Many experiments have been made in man and animals to gain information as to the actual disturbances of metabolism in fevers and to determine how recovery may be aided by diet.

Fevers are commonly due to bacterial toxins. These produce a disturbance of the regulation of heat, of which the most marked feature seems to be a diminution of the loss of heat from the skin. In many cases there is an increased production of heat, but it is much less in degree than was formerly thought.

The amount of oxygen used is not commonly raised more than 10-20 per cent, and some authorities regard this increase as due solely to muscular activity, evidenced in the restlessness of a feverish patient. The excretion of nitrogen in the urine is much greater than is normal for a healthy person fed upon a similar diet, and this is due to the breakdown of protein tissues. This breakdown appears to be, in some cases at all events, independent of the increase of oxidation, and in some degree independent of the temperature, for it may occur in septic infections when there is no rise of temperature. Artificial pyrexia does, however, produce both increased oxidation and protein breakdown. The destruction of protein is considered to take place chiefly in the muscles because potassium salts, neutral sulphur and creatinine, which are regarded as derived from muscle, are also passed out in greater quantity. The writer has published some evidence to show that uric acid is probably chiefly derived from non-muscular protein tissues : if this be also true in pyrexia, we may conclude that all the protein tissues share in the disintegration, for the uric acid excretion is greater in fever.

With these considerations in mind we may now look at the experiments upon diet in febrile conditions. In the first place, it has been shown that animals suffering from fever cannot endure deprivation of food so well as normal animals. This seems obvious enough, but strikes directly against the old practice of putting fever patients on as low a diet as possible, and shows that food should be supplied in fever so far as is compatible with the condition of the patient. The idea has been current that the food taken in fever is not properly utilized, but analyses of the faeces show that the absorption of protein, fat, and carbo-hydrate does not materially differ from that in health: various experimenters have found this in typhoid fever, pneumonia, tuberculosis, recurrent fever and diphtheria. It might be thought that carbo-hydrate and fat, although not appearing in the fasces, are broken up in the intestines by bacterial action, and their energy lost to the tissues, but this has not been shown to be the case. On the contrary, it has been proved that it is possible to supply sufficient food in fevers to prevent a loss of weight in some cases and to greatly diminish the loss in others. The heat value should amount, if possible, to about 25 calories per kilogramme, a quantity sufficient to maintain the weight of a resting person. Three pints of milk, containing 1,150 calories, would be adequate for a person of 46 kilogrammes or 7 1/4 stone, and is therefore less than should be supplied to persons of greater weight. Milk should form the basis of the diet, because it appears that it is digested with the least expenditure of energy on the part of the intestines, but it should be thickened by the addition of other foods.

Protein may be added to the milk in the form of plasmon or some similar product, or given as meat juice, or minced meat, or pounded fish. Experiment has shown that the loss of nitrogen from the body is much less when a fair quantity of protein is taken, but it has not been found possible in pneumonia or typhoid fever to attain nitrogenous equilibrium. It is, however, an advantage to supply protein freely, and Moraczewski, who observed the metabolism in eighteen cases of pneumonia, believes that convalescence is thereby shortened. Puritz came to the same conclusion as the results of his experiments on cases of enteric fever.

Fat is a valuable constituent of the food, but requires to be given with caution, since an excess so readily excites disgust, and hinders the patient from taking a sufficiency of other foods. Milk fat is easily absorbed, and as much cream and butter as possible should be taken. The cream may be given in soups or added to the milk. Butter may be taken on bread, in puddings, and with a little potatoes. Two ounces of butter and 1/4 pint of cream will add 750 calories to the daily ration.

Carbo-hydrate must necessarily form part of the febrile diet. It has been shown, both in animals and in man, that dextrose saves protein from being broken down in fever, as it does in health. It is clear, therefore, that starches and sugars should be given as freely as possible. Arrowroot is often well taken by patients with cream and sugar. Cornflour may be added to the milk, and, in the less severe cases, thin bread and butter or biscuits allowed.