- Email: [email protected]
The effect of peripheral chemodenervation on the ventilatory response to potassium
217
Respiration Physiology (1985) 60, 217-225 Elsevier
THE EFFECT OF PERIPHERAL CHEMODENERVATION ON THE VENTILATORY RESPONSE TO POTASSIUM
D.M. BAND, R.A.F. LINTON, R. KENT and F.L. KURER Laboratory of Applied Physiology, St. Thomas's Hospital Medical School, London SE1 7EH, U.K.
Abstract. The present experiments were designed to test the hypothesis that the increase in plasma potassium which occurs during exercise acts as a stimulus to respiration via the peripheral chemoreceptors. The effect of intravenous infusion of KC1 on ventilation was measured in anaesthetised cats while they were loaded with CO2 intravenously via a bubble gas exchanger, Ventilation during K + infusion was compared with that immediately before in 'intact' and peripherally chemodenervated cats. In the 'intact' group there was a highly significant increase in ventilation of approximately 25~o ( + 253 + 22 ml/min, P < 0.001), whereas in the chemodenervated grou p there was no significant change (+ 17 _+ 11 ml/min) in spite of similar increases in arterial K ÷ concentration. The results of these experiments indicate that K ÷ infusions stimulate ventilation and that this effect is abolished by peripheral chemodenervation.
Arterial chemoreceptors Cat Chemodenervation
Chemoreflexes Control of breathing Exercise
Hyperkalemia Potassium Ventilation
At the start of exercise there is a rapid rise in the concentration of plasma potassium due to its release from contracting muscle (Laurell and Pernow, 1966; Kilburn, 1966; Van Beaumont et al., 1973). In man, the magnitude and time course of the changes in arterial plasma potassium produced by moderate exercise have been demonstrated (Linton et al., 1984). The carotid body is sensitive to intra-arterial injections of KCI (Jarisch et al., 1952), and we have shown in anaesthetised cats that it is sensitive to transient changes of plasma potassium within the range of those produced by exercise in man (Linton and Band, 1985). The hypothesis that the rise in arterial potassium in exercise may form part of the chemical stimulus to ventilation under these circumstances has therefore to be considered. Brief injections of KCI as a test stimulus are unphysiological in that the arterial plasma potassium rises very rapidly and the duration of the stimulus is only for one or Accepted for publication 1 March 1985
218
POTASSIUM AND PERIPHERAL CHEMODENERVATION
two breaths. This is unlike the sustained rise which occurs during exercise in man (Linton et al., 1984). Since there may be fast adaptation by the chemoreceptor to elevated levels of potassium, we have tested the response to more sustained changes produced by infusions of KC1. In order to simulate the exercise changes more closely and to reduce the masking effect of changes in the gas tensions resulting from any response, KC1 infusions were given against a background of intravenous CO 2 loading. Unfortunately it was not technically possible to mimick the onset transient of CO 2 production with the gas exchanger. The ventilatory responses to brief potassium injections, as used previously (Linton and Band, 1985), diminish with time of anaesthesia, so that abolition of the response by peripheral chemodenervation would not provide conclusive evidence that it was mediated by the peripheral chemoreceptors. Under the conditions of the present experiments the ventilatory response to potassium persisted for many hours, allowing us to test the effect of peripheral chemodenervation.
Methods Cats weighing 2.4-3.7 kg were anaesthetised with pentobarbitone, 40 mg/kg intraperitoneally and supplements intravenously as necessary. A tracheostomy cannula was inserted and connected to a pneumotachograph; the integrated flow signal was recorded as inspiratory and expiratory tidal volumes. The left axillary artery and vein were cannulated for the measurement of blood pressure and infusion of KC1 respectively. A pH electrode catheter was inserted into the upper abdominal aorta via the right renal artery. The pH electrodes used were similar in design to the K + electrodes previously described (Linton et al., 1982) but with membranes containing tri-N-dodecylamine to make them selective for the hydrogen ion (Schulthess et al., 1981). An external Ag/AgC1 reference electrode was connected to a saline-filled catheter inserted via a femoral vein so that its tip made electrical contact with the blood in the inferior vena cava and lay close to the sensing tip of the pH electrode in the adjacent aorta. This arrangement reduces the tendency to record an ECG artefact on the pH trace. The pH electrode catheters have exceptional discrimination for small changes in pH and follow the breath-by-breath fluctuations which correspond to the alveolar CO2 fluctuations induced by tidal breathing. Changes in mean pH can be measured from the records to a precision beyond the discrimination of an in vitro electrode system. A large-bore cannula was tied into the abdominal aorta just above its bifurcation occluding distal flow. This was connected to a bubble gas exchanger consisting of a vertical column, defoaming chamber and reservoir. Gas passed through a perforated Perspex disc in the base of the column. Arterial blood flowed via a side arm into the bottom of the column where it was lifted up the column by the bubbles of gas. The blood then passed along a defoaming chamber, lined with silicone grease, to the reservoir. From here it was pumped at 25 ml/min by an electrically driven roller pump through a 40 #m screen filter (Pall, Ultipor) and returned to the inferior vena cava. The level
D.M. BAND etal.
219
of blood in the reservoir was kept constant by adjusting a screw clamp on the arterial input tubing. When the exchanger was gassed with C O / a n approximate doubling of the animal's CO 2 'production' resulted. The gas exchanger was primed with Hartmann's solution and the cats were given 3000 units of heparin before the circuit was opened and 1000 units hourIy thereafter. Arterial K + concentration was measured continuously using a K +-selective electrode, similar to that previously described (Linton et al., 1982) positioned in the arterial flow to the gas exchanger. Airway CO z concentration was measured with a mass spectrometer (Medishield MS2). Traces were recorded on a Gould ES 1000 electrostatic recorder. The gas exchanger was initially bubbled with nitrogen. When the Hartmann's solution had washed out of the circuit and the cat was stable, the gas exchanger was bubbled with COz. At least 2 min were required for the CO 2 load to stabilise. An infusion of KC1 was then started (5 ml/min for 0.5 min of 150 mmol/L). The effects of the infusions of KC1 were studied in two groups of cats. In the first group ('intact') 6 responses were measured in 5 cats. In cats 1 and 2 both aortic and carotid sinus nerves were intact; in cats 3, 4 and 5 the aortic nerves were cut and fine cannulae were positioned close to the carotid sinus nerves which were intact. In the second group (chemodenervated) 5 responses were measured in 4 cats which had been peripherally chemodenervated. In cats 3, 4 and 5 this was achieved by injection of lignocaine around the carotid sinus nerves via the cannulae which had been positioned prior to the measurement of the 'intact' responses. In this way the time interval between the 'intact' and chemodenervated responses could be kept as short as possible. In cat 6 both aortic and carotid sinus nerves were cut. Peripheral chemodenervation was considered to be complete if there was no increase in ventilation after administering 4 breaths of nitrogen.
Results Results are shown as mean values + SEM. The significance of the difference between mean values was assessed within groups using Student's t-test for paired data and between groups using Student's t-test for unpaired data. Figure 1 shows an 'intact' response and the results for the 'intact' group are given in table 1. An increase in arterial potassium of 3.9 + 0.5 mmol/L caused a highly significant increase in ventilation (253 + 22 ml/min, P < 0.001), which in turn produced a significant increase in mean arterial pH (0.023 + 0.008, P < 0.05). Figure 2 shows the experiment repeated in cat 4 after injection of lignocaine around the sinus nerves, and the results for the chemodenervated group are shown in table 2. There was no significant change in ventilation ( + 1 7 + l l m l / m i n ) or pH ( - 0.003 + 0.003) in spite of an increase in arterial K + concentration similar to that in the 'intact' group.
220
POTASSIUM AND PERIPHERAL C H E M O D E N E R V A T I O N
]100
7"32°[A pH
8
//j/
7"3601
7.400c' t~
Jll.,'//J///,J
1
K,,I-
i1 °
(mmol.t "l}
2 . . . .
! I li! i!i
'
IIU
I' ,!
,,
I ~i~,hr i
,!~ ,i,i I..... i
,,
i ! ~il
i!~l
10 sec
J
S.0 A I RWAY C02 °/0 0
Fig. 1. K ÷ infusion during intravenous CO 2 loading in 'intact' cat 4. The traces from above downwards are arterial pH (becoming more alkaline as the trace falls), arterial K +, tidal volume and airway CO2 concentration. As the arterial K ÷ rises at the start of the infusion (KCI 0.75 mmol/min) there is an increase in the ventilation and the pH becomes more alkaline. When the infusion is switched off the arterial K ÷ falls, causing ventilation to decrease and the pH to become more acid.
TABLE 1 'Intact' group. The upper half of the table shows control values (measured over a 5-breath period) of minute volume (~z) and arterial K + concentration during intravenous CO 2 loading immediately prior to infusion of KCI. The lower half shows changes in minute volume (A~'), arterial K + (AK +) and arterial pH (ApH) measured over a 5-breath period 15 sec after the arterial K + started to rise. Six responses were measured in 5 cats. In cats 1 and 2 the aortic and sinus nerves were intact. In cats 3, 4 and 5 the aortic nerves were cut and cannulae were positioned close to the intact sinus nerves (for later injection of lignocaine). Cat
Mean + SEM 1
1
2
3
4
5
x~ (ml/min)
995
1080
1001
850
1225
930
K ÷ (mmol/L)
3.6
3.9
2.5
4.7
3.0
2.4
AV (ml/min)
+283
+ 172
+332
+241
+263
+228
AK + (mmol/L)
+ 2.9
+ 2.8
+ 4.4
+ 5.6
+ 4.9
+ 3.0
+ 0.024
+ 0.052
+ 0.018
--
+ 0.009
+ 0.011
C02
1014 +53 3.4 + 0.4
CO 2 + K +
ApH
253 + 22 + 3.9 +0.5 + 0.023 _+0.008
D.M. BAND etal.
221
~.,,o[% %VV'vlN'
pH
Lf l
A
tJ
f f!
',L/:/:/',/ /~:?/. 11:/./, (mmot.t_tl 4
:IIII} iti!II!ii!
, q
!
I I'~'i!
I,
!~lili
I
iF
i'i,
i;',
i;l'
'
C02 %
i
10 sec Fig. 2. K ÷ infusion during intravenous CO2 loading in chemodenervated cat 4. There is now no change in ventilation or arterial pH in response to hyperkalaemia. TABLE 2 Chemodenervated group. The layout of the table is the same as for table 1. Five responses were measured in 4 cats. Chemodenervation was achieved in cat 6 by cutting both sinus and aortic nerves, and in cats 3, 4 and 5 by injecting lignocaine around the sinus nerves via the cannulae positioned prior to measurement of the intact responses in these cats. Cat
Mean _+SEM 3
4
5
6
6
(ml/min)
981
799
930
977
702
K ÷ (mmol/L)
4.4
2.5
2.2
3.6
5.6
-5
+43
+44
-9
+12
+ 3.5
+ 4.4
+ 2.6
+ 4.0
+ 4.5
--
- 0.002
+ 0.004
- 0.006
- 0.009
C02
878 +55 3.7 +0.6
CO 2 + K +
AV (ml/min) AK
÷
ApH
(mmol/L)
+17 +11 + 3.8 _+0.3 - 0.003 _ 0.003
A c o m p a r i s o n o f t h e c h a n g e s i n K + a n d v e n t i l a t i o n in t h e t w o g r o u p s o f c a t s is s h o w n i n fig. 3. T h e level o f v e n t i l a t i o n p r i o r t o i n f u s i o n o f KC1 w a s s i g n i f i c a n t l y l o w e r in t h e c h e m o d e n e r v a t e d g r o u p ( P < 0.05) a n d t h e e n d - t i d a l P c o 2 w a s c o r r e s p o n d i n g l y h i g h e r ( ' i n t a c t ' = 29.3 + 2.0 m m H g ; c h e m o d e n e r v a t e d
-- 37.0 + 2.1 m m H g ; P < 0.001). I n
s o m e e x p e r i m e n t s m e a n v a l u e s o f v e n t i l a t i o n a n d e n d - t i d a l P c o 2 w e r e d e r i v e d for t h e
222
POTASSIUM AND PERIPHERAL CHEMODENERVATION 1~00 1300 1200 VENTILATION (ml.min"~)
1100 1000
÷
900 800 J
3
t~
5
6
7
9
+
ARTERIAL K
(mm01.1-~) Fig. 3. Comparison of ventilatory responses to hyperkalaemia during intravenous CO 2 loading in 'intact' (n = 6) and chemodenervated (n = 5) cats. Mean values + 1 SEM are shown for the 'intact' ( 0 ) and chemodenervated (O) groups. Similar increases in arterial K + produced a highly significant increase in ventilation in the 'intact' group (P < 0.001) and no significant change in the chemodenervated group.
1400 1200 I000
VENTILATION
800
(ml.min-I)
600
Y
400 22
24
26
28
30
32
34
36
38
END TIDAL PCO 2 (mmHg)
Fig. 4. The effect of intravenous CO2 loading and K ÷ infusion on ventilation and end-tidal Pco2. Mean values for 'intact' ( O ) and ehemodenervated cats (O) are shown. Arrows indicate the experimental sequence: exchanger gassed with N2 ~ exchanger gassed with CO 2 --*exchanger gassed with CO 2 and intravenous infusion of KC1. For the 'intact' group, n = 4 for the first point and n = 6 for the second and third points. For the chemodenervated group n = 3 for the first point and n = 5 for the second and third points.
p e r i o d w h e n the gas e x c h a n g e r w a s g a s s e d w i t h n i t r o g e n i m m e d i a t e l y prior to switching to C O 2 . S u b s e q u e n t c h a n g e s in t h e s e v a r i a b l e s p r o d u c e d by C O 2 l o a d i n g a n d t h e n K ÷ infusion (during C O 2 l o a d i n g ) are s h o w n in fig. 4. F i g u r e 5 s h o w s the effect o f i n t r a v e n o u s C O 2 l o a d i n g o n the arterial p H r e c o r d in cat 5. T h e g r a d i e n t ( p H / s e c ) o f the a c i d - g o i n g p a r t o f the cycle i n c r e a s e d by 206~o.
D.M. BAND et al.
-I pH.sec
0.0097
223
0.020
Fig. 5. The effectof intravenous CO2 loadingon the arterial pH record in cat 5. CO2 loading increases the slope of the pH oscillation by 206~o,implyinga similar increase in CO2 production.
Discussion
These experiments show that increasing the plasma potassium concentration by intravenous infusion of potassium chloride enhances the ventilatory response of the anaesthetised cat to intravenous CO= loading. This enhancement is mediated via the peripheral chemoreceptors. The effect persists for the 0.5 min period of the infusions used in the present experiments. Hyperkalaemia due to potassium effiux from exercising muscle may therefore be a contributing factor to the ventilatory response to exercise. There is controversy concerning the effects of intravenous CO z loading alone as a stimulus to ventilation. The technique has been used extensively as a means of testing the hypothesis that oscillations in Pa¢o~ provide a controlling signal related to CO 2 production (Yamamoto and Edwards, 1960). When ventilation is plotted as a function of P a c o 2 the response curves vary from more than vertical, i.e. the P a c o ~ actually falls, to curves indistinguishable from the classical response curves to inhaled CO 2 mixtures (cf. Linton et al., 1976; Phillipson et al., 1981; Bennett et aL, 1984). The present experiments confh'm that CO2 loading does change the profile of the oscillations in pH and therefore P a c o :. The pH oscillations shown in fig. 5 have an increased rate of change during CO2 loading and are very similar to traces obtained from man during the onset of exercise (Band et al., 1980). The pH records provide indirect confn'mation that an approximate doubling of the CO2 production of the animals was achieved by the bubble exchanger. There are two factors which make it difficult to draw conclusions from the ventilatory responses to CO 2 infusions alone in the present study. Firstly, no attempt was made to control the Pao:. Recent work (Cross et al., 1984) shows that the response of the carotid body chemoreceptor in the anaesthetised cat to the oscillations in CO2 is strongly depressed by slight hyperoxia such as would occur during the present ventilatory responses. Secondly, it can be seen from fig. 4 that the control states of the groups were different. The A/V shunt and the bubble exchanger have an effect upon ventilation even without the CO 2 loading. This effect is dependent upon the integrity of the carotid sinus nerve and might involve a baroreceptor reflex.
224
POTASSIUM AND PERIPHERAL CHEMODENERVATION
The effects of the KCI infusions were clear-cut, every trial resulted in an obvious increase in ventilation. Chemodenervation abolished the response (fig. 3). The changes in arterial plasma potassium were large, but within the range of those observed in severe exercise in man. The responses may have been masked to some extent by changes in P a o 2. It is not possible to distinguish in the present study between an effect upon the mean firing rate of the chemoreceptors and an effect upon the sensitivity to oscillations in carbon dioxide. The interaction between K ÷ and other chemical stimuli at the peripheral chemoreceptors has yet to be defined. It is of interest that Acker and O'Regan (1984) have found that stimulation of the carotid body by hypoxia, cyanide and acetylcholine causes an increase in its extracellular K ÷ concentration. Their work supports the hypothesis that the transmitter within the carotid body is K ÷ (Biscoe, 1971). In conclusion, changes in plasma potassium affect chemoreceptor reflexes and the stimulating effect of hyperkalaemia may be important in the ventilatory response to exercise. There are, however, many other circumstances in which plasma potassium may be disturbed - injections of adrenaline cause gross changes; haemorrhage, particularly after prolonged anaesthesia, can cause rapid rises of several mmol/L (Treasure, 1977). The importance of potassium as a 'respiratory variable' may have been underestimated in recent years and unnoticed changes may have led to apparently conflicting results in some experiments. We suggest that potassium released from exercising muscle increases the peripheral drive to breathing and is the 'unknown humoral factor' postulated by earlier workers (Geppert and Zuntz, 1888).
Acknowledgements We are grateful for financial support from the British Heart Foundation, St. Thomas's Hospital Research Endowment Fund and the Association of Anaesthetists of Great Britain and Ireland.
References Acker, H. and R.G. O'Regan (1984). Extracellular and Ca + ÷ activities in the carotid body during chemoreceptor excitation and inhibition. J. PhysioL (London) 355: 45P. Band, D. M., C. B. Wolff,J. Ward, G. M. Cochrane and J. R. Prior (1980). Respiratoryoscillations in arterial carbon dioxide as a control signal in exercise. Nature (London) 283: 84-85. Bennett, M. F., R.D. Tallman Jr., and F.S. Grodins (1984). Role of V¢o2 in control of breathing of awake exercising dogs. J. Appl. Physiol. 56: 1335-1337. Biscoe, T.J. (1971). Carotid body: structure and function. Physiol. Rev. 51: 437-495. Cross, B.A., K.D. Leaver, S.J.G. Semple and R.P. Stidwell (1984). Response of carotid chemoreceptor discharge to alterations in the slope of the arterial pH oscillation in the cat. J. Physiol. (London) 357: 90P. Geppert, J. and N. Zuntz (1888). l~ber die Regulation der Atmung. Arch. Ges. Physiol. 42: 189-245.
D.M. BAND etal.
225
Jarisch, A., S. Landgren, E. Neff and Y. Zotterman (1952). Impulse activity in the carotid sinus nerve following intra-carotid injection of potassium chloride, veratrine, sodium citrate, adenosine-triphosphate and alpha-dinitrophenol. Acta Physiol. Scand. 25:195-211. Kilburn, K.H. (1966). Muscular origin of elevated plasma potassium during exercise. J. Appl. Physiol. 21: 675-678. Laurell, H. and B. Pernow (1966). Effect of exercise on plasma potassium in man. Acta Physiol. Scand. 66: 241-242. Linton, R. A. F., R. Miller and I. R. Cameron (1976). Ventilatory response to CO 2 inhalation and intravenous infusion of hypercapnic blood. Respir. Physiol. 26: 383-394. Linton, R. A. F., M. Lim and D. M. Band (1982). Continuous intravascular monitoring of plasma potassium using ion-selective electrode catheters. Crit. Care Med. 10: 337-340. Linton, R.A.F., M. Lim, C.B. Wolff, P. Wilmhurst and D.M. Band (1984). Arterial plasma potassium measured continuously during exercise in man. Clin. Sci. 67: 427-431. 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. Phillipson E.A., J. Duffin and J.D. Cooper (1981). Critical dependence of respiratory rhythmicity on metabolic CO 2 load. J. Appl. Physiol. 50: 45-54. Schulthess, P., Y. Shijo, H. V. Pham, E. Pretsch, D. Amman and W. Simon (1981). A hydrogen ion-selective liquid membrane electrode based on tri-n-dodecylamine as a neutral carrier. Anal. Chim. Acta 131: 111-116. Treasure, T. (1977). Studies of the application of ion-selective electrodes to the measurement of plasma potassium. M.S. Thesis, University of London. Van Beaumont, W., J.C. Strand, J. S. Petrofsky, S. G. Hipskind and J. E. Greenleaf (1973). Changes in total plasma content of electrolytes and proteins with maximal exercise. J. Appl. Physiol. 34: 102-106. Yamamoto, W. S. and M.W. Edwards (1960). Homeostasis of carbon dioxide during intravenous infusion of carbon dioxide. J. Appl. Physiol. 15: 807-818.
Respiration Physiology (1985) 60, 217-225 Elsevier
THE EFFECT OF PERIPHERAL CHEMODENERVATION ON THE VENTILATORY RESPONSE TO POTASSIUM
D.M. BAND, R.A.F. LINTON, R. KENT and F.L. KURER Laboratory of Applied Physiology, St. Thomas's Hospital Medical School, London SE1 7EH, U.K.
Abstract. The present experiments were designed to test the hypothesis that the increase in plasma potassium which occurs during exercise acts as a stimulus to respiration via the peripheral chemoreceptors. The effect of intravenous infusion of KC1 on ventilation was measured in anaesthetised cats while they were loaded with CO2 intravenously via a bubble gas exchanger, Ventilation during K + infusion was compared with that immediately before in 'intact' and peripherally chemodenervated cats. In the 'intact' group there was a highly significant increase in ventilation of approximately 25~o ( + 253 + 22 ml/min, P < 0.001), whereas in the chemodenervated grou p there was no significant change (+ 17 _+ 11 ml/min) in spite of similar increases in arterial K ÷ concentration. The results of these experiments indicate that K ÷ infusions stimulate ventilation and that this effect is abolished by peripheral chemodenervation.
Arterial chemoreceptors Cat Chemodenervation
Chemoreflexes Control of breathing Exercise
Hyperkalemia Potassium Ventilation
At the start of exercise there is a rapid rise in the concentration of plasma potassium due to its release from contracting muscle (Laurell and Pernow, 1966; Kilburn, 1966; Van Beaumont et al., 1973). In man, the magnitude and time course of the changes in arterial plasma potassium produced by moderate exercise have been demonstrated (Linton et al., 1984). The carotid body is sensitive to intra-arterial injections of KCI (Jarisch et al., 1952), and we have shown in anaesthetised cats that it is sensitive to transient changes of plasma potassium within the range of those produced by exercise in man (Linton and Band, 1985). The hypothesis that the rise in arterial potassium in exercise may form part of the chemical stimulus to ventilation under these circumstances has therefore to be considered. Brief injections of KCI as a test stimulus are unphysiological in that the arterial plasma potassium rises very rapidly and the duration of the stimulus is only for one or Accepted for publication 1 March 1985
218
POTASSIUM AND PERIPHERAL CHEMODENERVATION
two breaths. This is unlike the sustained rise which occurs during exercise in man (Linton et al., 1984). Since there may be fast adaptation by the chemoreceptor to elevated levels of potassium, we have tested the response to more sustained changes produced by infusions of KC1. In order to simulate the exercise changes more closely and to reduce the masking effect of changes in the gas tensions resulting from any response, KC1 infusions were given against a background of intravenous CO 2 loading. Unfortunately it was not technically possible to mimick the onset transient of CO 2 production with the gas exchanger. The ventilatory responses to brief potassium injections, as used previously (Linton and Band, 1985), diminish with time of anaesthesia, so that abolition of the response by peripheral chemodenervation would not provide conclusive evidence that it was mediated by the peripheral chemoreceptors. Under the conditions of the present experiments the ventilatory response to potassium persisted for many hours, allowing us to test the effect of peripheral chemodenervation.
Methods Cats weighing 2.4-3.7 kg were anaesthetised with pentobarbitone, 40 mg/kg intraperitoneally and supplements intravenously as necessary. A tracheostomy cannula was inserted and connected to a pneumotachograph; the integrated flow signal was recorded as inspiratory and expiratory tidal volumes. The left axillary artery and vein were cannulated for the measurement of blood pressure and infusion of KC1 respectively. A pH electrode catheter was inserted into the upper abdominal aorta via the right renal artery. The pH electrodes used were similar in design to the K + electrodes previously described (Linton et al., 1982) but with membranes containing tri-N-dodecylamine to make them selective for the hydrogen ion (Schulthess et al., 1981). An external Ag/AgC1 reference electrode was connected to a saline-filled catheter inserted via a femoral vein so that its tip made electrical contact with the blood in the inferior vena cava and lay close to the sensing tip of the pH electrode in the adjacent aorta. This arrangement reduces the tendency to record an ECG artefact on the pH trace. The pH electrode catheters have exceptional discrimination for small changes in pH and follow the breath-by-breath fluctuations which correspond to the alveolar CO2 fluctuations induced by tidal breathing. Changes in mean pH can be measured from the records to a precision beyond the discrimination of an in vitro electrode system. A large-bore cannula was tied into the abdominal aorta just above its bifurcation occluding distal flow. This was connected to a bubble gas exchanger consisting of a vertical column, defoaming chamber and reservoir. Gas passed through a perforated Perspex disc in the base of the column. Arterial blood flowed via a side arm into the bottom of the column where it was lifted up the column by the bubbles of gas. The blood then passed along a defoaming chamber, lined with silicone grease, to the reservoir. From here it was pumped at 25 ml/min by an electrically driven roller pump through a 40 #m screen filter (Pall, Ultipor) and returned to the inferior vena cava. The level
D.M. BAND etal.
219
of blood in the reservoir was kept constant by adjusting a screw clamp on the arterial input tubing. When the exchanger was gassed with C O / a n approximate doubling of the animal's CO 2 'production' resulted. The gas exchanger was primed with Hartmann's solution and the cats were given 3000 units of heparin before the circuit was opened and 1000 units hourIy thereafter. Arterial K + concentration was measured continuously using a K +-selective electrode, similar to that previously described (Linton et al., 1982) positioned in the arterial flow to the gas exchanger. Airway CO z concentration was measured with a mass spectrometer (Medishield MS2). Traces were recorded on a Gould ES 1000 electrostatic recorder. The gas exchanger was initially bubbled with nitrogen. When the Hartmann's solution had washed out of the circuit and the cat was stable, the gas exchanger was bubbled with COz. At least 2 min were required for the CO 2 load to stabilise. An infusion of KC1 was then started (5 ml/min for 0.5 min of 150 mmol/L). The effects of the infusions of KC1 were studied in two groups of cats. In the first group ('intact') 6 responses were measured in 5 cats. In cats 1 and 2 both aortic and carotid sinus nerves were intact; in cats 3, 4 and 5 the aortic nerves were cut and fine cannulae were positioned close to the carotid sinus nerves which were intact. In the second group (chemodenervated) 5 responses were measured in 4 cats which had been peripherally chemodenervated. In cats 3, 4 and 5 this was achieved by injection of lignocaine around the carotid sinus nerves via the cannulae which had been positioned prior to the measurement of the 'intact' responses. In this way the time interval between the 'intact' and chemodenervated responses could be kept as short as possible. In cat 6 both aortic and carotid sinus nerves were cut. Peripheral chemodenervation was considered to be complete if there was no increase in ventilation after administering 4 breaths of nitrogen.
Results Results are shown as mean values + SEM. The significance of the difference between mean values was assessed within groups using Student's t-test for paired data and between groups using Student's t-test for unpaired data. Figure 1 shows an 'intact' response and the results for the 'intact' group are given in table 1. An increase in arterial potassium of 3.9 + 0.5 mmol/L caused a highly significant increase in ventilation (253 + 22 ml/min, P < 0.001), which in turn produced a significant increase in mean arterial pH (0.023 + 0.008, P < 0.05). Figure 2 shows the experiment repeated in cat 4 after injection of lignocaine around the sinus nerves, and the results for the chemodenervated group are shown in table 2. There was no significant change in ventilation ( + 1 7 + l l m l / m i n ) or pH ( - 0.003 + 0.003) in spite of an increase in arterial K + concentration similar to that in the 'intact' group.
220
POTASSIUM AND PERIPHERAL C H E M O D E N E R V A T I O N
]100
7"32°[A pH
8
//j/
7"3601
7.400c' t~
Jll.,'//J///,J
1
K,,I-
i1 °
(mmol.t "l}
2 . . . .
! I li! i!i
'
IIU
I' ,!
,,
I ~i~,hr i
,!~ ,i,i I..... i
,,
i ! ~il
i!~l
10 sec
J
S.0 A I RWAY C02 °/0 0
Fig. 1. K ÷ infusion during intravenous CO 2 loading in 'intact' cat 4. The traces from above downwards are arterial pH (becoming more alkaline as the trace falls), arterial K +, tidal volume and airway CO2 concentration. As the arterial K ÷ rises at the start of the infusion (KCI 0.75 mmol/min) there is an increase in the ventilation and the pH becomes more alkaline. When the infusion is switched off the arterial K ÷ falls, causing ventilation to decrease and the pH to become more acid.
TABLE 1 'Intact' group. The upper half of the table shows control values (measured over a 5-breath period) of minute volume (~z) and arterial K + concentration during intravenous CO 2 loading immediately prior to infusion of KCI. The lower half shows changes in minute volume (A~'), arterial K + (AK +) and arterial pH (ApH) measured over a 5-breath period 15 sec after the arterial K + started to rise. Six responses were measured in 5 cats. In cats 1 and 2 the aortic and sinus nerves were intact. In cats 3, 4 and 5 the aortic nerves were cut and cannulae were positioned close to the intact sinus nerves (for later injection of lignocaine). Cat
Mean + SEM 1
1
2
3
4
5
x~ (ml/min)
995
1080
1001
850
1225
930
K ÷ (mmol/L)
3.6
3.9
2.5
4.7
3.0
2.4
AV (ml/min)
+283
+ 172
+332
+241
+263
+228
AK + (mmol/L)
+ 2.9
+ 2.8
+ 4.4
+ 5.6
+ 4.9
+ 3.0
+ 0.024
+ 0.052
+ 0.018
--
+ 0.009
+ 0.011
C02
1014 +53 3.4 + 0.4
CO 2 + K +
ApH
253 + 22 + 3.9 +0.5 + 0.023 _+0.008
D.M. BAND etal.
221
~.,,o[% %VV'vlN'
pH
Lf l
A
tJ
f f!
',L/:/:/',/ /~:?/. 11:/./, (mmot.t_tl 4
:IIII} iti!II!ii!
, q
!
I I'~'i!
I,
!~lili
I
iF
i'i,
i;',
i;l'
'
C02 %
i
10 sec Fig. 2. K ÷ infusion during intravenous CO2 loading in chemodenervated cat 4. There is now no change in ventilation or arterial pH in response to hyperkalaemia. TABLE 2 Chemodenervated group. The layout of the table is the same as for table 1. Five responses were measured in 4 cats. Chemodenervation was achieved in cat 6 by cutting both sinus and aortic nerves, and in cats 3, 4 and 5 by injecting lignocaine around the sinus nerves via the cannulae positioned prior to measurement of the intact responses in these cats. Cat
Mean _+SEM 3
4
5
6
6
(ml/min)
981
799
930
977
702
K ÷ (mmol/L)
4.4
2.5
2.2
3.6
5.6
-5
+43
+44
-9
+12
+ 3.5
+ 4.4
+ 2.6
+ 4.0
+ 4.5
--
- 0.002
+ 0.004
- 0.006
- 0.009
C02
878 +55 3.7 +0.6
CO 2 + K +
AV (ml/min) AK
÷
ApH
(mmol/L)
+17 +11 + 3.8 _+0.3 - 0.003 _ 0.003
A c o m p a r i s o n o f t h e c h a n g e s i n K + a n d v e n t i l a t i o n in t h e t w o g r o u p s o f c a t s is s h o w n i n fig. 3. T h e level o f v e n t i l a t i o n p r i o r t o i n f u s i o n o f KC1 w a s s i g n i f i c a n t l y l o w e r in t h e c h e m o d e n e r v a t e d g r o u p ( P < 0.05) a n d t h e e n d - t i d a l P c o 2 w a s c o r r e s p o n d i n g l y h i g h e r ( ' i n t a c t ' = 29.3 + 2.0 m m H g ; c h e m o d e n e r v a t e d
-- 37.0 + 2.1 m m H g ; P < 0.001). I n
s o m e e x p e r i m e n t s m e a n v a l u e s o f v e n t i l a t i o n a n d e n d - t i d a l P c o 2 w e r e d e r i v e d for t h e
222
POTASSIUM AND PERIPHERAL CHEMODENERVATION 1~00 1300 1200 VENTILATION (ml.min"~)
1100 1000
÷
900 800 J
3
t~
5
6
7
9
+
ARTERIAL K
(mm01.1-~) Fig. 3. Comparison of ventilatory responses to hyperkalaemia during intravenous CO 2 loading in 'intact' (n = 6) and chemodenervated (n = 5) cats. Mean values + 1 SEM are shown for the 'intact' ( 0 ) and chemodenervated (O) groups. Similar increases in arterial K + produced a highly significant increase in ventilation in the 'intact' group (P < 0.001) and no significant change in the chemodenervated group.
1400 1200 I000
VENTILATION
800
(ml.min-I)
600
Y
400 22
24
26
28
30
32
34
36
38
END TIDAL PCO 2 (mmHg)
Fig. 4. The effect of intravenous CO2 loading and K ÷ infusion on ventilation and end-tidal Pco2. Mean values for 'intact' ( O ) and ehemodenervated cats (O) are shown. Arrows indicate the experimental sequence: exchanger gassed with N2 ~ exchanger gassed with CO 2 --*exchanger gassed with CO 2 and intravenous infusion of KC1. For the 'intact' group, n = 4 for the first point and n = 6 for the second and third points. For the chemodenervated group n = 3 for the first point and n = 5 for the second and third points.
p e r i o d w h e n the gas e x c h a n g e r w a s g a s s e d w i t h n i t r o g e n i m m e d i a t e l y prior to switching to C O 2 . S u b s e q u e n t c h a n g e s in t h e s e v a r i a b l e s p r o d u c e d by C O 2 l o a d i n g a n d t h e n K ÷ infusion (during C O 2 l o a d i n g ) are s h o w n in fig. 4. F i g u r e 5 s h o w s the effect o f i n t r a v e n o u s C O 2 l o a d i n g o n the arterial p H r e c o r d in cat 5. T h e g r a d i e n t ( p H / s e c ) o f the a c i d - g o i n g p a r t o f the cycle i n c r e a s e d by 206~o.
D.M. BAND et al.
-I pH.sec
0.0097
223
0.020
Fig. 5. The effectof intravenous CO2 loadingon the arterial pH record in cat 5. CO2 loading increases the slope of the pH oscillation by 206~o,implyinga similar increase in CO2 production.
Discussion
These experiments show that increasing the plasma potassium concentration by intravenous infusion of potassium chloride enhances the ventilatory response of the anaesthetised cat to intravenous CO= loading. This enhancement is mediated via the peripheral chemoreceptors. The effect persists for the 0.5 min period of the infusions used in the present experiments. Hyperkalaemia due to potassium effiux from exercising muscle may therefore be a contributing factor to the ventilatory response to exercise. There is controversy concerning the effects of intravenous CO z loading alone as a stimulus to ventilation. The technique has been used extensively as a means of testing the hypothesis that oscillations in Pa¢o~ provide a controlling signal related to CO 2 production (Yamamoto and Edwards, 1960). When ventilation is plotted as a function of P a c o 2 the response curves vary from more than vertical, i.e. the P a c o ~ actually falls, to curves indistinguishable from the classical response curves to inhaled CO 2 mixtures (cf. Linton et al., 1976; Phillipson et al., 1981; Bennett et aL, 1984). The present experiments confh'm that CO2 loading does change the profile of the oscillations in pH and therefore P a c o :. The pH oscillations shown in fig. 5 have an increased rate of change during CO2 loading and are very similar to traces obtained from man during the onset of exercise (Band et al., 1980). The pH records provide indirect confn'mation that an approximate doubling of the CO2 production of the animals was achieved by the bubble exchanger. There are two factors which make it difficult to draw conclusions from the ventilatory responses to CO 2 infusions alone in the present study. Firstly, no attempt was made to control the Pao:. Recent work (Cross et al., 1984) shows that the response of the carotid body chemoreceptor in the anaesthetised cat to the oscillations in CO2 is strongly depressed by slight hyperoxia such as would occur during the present ventilatory responses. Secondly, it can be seen from fig. 4 that the control states of the groups were different. The A/V shunt and the bubble exchanger have an effect upon ventilation even without the CO 2 loading. This effect is dependent upon the integrity of the carotid sinus nerve and might involve a baroreceptor reflex.
224
POTASSIUM AND PERIPHERAL CHEMODENERVATION
The effects of the KCI infusions were clear-cut, every trial resulted in an obvious increase in ventilation. Chemodenervation abolished the response (fig. 3). The changes in arterial plasma potassium were large, but within the range of those observed in severe exercise in man. The responses may have been masked to some extent by changes in P a o 2. It is not possible to distinguish in the present study between an effect upon the mean firing rate of the chemoreceptors and an effect upon the sensitivity to oscillations in carbon dioxide. The interaction between K ÷ and other chemical stimuli at the peripheral chemoreceptors has yet to be defined. It is of interest that Acker and O'Regan (1984) have found that stimulation of the carotid body by hypoxia, cyanide and acetylcholine causes an increase in its extracellular K ÷ concentration. Their work supports the hypothesis that the transmitter within the carotid body is K ÷ (Biscoe, 1971). In conclusion, changes in plasma potassium affect chemoreceptor reflexes and the stimulating effect of hyperkalaemia may be important in the ventilatory response to exercise. There are, however, many other circumstances in which plasma potassium may be disturbed - injections of adrenaline cause gross changes; haemorrhage, particularly after prolonged anaesthesia, can cause rapid rises of several mmol/L (Treasure, 1977). The importance of potassium as a 'respiratory variable' may have been underestimated in recent years and unnoticed changes may have led to apparently conflicting results in some experiments. We suggest that potassium released from exercising muscle increases the peripheral drive to breathing and is the 'unknown humoral factor' postulated by earlier workers (Geppert and Zuntz, 1888).
Acknowledgements We are grateful for financial support from the British Heart Foundation, St. Thomas's Hospital Research Endowment Fund and the Association of Anaesthetists of Great Britain and Ireland.
References Acker, H. and R.G. O'Regan (1984). Extracellular and Ca + ÷ activities in the carotid body during chemoreceptor excitation and inhibition. J. PhysioL (London) 355: 45P. Band, D. M., C. B. Wolff,J. Ward, G. M. Cochrane and J. R. Prior (1980). Respiratoryoscillations in arterial carbon dioxide as a control signal in exercise. Nature (London) 283: 84-85. Bennett, M. F., R.D. Tallman Jr., and F.S. Grodins (1984). Role of V¢o2 in control of breathing of awake exercising dogs. J. Appl. Physiol. 56: 1335-1337. Biscoe, T.J. (1971). Carotid body: structure and function. Physiol. Rev. 51: 437-495. Cross, B.A., K.D. Leaver, S.J.G. Semple and R.P. Stidwell (1984). Response of carotid chemoreceptor discharge to alterations in the slope of the arterial pH oscillation in the cat. J. Physiol. (London) 357: 90P. Geppert, J. and N. Zuntz (1888). l~ber die Regulation der Atmung. Arch. Ges. Physiol. 42: 189-245.
D.M. BAND etal.
225
Jarisch, A., S. Landgren, E. Neff and Y. Zotterman (1952). Impulse activity in the carotid sinus nerve following intra-carotid injection of potassium chloride, veratrine, sodium citrate, adenosine-triphosphate and alpha-dinitrophenol. Acta Physiol. Scand. 25:195-211. Kilburn, K.H. (1966). Muscular origin of elevated plasma potassium during exercise. J. Appl. Physiol. 21: 675-678. Laurell, H. and B. Pernow (1966). Effect of exercise on plasma potassium in man. Acta Physiol. Scand. 66: 241-242. Linton, R. A. F., R. Miller and I. R. Cameron (1976). Ventilatory response to CO 2 inhalation and intravenous infusion of hypercapnic blood. Respir. Physiol. 26: 383-394. Linton, R. A. F., M. Lim and D. M. Band (1982). Continuous intravascular monitoring of plasma potassium using ion-selective electrode catheters. Crit. Care Med. 10: 337-340. Linton, R.A.F., M. Lim, C.B. Wolff, P. Wilmhurst and D.M. Band (1984). Arterial plasma potassium measured continuously during exercise in man. Clin. Sci. 67: 427-431. 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. Phillipson E.A., J. Duffin and J.D. Cooper (1981). Critical dependence of respiratory rhythmicity on metabolic CO 2 load. J. Appl. Physiol. 50: 45-54. Schulthess, P., Y. Shijo, H. V. Pham, E. Pretsch, D. Amman and W. Simon (1981). A hydrogen ion-selective liquid membrane electrode based on tri-n-dodecylamine as a neutral carrier. Anal. Chim. Acta 131: 111-116. Treasure, T. (1977). Studies of the application of ion-selective electrodes to the measurement of plasma potassium. M.S. Thesis, University of London. Van Beaumont, W., J.C. Strand, J. S. Petrofsky, S. G. Hipskind and J. E. Greenleaf (1973). Changes in total plasma content of electrolytes and proteins with maximal exercise. J. Appl. Physiol. 34: 102-106. Yamamoto, W. S. and M.W. Edwards (1960). Homeostasis of carbon dioxide during intravenous infusion of carbon dioxide. J. Appl. Physiol. 15: 807-818.