0
Original Research: CARDIOVASCULAR DISEASE |

Oxygen Uptake Efficiency Plateau Best Predicts Early Death in Heart FailureOxygen Uptake Efficiency: Early Death Predictor FREE TO VIEW

Xing-Guo Sun, MD; James E. Hansen, MD, FCCP; William W. Stringer, MD, FCCP; American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure)
Author and Funding Information

From the Department of Medicine, Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, David Geffen School of Medicine at UCLA, St. John’s Cardiovascular Research Center, Torrance, CA.

Correspondence to: Xing-Guo Sun, MD, State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100037, China, e-mail: xgsun@labiomed.org


Dr Sun is currently at the State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College (Bejing, China).

Funding/Support: The study was partially supported by Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (http://www.chestpubs.org/site/misc/reprints.xhtml).


© 2012 American College of Chest Physicians


Chest. 2012;141(5):1284-1294. doi:10.1378/chest.11-1270
Text Size: A A A
Published online

Background:  The responses of oxygen uptake efficiency (ie, oxygen uptake/ventilation = V˙ o2/V˙ e) and its highest plateau (OUEP) during incremental cardiopulmonary exercise testing (CPET) in patients with chronic left heart failure (HF) have not been previously reported. We planned to test the hypothesis that OUEP during CPET is the best single predictor of early death in HF.

Methods:  We evaluated OUEP, slope of V˙ o2 to log(V˙ e) (oxygen uptake efficiency slope), oscillatory breathing, and all usual resting and CPET measurements in 508 patients with low-ejection-fraction (< 35%) HF. Each had further evaluations at other sites, including cardiac catheterization. Outcomes were 6-month all-reason mortality and morbidity (death or > 24 h cardiac hospitalization). Statistical analyses included area under curve of receiver operating characteristics, ORs, univariate and multivariate Cox regression, and Kaplan-Meier plots.

Results:  OUEP, which requires only moderate exercise, was often reduced in patients with HF. A low % predicted OUEP was the single best predictor of mortality (P < .0001), with an OR of 13.0 (P < .001). When combined with oscillatory breathing, the OR increased to 56.3, superior to all other resting or exercise parameters or combinations of parameters. Other statistical analyses and morbidity analysis confirmed those findings.

Conclusions:  OUEP is often reduced in patients with HF. Low % predicted OUEP (< 65% predicted) is the single best predictor of early death, better than any other CPET or other cardiovascular measurement. Paired with oscillatory breathing, it is even more powerful.

Figures in this Article

Despite recent improvements in therapy, chronic left heart failure (HF) frequently requires hospitalization or is lethal.13 We postulated that management of such patients could be improved by adding more sensitive predictors, further refining well-known predictors, or by combining predictors. Currently, in such patients, gas exchange measurements during incremental cardiopulmonary exercise testing (CPET) such as peak oxygen uptake (peak V˙ o2) at standard conditions of 0ºC, 760 mm Hg, dry (STPD); anaerobic threshold (AT); peak oxygen pulse; lowest ratio of minute ventilation (V˙ e) at body temperature, ambient atmospheric pressure, saturated with water vapor to minute CO2 output (V˙ co2) STPD; V˙ e-vs-V˙ co2 slope; oscillatory breathing (OB) pattern310; and the slope of V˙ o2 to log V˙ e (oxygen uptake efficiency slope [OUES])1121 have been found to be useful in predicting mortality and morbidity and guiding therapy.310 However, the full exercise response pattern of oxygen uptake efficiency (OUE) (V˙ o2/V˙ e, mL/L) and specifically the measurement of its highest average plateau (OUEP) (mL/L) have not been evaluated.

As the CPET core laboratory for two St. Jude Medical multicenter studies3,22,23 involving 508 patients with HF, we previously reported the CPET parameters listed here, plus resting and exercise heart rates, systemic BP, 6-min walk distance, left ventricular ejection fraction, quality of life, and New York Heart Association classification for their value in predicting 6-month mortality and cardiac hospitalizations. Our recent report presenting OUEP and OUES reference values24 for normal subjects allowed us to now evaluate OUEP and OUES as additional prognosticating parameters in patients with HF.

Setting and Participants

With approval from the Institutional Review Board Committee for Pre-Clinical/Clinical Trials of Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center (12105-01), we acted as the CPET core laboratory for St. Jude Medical biventricular pacing and cardiac resynchronization therapy (BVP/CRT) trials: Resynchronization for Hemodynamic Treatment for Heart Failure Management (RHYTHM) and Cardiac Resynchronization Therapy in Patients With Heart Failure and Narrow QRS Complexes (ReThinQ) (ClinicalTrials.gov number, NCT00132977) as previously reported.3,22,23 After obtaining informed consent, satisfactory baseline CPET studies were obtained in 508 patients with low (< 35%) ejection fraction HF. Each received a biventricular implantable cardioverter defibrillator and was randomized into pacemaker OFF or pacemaker ON subgroups.

Quality Control, Protocol, Gas Exchange Measurement, and Follow-up

Sixty-nine laboratories, using six different commercial systems, sent raw breath-by-breath CPET data to our laboratory for monitoring/review/calculation/analyses. Repeat studies were requested and obtained for any studies considered questionable.3 Author-investigators were blinded to patient assignments, catheterization findings, and imaging results at the time of data analysis. The mortality outcome was all-cause death; morbidity outcome included death or a cardiac hospitalization for ≥ 24 h within the 6-month follow-up period.

The exercise protocol required multiple measurements during 3 min of rest, 3 min of unloaded cycling or comparable warm-up on the treadmill, followed by progressively increasing work by 5 to 15 W/min (usually 10 W/min) in a ramp pattern or 1-min step intervals to maximum tolerance. Fifty-eight percent of studies were performed on the cycle ergometer and 42% on the treadmill. For measurements other than OUE, the methodology and prognostic values of all other data listed in the introduction and CPET have been previously reported.3,22,2530

OUE Measurement

Three relationships between V˙ o2 and Ve data were graphically evaluated, identified, and calculated for each subject as follows: (1) OUEP was the 90 s average of the highest consecutive values for V˙ o2(mL/min, STPD)/V˙ e (L/min, at body temperature, ambient atmospheric pressure, saturated with water vapor) (Fig 1); (2) OUE at the AT (OUE@AT) was the 60 s average of consecutive values at and immediately before the AT (Fig 1); and (3) OUES was the slope of linear regression of V˙ o2(L/min)-vs-logV˙ e(L/min) for all exercise values. Percent of predicted values for OUEP and OUES were based on age, sex, and body size.24 The lowest V˙ e/V˙ co2 were determined over a 90-s period.3,26

Figure Jump LinkFigure 1. OUE patterns during cardiopulmonary exercise testing (CPET) in a normal subject and three patients with HF. The response patterns of OUE (ie, V˙ o2/V˙ e) from rest to peak exercise during CPET in the normal subject (gray solid circle)24 and patients with HF (black, dark gray, and gray inverted triangle) and are reasonably parallel. The OUE varies minimally during rest, increases during warmup, and then increases further to its highest level or plateau (90-s average oxygen uptake efficiency plateau, mL/L [OUEP]) just before the AT. Thereafter, the OUE invariably decreases as exercise intensity increases. As the severity of HF increases, the OUEP decreases. An oscillatory breathing (OB) pattern was found in many patients with severe and very severe HF. AT = anaerobic threshold; HF = chronic left heart failure; O2 = oxygen; OUE = oxygen uptake efficiency; V˙ o2/V˙ e = minute oxygen uptake/minute ventilation.Grahic Jump Location
Oscillatory Breathing

The breathing pattern was scored as oscillatory breathing positive (OB+) or negative (OB). The determination of an OB+ pattern required all of the following: (1) the amplitude of oscillatory ventilation must exceed 30% of concurrent mean ventilation3,7,8,31,32; (2) three or more consecutive cyclic fluctuations of ventilation or ≥ 60% of entire CPET period3,710,31,32; (3) an oscillatory frequency of 40 to 140 (56 ± 5.2) s3; and (4) visible oscillations of similar frequency in three or more of the following variables: oxygen pulse, V˙ o2, V˙ co2, V˙ e/V˙ co2, respiratory exchange ratio, or end-tidal pressures for oxygen (O2) and CO2.3

Statistical Analyses

Data, expressed as mean ± SD unless specially noted, were first analyzed by unpaired t and χ2 tests. A two-tailed P value < 0.05 was considered significant. Using all-cause mortality and morbidity as outcomes, receiver operator characteristic curves (ROCs) analyses were performed. Optimal cutoff values were determined by the shortest distance on each ROC for all CPET and non-CPET variables, including steps of whole numbers for OUEP, two decimals for OUES, and 1% for % predicted .3,33 Sensitivity, specificity and area under the curve of ROC (AUC) were calculated. Using optimal cutoff values of all CPET and non-CPET variables without predetermined priorities, univariate and multivariate regression analyses disclosed prognostic predictors for the optimal model.3,34,35 To evaluate OB, step-by-step repeated multivariate analyses were used to obtain the next model by excluding the prior most significant variables. OR were calculated for single and combined predictors and tested by Wald logistic regression model. Kaplan-Meier survival curves and log-rank tests were used to compare single and combined predictors. For the dual-predictors model, Kaplan-Meier survival curves compared all four options. For relationships of peak V˙ o2 with OUEP and OUES, linear regression analyses of patients with HF vs normal subjects were compared. SPSS Statistics 17.0, Origin 7, and Sigma Plot 8.0 statistical software were used; graphs were generated using Origin 7.

Demographics, Usual Cardiovascular, and OB Data

Satisfactory resting cardiovascular measurements and CPET were obtained for 355 male and 153 female stable patients with HF (Table 1).3 Thirty-six of the 508 baseline studies were not of optimal quality and needed to be repeated. The patients’ mean age, height, and body weight were 64 years, 172 cm, and 87 kg, respectively. New York Heart Association functional classifications were II (n = 52), III (n = 440), and IV (n = 16). There were 362 ON and 146 OFF for the BVP/CRT settings; 271 patients in RHYTHM (QRS interval ≥ 150 milliseconds) and 237 patients in RethinQ (QRS interval ≤ 130 milliseconds, including 67 nonrandomized patients in the registry group). All 508 patients received angiotensin-converting enzyme inhibitors (82%) or angiotensin receptor blockers (18%); 86% received β-adrenergic blockers. All patients were accounted for during 6-month follow-up: 19 patients (3.7%) died, and 107 patients (21%) died or had one or more cardiac hospitalizations lasting greater than 24 h.3

Table Graphic Jump Location
Table 1 —Demographics and Selected Resting and Exercise Measurements in Patients With HF

Data are mean ± SD unless otherwise specified. AT = oxygen uptake rate at anaerobic threshold; BVP/CRT-ON or -OFF = biventricular pacemaker/cardiac resynchronization setting is “on” or “off”; CPET = cardiopulmonary exercise testing; DBP = diastolic BP; HF = chronic left heart failure; HR = heart rate; LVEF = left ventricular ejection fraction; NYHA = New York Heart Association classification; O2 pulse = oxygen uptake per min/heart rate; OB = oscillatory breathing; RER = respiratory gas exchange ratio (ie, CO2 output per min [standard conditions of 0ºC, 760 mm Hg, dry]/oxygen update per min [standard conditions of 0ºC, 760 mm Hg, dry]); SBP = systolic BP; V˙ co2 = CO2 output per min (standard conditions of 0ºC, 760 mm Hg, dry); V˙ e = minute ventilation (body temperature, ambient atmospheric pressure, saturated with water vapor); V˙ e/V˙ co2 = ratio of minute ventilation (body temperature, ambient atmospheric pressure, saturated with water vapor) to CO2 output per min (standard conditions of 0ºC, 760 mm Hg, dry); V˙ e-vs-V˙ co2 slope = minute ventilation (body temperature, ambient atmospheric pressure, saturated with water vapor) as a function of CO2 output per min (standard conditions of 0ºC, 760 mm Hg, dry); V˙ o2 = oxygen uptake per min (standard conditions of 0ºC, 760 mm Hg, dry).

a 

Quality of life was evaluated with the Minnesota Living with Heart Failure Questionnaire, with scores ranging from 0 to 105 and with higher scores indicating a poorer quality of life.23

The mean resting heart rates and systolic and diastolic BPs were 77 beats/min and 118/71 mm Hg, respectively. Mean left ventricle ejection fractions were 26%; 6-min walking distances were 291 m. Mean peak V˙ o2, peak oxygen pulse, and the lowest V˙ e/V˙ co2 were, respectively, 1.00 L/min (53% predicted), 9.4 mL/beat (77% predicted), and 39.2 (137% predicted). The mean AT, detected in 491 patients, was 0.72 L/min (57% predicted). Two hundred fifty-eight patients were identified as OB+ and 250 as OB.3 The ORs, 95% CIs, and AUC analyses for mortality for resting measures (including cardiac catheterization and echocardiography) and many CPET measures excluding OUEP and OUES are shown in Table 2. Other than CPET measurements, only higher resting heart rate, lower resting systolic BP, and shorter 6-min walking tests had OR confidence limits over 1.00. For AUC, only 6-min walking distance was statistically significant. In contrast, OR and AUC analyses of CPET measures yielded highly statistically significant findings for absolute or % predicted peak V˙ o2, peak O2 pulse, AT and V˙ e:V˙ co2 relationships, and some statistical significance for peak BP, ventilation, and workload. Essential morbidity data are given in Figure 3 and its legend.

Table Graphic Jump Location
Table 2 —OR, AUC, With 95% CI, and Best Cutoff Values for Mortality in HF

AUC = area under the receiver operating characteristics curve. See Table 1 legend for expansion of other abbreviations.

OUE Responses

Figure 1 shows the response patterns of OUE (ie, ratio of V˙ o2/V˙ e in mL/L) from rest to peak exercise in a normal subject and three patients with different severity HF. In all patients, OUE increased when exercise started, soon reached a plateau (OUEP) just before the AT, and then decreased until exercise ended. Because gas exchange data were often oscillatory at rest and with early exercise, OUEP values for each individual were their 90-s average highest values, while OUE@AT values were averaged for 60 s at and before AT. Figure 2A indicates the near identity of the OUEP and OUE@AT in all 491 patients with HF who had a detectable AT (r = 0.99, P < .0001). Figure 2B shows the more scattered relationship between the OUEP and OUES.

Figure Jump LinkFigure 2. OUEP, OUE@AT, and OUES in patients with HF and healthy subjects. Each symbol is an individual patient with HF (open black inverted triangle), normal subject (solid gray circle), or very fit (open gray circle) subject. The regression lines are for normal subjects only (gray) or all patients with HF (black). The equations are for patients with HF only because those for normal subject data were previously presented.24 A, OUEP vs OUE@AT indicates their similarity for normal subjects or patients with HF. B, OUEP vs OUES. The wider distribution of data and differing regression lines indicate differences between the two groups. OUE@AT = 60-s average oxygen uptake efficiency at the anaerobic threshold (mL/L); OUES = oxygen uptake efficiency slope, slope of V˙ o2 vs −logV˙ e [L/min/log (L/min)]. See Figure 1 legend for expansion of other abbreviations.Grahic Jump Location

Table 3 compares OUEP and OUES values of patients with HF to normal subjects and each other. Values of both absolute and % predicted OUEP and OUES were markedly decreased in patients with HF (P < .001) and useful in predicting mortality, morbidity, OB, and severity (all P < .01 to .0001).

Table Graphic Jump Location
Table 3 —OUE Values Linked to Mortality, Morbidity, and Severity of HF

Values are mean ± SD (range). The P values between the positive and negative for mortality and morbidity are by unpaired t test. − = negative; + = positive; ANOVA = analysis of variance; OUE = oxygen uptake efficiency (ie, V˙ o2/V˙ e, mL/L); OUEP = 90-s average oxygen uptake efficiency plateau (mL/L); OUES = slope of V˙ o2 to logV˙ e [L/min/log(L/min)]. See Table 1 legend for expansion of other abbreviations. Comparisons of functional severity by ANOVA are

vs “~Normal” group:

a 

P < .05.

b 

P < .01.

c 

P < .001.

d 

P < .0001.

vs “Mild” group:

e 

P < .05.

f 

P < .01.

g 

P < .001.

h 

P < .0001.

vs “Moderate” group:

i 

P < .01.

j 

P < .001.

k 

P < .0001.

vs “Severe” group:

l 

P < .0001.

Both OUEP and OUES as either absolute or % predicted were powerful predictors (P < .001) for early death, but OUEP measurements were superior (Tables 3, 4). With the best cutoff of 25 mL/L and 65% predicted , OUEP had the highest AUC of 0.729 and 0.748, respectively, which significantly (P < .001) increased to 0.783 and 0.800 when combined with OB. For OUES, AUC were 0.680 (absolute) and 0.731 (% predicted ), with greater (P < .05) increases to 0.724 and 0.765, respectively, with OB. AUC % predicted values were always higher than absolute values (P < .05).

Table Graphic Jump Location
Table 4 —Best Cutoff Values, Sensitivity, Specificity, AUC, and 95% CI of OUE for Mortality in HF

See Table 1-3 legends for expansion of abbreviations.

The mortality OR analyses for OUEP were consistently higher than for OUES (P < .01) (Fig 3A), while OR of % predicted values were invariably higher than those of absolute values (P < .01). The OR of absolute and % predicted OUEP reached 9.8 and 13.0, respectively. Combined with OB, the % predicted OR reached 56.3. All ORs of absolute and % predicted OUEP with and without OB were significantly (P < .05) superior to any other CPET parameter (Table 2 and Fig 3 of this article and the previously published Fig 2A3) (P < .01), while other cardiovascular measurements were not significant for early death.3 The morbidity OR analyses for OUEP and OUES were statistically significant, but less than for mortality (Fig 3B). Adding OB+ increased morbidity ORs minimally.

Figure Jump LinkFigure 3. ORs of OUEP and OUES, alone and combined with OB+, for mortality and morbidity of patients with HF. A, Mortality. The left column shows the optimal absolute and % predicted cutoff values for each variable. The center column (gray bars) shows OR mean and 95% CI for single lower OUE measurements. The right column (black bars) shows OR for combined lower OUE and OB+ measurements. The stepwise selected optimal predictor model uses only two-predictor variables of decreased OUEP (<65% predicted ) with OB+ by stepwise multivariate regression analysis (P < .001) since all other CPET and non-CPET variables are then nonsignificant (P > .05). OR values were invariably higher using % predicted values and were always higher using OUEP than OUES. The best single predictor of early mortality is OUEP < 65% predicted (OR = 13.0) and the best combination is OUEP < 65% predicted with OB+ (OR = 56.3). All OR values of absolute and % predicted OUEP with and without OB were superior to all other key CPET parameters,3 respectively (P < .05). As shown in Figure 2A of our previous published data,3 the highest single OR values of lowest V˙ e/V˙ co2, V˙ e-vs-V˙ co2 slope, peak V˙ o2, peak O2 pulse, and AT, respectively, were 9.4, 7.6, 6.0, 5.8, and 5.1 for % predicted values, and 5.8, 5.7, 4.5, 4.6 and 5.1 for absolute values. Combined with OB+, they were 38.9, 19.7, 19.5, 16.2, and 18.1 for % predicted values, and were 22.6, 15.1, 12.7, 15.4, and 14.1 for absolute values, respectively. B, Morbidity. Data are displayed as in panel A. Although the OR values are lower, they parallel the mortality findings. None of the OR morbidity values for other CPET measures exceeded those of OUEP. OB+ = positive oscillatory breathing; pred = predicted. See Figure 1 and 2 legends for expansion of other abbreviations.Grahic Jump Location

Using Kaplan-Meier and stepwise multivariable regression analysis (reducing all other CPET and non-CPET measures except only OB to nonsignificance), a low OUEP (< 65% predicted ) was again the best single predictor of early mortality (Fig 4). The optimal predictor model was the combination of low % predicted (< 65% predicted ) OUEP with OB+. Six-month survival rates were 98.4% in patients with HF with OUEP ≥ 65% predicted, and 82.6% in those OUEP < 65% predicted. Survival rates for the four combinations of OUEP and OB ranged from 99.6% to 80.4% (P < .001).

Figure Jump LinkFigure 4. Kaplan-Meier survival curves of patients with HF using status of OUE, singly, and combined with OB. The numbers of patients at risk for each curve are given at 60-day intervals. A, Comparison of the OUEP ≥ 65% predicted (gray) vs OUEP < 65% predicted (black) (P < .001). B, Comparison from top to bottom: OUEP ≥ 65% predicted and OB−, OUEP ≥ 65% predicted and OB+, OUEP < 65% predicted and OB−, and OUEP < 65% predicted and OB+. The bottom curve values were significantly lower than the values of the upper two curves (P < .001), but not for values of the third curve (P > .05). OB− = negative oscillatory breathing. See Figure 1 and 3 legends for expansion of other abbreviations.Grahic Jump Location

Figure 5 identifies the early deaths and shows the relationships of individual OUEP and OUES values (x-axis) vs peak V˙ o2 values (y-axis), expressed in three ways, to compare patients with HF vs normal subjects. Using OUEP, the slopes and intercepts of the patients with HF differed markedly from those of the normal subjects. Using OUES, the slopes and intercepts of the patients with HF and normal subjects were almost identical.

Figure Jump LinkFigure 5. Relationships of OUEP and OUES vs peak V˙ o2 in HF. Each symbol represents an individual normal (solid gray circle) or very fit (open gray circle) healthy subject24 or survivor (open black inverted triangle) or deceased (solid black inverted triangle) patient with HF. The gray regression lines are for normal subjects24; the equations and black regression lines are for all patients with HF. A-C, OUEP vs peak V˙ o2. Linear regression line slopes and intercepts of patients with HF and normal subjects invariably differ. For HF, the OUEP varies over a wider range than peak V˙ o2; for normal subjects, peak V˙ o2 varies over a wider range than OUEP (P < .001). D-F, OUES vs peak V˙ o2. Regression lines and intercepts of normal subjects and HF are similar (P > .05). See Figure 1-3 legends for expansion of abbreviations.Grahic Jump Location

The life-sustaining function of the cardiovascular system is to transport O2 and CO2 between the lungs and the systemic organs. It is not surprising then that CPET measures such as peak V˙ o2, AT, peak oxygen pulse, lowest V˙ e/V˙ co2 and V˙ e-vs-V˙ co2 slope, and OUES are good measures of cardiovascular performance and predictive of survival in patients with HF.310 Another measurement, perhaps even more useful, is the OUEP. Surprisingly, to the best of our knowledge, the measurements of OUEP (or lowest V˙ e/V˙ o2 or highest V˙ o2/V˙ e), either for a fraction of a minute or longer durations, have not been carefully analyzed or reported for prognosis or severity of disease in the past.

Logically, by Fick’s principle,26,3638 the ratio of oxygen uptake to ventilation (ie, OUE = V˙ o2/V˙ e), should evaluate cardiovascular performance, as it is dependent on three factors: cardiac output, arterial-mixed venous oxygen content difference, and lung ventilation.26,3638 The OUEP tended to be higher in healthier individuals.24 For reasons that may not be intuitively obvious, in both normal subjects24 and patients with HF, the OUE was not highest at rest or at maximal exercise, but during moderate exercise (Fig 1). The highest OUE (OUEP) does not occur at rest because cardiac output is lowest; ventilation is less efficient (anatomic dead space volume is a higher proportion of tidal volumes); and arterial-mixed venous oxygen content difference is lowest because the oxygen saturation of mixed venous blood is higher both in normal subjects and patients at rest than during exercise (so that not as much O2 can be extracted from inhaled air at rest). In both normal subjects24 and patients with HF, the OUEP does not occur at heavy, very heavy, or maximal exercise, even though mixed venous oxyhemoglobin saturation is low and the ratio of anatomic dead space to alveolar ventilation is optimal. This is because, with more intense exercise, increasing production of lactate, metabolic CO2, excess CO2, and increasing H+ stimulate ventilation. Because ventilation does not normally limit exercise, this stimulation results in increasing CO2 output but less increasing O2 loading equally, lowering the alveolar and arterial CO2 partial pressure. This, in turn, further reduces ventilatory efficiency as more ventilation is required to remove CO2 to balance the declining bicarbonate levels and reduce acidemia. As exercise intensity increases, ventilatory drive to reduce acidemia decreases V˙ co2/V˙ e and V˙ o2/V˙ e (equivalent to increasing V˙ e/V˙ co2 and V˙ e/V˙ o2). The time of decreasing V˙ o2/V˙ e at AT (Fig 1) is earlier than that of decreasing V˙ co2/V˙ e at the ventilatory compensation point.26 Thus, the OUEP occurs during less than maximal exercise (below AT level) in both normal subjects24 and patients with HF. It assesses the optimal capacity to match the increasing lung perfusion of mixed venous partially deoxygenated blood with the increasing ventilation and to load O2 into the systemic circulation. In normal subjects and patients with HF without significant ventilatory limitation or primary lung disease, the OUEP appears to reflect overall cardiovascular function of loading and transporting O2 well. Initially, the OUEP was found to be dependent on age, sex, body size, and fitness in over 474 healthy subjects; reference equations from 417 normal subjects were presented.24 Now, it is evident that the OUEP also indicates disease severity in over 500 patients with HF (Fig 1, Table 3). In these patients, the % predicted OUEP, based on age, sex, and height, turns out to be the strongest single predictor of early survival (Figs 3A, 4). While the V˙ e-V˙ co2 relationship better reflects ventilatory efficiency,25,26 the OUEP appears to better reflect the efficiency of oxygen uptake, which depends on systemic and lung perfusion, just before ventilation is stimulated by exercise-induced acidemia.

In this group of patients with HF, it appears that values of OUEP have an advantage over OUES, peak V˙ o2, AT, V˙ e/V˙ co2, and V˙ e-vs-V˙ co2 values or any other invasive or noninvasive measurement in assessing early mortality and morbidity.3 This superiority is evident by AUC, OR, stepwise multivariable regression selection, and Kaplan-Meier analyses (Figs 2-4, Tables 2-4, and our prior report3). Multivariate analysis selected OUEP ahead of other CPET parameters, despite the overlapping of their physiologic information and relatively high correlations. Manually adding the other key CPET measures did not add significant prognostic importance to OUEP, which was superior to OUES. Alone, a decreased % predicted OUEP was superior to any other measurement (Fig 3A and our previous published Figure 2A3). For both OUEP and OUES, the % predicted values were superior to absolute values (Fig 3A), probably because % predicted lessens differences between sex, body size, and age. The fact that OUEP is measured during submaximal rather than maximal exercise is also advantageous in evaluating patients with HF.24,39,40

The fact that the intercepts of OUES vs peak V˙ o2 are so close to zero in both patients with HF and normal subjects (Fig 5) strongly suggests that OUES and peak V˙ o2 measure similar phenomenon, and thus may be somewhat redundant. In contrast, the dissimilarity of the slopes and intercepts of OUEP vs peak V˙ o2, suggests, that though they are related, they likely represent somewhat different phenomena.

OB is also a significant independent predictor,3,710,31,32 and in combination with OUE measurements, is even more powerful (P < .001). Adding OB increased the OR for both OUEP and OUES twofold to fourfold (Fig 3A). The OR of OUEP using 65% predicted as a cutoff with and without OB were 13 and 56, respectively, higher than those of OUES with or without OB. Although the number of deaths (n = 19) over the 6-month follow-up period were modest, the parallel changes in morbidity (n = 107) confirm the value of OUE measurements.

A limitation of this study is the small number of deaths in this population; however, the parallel findings in morbidity somewhat balance this limitation. The findings here, confined to patients with left ventricular dysfunction and wide QRS complexes, may or may not be found in larger series or in patients with other cardiac disorders.

In conclusion, this study finds that OUEP combined with OB strongly prognosticates an increased risk of early death and hospitalization in patients with HF. OUEP < 65% predicted is the best single predictor for early mortality with an OR of 13.0. Adding OB increases the OR to 56.3. CPET measurements of OUEP and OB, which are noninvasive, inexpensive, and do not require maximal exercise, should be useful in clinical trials and in evaluating and managing individual cardiac patients. Thus, it would be useful for other clinical investigators managing patients with cardiovascular disease, including those being evaluated for cardiac transplantation, to retrospectively or prospectively analyze their gas exchange data to ascertain whether the measurement of OUEP is useful in grading severity, prognosis, and/or guiding therapy for their patients.

Author contributions:Dr Sun: contributed to data collection, data analyses, and writing and revising the manuscript.

Dr Hansen: contributed to writing and revising the manuscript.

Dr Stringer: contributed to writing and revising the manuscript.

Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Role of sponsors: The sponsor had no role in the design of the study, the collection and analysis of the data, or in the preparation of the manuscript.

Other contributions: The clinical trials of RHYTHM (Resynchronization for Hemodynamic Treatment for Heart Failure Management) and RethinQ (Cardiac Resynchronization Therapy in Patients With Heart Failure and Narrow QRS Complexes)22,23 were funded by St. Jude Medical. We also gratefully acknowledge Tamara Shipman and the study groups’ staff at St. Jude Medical for their help in providing us with individual patient information. However, this manuscript has no financial support from St. Jude Medical.

AT

anaerobic threshold

AUC

area under the curve

BVP/CRT

biventricular pacing/cardioversion therapy

CPET

incremental cardiopulmonary exercise testing

HF

chronic left heart failure

O2

oxygen

OB

oscillatory breathing

OUE

oxygen uptake efficiency

OUE@AT

oxygen uptake efficiency at the anaerobic threshold

OUEP

oxygen uptake efficiency plateau

OUES

oxygen uptake efficiency slope

ReThinQ

Resynchronization Therapy in Patients With Heart Failure and Narrow QRS Complexes

RHYTHM

Resynchronization for Hemodynamic Treatment for Heart Failure Management

ROC

receiver operator characteristic curve

STPD

standard conditions of 0ºC, 760 mm Hg, dry

V˙ co2

minute CO2 output

V˙ e

minute ventilation

V˙ o2

minute O2 uptake

Hunt SA. American College of Cardiology American College of Cardiology American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure) American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure) ACC/AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure). J Am Coll Cardiol. 2005;466:e1-e82. [CrossRef] [PubMed]
 
Stewart S, MacIntyre K, Capewell S, McMurray JJ. Heart failure and the aging population: an increasing burden in the 21st century? Heart. 2003;891:49-53. [CrossRef] [PubMed]
 
Sun XG, Hansen JE, Beshai JF, Wasserman K. Oscillatory breathing and exercise gas exchange abnormalities prognosticate early mortality and morbidity in heart failure. J Am Coll Cardiol. 2010;5517:1814-1823. [CrossRef] [PubMed]
 
Gitt AK, Wasserman K, Kilkowski C, et al. Exercise anaerobic threshold and ventilatory efficiency identify heart failure patients for high risk of early death. Circulation. 2002;10624:3079-3084. [CrossRef] [PubMed]
 
Myers J, Arena R, Oliveira RB, et al. The lowest VE/VCO2 ratio during exercise as a predictor of outcomes in patients with heart failure. J Card Fail. 2009;159:756-762. [CrossRef] [PubMed]
 
Guazzi M, Reina G, Tumminello G, Guazzi MD. Exercise ventilation inefficiency and cardiovascular mortality in heart failure: the critical independent prognostic value of the arterial CO2 partial pressure. Eur Heart J. 2005;265:472-480. [CrossRef] [PubMed]
 
Brack T, Thüer I, Clarenbach CF, et al. Daytime Cheyne-Stokes respiration in ambulatory patients with severe congestive heart failure is associated with increased mortality. Chest. 2007;1325:1463-1471. [CrossRef] [PubMed]
 
Guazzi M, Raimondo R, Vicenzi M, et al. Exercise oscillatory ventilation may predict sudden cardiac death in heart failure patients. J Am Coll Cardiol. 2007;504:299-308. [CrossRef] [PubMed]
 
Corrà U, Pistono M, Mezzani A, et al. Sleep and exertional periodic breathing in chronic heart failure: prognostic importance and interdependence. Circulation. 2006;1131:44-50. [CrossRef] [PubMed]
 
Corrà U, Giordano A, Bosimini E, et al. Oscillatory ventilation during exercise in patients with chronic heart failure: clinical correlates and prognostic implications. Chest. 2002;1215:1572-1580. [CrossRef] [PubMed]
 
Baba R, Nagashima M, Goto M, et al. Oxygen uptake efficiency slope: a new index of cardiorespiratory functional reserve derived from the relation between oxygen uptake and minute ventilation during incremental exercise. J Am Coll Cardiol. 1996;286:1567-1572. [CrossRef] [PubMed]
 
Baba R, Nagashima M, Nagano Y, Ikoma M, Nishibata K. Role of the oxygen uptake efficiency slope in evaluating exercise tolerance. Arch Dis Child. 1999;811:73-75. [CrossRef] [PubMed]
 
Baba R, Tsuyuki K, Kimura Y, et al. Oxygen uptake efficiency slope as a useful measure of cardiorespiratory functional reserve in adult cardiac patients. Eur J Appl Physiol Occup Physiol. 1999;805:397-401. [CrossRef] [PubMed]
 
Van Laethem C, Bartunek J, Goethals M, Nellens P, Andries E, Vanderheyden M. Oxygen uptake efficiency slope, a new submaximal parameter in evaluating exercise capacity in chronic heart failure patients. Am Heart J. 2005;1491:175-180. [CrossRef] [PubMed]
 
Arena R, Brubaker P, Moore B, Kitzman D. The oxygen uptake efficiency slope is reduced in older patients with heart failure and a normal ejection fraction. Int J Cardiol. 2010;1441:101-102. [CrossRef] [PubMed]
 
Myers J, Arena R, Dewey F, et al. A cardiopulmonary exercise testing score for predicting outcomes in patients with heart failure. Am Heart J. 2008;1566:1177-1183. [CrossRef] [PubMed]
 
Arena R, Arrowood JA, Fei DY, Helm S, Kraft KA. Maximal aerobic capacity and the oxygen uptake efficiency slope as predictors of large artery stiffness in apparently healthy subjects. J Cardiopulm Rehabil Prev. 2009;294:248-254. [PubMed]
 
Defoor J, Schepers D, Reybrouck T, Fagard R, Vanhees L. Oxygen uptake efficiency slope in coronary artery disease: clinical use and response to training. Int J Sports Med. 2006;279:730-737. [CrossRef] [PubMed]
 
Van Laethem C, Van De Veire N, De Backer G, et al. Response of the oxygen uptake efficiency slope to exercise training in patients with chronic heart failure. Eur J Heart Fail. 2007;96-7:625-629. [CrossRef] [PubMed]
 
Gademan MG, Swenne CA, Verwey HF, et al. Exercise training increases oxygen uptake efficiency slope in chronic heart failure. Eur J Cardiovasc Prev Rehabil. 2008;152:140-144. [CrossRef] [PubMed]
 
Mourot L, Perrey S, Tordi N, Rouillon JD. Evaluation of fitness level by the oxygen uptake efficiency slope after a short-term intermittent endurance training. Int J Sports Med. 2004;252:85-91. [CrossRef] [PubMed]
 
Wasserman K, Sun XG, Hansen JE. Effect of biventricular pacing on the exercise pathophysiology of heart failure. Chest. 2007;1321:250-261. [CrossRef] [PubMed]
 
Beshai JF, Grimm RA, Nagueh SF, et al; RethinQ Study Investigators RethinQ Study Investigators Cardiac-resynchronization therapy in heart failure with narrow QRS complexes. N Engl J Med. 2007;35724:2461-2471. [CrossRef] [PubMed]
 
Sun XG, Hansen EJ, Stringer WW. Oxygen uptake efficiency plateau: physiology and reference values. Eur J Appl Physiol. 2012;1123:919-928. [CrossRef] [PubMed]
 
Wasserman K, Hansen JE, Sue DY, Stringer WW, Whipp BJ. Principles of Exercise Testing and Interpretation. 2005;4th ed Philadephia, PA Lippincott Williams &Wilkins
 
Sun XG, Hansen JE, Garatachea N, Storer TW, Wasserman K. Ventilatory efficiency during exercise in healthy subjects. Am J Respir Crit Care Med. 2002;16611:1443-1448. [CrossRef] [PubMed]
 
Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol. 1986;606:2020-2027. [PubMed]
 
Hansen JE, Sue DY, Wasserman K. Predicted values for clinical exercise testing. Am Rev Respir Dis. 1984;1292 pt 2:S49-S55. [PubMed]
 
Sun XG, Hansen JE, Oudiz RJ, Wasserman K. Exercise pathophysiology in patients with primary pulmonary hypertension. Circulation. 2001;1044:429-435. [CrossRef] [PubMed]
 
Sun XG, Hansen JE, Oudiz RJ, Wasserman K. Gas exchange detection of exercise-induced right-to-left shunt in patients with primary pulmonary hypertension. Circulation. 2002;1051:54-60. [CrossRef] [PubMed]
 
Ben-Dov I, Sietsema KE, Casaburi R, Wasserman K. Evidence that circulatory oscillations accompany ventilatory oscillations during exercise in patients with heart failure. Am Rev Respir Dis. 1992;1454 pt 1:776-781. [PubMed]
 
Ribeiro JP, Knutzen A, Rocco MB, Hartley LH, Colucci WS. Periodic breathing during exercise in severe heart failure. Reversal with milrinone or cardiac transplantation. Chest. 1987;923:555-556. [CrossRef] [PubMed]
 
Nguyen TT, Adair LS, He K, Popkin BM. Optimal cutoff values for overweight: using body mass index to predict incidence of hypertension in 18- to 65-year-old Chinese adults. J Nutr. 2008;1387:1377-1382. [PubMed]
 
Warner RM. Applied Statistics: From Bivariate Through Multivariate Techniques. 2008; Thousand Oaks, CA Sage Publications, Inc
 
Glantz AS, Slinker BK. Primer of Applied Regression and Analysis of Variance. 1990; New York, NY McGraw-Hill
 
Wasserman K. Cardiopulmonary Exercise Testing and Cardiovascular Health. 2002; Armonk, NY Futura
 
Sun XG, Hansen JE, Ting H, et al. Comparison of exercise cardiac output by the Fick principle using oxygen and carbon dioxide. Chest. 2000;1183:631-640. [CrossRef] [PubMed]
 
Stringer WW, Hansen JE, Wasserman K. Cardiac output estimated noninvasively from oxygen uptake during exercise. J Appl Physiol. 1997;823:908-912. [PubMed]
 
Kasikcioglu E. Which is the best parameter of submaximal cardiopulmonary exercise testing? Eur Heart J. 2006;2720:2483. [CrossRef] [PubMed]
 
Agostoni P. Cardiopulmonary exercise testing for heart failure patients: a hodgepodge of techniques, parameters and interpretations. In other words, the need for a time-break. Eur Heart J. 2006;276:633-634. [CrossRef] [PubMed]
 

Figures

Figure Jump LinkFigure 1. OUE patterns during cardiopulmonary exercise testing (CPET) in a normal subject and three patients with HF. The response patterns of OUE (ie, V˙ o2/V˙ e) from rest to peak exercise during CPET in the normal subject (gray solid circle)24 and patients with HF (black, dark gray, and gray inverted triangle) and are reasonably parallel. The OUE varies minimally during rest, increases during warmup, and then increases further to its highest level or plateau (90-s average oxygen uptake efficiency plateau, mL/L [OUEP]) just before the AT. Thereafter, the OUE invariably decreases as exercise intensity increases. As the severity of HF increases, the OUEP decreases. An oscillatory breathing (OB) pattern was found in many patients with severe and very severe HF. AT = anaerobic threshold; HF = chronic left heart failure; O2 = oxygen; OUE = oxygen uptake efficiency; V˙ o2/V˙ e = minute oxygen uptake/minute ventilation.Grahic Jump Location
Figure Jump LinkFigure 2. OUEP, OUE@AT, and OUES in patients with HF and healthy subjects. Each symbol is an individual patient with HF (open black inverted triangle), normal subject (solid gray circle), or very fit (open gray circle) subject. The regression lines are for normal subjects only (gray) or all patients with HF (black). The equations are for patients with HF only because those for normal subject data were previously presented.24 A, OUEP vs OUE@AT indicates their similarity for normal subjects or patients with HF. B, OUEP vs OUES. The wider distribution of data and differing regression lines indicate differences between the two groups. OUE@AT = 60-s average oxygen uptake efficiency at the anaerobic threshold (mL/L); OUES = oxygen uptake efficiency slope, slope of V˙ o2 vs −logV˙ e [L/min/log (L/min)]. See Figure 1 legend for expansion of other abbreviations.Grahic Jump Location
Figure Jump LinkFigure 3. ORs of OUEP and OUES, alone and combined with OB+, for mortality and morbidity of patients with HF. A, Mortality. The left column shows the optimal absolute and % predicted cutoff values for each variable. The center column (gray bars) shows OR mean and 95% CI for single lower OUE measurements. The right column (black bars) shows OR for combined lower OUE and OB+ measurements. The stepwise selected optimal predictor model uses only two-predictor variables of decreased OUEP (<65% predicted ) with OB+ by stepwise multivariate regression analysis (P < .001) since all other CPET and non-CPET variables are then nonsignificant (P > .05). OR values were invariably higher using % predicted values and were always higher using OUEP than OUES. The best single predictor of early mortality is OUEP < 65% predicted (OR = 13.0) and the best combination is OUEP < 65% predicted with OB+ (OR = 56.3). All OR values of absolute and % predicted OUEP with and without OB were superior to all other key CPET parameters,3 respectively (P < .05). As shown in Figure 2A of our previous published data,3 the highest single OR values of lowest V˙ e/V˙ co2, V˙ e-vs-V˙ co2 slope, peak V˙ o2, peak O2 pulse, and AT, respectively, were 9.4, 7.6, 6.0, 5.8, and 5.1 for % predicted values, and 5.8, 5.7, 4.5, 4.6 and 5.1 for absolute values. Combined with OB+, they were 38.9, 19.7, 19.5, 16.2, and 18.1 for % predicted values, and were 22.6, 15.1, 12.7, 15.4, and 14.1 for absolute values, respectively. B, Morbidity. Data are displayed as in panel A. Although the OR values are lower, they parallel the mortality findings. None of the OR morbidity values for other CPET measures exceeded those of OUEP. OB+ = positive oscillatory breathing; pred = predicted. See Figure 1 and 2 legends for expansion of other abbreviations.Grahic Jump Location
Figure Jump LinkFigure 4. Kaplan-Meier survival curves of patients with HF using status of OUE, singly, and combined with OB. The numbers of patients at risk for each curve are given at 60-day intervals. A, Comparison of the OUEP ≥ 65% predicted (gray) vs OUEP < 65% predicted (black) (P < .001). B, Comparison from top to bottom: OUEP ≥ 65% predicted and OB−, OUEP ≥ 65% predicted and OB+, OUEP < 65% predicted and OB−, and OUEP < 65% predicted and OB+. The bottom curve values were significantly lower than the values of the upper two curves (P < .001), but not for values of the third curve (P > .05). OB− = negative oscillatory breathing. See Figure 1 and 3 legends for expansion of other abbreviations.Grahic Jump Location
Figure Jump LinkFigure 5. Relationships of OUEP and OUES vs peak V˙ o2 in HF. Each symbol represents an individual normal (solid gray circle) or very fit (open gray circle) healthy subject24 or survivor (open black inverted triangle) or deceased (solid black inverted triangle) patient with HF. The gray regression lines are for normal subjects24; the equations and black regression lines are for all patients with HF. A-C, OUEP vs peak V˙ o2. Linear regression line slopes and intercepts of patients with HF and normal subjects invariably differ. For HF, the OUEP varies over a wider range than peak V˙ o2; for normal subjects, peak V˙ o2 varies over a wider range than OUEP (P < .001). D-F, OUES vs peak V˙ o2. Regression lines and intercepts of normal subjects and HF are similar (P > .05). See Figure 1-3 legends for expansion of abbreviations.Grahic Jump Location

Tables

Table Graphic Jump Location
Table 1 —Demographics and Selected Resting and Exercise Measurements in Patients With HF

Data are mean ± SD unless otherwise specified. AT = oxygen uptake rate at anaerobic threshold; BVP/CRT-ON or -OFF = biventricular pacemaker/cardiac resynchronization setting is “on” or “off”; CPET = cardiopulmonary exercise testing; DBP = diastolic BP; HF = chronic left heart failure; HR = heart rate; LVEF = left ventricular ejection fraction; NYHA = New York Heart Association classification; O2 pulse = oxygen uptake per min/heart rate; OB = oscillatory breathing; RER = respiratory gas exchange ratio (ie, CO2 output per min [standard conditions of 0ºC, 760 mm Hg, dry]/oxygen update per min [standard conditions of 0ºC, 760 mm Hg, dry]); SBP = systolic BP; V˙ co2 = CO2 output per min (standard conditions of 0ºC, 760 mm Hg, dry); V˙ e = minute ventilation (body temperature, ambient atmospheric pressure, saturated with water vapor); V˙ e/V˙ co2 = ratio of minute ventilation (body temperature, ambient atmospheric pressure, saturated with water vapor) to CO2 output per min (standard conditions of 0ºC, 760 mm Hg, dry); V˙ e-vs-V˙ co2 slope = minute ventilation (body temperature, ambient atmospheric pressure, saturated with water vapor) as a function of CO2 output per min (standard conditions of 0ºC, 760 mm Hg, dry); V˙ o2 = oxygen uptake per min (standard conditions of 0ºC, 760 mm Hg, dry).

a 

Quality of life was evaluated with the Minnesota Living with Heart Failure Questionnaire, with scores ranging from 0 to 105 and with higher scores indicating a poorer quality of life.23

Table Graphic Jump Location
Table 2 —OR, AUC, With 95% CI, and Best Cutoff Values for Mortality in HF

AUC = area under the receiver operating characteristics curve. See Table 1 legend for expansion of other abbreviations.

Table Graphic Jump Location
Table 3 —OUE Values Linked to Mortality, Morbidity, and Severity of HF

Values are mean ± SD (range). The P values between the positive and negative for mortality and morbidity are by unpaired t test. − = negative; + = positive; ANOVA = analysis of variance; OUE = oxygen uptake efficiency (ie, V˙ o2/V˙ e, mL/L); OUEP = 90-s average oxygen uptake efficiency plateau (mL/L); OUES = slope of V˙ o2 to logV˙ e [L/min/log(L/min)]. See Table 1 legend for expansion of other abbreviations. Comparisons of functional severity by ANOVA are

vs “~Normal” group:

a 

P < .05.

b 

P < .01.

c 

P < .001.

d 

P < .0001.

vs “Mild” group:

e 

P < .05.

f 

P < .01.

g 

P < .001.

h 

P < .0001.

vs “Moderate” group:

i 

P < .01.

j 

P < .001.

k 

P < .0001.

vs “Severe” group:

l 

P < .0001.

Table Graphic Jump Location
Table 4 —Best Cutoff Values, Sensitivity, Specificity, AUC, and 95% CI of OUE for Mortality in HF

See Table 1-3 legends for expansion of abbreviations.

References

Hunt SA. American College of Cardiology American College of Cardiology American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure) American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure) ACC/AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure). J Am Coll Cardiol. 2005;466:e1-e82. [CrossRef] [PubMed]
 
Stewart S, MacIntyre K, Capewell S, McMurray JJ. Heart failure and the aging population: an increasing burden in the 21st century? Heart. 2003;891:49-53. [CrossRef] [PubMed]
 
Sun XG, Hansen JE, Beshai JF, Wasserman K. Oscillatory breathing and exercise gas exchange abnormalities prognosticate early mortality and morbidity in heart failure. J Am Coll Cardiol. 2010;5517:1814-1823. [CrossRef] [PubMed]
 
Gitt AK, Wasserman K, Kilkowski C, et al. Exercise anaerobic threshold and ventilatory efficiency identify heart failure patients for high risk of early death. Circulation. 2002;10624:3079-3084. [CrossRef] [PubMed]
 
Myers J, Arena R, Oliveira RB, et al. The lowest VE/VCO2 ratio during exercise as a predictor of outcomes in patients with heart failure. J Card Fail. 2009;159:756-762. [CrossRef] [PubMed]
 
Guazzi M, Reina G, Tumminello G, Guazzi MD. Exercise ventilation inefficiency and cardiovascular mortality in heart failure: the critical independent prognostic value of the arterial CO2 partial pressure. Eur Heart J. 2005;265:472-480. [CrossRef] [PubMed]
 
Brack T, Thüer I, Clarenbach CF, et al. Daytime Cheyne-Stokes respiration in ambulatory patients with severe congestive heart failure is associated with increased mortality. Chest. 2007;1325:1463-1471. [CrossRef] [PubMed]
 
Guazzi M, Raimondo R, Vicenzi M, et al. Exercise oscillatory ventilation may predict sudden cardiac death in heart failure patients. J Am Coll Cardiol. 2007;504:299-308. [CrossRef] [PubMed]
 
Corrà U, Pistono M, Mezzani A, et al. Sleep and exertional periodic breathing in chronic heart failure: prognostic importance and interdependence. Circulation. 2006;1131:44-50. [CrossRef] [PubMed]
 
Corrà U, Giordano A, Bosimini E, et al. Oscillatory ventilation during exercise in patients with chronic heart failure: clinical correlates and prognostic implications. Chest. 2002;1215:1572-1580. [CrossRef] [PubMed]
 
Baba R, Nagashima M, Goto M, et al. Oxygen uptake efficiency slope: a new index of cardiorespiratory functional reserve derived from the relation between oxygen uptake and minute ventilation during incremental exercise. J Am Coll Cardiol. 1996;286:1567-1572. [CrossRef] [PubMed]
 
Baba R, Nagashima M, Nagano Y, Ikoma M, Nishibata K. Role of the oxygen uptake efficiency slope in evaluating exercise tolerance. Arch Dis Child. 1999;811:73-75. [CrossRef] [PubMed]
 
Baba R, Tsuyuki K, Kimura Y, et al. Oxygen uptake efficiency slope as a useful measure of cardiorespiratory functional reserve in adult cardiac patients. Eur J Appl Physiol Occup Physiol. 1999;805:397-401. [CrossRef] [PubMed]
 
Van Laethem C, Bartunek J, Goethals M, Nellens P, Andries E, Vanderheyden M. Oxygen uptake efficiency slope, a new submaximal parameter in evaluating exercise capacity in chronic heart failure patients. Am Heart J. 2005;1491:175-180. [CrossRef] [PubMed]
 
Arena R, Brubaker P, Moore B, Kitzman D. The oxygen uptake efficiency slope is reduced in older patients with heart failure and a normal ejection fraction. Int J Cardiol. 2010;1441:101-102. [CrossRef] [PubMed]
 
Myers J, Arena R, Dewey F, et al. A cardiopulmonary exercise testing score for predicting outcomes in patients with heart failure. Am Heart J. 2008;1566:1177-1183. [CrossRef] [PubMed]
 
Arena R, Arrowood JA, Fei DY, Helm S, Kraft KA. Maximal aerobic capacity and the oxygen uptake efficiency slope as predictors of large artery stiffness in apparently healthy subjects. J Cardiopulm Rehabil Prev. 2009;294:248-254. [PubMed]
 
Defoor J, Schepers D, Reybrouck T, Fagard R, Vanhees L. Oxygen uptake efficiency slope in coronary artery disease: clinical use and response to training. Int J Sports Med. 2006;279:730-737. [CrossRef] [PubMed]
 
Van Laethem C, Van De Veire N, De Backer G, et al. Response of the oxygen uptake efficiency slope to exercise training in patients with chronic heart failure. Eur J Heart Fail. 2007;96-7:625-629. [CrossRef] [PubMed]
 
Gademan MG, Swenne CA, Verwey HF, et al. Exercise training increases oxygen uptake efficiency slope in chronic heart failure. Eur J Cardiovasc Prev Rehabil. 2008;152:140-144. [CrossRef] [PubMed]
 
Mourot L, Perrey S, Tordi N, Rouillon JD. Evaluation of fitness level by the oxygen uptake efficiency slope after a short-term intermittent endurance training. Int J Sports Med. 2004;252:85-91. [CrossRef] [PubMed]
 
Wasserman K, Sun XG, Hansen JE. Effect of biventricular pacing on the exercise pathophysiology of heart failure. Chest. 2007;1321:250-261. [CrossRef] [PubMed]
 
Beshai JF, Grimm RA, Nagueh SF, et al; RethinQ Study Investigators RethinQ Study Investigators Cardiac-resynchronization therapy in heart failure with narrow QRS complexes. N Engl J Med. 2007;35724:2461-2471. [CrossRef] [PubMed]
 
Sun XG, Hansen EJ, Stringer WW. Oxygen uptake efficiency plateau: physiology and reference values. Eur J Appl Physiol. 2012;1123:919-928. [CrossRef] [PubMed]
 
Wasserman K, Hansen JE, Sue DY, Stringer WW, Whipp BJ. Principles of Exercise Testing and Interpretation. 2005;4th ed Philadephia, PA Lippincott Williams &Wilkins
 
Sun XG, Hansen JE, Garatachea N, Storer TW, Wasserman K. Ventilatory efficiency during exercise in healthy subjects. Am J Respir Crit Care Med. 2002;16611:1443-1448. [CrossRef] [PubMed]
 
Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol. 1986;606:2020-2027. [PubMed]
 
Hansen JE, Sue DY, Wasserman K. Predicted values for clinical exercise testing. Am Rev Respir Dis. 1984;1292 pt 2:S49-S55. [PubMed]
 
Sun XG, Hansen JE, Oudiz RJ, Wasserman K. Exercise pathophysiology in patients with primary pulmonary hypertension. Circulation. 2001;1044:429-435. [CrossRef] [PubMed]
 
Sun XG, Hansen JE, Oudiz RJ, Wasserman K. Gas exchange detection of exercise-induced right-to-left shunt in patients with primary pulmonary hypertension. Circulation. 2002;1051:54-60. [CrossRef] [PubMed]
 
Ben-Dov I, Sietsema KE, Casaburi R, Wasserman K. Evidence that circulatory oscillations accompany ventilatory oscillations during exercise in patients with heart failure. Am Rev Respir Dis. 1992;1454 pt 1:776-781. [PubMed]
 
Ribeiro JP, Knutzen A, Rocco MB, Hartley LH, Colucci WS. Periodic breathing during exercise in severe heart failure. Reversal with milrinone or cardiac transplantation. Chest. 1987;923:555-556. [CrossRef] [PubMed]
 
Nguyen TT, Adair LS, He K, Popkin BM. Optimal cutoff values for overweight: using body mass index to predict incidence of hypertension in 18- to 65-year-old Chinese adults. J Nutr. 2008;1387:1377-1382. [PubMed]
 
Warner RM. Applied Statistics: From Bivariate Through Multivariate Techniques. 2008; Thousand Oaks, CA Sage Publications, Inc
 
Glantz AS, Slinker BK. Primer of Applied Regression and Analysis of Variance. 1990; New York, NY McGraw-Hill
 
Wasserman K. Cardiopulmonary Exercise Testing and Cardiovascular Health. 2002; Armonk, NY Futura
 
Sun XG, Hansen JE, Ting H, et al. Comparison of exercise cardiac output by the Fick principle using oxygen and carbon dioxide. Chest. 2000;1183:631-640. [CrossRef] [PubMed]
 
Stringer WW, Hansen JE, Wasserman K. Cardiac output estimated noninvasively from oxygen uptake during exercise. J Appl Physiol. 1997;823:908-912. [PubMed]
 
Kasikcioglu E. Which is the best parameter of submaximal cardiopulmonary exercise testing? Eur Heart J. 2006;2720:2483. [CrossRef] [PubMed]
 
Agostoni P. Cardiopulmonary exercise testing for heart failure patients: a hodgepodge of techniques, parameters and interpretations. In other words, the need for a time-break. Eur Heart J. 2006;276:633-634. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543