DETERMINATION OF THE IN VIVO CARBON DIOXIDE TTTRATION CURVE OF ANAESTHETIZED MAN

Brit. J. Anaesth, (1966), 38, 500

DETERMINATION OF THE IN VIVO CARBON DIOXIDE TTTRATION CURVE OF ANAESTHETIZED MAN BY

C. PRYS-ROBERTS, G. R. KELMAN AND J. F. NDNN

Department of Anaesthesia, University of Leeds, England SUMMARY

The relationship between pH, Pco13 and the concentration of bicarbonate ion is denned by the Law of Mass Action, commonly expressed in the form of the Henderson-Hasselbalch equation. However, changes of pH and [HCO/] in response to changes of Pco, cannot be inferred from this equation, and must be determined experimentally. These changes may then be graphically represented as carbon dioxide titration or dissociation curves. Following the report by Shock and Hastings (1935) that the carbon dioxide titration curve of arterial blood in vivo was almost identical with the in vitro curve previously described by Henderson (1928), it has been generally assumed that the relationships derived from the in vitro curve can be applied to the behaviour of arterial blood in vivo (Astrup et al., 1960). Evidence contradicting this assumption has recently been presented by Siggaard-Andersen (1962a), Cohen, Brackett and Schwartz (1964) and Presented in part before the Anaesthetic Research Group, July 1965, and published in abstract form (Bnr. J. Anaesth. (1965), 37, 553).

Norman and Linden (1965), in the form of direct comparisons of the in vivo and in vitro carbon dioxide titration curves in dogs. More recently, Brackett, Cohen and Schwartz (1965) have presented similar findings in conscious man. None of these studies has covered the full range of Paoo, seen during clinical anaesthesia but some have suggested that there is an increasing divergence between the in vivo and in vitro curves as thePa
Downloaded from http://bja.oxfordjournals.org/ at Harvard University on July 1, 2015

A comparison has been made between the pH changes in the arterial blood of anaesthetized patients exposed to changes of Pco, above and below the normal range, and those in the same blood equilibrated with similar changes of Pco, in vitro. Changes in the non-respiratory component as represented by standard bicarbonate, were derived from the in vitro equilibrations. There was a linear relationship between log Paooi and pH, and between pH and the plasma bicarbonate concentration, over the whole range studied in vivo. In the presence of wide variations of carbon dioxide tension, nitrous oxide anaesthesia of more than 2 hours duration, supplemented with a muscle relaxant and a potent analgesic, did not produce any significant alteration of the non-respiratory component of acid-base balance. Although changes of standard bicarbonate occurred in response to alterations of P a c e these changes are not considered to represent true changes of the non-respiratory component. The application of parameters derived from in vitro equilibrations, to the behaviour of arterial blood in vivo is discussed. The mechanisms of extravascular buffering of induced changes of Pco, are reviewed, and related to the behaviour of arterial blood in equilibrium with extracellular and intracellular tissue fluids.

DETERMINATION OF IN VIVO CARBON DIOXIDE TTTRATION CURVE carbon dioxide titration curves during general anaesthesia has not been satisfactorily determined. We have therefore made a comparative study of the pH changes of arterial blood from anaesthetized patients exposed to changes of carbon dioxide tension, and the same blood equilibrated in vitro with similar changes of Pco,. METHODS

studied in five patients, the effects of hyperventilation alone in three patients, and the effects of hypercapnia alone in one patient. In the first group of five patients, the minute volume was kept constant at the level established during the control period, and carbon dioxide was added to the inspired gases so that a PE'OOJ of about 75 mm Hg was attained as rapidly as possible and maintained for periods between 40 and 60 minutes. After this period of hypercapnia, the minute volume was increased to 14 l./min and the administration of carbon dioxide discontinued Hyperventilation was then maintained for a further period of 40 to 60 minutes, following which the ventilation was readjusted to the control level. In the other group of patients, in whom only the effects of hypercapnia or hyperventilation alone were studied, the return to the control level followed a period of 1 hour at an elevated or reduced Pcoa, and was allowed to occur gradually over a period of 1 hour. Finger blood flow was qualitatively measured with a photoelectric device (Videograph Phase II, Medical and Industrial Equipment), and monitored simultaneously with the e.c.g. on an oscilloscope. An indwelling Teflon cannula (Becton Dickinson & Co, Rutherford, NJ.) was inserted percutaneously into a radial artery (Barr, 1961), and continuous pressure measurements were made with an Elema-Schonander EMT.34 pressure transducer. Samples of blood (3 ml) were withdrawn from the artery into heparinized glass syringes at 10-minute intervals throughout the study, additional samples being taken 5 minutes after the changes of carbon dioxide tension. The following measurements were carried out immediately on these samples. pH was measured anaerobically at 37 °C in a capillary microelectrode (Type AME.lb, Radiometer Corp., Copenhagen). Measurements of Pco2 were made by the microequihbration technique described by Siggaard-Andersen et al. (1960), and as a check, the PcOj was also measured with a carbon dioxide sensitive electrode as described by Severinghaus and Bradley (1958). The technique, and accuracy of these measurements in this laboratory have been described by Kelman, Coleman and Nunn (1966).

Downloaded from http://bja.oxfordjournals.org/ at Harvard University on July 1, 2015

This study was carried out on six male and three female patients, aged between 24 and 60, who were undergoing surgery of a minor but lengthy nature. Each patient had agreed to participate in the study, and had been examined by one of us to exclude any condition which might be associated with acid-base disturbances. All the patients were premedicated with papaveretum 10 mg and Droperidol (dehydrobenzperidol) 10 mg given intramuscularly 1 hour before the start of the study. There was no evidence that this premedication had produced any respiratory or circulatory depression. Anaesthesia was induced with thiopentone 200-300 mg, and maintained with 70 per cent nitrous oxide and 30 per cent oxygen, except during the period of carbon dioxide inhalation, when the concentration of nitrous oxide was reduced in order to maintain a constant concentration of oxygen in the inspired gases. Muscular relaxation was obtained with tubocurarine 30-45 mg and, following endotracheal intubation, intermittent positive pressure ventilation was maintained with a Manley ventilator (Blease Anaesthetic Equipment). Analgesia was supplemented where necessary with phenoperidine 2 mg initially and 1-mg doses were repeated at hourly intervals. During a 30-minute control period, the monitoring and measuring devices were set up and calibrated, and the minute volume of ventilation was regulated to produce a steady end-tidal carbon dioxide tension (PE'OO2) between 35 and 40 mm Hg, measured with a Hartmann and Braun URAS.4 rapid infrared analyzer sampling continuously from the endotracheal tube. Rapid adjustment of the PE'CO2 could then be made at any stage during the study, by alteration of the minute volume of ventilation, or by adding carbon dioxide to the inspired gases. Following the control period, the effects of hypercapnia followed by hyperventilation were

501

502

Calculations. In order to determine the slope of the overall in vivo carbon dioxide titration line, regression coefficients (A log Paoo2/A pH) were calculated from all the measured values for pH and Paooi from each patient. Regression coefficients were also calculated from the values obtained during hypercapnia and hyperventilation separately. The slope of the whole blood in vitro line (A log Paooj/A pH) for each patient was taken as the mean of all the equilibration lines on each blood sample by the Astrup technique. The slope of the in vitro line for each sample was checked at the time of measurement, and repeated if there was a difference of slope between consecutive samples of more than ±0.05 (equivalent to a difference in haemoglobin concentration of ± 1.75 gm/100 ml). The slope of the in vitro line for separated plasma was determined by carbon dioxide equilibration of centrifuged plasma from the pooled blood samples from each patient (Siggaard-Andersen, 1964). Values for standard bicarbonate and base excess were derived from the revised curve nomogram described by Siggaard-Andersen (1962b); values

for plasma bicarbonate were also derived from the same diagram. RESULTS

The range of pH measured during this study was 7.120 to 7.666, and the corresponding range of Paco2 was 84.0 to 15.7 mm Hg. The range of Pao3 was 89.0 to 135.0 mm Hg, although in most of the patients, the Pao2 remained within the range 120 to 135 mm Hg. Table I shows details of measurements of Paoo3 and pH, together with derived values for standard bicarbonate and actual plasma bicarbonate concentrations for each patient. There was no significant difference between the standard bicarbonate values of the initial and final control period (0.2
Downloaded from http://bja.oxfordjournals.org/ at Harvard University on July 1, 2015

Measurements of Po2 were made with a macrocathode oxygen electrode (Beckman Instruments Inc., Palo Alto, Calif., U.S.A.). The nasopharyngeal temperature was measured continuously with a thermocouple, and corrections described by Rosenthal (1948) and Nunn, and colleagues (1965) were applied to the pH, Pco2, and Po, values when there was a difference between the temperature of the patient and that of the measuring electrode. The haemoglobin concentration was measured by the alkaline-haematin method on samples taken at the beginning and end of each study, and haematocrit determinations were carried out on each sample using a microhaematocrit centrifuge (Hawksley & Son Ltd.). The lactic acid concentration in arterial blood was determined in four patients in the first group, by the enzymic method described by Lundholm, Mohme-Lundholm and Vamos (1963), using a kit supplied by Biochemica "Boehringer", Mannheim. Measurements were made at 366 nanometers in an Unicam SP.50O spectrophotometer, on samples taken before, during and after the changes of Pco3.

BRITISH JOURNAL OF ANAESTHESIA

Downloaded from http://bja.oxfordjournals.org/ at Harvard University on July 1, 2015

TABLE I

Details of measurements on individual patients. Quoted values for pH, Paw*, standard bicarbonate (SLHCO,), and actual plasma bicarbonate ([HOOl) refer to mean values during each phase. Control Hypercapnia Hyperventilation

pH No.

1 2 3 4 5 6 7 8 9

Mean

Paoo, StJICO, [HC0.1

(mm Hi) (m.cquiv/1.) (m.cqulv/1.)

7.41 7.38 7.36 7.36 7.45 7.44 7.40 7.46 7.38

40.1 41.2 43.9 45.5 34.5 31.2 40.9 32.3 35.8

7.40

38.4

SD

pH

Pa«,

SLHCO, [HCO.'l

Paoo,

pH

(mm Ha) (m.cquiv/l.) (m.equiv/lj

' (mm Hg) (m.equlv/I.) (m.equlv/l.)

24.6 23.4 23.1 23.8 24.5 22.8 24.3 24.0 21.2

24.6 23.6 24.0 25.0 23.2 19.0 24.7 22.1 20.4

7.35

51.0

22.2

27.8

7.21 7.16 7.17

70.7 84.0 72.0

22.0 21.8 21.1

28.0 29.5 25.6

7.19 7.13

72.0 73.2

21.5 18.5

27.0 23.6

23.5

23.2

7.20

70.5

21.2 ±1.37

27.2

±1.05

SLHCO, [HCOa1

7.52 7.56 7.55 7.54 7.60 7.65 7.64 7.62

26.9 19.6 23.4 24.6 20.6 18.0 18.1 18.5

7.59

21.2

Paoo,

7.45 7.36 7.39 7.40 7.41 7.39 7.40 7.34 7.35

35.0 39.6 39.4 39.0 38.5 39.3 41.0 45.3 40.0

24.2 21.5 23.4 23.4 24.0 23.0 24.4 23.0 21.4

23.4 21.7 23.2 23.6 23.7 23.2 24.8 23.9 21.4

24.9

19.5

7.39

39.7

23.2 ±1.08

23.5

of the carbon dioxide titration lines.

No.

Separated plasma in vitro

Whole blood in vitro

Arterial blood in vivo

In vitro

1 2 3 4 5 6 7 8 9

-1.444 -1.282 -1.176 -1.179 -1.068 -1.272 -1.155 -1.167 -1.180

-1.693 -1.681 -1.636 -1.544 -1.549 -1.489 -1.569 -1.605 -1.627

-1.520 -1.655 -1.434 -1.448 -1.472 -1.306 -1.435 -1.470 -1.322

0.898 0.985 0.876 0.938 0.950 0.877 0.915 0.916 0.810

-1.383 -1.425 -1.281

-1.202 ±0.114

-1.599 ±0.067

-1.451 ±0.129

0.907 ±0.051

-1.371 ±0.092

In vivo

SLHCO, [HCO.I

(mm He) (m.cquiv/1.) (m.cquiv/l.)

21.0 17.0 19.8 20.2 19.4 19.0 18.6 18.2

TABLE II

Mean SD

Final Control

24.8 25.2 24.2 24.5 25.2 25.5 25.0 24.5

±0.44

Details of the slopes

pH

Arterial blood in vivo hypercapnia

Arterial blood in vivo hyperventilation

-1.456

-1.602 -1.655 -1.434 -1.388 -1.426 -1.287 -1.465 -1.486

-1.356 -1.322

-1.468 ±0.116

BRITISH JOURNAL OF ANAESTHESIA

504 [HCO 3 *]

m .Eq/litre

40 -

35 "

30

20

PLASMA

IN VITRO

ARTERIAL BLOOD IN VIVO

15 WHOLE M.OOD IN VITRO

70

7-1

7-2

73

7-4

7-5

7-6

7-7 pH

1 Relationship between in vivo and in vitro carbon dioxide titration lines for all patients in the series. The points on the line representing the response of arterial blood in vivo are derived from the mean values quoted in table I, and the extreme values measured in individual patients. FIG.

AfHCO,] Whole blood in vitro = 33.0 m.equiv/l./pH Arterial blood in vivo =19.3 m.equiv/L/pH Plasma in vitro =11.0 m.equiv/L/pH

which represents the rise from the control level to the level of maximal hypercapnia follows a different course to that of the in vitro line. The lines which represent the fall of Paooa during hyperventilation and the rise of Paooa during the return to the control level are coincident, although the slopes of these lines are also different from that of the in vitro line. Figure 3 shows representative plots from two separate patients, one of whom was subjected to hypercapnia followed by a gradual return to the control level, while the other was hyperventilated prior to a gradual return to the control leveL The difference between the slopes of the in vivo res-

ponse to hypercapnia and hyperventilation was not significant (0.2
Downloaded from http://bja.oxfordjournals.org/ at Harvard University on July 1, 2015

25

Downloaded from http://bja.oxfordjournals.org/ at Harvard University on July 1, 2015

Pacoj-mm Hg

,- mm Hg

IN VITtO

80 -

npF i - 1i Dl C1 . ei al.urc \

In Vitro

SO

\

\

12

V) \ 60 -

V

VIVO -

UfPEICtPNIt SLOPE- - 1-31

60 •

V In Vi»o

•>

Slope«-1 30

\

\ (.0 20C

18

22 \ \ A

'

2t

28

SMMUID 8IUIB0»TE - mEq/litre

to

A

o _

A

D O

N<-IN

CONTROL HTPERCAPWIA

;• X220

»

\

\

HTPEtVENTIUTION •FTURH TO CONTIOL POST-JNWSTHETIC

IN VIVO RESPONSE HYPERCAPNIA AND HYPERVENTILATION

£ \ 26 \

C WPERC»PNI»- CONTROL O HTPEIUPNIt • HTPERCAPNIA- •ETUDN TO CtNTIBL

VIVO - HVPERVEHTILATIOH SLOPE- - V 4 9

21

\

TO

ST«NO«RO IICAISOKATf - m.Eq/litr«

\

\ \ A HVPEIVENTILtTION-CONTROL A HTPERVENTIUTION A HTPERVEMTILATIOM - RETURN TO CONTIOL

\

-

\

20 -

\ \

\

\ \

vL

IN VIVO CO, TITRATION CURVE \ i

7-2

i

7-3

la V110

Slope --1 IE

i

I

i

7-4

75

7-6

pH

FIG. 2 Relationship between in vivo and in vitro carbon dioxide titration lines in a typical patient exposed to hypercapnia and hyperventilation consecutively.

In Vitro \ Slope • - ) 5r

\

1

72

7-3

7-4

7-5

7-6

pH

FIG. 3 Relationship between in vivo and in vitro carbon dioxide titration lines In two separate patients, exposed to hypercapnia and hyperventilation

respectively.

BRITISH JOURNAL OF ANAESTHESIA

506 DISCUSSION

to changes of Pcoa. Brackett, Cohen and Schwartz (1965) have recently described a direct comparison of the in vivo and in vitro carbon dioxide titration curve in conscious subjects breathing spontaneously in an atmosphere containing an excess of carbon dioxide. A log Paco2/pH line derived from their in vivo data had a mean slope of -1.18, which represents a buffering response of the whole blood in vivo which is weaker than that of separated plasma in vitro, a finding which we are unable to interpret. We were unable to confirm their quoted linear relationship between Paooj and [H+] (nanomolar concentration) over the entire range that we have studied in vivo, although it was very nearly linear above a Paoo2 of 40 mm Hg. It appears, therefore, that although most workers are agreed that the in vivo response of arterial blood to changes of Paoo3 differs from that of the same blood in vitro, there is no general agreement on the magnitude. Furthermore, Bunker (1965) has drawn attention to the considerable disagreement as to the validity of parameters derived from in vitro measurements when these are applied to the response of the patient as a whole. The difference in the magnitude of this effect under different conditions has prompted us to control certain factors which might influence the acid-base response to changing Pco2 during anaesthesia. These factors are : (1) Maintenance of a constant oxygen saturation of arterial blood at a level slightly above the normal value. (2) Maintenance of constant conditions of ventilation during each phase of the study, by automatic ventilation of the lungs, with complete muscular relaxation. (3) Avoidance of the administration of flujd intravenously, such as blood, citrate, or dextrose solutions, in an attempt to prevent extraneous alterations in the non-respiratory component of acid-base equilibrium (Huckabee, 1958). That these measures have been successful is suggested by the remarkable stability of the in vivo titration line during the course of anaesthesia of more than 2 hours duration. Although the slope of the in vivo line was different from that of the in vitro line, there was little or no shift of the intercept of the former along the isobar representing a Paco, of 40 mm Hg. This was supported by

Downloaded from http://bja.oxfordjournals.org/ at Harvard University on July 1, 2015

Some idea of the carbon dioxide titration curve during anaesthesia in man during hyperventilation can be made from data collected in the studies by Robinson (1960), Papadopoulos and Keats (1959), and more recently by Millar and Marshall (1965). Our analysis of Robinson's data showed a series of log Paoon/pH lines having a mean slope of -1.28. Assuming that the patients had normal haemoglobin values, this slope would be indicative of an apparent metabolic alkalosis during hyperventilation. Papadopoulos and Keats (1959) also reported their findings as indicative of a metabolic acidosis arising during hyperventilation, although this only occurred after a period of more than 1 hour. Their control values, and those of samples measured up to 1 hour, he on an in vivo line having a mean slope of - 1.58 for twenty subjects. This is unlikely to represent a metabolic acidosis unless their subjects had a mean haemoglobin concentration of more than 16 gm/100 ml. The findings of Millar and Marshall (1965) are difficult to interpret owing to the wide scatter of values, calculated from their data, for the slope of consecutive in vitro lines in the same patient. Less information has been available on the effects of hypercapnia in anaesthetized man, though Holaday, Ma and Papper (1957) reported the onset of metabolic acidosis which tended to be proportional to the degree of carbon dioxide retention in patients who were allowed to hypoventilate for long periods. The data available on anaesthetized animals suggest that there may be a species difference in acid-base response. Shaw and Messer (1932) in a well controlled study, showed a difference between the response of cats and dogs while breathing 10-11 per cent carbon dioxide, although their results during hyperventilation were equivocal. Their results in dogs, however, have been well confirmed by Siggaard-Andersen (1962a), Morris and Millar (1962), Norman and Linden (1965), and Brown and Clancy (1965). Our analysis of their respective data gives a mean value of - 1.30 for the slope of their in vivo lines, this being equivalent to a haemoglobin concentration of about 5 gm/100 ml. Our data are in reasonable agreement with those of Cunningham, Lloyd and Michel (1962) who describe the in vivo responses of conscious man

DETERMINATION OF IN VIVO CARBON DIOXIDE TTTRATION CURVE

507

Downloaded from http://bja.oxfordjournals.org/ at Harvard University on July 1, 2015

the initial and final control values of standard environment of blood in a tonometer and that in bicarbonate, before and after the changes of Paoo2 a dynamic equilibrium with other body fluids. (table I), and implied that there had been no true Figure 1 shows the difference between the increchange of the non-respiratory component as a mental rise of plasma bicarbonate concentration of result of the variations of Paoo2whole blood and separated plasma in vitro to carIt would appear that despite deliberate wide bon dioxide-induced changes of pH. The resvariations of the Paoou the influence of the ponse of arterial blood in vivo lies between these, anaesthetic technique used in this study has been suggesting that it equilibrates with a fluid phase confined to changes in the respiratory component having a buffering capacity which is less than that of acid-base equilibrium. The source of non-res- of whole blood but greater than that of plasma. piratory changes during anaesthesia cannot thus Due to the efficiency of the lungs as tonometers, be related to ventilatory factors. the arterial Pco2 rapidly follows the changes of During hypercapnia there was a mean fall of alveolar Pco^ and further equilibration of the standard bicarbonate of 2.4 m.equiv/L in response blood with the extracellular fluid has been shown to a mean rise of Pacoj of 32.1 mm Hg. During to be very rapid (Brown and Clancy, 1965). If hyperventilation, there was a mean rise of standard blood were in equilibrium only with the extrabicarbonate of 1.3 m.equiv/1. in response to a fall cellular fluid, which has no efficient buffering of Paooa of 172 mm Hg (table I). It is clear that mechanism against changes of pH induced carbon these changes of standard bicarbonate do not rep- dioxide we should expect the carbon dioxide resent true alterations of the non-respiratory com- titration lines for arterial blood in such equiliponent. If standard bicarbonate is required to give brium to have a slope of about —1.30. This value an indication of the true non-respiratory state, is calculated by assuming a dilution of 5 1. of then allowance should be made for the phenome- whole blood containing 15 gm/100 ml of haemonon which we have described, and the measured globin, with about 10 1. of extracellular fluid. standard bicarbonate may be corrected as follows: The mean slope which we have found for arterial blood in vivo is -1.45 (table II), and thus Corrected standard bicarbonate must represent equilibration of the whole blood =Measured standard bicarbonate+ with a fluid phase having a buffering capacity 0.075 (Paoo 3 -40) which is higher than that of extracellular fluid Alternatively, the base excess may be corrected: alone. Equilibration would have reached an acute Corrected base excess steady-state within the period of 1 hour for which = Measured base excess + the changes of Pcoa were maintained in our study 0.0975 (Paoo 3 -40) (Giebisch, Berger and Pitts, 1955), and since no The assumption is made that base excess = 1.3 measurable deviation from the slope of our in vivo (standard bicarbonate - 24), which is only approxi- titration line has occurred after the first 5 minutes mately true as the ratio depends upon the actual within this period, we have assumed that intracelvalue of base excess and the haemoglobin concen- lular mechanisms must have contributed within tration (Siggaard-Andersen, 1964). these first few minutes to the buffering of the When the Paj does not deviate significantly extracellular changes of Pco2 and pH. This from the normal limits, it will be seen that the assumption is given support by the findings of error introduced by the use of derived parameters Brown and Clancy (1965), and the reviews of the such as standard bicarbonate is small, and for the intraceUular acid-base mechanisms by Elkinton purposes of clinical assessment its effect should not (1956) and Robin, Wilson and Bromberg (1961). be exaggerated. Adler, Roy and Relman (1965a, b) have demonThe explanation of the difference between the strated the titration lines of muscle cells in resresponse of arterial blood to in vivo and in vitro ponse to changes of extracellular Pco3 and equilibration with changes of Pcoa provides ample [HCOj'] in vitro. Intracellular pH remains relaopportunity for hypothesis. Schwartz and Relman tively constant when the extracellular Pco2 varies (1963) have drawn attention to the different between 40 and 70 mm Hg; this means that the

508

BRITISH JOURNAL OF ANAESTHESIA

Downloaded from http://bja.oxfordjournals.org/ at Harvard University on July 1, 2015

Bracken, N. C , Cohen, J. J., and Schwartz, W. B. (1965). Carbon dioxide titration curve of normal man: effect of increasing degrees of acute hypercapnia on acid-base equilibrium. New Engl. J. Med., 272, 6. Brown, E. B. jr., and Clancy, R. L. (1965). In vivo and in vitro CO, blood buffer curves. J. appl. Physiol., 20, 885. Bunker, J. P. (1965). Editorial: The great transatlantic acid-base debate. Anesthesiology, 26, 591. Cohen, J. J., Bracken, N. C , and Schwartz, W. B. (1964). The nature of the carbon dioxide titration curve in the normal dog. J. elm. Invest., 43, 777. Cunningham, D. J. C , Lloyd, B. B., and Michel, C C (1962). Acid-base changes in the blood during hypercapnia and hypocapnia in normal man. J. Physiol. (Lend.), 161, 26P. Elkinton, J. R. (1956). Whole body buffers in the regulation of acid-base equilibrium. Yale J. Biol. Med., 29, 191. Singer, R. B., Barker, E. S., and Clark, J. K. (1955). Effects in man of acute experimental respiratory alkalosis and acidosis on ionic transfers in the total body fluids. J. elm. Invest., 34, 1671. Such a migration of ions from within the cells Fenn, W. O. (1961). Carbon dioxide and intracellular during periods of altered carbon dioxide tension homeostasis. Ann N.Y. Acad. Set., 91, 547. has been reviewed by Fenn (1961), and can be Giebisch, G., Berger, L., and Pitts, R. F. (1955). The extrarenal response to acute acid-base disturbances predicted by application of the Donnan equiliof respiratory origin. J. elm. Invest., 34, 231. brium. Experimental confirmation of these ionic Henderson, L. J. (1928). Blood: A Study m General Physiology, p. 127. New Haven. shifts was given simultaneously by Elkinton and Holaday, D. A., Ma, D., and Papper, E. M. (1957). associates (1955) and Giebisch, Berger and Pitts The immediate effects of respiratory depression (1955) who demonstrated an exchange of sodium on acid-base balance in anesthetized man J. clin. Invest., 36, 1121. and other unidentified ions for hydrogen ions Huckabee, W. E. (1958). Relationships of pyruvate and across the cell membrane during acutely induced lactate during anaerobic metabolism. I : Effects acid-base disturbances. of infusion of pyruvate or glucose and of hyperventilation. J. elm. Invest., 37, 244. J0rgensen, K., and Astrup, P. (1957). Standard bicarbonate, its clinical significance and a new method ACKNOWLEDGEMENTS for its determination. Scand. J. clin. Lab Invest., 9, 122. We wish to express our thanks to Professor J. C Goligher and Mr. C G. dark who allowed us to Kelman, G. R., Coleman, A. J., and Nunn, J. F. (1966). Assessment of capillary pH electrode and microstudy patients under their care, and to Dr. R. Heytonometer system. J. appl. Physiol. (in press). worth of the Department of Chemical Pathology, University of Leeds, who kindly performed the estima- Lundholm, L., Mohme-Lundholm, E., and Vamos, N. (1963). Lactic acid assay with L( + ) lactic add tions of lactic acid concentration. dehydrogenase from rabbit muscle. Acta. physiol. This study was supported by a grant from the scand., 58, 243. Medical Research Council. Millar, R. A., and Marshall, B. E. (1965). Acid-base changes in arterial blood associated with spontaneous and controlled ventilation during anaesREFERENCES thesia. Brit. J. Anaesth., 37, 492. Adler, S., Roy, A., and Relman, A. S. (1965a). Intra- Morris, M. E., and Millar, R. A. (1962). Blood p H / plasma catecholamine relationships: respiratory cellular acid-base regulation. I : The response of acidosis. Brit. J. Anaesth., 34, 672. muscle cells to changes of CO, tension or extraNorman, J., and Linden, R. J. (1965). Hyperventilation cellular bicarbonate concentration. J. clin. Invest., and acid-base balance. Brit. J. Anaesth., 37, 290. 44, 8. (1965b). Intracellular acid-base regula- Nunn, J. F., Bergman, N. A., Bunatyan, A., and Coleman, A. J. (1965). Temperature coefficients for tion. I I : The interaction between CO, tension Pco, and Po, of blood in vitro. J. appl. Physiol., and extracellular bicarbonate in the determination 20, 23. of muscle cell pH. J. clin. Invest., 44, 20. Astrup, P., Jdrgensen, K., Siggaard-Andersen, O., and Papadopoulos, C. N., and Keats, A. S. (1959). The metabolic acidosis of hyperventilation produced by Engel, K. (1960). The acid-base metabolism: a controlled ventilation. Anesthesiology, 20, 156. new approach. Lancet, 1, 1035. Barr, P-O. (1961). Percutaneous puncture of the radial Robin, E. D., Wilson, R. J., and Bromberg, P. A. artery with a multipurpose Teflon cannula for (1961). Intracellular acid-base relations and intraindwelling use. Acta. physiol. scand., 51, 643. cellular buffers. Arm. N.Y. Acad. Set., 92, 539. buffering capacity of the cell is very high compared with that of other body fluids. The rise of [HCO,'] within the cell, for the range of Pco, change from 40 to 70 mm Hg, is therefore much greater than it is in blood. Since the cell membrane is relatively permeable to bicarbonate ions (Adler, Roy and Relman, 1965a), it is reasonable to assume that there will be a migration of bicarbonate ions from cells into the extracellular fluid in response to a concentration gradient. The rate and extent to which this occurs will influence the corresponding gradient across the capillary membrane, and thus the rate at which bicarbonate ions will migrate out of the blood (Shaw and Messer, 1932); this in turn will influence the slope of the carbon dioxide titration line for arterial blood in vivo.

DETERMINATION OF IN VIVO CARBON DIOXIDE TTTRATION CURVE

ETUDE DE LA COURBE DE TITRAGE DU GAZ CARBONIQUE IN VIVO CHEZ L'HOMME ANESTHESIE SOMMAIRE

On a fait une comparaison entre les modifications du pH Hung le sang arteriel de malades anesthesies exposes a des changements de la Pco, au-dessus et au-dessous de la normale, et ceux du meme sang equilibre par des changements analogues de la Pco, in vitro. Les changements de la composante non respiratoire comme ils sont representes par le bicarbonate standard, ont itt derives des etudes in vitro. II y a eu un rapport lineaire entre log Pawj et le pH, et entre

le pH et la concentration du plasma en bicarbonate, sur Pensemble des valeurs etudiees in vivo. En presence de grandes variations de la tension en gaz carbonique, l'anesthesie a l'oxyde nitreux pendant plus de 2 heures, completee par un myorelaxant et un analgesique puissant, n'a pas grandement alteri la composante non respiratoire de l'equilibre acidobasique. Bien qu'il y ait eu des changements du bicarbonate standard en reponse a des modifications de la Pacoi, on considere que ces changements ne representent pas les vrais changements de la composante non respiratoire. On dispute l'application des parametres derives des equilibrations in vitro au comportement du sang arteriel in vivo. On passe en revue les mecanismes du tampon extra-vasculaire des changements induits de la Pco,, et on les met en rapport avec le comportement du sang arteriel en equilibre avec les liquides extra- et intracellulaires. BESTIMMUNG DER IN VTVO-KOHLENDIOXYDTTTRATIONSKURVE BEIM NARKOTISIERTEN MENSCHEN ZUSAMMENFASSUNG

Die pH-Veranderungen im arteriellen Blut von narkotisierten Patienten, die Pco,-Verschiebungen nach oberhalb und unterhalb des Normbereiches ausgesetzt worden waren, wurden mit den in vitro-Veranderungen am gleichen Blut mit Shnlichen Pco,-Verschiebungen verglichen. Aus den in vitro-Berechnungen wurden die Veranderungen der nicht-respiratorischen Komponente, reprasentiert durch das Standardbikarbonat, abgeleitet. Uber den gesamten in vivo untersuchten Bereich fand sich eine Uheare Beziehung zwischen dem Logarithmus des Paoo, und dem pH und zwischen dem pH und der Bikarbonatkonzentration im Plasma. Eine Lachgasnarkose von mehr als zwebtiindiger Dauer mit Unterstiitzung durch ein Muskehelaxans und ein stark wirkendes Analgetikum fuhrt im Falle von grofJen Abweichungen der Kohlendioxydspannung zu keiner signinkanten Veranderung der nicht-respiratorischen Komponente des Saure-Basen-Gleichgewichtes. Obwohl Veranderungen des Standardbikarbonats in Reaktkm auf die Paoo,-Abweichungen auftraten, konnen diese Veranderungen nicht als echte Verschiebungen der nichtrespiratorischen Komponente angesehen werden. Die Anwendung der aus den in vitro-Gleichungen abgeleiteten Parameter auf das Verhalten des arteriellen Blutes in vivo wird besprochen. Die Mechanismen der extravaskularen Abpufferung von induzierten Veranderungen des Pco, werden angefuhrt und zu dem Verhalten des arteriellen Blutes im Gleichgewicht mit der extrazellularen und intrazellularen GewebsflufiiKkeit in Beziehung gesetzt.

Downloaded from http://bja.oxfordjournals.org/ at Harvard University on July 1, 2015

Robinson, J. S. (I960). Observations on the effects of passive hyperventilation in human subjects. MJD. Thesis, Liverpool University. Rosenthal, T. B. (1948). The effect of temperature on the pH of blood and plasma in vitro. J. biol. Chem., 173, 25. Schwartz, W. B., and Relman, A. S. (1963). A critique of the parameters used in the evaluation of acidbase disorders. New Engl. J. Med., 268, 1382. Severinghaus, J. W., and Bradley, A. F. (1958). Electrodes for blood Po, and Pco, determination. J. appl. PhysioL, 13, 515. Shaw, L. A., and Messer, A. C (1932). The transfer of bicarbonate ion between the blood and tissues caused by alterations of the carbon dioxide concentrations in the lungs. Amer. J. PhysioL, 100, 122. Shock, N. W., and Hastings, A. B. (1935). Studies of acid-base balance of blood. IV: Characterisation and interpretation of displacement of acid-base balance. J. biol. Chem., 112, 239. Siggaard-Andersen, O. (1962a). Acute experimental acid-base disturbances in dogs: an investigation of the acid-base and electrolyte content of the blood and urine. Scand. J. clin. Lab. Invest., 14, Suppl. 66. (1962b). The pH, log Pco, blood acid-base nomogram revised. Scand. J. clin. Lab. Invest., 14, 598. (1964). The acid-base Status of the Blood, 2nd ed. Chapter 3, p. 38. Copenhagen: Munksgaard. Engel, K., J0rgensen, K., and Astrup, P. (1960). A micro method for determination of pH, carbon dioxide tension, base excess, and standard bicarbonate in capillary blood. Scand. J. clin. Lab. Invest., 12, 172. Singer; R. B., and Hastings, A. B. (1948). Improved clinical method for estimation of disturbances of acid-base balance of human blood. Medicine, 27, 223.

509