ASAIO Journal 2015
Adult Circulatory Support
Left Ventricular Assist Devices Improve Functional Class without Normalizing Peak Oxygen Consumption Casey R. Benton,* Gabriel Sayer,† Ajith P. Nair,‡ Kimberly Ashley,‡ Michael J. Domanski,‡ Milena J. Henzlova,‡ Anelechi C. Anyanwu,§ and Sean P. Pinney‡
Heart failure patients supported with left ventricular assist devices (LVAD) enjoy improvements in functional capacity and quality of life. We reasoned that such improvements in exercise capacity should be reflected in an objective increase in peak oxygen consumption as measured by cardiopulmonary exercise testing (CPET). We performed a retrospective review of all recipients of a HeartMate II LVAD at our center from June 2009 to June 2012 who completed CPET. Thirtyseven patients completed CPET an average of 6 months after implantation. Of these, 10 patients had CPET performed before LVAD implantation. Overall, 91.4% of patients improved by at least two New York Heart Association classes, with 34.3% improving by three classes. Postimplant VO2 max was significantly less than predicted (14.7 ± 3.1 vs. 29.8 ± 6.6 ml/kg/min, p < 0.001; percent-predicted 51% ± 12%). For 10 patients with pre- and post-implant studies, VO2 max increased significantly from 11.6 ± 5.0 to 15.4 ± 3.9 ml/kg/min (p = 0.009). VO2 max improves significantly with LVAD support but fails to normalize to predicted values, in spite of improvements in functional class. The severity of preimplantation heart failure does not associate with the degree of VO2 max improvement. ASAIO Journal 2015; 61:237–243.
in their functional capacity and quality of life with over 50% of patients improving to New York Heart Association (NYHA) functional class I or II within 1 month after implantation and over 80% achieving this by 6 months. These improvements are sustained for at least 2 years after implantation.5,6 We hypothesized that such subjective increases in functional capacity should be associated with objective improvements in peak oxygen consumption (VO2 max), as measured by cardiopulmonary exercise testing (CPET). We also hypothesized that patients who were more critically ill at the time of implantation would not recover to the same extent as those with lesser critical presentations. Methods We performed a retrospective review of all patients implanted with a HeartMate II (Thoratec Corporation, Pleasanton, CA) LVAD at our center, as either a bridge to transplantation or destination therapy, from June 2009 to June 2012. Per protocol, all patients were scheduled to undergo transthoracic echocardiography, right heart catheterization, and CPET once functional capacity appeared maximal and no sooner than 3 months after implantation. Baseline characteristics including laboratory values, hemodynamic data, previous CPET results, NYHA functional class, and Intermacs Level were obtained before LVAD implantation. The principal outcomes were VO2 max, percent-predicted VO2 max, and NYHA functional class at the time of CPET. Patients were also stratified based on Intermacs level at implantation, with Intermacs levels 1 and 2 being compared with Intermacs levels ≥3. Those patients with CPET data before LVAD implantation had their pre-LVAD VO2 max compared with their post-LVAD VO2 max. Predicted VO2 max was calculated using the Wasserman/Hansen equation, which is a validated method of determining predicted VO2 max in heart failure patients.7,8 Patients were excluded from analysis if they failed to reach anaerobic threshold. No specific protocol was in place after LVAD implantation to adjust pump speed. In general, pump speed was increased to a level which alleviated exertional fatigue and dyspnea. We targeted a pump output which would be the equivalent of a cardiac index of at least 2.2 L/min/m2 while avoiding speeds which might produce suck down events. Although pump speed was not adjusted to a certain frequency of aortic valve opening, the speed was increased to reduce mitral regurgitation, if present, to a severity not greater than mild, and pump speed was lowered to avoid shifting of the interventricular septum past midline toward the left ventricle. Cardiopulmonary exercise testing was performed using the Innocor ergospirometry system (Medset Medical Technology, Hamburg, Germany) with treadmill exercise using the modified Naughton protocol. Those patients unable to follow the
Key Words: VO2 max, New York Heart Association Functional Classification, cardiopulmonary exercise testing, exercise capacity
M
echanical circulatory support with a left ventricular assist device (LVAD) is now a well-established therapy for patients with advanced heart failure.1 Newer continuous-flow LVAD systems have improved survival for patients waiting heart transplantation and for those receiving lifetime or “destination” therapy.2–5 LVAD-supported patients report marked improvements
From the *Department of Medicine, Icahn School of Medicine at Mount Sinai, New York, New York 10029; †Division of Cardiology, University of Chicago Medical Center, Chicago, Illinois 60637; ‡Zena and Michael A. Wiener Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, New York 10029; and §Department of Cardiothoracic Surgery, Icahn School of Medicine at Mount Sinai, New York, New York 10029. Submitted for consideration June 2014; accepted for publication in revised form December 2014. Disclosure: Dr. Pinney has served as a consultant for Thoratec, Inc. The authors have no other conflicts of interest to report. Correspondence: Sean P. Pinney, MD, Associate Professor of Medicine, Zena and Michael A. Wiener Cardiovascular Institute, Icahn School of Medicine at Mount Sinai One Gustave L. Levy Place, Box 1030, New York, NY 10029. Email: [email protected]. Copyright © 2015 by the American Society for Artificial Internal Organs DOI: 10.1097/MAT.0000000000000199
237
238 BENTON et al. modified Naughton protocol were exercised by a manual protocol at the discretion of the cardiologist supervising the CPET. Data generated from the Innocor system include serial measurements of VO2, minute ventilation (VE), carbon dioxide production (VCO2), and respiratory exchange ratio (RER) at 15 second intervals, allowing for calculation of anaerobic threshold and ventilatory efficiency (VE/VCO2 slope). Statistical analysis was performed using IBM SPSS Statistics 20. Descriptive statistics are presented as mean ± standard deviations or percentages where appropriate. Means between groups were compared with the Mann–Whitney U test for continuous variables and the Fisher’s exact test for categorical variables. The Pearson’s correlation coefficient was calculated when comparing Intermacs level and VO2. Statistical significance was defined as p < 0.05. Results One hundred six patients received a HeartMate II during the study period (Figure 1). Twenty-three patients died before CPET. Twenty-eight patients were transplanted and one patient had his device explanted before CPET. Fifteen patients declined or did not attend their CPET, and two patients were amputees who could not participate in CPET. Thirty-seven (29 male, mean age 56 ± 15 years) patients completed CPET an average of six months (178 ± 87 days) after implantation (Table 1). The modified Naughton protocol was used in 33 of the 37 patients, with the other four using a manual protocol. Two patients failed to reach anaerobic threshold and were excluded from
analysis. The mean duration of exercise was 7.5 ± 3.4 min, with a mean time to anaerobic threshold of 4.5 ± 1.6 min. Patients achieved a mean 3.9 ± 1.1 metabolic equivalents and a mean RER of 1.0 ± 0.1. Ten patients had CPET performed an average of 4 months (121 ± 82 days) before LVAD implantation. New York Heart Association functional class improved significantly after LVAD implantation (Figure 2A and B). Overall, 91.4% of patients improved by at least two NYHA classes, with 34.3% improving by three classes. All Intermacs level 1 or 2 patients (n = 17) improved by at least two NYHA classes, with 47.1% improving by three classes. Of those patients Intermacs level 3 and greater (n = 18), 85.5% improved by at least two classes, with 16.7% improving by three classes. For all 35 patients, post-LVAD VO2 max was significantly less than predicted (14.7 ± 3.1 vs. 29.8 ± 6.6 ml/kg/min, p < 0.001; percent-predicted 51% ± 12%). For the 10 patients with preand post-LVAD studies, VO2 max increased significantly from 11.6 ± 5.0 to 15.4 ± 3.9 ml/kg/min, p = 0.009, but remained markedly less than predicted (40% ± 17% vs. 54% ± 16%; p = 0.010; Figure 3). In these 10 patients, VO2 max increased, on average, 3.8 ± 3.6 ml/kg/min per patient. VO2 max for patients classified as Intermacs levels 1 or 2 (n = 17) did not differ significantly from those classified as Intermacs levels ≥3 (n = 18; 14.1 ± 2.7 vs. 15.2 ± 3.4 ml/kg/min; p = 0.309). Likewise, Intermacs level at implantation did not correlate with post-LVAD peak VO2 max (r = 0.213; p = 0.218). The mean VE/VCO2 slope was 41.3 ± 7.9, and did not differ significantly when stratified by Intermacs level (38.8 ± 7.1 for Intermacs levels 1 or 2 vs. 43.5 ± 8.3 for Intermacs levels ≥3; p = 0.094).
Figure 1. HeartMate II Implantations June 2009–June 2012. Asterisk Physical limitations included one above-knee and one below-knee amputee. CPET, cardiopulmonary exercise testing; LVAD, left ventricular assist device.
239
FUNCTIONAL CAPACITY AND VO2 IN LVAD PATIENTS Table 1. Baseline Characteristics of the 35 Patients
Characteristic Age (y) Male, n (%) Race, no. (%) White Black Hispanic Indian Pacific Islander Body mass index (kg/m2) Body surface area (m2) Heart failure etiology, n (%) Nonischemic Ischemic Acute myocardial infarction Other* New York Heart Association class, n (%) III IV Intermacs level, n (%) 1 2 3 4 Left ventricular ejection fraction (%) Inotropic support, n (%) Pressor requirement, n (%) VAD indication, n (%) Bridge Destination Patients transplanted, n (%) Hospital length of stay (days) ICU length of stay (days) Comorbidities, n (%) Hypertension Diabetes mellitus Tobacco use Coronary artery disease Atrial fibrillation Lung disease† Severe arthritis Neuropathy History of CVA Medications at CPET, n (%) β-blocker Angiotensin-converting enzyme inhibitor Angiotensin II receptor antagonist Aldosterone antagonist Digoxin Amiodarone Loop diuretic Laboratory values pre-LVAD Sodium (mEq/L) Blood urea nitrogen (mg/dL) Creatinine (mg/dL) Estimated glomerular filtration rate (mL/min) Aspartate aminotransferase (units/L) Total bilirubin (mg/dL) Hemoglobin (g/dL) Hemodynamic parameters pre-LVAD Arterial blood pressure (mm Hg) Systolic Diastolic Mean Heart rate (beats per min) Pulmonary artery pressure (mm Hg) Systolic Diastolic Mean
Value 56.3 ± 15.0 29 (83) 22 (63) 7 (20) 3 (9) 2 (6) 1 (3) 26.7 ± 4.3 2.0 ± 0.1 15 (43) 10 (29) 6 (17) 4 (11) 3 (9) 32 (91) 8 (23) 9 (26) 16 (46) 2 (6) 18 ± 7 26 (74) 4 (11) 32 (91) 3 (9) 27 (77) 31.6 ± 19.7 7.7 ± 6.4 19 (54) 16 (46) 17 (49) 18 (51) 13 (37) 4 (11) 5 (14) 1 (3) 4 (11) 15 (43) 8 (23) 10 (29) 14 (40) 6 (17) 11 (31) 16 (46) 134.5 ± 4.8 35.3 ± 18.3 1.5 ± 0.5 58.8 ± 26.8 38.2 ± 32.8 1.3 ± 0.9 11.5 ± 1.8 104.0 ± 13.7 65.4 ± 7.7 78.6 ± 8.6 85.9 ± 18.0 54.2 ± 13.7 26.3 ± 7.4 36.1 ± 8.7 (Continued)
Table 1. (Continued) Characteristic
Value
Right atrial pressure (mm Hg) Pulmonary-capillary wedge pressure (mm Hg) Cardiac index (L/min/m2) Mixed venous oxygen saturation (%) Right ventricular stroke-work index (g/m2/beat) Right ventricular failure risk score, n (%)‡ ≤3 4–5 ≥5.5
13.3 ± 5.4 25.5 ± 8.1 2.0 ± 0.7 50.8 ± 12.8 6.9 ± 3.3 31 (89) 2 (6) 2 (6)
Plus-minus values are mean ± SD. *The other etiologies of heart failure were valvular (2), ChurggStrauss (1), and Vasospasm (1). †The lung diseases were COPD (1), interstitial restrictive lung disease (2), and Churgg-Strauss (1). ‡The right ventricular failure risk score = sum of points awarded for vasopressor requirement (four pts). AST ≥ 80 units/L (two pts), bilirubin ≥ 2.0 mg/dl (2.5 pts), and creatinine ≥ 2.3 mg/dl (three pts).19 AST, aspartate transaminiase; COPD, chronic obstructive pulmonary disease; CPET, cardiopulmonary exercise testing; CVA, cerebrovascular accident; ICU, intensive care unit; LVAD, left ventricular assist devices.
VO2 max and VE/VCO2 slope also did not differ significantly between etiologies of heart failure (Table 2). Discussion The principle finding of our study is that VO2 max improves significantly with continuous-flow LVAD support but fails to normalize to predicted age-, gender-, and BMI-matched values, despite improvement in NYHA functional class. We found no difference in postimplantation VO2 max or VE/VCO2 when stratifying by Intermacs levels or between etiologies of heart failure. Measurement of peak oxygen consumption (VO2 max) is a well-established method of risk stratifying heart failure patients. In 1991, Mancini et al.9 showed that heart failure patients achieving VO2 max >14 ml/kg/min had 94% 1 year survival, comparable with cardiac transplant recipients, whereas those ≤14 ml/kg/min had only a 47% 1 year survival. VO2 max is an ideal marker of cardiopulmonary fitness in heart failure patients, combining cardiac output, oxygen delivery, and oxygen utilization by skeletal muscle. More recently, ventilatory efficiency, the relation between VE and carbon dioxide production (VE/VCO2 slope), has been shown to be of prognostic significance in chronic heart failure patients, and may be a better discriminator than VO2 max.10,11 When combined with VO2 max, VE/VCO2 slope improves the identification of the highestrisk patients at each level of peak oxygen consumption.12 Our results complement those of studies of first-generation pulsatile LVADs which demonstrated improvements in peak oxygen consumption of 3–4 ml/kg/min, 3–6 months after implantation.13–15 More recently, a few small studies demonstrated similar results with continuous-flow LVADs achieving peak oxygen consumption of 15–16 ml/kg/min, though none of these cohorts had preimplantation data for comparison.16–18 Two small studies with paired pre- and postimplant VO2 showed similarly small increases in VO2 max after 3–6 months of support.19,20 Recently, Nahumi et al.21 measured VO2 max and 6 min walk distance in 26 patients supported with a
240 BENTON et al.
Figure 2. A: Pre-LVAD NYHA class comparison. Overall and by Intermacs level. B: Post-LVAD NYHA class comparison. Overall and by Intermacs level. LVAD, left ventricular assist devices; NYHA, New York Heart Association.
continuous flow LVAD. When compared with a matched cohort of chronic heart failure patients, those supported with an LVAD were able to walk a longer distance in 6 min (80 m),
but failed to reach a similar peak oxygen consumption (VO2 max 12.4 vs. 15.0 ml/kg/min). This suggests that measurable differences in exercise capacity exist between heart failure
Figure 3. Matched changes in peak oxygen consumption (VO2) for 10 patients with preimplant studies. Average change is represented by the bold line with square markers. LVAD, left ventricular assist device.
241
FUNCTIONAL CAPACITY AND VO2 IN LVAD PATIENTS Table 2. Cardiopulmonary Exercise Test Results and Change in Functional Capacity
NYHA class (n = 35) VO2 max (n = 35) Predicted VO2 max %Predicted VO2 max (%) Intermacs 1&2 VO2 max (n = 17) Intermacs ≥3 VO2 max (n = 18) Chronic ischemic VO2 max (n = 10) Chronic nonischemic VO2 max (n = 15) Acute myocardial infarction VO2 max (n = 6) Ve/VCO2 slope (n = 35) Intermacs 1&2 Ve/VCO2 slope (n = 17) Intermacs ≥3 Ve/VCO2 slope (n = 18) Chronic ischemic Ve/VCO2 slope (n = 10) Chronic nonischemic Ve/VCO2 slope (n = 15) Acute myocardial infarction Ve/VCO2 slope (n = 6) VO2 max (n = 10) %Predicted VO2 max (%) Time before LVAD placement (days) Improvement by two NYHA classes Overall Intermacs 1 & 2 Intermacs ≥3 Improvement by three NYHA classes Overall Intermacs 1&2 Intermacs ≥3 CPET characteristics Time since LVAD placement (days) Protocol, n (%) Modified Naughton Manual Test duration (min) RER Time to anaerobic threshold (min) Metabolic equivalents Resting heart rate (beats per min) Peak heart rate (beats per min)
Pre-LVAD
Post-LVAD
p Value
3.9 ± 0.3
1.7 ± 0.5 14.7 ± 3.1 29.8 ± 6.6 51 ± 12 14.1 ± 2.7 15.2 ± 3.4 14.1 ± 1.9 14.9 ± 3.3 13.9 ± 3.0 41.3 ± 7.9 38.8 ± 7.1 43.5 ± 8.3 43.0 ± 8.9 40.9 ± 8.1 38.0 ± 2.5 15.4 ± 3.9 54 ± 16
Adult Circulatory Support
Left Ventricular Assist Devices Improve Functional Class without Normalizing Peak Oxygen Consumption Casey R. Benton,* Gabriel Sayer,† Ajith P. Nair,‡ Kimberly Ashley,‡ Michael J. Domanski,‡ Milena J. Henzlova,‡ Anelechi C. Anyanwu,§ and Sean P. Pinney‡
Heart failure patients supported with left ventricular assist devices (LVAD) enjoy improvements in functional capacity and quality of life. We reasoned that such improvements in exercise capacity should be reflected in an objective increase in peak oxygen consumption as measured by cardiopulmonary exercise testing (CPET). We performed a retrospective review of all recipients of a HeartMate II LVAD at our center from June 2009 to June 2012 who completed CPET. Thirtyseven patients completed CPET an average of 6 months after implantation. Of these, 10 patients had CPET performed before LVAD implantation. Overall, 91.4% of patients improved by at least two New York Heart Association classes, with 34.3% improving by three classes. Postimplant VO2 max was significantly less than predicted (14.7 ± 3.1 vs. 29.8 ± 6.6 ml/kg/min, p < 0.001; percent-predicted 51% ± 12%). For 10 patients with pre- and post-implant studies, VO2 max increased significantly from 11.6 ± 5.0 to 15.4 ± 3.9 ml/kg/min (p = 0.009). VO2 max improves significantly with LVAD support but fails to normalize to predicted values, in spite of improvements in functional class. The severity of preimplantation heart failure does not associate with the degree of VO2 max improvement. ASAIO Journal 2015; 61:237–243.
in their functional capacity and quality of life with over 50% of patients improving to New York Heart Association (NYHA) functional class I or II within 1 month after implantation and over 80% achieving this by 6 months. These improvements are sustained for at least 2 years after implantation.5,6 We hypothesized that such subjective increases in functional capacity should be associated with objective improvements in peak oxygen consumption (VO2 max), as measured by cardiopulmonary exercise testing (CPET). We also hypothesized that patients who were more critically ill at the time of implantation would not recover to the same extent as those with lesser critical presentations. Methods We performed a retrospective review of all patients implanted with a HeartMate II (Thoratec Corporation, Pleasanton, CA) LVAD at our center, as either a bridge to transplantation or destination therapy, from June 2009 to June 2012. Per protocol, all patients were scheduled to undergo transthoracic echocardiography, right heart catheterization, and CPET once functional capacity appeared maximal and no sooner than 3 months after implantation. Baseline characteristics including laboratory values, hemodynamic data, previous CPET results, NYHA functional class, and Intermacs Level were obtained before LVAD implantation. The principal outcomes were VO2 max, percent-predicted VO2 max, and NYHA functional class at the time of CPET. Patients were also stratified based on Intermacs level at implantation, with Intermacs levels 1 and 2 being compared with Intermacs levels ≥3. Those patients with CPET data before LVAD implantation had their pre-LVAD VO2 max compared with their post-LVAD VO2 max. Predicted VO2 max was calculated using the Wasserman/Hansen equation, which is a validated method of determining predicted VO2 max in heart failure patients.7,8 Patients were excluded from analysis if they failed to reach anaerobic threshold. No specific protocol was in place after LVAD implantation to adjust pump speed. In general, pump speed was increased to a level which alleviated exertional fatigue and dyspnea. We targeted a pump output which would be the equivalent of a cardiac index of at least 2.2 L/min/m2 while avoiding speeds which might produce suck down events. Although pump speed was not adjusted to a certain frequency of aortic valve opening, the speed was increased to reduce mitral regurgitation, if present, to a severity not greater than mild, and pump speed was lowered to avoid shifting of the interventricular septum past midline toward the left ventricle. Cardiopulmonary exercise testing was performed using the Innocor ergospirometry system (Medset Medical Technology, Hamburg, Germany) with treadmill exercise using the modified Naughton protocol. Those patients unable to follow the
Key Words: VO2 max, New York Heart Association Functional Classification, cardiopulmonary exercise testing, exercise capacity
M
echanical circulatory support with a left ventricular assist device (LVAD) is now a well-established therapy for patients with advanced heart failure.1 Newer continuous-flow LVAD systems have improved survival for patients waiting heart transplantation and for those receiving lifetime or “destination” therapy.2–5 LVAD-supported patients report marked improvements
From the *Department of Medicine, Icahn School of Medicine at Mount Sinai, New York, New York 10029; †Division of Cardiology, University of Chicago Medical Center, Chicago, Illinois 60637; ‡Zena and Michael A. Wiener Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, New York 10029; and §Department of Cardiothoracic Surgery, Icahn School of Medicine at Mount Sinai, New York, New York 10029. Submitted for consideration June 2014; accepted for publication in revised form December 2014. Disclosure: Dr. Pinney has served as a consultant for Thoratec, Inc. The authors have no other conflicts of interest to report. Correspondence: Sean P. Pinney, MD, Associate Professor of Medicine, Zena and Michael A. Wiener Cardiovascular Institute, Icahn School of Medicine at Mount Sinai One Gustave L. Levy Place, Box 1030, New York, NY 10029. Email: [email protected]. Copyright © 2015 by the American Society for Artificial Internal Organs DOI: 10.1097/MAT.0000000000000199
237
238 BENTON et al. modified Naughton protocol were exercised by a manual protocol at the discretion of the cardiologist supervising the CPET. Data generated from the Innocor system include serial measurements of VO2, minute ventilation (VE), carbon dioxide production (VCO2), and respiratory exchange ratio (RER) at 15 second intervals, allowing for calculation of anaerobic threshold and ventilatory efficiency (VE/VCO2 slope). Statistical analysis was performed using IBM SPSS Statistics 20. Descriptive statistics are presented as mean ± standard deviations or percentages where appropriate. Means between groups were compared with the Mann–Whitney U test for continuous variables and the Fisher’s exact test for categorical variables. The Pearson’s correlation coefficient was calculated when comparing Intermacs level and VO2. Statistical significance was defined as p < 0.05. Results One hundred six patients received a HeartMate II during the study period (Figure 1). Twenty-three patients died before CPET. Twenty-eight patients were transplanted and one patient had his device explanted before CPET. Fifteen patients declined or did not attend their CPET, and two patients were amputees who could not participate in CPET. Thirty-seven (29 male, mean age 56 ± 15 years) patients completed CPET an average of six months (178 ± 87 days) after implantation (Table 1). The modified Naughton protocol was used in 33 of the 37 patients, with the other four using a manual protocol. Two patients failed to reach anaerobic threshold and were excluded from
analysis. The mean duration of exercise was 7.5 ± 3.4 min, with a mean time to anaerobic threshold of 4.5 ± 1.6 min. Patients achieved a mean 3.9 ± 1.1 metabolic equivalents and a mean RER of 1.0 ± 0.1. Ten patients had CPET performed an average of 4 months (121 ± 82 days) before LVAD implantation. New York Heart Association functional class improved significantly after LVAD implantation (Figure 2A and B). Overall, 91.4% of patients improved by at least two NYHA classes, with 34.3% improving by three classes. All Intermacs level 1 or 2 patients (n = 17) improved by at least two NYHA classes, with 47.1% improving by three classes. Of those patients Intermacs level 3 and greater (n = 18), 85.5% improved by at least two classes, with 16.7% improving by three classes. For all 35 patients, post-LVAD VO2 max was significantly less than predicted (14.7 ± 3.1 vs. 29.8 ± 6.6 ml/kg/min, p < 0.001; percent-predicted 51% ± 12%). For the 10 patients with preand post-LVAD studies, VO2 max increased significantly from 11.6 ± 5.0 to 15.4 ± 3.9 ml/kg/min, p = 0.009, but remained markedly less than predicted (40% ± 17% vs. 54% ± 16%; p = 0.010; Figure 3). In these 10 patients, VO2 max increased, on average, 3.8 ± 3.6 ml/kg/min per patient. VO2 max for patients classified as Intermacs levels 1 or 2 (n = 17) did not differ significantly from those classified as Intermacs levels ≥3 (n = 18; 14.1 ± 2.7 vs. 15.2 ± 3.4 ml/kg/min; p = 0.309). Likewise, Intermacs level at implantation did not correlate with post-LVAD peak VO2 max (r = 0.213; p = 0.218). The mean VE/VCO2 slope was 41.3 ± 7.9, and did not differ significantly when stratified by Intermacs level (38.8 ± 7.1 for Intermacs levels 1 or 2 vs. 43.5 ± 8.3 for Intermacs levels ≥3; p = 0.094).
Figure 1. HeartMate II Implantations June 2009–June 2012. Asterisk Physical limitations included one above-knee and one below-knee amputee. CPET, cardiopulmonary exercise testing; LVAD, left ventricular assist device.
239
FUNCTIONAL CAPACITY AND VO2 IN LVAD PATIENTS Table 1. Baseline Characteristics of the 35 Patients
Characteristic Age (y) Male, n (%) Race, no. (%) White Black Hispanic Indian Pacific Islander Body mass index (kg/m2) Body surface area (m2) Heart failure etiology, n (%) Nonischemic Ischemic Acute myocardial infarction Other* New York Heart Association class, n (%) III IV Intermacs level, n (%) 1 2 3 4 Left ventricular ejection fraction (%) Inotropic support, n (%) Pressor requirement, n (%) VAD indication, n (%) Bridge Destination Patients transplanted, n (%) Hospital length of stay (days) ICU length of stay (days) Comorbidities, n (%) Hypertension Diabetes mellitus Tobacco use Coronary artery disease Atrial fibrillation Lung disease† Severe arthritis Neuropathy History of CVA Medications at CPET, n (%) β-blocker Angiotensin-converting enzyme inhibitor Angiotensin II receptor antagonist Aldosterone antagonist Digoxin Amiodarone Loop diuretic Laboratory values pre-LVAD Sodium (mEq/L) Blood urea nitrogen (mg/dL) Creatinine (mg/dL) Estimated glomerular filtration rate (mL/min) Aspartate aminotransferase (units/L) Total bilirubin (mg/dL) Hemoglobin (g/dL) Hemodynamic parameters pre-LVAD Arterial blood pressure (mm Hg) Systolic Diastolic Mean Heart rate (beats per min) Pulmonary artery pressure (mm Hg) Systolic Diastolic Mean
Value 56.3 ± 15.0 29 (83) 22 (63) 7 (20) 3 (9) 2 (6) 1 (3) 26.7 ± 4.3 2.0 ± 0.1 15 (43) 10 (29) 6 (17) 4 (11) 3 (9) 32 (91) 8 (23) 9 (26) 16 (46) 2 (6) 18 ± 7 26 (74) 4 (11) 32 (91) 3 (9) 27 (77) 31.6 ± 19.7 7.7 ± 6.4 19 (54) 16 (46) 17 (49) 18 (51) 13 (37) 4 (11) 5 (14) 1 (3) 4 (11) 15 (43) 8 (23) 10 (29) 14 (40) 6 (17) 11 (31) 16 (46) 134.5 ± 4.8 35.3 ± 18.3 1.5 ± 0.5 58.8 ± 26.8 38.2 ± 32.8 1.3 ± 0.9 11.5 ± 1.8 104.0 ± 13.7 65.4 ± 7.7 78.6 ± 8.6 85.9 ± 18.0 54.2 ± 13.7 26.3 ± 7.4 36.1 ± 8.7 (Continued)
Table 1. (Continued) Characteristic
Value
Right atrial pressure (mm Hg) Pulmonary-capillary wedge pressure (mm Hg) Cardiac index (L/min/m2) Mixed venous oxygen saturation (%) Right ventricular stroke-work index (g/m2/beat) Right ventricular failure risk score, n (%)‡ ≤3 4–5 ≥5.5
13.3 ± 5.4 25.5 ± 8.1 2.0 ± 0.7 50.8 ± 12.8 6.9 ± 3.3 31 (89) 2 (6) 2 (6)
Plus-minus values are mean ± SD. *The other etiologies of heart failure were valvular (2), ChurggStrauss (1), and Vasospasm (1). †The lung diseases were COPD (1), interstitial restrictive lung disease (2), and Churgg-Strauss (1). ‡The right ventricular failure risk score = sum of points awarded for vasopressor requirement (four pts). AST ≥ 80 units/L (two pts), bilirubin ≥ 2.0 mg/dl (2.5 pts), and creatinine ≥ 2.3 mg/dl (three pts).19 AST, aspartate transaminiase; COPD, chronic obstructive pulmonary disease; CPET, cardiopulmonary exercise testing; CVA, cerebrovascular accident; ICU, intensive care unit; LVAD, left ventricular assist devices.
VO2 max and VE/VCO2 slope also did not differ significantly between etiologies of heart failure (Table 2). Discussion The principle finding of our study is that VO2 max improves significantly with continuous-flow LVAD support but fails to normalize to predicted age-, gender-, and BMI-matched values, despite improvement in NYHA functional class. We found no difference in postimplantation VO2 max or VE/VCO2 when stratifying by Intermacs levels or between etiologies of heart failure. Measurement of peak oxygen consumption (VO2 max) is a well-established method of risk stratifying heart failure patients. In 1991, Mancini et al.9 showed that heart failure patients achieving VO2 max >14 ml/kg/min had 94% 1 year survival, comparable with cardiac transplant recipients, whereas those ≤14 ml/kg/min had only a 47% 1 year survival. VO2 max is an ideal marker of cardiopulmonary fitness in heart failure patients, combining cardiac output, oxygen delivery, and oxygen utilization by skeletal muscle. More recently, ventilatory efficiency, the relation between VE and carbon dioxide production (VE/VCO2 slope), has been shown to be of prognostic significance in chronic heart failure patients, and may be a better discriminator than VO2 max.10,11 When combined with VO2 max, VE/VCO2 slope improves the identification of the highestrisk patients at each level of peak oxygen consumption.12 Our results complement those of studies of first-generation pulsatile LVADs which demonstrated improvements in peak oxygen consumption of 3–4 ml/kg/min, 3–6 months after implantation.13–15 More recently, a few small studies demonstrated similar results with continuous-flow LVADs achieving peak oxygen consumption of 15–16 ml/kg/min, though none of these cohorts had preimplantation data for comparison.16–18 Two small studies with paired pre- and postimplant VO2 showed similarly small increases in VO2 max after 3–6 months of support.19,20 Recently, Nahumi et al.21 measured VO2 max and 6 min walk distance in 26 patients supported with a
240 BENTON et al.
Figure 2. A: Pre-LVAD NYHA class comparison. Overall and by Intermacs level. B: Post-LVAD NYHA class comparison. Overall and by Intermacs level. LVAD, left ventricular assist devices; NYHA, New York Heart Association.
continuous flow LVAD. When compared with a matched cohort of chronic heart failure patients, those supported with an LVAD were able to walk a longer distance in 6 min (80 m),
but failed to reach a similar peak oxygen consumption (VO2 max 12.4 vs. 15.0 ml/kg/min). This suggests that measurable differences in exercise capacity exist between heart failure
Figure 3. Matched changes in peak oxygen consumption (VO2) for 10 patients with preimplant studies. Average change is represented by the bold line with square markers. LVAD, left ventricular assist device.
241
FUNCTIONAL CAPACITY AND VO2 IN LVAD PATIENTS Table 2. Cardiopulmonary Exercise Test Results and Change in Functional Capacity
NYHA class (n = 35) VO2 max (n = 35) Predicted VO2 max %Predicted VO2 max (%) Intermacs 1&2 VO2 max (n = 17) Intermacs ≥3 VO2 max (n = 18) Chronic ischemic VO2 max (n = 10) Chronic nonischemic VO2 max (n = 15) Acute myocardial infarction VO2 max (n = 6) Ve/VCO2 slope (n = 35) Intermacs 1&2 Ve/VCO2 slope (n = 17) Intermacs ≥3 Ve/VCO2 slope (n = 18) Chronic ischemic Ve/VCO2 slope (n = 10) Chronic nonischemic Ve/VCO2 slope (n = 15) Acute myocardial infarction Ve/VCO2 slope (n = 6) VO2 max (n = 10) %Predicted VO2 max (%) Time before LVAD placement (days) Improvement by two NYHA classes Overall Intermacs 1 & 2 Intermacs ≥3 Improvement by three NYHA classes Overall Intermacs 1&2 Intermacs ≥3 CPET characteristics Time since LVAD placement (days) Protocol, n (%) Modified Naughton Manual Test duration (min) RER Time to anaerobic threshold (min) Metabolic equivalents Resting heart rate (beats per min) Peak heart rate (beats per min)
Pre-LVAD
Post-LVAD
p Value
3.9 ± 0.3
1.7 ± 0.5 14.7 ± 3.1 29.8 ± 6.6 51 ± 12 14.1 ± 2.7 15.2 ± 3.4 14.1 ± 1.9 14.9 ± 3.3 13.9 ± 3.0 41.3 ± 7.9 38.8 ± 7.1 43.5 ± 8.3 43.0 ± 8.9 40.9 ± 8.1 38.0 ± 2.5 15.4 ± 3.9 54 ± 16
