Lactic Acidosis:

An Experimental Model

Allen I. Arieff and Alice Kerian A model of spontaneous lactic acidosis was developed in alloxan diabetic mbbits by infusing intmvenously (I-hydroxybutyric acid followed by a continuous infusion of NaHCO,. In half of the animals, the arterial lactate/pyruvate mtio rose from 2.5 mM/ 0.19 mM to 20.4 mM/ 0.28 mM, and arterial pH fell to 7.16. In animals with lactic acidosis, the calculated mtio in blood of NAD/NADH was 1437 f 230, versus a normal value of 6754 f 1250.

Both arterial po, and blood pressure were normal. Continued infusion of NaHCOs led to increased blood lactate levels, with cardiorespimtory arrest in 36% of animals. lactic acidosis did not develop in normal mbbits who were similarly treated. It is concluded that spontaneous lactic acidosis can be produced in diabetic, but not in normal, mbbits by infusion of @-hydroxybutric acid followed by infusion of NaHCOs.

S

PONTANEOUS LACTIC ACIDOSIS is a clinical entity characterized by metabolic acidosis with marked tachypnea. It usually occurs concomittant with serious medical illness but there is no obvious medical reason for metabolic acidosis.’ The acidosis is primarily caused by an elevation in blood lactate (to at least 7 mM), without other significant cause for acidosis, such as uremia, ketoacidosis, or exogenous intoxication. Blood pyruvate in this disorder is usually normal or only modestly elevated, so that blood lactate/pyruvate ratio exceeds 10/l. Generally, spontaneous lactic acidosis occurs in the presence of normal arterial p02, with both blood pressure and cardiac output normal. It appears to occur far more frequently in diabetic than nondiabetic individuals’-3 and has been reported as complicating the treatment of ketoacidosis.*s4 The etiology of spontaneous lactic acidosis is not known, and mortality probably exceeds 90%.lm3The lack of an effective therapeutic regimen points to the need for an animal model of lactic acidosis, in order to better elucidate the pathophysiology of this disorder. Recently, there have been several reports of diabetic patients with metabolic acidosis associated with elevations in blood levels of both lactate and P-hydroxybutyrate, with normal blood acetoacetate. 5m7Such a biochemical constellation suggests depletion of skeletal muscle and liver mitochondrial NAD+, so that P-hydroxybutyrate cannot be converted to acetoacetate. Depletion of NAD+ could also limit conversion of lactate to pyruvate in liver cytoplasm and might predispose to lactate accumulation and lactic acidosis: acetoacetic acid + NADH + H+ \

,&hydroxybutyric

acid + NAD+ (1)

From the Kidney Research Laboratories, Veterans Administration Hospital and University of California School of Medicine. San Francisco, Calif Receivedforpublication July 3. 1975. Supported by Grant A M-18350 from NIA MDD. National Institutes of Health. Presented in part at 32nd Annual Meeting, American Federation for Clinical Research, Atlantic City, New Jersey, May 4. 1975. Reprint requests should be addressed to Allen I. Arieff, M.D.. Nephrology Service (1115). VA Hospital, 4150 Clement, San Francisco. Callf 94121. @ 1976 by Grune & Stratton, Inc.

Metabolism, Vol. 25, No. 3 (March), 1976

307

308

ARIEFF

pyruvic acid + NADH \

lactic acid + NAD+

AND

KERIAN

(2)

Such a mechanism might be the cause of lactic acidosis in some patients. It was therefore attempted to induce spontaneous lactic acidosis in animals by depletion of NAD + . MATERIALS

AND METHODS

Studieswere carried out in six groups of New Zealand white rabbits, 1.8-3.0 kg, as follows: (1) normal rabbits who were anesthetized for 4 hr; (2) normal rabbits infused for 4 hr with @-hydroxybutyric acid; (3) normal rabbits treated as in group 2, but then infused for 3 hr with NaHCOj at 0.1 mEq/min; (4) diabetic rabbits who were anesthetized for 4 hr; (5) diabetic rabbits infused for 4 hr with p-hydroxybutyric acid; (6) diabetic rabbits infused for 4 hr with p-hydroxybutyric acid, then 30-175 min with NaHC03. Half of the animals in group 6 developed lactic acidosis (group 6A), while in the other half, plasma HCOT was restored to normal (group 6B). Alloxan diabetes was produced in groups 3-6 as previously described.s All diabetic animals studied had a plasma glucose of at least 20 mM, and these animals were then maintained on ad lib food (Purina Chow) and water without insulin for lo-20 days. All animals were anesthetized with succinyl choline and sodium pentobarbitol, tracheostomized, and mechanically ventilated as previously described.8 Arterial pco, was maintained at 30-40 mm Hg. In groups 2, 3, 5, 6A, and 6B, I20 mM &hydroxybutyric acid in 115 mM NaCl was infused at an initial rate of 0.12 mmole/kg/min. Plasma bicarbonate was monitored every 15-30 min, and it fell to 8-12 mEq/L after (+SD) 122 f 36 min of infusion. Frequent small adjustments were made in the rate of infusion in order to maintain plasma HCO< at about 10 mEq/L. After 4 hr, the &hydroxybutyric acid infusion was stopped in groups 3 and 6 and infusion of 200 mM NaHC03 containing 20 mEq/L of potassium acetate was begun. The infusion rate was 0. I mEq/ min and it was continued until plasma HCOF was either normal or bad not risen after 3 hr. Measurements were made in arterial blood of pH, po,. pco,, and bicarbonate, and of the concentration of &hydroxybutyrate, lactate, and pyruvate. Lactate, pyruvate, and &hydroxybutyrate were measured in perchloric acid extracts as previously described.7+9 Methods for measurement of arterial pH, pcoz, poz, and bicarbonate have previously been published,” and the ratio in arterial blood of NAD +/NADH was calculated.“~12 RESULTS

Normal values for arterial pH, pcoz, bicarbonate, poz, lactate, and pyruvate are shown in Table 1. The ratio of NAD +/NADH was 6754 f 1250, and the lactate/pyruvate ratio was 2.5/0.19. In diabetic rabbits (group 4), the plasma glucose (*SD) was 23.9 f 2.3 mM and /3-hydroxybutyrate was < 2 mM. There was a modest rise in blood lactate and fall in pyruvate so that the NAD ‘/ NADH ratio decreased to 3318 f 564, a value significantly less (p < .Ol) than normal. Diabetic animals also had an elevated plasma bicarbonate, with a resultant metabolic alkalosis. In diabetic rabbits infused for 4 hr with /3-hydroxybutyric acid (group 5), arterial pH and bicarbonate (*SE) were 7.10 f .02 and 11.3 f 0.1 mEq/L, values significantly less than normal (p < O.Ol), and plasma Phydroxybutyrate was 18.7 f 2.2 mM. However, there were no important changes in blood lactate or pyruvate. In another group of diabetic animals (group 6), after 4 hr of /3-hydroxybutyric acid infusion, NaHCOj was begun. In six of 14 animals (group 6A), plasma HCO, rose from 11.3 to 21.7 mEq/L within 175 min, with the rise in plasma HCO, almost exactly paralleling the rate of infusion (Fig. 1). In eight of 14 animals (group 6B), however, plasma HCO? did not change significantly

309

LACTIC ACIDOSIS

Table 1. Effectsof Diobotm, Infusion of @-Hydroxybutyric Acid, or &Hydroxybutyric Acid + NoHCO, on Arieriol Blood Add-Base Status PC02

f-2

HCO,

(1) Contol,n 7.36

= 8

62

35 *1

*6

f 0.03

(4) Diabetic,n 7.53

19.8

2.5

0.19

f 1.o

zt0.6

zto.05

= 7

60

36 Al

z!z5

*O.Ol

pyruvate

mmol/L

mm Hg

PH

lactate

29.6

3.6

0.16

l 1.2

*to.5

kO.02

(5) Diabetes + B~HBA, n = 5 7.10 zto.02

60

38

11.3

3.9

ZkB

Ztl

Tko.1

zto.7

(6A) Diabetes + &OHBA 7.42

67

f 0.02

*0

+ NaHC03,

7.16 ztO.06

58

21.7

Ztl

l 0.9

31 *5

fll

Data are presented cts mean f

no lactic acidosis, n = 6

34

(68) Diabetes + @-OHBA + NaHCO;

0.16 *0.03

1.5 +0.5

0.10 zko.03

, lactic acidosis, n = 8 10.8 h1.7

20.4

0.20

zt3.7

ZtO.06

SE.

n = number of animals. j3-OHBA = &hydroxybutyric NaHC03

acid infusion.

= sodium bicorbonoh infusion.

during HCO; infusion. At the end of 175 min, plasma HCO; was not significantly different from preinfusion values (Fig. 1). In these eight animals, blood lactate increased fivefold from 3.9 f 0.7 to 20.4 f 3.7 mM (p < 0.01) while pyruvate rose less than twofold, from 0.16 f 0.02 to 0.28 + 0.06 mM (P < 0.05). The latter group of animals, then, had lactic acidosis. Arterial poz in these animals (58 f 11 mm Hg) was not significantly different from normal values (62 f 6 mm Hg). Plasma @-hydroxybutyrate level fell by a similar amount in both groups of animals (to 8.8 f 4.8 and 4.2 f 2.7 mM, respectively). In animals with lactic acidosis, blood NAD+/NADH ratio was 1437 f 230, significantly less (p < 0.01) than that observed in diabetic animals (3381 f 564) or diabetic animals infused with /3-hydroxybutyric acid (3528 f 878). In normal animals (group 2) who were infused with @-hydroxybutyric acid for 4 hr, blood lactate (1.6 f 0.3 mM) and pyruvate (0.11 f 0.02 mA4) were not different from normal values. When a similar group of animals (group 3) were infused with NaHC03 as was done in the diabetic animals, plasma HCO, returned to normal values within 24 hr in seven of seven animals, with no animal developing lactic acidosis. Spontaneous lactic acidosis developed in a total of 12 diabetic animals. Of the 12 animals with lactic acidosis, four abruptly developed cardiovascular collapse within 1 hr of the time lactic acidosis was noted. This occurred despite the fact that all animals were being mechanically ventilated via a tracheostomy. The four animals who died all had arterial blood samples taken within 15 min of

310

ARIEFF

AND

KERIAN

HCOC INFUSION >

LA

1

IO

50

100

125

150

I75

MINUTES Pig. 1. The effects of infusion of NaHCOs, 0.1 mmol/min, on 14 diabetic mbbits who had received an infusion of &hydroxybutyric acid (0.12 mmol/kg/min) for 4 hr. The mean plasma bicarbonate in these animals was 9.6 + 0.7 mM, and plasma @-hydroxybutymte was 18.7 * 2.2 mM. For the initial hour of infusion, plasma bicarbonate rose in all animals, essentially pamlleling the rate of infusion. In 6 animals (ketoacidosis (KA)], plasma bicarbonate continued to rise during the period of infusion, reaching normal values (21.7 & 0.9 mM) after about 3 hr. In the other 3 animals [lactic acidosis (LA)], plasma bicarbonate did not rise after 1 hr and started to decline after 2 hr, despite continued infusion of NaHCOs. After 3 hr, plasma bicarbonate in the latter group (LA) was not different from the preinfusion value, but blood lactate was 20.4 * 0.5 mM.

3.7 mM, while in the former group (KA), blood lactate was I.5 f

their demise. None of these animals are included in Table 1, but blood lactates were 14.9 f 3.5 mM and arterial pH 7.20 + 0.03. Thus, uncontrolled diabetes, per se, was associated with an increased ratio in blood of lactate/pyruvate, suggesting an altered redox potential which might predispose to lactic acidosis. When NAD+ was further depleted by infusion of #Lhydroxybutyric acid (see equation l), and then NaHCO, was infused (group 6), spontaneous lactic acidosis developed in over half the animals so treated. When normal animals were treated in a similar manner, none developed lactic acidosis. In all animals, arterial blood pressure was frequently monitored, and no important changes were noted in any of the groups evaluated. DISCUSSION

The data presented show that in the diabetic rabbit, spontaneous lactic acidosis can be induced by infusion of &hydroxybutyric acid followed by in-

LATIC ACIDOSIS

311

fusion of NaHCO,. In the model described, diabetes per se appears to induce an altered redox state, as suggested by an increase in the ratio of lactate/ pyruvate in blood. Similar observations have been reported in liver of rats with uncontrolled diabetes.“*13 In the animal model described, lactic acidosis was present while at the same time arterial po, and blood pressure were normal. There are several possible reasons for lactic acidosis to occur in the setting described. In patients with diabetic ketoacidosis, red blood cell levels of 2,3diphosphoglycerate are markedly reduced’4J5 and such a situation may be present in ketoacidotic animals as well. A deficit in this glycolytic intermediate compound would tend to impaired release of oxygen at the tissue level. Severe acidosis, by shifting the oxygen-hemoglobin disassociation curve to the right, tends to compensate for the effects of low red blood cell 2,3-diphosphoglycerate levels.15 However, intravenous bicarbonate, by raising arterial pH, increases the affinity of hemoglobin for oxygen, impairing delivery of oxygen to tissues. Such impaired peripheral oxygenation could lead to increased production of lactate and in fact, there are several reported instances where treatment of diabetic ketoacidosis with bicarbonate has resulted in increased levels of blood lactate.2*4*7In addition, intravenous bicarbonate would increase blood pH, and alkalosis per se increases production of lactate by the liver.3 However, despite similar biochemic abnormalities and therapy, only a minority of such patients developed lactic acidosis, and only about half the experimental animals so treated did so. Ordinarily, one would expect that the liver could adequately remove most of the excess lactate produced. l6 In lactic acidosis there is impairment in the ability of the liver to metabolize amino acids,” however, and it may be that hepatic removal of lactate is impaired by hypoperfusion or acidosis.‘8-20 Whatever the mechanism which occasionally leads to hyperlactemia in patients with ketoacidosis, there is probably interaction of several factors. In some series of ketoacidotic patients, administration of NaHC03 did not affect blood lactate levels,2’ while in others, up to 70% of patients receiving HCO; had an increase in plasma lactate. 4,7It may be that the hyperlactemia is related to circulatory insufficiency and hypotension, which often accompany diabetic ketoacidosis, or there may be different degrees of phosphate depletion among individual patients, with varying effects on tissue oxygen delivery. The state of tissue oxygenation determines whether or not excess lactate will be produced. Tissue oxygenation is determined by oxidative metabolism in mitochondria, which is dependent upon the intramitochondrial concentrations of NADH and NAD. The concept of “excess lactate” assumes that the ratio of NAD+/NADH in cells is approximated by the ratio of lactate/pyruvate in blood.22 However, there are at least three pools of NAD+/NADH in cells, one in cytoplasm (lactate/pyruvate), and two in mitochondria (glutamate/oxoglutarate and &hydroxybutyrate/acetoacetate).” The blood lactate/pyruvate ratio will not necessarily reflect the mitochondrial redox potential, but should reflect that in cytoplasm. “v23Even with these limitations in mind, the blood ratio of lactate/pyruvate is a useful indicator of tissue hypoxia, although not without sources of error. In the present study, the presence of increased arterial lactate/ pyruvate ratio with normal po, is certainly strongly suggestive of tissue hypoxia.3v22

312

ARIEFF

AND

KERIAN

There are other experimental models of lactic acidosis. Techniques have involved instillation of oil into pericardial sacz4 hemorrhage,22 hyperventilation,25 or certain pharmacologic agents.’ In all of the aforementioned situations, however, there is usually a concommitant increase in blood pyruvate, so that the lactate/pyruvate ratio is elevated. Therefore, these models are not similar to primary, or spontaneous, lactic acidosis. REFERENCES 1. Tranquada RE, Grant WJ, Peterson CR: Lactic acidosis. Arch Intern Med 117:192-202, 1966 2. Waters WC HI, Hall JD, Schwartz WB: Spontaneous lactic acidosis. Am J Med 35: 781-793, 1963 3. Ohva PB: Lactic acidosis. Am J Med 48: 209-225, 1970 4. Barta L: Lactic acidosis in diabetic coma. Lancet 2:779-780,197O 5. Pettersen JE, Landaas S, Eldjarn L: The occurrence of 2-hydroxybutyric acid in urine from patients with lactic acidosis. Clin Chem Acta 48:213-219, 1973 6. Marliss EB, Ohman JL, Aoki TT, Kozak GP: Altered redox state obscuring ketoacidosis in diabetic patients with lactic acidosis. N Engl J Med 283:978-980, 1970 7. Watkins PJ, Smith JS, Fitzgerald MG, Malins JM: Lactic acidosis in diabetes. Br J Med 1:744-747, 1969 8. Arieff AI, Kleeman CR: Cerebral edema in diabetic comas. II. Effects of hyperosmolality, hyperglycemia, and insulin in diabetic rabbits. J Clin Endocrinol Metab 38: 1057-1067, 1974 9. Sigma Chemical Company, Technical Bulletin #826-UV, 1974 IO. Arieff AI, Doerner T, Zelig H, Massry SG: Mechanisms of seizures and coma in hypoglycemia: Evidence for a direct effect of insulin on electrolyte transport in brain. J Clin Invest 541654-663, 1974 11. Williamson DH, Lund P, Krebs HA: The redox state of free nicotinamide-adenine dinucleotide in the cytoplasma and mitochondria of rat liver. Biochem J 103:514-527, 1967 12. Kaplan NO, Ciotti MM, Stolzenbach FE: Reaction of pyridine nucleotide analogues with dehydrogenases. J Biol Chem 221:833-844, 1956 13. Hohorst HJ, Kreutz FH, Reim M, Hubener HJ: The oxidation/reduction state of the extra-mitochondrial DPN/DPNH system in rat liver and the hormonal control of substrate

levels in vivo. Biochem Biophys Res Commun 4163-168, 1961 14. Guest GM: Organic phosphates of the blood and mineral metabolism in diabetic acidosis. Am J Dis Child 64:401-412, 1942 15. Bellingham AJ, Detter JC, Lenfant C: The role of hemoglobin affinity for oxygen and red-cell 2, 3-diphosphoglycerate in the management of diabetic ketoacidosis. Trans Assoc Am Phys 83:113-120, 1970 16. Berry MN: The liver and lactic acidosis. Proc Roy Sot Med 60: 1260-1262, 1967 17. Marliss EB, Aoki TT, Felig P: Altered amino acid metabolism in lactic acidosis. Diabetes 19(Supp 1):355, 1970 18. Hems RA, Ross BD, Berry MN, Krebs HA: Gluconeogenesis in the perfused rat liver. Biochem J 101:284-292, 1966 19. Cohen RD. Iles RA, Barnett D, Howell MEO, Strunin JM: The effect of changes in lactate uptake on the intracellular pH of the perfused rat liver. Clin Sci41:159-170, 1971 20. Lloyd MN, Iles RA, Simpson BR, Strunin JM, Layton JM, Cohen RD: The effect of simulated metabolic acidosis on intracellular pH and lactate metabolism in the isolated perfused rat liver. Clin Sci Mol Med 45:543-549, 1973 21. Assal JP, Aoki TT, Manzano FM, Kozak GP: Metabolic effects of sodium bicarbonate in management of diabetic ketoacidosis. Diabetes 23:4054 11, 1974 22. Huckabee WE: Relationships of pyruvate and lactate during anaerobic metabolism. I. Effects of infusion of pyruvate or glucose and of hyperventilation. J Clin Invest 37:244-254, 1958 23. Olson RE: “Excess lactate” and anaerobiosis. Ann Intern Med 59:960-963, 1963 24. Alpert NR: Lactate production and removal and the regulation of metabolism. Ann NY Acad Sci 119:995-1012. 1965 25. Eichenholz A, Mulhausen RO, Anderson WE, MacDonald FM: Primary hypocapnia: A cause of metabolic acidosis. J Appl Physiol 17:283-288, 1962