SUPERIMPOSED CAVITIES, THE POS- SIBLE CAUSE OF CRACKED POT SOUND * JOSEPH WALSH, M.D. PHILADELPHIA Cracked pot tympany (bruit de pot fele of Laennec) is so well known to clinicians that a special description is unnecessary. It was first called attention to by Laennec about 1820, and in 1860 was so fully described by Walshe1 that nothing has been added since. It is found over the following conditions: Pulmonary cavities. Osler2 says "only over toler- ably large cavities with thin walls"; Fowler and God- lee,3 superficial cavity with slightly yielding walls and free bronchial communication, the mouth being open at the time ; Cornet4 in Nothnagel's Practice, cavities, especially at the apex communicating by a narrow orifice with an open bronchus. Pleural effusion (skodaic cracked pot, though ordi- nary tympany or hyperresonance is much more com- mon ). Occasional cases of pneumonia when the consolida- tion is at its height. Occasional cases of bronchitis, especially in children. Normal lungs of crying children during expiration. Cornet mentions relaxed or infiltrated or tubercu- lous lung, or open pneumothorax. A similar sound can be produced by filliping the cheek with the mouth open, or striking the back of the hands, loosely folded across each other, against the knee, the contained air being forced out abundantly * Read at the Conference of the Medical Staff, U. S. Army General Hospital No. 17, Markleton, Pa., Feb. 12, 1919. 1. Walshe: Diseases of the Lungs, Philadelphia, Blanchard & Lea, 1860. 2. Osler, William: The Principles and Practice of Medicine, New York, D. Appleton & Co., Ed. 8, 1917, p. 201. 3. Fowler and Godlee: Diseases of the Lungs, New York, Longmans, Green & Co., 1898 edition, p. 57. 4. Cornet in volume Tuberculosis Nothnagel's Encyclopedia of Prac- tical Medicine, Philadelphia, W. B. Saunders Company, 1904, p. 424. 2 and quickly between the fingers at each blow, or by filling the hand with coins, shutting it tight enough to allow only a slight space for the coins to move, then shaking the hand.5 It is commonly thought that the larger the cavity, and the nearer to the anterior wall, the more likelihood of its production. Yet we often see cavities of the largest size immediately underlying the surface not producing it, so that these circumstances alone do not explain it. Walshe states that in a case showing it, it will cease to be produced if the mouth and nose be closed, though p.ercussion will still show tympany. He contends that the fair interpretation of this fact seems to be that the sudden rush of air from the cavity out- ward, produced by the forcible blow on the yielding parietes in the ordinary open state of the mouth and nose, but completely prevented by their closure, is the real cause of the phenomenon. It has, therefore, been considered necessary for the cavity to connect freely with bronchi, thereby allowing the air in the cavity to be expelled freely and instantly. Yet we often find at necropsy large cavities apparently connecting freely with large bronchi without the sound having been pro- duced (Case 439, U. S. Army General Hospital No. 17). It has also been asserted that it depends on the colli- sion of liquid and air in the cavity ; but Walshe called attention to cavities which with fluid did not produce it and without fluid did produce it. In addition, in the case of filliping the cheek there is no fluid to explain it. It is ordinarily stated, therefore, that the pathologic conditions producing it are a large cavity communi- cating freely with a bronchus, and relaxation of the pulmonary tissue. Dec. 1, 1918, Case 296 at General Hospital No. 17 showed cracked pot in the first interspace near the sternum on the left with dulness to the fourth inter- space. December 5, it showed ordinary tympany in the first interspace and cracked pot in the third interspace 1% inches from the sternum. A diagnosis of two cavities was made. The roentgen ray, however, showed the appearance of four cavities, two superimposed on two others. 5. Sahli, H.: Treatise on Diagnostic Methods of Examination, Phila- delphia, W. B. Saunders Company, 1905, p. 158. 3 The possibility was suggested that the double cavity might have some influence in the production of cracked pot, and further cases were sought. Only one (Case 17) was found. This manifested cracked pot in the first interspace on the left, and the roentgen ray showed again the appearance of two cavities superimposed. This patient is still living. Subsequent examinations of Case 296 revealed the cracked pot tympany sometimes in one and sometimes in the other of the two previously mentioned places and sometimes absent in both. For instance, December 14, there was ordinary tympany between the first and second ribs, and dulness between the third and fourth. When present, the cracked pot sound could be elicited both with the mouth open and with it closed, though more satisfactorily when open. Closing of the nasal passages was not tried. December 30, this patient died. The necropsy dis- closed not four cavities, but three, yet so situated that the first underlay the second and the second the third. The first extended from the apex to the second rib, the second from the first to the third rib, the third from the second interspace to the fourth rib. In other words, though only three cavities were present, they were so placed that in the two situations in which cracked pot sound occurred, there was a part of a cavity superim- posed on another. On the strength of this finding, the conclusion was reached that the superimposing of cavi- ties may, at least, aid in the production of cracked pot sound, and investigations were made by artificially introducing cavities into the removed lungs of animals. It was found that percussion of two inflated bladders tied so as to prevent the escape of air, either introduced into an animal's lung, or placed over one another out- side, gave tympany, but not cracked pot. In other words, communication with the air is evidently neces- sary; the mere rattling of cavities on one another is not sufficient. An open trachea was percussed both within and without the lung, and found to produce it, though not so typically as occasionally heard; two open tracheas, however, superimposed produced the sound exquisitely. The application of these facts to its occurrence in pneumonia, above a pleural effusion, and in crying children, affords a more satisfactory explanation than 4 the one usually offered, namely, relaxation of the lung. In pneumonia, the tissue is consolidated around the bronchial tubes, so that they act practically like super- imposed cavities, sometimes open, sometimes occluded, only in the former event producing the cracked pot sound. Above a pleural effusion the normal lung is compressed about the bronchial tubes, and since there is no secretion to occlude them, the sound is very common, much more common, for instance, than in pneumonia. An analogous condition prevails in crying children. Fluoroscopic examination reveals that if we exhale all the air possible and then cough several times, we can raise the diaphragm to the third and even occa- sionally the second rib, thereby compressing the lung to this extent. The effort of the crying child is practically entirely expiratory, producing a similar condition and one not unlike that above a pleural effusion. In addi- tion, the bronchial tubes of the child are much larger in proportion to the lung space, and give the sound of cavity more evidently. I have never heard it over pneumothorax, though I have examined and necropsied more than twenty cases, several of which (for instance, Case 210) com- municated with the lung by large openings. Stanton6 reports sixty-one cases without cracked pot ever being noted. With the idea of superimposed cavities in mind we would expect it only over a small pneumothorax communicating freely with the lung with a large cavity behind it. 2026 Chestnut Street. 6. Stanton: Fifth Annual Report of the Henry Phipps Institute, p. 251. Reprinted from The Journal of the American Medical Association Nov. 29, 1919, Vol. 73, pp. 1656 and 1657 Copyright, 1919 American Medical Association, 535 N. Dearborn St., Chicago