be needed. There was no evidence of disease of the biliary tract or appendix. The advisibility of further study was explained to the patient and he was urged to remain in town for that purpose, but he insisted on his trip to the West so he was given a modified ulcer treatment. Numerous reports from him, including one of a few days ago, show that he has been entirely free from pain and vomiting and with only an occasional stomach symptom of any kind.

Certainly the results of this case have been very different from those that I had anticipated, and it well illustrates the ease with which wrong opinions may be formed and unsuitable advice given as a result of insufficient diagnostic work.

Hemorrhagic Osteomyelitis.

ERNST H. ARNOLD, M.D., New Haven.

Several reasons prompt me to bring to your attention the above subject at the present time. Hemorrhagic osteomyelitis is apt to exist without symptoms subjective or objective and may go, therefore, unrecognized. Fracture without what would be to a healthy bone sufficient provocation may take place. A bone having thus been broken and hemorrhagic osteomyelitis yet not recognized as the remote causative agent, delayed or nonunion may result.

If the lesion be situated near a joint which from the predilection of the lesion for spongy bone is likely to be the case, it may be mistaken for a joint lesion such as tuberculosis. Conservative treatment for the latter, braces for instance, have no influence on hemorrhagic osteomyelitis, whereas prompt surgical interference would speedily restitute the part at integram. With the former the patient would face a long drawn out and inefficient course of treatment.

Finally hemorrhagic osteomyelitis may be mistaken for a malignant bone lesion and drastic and mutilating surgical measures taken which would be altogether unnecessary.

In view of these reasons a recital of the nature and behaviour of this condition together with the relation of a few illustrative cases may not be out of place.

ETIOLOGY.

The condition is not very frequent. This infrequency is in part undoubtedly due to the fact that the lesion exists often unrecognized and undiagnosed. Its preference for spongy bone explains its occurrence in childhood and youth. I myself have not come across a case older than twenty-five years. The pathology being that of a process of repair, the injury to the tissue calling for repair must in the absence of infection be traumatic. That seems to be the case history as the history of trauma is frequently given.

PATHOLOGY.

The gross pathology of the lesion usually presents a cavity in the spongy part of a long bone. When the formation of this cavity has progressed for some time it consists of one chamber of very regular, often oval outline. If bone absorption has not yet advanced much, the unabsorbed trabeculae may subdivide this chamber in several compartments. The cavity is found filled with a dark brownish mass of jelly-like consistency from which uncoagulated blood oozes. A membrane brownish red in appearance, fairly thick one eighth inch and more and which is very friable, lines the cavity. The cavity expands driving the cortica before it, but unless there be spontaneous fracture rarely ruptures through it. The lesion is noticed then as a swelling of regular shape and smooth surface. There are no signs of inflammation, no reddening, no heat till the pressure from underneath causes the skin reaction to show these signs. Microscopically the mass is found to be a hemorrhagic extravasation in which light colored areas consisting of young fibroblastic tissue are noticed. In these are found many multinucleated giant cells of the foreign. body type of cell. They are derived from the fusion of leukocytes proliferated from vascular endothelium. The giant cell formation is due to the presence of much hemorrhagic granular tissue. It is also favored by the necrosis in the extravasated areas. The tissue necrosis is likewise responsible for the appearance of fat globules in and around the endothelium cells. Numerous fibroblasts are seen forming collagen fibrils. Where the process. becomes chronic this collagen formation is well marked, proceeding to fibrous contraction and cyst formation. This cyst formation and the presence of giant cells were undoubtedly responsible for the view held in many quarters until quite recently, that the condition was one of tumor formation, bone cysts of this variety being classed as giant cell sarcoma. It is now, however, recognized that the process is one of a low grade inflammation usually chronic in character in response to trauma. The giant cells have not a tumor forming but as Barrie, who has done a great deal of excellent work on this condition puts it, scavenger function.

SYMPTOMS.

There may not be any subjective symptoms. Pain is present only when the expansion has gone far enough to make the cortical support of ligament or muscle attachment insufficient. We have then pain on joint movement in the first, or muscle contraction in the latter case. Likewise may there be absence of objective symptoms and the condition is only accidentally revealed either when X-rays are employed for other conditions in the neighborhood of the lesion or by the fracture of the bone. With the expansion, however, of the lesion we have a distinct enlargement of the bone. Usually there is neither heat or redness to be noted. Only when the pressure from below becomes keen do these set in. There may be enlarged superficial veins over the swelling due to pressure interference with the venous circulation. In the lower extremity limping may be associated with the condition. The condition being strictly a local one, no general symptoms such as fever, anemia or loss of weight need be looked for. Atrophy of the soft parts in the neighborhood is absent in the beginning and not particularly marked in the advanced states. If the lesion is very near the joint there is some limitation of movement but passive movement within the limitation is painless. Protective muscle spasms are absent or only slightly marked.

DIAGNOSIS.

The youth of the patient, the location in spongy bone, the history of trauma, the absence of marked pain, limitation of movement and muscle spasm, the long duration without exacerbations or metastasis make the diagnosis likely. The X-ray findings are sufficiently characteristic to confirm the tentative clinical diagnosis.

The X-ray shows a cavity usually very regular in shape, sharply outlined. The cavity shows no bone proliferation. The cortica, unless broken by violence, is intact. If broken no bony tumor mass is shown outside the cortica.

The differential diagnosis is important in regard to malignant bone tumor (sarcoma) and bone tuberculosis. Hemorrhagic. osteomyelitis has the age of the patient in common with both.

The location is the same in tuberculosis and may be the same in

sarcoma.

The history of trauma may be had in all three. The presence of early and great pain distinguishes tuberculosis and sarcoma from it. Limitation of movement is an early and constant symptom of tuberculosis; it is marked in sarcoma near joints. Muscle spasms are marked and constant in tuberculosis and sarcoma near the joint. The local swelling of tuberculosis is diffuse and involves the soft parts. In sarcoma it may be circumscript but is irregular. The tubercular joint is hot. Sarcoma and hemorrhagic osteomyelitis only show redness and heat when the skin becomes involved. Tuberculosis and sarcoma are progressive; hemorrhagic osteomyelitis may be stationary for years.

Atrophy of soft parts adjacent above and below the lesion is present in tuberculosis and sarcoma. It is absent in hemorrhagic osteomyelitis except where long disuse has brought it about. No fever or general marasmus is noticed in hemorrhagic osteomyelitis. The X-ray usually clinches the matter. Tuberculosis presents no cavity, sarcoma not one of such regular shape and smooth outline. Sarcoma involves the cortica, breaks through it and usually shows bony mass outside of it.

TREATMENT.

A sufficient opening through the cortica to allow the complete curettage of the hemorrhagic membrane and of the whole cavity and recesses if there be such. If the cavity is large, filling it with bone chips taken from adjacent healthy cortica is helpful. Suture of periosteum and skin. No drainage. No use of tissue destroying swabbing such as carbolic acid. No bone wax filling. Perhaps a mildly stimulating application of dilute tincture of iodine may be had. If the cavity be large and the cortica thin some protecting splint must be worn even in the upper extremity. In the lower extremity a protecting and supporting brace is indispensable till the X-ray shows sufficient bone deposit in the cavity to prevent untoward accidents. The X-ray taken at fairly frequent intervals (six to eight weeks) will control the recovery and indicate when protection may be done away with. The prognosis is uniformly good.