J o u r n a l of Analytical T o x i c o l o g y , Vol.

14, March/April 1990

Analysis of Corticosteroids in Urine

by HPLC and Thermospray LCIMS

Song-Ja Park, Yun-Je Kim, Hee-Soo Pyo, and Jongsei Park*

Doping Control Center, Korea Institute of Science and Technology, P.O. Box 131, Cheongryang, Seoul, Korea

Recently, several HPLC procedures were reported for the

Abstract J detection and assay of synthetic or natural corticosteroids in
biological fluids (1,7-10). Because there is no possibility of
The method for simultaneous determination and confirmation thermal decomposition, HPLC offers the advantage that the
of nine corticosterolde in urine by high-performance liquid corticosteroids can be analyzed without derivatization. HPLC
chromatography with s diode array detector and thermoeprey interfaced to a mass spectrometer (thermospray LC/MS) offers
liquid chromatography-mass spectrometry was studied. the further advantage of highly specific detection of these
HPLC was performed on a C1= column with a gradient mobile steroids in biological extracts.
phase system of water and acetonltrtle. The calibration curve
was linear from 20 ng/mL to 1.0 ~g/mL for each corticosterold,
The present paper describes a memoa zor the extraction
end the detection limit was 10 ng/mL in 5 mL of urine. The and simultaneous determination of nine corticosteroids in
extraction recovery of each corticosteroid from the spiked human urine by using H P L C / D A D and thermospray liquid
urine was equal to or greater than 85% by liquid-liquid chromatography-mass spectrometry (LC/MS) for on-line
extraction (LLE) at pH 9 using dlethyl ether end separation and identification of corticosteroids.
approximately 79% or more by solid-phass extraction (SPE)
with a Sep-Pak C1= cartridge. The mass spectra obtained
with the positive Ion mode showed protonated molecular
species [M + H] +, ammonium adduct ion [M + NH4] +,
[MH-60] +, [MH-30]*, and [MH-18] +, The LC/MS detection
.o c.----.~ 2~
limits ranged from 10 to 50 ng in the scan mode and from
1 to 5 ng in the selected ion monitoring (SIM) mode.

Cortisone
B elornelhosolte Cotllcoslerone
Introduction
Both naturally occurring and synthetic corticosteroids are used
primarily as anti-inflammatory drugs that also relieve pain. They
affect the blood level of natural corticosteroids.
It was known that athletes in sports such as cycling and
~ _~OH ~H2CH

weightlifting used corticosteroids to improve performance. Since 0eoxycorlicostcrone Hydrocortiscne

1975, the IOC Medical Commission has restricted such use ex-
cept for legitimate medical purposes.
Chemical structures common to corticosteroids included A', HO-~-~OH

3-keto, 20-keto, and 21-hydroxy group (Figure 1). When ad-

ministered to humans, corticosteroids are bound to protein.
Their major metabolites are glucuronides and they are excreted
as conjugated form or parent form in urine.
I" = "Hydroxypro~eslerone Prednisolone Prednisone
Analyses based on radioimmunoassay (RIA) procedures are
characterized by high sensitivity, but this benefit may be offset by
~HzOH H 0
the lack of specificity due to cross-reactivity of related compounds
. ~c 3
(1). Analyses of corticosteroids by GC and G C / M S methods
(2-6), while highly sensitive and specific, require derivatization
prior to injection because the compounds are thermally labile
and their volatility is too low for direct GC analysis. T rk:mcinolone Tr ~l~nN~e Gcelonlde

Figure 1. Chemical structures of corticosteroids.

9 Author tO whom requests for reprints should be addressed

102 Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission.
Journal of Analytical Toxicology, Vol. 14, March/April 1990

Experimental at the rate of 0.5~ held for 1 min, decreased to 86~

Reagents. Corticosteroid standards were purchased from
Sigma Chemical. HPLC grade acetonitrile and methanol were
purchased from Burdick & Jackson or E. Merck. Purified water
for HPLC was obtained with the Milli-RO and Milli-Q water
systems. Peroxide-free diethyl ether was obtained by distilla-
tion over Call=. All other reagents were analytical grade. The
thermospray ionization solution was 0.15M ammonium acetate
and the standard tuning solution for standard mass range was
polypropylene glycol (Hewlett-Packard).
High-performance liquid chromatography. A Hewlett-
Packard Model 1090M HPLC with a diode array detector
(DAD), autosampler, autoinjector, and DR-5 solvent delivery
system was used for the analyses. The data system used was an
HP 9000-300 computer with HP 9133 disc drive. The chromato-
graphic column was a Hypersil-ODS (100 x 4.6 mm i.d. 5/tm);
it was maintained at 40~ The mobile phase was water and
acetonitrile at a flow rate of 1.0 m L /m i n . The solvent gradient
program was as follows: Initial acetonitrile was 4%, increased
to 30~ for 10 min, then for 5 min to 450/0, and finally for
3 min to 50~ The UV absorbance was monitored at 246 nm
for all corticosteroids. Total run time was 17 min.
Liquid-liquid extraction (LLE). A 3-mL urine sample was
pipetted into a 15-mL centrifuge tube with betamethasone (0.5
#g/mL) as an internal standard (IS). A solid phosphate buffer
(K=HPO,, 0.1 g) was added to adjust pH to 9; 0.5 g anhydrous
sodium sulfate was added. Corticosteroids were extracted into
5 mL of diethyl ether by mechanical shaking for 10 min and
the tube was immediately centrifuged at 2500 g for 5 min. The
at the rate of 1~ and held for 5 min. Other MS parameters
were as follows: ion source temperature, 276~ emission
current, 150 #A; electron energy, 955 eV; mode, positive ion
and filament-on.
Results and Discussion
Separation and calibration by HPLC/DAD
Figure 2 indicates the chromatogram of a mixed cortico-
steroid standard. Under these conditions, prednisolone and
prednisone were not separated, but they could be resolved by
substituting methanol for acetonitrile. The result is shown inset
in Figure 2. All corticosteroids tested have 3-keto and 4-en (c~,
~-unsaturated ketone group); thus they showed the maximum
absorption near 246 nm, which was used for quantitation. The
diode-array detector system gave advantages over the conven-
tional UV detectors. For example, it could show the UV spec-
trum and its first derivative spectrum for each peak in the
chromatogram, and therefore it played an important role in the
identification of each compound together with the relative re-
tention time (Table I).
The standard calibration curve for each corticosteroid was
constructed by the least square method for the peak area ratios
of corticosteroid to internal standard versus the concentration
LC n 24G.,t !~50.]BF] of COCr.II.ts

organic layer was transferred to a second tube and evaporated

to dryness under vacuum. The residue was dissolved in 200 ~tL
methanol and filtered through the sample clarification kit
(pore size, 0.45 p.m) into a vial, and 15 #L was injected into
the HPLC. i iml (rain&) I

Solid-phase extraction (SPE). Sep-Pak C~ cartridge columns

(Waters) were mounted on a specially designed vacuum manifold
and washed twice each with 5 mL methanol and twice each with
5 mL water by applying the vacuum. A 3-mL urine sample con-
taining 0.5 /~g/mL of internal standard was loaded onto the
column and drawn through the column with the vacuum. The Figure 2. (A) Chromatogram of corticosteroid standard mixture
column was washed with 5 mL of water and 1 mL of n-hexane (500 ng/mL each). Peaks: (1) triamcinolone, (2A) prednisone, (2B) pred-
to remove the polar or nonpolar interferences. The collection nisolone, (3) hydrocortisone, (4) cortisone, (5) betamethasone (IS), (6)
tube was then inserted in the vacuum manifold, and the adsorbed corticosterone, (7) triamcinolone acetonide, (8) 11-a-hydroxypro-
gesterone, (9) deoxycorticosterone.
corticosteroids were eluted with 2 mL of diethyl ether (• 3).
The collected extract was evaporated to dryness under vacuum
and the residue was dissolved in 200/zL of methanol and filtered
through the sample clarification kit (pore size, 0.45 #m). Finally, Table I. Relative Retention Times (RtR)" and Typical
15/~L was injected into the HPLC. Linear Regression Equations (y = Ax + B)* *
of Corticosteroids
Thermospray LC/MS. The L C / M S system was a Hewlett-
Packard Model 1090A HPLC linked via capillary tubing and Corticosterolds RtR A D r
thermospray vaporizer probe to an HP 5988A mass spectrom-
eter. The data system was an HP 9000-300 computer with an Triamcinolone 0.716 0.872 -0.020 0.999
Prednisone 0.870 1.020 +0.010 0.999
HP 7946 disc drive. The MS was tuned daily with polypropylene Prednisolone 0.870 0.990 +0.008 1.000
glycol solution. The LC column was a HypersiI-ODS (60 x 4.6 Hydrocortisone 0.882 1.160 0.008 0.999
mm i.d., 3/Lm), and the mobile phase was 0.15M ammonium Cortisone 0.893 1.160 -0.007 0.999
acetate and methanol with a flow rate of 0.8 m L /m i n . The sol- Corticosterone 1.048 1.130 +0.033 0.999
Triamcinolone acetonide 1.077 0.886 0.008 0.999
vent composition was as follows: Initial methanol was 40%, 11-a-Hydroxyprogesterone 1.143 1.240 0.000 0.999
increased to 50~ for 6 min, held for 1 min, then increased for Deoxycorticosterone 1.290 1.300 +0.034 0.999
3 min to 60~ and held for 5 min. The optimum vaporizer probe 9Retention time of ISTD (betamethasone), 12.83 rain (RtR = 1.000),
temperature was programmed according to the composition of " Concentration range; 0.O20-1.O vg/mL, y is the peak area ratio of corlico-
the mobile phase. The typical probe temperature program was steroid to ISTD (0.5 ~g/rnL), x is the concentration ratio of corticosteroi~ to
ISTD (0.5 ~g/mL}, and 9is the correlation coefficient.
as follows: Initial stem temperature was 92~ decreased to 89~

103
 Journal of Analytical Toxicology, Vol. 14, March/April 1990

ratios of drug to internal standard. They indicated good linearity calculations were carried out on the basis of peak-area ratios
(r = 0.999) over the concentration range from 0.020/zg/mL between corticosteroids and the internal standard. The ratio
to 1.0 t~g/mL in the spiked urine sample (Table I). The detec- from extracted urine samples were compared to the peak-area
tion limits for the LLE method were about 3 ng as an injection ratios obtained from unextracted solutions of corticosteroids
amount, which corresponds to the 10-ng/mL level for 5 mL at known concentrations, to determine the percent recovery.
of urine sample. Except for triamcinolone, the recoveries with the LLE method
were higher than those with the SPE procedure.
Recovery study
Recovery studies were performed for 3 mL of urine contain- Analysis by thermospray LC/MS
ing 0.5 #g/mL of each corticosteroid and 0.5/zg/mL of internal U n d e r the described L C / M S conditions, the total ion
standard. The extracting solvent, diethyl ether showed good ex- chromatogram and mass spectrum o f each c o m p o u n d are shown
traction efficiency, least interfering substances, and quick con- in the appendix following the conclusion. The fragment or
centration time. Figure 3 shows the typical HPLC chromato- adduct ions of corticosteroids are listed in Table Ill. Protonated
grams for extracts of (A) blank urine and (B) spiked urine at molecular ion species [MH] § were found in all corticosteroids
pH 9 9 shows much less interference than acidic pH. commonly having 3-keto and 17-keto groups. The base peak
Table II shows the recoveries by LLE and SPE methods. All [MH] § was found in compounds containing no 17-hydroxy
group. Corticosteroids containing a 17-hydroxy group such as
betamethasone, cortisone, hydrocortisone, prednisolone, and
L.C i:1 2 4 6 . 4 5~]. 10~ of COULFrlIIIr
prednisone produced a base peak [MH-60] + from the bond-
cleavage between 17- and 20-carbon, and [MH-30] § due to
4O
30
'Ha
30
the loss of CH20. This can be explained by the elimination
2el a13 mechanism (14) proposed by Vandenheuvel. In addition
113
[MH-lg] § from the loss of H~O was observed. Ammonium
LC 'R 2 ' ~ 6 , 4 55~.1~ or CosPI)<EI.U G~ adduct ion [MNH,] § was found in corticosteroids with no
5e B 23 5 8 5E9 double bond between l- and 2-carbon, except for triamcinolone
41a 4 6 ,tel
313 1 7 9 3el acetonide. Both triamcinolone and triamcinolone acetonide
20 ' 20 showed m / z 359 and m / z 377 due to the side chain cleavage
9between 15- and 16-carbon. Figures 4 and 5 show the total ion
Time (mln.)
chromatogram and single ion chromatograms of a cortico-
Figure 3. Typical chromatograms for extracts of (A) blank urine and steroids mixture obtained in the SIM mode. The retention time
(8) spiked urine (pH 9), Peaks are the same as in Figure 2. and characteristic mass fragment ions [MH § [MH +-30],
and [MH+-Ig], as well as the protonated ion [MH]+and
ammonium adduct ion [MNH,] + could be used for the con-
firmation of each corticosteroid. The detection limits (by scan
Table II. Extraction Recoveries of Corticosteroids*
mode) were 10 ng for I 1-a-hydroxyprogesterone, prednisolone,
Uquld-Liquld Solid-Phase prednisone, and triamicinolone; 30 ng for cortisone and hydro-
Extraction (%) Extraction(~) cortisone; and 50 ng for the other corticosteroids tested. Sen-
Cortlcosterolds X*" RSD X'* RSD sitivity could be increased by using the selected ion monitoring
(SIM) mode with two or three ions producing the strongest
Triamcinolone 85.0 3.2 103 4,0
Prednisone 91.6 1.8 79.2 5.0
intensity; the detection limit was thus decreased to approximately
Prednisolone 99.1 1.1 89.9 3.1 1-5 ng for each corticosteriod.
Hydrocortisone 92.9 2.4 83.3 1.9
Cortisone 90.6 3.4 83.6 2.2 Case study: Determination of cortisohcortisone ratio in urine
Corticosterone 95.0 1.2 89.5 1.1 During the 24th Seoul Olympic Games, 1601 samples of
Triamcinolone acetonide 97.3 2.8 83.1 1.5
11-a-OH-Progesterone 87.3 3.2 86.4 1.6 athletes' urine, excreted after the end of competition, were
Deoxycorticosterone 98.7 2.3 85.6 2.3 analyzed for corticosteroids and to measure the distribution of
* IS: Betamethasone (LLE, 98.0%; SPE, 83.5%).
the cortisol:cortisone ratio. Urine (3 mL) was extracted by the
"" 1"1=5, LLE method, and the concentrations of cortisol and cortisone
were determined by using HPLC/DAD.

Table III. Mass Fragment Ions of Corticosteroids

Mass fragment ions, mlz
Cortlcosteroids Molecular weight Base peak MH* MNHt MH*-60 MH+-30 MH+-18
Betamethasone 392.45 333 393 333 363 375
Corticosterone 346.00 347 347 364 - -
Cortisone 360.47 301 361 378 301 331 343
De0xycorticosterone 330.45 331 331 348 - -
Hydrocortisone 362.47 303 363 380 303 333 345
11-a-OH-Progesterone 330.45 331 331 348 - -
Prednisolone 360.44 301 361 301 331 343
Prednisone 358.44 299 359 299 329 341
Triamcinolone 394.45 347 395 335 365 377
Triamcinolone acetonide 434.44 435 435 452 - 417

104
Journal of AnalyticalToxicology,Vol. 14, March/April 1990

Table IV. Urinary C o r t i s o l Levels of S a m p l e s f r o m

= f t . BE4 TIC =r DQTR, CHI xR--;. D the 24th O l y m p i c Games
';~.r~r:,t (a)
?.,. 1 A2 3 Concentration range
(~glmL) Male Female Unknown Total
~. ~~~']" ~ ,' L ~ L__J k___,__.J k_
L L,
a 1-1~= (ml~.)
8
IB
0.0-0.09 464 187 37 688
= ~ I=,, 3a?.o~ =~u. ~ram DRTR:CMIXR--S.D
0.1-0.19 342 148 33 523
0.2-0.29 108 41 5 154
:.....
] /L 0.3-0.39
0.4-0.49
31
16
9
8
6
0
46
24
L 0.5-0.59 8 4 0 12
0.6-0.69 2 6 0 8
9 I=,~ 299.E)c} =~,u. fr=m DRTR:CMIXR--S.D
0.7-0.79 2 1 0 3
0.8-0.89 0 0 0 0
0.9-0.99 0 3 0 3
1.0-1.99 3 1 0 4
2.0-2.99 0 1 0 1
3.0-3.99 0 0 0 0
~. ~=, (d) 4.0-4.99 1 2 0 3
I".....
......l
i.oi L >5.0 0 0 0 0
uncountable 89 34 9 132
Total* 1066 445 90 1601
9Mean:total,0.14/~g/mL(SD • 0.24);male,0.13t~g/mL(SD • 0.19);and
t m~t~u
female,O.17pg/mL(SD • 0.36).
~ ~ "~ .... LL, ~ '~

Figure 4. Total ion chr0mat0gram and singleion chr0matograms 01

corticosteroidsstandardmixture(500 no each).Peaks:(1) triamcinol-
one, (2) prednisone,(3) prednisolone,(4) corticosterone, and (5) Table V. Urinary C o r t i s o n e Levels of Samples f r o m
11a-hydroxyprogesterone. the 24th O l y m p i c Games
Concentration range
OwglmL) Male Female Unknown Total

0.0-0.09 94 35 9 128
TIC =r DRTR:CHIXB-S.D 0.1-0.19 317 160 30 307
0.2-0.29 345 130 25 490
0.3-0.39 140 51 12 203
c:~. o t 4.
0.4-0.49 49 14 4 67
0.5-0.59 20 10 1 31
4
TIma
G
(~In.)
8 I~ 0.6-0.69 7 3 0 10
0.7-0.79 4 1 0 5
0.8-0.89 3 2 0 5
0.9-0.99 0 2 0 2
2 t3t-,t~ 1.0-1.99 4 1 1 6
Z O . . . .
2.0-2.99 0 1 0 1
I=,, 3EI3.~EI =.,u. Cram ~RTR=CMIXB-~.O
3.0-3.99 0 1 0 1
4.0-4.99 0 0 0 0
D,~c~.,~' (c) >5.0 0 0 0 0
uncountable 83 34 8 125

9 Ia,, 333,Q~ m,r Crom DRTA:CHIXB--S.n

Total* 1066 445 90 1601
9 Mean:total,0.24~g/mL(SD • 0.18);male,0.23lag/mL(SD • 0,14); and

# 2~uo
4elu~

4
r,,.j,,..
Ttmm
g
(mlh,)
L 8 IO
female,0.24~g/mL(SD • 0.25).

.... I (e) .................... ~ • .....

Table IV shows the concentration distribution of cortisol. The
Z 4 G O I~ l~
mean values were 0.14 #g/mL for totally countable measure-
Tin= (mln.)
It,', 331,GE4 i,nu. from flRTR:CMIXB--•.D ments, 0.13 #g/mL for male, and 0.17 #g/mL for female
athletes. The cortisone concentration distribution is listed in
ot ~L4"
u2.st:, .t If) J
~1 ~ t ~ r a a , =: ~ tin]~ Table V. The mean values were 0.24 #g/mL, 0.23 #g/mL, and
0.24/zg/mL for total, male, and female athletes, respectively.
As shown in Tables IV and V, the concentrations of cortisol
and cortisone appeared to be independent of sex. The concen-
Figure 5. Total ion chromatogramand single ion chromatogramsof tration distributions of cortisol and cortisone appear as a
corticosteroidsstandardmixture(500 ng each). Peaks: (1) cortisone, histogram in Figures 6 and 7.
(2) hydrocortisone,(3) betamethasone,(4) triamcinoloneacetonide, P.V. Bertrand and co-workers (15) measured the cortisoi in
and (5) deoxycorticosterone.
two consecutive overnight urine collections from 103 healthy

105
 Journal of Analytical Toxicology, Vol. 14, March/April 1990.

T a b l e Vl. C o n c e n t r a t i o n Ratio of Cortisol to C o r t i s o n e

~oo

Ratio
I : Male corUsol:codisone Male Female Unknown Total
400
: Female 0.0-0.09 66 33 6 105
3so. 0.1-0.19 34 6 2 42
0.2-0.29 102 35 6 143
aoo 0.3-0.39 153 60 13 226
!~o
0.4-0.49 135 62 6 203
z 0.5-0.59 136 62 15 213
2oo 0.6-0.69 125 41 10 176
0.7-0.79 70 23 4 97
0.8-0.89 41 23 6 70
0.9-0.99 15 10 1 26
1oo.
1.0-1.99 68 35 11 114
2.0-2.99 13 8 1 22
3.0-3.99 4 0 0 4
o~ . . . . . . . . . . 4.0-4,99 1 3 0 4
o.z o.a o.a o.4 o.~ o.6 o.7 o.e o.g t.o
ConcentPllton Range (ug/mL) >5.0 2 4 0 6
Figure 6. Concentration distribution of cortisol. uncountable 101 40 9 150
Total* 1066 445 90 1601
9 Mean: total, 0.64 pg/mL (SD + 1.86); male, 0.57 ~g/mL (SD + 0.82); and
female, 0.82 ~glmL (SD + 3.39).

$00.

4~0-
I : Male
~ooq
400-
: Female
taoq
aao-
I : Male

aoo. xao] ~ : Female

t40~
E a~o.

t2oq
eoo-

l~o. xooJ

eo 1
1oo,

~o

o Lo.~ 0.2 o.a o.4 o.e o.e 0.7

C o n c l n t . r s t 1o~ R l n g e
o.e
(ug/mr,.)
o.9 ~*-0
eo ~

"~
ao 4

Figure 7. Concentration distribution of cortisone. 0

o 0.1 0.2 o.a o., o.~ 0.6 o.7 o.a o.Q ~.0
Cort~sol/Cortia=ne ne~o

Figure 8. Distribution of c0~isol:cortisone ratio.

school children, ages 7 to 18.5 years. They used a competitive
protein binding assay technique and reported that the mean
cortisol concentration was 194 n m o l / L (approximately 0.070
/zg/L). But the measurement range of concentration of cortisol
appropriate to 95~ of normal children was from 83 to 396
nmol/L (approximately 0.030-0.143 #g/mL). They also found
that the cortisol concentration was independent of sex, age, and
body weight.
Table VI and Figure 8 show the distribution for the concen-
tration ratio of cortisol to cortisone according to sex. The mean
ratios were 0.64 for total available data, 0.57 for male athletes
and 0.82 for female athletes.
During the '86 Seoul Asian Games, urine samples from 598
athletes were tested and the mean values of cortisol and cor-
tisone were 0.11 #g/mL and 0.17 #g/mL, respectively. The
mean ratio of cortisol to cortisone concentration was 0.64.
A few cases of high cortisol concentrations (above 4/~g/mL)
were found in the urine samples during the 24th Olympic Games
and their cortisohcortisone ratios ranged from 5.6 to 8.9. As
shown in Table VI, those values are considerably high compared
to the ratios for most of the other samples. It is necessary to
further research the relationship between the cortisol:cortisone
ratio for the useful prediction of abnormality of corticosteriod
use in the doping test.
Conclusion
We described the method for resolving and quantitating
corticosteroids in human urine by reversed-phase HPLC. This
method allows the simultaneous determination of at least ten
corticosteroids with a very simple, reproducible, and reliable
analytical procedure. The thermospray LC/MS technique pro-
duced useful information concerning the molecular weight of
corticosteroids without the need for derivatization and detec-
tion limits that could be decreased by using SIM mode.

106
Journal of Analytical Toxicology, Vol. 14, March/April 1990

Appendix
Scan

i{38j

~
]
5{3
j

3{30
165
333

.
,,,/I ~ ~

!.
(7.220 m i n )
S U B T R R C T E ~ SCALE[3

340
375 253

m4r/ I
of DATR:BETRME-5.D

.........~..,.,,; .....I. . . . . . . . . . . . . . . .
360 380 400
434

428
~

440
451/

46D
,00]j I
g:
Scan

50 t

e ....
168

I
47

,.
(7.353

SU[3TRRCTE[3

362

.,
min)

I,
364

.
o4" [3ATR:CORTICOI.DD

SCNLED

,, . . . .
35{} 368 378 388 398 4{3{3 41'8
Mass/Charge Ha=s/Charge
TIC o~ DATA:BETRME-S.D TIC of DRTA:CORTICOI.D
u4,DE4 ~8.{3E4

"1
~3.~E4
~2.0E4
~10000

2
Betamethasone

~
Ime (min.)
6
L 8 10
~6.eE41
{3Eq
O~ .
2
. . ~. .
~
Corticosterone

ime
.
(min.)
6
~
8
" --c
IO

Scan 87 (4.451 mln) of D R T A : C O R T N - I . O Scan 196 (18.083 min) of D A T A : D E O X Y - I . D

u~ 100] 3~1/~" SUBTAAC~j~J SCALED ~31 SUBTRACTED SCNLED
348
7 5oj 58
9 263 390/
~: o ,11.. , 14 . ,:,..... .1~ ..... {h....... ,/,, ..... .ft.. ,
300 330 340 360 3{30 40~ 330 340 350 360 370 380 390
Mass/Charge Mass/Chamge
TIC o~ [3RTA:CORTN-I.O TIC o f DRTR:DEOXY-I.D

~ 1. {3E5 u2.0E4~
Cortisone ~.SE4j
10000J Deoxycorticosterone
5. 0E4
500D~
8J . . . . . . . . . . . . . . -
2 4 6 8 10 2 4 G 8 IO 12
Time (mln.) Time (min.)

Scan 103 (5.10] mln) of DRTR:CORTL-I.D Scan IS6 48,782 min) of DRTR:PROGE-I.D
3 SUBTRACTED SCALED

eJ ..
__3 3{38
_ . ~ , . _ - , . I 4 . . . ! , . . . . . . .11,. . . . . ( . . . . . . .~ .
404 ~oj 0 ' ,I
/
,
~3 ~0~ 363
,
/
,, ,
/
h .
390/
'D
300 320 340 360 380 400 330 340 350 360 37{a 38{~ 35{3
Mass/Charge M ~ s s / C h apse
TIC of D ~ T A : C O R T L - I . D TIC o f DATR:PROGE-I.DD
b.{3E0i
~4.{3E4
~L_ r
~4.{3E4
~3.0E4
~2.0E4
Hydrocortisone ~3.OE41
~{3E41
,,i?~ e-~
~10080~ ~10000~
D" , - , , - , ,
2 4 6 8 10 2 4 S 8 10
T Ime (ml~.) Time ( m i n . )
Scan 116 ( 5 . 165 m i n ) O~ DRTR:PREDNSLI .D Scan 89 44. ]s min) of DRTR:PREDDNSNI .D
I{301 iS % T R A C T E [ 3 SCALED 1001 ~99 SUBTRACTED/
/ SCF~Ep
343 sD I ~ 3s3
/. ... / / , 3~6 325 , / 3~

I{301 o ,
3{3{3

TIC
II,

o?
,,
320
,,~J.l.

Mess/Charge
[3ATR:PREDNSLI.D
34o
.. hL. ,J,.
36D
/
380 300

TIC
320 340
Me~:;s/Ch s r g e
360

o f EIFtTA:PREDNSN! .[3
380

I,OESJ
e. DE4 t

•2 . DE4
J 9

2
, .
Prednisolone

4
, .
i~

6
.

8
,
ff
~,0E4 t
4 . DE4
~.DE41
N"
2 4 S 8
Time (mln.) Time (min.)

Scan 43 42.677 min) o# DATA:TRIRM-3.D Scan 163 47.832 mln) of DRTR:TRIAMRCI.D

E
7 5el
0
3~s
..... .........~,L~ 9
,
RCTE[3 SCALE[3

~5
: . .+ , ~,......
~77
. .
10D 1
5o
o
34,
SUBTRACTED

2$5 )77
/ ...... I~,....... ,.. . . . . . . . . .
SCALED

~ .~.
@35

L
~.
320 340 360 388 400 4~D 440 340 360 380 408 420 440 468
Ma=s/Charge Mass/Charge
TIC of OATA:TRIRM-3.[3 TIC of [3ATA:TRIAMACI.D

1,OE5
5.DE4 Triamcinolone ~
gl.SE41
~2;8E4

1 O D O 8

J 5DODI
2 4 6 8 18
Ttme (mln.) T ime 4min.)

Total ion chromatograms and mass spectra of corticosteroids by thermospray L C / M S .

107

References
1. R.W. Kuhn and M.E, Deyman. Simultaneous quantification of
natural glucocorticoids and progestins in serum. J. Chromatogr.
421:123-29 (1987).
2. E. Houghton, P. Teale, and M.C. Dumasia. Improved capillary gas
chromatographic/mass spectrometric method for the determina-
tion of anabolic steroid and corticosteroid metabolites in horse
urine using on-column injection with high-boiling solvents. Analyst
109:273-75 (1984).
3. Y. Kasuya, J.R. Althaus, J.P. Freeman, R.K. Mitchum, and J.P.
Skelly. Quantitative determination of dexamethasone in human
plasma by stable isotope dilution mass spectrometry. J. Pharm.
Sci. 73:446-51 (1984).
4. J.J. Vrbanac, C.C. Sweeley, and J.D. Pinkstone. Automated meta-
bolic profiling analysis of urinary steroids by a gas chromatog-
raphy/mass spectrometry data system. Biomed. Mass Spec. 10:
155-61 (1983).
5. J.A. Jonckheere and A.P. De Leenheer. Candidate definitive
method for the determination of cortisol in human serum. Biomed.
Mass Spec. 10:197-202 (1983).
6. E. Houghton, P. Teale, M.C. Dumasia, and J.K. Welby. The use
of capillary column gas chromatography and negative ion
chemical ionization mass spectrometry to confirm the administra-
tion of synthetic corticosteroids to horses. Biomed. Mass Spec.
1 0 : 4 5 9 - 6 5 (1982).
7. C.J.H. Woodward and P.W. Emery. Determination of plasma cor-
ticosterone using high-performance liquid chromatography. J.
Chromatogr. 419:280-84 (1987).
8. M. Schonessh0fer, A. Kage, B. Eisenschmid, P. Heilmann, T,K.
Dhar, and B. Weber. Automated liquid chromatographic determina-
108
Journal of AnalyticalToxicology,Vol. 14, March/April 1990
tion of the 20-dihydro isomers of cortisol and cortisone in human
urine. J. Chromatogr. 380:267-74 (1986).
9. W.F. Ebling, S,J. Szefler, and W.J. Jusko. Analysis of cortisol,
methylprednisolone and methylprednisolone hemisuccinate. J.
Chromatogr. 305:271-80 (1984).
10. V.K. Prasad, B. Ho, and C. Haneke. Simultaneous determination
of prednisolone acetate prednisolone, prednisone, cortisone, and
hydrocortisone in swine plasma using solid-phase and liquid-
liquid extraction techniques. J. Chromatogr. 378:305-16 (1986).
11. G.W. Taylor and D. Watson. Biomedical applications of thermo-
spray liquid chromatography-mass spectrometry. J. Chromatogr.
394. 135-46 (1987).
12. D.S. Skrabalak, T.R. Covey, and J.D. Henion. Qualitative detec-
tion of corticosteroids in equine biological fluids and the com-
parison of relative dexamethasone metabolite/dexamethasone
concentration in equine urine by micro-liquid chromatography-
mass spectrometry. J. Chromatogr. 315:359-72 (1984).
13. E. Houghton, M.C. Dumasia, and J.K. Wellby. The use of com-
bined high performance liquid chromatography negative ion
chemical ionization mass spectrometry to confirm the administra-
tion of synthetic corticosteroids to horses. Biomed. Mass Spec.
8 : 5 5 8 - 6 4 (1981).
14. K. Tsuji. GLC and HPLC Determinatio n of Therapeutic Agents,
Pt. 2. Marcel Dekker, New York, 1980. p. 865.
15, P.V.Bertland, B.T. Rudd, R.H. Weller, and A.J. Day, Free cortisol
and creatinine in urine of healthy children. Clin. Chem. 33:
2047-51 (1987).
Manuscript received April 20, 1989;
revision received January 3, 1990.