William Osler.

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

. (page 73 of 126)
Online LibraryWilliam OslerModern medicine : its theory and practice, in original contributions by American and foreign authors (Volume v. 5) → online text (page 73 of 126)
Font size
QR-code for this ebook

nearly always regular, quickened, and becomes progressively more rapid,
being 100 to 120 in the early stages, and becoming running, 170, or uncount-
able, toward the end. In some instances the rate of the pulse may be little
if at all increased, although there is extensive purulent peritonitis; this has
been ascribed to interference with toxic absorption from blocking of the
stomata and lymphatics, but is not necessarily a sign that the patient is
improving. There is no correspondence between the rate of the pulse and
the temperature; the temperature may be low and the pulse extremely rapid.
The pulse is small and hard in the early stages, and is spoken of as wiry;
Crile,^ who found the average blood pressure in 20 cases to be 166 mm. of
mercury, ascribes the small volume to the accumulation of blood in the
splanchnic area and the hardness to increased blood pressure induced by
reflex stimulation of the vasomotor centres from the peritoneum. In col-
lapse and in the later stages the blood pressure falls and the pulse becomes
thready and eventually perhaps imperceptible.

The temperature in peritonitis is so extremely variable as to be of little
or no value in diagnosis or as to the degree and gravity of the peritonitis.
In peritonitis there is usually a more considerable difference between the
rectal and the surface temperatures, such as the axillary, than in other dis-
eases; this is connected with the stagnation of blood in the splanchnic area
and the correspondingly small amount of blood in the cutaneous vessels.

In cases of perforative peritonitis the initial shock nearly always depresses
the temperature; this is most noticeable in typhoid fever, in which the writer
has seen a fall of eight degrees. But this is not an absolute rule, the varia-
tion probably depends on how much of the peritoneum is inundated by the
contents of the bowel. It is possible that the form of infection may have
something to do with the temperature, and that a particularly virulent microbic
invasion may so paralyze the power of resistance that there is less febrile
reaction than in a less severe form of infection. In the vast majority of the
cases of diffuse peritonitis the temperature is raised in some period of its
course; it may be raised continuously, usually not to any. great extent, and
very seldom above 104°. A relatively high temperature, especially in the
early stages, is more likely to occur in healthy and vigorous individuals.
A persistently low temperature may depend on the severity of the infection
and on failure of the organism to react, and is therefore a grave sign.

Blood. — There is anaemia, but the erythrocyte count is not very greatly
diminished; probably the concentration of the blood accounts for this. In
16 cases examined by Da Costa, the average was 3,970,000 reds, and the
color-index 0.78. In fulminating cases in which the resisting powers of
the body are paralyzed by the intensity of the toxaemia, leukocytosis is
absent and there may be leukopenia. In less severe peritonitis there is a
leukocytosis which in some instances may be extremely marked. It is of
the ordinary polymorphonuclear neutrophile form. According to Locke^
and Gulland^ the glycogenic reaction seen in the polymorphonuclear neu-

^ Blood Pressure in Surgery, Philadelphia, 1903.

^ Boston Medical and Surgical Journal, 1902, cxlvii, 287.

^ British Medical Journal, 1904, i, 880.


trophiles is much more trustwortliy, for it is present in cases of peritonitis
with such a "brutal" infection that leukocytosis does not appear. Cabot' lays
stress on the importance of the increase of the fibrin network (hyperinosis)
in the diagnosis from mechanical obstruction, for this increase does not occur
in obstruction, while leukocytosis does.

Blood Cultures. — Although the fatal results of peritonitis are usually con-
sidered to depend upon septiciiemia as well as on tox?emia, and although
organisms can often be found after death in the heart's blood, it does not
ap]>ear that the presence of organisms in the blood during life has been
proven. In this connection it may be pointed out that in human septiciemia
pyogenetic cocci are demonstrated with difficulty in the blood, thus contrast-
ing with experimental peritoneal infection and septiciemia in animals.
Libman^ found that in 25 cases of peritonitis, mainly of intestinal origin,
the blood was sterile, although in some instances the blood cultures were
made shortly before death.

The urine is scanty, high colored, and may contain a small quantity of
albumin, diacetic acid, and acetone. The most important point, however,
is the large quantity of indican, which Nothnagel regards as practically
constant and as present in larger amounts than in any other disease except
intestinal obstruction. The normal amount in adults is from 5 to 20 mg.
in the twenty-four hours, but in pathological conditions it may be increased
up to 150 mg. Indicanuria takes some twelve to twenty-four hours to
appear after the onset of perforative peritonitis.

According to Lennander, the urine contains the organisms responsible for
the peritonitis. Their presence, especially as he states they may appear
within tw^enty-four hours of the onset of the disease, may possibly in the future
be of some value in directing the treatment. Unfortunately, as regards
gonococcal peritonitis, the presence of gonococci in the urine could hardly
be regarded as an infallible guide to the nature of the peritoneal infection,
as there might be a mixed infection.

From the presence of peritonitis over the bladder micturition is painful,
and usually urine is retained, as if from paralysis of that viscus resembling
that of the intestines. Occasionally, on the other hand, micturition is more
frequent than normal.

The mental faculties are often clear to the end, but sometimes they are
somew^hat obscured by the progressive toxaemia. In cases of peritonitis
supervening in the course of typhoid fever the mental obscuration due to the
primary disease may be expected. Sleeplessness is necessarily the rule,
headache is common, and great anxiety and a painful form of restlessness
are often present.

Death may come quite suddenly from cardiac failure brought about by
the action of toxins on the heart muscle, and may follow almost immediately
after copious vomiting. Death sometimes comes gradually from progressive

The duration is very variable and depends on the cause and on the nature
of the infection, the previous state of the patient, young and vigorous, or
old and the subject of some exhausting disease, and on the treatment.
Thus, in the most virulent and fulminating forms, as in some cases of

^ Examination of the Blood, 1904, p. 282.

^ Johns Hopkins Hospital Bulletin, 1906, xvii, 221.


appendicular and puerperal infection, death may occur in thirty-six to
forty-eight hours from the onset, whereas the more usual and less severe
forms generally last four, five, or even more days. A previously healthy
boy may live through several days with severe peritonitis, while a patient
with carcinoma of the stomach may only survive the onset of perforative
peritonitis for eighteen to twenty-four hours. In pneumococcal peritonitis
the duration is very considerably prolonged, and, as pointed out elsewhere,
the peritonitis may become localized into residual abscesses, so that the
symptoms abate without disappearing. In the progressive fibrinopuru-
lent peritonitis, described by Mikulicz, which constitutes a transition from
circumscribed to diffuse peritonitis, the disease may go on for weeks.

Pneumococcal Peritonitis. — This is a rare form, but presents some rather
special features which necessitate a separate description. In 140 cases of
peritonitis examined bacteriologically, Netter found this form only twice.
Jensen^ has collected 106 cases. The disease is much commoner in children
than in adults. It may occur very early in life; Gossage^ reported 4 fatal
cases of primary pneumococcal peritonitis in infants under three months,
and Dudgeon and Sargent report its occurrence in a male infant aged seven
weeks. Under the age of fifteen years it occurs more often in girls than in
boys, according to Annand and Bowen,^ who have collected 91 cases, in the
proportion of 3 to 1, but after that age its incidence is equal in the two sexes.

Pneumococcic peritonitis may be primary, secondary, or, as Bowen and
Annand point out, it may be difficult from the rapid spread of the infection
to decide whether it is primary or secondary. The following channels have
been regarded by various authors as enabling the pneumococci to reach
the peritoneum: (1) Through the blood stream, the organisms being
absorbed from foci of pneumococcal infection in the thorax or possibly from
the throat or ear; (2) from the stomach (pneumococcal gastritis), intestines
in inflammatory conditions, appendicitis, or foci in the abdominal viscera;
(3) from the pleurse through the diaphragm; and (4) through the Fallopian
tubes. In favor of the last path, the greater frequency of the process in
the lower half of the abdomen and its preference for young girls were at one
time urged. Against this hypothesis it has been argued that there is generally
no evidence of vulvovaginitis clinically or at autopsy. Bond's* observations
on ascending mucous currents in the uterus and Fallopian tubes, however,
make it probable that the pneumococcus may be rapidly carried up the
genital passages in a few hours, so that there is not sufficient time for local
inflammation to become evident. Probably infection of the peritoneum
occurs by different channels in different cases. There may be associated
pneumococcal lesions elsewhere, such as pneumonia, empyema, broncho-
pneumonia, pericarditis, otitis, but in the majority of instances the peri-
toneum is the first part attacked. Diffuse peritonitis is rare in pneumonia;
in 4454 cases^ of lobar pneumonia in patients over ten years of age in the
hospitals of London there were 11 fatal cases, or 0.25 per cent. Pneumo-
coccal infection of several serous membranes, the pleura, pericardium, and
peritoneum, pneumococcal polyorrhymenitis, may occm-, especially in young

1 Arch. f. klin. Chir., 1903, xx, 91.

^ Proceedings of the Royal Society of Medicine, Medical Section, London, 1908, i, 64.
3 Lancet,^ 1906, i, 1591. * British Medical Journal, 1906, ii, 1689.

^ Statistics of Hadley, Pasteur, Fawcett, Owen, Gossage, Proceedings of the Royal
Society of Medicine, Medical Section, London, 1908, i, 61 et seq.
voL.v. — 34


children. In most cases the peritoneum has been previously healthy, and
there is no obvious reason why it has been picked out for infection; in a few
instances there has been ascites due to cirrhosis or to chronic nephritis
(Sevestre and Aubertin'). Pncumococcic peritonitis was i)rimary, that is,
the chief or only focus of pneumococ-cal infection in 47 out of 74 cases, and in
34 of these 47 cases the peritonitis was encysted (Lciiormant and Lecene^).

In about half the cases the peritonitis is encysted or localized; in the other
half it is diffuse or widcsprcatl. A localized abscess may either be primary
or a result of diffuse pneumococcal peritonitis. The exudate is like that
of a pneumococcal empyema; it is highly fibrinous and prone to coagulate,
yellowish green in color, and odorless. The character of the exudation
renders thorough drainage of the peritoneum difficult. In some instances,
however, a much more fluid exudate has been foimd. When the peritonitis
is localized — the abdominal empyema of Lennander — an abscess results
which is usually below the umbilicus, and is particularly prone to point at
the umbilicus.

Clinically the disease has three stages: (1) Without any prodromal
symptoms there is an acute onset with vomiting, followed l)y signs of peri-
tonitis. As the inffaramation is often mainly in or near the pelvis, appen-
dicitis is not uncommonly diagnosed. (2) After a few days the symptoms
subside, the temperature falls, but diarrhoea is usually present, and for this
reason typhoid fever may be suspected. (3) A prolonged stage in which
residual abscesses may form, usually in the upper part of the abdomen,
giving rise to continued fever and imitating tuberculous peritonitis. As
bearing on the resemblance of the manifestations to those of pneumonia,
it may be mentioned that herpes labialis has been noted in some cases and
that the temperature occasionally fails by crisis.

It is one of the less severe forms of peritonitis; the prog)io.sii in primary
encysted pneumococcal peritonitis is fairly good, but it is less favorable
in the diffuse form; thus, Bowen and Annand found that recovery occurred
in 86 per cent, of the encysted and in only 14 per cent, of the diffuse form.

Gonococcal Peritonitis. — As the result of comparatively recent work,
especially that of Gushing^ and Hunner,* it is clear that the gonococcus
alone, i. e., in pure culture, is capable of setting up diffuse peritonitis. The
peritonitis is not, however, of a severe character, and it has been suggested
that this is because the gonococcus flourishes best at from 91° to 98° F.,
or about the normal temperature of the body, and that a higher temperature,
such as would be met with in the peritoneal cavity, inhibits its activity
(Hale White^). When the normal resistance of the peritoneum is diminished
by menstruation or by cold, gonococci in the Fallopian tubes have a better
chance of infecting the peritoneum. Gonococci enter the peritoneum from
the ostia of the Fallopian tubes, but gonococcal peritonitis has followed
leakage from a pyosalpinx during removal by operation. In males infection
is said to travel up the lymphatics of the spermatic cord.

The disease is very rare in men and its frequency in women has not yet

' Bull, et mem. Soc. med. d. hop., Paris, 1906, xxiii, 215.
' Rev. de gijn. et de chir. abdom,., Paris, 1905, ix, 225.

^ Johns Hopkins Hospital Bulletin, 1899, x, 75. (This article contains the detailed
historv of gonococcal peritonitis.)
* Ibid., 1902, xiii, 247.
' System of Medicine (Allbutt and Rolleston), 1905, i, 855.


been established by extensive statistics; but diffuse gonococcal peritonitis
is not common. In 1907 Goodman^ collected 75 cases, but only 30 of these
had been established by bacteriological examination at autopsy or operation.
It may attack girls under puberty (Northrup,^ Comby^), and it is highly
probable that it is often overlooked or not recognized. It usually follows
or precedes menstruation, and may appear shortly after delivery. The
anatomical lesions, as pointed out by Gushing, are comparatively slight;
there is a dry, fibrinous exudate without pus or serum, and the peritoneum
is uniformly injected in the diffuse cases.

Clinically, gonococcal peritonitis comes on suddenly with great severity,
which usually becomes greatly mitigated in twenty-four or forty-eight hours.
Gomby speaks of the condition as one of peritonism rather than of peri-
tonitis. It appears to be more fatal in children than in adults; the mortality
in children has been estimated at 20 per cent. It may imitate appendicitis,
as it mainly affects the lower part of the abdomen. Abdominal distention
is usually not marked, and muscular rigidity is absent. The existence of
gonococcal peritonitis should be suspected in the presence of gonorrhoea
when the peritonitis is mainly in the lower part of the abdomen. At the
onset of the disease the gonorrhoeal discharge may be slight, and as its
existence is often firmly denied careful examination is necessary to avoid
overlooking it. It is usually agreed that the patients should not be operated
upon, but in 20 cases in which laparotomy was performed there were four
deaths only (Goodman). The difficulty is to decide in a given case of a
woman with a vaginal discharge containing gonococci whether the peritonitis
is due to a pure infection or to a mixed infection, such as may result from
rupture of a pyosalpinx.

Puerperal Peritonitis. — The infection gains entry through tears or wounds
of the genital passages, and is much commoner in primiparee than in mul-
tiparse. It is not uncommon in criminal abortion. The infection spreads
from the genital organs by the lymphatics, and may be localized or wide-
spread in the peritoneal sac. The organism is most commonly Strepto-
coccus pyogenes, but staphylococci or the gonococcus may be the causal
agent. The exudation may be purulent, sanious, or putrid.

Clinically, puerperal peritonitis is characterized by the following features :
there is grave intoxication, usually diarrhoea; vomiting is less prominent
than in other forms; meteorism is marked, partly no doubt from the weak-
ened and relaxed condition of the abdominal walls; and a high mortality,
which may be correlated with the presence of Streptococcus pyogenes. At
its onset the lochia may be noticed to be offensive and the mammary secre-
tions to be suppressed. Gomplications are prone to occur, and include
septicaemia, pyaemia, empyema, pericarditis, arthritis, and phlebothrom-

Diagnosis. — In characteristic cases the association of the prominent
symptoms of acute pain, tenderness, distention, and immobility of the abdo-
men, vomiting, constipation, prostration and collapse, together with the
history, make the existence of acute peritonitis clear. But in many cases the

^ Annals of Surgery, 1907, xlvi, 111.

* Archives Pediatrics, New York, 1903, 910, and Transactions of Association of
Am.erican Physicians, 1903, xviii, 203.
^ Arch. mat. d. enfants, 1901, iv, 513.


decision is far from easy; when peritonitis supervenes late in the course of
chronic nephritis it may he almost latent, when it occurs in typhoid fever
it may be diffieult to distino-uish from tympanites, and in acute intestinal
obstruction its manifestations are com))ined with and obscured by those of
the primary disease. Further, the symptoms may be almost entirely masked
by opiimi or morphine. In some cases the majority of the signs and symp-
toms may be absent, although there is advanced peritonitis. This masked
or latent form of peritonitis may occur in patients in the last stages of ex-
hausting diseases, such as cancer. After operations on the abdomen the
only evidence of peritonitis may be vomiting and a rapid jiulse.

Diagnosis of Acute Peritonitis from Conditions Simulating It. — In
hysterical subjects a condition of peritonism may occur and give rise to
considerable anxiety, as tlae diagnosis between it and genuine peritonitis
occurring in an hysterical patient may be difficult. The history of the
patient and any concomitant nervous symptoms must be taken into account.
As an important point in the history, the occurrence of exactly similar
attacks in the past should be specially noted. In such cases there is
cutaneous hyperresthesia, deep tenderness is not more marked than the
superficial hypersesthesia, and may disappear if the patient's attention be

Severe colic, wdiether due to lead, constipation, or the passage of a gall-
stone or renal calculus, may to some extent imitate acute peritonitis, since
the exacerbations of pain in peritonitis are probably of the same nature.
In colic there is, as a rule, no real tenderness, and the pa in may be relieved
by pressure. In severe lead colic there may, however, be tenderness, but,
as T. Oliver^ has pointed out, the pupils are often unequal and there may be
tenderness over the vagus in the neck. In a doubtful case with abdominal
tenderness the possibility of lead poisoning should always be considered.
In colic due to the passage of a calculus vomiting occurs, but it is rare in
the other forms. The history of the case and examination for evidence of
lead poisoning, renal, or hepatic disease should be carefully considered.
A raised tempei'ature at the onset is strongly in favor of peritonitis and against
ordinary colic. Colic, however, is more likely to be confused with the
causes of peritonitis — such as appendicitis — than with the disease itself.
The rare and puzzling condition of enteralgia may also imitate peritonitis;
here again the history of repeated similar attacks is of importance. Ex- (_^^
tremely acute enterocolitis is accompanied by the same collapse as in peri-
tonitis, but there is marked diarrhoea and the pain is of a colicky character.

Tympanites should be readily distinguished from acute peritonitis by the
absence of tenderness and vomitina;.

In acute mechanical intestinal obstruction acute peritonitis supervenes
after some three or four days, and, conversely, acute peritonitis produces
intestinal paralysis and the symptoms of obstruction, so that the diagnosis
as to the primary lesion may be difficult. In acute obstruction the tempera-
ture is not raised, and the abdomen is not tender or rigid, before the onset
of secondary peritonitis. In mechanical obstruction fseces and wind are
not passed, while in peritonitis an enema will often bring some away. In
mechanical obstruction vomiting is more profuse, the pain is colicky and not
continuous, as in peritonitis.

' System of Medicine (AUbutt and Rolleston), 1906, ii, part i, 1045.


Extensive hemorrhage into the pcriton(!um, usually due to rupture of
an ectopic gestation, may imitate perforative peritonitis very closely; in fact,
the manifestations in both instances are thf)se of shock. A similar clinical
condition of peritonism may follow hemorrhage into tlu; m{;sentery, torsion
of an undescended testis, of an ovarian cyst, of a floating kidney, or of a
wandering spleen. In a woman likely to be pregnant this train of symptoms,
especially the ansemia progressing until it becomes extreme, should suggest
the possibility of rupture of an ectopic gestation, while the presence of an
ovarian tumor should arouse a suspicion of torsion. The distinction, how-
ever, is hardly essential, for laparotomy is the proper treatment in either case.
Rupture of an abdominal aneurism and embolism of the superior mesenteric
artery have been known to imitate acute peritonitis (Osier).

Acute hemorrhagic pancreatitis may imitate acute perforative peritonitis;
great collapse, abdominal distention, complete constipation, and vomiting
follow the sudden onset of pain in the epigastric region. If the abdomen
be opened, turbid fluid and fat necrosis may be found. It may be noted,
however, that fat necrosis may be due to the escape of pancreatic fluid through
the perforation of a duodenal ulcer,^ so that its discovery, although almost
pathognomonic of pancreatitis, does not absolutely exclude perforation.
In pancreatitis there is extreme feebleness of the pulse, the temperature
is often not raised, and the symptoms are more suggestive of mechanical
obstruction than of peritonitis.

Acute inflammatory conditions in the thorax, such as lobar pneumonia
or pleurisy, may, especially at the onset before physical signs have appeared,
imitate peritonitis. The sharp pain of pleurisy, the onset of which may
wake a patient up in the night, has been mistaken for perforation of a
gastric ulcer or of a gangrenous appendix. Difficulty is especially likely
to arise in children in whom the appearance of the physical signs of pneu-
monia, particularly when apical, is often much delayed. Thus pain referred
to the right iliac fossa may imitate acute appendicitis, and the' fever, consti-
pation, and abdominal distention suggest peritonitis. It is, therefore, most
important to examine the chest, to note the respiration rate and temperature,
which are higher and more persistent in pneumonia than in peritonitis, and
to inquire for cough. In pneumonia the respirations may be jerky and
grunting. The rigidity of the abdominal muscles sometimes seen in pneu-
monia, especially in children, will be found to relax between the respirations,
and may even disappear on prolonged and deep pressure (Clogg^), while
the referred pain of thoracic disease is mainly superficial and not increased
on deep pressure, as in peritonitis. The temperature may be of some
assistance; low at the onset of perforative peritonitis, it then rises, but may
fall after some hours, although the patient's condition is not improved, it
thus dift'ers from the continuously raised temperature of pneumonia. Indican
in the urine is in favor of peritonitis. Tuberculous peritonitis in rare cases
comes on with such acute symptoms in children as to imitate very closely
peritonitis due to appendicitis ; this has been thought to be due to thrombosis

Online LibraryWilliam OslerModern medicine : its theory and practice, in original contributions by American and foreign authors (Volume v. 5) → online text (page 73 of 126)