DR. SIMON FLEXNER, New York (closing discussion).—I recognize the criticism Dr. Koplik makes, that meningitis is a very variable disease; it may not only vary in different seasons, but in different localities. It has not been possible in every instance to get comparative statistics; that is, cases occurring at the same time and of the same age, where nothing, or only lumbar puncture has been done and then those in which the serum has been used systematically.

The largest and the best statistics I have gotten from Belfast, and this is fortunate, because it has been for a year past passing through an epidemic. During the height of the epidemic a mortality of 75 per cent. was reported. I cannot say whether lumbar puncture was used only for diagnostic purposes or also as a therapeutic measure. Dr. Robb has secured for me a record of cases treated outside of the fever hospitals during the same period as those treated with serum in the hospitals. The mortality had been 80 per cent. at a time when it was only 26 per cent. in the hospitals. Of course, the cases outside were probably not treated so well as those in the hospitals. There is also this difference to consider, that the cases outside may have included a greater number of fulminating examples, but the difference between 80 per cent. and 26 per cent. is so great that something else than this must be invoked to explain it. Akron, O., passed through a small epidemic the year after the New York one. There were 26 cases there and in the surrounding locality considerably more. Two sets of figures are to be considered here-12 cases treated with the serum with 3 deaths and 9 recoveries, and 9 cases treated without the serum with 8 deaths and I recovery.

I only state these points for what they are worth. I do not wish to be put in the position of making out an overstrong case for the serum, though I believe in it.

As to the manner of production of the serum, the serum belongs to the class of bacteriolytic sera, in which we are obliged to use the entire constituents of the organisms in its preparation. We know now at least three antitoxic bacterial sera-of diphtheria, tetanus and dysentery. They are the most useful; we all know what they accomplish. You can give an animal any amount of tetanus toxin you please, and if you give a corresponding amount of antitoxin the animal will be saved. The same thing appears to be true of the dysenteric serum. If you multiply the toxic and antitoxic sera equally you secure neutralization. This is not true of the antimeningitis serum. In making the serum the organisms are subjected to unfavorable conditions, when they undergo autolysis and yield an extracellular toxin. With that preparation the process of immunization of horses is begun. It takes many months to accustom the horses to a considerable dose. After this the living organisms are employed to secure the full value of the microorganisms' immunization.

Now when you test the serum on animals, it is found that it does not produce neutralization of the toxin according to the law of multiples, as is the case in the diphtheric and tetanic sera. The guinea pig can be protected up to a certain amount, but a point is quickly reached beyond which multiples of the serum do not neutralize.

In employing a serum of this class it should be borne in mind that the results depend, to a considerable degree, on the concentration. The antibody should be present in a certain state of concentration. When it is diluted too much by the blood and lymph it is powerless to produce the most favorable results.

I believe that the success which has been achieved by this serum has been in virtue of the fact that you bring a bacteriolytic and somewhat antitoxic substance in contact with the focus of the disease. I think this the secret of the matter so far as efficiency goes. Given into the circulation the secretion into the subdural space is slow and imperfect. The thing to be emphasized is that you require this fluid in a certain high state of concentration, and hence injections made directly into the spinal membranes is the obvious way of accomplishing this desideratum.

There is a larger question involved in this study, which is that it is an indication of the value of local application of antisera. If a means could be found of treating pneumonia by some such method as this it would be one of the things by all means to be tried. To return to meningitis. In employing the antiserum, your problem is not to neutralize a poison that is already attached to the viscera. The poison of the diplococcus is not a general, systemic one, but a local one, chiefly acting on the nervous system. Hence the endeavor to bring the agent directly in contact with the focus of the disease, which idea involves, I believe, a slight departure from the general notion of the manner in which infectious processes are influenced by curative sera. The reason we have failed with sera in certain diseases is perhaps because we have tried to reach the local processes through the general circulation. Now in meningitis the general viscera show few evidences of injury at autopsy, thus supporting the notion that the systemic effects of the poison are inconspicuous.

A REPORT UPON ONE THOUSAND TUBERCULIN

TESTS IN YOUNG CHILDREN.

BY L. EMMETT HOLT, M.D., LL.D.,

Professor of Diseases of Children in the College of Physicians and Surgeons, Columbia University, New York.

The observations included in the following report were all made at the Babies' Hospital upon ward patients. Very few of the children were over three years of age, the majority being under two years. Nearly all of the observations have been made. in the past year. I desire at the outset to express my indebtedness to Dr. Josephine Hemenway, house physician at the Babies' Hospital, for her invaluable assistance in the work, the results of which are embodied in this paper. In the early part of the year, unless some positive contraindication existed, some test, most frequently the eye test, was used as a routine measure, in order to determine whether and under what circumstances reactions were obtained in healthy children or in those at least presumably non-tubercular. During the latter half of the period the tests have been chiefly used when some grounds for suspecting tuberculosis existed. Routine tests proved of considerable value in revealing cases of tuberculosis not hitherto suspected. A positive reaction to a skin or eye test was immediately followed up by careful clinical study of the case to discover, if possible, any other evidence of tuberculosis. In a large proportion of such patients bacilli were found in the sputum; though in many only after prolonged search and repeated examinations, thus establishing the accuracy and the value of the test. In many patients further evidence was afforded by the development of definite signs of pulmonary disease.

The Ophthalmic Reaction.-The ophthalmic test was made 615 times; in most instances one test only was made in a patient, although in a few children, where the results were questionable, the test was repeated. The ophthalmic tests were all made with tuberculin obtained from the Rockefeller Institute, which had been precipitated with 65 per cent. alcohol. Before using it was freshly

dissolved in a sterile saline solution.

For the first half of these tests a I per cent. solution was used, for the latter half a 1⁄2 per cent. Especial care was taken not to use the tuberculin in an eye which was the seat of any form of disease, tuberculous or otherwise. As a further precaution the hands of the children. were confined during the first twelve hours to prevent any rubbing of the eye.

In the preceding table are given the reactions in the different types of cases. In the group marked probably not tuberculosis are included those in which no evidence of tuberculosis existed other than the reaction.

The clinical course which the reaction follows is its most diagnostic feature and hence rather close observation of a patient is necessary or some of the milder reactions will be missed. There is usually congestion of the conjunctiva, with some swelling and an increased secretion of mucus, which is frequently very abundant, so that the lids are adherent. There are very seldom pus cells and repeated cultures made revealed no micro-organisms. The symptoms usually begin in from six to eight hours after the tuberculin is used and reach their height in the first twenty-four hours, gradually fading. The usual duration of symptoms is from one to three days. In about 12 per cent. of the cases the reaction lasted somewhat longer than three days; in one case it lasted ten days, but in no case was the test followed by any unpleasant results. Having the opposite eye for comparison enables one, in most cases, to be certain as to the existence

of a reaction. There were, however, a few cases in which the symptoms were so slight that the results must be classed as doubtful.

I am aware that serious results with the ophthalmic test have occasionally been reported by other observers. The precautionary measures taken in this group of cases are, I believe, of considerable importance and largely responsible for our freedom from unpleasant results. On account of the kind of observation necessary, and the possible dangers connected with the eye test, it is not wise to employ it indiscriminately as among the outpatients of a hospital.

The statement has been repeatedly made that young infants do not respond to the eye test. Such was not our experience. Of this series of cases positive reactions were obtained in 14 patients under one year old, 6 of whom were under six months and I only two months old. So far as we could see, the character of the reaction and the frequency with which it occurred were not affected by the age of the patients nor by the extent nor activity of the pathological process, but only by the susceptibility of the patient. In no cases were positive reactions obtained in dying children or those suffering from extreme prostration. In this respect it corresponded with the other tests.

The Skin Reaction.-The skin test of Von Pirquet was employed 217 times. It was made with crude tuberculin obtained from the laboratory of the New York Health Department, which was simply diluted with sterile water to a 25 per cent. strength. With a sterile needle three short linear scratches were made, usually upon the extensor surface of the forearm and into the middle one the tuberculin was rubbed. The skin was allowed to dry and in some cases was covered for a few hours with a piece of sterile gauze. The first sign of reaction was usually apparent in from six to twelve hours; rarely later than twenty-four hours. There is seen a rather bright aureola slowly spreading from the line of scarification to a distance, varying with the intensity of the reaction, from 1/16 to 1/4 inch. The more pronounced reactions are accompanied by an induration readily palpable. The reaction reaches its height in most cases in from twenty-four to thirty-six hours. It then fades rather slowly, usually lasting two or three days, but not infrequently for a week or even longer. It may be followed by a slight desquamation.