Abstract
It has been believed for a long time that diabetes predisposes human beings to infections, including tuberculosis, and that when a diabetic individual develops tuberculosis the disease is extremely severe, and rapidly fatal. The incidence of tuberculosis is almost 3 times as great in diabetics as in the general population (Banyai 1 ). Fitz 2 found that 63% of his cases died in less than one year (19 of 31 cases). While most writers agree on these findings (Gotten, 3 Adams 4 ), they have been disputed by many others. Joslin, 5 and Rosenberg and Wolf 6 feel that the prognosis is not bad in cases of tuberculosis complicated by diabetes. It is, therefore, obvious that much confusion prevails concerning the increased susceptibility of the diabetic to tuberculosis. It was hoped that animal experimentation might help to clarify this problem.
Our experiments include first a study of tuberculosis in the diabetic dog. A series of dogs was pancreatectomized. Eight survived, 7 of which developed severe diabetes while one had only a slight increase in blood sugar. As noted on the accompanying chart the animals survived from 16 to 95 days after infection. The average blood sugar is also included in the chart. Small doses of insulin were given daily to offset coma, and at the same time the animals remained severely diabetic. To replace some of the enzymes of which pancreatectomized animals are deprived, we fed them daily with desiccated pancreatic substance incorporated in their food. Twenty-four hours after operation, the animals were injected intraperitoneally with a human strain of tubercle bacillus (H37) in doses of 1 mg. per pound of weight.
As recorded on the chart, 7 of the 8 animals were macroscopically positive for tuberculosis (87.5%). In 6 of the 7 that have so far been checked histologically the diagnosis was confirmed, while one animal was negative both macroscopically and microscopically.
Get full access to this article
View all access options for this article.