Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Clinical Science and Molecular Medicine (1976) 50, 177-184. The haemodynamic effects of metabolic acidosis in the rat J . YUDKIN, R. D. COHEN AND BARBARA SLACK The Metabolic and Endocrine and Medical Units, The London HospitaI Medical College, University of London (Received 21 August 1975) Summarv 1. The effect of metabolic acidosis of 4-6 h duration on cardiac output, blood pressure, heart rate, and hepatic and renal blood flow has been studied in the rat. 2. In anaesthetized rats, blood pressure and heart rate fell linearly with blood pH in both shamoperated and nephrectomized rats. There was no significant difference between the two groups in the effect of acidosis on either variable. 3. Cardiac output showed a significant fall with increasing acidosis in the conscious rat. 4. Estimated hepatic blood flow in wnscious rats showed a significant positive correlation with blood pH in both sham-operatedand nephrectomized animals. There was no significant difference in estimated hepatic blood flow between the two groups of animals at any blood pH. 5. In conscious rats, increasing acidosis caused a progressive decrease in estimated renal blood flow. 6. It is concluded that the increase in the previously described apparent renal contribution to lactate removal in the acidotic rat cannot be explained by any circulatory effect mediated by the kidney. The possible relevance of the findings to lactate homeostasis is discussed. Key words: acidosis, blood pressure, cardiac Correspondence: Professor R. D. Cohen, Unit of Metabolism and Endocrinology, The London Hospital, Whitechapel, London El 1BB. C output, hepatic blood flow, lactate metabolism, renal blood flow. Introdwtion In a previous study (Yudkin & Cohen, 1975) it was shown that in the consciousrat at normal blood pH, nephrectomy resulted in approximately30% slowing of absolute lactate removal after administration of an intravenous lactic acid load. Furthermore, in severely acidotic rats, nephrectomy produced more marked slowing of absolute lactate removal than occurred in rats of normal acid-base status. In the same study it was demonstrated that renal excretion of lactate ions could account for only a small fraction of the renal component of lactate removal. The slowing of lactate removal with nephrectomy could theoretically be due to two other mechanisms. First, the kidney might contribute directly to the metabolic removal of lactate ions. Secondly, nephrectomy might iduence lactate removal in the remainder of the body, either through circulatory changes, or by some humoral mechanism. In the present experiments the haemodynamic effects of acidosis and nephrectomy in the rat have been studied in order to determine whether the changes in lactate metabolism seen in the acidotic animal could result from circulatory changes. The studies also had the more general purpose of establishing the detailed pattern of haemodynamic response to acidosis and included measurement of hepatic and rend blood flow as well as of heart rate, blood pressure and cardiac output. 177 178 J. Yudkin, R. D. Cohen and Barbara Slack Methods Experimental design Experiments were performed on male SpragueDawley rats weighing between 200 and 350 g. There were no significant differences in mean weight in the various groups of animals studied. The animals were maintained on a standard laboratory diet (Dixon's Rat Cake type 41B; E. Dixon and Sons, Ware, Herts.) and were permitted free access to food and water. All animals studied in the conscious state had previously been accustomed to restraining cages. Measurements were performed on rats of normal acid-base status, and on rats made acidotic to various degrees by oral administration of ammonium chloride solution (1.87 mol/l) 4-6 h earlier, in doses of 15-40 mmollkg body weight. Blood pressure and heart rate measurements Measurementswere made on both nephrectomized and sham-operated rats; these renal operations were performed by methods previously described (Yudkin & Cohen, 1975) 4-6 h before the measurements were taken. Blood pressure was recorded directly from a carotidcannula, inserted immediately beforehand under light ether anaesthesia,with a semiconductor strain-gauge transducer (type S.E.3/81; S.E. Laboratories Ltd) and ultraviolet recorder (type S.E. 2005). Heart rate was measured from the blood pressure trace. Blood pH was measured on a sample withdrawn from a right atrial cannula implanted 1-3 days previously. Cardiac output measurements The changes in cardiac output during induction of metabolic acidosis with ammonium chloride were determined in three rats; two of these animals were also studied after administration of similar amounts of sodium chloride. An electromagnetic flowmeter (Browning, Pelling & Ledingham, 1969) was implanted on the ascending aorta 2 weeks before the experiment, and an aortic polypropylenecannula (Browning, Ledingham & Pelling, 1970) was inserted 1 week later. Measurements were made on conscious animals before, and at hourly intervals after, administration of sodium chloride solution (1 -87 mol/l) or ammonium chloride solution (1.87 mol/l) in a dose of 40 mmol/kg body weight. An interval of 8 days separated the studies on rats which were given both solutions. Aortic blood pH was measured at each time, the volumes being replaced with rat donor blood. Heart rate was measured from the cardiac output trace. Hepatic and renal bloodflow The small blood volume of the rat imposes limitations on the volume of blood samples obtained from these animals; even minor degrees of hypovolaemia substantially reduce hepatic and renal blood flow (Sapirstein, Sapirstein & Bredemeyer, 1960). An experimental protocol was therefore devised which required only two blood samples to be taken. The techniques used depended on analysis of the rate of decline of blood concentration of intravenouslyadministeredIndocyanine Green (ICG) and sodium p-aminohippurate (PAH). In man, the decay curve of ICG") after an intravenous injection is of single exponential form (Caesar, Shaldon, Chiandussi, Guevara & Sherlock, 1961); it has been assumed that this form of decay curve also occurs in the rat. The fractional removal rate (KIco)and volume of distribution (VIco) were determined from the slope and the zero intercept respectively of the line joining two values of ICG concentration obtained at two time-intervals after injection and plotted on semilogarithmicpaper. Hepatic extraction of the dye (EICG)at various blood pH values was measured in a separate group of animals and these values were used to obtain an estimate of hepatic blood flow (ml min-I kg-I body wt.) as follows: Hepatic blood flow (min-I) - VICG (ml/kg body wt.) x Klcc Eicc x (1 - PCV) where PCV represents packed cell volume, assumed to be 0.42 (Yudkin & Cohen, 1975). The decay curve of PAH after an intravenous injection in dogs is of double-exponential form (Mandel, Vidt & Sapirstein, 1955), and evidence was presented by these workers suggesting that the more rapid exponential represents distribution of the substance, and the slower exponential represents excretion. It has been assumed that a similar decay curve occurs in the rat, and that in the nephrectomized animal the decay curve is of single exponential form to a plateau concentration which represents equilibration of the substance in the total volume of distribution. It has further been assumed that at any (') Abbreviations: ICG, p-aminohippurate. Indocyanine Green; PAH, Haemdynamic effectsof acihsis value of blood pH the volumes of distribution of PAH and the distribution constant are unaffected by nephrectomy. Given these assumptions, the principle of the method is to obtain the characteristics of the fist exponential (i.e. distribution) from observations in the nephrectomized animal at different blood pH values and to use this information to extract from the data obtained in the shamoperated animals the characteristics of the second exponential (i.e. excretion). Values were obtained for plasma concentration of PAH after complete distribution ( Pm ) by withdrawing a single blood sample from the nephrectomized rat 30 rnin after an intravenous injection of PAH. The total volume of distribution of PAH (VmPAH) may be calculated from these values. In another group of nephrectomized animals, blood samples were withdrawn at 4 and 8 rnin after administration of PAH. From the measurements of PAH concentrations at these times, and the extrapolated value at t = ffi (Pm), it is possible to calculate (Yudkin, 1975) the following three values by the method shown in the table deposited with the Librarian, The Royal Society of Medicine, 1 Wimpole Street, London W1M 8AE (Clinical Science and Molecular Medicine, Table 7518; copies may be obtained by application to the Librarian): the theoretical plasma concentration of PAH at zero time (Po),the theoretical initial volume of distribution (VopAH),and the distributional constant (KZpAH).For sham-operated rats, blood concentrations of PAH were determined at 4 and 8 min after intravenous administration of PAH, and by using these figures supplementedwith appropriate vdues ofPo,P mand KIPAH obtained from regression lines of these variables against blood pH obtained from observations on nephrectomized animals, the excretoryfraction removal rate (KZpAH) was calculated. Renal blood flow was derived as in the deposited Table (see above) from the values of yopAH and KZpAH, and from estimates of renal extraction of PAH ( E p A H ) at various blood pH values made in a separate group of animals: VOPAH x KjPAH Renal blood flow = EPAH x (1 - PCV) where PCV is the packed cell volume, assumed to be 0.42 in all experiments. The experimental design consisted of administration of a bolus of ICG and PAH to the conscious resting rat through a right atrial cannula 4-6 h 179 after either nephrectomy or sham-operation, and after administration of ammonium chloride solution as described previously. The doses given were ICG 5.16 mollkg body wt. and PAH 51-5 mollkg body wt. in a volume of 2.5 mllkg body wt. Blood pH was measured on a preinfusion sample. At 4 min and 8 min after the injection, a sample of blood (0.5 ml) was withdrawn from the cannula in the opposite jugular vein for determination of ICG and PAH concentrations. In a separate group of bilaterally nephrectomized rats, a single sample was withdrawn 30 rnin after injection for determination of PAH concentration. Af'ter completing the observations on each shamoperated rat, the animal was anaesthetized with sodium pentobarbitone (0.24 molll), 1 mlFg body wt. by intraperitoneal injection. A constant infusion of a mixture of ICG and PAH was commenced through a right atrial cannula to administer approximately ICG, 0,323 m o l min-' k g - I body wt. and PAH, 3.09 rmol min-1 kg-I body wt. A tracheostomy was performed and the animal ventilated. Approximately 10 min after commencing the constant infusion, an abdominal and thoracic incision was made and blood samples (0.5 ml) were withdrawn in turn from the right renal vein and the abdominal aorta. Clamps were then placed on the inferior vena cava just caudal to the liver, and on the inferior vena cava at its junction with the right atrium, and a further blood sample (0.5 ml) was withdrawn from the inferior vena cava between the clamps. The whole procedure took less than 1 min in each experiment. Aortic blood was analysed for pH, and for ICG and PAH concentration; renal vein and inferior vena cava ('hepatic vein') blood were analysed for PAH and ICG concentrations respectively. The percentage extractions of ICG and PAH (EIco and EPAH) were then calculated. Analytical methods Blood pH was measured with a blood microsystem acid-base analyser (model BMS-3; Radiometer, Copenhagen) standardized with buffers of nominal pH 6.841 and 7.383. ICG was determined in diluted plasma within 3 h of taking the blood sample by measurement of extinction at 805 MI. Readings were unaffected by alteration of pH between 5.5 and 7.5, and by the presence of PAH in the plasma in concentrations above those reached in the current experiments. J. Yudkin, R. D. Cohen and Barbara Slack 180 The determination of PAH was made on plasma samples diluted 1:2 in sodium chloride solution (154 mmol/l) by the Autoanalyser method of Harvey & Brothers (1962). Alteration of pH between 6.6 and 7.4 had no effect on the readings of PAH obtained, and the presence of ICG in plasma in concentrations covering the range found in the current experiments was also without effect on the readings of concentration of PAH. Statistical methods Statistical analysis of the data was performed by non-paired Student's t-tests, or by the non-parametric Mann-Whitney U test (Siegel, 1956) for non-normally distributed data or when groups were compared in which variances were unequal as judged by Fisher's F-test. Simple linear regression analysis was also used (Snedecor & Cochran, 1967). Comparison of two regression lines was performed by computer with the methods described in Geigy Scientific Tables, 7th edn, pp. 178-179. Mean values are expressed f 1 SEM. Results Effect of acidosis and nephrectomy on blood pressure and heart rate Both blood pressure and heart rate fell linearly and highly significantly with pH (Fig.1). The regression 122 r( a ) 0 / 0 lines for sham-operated and nephrectomized rats were not significantly different (blood pressure P > 0.05, heart rate P >0.4). Efect of acidosis on cardiac output The electromagnetic flowmeter technique used in these studies is not able to provide absolute values for cardiac output, but can measure changes in cardiac output in the same animal (Browning et al., 1969). In two rats to which sodium chloride solution was given, there was no significant trend in cardiac output over 6 h; the mean value varied between 77.3% and 103.3% of control values. In three conscious rats made acidotic with ammonium chloride, the cardiac output showed a significant fall during the 5 or 6 h of observation. There was a significant positive regression of cardiac output on aortic blood pH as shown in Fig. 2(a) (datafrom all animals). Heart rate in these conscious animals is shown in Fig. 2(b) and also slows significantly with acidosis. There was a fall in heart rate of approximately42Z between pH 7-4 and 6.7, whichis similar to the findings in anaesthetized rats, shown in Fig. I@). Efect of acidosisandnephrectomyon hepatic bloodflow Hepatic extraction of Indocyanine Green (EIcG) was measured in anaesthetized rats, and fell signific- @/ 4oot 00.0. 0 0 0 9 0 1 67 I 6.8 I I I 6.9 7.0 7.1 I 7.2 01 7.3 1 I 7-4 7.5 200 67 6 8 6.9 7.0 7.1 7.2 7.3 7.4 7.5 Preinfusion right alrbl blood pH FIG. 1. (a) Relation of mean carotid artery pressure to blood pH in sham-operated ( 0 )and nephrectomized ( 0 ) rats. pH was measured on right atrial blood. The linear regression lines are: for sham-operated rats, blood pressure = -70.2f24.8 (pH)(Pe0.02, r = 0.61) and, for nephrectomized rats, blood pressure = 133.0f 33.6 (pH) (P i0.01, r = 074). (b) Relation of heart rate to right atrial blood pH in sham-operated (0)and nephrectomized (0)rats. The linear regression lines are: for sham-operated rats, rate = -875+168 (pH) (P<OOOl, r = 0.92) and, for nephrectomized rats, rate = - 1089+198 (pH)(P<OOOI, r = 092). - 181 Haemodynamic effects of acidosis 110- 110- loo - o r / A - 5 _e 90- 5 's -z 00- 5, 7060- 'i 8b // loo- : ;5 90 ::/ 60- 50- u 5040 - I 0 1 1 1 40 1 1 1 1 1 1 1 antly with increasing acidosis (P<OO2, n = 19); the mean values of EICGfrom the regression line are 69.9% at pH 7-4 and 49.9% at pH 6.7. The relationship between KIc0 and right atrial blood pH is shown in Fig. 3(a). It m a y be seen that in both sham-operated and nephrectomized rats, KIc0 was unaffected by a reduction in blood pH until the value for pH falls below 7.1. At greater degrees of acidosis there was a similar fall in Krc0 in both groups. The values of estimated hepatic blood flow inconscioussham-operatedandnephrectomized rats at varying blood pH is shown in Fig. 3(b). There is a significant positive regression of estimated hepatic blood flow on right atrial blood pH in both groups of rats, and the gradients of these regression lines are very similar (sisnificance of difference between gradients P 0.3). Estimated hepatic blood flow in rats at normal pH was approximately 55 ml min-' kg-l body wt. in both groups and this fell to between 10 and 15 ml min-I kg-1 body wt. at pH 6.7. There is no plateau apparent for the values of estimated hepatic blood flow in either group between blood pH 7.1 and 7-5. It should be noted that the calculation of estimated hepatic blood flow assumes a constant packed cell volume of 0.42 whereas in acidotic animals the packed cell volume is approximately0.05 higher than at normal pH (Yudkin & Cohen, 1975). This assumption will produce an underestimate of esti- =- - I 0 I I I I I I I I I mated hepatic blood flow of 9% in severe acidosis for which no correction has been made. Effect of acidosis on renal bloodfTow The total volume of distribution of PAH (VmPAH) in nephrectomized rats did not vary significantly with pH (P>O*6, n = 12); the values derived from the regression line are 332 ml/kg body wt. at pH 7.4 and 355 ml/kg body wt. at pH 6-7. In a separate group of nephrectomized rats, the theoretical volume of distribution of PAH at zero time (VopAH)and the distributional constant for PAH (KIPAH)were determined. The gradients of the regression lines of VoPAHand KZPAH on right atrial blood pH do not differ significantly from zero (P>O.l, n = 20 for VOPAH and P > 0 - 4 , n = 20 for KIPAR). The mean values for VOPAH derived from the regression lines are 164 ml/kg body wt. at pH 7.4 and 138 ml/kg body wt. at pH 6.7. The corresponding values for KIPAH are 0.155 min-' at pH 7.4 and 0.130 min-I at pH 6.7. The renal extraction of PAH (EpAw) was determined in anaesthetized rats and fell significantly with increasing acidosis (P<OOZY n = 23); the mean values of E p A H from the regression line are 83.4% at pH 7.4 and 68.7% at pH 6.7. The derivation of excretory fractional removal rate (KzPAH)for sham-nephrectomized rats was J. Yudkin, R . D. Cohen and Barbara Slack 182 0 0 8, 70- - c W L I I I I I I 6 6 6? 6 8 6.9 7.0 7.1 I I 7-2 7.3 1 1 7.4 7 5 1 1 1 1 1 1 65 6 6 6? 66 6.9 7-0 7.1 1 1 1 1 7.2 7.3 7.4 7.5 Right driol blood pH FIG.3. (a) Relation of fractional removal rate of Indocyanine Green (KIcO)to blood pH in sham-operated (0) and nephrectomized (0)rats. pH was measured on preinfusion right atrial blood. (b) Relation of estimated hepatic blood flow to right atrial blood pH in sham-operated (0) and nephrectomized (0)rats. pH was measured on preinfusion right atrial blood. Estimated hepatic blood flow (EHBF) is expressed per kg body weight. The linear regression lines are: for sham-operated rats, EHBF = - 3 4 4 6 + 53.9 (pH) (P<0.001,r = 0.83) and, for nephrectomized rats, EHBF = 424.1 64.7 (pH) (P<0,001,r = 0.83). - + made according to the method described above. The values for estimated renal blood flow in shamnephrectomized rats at various blood pH values were then calculated from KZPAH and are shown in Fig. 4. There is a significant positive regression of estimated renal blood flow on blood pH. The mean value for estimated renal blood flow falls from 98 ml min-' kg-' body wt. at pH 7.4 to 50 ml min-' kg- body wt. at pH 6-7. I40 1 6-5 6 6 6.7 6% 6.9 7Q 7.1 7.2 7.3 7 4 7.5 Right atrial blood pH Fro. 4. Relation of estimated renal blood flow to preinfusion right atrial blood pH in sham-nephrectomized rats. Estimated renal blood flow (ERBF) is expressed per kg body weight. The linear regression line is ERBF = -407*9+68-3 @H) (Pc0.001,r = 0.63). The calculation of estimated renal blood flow assumes a constant packed cell volume of 0.42, and, as noted fof hepatic blood flow, the assumption produces an underestimate of renal blood flow of 9% in severe acidosis. Discussion In the rat, metabolic acidosis of 4-6 h duration produced marked effects on cardiac output, heart rate, blood pressure, and hepatic and renal blood flow. At a blood pH of 6.7 cardiac output was reduced by 40% compared to that found at pH 7.4. For the same decrease in pH, heart rate fell by 34% in anaesthetized rats and 42% in conscious rats; mean arterial blood pressure was reduced by 17%, hepatic blood flow by 76% and renal blood flow by 49%. The changes in blood pressure, heart rate and hepatic blood flow with acidosis were similar in nephrectomized and in sham-operated rats. The fall in cardiac output in acidosis was mainly due to a fall in heart rate, with no change in stroke volume. However, the lack of compensatory rise of stroke volume suggests a negative inotropic action of acidosis (see also Smith & Corbascio, 1966; Wildenthal, Mierzwiak, Myers & Mitchell, 1968) in addition to the chronotropic effect. Haemodynamic efects of acidosis Measurements of blood pressure were performed on anaesthetizedrats whereasall otherhaemodynamic studies employed conscious resting animals. In rats of normal acid-base status, blood pressure measurements by tail plethysmography under light ether anaesthesia correlate well with measurements by an ankle cuff and photoelectric cell in conscious a n i d s (Floyer, 1951). It should be noted that in the studies in conscious animals, previous operation had been required for sham- or actual nephrectomy, insertion of cannulae and flowmeters; the influence of such procedures on the results is unknown. The necessity for restriction of blood sampling in the measurement of hepatic and renal blood flow precluded the usual technique of obtaining many samples at intervals and applying an exponential curve-fitting analysis. The method employed to measure hepatic blood flow depends on two assumptions: first, that the decay curve of ICG in the rat is of single exponential form, and secondly that hepaticclearanceof ICG is &&ed by anaesthesia. There is no evidence in the literature on these points. The calculation of renal blood flow requires four assumptions to be made. First, the extraction of PAH by the kidney should be unaffected by anaesthesia. Secondly, the double-exponential decay curve of PAH concentration seen in the dog should also apply in the rat, and the two components of this curve should represent distribution and excretion of the substance. Thirdly, the volumes of distribution of PAH and intercompartmental exchange should be unaffected by nephrectomy. Finally, the tubular transport mechanism for PAH should not be saturated at the plasma concentrations attained. We have no formal evidence to subtantiate the first three assumptions, but the fourth is substantiated by the &ding of Friedman, Polley & Friedman (1947) that PAH transport mechanisms in the rat under normal acid-base conditions were only saturated at plasma concentrations in excess of those used in the present study. Most of the previousstudieson the haemodynamic effects of acidosis have been performed in the anaesthetized ventilated dog at constant P a z and thus differ from the present studies in which respiratory compensation for metabolic acidosis occurs. In the anaesthetizeddog, cardiac output and heart rate fell with metabolic acidosisbut the changes of blood pressure were variable (Smith & Corbascio, 1966; Kittle, Aoki & Brown, 1965). Hepatic blood flow in the anaesthetized dog 183 ventilated at constant rate did not alter with reduction of blood pH to a mean value of 7.1 (Goldstein, Simmons & Tashkin, 1972). Measurements were not made at lower pH values. It was noted in the present experiments that KIcoremained constant with reduction in blood pH to 7-1 but no such plateau was apparent in estimated values of hepatic blood flow (Fig. 3a and Fig. 3b). The calculation of estimated hepatic blood flow from KIcorequires multiplication by the volume of distribution of ICG; it is possible that the greater variability in values of estimated hepatic blood flow introduced by this calculation might disguise any apparent plateau in the relation of estimated hepatic blood flow to blood pH. The estimata of hepatic blood flow at normal blood pH were similar to those found by Ossenberg, Denis & Benhamou (1974) in anaesthetized rats (66 ml min-' kg-I body wt.) In the anaesthetized dog, renal blood flow showed little change when blood pH was r e d u d to 7.1-7.2 by infusion of hydrochloric acid (Kittle et al., 1965; Bersentes 8 Simmons, 1967) but with further reduction in blood pH, renal blood flow fell significantly. Estimated values of renal blood flow in rats of normal acid-base status were slightly higher than those derived from the figures of Friedman et al. (1947); estimated renal blood flow may be calculated as 77 ml min-l kg-' body wt. from their data. Studies of lactate handling by the liver and kidney in the dog have demonstrated important differences between the two organs. If hepatic blood flow is reduced by obstruction of hepatic artery and portal vein, lactate uptake by the liver remains constant until the reduction of total hepatic blood flow exceeds 75% (Tashkin, Goldstein & Simmons, 1972). In the dog kidney, however, graded reduction in blood flow produced proportional reduction in renal lactate uptake (Levy, 1962). In a previous study (Yudkin & Cohen, 1975) it has been demonstrated that, in the severely acidotic rat, the absolute renal contribution to lactate removal increased while the extrarenal component of lactate removal was reduced. If it is assumed that hepatic lactate handling in the rat is similar to that in the dog, then part of the reduction in extrarenal lactate removal in severely acidotic animals may have resulted from the reduction of hepatic blood flow to critical levels. However, Lloyd, Iles, Simpson, Strunin, Layton & Cohen (1973) demonstrated reduced uptake of lactate by the isolated rat liver perfused with medium 184 J. Yudkin, R . D. Cohen and Barbara Slack below pH 7.1 in spite of constant perfusate flow rate, demonstrating that acidosis has an effect on lactate uptake other than that mediated via hepatic blood flow. There is no evidence from the present work to suggest that the increase in the apparent renal contribution to lactate removal in the acidotic rat is mediated through any general haemodynamic effect of the kidney. The increase in this renal contribution in severe acidosis occurs in spite of a substantial reduction in renal blood flow, suggesting that lactate extraction by the kidney increases markedly in the acidotic animal. These findings may be relevant to the clinical condition of lactic acidosis. It is possible that in idiopathic or type B lactic acidosis (Lancet Editorial, 1973) the terminal shock which is sometimes observed may result from the effect of acidosis on cardiac output. Furthermore, hepatic blood flow in the acidotic rat was more sensitive to blood pH than was renal blood flow; if the same applies to acidosis in man, the kidney may assume an increasingly important role in removal of lactate from the circulation in acidosis. In lactic acidosis associated with circulatory failure, the ability of the kidney to remove lactate from the circulation may be impaired by a reduction in renal blood flow. Sriussadaporn & Cohn (1968) found net renal lactate output in four of eight patients with lactic acidosis due to shock, and net hepatic lactate output was Seen in two of four patients with this condition. AcLnowledgments We are grateful to Mr C. Browning for valuable assistance during the studies of cardiac output. We are also most grateful to Mr Michael Rablen, who derived the equations for the calculation of renal blood flow, and to Mr Ashley Hyams, who constructed computer programs to solve the equations for renal blood flow by the method described. References BERSENTES, T.J. & SIMMONS, D.H. (1967) Effects of acute acidosis on renal haemodynamics. American Journal of Physiology, 212,633-640. BROWNING, C.. LEDINOHAM, J.M. & PELLING, D. (1970) Simultaneous measurement of cardiac output and mean arterial pressure changes in unanaesthetized rats. Journal of Physiology (London), u)8,ll~-12~. BROWNING, C., PELLING, D. & LEDINGHAM, J.M. (1969)An electromagnetic flowmeter for studying changes of cardiac output in unanaesthetized rats. Medical and Biological Engineering, 7, 549-558. CAESAR, J., SHALDON, S., CHIANDUSSI, L., GUEVARA, L. & SHERLOCK, S . (1961)The use of indocyanine green in the measurement of hepatic blood flow and as a test of hepatic function. Clinical Science, 21,43-57. FLOYER, M.A. (1951)The effect of nephrectomy and adrenalectomy upon the blood pressure in hypertensive and normotensive rats. Clinical Science, 10,405-421. C.L. (1947) FRIEDMAN, S.M., POLLEY,J.R. & FRIEDMAN, The clearance of inulin and sodium p-aminohippurate in the rat. American Journal of Physiology, 150, 340-352. GOLDSTEIN,P.J.. SIMMONS,D.H. & TASHKIN, D.P. (1972) Effect of acid-base alterations on hepatic lactate utilization. Journal of Physiology (London), 223,261-278. HARVEY, R.B. & BROTHERS,A.J. (1962) Renal extraction of para-aminohippurate and creatinine measured by continuous in uivo sampling of arterial and renal-vein blood. Annals of the New York Academy of Sciences, 102,4654. KITTLE,C.F., AOKI,H. & BROWN,E.B., JR (1965)The role of pH and COz in the distribution of blood flow. Surgery, 57, 139-154. LANCET EDITORIAL(1973)Lactic acidosis. Lancez, ii, 27-28. LEVY,M.N. (1962) Uptake of lactate and pyruvate by intact kidney of the dog. American Journal of Physiology, 202, 302-308. LLQYD,M.H., ILES,R.A., SIMPSON.B.R.. STRUNIN,J.M., LAYCON, J.M. & COHEN, R.D. (1973)The effect of simulated metabolic acidosis on intracellular pH and lactate metabolism in the isolated perfused rat liver. Clinical Science and Molecular Medicine, 45, 543-549. MANDEL,M.J., VIDT, D.G. & SAPIRSTEIN, L.A. (1955) Disappearance of para-aminohippuric acid from the plasma of the dog after single intravenous injection. American Journal of Physiology, 182,428-432. OSSENBERO, F.-W.,DENIS,P. & BENHAMOU, J.-P. (1974) Hepatic blood flow in the rat: effect of portacaval shunt. Journal of Applied Physiology, 31, 806808. SAPIRSTEIN, L.A., SAPIRSTEIN,E.H. & BREDEMEYER, A. (1960)Effect of hemorrhage on the cardiac output and its distribution in the rat. Circulation Research, 8, 135-148. SIEGEL,S. (1956) Non-Parametric Statistics for the Behavioral Sciences. McGraw-Hill, New York. SMITH, N.T. & CORBASCIO, A.N. (1966)Myocardial resistance to metabolic acidosis. Archives of Surgery, 92,892-897. SNEDECOR,G.W. & COCHRAN, W.G. (1967) Statistical Methods, 6th edn. Iowa State University Press, Am-, Iowa. SRIUSSADAPORN, S. & Corn, J.N. (1968) Regional lactate metabolism in clinical and experimental shock. Circulation, 38, Suppl. 6, 187. TASHKIN,D.P., GOLDSTEIN, P.J. & SIMMONS, D.H. (1972) Hepatic lactate uptake during decreased liver perfusion and hypoxemia. American Journal of Physiology, W , 968-974. WILDENTHAL, K., MIERZWIAK, D.S., MYERS, R.W. & MITCHELL, J.H. (1968) Effects of acute lactic acidosis on left ventricular performance. AmericanJournal of Physiology. 214,1352-1359. YUDKIN.J. (19751 M.D. thesis. University of Cambridge. YUDKIN;J. & COHEN,R.D. (1975) The contribution of the kidney to the removal of a lactic acid load under normal and acidotic conditions in the conscious rat. Clinical Science and Molecrtlar Medicine, 48, 121-131.