AXERICAN
JOURNAL
Vol. 228, No. 4, April
01'.
hYSSOLOGY
1975.
Printed in U.S.A.
Distribution of H+ and HC03- between CSF and blood during respiratory acidosis in dogs E. G. PAVLIN AND T. F. HORNBEIN (With Anesthesia Research Center, Department of Anesthesiology University of Washington, Seattle, Washington 98195
balance; regulation;
cisternal and lumbar passive distribution;
of Carol Stowers) and Biophysics,
Six dogs were anesthetized with pentobarbital (25 mg/kg), paralyzed with gallamine, and intubated and mechanically ventilated with 100 % oxygen. Methods of measurement of the CSF/plasma DC potential difference (E) and acid-base values have been described (13). A control state at pH, at 7.40 was sustained for 1.5-2 h. Following control measurements, respiratory acidosis was induced by increasing the FI~O% sufficiently to decrease pH, about 0.20 pH U from control (Fig. 1). Pacoz was kept constant thereafter. Plasma [HCOa-] rose about 1.6 meq/kg Hz0 with induction of respiratory acidosis due to buffering by hemologin and was thereafter maintained at that level by small infusions of 0.6 N HCl or 0.9 N sodium bicarbonate as necessary. Arterial and cisternal and lumbar CSF samples were obtained approximately 1h at after the increase in Flcoz when PacoZ had stabilized the new level. This sample was defined as time 0. Subsequent samples were obtained 3, 4.5, and 6 h later. CSF/ plasma E was recorded at each time of sampling. Electrochemical potential differences (p) of H+ and HCO3were calculated ( 13).
CSF; cerebrospinal CSF/plasma DC
have observed that during chronic respiratory acidosis in man, the pH of CSF deviates less from normal than the pH of blood (10, 16). Studies in animals in which other variables, such as hypoxia and circulatory changes, are more readily controlled reveal similar mitigation of CSF pH change (10). This paper describes the changes in CSF acid-base state occurring during respiratory acidosis sustained for 6 h in dogs. The purpose of this investigation was to evaluate the mechanisms regulating concentrations of H+ and HCO3ions in brain extracellular fluid, in particular whether this regulation is effected by active ion transport or by passive distribution. The approach was like that used to study H+ and HCO, distribution between cisternal and lumbar CSF and blood during metabolic acidosis (13) and alkalosis (14). The response of the electrochemical potential difference (p) for these ions was followed after induction of an acid-base change until a steady state was achieved. Return of p for H+ and HCO3to control during both metabolic derangements was interpreted as compatible with a process of passive distribution for these ions. The rationale has been discussed previously (13, 17). SEVERAL
Assistance
of Physiology
METHODS
PAVLIN, E. G., AND T. F. HORNBEIN. Distribution of H+ and HCO3 betzueen CSF and blood during respiratory acidosis in dogs. Am. J. Physiol. 228(4): 1145-I 148. 1975.-To evaluate the regulation of [H+] and [HCOa-] in brain extracellular fluid during respiratory acidosis, the changes in cisternal and lumbar CSF acid-base state were assessed in six anteshetized, paralyzed, mechanically ventilated dogs rendered hypercapnic by increase in FIcoz. Arterial [HCOa-] was held constant. The electrochemical potential difference (cl) between CSF and blood for Hf and HCOawas calculated from values for [H+] and [HCOS-] in CSF and arterial plasma and the simultaneously measured CSF/plasma DC potential difference. Measurements were made at pH, = 7.40, after stable arterial values of pH, of about 7.2 were attained and 3, 4.5, and 6 h thereafter. A steady state for ion distribution was attained by 4.5 h. Values of p for H+ and HCO3 at 6 h had for cisternal CSF and returned to +0.7 and - 0.7 mV of control +1.3 and -0.6 mV of control for lumbar CSF. The attainment of steady-state values for ,u close to control is comparable with passive distribution of these ions between CSF and blood. acid-base fluid; ion potential
the Technical and Department
INVESTIGATORS
RESULTS
With respiratory acidosis (Fig. l), we observed a change in the CSF/plasma AE/ApH, of -29.8 rnV/pH. at the cisterna magna and - 19.6 at the lumbar site (Fig. 2). Maintaining pH, constant resulted in a constant E over the next 6 h. Our cisternal AE/ApH. is close to the -30.6 mV/pH. observed by Held et al. (6), but our lumbar value is considerably greater than the -5.6 mV/pH, reported by Hornbein and S$rensen (8). Considering the commonality of one investigator, the reason for this difference in lumbar AE/ApH, is baffling. At a constant Pace, from time 0, CSF Pco2 rose progressively over the next 6 h (Fig. 1). With a constant arterial pH, cisternal CSF pH decreased with initiation of respiratory acidosis, but then increased to a stable value by 4.5 h (Fig. 1, Table 1). Lumbar CSF pH (Fig. 1) showed no return toward control over the 6 h, in part consequent to a greater rise in CSF PCO~ and in part because of a lesser rise in [HCO 3-I (Fig. 1). We permitted the initial rise in [HCOS-] a resulting from titration of hemoglobin with CO2, but thereafter held [HCOZ-]a constant by HCl infusion to offset the usual
1145
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.
E.
G.
PAVLIN
T
,
AND
T.
Ci sterna
I=-29.8
Lumbar
= - 19.6
O
?
a’\ \\ - I **.. -.* -.I
7.20-
Cistern0
0
=\
I 0-Li
-5-
. . ..- ...*
l
..-•
...........
.@ e-0 T 0) L 0 I
/)0
....
//
mV/pH mV/pH
0
\ 0@
HORNREIN
Lumbar
...
7 25-
F.
. ..*
. ..*. l *-*--* \
I
Lumbar
I 6
c 32-
I
1
I
o
1
2 Time
1
I
1
1
3 ( hours)
4
5
6
FIG. 2. CSF/plasma DC potential difference (E) and electrochemical potential differences for H+ (pn+ ) and HC03(pncos-) vs. time during respiratory acidosis in 6 anesthetized dogs. Values o AE/ApH. were calculated for change in time 0 sample from control. Values of &kHt and pHC03are calculated for concentrations in CSF and mean capillary plasma water. Vertical bars indicate & 1 SEM.
, Arterial
Comparing time course
I ,
I
C
0
1
1
1
2 Time
I
3 (hours)
I
1
4
1
5
6
FIG. 1. Pco2, pH, and [HCOs-] of arterial plasma, cisternai m and lumbar CSF vs. time during respiratory acidosis in 6 anesthetized dogs. Vertical bars indicate =t 1 SEM.
.
compensatory processes (Fig. 1). The [HC03-]esF rose from 26.3 to 33.3 meq/kg Hz0 at the cisternal site and 26.7 to 31.7 meq/kg Hz0 in the lumbar CSF, a steady state being achieved at both locations by 4.5 h. Values for pH+ and pHe03- returned essentially to their control values by 4.5 h (Fig. 2, Table 2). DISCUSSION
The initial increase in plasma [HCOS-] with induction of respiratory acidosis exhibited a AIHCOa-]/ApH, of 9.1 slykes, which agrees well with in vivo values reported in the literature (18). Accepting this initial rise, we kept pl asma [HCO 3-1 constant thereafter to prevent a slow compensatory rise from possibly delaying establishment of a steady state of ion concentration in CSF.
lumbar and cisternal sites, we observed a of acid-base changes and return to control that were similar at both locations as well pH+ and PHCOsas identical to the behavior observed during the other types of acid-base derangements studied (7, 13, 14). The significance of this common behavior at the two locations in the presence of differing magnitudes of change in CSF/ plasma potential difference was discussed earlier (13). Although the CSF-arterial PCO~ difference at time 0 remained essentially unchanged from control (Fig. 1 ), the slope of the blood CO2 dissociation curve is such that the arterial-venous CO 2 content difference (assuming venous and CSF Pco:! to be equal) decreased. We estimate that at time 0, 1 h after onset of hypercapnia, brainstem blood flow was increased by 17 % and lumbar spinal cord flow by 9%. By 6 h, blood flow at both locations had actually fallen below control, by 9 % at the brainstem, and 16% in the lumbar area. Bleich et al.‘s (3) data seem to indicate similar changes in dogs maintained hypercapnic for 5 days. The decrease in flow as [HCO &BP increased is qualitatively compatible with Fencl et al.‘s (5) observations in man that cerebral blood flow is inversely proportional to [HCO Q-]cBF. Quantitatively our estimated blood flow ended up below control at a time when CSF pH remained distinctly acid, an observation hard to reconcile with the hydrogen ion theory of cerebral blood flow regula-
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.
CSF-BLOOD
H+
AND
HC03-
WITH
RESPIRATORY
1147
ACIDOSIS
1. Values for acid-base parameters in arterial plasma, cisternal and lumbar CSF, CSF/plasma E, and calculated electrochemical potential dijferences (p> for H+ and HCO, in six dogs during respiratory acidosis with constant plasma [HCOZ-] TABLE
Time, h Control
0
3
4.5
6
Location
[HCOs-1, meq/kg H20
PH
[Lactate-],
Pco2, torr
meq/kg
H20
E, mV
pH+# mV
I”HCO-3’
mV
Arterial CSF-cistern CSF-lumbar
7.406 7.350 7.337
=f= 0.001 =t 0.004 =t 0.007
22.6 26.3 26.7
zk 0.7 rt 0.3 + 0.3
33.6 51.0 53.5
zk * *
1.0 1.1 1.2
1.2 =t 1.8 + 1.8.h
0.2 0.1 0.1
2.6 4.4
zt +
0.4 0.9
2.0 4.3
It *
0.5 1.1
-0.1 -1.9
& *
0.6 1.2
Arterial CSF-cistern CSF-lumbar
7.230 7.217 7.188
=t =t &
0.004 0.013 0.012
24.2 28.7 28.4
zt =t rt
0.6 0.7 0.7
55.8 75.0 79.2
=t =t &
1.4 1.3 1.5
0.8 1.3 1.5
+ + *
0.1 0.1 0.1
7.8 7.9
& 0.5 =i= 1.3
5.8 6.9
=t +
0.9 1.3
-4.8 -5.3
zt zt
1.2 1.4
Arterial CSF-cistern CSF-lumbar
7.220 7.245 7.196
=t & +
0.010 0.009 0.019
24.4 31.6 30.1
=t =t =t
0.6 0.4 0.6
57.5 77.5 84.5
zk 1.6 + 1.9 zt 3.1
0.7 1.4 1.4
zk 0.1 ZiZ 0.1 =t 0.2
7.9 7.9
* zt
0.4 1.3
3.4 5.6
zk 0.7 =t 1.4
-2.3 -3.4
+ It
0.8 1.4
Arterial CSF-cistern CSF-lumbar
7.221 7.247 7.190
rt =t =t
0.010 0.009 0.010
24.2 32.8 31.5
rt zt &
0.8 0.7 0.4
57.1 79.7 88.3
zt 2.5 zt 2.5 zk 2.0
0.7 1.5 1.5
zk 0.1 =t 0.2 St: 0.2
7.7 7.9
* =t
0.4 1.3
2.5 5.4
=t zt
0.7 1.5
-1.1 -2.6
=t =t
0.6 1.4
Arterial CSF-cistern CSF-lumbar
7.225 7.246 7.188
zt rk &
0.004 0.015 0.014
24.4 33.3 31.7
zt =t +
0.5 0.8 0.6
56.8 81.2 88.0
zt 1.3 3~ 1.8 dz 2.3
0.7 1.7 1.5
It It It
7.7 7.8
Ik 0.5 IIZ 1.1
2.7 5.6
zk 0.6 If 1.5
-0.8 -2.5
zt =t
0.6 1.3
Values are means & in METHODS of ref. 13.
SE.
The
p for
H+
and
HCOs-
2. Cisternal-plasma p for H+ and HCOa-: between sustained respiratory acidosis and normal mV
Bleich
et al.
(3)
-1.6
-
Adaro
et al.
(1)
+1.4 +2.0
-1.7 -2.2
Dogs, 4 h, E simultaneously
Dogs,
5 days measured
Kazemi
et al.
(9)
+0.7
-2.0
Dogs,
hlesseter
and
SiesjB
-0.6
- 1.8
Rats, 3 days, AE/ApH measured separately
Chazan
et al.
(4)
+1.4
-1.3
Dogs,
Present
study
+0.7
-0.7
Dogs, 6 h, E simultaneously
(12)
(Ap)
Comments
pHCO3-9 mv
1 .o
calculated
diference ’
TABLE
&a+,
are
5 h
5 days measured
In all studies but the present one, metabolic compensation was permitted. E in those instances where not measured (3, 4, 9) was estimated assuming a AE/ApH of -30.5 mV for the acute respiratory component and of -42.3 mV for the compensating metabolic change. Separation of respiratory and metabolic contributions to the change in pH, and E was made assuming an in vivo buffer value of 10 slykes.
tion, yet very like what Alexander et al. (2) reported in goats during respiratory alkalosis. In both cases a blood flow lower than would be anticipated for the existing CSF pH may be simply a consequence of the decrease in cerebral blood flow with time (15) observed by us in varying degree in every acid-base derangement studied (7, 13, 14). The return of CC& and pHC03- to control values is, as we have discussed previously (13), compatible with a passive movement of H+ and HCOSbetween CSF and blood, assuming a constant rate of production of H+ by brain tissue during respiratory acidosis. In Table 2, we compare our values for p with those either provided by or derived from other studies of CSF acid-base status during
from
CSF-mean
0.1 0.3 0.2
capillary
concentration
differences
as described
sustained hypercapnia. In none of these other studies was plasma [HCO 3-1 held constant, possibly delaying achievement of steady-state conditions. Only Adaro et al. (1) measured CSF/plasma E directly, while Messeter and Siesjo (12) applied values for E derived from previous studies in rats (11). For the other studies in dogs we have applied the values of Held et al. (6) for AE/ApH, during respiratory and metabolic changes to permit an estimate of change in p. Admittedly such estimates of 1-1must be imprecise when actual values of E and assurance of steady state for both control and final values are lacking. Nevertheless, the closeness of final values for p to controls in all these studies is impressive, indicating that in spite of differences in experimental format, [HCO3-] and [H+] change in a manner expected for simple passive distribution. The studies of Adaro et al. (I), where E was actually measured, show a slightly poorer return of 1-1than the other studies, but his data after 4 h of respiratory acidosis superimposed on metabolic alkalosis do not demonstrate a steady state. On the other hand, Kazemi et al.‘s (9) final sample would be expected to be close to a steady state according to our observations. Between 0 and 6 h cisternal CSF pH rose from 7.2 17 to 7.246. This return toward the control pH CSF while arterial [HCO 3-1 and PCO~ were kept constant resulted from an increase in [HC03-lCSF engendered primarily by the rise in CSF/plasma E. The failure of lumbar CSF pH to return toward control in the face of passive distribution of H+ and HCOScan be accounted for by two factors. First, as a consequence of a lesser change in E at the lumbar location, the rise in CSF [HCOS-] was less (Fig. I). Second, CSF PCO~ increased more in lumbar than cisternal CSF. The effect of changes in the CSF-arterial Pcoz difference on the relationship between pH+ and pHCo3- will be discussed in the final paper (7). As [HCO,]. rises in response to renal and other com-
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.
1148
E.
pensatory mechanisms, [HCOs-lCSF will also rise, for the fall in CSF/plasma E with increasing blood pH is not sufficient to prevent totally the influx of HCOSinto brain extracellular fluid. This behavior is clearly illustrated by our observations during metabolic alkalosis at constant PacOp (14). Thus pH CSF will return closer to normal as systemic compensation for the respiratory acidosis occurs. The data of Bleich et al. (3), representing a more extreme the greater return of pHe*p respiratory acidosis, illustrate toward normal when [HCO& is permitted to rise. vlesseter and Siesjo’s (12) serial observations in rats also illustrate this direct relationship between CSF pH and plasma [HCOS-1. 0 ne would predict that because of the
G.
PAVLIN
AND
T.
F.
HORNBEIN
larger value for AE/ApH, occurring during metabolic compensation than with the initial respiratory acidosis, CSF pH will be below normal even when systemic compensation is complete, i.e., when pH, equals 7.40 (Fig. 3 of ref. 7). We thank Ms. Watson for patience This work was 15991-05 from the National Institutes T. F. Hornbein Award Grant 5 K03 Received
Patricia Reynolds, Mica1 Middaugh, and Mary and help. supported by Public Health Service Grant GM National Institute of General Medical Sciences, of Health. is supported by Research Career Development HE0961 7-09.
for publication
3 June
1974.
REFERENCES 1. ADARO, BERG
F. V. M.,
E. E. ROEHR, A. R. VIOLA, AND Acid-base equilibrium between fluid in acute hypercapnia. J. A@.
DE OBRUTZKY.
cerebrospinal 271-275, 1969. 2. ALEXANDER, S. C., B. E. MARSHALL, blood flow in the goat with sustained Lab.
Invest.
C.
WYMERSZ-
blood
and
Physiol.
27:
AND A. AGNOLI. Cerebral hypocarbia. Stand. J. C/in.
102, 1968.
sufq!d.
3. BLEICH, H. L., P. M. BERKMAN, AND W. B. SCHW~TZ. The response of cerebrospinal fluid composition to sustained hypercapnia. J. Clin. Invest. 43 : 11-16, 1964. 4. CHAZAN, J. A., F. M. APPLETON, A. M. LONDON, AND W. B. SCHWARTZ. Effects of chronic metabolic acid-base disturbances on the composition of cerebrospinal fluid in the dog. Clin. Sci. 36 : 345-358, 1969. V., J. R., VALE, AND J. A. BROCH. Respiration and 5. FENCL, cerebral blood flow in metabolic acidosis and alkalosis in humans. J. A#.
Physiol.
27 : 67-76,
1969.
6. HELD, D., V. FENCL, AND J. R. PAPPENHEIMER. Electrical potential of cerebrospinal fluid. J. Neurophysiol. 27: 942-959, 1964. T. F., AND E. G. PAVLIN. Distribution of H+ and 7. HORNBEIN, HCOabetween CSF and blood during respiratory alkalosis in dogs. Am. J. Physiol. 228: 1149-1154, 1975. 8. HORNBEIN, T. F., AND S. C. S$RENSEN. d-c Potential difference between different cerebrospinal fluid sites and blood in dogs. Am. J. Physiol. 223: 4.15-418, 1972. H., D. C. SHANNON, AND E. CARVALLO-GIL. Brain CO2 9. KAZEMI, buffering capacity in respiratory acidosis and alkalosis. J. App&. Physiol.
22 : 24-l-246,
1967.
of cerebrospinal fluid composition with 10. LEUSEN, I. Regulation. reference to breathing. Physiol. Rev. 52 : l-56, 1972. 11. MESSETER, K., AND B. K. SIESJ~. The DC potential between CSF and plasma in respiratory acidosis. Acta Physiol. Stand. 83 : 13-20, 1971. of CSF pH in acute 12. MESSETER, K., AND B. K. SIESJ~ Regulation and sustained respiratory acidosis. Acta Physiol. Stand. 83: 21-30, 1971. E. G., AND T. F. HORNBEIN. Distribution of H+ and 13. PAVLIN, between CSF and blood during metabolic acidosis in HC03dogs. Am. J. Physiol. 228; 1134-l 140, 1975. 14. PAVLIN, E. G., AND T. F. HORNBEIN. Distribution of H+ and HC03between CSF and blood during metabolic alkalosis in dogs. Am. J. Physiol. 228: 1141-l 144, 1975. M. E., J. B. POSNER, AND F. PLUM. Cerebral blood flow 15. RAICHLE, during and after hyperventilation. Arch. Neural. 23: 394-403, 1970. 16. SAUNIER, C., M-C AUG-LAXENAIRE, M. SCHIBI, AND P. SADOUL. Acid base and electrolyte equilibrium of arterial blood and cerebra-spinal fluid in respiratory insufficiency. Respiration 26 : 81-101, 1969. 17. SIESJ~, B. K., AND A. KJALLQUIST. A new theory for the regulation of the extracellular pH in the brain. gand. J. Clin. Lab. Invest. 24 : l-9, 1969. 18. WOODBURY, W. Regulation of pH. In: Physiology and Biophysics (20 ed.), edited by T. C. Ruth and H. D. Patton. Philadelphia: Saunders, chapt. 27. In press.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.
JOURNAL
Vol. 228, No. 4, April
01'.
hYSSOLOGY
1975.
Printed in U.S.A.
Distribution of H+ and HC03- between CSF and blood during respiratory acidosis in dogs E. G. PAVLIN AND T. F. HORNBEIN (With Anesthesia Research Center, Department of Anesthesiology University of Washington, Seattle, Washington 98195
balance; regulation;
cisternal and lumbar passive distribution;
of Carol Stowers) and Biophysics,
Six dogs were anesthetized with pentobarbital (25 mg/kg), paralyzed with gallamine, and intubated and mechanically ventilated with 100 % oxygen. Methods of measurement of the CSF/plasma DC potential difference (E) and acid-base values have been described (13). A control state at pH, at 7.40 was sustained for 1.5-2 h. Following control measurements, respiratory acidosis was induced by increasing the FI~O% sufficiently to decrease pH, about 0.20 pH U from control (Fig. 1). Pacoz was kept constant thereafter. Plasma [HCOa-] rose about 1.6 meq/kg Hz0 with induction of respiratory acidosis due to buffering by hemologin and was thereafter maintained at that level by small infusions of 0.6 N HCl or 0.9 N sodium bicarbonate as necessary. Arterial and cisternal and lumbar CSF samples were obtained approximately 1h at after the increase in Flcoz when PacoZ had stabilized the new level. This sample was defined as time 0. Subsequent samples were obtained 3, 4.5, and 6 h later. CSF/ plasma E was recorded at each time of sampling. Electrochemical potential differences (p) of H+ and HCO3were calculated ( 13).
CSF; cerebrospinal CSF/plasma DC
have observed that during chronic respiratory acidosis in man, the pH of CSF deviates less from normal than the pH of blood (10, 16). Studies in animals in which other variables, such as hypoxia and circulatory changes, are more readily controlled reveal similar mitigation of CSF pH change (10). This paper describes the changes in CSF acid-base state occurring during respiratory acidosis sustained for 6 h in dogs. The purpose of this investigation was to evaluate the mechanisms regulating concentrations of H+ and HCO3ions in brain extracellular fluid, in particular whether this regulation is effected by active ion transport or by passive distribution. The approach was like that used to study H+ and HCO, distribution between cisternal and lumbar CSF and blood during metabolic acidosis (13) and alkalosis (14). The response of the electrochemical potential difference (p) for these ions was followed after induction of an acid-base change until a steady state was achieved. Return of p for H+ and HCO3to control during both metabolic derangements was interpreted as compatible with a process of passive distribution for these ions. The rationale has been discussed previously (13, 17). SEVERAL
Assistance
of Physiology
METHODS
PAVLIN, E. G., AND T. F. HORNBEIN. Distribution of H+ and HCO3 betzueen CSF and blood during respiratory acidosis in dogs. Am. J. Physiol. 228(4): 1145-I 148. 1975.-To evaluate the regulation of [H+] and [HCOa-] in brain extracellular fluid during respiratory acidosis, the changes in cisternal and lumbar CSF acid-base state were assessed in six anteshetized, paralyzed, mechanically ventilated dogs rendered hypercapnic by increase in FIcoz. Arterial [HCOa-] was held constant. The electrochemical potential difference (cl) between CSF and blood for Hf and HCOawas calculated from values for [H+] and [HCOS-] in CSF and arterial plasma and the simultaneously measured CSF/plasma DC potential difference. Measurements were made at pH, = 7.40, after stable arterial values of pH, of about 7.2 were attained and 3, 4.5, and 6 h thereafter. A steady state for ion distribution was attained by 4.5 h. Values of p for H+ and HCO3 at 6 h had for cisternal CSF and returned to +0.7 and - 0.7 mV of control +1.3 and -0.6 mV of control for lumbar CSF. The attainment of steady-state values for ,u close to control is comparable with passive distribution of these ions between CSF and blood. acid-base fluid; ion potential
the Technical and Department
INVESTIGATORS
RESULTS
With respiratory acidosis (Fig. l), we observed a change in the CSF/plasma AE/ApH, of -29.8 rnV/pH. at the cisterna magna and - 19.6 at the lumbar site (Fig. 2). Maintaining pH, constant resulted in a constant E over the next 6 h. Our cisternal AE/ApH. is close to the -30.6 mV/pH. observed by Held et al. (6), but our lumbar value is considerably greater than the -5.6 mV/pH, reported by Hornbein and S$rensen (8). Considering the commonality of one investigator, the reason for this difference in lumbar AE/ApH, is baffling. At a constant Pace, from time 0, CSF Pco2 rose progressively over the next 6 h (Fig. 1). With a constant arterial pH, cisternal CSF pH decreased with initiation of respiratory acidosis, but then increased to a stable value by 4.5 h (Fig. 1, Table 1). Lumbar CSF pH (Fig. 1) showed no return toward control over the 6 h, in part consequent to a greater rise in CSF PCO~ and in part because of a lesser rise in [HCO 3-I (Fig. 1). We permitted the initial rise in [HCOS-] a resulting from titration of hemoglobin with CO2, but thereafter held [HCOZ-]a constant by HCl infusion to offset the usual
1145
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.
E.
G.
PAVLIN
T
,
AND
T.
Ci sterna
I=-29.8
Lumbar
= - 19.6
O
?
a’\ \\ - I **.. -.* -.I
7.20-
Cistern0
0
=\
I 0-Li
-5-
. . ..- ...*
l
..-•
...........
.@ e-0 T 0) L 0 I
/)0
....
//
mV/pH mV/pH
0
\ 0@
HORNREIN
Lumbar
...
7 25-
F.
. ..*
. ..*. l *-*--* \
I
Lumbar
I 6
c 32-
I
1
I
o
1
2 Time
1
I
1
1
3 ( hours)
4
5
6
FIG. 2. CSF/plasma DC potential difference (E) and electrochemical potential differences for H+ (pn+ ) and HC03(pncos-) vs. time during respiratory acidosis in 6 anesthetized dogs. Values o AE/ApH. were calculated for change in time 0 sample from control. Values of &kHt and pHC03are calculated for concentrations in CSF and mean capillary plasma water. Vertical bars indicate & 1 SEM.
, Arterial
Comparing time course
I ,
I
C
0
1
1
1
2 Time
I
3 (hours)
I
1
4
1
5
6
FIG. 1. Pco2, pH, and [HCOs-] of arterial plasma, cisternai m and lumbar CSF vs. time during respiratory acidosis in 6 anesthetized dogs. Vertical bars indicate =t 1 SEM.
.
compensatory processes (Fig. 1). The [HC03-]esF rose from 26.3 to 33.3 meq/kg Hz0 at the cisternal site and 26.7 to 31.7 meq/kg Hz0 in the lumbar CSF, a steady state being achieved at both locations by 4.5 h. Values for pH+ and pHe03- returned essentially to their control values by 4.5 h (Fig. 2, Table 2). DISCUSSION
The initial increase in plasma [HCOS-] with induction of respiratory acidosis exhibited a AIHCOa-]/ApH, of 9.1 slykes, which agrees well with in vivo values reported in the literature (18). Accepting this initial rise, we kept pl asma [HCO 3-1 constant thereafter to prevent a slow compensatory rise from possibly delaying establishment of a steady state of ion concentration in CSF.
lumbar and cisternal sites, we observed a of acid-base changes and return to control that were similar at both locations as well pH+ and PHCOsas identical to the behavior observed during the other types of acid-base derangements studied (7, 13, 14). The significance of this common behavior at the two locations in the presence of differing magnitudes of change in CSF/ plasma potential difference was discussed earlier (13). Although the CSF-arterial PCO~ difference at time 0 remained essentially unchanged from control (Fig. 1 ), the slope of the blood CO2 dissociation curve is such that the arterial-venous CO 2 content difference (assuming venous and CSF Pco:! to be equal) decreased. We estimate that at time 0, 1 h after onset of hypercapnia, brainstem blood flow was increased by 17 % and lumbar spinal cord flow by 9%. By 6 h, blood flow at both locations had actually fallen below control, by 9 % at the brainstem, and 16% in the lumbar area. Bleich et al.‘s (3) data seem to indicate similar changes in dogs maintained hypercapnic for 5 days. The decrease in flow as [HCO &BP increased is qualitatively compatible with Fencl et al.‘s (5) observations in man that cerebral blood flow is inversely proportional to [HCO Q-]cBF. Quantitatively our estimated blood flow ended up below control at a time when CSF pH remained distinctly acid, an observation hard to reconcile with the hydrogen ion theory of cerebral blood flow regula-
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.
CSF-BLOOD
H+
AND
HC03-
WITH
RESPIRATORY
1147
ACIDOSIS
1. Values for acid-base parameters in arterial plasma, cisternal and lumbar CSF, CSF/plasma E, and calculated electrochemical potential dijferences (p> for H+ and HCO, in six dogs during respiratory acidosis with constant plasma [HCOZ-] TABLE
Time, h Control
0
3
4.5
6
Location
[HCOs-1, meq/kg H20
PH
[Lactate-],
Pco2, torr
meq/kg
H20
E, mV
pH+# mV
I”HCO-3’
mV
Arterial CSF-cistern CSF-lumbar
7.406 7.350 7.337
=f= 0.001 =t 0.004 =t 0.007
22.6 26.3 26.7
zk 0.7 rt 0.3 + 0.3
33.6 51.0 53.5
zk * *
1.0 1.1 1.2
1.2 =t 1.8 + 1.8.h
0.2 0.1 0.1
2.6 4.4
zt +
0.4 0.9
2.0 4.3
It *
0.5 1.1
-0.1 -1.9
& *
0.6 1.2
Arterial CSF-cistern CSF-lumbar
7.230 7.217 7.188
=t =t &
0.004 0.013 0.012
24.2 28.7 28.4
zt =t rt
0.6 0.7 0.7
55.8 75.0 79.2
=t =t &
1.4 1.3 1.5
0.8 1.3 1.5
+ + *
0.1 0.1 0.1
7.8 7.9
& 0.5 =i= 1.3
5.8 6.9
=t +
0.9 1.3
-4.8 -5.3
zt zt
1.2 1.4
Arterial CSF-cistern CSF-lumbar
7.220 7.245 7.196
=t & +
0.010 0.009 0.019
24.4 31.6 30.1
=t =t =t
0.6 0.4 0.6
57.5 77.5 84.5
zk 1.6 + 1.9 zt 3.1
0.7 1.4 1.4
zk 0.1 ZiZ 0.1 =t 0.2
7.9 7.9
* zt
0.4 1.3
3.4 5.6
zk 0.7 =t 1.4
-2.3 -3.4
+ It
0.8 1.4
Arterial CSF-cistern CSF-lumbar
7.221 7.247 7.190
rt =t =t
0.010 0.009 0.010
24.2 32.8 31.5
rt zt &
0.8 0.7 0.4
57.1 79.7 88.3
zt 2.5 zt 2.5 zk 2.0
0.7 1.5 1.5
zk 0.1 =t 0.2 St: 0.2
7.7 7.9
* =t
0.4 1.3
2.5 5.4
=t zt
0.7 1.5
-1.1 -2.6
=t =t
0.6 1.4
Arterial CSF-cistern CSF-lumbar
7.225 7.246 7.188
zt rk &
0.004 0.015 0.014
24.4 33.3 31.7
zt =t +
0.5 0.8 0.6
56.8 81.2 88.0
zt 1.3 3~ 1.8 dz 2.3
0.7 1.7 1.5
It It It
7.7 7.8
Ik 0.5 IIZ 1.1
2.7 5.6
zk 0.6 If 1.5
-0.8 -2.5
zt =t
0.6 1.3
Values are means & in METHODS of ref. 13.
SE.
The
p for
H+
and
HCOs-
2. Cisternal-plasma p for H+ and HCOa-: between sustained respiratory acidosis and normal mV
Bleich
et al.
(3)
-1.6
-
Adaro
et al.
(1)
+1.4 +2.0
-1.7 -2.2
Dogs, 4 h, E simultaneously
Dogs,
5 days measured
Kazemi
et al.
(9)
+0.7
-2.0
Dogs,
hlesseter
and
SiesjB
-0.6
- 1.8
Rats, 3 days, AE/ApH measured separately
Chazan
et al.
(4)
+1.4
-1.3
Dogs,
Present
study
+0.7
-0.7
Dogs, 6 h, E simultaneously
(12)
(Ap)
Comments
pHCO3-9 mv
1 .o
calculated
diference ’
TABLE
&a+,
are
5 h
5 days measured
In all studies but the present one, metabolic compensation was permitted. E in those instances where not measured (3, 4, 9) was estimated assuming a AE/ApH of -30.5 mV for the acute respiratory component and of -42.3 mV for the compensating metabolic change. Separation of respiratory and metabolic contributions to the change in pH, and E was made assuming an in vivo buffer value of 10 slykes.
tion, yet very like what Alexander et al. (2) reported in goats during respiratory alkalosis. In both cases a blood flow lower than would be anticipated for the existing CSF pH may be simply a consequence of the decrease in cerebral blood flow with time (15) observed by us in varying degree in every acid-base derangement studied (7, 13, 14). The return of CC& and pHC03- to control values is, as we have discussed previously (13), compatible with a passive movement of H+ and HCOSbetween CSF and blood, assuming a constant rate of production of H+ by brain tissue during respiratory acidosis. In Table 2, we compare our values for p with those either provided by or derived from other studies of CSF acid-base status during
from
CSF-mean
0.1 0.3 0.2
capillary
concentration
differences
as described
sustained hypercapnia. In none of these other studies was plasma [HCO 3-1 held constant, possibly delaying achievement of steady-state conditions. Only Adaro et al. (1) measured CSF/plasma E directly, while Messeter and Siesjo (12) applied values for E derived from previous studies in rats (11). For the other studies in dogs we have applied the values of Held et al. (6) for AE/ApH, during respiratory and metabolic changes to permit an estimate of change in p. Admittedly such estimates of 1-1must be imprecise when actual values of E and assurance of steady state for both control and final values are lacking. Nevertheless, the closeness of final values for p to controls in all these studies is impressive, indicating that in spite of differences in experimental format, [HCO3-] and [H+] change in a manner expected for simple passive distribution. The studies of Adaro et al. (I), where E was actually measured, show a slightly poorer return of 1-1than the other studies, but his data after 4 h of respiratory acidosis superimposed on metabolic alkalosis do not demonstrate a steady state. On the other hand, Kazemi et al.‘s (9) final sample would be expected to be close to a steady state according to our observations. Between 0 and 6 h cisternal CSF pH rose from 7.2 17 to 7.246. This return toward the control pH CSF while arterial [HCO 3-1 and PCO~ were kept constant resulted from an increase in [HC03-lCSF engendered primarily by the rise in CSF/plasma E. The failure of lumbar CSF pH to return toward control in the face of passive distribution of H+ and HCOScan be accounted for by two factors. First, as a consequence of a lesser change in E at the lumbar location, the rise in CSF [HCOS-] was less (Fig. I). Second, CSF PCO~ increased more in lumbar than cisternal CSF. The effect of changes in the CSF-arterial Pcoz difference on the relationship between pH+ and pHCo3- will be discussed in the final paper (7). As [HCO,]. rises in response to renal and other com-
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.
1148
E.
pensatory mechanisms, [HCOs-lCSF will also rise, for the fall in CSF/plasma E with increasing blood pH is not sufficient to prevent totally the influx of HCOSinto brain extracellular fluid. This behavior is clearly illustrated by our observations during metabolic alkalosis at constant PacOp (14). Thus pH CSF will return closer to normal as systemic compensation for the respiratory acidosis occurs. The data of Bleich et al. (3), representing a more extreme the greater return of pHe*p respiratory acidosis, illustrate toward normal when [HCO& is permitted to rise. vlesseter and Siesjo’s (12) serial observations in rats also illustrate this direct relationship between CSF pH and plasma [HCOS-1. 0 ne would predict that because of the
G.
PAVLIN
AND
T.
F.
HORNBEIN
larger value for AE/ApH, occurring during metabolic compensation than with the initial respiratory acidosis, CSF pH will be below normal even when systemic compensation is complete, i.e., when pH, equals 7.40 (Fig. 3 of ref. 7). We thank Ms. Watson for patience This work was 15991-05 from the National Institutes T. F. Hornbein Award Grant 5 K03 Received
Patricia Reynolds, Mica1 Middaugh, and Mary and help. supported by Public Health Service Grant GM National Institute of General Medical Sciences, of Health. is supported by Research Career Development HE0961 7-09.
for publication
3 June
1974.
REFERENCES 1. ADARO, BERG
F. V. M.,
E. E. ROEHR, A. R. VIOLA, AND Acid-base equilibrium between fluid in acute hypercapnia. J. A@.
DE OBRUTZKY.
cerebrospinal 271-275, 1969. 2. ALEXANDER, S. C., B. E. MARSHALL, blood flow in the goat with sustained Lab.
Invest.
C.
WYMERSZ-
blood
and
Physiol.
27:
AND A. AGNOLI. Cerebral hypocarbia. Stand. J. C/in.
102, 1968.
sufq!d.
3. BLEICH, H. L., P. M. BERKMAN, AND W. B. SCHW~TZ. The response of cerebrospinal fluid composition to sustained hypercapnia. J. Clin. Invest. 43 : 11-16, 1964. 4. CHAZAN, J. A., F. M. APPLETON, A. M. LONDON, AND W. B. SCHWARTZ. Effects of chronic metabolic acid-base disturbances on the composition of cerebrospinal fluid in the dog. Clin. Sci. 36 : 345-358, 1969. V., J. R., VALE, AND J. A. BROCH. Respiration and 5. FENCL, cerebral blood flow in metabolic acidosis and alkalosis in humans. J. A#.
Physiol.
27 : 67-76,
1969.
6. HELD, D., V. FENCL, AND J. R. PAPPENHEIMER. Electrical potential of cerebrospinal fluid. J. Neurophysiol. 27: 942-959, 1964. T. F., AND E. G. PAVLIN. Distribution of H+ and 7. HORNBEIN, HCOabetween CSF and blood during respiratory alkalosis in dogs. Am. J. Physiol. 228: 1149-1154, 1975. 8. HORNBEIN, T. F., AND S. C. S$RENSEN. d-c Potential difference between different cerebrospinal fluid sites and blood in dogs. Am. J. Physiol. 223: 4.15-418, 1972. H., D. C. SHANNON, AND E. CARVALLO-GIL. Brain CO2 9. KAZEMI, buffering capacity in respiratory acidosis and alkalosis. J. App&. Physiol.
22 : 24-l-246,
1967.
of cerebrospinal fluid composition with 10. LEUSEN, I. Regulation. reference to breathing. Physiol. Rev. 52 : l-56, 1972. 11. MESSETER, K., AND B. K. SIESJ~. The DC potential between CSF and plasma in respiratory acidosis. Acta Physiol. Stand. 83 : 13-20, 1971. of CSF pH in acute 12. MESSETER, K., AND B. K. SIESJ~ Regulation and sustained respiratory acidosis. Acta Physiol. Stand. 83: 21-30, 1971. E. G., AND T. F. HORNBEIN. Distribution of H+ and 13. PAVLIN, between CSF and blood during metabolic acidosis in HC03dogs. Am. J. Physiol. 228; 1134-l 140, 1975. 14. PAVLIN, E. G., AND T. F. HORNBEIN. Distribution of H+ and HC03between CSF and blood during metabolic alkalosis in dogs. Am. J. Physiol. 228: 1141-l 144, 1975. M. E., J. B. POSNER, AND F. PLUM. Cerebral blood flow 15. RAICHLE, during and after hyperventilation. Arch. Neural. 23: 394-403, 1970. 16. SAUNIER, C., M-C AUG-LAXENAIRE, M. SCHIBI, AND P. SADOUL. Acid base and electrolyte equilibrium of arterial blood and cerebra-spinal fluid in respiratory insufficiency. Respiration 26 : 81-101, 1969. 17. SIESJ~, B. K., AND A. KJALLQUIST. A new theory for the regulation of the extracellular pH in the brain. gand. J. Clin. Lab. Invest. 24 : l-9, 1969. 18. WOODBURY, W. Regulation of pH. In: Physiology and Biophysics (20 ed.), edited by T. C. Ruth and H. D. Patton. Philadelphia: Saunders, chapt. 27. In press.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.237.035.237) on January 19, 2019.
