Noninvasive Assessment of Cardiac Pumping Capacity During Exercise Predicts Prognosis in Patients with Congestive Heart Failure

Noninvasive Assessment of Cardiac Pumping Capacity During Exercise Predicts Prognosis in Patients With Congestive Heart Failure* Christoph Scharf, MD; Tobias Merz, MD; Wolfgang Kiowski, MD; Erwin Oechslin, MD; Christoph Schalcher, MD; and Hans Peter Brunner-La Rocca, MD

Background: Prognostic parameters in patients with congestive heart failure (CHF) are important for guiding therapeutic options. Maximal oxygen uptake (V˙O2max) is a widely used parameter for prognostic assessment in patients with CHF and correlates with exercise cardiac output; however, afterload is not taken into account. Methods: The concept of a noninvasive surrogate of cardiac power output combines exercise systolic BP (SBP), as an estimate of afterload, with V˙O2max, as an estimate of exercise cardiac output neglecting preload. Thus, a variable termed exercise cardiac power (ECP) is defined as the product of V˙O2max (expressed as a percent predicted value) and SBP (ECP, expressed as %mm Hg, is the product of V˙O2max, expressed as percentage of predicted maximum, times systolic pressure. The prognostic value of ECP obtained during routine treadmill ergospirometry was assessed in patients referred to our heart failure clinic. Patients undergoing heart transplantation were censored at the time of transplantation. Results: One hundred fifty-four patients were followed prospectively for a mean (ⴞ SE) duration of 625 ⴞ 32 days. Thirty-two patients (21%) died. ECP was the most powerful predictor of mortality, was the combined end point of mortality or hospitalization for worsening heart failure (all p < 0.001), and was an independent predictor in multivariate analysis. An ECP of < 5,000 %mm Hg indicated a poor prognosis with a 1-year mortality rate of 37%, whereas only 2% of the patients having an ECP of > 9,000 %mm Hg died during the first year. Conclusion: The integration of afterload and V˙O2max improves the prognostic value of each indicator, and provides an easily available and independent predictor of mortality and morbidity in CHF patients. This integrative concept of cardiac hydraulic performance is superior to V˙O2max and can be used in routine ergospirometry. (CHEST 2002; 122:1333–1339) Key words: cardiac assessment; congestive heart failure; exercise capacity; maximal oxygen uptake Abbreviations: CHF ⫽ congestive heart failure; CPO ⫽ cardiac power output; ECP ⫽ exercise cardiac power; ECW ⫽ exercise cardiac work; MAP ⫽ mean arterial pressure; NYHA ⫽ New York Heart Association; % mm Hg ⫽ the product of maximal oxygen uptake and systolic BP multiplied by 100; SBP ⫽ systolic BP; SWI ⫽ stroke work index; V˙O2max ⫽ maximal oxygen uptake

heart failure (CHF) is a major cause of C ongestive mortality and morbidity in Western countries. 1

Despite improved medical therapy, heart transplantation is required in patients with end-stage CHF to *From the Heart Failure and Cardiac Transplantation Unit, Division of Cardiology, University Hospital of Zu¨rich, Switzerland. Supported in part by Swiss National Funds grant No. 3200056913. Manuscript received June 18, 2001; revision accepted April 17, 2002. Correspondence to: Christoph Scharf, MD, Department of Cardiac Electrophysiology, University of Michigan, 1500 E Medical Center Dr, Ann Arbor, MI 48109; e-mail: [email protected] www.chestjournal.org

improve survival in suitable candidates. As organ shortage is a major problem, waiting time is increasing.2 Therefore, the early and reliable prediction of prognosis is important in patients with CHF.3 Major improvements in the characterization of circulatory function have been achieved by the progression from static parameters (eg, ejection fraction, filling pressures, and resting hemodynamics) to dynamic integrals of pumping capacity during exercise or inotropic stimulation. Maximal oxygen uptake (V˙o2max) measured during ergospirometry as a noninvasive approximation of cardiac output during exercise4 has proven to be one of the best measures of CHEST / 122 / 4 / OCTOBER, 2002

1333

prognosis in patients with CHF.5 However, V˙o2max has well-known limitations as a parameter of cardiovascular capacity. Of these, the neglecting of afterload is probably the most important one, since cardiac output critically depends on it. Afterload may vary significantly between patients, and recent improvements in medical therapy have changed the characteristics of CHF patients. Thus, relying only on V˙o2max may be insufficient in assessing cardiac pumping reserve. Afterload is integrated in invasively measured parameters such as cardiac power output (CPO) or stroke work index (SWI), allowing a better characterization of cardiac hydraulic pumping capacity. These parameters have been shown to be very powerful predictors of mortality in patients with cardiogenic shock,6 as well as in those with chronic CHF,7 especially when they are measured during exercise or inotropic stimulation. However, they require right heart catheterization, which is not suitable for routine clinical use in ambulatory exercise testing. Moreover, invasive hemodynamic recordings may be altered during exercise by artifacts from movement, respiration, or catheter dislocation. Thus, having the important advantages of CPO in mind, we developed an easily applicable concept that assesses cardiac pumping capacity noninvasively by integrating surrogates of exercise cardiac output and afterload. This concept was tested in patients who had been referred to our heart failure clinic. Materials and Methods The concept of the noninvasive approximation of cardiac hydraulic power reserve is closely related to the principles applied in the calculation of SWI and CPO (Table 1). CPO is the product of the cardiac output and the pressure gradient generated by both ventricles (mean arterial pressure [MAP] ⫺ central venous pressure), representing hydraulic energy per time-unit.8 We derived our concept from the following assumptions. First, V˙o2max, expressed as a percentage of predicted maximum, is used to approximate exercise cardiac output. Second, systolic BP (SBP) is considered to be a measure for afterload. In order to avoid central venous cannulation, we had to neglect the component of preload contributing to afterload or to pressure gradients. Constants and other factors used in the calculation of CPO were ignored in order to obtain an easily calculable variable.

This concept was validated in patients who were referred to our heart failure clinic from February 1996 to July 1999. The primary end point was all-cause mortality. The secondary end point was the combination of all-cause mortality and hospitalization due to worsening heart failure. Patients undergoing heart transplantation were withdrawn from further analysis at the time of transplantation. Ergospirometry Treadmill exercise testing was performed on a treadmill using a two-step protocol, as described previously.9 The ECG was monitored continuously (CASE 12 monitor; Marquette Corporation; Milwaukee, WI). Brachial BP at the heart level was measured every minute with automated cuff manometry, and pulse waves were visualized for the monitor operator. Gas exchange was assessed breath by breath (CPX/d system; Medical Graphics Corporation; St Paul, MN). The system was calibrated before each test. Oxygen levels in the expired air were analyzed by a rapidly responding zirconia fuel cell, and carbon dioxide was analyzed by an infrared analyzer. Flow measurements were performed using a disposable pneumotachograph. Initially, the patients walked at 1.0 mile per hour with an elevation of 6% for 6 min, corresponding to approximately 0.5 W per kilogram of body weight. Thereafter, both speed and elevation were increased to augment workload by approximately 0.15 W per kilogram of body weight per minute until exhaustion. Workload was assessed by calculating the power to overcome the elevation (speed ⫻ tan[grade] ⫻ g) and cover the distance.10 V˙o2max was expressed as a percentage of the predicted V˙o2max, which takes age, body mass index, and gender into account.11 Statistical Analysis Values are expressed as the mean ⫾ SE or as frequencies, when indicated. Survival analysis was performed according to the Kaplan-Meier method. The significance of each variable as a prognostic marker was first tested univariately using log-rank test for categoric and Cox proportional hazards model for continuous variables. Variables identified as significantly associated with the outcome were examined multivariately using the stepwise Cox proportional hazards model. Wald statistic coefficients express the magnitude of the predictive value of each significant variable. The significance of each variable was tested for overall mortality as well as for the combined end point of mortality and hospitalization due to worsening heart failure. Receiver operating characteristics were used to test the predictive value of the ergospirometric parameters regarding the 1-year mortality rate. A p value of ⱕ 0.05 was considered to show a statistically significant difference. All analyses were performed using a commercially available statistical package (SPSS for Windows 9.0; SPSS; Chicago, IL).

Results Table 1—Formulas for CPO and SWI and Its Noninvasive Surrogates During Exercise* CPO (W) ⫽ CO ⫻ (MAP–CVP) ⫻ 2.2167/1,000 SWI (g ⫻ m/m2) ⫽ SVI ⫻ (MAP ⫺ PCWP) ⫻ 0.0136 ECP (%mm Hg) ⫽ % predicted Vo2max ⫻ SBPmax ECW (%mm Hg/min) ⫽ % predicted Vo2max ⫻ SBPmax/HR *SBPmax ⫽ maximal SBP; PCWP ⫽ pulmonary capillary wedge pressure; HR ⫽ heart rate. 1334

Patient Characteristics at Baseline A total of 154 patients (135 men [88%]; mean age, 51 ⫾ 0.8 years) with an ejection fraction of ⱕ 40% (mean, 28 ⫾ 1%), measured by angiography or echocardiography, were included in the study. The patients had been in stable condition for at least 1 month. The main underlying diseases were coronary artery disease in 83 patients (54%), dilated cardioClinical Investigations

myopathy in 51 patients (33%), hypertensive cardiomyopathy in 9 patients (6%), and miscellaneous in 11 patients (7%). The initial evaluation included laboratory testing and ergospirometry in all patients, chest radiographs in 112 patients (73%), echocardiography in 90 patients (58%), and right heart catheterization in 106 patients (69%). All but two patients were either receiving angiotensin-converting enzyme inhibitors (139 patients; 90%) or angiotensin II type-1 receptor antagonists (13 patients; 9%), 140 patients (91%) were receiving diuretics, 91 patients (59%) were receiving digoxin, 50 patients (32%) were receiving either nitrates or molsidomine, 47 patients (31%) were receiving ␤-blockers, 36 patients (23%) were receiving amiodarone, 7 patients (5%) were receiving calcium channel blockers, 26 patients (17%) were receiving aspirin, and 112 patients (73%) were receiving oral anticoagulants. Most patients (94%) were in New York Heart Association (NYHA) class II or III. The V˙o2max was relatively preserved in a large number of patients. Thus, only 36 patients (23%) had a V˙o2max of ⬍ 14 mL/kg/min, and 11 patients (7%) had a V˙o2max of ⬍ 12 mL/kg/min. The baseline characteristics of patients in the study are summarized in Table 2. Follow-up and Predictors of Survival Thirty-two patients (21%) died after a mean time of 625 ⫾ 32 days (median, 550 days). The actual Table 2—Baseline Characteristics of All Patients* Parameter

Values

Age, yr Gender Male Female NYHA class I and II NYHA class III and IV Heart rate at rest, beats/min SBP at rest, mm Hg Maximal heart rate during exercise, beats/min SBPmax during exercise, mm Hg Serum sodium, mmol/L Serum creatinine, ␮mol/L Cardiac thoracic ratio, % PCWP, mm Hg Cardiac index, beats/min/m2 V˙o2max mL/kg/min % predicted Respiratory exchange ratio at exercise Exercise capacity, W/kg V˙e/V˙co2 ratio ECP, %mm Hg ECW, %mm Hg/min

51.7 ⫾ 0.8 135 (88) 19 (12) 81 (53) 73 (47) 84.5 ⫾ 1.5 104.4 ⫾ 1.3 137.2 ⫾ 2.2 128.1 ⫾ 2.3 138.9 ⫾ 0.3 111.7 ⫾ 2.9 54.8 ⫾ 0.6 17.0 ⫾ 0.8 2.53 ⫾ 0.05 18.8 ⫾ 0.4 63.5 ⫾ 1.7 1.12 ⫾ 0.01 1.48 ⫾ 0.05 36.9 ⫾ 0.6 8,390 ⫾ 330 3,680 ⫾ 130

*Values given as mean ⫾ SEM or No. (%). V˙e ⫽ minute ventilation; V˙co2 ⫽ carbon dioxide output. See Table 1 for abbreviations not used in the text. www.chestjournal.org

survival rates after 1, 2, and 3 years were 86 ⫾ 3%, 79 ⫾ 4%, and 75 ⫾ 4%, respectively. As depicted in Table 3, a variety of clinical and ergospirometric variables were related to prognosis in this population. Exercise cardiac power (ECP, expressed as %mm Hg, which is the product of SBP times the percentage of predicted V˙o2max) emerged as the most powerful univariate predictor of mortality, even better than V˙o2max (expressed as a percent predicted value) and SBP at peak exercise in univariate analysis. In a bivariate Cox regression, they were independently related to mortality (V˙o2max hazard ratio ⫽ 0.75 [per 10%]; and SBP hazard ratio ⫽ 0.82 [per 10 mm Hg], respectively [both p ⬍ 0.02]). However, their predictive value in the bivariate analysis was equal to ECP alone. In a multivariate Cox-regression analysis, ECP was the only independent predictor of mortality. Survival curves of patients with ECP ⬍ 5,000 %mm Hg, 5,000 to 9,000 %mm Hg, and ⬎ 9,000 %mm Hg are depicted in Figure 1. An ECP ⬍ 5,000 %mm Hg indicated a poor prognosis, with a 1-year mortality rate of 37%. On the other hand, patients with an ECP of ⱖ 9,000 %mm Hg had a good prognosis with a 1-year survival rate of 98%. Figure 2 shows that there was a considerable overlap between V˙o2max and ECP in the three groups. Importantly, of 31 patients with a poor prognosis that had been indicated by an ECP of ⬍ 5,000 %mm Hg, only 7 (23%) were correctly identified by a V˙o2max of ⬍ 12 mL/min/kg alone. Conversely, of 36 patients with V˙o2max of ⬍ 14 mL/kg, 10 patients were in lower risk groups with an ECP between 5,000 and 9,000 %mm Hg and 4 patients had an excellent prognosis with an ECP of ⬎ 9,000 %mm Hg. The survival rates free of hospitalization were 79 ⫾ 3%, 68 ⫾ 4%, and 57 ⫾ 5%, respectively, after 1, 2, and 3 years. Significant univariate predictors for the combined end point of mortality and hospitalization due to worsening heart failure are summarized in Table 4. In multivariate analysis, ECP emerged as

Table 3—Univariate Predictors of Mortality* Predictors

Hazard Ratio Wald Statistic p Value

ECP (1,000 %mm Hg) ECW (1,000 %mm Hg/min) Exercise SBP (10 mm Hg) V˙o2max (10% predicted) Exercise capacity V˙e/V˙co2 ratio V˙o2max Resting SBP (10 mm Hg) NYHA class Serum sodium

0.76 0.52 0.76 0.68 0.35 1.07 0.91 0.71 1.94 0.91

13.4 13.4 12.4 11.1 7.3 8.6 6.1 6.3 3.7 3.1

0.0002 0.0002 0.0005 0.0008 0.007 0.004 0.02 0.02 0.06 0.08

*See Table 2 for abbreviations not used in the text. CHEST / 122 / 4 / OCTOBER, 2002

1335

Table 4 —Univariate Predictors for the Combined End Point of All-Cause Mortality and Hospitalization Due to Worsening Heart Failure* Predictors

Hazard Ratio Wald Statistic p Value

ECW (1,000 %mm Hg/min) ECP (1,000 %mm Hg) Exercise SBP (10 mm Hg) V˙o2max (10% predicted) Exercise capacity Resting SBP (10 mm Hg) Serum sodium V˙o2max V˙e/V˙co2 ratio NYHA class

Figure 1. Actuarial survival in patients with ECP levels of ⬍ 5,000 %mm Hg, 5,000 to 9,000 %mm Hg, and ⬎ 9,000 %mm Hg is shown. While patients with preserved ECP levels had an excellent prognosis, those with limited ability to increase cardiac exercise power were at a considerable risk of death. Censored ⫽ withdrawn from further analysis at time of transplantation.

the most powerful predictor of the combined end point. Again, the results of a bivariate analysis of V˙o2max and exercise SBP at peak exercise were not superior to the predictive value of ECP alone. As depicted in Figure 3, patients with ECP of ⬍ 5,000 %mm Hg had a very high likelihood of reaching the combined end point (ie, almost 50% of these patients reached the combined end point within 1 year). Receiver operating characteristic curves showed good predictive value for 1-year mortality by ECP (134 patients [surviving patients with follow-up of ⬍ 1 year were excluded]). The mean area under the curve was 0.73 ⫾ 0.06 (p ⬍ 0.001). The resulting values for

Figure 2. V˙o2max values normalized to body weight in patients with ECP levels of ⬍ 5,000 %mm Hg, 5,000 to 9000 %mm Hg, and ⬎ 9,000 %mm Hg are shown. Despite significant differences (p ⬍ 0.001), there was a considerable overlap between the groups. 1336

0.80 0.87 0.84 0.79 0.46 0.77 0.89 0.96

11.5 11.0 10.1 9.4 8.2 7.4 6.8 2.9

0.0007 0.0009 0.002 0.003 0.005 0.007 0.01 0.08 ⬎ 0.1 ⬎ 0.1

*See Table 2 for abbreviations not used in the text.

V˙o2max and SBP at peak exercise were 0.70 ⫾ 0.07 (p ⬍ 0.005) and 0.70 ⫾ 0.06 (p ⬍ 0.006), respectively. The additional division of ECP by heart rate, termed exercise cardiac work (ECW), did not provide any further improvement of prognostic power (Tables 3, 4). Discussion Our study indicated that ECP, an easily obtainable surrogate of peak CPO, is superior to V˙o2max alone in prognostic assessments of patients with CHF. The quantitative integration of afterload, using exercise SBP as a surrogate, has been proven to increase the prognostic power of ergospirometry. The product of SBP at peak exercise and the percentage of the predicted V˙o2max can be quantified in every patient undergoing routine ergospirometry and should be prospectively evaluated in a larger number of CHF patients.

Figure 3. Survival free of hospitalization with an ECP level of ⬍ 5,000 %mm Hg, 5,000 to 9,000 %mm Hg, and ⬎ 9,000 %mm Hg is shown. Nearly two thirds of patients with limited ability to increase cardiac power reached one of these end points. Censored ⫽ withdrawn from further analysis. Clinical Investigations

Conceptual Considerations So far, V˙o2max obtained by ergospirometry is the most widely accepted parameter with which to assess prognosis in CHF patients.12 This can be illustrated by the good correlation of V˙o2max to invasively measured exercise cardiac output.7,13,14 However, cardiac output, which is critically dependent on preload and afterload, provides quite a limited and insensitive assessment of ventricular function or of myocardial contractility.15 Accordingly, the value of oxygen uptake measurements alone maybe limited in CHF patients with dilated ventricles, which are extremely responsive to changes in afterload. For instance, in CHF patients with an elevated adrenergic tone and an inappropriately constricted arterial bed, cardiac output can be lowered in the presence of disproportionately elevated BP (ie, afterload).16 As a consequence, V˙o2max may be severely reduced and may underestimate cardiac pumping capacity in a substantial number of patients. Inversely, V˙o2max may be relatively preserved in patients on optimal afterload-lowering medications (angiotensinconverting enzyme inhibitors), despite severely reduced pumping capacity. In addition, patients lacking physical training may present with severely reduced V˙o2max, and cardiac pumping capacity may be quantified correctly only when disproportionately elevated BP values are taken into account. The hydraulic work of the ventricles can be calculated by the product of the stroke volume and the pressure gradients, resulting in the SWI (Table 1). In patients with acute cardiogenic shock, SWI is one of the most powerful predictors of mortality.6 A further integration of time and both ventricles leads to CPO, which represents hydraulic energy output per timeunit. Both parameters consider afterload and have shown excellent correlations to exercise capacity8 and prognosis in patients with CHF.17 Because of the inherent problems of invasive hemodynamic measurements, particularly during exercise, we sought to derive a concept of noninvasive, easily available surrogates of CPO. However, this goal could be achieved only at the expense of certain assumptions. Thus, the original formulas of CPO and SWI (Table 1) were simplified based on the following considerations. According to the Fick principle, cardiac output can be quantified by oxygen uptake, with a linear correlation during exercise.7,13,14,18,19 In addition, V˙o2max correlates to forward cardiac output in patients with mitral valve regurgitation, and not to the regurgitant fraction.20 The adjustment of V˙o2max to age, body mass index, and gender has been shown to be advantageous over normalization only to body weight, for prognostic assessment in our www.chestjournal.org

study and others,21 but not in all studies.22 Nevertheless, comparisons between different patient populations (eg, adipose vs asthenic constitution) may be more accurately performed by normalizing V˙o2max not only to body weight, but to lean body mass,23 which was an indicator that was not used in the present study. In the original formula for CPO, the pressure gradient between MAP and central venous pressure is used to quantify the hydraulic force that is generated by both ventricles. In order to avoid central venous cannulation, we had to neglect the latter, which can be justified by the fact that dilated ventricles contracting on the flat side of the FrankStarling curve become more dependent on afterload than on preload. Moreover, central venous pressure undergoes smaller variations than do arterial pressures, and its measurements are more susceptible to fluid intake, diuretic administration, and respiratory artifacts during exercise. Therefore, inaccuracy may not be substantially increased by neglecting cardiacfilling pressures. By neglecting preload, we used SBP as the best surrogate of maximal transmural wall tension, which is called afterload. The potential overestimation of SBP due to aortic stiffness may be less important in the interesting range of SBP levels ⬍ 120 mm Hg at lower stroke volumes. No constants were included, because we did not calculate a precise index such as SWI or CPO. Prognostic Value of ECP ECP emerged as the most powerful and independent predictor of morbidity and mortality in our study population of CHF patients and was superior to V˙o2max alone. The overall 1-year mortality rate of 14% rose to 37% in patients with an ECP of ⬍ 5,000 %mm Hg, indicating that this parameter identified correctly a high-risk group from a group of patients with advanced heart failure, although resting cardiac output or and V˙o2max was still relatively preserved (ie, ⬎ 14 mL/kg) in a substantial number of patients. Therefore, the listing for transplantation may be considered earlier in these patients in view of the current longer waiting times for organ transplantation. We used this cutoff value because previous studies reported bad outcomes in patients with V˙o2max levels of ⬍ 50% of the predicted value21 and a peak exercise SBP of ⬍ 100 mm Hg.24 Vice versa, an ECP of ⱖ 9,000 %mm Hg identified patients with an excellent survival rate of 95% at 4 years, even in patients with a V˙o2max of ⬍ 14 mL/kg. Overall, V˙o2max normalized to body weight did not distinguish equally well between these patients groups, illustrating the impact of afterload on prognostic CHEST / 122 / 4 / OCTOBER, 2002

1337

assessments in CHF patients. On the other hand, this may be in part related to the fact that V˙o2max normalized to body weight does not take differences in body fat into account. The correction of V˙o2max to lean body mass significantly improved the prognostic power of exercise testing.23 This is in line with our results, which showed that V˙o2max normalized not just to body weight is superior as a prognostic marker. There was no improvement in the prognostic power of ECP by additional division by heart rate. This is in line with results indicating that there is no advantage in dividing V˙o2max by heart rate, thus estimating stroke volume instead of cardiac output noninvasively.25 The integration of heart rate, as in CPO and ECP, is essential in global cardiac function testing. Although a cross-sectional study is not able to investigate the influence of medication on prognostic markers, it is tempting to speculate that ECP may be less dependent on medication than is V˙o2max. This may be particularly true for medication that directly influences afterload. Although the effects of shortterm therapy with ␤-blockers with respect to exercise capacity (in particular V˙o2max) have not been uniform,26 –28, it may well be that ␤-blockers increase both V˙o2max and the maximally achievable BP in the long term by improving left ventricular performance. ECW might be superior in this regard since the heart rate-reducing properties of ␤-blockers lead to larger increases of ECW and SWI than of CPO and ECP. However, our cross-sectional study is unable to prove this hypothesis, and prospective studies simultaneously investigating the influence of medical therapy on both exercise performance and prognosis are needed to properly address this issue. Limitations Although our patients underwent routine ergospirometry at the initial consultation and were observed prospectively, the retrospective evaluation of our concept may lead to systemic error. Therefore, the validation of the calculation of ECP in studies with prospective designs is required to assess its true prognostic power in CHF patients. In addition, various study populations need to be tested to see whether the prognostic power of these parameters is dependent on patient selection. However, given the simple calculation of the proposed parameter that is available during routine ergospirometry, such prospective information should be readily obtainable in ongoing or future studies. In addition, the differences in the predictive values of V˙o2max, exercise SBP, and ECP were relatively small. However, in a bivariate Cox regression analysis, V˙o2max and exercise BP independently con1338

tributed to the prognostic assessment both of mortality and of the combined end point. Therefore, and because of the aforementioned theoretical considerations, their combination may be advantageous, although larger studies have to confirm the superiority of our concepts over the measurement of V˙o2max alone. Finally, this concept has the same limitations as ergospirometry (ie, in patients with significant valvular heart disease and pulmonary disease). Conclusion The quantification of a readily available surrogate of CPO that integrates V˙o2max and afterload significantly improved the prognostic value of ergospirometry. The concept should be independent of medication, physical training, or inappropriate vasoconstriction and seems promising in CHF patients with dilated ventricles who are more responsive to changes in afterload than to those in preload. An ECP value of ⬍ 5,000 %mm Hg indicates a first-year mortality rate of 37% and may justify an early listing of a patient for transplantation, even when V˙o2max is still relatively preserved (ie, ⬎ 14 mL/kg/min). However, prospective studies are needed to prove this concept and the superiority of measuring V˙o2max alone. References 1 Ho KK, Anderson KM, Kannel WB, et al. Survival after the onset of congestive heart failure in Framingham Heart Study subjects. Circulation 1993; 88:107–115 2 Rodeheffer RJ, Naftel DC, Stevenson LW, et al. Secular trends in cardiac transplant recipient and donor management in the United States, 1990 to 1994: a multi-institutional study; Cardiac Transplant Research Database Group. Circulation 1996; 94:2883–2889 3 Oechslin E, Brunner-La Rocca HP, Solt G, et al. Prognosis of medically treated patients referred for cardiac transplantation. Int J Cardiol 1998; 64:75– 81 4 Tan LB, Bain RJ, Littler WA. Assessing cardiac pumping capability by exercise testing and inotropic stimulation. Br Heart J 1989; 62:20 –25 5 O’Connell JB, Bourge RC, Costanzo-Nordin MR, et al. Cardiac transplantation: recipient selection, donor procurement, and medical follow-up; a statement for health professionals from the Committee on Cardiac Transplantation of the Council on Clinical Cardiology, American Heart Association. Circulation. 1992; 86:1061–1079 6 Tan LB, Littler WA. Measurement of cardiac reserve in cardiogenic shock: implications for prognosis and management. Br Heart J 1990; 64:121–128 7 Metra M, Faggiano P, D’Aloia A, et al. Use of cardiopulmonary exercise testing with hemodynamic monitoring in the prognostic assessment of ambulatory patients with chronic heart failure. J Am Coll Cardiol 1999; 33:943–950 8 Bain RJ, Tan LB, Murray RG, et al. The correlation of cardiac power output to exercise capacity in chronic heart failure. Eur J Appl Physiol 1990; 61:112–118 Clinical Investigations

9 Brunner-La Rocca HP, Weilenmann D, Follath F, et al. Oxygen uptake kinetics during low level exercise in patients with heart failure: relation to neurohormones, peak oxygen consumption, and clinical findings. Heart 1999; 81:121–127 10 American College of Sports Medicine. Guidelines for grades exercise testing, and prescription. 3rd ed. Philadelphia, PA: Lea & Febiger, 1986; 157–172 11 Wasserman K, Hansen JE, Sue DY, et al. Principles of exercise testing and interpretation. 2nd ed. Philadelphia, PA: Lea & Febiger, 1994 12 Mancini D, LeJemtel T, Aaronson K. Peak VO(2): a simple yet enduring standard. Circulation 2000; 101:1080 –1082 13 Weber KT, Kinasewitz GT, Janicki JS, et al. Oxygen utilization and ventilation during exercise in patients with chronic cardiac failure. Circulation 1982; 65:1213–1223 14 Saxon LA, Stevenson WG, Middlekauff HR, et al. Predicting death from progressive heart failure secondary to ischemic or idiopathic dilated cardiomyopathy. Am J Cardiol 1993; 72: 62– 65 15 Braunwald E. Heart disease: assessment of cardiac function. 4th ed. Philadelphia, PA: WB Saunders, 1992; 419 – 440 16 Zelis R, Flaim SF. Alterations in vasomotor tone in congestive heart failure. Prog Cardiovasc Dis 1982; 24:437– 459 17 Roul G, Moulichon ME, Bareiss P, et al. Prognostic factors of chronic heart failure in NYHA class II or III: value of invasive exercise haemodynamic data. Eur Heart J 1995; 16:1387– 1398 18 Mancini D, Katz S, Donchez L, et al. Coupling of hemodynamic measurements with oxygen consumption during exercise does not improve risk stratification in patients with heart failure. Circulation 1996; 94:2492–2496 19 Szlachcic J, Massie BM, Kramer BL, et al. Correlates and prognostic implication of exercise capacity in chronic congestive heart failure. Am J Cardiol 1985; 55:1037–1042 20 Leung DY, Griffin BP, Snader CE, et al. Determinants of

www.chestjournal.org

21

22

23

24

25

26

27

28

functional capacity in chronic mitral regurgitation unassociated with coronary artery disease or left ventricular dysfunction. Am J Cardiol 1997; 79:914 –920 Stelken AM, Younis LT, Jennison SH, et al. Prognostic value of cardiopulmonary exercise testing using percent achieved of predicted peak oxygen uptake for patients with ischemic and dilated cardiomyopathy. J Am Coll Cardiol 1996; 27:345–352 Aaronson KD, Mancini DM. Is percentage of predicted maximal exercise oxygen consumption a better predictor of survival than peak exercise oxygen consumption for patients with severe heart failure? J Heart Lung Transplant 1995; 14:981–989 Osman AF, Mehra RM, Lavie CL, et al. The incremental prognostic importance of body fat adjusted peak oxygen consumption in chronic heart failure. J Am Coll Cardiol 2000; 7:2126 –2131 Osada N, Chaitman BR, Miller LW, et al. Cardiopulmonary exercise testing identifies low risk patients with heart failure and severely impaired exercise capacity considered for heart transplantation. J Am Coll Cardiol 1998; 31:577–582 Cohen-Solal A, Barnier P, Pessione F, et al. Comparison of the long-term prognostic value of peak exercise oxygen pulse and peak oxygen uptake in patients with chronic heart failure. Heart 1997; 78:572–576 Randomised, placebo-controlled trial of carvedilol in patients with congestive heart failure due to ischaemic heart disease: Australia/New Zealand Heart Failure Research Collaborative Group. Lancet 1997; 349:375–380 Genth-Zotz S, Zotz RJ, Sigmund M, et al. MIC trial: metoprolol in patients with mild to moderate heart failure; effects on ventricular function and cardiopulmonary exercise testing. Eur J Heart Fail 2000; 2:175–181 Metra M, Giubbini R, Nodari S, et al. Differential effects of beta-blockers in patients with heart failure: a prospective, randomized, double-blind comparison of the long-term effects of metoprolol vs carvedilol. Circulation 2000; 102:546–551

CHEST / 122 / 4 / OCTOBER, 2002

1339