It has long been known that active exercise of the muscles decreases the amount of glycosuria in diabetes mellitus. This idea has been enlarged upon by Bouchardat (1841), by Kulz (1874), and by von Mering, who noticed that when glycosuria is lessened by physical exercise the quantity of urine secreted is not increased.

Just as the clinical picture of diabetes mellitus and glycosuria presents endless variations, so does the therapeutic value of physical exercise vary for individual patients.

There are many individuals who never, or only occasionally, have an appreciable amount of sugar in the urine, but who frequently or constantly have some glycosuria after a rich meal; these individuals should take much active physical exercise, and are often very suitable cases for the terrain-cure, not only or chiefly because of the glycosuria, but rather because of gout, obesity, heart weakness, and other results of a sedentary life.

This is also the case with the mild cases of diabetes, i.e., those who can take some carbohydrate without the appearance of glycosuria, or who at least are free from sugar when carbohydrates are excluded from the diet. In many cases where the diabetes is complicated by the presence of other ailments, especially obesity, the addition of physical exercise to the routine of daily life is beneficial. Although the patient has often a deceptive appearance of good health, the greatest care must be taken not to overtax his strength or diminish his weight too rapidly.

Advanced cases of diabetes, i.e., those cases in which, under any diet, sugar as well as acetone, diacetic acid, and often β-oxybutyric acid are present in the urine, are always critical cases and liable to end in coma, especially if β-oxybutyric acid is found in large quantities. These are seldom suitable cases for gymnastics of any kind, and least of all for the terrain-cure, since a short walk causing the slightest fatigue may induce a fatal termination.

It is obvious that muscle massage may have the same effect as physical exercise in decreasing the amount of glycosuria in diabetes. Finkler of Bonn and Brockhaus of Konigsberg have shown this to be the case. Experiments were made in five cases of advanced, six cases of slight diabetes, and three cases of simple glycosuria. Massage consisting of muscle kneading over the whole body was given for periods of twenty minutes, at first once and later twice daily. The amount of carbohydrate taken by the patient was not determined, but was quite unlimited, and the patients were allowed to eat bread and potatoes, with plenty of fat and meat, as well as to drink ale. Some of the patients were up and taking vigorous physical exercise; others were in bed. In all cases a diminution in the secretion of sugar took place, the amount of urine diminishing, but the percentage of sugar remaining the same for a long time. In some cases the decrease was not noticeable; on the average, however, during three months' observation there was a decrease from 450 to 120 grm. per day ! ! ? In the advanced cases the average for 100 days was 200 grm. per day, where before treatment it had been 730 grm. ! In one case the sugar disappeared, in spite of mixed diet, after three months' treatment, and three months later reappeared to the amount of 5 grm. daily. The thirst disappeared, perspiration and body weight increased. One of the patients died from pneumonia, another from diabetic coma, during treatment.

Latterly I have made experiments in Carlsbad, where I treat a number of diabetics every year, in regard to the effects of massage on the excretion of sugar. Most patients who come to Carlsbad belong to the milder type of the disease, and many tolerate carbohydrates well. It is difficult, on account of the great variation in tolerance, without any obvious reason, which is found in this disease, to carry out in a short time reliable investigations on this point. I have, however, come to the conclusion, in comparing the limit of assimilation in cases treated with and without massage, that general massage, carried out in long seances, does really by itself in some slight degree diminish glycosuria. Such a marked diminution as that noted by Finkler and Brockhaus I have never seen approached; and I do not consider it possible to produce this diminution by massage alone, even if the patient is massaged three or four hours daily. R. Seichten (1895) failed to find a decrease in the amount of sugar in the urine after massage. And I am certain that, in order to make adequate observation, massage must be given in thorough and prolonged seances. Quite apart from the diminution of glycosuria, the treatment often exercises a markedly beneficial effect in every respect on the patient. One of my patients, a young lady of some twenty years of age, had suffered from severe diabetes for two years (after a broken-off engagement), and, although on a strict diet (the diet was only given for a few days for the purpose of examination), a great amount of sugar was secreted, and her vitality was much lowered; her urine showed the well-known port wine reaction with ferric chloride. After a mineral water cure on the advice of another physician, she underwent treatment by general massage, during which she died of diabetic coma. This has happened to another of my patients, and also to one of Finkler's. It is hardly nesessary to point out that even massage cannot work miracles.

I do not know definitely what effect general massage in diabetes has upon the quantity of acetone, diacetic acid and β-oxybutyric acid in the blood. It has been several times my experience that patients suffering from advanced acidosis have died of coma immediately after the cessation of treatment by massage. It is probable that massage promotes the elimination of these toxins by the urine, especially of the dangerous product β-oxybutyric acid, so that the amount of it in the blood is lessened; also that muscle massage helps to complete the oxidation of these substances to carbonic acid and water.

The diet in diabetes is a complicated and much-discussed subject which I shall deal with very briefly. The diet of diabetic patients in my practice resembles that described above for obesity, but differs from it in that the fat in the food is never limited in diabetes, while carbohydrates are occasionally in the earlier stage absolutely excluded for a time. In advanced cases with acidosis (i.e., the presence in the blood of diacetic and β-oxybutyric acid) the absolute exclusion of carbohydrates increases the danger of coma; and this method, chiefly advocated by Naunyn and his school, must, in spite of his reputation, be condemned. At this stage I constantly allow a certain amount of carbohydrate; even in the early cases the "strict diet" should only be used periodically, and the daily amount of carbohydrates should not generally be less than 70 grm.

In Sweden Rickets has occasionally been treated by general massage. This has been more particularly the case in Stockholm, where the atonic condition of the intestines, so often associated with this disease, has led to the use of abdominal kneading, a treatment now very popular. Next to abdominal kneading in value comes deep massage, especially effleurage, of the extremities. Treatment lasts about a quarter of an hour, and is given with the patient undressed. Dr. P. Silfverskjold * has given special attention to this treatment of rickets, and is quite satisfied with its results. He gives the maximum time for a complete return to health as four to six weeks. Massage may well be accompanied by passive gymnastic movements, and must of course be accompanied by dietetic and other treatment.