Duration.—There is so little resem blance between the various cases that an average duration, even supposing that it could. be rigorously established, would be of no importance. It suffices to say that in a general way the average dura tion of diabetes is several years.
I am consequently much surprised at the results given by Griesinger concern ing 100 cases. In 13 the disease only lasted from 6 months to 1 year; in 39, 1 to 2 years; and in 20, 2 to 3 years, which would make the duration of the disease in three-fourths of the cases from 6 months to 3 years.
In order to explain such remarkable figures it must be supposed that the dia betes was latent, in the beginning, in a large number of the patients, and that these statistics include a great many seri ous cases.
The duration in children varies greatly. Out of 34 cases the shortest duration was two days; the longest had not termi nated at the end of five years. In 7 cases it did not last one month, and of these I was cured. Seventeen lasted less than a year, and of these 7 were cured. Ten lasted over a year, and not one of these recovered, and it may be said that re coveiy scarcely occurs where the dura tion is more than one year. C. Stern (Archiv f. Kinderh., 13. 11, H. 2, '59).
The main prognostic features are: Age, power of assimilation of carbohy drates, early recognition of the affec tion, the presence of intercurrent and complicating diseases, condition in life, state of the urine, and the power of ab sorption of other foodstuffs than carbo hydrates. H. S. Starr (Med. Record, Apr. 6, 1901).
Termination.—It is evident that dia betes, which is but seldom cured, ally ends in death. In explanation of this rather naive statement, which might lead to a false interpretation, it must be borne in mind that the duration of the disease is a long one, and that in a great number of cases mild diabetes allows the patient to live to an advanced age.
In referring to the complications of the disease, I have already mentioned the frequency of phthisis, and the even greater prevalence of coma, in diabetics enjoying a certain affluence. To these should be add.ed gangrene, pneumonia, and the numerous complications which may affect the organism when already debilitated by diabetes.
In a certain number of cases, particu larly in arthritic subjects, the diabetes may be changed into another malady. Following traumatisms, it may end (after a certain duration of the glycosuria) in simple polyuria.
Principal reasons why diabetes inter feres with operative success: 1. The
sugar circulation in the blood is hygro scopic, and it draws water from all the tissues of the body until the tissues are actually too dry. This must interfere with the normal process of repair, and it probably does so in several different ways. 2. The surgeon must give these cases special attention, because the fluids of a wound loaded with sugar are, in all probability, excellent culture media and particularly susceptible to the attacks of bacteria. Rigid asepsis is, therefore, deinanded. 3. Certain anfesthetics may precipitate an im pendin nephritis because of the un usual labor involved in excreting sugar. In these cases nitrous oxide and oxygen used instead of the other arnesthetics, especially avoiding the use of ether. R. T. Morris (Med. News, June 29, 1901).
Prognosis.—It may be inferred from the preceding statements that it is diffi cult to speak of the prognosis of diabetes in general; this can only be established in each individual case.
E may be said, however, that arthritic diabetes and many cases of nervous dia betes are us-ually not very severe.
in the nervous variety the glycosuria is often quite moderate, and ma,y even disappear. leaving behind a simple poly uria. The type developed under the in fluence of gout in arthritic subjects is associated with an intermittent, but abundant, glyeosuria, and is compara tively benign. Certain diseases of the pancreas, such as calculi of Wirsung's canal, and sclerosis of the whole paren chyma, may be followed by a rapid and dangerous diabetes. There are other varieties difficult to classify. LC•pine (Sem. MM., Aug. 27, '97).
Deductions based on twenty-two origi nal observations fIS well as the literature of the subject in respect to the influence of diabetes upon the functions of the female organs of reproduction. ln dia betes mellitus menstruation is generally diminished, but not always to a degree parallel to the sugar in the urine. Preg nancy in 66 per cent. is undisturbed, in the remainder is prematurely inter rupted, but more often by miscarriage (seven Or eight months) than by abor tion. The prognosis for the mother is likewise doubtful. Pruritus vulvm, boils, and acuminate condylomata are well known diabetic symptoms. Affections of the vaginal mucous membrane and uter ine, and necrosis of the ovaries are not so CO111111011. Kleinwachter (Zeitschrift f. G. u. G., xxxviii, H. 2, '98).