Georges Dieulafoy.

A text-book of medicine, Volume 1 online

. (page 35 of 129)
Online LibraryGeorges DieulafoyA text-book of medicine, Volume 1 → online text (page 35 of 129)
Font size
QR-code for this ebook

pleura. We have the proof of thLs in the case quoted above.

Landouzy's patient while in good health was seized with apparently
primary tubercular pleurisy ; he died suddenly, and the autopsy revealed a
small focus in the lung. My two patients who died suddenly had both been
taken ill with apparently primary pleurisy. At the autopsy we found a
small focus in the lung, which had given rise to infection of the pleura.

These cases prove that true primary pleuro-tuberculosis, associated with
no pre-existing lesions in the lung, and secondary pleuro-tuberculosis, set up

Digitized by



by a small latent focus in the lung, may both assume the symptoms of so-
called frank pleurisy. In many cases it is not possible to distinguish
them ; clinically, they may show no differences, and cyto-diagnosis includes
them in the same cellular formula.

These two varieties of tubercular pleurisy, however, are not comparable
as regards prognosis ; one is less grave than the other. In the primary form
the lesion in the pleura may recover without producing general infection of
the lung, or of the other organs. If the lung is already affected, although the
lesion is small, the prognosis is not so good, for we have to cure tuberculosis,
both of the pleura and of the lung. The prognosis in acute cases of long
duration is evidently uncertain.

Treatment — I would refer the reader to the preceding section ; I have,
however, some remarks to add. Acute pleuro-tuberculosis is generally
accompanied by much effusion. Perhaps this effusion is a mode of defence ;
perhaps the lung which is compressed by the fluid has less tendency to be
infected from the pleura. If this hypothesis be true, it would be better not
to perform thoracentesis too hastily ; but yet, on the other hand, we know
how dangerous it is to allow too much fluid to accumulate in the pleura :
sudden death may be the consequence, whatever be the theory employed to
explain it. It is therefore necessary to perform thoracentesis in good time.

There is another question which is also of importance. In the case of
acute tubercular pleurisy the fluid may reform rapidly, even when it appears
to have been drained away by thoracentesis. I have found that this rapid
and obstinate reproduction of fluid is much less marked in acute infective
pleurisies that are not tubercular. I have often in tubercular cases had
occasion to draw off 4 or 5 pints of sero-fibrinous fluid by two or three
successive punctures. The effusion seemed to cease for the moment, and the
patient was considered cured ; but yet the fluid reformed without fever,
dyspnoea, or pain, and in a few days amounted to 3 pints or more. The
patient must be kept under observation, even if the acute phase appear to be
ended ; and w6 must not forget that fluid may reform rapidly after puncture,
and cause sudden death, if we be not forewarned.

I have just given my recommendations in the acute phase of pleurisy,
but treatment does not stop there. The patient is convalescent, but the
tubercular lesion lies hidden, though health is apparently regained. What
will happen in this case ? Will it be cured without leaving any sequelae, or
may it not rather be the first stage of tubercular infection, which will later
attack the lung or the other organs ? We know nothing of this, but we do
know that the patient has tuberculosis, and we should place him under the
best therapeutic and hygienic conditions.

An individual who is convalescent from acute pleurisy should for a long
while take care of himself, even though he be considered as cured of active

Digitized by



disease. Tubercular infection lies in wait for him. Years must pass before
he can be considered free from all risk of tuberculosis.

Special attention must be paid to hygiene. The patient should avoid
all causes of over-fatigue. Nourishment should be substantial and varied ;
food and diink which excite the appetite should be chosen. As regards
residence, preference should be given to high altitudes, and life should be
passed in the open air. All kinds of exercise are permissible, provided they
are never carried to excess.

As regards tonic and constitutional remedies, cod-liver oil should be given
in increasing doses — e.g., 2 to 4 ounces daily — if it is well tolerated ; many
patients will swallow a timiblerf ul of cod-liver oil before meals. In order to
render it less disagreeable, it may be cooled by placing the glass in ice.

In patients who do not take cod-liver oil well we should recommend fatty
foods, such as cream or bread and butter. Oysters, caviare, sardines in oil,
tunny fish, smoked fish, and meat should form part of the diet. Raw meat
and meat- juice are of benefit in cases of tuberculosis (Richet and Hericourt,
Josias and Roux). Injections of cacodylate of soda should also be given.


General Considerations. — For many years while I was occupied with the
histological examination of fluid from acute pleurisies, I had seen that my
specimens contained some thousands of red corpuscles per cubic millimetre.

With 1,500, 2,000, and 3,000 red corpuscles per cubic millimetre the
colour of the fluid was not sensibly altered ; the colour only becomes rosy
when the fluid contains 5,000 to 6,000 red corpuscles per cubic millimetre.
I have called these pleurisies histologically haBmorrhagic, in order to
differentiate them from true haemorrhagic pleurisies, which are quite distinct.
Fluid which is very rich in red cells may remain histologically haemorrhagic
without becoming hsemorrhagic in the true sense of the word.

In this section I shall leave out hasmorrhage into the pleura from injury,
and shall only take count of hsemorrhagic pleurisy from the medical aspect.

It is customary to include various morbid conditions under the term
" hsemorrhagic pleurisy." Haemorrhagic effusions into the pleura which
are consecutive to tubercular or to cancerous lesions are the most frequent.
These effusions are sometimes only symptomatic, and develop as a complica-
tion in the course of cancer, or of pleuro-pulmonary tuberculosis ; at other
times they attract attention from the first, and appear as the prodromal a
of hidden tubercular or cancerous lesions. In some cases haemorrhagic
effusions into the pleura appear independent of tuberculosis or cancer ; they
seem to he simple haematomata of the pleura. This simple haematoma,
however, must be extremely rare, and the more I study the question the more

Digitized by



I believe that the hfiematoma is only a benign or curable haemorrhagic
tubercular pleurisy.

We do not see therefore one, but several kinds of haemorrhagic pleurisy.
The fluid is reddish or blackish, and contains fibrin, haematin, red cor-
puscles, and dissolved elements ; the composition depends upon the nature
of the pleurisy and on the abundance and the age of the fluid.

I may say in advance that the haemorrhagic nature of the pleural fluid
usually comes as a surprise ; thoracentesis is performed, and the fluid is
found to be haemorrhagic. It is practically impossible to afllrm before
thoracentesis that pleurisy is haemorrhagic. What are the signs and
symptoms which would lead to such a diagnosis ? In the great majority of
cases haemorrhagic pleurisy is just like the sero-fibrinous form; I see no dis-
tinctive signs between them : the course may in both cases be acute, sub-
acute, or latent. On palpation the same modifications of the vocal fremitus ;
on percussion, the same character of the dullness ; on auscultation, the same
tubular breathing and aegophony, as well as aphonic pectoriloquy which has
been given as a distinctive sign between sero-fibrinous and purulent or
haemorrhagic effusions. I have found aphonic pectoriloquy in riiost of
my cases, and it was very clearly marked in a case of haemorrhagic pleurisy
described by Jaccoud, and hence I repeat the haemorrhagic nature of the
fluid is a surprise. We perform thoracentesis, thinking to draw off sero-
fibrinous fluid from the pleizra, and we are often astonished to find it

Under some conditions haemorrhagic pleurisy may simulate empyema ;
the general symptoms which lead to this error in diagnosis are due to the
tubercular or to the cancerous lesions .which have set up pleurisy. The
patient is feeble, has an earthy colour, and shows oedema of the lower limbs
and of the chest- wall ; thoracentesis is performed with the idea that pus will
result, but here again we are much astonished to withdraw blood-stained
fluid. We make this mistake because we are too accustomed to consider
oedema of the chest- wall as a sign of suppuration ; it is, indeed, a valuable
sign, but it is not limited to purulent effusions. It is also met with in
haemorrhagic and even in some sero-fibrinous effusions.

Haemorrhagic pleurisy may at times be suspected beforehand — e.g., when
the trouble develops in a cancerous patient. Whether the cancer be
primary or secondary, we may prophesy in such a case that the effusion is
perhaps haemorrhagic. I say perhaps, for effusion of cancerous origin is
sero-fibrinous in at least one-third of the cases.

In short, the diagnosis of the haBmorrhagic nature of the fluid rests upon
no certain sign ; its existence may be suspected and reservations made as to
the qualities of an effusion which shows unusual characters, but it is impossible
to afiirm the haemorrhagic nature. After these few remjurks it will be

Digitized by



evident that the study of haemorrhagic pleurisies is surrounded by difficulties ;
therefore, in order to facilitate the description, I shall divide them into four
groups :

First Group. — These cases supervene in the course of hepatic cirrhosis
and of Bright's disease, or appear as a pleural haemorrhage in the course
of scurvy and the eruptive fevers. In this group I shall also place ha)mor-
rhage from the opening of an aortic aneurysm, or from the rupture of an
atheromatous aorta. This group, then, contains the most dissimilar

Second Group. — This group comprises the tubercular pleurLsies. Three
varieties must be distinguished. In the first variety the condition forms
part of an acute tuberculosis ; in the second variety pleurisy supervenes
in the course of ordinary chronic tuberculosis ; in the third variety
ha3morrhagic pleurisy appears as the first sjonptom of tuberculosis : it is the
result of local or primary tuberculosis of the pleura.

Third Group. — To this category belong cancerous pleurisies, whether
the cancer be primary or secondary.

Fourth Group.— Simple haematoma of the pleura forms the fourth group.

First Group.

Description. — The most dissimilar effusions are found in this group.
Does cirrhosis of the liver deserve the place assigned to it in the pathogenesis
of haemorrhagic pleurisy ? I think not. In Moutard-Martin's remark-
able work two cases of ha^morrhagic pleurisy are, in my opinion, wrongly con-
sidered as dependent on cirrhosis of the liver. One of them is taken from
Laennec's famous memoir, in which the lesions of atrophic cirrhosis were
first described. A patient with atrophic cirrhosis had also haemorrhagic
pleurisy on the left side. Laennec, however, did not say that the pleurisy
resulted from the cirrhosis ; I am more inclined to believe that the pleurisy
was tubercular in nature, for at the autopsy " the deep layer of the pleura
contained innumerable greyish tubercles." The other case which has also
been considered as dependent on cirrhosis of the liver may, I think, have
been due to independent lesions of the pleura ; for if we look up the details of
the autopsy we shall agree that it is difficult to admit atrophic cirrhosis in a
liver of " normal size, which showed remarkable friability, and broke up on
pressure with the finger into a pulp."

I do not deny, of course, the haemorrhagic form of pleurisy in the course
of hepatic cirrhosis, for I have seen several cases ; but I think that it is rare as
opposed to the sero-fibrinous form, which is fairly common.

I also regard hemorrhagic pleurisy associated with Bright's disease as
exceptional, though Bright's disease predisposes on the one hand to effusion,
and on the other to haemorrhage.

Digitized by



In pleuro-pulmonary inflammations of infectious origin (influenzal
pleuro-pneumonia, typhoid fever), the fluid is sometimes haemorrhagic.

In the hfiBmorrhagic forms of the eruptive fevers haemorrhagic eflhision is
sometimes met with, but it is here a case of haemorrhage into the pleura
rather than that of an inflammatory condition, properly speaking.

Haemorrhagic effusion may also result from opening of an aortic
aneurysm, or from the rupture of an atheromatous aorta. Several cases
have been published ; the following case is given by Bibail :

A man, thirty-five years of age, suffering from palpitation, breathlessness, and
angim^ pectoris, came into the Beaujon Hospital, mider Gombault. The diagnosis of
aortic aneiirysm, with aortic insufficiency, was made. A month later the patient felt a
sharp pain on the left side. Pleural effusion was recognized, and punctures gave
issue to 12 ounces of bloody fluid on the first, 16 oimces on the second and third occa-
sions. The patient died suddenly from angina pectoris. Post mortem, the left pleura
was found covered by a clot, which was continuous with the clot in an aortic aneurysm.

Second Group.

Description. — This group includes haemorrhagic pleurisies of tuber-
cular nature. I shall divide them into three varieties. The first variety
is associated with acute granular tuberculosis, or with acute tubercular
broncho-pneumonia. The lesions in the pleura and in the lung appear
together ; the general symptoms are usually very marked : fever is acute,
temperature is very high, dyspnoea is severe and continuous, or sometimes
paroxysmal. The estimation of the quantity of fluid is very diflficult, because
the signs of pleurisy are distorted by the subjacent lesions in the lung.

The dyspnoea is sometimes so violent and the quantity of fluid appears
so large that thoracentesis is performed ; 1 or 2 pints of haemorrhagic fluid
are withdrawn, but practically no relief follows, because the dyspnoea, like
all the other symptoms, is due rather to the lung trouble than to the effusion.

The effusion, however, either from its early appearance or its abundance,
sometimes appears to be the chief lesion. The patieiit experiences some relief
after the evacuation of the fluid, and may even ask for a second or a third
operation; but the severity of the general symptoms, the elevation of
the temperature, the persistent or rapid reappearance of dyspnoea after
evacuation of the fluid, the wasting and the signs found on auscultation,
prove that the effusion is associated with acute tuberculosis of the limg and
pleura. The sputum must be examined for bacilli. The prognosis is nearly
always fatal in these forms.

In the second variety haemorrhagic pleurisy is associated with the
ordinary chronic forms of phthisis, and I am surprised that Moutard-Martin
has stated that it never coexists with chronic tuberculosis. I have collected
several cases which, on the contrary, prove that the chronic as well as the
acute forms of tuberculosis may cause haemorrhagic pleurisy.

Digitized by



The pathogenic diagnosis is very simple. The patient presents both the
symptoms of pulmonary tuberculosis and of pleurisy. Fever, pain, and
djrspnoea may be absent ; the mischief may end after one or more punctures,
because it has been only an incident in the course of the tuberculosis, just as
sero-fibrinous pleurisy may be.

In the third variety — and I draw special attention to this point — pleurisy
appears as the initial symptom of tuberculosis. It is the result of primary
tuberculosis of the pleura. Tuberculosis may commence in the pleura, just
as it may in the synovial membrane, testis, prostate, eye, skin, pericardium,
etc. ; remain localized for a lengthy period, and recover without becoming

As we have seen in the section on sero-fibrinous pleurisy, it often
happens that pleurisy is met with in an individual who recovers, but shows
signs of pulmonary tuberculosis some months or years later. In this case
pleurisy, though simple in appearance, was only the result of tuberculosis,
which showed itself by effusion, and then became generalized throughout the

Haemorrhagic pleurisy, therefore, may result from local or from initial
tuberculosis of the pleura ; and just as persons have haemoptysis long
before other signs of tuberculosis, so others have haemorrhagic pleurisy as
the first symptom, and, if I may use the expression, these people ** reject
their haemoptysis into their pleura." These cases may present all the signs
of sero-fibrinous pleurisy, and the haemorrhagic nature of the fluid is only
recognized on pimcture ; thoracentesis is performed once, twice, three, or
four times, the fluid is drawn off, the pleurisy cured, and the case thought
to be one of simple haematoma of the pleura ; but yet signs of pulmonary
tuberculosis appear a few months later, and show the error in diagnosis.

These considerations show that the pathogenic diagnosis of this variety
may be fairly easy or very difficult. It is easy if the patient has signs of
acute or of chronic pulmonary tuberculosis ; if the pleurisy arises during
apparently good health, the diagnosis cannot be settled either by the quality
or by the quantity of the fluid, or by the course of the pleurisy, which may
be acute, subacute, or latent. In such a case the various methods of labora-
tory research given above must be employed.

In a patient under my care for diabetes and haemorrhagic plem'isy, the
lymphocytosis demonstrated the tubercular nature of the pleurisy.

This form of pleurisy, when accompanied by fever, becomes much more
serious, and the gravity arises from the lesions in the lung. Nevertheless,
the condition may recover perfectly after one or several punctures. I
have pubUshed cases, and LercbouUet has quoted others. The patient is
sometimes definitely cured, in which case it is probable that the haemorrhagic
pleurisy was the result of local tuberculosis of the pleura ; at other times

Digitized by



the patient, after recovering from pleurisy, subsequently develops tuber-
culosis in the lung.

Pathological Anatomy. — The lesions show some peculiarities. Some-
times the lesion is found at the same time in the lung, in the pleura, beneath
the pleura, or in the false membranes ; at other times it is limited to the
pleura or the false membranes. The walls of the vessels show coagulation
necrosis, and Kelsch thinks that the haemorrhage is due to this change.
Numerous vessels are obstructed by hyaline thrombi ; the vessel walls are
no longer distinct, and are surrounded by fibroid networks.

The newly-formed membranes are generally stratified and rich in vessels,
friable if young, thick and firm if old. These membranes are composed of
granulation tissue, and of deeper layers that are made up of lymphatic cells,
connective bundles, and fibrous tissue (Malassez).

Third Group.

Description. — Pleurisy in the course of pleuro-piilmonary cancer is
not always hsemorrhagic ; the fluid is sero-fibrinous in at least one-third of
the cases, and this fact must carefully be. borne in mind, for it would be
wrong to reject the hypothesis of cancer because the efiEusion was sero-
fibrinous. The haemorrhagic form alone, however, must now occupy our
attention. Hsemorrhagic pleurisy in cancer may arise quite suddenly, like
acute pleurisy, or have an insidious onset, so that the patient finds some
difficulty in fixing the date. These different varieties are foimd, moreover,
in sero-fibrinous, haBmorrhagic, purulent, tubercular, or cancerous pleurisy.

Prom the clinical point of view I shall divide haemorrhagic pleurisy in
cancer into two varieties.

In the first variety, pleurisy appears in an individual who has obvious
cancer. We find in one patient cancer of the stomach, oesophagus, intes-
tines, rectum, omentum, liver, kidney, bladder, prostate, testis, eye, skin,
or of one of the vertebrae ; in another patient we see cancer of the uterus or
of the breast. Cough, thoracic pain, continuous or paroxysmal dyspnoea
and currant-jelly expectoration, appear during the course of these cancers.
Pleural effusion is then discovered, and thoracentesis gives vent to haemor*
rhagic fluid. In such a case the pathogenic diagnosis is clear — ^viz., secondary
cancer of the lung and of the pleura ; and it may be stated that the haemor-
rhagic pleurisy is of cancerous origin.

In some cases we do not witness the evolution of the cancerous lesions,
but the patient shows traces of a more or less recent scar, resulting from an
operation for epithelioma of the nose or of the lip, for cancer of the breast
or of the testis, or for osteo-sarcoma. Pleurisy then appears and thora-
centesis yields haemorrhagic fluid. The lung and the pleura have evidently
been attacked by secondary cancer.


Digitized by



The pathogenic diagnosis, however, is not always so simple. In the
cases which constitute the second variety, hsemorrhagic pleurisy is not
preceded by otherwise appreciable cancerous lesions. Primary cancer may
aifect the pleura and only give rise to symptoms of pleurisy, which may be
acute or insidious in its onset. The pathogenic diagnosis is sometimes

If haemorrhagic pleurisy, consecutive to mediastino-pulmonary cancer,
were always accompanied by special symptoms, such as dysphagia, aphonia,
oedema of the arm or of the face, and well-marked collateral circulation, which
are so common in tumours of the mediastinum, and if the patient suffering
from pleurisy showed supraclavicular glands, currant- jelly expectoration, and
violent attacks of dyspnoea, which are seen in cancer of the lung, the patho-
genic diagnosis of the pleurisy would be signally simplified; there are
cases in which nothing leads us to suppose the existence of cancer of the
mediastinum or of the lung. There are also cases in which cancer of the
pleura is primary, or associated with early cancer of the lung, which may
pass unnoticed ; the pleural effusion is then the chief feature, and we find
hsemorrhagic pleurisy which presents much difficulty as to its origin.

The following signs and symptoms helped me to make a diagnosis in a
case of haemorrhagic pleurisy, consecutive to primary cancer of the lung, in
a man twenty-two years old :

Pain constitutes an important symptom ; it is frequent, sometimes
sharp, persistent, unlike the " stitch in the side " of common pleurisy. It
may be worse at the base of the thorax, and raldiate to the shoulder, the
arms, and the wrists, so that patients beUeve themselves to be suffering
from rheumatism. Acuteness and radiation of the pain are fairly frequent
symptoms in pleuro-pulmonary cancer. Neuralgia of the brachial plexus
was the chief symptom in one of B6hier's cases. One of Lancereaux's
patients complained of " a sharp pain in the left side of the neck, and in the
shoulder on the same side," and later swelling of the joints of the left arm
supervened. In several of my cases I have noted pains in the joints, so
that I have asked myself whether pseudo-rheumatism may not be one of
the manifestations of cancer. These pains are not found in h»morrhagic
tubercular pleurisy.

Dyspnoea is one of the usual symptoms of cancerous pleurisy. It may
be continual or paroxysmal, and is relieved by thoracentesis, but the relief
is only of short duration. This dyspnoea is found in most of the cases of
cancerous pleurisy, and I have seen it cause terrible agony on three occa-
sions ; it depends chiefly upon the cancerous lesions in the mediastinum
and the lung. Similar dyspnoea is not found in chronic tubercular haemor-
rhagic pleurisy. Some cases of haemorrhagic pleurisy, associated with
acute tuberculosis of the limg and of the pleura, may be accompanied by

Digitized by



acute dyspnoea ; in these cases, however, the fever is high ; this does not
happen in cancerous pleurisy.

Pennanent displacement of the heart is seen in left cancerous pleurisy,
at first on account of the fluid, and later because the growth in the pleura
and the lung may help to cause deviation. Further, the heart sounds are

Online LibraryGeorges DieulafoyA text-book of medicine, Volume 1 → online text (page 35 of 129)