Since the adoption of exchange transfusion as a method in the treatment of newborn infants with severe erythroblastosis (1-3) the question of the possible toxic effects of sodium citrate added to the administered blood, to prevent coagulation, has arisen (4). In this procedure from 85 to 95% of the infant's Rh-positive erythrocytes are replaced by the donor's Rh-negative erythrocytes depending upon the volume of blood that is administered and simultaneously removed (5). In order to effectuate an exchange of 85%o, a quantity of blood equivalent to twice the infant's blood volume must be used, and for an exchange of 98%o, a quantity of blood equivalent to four times the blood volume must be used. Since the average newborn infant has a blood volume in the range of 250-300 ml. the procedure involves the use of approximately 500 ml. and approximately 1,000 ml., respectively. With the use of sodium citrate as an anticoagulant the infant receives as much as 60 to 120 ml. of a 3%o solution of sodium citrate intravenously over a period of about 1½/2 hours.In the experimental animal, the intravenous administration of large quantities of sodium citrate has been shown to produce convulsions (6). This has been ascribed to the formation of a poorly ionized calcium citrate complex which removes calcium ion from the blood stream (7-9). Despite this consideration, exchange transfusions have been successfully carried out in a large number of cases, although occasionally deaths have been reported in the very severely affected cases (10).In order to study the adjustment that the newborn makes to the physiological strains to which it is subjected when an exchange transfusion is 1 The third in a series on the mechanism controlling citric acid levels in the blood. The first two papers in this series appeared in the Journal of Clinical Investigation, Vol. 27, pages 446 and 450, 1948. 2 This paper was presented at the Washington meeting of the American Chemical Society, August 30, 1948. carried out, citric acid, calcium, phosphorus, protein and total base levels were studied before, during and after such a transfusion on a newborn with no symptoms of erythroblastosis and with apparent normal hepatic and kidney function. The newborn studied was a mongolian idiot born in normal delivery. In addition citric acid determinations were carried out on serial bloods withdrawn from eight erythroblastotic infants during exchange transfusion. 500 ml. of blood were freshly drawn from a compatible donor. To this were added 60 ml. of 3%o sodium citrate dehydratee of tri-sodium citrate). This made a total volume of 560 ml. Over a period of approximately 60 minutes 500 ml. of. this citrated blood were infused into the saphenous vein at the ankle. After 60 ml. had been infused the radial artery was cut and the blood was collected in 60 ml. portions, during the progress of the transfusion. The first 60 ml. were added without removal of blood in order to dilate the blood vessels and make the radial artery more apparent. The cells were separated by cen...

It has been demonstrated in animal and plant tissue studies that citric acid metabolism is intimately related to carbohydrate metabolism (1-5).

tiple doses of the drug and that injury of this degree was never seen in many other dogs, including 2 littermates of dog 132, which survived a single dose. Serial biopsies showed, i n agreement with earlier reports and with our m observations on other dogs, that the primary effect of alloxan was on the tubules. The glomerular changes lagged behind the tubular degeneration and they appeared both in time and in type to be consequent to the latter. The histologic picture suggests that the glomerular damage originated from a failure of drainage, with resulting dilatation of Bowman's space and atrophy of the tuft. The absence of epithelial and enddthelial proliferation and the presence of periglomerular fibrosis are consistent with this interpretation. Summary.In 2 dogs which did not become diabetic following repeated doses of alloxan, serial biopsies of the kidney showed the late development of glomerular lesions. These lagged behind the usual tubular degeneration and appeared both in time and in .type to be consequent to'the latter. Similar studies on dogs receiving a single dose of alloxan and surviving for comparable periods of time showed only minimal glomerular changes.

In the first paper of this series (1) it was demonstrated in vivo in humans that the normal response to active carbohydrate metabolism, as in the postabsorptive period for glucose, or after injection of insulin, is a lowering of blood citric acid levels.In the studies presented below, the nature of the citric acid response curve in abnormal conditions is explored with the hope of throwing more light on the nature of the mechanism linking citric acid metabolism with glucose removal from the blood stream.When abnormalities occurred they were apparent during the first three hours of this test. For this reason and for the convenience of the patient, we discarded the six-hour test and substituted a three-hour test, seeking the contour of the first half of the citric acid curve.

No abstract