It is important at the outset to stress once more that it is absolutely impossible with 100% accuracy to differentiate between benign and malignant ulcers of the stomach by any means short of laparotomy and pathologic examination of the removed tissue. Many clinical, laboratory, and x-ray features have been described that characterize one or the other type, and the finding of a positive cytology may settle the issue in the individual case. These factors should, of course, be looked for and taken into consideration in the differential diagnosis. In any large case series they will substantially reduce the number of misdiagnosed lesions. In the individual case, however, the chances of mistaking an ulcerating cancer for a benign ulcer are ever present. In view of the grave consequences of such an error, one must bend all efforts toward avoiding this tragedy-even to accepting the possibility of advising surgery for ulcers that may well prove to be benign.

A number of authorities resolve this dilemma quite simply and decisively by insisting on operation in the case of all gastric ulcers. They argue:

1. That it is impossible to be 100% sure about the nature of an individual lesion by any means short of operation.

2. That, in any event, surgery is good treatment for gastric ulcers, benign or malignant, when first seen.

3. That in the long run surgery will be required for many benign gastric ulcers that prove to be intractable or recurrent on a medical regimen.

There are a number of authorities who take an almost diametrically opposite view; i.e., it is only rarely that surgery is required. They argue:

1. That in experienced hands the cases in which an accurate clinical differentiation between benign and malignant ulcer is not possible are extremely few in number.

2. That in any event a therapeutic trial of a strict medical regimen with meticulous follow-up will separate the healers from the nonhealers and reduce the number of patients requiring surgery to a minimum. No harm can come from a carefully controlled medical trial of this sort, they insist.

3. That the morbidity and mortality attending gastric resection, although quite low, still must be taken into consideration. The well-recognized possibility of postoperative physiologic sequelae in the form of impairment in the absorption or utilization of various nutrients and the dumping syndrome must be remembered. In the case of cancer, these possibilities are quite acceptable. They may, however, be a high price to pay for removal of a benign lesion.

Numerous papers supporting each of these viewpoints could be cited and summarized. Suffice it to say, however, that after decades of controversy this problem remains unresolved. An attempt will be made to present a middle-of-the-road viewpoint after first considering some of the features that are said to characterize benign and malignant lesions. The differences between benign and malignant ulcers are given in Table 10.

When the practitioner is confronted by a gastric ulcer of unknown pathology, the following sequence of steps in management may prove helpful:

1. Carry out the major steps to diagnosis previously described:

(a) Gastrointestinal series

(b) Cytology

(c) Gastroscopy (as indicated)

2. Carry out the lesser steps to diagnosis:

(a) Gastric analysis

(b) Guaiactest

(c) Hemoglobin or hematocrit determination

3. If the bulk of evidence when compared with the points in the differential diagnosis clearly favors the diagnosis of a malignant ulcer, immediate exploratory laparotomy should be advised. If tissue proof in the form of positive cytology or a gastroscopic biopsy has been obtained, this recommendation should be all the stronger.

If, however, the bulk of evidence favors the diagnosis of a benign ulcer and the patient is cooperative, then a strict medical regimen may be advised.

4. The therapeutic trial must include the following:

(a) Hospitalization

(b) A strict medical regimen of diet, omission of tobacco and alcohol, antisecretory and antispasmodic agents, antacids, and sedatives

(c) Provision for repeat gastrointestinal series after three weeks and six weeks and periodically thereafter (some insist on gastroscopic follow-up)

5. Surgery must be advised if the lesion at three weeks does not show symptomatic and x-ray evidence of pronounced regression and at six weeks does not show complete disappearance on x-ray examination. 6. Surgery should, of course, be advised for all patients who will not cooperate fully in such a medical program or whose ulcer, even although it be almost certainly benign, proves to be intractable under such a program.

Table 10. Differential Diagnosis Of Gastric Ulcers*

*The remarks herein are all relative statements and represent guidelines only. There are exceptions to every general remark made in this table, which must be considered in conjunction with the text.

7. Surgery should be seriously considered and probably recommended if there is recurrence of the symptoms and x-ray evidence of a gastric ulcer after a period of months or years of quiescence.

8. All patients not subjected to surgery should be followed carefully for a period of at least one year, preferably longer.

Other Sites Of Upper Gastrointestinal Tract

Although the stomach is the major site of cancer in the upper gastrointestinal tract (estimated 1961 incidence-26,000, or 5.1% of all new cancer cases), there are three other sites which also present important cancer problems and occasional opportunities for earlier diagnosis: esophagus, pancreas, and liver and biliary tract (Table 11).

Table 11. Incidence Of Cancer Of Esophagus, Pancreas, Liver And Biliary Tract

Almost always, diagnosis of cancer of these sites depends upon awareness of the possible significance of diagnostic leads obtained in the history or upon physical examination, such as persistent heartburn or dysphagia (esophagus), painless jaundice (head of pancreas or lower biliary tract), or otherwise unexplained upper abdominal mass or discomfort. There are a few etiologic factors suggested by epidemiologic studies that can help in raising suspicion before symptoms and thus serve as a basis for screening programs.

Esophagus

There is evidence that esophageal cancer is more frequent in smokers- particularly cigar and pipe smokers-and in those with a history of chronic high alcohol consumption. Prolonged nutritional deficiences (nonspecific) appear to bear a relationship to esophageal cancer. Patients with Plummer-Vinson syndrome (Sweden) and certain native groups in South Africa have high rates of esophageal cancer.

For such groups and for the patient who presents a history suggesting the possibility of cancer of the esophagus, there are several diagnostic procedures available:

1. X-ray-esophagram

2. Cytology-usually washing or brush at the time of esophagoscopy, but "nylon pledget on a string" is an additional possibility

3. Biopsy of the suspicious area or of a tumor seen on esophagoscopy

Pancreas

The diagnosis of cancer of the pancreas is rarely suggested before the presence of painless jaundice, upper abdominal mass, or midabdominal or mid-back pain. There are no known epidemiologic clues, although the syndrome of migratory thrombophlebitis is considered by many as sufficient indication to search for cancer of the pancreas. Cytologic study of duodenal washings may aid in a preoperative diagnosis by revealing malignant cells from the pancreas. Definitive diagnosis, together with the decision in regard to the possibility of curative or palliative surgery, requires laparotomy.

Liver And Biliary Tract

Admittedly, the deaths attributed to disease of the liver and biliary tract include a substantial number of cancers originating elsewhere, since primary carcinoma of the liver is a relatively rare disease in the United States, and biliary tract cancer is not common. However, these diseases do occur in recognizable settings and forms. Cancer of the gallbladder almost never occurs in the absence of biliary calculi and/or chronic cholecystitis, which are thus listed by some as precancerous conditions. When cancer of the liver occurs, it is typically in association with a long-standing cirrhosis, except in special settings such as in the native population of parts of South Africa, where it is a frequent form of cancer presumably related to nutritional deficiencies and liver impairment dating from early life.

The preoperative diagnosis of cancer of the biliary tract may be aided by the same type of cytologic studies of duodenal washings that are helpful in carcinoma of the pancreas. Again, laparotomy is the key to definitive diagnosis and therapy.