Chest ACCP Education Calendar
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Guest Access | Sign In via User Name/Password
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Thin, A. G.
Right arrow Articles by McLoughlin, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Thin, A. G.
Right arrow Articles by McLoughlin, P.
(Chest. 2002;121:1761-1770.)
© 2002 American College of Chest Physicians

Use of the Gas Exchange Threshold to Noninvasively Determine the Lactate Threshold in Patients With Cystic Fibrosis*

Alasdair G. Thin, PhD; Seamus J. Linnane, MB; Edward F. McKone, MB; Rosemarie Freaney, PhD; Muiris X. FitzGerald, MD, FCCP; Charles G. Gallagher, MD, FCCP and Paul McLoughlin, MB, PhD

* From the Departments of Physiology (Drs. Thin and McLoughlin) and Medicine and Therapeutics (Drs. Linnane, McKone, and FitzGerald), Conway Institute of Biomedical and Biomolecular Research, University College, Dublin; and Education and Research Center (Dr. Freaney) and Department of Respiratory Medicine and National Adult Cystic Fibrosis Unit (Dr. Gallagher), St Vincent’s University Hospital, Dublin, Ireland.

Correspondence to: Paul McLoughlin, MB, PhD, Department of Physiology, Conway Institute of Biomedical and Biomolecular Research, University College Dublin, Earlsfort Terrace, Dublin 2, Ireland; e-mail: paul.mcloughlin{at}ucd.ie


   Abstract
TOP
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Conclusion
 References
 
Objective: The anaerobic threshold (AT) is a submaximal index related to endurance exercise performance, which is usually determined by the measurement of blood lactate concentration during an incremental exercise test (lactate threshold [LT]). The LT, and thus the AT, can also be detected noninvasively in normal subjects by means of the gas exchange threshold (GET). This study was undertaken to validate the use of GET in patients with cystic fibrosis (CF) with a wide range of disease severity, and to assess the reproducibility of this index.

Methods: In patients with CF (FEV1 range, 23 to 118% of predicted) and control subjects, gas exchange was measured breath by breath during the incremental exercise tests to allow determination of the GET. Arterialized-venous blood was sampled for determination of the LT. The GET and LT were determined in a blinded manner.

Results: The mean differences (GET - LT) for control subjects (n = 18) and patients with CF (n = 23) were - 40 mL/min and + 10 mL/min, respectively, neither being significantly different from zero. The limits of agreement were ± 550 mL/min and ± 410 mL/min, respectively. The mean test-retest differences in GET for control subjects (n = 14) and patients with CF (n = 12) were - 50 mL/min and 0 mL/min, respectively, neither being significantly different from zero; the respective limits of reproducibility were ± 450 mL/min and ± 350 mL/min.

Conclusions: This study demonstrates that in patients with CF, the GET can be used to obtain an unbiased estimate of the LT, and that the GET is reproducible.

Key Words: cystic fibrosis • gas exchange threshold • lactate threshold • noninvasive • submaximal


   Introduction
TOP
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Conclusion
 References
 
Higher levels of aerobic fitness (peak oxygen uptake [O2]) in patients with cystic fibrosis (CF) have been found to be associated with increased longevity.1 More recently, it has been reported that regular exercise attenuates the progressive decline in lung function.2 Physical activity also promotes sputum clearance,3 improves aerobic fitness,4 5 and has psychological benefits.6 For these reasons, it is recommended that regular exercise should form an important part of a CF treatment program.6 7 8 In order to monitor responses to training programs and to investigate the mechanisms by which aerobic exercise exerts its beneficial effects, methods of quantifying aerobic fitness in patients with CF are required that are not excessively stressful and are readily acceptable to patients.

The anaerobic threshold (AT) is a submaximal index of aerobic fitness that is closely related to endurance performance.9 10 11 The AT may be determined both by the measurement of blood lactate concentration (the lactate threshold [LT]) and noninvasively by a hyperventilatory response to the exercise-induced metabolic acidosis (the ventilatory threshold [VentT]).12 Subjects with significant obstructive lung disease are frequently unable to produce this hyperventilatory response. However, in these subjects, it may be still possible to detect an increase in the rate of carbon dioxide output (CO2) over the rate of O2 (the gas exchange threshold [GET]).13 14 15

The response of patients with CF to exercise is characterized by an increased ventilatory equivalents for oxygen (minute ventilation [E]/O2) and an increased ventilatory equivalents for carbon dioxide (E/CO2), and an elevated physiologic dead space to tidal volume ratio.16 17 18 Despite this inefficiency of carbon dioxide excretion, it has been demonstrated in CF patients with mild lung disease that the GET may be used to detect the AT noninvasively, as indicated by the agreement between GET and the invasively determined LT.19 Although in patients with mild CF ventilatory capacity is normal, with declining lung function there is evidence of ventilatory limitation.20 As the ventilatory and gas exchange abnormalities worsen with disease progression, it may no longer be possible to use the GET to detect the AT in these subjects. Furthermore, if the GET is to be used to monitor changes in aerobic fitness on an individual basis, then information on the reproducibility of the GET is required—in particular, how large the difference between two separate determinations of the GET performed in the same laboratory must be before one can be confident that a real change has occurred.21

We hypothesized that the GET could be used to provide an accurate estimate of the LT in patients with CF, even when lung disease is severe. The present study was therefore undertaken to determine if the onset of exercise-induced lactic acidosis during an incremental exercise test could be detected noninvasively by means of the GET in CF patients with more severe lung disease, thereby providing an easily determined index of aerobic fitness in this patient group. We also assessed, using the technique of Bland and Altman,21 the reproducibility of the GET in patients with CF.


   Materials and Methods
TOP
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Conclusion
 References
 
Patients with CF diagnosed on the basis of clinical history and abnormal sweat electrolyte measurements (chloride level > 60 mmol/L) were recruited from the National Referral Center for Adult Cystic Fibrosis, St. Vincent’s University Hospital. All patients were free of acute pulmonary exacerbations for at least 2 months. Nonsmoking subjects with no history of lung or cardiovascular disease were recruited as control subjects. The study protocol was approved by the Ethics Committee of St. Vincent’s University Hospital, and all subjects gave written informed consent.

Anthropometric measurements were undertaken, and spirometry was performed using the spirometry module of the Vmax 229 (Sensor Medics; Yorba Linda, CA). Predicted maximum voluntary ventilation (MVV) was estimated as 37.5 times FEV1.22 The CF subjects were classified into groups according to the severity of their lung disease as follows: mild CF, FEV1 > 70% of predicted; moderate CF, FEV1 > 40% and <= 70% of predicted; and severe CF, FEV1 <= 40% of predicted.

The cycle ergometer (Excalibur; Lode BV; Groningen, the Netherlands) was electromagnetically braked, and the subjects were able to select their own cadence, anywhere from 50 to 80 revolutions per minute. The exercise protocol consisted of 1 min of unloaded cycling, a 4-min warm-up period, and then a continuous work rate (WR) ramp to volitional exhaustion, selected for the CF patients according to stature, FEV1 percent predicted, and level of habitual physical activity. Bipolar ECG and ear lobe pulse oximetry (Sat-Trak; SensorMedics) were monitored and recorded throughout the test. Subjects wore a nose clip and a mouthpiece connected to a hot-wire anemometer for measurement of flow that was digitally integrated to obtain tidal volume. Oxygen and carbon dioxide were measured continuously using fast-response paramagnetic oxygen and nondispersive infrared carbon dioxide analyzers (Vmax 229; Sensor Medics). All signals were digitized and stored on a breath-by-breath basis for later analysis. Calibration was carried out before each test using a calibration syringe and precision oxygen and carbon dioxide mixtures.

Arterialized-venous blood was drawn (4 to 5 mL) over the last 10 s of each completed minute as previously described.23 In brief, a Teflon cannula was inserted into a vein on the dorsum of the hand, and an extension catheter and a three-way tap attached. The hand was then immersed in a water bath maintained at 42°C for the duration of the test, with 10 min allowed to elapse before sampling began. Patency was maintained by intermittent flushing with heparinized saline solution. Two milliliters of each sample were immediately aliquoted into glass tubes containing 12.5 mg of sodium fluoride and 10 mg of potassium oxalate (Gray Vacutainer; Becton Dickinson; Meylan, France) for the measurement of blood lactate concentration. The assay was performed on an automated spectrophotometric analyzer (Cobas Mira Bio; Roche; Basel, Switzerland) as previously described.24 An increase in blood lactate concentration of > 2 mmol/L above the resting value was considered to indicate that an exercise intensity above the LT had been achieved. The remainder of each sample was stored anaerobically on ice for measurement of arterialized-venous PCO2 (PavCO2) within 90 min using an automated, self-calibrating blood gas analyzer (Model IL 1312 Blood Gas Manager; Allied Instrumentation Laboratory UK; Cheshire, UK). This method of arterialized-venous blood yields PavCO2, pH, and blood lactate concentration values that are in close agreement with arterial values.23

Calculations
Peak WR was taken as the highest value achieved during the test. Peak E and peak O2 were calculated as the average of the 30 s breath-by-breath data immediately prior to the peak WR. Peak O2 was expressed as a percentage of predicted value,25 since the expression of these indexes per unit of body mass may be affected by nutritional status, particularly in CF patients with more severe lung disease. One patient with CF (FEV1 97% of predicted) exhibited intermittent breath holding at maximal exertion; therefore, in order to obtain more representative values for peak E and peak O2 in these patients, they were calculated on the basis of the highest values obtained from a seven-breath running average.

Threshold Determination
All thresholds were determined by a reviewer in a blinded manner using plots approximately 15 x 15 cm. In addition to not identifying the subjects, no information was given on corresponding threshold determinations. The LT was determined using the log-log transform method.26 Log(blood lactate concentration) is plotted against log(O2), and the point of inflection was chosen visually (Fig 1 , top, A). Using least-squares regression, a straight line was fitted to data points below and including the chosen inflection point, and a separate line to data points above and including the inflection point. The intersection of these two lines indicates the LT. The GET was determined manually using the modified V-slope method,14 and is illustrated in Figure 1 , bottom, B. CO2 is plotted against O2, and a line parallel to the line of identity is plotted through the lower data points. This maneuver identifies the point during the exercise test where CO2 begins to rise disproportionately faster than O2, due to the bicarbonate buffering of the lactic acidosis. Any inflection point occurring < 2 min into the WR ramp was rejected on the basis that it may represent a "pseudothreshold"27 due to transient changes in body carbon dioxide stores.28 29 The VentT was determined by identifying the point at which E/O2 began to increase after a period when they had been unchanging or progressively falling and without a concomitant increase in E/CO2.12



View larger version (13K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Top, A: Sample LT determination for one patient with CF (FEV1 52% of predicted) using the method of Beaver et al.26 Log(blood lactate concentration) is plotted against log(O2), and the point of inflection was chosen visually. Using least-squares regression, a straight line was fitted to data points below and including the chosen inflection point, and a separate line to data points above and including the inflection point. The intersection of these two lines indicates the LT. Bottom, B: Sample determination of the GET using the method of Sue et al14 for the same CF patient. CO2 is plotted against O2, and a line parallel to the line of identity is plotted through the lower data points. This maneuver identifies the point during the exercise test where CO2 begins to rise disproportionately faster than O2, due to the bicarbonate buffering of the lactic acidosis. Arrows indicate the selected GET.

 
Comparison of the Exercise Data at Physiologically Equivalent Workloads
In order to compare the relative impairment of gas exchange during exercise in the three CF groups, blood lactate concentration and PavCO2 data were compared at physiologically equivalent workloads: (1) rest, (2) 70% LT, (3) LT, (4) the last blood sample (peak exercise), (5) halfway between LT and the last blood sample (peak exercise). As workloads at 70% LT, LT, and halfway between LT and the last blood sample (peak exercise) were defined for each individual in terms of O2, they do not necessarily correspond to times when blood samples were obtained. Therefore, blood lactate concentration and PavCO2 were plotted against O2, and the value of these variables at the defined physiologically equivalent workloads were obtained by interpolation.

Reproducibility of the GET
The reproducibility of the GET in patients was assessed by means of two additional incremental exercise tests separated by 4 to 8 days as described above except without blood sampling. The control subjects underwent a second identical exercise test 1 week later (except one subject for whom the interval was 2 weeks). Each exercise test took place at the same time of day, and subjects were instructed not to change their diet or pattern of physical activity between tests. Anthropometric measurements were undertaken, spirometry was performed, peak exercise variables were calculated, and GET was determined in a blinded manner, as described above.

Statistical Tests
Agreement between the determined GETs and LTs was assessed using the technique of Bland and Altman.21 For each subject, the difference between the O2 at the GET and that at the LT is plotted against the average of the two values. The mean of all the differences was calculated. If the GET provided an unbiased estimate of the LT, then the mean of these differences would not have been significantly different from zero. This null hypothesis was tested using a t test. The limits of agreement were calculated by adding and subtracting from the mean value 1.96 times the SD of the differences. These limits indicate the 95% confidence interval (CI) for the size of the difference between the two methods of determining the AT. This provides an estimate of the maximum amount by which the GET may differ above or below the LT in 95% of determinations.

Statistical analysis between the groups was undertaken using analysis of variance in both single and repeated-measures forms. Where the variance increased with the size of the mean (E and blood lactate concentration), a logarithmic transformation was performed to normalize the distribution of the data. When the overall analysis of variance was statistically significant, the significance of the differences between specific means was assessed using Student Newman-Keul post hoc testing. Test-retest comparisons within groups were made using a paired t test.

The reproducibility of the GET was again assessed using the technique of Bland and Altman.21 For each subject, the GET test-retest difference was plotted against the average of the two values. If there was no systematic difference between the two tests undertaken on separate occasions (ie, the GET was reproducible), then the mean of the test-retest differences should not have been significantly different from zero. This hypothesis was tested using a t test. The limit of reproducibility, defined as the 95% CI for the differences (1.96 SD of the differences), was then calculated. This interval indicates the 95% CI for no systematic change in the GET. Conversely, in order to be confident at the 95% level that two separate determinations of the GET indicate a real change, the measured difference must exceed either the lower or upper limit. The reproducibility of peak O2 was also assessed in the same manner.


   Results
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Conclusion
 References
 
Profile of Subjects
In total, 36 patients with CF and 18 control subjects were recruited to one or both of the protocols involving the assessment of the agreement between the GET and the LT, and the assessment of the reproducibility of the GET. Six patients were not included in the analyses for a variety of reasons detailed below. In addition, one patient was included in the reproducibility part of the study and at a later stage underwent the agreement part, but was excluded from the analysis of the agreement between the GET and the LT. Anthropometric and pulmonary function data relating to the 18 control subjects and remaining 30 CF patients are shown in Table 1 . There were no significant differences between the mean age and height of the CF and control groups (Table 1) . The mean body mass index of the moderate and severe CF groups was significantly less than that of the control group and the mild CF group. As the CF patients were classified according to FEV1 percent predicted, and since other spirometric indexes correlate with this variable, no statistical analysis of the spirometric data presented in Table 1 was undertaken. Furthermore, there were no significant postexercise falls in FEV1 in any of the subjects.


View this table:
[in this window]
[in a new window]

  		Table 1. Anthropometric Data Classified by Disease Severity*

 
Data relating to the peak exercise capacity of each group are shown in Table 2 . Mean peak WR, peak E, peak O2, and peak O2 percent predicted were all significantly lower in all CF groups than in the control group, and also significantly lower in the severe CF group than in the mild CF group. Conversely, peak E expressed as a percentage of predicted MVV was systematically higher as lung function declined.


View this table:
[in this window]
[in a new window]

  		Table 2. Peak Exercise Data Classified by Disease Severity*

 
Agreement Between GET and LT
Altogether, 28 CF patients and 18 control subjects initially participated in the assessment of the agreement between the GET and LT. Data from a single CF subject (FEV1 71% of predicted) had to be excluded because of a technical problem with the measurement of breath-by-breath gas exchange. Two CF subjects (FEV1 39% and 72% of predicted, respectively) terminated the exercise test before the development of exercise-induced lactic acidosis and were also excluded. Two additional CF subjects (FEV1 80% and 88% of predicted, respectively), who both develop significant exercise-induced lactic acidosis, demonstrated a pseudothreshold and no GET could be identified; they too were not included in the analysis.

The O2 value at the determined GET was plotted against that at the LT for the control subjects (Fig 2 , top left, A) and for the patients with CF (Fig 2 , bottom left, B). The mean values of the GET and LT determined in each group declined with increasing severity of lung disease and are shown in Table 3 . Figure 2 , top right, C, and bottom right, D shows the difference between the O2 at the GET and that at the LT for each subject, plotted against the average of the two values for the control subjects and CF patients, respectively. For the control subjects, the mean difference (GET - LT) was - 40 mL/min, a value not significantly different from zero, and the limits of agreement were ± 550 mL/min. The mean difference for the CF subjects was + 10 mL/min, a value not significantly different from zero, and the limits of agreement were ± 410 mL/min.



View larger version (30K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Assessment of the agreement between the GET and the LT for the control subjects (top left, A, and top right, C) and the patients with CF (bottom left, B, and bottom right, D). • = control subjects (n= 18); {square} = mild CF (FEV1 > 70% of predicted; n = 6); {Delta} = moderate CF (FEV1 > 40% and <= 70% of predicted; n = 11); {diamond} = severe CF (FEV1 <= 40% of predicted; n = 6). Top left, A, and bottom left, C: O2 values at the determined GET plotted against that at the LT for the two groups. Top right, C, and bottom right, D: Difference between the O2 at the GET and that at the LT for each subject, plotted against the average of the two values. The mean of the differences, indicated by the solid line, is not significantly different from zero. Dashed lines indicate limits of agreement (ie, ± 1.96 SD of the differences; see text for details).

 

View this table:
[in this window]
[in a new window]

  		Table 3. Mean LT and GET Data Classified by Disease Severity*

 
There was no correlation between the individual GET - LT differences in the CF subjects and FEV1 percent predicted. In addition, none of the mean differences separately calculated for the mild, moderate, and severe CF groups ( + 60, + 10, and - 50 mL/min, respectively) were significantly different from zero.

Performance of the VentT
In many patients with CF, no increase in E/O2 was seen, and thus the VentT could not be identified. A typical plot illustrating the absence of a rise in E/O2 is shown in Figure 3 . A VentT was only detected in 11 of 23 patients with CF (48%) compared to 16 of 18 control subjects (89%). Furthermore, a VentT was not detected in any of the patients in the severe CF group. The high failure rate in the detection of the VentT in the patients with CF meant that agreement between the VentT and the LT could not be meaningfully assessed.



View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Sample plot of ventilatory equivalents for oxygen plotted against O2 for a patient with CF (FEV1 36% of predicted) in whom it was not possible to determine a VentT. • = E/O2; {circ} = E/CO2.

 
Comparison of the Exercise Data at Physiologically Equivalent Workloads
Comparisons of the exercise data at physiologically equivalent workloads are shown in Figure 4 . In all groups, mean blood lactate concentration values were elevated at peak exercise. The mean value in the control group was significantly greater than that in the severe CF group, but not significantly different from the other two groups. At peak exercise, the control group displayed a fall in PavCO2 below the value at the LT. There was no significant reduction in PavCO2 in the mild and moderate CF groups, while in the severe CF group, the mean PavCO2 rose significantly above the value at the LT. These data indicate a lack of respiratory compensation for the exercise-induced metabolic acidosis in the CF subjects.



View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Plot of blood lactate concentration (top) and PavCO2 (bottom) at physiologically equivalent workloads (see text for details of calculations). Data are shown as group means with error bars indicating SEM. • = control subjects (n = 18); {square} = mild CF (FEV1 > 70% of predicted; n = 6); {Delta} = moderate CF (FEV1 > 40% and <= 70% of predicted; n = 11); {diamond} = severe CF (FEV1 <= 40% of predicted; n = 6); * = significantly different from severe CF group (p < 0.05); {dagger} = significantly different from rest and all submaximal workloads (p < 0.05); {ddagger} = significantly different from all other groups (p < 0.05); § = significantly different from LT (p < 0.05); 50%Delta = halfway between LT and the last blood sample.

 
Reproducibility of the GET
Fourteen of the control subjects also participated in the assessment of the reproducibility of the GET. Fourteen CF patients were also incorporated into this part of the study, 5 of whom were included in the analysis of the agreement between the GET and the LT. One CF patient (FEV1 89% of predicted) was withdrawn from the study due to Pseudomonas cepacia infection, and data from one CF patient (FEV1 36% of predicted) had to be excluded because of a technical problem with the measurement of breath-by-breath gas exchange, leaving 12 CF patients included in the analysis of the reproducibility of the GET. In eight of the these CF subjects, the data were obtained during the course of a previously published study.30

Mean FEV1 and FVC were not different between the two occasions, in either the control or CF groups (data not shown), indicating that the CF patients were in clinically stable condition for the period of testing. Bland and Altman21 plots assessing the reproducibility of the GET in the control and CF groups are shown in Figure 5 . In both groups, the mean test2 - test1 differences, shown by the solid lines, were not statistically significantly different from zero. The reproducibility of peak O2 was also assessed, and the mean test2 - test1 differences for the control subjects and patients with CF were - 50 mL/min and + 80 mL/min, respectively, and were not statistically significantly different from zero. The limits of reproducibility were 520 mL/min and 350 mL/min, respectively.



View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. Assessment of the reproducibility of the GET for the control subjects (top, A) and CF patients (bottom, B). • = control subjects (n = 14); {square} = mild CF (FEV1 > 70% of predicted; n = 5); {Delta} = moderate CF (FEV1 > 40% and <= 70% of predicted; n = 6); {diamond} = severe CF (FEV1 <= 40% of predicted; n = 1). For each subject the test2 - test1 difference is plotted against the average of the two values. The means of the differences, indicated by the solid line, are not significantly different from zero. Dashed lines indicate limit of reproducibility (ie, ± 1.96 SD of the differences; see text for details).

 

   Discussion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Conclusion
 References
 
We have demonstrated that the noninvasive GET can be used to obtain an unbiased estimate of the LT in CF patients with mild, moderate, and severe lung disease, and therefore is a useful submaximal index of aerobic fitness. In contrast, the VentT cannot be used to reliably assess fitness in patients with CF. In addition, we have also shown that GET was reproducible in patients with CF.

Rate of Successful GET Determination
It is well recognized that inflection points in the plot of CO2 against O2 that occur very early (< 2 min) into an incremental exercise test frequently do not result from the onset of lactic acidosis15 27 ; rather, such inflection points may reflect the end of an initial rapid filling of body carbon dioxide stores, and have been termed pseudothresholds.27 In normal subjects, the LT is expected to occur beyond this point in the test, and the associated GET would be more readily apparent than any pseudothresholds. If, however, the onset of lactic acidosis occurs within this critical period, retention of the carbon dioxide produced by bicarbonate buffering prevents detection of the LT by gas exchange measurements. It is for this reason that the three patients who demonstrated an early inflection point were excluded, as previously recommended.15

In the present study, the frequency with which the GET could be detected in the CF patients (approximately 85%) was substantially higher than 31%31 and 60%14 in studies with COPD patients. There appear to be two major reasons for this greater success rate. First, in COPD patients, a common cause of failure to identify a GET is because patients do not have lactic acidosis develop before the exercise test is terminated.14 31 This is presumably because exercise is limited by dyspnea32 or other factors that cause the subject to stop prematurely. In contrast, in the present study, only two individuals (ie, approximately 7%) with FEV1 of 39% and 72% of predicted, respectively, did not have exerciseinduced lactic acidosis develop. Furthermore, in neither case was a valid GET identified.

The second major reason for failure to detect the GET in patients with COPD is the early onset of lactic acidosis15 at the time when body carbon dioxide stores are rapidly changing. In patients selected on the basis of their ability to have lactic acidosis develop, blood lactate concentration rose significantly before 2 min had elapsed in approximately 25% of cases. In the present study, the early onset of lactic acidosis was evident in only two CF patients (approximately 7%). Both of these patients had only mild abnormalities of lung function (FEV1, 80% and 88% of predicted, respectively), and therefore a WR increment of 15 W/min was selected. It is likely that use of a lower WR increment may have allowed successful GET identification in these two individuals.

Size of the Limits of Agreement
The sizes of the limits of agreement between GET and the LT were similar in the CF and control subjects (Fig 2) and similar to those that we have previously reported in CF patients with relatively mild lung disease.19 Furthermore, as lung function declined in the CF patients, there was no evidence of a failure of agreement between the GET and the LT (Fig 2 , bottom left, B and bottom right, D). This indicates that the GET can be used to noninvasively determine the AT, and hence provides a useful index of aerobic fitness in CF patients with a wide severity of lung disease.

In a report assessing the agreement between the GET and the LT in patients with documented myocardial infarction,33 the SD of the differences was reported, which gives an estimate of ± 300 mL/min for the size of the limits of agreement, and is comparable to those in the present study. In patients with COPD, after removal of those patients with pseudothresholds, the agreement between the GET and the inflection point on a log-log plot of standard bicarbonate vs O2, the limits of agreement were ± 180 mL/min.15 Although this value is somewhat smaller than the estimates in the present study, the two values are not significantly different.

The GET and LT are two independent methods of determining AT. The present study shows that the two methods agree with one another in both control subjects and CF patients. The variability inherent in each method limits the amount of agreement that is possible between the two methods.21 In normal subjects, we have found the reproducibility of the LT and GET to be identical (unpublished data), implying that the random variability of each method is equal. Therefore, neither method provides a more precise estimate of the AT. Thus exercise prescriptions based on the GET are as useful as those based on the LT. If a more precise estimate of AT is required, then repeated testing with either method could provide this. Obviously, the noninvasive nature of the GET permits repeated testing to be carried out easily and conveniently. While it must be recognized that any single GET determination may underestimate or overestimate the AT, our results suggest that regular repeated assessment of the GET will allow a more precise determination of AT. This will permit training intensity to be optimized throughout an exercise program.

Absence of VentT in CF
There was no evidence of respiratory compensation for the exercise-induced metabolic acidosis in the moderate and severe CF groups, as indicated by the absence of a fall in PavCO2 (Fig 4 , bottom). Indeed, the increase in PavCO2 above the LT in these two groups is indicative of a degree of alveolar hypoventilation leading to carbon dioxide retention. The failure to detect a VentT in more than half of the CF patients can therefore be attributed to the absence of a significant hyperventilatory response in many of these patients.

Reproducibility of the GET in CF
The CF patients were in clinically stable condition over the period of testing, as indicated by the absence of any change in spirometric indexes (Table 4 ). There are a small number of studies that have reported either the individual test results34 35 or SD of the differences36 37 38 39 for the LT, GET, or VentT. Similar to above, the results from these reports can be used to calculate limits of reproducibility. The results of this analysis yield estimates for the invasive and noninvasive thresholds ranging 200 to 580 mL/min, and are of a similar size to those obtained in the present study.


View this table:
[in this window]
[in a new window]

  		Table 4. Test-Retest Values of FEV1 and FVC*

 

   Conclusion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Conclusion
References
 
This study has demonstrated that in CF patients with mild, moderate, and severe lung disease, the GET may be used to identify the LT, providing inflection points occurring < 2 min into the WR ramp are rejected. In contrast, the VentT cannot be used to reliably assess fitness in these patients due to the absence of a respiratory compensation for the exercise-induced metabolic acidosis. The noninvasive nature of the GET makes it more acceptable to patients than the determination of the LT, which requires blood sampling. Furthermore, an assessment of the reproducibility of the GET indicated that the limits of reproducibility are similar in CF and control subjects.


   Footnotes
 
Abbreviations: AT = anaerobic threshold; CF = cystic fibrosis; CI = confidence interval; GET = gas exchange threshold; LT = lactate threshold; MVV = maximum voluntary ventilation; PavCO2 = arterialized-venous PCO2; CO2 = carbon dioxide output; E = minute ventilation; VentT = ventilatory threshold; E/CO2 = ventilatory equivalents for carbon dioxide; E/O2 = ventilatory equivalents for oxygen; O2 = oxygen uptake; WR = work rate

Supported by Health Research Board (Ireland) and Cystic Fibrosis Association (Ireland).

Work performed at Education and Research Centre, St. Vincent’s University Hospital, Dublin, Ireland.

Received for publication May 8, 2001. Accepted for publication January 17, 2002.


   References
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Conclusion
 References
 

  1. Nixon, PA, Orenstein, DM, Kelsey, SF, et al (1992) The prognostic value of exercise testing in patients with cystic fibrosis. N Engl J Med 327,1785-1788[Abstract]
  2. Schneiderman-Walker, J, Pollock, SL, Corey, M, et al (2000) A randomized controlled trial of a 3-year home exercise program in cystic fibrosis. J Pediatr 136,304-310[CrossRef][ISI][Medline]
  3. Baldwin, DR, Hill, AL, Peckham, DG, et al (1994) Effect of addition of exercise to chest physiotherapy on sputum expectoration and lung function in adults with cystic fibrosis. Respir Med 88,49-53[CrossRef][ISI][Medline]
  4. Orenstein, DM, Franklin, BA, Doershuk, CF, et al (1981) Exercise conditioning and cardiopulmonary fitness in cystic fibrosis: the effects of a three-month supervised running program. Chest 80,392-398[Abstract/Free Full Text]
  5. de Jong, W, Grevink, RG, Roorda, RJ, et al (1994) Effect of a home exercise training program in patients with cystic fibrosis. Chest 105,463-468[Abstract/Free Full Text]
  6. Boas, SR (1997) Exercise recommendations for individuals with cystic fibrosis. Sports Med 24,17-37[ISI][Medline]
  7. Orenstein, DM (1998) Exercise testing in cystic fibrosis. Pediatr Pulmonol 25,223-225[CrossRef][ISI][Medline]
  8. Webb, AK, Dodd, ME, Moorcroft, J (1995) Exercise and cystic fibrosis. J R Soc Med 88(Suppl 25),30-36
  9. Tanaka, K, Matsuura, Y (1984) Marathon performance, anaerobic threshold, and onset of blood lactate accumulation. J Appl Physiol 57,640-643[Abstract/Free Full Text]
  10. Sjodin, B, Jacobs, I (1981) Onset of blood lactate accumulation and marathon running performance. Int J Sports Med 2,23-26[ISI][Medline]
  11. Farrell, PA, Wilmore, JH, Coyle, EF, et al (1979) Plasma lactate accumulation and distance running performance. Med Sci Sports 11,338-344[ISI][Medline]
  12. Reinhard, U, Muller, PH, Schmulling, RM (1979) Determination of anaerobic threshold by the ventilation equivalent in normal individuals. Respiration 38,36-42[ISI][Medline]
  13. Beaver, WL, Wasserman, K, Whipp, BJ (1986) A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol 60,2020-2027[Abstract/Free Full Text]
  14. Sue, DY, Wasserman, K, Moricca, RB, et al (1988) Metabolic acidosis during exercise in patients with chronic obstructive pulmonary disease: use of the V-slope method for anaerobic threshold determination. Chest 94,931-938[Abstract/Free Full Text]
  15. Patessio, A, Casaburi, R, Carone, M, et al (1993) Comparison of gas exchange, lactate, and lactic acidosis thresholds in patients with chronic obstructive pulmonary disease. Am Rev Respir Dis 148,622-626[ISI][Medline]
  16. Godfrey, S, Mearns, M (1971) Pulmonary function and response to exercise in cystic fibrosis. Arch Dis Child 46,144-151
  17. Cerny, FJ, Pullano, TP, Cropp, GJ (1982) Cardiorespiratory adaptations to exercise in cystic fibrosis. Am Rev Respir Dis 126,217-220[ISI][Medline]
  18. Goldring, RM, Fishman, AP, Turino, GM, et al (1964) Pulmonary hypertension and cor pulmonale in cystic fibrosis of the pancreas. J Pediatr 65,501-524[CrossRef][ISI][Medline]
  19. McLoughlin, P, McKeogh, D, Byrne, P, et al (1997) Assessment of fitness in patients with cystic fibrosis and mild lung disease. Thorax 52,425-430[Abstract]
  20. Marcotte, JE, Grisdale, RK, Levison, H, et al (1986) Multiple factors limit exercise capacity in cystic fibrosis. Pediatr Pulmonol 2,274-281[ISI][Medline]
  21. Bland, JM, Altman, DG (1986) Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1,307-310[CrossRef][ISI][Medline]
  22. Carter, R, Peavler, M, Zinkgraf, S, et al (1987) Predicting maximal exercise ventilation in patients with chronic obstructive pulmonary disease. Chest 92,253-259[Abstract/Free Full Text]
  23. McLoughlin, P, Popham, P, Linton, RA, et al (1992) Use of arterialized venous blood sampling during incremental exercise tests. J Appl Physiol 73,937-940[Abstract/Free Full Text]
  24. Thin, AG, Hamzah, Z, FitzGerald, MX, et al (1999) Lactate determination in exercise testing using an electrochemical analyser: with or without blood lysis? Eur J Appl Physiol 79,155-159[CrossRef]
  25. Wasserman, K, Hansen, JE, Sue, DY, et al (1994) Principles of exercise testing and interpretation 2nd ed. ,112-118 Lea and Febiger Philadelphia, PA.
  26. Beaver, WL, Wasserman, K, Whipp, BJ (1985) Improved detection of lactate threshold during exercise using a log-log transformation. J Appl Physiol 59,1936-1940[Abstract/Free Full Text]
  27. Whipp, BJ, Lamarra, N, Ward, SA (1987) Required characteristics of pulmonary gas exchange dynamics for non-invasive determination of the anaerobic threshold. Benchetrit, G Baconnier, P Demongeot, J eds. Concepts and formulations in the control of breathing ,185-200 Manchester University Press Manchester, UK.
  28. Wasserman, K (1984) Coupling of external to internal respiration. Am Rev Respir Dis 129,S21-S24[ISI][Medline]
  29. Ozcelik, O, Ward, SA, Whipp, BJ (1999) Effect of altered body CO2 stores on pulmonary gas exchange dynamics during incremental exercise in humans. Exp Physiol 84,999-1011[Abstract]
  30. McKone, EF, Barry, SC, FitzGerald, MX, et al (1999) Reproducibility of maximal exercise ergometer testing in patients with cystic fibrosis. Chest 116,363-368[Abstract/Free Full Text]
  31. Belman, MJ, Epstein, LJ, Doornbos, D, et al (1992) Noninvasive determinations of the anaerobic threshold. Reliability and validity in patients with COPD. Chest 102,1028-1034[Abstract/Free Full Text]
  32. Punzal, PA, Ries, AL, Kaplan, RM, et al (1991) Maximum intensity exercise training in patients with chronic obstructive pulmonary disease. Chest 100,618-623[Abstract/Free Full Text]
  33. Dickstein, K, Barvik, S, Aarsland, T, et al (1990) A comparison of methodologies in detection of the anaerobic threshold. Circulation 81,II38-II46
  34. Hughson, RL, Weisiger, KH, Swanson, GD (1987) Blood lactate concentration increases as a continuous function in progressive exercise. J Appl Physiol 62,1975-1981[Abstract/Free Full Text]
  35. Cooper, CB, Beaver, WL, Cooper, DM, et al (1992) Factors affecting the components of the alveolar CO2 output-O2 uptake relationship during incremental exercise in man. Exp Physiol 77,51-64[Abstract]
  36. Meyer, K, Hajric, R, Westbrook, S, et al (1996) Ventilatory and lactate threshold determinations in healthy normals and cardiac patients: methodological problems. Eur J Appl Physiol 72,387-393[CrossRef]
  37. Bingisser, R, Kaplan, V, Scherer, T, et al (1997) Effect of training on repeatability of cardiopulmonary exercise performance in normal men and women. Med Sci Sports Exerc 29,1499-1504[ISI][Medline]
  38. Smith, DA, O’Donnell, TV (1984) The time course during 36 weeks’ endurance training of changes in O2 max. and anaerobic threshold as determined with a new computerized method. Clin Sci 67,229-236[Medline]
  39. Dickstein, K, Barvik, S, Aarsland, T, et al (1990) Validation of a computerized technique for detection of the gas exchange anaerobic threshold in cardiac disease. Am J Cardiol 66,1363-1367[CrossRef][ISI][Medline]



This article has been cited by other articles:


Home page
 ChestHome page
J. E. Hansen, X.-G. Sun, Y. Yasunobu, R. P. Garafano, G. Gates, R. J. Barst, and K. Wasserman
Reproducibility of Cardiopulmonary Exercise Measurements in Patients With Pulmonary Arterial Hypertension
Chest, September 1, 2004; 126(3): 816 - 824.
[Abstract] [Full Text] [PDF]

Home page
 ChestHome page
W. P. Sexauer, H.-K. Cheng, and S. B. Fiel
Utility of the Breathing Reserve Index at the Anaerobic Threshold in Determining Ventilatory-Limited Exercise in Adult Cystic Fibrosis Patients
Chest, October 1, 2003; 124(4): 1469 - 1475.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Thin, A. G.
Right arrow Articles by McLoughlin, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Thin, A. G.
Right arrow Articles by McLoughlin, P.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS