from the wound, and since it is imperative that the line of incision be kept thoroughly clear of blood in order that the capsule be exposed to view, I cannot see that it is accomplished more easily in one position than in another. Personally, the child on its back with head to one side, facing the operator is preferable, and that is only because the work has been done in that position more often and the writer can best orient himself when the patient is so placed.

With the laity the removal of the tonsils has come to be more or less discredited simply because we have failed many times to do the work in a creditable manner, and tonsilitis attacks the child about as often as it did before the attempted removal of the tonsils.

If a child has tonsilitis any time after their removal it is positive evidence that all the tonsillar tissue was not removed, or again, if any hypertrophied tonsillar tissue is found in the fossa, it is again evidence that the gland was not removed completely.

That so many disappointments come to both patient and physician in this work is due to the fact that it is more difficult to do a complete tonsillectomy than we have led ourselves to believe and prepare for, but there seems to be no plausible reason why we cannot acquire a technique to remove tonsils as well as a surgeon removes the appendix.

Occult Blood in the Stool.

*By A. D. DUNN, M D., Omaha.

Since Boaz's investigations (D. M. W. 1901-1903) the test for occult blood in the feces has become firmly established in our best medical clinics, but I feel that as a routine procedure it has not received the attention generally that it deserves. The reaction has the definiteness and precision common to chemical tests, its significance is limited, and the technique of the tests comparatively simple. I shall endeavor, first, to show sketchwise the semeiologic importance of occult blood in the feces and its interpretation; secondly, some fallacies, and, thirdly, the relative values of the various tests.

Blood in the stools may be (1) macroscopic; it is then patent to the naked eye as evident blood or its presence is strongly suggested by a black, brownish or tarry stool; or (2) it may be occult, i. e., not demonstratable nor even suggested to the naked

*Read before the Nebraska State Medical Association, Omaha, May 2, 3 and 4, 1911.

eye.

In the latter case the blood either comes from high in the gastro-intestinal tract or if from lower down the amount is so slight as to escape ocular detection. Blood in the stool, whether occult or manifest, if not swallowed, for clinical purposes means. a break in the mucosa of the bowel, which opens a vessel, and this lesion is usually an ulcer. Of course we have the rare types of parenchymatous mucosal bleedings, such as occur in vicarious menstruation, hemorrhagic diathesis, scurvy, profound icterus, sepsis, etc., but it is questionable if minute ulcers do not actually exist in these cases. Therefore, for practical purposes blood in the stools, if ingested blood can be excluded, signifies a solution of continuity somewhere in the gastro-intestinal mucosa. The location and nature of the break in the mucosa then becomes the question to be solved.

Oesophagus-Bleeding from the oesophagus occurs in the order of its frequency from carcinomata, varices secondary to hepatic cirrhosis, ulcers and traumata. Symptoms and findings of oesophageal stenosis plus occult blood in the stools practically spells neoplasm. Oesophageal varices often ooze blood and give rise to a positive blood reaction in the stools. The presence of occult blood with an atrophic liver or with a liver whose consistency is increased should suggest hepatic cirrhosis and it may be of decided confirmative value in making this diagnosis. Ulcers of the oesophagus are rare, save the cardiac ulcers, which commonly involve the oesophagus and produce characteristic symptoms. The stenosing or ulcerating processes which arise from structures about the oesophagus, such as aneurysms, neoplasms of the trachea, bronchi, lung and mediastinal glands, tuberculous peri-bronchial lymph glands, nearby abscesses occasionally bleed into the oesophagus and give rise to occult blood in the stools. These processes, however, are usually accompanied by symptoms which suggest the morbific factor.

Gastric and Duodenal Ulcers-In all patients with gastrointestinal complaints a routine examination of the stools should be made for occult blood. In this field a blood test of the stools is just as important as a chemical or microscopic examination of the urine in patients complaining of genito-urinary symptoms. This is true even if the complaints are mild and of apparently minor significance. I recently sectioned a case in which a failure to make this examination led to a rather disagreeable fluke. An old man, hard of hearing, confused mentally, anemic, emaciated and visibly all in, entered St. Joseph's hospital. His complaints were almost nil. Diarrohea was

present. An examination of the stomach contents showed a

total acidity of 6 and the diagnosis of achylia gastrica was made. The autopsy revealed a non-palpable ulcerating carcinoma of the lesser curvature. Hemorrhages must have been present aud blood was just the clue that was needed to set the examiner on the right trail. Occult blood does not make the diagnosis, but it presupposes an ulcerative process and with other symptoms proves its existence almost to a certainty. This leaves only the question of location, nature and complications of the ulcer to be solved.

It would seem that there has been too much diagnosing of stomach ulcers on history and tender areas, too much unwarrantable and categorical assumption of the existence of a duodenal ulcer on the so-called "hunger pain," eased by eating, drinking or taking soda and occurring in periodic attacks. We read: If the patient has eaten between 1 and 2 p. m. the pain occurs at 4. Patients are invariably wakened at night at about 2 a. m. There is a feeling of fullness and distress in the epigastrium preceding the pain. The pain occurs earlier on liquid diet and is apparently made worse by such diet, although persistence in a liquid diet will ultimately bring improvement. The attacks are prone to occur in the winter months. There need be no tender areas or objective findings. The history alone suffices to make the diagnosis, etc.

Monyhan of Leeds, England, has dogmatized the above complex as pathognomonic of duodenal ulcer and needs no further encouragement to perform gastro-enterostomy. I have suffered from the above symptoms at intervals for the last 27 years and should hate to have my alimentary canal short circuited on such flimsy grounds. The above symptom complex, however, plus occult blood, when we are justifiably sure that the blood has no other origin, may be considered presumptive of duodenal ulcer. I believe that one is more justified in dogmatizing in an opposite sense. Epigastralgia without occult blood is not ulcer. When one considers the storm center that the epigastrium affords for all kinds of nervous manifestations one cannot too highly appreciate any precise objective test which will help to differentiate functional from organic disease. Blood often disappears rapidly under dietetic restriction and its absence in the absence of symptoms should not be taken to exclude ulcer. Repeated examinations must be insisted upon and if the reaction is continuously absent the diagnosis must be revised. Ewald says (D. M. W., April 6th, 1911): "I have found it (occult blood) almost without exception. One should never be satisfied with a single examination, but in questionable cases the

stools should be examined daily, avoiding the usual sources of error."

Ulcerations Lower in the Tract-The importance of the test in lesions lying beyond the stomach and duodenum is not so great because ulcers here do not admit of as much diagnostic precision. In typhoid the presence of a sharp chemical reaction should suggest an impending hemorrhage; in pulmonary tuberculosis it indicates intestinal involvement with consequent clouding of the prognosis. The nature of tuberculous ulcers is such that they rarely lead to hemorrhages or perforation, but the test finds occasional use here inasmuch as the demonstration of organisms is fallacious on account of the habit of these patients of swallowing their sputum. In gradually developing intestinal obstruction after middle age occult blood, with proper restrictions, points quite definitely to malignancy. The same is largely true of its persistent presence with suspicious tumor masses. Few tumors except carcinoma produce bowel ulcerations. In colonic cases the reaction is valuable in distinguishing a simple catarrh or mucous colitis from a dysentery-a diagnosis which occasionally presents difficulties with, therapeutic indications directly at variance. The reaction is of no importance in diagnosing the nature of ulcerations of the large gut, such as decubitus, gonorrheal, syphilitic, tuberculous, etc. It points out the existence of ulceration and the diagnosis must be made ex juvantibus.

Fallacies. As in the case of all laboratory methods, results must be interpreted with discrimination. Occult blood may be considered as (1) endogenous: when it is derived from the patient's alimentary tract; (2) exogenous when it is ingested; 2-5 c. c., of swallowed blood will give a positive reaction. Therefore meat must be excluded from the diet for several days before a positive reaction may be interpreted as meaning hemorrhage.

In event of a positive reaction the patient should be fed on a meat free diet, the stools marked with 5 grain carmine or charcoal capsules and the dejecta following the appearance of the coloring matter examined.

A positive reaction should never be interpreted as significant of hemorrhage from the stomach or duodenum without a microscopic examination. Blood corpuscles or shadow corpuscles point definitely to a colonic or rectal origin (hemorrhoids). Simple inspection of the feces often determines the low origin of the blood.

A negative reaction may give rise to interpretative fallacies.

When symptoms point to ulceration, repeated examinations should be made before a negative result is finally accepted. Carcinoma of the gastro-intestinal tract must not be excluded on a negative blood test in the absence of symptoms pointing clearly to that diagnosis. A case in mind in which the diagnosis of achylia gastrica was based on a classical history, a three months' period of improvement, a fitting stomach content analysis and the continuous absence of blood in the stools and stomach contents showed a non-ulcerating infiltrating scirrhus carcinoma of the plyorus and pars pylorica at autopsy. The tumor was situated high under the costal arch and liver and could not be palpated.

It is hardly necessary to mention the dangers of contamination from the urine, menses, unclean vessels, dirty test tubes, etc. For instance, we have found it absolutely necessary not to use any tubes that have been used for the Wassermann test or washed in the same water, although such tubes may have. been thoroughly washed and cleaned and though they are to all appearances perfectly clean. In making ethereal extracts sweat from the worker's hands should not enter the extract.

The adventitious presence of certain ingested substances which may give the reaction should be mentioned: Formol, manganese dioxide, iron compounds. Pus will give the guaiac reaction.

Tests. All of the tests depend on the same chemical principle, i. e., a color reaction produced by the oxidization of a given substance. The oxidization is induced by the hematin acting as a catalytic agent in the presence of an excess of oxygen which has been added in the form of H2 02 or ozonized turpentine. The tests used in order of their age are the guaiac, aloin, benzidine and phenolphthalein tests. The hematin is usually obtained by treating the feces with acetic acid and extracting with ether, although the process of extraction is not necessary in the benzidine or phenolphthalein tests. The reagent is made in the guaiac and aloin test by taking a pen knife point of powdered aloin, dissolve in 2-3 c. c., of alcohol, then adding 1-2 c. c., of hydrogen peroxide; a contact test is then made with an ether extract. In the presence of blood a dark bluish or purple color appears in the first and a cherry red color in the latter. The reaction takes place almost immediately and a reaction that has occurred after ten minutes should not be considered positive. These tests will demonstrate blood in about 1:10,000.

The benzidine test is simpler and owes its chief disadvantage to its delicacy. A knife point of C. P. benzidine (Merck)