It is known that hypopotassemia may appear during therapy of diabetic acidosis or coma, at times associated with muscular paralysis (1-9). This is not, however, an invariable occurrence, nor have the origins of such decreases in the serum and interstitial fluid potassium concentrations been clearly defined. It seems probable that one or more mechanisms are responsible. These include the possibilities that hypokaliemia 1 develops as a result of a continued loss of potassium in urine, that it is a manifestation of dilution of body fluids by potassium-free solutions administered during therapy, and, finally, that it results from movements of potassium into body cells. The studies herein reported evaluate the role of each of these factors in a limited number of patients, chiefly pre-adult, in whom intensive study using the balance method was possible.
MATERIALS AND METHODSNine patients have been studied during a total of 11 admissions for the treatment of diabetic acidosis. Two of the subjects were young adults; the remainder were children seven to 16 years of age. Over-breathing was definitely present in each case; the serum carbon dioxide content ranged from 2.6 to 10.4 m.eq. per liter on admission. All patients were conscious, though difficult to arouse. Definite vascular collapse was present in only one patient, R. J., requiring three transfusions. Anuria did not develop in any of the subjects. Insulin was administered subcutaneously to all patients, although J. K. and T. W. also received large intravenous injections initially. Data on the retention of potassium salts administered in large amounts during convalescence to eight of the cases have been presented in a previous communication (9). The studies herein reported are based on observations from the time of admission and continued until convalescence was well established. During these inter-1 This term has been taken to refer solely to diminished concentrations of potassium in blood serum or plasma and the contiguous interstitial fluid, and does not necessarily indicate any decrease in the total amount of potassium in these fluids. vals the patients received insulin, glucose, saline, and only little, if any, potassium.The experimental procedure employed and the methods of calculation have been previously described in detail (9-12). Briefly they may be summarized as follows: the chloride space, corrected for the external balance of this ion and for changes in the chloride concentration in serum, has been used as an index of alterations in the volume of extracellular fluid. The intake and the urinary excretion of sodium, potassium, chloride, glucose, and nitrogen were measured during intervals marked at the start and finish by determinations of the values of these various components in serum. The retention or loss of these cations and of nitrogen has been apportioned between the extracellular fluid and the cells. In the case of nitrogen, suitable corrections for changes in the concentration of the whole blood nonprotein nitrogen were made. The values for the c...
