Jaccoud, Moizard, and many of the French writers attach much importance to a purely milk diet (regime lacte absolu). This should be invariably used in the febrile stage, and should be continued with only slight addition, such as chocolate and coffee, for a full week after the temperature is normal. Thereafter eggs, meat and vegetables are allowed. Fish, which in this country we usually prescribe in preference to meat, is excluded altogether, being placed in the same category as pork. Even after the diet has been thus extended the patient is expected to drink from 2-4 pints of milk daily. A weak point in this diet is the fact that the nitrogenous additions are first made in the third week, at the very time, in other words, that we are beginning to expect nephritis. Jaccoud claims, however, that this condition never follows in cases so dieted. This is not in accordance with my own experience, as I have seen nephritis supervene in cases in which nothing but milk has been given. The method of Forchheimer is more logical. He allows fairly liberal feeding during the febrile stage when most of us are content with a fluid diet. From the beginning, however, of the second week until the end of the fourth, he restricts the diet to milk alone. He insists on the importance of supplying much fluid to assist the elimination of waste products, and considers that milk, "when subjected to the action of bacteria, will produce fewer harmful end products of digestion than any other article of food." At the end of the fourth week carbo-hydrates are first tried. If no harm results meat is added, and normal diet is resumed. Even stricter dietaries than the above have been suggested, such as milk and milk alone for six weeks, or milk with carbo-hydrates, such as rice and arrowroot.
There is certainly little evidence that the occurrence of nephritis is in any way influenced by the use or by the restriction of nitrogenous substances. On the other hand, it would appear only reasonable in a disease in which nephritis is liable to occur to put as little tax as possible on the excretory powers of the kidney. The complication may occur at any time during the fever and convalescence, but the usual period at which it appears is from the sixteenth to the twenty-sixth days. For instance, in a series of 134 consecutive cases of nephritis, 88 occurred within the 10 days named; and of these 68 had their first symptoms between the twentieth and twenty-fourth days inclusive. Whether we regard the cause of the inflammatory process as due to the direct invasion of the kidney by the streptococcus, and that micro-organism has undoubtedly been demonstrated in the organ, or whether we prefer to believe that for some reason or other there is a specially plentiful discharge of irritating toxins about this particular period, the excretion of which proves too much for the kidney, any efforts to prevent the renal tissue becoming depressed either by chill or by hard work, seem perfectly logical. We do not believe nowadays that chill alone will cause nephritis, but it may certainly facilitate its occurrence by lowering the resistance of the kidney. May not the work entailed by the excretion of large quantities of nitrogenous food have a similar effect? Recently it has been suggested by those who hold the streptococcus responsible for the inflammation that prophylaxis is best directed to the antiseptic treatment of the throat. This is no doubt excellent, but, to my mind, the difficulty of accepting this theory of causation is the extraordinary regularity with which nephritis occurs at a definite period of the fever, the end of the third week. The streptococcus has been in the blood from the first. Why has it not attacked the kidney earlier? It has shown no such delay in causing arthritis, usually a much earlier complication. Inflammations of the middle ear occur throughout the fever, early and late, but do not group themselves into a particular period of about 10 days. The connexion of the throat and ear, moreover, is much more apparent than that of the throat and kidney. I am therefore inclined to regard nephritis as due to an excretory difficulty and, as such, likely to be influenced by the diet employed. Still, it is only fair to admit that, as far as statistics go, there is no proof that diet exerts such an influence in prophylaxis, for whereas Foord Caiger reports an incidence of 11.9 per cent of nephritis and late albuminuria in 10,983 consecutive cases of scarlatina, dieted presumably on the liberal lines indicated above, of 4,436 patients of my own, who had a more restricted dietary, 11.02 per cent developed nephritis or albuminuria, a difference which can be totally disregarded.
If, however, the more moderate diet is considered both sufficiently adequate and theoretically safer, it is always as well to consider the age of the patient. Most cases of nephritis occur between the ages of 3 and 8 years, and in older children and young adults more latitude may be allowed in increasing the amount and character of the food allowed.
On this question there is little to be said. With the exception of nephritis, the complications which are liable to occur in the convalescence of the fever demand, as a rule, no modification in the diet. Should considerable pyrexia accompany any of them, it is only reasonable to return to fluid food, or even to restrict the patient entirely to milk. If the appetite remains good, a slight degree of pyrexia may be disregarded and the ordinary diet of convalescence continued. In nephritis, however the patient has been fed previously, a milk diet should at first be insisted on. For a day or two, even, a little arrowroot and water with no milk may be given, or it may be advantageous to limit the patient to whey. As long as the urine is restricted in amount, or blood remains in it, it is wiser not to go beyond the milk diet. Fluids must be liberally supplied in order to assist the elimination of waste material by the skin and bowels. A useful drink to give is the so-called "Imperial drink," which is made by dissolving a drachm of acid tartrate of potash in a pint of boiling water, and flavouring with lemon juice and sugar. The mixture is allowed to cool before use. Barley water may also be used. Recently the question of limiting the amount of chlorides given to the patient has been much considered, and if the food is increased as the complication continues, it is probably wiser to limit the amount of salt taken. I am not, however, in the habit of adding much to the milk diet if the amount of albumin remains considerable, unless the condition is prolonged beyond three weeks. If at that time the albuminuria still persists, the best way to prevent it becoming chronic is undoubtedly to feed. Assuming that the patient has hitherto had nothing more than milk and milk puddings, the addition of one or two boiled eggs to the day's diet is my first step. Caiger points out that in this way the loss of albumin may be to some extent replaced, and since first trying his method some years ago I have always used it, being entirely satisfied with the great improvement in the colour and strength of the patients and the rapid disappearance of the albuminuria. It is just at this time, also, that we are accustomed to prescribe iron, and it is probable that the comparative richness in iron possessed by eggs may have something to do with the good results obtained.
The question of providing a diet in nephritis from which chlorides are, so far as possible, excluded may be worth considering. Delearde recommends such a regime for scarlatinal nephritis and considers that milk, containing as it does 15 grains of salt, or more, to the pint, is by no means an ideal diet for this complication. He suggests as a prophylactic against nephritis a dietary containing eggs, dried or fresh vegetables, and even a little meat. Fish, fats, jam, and bread made without salt are also allowed. As the diuretic effect of milk may be missed, other diluent drinks must be given freely. When the nephritis has occurred a very similar dietary may be employed, one of Delearde's patients receiving two eggs, puree of rice with sugar, and potatoes. A diet so elaborate as to require the bread being especially made is too troublesome to give when so comparatively few cases develop nephritis. In the presence of the complication, however, especially if there is much oedema, the system deserves a trial.