at a time when their medical therapy and LVAD support were in a stable and optimally treated condition according to clinical, LVAD, and echocardiographic data. Before testing, 2 minutes of gas exchange data were collected during seated rest. Exercise was performed to a symptom-limited maximum using a 10 watts/min ramp protocol with a goal of at least 6 minutes of exercise. The ramp was reduced to 5 watts/min for patients with limited exercise capacity. Oxygen consumption (VO2), carbon dioxide production (VCO2), minute ventilation (VE), and respiratory exchange ratio (RER = VCO2/VO2) were measured at rest on a breath-to-breath basis using a SensorMedics VMax Encore (Model E29C, software version 20-5B [CareFusion, San Diego, CA]) metabolic cart. Additional measurements made during the exercise test included heart rate (HR), peak power, oxygen saturation via pulse oximetry, tidal volume, and respiratory rate. Cardiopulmonary exercise variables were interpreted using tabular data averaged every 20 seconds. Peak exercise values were the highest interval value during the last minute of exercise. Anaerobic threshold was derived using the V-slope method.14 Parameters extracted from the exercise test were forced vital capacity, forced exhaled volume in 1 second, peak VO2, percentage of the predicted peak VO2, maximum VE, ventilatory efficiency (i.e., VE/VCO2), and breathing reserve.
Unpaired patient data were separated into preimplant (n = 25), 3–6 months (n = 31), 6 months to 1 year (n = 16), and >1 year (n = 10) time intervals. For patients who completed multiple exercise tests, preimplant, the test closest to implantation was used. For all other time intervals, the average value was used. Data for each time period for each data element were reported as average value and standard deviation (SD). Statistical analysis was done using standard two-tailed, unpaired t-tests for each data element by comparing the data set for the preimplant individually with each data set for the other three time periods (3–6 months, 6 months to 1 year, and >1 year). Analysis of variance on ranks was performed for the peak VO2, percentage of the predicted peak VO2, and the peak VO2 per kilogram.
Data from nine patients who had preimplant and the 3–6 month test data were used for paired analysis. Statistical analysis using two-tailed, paired t-tests was performed to compare peak VO2, percentage of the predicted peak VO2, and VE/VCO2 at 3–6 months with preimplant data.
Subgroup analysis was done for both age (<60 years old vs. ≥60 years old) and indication (BTT or DT) for VO2 max per kilogram for each time period, and was reported as average value and SD. Statistical analysis was done using standard t-tests for comparing the data set for the preimplant individually with each data set for the other three time periods for each subgroup and for comparing each subgroup at each time period.
The data included in this analysis were obtained in the course of routine care. The study was approved by the institutional review board of the Penn State Hershey Medical Center.
Patient demographic data are displayed in Table 1. Patients were predominantly implanted with a HeartMate II cfLVAD, and the percentage of patients supported with HeartMate II increased at longer time points (100% of patients were supported with HeartMate II at >1 year support). Approximately one third were implanted for DT, but the percentage increased over time and at >1 year, 90% were DT.
The complete exercise test results for each time period are shown in Table 2. Figure 1A shows the exercise capacity in patients before implant and after implant at 3–6 months, 1 year, and >1 year. There was a slight increase in peak VO2 at 3–6 months that was not statistically significant. At 1 year and at >1 year, peak VO2 decreased, but there was no significant difference when compared with preimplant. To account for differences in patient age, sex, and body weight between the time points, exercise capacity as percentage of the predicted peak VO2 is shown in Figure 1B. At all time points, there was an increase in the percentage of the predicted peak VO2 that reached significance at 3–6 months and >1 year.
Ventilatory efficiency at anaerobic threshold improved after LVAD implantation over time, and after 1 year postimplantation, there was a significant improvement when compared with preimplant (Figure 1C). The mean RER was greater than 1.1 at all time points, indicating that sufficient effort was exerted during the exercise testing (Table 2).
The subgroup analysis results for age and indication are shown in Table 3. Patients younger than 60 years had higher VO2 at preimplant and 3–6 months postimplant when compared with patients older than 60 years. However, the results did not reach statistical significance. BTT patients also had a greater exercise capacity at preimplant and 3–6 months postimplant when compared with DT patients, but the results were not statistical significant.
The exercise testing results for patients with paired results at preimplant and 3–6 months postimplant are shown in Figure 2. The mean peak VO2 and the mean percentage of the predicted peak VO2 for the paired patient population (10.1 ± 3.4 ml/kg/min and 34.2 ± 7.4%, respectively, at preimplant; 13.4 ± 5.5 ml/kg/min and 43.9 ± 11.1%, respectively, at 3–6 months) were not significantly different from the cumulative data in Table 2. However, there was a significant increase in peak VO2 and the percentage of the predicted peak VO2 at 3–6 months for the paired patient population (Figures 2, A and B). Eight of the nine patients had an increase in peak VO2 and percentage of the predicted peak VO2. Ventilatory efficiency for the paired population (44.1 ± 9.2 and 40.6 ± 7.2 at preimplant and 3–6 months, respectively) was not significantly different from the cumulative data in Table 2, and there was no significant improvement in ventilator efficiency at 3–6 months (Figure 2C).
Our retrospective analysis of patients with cfLVADS found that despite the established relief from heart failure symptoms and improved quality of life, there was little quantifiable improvement in exercise capacity. The data presented here are unique in that the population studied was older and included more patients implanted with cfLVADS as DT. In “Clinician’s Guide to Cardiopulmonary Exercise Testing,”19 put forth by the American Heart Association, Balady et al. state that for patients with heart failure, the criteria for severe functional limitation and poor prognosis are peak VO2 ≤ 14 ml/kg/min, percentage of the predicted peak VO2 ≤50%, and VE/VCO2 ≥34. In our patient population, there was no improvement in peak VO2 at 1 year postimplant, and although percentage of the predicted peak VO2 did improve at 1 year, it remained at the threshold for severe functional limitation. Likewise, ventilatory efficiency improved after 1 year but remained higher than the recommended threshold for severe functional limitations. In our limited paired patient population, only two of the nine patients exceeded the stated thresholds for peak VO2 and percentage of the predicted peak VO2 at 3–6 months postimplant. Only one of these patients had improved ventilatory efficiency at 3–6 months postimplant below the recommended threshold.
During normal exercise, a number of physiologic changes occur. Venous constriction and other venous return mechanisms lead to an increase in preload. Through increased HR, combined with the Frank-Starling mechanism and increased inotropy, cardiac output increases. In regard to support with a cfLVAD, the flow through the pump is dependent on the speed of the pump, the inlet pressure presented from the left ventricle, and the systemic resistance distal to the pump. During exercise with a cfLVAD operating at a fixed speed, a modest increase in pump flow may occur that can be attributable to increased preload from the left ventricle and increased ratio of systole to diastole as HR increases.16,17 However, the cfLVAD response is limited when compared with the Frank-Starling response of a healthy ventricle.18 In patients with a moderate amount of intrinsic heart function, the native left ventricle may contribute significantly to the cardiac output, both by increasing cfLVAD inlet pressure through the pump and by direct ejection through the aortic valve. In an advanced failing heart that cannot eject directly through the aortic valve, blood flow during exercise is limited to what is delivered by the pump. In this patient population, a preload-sensitive control algorithm may be necessary to improve pump flow during exercise.18,19 In a limited patient population,16 an automatic control system that increased pump speed during exercise led to increased pump flow, reduced pulmonary capillary wedge pressure, and improved exercise capacity.
Previous reports of exercise capacity in patients with cfLVADs are mostly from a relatively young patient population that is typically implanted as BTT.9–14 Our cohort of patients included many patients implanted for DT and their average age was older than that in previous reports. Jakovljevic et al.12,13 reported a peak VO2 of 19.8 ± 5.8 ml/kg/min for patients supported with cfLVAD at the Harefield Hospital, which is much higher than that in our patient population. In addition, the ventilatory efficiency during cfLVAD support in that study (30.9 ± 9.1) was much improved compared with our study. However, the patients in the study by Jakovljevic et al.12,13 were much younger (39 ± 14 years) than our patient population and had a high incidence of myocardial recovery with the mean left ventricular ejection fraction of the group being 50 ± 8%. Indeed, the preimplant peak VO2 in the heart failure cohort from that study was 15.6 ± 4.7 ml/kg/min, which is significantly higher than the preimplant baseline in our study. The discrepancy in exercise tolerance between our patient population and that in the study by Jakovljevic et al.12,13 is most likely a result of increased residual cardiac function in the latter group.
In a small cohort of patients closer to the age of our population (50.1 ± 13.2 years), Jacquet et al.11 reported a peak VO2 during cfLVAD support that was more similar to the peak VO2 in our study (15.8 ± 6.2 ml/kg/min).11 In addition, they found that changes in pump flow had a limited impact on total cardiac output and that residual cardiac function accounted for increased flow. Accordingly, they found a proportional decrease in cardiac output during exercise with increasing patient age. Kugler et al.8 reported in a study of patients slightly younger than our population (47 ± 13 years) a 45% predicted peak VO2 at 6 weeks, which is similar to our findings.8 In our patient cohort, patients receiving BTT and younger patients had increased exercise capacity at preimplantation and 3–6 months postimplantation when compared with patients receiving DT and older patients, respectively. However, the increased exercise capacity was not statistically significant, which may be attributable to the small sample size. Alternatively, it may be attributable to the overall advanced age and poor cardiac function of our entire patient population.
The lack of improvement in exercise capacity in our patient cohort may be attributed to age and minimal residual cardiac function. A control system that can sense changes in demand and respond with a Starling-type mechanism, similar to a native ventricle, may be beneficial to patients undergoing DT with very little residual cardiac function.18
There are limitations in our study that must be addressed. The study was an observational, retrospective analysis that had many missing data points and a small patient population. However, including data from a consecutive series of patients helps to avoid selection bias. Another shortcoming of this data is the lack of a fully paired data set, which is attributed in part to the large number of patients not well enough for exercise testing before implantation with a median INTERMACS profile of 2. Although there are only a limited number of paired data sets, they were consistent with the other findings. Therefore, we believe that our results are unlikely to be attributed to the unpaired data analysis and are a true representation of our patient population. In addition, peak aerobic capacity was poor after cfLVAD implantation, regardless of preimplant values.
Only patients healthy enough to tolerate exercise testing were studied at preimplant. This is reflected in the low representation of patients undergoing DT before implant that increases after implant. It is expected that the pre-LVAD exercise tolerance is actually lower in our patient population that we report and that some exercise improvement may have occurred after LVAD implantation, particularly in the sickest patients. However, the post-LVAD improvement increased exercise performance neither to the levels reported in previous studies8–13 nor to the critical values recommended by Balady et al.15
The data are from a single-center experience and may not be applicable to other centers that may differ in patient population and care management. Exercise studies rely on consistent patient effort that may be difficult to ensure, especially when dealing with unpaired patient populations. In our study, the mean RER was greater than 1.1 at all time points, indicating that sufficient effort was exerted during the exercise testing.15 In addition, there was no significant difference in RER postimplant compared with RER preimplant; and therefore, the lack of improvement in peak VO2 cannot be attributed to reduced effort.
In addition, the INTERMACS database was queried for all patient data with VO2 values recorded at any time period. The Data Access, Analysis, and Publications Committee of INTERMACS believed that the frequency of the data requested was very low and unlikely to be of sufficient frequency to provide meaningful data analysis, which was more evident with paired assessments.
In our cohort of patients with cfLVAD who were older with more patients undergoing DT than those previously reported in the literature, peak VO2 during exercise was limited. For our patients, the maximum achieved VO2 was 12.7 ml/kg/min at the 3–6 month period after implantation, but it dropped back to 11.2 ml/kg/min after 1 year, which is consistent with the aerobic capacity seen in advanced heart failure. We hypothesize that the mechanism for this is the limited ability of a continuous-flow pump at a fixed speed to increase flow during exercise combined with a population with limited residual cardiac function. The mechanisms, treatments, and clinical implications of these findings will require further study.
1. Roger VL, Go AS, Lloyd-Jones DM, et al.American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Executive summary: Heart disease and stroke statistics–2012 update: A report from the American Heart Association. Circulation. 2012;125:188–197
2. Miller LW. Left ventricular assist devices are underutilized. Circulation. 2011;123:1552–1558 discussion 1558
3. . Organ procurement and transplantation network and scientific registry of transplant recipients 2010 data report. Am J Transplant. 2012;12(Suppl 1):1–156
4. Miller LW, Pagani FD, Russell SD, et al.HeartMate II Clinical Investigators. Use of a continuous-flow device in patients awaiting heart transplantation. N Engl J Med. 2007;357:885–896
5. 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
6. Krishnamani R, DeNofrio D, Konstam MA. Emerging ventricular assist devices for long-term cardiac support. Nat Rev Cardiol. 2010;7:71–76
7. Terracciano CM, Miller LW, Yacoub MH. Contemporary use of ventricular assist devices. Annu Rev Med. 2010;61:255–270
8. Kugler C, Malehsa D, Tegtbur U, et al. Health-related quality of life and exercise tolerance in recipients of heart transplants and left ventricular assist devices: A prospective, comparative study. J Heart Lung Transplant. 2011;30:204–210
9. de Jonge N, Kirkels H, Lahpor JR, et al. Exercise performance in patients with end-stage heart failure after implantation of a left ventricular assist device and after heart transplantation: an outlook for permanent assisting? J Am Coll Cardiol. 2001;37:1794–1799
10. Haft J, Armstrong W, Dyke DB, et al. Hemodynamic and exercise performance with pulsatile and continuous-flow left ventricular assist devices. Circulation. 2007;116(11 Suppl):I8–15
11. Jacquet L, Vancaenegem O, Pasquet A, et al. Exercise capacity in patients supported with rotary blood pumps is improved by a spontaneous increase of pump flow at constant pump speed and by a rise in native cardiac output. Artif Organs. 2011;35:682–690
12. Jakovljevic DG, George RS, Donovan G, et al. Comparison of cardiac power output and exercise performance in patients with left ventricular assist devices, explanted (recovered) patients, and those with moderate to severe heart failure. Am J Cardiol. 2010;105:1780–1785
13. Jakovljevic DG, George RS, Nunan D, et al. The impact of acute reduction of continuous-flow left ventricular assist device support on cardiac and exercise performance. Heart. 2010;96:1390–1395
14. Sue DY, Wasserman K, Moricca RB, Casaburi R. Metabolic acidosis during exercise in patients with chronic obstructive pulmonary disease. Use of the V-slope method for anaerobic threshold determination. Chest. 1988;94:931–938
15. Balady GJ, Arena R, Sietsema K, et al. Clinician’s guide to cardiopulmonary exercise testing in adults: A scientific statement from the American Heart Association. Circulation. 2010;122:191–225
16. Akimoto T, Yamazaki K, Litwak P, et al. Rotary blood pump flow spontaneously increases during exercise under constant pump speed: Results of a chronic study. Artif Organs. 1999;23:797–801
17. Schima H, Vollkron M, Jantsch U, et al. First clinical experience with an automatic control system for rotary blood pumps during ergometry and right-heart catheterization. J Heart Lung Transplant. 2006;25:167–173
18. Salamonsen RF, Mason DG, Ayre PJ. Response of rotary blood pumps to changes in preload and afterload at a fixed speed setting are unphysiological when compared with the natural heart. Artif Organs. 2011;35:E47–E53
19. Vollkron M, Schima H, Huber L, Benkowski R, Morello G, Wieselthaler G. Development of a reliable automatic speed control system for rotary blood pumps. J Heart Lung Transplant. 2005;24:1878–1885
Keywords:Copyright © 2013 by the American Society for Artificial Internal Organs
ventricular assist device; exercise capacity; oxygen consumption; rotary blood pump