JULY 24, 1954THE CHILD WITH HEART DISEASE
- McKeown, T., and Record, R. G. (1953).
Parkinson, J. (1945).
Lancet, 2, 657.
Rammelkamp, C. H., Wannamaker, t. W., and Denny, F. W.
Bull. N.Y. Acad. Med., 28, 321.
Rantz, L. A., Maroney, M., and DiCaprio,
Fever, edited by Lewis Thomas. p. 90.
Read, F. E. M., Ciocco, A., and Taussig, H. B. (1938).
Record, R. G., and McKeown, T. (1953).
Selwentker, F. F. (1952).
In Rheumatic Fever, edited by Lewis Thomas,
p. 17.Oxford Univ. Press, London.
Stevenson, A. C., and Cheeseman, E. A. (1953).
Swan, C., Tostevin. A. L., Moore, B., Mayo, H., and Black, G. H. B.
Med. J. Aust., 2, 201.-
Taran, L. M. (1953).
Bull. St. Francis Sanat., 10, 13.
Warkany, J., and Schraffenberger, E. (1944).
Wilson, M. G. (1940). Rheumatic Fever. Commonwealth Fund, New York.
Whipham, T. (1888).
British Medical Journal,
Brit. J. soc. Med., 6, 178.
Brit. Heart J., 15, 121.
J. M. (1952).
Oxford Univ. Press, London.
Amer. J. Ilyg.,
Brit. Heart J., 15, 376.
Ann. Eugen., 17, 177.
J. Nutr., 27. 477.
A NEW TABLET TEST FOR BILIRUBIN
J. A. TALLACK, M.B., Ch.B.
SHEILA SHERLOCK, M.D., F.R.C.P.
Physician and Lecturer
Department of Medicine, Postgraduate Medical School of
This paper describes a simple method for the detection
of small amounts of bilirubin in the urine.
first concentrated by adsorption on a special test mat
and then allowed to react with a stable diazonium com-
pound incorporated in a tablet.
urine from normal and jaundiced subjects have been
compared with those using two standard routine methods
for urinary bilirubin, Fouchet's test and the iodine ring
The tablet test is performed as follows: (1) Place 5 drops
of urine on the test mat, which is 19 mm. square and
2 mm. deep, and is composed of a mixture of asbestos
and cellulose fibres.
of the moistened area of the mat.
dye (p-nitrobenzene diazonium p-toluene sulphonate) and
also sulphosalicylic acid, sodium bicarbonate, and boric
(4) The colour developing on the mat is recorded
within 30 seconds.
The bilirubin is concentrated on the test
The diazo dye is taken into solution by the water,
aided by the effervescence caused by the presence of sul-
phosalicylic acid and sodium bicarbonate in the tablet;
the boric acid is a vehicle.
urine the mat around the tablet turns purple and the amount
of bilirubin is roughly proportional to the speed of develop-
ment and intensity of the colour.
after 30 seconds is ignored.
satisfactory colour reaction
water is allowed to remain on the tablet for a few seconds
before the second drop is put on to wash the fluid on to
the test mat.
Fouchet's test is performed by adding 2 ml. of 10%
barium chloride to 10 ml. of urine, filtering and adding
Fouchet's reagent to the filter paper.
indicated by a green colour.
The iodine ring test is performed by layering 2 ml. of
tincture of iodine diluted
of urine in a test-tube.
A positive result
by a green ring at the interface.
A modification of the tablet test has been employed, using
5 drops of urine on the test mat as before, but replacing
the tablet with
result is indicated by the development of a green colour
on the mat.
Results obtained with
(2) The tablet is placed on the centre
It contains a stable diazo
(3) Two drops of water are allowed to flow on the
If bilirubin is present in the
Any colour developing
It is found that the most
is given if the first drop of
A positive result is
1 in 4 with water on to 10 ml.
1 drop of Fouchet's reagent.A positive
Quantitative urinary bilirubin estimations were performned
by the method of Golden and Snavely (1948).
Schlesinger's test with alcoholic zinc acetate and by Ehrlich's
Serum bilirubin levels were estimated by
the method of King and Coxon (1950).
Samples of urine from 100 non-jaundiced patients picked
at random from the wards of a general hospital and 200
urines from jaundiced patients were tested.
urines containing urobilin or an excess of urobilinogen but
no bilirubin were also studied.
bilirubin level was estimated at the time of testing the
The 100 urines from non-jaundiced patients were tested
to determine whether false-positive reactions might result
from the administration of drugs or from the presence of
other abnormal substances in the urine.
receiving many varied forms of treatment and suffering from
a wide range of disorders.
give a positive tablet reaction, and it therefore seems unlikely
that false-positive reactions occur owing to the presence of
commonly administered drugs in the urine.
there was seen on the mat a faint orange-brown colour
which became more conspicuous after the mat had been
standing for some time.
This colour did not interfere
with the reading of the test, and was presumably due to
in the urine of diazotizable non-bilirubin
The presence of urobilin or an excess of urobilinogen
does not give
36 urines with a positive Ehrlich or Schlesinger reaction
were negative both by the tablet test and by Fouchet's
These results agree with those previously reported with
the tablet test.
Free and Free (1953) tested approximately
2,000 urine samples from general-hospital patients and failed
to find any substance giving false-positive reactions.
also tested 100 urine samples from normally healthy adults;
99 gave negative reactions and one an unexplained trace
reaction. Klatskin and Bungards (1953) found two false-
positive tablet reactions in a series of 478 urines, and in both
these instances there might have been some hepatic dysfunc-
Giordano and Winstead (1953) found that the pre-
sence of increased amounts of urobilinogen in the urine
did not mask the colour reaction as it did in the Harrison
spot test, in which Fouchet's reagent is used.
Of the 200 urines tested which gave either a positive
Fouchet or a positive tablet reaction, 185 were positive for
both methods; 192 gave a positive Fouchet and 193 a
positive tablet reaction, while only 135 were positive with
the iodine method.
Ninety urines giving a positive tablet reaction were also
tested with the modification using the test mat and replacing
the tablet by Fouchet's reagent: 58 gave a positive result
and 32 a negative one.
These results suggest that the tablet test and Fouchet's
method are of about equal sensitivity and are considerably
superior to the iodine ring test and to the modification
for the detection of bilirubin in the urine.
found that the tablet test was more sensitive than methods
using Fouchet's reagent (Klatskin and Bungards,
Giordano and Winstead, 1953).
Quantitative analysis of bilirubin in urine is made difficult
by the presence of interfering substances.
attempt was made to compare the quantitative urinary
bilirubin values with the results for the qualitative tests.
Eighteen urines were analysed and then diluted serially, and
the lowest concentration giving a positive qualitative reac-
tion was recorded (Table I).
the tablet test could usually detect bilirubin between 0.1
In 92 instances the serum
The patients were
In no instance did the urine
Altho>ugh -the results varied,
JULY 24, 1954
NEW TABLET TEST FOR BILIRUBIN
and 0.15 mg. per 100 ml.
slightly less sensitive, usually detecting concentrations be-
tween 0.15 and 0.20 mg. per 100 ml.
was always less sensitive than the other tests, and usually
over 1 mg. per 100 ml. was needed for a positive result. The
sensitivity did not alter with the cause of the bilirubinuria,
whether due to obstruction of the bile passages, to infective
hepatitis, or -to portal or biliary cirrhosis.
These results agree well with those of Sobotka and his
co-workers (1953), who found that the lowest concentration
of urinary bilirubin giving a colour varied from 0.04 to
TABLE I.-Comparison of the Test Results with the Quantitative
Analysis of Urinary Bilirubin
The Fouchet method was only
The iodine method
Lowest concentration of bilirubin
in diluted urine
giving positive test
0.1 mg. per 100 ml.
0.1 mg. of bilirubin per 100 ml. of urine usually gave a
Klatskin and Bungards (1953) found that
the first definite colour change was at 0.05 mg. per 100 ml.,
but 0.9 mg. was needed*for a definite positive.
workers found that albumin in the urine would interfere
with the sensitivity of the tablet test.
not be confirmed in the present study.
Relation to Serum Bilirubin Level
Simultaneous serum bilirubin estimations and urine tests
were performed on 92 occasions.
serum bilirubin level was greater than
Free and Free (1953) stated that
This finding could
In every instance the
1 mg. per 100 ml.
TABLE II.-Comparison of Urinary Test Results with the Simul-
taneous Serum Bilirubin Concentration
Serum Bilirubin (mg./100 ml.)
No. of urines tested
Clinical icterus is apparent at 2 to 3 mg. of serum bili-
rubin per 100 ml., and at this level the tablet and Fouchet
reactions may or may not be positive while the iodine
reaction is usually negative.
per 100 ml. all these methods may give negative results.
At serum bilirubin levels above 6 mg. per 100 ml. the reac-
tions were usually positive.
The failure to detect bile pigment in the urine at low
serum bilirubin levels can be attributed to the small number
of patients with early infective hepatitis included in the
In these subjects bilirubin occurs in the urine at low
serum bilirubin levels.
Cases of obstructive jaundice with
low serum bilirubin values were also few in number., The
majority of the patients studied when their serum bilirubin
value was low were recovering from infective hepatitis, in
which disease bilirubin is known to disappear from the urine
in the presence of a raised serum bilirubin level.
For the early detection of bilirubin in the urine in the
jaundiced subject the tablet test
Even at values of 3 to 6 mg.
is at least as good as
enough to justify its continued routine performance in ward
The modification with the mat and Fouchet's
reagent but no tablet is superior to the iodine ring method
although inferior to the usual Fouchet's method and to the
A new tablet test for urinary bilirubin is described
which is specific for bilirubin, and of at least equal sensi-
tivity to the Fouchet test while being quicker and easier
The iodine ring test is of little value as a
method of detecting small amounts of bilirubin in the
We wish to thank Mr. J. A. Bunce, of Don S. Momand Ltd.,
who provided the tablets and test mats for the trial, and Dr.
B. H. Billinig for the quantitative urinary bilirubin estimations.
The iodine method is not sensitive
Free, A. H., and Free, H. M. (1953).
Giordano, A. S., and Winstead, M. (1953).
Golden. W. R. C., and Snavely, J. G. (1948).
King, E. J., and Coxon, R. V. (1950).
Klatskin, G., and Bungards, L. (1953).
H., Luisada-Opper, A.
clin. Path., 23, 607.
Gastroenterology, 24, 414.
Amer. J. clin. Path., 23, 610
J. Lab. clin. Med., 33, 890.
J. clin. Path., 3, 248.
New. Engl. J. Med., 248, 712.
V., and Reiner, M.
Scrotal Pneumocele Complicating Induction of
Although several cases of scrotal pneumocele complicating
artificial pneumonperitoneum have been reported it has not
been possible to find any report of this occurring as a
following case is therefore recorded.
A man aged 40 was admitted to the London Hospital
Annexe on March
17, 1952, with advanced bilateral pul-
monary tuberculosis, and a previous history of a right
inguinal- herniorrhaphy some 25 years before.
ment with bed rest and full courses of chemotherapy
was decided to attempt to induce a right artificial pneumo-
thorax, although at the time it was realized that this might
not be possible as extensive adhesions were to be expected.
Induction of a right artificial pneumothorax was carried
out at 2 p.m. on December 17.
pressures were -Il and -3 mm. of water, and there was a
free rise and fall in the manometer.
introduced, the final pressures being -11 and 0 mm. of
During the course of the induction the patient had an
attack of coughing, but did not complain of any difficulty
in breathing or of pain, and there was no surgical emphy-
sema at the completion of the induction.
wards another bout of coughing began, again with no
distress, but this time it was accompanied by the appearance
of minimal surgical emphysema around the
The cough was settled by the administration of
linct. terp. heroin co., 1 dr. (3.5 ml.), and no more surgical
However, at 9 p.m. another bout of coughing began,
accompanied by some slight breathlessness.
went to sleep after this, but awoke at 11 p.m. to find that
his scrotum was distended but not painful. On examination
he was not distressed or dyspnoeic.
minimal down the right side of the trunk and along the
line of the right inguinal canal to the right side of the
scrotum, which was distended to about 5 by 3 by 2 in.
(12.5 by 7.5 by 5 cm.).
The patient was sedated and slept well, but by 10 a.m.
the next morning surgical emphysema was extensive in the
right axilla and the right side of the chest, pitting up to
It was minimal in the neck, but
The initial intrapleural
350 c.cm. of air was
site of the
Other than a small
surgical emphysema was
in. (2.5 cm.).