| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Pulmonary Physiology Laboratory, Veterans Affairs Medical Center; and Department of Medicine, George Washington University, Washington, DC.
Correspondence to: Ann E. Medinger MD, FCCP, 5605 Park St, Chevy Chase, MD 20815
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Design: Retrospective analysis of consecutive patients with sarcoidosis referred to the Pulmonary Physiology Laboratory between 1992 and 1997, who completed at least 6 min of progressive bicycle exercise. Resting and exercise pulmonary function measurements were compared to radiographic stage of disease.
Setting: Pulmonary Physiology Laboratory at Washington, DC, Veterans Affairs Medical Center.
Patients: Forty-eight outpatient veterans with biopsy specimen-proven sarcoidosis.
Results: Across all radiographic stages of sarcoidosis, total lung capacity, resting diffusing capacity, and exercise gas exchange measurements had a significant variance with radiographic stage. Across the early radiographic stage disease (stages 0 to 2), the change in alveolar-arterial oxygen pressure gradient between rest and exercise, normalized for oxygen uptake, was the most significant measurement in its variation with radiographic stage.
Conclusions: Changes in gas exchange with exercise may be the most sensitive physiologic measurements to assess the extent of disease in early radiographic stages of sarcoidosis.
Key Words: clinical exercise testing • extent of disease • pulmonary function testing • radiographic stage • sarcoidosis
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Sensitive and accurate diagnostic tools are needed to measure the extent of pulmonary parenchymal disease and to detect progressive functional impairment over the course of the disease. The plain chest radiograph is often used for surveillance because of its low cost and ready availability in the community. Pulmonary function measurements are also readily available and relatively inexpensive. In patients with early radiographic stages of sarcoidosis, resting pulmonary function studies have a better correlation with lung parenchymal histology than does the chest radiograph.1 2 Pulmonary parenchymal disease may be present without appearing on the chest radiograph. The diffusing capacity of the lung for carbon monoxide (DLCO) has been found to be the most sensitive of the resting physiologic measurements.1 2 Although resting lung function measurements have not been shown to be highly sensitive for detecting lung disease in patients with radiographic stage 0 to 1 sarcoidosis, patients with normal resting pulmonary function have less active inflammation and fibrosis observed on lung biopsy specimens.1
Looking at patients with a variety of interstitial lung diseases and
comparing resting and exercise pulmonary physiologic measurements to
histologic findings on lung biopsy specimens, Fulmer and
Crystal3
found that the measurements correlating best with
histopathologic changes in the lung biopsy specimen were the change in
oxygen tension (
PO2) and the
change in the alveolar-arterial oxygen pressure gradient between rest
and exercise (P[A-a]O2).3
Because the lesions of sarcoidosis favor the upper lobes of the lung,
physiologic lung function measured during exercise may have a higher
sensitivity for detecting the presence and progression of parenchymal
pulmonary disease. At rest, the apical regions of the lung contribute
relatively little to ventilation and gas exchange. During exercise,
apical pulmonary circulation is normally recruited and contributes
significantly to the capacity of the lung to increase the rate of gas
exchange. Hence, the regional gas exchange abnormalities of sarcoidosis
should be more easily detected during exercise than at rest,
particularly in early-stage disease. Several other
investigators4
5
have measured responses to low-level
exercise in patients with sarcoidosis; they have reported a lower
frequency of exercise gas exchange abnormalities in sarcoidosis than in
patients with other chronic interstitial lung diseases. Using
progressive symptom-limited exercise (SLE) testing to evaluate patients
with sarcoidosis with normal or mildly abnormal spirometry results,
Sietsema and Kraft6
found abnormal exercise measurements
with low peak oxygen uptake
(
O2max) and low anaerobic
threshold (AT). They concluded that many of these patients had
unsuspected cardiac circulatory impairment.6
Miller and
Brown7
studied a similar group of patients with
sarcoidosis who had normal spirometry findings, and found that those
with abnormal resting DLCO results were more likely to have
abnormal exercise gas exchange.7
To further evaluate the
use of progressive SLE for assessment of extent of lung disease in
patients with sarcoidosis, we studied veterans with sarcoidosis who
were referred to the Pulmonary Physiology Laboratory at the Washington,
DC, Veterans Affairs Medical Center for progressive SLE testing.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Prior to exercise, each patient underwent the following resting pulmonary function measurements: air flow (FEV1/FVC), plethysmographic lung volume (total lung capacity [TLC]), single-breath DLCO, and maximum voluntary ventilation (MVV) (6200 Autobox; SensorMedics; Yorba Linda, CA). Measurements were made according to American Thoracic Society standards for equipment and patient performance.9 10 MVV was performed by coaching the individual to hyperventilate as vigorously as possible for 10 s, aiming for a minimum frequency of 80 breaths/min and measuring ventilatory output as liters per minute. Each individual then completed a minimum of 6 min of exercise, pedaling a bicycle ergometer (Ergoline 800; SensorMedics) with progressively increasing work rate (WR), pedaled at 60 revolutions per minute plus or minus 5 revolutions per minute. The WR increment for each ramped exercise test was individualized on the basis of each patient’s pretest activity level (range, 10 to 25 W/min), with the objective of achieving 6 to 12 min of progressive exercise before stopping. Serial exercise measurements included 12-lead ECG (Marquette Max 1; Marquette Electronics; Milwaukee, WI), BP (Hewlett Packard 7000; Hewlett Packard; Andover, MA), WR (Ergoline 800; Sensormedics ), tidal oxygen and carbon dioxide tensions, tidal airflow, and respiratory rate (2900C; SensorMedics).
Arterial blood gas levels were measured at rest, at peak performance,
and at 2 min postexercise, from an indwelling radial arterial line;
initial hemoglobin and carboxyhemoglobin (COHb) were recorded (analysis
on Radiometer 520 Radiometer Medical; Brønshøj, Denmark). The
change in oxygen consumption
(O2) was calculated as the
difference in oxygen uptake between peak exercise and resting
measurements of oxygen consumption.
PO2 was calculated as the
difference in arterial oxygen tension between peak exercise and resting
measurements. To normalize PO2 for
work performed during exercise, we also calculated
PO2 as a ratio of the
O2
(PO2/O2).
An alveolar-arterial oxygen pressure gradient
(P[A-a]O2) was calculated for each arterial
blood sample using measured respiratory quotients; they were expressed
as the difference between peak exercise and resting measurements
(P[A-a]O2) and normalized for
O2 achieved during exercise
(P[A-a]O2/O2).
Standard exercise calculations included
O2max, peak exercise expired
minute ventilation, AT, and ventilatory equivalent for carbon dioxide
at AT (VE/VCO2
AT).11
AT was determined by V-slope method, and reported
as percentage of predicted
O2max.12
Breathing reserve (BR) was calculated as the difference between
measured resting MVV and the peak exercise ventilation, expressed as a
percentage of MVV and normalized for the achieved oxygen uptake. Heart
rate reserve (HRR) was calculated as the difference between peak and
resting heart rate, normalized for achieved oxygen uptake. The age-,
gender-, and weight-dependent normal predicted
O2max was derived from Hansen
et al,13
and reported as percentage of predicted
O2max. Normal ranges for
P(A-a)O2/O2
and
PO2/O2
were not available to us. Predicted values for resting pulmonary
function were derived as follows: TLC from Goldman and
Becklake,14
FEV1/FVC from Crapo et
al,15
and single-breath DLCO from Miller et
al.16
Predicted resting lung volume reference values were
adjusted for African American individuals.17
Measured
single-breath DLCO was adjusted for abnormal hemoglobin and
COHb concentrations.18
The FEV1/FVC
percentage was reported as the difference from predicted percentage
(FEV1/FVC%). The physiologic measurements
were analyzed for variation with radiographic stage employing
descriptive statistics, one-way analysis of variance (ANOVA) for each
factor in relation to radiographic stage in the full data set, and
analysis of covariance for evaluating single-breath DLCO
and exercise gas exchange in relation to radiographic stage in those
subjects with complete data sets
(P[A-a]O2/O2
as dependent variable, stage as factor, and percentage of predicted
single-breath DLCO as covariate). Exercise gas exchange
measurements were also correlated with resting DLCO in
patients with complete data sets, using simple linear regression
calculations. The statistical package employed was the Microsoft Excel
5.0 Analysis Toolpak program (Gray Matter International; Cambridge, MA)
and SAS version 6.12 (SAS Institute; Cary, NC).
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
FEV1/FVC%,
O2max, AT, HRR, BR, or
VE/VCO2 AT. A significant
association was found between all radiographic stages and TLC,
DLCO, P(A-a)O2, and
PO2. Across all radiographic
stages of sarcoidosis, the most significant association was found
between radiographic stage and the
PO2 with exercise (Table 1)
. For
patients with radiographic stage 0 to 2 disease, the physiologic
measurements most significantly associated with radiographic stage were
the P(A-a)O2 and
P(A-a)O2/O2
(Table 1)
. In this group with less extensive disease by chest
radiographic criteria, TLC did not vary significantly with radiographic
stage and the DLCO had a weaker association than it had
across all radiographic stages. Excluding from analysis the patients
with COHb evidence of antecedent smoking did not significantly alter
these findings, nor did elimination of the patient receiving calcium
channel-blocking medication. The 30 individuals who had complete data
sets of both resting and exercise measurements had the same highly
significant association of
P(A-a)O2/O2
with radiographic stage as did the whole group (p < 0.000331 for
stages 0 to 2; p < 0.000967 for stages 0 to 4). For this group of 30
patients, the R2
coefficient of determination for single-breath DLCO and
PO2 was 0.52 for all stages and
0.33 for those with stage 0 to 2 disease (Fig 1
). The R2 coefficient for single-breath
DLCO and P(A-a)O2 was
0.58 for all radiographic stages and 0.47 for stages 0 to 2 (Fig 2
). The best correlation of exercise measurements with
DLCO was
P(A-a)O2/O2,
where R2 was 0.78 for all radiographic
stages and 0.66 for stages 0 to 2 (Fig 3 ). Further analysis of the 30 patients with matched data across all
stages, considering
P(A-a)O2/O2
as dependent variable, radiographic stage as a factor and percentage of
predicted single-breath DLCO as covariate also
gave a statistically significant model (p = 0.0001). Both variables
together accounted for 71.7% of the variation in
P(A-a)O2/O2.
There was no improvement in prediction of exercise gas exchange gained
by factoring in the radiographic stage to the percent of
DLCO.
|
|
|
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
At high levels of exercise performance, the normal individual experiences a fall in mixed venous oxygen tension, decrease in pulmonary capillary transit time, and rise in P(A-a)O2. However, overall ventilation/perfusion matching improves so that both the physiologic shunt fraction and dead space to tidal volume ratio improve. This is achieved through recruitment of normally closed pulmonary vascular channels, increasing pulmonary capillary surface area, increasing the rate of gas exchange commensurate with increased blood flow through the lung, and increased ventilation. Hence, while oxygen consumption rises dramatically during exercise in normal individuals, arterial oxygen tension remains relatively unchanged.
Arterial blood gas levels are a good measure of gas exchange across the lung, and a normal P(A-a)O2 reflects normal ventilation/perfusion function of the lung. Abnormalities in P(A-a)O2 are specific for lung disease or less commonly right-to-left cardiovascular shunt. The patient with interstitial lung disease has fewer pulmonary vascular channels and hence a smaller reserve pulmonary vascular bed is available for recruitment to increase the lung surface area for gas exchange during exercise. Hence, with progressively increasing workload and increased peripheral oxygen uptake, both mixed venous oxygen tension and arterial oxygen tension fall. The degree of fall in arterial oxygen tension depends on both the extent of interstitial lung disease and the amount of work performed. Patients with interstitial lung disease may be found to have a normal peak exercise arterial oxygen tension when they have minimal disease or when they perform a low level of exercise during the test. Studies in patients with sarcoidosis, which have shown a poor correlation of exercise measurements with other measures of disease extent and progression, have tested the patients at low levels of exercise.4 5 Because peak WR with exercise testing is a subjective end point that can be influenced by the coach-technician and the patient’s own willpower, all measurements obtained at peak WR have a subjective element. This may confound their correlation with extent of pulmonary disease. Normalizing peak exercise cardiac and gas exchange measurements for achieved oxygen uptake helps remove this subjective bias.
In normal individuals, the
O2 with exercise has a
constant linear relationship to total work performed
(O2/WR). This makes change in
oxygen uptake between rest and exercise
(O2) generally a good
measurement of the level of exercise performed. However, both
O2max and
O2 with exercise are
determined by multiple factors: gas exchange across the lung, oxygen
content of blood, oxygen delivery to tissues, and oxygen uptake in the
tissues, in addition to the subjective coaching and patient’s
willingness to work hard during the test. The peak WR achieved is
influenced by body weight in addition to the above-named factors.
Impairment in oxygen flow can alter the slope of
O2/WR and affect our ability
to normalize the exercise blood gas measurements by using delta
O2. Thus, when significant
heart, peripheral vascular, or metabolic muscle disease is present, use
of the O2 to normalize the
P(A-a)O2 for work achieved during testing may
decrease its sensitivity and specificity for measuring extent of lung
disease.
In this study, radiographic and physiologic measurements and interpretations were performed independent of each other. To achieve consistency of data analysis, chest images and physiologic measurements were interpreted by one individual or pair of individuals. Analysis of the results shows a significant association of exercise gas exchange with radiographic stage in patients with sarcoidosis who can perform at least 6 min of progressive exercise. The study has the limitations of any retrospective analysis, including the heterogeneity of medications and the antecedent cigarette smoking used by patients. Some substances taken prior to exercise testing can alter exercise test measurements and reduce their validity (such as calcium channel blockers and cigarette smoking). When we eliminated the patients with these factors from the analysis, we found no change in the significance of the associations between physiologic and radiographic measurements.
Although clinical evidence of myocardial involvement is only present in about 5% of patients with sarcoidosis, unsuspected cardiac involvement may be found in as many as 10 to 15% of patients with sarcoidosis.6 19 20 Sarcoidosis patients with chronic respiratory failure commonly develop cor pulmonale. Patients with radiographic stage 0 to 2 disease are less likely than stage 3 to 4 patients to have cardiac impairment from cor pulmonale. Hence, exercise measurements may be a more specific reflection of extent of pulmonary disease in the radiographic stage 0 to 2 patients with sarcoidosis.
In this study, the correlation of exercise gas exchange with
radiographic stage was less significant when stage 3 to 4 patients were
included in the analysis. Although we had no antecedent clinical
evidence of heart disease in any of our patients, 5 of the 48 patients
had exercise evidence of an oxygen flow problem (3 stage 2 patients; 2
stage 4 patients). These individuals were younger than average for the
whole group, and showed good effort during the test; they achieved a
peak heart rate within 15 beats of the maximum predicted or had a very
low BR at peak performance with a peak
O2 < 84% of predicted. All
reached AT at an oxygen uptake < 40% of predicted
O2max. All had a low oxygen
pulse, and three patients had a low slope of
O2/WR. No patients developed
arrhythmias or ischemic ECG changes during exercise. One radiographic
stage 4 patient stopped the exercise test because of leg pain. In the
other stage 4 patient, PO2 fell from
97 to 51 mm Hg at peak exercise, suggesting that cor pulmonale was
likely to be present. No significant change in the peak exercise
PO2 and no leg pain was noted in the
stage 2 patients with exercise evidence of oxygen flow problems. An
echocardiogram was normal in one of the stage 2 patients, but no other
cardiac imaging was available to further evaluate right heart function
or identify myocardial sarcoidosis. When we excluded from the analysis
the patients with exercise evidence of an oxygen flow problem, the
significance of the association between radiographic stage and
physiologic measurements did not change.
Other authors7
have emphasized the value of the
single-breath DLCO to measure extent of disease in
sarcoidosis. Our measurements confirm the value of single-breath
DLCO as a measurement of resting pulmonary function that
varies significantly with radiographic stage in sarcoidosis. However,
in our patients with radiographic stage 0 to 2 sarcoidosis, we found
the stage to be more significantly associated with exercise gas
exchange measurements than with single-breath DLCO
(p < 0.0005 for P[A-a]O2 vs p < 0.039
for DLCO). The exercise gas exchange measurement was even
more discriminating among radiographic stages 0 to 2 when normalized
for work performed during testing (p < 0.0000493 for
P[A-a]O2/O2).
Figures 4 ,
5
show the relative changes in percentage of predicted single-breath
DLCO and
P(A-a)O2/O2
across radiographic stages 0 to 2. The 95% confidence intervals of the
percentage of predicted single-breath DLCO overlap for all
radiographic stages. This suggests that the percentage of predicted
single-breath DLCO is not able to separate patients with
and without radiographic evidence of pulmonary parenchymal disease (Fig 4)
. The 95% confidence intervals for
P(A-a)O2/O2
distinguish patients with radiographic evidence of pulmonary
parenchymal disease (stage 2) from those without (stages 0 to 1; Fig 5
). The clinician is most concerned to have a sensitive test for
quantifying extent of disease in asymptomatic patients with sarcoidosis
who have radiographic stage 0 to 2, so that progression can be detected
on sequential measurements. One seeks to withhold treatment from those
who are remitting spontaneously but initiate with conviction in those
who are progressing.
P(A-a)O2/O2
may be more sensitive than single-breath DLCO for
monitoring disease extent and progression in early radiographic stages
of sarcoidosis.
|
|
P(A-a)O2/VO2
is needed. Normal ranges and reproducibility need to be established.
These may enable us to use this measurement to detect and quantify lung
impairment with greater sensitivity in patients with early radiographic
stages of sarcoidosis. Sequential physiologic measurements are most
useful in the management of patients with sarcoidosis, and additional
study is needed to follow changes in
P(A-a)O2/VO2
over the course of the disease. We need to correlate changes in this
measurement with other longitudinal markers of disease extent and
activity, with and without treatment. We have found the
P(A-a)O2/VO2
to have a promising correlation with radiographic stage. Further
investigation is needed to determine whether we can use it as a
reliable marker of the extent of disease in patients with pulmonary
sarcoidosis.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
FEV1/FVC% = difference from
predicted percentage FEV1/FVC; HRR = heart rate reserve;
MDI = metered-dose inhaler; MVV = maximum voluntary ventilation;
P(A-a)O2 = change in alveolar-arterial oxygen pressure
gradient between rest and exercise;
PO2 = change in oxygen tension;
P(A-a)O2 = alveolar-arterial oxygen pressure gradient;
SLE = symptom-limited exercise; TLC = total lung capacity;
VE/VCO2 AT = ventilatory
equivalent for carbon dioxide at anaerobic threshold;
O2 = change in oxygen consumption;
O2max = peak oxygen uptake;
WR = work rate Received for publication March 22, 2000. Accepted for publication December 28, 2000.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. P. Baughman, B. K. Sparkman, and E. E. Lower Six-Minute Walk Test and Health Status Assessment in Sarcoidosis Chest, July 1, 2007; 132(1): 207 - 213. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M A Spruit, M J Thomeer, R Gosselink, T Troosters, A Kasran, A J T Debrock, M G Demedts, and M Decramer Skeletal muscle weakness in patients with sarcoidosis and its relationship with exercise intolerance and reduced health status Thorax, January 1, 2005; 60(1): 32 - 38. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Sette, J. A. Neder, L. E. Nery, J. Kavakama, R. T. Rodrigues, M. Terra-Filho, S. Guimaraes, E. Bagatin, and N. Muller Thin-Section CT Abnormalities and Pulmonary Gas Exchange Impairment in Workers Exposed to Asbestos Radiology, July 1, 2004; 232(1): 66 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Lamberto, H. Nunes, P. Le Toumelin, F. Duperron, D. Valeyre, and C. Clerici Membrane and Capillary Blood Components of Diffusion Capacity of the Lung for Carbon Monoxide in Pulmonary Sarcoidosis: Relation to Exercise Gas Exchange Chest, June 1, 2004; 125(6): 2061 - 2068. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X.-G. Sun, J. E. Hansen, N. Garatachea, T. W. Storer, and K. Wasserman Ventilatory Efficiency during Exercise in Healthy Subjects Am. J. Respir. Crit. Care Med., December 1, 2002; 166(11): 1443 - 1448. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. Hansen, A. E. Medinger, S. Khouri, and P. K. Rohatgi Sarcoidosis and Gas Exchange Measures Chest, March 1, 2002; 121(3): 1004 - 1005. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
