Liberal diabetic diet the days before these:

On the fifth day of a preliminary fast the D : N ratio of this man was 2.95. Then for four days he was given a mixed diet, moderate in quantity. After this followed a diet containing about 100 grams of protein and the D : N rose to 3.97, 4.01, and 3.87 on three successive days. On the first of these days there was distinct drowsiness; there were 36.4 grams of nitrogen in the urine after an intake of 19 grams in the food; 71 grams of β-oxybutyric acid were eliminated in the urine or about the quantity which could have arisen from the oxidation of fat during the period; the carbonic acid tension in the blood was one-half the normal, and the respiratory quotient was 0.687. The metabolism presented the picture of complete diabetes (see p. 469). Several who saw the patient pronounced the outlook hopeless. Joslin, who happened to be in New York at the time, gave a favorable prognosis. This prognosis was correct. A period of fasting interrupted by days of very low diets resulted in the complete disappearance of glucose and of high nitrogen elirnination in the urine within ten days, and on the eleventh day only 0.2 gram of β-oxybutyric acid was eliminated.

Joslin has for a long time privately informed the writer that he would not place a patient upon a diet consisting of protein and fat alone on account of the deleterious effects which might be produced, and he has stated in personal conversation that fatal results might ensue if the diet were long continued. The experiment on C. K. as well as one unpublished experiment not here described show most clearly that Joslin is correct as regards the evil effect of even a moderately high protein intake upon the diabetic patient.

The case of C. K. exemplifies the method of modern treatment of diabetes known as the "Allen method." Wein-traud,1 in the clinic of Naunyn, was the first to recommend the interpolation of occasional fasting days for the benefit of the diabetic patient, and Naunyn2 practised the reduction of the body weight of the patient, both by interposing single fasting days and by giving 600 grams of green vegetables which contain little nourishment. Allen, however, on the basis of many fasting experiments with depancreatized dogs,1 obtained information which he was subsequently able to apply to many diabetic patients2 in the Rockefeller Hospital. It is certain that he was the first to introduce a rigorous regime of fasting until the diabetic patient becomes free from urinary glucose and from acidosis. Frequently whisky was administered as the only nourishment. Benedict and Torok3 were able to reduce the aceton excretion, as well as that of nitrogen and glucose, after administering alcohol to a diabetic. The experiments of Otto Neubauer4 showed that red wine reduced the sugar output and the acidosis in diabetes, and Allen and Du-Bois find indications that the administration of whisky favors the oxidation of glucose in the diabetic. Joslin5 and Allen and DuBois report that during the clearing up of the diabetes in a fasting patient respiratory quotients are found which are higher than could possibly be obtained from the oxidation of body protein and fat alone. This may possibly be due to the oxidation of the accumulated aceton bodies, the respiratory quotient of β-oxybutyric acid being 0.89, an explanation given by Joslin, who has found respiratory quotients in fasting diabetics as high as 0.8.

1 Weintraud: "Centralblatt fur klinische Medizin," 1893, xiv, 737. 2 Naunyn: "Zeitschrift fur arztliche Fortbildung," 1908, v, 737.

Although the great majority of diabetic patients are cured by the fasting treatment, so that they may live on a carefully regulated diet without showing glycosuria, still there are some cases that do not yield to such treatment. Joslin6 has reported concerning 14 such patients who were diabetics of long standing, or very severe or complicated cases.

1 Allen: "Glycosuria and Diabetes," Boston, 1913.

2 Allen: "Journal of the Amer. Med. Assoc," 1914, lxiii, 939; Ibid., 1916, lxvi, 1525.

3 Benedict and Torok: "Zeitschrift fur klinische Medizin," 1906, lx, 329. 4 Neubauer, O.: "Munchener med. Wochenschrift," 1906, liii, 791.

5 Joslin: "American Journal of the Medical Sciences," 1915, cl, 485.

6 Joslin: "Boston Medical and Surgical Journal," 1916, xlxxiv, 371 and 425.

One of the most interesting researches upon diabetes ever accomplished is that of H. O. Mosenthal in Janeway's Baltimore clinic. At this writing the results have not been published, but are presented here by special permission.

The following table shows the results obtained from a fasting diabetic with a D : N ratio closely approximating that found in the phlorhizinized dog, and resembling a phlorhizin. ized man (see p. 455) both as regards the D : N ratio and in the quantities of ammonia and β.oxybutyric acid eliminated.

Diabetic Patient Practically Fasting

(One egg and green vegetables containing 2 to 6 grams carbohydrate per day during last five days. Wh. = whisky; Wi. = wine. Bicarbonate of soda also administered. The D : N is calculated after subtracting ingested carbohydrate).

* And 360 c.c. wine.

At the beginning of the investigation the blood sugar of the diabetic was 0.33 per cent., but fell on the eighth and ninth days to 0.25 and 0.24 per cent.

The advent of the Allen fasting treatment seemed at first likely to make it possible to dispense with laboratory records concerning diabetic cases, but it is still necessary in 10 per cent, and perhaps more of the cases to follow the sugar output, and the intensity of the acidosis for these do not invariably diminish.

Joslin's method2 of applying the Allen treatment is here reproduced:

2 Consult Joslin: "The Treatment of Diabetes Mellitus," 1916, p. 243.