Cystic fibrosis (CF) is a multiorgan disease based on mutations in the gene encoding for the cystic fibrosis transmembrane regulator Cl− channel. Quality of life and life expectancy of affected individuals are often limited by progressive lung disease. In addition, many patients with CF suffer from pancreatic insufficiency, which may result in malnutrition if not treated appropriately. Furthermore, some studies suggest changes in metabolic and functional muscle characteristics associated with CF (5,20 20
Normal habitual activities, especially in children, are characterized by frequent changes in metabolic demand. A fast adaptation of the cardiorespiratory system to the changing metabolic demand is thought to be advantageous in these situations. For example, a slowed response of oxygen uptake (V̇O2 ) after an increase in exercise intensity results in a relatively higher oxygen debt, which has been linked to premature fatigue.
V̇O2 on-kinetics are often studied to assess an individual's ability to adapt to changes in work rate. In most studies, a steplike increase in exercise intensity is employed to trigger adaptations. V̇O2 response is then best described by a three-phase model (33 2 increases rapidly as cardiac output and pulmonary blood flow increase at the start of exercise . Phase I ends with the arrival of blood with a greater level of deoxygenation from the exercising muscle in the lungs. Phase II is the further increase in V̇O2 as venous return continues to increase and more O2 has been extracted at the exercising muscles. In moderate-intensity exercise , the phase II V̇O2 response may be described by a monoexponential equation, which provides estimates of the delay, the amplitude, and the time constant tau of the response. Phase III is the steady state. In exercise above the anaerobic threshold, a second monoexponential with a longer time constant may be observed, the slow component. In this case, the steady state may be delayed or may not be reached before the subject is fatigued.
A large number of studies have been performed using the above approach in healthy individuals and in those with a chronic health condition, covering all age groups from children to the elderly. In consequence, the understanding of factors limiting oxygen uptake kinetics is derived mainly from these studies. In healthy individuals, some controversy exists whether phase II kinetics are limited by oxygen delivery or muscle O2 metabolism (31,32 18,24 8,29 7,12,31
So far, V̇O2 kinetics have not been studied in individuals with CF using the above methodology. The few available studies on V̇O2 kinetics in CF employed different experimental designs and show conflicting results: Braggion et al. (4 1 2 response after an increase in work rate. In contrast, Massin et al. (17 15 2 kinetics between patients with CF and controls when using the pseudorandom binary sequence method (PRBS). However, both approaches do not distinguish between the phase I and II response. Furthermore, using PRBS, changes in V̇O2 during recovery are also included in the measurement of kinetics. In consequence, cardiodynamic factors will have considerable impact on the measures of V̇O2 kinetics using the latter techniques. Thus, the physiological interpretation of slowed kinetics in CF using these methods is difficult.
The objective of the present study was to determine phase II V̇O2 on-kinetics in patients with CF and healthy controls. Limitations in pulmonary function, right heart dysfunction, and reduced aerobic capacity have been repeatedly reported in CF. Because these factors have been associated with slowed V̇O2 kinetics, we hypothesized that patients with CF would have prolonged kinetics. To better understand factors that could be limiting V̇O2 on-kinetics in CF, we intended to identify correlates of V̇O2 time constants.
MATERIALS AND METHODS
Subjects.
A total of 24 patients with CF, aged 10–33 yr, volunteered to participate in this study. CF was diagnosed in all subjects on the basis of characteristic symptoms, two pathological sweat tests, and typical findings on chest x-ray. Scores of disease severity (Shwachman score and Crispin Norman score) and status of pancreatic sufficiency were derived from the visit to the CF clinic of the hospital closest to the time of testing. Results of genotyping were taken from the patients' charts.
Six patients with CF were excluded from analysis because the V̇O2 data were too scattered to fit a reliable monoexponential to the phase II response (for details, see below). These patients had similar characteristics compared with the 18 patients contributing data to the analysis. Table 1 summarizes the descriptive characteristics of the 18 patients included in the analysis. Nine of the 18 patients were homozygous for ΔF508 (class 2 mutation), and one patient was heterozygous for ΔF508 and the class 1 mutation G542X. Three patients were heterozygous for ΔF508 and a class 3 (G551D) or class 4 (R334W, R347P) mutation. One patient was heterozygous for R553X (class 1) and 3849 + 10 kb C→T (class 5). Three patients were heterozygous for ΔF508 and an unknown mutation; one was heterozygous for N1303K (class 2) and an unknown mutation. Based on the observation that patients with a “severe” (Type 1 or 2) mutation in addition to ΔF508 have a lower aerobic fitness than those with ΔF508 and a Type 3, 4, or 5 mutation (27
TABLE 1: Subjects' characteristics.
Nineteen healthy volunteers served as controls. None of the controls suffered from any acute or chronic health condition at the time of testing, none took medication, and all were nonsmokers. Data of four controls were excluded from analysis (details see below). These four controls were not different in any descriptive variable compared with the remaining 15 controls who contributed data for analysis. Characteristics of these latter 15 subjects are summarized in Table 1 .
Design and procedures.
Subjects came to the laboratory in the afternoon for testing. Upon arrival at the laboratory, the study's objectives and methodology were explained to the subjects. The study protocol was approved by the ethics committee of the Medical Faculty of Würzburg University. Written informed consent was obtained from the subjects and their guardians, if appropriate. A brief medical history was taken and a medical examination performed to identify and exclude controls with any acute or chronic health condition and patients with acute exacerbations. Height and weight were determined to the nearest 0.1 cm and 100 g with the subjects wearing light exercise clothing but no shoes. Z-scores for body mass index (BMI = weight·height−2 ) were calculated according to Rolland-Cachera et al. (26
Subjects performed at least three forced expiratory maneuvers to determine forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV1) (CPX/D, MedGraphics, St. Paul, MN). Data were expressed as percentage of predicted (28 exercise tasks, with a break of at least 10 min between tasks. Subjects started pedaling at a work rate of 20 W and with a cadence of 50 rpm in a semisupine position (stage 1; Ergometrics 900L, Ergoline Europa, Bitz, Germany). This body position was chosen to minimize movement artifacts affecting pulse oximetry. After 2 min, exercise intensity was increased to 1.4 W·kg−1 body weight in male subjects or 1.3 W·kg−1 in females (stage 2). Subjects were instructed to keep the cadence constant. Subjects continued cycling at this second level of exercise intensity for 3 min. There were two reasons to choose a 3-min duration for stage 2: First, we presumably exercised our subjects at a low enough intensity not to induce a slow component. To fit a monoexponential equation, 3 min of data are sufficient if the time constants are not too long. Preliminary work with CF patients showed time constants of 30–40 s; 3-min stages, thus, should have resulted in a time sufficient to include phase I plus four times the time constant. Second, we could test our subjects who partly came from far away only on one occasion, and we wanted to complete several repetitions, plus a V̇O2peak test, on that day. Longer stages would have resulted in failure to recruit enough subjects because of long visits and, possibly, premature fatigue. With the approach chosen, there was no evidence of fatigue. Heart rate (HR) was monitored by ECG (Custo card m, custo med, München, Germany) and oxygen saturation (SpO2 ) by pulse oximetry (NP10 forehead sensor, NPB 290 pulse oximeter, Nelcor Puritan Bennet, Pleasanton, CA). Subjects breathed through a mouthpiece with a saliva trap and a lightweight, low dead space pneumotach during the test (Pneumotach; MedGraphics; total dead space including mouthpiece 85 mL). Ventilatory and respiratory variables were determined breath by breath using a commercially available metabolic cart calibrated before and after each test with gases of known concentrations (CPX/D, MedGraphics). Averages of HR, V̇O2 , respiratory exchange ratio (RER), oxygen pulse (V̇O2 /HR), and SpO2 were computed for the final 30 s of stage 1 and for the final 30 s of stage 2 for each individual using the measurements taken during all submaximal exercise tasks of this subject.
After a rest of at least 10 min, a continuous incremental cycling task was performed. Starting with 0.6 W·kg−1 body weight, work rate was increased every 2 min by 0.7 W·kg−1 body weight until the subjects could not maintain cycling cadence despite verbal encouragement. HR and SpO2 were monitored as described above. V̇O2 , carbon dioxide output, and ventilation were measured breath by breath. V̇O2peak was determined as the highest V̇O2 during two consecutive 15-s periods, and expressed as percentage of predicted calculated from height and gender (23 10
Modeling phase II V̇O2 kinetics.
V̇O2 data of the submaximal exercise tasks were interpolated second by second, time aligned, and averaged. A 5-s moving average was used to smooth the signal, as described elsewhere (9 2 without a concomitant increase in V̇CO2 was first identified in each individual. Thereafter, monoexponential equations (program 3R, BMDP Statistical Software Version 7.0, Cork, Ireland) were fitted to the V̇O2 –over-time relationship, including first only data gathered well within phase II (the approximate end of phase I +10 s). The fitting was then repeated 20 times for each data set, including one extra data point at a time (approximate end of phase I +9 s, +8 s, and so on until −10 s). The definite beginning of phase II was defined as the time when 1) the lowest mean residual square was reached, and 2) further shortening of phase I resulted in a sudden increase in tau. The monoexponential equation describing the entire phase II V̇O2 response was used to determine amplitude (mL O2 ·W−1 ) and time constant (tau) of the V̇O2 response (9
For each individual, measured V̇O2 and the fitted monoexponential regression line were plotted over time. A plot of residuals over time was also prepared. Three blinded observers were asked to determine whether they felt that the data were too scattered to be adequately described by the regression line, or whether the regression line did not fit the data. If one of three observers rated the time-aligned and averaged data as too scattered for fitting a monoexponential regression line, or if the reviewer was concerned that a slow component was present, the individual was excluded from analysis. To further search for a slow component V̇O2 , the fitting of the monoexponential equation was repeated using only the data of the phase II response until 120 s into stage 2. In none of the subjects, a time constant was observed that was below the 95% confidence interval of the time constant derived from the analysis of the entire data set.
Statistical analysis.
ANOVA revealed no significant effect of gender on the entire V̇O2 reponse during stage 2, or on tau and amplitude of the phase II V̇O2 response. There was also no significant gender × disease state interaction so that female and male subjects were pooled into one control group and one patient group. Differences between patients with CF and healthy controls were analyzed using Student's t -test. Testing size-related variables such as oxygen pulse for differences between the groups was performed using ANCOVA with body weight as covariate. A two-way ANOVA for one repeated measure was used to compare the V̇O2 response between healthy controls and patients with CF 1) during the entire 3 min of stage 2, and 2) during the final minute of the stage.
Least square linear regression analysis was performed to identify significant predictors of the time constant tau in the entire sample and in the subsamples of patients with CF and controls separately. The following variables were included as independent variables in the analyses: age (yr), gender, height (cm), weight (kg), BMI (z-score), FVC (%predicted), FEV1 (%predicted), V̇O2peak (%predicted and mL·min−1 ·kg−1 ), Wmax (W·kg−1 ), VAT (%V̇O2peak and mL·min−1 ·kg−1 ), O2 saturation at peak exercise during the incremental cycling task, exercise intensity during the second stage of the submaximal exercise task to determine V̇O2 kinetics (%V̇O2peak and %VAT), O2 saturation at the end of the two stages (1 and 2) of the submaximal exercise task, and amplitude of the phase II V̇O2 response. For the analysis in CF patients alone, three additional variables were used: Crispin Norman score, Shwachman score, and type of mutation. ANCOVA was employed to determine between group differences with significant correlates of tau as covariates.
All statistical analysis was performed using BMDP statistical software version 7.0 (BMDP Statistical Software). Statistical significance was taken at P < 0.05.
RESULTS
Table 2 summarizes the exercise responses of the patients with CF and the controls during stage 1 and stage 2 of the submaximal exercise tasks. There was no difference between groups in relative exercise intensity of stage 2, as expressed as percent V̇O2peak or percent VAT. HR was similar in patients with CF and controls during both stages. Although oxygen pulse appeared to be lower in patients with CF during stage 2, ANCOVA with body mass as covariate revealed no difference between groups in stage 2 oxygen pulse. There was also no difference in the relative increase in oxygen pulse with increasing exercise intensity between the groups. There were, however, two variables in which between-group differences during the submaximal exercise were observed: RER was significantly higher and SpO2 was lower during cycling at both submaximal intensity levels in individuals with CF compared with controls (Table 2 ).
TABLE 2: Characteristics of the two-level submaximal exercise task in CF patients and controls.
Figure 1 shows the average V̇O2 response during stage 2 of the submaximal exercise in healthy controls and patients with CF. ANOVA revealed no significant difference between groups (P = 0.23) but did show a significant group × time interaction (P < 0.001), indicating a significantly different time course of V̇O2 in the two groups. When only the V̇O2 data of the final minute of the 3-min stage were included in the analysis, neither the difference between groups (P = 0.91) nor the group × time interaction (P = 0.78) were significant.
The individual V̇O2 responses of two healthy subjects and two patients with CF are presented in Figure 2 . A graphical presentation of the monoexponential functions fitted to the phase II response and the pattern of residuals are also shown for each individual.
There was no difference in duration of phase I between patients with CF and controls (19.9 ± 6.3 vs 19.1 ± 6.8 s). Table 3 summarizes amplitude and tau of the phase II V̇O2 response in CF patients and controls. Tau was significantly prolonged in CF.
TABLE 3: Amplitude and time constant tau of the phase II V̇O2 response in CF patients and controls; data are means ± SD of the means (range).
Age (r = 0.68, P = 0.003), height (r = 0.52, P = 0.032), weight (r = 0.52, P = 0.033), FEV1 (r = −0.53, P = 0.029), V̇O2peak (mL·min−1 ·kg−1 ; r = −0.59, P = 0.013), SpO2 at the end of stage 1 (r = −0.64, P = 0.006) and at the end of stage 2 (r = −0.69, P = 0.002), and SpO2 at peak exercise (r = −0.72, P = 0.001) were identified as significant correlates of tau in the patients with CF. Clinical scores of disease severity (Crispin Norman and Shwachman score), genotype, or BMI z-score as index of nutritional status were not associated with tau.
In healthy controls, only age (r = 0.72, P = 0.002), height (r = 0.57, P = 0.027), and exercise intensity during stage 2 of the submaximal exercise test expressed as percentage of VAT (r = −0.55, P = 0.033) correlated significantly with tau.
Combining the two groups revealed age (r = 0.45, P = 0.008), FVC (r = −0.40, P = 0.021), FEV1 (r = −0.54, P = 0.001), SpO2 at the end of stage 1 (r = −0.60, P < 0.001) and at the end of stage 2 (r = −0.63, P < 0.001; Fig. 3 ), and SpO2 at peak exercise (r = −0.66, P < 0.001) as significant correlates of tau. When ANCOVA was used to adjust for the effects of FEV1, FVC, SpO2 at the end of stage 1, SpO2 at the end of stage 2, or SpO2 at peak exercise , on tau, the difference in tau between patients with CF and controls vanished. However, differences in age could not explain the differences in tau between the groups.
DISCUSSION
In the present study, cystic fibrosis was associated with slowed on-transients of oxygen uptake. This finding is in contrast to two other studies (1,4 exercise intensity to determine V̇O2 on-kinetics. Furthermore, they used a different mathematical model to analyze the data: In the studies by Braggion et al. (4 1 2 . This approach results in significantly different time constants, and may lead to different conclusions compared with an analysis of phase II alone (9,33 2,3,33 2 responses by modeling phase I and II separately.
In three other studies, impaired V̇O2 kinetics were observed in patients with CF compared with healthy controls (11,15,17 exercise intensities. Spectral analysis was used to determine amplitude and phase shift of the V̇O2 response at various harmonic frequencies. This type of analysis integrates the phase I and II V̇O2 on-responses as well as the recovery V̇O2 response.
The protocol employed in the present study does not allow a mathematical proof that there was no slow component V̇O2 during the second stage of the submaximal exercise task in some individuals. It could, thus, be argued that, if a slow component V̇O2 response was present in some patients with CF but not in healthy controls, this slow component might have resulted in falsely prolonged time constants in the former. However, three reviewers who inspected the plots of raw data including the fitted monoexponential equations and the residual plots did not detect a slow component in any of the subjects included in the analysis. Furthermore, no significant shortening of time constants was observed in any subject when the equation was fitted only to the V̇O2 data gathered during the first 120 s of stage 2. And finally, there was no difference in the time course of V̇O2 between the controls and the patients with CF during the final minute of the submaximal exercise task (Fig. 1 ). Thus, we are confident that a slow component, if it had occurred undetected, cannot explain the between group differences in V̇O2 kinetics.
We employed a semisupine position for cycling to minimize movement artifacts. To our knowledge, no studies have compared the time constants of V̇O2 kinetics in semisupine and upright cycling. However, it has been shown that phase II V̇O2 on-transients are slower in a fully supine compared with an upright body position (13
The reason for slowed oxygen uptake kinetics in patients with CF is not easy to discern. Because impaired lung functions are most obvious in patients with this disease, it would seem logical to assume that kinetics are slowed due to an impaired pulmonary oxygen uptake. There are several findings that would support this hypothesis: In our study, differences in FVC and FEV1 between patients with CF and controls could be related to the differences in tau. In line with these findings, Massin et al. (17 2 kinetics (amplitude ratio) in patients with CF. Furthermore, in the present study, a significant relationship between SpO2 at the end of the second stage of the submaximal exercise test (and the SpO2 at peak exercise ) and tau was observed in patients with CF, and the CF and control groups combined. The slightly lower capillary oxygen saturation in the patients with CF, which reflects a decrease in oxygen pressure of more than 10 mm Hg might have contributed to an altered redox potential and phosphorylation state of the exercising muscle, thereby slowing V̇O2 kinetics (31 2 and tau does not reflect causality but results from a significant relationship of both variables to a third, yet unknown. Nevertheless, there is evidence for a causal relationship between oxygen delivery (SpO2 ) and V̇O2 kinetics: In healthy adults breathing air with low oxygen content, prolonged phase II V̇O2 time constants have been observed compared with normoxia (6,21 2 kinetics in some studies (24
Kusenbach et al. (15 2 kinetics in CF, and no effect of supplemental oxygen on slowed oxygen uptake kinetics was observed in their patients despite a considerable increase in blood oxygen saturation during exercise . Based on these findings, the authors concluded that oxygen uptake kinetics in CF might be predominantly limited by peripheral factors and not by impaired oxygen delivery (15 15 2 response. The slowed kinetics observed in their study might, thus, reflect cardiovascular factors. Furthermore, Kusenbach et al. (15 2peak was not determined in their subjects. It might well be that their patients with CF were quite unfit. We, however, observed only a relatively small difference in V̇O2peak between CF patients and controls. Possibly, our CF and control groups were more alike in their peripheral oxidative potential that allowed us to identify a relationship between SpO2 and tau.
V̇O2peak was only slightly lower in our patients with CF compared with the controls, and no association was observed between tau and V̇O2peak when both groups were combined. Because V̇O2peak is often thought to be limited mainly by oxygen delivery, this finding seems in contrast to the hypothesis of an impaired oxygen delivery causing slowed O2 kinetics in patients with CF. However, an increased inspiratory oxygen concentration does not raise V̇O2peak in patients with CF despite a decrease in minute ventilation and an increase in SpO2 (22 2peak , in addition to central factors.
In patients with CF, especially those with poor lung functions, heart function may be impaired (14 25 2 kinetics observed in our study and that by Massin et al. (17 2 kinetics. However, there was no difference between groups in oxygen pulse at both exercise -intensity levels during the submaximal exercise tasks, once differences in body mass between CF patients and controls were taken into account. Both groups could increase their oxygen pulse from stage I to II of the test by 40–45% (Table 2 ). Furthermore, there was no difference in the duration of the cardiodynamic phase, phase I, between the patients with CF and the healthy controls.
RER was higher in CF compared with controls during both intensity levels of the submaximal exercise task (Table 2 ). This finding is in line with reports on elevated RER in CF at rest (30 exercise (15 16 exercise intensity during dynamic bulb-squeezing exercise up to 55% of maximal voluntary contraction (e.g., at 30% of max voluntary contraction: CF 18,6 vs controls 33,6 L·mmol−1 ) (5 2 kinetics at the onset of exercise (31 2 on-transients should be speeded if oxidative metabolism during exercise relies on carbohydrates. Indeed, some data on healthy adults exercising at low intensities suggest that average phase II V̇O2 kinetics are fastest with RER being around 0.97 (19 2 time constants in the total sample or any subgroup. Furthermore, the patients with CF showed prolonged, not shortened time constants. It is, thus, unlikely that differences in substrate utilization might explain the prolonged time constants in the patients with CF.
There is accumulating evidence that oxygen uptake kinetics can be limited by peripheral factors in patients with cardiopulmonary disease (8,29 5,20 15 2peak , peak work rate (WR), and ΔV̇O2 /ΔWR, were not related to tau. There was also no association between nutritional status, as measured by BMI z-score, and tau. The fact that we could not detect a peripheral limitation of V̇O2 kinetics might be attributed to the relatively good aerobic fitness and nutritional status of our patients. However, our study was not designed to detect limitations of muscle metabolism that could be causing slowed oxygen uptake kinetics. It is, thus, possible that more elaborate techniques to study muscle metabolism may have discerned a peripheral limitation of oxygen uptake kinetics in CF. As stated above, central and peripheral factors most likely interact in limiting oxygen uptake kinetics.
This study is the first to show slowed phase II V̇O2 kinetics in patients with CF. This limitation may result in premature fatigue during tasks with frequent changes in metabolic demand. By using a steplike increase in work rate and a monoexponential fit to the phase II V̇O2 response, it was possible to search for factors associated with slowed kinetics. For the first time, an association between the oxygen saturation during submaximal exercise and the time constant of the phase II V̇O2 response was shown, which could explain, at least in part, the prolonged time constants in patients with CF. Thus, the slowed V̇O2 kinetics in CF may partly be attributed to an impairment of oxygen delivery. Peripheral factors may also play a role. Further studies are needed to understand better the mechanisms underlying the slowed V̇O2 kinetic in CF.
REFERENCES
1. Barker, M. Kinetics of the ventilatory responses to physical
exercise in healthy children and adolescents and those with cystic fibrosis. Doctoral Dissertation. Rheinisch-Westfälische Technische Hochschule, Aachen, Germany, 1997.
2. Barstow, T. J., R. Casaburi, and K. Wasserman. O
2 uptake kinetics and the O
2 deficit as related to
exercise intensity and blood lactate.
J. Appl. Physiol. 75:755–762, 1993.
3. Barstow, T. J., and P. A. Molé. Linear and nonlinear characteristics of oxygen uptake kinetics during heavy
exercise .
J. Appl. Physiol. 71:2099–2106, 1991.
4. Braggion, C., M. Cornacchia, A. Miano, F. Schena, G. Verlato, and G. Mastella.
Exercise tolerance and effects of training in young patients with cystic fibrosis and mild airway obstruction.
Pediatr. Pulmonol. 7:145–152, 1989.
5. DeMeer, K., J. A. L. Jeneson, V. A. M. Gulmans, J. van der Laag, and R. Berger. Efficiency of oxidative work performance of skeletal muscle in patients with cystic fibrosis.
Thorax 50:980–983, 1995.
6. Engelen, M., J. Proszasz, M. Riley, K. Wasserman, K. Maehara, and T. J. Barstow. Effects of hypoxic hypoxia on O
2 uptake and heart rate kinetics during heavy
exercise .
J. Appl. Physiol. 81:2500–2508, 1996.
7. Grassi, B. Regulation of oxygen consumption at
exercise onset: is it really controversial?
Exerc. Sport Sci. Rev. 29:134–138, 2001.
8. Grassi, B., C. Marconi, M. Meyer, M. Rieu, and P. Cerretelli. Gas exchange and cardiovascular kinetics with different
exercise protocols in heart transplant recipients.
J. Appl. Physiol. 82:1952–1962, 1997.
9. Hebestreit, H., S. Kriemler, R. L. Hughson, and O. Bar-Or. Kinetics of oxygen uptake at the onset of
exercise comparing boys and men.
J. Appl. Physiol. 85:1833–1842, 1998.
10. Hebestreit, H., B. Staschen, and A. Hebestreit. Ventilatory threshold: a useful method to determine aerobic fitness in children?
Med. Sci. Sports Exerc. 32:1964–1969, 2000.
11. Hoffmann, U., G. Kusenbach, R. Wieching, M. Barker, and D. Essfeld. Comparison of V̇O
2 kinetics and V̇O
2max in adolsecents with cystic fibrosis (Abstract).
Med. Sci. Sports Exerc. 25:S46, 1994.
12. Hughson, R. L., M. E. Tschakovsky, and M. E. Houston. Regulation of oxygen consumption at the onset of
exercise .
Exerc. Sport Sci. Rev. 29:129–133, 2001.
13. Hughson, R. L., H. C. Xing, C. Borkhoff, and G. C. Butler. Kinetics of ventilation and gas exchange during supine and upright cycle
exercise .
Eur. J. Appl Physiol. Occup Physiol. 63:300–307, 1991.
14. Ionescu, A. A., A.-A. Ionescu, N. Payne, I. Obieta-Fresnedo, A. G. Fraser, and D. J. Shale. Subclinical right ventricular dysfunction in cystic fibrosis.
Am. J. Respir. Crit. Care Med. 163:1212–1218, 2001.
15. Kusenbach, G., R. Wieching, M. Barker, U. Hoffmann, and D. Essfeld. Effects of hyperoxia on oxygen uptake kinetics in cystic fibrosis patients as determined by pseudo-random binary sequence
exercise .
Eur. J. Appl. Physiol. 79:192–196, 1999.
16. Larson, D. E., R. L. Hesslink, M. I. Hrovat, R. S. Fishman, and D. M. Systrom. Dietary effects on exercising muscle metabolism and performance by
31 P-MRS.
J. Appl. Physiol. 77:1108–1115, 1994.
17. Massin, M. M., J. Leclercq-Foucart, and J.-P. Sacre. Gas exchange and heart rate kinetics during sequence
exercise in cystic fibrosis.
Med. Sci. Monit. 6:55–62, 2000.
18. Mettauer, B., Q. M. Zhao, E. Epailly, et al. VO
2 kinetics reveal a central limitation at the onset of subthreshold
exercise in heart transplant recipients.
J. Appl. Physiol. 88:1228–1238, 2000.
19. Molé, P. A., and J. J. Hoffmann. VO
2 kinetics of mild
exercise are altered by RER.
J. Appl. Physiol. 87:2097–2106, 1999.
20. Moser, C., P. Tirakitsoontorn, E. Nussbaum, R. Newcomb, and D. M. Cooper. Muscle size and cardiorespiratory response to
exercise in cystic fibrosis.
Am. J. Respir. Crit. Care Med. 162:1823–1827, 2000.
21. Murphy, P. C., L. A. Cuervo, and R. L. Hughson. A study of cardiorespiratory dynamics with step and ramp
exercise tests in normoxia and hypoxia.
Cardiovasc. Res. 23:825–832, 1989.
22. Nixon, P. A., D. M. Orenstein, S. E. Curtis, and E. A. Ross. Oxygen supplementation during
exercise in cystic fibrosis.
Am. Rev. Respir Dis. 142:807–811, 1990.
23. Orenstein, D. M. Assessment of
exercise pulmonary function. In:
Pediatric Laboratory Exercise Testing: Clinical Guidelines , T. W. Rowland (Ed.). Champaign, IL: Human Kinetics, 1993, pp. 141–163.
24. Palange, P., P. Galassetti, E. T. Mannix, et al. Oxygen effect on O
2 deficit and VO
2 kinetics during
exercise in obstructive pulmonary disease.
J. Appl. Physiol. 78:2228–2234, 1995.
25. Pianosi, P., and A. Pelech. Stroke volume during
exercise in cystic fibrosis.
Am. J. Respir. Crit. Care Med. 153:1105–1109, 1996.
26. Rolland-Cachera, M. F., T. J. Cole, M. Sempé, J. Tichet, C. Rossignol, and A. Charraud. Body mass index variations: percentiles from birth to 87 years.
Eur. J. Clin. Nutr. 45:13–21, 1991.
27. Selvadurai, H. C., K. O. McKay, C. J. Blimkie, P. J. Cooper, C. M. Mellis, and P. P. Van Asperen. The relationship between genotype and
exercise tolerance in children with cystic fibrosis.
Am. J. Respir. Crit. Care Med. 165:762–765, 2002.
28. Sherrill, D. L., M. D. Lebowitz, R. J. Knudson, and B. Burrows. Continuous longitudinal regression equations for pulmonary function measures.
Eur. Respir. J. 5:452–462, 1992.
29. Somfay, A., J. Pórszász, S. M. Lee, and R. Casaburi. Effect of hyperoxia on gas exchange and lactate kinetics following
exercise onset in nonhypoxemic COPD patients.
Chest 121:393–400, 2002.
30. Spicher, V., M. Roulet, and Y. Schutz. Assessment of total energy expenditure in free-living patients with cystic fibrosis.
J. Pediatr. 118:865–872, 1991.
31. Tschakovsky, M. E., and R. L. Hughson. Interaction of factors determining oxygen uptake at the onset of
exercise .
J. Appl. Physiol. 86:1101–1113, 1999.
32. Whipp, B. J., and S. A. Ward. Physiological determinants of pulmonary gas exchange kinetics during
exercise .
Med. Sci. Sports Exerc. 22:62–71, 1990.
33. Whipp, B. J., S. A. Ward, N. Lamarra, J. A. Davies, and K. Wasserman. Parameters of ventilatory and gas exchange dynamics during
exercise .
J. Appl. Physiol. 52:1506–1513, 1982.
