William Osler.

Modern medicine : its theory and practice, in original contributions by American and foreign authors (Volume v. 5) online

. (page 32 of 126)
Online LibraryWilliam OslerModern medicine : its theory and practice, in original contributions by American and foreign authors (Volume v. 5) → online text (page 32 of 126)
Font size
QR-code for this ebook

Boas-Oppler bacilli amid mucus which may plug the introduced tube,
also food remains and heaps of leukocytes. These early signs he attributes
to cancerous invasion of the lesser curvature, hence the rigidity, the lessened
peristalsis, and slight stasis.

Sarcince are uncommon unless free HCl is present; therefore, as a
rule, they are only preseilt when the cancer has formed on an old ulcer.
The presence of sarcinse is no proof that cancer is not present, although if
they are in abundance the diagnostic value is of some importance as sug-
gesting a benign rather than a malignant condition.

A few years ago Cohnheim recorded six cases of cancer with trichomonas
and megastoma entericum. He attaches diagnostic importance to their
presence in early stages before the lactic acid forms. They cannot live
where lactic acid exists, and are hence more likely to be found in cancer of
the cardia and lesser curvature, for in these cases often the lactic acid de-
velops late. Others partially confirm these findings (Heuser, Strube,

Amoeboe andflagellata are found in cases of cancer of the stomach in which
no stasis exists, but in which an alkaline reaction obtains on the gastric
surface, i. e., non-pyloric cases. Cohnheim thinks them an early sign, and
that they are found before the ulcerating tumor has become gangrenous.
Later they disappear.

Yeast cells sometimes occur. Meat fibers are found because of delayed
and lessened proteid digestion. Starchy foods may be abundant, but are
found to be more digested than the meats. Fats are likewise present.

Diagnosis. — Inasmuch as surgical intervention at the commencement of
the disease is essential to a cure, the earliest possible diagnosis is necessary
for rational treatment, and one must endeavor to determine not only the
location of the disease, but the extent of its development, the presence or
absence of complications, and the possibilities of satisfactory treatment.
Where possible, a diagnosis should be established before a tumor is detected,
although from recent surgical statistics this is not essential for a radical
extirpation with cure.

Many tumors, even though palpable, are sujffieiently localized to extirpate,
and even when fairly large, a successful resection of the stomach has been
accomplished. Then, too, many tumors feel larger than they really are,
a,nd vice versa. As regards palpation, it does not follow that because a.


tumor is not yet ]ial])alile that the disease is in an early stage. Hiifje cancers
with adhesions and metastases may liave ah'eady formed and he entirely
covered antl inaecessihle to palpation, and, vice versa, an advanced condition
may remain locaUzed for a long time and grow slowly long after the diagnosis
is made.

The main evidence in the early cases lies in the findings of functional
disorder. There is no single pathognomonic sign of cancer of the stomach,
and one must consider the whole symptom comj)lex, using for his purposes
not only the history, but a careful examination of the symptoms, signs, the
physical examination, the chemical and microscopic examination of gastric
contents and stools, and such investigations recjuire often prolonged, patient,
and thorough observations. In spite of all there are numerous cases in
which it is quite impossible to diagnose. One general rule holds in most
cases, viz., the absence of all gastric symptoms and signs is important evidence
against cancer.

Penzoldt has well said that in individuals over forty years of age with
gastric trouble, one should not rest until one has satisfactorily decided for
or against cancer. Klemperer advocates as a therapeutic test that elderly
patients in doubtful early states should be ]:)laced on a rest cure with a pro-
tective diet for several weeks. If with this method improvement does not
supervene, cancer may be strongly suspected.

General Signs. — When no tumor is evident one must rely on the age,
family, and personal history, the history of the illness, especially its onset
in the midst of good health, the anorexia especially for meats, the hfema-
temesis, weakness, and emaciation in spite of apparently sufficient food,
the motor insufficiency, and the results of chemical analysis and other investi-
gations of the vomitus and contents from test meals, and upon the presence
of persistent occult blood in the freces. In the stomach contents, the cardinal
features are evidences of motor insufficiency, the persistent diminution of
HCl, the presence of lactic acid and of blood.

Tumor. — The possibility of palpating a growth depends much on the
topographical conditions of the stomach and of the tumor, e. g., cancers with
adhesions, with extension upward of the growth, etc., may not be accessible
to palpation. Where tumor is evident, one is called upon to decide (1)
whether or not it be gastric, and (2) if gastric, whether it be of a cancerous
nature or otherwise.

Certain general tests have been recommended as aids in the early diag-
nosis of gastric cancer before a tumor is palpable, although none of these
have justified the claims originally made for them. The tryoptophan reac-
tion is based on the effect of chlorine or bromine to produce a red-violet
color on the proteinochrom which forms as a result of the great albumin
destruction. The test is not constant, having failed in 13 out of 15 cases
(Sigel), and having been found positive in other conditions, e. g., ulcer.
Kuttner's opinion was similar to that of Sigel.

Salomon believed an earlier diagnosis possible by testing the lavage water
for nitrogen and albumin, his test being based on the idea that in gastric
cancer owing to the great destruction of tissue, an albuminous serum is
poured into the stomach. The stomach is first carefully washed on the
evening before testing, after a preliminary non-albuminous fluid diet
for twenty-four hours. On the next morning the stomach is washed
thoroughly with normal saline solution (400 cc), the sa,me fluid being re-


peatedly used and then tested by Esbach's and Kjeldahl's tests. More than
20 mg. of nitrogen to 100 cc. or 0.5 gm. of albumin, suggest the presence of
cancer. In negative cases there should be little or no turbidity with Esbach's
reagent, and the nitrogen by Kjeldahl's method is not greater than 15 mg.
Minkowski is not in accord with these views. Gerster, however, regards
this test as useful in cancer of the lesser curvature without stenosis, unless
the cancer has formed on an old ulcer, in which case the little HCl present
would digest the albumin present. Reicher, Sigel, Tabora, and Zirkelbach
confirm the test as being at all events of some value, especially as indicating
an ulceration of the gastro-intestinal tract, although not necessarily differ-
entiating malignant from benign conditions. Furthermore, it may assist
in diagnosing cancer from chronic gastritis unless the cancer be of an early
diffusely infiltrating character.

Gluzinski's test for the relative diminution of HCl has already been

0. Reissner's contention that the diagnosis of gastric cancer is aided by
the determination of the chlorides in the stomach contents, is worthy of
mention. In carcinoma ventriculi the chlorides are apparently much
increased, despite the fact that HCl is itself deficient, the alkaline cancer
juices evidently themselves containing the chloride excess. The method of
Liitke Martins is employed, and while in other gastric diseases 100 cc. of
gastric contents have a corresponding chloride value of 24 to 40 cc. of ^
silver solution, in carcinoma 50 to 70 cc. y^ silver solutions are required.

DifEerential Diagnosis. — Two main types of condition may be con-
sidered :

A. Gastric cancer without evidence of tumor.

B. Gastric cancer with evidence of tumor.

A. Where no tumor can be detected, one must take cognizance of two
classes of cases:
I. Those with little or no motor insufficiency.
II. Those with definite gastrectasis.
A third class without tumor exists, namely, cancer at the cardiac orifice,
which will be dealt with separately.

1. Where no motor insufficiency exists, the cases are easier to diagnose,
because we can depend on the presence of lactic acid, which is rare in the
other gastric diseases that have no atony. Therefore, if we happen to get
lactic acid early, cancer is probably present. The main diseases in this
group are:

(a) Ulcer without tumor or stenosis. Here the HCl and pepsin are usually
normal or increased. No lactic acid exists, and there are other positive signs
of ulcer, e. g., the age and nutrition of the patient, the character of the pain
and of the vomited matter, the course of the malady, effects of therapy, etc.

{h) Neuroses. — HCl and pepsin are usually normal in amount, altlaough
not always, and thus prolonged observation is necessary sometimes before
deciding, especially in elderly people. Frequent analyses are essential,
and the effects of therapy will materially aid the diagnosis. There are,
moreover, no occult hemorrhages in the gastric contents or stools.

(c) Chronic Gastritis. — The history of a cause (alcohol), the gradual

onset and slower progress, with remissions and exacerbations, are important

features. The secretions are disturbed later in the disease, and the features

are more local than constitutional. Hydrochloric acid is less constantly

VOL. v. — 16


absent and is often only diminished. INlotor insufficiency is later too,
progresses less rapidly, and there is less gastrectasis; there is also less stag-
nation of solids, and occult bleedings are less ])ersistent, if at all present.

(d) Atrophy of the mucous membrane (including pernicious anaemia,
primary atrophy of the gastric follicles, terminal stage of advanced chronic
gastritis, cancer of distant organs, e. g., breast, rectum, uterus, Addison's
disease, etc.).

Primary atrophy runs a protracted course with gradual onset, and the
motor ])o\vcr is unimpaired for a long time. There is no lactic acid, and
little emaciation; amiemia, however, exists in this as in cancer. Advanced
gastriths has a similar series of signs and symptoms, but some exceptions
occur, and then the diagnosis is impossible. There is, however, no cachexia.
Lactic acid is less constant, Salomon's test may be negative, and occult
hemorrhages are far less constant. Achylia gastrica: There is no cachexia
or emaciation, a fair general condition exists with good digestive power.
Lactic acid is absent and there are no occult hemorrhages. Pernicious
anaemia shows little emaciation, as a rule, although sometimes it is extreme.
HCl may be absent in the contents, and sometimes even lactic acid is present.
Even the blood counts may be hard to distinguish, but the cachexia and the
aniemia do not go pari passu. There is no haematemesis, and occult bleedings
are usually absent. Remissions and exacerbations are marked. Addison s
disease resembles cancer because of the general malaise, weakness, dyspeptic
signs, etc. But the pigmentary changes are present as a feature of the mal-
ady. Cancer of distant organs induces no motor insufficiency, and lactic
acid is not present. In many cases, however, HCl is absent, as has been
repeatedly observed in recent years.

II. Cases with Gastrectasis. — With gastrectasis and pyloric stenosis it is
harder to differentiate, because the lactic acid is so often present in gastric
diseases (other than cancer) with atony, and therefore its presence has less

The main diseases to be differentiated are: Scarred ulcer of the pylorus
tvithout palpable tumor (which may lie beneath the liver or be too small
for palpation). The history of previous ulcer, the longer duration of the
malady, the presence of HCl (perhaps even in excess), the absence of lactic
acid, the eructations of gas containing HjS, and the findings of sarcinse,
all suggest the benign condition; and the absence of cachexia aids in excluding
carcinoma. Hypertrophic stenosis of the pylorus (Boas' stenosing gastritis),
if acquired, is usually due to ulcer and gives signs similar to the above.
Sometimes a carcinoma developing on an old ulcer must be differentiated,
and the diagnosis is not always easy; indeed, often it is cjuite impossible. As
a rule, there is a history of old ulcer and the symptoms change in character,
becoming more persistent, and resistant to treatment. There is, moreover,
greater wasting, anaemia, and pain. Bleeding and perforations are not
uncommon. HCl is present, often in excess, and there may be hypersecretion
(continuous or alimentary). Gluzinski's tests are perhaps useful in these
cases; at all events, if positive, they indicate the presence of cancer, and if
negative, they do not prove its absence. Cancer developing on an ulcer
shows a tumor more easily because of the usual perigastritis and pyloric

B. When Tumor Exists. — Two questions arise: (1) Is the growth gastric
or extragastric ? (2) If gastric, is it benign or malignant?


It is often well, as an aid to diagnosis, to palpate the abdomen with the
patient in a warm bath or under ether, unless, indeed, an exploratory laparot-
omy be undertaken. Minkowski suggested inflation of the stomach, and
at the same time distention of the colon with water, for differential diagnosis
of gastric and perigastric tumors. With inflation of the stomach, pyloric
tumors usually move downward and to the right; those of the lesser curvature
disappearing, while those on the anterior wall and lesser curvature together
become broader but less defined. Under similar conditions, tumors of the
liver would move upward and to the right, those of the spleen, intestines, or
omentum would move downward, while if the pancreas were the seat of
the neoplasm, it would entirely disappear as the gas filled the distending
stomach. Renal tumors would be unaffected by the inflation.

Inflation or distention of the colon with water would displace gastric
tumors upward, as also those of the spleen. Tumors of the kidney dis-
appear behind an inflated colon. Tumors of the liver would vary according
to size and position, while those of the omentum would be displaced down-

The consideration as to the location of the neoplasm in or outside of the
stomach has already been dealt with in the discussion of the symptoms and
physical signs.

Growths Located Outside the Stomach. — The tumors simulating gastric
growths, but existing in reality outside the stomach, are mainly:

1. Perigastritis with an exudate and perhaps adhesions about the pylorus,
or the lesser curvature, may induce chronic pain with dyspepsia and even
dilatation, all simulating cancer. The main features against cancer are
the long duration without sufficient corresponding emaciation and cachexia,
and the difPerent chemical signs on analysis. (See Complications of Gastric

2. Cancer of the Duodenum. — This condition is rather rare. Schlesinger
found only 7 primary cases in 25,000 histories. Rolleston could collect only
41 cases in the literature; of these, 8 were in the first portion, 5 in the first and
second parts, 24 in the second portion, and 4 in the third portion of the
duodenum; 10 were in females and 31 in males. Here the difficulties are
great indeed; it may be impossible to diagnose the true condition. One
may have diminution or absence of HCl. Lactic acid, on the other
hand, may be present, as, indeed, may most of the other signs of gastric

3. Omental or general peritoneal tumors often cause difficulty in the differ-
ential diagnosis, especially if accompanied by ascites. Peritoneal cancer from
any origin other than gastric may simulate cancer of the stomach, especially
if in aged people with cachexia, and enlarged inguinal glands, pressure on
the portal vein, and hemorrhagic ascitic fluid. But these neoplasms are less
mobile with respiration, and usually give negative chemical tests on gastric
analysis. The rectum should always be examined in doubtful cases of the

4. Peritoneal tuberculosis with ascites may simulate gastric cancer with
secondary peritoneal involvement. However, it occurs usually in younger
subjects, runs a more protracted course, with exacerbations and remissions,
and is accompanied by more persistent irregular fever. No occult bleedings
occur. Dock has drawn attention to the value of cytological diagnosis of
the fluid with its characteristic cells, in which mitoses are very common.


5. Tumor.t of the transverse colon usually o;ive cvideuce of some degree
of intestinal obstruction; the stools may be bloody. Examination of the
stomach itself by test meals and inflation shows the absence of gastric

6. Tumors of the gall-bladder are situated in their ap])ropriate place, are
somewhat movable, and may by pressure upon the surrounding ducts cause
early jaundice. At times, however, the differential diagnosis is extremely
difficult. While there need be no motor insufficiency, yet adhesions in the
neighborhood of the gall-blailder tumor may cause pyloric stenosis and
certain signs of gastric cancer. The absence of resjnratory mobility may
suggest a gastric cancer rather than one of the gall-bladder; but a test meal,
when possible, will usually reveal the absence of the functional and chemical
signs of gastric cancer. Palpation w ith a sound introduced into the stomach
may aid the diagnosis.

7. Cancer of the liver is usually secondary, and the tumor is less easily
fixed W'ith expiration than are tumors of the stomach. A satisfactory means
of differentiating is, where possible, to place the hand above the tumor.
Then, if the liver can be felt still higher, the tumor is probably gastric;
at all events, it is not hepatic. The same is true if, w'ith inflation of the
stomach, the tumor alters its position. When the cancer of the liver is
primary, the liver becomes large rapidly; jaundice is early and pronounced.
The gastric signs are slight or absent, and there is no lijiematemesis nor
gastrectasis. Moreover, the usual signs after a test meal are not found in
the contents of the stomach.

8. The 'pancreas, either normal or pathological. When normal, it is
usually found deeply situated in the median line, fixed and immovable with
respiration, and disappearing upon inflation of the stomach. When it
is diseased, and especially when it is the seat of a cancerous growth, one
may find clayey, fatty stools, even wdien there is at the same time no jaundice.
Jaundice may, however, be present from pressure of the growth upon the
common duct. The portal vein may be pressed upon, and the resulting
obstruction may give rise to ascites. There may be glycosuria, and, as a
rule, the course is a rapid one. Armstrong recently reported a case in which
a large pancreatic cyst simulated cancer of the stomach, even to the analysis
of the gastric contents. The stools, however, had not been observed.

9. Other conditions which may simulate a gastric tumor are gumma of
the left lobe of the liver, aneurism of the abdominal aorta, sw^ollen glands
about this artery, a movable kidney, and enlarged spleen, the last simulating
a cancer of the fundus.

Tumors within the Stomach. — Gastric tumors arise either from carcinoma
or sarcoma, cicatrized ulcer of the stomach or duodenum, with or without
a perigastritis, thickened or spastic pylorus, fibroma, lipoma, or foreign
bodies (hair ball, gastroliths, etc.).

The differential diagnosis depends upon the history, the symptom complex,
and the summing up of the results of physical examination:

1. In scarred pyloric ulcers, stenosis and gastrectasis occur with the
results common to all cases of pyloric obstruction, but the history of previous
ulcer, the long duration, the presence of a smoother and more movable
tumor, the character of the vomitus with proteids well digested, and chemical
findings typical of benign stenosis, usually render the diagnosis clear. There
are, moreover, no metastases.


2. A hypertrophic stenosis of the pylorus is usually acquired, but occurs,
however, oftenest in early infancy. The cases late in life usually have
diagnostic features common to scarred pyloric ulcers with obstruction
with which, too, they have usually a causal connection. Simple hypertrophy
is undoubtedly rare. Sometimes it occurs with mere hypersecretion. One
relies on the history and duration, the chemical analyses, and absence of
general signs of cancer. Occasionally it may be impossible to differentiate
the two conditions.

3. Spasm of the pylorus may possibly induce gastrectasis, and if so will
manifest few if any of the signs of malignant disease. The condition will
be of long duration, with exacerbations and remissions, and will present
no chemical signs in the gastric contents of malignant disease nor occult
bleedings in the stools.

4. Fibromata and lipomata can only be inferred and not definitely diag-
nosed in the present state of our knowledge.

Regional Diagnosis of Tumors within the Stomach. — Certain differential
features are useful to determine the exact site of the neoplasm in the stomach.

Cancer of the lesser curvature usually presents achylia from the beginning.
There is less motor insufficiency until cancerous infiltration occurs. Peri-
stalsis is absent, but one may get tonic contraction of muscle (gastric rigidity).
When the neoplasm extends to the pylorus, we get pyloric rigidity with insuffi-
ciency, i. e., a paradoxical condition of motor insufficiency with pyloric
insufficiency, which may be tested by inflation. It is usually a late sign.

Cancer of the lesser curvature is palpable only if ptosis exists or if the tumor
has attained a considerable size. The weight is often maintained for a long
time, because there is less motor insufficiency. Salomon's test has special
value in cancer of the lesser curvature without pyloric stenosis, although
it is useless if the cancer has developed on an old ulcer, because small amounts
of HCl will digest the albumin. Estimation of the ferments aids regional
diagnosis sometimes. The fact that the fundus yields pepsin and rennet,
while the pylorus gives merely pepsin, affords an opportunity by ferment
estimates of locating the probable site. If rennet is preserved while the
pepsin is relatively diminished, it implies involvement of the pylorus, while
if both are lessened it signifies a tumor of the fundus.

Cancer of the Cardiac End. — Statistics of the Middlesex Hospital (1854
to 1904) record 227 cases of cancer of the stomach, of which 19 had their
origin at the cardia; 13 were in males and 6 in females. Certainly the condi-
tion is uncommon, and many cases supposedly originating in the cardia have
probably commenced in the oesophagus. Fagge's views expressed in his
reports from Guy's Hospital on the frequency of gastric cancer suggests that
"almost all cases that have been set down as examples of cancer affecting
the cardia have really been instances of cancer of the end of the oesophagus
extending into the adjacent parts of the stomach. . . . Indeed, on
a priori grounds we should expect that a part at which the digestive tube is
opening out into a large cavity should have little or no tendency to be affected
with the disease in comparison with the narrow passage above it."

The type of cancer originating at the cardia is usually adenocarcinomatous.
Sometimes a cardiac cancer with stenosis is accompanied by another separate
and independent cancer of the pylorus, and sometimes, too, cancer of the
cardia is accompanied by a secondary stenosis of the pylorus from pressure
of metastatic periportal glands.


Sijmpioms of Cancer at the Cardia. Not infrequently these remain latent
for a long time and certainly in view of their position beneath the liver and
ribs they are not accessible to jialpation unless they have become extensive
or unless there be an accompanying gastroptosis. An interesting example
under personal observation occurred in a man, aged forty-eight years, whose
death occurred within four weeks of the onset of the symptoms. His
earliest symptoms were those of abdominal distention, swelling of the left
leg, and some weakness. Careful inciuiry elicited the fact that while for
three weeks he had felt general abdominal pain with some anorexia, there
had been no other subjective evidence of gastric disease. At the end of
three weeks (on admission to hospital) there was enormous distention from
ascites, and he died two days later, sudden abdominal pain appearing; the

Online LibraryWilliam OslerModern medicine : its theory and practice, in original contributions by American and foreign authors (Volume v. 5) → online text (page 32 of 126)