Detecting the Threshold of Anaerobic

Metabolism in Cardiac Patients

During Exercise*
KARLMAN WASSERMAN,M.D.~ and MALCOLM B. MCILROY, M.D.$

Palo Alto and San Francisco, California

H
ILL ET AL.l have shown that large quantities While Harrison and Pilcher did not detect
of lactic acid form when muscles lack significant increases in the gas exchange ratio
oxygen. Formation of lactic acid during exer- (R)$ of their normal subjects at the low loads
cise permits oxidation to proceed anaerobically they studied, metabolic acidosis does develop in
and accounts for most of the oxygen debt which normal subjects during work of greater intensity
accumulates during heavy exercise.2J Conse- and R increases.‘j-8
quently, measurement of accumulated lactate The onset of anaerobic metabolism during
has been used as an index of anaerobic metab- exercise can thus be detected in three ways:
olism.4s5 Because of the relatively high dis- (1) as an increase in the lactate concentration in
sociation constant of lactic acid, virtually all of blood, (2) as a decrease in arterial blood bicar-
it is ionized in the physiologic range of pH. bonate and pH and (3) as an increase in the
Therefore, when lactic acid is produced during respiratory gas exchange ratio (R). In evaluat-
anaerobic metabolism, it is completely buffered ing cardiovascular performance during work, a
in the blood, and the level of bicarbonate is method which measures R has the advantage of
reduced,6-8 as in other metabolic acidoses. avoiding blood sampling. Furthermore, equip-
Pilcher and associates9 have shown that patients ment which is available in many cardiopul-
in acute heart failure have arterial pH values monary laboratories can be used to measure R
that are extraordinarily low, and Huckabeelo breath by breath. Thus, it is possible for the
has reported an increased concentration of lactate examiner to detect the threshold of anaerobic
in the blood of patients with circulatory failure. metabolism during the work test and to avoid
Harrison and Pilcher” found that patients exhaustive and potentially dangerous exercise of
with heart failure produced more carbon dioxide patients with heart disease under study. A
(CO*) during exercise than normal subjects method for the evaluation of cardiac perform-
performing the same exercise. Consequently, ance by analysis of R breath by breath is re-
their gas exchange ratio (CO2 production to 02 viewed in this paper.
consumption) was increased. They concluded
that the excess CO2 could not be the end product METHODS
of aerobic metabolism and must be evolved from
CO2 stores. They reasoned that the excess CO2 The method used has been described by Naimark
was released from bicarbonate when acids formed and associates.’ It involves a standardized exercise
during anaerobic metabolism were buffered : test in which the subject pedals an ergometer or walks
on a treadmill for four minutes at each of several
TISSUES BLOOD LUNGS
I 02 c--- __ graded workioads. Heart rate, minute ventilation,
02 t-1 02
1
+
oxygen consumpti,on and end-tidal CO2 and Nt con-
Metabolic j centrations are measured. The respiratory gas ex-
change ratio can be calcuiated from the end-tidal gas
--t Lactic acid +
NaHCOg + Na lactate
f Hz0 + COZ-,+ CO2 (excess) $R = CO2 production/Oz consumption = +I,,,/
p con --,+ co* eo? = pulmonary respiratory quotient.
* From the Pulmonary Function Laboratory, Department of Medicine, Stanford University School of Medicine,
Palo Alto, t and the Cardiovascular Research Institute, University of California Medical Center, San Francisco, Calif. 1
This investigation was supported in whole by U. S. Public Health Service Research Grants HE 06591 and HE 06285
from the National Institutes of Health.

844 THE AMERICANJOURNAL OF CARDIOLOOY

 Anaerobic Metabolism During Exercise 845

level of oxygen consumption at which anaerobic

metabolism becomes important, and we call this
the “threshold” of anaerobic metabolism. It
corresponds to the point at which the concentra-
tion of bicarbonate in the arterial blood de-
creases and the concentration of lactate rises.‘**
Naimark et al. 7 found the threshold of anaerobic
metabolism to be at a level of oxygen consump-
tion of about 1.2 L./min. in normal sedentary
men and about 0.8 in normal sedentary women.
They also found that R increased at lower levels
of work in patients with heart disease.
The results of exercise tests in which the
threshold of anaerobic metabolism was meas-
F’N2 1.0 R ured in 37 patients with heart disease are shown
1.1
in Figure 4. The level of oxygen consumption
at which the A R-Go, curve showed the steepest
- ,.o- 1.2
slope (anaerobic threshold) has been plotted in
patients with different types of heart disease.
-P.O- In addition, the functional limitation of the
13
patient is noted according to the New York
Heart Association classification. The relation
-X0-
20
between the threshold of anaerobic metabolism
and the “wedge” pressures at rest are plotted in
patients with mitral stenosis. In these patients
3.0 40 5:0 do <O do the discrepancies between the level of “wedge”
EN0 TIDALCOP W pressure and the severity of symptoms are as
FIG. 1. Nomogram for determining R from end-tidal NZ numerous as those between the threshold of
and COZ. FINKequals inspired NZconcentration. anaerobic metabolism and symptoms. This is
not surprising in view of the multiplicity of
factors determining exercise tolerance. In the
concentrations, using the following equation:
other groups, only the severity of symptoms is
FACO~ shown. In this limited series, the patients with
R=
1.26 FAN2 - 1 + FAN2 the less severe symptoms had higher thresholds
of anaerobic metabolism.
where
F,COz = the end-tidal COZ concentration and ILLUSTRATIVE CASES
FAN2 = the end-tidal N2 concentration.
The clinical usefulnessof determining the threshold
This equation is derived from the alveolar ventilation of anaerobic metabolism is illustrated by the following
and alveolar gas equations.7J2 R can be determined case histories.
quickly by reference to a nomogram calculated from
CASE 1. Acromegaly. A 25 year old male insurance
this equation (Fig. 1). salesman was admitted to Stanford University Hos-
pital, Palo Alto in May 1962 for evaluation of cardio-
RESULTS
megaly. His past history revealed that he was the
An example of the tracings of Nz, COs, heart product of a normal delivery and had frequent colds
rate and minute ventilation for the last 30 sec- in infancy. At the age of 14 months he was admitted
onds of exercise at each workload during the test to the hospital for the evaluation of hirsutism. At
is shown in Figure 2. The values of R were the age of 3 years he had a cardiac arrest during a
determined from the nomogram. It will be tonsillectomy and required artificial respiration. At
7 years of age he was admitted to the hospital with an
seen that the increase in R is evident principally
attack of lobar pneumonia, and a heart murmur was
as a decrease in the end-tidal Nz concentration.
heard for the first time. Digitalis was started at this
When R, during the last 30 seconds of each time. At the age of 14 he was in the hospital again
workload, is plotted against oxygen consump- because of the onset of ankle edema. Hirsutism,
tion, a sigmoid curve is usually obtained (Fig. marked muscular development and a systolic mur-
3). The steepest part of this curve indicates the mur of variable quality, loudness and duration were

VOLUME14, DECEMBER1964
846 Wasserman and McIlroy

870 //oo /a ;Ilr 07.m aa

i! \\ ’
i

FIG. 2. End-tidal Nt and CO2 concentrations, heart rate and ventilation rate during graded exercise. 02 consump-
tion and R are noted on top. The anaerobic threshold is between 1,100 and 1,385 ml. Oz/min.

noted. A chest roentgenogram was interpreted as 4 L./min./M2. There was no change in the patient’s
showing slight left ventricular hypertrophy. Circula- clinical condition in the next three years. The only
tion time, venous pressure and arterial blood pressure symptom was mild intermittent ankle edema which
were normal. The electrocardiogram was normal ex- responded to treatment with chlorothiazide.
cept for evidence of digitalis effect. The drug was During the present admission there was no exertional
stopped at this time. dyspnea, and the electrocardiogram was normal, as
At the age of 16 he was admitted to the hospital for were the third cardiac catheterization findings, ex-
cardiac catheterization. Normal intracardiac pres- cept for the high cardiac output noted on the previous
sures and normal arterial oxygen saturation were occasions. The cardiac output on this occasion was
found, and there was no evidence of intracardiac 7 L./min./Ms., with a stroke volume of 152 ml. The
shunting. The cardiac output was increased, measur- total blood volume was also increased at 4.3 L./MZ.
ing 6.5 L./min./Ms. He was cautioned against (predicted, 2.5). Both the red cell mass and the
overexertion, although the precise cardiac defect plasma volume were increased. An extensive evalua-
was not clear. At the age of 22 he was investigated tion of the erythropoietic, endocrine, renal, skeletal
again because of three episodes of hemoptysis. The and metabolic systems did not reveal the cause of the
hemoptysis was mild and associated with prolonged patient’s enlarged heart and high cardiac output.
episodes of coughing. The heart was found to be The patient exercised on a treadmill and was found to
enlarged, and there was a loud sound of pulmonary have a threshold of anaerobic metabolism of 3 L.
valve closure. An early systolic murmur was heard Op/min. in contrast to the values of about 1.2 L./min.
along the left sternal edge, and purplish mottling of in untrained normal men. The patient’s heart rate
the lower legs was noticed. Arterial and venous blood reached only 138 during this exercise test, and linear
pressures were normal; hematocrit was 55 per cent. extrapolation of his heart rate response gave a value
The chest x-ray film showed biventricular enlarge- for the oxygen consumption of 4.6 L./min. at a heart
ment. The results of a second cardiac catheterization rate of 170. These measurements are similar to those
at this time were normal. The cardiac output was seen in well trained athletes. The only hormone assay

THE AMERICAN JOURNAL OF CARDIOLOGY

 Anaerobic Metabolism During Exercise

1.0

0.9,

0.0

0.7

Fm. 3. The chanae in R and bicarbonate from rest during uninterrupted graded ergometer exercise in an asympto-
matic 20 year old boy with a ventricular septal defect.

which proved to be abnormal in this patient was diastolic murmur, heard best between the apex and
that of growth hormone. left lower sternal border throughout ,diastole with
The provisional diagnosis was that the patient had presystolic accentuation. The electrocardiogram was
an unusual variant of acromegaly, possibly involving normal. Chest x-ray films showed no overall cardiac
only the cardiovascular and muscular systems. The enlargement, although the left atrium appeared to be
measurement of the threshold of anaerobic metabolism enlarged.
in this patient gave the clue to his supernormal exer- Cardiac catheterization showed a mean pulmonary
cise performance. artery wedge pressure of 14 mm. Hg, with a cardiac
output of 4.3 L./min. During exercise (oxygen con-
CASE 2. Mitral stenosis. A 24 year old married sumption of 566 ml./min., with a heart rate of 150/
woman was admitted to the hospital for evaluation min.) her cardiac output increased to 5.8 L./min.,
of a heart murmur. Six weeks before, she had become while the mean pulmonary artery wedge pressure in-
short of breath and coughed up blood-tinged sputum creased to 36 mm. Hg. Pulmonary vascular resist-
while taking swimming lessons. She also noted chest ance was calculated to be normal. The mitral valve
pain and fatigue. Chest x-ray films showed cardio- area was calculated to be 1.25 sq. cm. These results
megaly and evidence of pulmonary edema. She was were interpreted as showing moderate mitral stenosis.
treated with digoxin, chlorothiazide and penicillin. Because of her severe symptoms, plans were made
She improved with thii therapy; but when admitted for surgical correction at an early date.
for evaluation, she complained of dyspnea on mild Treadmill exercise studies before surgery and after
exertion, fatigue and orthopnea and required two optimal medical management revealed that this pa-
pillows at night. Her past history revealed that she tient’s anaerobic threshold was between 550 and 620
had been slightly short of breath on exertion during ml. of oxygen/mm This is a low value and suggested
childhood. She had had two pregnancies which she that this patient had a reduced capacity for exercise
tolerated well except for some dyspnea and orthopnea without developing a metabolic acidosis. At surgery,
during the second one. There was no history of pre- severe mitral stenosis without calcification of the valve
vious rheumatic fever. was found. The mitral valve orifice barely admitted
The physical findings were normal except for the the tip of the surgeon’s finger. The results of the exer-
heart, which was slightly enlarged. A diastolic thrill cise studies in thii patient were in keeping with her
was felt at the apex. There was a grade 4/6 harsh symptoms and seemed to give a better indication of

VOLUME 14, DECEMBER 1964

848 Wasserman and McIlroy

A :
A
00
0. 0

0 f+ . .

I-
Z’
A A
A
: B

f
0 0 00
,-
A
L UITRAL SrERosls -I
,-

FIG. 4. Threshold of anaerobic metabolism in 37 patients with heart disease.

the severity of her lesion than the resting cardiac was regular. The electrocardiogram was abnormal
catheterization data from which the valve area was and showed prolonged A-V conduction with intra-
calculated. ventricular conduction delay, large voltage in the left
CASE 3. Mitral insu&%iency. A 41 year old man precordial leads suggesting left ventricular hyper-
was admitted to the hospital for the evaluation of a trophy, and digitalis effect. Chest x-ray films showed
heart murmur. The murmur was first discovered mild left ventricular and possible left atria1 enlarge-
eight years before when he applied for life insurance. ment suggestive of mitral valve disease.
He remained asymptomatic until 18 months before Cardiac catheterization showed a cardiac output of
admission, when he began to have chest pain un- 5.6 L./min., with a v wave in the “wedge” pressure of
related to exertion and increasing exertional dyspnea. 43, a y of 14 and a mean pressure of 21 mm. Hg. With
His physician prescribed bed rest, a low salt diet, exercise during cardiac catheterization (90, = 795
digitalis and diuretics. With this regimen the patient ml./min. and heart rate = 98), the cardiac output
was symptom-free, having shortness of breath only increased to 7.3 L./min., while the wedge pressure
after heavy drinking or hard exercise. There was no increased to a mean of 54 mm. Hg with a v wave of
cough, hemoptysis, edema nor cyanosis and no past 88. It was concluded that the patient had a marked
history of rheumatic fever. degree of mitral insufficiency.
On physikal examination the heart was found to be Exercise studits on the treadmill showed normal results.
enlarged, with a loud, blowing systolic murmur best The patients’ anaerobic threshold was about 1.3
heard at the apex and radiating to the axilla. There L./min. VO,. The highest oxygen consumption at
was no diastolic murmur, and the cardiac rhythm which the patient exercised was 1.59 L./min. (heart

THE AMERICAN
JOURNALOF CARDIOLOGY
 Anaerobic Metabolism During Exercise 849

1.0

FIG. 5. The reproducibility of

0.9
effect of workload on R measured
during continuous exercise. The
subject is exercised between each
of the three series at work loads
R
less than 800 ml./min. until a
steady state in R is reached (re-
covery exercise). This required
approximately 10 minutes. The
numbers indicate the order of the
tests.
0.8

I I I I I
400 800 I200 1600 PO00

QO, (ML/M/N - StPD)

rate 125). These results were inconsistent with the (Fig. 5). Between each test the subject exer-
interpretation of the cardiac catheterization data cised continuously for about 10 minutes at a
but were in keeping with the patient’s relative lack of
workload below the “knee” of the A R-30,
symptoms. At surgery a cleft was found in the aortic
leaflet of the mitral valve, and the chordae tendineae curve-“recovery exercise”.13 Newman et a1.14
appeared normal. have shown that the rate of removal of lactic
acid from the blood is more rapid if the subject
DISCUSSION exercises at a light load instead of resting during
The cases presented show the type of informa- the recovery period. This also permits a more
tion that can be obtained from exercise tests in rapid recovery of R after work. The three
which the threshold of anaerobic metabolism is R-v,, curves were similar. R decreased when
determined. The workload at which meta- the workload was reduced, despite the continua-
bolic acidosis develops during exercise can tion of exercise. Furthermore, the threshold of
be seen as the exercise test is being performed, anaerobic metabolism was approximately the
and it is unnecessary to ask the patient to exer- same in the three exercise tests. This study
cise until exhausted. An awareness of the shows that the R-Oo, curve was independent of
changes in R during the test makes it possible to the cumulative effects of a period of exercise
choose appropriate loads for the subject, and the lasting one and a half hours.
discomfort of blood sampling is avoided. The Effect of Hyperventilation on R: Factors other
results obtained provide an objective measure of than metabolic acidosis which may cause R to
an aspect of exercise performance which supple- increase have been discussed elsewhere.’ The
ments the clinical and hemodynamic data. The most troublesome is hyperventilation. Its effect
test can be repeated at intervals to follow the on R is transientI and is also less during exercise
patient’s progress. when the rate of oxygen consumption is in-
To demonstrate that the observed changes in creased. We have calculated the order of
R are not due to the duration of exercise or the magnitude of the increase in R caused by hyper-
total energy expended, we repeated the work ventilation during exercise. Figure 6 shows the
test (ergometer exercise) three times without change in R at various oxygen consumptions for
stopping following a 10 minute “warm-up” different degrees of hyperventilation measured

VOLUME 14, DECEMBER 1964

850 Wasserman and McIlroy

+AR
0.4
drawing any conclusions about the development
of metabolic acidosis.
Anaerobic Metabolism us. Oxygen Debt:
Throughout this report we have spoken of the
use of the gas exchange ratio (R) to detect the
-Ahco,
threshold of anaerobic metabolism. Anaerobic
0.: metabolism should be differentiated from oxygen
debt. Oxygen debt includes not only the
oxygen deficit due to anaerobic oxidative proc-
esses during the conversion of pyruvate to
lactate but also the oxygen deficit which occurs
0.2
during exercise and which has been called by
Margaria et al. the “alactic acid” oxygen debt.lg
It occurs at all workloads during the first few
minutes of exercise and is rapidly repaid during
recovery.13 ,20 This is probably not an oxygen
debt resulting from anaerobic metabolism, but
0.1 rather it is due to depletion of the oxygen stores
in the body. Christensen et alzl have studied
intermittent heavy work of short duration and
showed that lactic acid did not increase in the
blood in spite of the accumulation of an oxygen
debt. When the rest period was sufficiently
l
long and the exercise of short duration, the
voptMLfMIn- PlPD) oxygen debt could be completely “alactic” and
be repaid during the rest period.
FIG. 6. The change in R at various 02 consumptions
Dawsonz2 questioned the source of the credi-
after two minutes of hyperventilation. The curves for
hypoventilation are the same except the signs for AR and tors for the oxygen debt. Certainly lactic acid
P*COz are reversed. is a creditor during heavy exercise. The credi-
tors responsible for the “alactic” oxygen debt
must be in part (1) the myoglobin which gives
as decrease in alveolar PcoZ. We have arbitrar- up its oxygenz3 (2) hemoglobin which gives up
ily assumed a period of hyperventilation lasting its oxygen when the mixed venous oxygen
two minutes and used the dissociation curve of saturation decreases during exercisez4 and (3)
the CO2 stores in the body and the dissociation the tissues which lose oxygen when oxygen ten-
rate data of Fahri and Rahn.15 While their sion in the tissues falls during exercise. These
measurements were made on dogs, the values creditors can account for approximately one
for the decrease in CO2 stores per kilogram of half of the “alactic” oxygen debt estimated from
body weight for each millimeter of Hg reduction the size of this debt reported by Dill.*O Re-
in alveolar CO2 tension are similar to those de- plenishment of the stores of creatine phosphate
termined by Vance and Fowler in man.16 It and adenosine triphosphate during recovery also
can be seen from Figure 6 that hyperventilation requires oxygen in excess of that required for
increases R most at low levels of oxygen con- The high energy phosphate acts
basal needs.
sumption. The one factor not considered in
as stored energy which is used during exercise
this calculation is the increase in cardiac output
and is restored as a result of aerobic oxidation
that occurs during exercise. It would tend to during recovery-another “alactic” creditor.
speed the rate of dissociation of CO2 stores and A graph showing the relationship between
cause the curves in Figure 6 to be shifted down.
oxygen debt and AR to oxygen consumption is
We, as well as others, seldom find a decrease in shown in Figure 7. The subject exercised for
arterial COZ tension during submaximal exer-
10 minutes at each workload, and 50 minutes
cise $7J’ ~8 so hyperventilation is not a common
were allowed for recovery. The oxygen debt
problem. Certain patients with mitral stenosis
may hyperventilate during exercise. In such was measured during the first 15 minutes of
instances it is necessary to determine how much recovery. This graph shows that the “knee” of
the hyperventilation would increase R before the AR-G,, curve (B) approximately coincided

THE AMERICAN JOURNAL OF CARDIOLOGY

 Anaerobic Metabolism During Exercise 851

toot

02 DEBT
WLJ

moo

.I5

+AR JO

.os,

0,

FIG. 7. A, the oxygen debt incurred for 10 minutes of exercise at the workloads indi-
cated by the oxygen consumption on the abscissa. B, AR for uninterrupted graded exer-
cise in the same subject.

with the onset of the steepest part of the oxygen how much work a subject can do before the
debt curve (A). The oxygen debt for workloads heart fails to meet the tissue oxygen require-
below the “knee” of the curve was paid off ments. The fact that this information can be
within four minutes of stopping work, but repay- obtained without blood sampling and at the
ment of oxygen debt was incomplete after 15 time the exercise is being performed has great
minutes of recovery for the workloads above the merit from both the patient’s and examiner’s
“knee.” Dill*O has pointed out that the bend in viewpoint.
the oxygen-debt curve is the point at which the
lactic acid oxygen debt becomes detectable. SUMMARY
The studies presented here indicate that the The measurement of the respiratory gas ex-
measurement of the ventilatory gas exchange change ratio (R) during a standard exercise test
ratio during exercise is a useful test of cardio- is used to detect the onset of anaerobic metab-
vascular function. It answers the question of olism during exercise, which results from fail-

VOLUME 14, DECEMBER 1964

852 Wasserman and McIlroy

ure of the cardiovascular system to supply the 10. IIUCKABEE,W. E. Abnormal resting blood lactate.
I. The significance of hyperlactatemia in hos-
oxygen requirements of the tissues. The
pitalized patients. Am. J. Med., 30: 833. 1961.
method described uses end-tidal gas concentra- 1 1. I IARRISON,7’. R. and PILCHER, C. Studies in con-
tions to calculate R while the exercise test is gestive heart failure. II. The respiratory exchange
taking place. Blood sampling is unnecessary, during and after exercise. J. Clin. Invest., 8: 291,
and the results can be determined during the 1930.
12. RAIIN, H. and FENN, W. 0. A Graphical Analysis
test, thus avoiding exhaustive exercise. The
of the Respiratory Gas Exchange, p. 19. Wash-
method provides an objective measurement of ington, D. C., 1955. American Physiological
one of the factors influencing exercise tolerance. Society.
13. DILL, D. B. and SACKTOR, B. Exercise and the
ACKNOWLEDGMENT oxygen debt. J. Sports Med. @ Phys. Fitness,
2: 66, 1962.
The authors wish to acknowledge the excellent assist-
14. NEWMAN, E. V., DILL, D. B., EDWARDS, H. T. and
ance of Antonius L. van Kessel during most of this study
WEBSTER,F. A. The rate of lactic acid removal in
and to thank Dr. Julius H. Comroe, Jr. and Dr. Herbert
exercise. Am. J. Physiol., 118: 457, 1937.
N. Hultgren for their helpful comments during the
15. FAHRI, L. E. and RAHN, H. Gas stores of the body
preparation of this paper.
and the unsteady state. J. Appl. Physiol., 7: 472,
1955.
REFERENCES 16. VANCE, J. W. and FOWLER, W. S. Adjustment of
1. HILL, A. V., LONG, C. N. H. and LUPTON, H. Mus- stores of carbon dioxide during voluntary hyper-
cular exercise, lactic acid and the supply and ventilation. Dis. Chest, 37: 304, 1960.
utilisation of oxygen. Proc. Roy. Sot. London, s. B, 17. MCILROY, M. B. and HOLMGREN,A. Arterial blood
96: 438, 1924; 97: 84, 1924; 97: 155, 1924. gas tensions during exercise in normal subjects.
2. HILL, A. V. The revolution in muscle physiology. Fed. Proc., 20: 423, 1961.
Phyriol. Rev., 12: 56, 1932. 18. RILEY, R. L. Pulmonary function in relation to
3. MARGARIA, R., CERRETELLI,P., DI PRAMPERO,P. E., exercise. In: Science and Medicine in Exercise
MASSARI, C. and TORELLI, G. Kinetics and and Sports, p. 162. Edited by JOHNSON,W. E.
mechanism of oxygen debt contraction in man. New York, 1960. Harper & Brothers.
J. Appl. Physiol., 18: 371, 1963. 19. MARGARIA, R., EDWARDS, H. T. and DILL, D. B.
4. HUCKABEE,W. E. Anaerobic energy metabolism in The possible mechanisms of contracting and pay-
man. Boston M. Quart., 9: 1, 1958. ing the oxygen debt and the role of lactic acid in
5. WELLS, F. G., BALKE, B. and VON FOSSAU, D. D. muscular contraction. Am. J. Physiol., 106: 689
Lactic acid accumulation during work. A sug- 1933.
gested standardization of work classification. J. 20. DILL, D. B. Fatigue and physical fitness. In:
Appl. Physiol., 10: 51, 1957. Ref. 18, p. 387.
6. DE LANNE, R., BARNES,J. R., BROUHA,L. and MAS- 21. CHRISTENSEN,E. H., HEDMAY, R. and SALTIN, B.
SART, F. Changes in acid-base balance and Intermittent and continuous running. Acta
blood gases during muscular activity and recovery. physiol. scandinou., 50: 269, 1960.
J. Appl. Physiol., 14: 328, 1959. 22. DAWSON, P. M. The Physiology of Physical Educa-
7. NAIMARK,A., WASSERMAN,K. and MCILROY, M. B. tion. D. 365. Baltimore. 1935. Williams &
Continuous measurement of ventilatory exchange Will&s.
ratio during exercise: A test of cardiovascular 23. ASTRAND, I., ASTRAND, P., CHRISTENSEN, E. H. and
function. J. Appl. Physiol., 19: 644, 1964. HEDMAN, R. Myohemoglobin as an oxygen-
8. ISSEKUTZ, B., JR. and RODAHL, K. Respiratory store in man. Acta physiol. scandinau., 48: 454,
quotient during exercise. J. Appl. Physiol., 16 : 1960.
606, 1961. 24. DONALD, K. W., BISHOP, J. M., GUMMING,G., and
9. PILCWER,C., CLARK, G. and HARRISON,T. R. The WADE, 0. L. Effect of exercise on cardiac output
buffering power of the blood and tissues. J. Clin. and circulatory dynamics of normal subjects.
Invest., 8: 317, 1930. Clin. SC., 14: 37, 1955.

THE AMERICAN JOURNAL OF CARDIOLOGY