Extended evaluation of the efficacy of a proactive forced oscillation technique-based auto-CPAP algorithm.

PURPOSE: Previous studies revealed substantially varying therapy efficacy of automatic continuous positive airway pressure (APAP) devices in the treatment of obstructive sleep apnea (OSA). We evaluated the efficacy of a new APAP device using the forced oscillation technique (FOT) to evaluate upper airway obstruction during apneas and flow contour analyses during hypopneas. METHODS: Forty-six initially diagnosed OSA patients were included and the pressure range was set from 5 to 20 hPa. Therapy efficacy was assessed based on the reduction of apnea-hypopnea index (AHI), improvement of objective sleep quality parameters, and the appropriateness of the device's pressure regulation. RESULTS: AHI and arousal index significantly decreased during APAP therapy (median [interquartile range]: AHI 36 [23-55] vs. 2 [1-6]/h, arousal index 30 [22-45] vs. 15 [10-19]/h, both p < 0.001). The amount of slow wave sleep (SWS) and rapid-eye-movement (REM) sleep significantly increased (SWS 20 [14-29] vs. 29 [19-34]%, REM 16 [11-21] vs. 24 [14-30]%, both p < 0.01). Most residual respiratory events during therapy were of central etiology and attributable to five patients, who presented with treatment-emergent central sleep apnea. The device's pressure regulation abolished most obstructive respiratory events (n = 6.7 residual obstructive events per patient). Of central respiratory events, 534/646 (83%) did not lead to pressure increases. CONCLUSION: This pilot study provides a proof of concept that the APAP device combining FOT and evaluation of flow contour allows for the suppression of obstructive events without relevant false reactions.

Feasibility of noninvasive single-channel automated differentiation of obstructive and central hypopneas with nasal airflow.

BACKGROUND: The identification of obstructive and central hypopneas is considered challenging in clinical practice. Presently, obstructive and central hypopneas are usually not differentiated or scores lack reliability due to the technical limitations of standard polysomnography. Esophageal pressure measurement is the gold-standard for identifying these events but its invasiveness deters its usage in daily practice. OBJECTIVES: To determine the feasibility and efficacy of an automatic noninvasive analysis method for the differentiation of obstructive and central hypopneas based solely on a single-channel nasal airflow signal. The obtained results are compared with gold-standard esophageal pressure scores. METHODS: A total of 41 patients underwent full night polysomnography with systematic esophageal pressure recording. Two experts in sleep medicine independently differentiated hypopneas with the gold-standard esophageal pressure signal. Features were automatically extracted from the nasal airflow signal of each annotated hypopnea to train and test the automatic analysis method. Interscorer agreement between automatic and visual scorers was measured with Cohen's kappa statistic (ĸ). RESULTS: A total of 1,237 hypopneas were visually differentiated. The automatic analysis achieved an interscorer agreement of ĸ = 0.37 and an accuracy of 69% for scorer A, ĸ = 0.40 and 70% for scorer B and ĸ = 0.41 and 71% for the agreed scores of scorers A and B. CONCLUSIONS: The promising results obtained in this pilot study demonstrate the feasibility of noninvasive single-channel hypopnea differentiation. Further development of this method may help improving initial diagnosis with home screening devices and offering a means of therapy selection and/or control.

Comparison of upper airway respiratory resistance measurements with the esophageal pressure/airflow relationship during sleep.

Measurement of upper airway resistance is of interest in sleep disordered breathing to estimate upper airway patency. Resistance is calculated with the airflow and respiratory effort signals. However, there is no consensus on a standard for upper airway resistance measurement. This study proposes a new benchmarking method to objectively compare different upper airway resistance measurement methods by objectively differentiating between breaths with inspiratory flow limitation (high resistance) and non-limited breaths (low resistance). Resistance was measured at peak-Pes, at peak-flow, at the linear portion of a polynomial equation, as an area comparative and as average resistance for an inspiration. A total of 20 patients with systematic, gold-standard esophageal pressure and nasal airflow acquisition were analyzed and 109,955 breaths were automatically extracted and evaluated. Relative resistance values in relationship to a reference resistance value obtained during wakefulness were also analyzed. The peak-Pes measurement method obtained the highest separation index with significant (p < 0.001) differences to the other methods, followed by the area comparative and the peak-flow methods. As expected, average resistances were significantly (p < 0.001) lower for the non-IFL than for the IFL group. Hence, we recommend employing the peak-Pes for accurate upper airway resistance estimation.

Automatic non-invasive differentiation of obstructive and central hypopneas with nasal airflow compared to esophageal pressure.

The differentiation of obstructive and central respiratory events is a major challenge in the diagnosis of sleep disordered breathing. Esophageal pressure (Pes) measurement is the gold-standard method to identify these events but its invasiveness deters its usage in clinical routine. Flattening patterns appear in the airflow signal during episodes of inspiratory flow limitation (IFL) and have been shown with invasive techniques to be useful to differentiate between central and obstructive hypopneas. In this study we present a new method for the automatic non-invasive differentiation of obstructive and central hypopneas solely with nasal airflow. An overall of 36 patients underwent full night polysomnography with systematic Pes recording and a total of 1069 hypopneas were manually scored by human experts to create a gold-standard annotation set. Features were automatically extracted from the nasal airflow signal to train and test our automatic classifier (Discriminant Analysis). Flattening patterns were non-invasively assessed in the airflow signal using spectral and time analysis. The automatic non-invasive classifier obtained a sensitivity of 0.71 and an accuracy of 0.69, similar to the results obtained with a manual non-invasive classification algorithm. Hence, flattening airflow patterns seem promising for the non-invasive differentiation of obstructive and central hypopneas.

Automatic differentiation of obstructive and central hypopneas with esophageal pressure measurement during sleep.

The differentiation between obstructive and central respiratory events is one of the most recurrent tasks in the diagnosis of sleep disordered breathing. Esophageal pressure measurement is the gold-standard method to assess respiratory effort and identify these events. But as its invasiveness discourages its use in clinical routine, non-invasisve systems have been proposed for differentiation. However, their adoption has been slow due to their limited clinical validation, as the creation of manual, gold-standard validation sets by human experts is a cumbersome procedure. In this study, a new system is proposed for an objective automatic, gold-standard differentiation between obstructive and central hypopneas with the esophageal pressure signal. First, an overall of 356 hypopneas of 16 patients were manually scored by a human expert to create a gold-standard validation set. Then, features were extracted from each hypopnea to train and test classifiers (Discriminant Analysis, Support Vector Machines and adaboost classifiers) to differentiate between central and obstructive hypopneas with the gold-standard esophageal pressure signal. The automatic differentiation system achieved promising results, with a sensitivity of 0.88, a specificity of 0.93 and an accuracy of 0.90. Hence, this system seems promising for an automatic, gold-standard differentiation between obstructive and central hypopneas.

Automatic classification of inspiratory flow limitation assessed non-invasively during sleep.

Detection of inspiratory flow limitation (IFL) is being recognized of increasing importance in order to diagnose pathologies related to sleep disordered breathing. Currently, IFL is usually identified with the help of invasive esophageal pressure measurement, still considered the gold-standard reference to assess respiratory effort. But the invasiveness of esophageal pressure measurement and its impact on sleep discourages its use in clinical routine. In this study, a new noninvasive automatic system is proposed for objective IFL classification. First, an automatic annotation system for IFL based on pressure/flow relationship was developed. Then, classifiers (Support Vector Machines and adaboost classifiers) were trained with these gold-standard references in order to objectively classify breaths non-invasively, solely based on the breaths' flow contours. The new non-invasive automatic classification system seems to be promising, as it achieved a sensitivity of 0.92 and a specificity of 0.89, outperforming prior classification results obtained by human experts.

Characterization of inspiratory flow limitation during sleep with an exponential model.

Assessing incidence and severity of inspiratory flow limitation (IFL) is of importance for patients suffering of sleep disordered breathing (SDB) in order to diagnose a spectrum of different pathologies. In this study a new exponential equation is proposed to characterize the pressure/flow relationship of IFL and non-IFL breaths. Classical and alternative criteria are applied on the model's predictions in order to assess IFL, and its outcome is compared to the outcome of other models. The newly proposed exponential model seems to be promising, as it outperforms other models by achieving a global average sensitivity of 93% and specificity of 91%, and the lowest mean square error when estimating resistance at peak pressure. Additional statistical tests were performed on the exponential model's coefficients in order to determine if a coefficient based classification is possible.

The wireless monitoring of vital parameters: a design study.

This project shows the way vital parameters can be transmitted and visualized with no connecting cables necessary to the PDA. This was realized using a sensor developed with an integrated Bluetooth interface and a PDA, also equipped with Bluetooth. This radio connection can span up to 10 m, and parameters, such as pulse frequency, oxygen saturation in blood, ECG measurements and plethysmograms, can be transmitted. Using the software introduced in this work, the transmitted measurements can be displayed numerically or graphically on the PDA. The software simultaneously checks for any limits and sends a warning message if these limits are exceeded. All received data are additionally documented.

Miniaturized module for the wireless transmission of measurements with Bluetooth.

The wiring of patients for obtaining medical measurements has many disadvantages. In order to limit these, a miniaturized module was developed which digitalizes analog signals and sends the signal wirelessly to the receiver using Bluetooth. Bluetooth is especially suitable for this application because distances of up to 10 m are possible with low power consumption and robust transmission with encryption. The module consists of a Bluetooth chip, which is initialized in such a way by a microcontroller that connections from other bluetooth receivers can be accepted. The signals are then transmitted to the distant end. The maximum bit rate of the 23 mm x 30 mm module is 73.5 kBit/s. At 4.7 kBit/s, the current consumption is 12 mA.

Evaluation and optimization of different spectral estimation methods for EEG signals.

The goal of this work was the evaluation of various spectral estimation methods with regard to their suitability for classifying EEG data. A test environment was implemented in which the algorithms are optimized and evaluated using various artificial and real EEG data. The methods are based on autoregressive approaches, as well as from FFT, wavelet, and matching pursuit-based spectral estimations. The evaluation showed that the quality of the results strongly correlate with the computational effort of the algorithm. The matching pursuit algorithm (MP) was implemented and further optimized since it had the best test result and had good scalability. Even under a sufficiently low runtime, it still gave good results.

Automated sleep stage detection with a classical and a neural learning algorithm--methodological aspects.

For classification tasks in biosignal processing, several strategies and algorithms can be used. Knowledge-based systems allow prior knowledge about the decision process to be integrated, both by the developer and by self-learning capabilities. For the classification stages in a sleep stage detection framework, three inference strategies were compared regarding their specific strengths: a classical signal processing approach, artificial neural networks and neuro-fuzzy systems. Methodological aspects were assessed to attain optimum performance and maximum transparency for the user. Due to their effective and robust learning behavior, artificial neural networks could be recommended for pattern recognition, while neuro-fuzzy systems performed best for the processing of contextual information.

[Development of Bluetooth wireless sensors]. / Entwicklung drahtloser Sensoren auf Basis von Bluetooth.

Wireless communication could help to overcome current obstacles in medical devices and could enable medical services to offer completely new scenarios in health care. The Bluetooth technology which is the upcoming global market leader in wireless communication turned out to be perfectly suited not only for consumer market products but also in the medical environment [1]. It offers a low power, low cost connection in the medium range of 1-100 m with a bandwidth of currently 723.2 kbaud. This paper describes the development of a wireless ECG device and a Pulse Oximeter. Equipped with a Bluetooth port, the measurement devices are enabled to transmit data between the sensor and a Bluetooth-monitor. Therefore, CSR's Bluetooth protocol embedded two-processor and embedded single-processor architecture has been used.

Design aspects for novel, telemedical unattended diagnosis and therapy control systems for sleep disorders.

Medical research confirmed the relevance of diagnosis and therapy of sleep-related breathing disorders, revealing prevalence and patho-physiological mechanisms [1-10]. Reliable sleep assessment currently demands examinations in the sleep-lab, which is very technical and time-consuming. Thus, expanding the conventional procedure to meet capacity demands does not seem realistic regarding cost aspects. As a solution, the authors present telemetric system concepts for the unattended assessment of a patient's sleep profile and cardio-respiratory parameters. Taking initial experience with home polysomnography into consideration, the presented research work focuses on easily applicable sensors, the corresponding sleep profile assessment rules, wireless data recording, intelligent signal processing algorithms and telemedical information logistics to include sleep-labs in the diagnosis and treatment process as competence centers.

[Artificial intelligence in sleep analysis (ARTISANA)--modelling visual processes in sleep classification]. / Artificial Intelligence in Sleep Analysis (ARTISANA)--Modellierung des visuellen Vorgehens bei der Schlafklassifikation.

We describe a novel approach to the problem of automated sleep stage recognition. The ARTISANA algorithm mimics the behaviour of a human expert visually scoring sleep stages (Rechtschaffen and Kales classification). It comprises a number of interacting components that imitate the stepwise approach of the human expert, and artificial intelligence components. On the basis of parameters extracted at 1-s intervals from the signal curves, artificial neural networks recognize the incidence of typical patterns, e.g. delta activity or K complexes. This is followed by a rule interpretation stage that identifies the sleep stage with the aid of a neuro-fuzzy system while taking account of the context. Validation studies based on the records of 8 patients with obstructive sleep apnoea have confirmed the potential of this approach. Further features of the system include the transparency of the decision-taking process, and the flexibility of the option for expanding the system to cover new patterns and criteria.

Delayed VO2 kinetics during ramp exercise: a criterion for cardiopulmonary exercise capacity in chronic heart failure.

PURPOSE: Kinetics of VO2 at onset of constant work rate exercise was previously shown to be slowed in patients with chronic heart failure (CHF) compared with that in healthy normals. Because bicycle ergometry with ramp protocol is usually used for exercise testing with CHF patients, it would be of practical importance if it can be shown that a delay in the time interval of linear increase of VO2 (TILIV) to work rate occurs after beginning ramp exercise. Data of central hemodynamics (CHF) and noninvasive cardiopulmonary parameters (CHF, normals) should also correlate with VO2 delay time if this parameter is related to cardiopulmonary exercise capacity. METHODS: Fifteen males with CHF (mean +/- SEM: age 52 +/- 2 yr; ejection fraction 32 +/- 4%; peak cardiac index 3.9 +/- 0.3 L x m(-2) x min(-1)) and 28 healthy males (50 +/- 1 yr) were assessed. During ramp bicycle ergometry (3 min unloaded, work rate increments of 12.5 W x min(-1)), VO2 was measured breath by breath. RESULTS: After the onset of ramp exercise, there was a difference in the TILIV between patients and normals (83.7 +/- 3.6 vs 66.8 +/- 2.9 s; P < 0.001). Significant differences between both groups were also found for VO2 at ventilatory threshold (VT) (10.1 +/- 0.1 vs 15.2 +/- 0.7 mL x kg(-1) x min(-1); P < 0.0001), VO2 at VT relative to predicted VT (58 +/- 4 vs 97 +/- 4%; P < 0.0001), peak VO2 (13.2 +/- 1.0 vs 34 +/- 1.4 mL x kg(-1) x min(-1), P < 0.001), and increase of systolic blood pressure (36 +/- 7 vs 71 +/- 5 mm Hg; P < 0.0001). In CHF, the TILIV correlated significantly with peak cardiac index and VO2 at VT (r = -0.71; P < 0.005 each), relative value of VO2/kg at VT (r = -0.61; P < 0.03), peak VO2/kg (r = -0.63; P < 0.01), and increase of systolic blood pressure (r = -0.52; P < 0.02). In the normals only VO2/kg at VT correlated significantly with TILIV (r = -0.41; P < 0.03). In patients, stepwise regression analysis identified three predictors which could explain 79% of the variance of TILIV: VO2/kg at VT (r2 = 0.51), peak cardiac index (r2 = 0.20), and peak VO2/kg (r2 = 0.08). CONCLUSION: TILIV, determined at the onset of ramp exercise, is prolonged in CHF patients compared with that in normals and reflects severity of functional impairment because of reduced cardiac index and aerobic capacity. TILIV can provide information about changes in cardiopulmonary exercise capacity and thus can be used for follow-up and treatment studies in CHF.

Predictors of response to exercise training in severe chronic congestive heart failure.

We prospectively assessed whether baseline central hemodynamics and exercise capacity can predict improvement of VO2 at ventilatory threshold (VT) after exercise training in patients with severe chronic congestive heart failure. Eighteen patients (mean +/- SEM; age 52 +/- 2 years), half of them listed for transplant, underwent 3 weeks of exercise training (interval cycle and treadmill walking; 5 x/week) and 3 weeks of activity restriction in a random-order crossover trial. Baseline data were not significantly different for groups with exercise training first and activity restriction first: cardiac index at rest (2.1 +/- 0.1 L/m2/min), maximum cardiac index (3.1 +/- 0.2 L/m2/min) (Fick), and echocardiographic ejection fraction (21 +/- 1%). The same was true for cardiopulmonary exercise data (cycle ergometry; up 12.5 W/min): VO2 at VT (9.3 +/- 0.4 ml/kg/min), maximum VO2 (12.2 +/- 0.7 ml/kg/min), VT in percentage of predicted maximum VO2 (31 +/- 2%), heart rate at VT (95 +/- 4 beats/min), and decrease of dead space-to-tidal volume ratio from rest to VT (33 +/- 1 --> 29 +/- 1). Improvement of VO2 at VT after training (2.2 +/- 0.4 ml/kg/min; p <0.001) was not related to baseline central hemodynamics (r = <0.10 for each), but was greater in patients with a lower baseline VO2 at VT (r = -0.65; p <0.01), peak VO2 (r = -0.66; p <0.01), VT in percentage of predicted maximum VO2 (r = -0.74; p <0.001), heart rate at VT (r = -0.63; p <0.01), and smaller decrease of dead space-to-tidal volume ratio from rest to VT (r = 0.65; p <0.01). Ejection fraction after exercise training (24 +/- 2%) and activity restriction (23 +/- 2%) did not differ significantly compared with baseline, and patient status (heart failure and cardiac rhythm) remained stable. Three parameters accounted for 84% of the variance of improvement in VO2 at VT: VO2 at VT in percent predicted maximum VO2, decrease of dead space-to-tidal volume ratio, and heart rate at VT. The findings suggest that there was a greater increase in VO2 at VT after exercise training in patients with greater peripheral deconditioning at baseline. The improvement was unrelated to central hemodynamics. Clinically stable patients with severe chronic congestive heart failure, potential heart transplant candidates, and those awaiting transplantation may benefit from involvement in a short-term exercise training program.

Short-term reproducibility of cardiopulmonary measurements during exercise testing in patients with severe chronic heart failure.

Eleven men with severe chronic heart failure (peak cardiac index 4.0 +/- 0.2 L/m2/min), six on a heart transplantation waiting list, were prospectively assessed. To determine reproducibility of cardiopulmonary and hemodynamic variables for clinical purposes during ramp bicycle ergometry, the patients underwent two ramp bicycle ergometer tests (3 minutes unloaded, work rate increments of 12.5 W/min) with a 1-week interval between tests. Oxygen uptake (VO2) carbon dioxide production (VCO2), and ventilation were measured breath by breath, and calculations were performed to determine gas exchange ratio, oxygen pulse, ventilatory equivalents of oxygen and carbon dioxide, and end-tidal partial pressure for oxygen and carbon dioxide. Additionally, heart rate, blood pressure, and lactate levels were assessed. Measurements were performed at submaximum work rate levels of 25 W, 50 W, and 75 W at ventilatory threshold and at peak work rate. At all measurement points, the coefficient of variation for cardiopulmonary variables was between 1.4% and 7.1% for submaximum work rate levels, between 1.2% and 4.4% at ventilatory threshold, and between 2.4% and 7.1% at peak work rate. For heart rate, blood pressure, and lactate levels, coefficient of variation was between 2.7% and 5.7% for submaximum work rate levels, between 1.4% and 6.1% at ventilatory threshold, and between 1.2% and 5.5% at peak work rate. The data suggest high reproducibility for duplicate measurements of cardiopulmonary and hemodynamic variables during ramp bicycle ergometry in patients with severe chronic heart failure. The results may be used to determine whether any variable in a single patient is significantly different from that obtained in a previous exercise test or if the change is within experimental error.

Effects of exercise training and activity restriction on 6-minute walking test performance in patients with chronic heart failure.

Eighteen hospitalized patients with severe chronic heart failure (ejection fraction [mean +/- SEM] 21% +/- 1%) underwent 3 weeks of exercise training (interval bicycle ergometer and treadmill walking training exercises) and 3 weeks of activity restriction in a random-order crossover trial. Before and after exercise training and after activity restriction, a 6-minute walking test was performed to determine the maximum distance walked, hemodynamic and cardiopulmonary responses, norepinephrine levels, and ratings of leg fatigue and dyspnea while walking. A ramp test on bicycle ergometer (increments of 12.5 W/min) was performed before and after exercise training and activity restriction to determine peak oxygen uptake. After training, the maximum distance walked was increased by 65% (from 232 +/- 21 m at baseline to 382 +/- 20 m; p < 0.001), whereas after activity restriction (253 +/- 19 m) there was no significant difference compared with baseline. No significant differences in hemodynamic and cardiopulmonary parameters (with the exception of the ventilatory equivalent for carbon dioxide and perceived exertion) or norepinephrine levels were observed during walking tests. Improvement in maximum distance walked correlated significantly with training-induced increase in peak oxygen uptake measured during bicycle ergometry (r = 0.47, p < 0.05). The lower the maximum distance walked at baseline, the more pronounced the training-induced prolongation of maximum distance (r= -0.73; p < 0.001). These data support the value of exercise training in patients with severe chronic heart failure for improving maximum distance walked, as documented by the 6-minute walking test. The impairment of walking test performance during activity restriction suggests a need for long-term exercise training programs.

Interval training in patients with severe chronic heart failure: analysis and recommendations for exercise procedures.

This study analyzes a new exercise training procedure, which includes interval exercise training on cycle ergometer (IntCT) (30-s work phases/60-s recovery phases) and on treadmill (60-s work and recovery phases each). Training was applied for 3 wk in 18 patients with severe chronic heart failure (CHF) ((mean +/- SEM) age 52 +/- 2 yr, ejection fraction 21 +/- 1%). Peak VO2 was increased from 12.2 +/- 0.7 to 14.6 +/- 0.7 ml-kg-1 min-1 owing to training (P < 0.001). A specific steep ramp test (work rate increments 25 W.10 s-1) was developed to derive exercise intensity for work phases in IntCT, which was 50% of the maximum work rate achieved. Steep ramp test was performed at the start of the study to determine the initial training work rate, then weekly to readjust it. Since the maximum work rate achieved from this test increased weekly (144 +/- 10 W -->172 +/- 10 W-->200 +/- 11 W; P < 0.001), the training work rate also increased (72 +/- 4 W-->86 +/- 6 W-->100 +/- 7 W; P < 0.001). Physical responses to IntCT were measured. There was no significant change in heart rate, blood pressure, and ratings of perceived exertion (RPE) using a Borg Scale between the first and the third week of training (heart rate 88 +/- 3 b.min-1; blood pressure 115 +/- 4/80 +/- 2 mm Hg; leg fatigue 12 +/- 1; dyspnea 10 +/- 1). Mean lactate concentration (1.70 +/- 0.09 mmol-1-1) indicated an overall aerobic range of training intensity. When compared with the commonly used intensity level of 75% peak VO2 from an ordinary ramp test (work rate increments 12.5 W.min-1), the performed training work rate was more than doubled (240%; P < 0.0001) while cardiac stress was lower (86%; P < 0.01). Values of norepinephrine and epinephrine as well as of RPE corresponded to those measured at 75% peak VO2. Interval exercise training is thus recommended for selected patients with CHF as it allows intense exercise stimuli on peripheral muscles with minimal cardiac strain. Using a steep ramp test, training work rate can be determined and readjusted weekly during initial training period.

Cardiopulmonary determinants of functional capacity in patients with chronic heart failure compared with normals.

BACKGROUND: Patients with chronic heart failure (CHF) are characterized by abnormal gas exchange and ventilatory responses to exercise. HYPOTHESIS: This study compares variables obtained from cardiopulmonary exercise testing in 35 patients with CHF with 35 age- and weight-matched healthy subjects. A second goal was to obtain cardiopulmonary variables measured at ventilatory threshold to distinguish patient changes from those of healthy subjects. METHODS: Exercise testing was carried out using bicycle ergometry with ramplike protocol (work rate increments 12.5 W/min). Gas exchange and ventilation were measured breath by breath. RESULTS: Compared with healthy subjects, the VO2 in patients was lower at identical work rates (p < 0.004) and at ventilatory threshold (p < 0.0001), and the slope of the VO2 curve during incremental exercise was flatter (p < 0.05). With the exception of heart rate, the variables for VO2, VCO2, ventilation, O2 pulse, ventilatory equivalents for O2 and CO2, and VD/VT (physiologic deadspace to tidal volume ratio), as well as lactate differed significantly at identical work rates. With the exception of VD/VT, all cardiopulmonary variables showed significant differences in their slopes during exercise. By means of a discriminant analysis, VCO2 and ventilation proved to be the most distinguishing variables at ventilatory threshold between patients with CHF and healthy subjects. CONCLUSIONS: These results indicate the clinical usefulness of cardiopulmonary exercise testing when assessing functional impairment due to CHF. For treatment evaluation, not only VO2 but also VCO2 and ventilation responses to exercise should be considered.