have rapid respiration, and one thinks that they are cases of acute lobar pneumonia in which signs of consolidation are not present. In other cases, a diagnosis of meningitis is made. Pyelitis is a disease in which we are apt to get the condition known as meningismus. As the result of the action of the toxin on the brain, you have vomiting, and may get definite signs suggesting meningitis, like Kernig's sign, retraction of the head and changes in the reflexes. If there is any disease that emphasizes the necessity for routine examination of the urine in every case, this is the disease; since without routine examinations of the urine, the cases are very frequently overlooked.

THE PRESIDENT: Will someone else further discuss these two papers? If not, I will call on Dr. Bartlett to close the discussion.

DR. CHARLES J. BARTLETT (New Haven): I have only a word to add. The possibility of recognizing these obscure cases of pyelitis when careful examinations are made is well illustrated by a recent report of Vanderhoof, published under the heading of "Pseudo-Malarial Types of Pyelitis." This is based upon the study of 2,500 patients in private practice. Of this number he found forty-seven cases, or about two per cent., with pyelitis, and of these, twenty-one had been treated for malaria. He thinks fifty per cent. of the cases of pyelitis have nothing indicating the location of the disease upon physical examination.

Regarding the use of drugs that may be excreted as urinary antiseptics, one thing that has been said this afternoon should be further emphasized, and that is the use of urotropin in connection with any drug that will produce an alkaline urine. That is contrary to sound policy, as shown experimentally. When the urine is faintly acid, there may be some free formaldehyde formed; but when the urine is alkaline, there will be none formed. Hence, the results are more marked when some drug like acid sodium phosphate is given in sufficient quantity to render the urine acid, if possible.

One other thing that I was interested in from Jordan's work was the selective action shown by some drugs in connection with certain bacteria. In attempting to find some drugs that would act as urinary antiseptics when given by mouth, he found that sandalwood oil had an effect on staphylococci, but not on colon bacillus or most other bacteria. In both acid and alkaline urine, it had the same specific action on staphylococci. Urotropin, however, will not act in an alkaline urine. Formaldehyde is only set free from this in an acid medium.

THE PRESIDENT: I will now call on Dr. Fritz C. Hyde of Greenwich to make some remarks in closing.

DR. FRITZ C. HYDE (Greenwich): I think I shall not add anything further.

The Use of Artificial Pneumothorax in the

Treatment of Pulmonary Tuberculosis.

DAVID R. LYMAN, M.D., WALLINGFORD.

Rest has long been recognized as probably the cardinal factor in the successful treatment of tuberculosis. Whatever the organ affected, the rule holds good that the nearer we can approach to placing the member at absolute rest the better are our final results. In tuberculosis of the bones and joints this is the treatment upon which we most rely. In laryngeal cases if we can only get the patient to stop using his voice-to keep the larynx absolutely quiet-we often meet with astonishing success after other means have failed. In pulmonary tuberculosis, having been without means of arresting the movement of the diseased lung, we have endeavored to keep it comparatively quiet by putting the patient himself at rest during the active stages of the disease and for varying periods thereafter. By so doing we decrease the rate and depth of respiratory movement and the rapidity of the circulation. In other words we put the lungs at comparative rest and decrease the absorption of toxic products from the diseased areas. It is true that many patients are so fortunate as to recover without a vigorous rest cure, but it is equally true that those who adhere most closely to such a course give us a far greater average of favorable results than those who do not; and especially is this so of those in whom the disease was active at the beginning of treatment. The beneficial effects which sometimes follow the appearance of a pleural effusion have at times been noted by most physicians; so much so that there seems to be a growing tendency against the immediate removal of such effusions as cause no distressing symptoms. In spontaneous pneumothorax several cases have been reported in which the appearance of an extensive effusion, which was not withdrawn, has been followed by marked decrease of active symptoms, improvement of general condi

In

tion of the patient and subsequent arrest of the disease. recent years I have had two such cases. The patients both had actively spreading disease before the occurrence of the pneumothorax. Following the effusion, which in each case extended up above the level of the third rib, there was a sudden drop of the temperature, a decrease of cough and expectoration, and the patient began to improve. The fluid was allowed to remain for nearly a year, small amounts being drawn off only when pressure symptoms presented. It was finally removed by gradual stages. In each case the lung to a large extent re-expanded, and the patient secured an arrest of the disease. It is on cases such as these that the treatment of pulmonary tuberculosis by artificial pneumothorax has been based-the object being to fill the pleural cavity with an inert gas, and by direct pressure on the lung to place this organ as nearly in a quiescent state as possible.

The theoretical advantages of such a procedure are many. The compression of the lung not only limits its mobility but also limits the flow of the blood, and the absorption of toxic products through the lymph channels is controlled; cavities are reduced. or obliterated and the absorption from the secondary infections greatly lessened. Most important of all, the putting of the lung at rest promotes the formation of connective tissue, and thus contributes to the limitation of the process and its final control. This increased formation of connective tissue has been demonstrated in autopsy reports on patients treated by artificial pneumothorax.

Nitrogen is the gas most generally used for this operation as it is more slowly absorbed than either oxygen or sterile air, both of which have been employed. The apparatus consists of two jars; one is stationary and contains the gas; the other, on a sliding rachet to allow for its elevation, is filled with fluid with which to obtain the needed pressure for forcing the gas into the pleural cavity. The jar of gas is connected by a tube to the aspirating needle, and this tube has a pass-cock leading to a water manometer on which the respiratory oscillations, the intra-thoracic pressure and the pressure exerted by the fluid

behind the gas may in turn be noted. The needle is equipped with a stilet in the head, so arranged that it can be withdrawn and re-inserted without disconnecting the gas. Dr. Baldwin, of Saranac, has recently added an adjustable guard which is fastened over the shaft of the needle and which enables the operator by firm pressure against the chest wall to maintain the point of the needle at the proper depth, thus avoiding accidental injury to the lung through a sudden slip, or, what was a more common incident, the withdrawal of the needle from the pleural cavity and the consequent injection of gas into the chest wall.

THE OPERATION.

There are two methods in vogue for performing the operation-the puncture method of Foralini, who in 1882 first suggested the possibilities of the treatment, and the incision or Brauer-Murphy method. In the latter operation an incision is made under local anesthesia and the tissues dissected until the parietal pleura is exposed; the needle is then inserted and the gas introduced under the control of the sight of the operator. This method has some advantages. It enables one to see just where the point of the needle is and to avoid injury to the lung and the possibility of gas embolism through injection of gas into a blood vessel. If the pleura be found adherent the wound is closed and another incision made. At the completion of the injection the intercostal muscles are stitched together over the puncture in the pleura and the wound is closed. Where the operation has been successful and the pleural surfaces are separated by gas, subsequent injections are made by the puncture method. The Brauer-Murphy procedure has several distinct disadvantages. The incision impresses the patient as far more of an operation than the simple puncture method. There is much greater danger of infection and more liability of the escape of the gas out into the tissues of the chest wall. The latter is never a serious occurrence, but is at times very uncomfortable for the patient and renders it impossible to maintain the intra-thoracic pressure at the desired

point. Recent improvements in the technique of the puncture method of Foralini, a due appreciation of its possible dangers, and a rigid observance of the rules for its use, combine to make this operation fully as safe as the more formidable BrauerMurphy procedure, and it is now the one generally in use. The danger of infection is much less. The patient has no dread of the repeated operations, and the escape of gas into the chest wall occurs much more rarely. Its great dangers, injury to the lung or the production of gas embolism, can be avoided by the use of the water manometer, care being exercised never to inject the gas until the oscillations in the tube show beyond doubt that the point of the needle is between the layers of the pleura.

The technique of the operation is as follows: The patient should first be given a small hypodermic of morphine to allay the nervous apprehension. The site at which the injection is to be made must be largely determined by the physical signs; the only general rule being to select that area over the lower portion of the side, either front or back, which seems to present the least liability of pleuritic adhesions. The results of such selections in any series of cases will make one much more lenient toward the diagnostic errors of his fellow practitioners. One of my cases with general involvement of the whole side, of three years' duration, and in which I had grave doubts of being able to get into the pleural sac, proved readily collapsible on the first attempt, while in another, in whom my notes for three months showed no apparent abnormality over the lower third of the lung, all attempts to produce a pneumothorax failed on account of a closely adherent pleura over the whole base. The site having been determined upon, the skin is anesthetized with a solution of I per cent. novocaine with I-10,000 adrenaline. Through this a long hypodermic needle is inserted and the same solution injected as the needle advances straight through to the parietal pleura. Cocaine should not be used for this purpose. This local anesthesia is of paramount importance as the deadening of sensation in the pleura practically abolishes the danger of pleural shock-that grave acciden