22 April 1967
Histocompatibility Tests-Johnson and Russell
Bach, F., and Hirschhorn, K. (1964). Science, 143, 813.
Bain, B., and Lowenstein, L. (1964).
Vas, M. R., and Lowenstein, L. (1964). Blood, 23, 108.
Brent, L., and Medawar, P. B. (1963).
Nature (Lond.), 204, 90.
Bridges, J. M., Nelson, S. D., and McGeown, M. G. (1964). Lancet, 1,
Chen, P. S. (1958).
Proc. Soc. exp. Biol. (N.Y.), 98, 546.
Gray, J. G., and Russell, P. S. (1963). Lancet, 2, 863.
(1965). In Histocompatibility Testing, edited by P. S. Russell
and H. J. Winn. Publication No. 1229, Nat. Acad. Sci., Washington,
Ibid., 145, 1315.
Brit. med. 7., 2, 269.
Harris, R., Clarke, C. A., Jones, A. L., Sheppard, P. M., Lehane, D.,
McCarthy, M., Lawler, S. D., and Shatwell, H. S. (1966).
med. 7., 1, 509.
Hirschhorn, K. (1965).
In Histocompatibility Testing, edited by P. S.
Russell and H.
Washington, p. 177.
Huggins, C. E. (1964). Ann. Surg., 160, 643.
Johnson, G. J., and Russell, P. S. (1965). Nature (Lond.), 208, 343.
Moorhead, J. F., and Patel, A. R.(1964).
Russell, P. S. (1966).
In Histocompatibility Testing 1965, edited by HE
Balner, F. J. Cleton, and J. G. Eernissc, p. 233. Copenhagen.
Streilein, J. W. (1966).
Ibid., p. 241.
J. Winn, Publication No. 1229, Nat. Acad.
Brit. med. 7., 2, 1111.
Conversion of Cortisone to Cortisol and Prednisone to Prednisolone
J. S. JENKINS,* M.D., M.R.C.P.; P. A. SAMPSON,t M.B., CH.B.
Brit. med. J., 1967, 2, 205-207
Though cortisone has been widely used since Hench et al.
arthritis, there is now considerable evidence that cortisone and
other steroids which possess the 1 1-oxo group instead of a
1l1l-hydroxyl are themselves biologically inactive.
application of cortisone to the skin is ineffective in the treat-
ment of skin diseases responsive to cortisol (Robinson and
Robinson, 1956), and when cortisone is injected locally into
inflamed joints its anti-inflammatory action is slight compared
with that of cortisol (Hollander et al., 1951).
11/3-hydroxydehydrogenase which renders the 11-oxo steroids
cortisone and its synthetic analogue prednisone biologically
active when they are administered systemically.
reaction appears to take place mainly in the liver (Jenkins,
1966). Several investigators have made clinical comparisons
of the therapeutic effects of cortisone and cortisol and of
prednisone and prednisolone, but there is less information about
the actual amounts of 11,8-hydroxysteroids which are formed
from the 11-oxo compounds. Peterson et al. (1957b) described
the conversion of cortisone to cortisol in one patient, but
the results were complicated by the prior administration of
9a-fluoroprednisolone to the patient.
studied the reduction of prednisone, again in only one patient,
but no comparison was made with the levels of prednisolone in
the plasma after prednisolone itself. The present paper describes
the extent to which oral cortisone and prednisone are con-
verted to cortisol and prednisolone in a series of normal subjects,
and also in patients suffering from liver disease.
its value in the treatment of rheumatoid
It is the reduc-
1 1/-hydroxyl by the enzyme
In man this
Bush and Mahesh (1964)
Procedure and Methods
Nine normal subjects, three patients with Addison's disease,
and four patients suffering from disease of the liver, were given
single oral doses of cortisone or prednisone as crushed tablets
in amounts ranging from 25 to 200 mg.
sions a similar dose of cortisol or prednisolone was given to the
At hourly intervals after the administration of
cortisone and cortisol the levels of cortisol in the plasma were
Similarly, the levels of prednisolone in the plasma
were measured after giving prednisone and prednisolone. Corti-
sone was administered mostly as the acetate, since this is the
usual form available for therapeutic use, but in some instances
the free alcohol was also given for comparison.
prednisone, and prednisolone were given as the free alcohol.
On subsequent occa-
After their oral administration the absorption of each of the
pair of steroids was compared by measuring the total 24-hour
excretion of 17-oxogenic steroids in the urine.
Cortisol in Plasma.-A fluorimetric technique (Rudd et al.,
1963) was used, by means of which cortisol can be measured
without interference from cortisone or other metabolites.
some cases plasma cortisol, tetrahydrocortisol, cortisone, and
tetrahydrocortisone were estimated
chromatographic technique similar to that described for the
estimation of plasma prednisolone, the appropriate standard
steroids being used.
plasma was extracted with 25 ml. of methylene chloride, and
the extract was washed first with 2 ml. of 0.1 normal sodium
hydroxide, followed by 0.1 normal acetic acid and finally with
The extract was evaporated to dryness, redissolved in
a small volume of a mixture of methanol and methylene
chloride, and applied to Whatman No. 1 chromatography paper.
Chromatograms were run for four hours in the benzene-
methanol-water (100:50:50) system of Bush (1952).
prednisolone area was located by ultraviolet light and eluted
with methanol from the chromatogram.
to dryness, redissolved in 0.5 ml. of methylene chloride, and
the steroid estimated, the phenylhydrazinesulphuric acid reagent
of Porter and Silber (1950) being used according to the tech-
nique of Peterson et al. (1957a).
solone were taken through the whole procedure.
17-Oxogenic steroids in the urine were estimated by the
method of Appleby et al. (1955).
specifically by a paper
The eluate was taken
Standard amounts of predni-
* Consultant Physician, St. George's Hospital, London S.W.1.
t Research Assistant, St. George's Hospital, London S.W.l.
Conversion of Cortisone to Cortisol
The mean levels of cortisol in the plasma after the oral ad-
ministration of 100 or 200 mg. of cortisone and cortisol to
normal subjects and 25 mg. of these steroids to patients with
Addison's disease are shown in Fig. 1.
ence was seen in the levels obtained with cortisone acetate and
the free alcohol when both were given to some of the subjects
on different occasions, and they have therefore not been recorded
The peak level after cortisone was normally at two
hours, but in all cases was lower than the peak after cortisol
This difference became disproportionately greater as the dosage
increased up to 200 mg., when plasma cortisol after cortisone
was only a third to half the level after cortisol.
unconjugated steroids present in the plasma of a normal subject
two hours after 200 mg. of cortisone and cortisol were examined
chromatographically, and are shown in Table I.
No significant differ-
22 April 1967
Cortisone and Prednisone-7enkins and Sampson
level after cortisone in this subject was approximately one-half
that after cortisol, but this difference was represented to only
a small extent by unreduced cortisone.
cortisone a greater amount of tetrahydrocortisone was present
than after cortisol, so that much of the cortisone was reduced
at ring A without reduction of the 11-oxo group.
It can be seen that after
TABLE I.-Plasma Unconjugated Steroids in Normal Subject Two Hours
After Oral Cortisone and Cortisol (jg./ 100 ml.)
Steroid in Plasma
as the acetate or the free alcohol were 90% or more of those
obtained after cortisol, indicating that cortisone was absorbed
only slightly less well than cortisol.
1 7-oxogenic steroids after cortisone given either
Conversion of Prednisone to Prednisolone
The levels of plasma prednisolone after the administration
of 100 mg. of prednisone and 100 mg. of prednisolone are
shown for three normal subjects in Fig. 2.
for these same subjects when previously given cortisone and
cortisol in the same dosage is also shown for comparison.
can be seen that, in contrast to the relatively low levels of
cortisol achieved after cortisone, the levels of prednisolone in
the plasma after prednisone closely followed those obtained
after prednisolone itself.
value after prednisolone was twice the cortisol value after a
similar dose of
obtained with prednisone was three to four times that obtained
with an equal amount of cortisone.
The plasma cortisol
It is also of interest that the peak
Conversion of l1-Oxo to 11/8-Hydroxysteroids in Liver
Cortisone and Cortisol.-Fig. 3 shows the plasma cortisol
levels after the administration of 100 mg. of cortisone and
cortisol to a patient suffering from severe infectious hepatitis.
The values after cortisone fell within the normal range, though
FIG. 3.-Plasma cortisol levels in a patient with infectious
hepatitis after 100 mg. of cortisone indicated by *-4
of cortisol indicated by 0- -- 0.
normal range after 100 mg. of cortisone (lower curve)
and cortisol (upper curve) is also shown.
values after cortisol rose above normal.
plasma unconjugated steroids estimated chromatographically
two hours after 200 mg. of cortisone and cortisol were given
to a patient suffering from hepatic cirrhosis, and also to a
The patient with hepatic cirrhosis showed
good conversion of cortisone to cortisol, and in fact at this
point of time the plasma cortisol was greater than in the normal
Table II shows the
FIG. 1.-Plasma cortisol
indicated by 0-S0, and after cortisol indicated
by 0 --- Q.
The mean and range of values
are for five normal subjects given 200 mg. of
each steroid, four normal subjects given 100 mg.,
and three patients with Addison's disease given
normal subjects, a, b, c, after 100 mg. of predni-
sone indicated by
-O and after 100 mg. of
prednsolone indicated by 0 --- 0.
two curves show plasma cortisol values in the
22 April 1967
Cortisone and Prednisone-enkins
unconverted, and both findings can be explained on the impair-
ment of ring A reduction, shown by the much lower values for
tetrahydrocortisol in the cirrhotic patient.
On the other hand, much more cortisone remained
TABLE II.-Plasma Unconjugated Steroids in a Normal and a Cirrhotic
Subject Two Hours After Oral Cortisone and Cortisol (,ug./100 ml.)
Steroid in Plasma
After cortisone (200 mg.)
After cortisoi (200 mg.)
Prednisone and Prednisolone.-Fig. 4 shows the conversion
of prednisone to prednisolone in a patient with infectious hepa-
ttis and another with hepatic cirrhosis who were given 50 mg.
of prednisone and prednisolone.
solone levels were almost identical after the two steroids, and
in the other patient conversion of prednisone to prednisolone
In one patient plasma predni-
FIG. 4.-Plasma prednisolone levels in two patients with
liver disease given 50 mg. of prednisone indicated by
The lower cortisol level found in the plasma after the admini-
stration of cortisone, which amounted to only a third to a half
that obtained with cortisol at the 200-mg. dosage, was not due
simply to large differences in absorption, since in agreement
with Cope and Black (1958) this difference was only about 10%.
Chromatographic analysis revealed that the lower peak value
after cortisone was due only in part to unmetabolized cortisone.
The presence of large amounts of tetrahydrocortisone indicated
that some of the cortisone was wasted, from a biological point
of view, by reduction in ring A instead of at the 11-oxo group.
It seemed that the capacity of cortisone to be converted to
cortisol by 11fP-hydroxydehydrogenase in the liver was limited
by competition from the liver enzymes concerned with reduc-
tion of the A ring, and this was particularly apparent when
large doses of cortisone were given.
was converted to prednisolone with a much higher degree of
efficiency than cortisone to cortisol, and this would follow
from the relative lack of ring A reduction which is known to
occur in these 1,2-dehydrosteroids (Sandberg and Slaunwhite,
The much higher levels achieved in the plasma after
of prednisolone compared with plasma
cortisol after cortisol administration may be an indication of
the differences in the degree to which the A ring is reduced.
These results are at variance with those of Bush and Mahesh
(1964), who, in a single patient, claimed that the 11-oxo group
In contrast, prednisone
of prednisone was reduced to a less degree than was the case
with cortisone, but this would imply that prednisone was bio-
logically less active in relation to prednisolone than cortisone
was in relation to cortisol.
The results reported here, however,
are in accordance with previous clinical findings; Bunim et al.
(1955) found that prednisone and prednisolone were approxi-
mately equivalent in their therapeutic effects.
found that cortisone, either as the free alcohol or as the acetate,
was only one-half to two-thirds as effective as cortisol at the
40-60-mg. dosage in the maintenance treatment of rheumatoid
The results obtained from the patients with liver disease are
of interest, since, although the liver is thought to be the main
site for reduction of the 11-oxo group, there was considerable
preservation of the ability to carry out this reaction in all cases.
There was, on the other hand, marked impairment of ring A
reduction, as reported previously by other investigators (Peter-
of liver disease than those studied would show more evidence of
an impairment of 11-oxo reduction.
of view these results indicate that even in the presence of normal
liver function much of the cortisone
cortisol, and there would seem to be a good reason for using
cortisol itself whenever this steroid is required orally, whereas
prednisone and prednisolone are almost equivalent in their
Thouah liver disease appeared to have little significant
effect on the reduction of the 1 1-oxo group of cortisone and
prednisone, the number of cases was limited, and if steroid
therapy is required in the presence of severe liver disease it is
prednisolone in order to ensure a full therapeutic effect.
It is of course possible that more severe degrees
From the practical point
is not converted to
The conversion of cortisone and prednisone to the biologically
active steroids cortisol and prednisolone has been studied in
normal patients and in liver disease.
blood in normal subjects after oral cortisone were much lower
than after cortisol itself, especially when high doses were used.
In contrast, prednisone was converted to prednisolone with
a high degree of efficiency.
The difference appeared to be due
to reduction of some of the cortisone to tetrahydrocortisone,
whereas prednisone does not readily form tetrahydro derivatives.
In four patients with liver disease, reduction of the 1 1-oxo
group of cortisone and prednisone, which is necessary for
biological activity, was well maintained.
The cortisol levels in the
We wish to thank Roussel Laboratories for the supply of cortisone
One of us (J. S. J.) was in receipt of a research
grant from the Governors of St. George's Hospital, and the other
(P. A. S.) was supported by a Medical Research Council grant.
Appleby, J. I., Gibson, G., Norymberski, J. K., and Stubbs, R. D. (1955).
Biochem. Y., 60, 453.
Boland, E. W. (1952).
Bunim, J. J., Pechet, M. M., and Bollet, A. J. (1955).
Ass., 157, 311.
Bush, I. E. (1952). Biochem. 7., 50, 370.
and Mahesh, V. B (1964).
Cope, C. L., and Black, E. G. (1958). Clin. Sci., 17, 147.
Hench, P. S., Kendall, E. C., Slocumb, C. H., and Policy, H. F. (1949).
Proc. Mayo Clin., 24, 181.
Hollander, J. L., Brown, E. M., Jessar, R. A., and Brown, C. Y. (1951).
7. Amer. med. Ass., 147, 1629.
Jenkins, J. S. (1966).
Peterson, R. E. (1960).
Karrex, A., and Guerra, S. L. (1957a). Analyt. Chem., 29, 144.
-Pierce, C.E.,Wvngaarden, J. B., Bunim, J. J., and Brodie, B. B.
Porter, C. C., and Silber, R. H. (1950).
Robinson, R. C. V., and Robinson, H. M. (1956). Sth. med. 7. (Bgham,
Ala.), 49, 260.
Rudd, B. T., Sampson, P., and Brooke, B. N. (1963).
Sandberg, A. A., and Slaunwhite, W. R. (1957).
Brit. med. 7., 1, 559.
7. Amer. med.
Ibid., 93, 236.
7. Endocr., 34, 51.
7. clin. Invest., 39, 320.
X. cdn. Invest., 36, 1301.
7. biol. Chem., 185, 201.
7. Endocr., 27,
7. clin. Endocr., 17,