| 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 |
|||||||||
Queen's University Kingston, ON, Canada
To the Editor:
In their recent article, Murariu and colleagues (October 1998)1 report that exercise performance in patients with obstructive lung disease is likely linked to the degree of resting lung hyperinflation as indirectly measured by inspiratory capacity (IC). Certainly, there is increasing evidence that dynamic lung hyperinflation during exercise (as assessed by serial IC measurements) contributes to the intensity and quality of dyspnea, and thus exercise limitation in COPD.2 ,3 ,4 Furthermore, increases in dynamic IC have been shown to correlate well with improvement in dyspnea and exercise endurance following pharmacologic2 ,3 and surgical volume reduction.5 The study by Murariu and colleagues is, however, the first to claim an association between the resting IC and exercise performance in patients with obstructive lung disease (COPD and asthma).
The authors report a correlation coefficient of 0.81
(p < 0.001) between maximal exercise power output (
max) and
IC in 25 patients with mild COPD (FEV1, 0.68 ± 21) who
had, on average, moderate exertional dyspnea (Borg 4 ± 1) and
exercise curtailment max (62 ± 29% predicted). There is,
however, a significant confounding factor here that the authors did not
correct for: the absolute max and the IC would be expected to be
linearly related, even in normal subjects. The predicted IC and
max are both functions of age, sex, and height. Simply stated,
taller, younger male subjects would be expected to have high max
and IC, while shorter, older female subjects would have lower
max and IC. In fact, predicted max is a linear equation
of age, sex, and height.6
This explains the significant
correlation between height and max as reported by Murariu and
colleagues (correlation coefficient, 0.61; p < 0.005).
To establish a causal relationship between reduced IC and diminished
max, both IC and max should be normalized and expressed
as percent predicted. In their article, Murariu and coworkers provided
us with the respiratory function and max data (reported as
percent predicted) for all study subjects. When viewed in this way, the
relationship is not as strong (Fig 1
).
In fact, the correlation coefficient is only 0.294 (p = 0.154).
|
Correspondence to: Marc Newton, MD, Division of Respiratory and Critical Care Medicine, Queen's University, 102 Stuart St, Kingston, ON, Canada; e-mail: fmnewton@istar.ca
References
Meakins-Christie Laboratories McGill University Montreal, QC, Canada
To the Editor:
In normal subjects, there is a close association between maximal
exercise power output (max) and vital capacity, the variation
coefficient (r2) amounting to 0.79.1-1
Nevertheless, in our patients with obstructive lung disease (OLD), the
variation coefficient of max to FVC was substantially weaker
than that of inspiratory capacity (IC) (0.41 vs 0.66). Furthermore, in
21 patients with restrictive lung disease (RLD) who were included in
our original manuscript, there was no significant correlation of
max to IC, though their exertional dyspnea and exercise
curtailment were similar to those of the OLD patients. Thus, in OLD but
not RLD patients, the resting IC is a better predictor of max
than either FVC or FEV1.
Although we argue that there is predictably a close association between
max and IC in OLD patients because of expiratory flow
limitation, we do not state that "a particular patient's exercise
limitation can be strongly predicted by IC." In fact, the
r2 between max and IC was only 0.66, reflecting
substantial scatter of the data.
Prediction of max and IC is a nettlesome problem in old
individuals, such as our OLD population. In fact, there are many
predictive equations for max, each giving a different
value.1-1
Furthermore, the predicted normal values of IC are
obtained as the difference between predicted total lung capacity
and FRC. Thus, assessment of the relationship of max (%
predicted) to IC (% predicted) may be problematic, particularly in
small numbers of patients with moderate exercise limitation. In 34 OLD
patients with a more marked decrease in max (39% predicted on
average), a significant correlation was found both between the absolute
and the predicted values of max and IC, with IC being the only
significant contributor to max (O. Diaz Patiño, MD, C.
Lisboa, MD, J. Milic-Emili, MD; unpublished observations, July
1998).
In conclusion, we think that resting IC, the Cinderella of lung
function testing, not only provides useful information about the
effects of bronchodilators1-2
and surgical
treatment1-3
on hyperinflation in OLD patients, but it also
gives a useful estimate of a patient's exercise capacity. If the
actual max in a given OLD patient is lower than that predicted
by our regression equation, it is likely that the patient should
benefit substantially by exercise rehabilitation because the exercise
performance is probably mostly limited by mechanisms other than lung
function impairment.
Correspondence to: Joseph Milic-Emili, MD, Meakins Christie Laboratories, McGill University, 3626 St. Urbain Street, Montreal QC H2X 2P2, Canada; e-mail: milic@meakins.lan.mcgill.ca
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
