DIABETIC COMA 561 



depends upon an excess of these substances in the blood — i. e., is 

 an acid intoxication — for the followinj^ reasons: (1) The coma usually 

 appears when the amount of organic acids in the urine is highest, and 

 is absent when there is little or none of them in the urine. (2) Be- 

 cause of the resemblance of the symptoms to those of experimental acid 

 intoxication. (3) Because of the repeated demonstration of a re- 

 duced amount of alkali in the blood, as determined by titration, and 

 a great reduction of the amount of CO2 carried in the venous blood. 

 The capacity of the serum to absorb CO2 in vitro is also greatly re- 

 duced, from a normal 55 to 75 per cent, to as low as 20 per cent. (Van 

 Slyke). By means of gas-chain measurements Roily ^' found that 

 by far the lowest OH values ever observed in the blood are in dia- 

 betic coma;^- and Sellards^^ showed that in diabetes the tolerance 

 for alkalies may be increased. He states^'* that a deficit of 20 to 30 

 grams of sodium bicarbonate produces a degree of acidosis demonstra- 

 ble by blood examination onlj'-, 40 to 50 grams deficit usually ciiuses 

 only dyspnoea on exertion, 75 to 100 grams deficit may produce 

 persistent dyspnoea, 150 grams deficit may accompany coma, while 

 the maximum deficits reported have been about 200 grams. (4) 

 The marked improvement that sometimes results from the administra- 

 tion of alkalies (usually sodium bicarbonate). Associated with this 

 improvement is an elimination of greatly increased amounts of organic 

 acids, indicating their previous retention in the body because of lack 

 of alkali with which they could combine. 



But there are cases of diabetic coma without typical air hunger, 

 and it is the exception rather than the rule for alkali therapy to 

 produce a marked improvement in the fully developed coma of 

 diabetes. Furthermore, coma may occur in diabetics who are pro- 

 ducing no such quantity of organic acids as would seem theoreti- 

 cally to be necessary to cause enough acid intoxication to result in 

 acidosis, and coma develops in diabetics who are being supphed with 

 sufficient bases for all requirements. Hence it must be concluded 

 that only a part of the symptomatology of diabetic coma depends on 

 acids as such, but as yet we do not know what other agents are acting.^'* 



/3-oxybutyric and diacetic acid, according to many authorities, 

 seem to have no specific poisonous effects, but act simply as acids 

 in the blood. Acetone does not have this eff'ect, not being an acid, 

 and seems not to be toxic to any considerable degree; doses of 4 grams 

 per kilo cause effects similar to ethyl alcohol in dogs, 8 grams per 

 kilo being fatal, which corresponds to a dose of 500 grams for an adult 



'1 Munch, med. Woch., 1912 (59), 1201. 



^^ Menten, however, reports that in diabetes before acidosis the OH concen- 

 tration is usually somewhat above normal (Jour. Cancer Res., 1917 (2), 179). 



^3 Johns Hopkins Hosp. Bull., 1912 (23), 289. 



^■* Pribram and Loewy (Zeit. klin. Med., 1913 (77), 384) suggest that abnor- 

 mal products of protein cleavage are responsible, and Rosenbloom (N. Y. Med. 

 Jour., Aug. 7, 1915) reports cases of typical diabetic coma without acetone bodies 

 in the urine. 

 36 



