- Email: [email protected]
Ventilatory and cardiovascular responses to hypoxic and hyperoxic statis handgrip exercise in man
Respiration Physiology, 81 (1990) 189-202
189
Elsevier
RESP 01675
Ventilatory and cardiovascular responses to hypoxic and hyperoxic static handgrip exercise in man M. Pokorski, A. Masuda, P-E. Paulev, Y. Sakakibara, B. Ahn, S. Takaishi, Y. Nishibayashi and Y. H o n d a Department of Physiology, School of Medicine, Chiba University, Chiba, 280 Japan (Accepted 7 April 1990) Abstract. The purpose of this study was to evaluate the ventilatory and cardiovascular responses to static
handgrip exercise at different levels of arterial chemoreceptor activation. The study was done on 10 healthy subjects. They performed handgrip of 50 ~ of maximal voluntary contraction on a background of either hypoxia (PEb2 ~ 47 mm Hg) or hyperoxia (PE~2 ~ 216 mm Hg), i.e., enhanced or suppressed chemoreceptor activity. The subjects were able to sustain the handgrip for 50-60 sec, during which time no steady-state responses were attainable. Minute ventilation ('(/1), cardiac output (t)), heart rate (HR), and a number of other variables were recorded. Handgrip exercise resulted in a rapid initial VI rise followed by a subsequent slow increase. Hyperoxia diminished the ~zl response over the exercise range. The ventilatory response was associated with an HR acceleration, increased arterial pressure and peripheral vascular resistance. No appreciable changes in t~ were noted, nor was there any particular relationship between ventilatory and circulatory changes. These results provide no support for the t) mediated ventilatory stimulus during static handgrip exercise in man. It is concluded that the ventilatory and cardiovascular responses are of independent nature.
Animal, man; Control of breathing, arterial chemoreceptors; Exercise, static; Exercise, ventilatory and cardiovascular responses to - ; Hyperoxia, and venti/atory responses to exercise; Hypoxia, and ventilatory responses to exercise
Static exercise, like dynamic, results in a rapid heart rate (HR) acceleration but is associated with a more pronounced arterial pressure increase and variable cardiac output (0) changes hinged on changes in vascular resistance (Martin et aL, 1974). An acutely increased afterload due to sympathetic vasoconstriction in response to an intense isometric contraction limits t) increase. The HR acceleration, known as a muscle-heart reflex (Gelsema et aL, 1983; Mitchell and Schmidt, 1983) originates in muscle receptors and is mediated by group III and IV afferents. The HR acceleration Correspondence to: M. Pokorski, Department of Neurophysiology, Polish Academy of Sciences Medical Research Center, 00-784 Warsaw, Dworkowa 3, Poland. 0034-5687/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
190
M. POKORSKIet al.
is accompanied by an hyperpnea (Paulev, 1971, 1973; Muza et al., 1983), whose origin and the contribution to it of chemical drive, mediated by the arterial chemoreceptors, are uncertain. The hyperpnea of exercise is known to be linked to metabolic rate changes. Close matching of the respiratory and cardiovascular events, yielding constant arterial blood gas content has been the mainstay of the cardiodynamic concept, linking the rapid rise in ventilation atthe onset of exercise to increased 0 (Wasserman et al., 1974). This concept has never been unequivocally proved, nor is it ubiquitously accepted (Adams et al., 1987). In view of the controversial mechanisms, we designed the present study in an attempt to gain further insight into the reflex cardiorespiratory output events. We hypothesized that isometric contraction of small, but providing an intense receptor stimulus, muscle group would be likely to resolve aspects of the interplay among heart rate, cardiac output, peripheral neurogenic drive, and chemical drive in the mechanism of exercise hyperpnea. We therefore used a handgrip test, in contrast to the previous work, on a background of either hypoxia or hyperoxia, i.e., with enhanced or suppressed arterial chemoreceptor activity, in the same subjects. The specific purpose of the study was to characterize the ventilatory and cardiovascular responses to step changes from rest to static handgfip exercise, and to determine the role of chemicai drive in these responses. Overall, we found that the increase in ventilation was unaccompanied by appreciable changes in t) and that inactivation of the arterial chemoreceptors had a substantial inhibitory effect on the response. The study emphasizes the importance of chemical drive and shows that mechanisms other than those involving a cardiac link predominate in the ventilatory response to handgrip exercise.
Methods
The study was performed on 10 conscious, healthy, fit, middle-aged men. They consented to study procedure after having its general nature explained. The study was approved by an institutional ethics committee. The subject was connected with a mouthpiece and noseclip to a breathing circuit consisting of an inspiratory and expiratory line separated by a one-way Lloyd valve. He breathed from a bag containing a humidified hypoxic or hyperoxic gas mixture, produced manually by adding O 2 or N2 to air. End-tidal 0 2 , and CO2 (pE ¢,o : ) pressures were measured with a San-ei IH2I rapidly re(PEo2) sponding 0 2 and CO2 analyzer in a sample drawn continuously from the valve. The analyzer was calibrated with predetermined gas concentrations. Breath-by-breath inspired airflow was measured with a hot wire flowmeter (Minato RF-H), which was inserted between the mouthpiece and the valve and calibrated with a 2 L piston. Breath tidal volume (VT) was integrated from the flow signal, and inspiratory (TI) and expiratory (TE) times computed by an analog calculator. Total respiratory cycle time Recording procedure.
VENTILATION IN STATIC EXERCISE
191
(TT) and breath frequency (f = 60/TT) were calculated. From the above variables instantaneous inspiratory minute ventilation (VI = VT X O, mean inspiratory flow (VT/TI, and fractional inspiratory time (TI/TT) were calculated. The cardiac variables heart rate (HR), stroke volume (SV) and cardiac output (0) were measured with a Minnesota impedance cardiograph. The impedance signal was ensemble-averaged every 5 heart beats to improve the time resolution of the measurement. The HR was additionally recorded along with arterial oxygen saturation with an ear oximeter (Ohmeda, Biox III). Oxygen uptake (~'o2) and C O 2 output (Vco2) were calculated before and in the latter half of the exercise period (fig. 1) from the volume of expired gas and 02 and CO2 fractional concentrations in it measured with a San-ei analyzer. Arterial blood pressure was measured at the arm with a Riva-Rocci cuff manometer at the same time points plus during the first 20 sec after the end of exercise (fig. 1). Total perip.heral vascular resistance was calculated as the ratio of mean arterial pressure over Q. All continuously recorded variables along with the force of the handgrip concentration, performed with a dynamometer (SPR-655, Sakai), were displayed on a multi-channel strip-chart hot stylus recorder (San-ei Instruments).
Experimentalprotocol. The handgrip was performed with the fight hand in the upright sitting posture. The experiment started from the determination of maximal voluntary contraction (MVC), taken as the highest value achieved in 2 trials. The mean MVC was 38.6 + 1.8 kg. The subject was then instructed to develop a grip tension of about 50~o his MVC and maintain it as closely stable as possible by observing the tension trace at the oscilloscope screen for 50-60 sec. The mean tension developed was 47 + 1~o MVC in hypoxia and 48 + 1 ~o MVC in hyperoxia. Tensions above 10-25 ~o MCV result in fatigue, limiting the exercise period and making steady-state responses unattainable (Asmus sen, 1981). To minimize the potentially confounding influence of fatigue, 50 sec was taken as a cutoff point in the handgrip test. Each subject was interrogated after the test about his subjective feeling of effort to maintain the grip tension. The subject was connected to the breathing circuit for familiarization with the experimental setup while breathing room air. At this stage, to compare later the control resting hypoxic and
CONTROL PERIOD
HANDGRIPEXERCISE
TIME { s }
0
50
I
IJ}lll,,, BIN I:I~II
0
1
2
3
4
5
RECE~Y PERIOD
6
I 7
i 8
Meosurerneet of blood pressure
CoIiect~ ~' expired gas for 5 breaths in order to ~Ic,,,,aote the
V02 and VCO2
Fig. l. Diagram of experimental procedure. See text for details.
192
M. POKORSKI et al.
hyperoxic ventilations while having only one chemical stimulus changed, PESo 2 was set a few mm Hg above the spontaneous normoxic value to a level of 41 + 1 mm Hg. This initial PESo 2 increase allowed the possibility to lower it when it slightly increased on switching to hyperoxia and to hold it constant in the control periods of both hypoxia and hyperoxia. The PESo ~ was adjusted by closing off the expiratory line and allowing the exhaled air with its higher CO 2 content to be rebreathed partially straightaway and partially after passing through a CO 2 scrubber connected in parallel to the circuit. Once the handgrip started, no attempt at further PESo ~ adjustments was made. The protocol consisted then of two sequences of the handgrip run, one in hypoxia ( P E ~ - 47 + 2 mm Hg) and one in hyperoxia ( P E ~ = 216 + 3 mm Hg). The order of gas condition was randomized. The handgrip always started from the end of expiration at a verbal signal and was preceded by at least a 5 min control period of breathing a given gas mixture on the circuit. The handgrip runs were separated by a 6-8 min rest period of breathing room air while remaining connected to the circuit. All data were collected off-line from the stripchart recording and stored in a computer (NEC PC-9801) for further computations. The mean value of each variable in the period of 10 full breath cycles immediately preceding the exercise run was taken as control (bin 0). The handgrip exercise period was divided into five 10 sec bins, and the recovery period into three 10 sec bins (fig. 1). The sequential bins were coded in arabic numerals from 1 to 8. The breath-by-breath data were averaged over each bin by the computer. The averaging was performed to minimize the effect of the breath-tobreath fluctuations on the response pattern. Means + SE of bins with equal sequential numbers were calculated for each variable. Comparisons of the means for the ten subjects and across the nine time bins covering three experimental conditions: rest, exercise, recovery were made with a random block analysis of variance. If the analysis showed significant differences among the means (F-ratio exceeding the critical Fo.os value), they were further compared with a paired t-test and the Bonferroni method of multiple comparisons. Since 10 pairwise comparisons out of the possible 36 among the 9 bins were of interest, a Bonferroni correction of 10 was used. Differences between two mean values of bins with equal sequential numbers in hypoxia and hyperoxia were determined with a paired t-test. P < 0.05 was deemed significant. Data treatment.
Results
Handgrip exercise induced a substantial change in metabolic rate at both 0 2 levels, which was evident from the increases in "V'o2and 'V'co2 (fig. 2). These increases were smaller in hyperoxia than in hypoxia: AVo2 was 0.105 + 0.04 and 0.144 + 0.05 L . m i n -1, and A~'co ~ was 0.128 + 0.04 and 0.152 + 0.06 L . m i n -~, respectively. These differences were insignificant (P > 0.05). The ventilatory and cardiac responses studied did not attain steady state within the 50 sec duration of the exercise bout. The responses are summarized in fig. 3 which shows the time course of the average
VENTILATION IN STATIC EXERCISE
HYPERCIKIA
H~'F~
I ~
,,
I 02
J 03 0,~ 02 Uptoke {I STPD. rmn"1 }
-
193
REST EXERCISE
" 0.5
t HYPC~
0,2
03 0.4 CO20~tpu! {I STPO.min-~)
0.5
Fig. 2. Metabolic rate changes in response to static handgrip exercise. Differencesbetween changes in hypoxia and hyperoxiawere not significant. *P < 0.05 vs correspondingrest (paired t-test). ventllatory and cardiac variables at sequential 10 sec bins of the handgrip test. The characteristics of the responses differed for the two gas conditions. Ventilatory response. The ~/I was significantly higher in hypoxia during the control rest period (P < 0.05; fig. 3). The general pattern of the ventilatory response to handgrip exercise was one of an abrupt rise in VI followed by a further gradual increase. The mean absolute VI rise in the first 10 sec (bin 1) was significantly higher in hypoxia than in hyperoxia (P = 0.05; fig. 4). The early component accounted, however, for a similar fraction of the total ~/I increment concluding the hypoxic and hyperoxic exercise runs; 81 ~o (range 17-176Yo) and 81 ~ (range 6-209~o), respectively. The difference between bin 1 VI increases (AVI) in hypoxia and hyperoxia amounted to 3.8 + 2.0 L. min - 1 (fig. 4). The difference rose progressively over the exercise range, reaching 9.0 + 2.5 L- min - 1 towards its end, i.e., bin 5. The rise of A~h was due largely to a gradual VI increase in hypoxia, which was virtually absent in hyperoxia. A least squares regression analysis applied to the data of bin 1-5 showed that the slope of the ~'I response was significant in hypoxia: 0.84 L. min - 1 per bin, r - 0.95, whereas that in hyperoxia was not: - 0.01 L. min - 1 per bin, r = - 0.03. The hypoxic and hyperoxic ~/I increments were due mostly to increases in the tidal component. The VT increased abruptly at the transition from rest to exercise and remained stable throughout the exercise bout at both O 2 levels. The increase was significantly higher in hypoxia (P < 0.05; fig. 3) and was associated with a tendency for f to increase, which was absent in hyperoxia. The analysis of VT/TI shows that respiratory drive was significantly higher in hypoxia than in hyperoxia over the range of handgrip exercise (fig. 3). The modulatory effect of chemical drive is also seen when one examines the components of breath timing. This is depicted in fig. 5 where sequential bin TE is plotted against its corresponding TI. Whereas hypoxia increased f shortening both TE and TI, hyperoxia tended to decrease f lengthening TE with no change in TI and correspondingly lower TI/TT.
194
M. POKORSKI
A
et al.
~
!I! n
.~
20
_-
d," ~" ~" ~"
~J ~ " ~ ' ~ ,
,~
q, ~' ~"
mL~'1
<~'~
,~_~,
[
2.5
"
"'
.
.
.
40'
~ ~
.
Q5 ~ 1.0
Time(s) Bin#
.
.
.
.
o
.
J
.I .....
I
,
i
c> ~" c~. ~. Cl C¢' #', ~' 0
=0.
.
i
Z>' ~"
C 010 01
0
0.3
1 ~'~, 30
5010
2345
678
EXERCISE
30
Tmae{s) Bin#
..... 0
10
.
30
..
,__,
50 10
30
0 1 2 3 4 5 6 ? 8
RECOVERY
EXERCISE
RBOCNERY
Fig. 3. Average minute ventilation ('~/I), frequency (f), tidal volume (VT), mean inspiratory flow (VT/TI), fractional inspiratory time (TI/TT), cardiac output ((~), heart rate (HR), and stroke volume (SV) for 10 subjects at rest (control), at sequential 10 see bins ofhandgrip exercise, and during recovery in hypoxia (squares) and hyperoxia (triangles). The control is coded bin 0, the first 10 sec of exercise bin 1, etc. Onand off-transitions are symbolized with heavy vertical lines. Bars on symbols are SE; lack of bar denotes SE smaller than the symbol size. The overall F-test indicated no differences among the means of the f, (~, and TI/TT. *P < 0.05 vs bin 0 in either gas condition (F-test and Bonferroni's multiple comparisons); i" P < 0.05 vs. corresponding hypoxic bin (paired t-test); ~.P < 0.05 vs bin 5 (F-test and Bonferroni's multiple comparisons).
HYPEROXlA
Or'I
HYPOXIA
Q i
0
I' I
5 A t. rnin -1
J
10
Fig. 4, Concurrent early (bin 1) increments in Vi (open bars) and (~ (solid bars) in response to a step change from rest to handgrip exercise in hypoxia and hyperoxia. Values are means + SE. * ~'I increment in hypoxia was significantly higher than that in hyperoxia (P = 0.05; paired t-test). Both were associated with unappreciable changes in (~.
195
V E N T I L A T I O N IN STATIC EXERCISE 5 10 xx
/
0.3 T,/Tr
x x / 4 •f
/
/ ~'x
0.4
,
/ "-, -
/
/
IZ] HYPOXIA A HYPEROXlA
1
"x
I
I
2
3
Tz
(see)
Fig. 5. Average expiratory time (TE) plotted against corresponding inspiratory time (TI) at sequential bins of handgrip exercise. Solid symbols represent control values at rest. SE bars are omitted for clarity. Dashed lines are frequency isolines, and continuous lines are TI/TT isolines,
The ventilatory response was accompanied by little fluctuations in PESo 2when 10 sec averages were considered (table 1). The PESo 2 remained closely at its prior control level over the exercise test at both 0 2 levels. A tendency to hyperventilation in hypoxia with TABLE 1 Average data for end-tidal gas tensions for the control state and sequential 10 sec bins of static exercise during hypoxia and hyperoxia. Hypoxia
Rest Exercise
Recovery
Hyperoxia
Bin No.
PE~2 ( m m Hg)
PESo 2 (mm Hg)
PE~2 ( m m Hg)
PESo 2 (mm Hg)
0 1 2 3 4 5 6 7 8
47.3 48.1 49.4 49.7 50.0 45.2 52.6 49.8 46.7
40.9 41.0 40.8 40.9 40.3 40.1 39.5 40.8 41.5
216.0 218.4 216.9 214.5 212.7 212.3 215.9 215.0 212.3
40.5 40.2 40.4 40.6 41.3 41.9 39.0 40.5 41.1
+ 1.5 + 2,1 + 2,8 + 3,3 + 3.9 + 3.3 + 3.0 ___2.3 _+ 2.2
+ + + + + + + + +
0.8 1.1 1.3 1.3 1.5 1.6 1.6 1.5 1.3
+ 3.1 + 3.6 + 3.6 + 3.8 + 4.3 _+ 6.0 + 4.6 + 4.8 + 5.0
+ 0.8 + 0.9 + 1.0 + 1.2 + 1.4 + 2.0 _+ 1.5 + 1.1 + 1.1
Values are means + SE for 10 subjects. F-test showed insignificant differences among the m e a n s of either end-tidal gas tension in hypoxia and in hyperoxia,
196
M. POKORSKI et al.
PE~2 increasing and PESo 2decreasing, and the converse in hyperoxia were seen towards the end of the exercise bout. In the recovery period the variables were gradually returning toward the preexercise values in about 30 see. The characteristic feature of VI was that it kept on increasing for some 10 sec after cessation of exercise. Cardiovascular response. The control resting heart rate and cardiac output were significantly lower in hyperoxia than in hypoxia (fig. 3). The cardiac response was qualitatively different from that described above for ventilation. The onset of isometric hand muscle activity was accompanied by an abrupt HR acceleration at both O2 levels. The magnitude of the early HR rise was not affected by the 0 2 level. It amounted to 8.5 + 2.2 beats- rain- 1 in hypoxia and 9.4 + 2.3 beats, rain - ~ in hyperoxia (P < 0.05), which accounted for 61 ~ (range 12-193 ~o) and 63 ~o (range 20-1119'o), respectively, of the HR increment concluding the exercise bout. The HR remained lower in hyperoxia over the exercise range in proportion to the difference present in the control period (fig. 3). It was steadily increasing starting from the initial acceleration at both 0 2 levels. Regression analysis applied to the data of bin 1-5 revealed similar and significant slopes of the HR responses in hypoxia and hyperoxia; 1.7 beats, min- ~ per bin, r = 0.97 and 1.9 beats, min - 1 per bin, r = 0.94, respectively. Changes in H R were associated with decreases in SV, which were greater in hyperoxia. As a result, Q did not change appreciably over the exercise range at both 0 2 levels. In fact, small reductions in (~ inclusive of the time period corresponding to the early VI increase were noted (fig. 3). In addition to the above cardiac alterations, handgrip exercise resulted in pronounced increases in total peripheral resistance and mean arterial blood pressure (fig. 6). The HR reverted abruptly to the resting level at the off-transition with no overshoot characteristics.
Linear correlation and regression analysis revealed no relationship between the initial ~/I increase and the corresponding HR increase, 0 alterations, control HR and control V1 at either 02 level. Particularly, this "v'xincrease was unaccompanied by an increase in 0 (fig. 4). When the whole exercise range was examined, a positive correlation emerged in hypoxia between the size of the sequential bin WI increment and the corresponding H R increment (r = 0.86; fig. 7). The good correlation disappeared in hyperoxia. Cardiorespiratory interplay.
Discussion
This study demonstrated that ventilation during static exercise in man was not linked to a cardiac output mediated stimulus and that hypoxic drive augmented the dynamics of the ventilatory response. The hyperpneic response unaccompanied by a rise in (~ was rather unexpected in the light of the cardiodynamic theory (Wasserman et al., 1974). This theory presumes that the increase of ~/I is coupled to a rapid increase of (~ at the onset of exercise, which
197
VENTILATION IN STATIC EXERCISE '-"
i REST I---I EXEROSE l ~ RECOVERY
~.. •
HYPEROXIA
i
//////L.
I
I
90
f
I - -
HYPEROXIA
I
110 130 Mean Arterial Pressure (mmHg)
150
1
/////A~
I---
i
~,
I
18 24 Tofol Per/pherol ~scul~,
I
I
30
36
ResistoncefmrnHcj.14.rr~)
Fig. 6, Mean arterial pressure (panel A) and peripheral vascular resistance (panel B) (means + SE) in consecutive periods ofhandgrip procedure in hypoxia and hyperoxia. *P < 0.05 vs rest and recovery (F-test and Bonferroni's multiple comparisons).
t0 i___ E
n
-
/
i'i HYI:~XIA A HYPEROXIA I
0
10
, I
20
6 HR ( bears - min'lJ
Fig. 7. Relationship between the mean responses of Vl and HR to hypoxic and hyperoxic handgrip exercise. Symbols represent increments above average resting value at sequential bins of exercise for 10 subjects. Lines are corresponding linear regressions. The slope of ~'l on HR in hypoxia is significant (P < 0.05).
would explain the proportionality of the ventilatory response to the gas exchange at the lung yielding constant blood gas tensions. Our findings argue against the cardiac mechanism being operative in increasing VI and lend support for those few reports which show that t) rise is not a prerequisite for '~'t rise. Huszczuk et al. (1986) exercised
198
M. POKORSKI et al.
calves with the Jarvik artificial heart implant, maintaining a constant Q, and found an isocapnic hyperpneic response similar to that when Q was allowed to increase. Banner et al. (1988) reported a dissociation of the ventilatory and cardiac responses to the onset of dynamic or electrically induced leg exercise in patients with heart or heart-lung transplants. The patients had a normal ~/I response, although (~ and H R did not change appreciably. Adams et al. (1987) found that the ventilatory response virtually preceded the rise of 0 in man. A parallel but not coupled stimulus to both respiratory and cardiovascular events is conceivable, which may affect the central controllers of both functions with disparate input kinetics or may not affect them concurrently at all. The initial ~/I increase would then be an independent phenomenon of cardiac events. The good correlation found between the magnitudes of the ~'I rise and the HR acceleration in hypoxia reflects likely a fortuitous association, since we noted important disparities. Hyperoxia had an inhibitory effect on the VI response, but it did not affect the H R response, causing a loss of the correlation between the two. At cessation of exercise the ~q increase continued for some time, but HR declined abruptly. In the present study on conscious man both descending command from the rostral brain and peripheral neurogenic drive could be operative in controlling the cardiorespiratory events. Central command alone can be an initiator of the cardiovascular and ventilatory responses (Eldridge and Millhorn, 1986). Our subjects did not report an increased perception of effort in the short exercise period studied. That the handgrip was not fatiguing enough to substantially activate central command may also be judged from the constancy of PESo 2, showing that no hyperventilation exceeding metabolic needs developed. Anticipatory or conditioning process to the repeat verbal signal to exercise was unlikely, since the initial hyperpnea was also present when the subject started handgrip without any prior signal. Peripheral neurogenic drive ascends from the muscle receptors known to respond to noxious mechanical and chemical stimulation (Mense, 1977; Kumazawa and Mizumura, 1977), a feature of isometric contraction. Feedback from muscle receptors mediated by group III and IV afferents has been recognized as an inducer of both exercise hyperpnea (Senapati, 1966; Kalia et al., 1972) and HR acceleration (Gelsema e t a l . , 1983). The disappearance of hyperpnea after anesthetic blockade of these afferents (McCloskey and Mitchell, 1972) strongly supports the peripheral neurogenic mechanism. An extended ventilatory response after cessation of exercise we found accords with the observation of Senapati (1966) made during stimulation of group III and IV afferents from the exercising muscles. The part played by peripheral or central drive could not be discerned in our study. The role of the arterial chemoreceptors in the ventilatory response to static exercise is imprecisely defined. Muza et al. (1983) reported responses similar to ours at the beginning of handgrip of 25 ~o MVC sustained for 5 rain in hyperoxia. Similar responses were also observed with 50~ MVC in normoxia by Myhre and Andersen (1971). This is not unexpected, given the little effect on ventilation of carotid chemoreceptors in the normoxic range. We have found no report of comparison of the ventilatory responses to static exercise at enhanced and abolished chemoreceptor activity. In the present study
VENTILATION IN STATIC EXERCISE
199
we observed a substantial augmenting influence of carotid chemoreceptor activation on the ventilatory response to static exercise. The mechanisms of such influence may differ at different stages of exercise and are not easy to explain. The early V1 rise was present at both 02 levels, which implies that the basic hyperpneic stimulus is extrachemoreceptor. This rise was, however, diminished when carotid chemoreceptor drive was absent in hyperoxia. The difference between the hypoxic and hyperoxic VI rise could thus point to the effect mediated by the arterial chemoreceptors. How they would do this is unknown. The rapidity of the initial VI rise and a fair stability of PESo: makes a feedback correction of a downstream chemical signal error improbable; an action that is likely later in the course of exercise. The arterial chemoreceptors assumed greater importance over time, which was evident from the inhibitory effect of hyperoxia on the progressive "~I increase seen in hypoxia. The carotid chemoreceptor involvement in the ventilatory response might have to do with a rapid release of potassium from the exercising muscles (Kilburn, 1966; Lim et al., 1981; Laurell and Pernov, 1966). Arterial plasma potassium concentration and ventilation are closely related in exercise (Newstead, 1988; Conway et al., 1988). Exogenously administered potassium in a dose matching that produced endogenously in moderate dynamic exercise stimulates both afferent carotid chemoreceptor activity and ventilation (Linton and Band, 1985). The potassium effect on the chemoreceptor activity is potentiated by hypoxia (Burger et al., 1986; Band and Linton, 1987), which might explain the potentiation of the hypoxic ventilatory response to static exercise in our study. Sneyd et al. (1988) found, however, no difference in the ventilatory effects of potassium during hypoxia, normoxia, and hyperoxia. We did not measure the arterial plasma concentration of potassium. If measured, it would have been difficult to ascertain to what extent it reflected the local concentration of potassium, since an appreciable negative arterio-venous difference exists and is augmented on exercise (Laurell and Pernov, 1966). While this study has not resolved the much debated mechanism of exercise hyperpnea, we believe we have shown that the ventilatory response to a step change from rest to static handgrip exercise is not cardiodynamic. The origin of the response is extrachemoreceptor, although the arterial chemoreceptors play an increasing role in its course. We conclude that the ventilatory and cardiovascular output events were driven in parallel. Further studies are needed to explore the proposed mechanism.
Acknowledgements. M. Pokorski and P-E. Paulev were Visiting Professors from the Department of Neurophysiology,Polish Academyof Sciences Medial Research Center, 00-784 Warsaw, 3, Dworkowa, Poland, and the Department of Physiology,Sports/Cybernetics,Universityof Copenhagen,Panum4: 2: 7, 2200 Copenhagen N., Denmark, recipients of Japan Society for the Promotion of Science and of Scandinavian-Japan Sasakawa FoundationFellowships,respectively.
200
M. POKORSKI et al.
References Adams, L., A. Guz, J.A. Innes and K. Murphy (1987). The early circulatory and ventilatory response to voluntary and electrically induced exercise in man. J. Physiol. (London) 383: 19-30. Asmussen, E. (1981). Similarities and dissimilarities between static and dynamic exercise. Circ. Res. 48 (Suppl. I): 3-10. Band, D.M. and R. A.F. Linton (1987). The interaction of hypoxia and K + on the carotid chemoreceptor in anesthetized cats. J. Physiol. (London) 394: 65P. Banner, N., A. Guz, R. Heaton, J. A. limes, K. Murphy and M. Yacoub (1988). Ventilatory and circulatory responses at the onset of exercise in man following heart or heart-lung transplantation. J. Physiol. (London) 399: 437-449. Burger, R.E., J.A. Estavillo, P. Kumar and P.C.G. Nye (1986). The excitation of cat carotid body chemoreceptors by hyperkalaemia depends on Po2 and Pco2. J. Physiol. (London) 374: 25P. Conway, J., D.J. Paterson, E.S. Petersen and P.A. Robbins (1988). Changes in arterial potassium and ventilation in response to exercise in humans. J, Physiol. (London) 399: 36P. Eldridge, F.L. and D.E. MiUhorn (1986). Hypothalamic central command and exercise hyperpnea. In: Neurobiology of the Control of Breathing, edited by C. yon Euler and H. Lagercrantz. New York: Raven Press, pp. 35-43. Gelsema, A.J., G. de Groot and L.N. Bouman (1983). Instantaneous cardiac acceleration in the cat elicited by peripheral nerve stimulation. J. Appl. PhysioL 55: 703-710. Huszczuk, A., B.J. Whipp, K. Wasserman, T.D. Adams, A.G. Fisher, D.B. Olsen, R.O. Crapo and C.G. Elliott (1986). Ventilatory control during exercise in calves with artificial (Jarvik) hearts. Fed. Proc. 45: 1127. Kalia, M., J. M. Senapati, B. Parida and A. Panda (1972). Reflex increase in ventilation by muscle receptors with nonmeduUated fibers (C fibers). J. Appl. Physiol. 32: 189-193. Kilburn, K.H. (1966). Muscular origin of elevated plasma potassium during exercise. J. Appl. Physiol. 21: 675-678. Kumazawa, T. and K. Mizumura (1977). Thin-fiber receptors responding to mechanical, chemical, and thermal stimulation in the skeletal muscle of the dog. J. Physiol. (London) 273: 179-194. Laurell, H. and B. Pernow (1966). Effect of exercise on plasma potassium in man. Acta Physiol. Scand. 66: 241-242. Lim, M., R.A.F. Linton, C.B. Wolff and D.M. Band (1981). Propranolol, exercise, and arterial plasma potassium. Lancet 2: 591. Linton, R.A.F. and D.M. Band (1985). The effect of potassium on carotid chemoreceptor activity and ventilation in the cat. Respir. Physiol. 59: 65-70. Martin, C.E., J.A. Shaver, D.F. Leon, M.E. Thompson, P. S. Reddy and J.J. Leonard (1974). Autonomic mechanisms in hemodynamic responses to isometric exercise. J. Clin. Invest. 54:104-115. McCloskey, D.I. and J.H. Mitchell (1972), Reflex cardiovascular and respiratory responses originating in exercising muscle. J. Physiol. (London) 224: 173-186. Mense, S. (1977), Muscular nociceptors. J. Physiol. (Paris) 73: 233-240. Mitchell, J. H. and R. F. Schmidt (1983). Cardiovascular reflex control by afferent fibers from skeletal muscle receptors. In: Handbook of Physiology. Section 2. The Cardiovascular System. Vol. III, Peripheral Circulation, edited by the American Physiological Society: Bethesda, MD, pp. 623-658. Muza, S. R., L.Y. Lee, R. L. Wiley, S. McDonald and F.W. Zechman (1983). Ventilatory responses to static handgrip exercise. J. Appl. Physiol. 54: 1457-1462. Myhre, K. and K.L. Andersen (197 i). Respiratory responses to static muscular work. Respir. Physiol. 12: 77-89. Newstead, C. G. (1988), The relationship between arterial potassium and ventilation during exercise in man. J. Physiol. (London) 403: 101P. Paulev, P-E. (1971), Respiratory and cardiac responses to exercise in man. J. Appl. Physiol. 30: 165-172.
VENTILATION IN STATIC EXERCISE
201
Paulev, P-E. (1973). Cardiac rate and ventilatory volume rate reactions to a muscle contraction in man. J. Appl. Physiol. 34: 578-583. Senapati, J.M. (1966). Effect of stimulation of muscle afferents on ventilation of dogs. J. Appl. Physiol. 21: 242-264. Sneyd, J.R., R.A.F. Linton and D.M. Band (1988). Ventilatory effects of potassium during hyperoxia, normoxia and hypoxia in anaesthetized cats. Respir. Physiol. 72: 59-64. Wasserman, K., B.J. Whipp and J. Castagna (1974). Cardiodynamie hyperpnea: hyperpnea secondary to cardiac output. J. Appl. Physiol. 36: 457-464.
189
Elsevier
RESP 01675
Ventilatory and cardiovascular responses to hypoxic and hyperoxic static handgrip exercise in man M. Pokorski, A. Masuda, P-E. Paulev, Y. Sakakibara, B. Ahn, S. Takaishi, Y. Nishibayashi and Y. H o n d a Department of Physiology, School of Medicine, Chiba University, Chiba, 280 Japan (Accepted 7 April 1990) Abstract. The purpose of this study was to evaluate the ventilatory and cardiovascular responses to static
handgrip exercise at different levels of arterial chemoreceptor activation. The study was done on 10 healthy subjects. They performed handgrip of 50 ~ of maximal voluntary contraction on a background of either hypoxia (PEb2 ~ 47 mm Hg) or hyperoxia (PE~2 ~ 216 mm Hg), i.e., enhanced or suppressed chemoreceptor activity. The subjects were able to sustain the handgrip for 50-60 sec, during which time no steady-state responses were attainable. Minute ventilation ('(/1), cardiac output (t)), heart rate (HR), and a number of other variables were recorded. Handgrip exercise resulted in a rapid initial VI rise followed by a subsequent slow increase. Hyperoxia diminished the ~zl response over the exercise range. The ventilatory response was associated with an HR acceleration, increased arterial pressure and peripheral vascular resistance. No appreciable changes in t~ were noted, nor was there any particular relationship between ventilatory and circulatory changes. These results provide no support for the t) mediated ventilatory stimulus during static handgrip exercise in man. It is concluded that the ventilatory and cardiovascular responses are of independent nature.
Animal, man; Control of breathing, arterial chemoreceptors; Exercise, static; Exercise, ventilatory and cardiovascular responses to - ; Hyperoxia, and venti/atory responses to exercise; Hypoxia, and ventilatory responses to exercise
Static exercise, like dynamic, results in a rapid heart rate (HR) acceleration but is associated with a more pronounced arterial pressure increase and variable cardiac output (0) changes hinged on changes in vascular resistance (Martin et aL, 1974). An acutely increased afterload due to sympathetic vasoconstriction in response to an intense isometric contraction limits t) increase. The HR acceleration, known as a muscle-heart reflex (Gelsema et aL, 1983; Mitchell and Schmidt, 1983) originates in muscle receptors and is mediated by group III and IV afferents. The HR acceleration Correspondence to: M. Pokorski, Department of Neurophysiology, Polish Academy of Sciences Medical Research Center, 00-784 Warsaw, Dworkowa 3, Poland. 0034-5687/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
190
M. POKORSKIet al.
is accompanied by an hyperpnea (Paulev, 1971, 1973; Muza et al., 1983), whose origin and the contribution to it of chemical drive, mediated by the arterial chemoreceptors, are uncertain. The hyperpnea of exercise is known to be linked to metabolic rate changes. Close matching of the respiratory and cardiovascular events, yielding constant arterial blood gas content has been the mainstay of the cardiodynamic concept, linking the rapid rise in ventilation atthe onset of exercise to increased 0 (Wasserman et al., 1974). This concept has never been unequivocally proved, nor is it ubiquitously accepted (Adams et al., 1987). In view of the controversial mechanisms, we designed the present study in an attempt to gain further insight into the reflex cardiorespiratory output events. We hypothesized that isometric contraction of small, but providing an intense receptor stimulus, muscle group would be likely to resolve aspects of the interplay among heart rate, cardiac output, peripheral neurogenic drive, and chemical drive in the mechanism of exercise hyperpnea. We therefore used a handgrip test, in contrast to the previous work, on a background of either hypoxia or hyperoxia, i.e., with enhanced or suppressed arterial chemoreceptor activity, in the same subjects. The specific purpose of the study was to characterize the ventilatory and cardiovascular responses to step changes from rest to static handgfip exercise, and to determine the role of chemicai drive in these responses. Overall, we found that the increase in ventilation was unaccompanied by appreciable changes in t) and that inactivation of the arterial chemoreceptors had a substantial inhibitory effect on the response. The study emphasizes the importance of chemical drive and shows that mechanisms other than those involving a cardiac link predominate in the ventilatory response to handgrip exercise.
Methods
The study was performed on 10 conscious, healthy, fit, middle-aged men. They consented to study procedure after having its general nature explained. The study was approved by an institutional ethics committee. The subject was connected with a mouthpiece and noseclip to a breathing circuit consisting of an inspiratory and expiratory line separated by a one-way Lloyd valve. He breathed from a bag containing a humidified hypoxic or hyperoxic gas mixture, produced manually by adding O 2 or N2 to air. End-tidal 0 2 , and CO2 (pE ¢,o : ) pressures were measured with a San-ei IH2I rapidly re(PEo2) sponding 0 2 and CO2 analyzer in a sample drawn continuously from the valve. The analyzer was calibrated with predetermined gas concentrations. Breath-by-breath inspired airflow was measured with a hot wire flowmeter (Minato RF-H), which was inserted between the mouthpiece and the valve and calibrated with a 2 L piston. Breath tidal volume (VT) was integrated from the flow signal, and inspiratory (TI) and expiratory (TE) times computed by an analog calculator. Total respiratory cycle time Recording procedure.
VENTILATION IN STATIC EXERCISE
191
(TT) and breath frequency (f = 60/TT) were calculated. From the above variables instantaneous inspiratory minute ventilation (VI = VT X O, mean inspiratory flow (VT/TI, and fractional inspiratory time (TI/TT) were calculated. The cardiac variables heart rate (HR), stroke volume (SV) and cardiac output (0) were measured with a Minnesota impedance cardiograph. The impedance signal was ensemble-averaged every 5 heart beats to improve the time resolution of the measurement. The HR was additionally recorded along with arterial oxygen saturation with an ear oximeter (Ohmeda, Biox III). Oxygen uptake (~'o2) and C O 2 output (Vco2) were calculated before and in the latter half of the exercise period (fig. 1) from the volume of expired gas and 02 and CO2 fractional concentrations in it measured with a San-ei analyzer. Arterial blood pressure was measured at the arm with a Riva-Rocci cuff manometer at the same time points plus during the first 20 sec after the end of exercise (fig. 1). Total perip.heral vascular resistance was calculated as the ratio of mean arterial pressure over Q. All continuously recorded variables along with the force of the handgrip concentration, performed with a dynamometer (SPR-655, Sakai), were displayed on a multi-channel strip-chart hot stylus recorder (San-ei Instruments).
Experimentalprotocol. The handgrip was performed with the fight hand in the upright sitting posture. The experiment started from the determination of maximal voluntary contraction (MVC), taken as the highest value achieved in 2 trials. The mean MVC was 38.6 + 1.8 kg. The subject was then instructed to develop a grip tension of about 50~o his MVC and maintain it as closely stable as possible by observing the tension trace at the oscilloscope screen for 50-60 sec. The mean tension developed was 47 + 1~o MVC in hypoxia and 48 + 1 ~o MVC in hyperoxia. Tensions above 10-25 ~o MCV result in fatigue, limiting the exercise period and making steady-state responses unattainable (Asmus sen, 1981). To minimize the potentially confounding influence of fatigue, 50 sec was taken as a cutoff point in the handgrip test. Each subject was interrogated after the test about his subjective feeling of effort to maintain the grip tension. The subject was connected to the breathing circuit for familiarization with the experimental setup while breathing room air. At this stage, to compare later the control resting hypoxic and
CONTROL PERIOD
HANDGRIPEXERCISE
TIME { s }
0
50
I
IJ}lll,,, BIN I:I~II
0
1
2
3
4
5
RECE~Y PERIOD
6
I 7
i 8
Meosurerneet of blood pressure
CoIiect~ ~' expired gas for 5 breaths in order to ~Ic,,,,aote the
V02 and VCO2
Fig. l. Diagram of experimental procedure. See text for details.
192
M. POKORSKI et al.
hyperoxic ventilations while having only one chemical stimulus changed, PESo 2 was set a few mm Hg above the spontaneous normoxic value to a level of 41 + 1 mm Hg. This initial PESo 2 increase allowed the possibility to lower it when it slightly increased on switching to hyperoxia and to hold it constant in the control periods of both hypoxia and hyperoxia. The PESo ~ was adjusted by closing off the expiratory line and allowing the exhaled air with its higher CO 2 content to be rebreathed partially straightaway and partially after passing through a CO 2 scrubber connected in parallel to the circuit. Once the handgrip started, no attempt at further PESo ~ adjustments was made. The protocol consisted then of two sequences of the handgrip run, one in hypoxia ( P E ~ - 47 + 2 mm Hg) and one in hyperoxia ( P E ~ = 216 + 3 mm Hg). The order of gas condition was randomized. The handgrip always started from the end of expiration at a verbal signal and was preceded by at least a 5 min control period of breathing a given gas mixture on the circuit. The handgrip runs were separated by a 6-8 min rest period of breathing room air while remaining connected to the circuit. All data were collected off-line from the stripchart recording and stored in a computer (NEC PC-9801) for further computations. The mean value of each variable in the period of 10 full breath cycles immediately preceding the exercise run was taken as control (bin 0). The handgrip exercise period was divided into five 10 sec bins, and the recovery period into three 10 sec bins (fig. 1). The sequential bins were coded in arabic numerals from 1 to 8. The breath-by-breath data were averaged over each bin by the computer. The averaging was performed to minimize the effect of the breath-tobreath fluctuations on the response pattern. Means + SE of bins with equal sequential numbers were calculated for each variable. Comparisons of the means for the ten subjects and across the nine time bins covering three experimental conditions: rest, exercise, recovery were made with a random block analysis of variance. If the analysis showed significant differences among the means (F-ratio exceeding the critical Fo.os value), they were further compared with a paired t-test and the Bonferroni method of multiple comparisons. Since 10 pairwise comparisons out of the possible 36 among the 9 bins were of interest, a Bonferroni correction of 10 was used. Differences between two mean values of bins with equal sequential numbers in hypoxia and hyperoxia were determined with a paired t-test. P < 0.05 was deemed significant. Data treatment.
Results
Handgrip exercise induced a substantial change in metabolic rate at both 0 2 levels, which was evident from the increases in "V'o2and 'V'co2 (fig. 2). These increases were smaller in hyperoxia than in hypoxia: AVo2 was 0.105 + 0.04 and 0.144 + 0.05 L . m i n -1, and A~'co ~ was 0.128 + 0.04 and 0.152 + 0.06 L . m i n -~, respectively. These differences were insignificant (P > 0.05). The ventilatory and cardiac responses studied did not attain steady state within the 50 sec duration of the exercise bout. The responses are summarized in fig. 3 which shows the time course of the average
VENTILATION IN STATIC EXERCISE
HYPERCIKIA
H~'F~
I ~
,,
I 02
J 03 0,~ 02 Uptoke {I STPD. rmn"1 }
-
193
REST EXERCISE
" 0.5
t HYPC~
0,2
03 0.4 CO20~tpu! {I STPO.min-~)
0.5
Fig. 2. Metabolic rate changes in response to static handgrip exercise. Differencesbetween changes in hypoxia and hyperoxiawere not significant. *P < 0.05 vs correspondingrest (paired t-test). ventllatory and cardiac variables at sequential 10 sec bins of the handgrip test. The characteristics of the responses differed for the two gas conditions. Ventilatory response. The ~/I was significantly higher in hypoxia during the control rest period (P < 0.05; fig. 3). The general pattern of the ventilatory response to handgrip exercise was one of an abrupt rise in VI followed by a further gradual increase. The mean absolute VI rise in the first 10 sec (bin 1) was significantly higher in hypoxia than in hyperoxia (P = 0.05; fig. 4). The early component accounted, however, for a similar fraction of the total ~/I increment concluding the hypoxic and hyperoxic exercise runs; 81 ~o (range 17-176Yo) and 81 ~ (range 6-209~o), respectively. The difference between bin 1 VI increases (AVI) in hypoxia and hyperoxia amounted to 3.8 + 2.0 L. min - 1 (fig. 4). The difference rose progressively over the exercise range, reaching 9.0 + 2.5 L- min - 1 towards its end, i.e., bin 5. The rise of A~h was due largely to a gradual VI increase in hypoxia, which was virtually absent in hyperoxia. A least squares regression analysis applied to the data of bin 1-5 showed that the slope of the ~'I response was significant in hypoxia: 0.84 L. min - 1 per bin, r - 0.95, whereas that in hyperoxia was not: - 0.01 L. min - 1 per bin, r = - 0.03. The hypoxic and hyperoxic ~/I increments were due mostly to increases in the tidal component. The VT increased abruptly at the transition from rest to exercise and remained stable throughout the exercise bout at both O 2 levels. The increase was significantly higher in hypoxia (P < 0.05; fig. 3) and was associated with a tendency for f to increase, which was absent in hyperoxia. The analysis of VT/TI shows that respiratory drive was significantly higher in hypoxia than in hyperoxia over the range of handgrip exercise (fig. 3). The modulatory effect of chemical drive is also seen when one examines the components of breath timing. This is depicted in fig. 5 where sequential bin TE is plotted against its corresponding TI. Whereas hypoxia increased f shortening both TE and TI, hyperoxia tended to decrease f lengthening TE with no change in TI and correspondingly lower TI/TT.
194
M. POKORSKI
A
et al.
~
!I! n
.~
20
_-
d," ~" ~" ~"
~J ~ " ~ ' ~ ,
,~
q, ~' ~"
mL~'1
<~'~
,~_~,
[
2.5
"
"'
.
.
.
40'
~ ~
.
Q5 ~ 1.0
Time(s) Bin#
.
.
.
.
o
.
J
.I .....
I
,
i
c> ~" c~. ~. Cl C¢' #', ~' 0
=0.
.
i
Z>' ~"
C 010 01
0
0.3
1 ~'~, 30
5010
2345
678
EXERCISE
30
Tmae{s) Bin#
..... 0
10
.
30
..
,__,
50 10
30
0 1 2 3 4 5 6 ? 8
RECOVERY
EXERCISE
RBOCNERY
Fig. 3. Average minute ventilation ('~/I), frequency (f), tidal volume (VT), mean inspiratory flow (VT/TI), fractional inspiratory time (TI/TT), cardiac output ((~), heart rate (HR), and stroke volume (SV) for 10 subjects at rest (control), at sequential 10 see bins ofhandgrip exercise, and during recovery in hypoxia (squares) and hyperoxia (triangles). The control is coded bin 0, the first 10 sec of exercise bin 1, etc. Onand off-transitions are symbolized with heavy vertical lines. Bars on symbols are SE; lack of bar denotes SE smaller than the symbol size. The overall F-test indicated no differences among the means of the f, (~, and TI/TT. *P < 0.05 vs bin 0 in either gas condition (F-test and Bonferroni's multiple comparisons); i" P < 0.05 vs. corresponding hypoxic bin (paired t-test); ~.P < 0.05 vs bin 5 (F-test and Bonferroni's multiple comparisons).
HYPEROXlA
Or'I
HYPOXIA
Q i
0
I' I
5 A t. rnin -1
J
10
Fig. 4, Concurrent early (bin 1) increments in Vi (open bars) and (~ (solid bars) in response to a step change from rest to handgrip exercise in hypoxia and hyperoxia. Values are means + SE. * ~'I increment in hypoxia was significantly higher than that in hyperoxia (P = 0.05; paired t-test). Both were associated with unappreciable changes in (~.
195
V E N T I L A T I O N IN STATIC EXERCISE 5 10 xx
/
0.3 T,/Tr
x x / 4 •f
/
/ ~'x
0.4
,
/ "-, -
/
/
IZ] HYPOXIA A HYPEROXlA
1
"x
I
I
2
3
Tz
(see)
Fig. 5. Average expiratory time (TE) plotted against corresponding inspiratory time (TI) at sequential bins of handgrip exercise. Solid symbols represent control values at rest. SE bars are omitted for clarity. Dashed lines are frequency isolines, and continuous lines are TI/TT isolines,
The ventilatory response was accompanied by little fluctuations in PESo 2when 10 sec averages were considered (table 1). The PESo 2 remained closely at its prior control level over the exercise test at both 0 2 levels. A tendency to hyperventilation in hypoxia with TABLE 1 Average data for end-tidal gas tensions for the control state and sequential 10 sec bins of static exercise during hypoxia and hyperoxia. Hypoxia
Rest Exercise
Recovery
Hyperoxia
Bin No.
PE~2 ( m m Hg)
PESo 2 (mm Hg)
PE~2 ( m m Hg)
PESo 2 (mm Hg)
0 1 2 3 4 5 6 7 8
47.3 48.1 49.4 49.7 50.0 45.2 52.6 49.8 46.7
40.9 41.0 40.8 40.9 40.3 40.1 39.5 40.8 41.5
216.0 218.4 216.9 214.5 212.7 212.3 215.9 215.0 212.3
40.5 40.2 40.4 40.6 41.3 41.9 39.0 40.5 41.1
+ 1.5 + 2,1 + 2,8 + 3,3 + 3.9 + 3.3 + 3.0 ___2.3 _+ 2.2
+ + + + + + + + +
0.8 1.1 1.3 1.3 1.5 1.6 1.6 1.5 1.3
+ 3.1 + 3.6 + 3.6 + 3.8 + 4.3 _+ 6.0 + 4.6 + 4.8 + 5.0
+ 0.8 + 0.9 + 1.0 + 1.2 + 1.4 + 2.0 _+ 1.5 + 1.1 + 1.1
Values are means + SE for 10 subjects. F-test showed insignificant differences among the m e a n s of either end-tidal gas tension in hypoxia and in hyperoxia,
196
M. POKORSKI et al.
PE~2 increasing and PESo 2decreasing, and the converse in hyperoxia were seen towards the end of the exercise bout. In the recovery period the variables were gradually returning toward the preexercise values in about 30 see. The characteristic feature of VI was that it kept on increasing for some 10 sec after cessation of exercise. Cardiovascular response. The control resting heart rate and cardiac output were significantly lower in hyperoxia than in hypoxia (fig. 3). The cardiac response was qualitatively different from that described above for ventilation. The onset of isometric hand muscle activity was accompanied by an abrupt HR acceleration at both O2 levels. The magnitude of the early HR rise was not affected by the 0 2 level. It amounted to 8.5 + 2.2 beats- rain- 1 in hypoxia and 9.4 + 2.3 beats, rain - ~ in hyperoxia (P < 0.05), which accounted for 61 ~ (range 12-193 ~o) and 63 ~o (range 20-1119'o), respectively, of the HR increment concluding the exercise bout. The HR remained lower in hyperoxia over the exercise range in proportion to the difference present in the control period (fig. 3). It was steadily increasing starting from the initial acceleration at both 0 2 levels. Regression analysis applied to the data of bin 1-5 revealed similar and significant slopes of the HR responses in hypoxia and hyperoxia; 1.7 beats, min- ~ per bin, r = 0.97 and 1.9 beats, min - 1 per bin, r = 0.94, respectively. Changes in H R were associated with decreases in SV, which were greater in hyperoxia. As a result, Q did not change appreciably over the exercise range at both 0 2 levels. In fact, small reductions in (~ inclusive of the time period corresponding to the early VI increase were noted (fig. 3). In addition to the above cardiac alterations, handgrip exercise resulted in pronounced increases in total peripheral resistance and mean arterial blood pressure (fig. 6). The HR reverted abruptly to the resting level at the off-transition with no overshoot characteristics.
Linear correlation and regression analysis revealed no relationship between the initial ~/I increase and the corresponding HR increase, 0 alterations, control HR and control V1 at either 02 level. Particularly, this "v'xincrease was unaccompanied by an increase in 0 (fig. 4). When the whole exercise range was examined, a positive correlation emerged in hypoxia between the size of the sequential bin WI increment and the corresponding H R increment (r = 0.86; fig. 7). The good correlation disappeared in hyperoxia. Cardiorespiratory interplay.
Discussion
This study demonstrated that ventilation during static exercise in man was not linked to a cardiac output mediated stimulus and that hypoxic drive augmented the dynamics of the ventilatory response. The hyperpneic response unaccompanied by a rise in (~ was rather unexpected in the light of the cardiodynamic theory (Wasserman et al., 1974). This theory presumes that the increase of ~/I is coupled to a rapid increase of (~ at the onset of exercise, which
197
VENTILATION IN STATIC EXERCISE '-"
i REST I---I EXEROSE l ~ RECOVERY
~.. •
HYPEROXIA
i
//////L.
I
I
90
f
I - -
HYPEROXIA
I
110 130 Mean Arterial Pressure (mmHg)
150
1
/////A~
I---
i
~,
I
18 24 Tofol Per/pherol ~scul~,
I
I
30
36
ResistoncefmrnHcj.14.rr~)
Fig. 6, Mean arterial pressure (panel A) and peripheral vascular resistance (panel B) (means + SE) in consecutive periods ofhandgrip procedure in hypoxia and hyperoxia. *P < 0.05 vs rest and recovery (F-test and Bonferroni's multiple comparisons).
t0 i___ E
n
-
/
i'i HYI:~XIA A HYPEROXIA I
0
10
, I
20
6 HR ( bears - min'lJ
Fig. 7. Relationship between the mean responses of Vl and HR to hypoxic and hyperoxic handgrip exercise. Symbols represent increments above average resting value at sequential bins of exercise for 10 subjects. Lines are corresponding linear regressions. The slope of ~'l on HR in hypoxia is significant (P < 0.05).
would explain the proportionality of the ventilatory response to the gas exchange at the lung yielding constant blood gas tensions. Our findings argue against the cardiac mechanism being operative in increasing VI and lend support for those few reports which show that t) rise is not a prerequisite for '~'t rise. Huszczuk et al. (1986) exercised
198
M. POKORSKI et al.
calves with the Jarvik artificial heart implant, maintaining a constant Q, and found an isocapnic hyperpneic response similar to that when Q was allowed to increase. Banner et al. (1988) reported a dissociation of the ventilatory and cardiac responses to the onset of dynamic or electrically induced leg exercise in patients with heart or heart-lung transplants. The patients had a normal ~/I response, although (~ and H R did not change appreciably. Adams et al. (1987) found that the ventilatory response virtually preceded the rise of 0 in man. A parallel but not coupled stimulus to both respiratory and cardiovascular events is conceivable, which may affect the central controllers of both functions with disparate input kinetics or may not affect them concurrently at all. The initial ~/I increase would then be an independent phenomenon of cardiac events. The good correlation found between the magnitudes of the ~'I rise and the HR acceleration in hypoxia reflects likely a fortuitous association, since we noted important disparities. Hyperoxia had an inhibitory effect on the VI response, but it did not affect the H R response, causing a loss of the correlation between the two. At cessation of exercise the ~q increase continued for some time, but HR declined abruptly. In the present study on conscious man both descending command from the rostral brain and peripheral neurogenic drive could be operative in controlling the cardiorespiratory events. Central command alone can be an initiator of the cardiovascular and ventilatory responses (Eldridge and Millhorn, 1986). Our subjects did not report an increased perception of effort in the short exercise period studied. That the handgrip was not fatiguing enough to substantially activate central command may also be judged from the constancy of PESo 2, showing that no hyperventilation exceeding metabolic needs developed. Anticipatory or conditioning process to the repeat verbal signal to exercise was unlikely, since the initial hyperpnea was also present when the subject started handgrip without any prior signal. Peripheral neurogenic drive ascends from the muscle receptors known to respond to noxious mechanical and chemical stimulation (Mense, 1977; Kumazawa and Mizumura, 1977), a feature of isometric contraction. Feedback from muscle receptors mediated by group III and IV afferents has been recognized as an inducer of both exercise hyperpnea (Senapati, 1966; Kalia et al., 1972) and HR acceleration (Gelsema e t a l . , 1983). The disappearance of hyperpnea after anesthetic blockade of these afferents (McCloskey and Mitchell, 1972) strongly supports the peripheral neurogenic mechanism. An extended ventilatory response after cessation of exercise we found accords with the observation of Senapati (1966) made during stimulation of group III and IV afferents from the exercising muscles. The part played by peripheral or central drive could not be discerned in our study. The role of the arterial chemoreceptors in the ventilatory response to static exercise is imprecisely defined. Muza et al. (1983) reported responses similar to ours at the beginning of handgrip of 25 ~o MVC sustained for 5 rain in hyperoxia. Similar responses were also observed with 50~ MVC in normoxia by Myhre and Andersen (1971). This is not unexpected, given the little effect on ventilation of carotid chemoreceptors in the normoxic range. We have found no report of comparison of the ventilatory responses to static exercise at enhanced and abolished chemoreceptor activity. In the present study
VENTILATION IN STATIC EXERCISE
199
we observed a substantial augmenting influence of carotid chemoreceptor activation on the ventilatory response to static exercise. The mechanisms of such influence may differ at different stages of exercise and are not easy to explain. The early V1 rise was present at both 02 levels, which implies that the basic hyperpneic stimulus is extrachemoreceptor. This rise was, however, diminished when carotid chemoreceptor drive was absent in hyperoxia. The difference between the hypoxic and hyperoxic VI rise could thus point to the effect mediated by the arterial chemoreceptors. How they would do this is unknown. The rapidity of the initial VI rise and a fair stability of PESo: makes a feedback correction of a downstream chemical signal error improbable; an action that is likely later in the course of exercise. The arterial chemoreceptors assumed greater importance over time, which was evident from the inhibitory effect of hyperoxia on the progressive "~I increase seen in hypoxia. The carotid chemoreceptor involvement in the ventilatory response might have to do with a rapid release of potassium from the exercising muscles (Kilburn, 1966; Lim et al., 1981; Laurell and Pernov, 1966). Arterial plasma potassium concentration and ventilation are closely related in exercise (Newstead, 1988; Conway et al., 1988). Exogenously administered potassium in a dose matching that produced endogenously in moderate dynamic exercise stimulates both afferent carotid chemoreceptor activity and ventilation (Linton and Band, 1985). The potassium effect on the chemoreceptor activity is potentiated by hypoxia (Burger et al., 1986; Band and Linton, 1987), which might explain the potentiation of the hypoxic ventilatory response to static exercise in our study. Sneyd et al. (1988) found, however, no difference in the ventilatory effects of potassium during hypoxia, normoxia, and hyperoxia. We did not measure the arterial plasma concentration of potassium. If measured, it would have been difficult to ascertain to what extent it reflected the local concentration of potassium, since an appreciable negative arterio-venous difference exists and is augmented on exercise (Laurell and Pernov, 1966). While this study has not resolved the much debated mechanism of exercise hyperpnea, we believe we have shown that the ventilatory response to a step change from rest to static handgrip exercise is not cardiodynamic. The origin of the response is extrachemoreceptor, although the arterial chemoreceptors play an increasing role in its course. We conclude that the ventilatory and cardiovascular output events were driven in parallel. Further studies are needed to explore the proposed mechanism.
Acknowledgements. M. Pokorski and P-E. Paulev were Visiting Professors from the Department of Neurophysiology,Polish Academyof Sciences Medial Research Center, 00-784 Warsaw, 3, Dworkowa, Poland, and the Department of Physiology,Sports/Cybernetics,Universityof Copenhagen,Panum4: 2: 7, 2200 Copenhagen N., Denmark, recipients of Japan Society for the Promotion of Science and of Scandinavian-Japan Sasakawa FoundationFellowships,respectively.
200
M. POKORSKI et al.
References Adams, L., A. Guz, J.A. Innes and K. Murphy (1987). The early circulatory and ventilatory response to voluntary and electrically induced exercise in man. J. Physiol. (London) 383: 19-30. Asmussen, E. (1981). Similarities and dissimilarities between static and dynamic exercise. Circ. Res. 48 (Suppl. I): 3-10. Band, D.M. and R. A.F. Linton (1987). The interaction of hypoxia and K + on the carotid chemoreceptor in anesthetized cats. J. Physiol. (London) 394: 65P. Banner, N., A. Guz, R. Heaton, J. A. limes, K. Murphy and M. Yacoub (1988). Ventilatory and circulatory responses at the onset of exercise in man following heart or heart-lung transplantation. J. Physiol. (London) 399: 437-449. Burger, R.E., J.A. Estavillo, P. Kumar and P.C.G. Nye (1986). The excitation of cat carotid body chemoreceptors by hyperkalaemia depends on Po2 and Pco2. J. Physiol. (London) 374: 25P. Conway, J., D.J. Paterson, E.S. Petersen and P.A. Robbins (1988). Changes in arterial potassium and ventilation in response to exercise in humans. J, Physiol. (London) 399: 36P. Eldridge, F.L. and D.E. MiUhorn (1986). Hypothalamic central command and exercise hyperpnea. In: Neurobiology of the Control of Breathing, edited by C. yon Euler and H. Lagercrantz. New York: Raven Press, pp. 35-43. Gelsema, A.J., G. de Groot and L.N. Bouman (1983). Instantaneous cardiac acceleration in the cat elicited by peripheral nerve stimulation. J. Appl. PhysioL 55: 703-710. Huszczuk, A., B.J. Whipp, K. Wasserman, T.D. Adams, A.G. Fisher, D.B. Olsen, R.O. Crapo and C.G. Elliott (1986). Ventilatory control during exercise in calves with artificial (Jarvik) hearts. Fed. Proc. 45: 1127. Kalia, M., J. M. Senapati, B. Parida and A. Panda (1972). Reflex increase in ventilation by muscle receptors with nonmeduUated fibers (C fibers). J. Appl. Physiol. 32: 189-193. Kilburn, K.H. (1966). Muscular origin of elevated plasma potassium during exercise. J. Appl. Physiol. 21: 675-678. Kumazawa, T. and K. Mizumura (1977). Thin-fiber receptors responding to mechanical, chemical, and thermal stimulation in the skeletal muscle of the dog. J. Physiol. (London) 273: 179-194. Laurell, H. and B. Pernow (1966). Effect of exercise on plasma potassium in man. Acta Physiol. Scand. 66: 241-242. Lim, M., R.A.F. Linton, C.B. Wolff and D.M. Band (1981). Propranolol, exercise, and arterial plasma potassium. Lancet 2: 591. Linton, R.A.F. and D.M. Band (1985). The effect of potassium on carotid chemoreceptor activity and ventilation in the cat. Respir. Physiol. 59: 65-70. Martin, C.E., J.A. Shaver, D.F. Leon, M.E. Thompson, P. S. Reddy and J.J. Leonard (1974). Autonomic mechanisms in hemodynamic responses to isometric exercise. J. Clin. Invest. 54:104-115. McCloskey, D.I. and J.H. Mitchell (1972), Reflex cardiovascular and respiratory responses originating in exercising muscle. J. Physiol. (London) 224: 173-186. Mense, S. (1977), Muscular nociceptors. J. Physiol. (Paris) 73: 233-240. Mitchell, J. H. and R. F. Schmidt (1983). Cardiovascular reflex control by afferent fibers from skeletal muscle receptors. In: Handbook of Physiology. Section 2. The Cardiovascular System. Vol. III, Peripheral Circulation, edited by the American Physiological Society: Bethesda, MD, pp. 623-658. Muza, S. R., L.Y. Lee, R. L. Wiley, S. McDonald and F.W. Zechman (1983). Ventilatory responses to static handgrip exercise. J. Appl. Physiol. 54: 1457-1462. Myhre, K. and K.L. Andersen (197 i). Respiratory responses to static muscular work. Respir. Physiol. 12: 77-89. Newstead, C. G. (1988), The relationship between arterial potassium and ventilation during exercise in man. J. Physiol. (London) 403: 101P. Paulev, P-E. (1971), Respiratory and cardiac responses to exercise in man. J. Appl. Physiol. 30: 165-172.
VENTILATION IN STATIC EXERCISE
201
Paulev, P-E. (1973). Cardiac rate and ventilatory volume rate reactions to a muscle contraction in man. J. Appl. Physiol. 34: 578-583. Senapati, J.M. (1966). Effect of stimulation of muscle afferents on ventilation of dogs. J. Appl. Physiol. 21: 242-264. Sneyd, J.R., R.A.F. Linton and D.M. Band (1988). Ventilatory effects of potassium during hyperoxia, normoxia and hypoxia in anaesthetized cats. Respir. Physiol. 72: 59-64. Wasserman, K., B.J. Whipp and J. Castagna (1974). Cardiodynamie hyperpnea: hyperpnea secondary to cardiac output. J. Appl. Physiol. 36: 457-464.