- Email: [email protected]
Impact of Optimal Heart Failure Medical Therapy on Heart Transplant Listing
CANDIDATES Impact of Optimal Heart Failure Medical Therapy on Heart Transplant Listing C. Geisberg, J. Goring, J. Listerman, M.A. Nading, R.L. Huang, and J. Butler ABSTRACT The data assessing the prognostic value of peak exercise oxygen consumption (VO2) in heart failure (HF) patients is largely derived from cross-sectional studies in which medical therapy was not maximized in all eligible patients and no clear explanation was given as to why such was the case. To assess the relative prognostic value of peak VO2 with respect to baseline medical therapy and its potential impact on transplant listing, 1-year event-free (death or left ventricular assist device placement) survival was compared among 341 HF patients, stratified in three groups based on peak VO2 (⬍10, 10 to 14, and ⬎14 mL/min/kg). Similar analysis was performed on a subset of 288 patients who were on optimal medical therapy within this group. Average age of the study population was 55 ⫾ 11 years, ejection fraction was 23% ⫾ 08%, and peak VO2 was 12.4 ⫾ 3.6 mL/min/kg. One-year event-free survival for the overall cohort was: peak VO2 ⬍ 10 (n ⫽ 87), 63.2%; 10 to 14 (n ⫽ 141), 81.1%; and ⬎14 mL/min/kg (n ⫽ 113), 90.2%. Patients with the same groups who were on optimal therapy had an event-free 1-year survival as follows: ⬍10 (n ⫽ 69), 72.4%; VO2 10 to 14 (n ⫽ 127), 91.5%; and ⬎14 mL/min/kg (n ⫽ 92), 94.6%. In conclusion, cross-sectional assessment of HF prognosis may be misleading. In the intermediate risk group, this can significantly impact on medical decisions (eg, transplant listing). Optimization of therapy and long-term follow-up by a specialist may impact transplant listing.
C
URRENTLY, HEART FAILURE (HF) patients with peak exercise oxygen consumption (VO2) ⬍14 mL/ min/kg are considered eligible for cardiac transplantation, based on the assumption that these patients will have a better survival with transplantation than with medical therapy.1 Several recent reports, however, have suggested that as prognosis for patients with HF has improved with newer therapies, the outcomes for patients with similar peak VO2 level has also improved.2– 4 These studies, however, were either retrospective or cross-sectional, and medical therapy was not maximized in all patients. Thus these results may be subject to change with optimization of medical therapy. In this study, we assessed the prognostic value of peak VO2 in HF prognosis determination with respect to baseline therapy.
defined within this group (n ⫽ 288) and qualitatively compared to the overall cohort. Optimal therapy was defined as therapy with angiotensin converting enzyme inhibitors or angiotensin receptor blockers, at least 50% of recommended beta-blocker dose (unless systolic blood pressure ⬍100 mm Hg or heart rate ⬍60 beats per minute), spironolactone (RALES trial criteria), defibrillator (MADIT II trial criteria with QRS duration ⬎120 ms), and biventricular pacemaker (COMPANION trial criteria), and compared to the overall group.5– 8 Any physician-documented reason for lack of therapy, including “intolerance,” was accepted and the patient was considered on optimal therapy with respect to that particular drug or device. We decided to compare the cohort on optimal therapy within the overall cohort rather than comparing those on optimal therapy versus those who were not. We expected to see the patients not on optimal therapy to do poorly compared to those who were aggres-
METHODS The study population consisted of 341 HF patients with ejection fraction ⬍40% who underwent exercise testing between June 2002 and February 2004. These patients were divided into three groups based on peak VO2: those with values ⬍10, 10 to 14, and ⬎14 mL/min/kg. A cohort of optimally treated patients was
From the Cardiology Division, Vanderbilt University, Nashville, Tennessee, USA. Address reprint requests to Javed Butler, MD, Cardiology Division, 383 PRB, Vanderbilt University Medical Center, Nashville TN 37232-6300. E-mail: [email protected]
© 2006 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
0041-1345/06/$–see front matter doi:10.1016/j.transproceed.2006.02.099
Transplantation Proceedings, 38, 1493–1495 (2006)
1493
1494
GEISBERG, GORING, LISTERMAN ET AL
sively managed. However, our purpose was to quantify the “dilution” of prognosis seen in larger databases when “overall” results are reviewed as compared to those on optimal therapy. We did, however, include patients in the “optimal therapy” group who were not necessarily on optimal therapy but had documented intolerance or any physician-documented reason for not prescribing a particular therapy. Thus this analysis does not bias the results toward an “ideal” treatment paradigm, but to “real-life” setting where many patients on “optimal therapy” arm were not on maximum therapy. The study outcome was 1-year event-free survival (without the need for left ventricular assist device). Chi-square test for categorical and t test for continuous variables were used where necessary. Kaplan-Meier method was used to assess survival. Seven patients who underwent transplantation were censored.
RESULTS Patient Population
Baseline characteristics for patients in the overall cohort and those who were optimally treated within this group are shown in Table 1. Patients optimally treated had lower mean blood pressures (80 ⫾ 11 vs 86 ⫾ 12 mm Hg, P ⫽ .01) and heart rate (74 ⫾ 14 vs 88 ⫾ 16, P ⬍ .01). Outcomes
One-year event-free survival for patients based on peak VO2 is shown in Fig 1 and was as follows: ⬍10 (n ⫽ 87), 63.2%; 10 to 14 (n ⫽ 156), 81.1%; and ⬎14 (n ⫽ 119), 90.2%. Patients within the same groups who were on optimal therapy had an event-free 1-year survival as follows: Table 1. Patient Characteristics and Medical Treatment
Characteristics
Age (y), mean ⫾ SD Gender (% males) Race (% Caucasians) Etiology (% ischemic) Ejection fraction (%), mean ⫾ SD Heart rate (beats per minute), mean ⫾ SD Mean blood pressure (mm Hg), mean ⫾ SD Serum sodium (meq/L) Serum creatinine (mg/dL) Diabetes (%) Peak VO2 (mL/min/kg) ⬍10 (%) 10–14 (%) ⬎14 (%) Medications (%) ACE inhibitor or ARB Diuretic Digoxin Beta-blocker Spironolactone Defibrillator Biventricular pacemaker
Overall (n ⫽ 341)
Optimally Treated (n ⫽ 288)
52 ⫾ 11 74 85 52 23 ⫾ 9
54 ⫾ 11 76 85 56 21 ⫾ 8
NS NS NS NS NS
88 ⫾ 16
74 ⫾ 14
⬍.01
86 ⫾ 11
80 ⫾ 12
⬍.01
137 ⫾ 4 1.5 ⫾ 0.7 29 13.0 ⫾ 3.3 20 45 35
138 ⫾ 4 1.4 ⫾ 0.6 28 12.8 ⫾ 3.2 23 42 35
NS NS NS NS
90 96 93 75 68 36 15
97 98 94 88 78 40 18
— — — — — — —
P
NS
Fig 1. Event-free survival for the overall cohort and those patients optimally treated.
⬍10 (n ⫽ 40), 72.4%; VO2 10 to 14 (n ⫽ 74), 91.5%; and ⬎14 (n ⫽ 42), 94.6%. DISCUSSION
Our data suggest that prognosis of HF patients at any particular level of exercise capacity can vary depending on baseline therapy. This was clinically most meaningful for patients in the intermediate risk group (VO2 10 to 14 mL/min/kg). In this group, 1-year event-free survival improved from what would typically be considered less than that of posttransplant outcomes to comparable or better with optimal therapy. This is especially important since even our “optimal therapy” patients were not on optimal therapy by the current standards of care (eg, more liberal defibrillator indication currently), which is likely to improve outcomes even further. This is an important issue since most of the HF care is provided in the primary care setting. Many insurance companies do not permit longitudinal management by specialists when patients are referred for transplantation. Instead, a few visits are authorized to “complete the transplant evaluation.” Plenty of data suggest that care provided by the specialist is associated with increased compliance with standard of care.9 Thus rather than making a decision based on crosssectional data, longitudinal management of these patients by a specialist may change the eventual outcomes for patients. Another issue is what the minimum baseline medical therapy should be for patients who are enrolled in clinical trials. Many patients enrolled in trials are drawn from non-HF specialty clinics. It is not uncommon for even recent trials to show beta-blocker use in the 70% range, whereas specialty practices have shown a rate closer to 90% in practice setting.10 If patients enrolled in the trial are not on maximal medical therapy, relatively modest drugs and device benefits may show “positive” results, which may not be the case if patients were optimally treated. There are no standards of why patients are not on certain therapies prior to enrollment in clinical trials and this is left up to the site principal investigator. Thus one potential consideration is a specialist evaluation prior to enrollment of such candidates in trials to get more accurate results. Our study has several limitations. The reasons for lack of
OPTIMAL HEART FAILURE MEDICAL THERAPY
therapy could not be ascertained in all cases unless obvious or documented. Peak VO2 determinations were not assessed at specific times during treatment but were dependent on individual provider’s discretion. Finally, the “standard of care” is a moving target, and our estimations may not be accurate as newer therapies are discovered. In conclusion, our study suggests a significant, clinically meaningful effect of maximization of medical therapy on outcomes among HF patients. For transplant evaluation or enrollment in clinical trial, longitudinal management by a specialist should be considered.
REFERENCES 1. Costanzo MR, Augustine S, Bourge R, et al: Selection and treatment of candidates for heart transplantation. Circ 92:3593, 1995 2. Butler J, Khadim G, Paul KM, et al: Selection of patients for cardiac transplantation in the current era of heart failure therapy. J Am Coll Cardiol 43:787, 2004
1495 3. Pohwani AL, Murali S, Mathier MM, et al: Impact of beta-blocker therapy on functional capacity criteria for heart transplant listing. J Heart Lung Transplant 22:78, 2003 4. Peterson LR, Schechtman KB, Ewald GA, et al: The effect of beta blockers on prognostic value of peak oxygen uptake in heart failure. J Heart Lung Transplant 22:70, 2003 5. Metoprolol CR/XL Randomized Intervention Trial in Congestive Heart Failure: Effect of metoprolol CR/XL in chronic heart failure. Lancet 353:2001, 1999 6. Moss AJ, Zareba W, Hall WJ, et al: Prophylactic implantation of a defibrillator in patients with MI and reduced ejection fraction. New Engl J Med 346:877, 2002 7. Pitt B, Zannad F, Remme WJ, et al: The effect of spironolactone on morbidity and mortality in patients with severe heart failure. N Engl J Med 341:709, 1999 8. Bristow MR, Saxon LA, Boehmer J, et al: Cardiacresynchronization therapy with or without a defibrillator in advanced heart failure. N Engl J Med 350:2140, 2004 9. Hanumanthu SK, Butler J, Chomsky DB, et al: Effect of a heart failure program on hospitalization frequency and exercise tolerance. Circ 96:8, 1997 10. Butler J, Khadim G, Belue R, et al: Tolerability to betablocker therapy among heart failure patients in clinical practice. J Card Fail 9:203, 2003
C
URRENTLY, HEART FAILURE (HF) patients with peak exercise oxygen consumption (VO2) ⬍14 mL/ min/kg are considered eligible for cardiac transplantation, based on the assumption that these patients will have a better survival with transplantation than with medical therapy.1 Several recent reports, however, have suggested that as prognosis for patients with HF has improved with newer therapies, the outcomes for patients with similar peak VO2 level has also improved.2– 4 These studies, however, were either retrospective or cross-sectional, and medical therapy was not maximized in all patients. Thus these results may be subject to change with optimization of medical therapy. In this study, we assessed the prognostic value of peak VO2 in HF prognosis determination with respect to baseline therapy.
defined within this group (n ⫽ 288) and qualitatively compared to the overall cohort. Optimal therapy was defined as therapy with angiotensin converting enzyme inhibitors or angiotensin receptor blockers, at least 50% of recommended beta-blocker dose (unless systolic blood pressure ⬍100 mm Hg or heart rate ⬍60 beats per minute), spironolactone (RALES trial criteria), defibrillator (MADIT II trial criteria with QRS duration ⬎120 ms), and biventricular pacemaker (COMPANION trial criteria), and compared to the overall group.5– 8 Any physician-documented reason for lack of therapy, including “intolerance,” was accepted and the patient was considered on optimal therapy with respect to that particular drug or device. We decided to compare the cohort on optimal therapy within the overall cohort rather than comparing those on optimal therapy versus those who were not. We expected to see the patients not on optimal therapy to do poorly compared to those who were aggres-
METHODS The study population consisted of 341 HF patients with ejection fraction ⬍40% who underwent exercise testing between June 2002 and February 2004. These patients were divided into three groups based on peak VO2: those with values ⬍10, 10 to 14, and ⬎14 mL/min/kg. A cohort of optimally treated patients was
From the Cardiology Division, Vanderbilt University, Nashville, Tennessee, USA. Address reprint requests to Javed Butler, MD, Cardiology Division, 383 PRB, Vanderbilt University Medical Center, Nashville TN 37232-6300. E-mail: [email protected]
© 2006 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
0041-1345/06/$–see front matter doi:10.1016/j.transproceed.2006.02.099
Transplantation Proceedings, 38, 1493–1495 (2006)
1493
1494
GEISBERG, GORING, LISTERMAN ET AL
sively managed. However, our purpose was to quantify the “dilution” of prognosis seen in larger databases when “overall” results are reviewed as compared to those on optimal therapy. We did, however, include patients in the “optimal therapy” group who were not necessarily on optimal therapy but had documented intolerance or any physician-documented reason for not prescribing a particular therapy. Thus this analysis does not bias the results toward an “ideal” treatment paradigm, but to “real-life” setting where many patients on “optimal therapy” arm were not on maximum therapy. The study outcome was 1-year event-free survival (without the need for left ventricular assist device). Chi-square test for categorical and t test for continuous variables were used where necessary. Kaplan-Meier method was used to assess survival. Seven patients who underwent transplantation were censored.
RESULTS Patient Population
Baseline characteristics for patients in the overall cohort and those who were optimally treated within this group are shown in Table 1. Patients optimally treated had lower mean blood pressures (80 ⫾ 11 vs 86 ⫾ 12 mm Hg, P ⫽ .01) and heart rate (74 ⫾ 14 vs 88 ⫾ 16, P ⬍ .01). Outcomes
One-year event-free survival for patients based on peak VO2 is shown in Fig 1 and was as follows: ⬍10 (n ⫽ 87), 63.2%; 10 to 14 (n ⫽ 156), 81.1%; and ⬎14 (n ⫽ 119), 90.2%. Patients within the same groups who were on optimal therapy had an event-free 1-year survival as follows: Table 1. Patient Characteristics and Medical Treatment
Characteristics
Age (y), mean ⫾ SD Gender (% males) Race (% Caucasians) Etiology (% ischemic) Ejection fraction (%), mean ⫾ SD Heart rate (beats per minute), mean ⫾ SD Mean blood pressure (mm Hg), mean ⫾ SD Serum sodium (meq/L) Serum creatinine (mg/dL) Diabetes (%) Peak VO2 (mL/min/kg) ⬍10 (%) 10–14 (%) ⬎14 (%) Medications (%) ACE inhibitor or ARB Diuretic Digoxin Beta-blocker Spironolactone Defibrillator Biventricular pacemaker
Overall (n ⫽ 341)
Optimally Treated (n ⫽ 288)
52 ⫾ 11 74 85 52 23 ⫾ 9
54 ⫾ 11 76 85 56 21 ⫾ 8
NS NS NS NS NS
88 ⫾ 16
74 ⫾ 14
⬍.01
86 ⫾ 11
80 ⫾ 12
⬍.01
137 ⫾ 4 1.5 ⫾ 0.7 29 13.0 ⫾ 3.3 20 45 35
138 ⫾ 4 1.4 ⫾ 0.6 28 12.8 ⫾ 3.2 23 42 35
NS NS NS NS
90 96 93 75 68 36 15
97 98 94 88 78 40 18
— — — — — — —
P
NS
Fig 1. Event-free survival for the overall cohort and those patients optimally treated.
⬍10 (n ⫽ 40), 72.4%; VO2 10 to 14 (n ⫽ 74), 91.5%; and ⬎14 (n ⫽ 42), 94.6%. DISCUSSION
Our data suggest that prognosis of HF patients at any particular level of exercise capacity can vary depending on baseline therapy. This was clinically most meaningful for patients in the intermediate risk group (VO2 10 to 14 mL/min/kg). In this group, 1-year event-free survival improved from what would typically be considered less than that of posttransplant outcomes to comparable or better with optimal therapy. This is especially important since even our “optimal therapy” patients were not on optimal therapy by the current standards of care (eg, more liberal defibrillator indication currently), which is likely to improve outcomes even further. This is an important issue since most of the HF care is provided in the primary care setting. Many insurance companies do not permit longitudinal management by specialists when patients are referred for transplantation. Instead, a few visits are authorized to “complete the transplant evaluation.” Plenty of data suggest that care provided by the specialist is associated with increased compliance with standard of care.9 Thus rather than making a decision based on crosssectional data, longitudinal management of these patients by a specialist may change the eventual outcomes for patients. Another issue is what the minimum baseline medical therapy should be for patients who are enrolled in clinical trials. Many patients enrolled in trials are drawn from non-HF specialty clinics. It is not uncommon for even recent trials to show beta-blocker use in the 70% range, whereas specialty practices have shown a rate closer to 90% in practice setting.10 If patients enrolled in the trial are not on maximal medical therapy, relatively modest drugs and device benefits may show “positive” results, which may not be the case if patients were optimally treated. There are no standards of why patients are not on certain therapies prior to enrollment in clinical trials and this is left up to the site principal investigator. Thus one potential consideration is a specialist evaluation prior to enrollment of such candidates in trials to get more accurate results. Our study has several limitations. The reasons for lack of
OPTIMAL HEART FAILURE MEDICAL THERAPY
therapy could not be ascertained in all cases unless obvious or documented. Peak VO2 determinations were not assessed at specific times during treatment but were dependent on individual provider’s discretion. Finally, the “standard of care” is a moving target, and our estimations may not be accurate as newer therapies are discovered. In conclusion, our study suggests a significant, clinically meaningful effect of maximization of medical therapy on outcomes among HF patients. For transplant evaluation or enrollment in clinical trial, longitudinal management by a specialist should be considered.
REFERENCES 1. Costanzo MR, Augustine S, Bourge R, et al: Selection and treatment of candidates for heart transplantation. Circ 92:3593, 1995 2. Butler J, Khadim G, Paul KM, et al: Selection of patients for cardiac transplantation in the current era of heart failure therapy. J Am Coll Cardiol 43:787, 2004
1495 3. Pohwani AL, Murali S, Mathier MM, et al: Impact of beta-blocker therapy on functional capacity criteria for heart transplant listing. J Heart Lung Transplant 22:78, 2003 4. Peterson LR, Schechtman KB, Ewald GA, et al: The effect of beta blockers on prognostic value of peak oxygen uptake in heart failure. J Heart Lung Transplant 22:70, 2003 5. Metoprolol CR/XL Randomized Intervention Trial in Congestive Heart Failure: Effect of metoprolol CR/XL in chronic heart failure. Lancet 353:2001, 1999 6. Moss AJ, Zareba W, Hall WJ, et al: Prophylactic implantation of a defibrillator in patients with MI and reduced ejection fraction. New Engl J Med 346:877, 2002 7. Pitt B, Zannad F, Remme WJ, et al: The effect of spironolactone on morbidity and mortality in patients with severe heart failure. N Engl J Med 341:709, 1999 8. Bristow MR, Saxon LA, Boehmer J, et al: Cardiacresynchronization therapy with or without a defibrillator in advanced heart failure. N Engl J Med 350:2140, 2004 9. Hanumanthu SK, Butler J, Chomsky DB, et al: Effect of a heart failure program on hospitalization frequency and exercise tolerance. Circ 96:8, 1997 10. Butler J, Khadim G, Belue R, et al: Tolerability to betablocker therapy among heart failure patients in clinical practice. J Card Fail 9:203, 2003