Alternative measurements of the metabolic component of the acid-base status in acute clinical situations.

C . T. KAPPAGODA, R. J. LINDEN, H. M. SNOW AND J. B. STOKER, Cardiovascular Unit, Department of Physiology, The University, Lee&, 6 August 1973 In an attempt to decide on a clinically acceptable method of assessing the metabolic component of the acid-base status of blood, Lewis & Stoddart [Clinical Science (1973) 44, 297-3001 compared two indices of the non-respiratory state of blood in vivo. These were the nonrespiratory pH (Kappagoda, Linden & Snow, 1970; Stoker, Kappagoda, Grimshaw & Linden, 1972) and the ‘corrected’ base excess (Prys-Roberts, Kelman & Nunn, 1966). Lewis & Stoddart (1973) were able to demonstrate a correlation between these two indices and concluded that the corrected base excess was the method of choice ‘because it retained the familiar terminology of Astrup, Jorgensen, Siggaard-Andersen & Engel (1960)’. However, any method which is recommended for clinical use in the assessment of nonrespiratory acid-base disturbances must satisfy one important criterion, i.e. the original data upon which the method is based must refer to changes which occur in blood in vivo. This requirement is based upon the finding that the slope of the COz titration curve of blood in vivo differs from that of the COz titration curve in vitro. Thus, concepts such as base excess and buffer anion concentration (Singer & Hastings, 1948), which are dependent on data obtained from blood in vitro, cannot be used to diagnose disturbances which occur in the body. In an attempt to circumvent these objections, Prys-Roberts et al. (1966) attempted to ‘correct’ the base excess values obtained from the conventional Astrup et al. (1960) nomogram using the following expression : corrected base excess = measured base excess 0.0975 (Pa,coz 40). It was their intention to compensate for the difference between the slope of the C02 titration curve in vivo and that of the COz titration curve in vitro. This equation, which is based upon results obtained from subjects without a non-respiratory disturbance, implies that the ‘correction factor’ is solely dependent on Pa,co,. It has, however, been shown that the slope of the COz titration curve changes with the degree of non-respiratory acidaemia (Kappagoda et al., 1970; Stoker et al., 1972) and therefore the ‘corrections’ suggested by Prys-Roberts et al. (1966) would not be applicable in these situations. Further, the existence of a correlation merely indicates a relationship between the two parameters considered and does not permit any comment to be made about the relative merits of any one index as suggested by Lewis & Stoddart (1973). For example, we have re-examined our acid-base data obtained from a series of children with congenital heart disease and have found equally significant correlations between the non-respiratory pH and a variety of other indices of the non-respiratory status of the blood, as shown below: Regression