Chapter XI: Liver Abscess

HISTORY AND GEOGRAPHICAL DISTRIBUTION

=History.=—Although Hippocrates noted the method of evacuating abscess of the liver by caustics there was very little known about the condition until during the last century.

The history in connection with the finding of amoebae in liver abscesses is of very recent date (1887) and is taken up under the history of amoebic dysentery.

=Geographical Distribution.=—Concisely one may state that the distribution of liver abscess is in relation to the existence of amoebic dysentery. It is particularly prevalent in those centers of amoebic infection where there are many white men having little knowledge of the conditions necessary for the maintenance of health in the tropics.

In liver abscess, as with blackwater fever, it is education rather
than acclimatization that brings about a diminution of these
tropical diseases.

For several years subsequent to the American occupation of the
Philippines amoebic dysentery and liver abscess were common but in
more recent years liver abscess has become rare in Americans and
amoebic dysentery much reduced in prevalence.

More temperate living results in less storing up of fat in the
liver and an organ more resistant to infection.

ETIOLOGY AND EPIDEMIOLOGY

=Etiology.=—The dislodgment of amoebae-containing material from amoebic intestinal ulcerations and the plugging of the portal capillaries by such emboli give us the starting point of a liver abscess. The exciting cause is _Entamoeba histolytica_ which in the liver continues the same production of a gelatinous necrosis as is carried on in the submucosa of the large intestine or appendix.

This pathogenic amoeba is fully described under amoebic dysentery.

As to obtaining a history of amoebic dysentery in liver abscess
cases we have the following statistics:

500 cases with dysentery findings in 60% (Kartulis).
444 cases with dysentery findings in 59% (Zancarol).
500 cases with dysentery findings in 85% (Kelsch and Kiener).
63 cases with dysentery findings in 90.5% (Rogers).
38 cases with dysentery findings in 85% (Seamen’s hospital
autopsies).

Amoebic liver abscess is exceedingly rare among children and
probably 10 times less common among women than men.

Of 40 cases of liver abscess Waring noted intemperance in 67.5% and authorities generally insist upon the importance of the abuse of alcohol as a predisposing factor.

Natives of India very rarely develop liver abscess but it has been
noted that when they begin to follow the customs of Europeans, as
to eating and drinking, such lesions become more common in them.

As to the proportion of cases of amoebic dysentery which give
rise to liver abscess only the statistics of those who have
differentiated between bacillary and amoebic dysentery are of any
value. Such statistics would indicate that about 20% of the cases
of amoebic dysentery are complicated by liver abscess.

Liver abscess may be present without demonstrable lesions in the
large intestines, such lesions having healed or the intestinal
involvement having been so slight as not to have caused other than
microscopic changes.

It is a well-known fact that liver abscess may set in years after a
patient has left the tropics and years after the occurrence of any
dysenteric manifestations.

PATHOLOGY

There seems little doubt but that the amoebae in the thrombosed terminals of the portal vein are carried by way of the upward current into the liver where they lodge in the liver capillaries, Councilman and Lafleur having found amoebae in such emboli.

Another view is that the amoebae may wander across the abdominal
cavity and enter the liver in this way. This seems as improbable as
that view which considers a possible entrance by way of the bile
duct. Bile is toxic to amoebae and it would be difficult to explain
their presence in the small intestines.

In 639 cases Roux found the abscess in the right lobe in 70% of the
cases.

Other statistics give about 75% for the right lobe, 10% for the
left lobe, 4% for the lobus Spigelii and in about 10% of cases
abscesses are found in both right and left lobes.

In 562 cases Zancarol found a single abscess in 60% of the cases.

In 288 cases Waring found a single abscess in 61.5%, double abscesses in 11.5% and multiple abscesses in 27%. The favorite site of liver abscess is the superior and posterior part of the right lobe and near its surface.

The abscesses vary enormously in size, some being no larger
than a walnut while others may contain a quart or more of pus,
exceptionally as much as a gallon. The pus is typically of a
chocolate color and contains degenerated liver cells, granular
débris and often haematoidin and Charcot-Leyden crystals. There is
an absence of polymorphonuclears. It may however be creamy in color.

In Strong’s cases about 50% of the abscesses showed bacteria upon
culturing, the organisms noted being staphylococci, streptococci,
_B. coli_ and _B. pyocyaneus_.

The walls of liver abscesses are rather shaggy and the amoebae are
found deeply located.

It is probable that the necrotic process, set up by the amoebae,
begins in the interlobular capillaries although it may at times
begin within the lobule.

Microscopically, the necrotic abscess wall shows amoebae in its depths but necrosis of the surrounding tissue beyond the zone of the amoebae is noticeable which would suggest the elimination by the amoebae of some toxic substance. There is an absence of polymorphonuclear infiltration around the abscess.

Surrounding the abscess wall there is a zone of marked hyperaemia.
Amoebae may be found in this area as well as in the abscess wall.

If the liver abscess is not treated by emetine or with this drug
and some surgical procedure the tendency is for rupture to occur
and Cyr’s statistics show that of 159 cases rupture occurred
as follows: lungs 59, pleural cavity 31, peritoneal cavity 39,
intestines 8, stomach 8, vena cava 3, kidneys 2, bile ducts 4,
pericardium 1 and externally 2.

SYMPTOMATOLOGY

Although the statistics would indicate that a history of amoebic dysentery has been obtained in only from 60 to 90% of cases of liver abscess, yet, when we consider that amoebic lesions of the large intestines have been frequently noted at autopsy in those who had never shown symptoms of dysentery during life, we are forced to believe that amoebic lesions of the appendix or large intestines are necessary factors in the production of liver abscess. Consequently, a history of amoebic dysentery is one of the most important points to consider in the making of a diagnosis of tropical liver abscess.

=Tropical Liver.=—There is also much evidence to be obtained from
statistics and otherwise to support the view that the amoebic
infection of the liver is only possible in a person whose liver
has been functionally impaired. To this condition the designation
tropical congestion of the liver or simply tropical liver has been
applied. There is much to support the view that, in the tropics,
the intestines and liver take the place of the thoracic organs
in being subject to congestion. In temperate climates excesses
and exposure to debilitating influences result in coryza or
pneumonia. In the tropics we have diarrhoea and congestion of the
liver. Tropical liver is recognized by vague digestive troubles,
high-colored urine, loss of energy, irritability, with a sensation
of fullness in the region of the liver which is generally described
by the patient’s statement that he feels his liver. There may be
pain referred to the right shoulder and the liver may be tender on
palpation.

By the discontinuance of alcohol and highly spiced foods, with treatment by phosphate of soda or sodium sulphate, together with general care of the health, the patient may recover completely.

Rogers recognizes a condition which he terms the _pre-suppurative
stage of amoebic hepatitis_ in which the amoebae from dysenteric
lesions have lodged in the portal terminals of the liver but in
which abscess formation has not taken place.

At this stage we have a leucocytosis in which the
polymorphonuclears are but little increased in percentage with a
low remittent fever. At this time Rogers considers that the disease
may be cured by emetine or ipecac and liver abscess avoided.

=A Typical Case of Liver Abscess.=—Following a case of amoebic dysentery, during the period of convalescence or subsequently, a rather irregular type of fever is noted, which shows an evening rise with sweatings which tend to become colliquative. From a marked feeling of weight in the region of the liver there may later develop tenderness or pain upon palpation of the liver. Of importance is the fact that there is no associated splenic enlargement. In the majority of cases the right side of the liver enlarges in an upward direction. A tape-measure will often show enlargement of the right side. Pain referred to the right shoulder is often complained of when the abscess is located in the upper convex part of the liver but, when nearer the inferior concave surface, there may be pain referred to the region of the appendix. When located in the left lobe the symptoms may be considered as of gastric origin.

The upward enlargement of the liver as shown by X-rays is of great value in diagnosis, but an abscess located in the center of the liver is not indicated by such rays.

There is a marked tendency to splint the liver so that the patient tends to lie towards the right side and when walking applies his right arm and forearm to his side, which led Koch to remark, “It is as if he carried his abscess under his arm.” The right rectus often shows rigidity.

Auscultation of the base of the right lung reveals a moist crepitation which, together with a dry cough (tussis hepatica), the fever, evening sweats, anaemia and emaciation, may suggest tuberculosis. The respirations are shallow as deep inspiration tends to cause pain. It must be remembered that cases of liver abscess have been reported where there were practically no symptoms.

Insomnia is a marked feature in many cases. Jaundice is rare,
but an earthy color or subicteroid tinging is often noted. The
superficial veins may be enlarged.

The urine is scanty and high-coloured, frequently with a marked
increase in the ammonia nitrogen. Urobilin may be present in
considerable amount.

There is a rather constant but low leucocytosis of from 12 to 20 thousand, which shows only about 70% of polymorphonuclears with an increase in large mononuclears up to 10 to 15%.

The final proof is the obtaining of the chocolate-coloured or anchovy sauce-like pus by exploratory puncture.

This pus does not contain pus cells but only granular débris,
cholesterin crystals and is often bacteriologically sterile. The
amoebae, being in the abscess wall, are not apt to be found when
pus is at first withdrawn. Owing to the tendency of liver abscess
to rupture into the lungs the first indication of the true nature
of a prolonged hectic fever may be obtained when the characteristic
pus is expectorated by the patient.

Attended with progressive emaciation and exhaustion the patient,
as a rule, after a prolonged illness, dies, unless operative
procedures cure him or some intercurrent disease brings about his
death.

Symptoms in Detail

_Onset and the Fever Chart._—The onset is at times so insidious
that there may be no symptoms and yet a liver abscess be found at
autopsy. Usually following convalescence after amoebic dysentery an
irregular fever sets in which becomes hectic in character. Profuse
sweats accompany the evening rise. The morning temperature is
frequently normal and there may be frequent apyretic intervals.

_The Respiratory System._—Crepitation at the right base, a dry
cough (tussis hepatica) and shallow respirations are features of
the disease.

_The Nervous System._—Pains in the right shoulder are connected
with irritation of the branches of the phrenic nerve.

The patient is irritable and often complains of insomnia.

The right rectus tends to be somewhat rigid and decubitus is dorsal
or toward the right side.

_The Liver and Spleen._—The liver is tender and as the abscess in
nine-tenths of the cases is located in the right lobe and generally
toward the upper convex surface we have an enlargement upward.
There is very rarely any jaundice.

The urine shows an excess of urobilin and of nitrogen eliminated
as ammonia. When the abscess is in the left lobe the condition is
apt to be considered as some gastric disturbance. The spleen, as a
rule, shows no enlargement.

_Examination of the Blood._—There is usually a moderate
leucocytosis with normal polymorphonuclear percentage and increase
in the large mononuclears.

DIAGNOSIS

=Clinical Diagnosis.=—Of greatest importance is a history of a previous dysentery although it must be remembered that liver abscess may appear in one who has never had dysenteric symptoms. Fever of a hectic type with crepitation at right base, pain and upward enlargement of the liver are most significant.

The X-ray may confirm the diagnosis of upward enlargement which may
be as high as the angle of the scapula. The majority of conditions
causing enlargement of the liver give a downward enlargement.

The amelioration of symptoms by giving emetine hypodermically for two or three days is diagnostically exceedingly important.

Syphilitic gummata may give the picture of liver abscess,
especially as regards the fever and loss of weight. Iodide of
potash is said to be of use in differentiating, as it controls the
fever of syphilis. The gummatous enlargement, however, is irregular
and projects downward.

Suppurative pylephlebitis generally arises from infections of the
colon or appendix. This condition as well as pyaemic (multiple)
abscesses of the liver is apt to show jaundice.

In suppurative cholangitis and cholecystitis we get a history of
biliary colic, jaundice and usually a marked point of tenderness
at the tip of the ninth rib and a tumor in the region of the gall
bladder. Abscess of the left lobe may give the symptoms of gastric
trouble.

In differentiating empyema we usually have a history of pleurisy or
pneumonia.

Suppurating hydatid cyst which may be confused with liver abscess
is most surely differentiated by finding echinococcus hooklets.

Then too the complement fixation test for hydatid disease will
differentiate.

Tuberculosis is often thought of, particularly when a liver
abscess ruptures through the lungs. Malaria is also usually
suspected. Abscess in the kidney or perinephritic region may be
very confusing. In an abscess of the abdominal wall an exploring
needle does not move up and down with respiration as it does when
penetrating a liver abscess cavity.

=Laboratory Diagnosis.=—The chocolate-colored pus of a liver abscess, when there has been no bacterial contamination, shows an absence of polymorphonuclears and does not at first show amoebae. These appear in the pus coming from the drainage tube about the third day. Cholesterin and haematoidin crystals may be found in the granular débris of a fresh drop of pus used for examination for amoebae.

Liver abscess usually shows a moderate leucocytosis with a normal polymorphonuclear percentage and an increase of large mononuclears and transitionals to about 12 to 20%.

According to Schilling-Torgau we may have a perfectly normal white
count and polymorphonuclear percentage and yet have evidence of the
presence of liver abscess in his modification of Arneth’s index, so
that in an apparently normal differential count we may find that
½ or more of the polymorphonuclears are of a less mature type and
in cases where there are many immature polymorphonuclears we have
indications which force a very cautious or unfavorable prognosis.

Thus a differential count of 33% band-form polymorphonuclears
and 39% of normal nucleated ones would make us give a cautious
prognosis, while one with 1% myelocytes, 22.5% immature
polymorphonuclears, 21% band-form nucleated ones and 30% of normal
ones would make for a very bad prognosis. We have a displacement to
the left. Normally there are 63% of normal polymorphonuclears, 4%
of band-form and no immature ones or myelocytes.

One may find an iodophilia in liver abscess.

Of the functional liver tests we may determine the ammonia
quotient, the percentage of N eliminated as ammonia being increased
in abscess of the liver. The same is true of the lipase test.
Probably the most specific test for disturbances of the hepatic
function is that for urobilinogen. The test is made by adding 5 to
10 drops of Ehrlich’s aldehyde reagent to 5 cc. of perfectly fresh
urine when a positive reaction gives a fine cherry-red color.

PROPHYLAXIS AND TREATMENT

=Prophylaxis.=—The prophylaxis is the same as that for amoebic dysentery plus avoidance of anything which reduces the functional power of the liver, such as overfeeding, alcoholic excesses, etc.

It is well to remember that abscesses may occur months or even two
or three years after an attack of amoebic dysentery, consequently
it is well to give a grain of emetine on two or three successive
days of each month following an acute attack.

=Treatment.=—Leaving out of consideration the pre-suppurative stage of amoebic hepatitis which, according to many authorities, responds to injections of emetine, it may be stated that the treatment of liver abscess is entirely surgical and such treatment should be instituted the moment the diagnosis is made. The earlier a liver abscess is drained the less run down will be the patient, the more rapid the convalescence and the better the prognosis.

Until recently surgical authorities condemned severely the trocar
and cannula method of operation, but with the introduction of
emetine there are now those who believe that such a procedure may
suffice and a more radical operation not be necessitated.

Prior to introducing the trocar and cannula the usual procedure
is to use an aspirating needle of about ⅛ inch bore and 3½ inch
length. If the needle happens to be longer it should not be passed
deeper than 3½ inches, in a person with a 32-inch chest, in order
surely to avoid the vena cava. If there are no distinct localizing
signs the needle should first be introduced in the eighth or ninth
interspaces in the anterior axillary line and pushed backward,
inward and slightly upward. Manson recommends at least 6 punctures
before abandoning exploration. Cantlie does not think that a
moderate degree of haemorrhage from the puncture of the liver will
do harm in a case which is simply a liver congestion. One should
always be ready to operate in case pus be found in the exploring
needle. Leaving the needle in situ a small skin incision is made
and a 4 or 5 inch by ⅜ inch trocar and cannula introduced along
the line of the needle. Withdrawing the trocar some of the pus is
allowed to escape through the cannula and there is then introduced
a 6 × ½ inch piece of strong rubber drainage tubing, one end of
which has lateral fenestrations but a closed tip in order that a
long steel pin may put the tubing on the stretch so that it passes
the smaller lumen of the cannula.

The cannula is then slipped out over the tubing and the external
stretched end of the tubing released so that the contracting rubber
fills the puncture. The steel pin used for introducing the rubber
tube is then withdrawn and the tubing transfixed close to the skin
with a safety pin.

After the cavity has drained of pus a dressing is applied. There
are some who advocate aspiration alone without subsequent drainage.
The dressing should be changed frequently and a connecting tube,
draining into an antiseptic-containing bottle, should be attached
to the tube in the cavity in order to obtain a syphoning action.
Some aspirate and inject into the cavity about 2 ounces of 1 to
1000 emetine solution.

Some report favorably from the use of 1 to 1000 quinine
irrigations. At present the hypodermic use of emetine will probably
obviate the necessity of any irrigation.

There are those who think that a preliminary aspiration, followed
by incision, after a few days of improvement in general condition,
is the best method in serious cases.

It is usual to recommend a general anaesthetic when introducing the
aspirating syringe or trocar and cannula. Local anaesthesia with
quinine and urea hydrochloride, however, will usually suffice and
lessen the dangers of shock in bad cases. Rib resections and even
intra-abdominal procedures are best done under local anaesthesia
provided the operator is familiar with the technic.

Newman has recently warned against the use of the small aspirator
for diagnosis, pointing out that it is unreliable and that the
diagnosis should be made by other diagnostic aids, including
hypodermic use of emetine. He notes the occurrence of death from
internal haemorrhage, the interference of the needle with the
surgical incision and, further, the obscuration of the field of
operation by pus where no adhesions exist and, finally, the danger
of general peritoneal infection from a leak. He notes that the
cavity may be under tension and that the pus may force itself along
the track of the needle. He recommends incision and packing with
gauze where adhesions do not exist and the exploration of the liver
with dressing forceps instead of cutting into the liver with the
knife.

USUAL OPERATION FOR LIVER ABSCESS

Either a vertical incision about the middle of the right rectus (Bevan) or a Kocher incision, parallel with the costal margin, may be used. The latter incision favors hernia if prolonged drainage is required. The hand is introduced into the abdominal cavity and the liver palpated. Often the borders of the site of a liver abscess give a hard feeling on palpation. If adhesions are not present the area should be packed off with gauze and the cavity opened by a dressing forceps, haemostat or thermo-cautery. It is often advisable to introduce a trocar and cannula and to drain off the excess of pus.

Where the abscess is situated far back or high up in the liver
the transpleural route is to be preferred to the abdominal one.
Make a 3 or 4 inch incision over the 9th rib with its center in
the line of the angle of the scapula. Excise about 2 inches of
rib subperiosteally. An assistant presses a roll of gauze against
the tissues above the line of incision to prevent air entering
the pleural cavity. Later the upper flap may be sutured to the
endothoracic fascia. Even if the pleural cavity should be opened
and air enter no serious result will follow although it is an
accident to be avoided if possible. The diaphragm is now cut
through and the liver exposed and after packing gauze around the
area to be opened, the abscess cavity is entered and drained as
previously described.

McDill prefers to resect 2 or 3 inches of both 9th and 10th ribs
in the midaxillary line. The muscle wounds made in resection are
then closed by catgut. This movable wall is now forced against the
diaphragm with a roll of gauze pressed inward by an assistant.
A 3-inch incision is now made through this bone-free wall near
the upper border of the 11th rib, going through diaphragm and
exposing the liver. The edges of the wound in the thoracic wall
and diaphragm are now clamped together by haemostats to close off
the pleural cavity. Later catgut sutures are substituted for the
forceps. The liver often bulges into the wound. Finding the abscess
by a palpating finger as a rather firm area in a less resistant
liver surface we introduce a forceps or aspirating needle or trocar
and proceed as above noted.

There are indications that the use of emetine subcutaneously may
make the more radical operations unnecessary. In a recent symposium
on liver abscess many of the papers would indicate a preference for
aspiration without drainage coupled with emetine subcutaneously.

Rogers, in a recent article, notes that a case of liver abscess was
cured by emetine without any form of operation. Recent experience,
however, would indicate that it is necessary to evacuate the pus to
effect a cure.