XENOBIOTICA, 1986, VOL. 16, NO.

3, 239-249

The metabolism of aspirin in man: a population study

A. J. HUTT,? J. CALDWELLS and R. L. SMITH
Department of Pharmacology, St Mary’s Hospital Medical School,
London W2 lPG, UK

Received 1 7 July 1984 revised 1 7July 1985

Xenobiotica Downloaded from informahealthcare.com by UB der LMU Muenchen on 06/18/14

1. T h e metabolism of a 900 mg oral dose of aspirin has been investigated in 129 healthy
volunteers. For this purpose, the 0-12 h urine was collected and analysed for the following
excretion products: salicylic acid, its acyl and phenolic glucuronides, salicyluric acid, its
phenolic glucuronide and gentisic acid.
2. T h e total excretion of salicylate and metabolites was normally distributed within the
population group studied, showing a 2.5-fold variation: a mean of 68.1 yo of the dose was
recovered in 12 h .
3. T h e excretion of salicylic acid was found to be highly variable within the study panel
(1.3-31’), of dose in 12 h), and was related to both urine volume and pH.
4. Salicyluric acid was the major metabolite in the majority of the volunteers and its
excretion was normally distributed amongst the study panel. T h e elimination of this
metabolite ranged from 19.8 to 65O;, of the dose and was related to the total recovery of
For personal use only.

salicylate.
5. T h e excretion of the two salicyl glucuronides was highly variable, ranging from 0.8 to
42YA of the dose. The elimination of the glucuronides was inversely related to that of
salicyluric acid.
6. Gentisic acid and salicyluric acid phenolic glucuronide were minor metabolites of
salicylate, accounting for 1 and 3?, of the dose, respectively.
7. T h e recovery of gentisic acid was statistically significantly greater in female subjects
than in males, whilst the opposite was found for salicyluric acid and total salicylate.
However, these differences were small in magnitude.

Introduction
Variability in metabolism is known to be an important determinant of an
individual’s response to a drug. There is thus considerable interest in identifying,
and where possible quantifying, the significance of discrete factors as sources of
human variation in drug metabolism and response. It is accepted that inter-
individual variations in biotransformation have their origins in differences between
subjects in the nature and activities of the various drug-metabolizing enzymes,
which are subject to regulation by a plethora of genetic, environmental, physiological
and pathological factors.
Inter-individual variability in drug metabolism has been almost exclusively
studied from the viewpoint of the oxidative reactions, and very little attention has
been paid to the metabolic conjugation reactions. Variability in the activity of these
latter pathways may be more complicated, as factors governing the supply of the
endogenous conjugating agent will be involved in addition to those governing
enzymic activity.

tPresent address: Department of Pharmacy, Brighton Polytechnic, Moulsecoomb, Brighton, East

Sussex, UK.
$To whom correspondence should be addressed.
 240 A. J . Hutt et al.

In order to assess inter-individual variability in metabolic conjugation processes

in the population, we have carried out a study on aspirin (acetylsalicylic acid). This
agent was chosen as a model compound as, following hydrolysis to salicylic acid, the
latter undergoes extensive metabolic conjugation with both glycine and glucuronic
acid to form four distinct conjugates (see figure 1). In addition to these conjugation
pathways, a small amount of salicylate undergoes aromatic hydroxylation yielding
gentisic (2,5-hydroxybenzoic) acid, which may in turn be conjugated with glycine to
yield gentisuric acid, and a variable amount is excreted unchanged (see Hutt,
Caldwell and Smith 1982, and references therein).
For many years the literature has contained suggestions that the metabolism of
salicylate is highly variable in the population (Alpen et al. 1951), and numerous
Xenobiotica Downloaded from informahealthcare.com by UB der LMU Muenchen on 06/18/14

factors have been associated with this variability, including gender (Menguy et al.
1972, Rainsford et al. 1980, Miaskiewicz, Shively and Vesell 1982), age (Montgom-
ery and Sitar 1981), genetic factors (Furst, Gupta and Paulas 1977), disease
processes (Kapp and Coburn 1942, Graham et al. 1977), auto-induction (Furst et al.
1977, Day, Shen and Azarnoff 1983) and glycine availability (Notarianni et al. 1983).
Of particular note is the study of Furst et al. (1977), who showed, using pairs of
identical twins, that genetic factors were significant in determining inter-individual
differences in salicylate disposition. However, an earlier study showed that the
plasma level of salicylate following administration of sodium salicylate to tubercu-
lous patients was found to be normally distributed amongst the study population
For personal use only.

(Evans 1960, quoted in Evans and Clarke 1961), suggesting that the existence of a
major polymorphism in the metabolism of salicylate is unlikely. However, taken
alone, this conclusion must be treated with some caution, as there do occur instances
of apparently normal distributions concealing polymorphically distributed variables
among the population.
Recently, in a preliminary study, Caldwell et al. (1980) have found the

Hydrolysis

Figure 1. The metabolic pathways of aspirin (acetylsalicylic acid) in man.

 Population study of aspirin metabolism 241

metabolism of salicylate to be highly variable amongst a panel of 85 healthy young

volunteers, in particular noting a 12-fold variation in the excretion of salicyluric acid.
Other reports (Angelo-Khatter, Al-Kharo and Al-Nassar 1981, Bennett et al. 1982)
have confirmed this variability in the formation of salicyluric acid. T h e present study
was initiated to define the extent of variability in the various metabolic pathways of
salicylate within a group of healthy medical students and nurses by monitoring the
excretion of all known metabolites of this compound.

Materials and methods

Compounds
Xenobiotica Downloaded from informahealthcare.com by UB der LMU Muenchen on 06/18/14

Salicylic (2-hydroxybenzoic) acid was obtained from the Pharmacy of S t Mary’s Hospital (London
W2, UK). The following compounds were purchased: aspirin (acetylsalicylic acid), gentisic (2,s-
dihydroxybenzoic) acid (Sigma Chemical Co., Poole, Dorset, UK); salicyluric (2-hydroxyhippuric) acid,
2-methoxybenzoic acid, hydroxylamine hydrochloride (Aldrich Chemical Co., Gillingham, Dorset,
UK); salicylhydroxamic acid (Koch-Light Laboratories, Colnbrook, Middlesex, UK). Soluble aspirin
tablets B.P., containing 300mg aspirin (Disprin, Reckitt and Colman, Hull, UK) were supplied by the
Pharmacy of St Mary’s Hospital.

Human investigation
The investigation used a panel of healthy volunteers, drawn from the students and staff of St Mary’s
Hospital and Medical School, made up of 67 males and 62 females, ages 19-35 years (mean 21.1), body wt
41-92 kg (mean 65.4). The subjects had taken no drugs for at least seven days before their participation in
the study; no other exclusion criteria were applied.
Each subject took a single oral dose of 900 mg aspirin dissolved in water (3 x 300 mg soluble aspirin
For personal use only.

tablets B.P., equivalent to 690mg salicyclic acid) at 9 a.m. following a light breakfast. They collected their
urine for 12 h after dosing. The volume and pH of the sample was recorded at the time of collection, and
samples stored at -20°C until analysed.
The study was approved by the local Ethics Committee, and all volunteers gave their informed consent
to the experimental procedure.

Analysis of urinary metabolites of aspirin

High-pressure liquid chromatography. This employed a Waters Associates (Harrow, Middlesex, UK)
WISP Model 710B autoinjector, model 6000A pump and model 720 systems controller, l i n k d with a
Cecil Instruments (Cambridge, UK) 2012UV detector set at 310nm, and a Waters model 730 data
module. The column was of stainless steel, 250 x 5 mm, packed with ODs-Hypersil 5 p (Shandon
Southern Products, Runcorn, Cheshire, UK) and the mobile phase was methanol-glacial acetic
acid-water (25:4: 71, by vol.), flow rate 1.5 mljmin. The retention times (min) of standard compounds in
this system were as follows: gentisic acid, 4.7; salicyluric acid, 6.7; 2-methoxybenzoic acid, 9.4; salicylic
acid, 16.8.
Assay of salicylic, salicyluric andgentisic acids. T o aliquots (1 or 2 ml) of urine were added the internal
standard, 2-methoxybenzoic acid (1 ml of an 0.5 mg/ml solution), the pH of the mixture adjusted to 1 by
the drop-wise addition of 6~ HC1 and the whole extracted with diethyl ether (2 x4ml). The ethereal
extracts were combined and evaporated to dryness, care being taken to prevent losses due to the volatility
and sublimation of salicylic acid (Goehl, DeWoody and Sunclaresen 19811, and the residue was taken up
in h.p.1.c. mobile phase (2ml) before examination by h.p.1.c. (injection volume 2 0 4 ) . The quantity of
each metabolite present was determined by reference to previously established calibration curves relating
concentration to peak area ratios for each compound to the internal standard. Regression analysis of these
lines gave calibration coefficients of not less than 0.998. The calibration curves were routinely established
each time the assay was performed and all samples were analysed in duplicate. Under these conditions the
recoveries of the compounds was as follows (mean+S.D.): salicylic acid (lOOpgjml), 9 2 k 1.6O/,;
salicyluric acid (500pg/ml), 8 6 k 1.5%; gentisic acid (lOOpgjml), 91 1.0%; 2-methoxybenzoic acid,
92+ 1.2% ( n = 4 in all cases).
Assay of total salicylic acid-content of urine. Urine (1 or 2 ml, depending on the vol. of the sample) was
mixed with 6~ HC1 (2ml) and the whole heated at 120°C for 17 h in a sealed tube. Following this, the
contents of the tube were assayed for salicylic acid as described previously.
Assay of salicyl acyl glucuronide. This used a modification of the method of Schachter (1957a,b).
Neutral hydroxylamine soh (0.5 ml of a 2 M soh in 0.2 M acetate buffer, p H 7) was added to urine (2 ml)
and the mixture left to stand at room temp. for two hours, after which 6 M HCI (0.5 ml) and ferric chloride
(0.5 ml of a 5 % s o h in 0.1 M HCI) were added. The optical density of the s o h was determined at 540nm
(Pye Unicam SP30 spectrophotometer) and the salicylhydroxamic acid formed determined by reference
to a previously established calibration curve prepared using salicylhydroxamic acid dissolved in blank
 242 A. J . Hutt et al.
urine (see Schachter 1957a, b). These curves were linear over the concentration range 0-1OOpg salicylic
acid equivalents/ml, with correlation coefficients always > 0,997.
Assay of salicyluric acid phenolic glucuronide. Urine ( 2 ml) was adjusted to p H 1 by the drop-wise
addition of 6 M HCI and extracted with diethyl ether (3 x 4 ml) and the organic extracts discarded. T h e p H
of the aq. phase was then adjusted to 5 with 5 M NaOH, and acetate buffer (1 ml of 0.5 M , p H 5 ) and p-
glucuronidase (Glucurase, 5000 units/ml; Sigma Chemical Co.) were added. T h e whole was incubated at
37°C for 17 h, and then subjected to the previously described assay for salicyluric acid. These conditions
were known to be optimal for the hydrolysis of the conjugate, as established in our previous study with
14C-aspirinin man (Hutt et al. 1982). Control experiments established that all of the free salicyluric acid
present in urine was extracted into ether under the conditions empioyed, and the identity of the conjugate
as the glucuronide of salicyluric acid confirmed by the complete inhibition of its hydrolysis under these
conditions by saccharo-1,4-lactone (Hutt et al. 1982).
Determination of salicyl phenolicglucuronide. In the absence of an authentic sample of this compound,
this metabolite was estimated by the difference between the total salicylic acid content of urine,
determined by acid hydrolysis (see above), and the sum of all other salicyl-containing metabolites,
Xenobiotica Downloaded from informahealthcare.com by UB der LMU Muenchen on 06/18/14

determined by h.p.1.c. or derivative formation.

Validation of assay methods. I t is known from a study with 14C-aspirin that the methods described here
assay all known metabolites (Hutt et al. 1982). Urines from that study were analysed for salicylic acid and
its metabolites with the present methods. The results obtained with the assay methods used here
correlated very well (r=0.96) with those obtained with independent techniques in the 14C study. All
assays in this study were performed in duplicate, and replicates were always within 6'g of each other.

Statistical analysis of data

Data were analysed for significance by the unpaired Student t test, and the Spearman Rank
Correlation, as appropriate. T h e normality of distributions was tested by probit analysis according to
Finney (1962).

Results
For personal use only.

Total recovery of salicylate and metabolites

T h e total recovery of salicylate and its metabolites in the 0-12h urine of
volunteers taking 900mg aspirin varied between 42 and 104% of the dose, with a
mean of 68.1 yo.T h e recovery of salicylate and metabolites was normally distributed
in the population studied (figure 2).

Excretion of salicylate and metabolites

T h e excretion of salicylic acid was highly variable within the group, accounting
for between 1.3 and 31 yo of the dose (mean 8.3y0),and exhibited a skewed normal
distribution (figure 3).
T h e major metabolite in the urine of the volunteers was salicyluric acid, present
as such and also as its phenolic glucuronide. T h e recovery of free salicyluric acid
varied between 19.8 and 65.3% (mean 45.0%) of the dose, and was normally
distributed within the population studied (figure 4). In addition, the double
conjugate salicyluric acid phenolic glucuronide was present in the urine of 114 of the
129 subjects, accounting for 0.5 to 6.3% (mean 3.2%) of the dose. T h e recovery of
this secondary metabolite exhibited a skewed normal distribution (data not shown).
T h e excretion of salicyl acyl glucuronide varied between 1 and 12% (mean 5.6%)
of the dose, assayed by the formation of salicylhydroxamic acid upon treatment of
the urine with hydroxylamine. T h e excretion of this metabolite exhibited a skewed
normal distribution in the study group, and this is shown in figure 5.
Evidence was obtained for the presence of salicyl phenolic glucuronide in the
urine of 122 of the 129 individuals studied, the recoveries ranging from 0 (in seven
subjects) to 37.9% (mean 7.2%) of the dose. T h e excretion of this metabolite
exhibited a skewed normal distribution in this population (figure 6).
T h e oxidation product, gentisic acid, was a minor metabolite in the urine of 117
of the 129 volunteers, varying between 0.4 and 2.3% (mean 1.0%) of the dose, and
following a normal distribution in the study group.
 Population study of aspirin metabolism 243

30.

**

20
+
0
z
ILI
3
Q n
w
CY
LL

I
Xenobiotica Downloaded from informahealthcare.com by UB der LMU Muenchen on 06/18/14

lo

-
I.

0 20 40 60 80 100 120
% TOTFlL RECOVERY

Figure 2. Frequency distribution of the total recovery of salicylate and metabolites in the 0-12 h urine
following a 900mg oral dose of aspirin to 129 subjects.
For personal use only.

30

20
+
u
z
W
3
,a
W
n
E
LL
10

0
10 20 30 40 50
% DOSE EXCRETED FlS SRLICYLIC RCIO
Figure 3. Frequency distribution of the excretion of salicylic acid in the 0-12h urine of 129 subjects
given an oral dose of 900 mg aspirin.

Table 1 presents quantitative details of the urinary excretion of salicylate and its
metabolites in this panel of subjects.

Factors affecting the elimination of salicylate and its metabolites

Urine volume and p H . Possible correlations between the percentage of dose
excreted as each metabolite and the volume and pH of the urine were examined with
 244 A. J . Hutt et al.

30 3

7
x x
x
x D
20 m
*
0
SOm
u
z -4
W ul
3
Q
W
CL 3
LL
1C
Xenobiotica Downloaded from informahealthcare.com by UB der LMU Muenchen on 06/18/14

1
1

c
0 20 40 60 80 100
% DOSE EXCRETED f l S S A L I C Y L U R I C A C I D

Figure 4. Frequency distribution of the excretion of salicyluric acid in the 0-12 h urine of 129 subjects
given an oral dose of 900 mg aspirin.
For personal use only.

30

20
+
0
z
w x
3
0 x a
W
CY
LL
1C

Z DOSE EXCRETED AS S A L I C Y L RCYL GLUCURONIDE

Figure 5. Frequency distribution of the recovery of salicyl acyl glucuronide in the 0-12 h urine of 129
volunteers given an oral dose of 900mg aspirin.

the Spearman Rank Correlation. T h e only statistically significant relationships to

emerge with urine volume were with salicylic acid excretion (R(s)0.539, P < O . O O l ) ,
and the total recovery of salicylate and its metabolites (R(s)0.429, P < O . O O l ) .
The p H of the urine was significantly positively correlated with salicylic acid
excretion (R(s)0.750, P < O . O O l ) . There was also a weak but statistically significant
negative correlation between urine p H and the percentage of dose eliminated as
 Population study of aspirin metabolism 245

30 I i9
Xenobiotica Downloaded from informahealthcare.com by UB der LMU Muenchen on 06/18/14

0 10 20 30 40 50
% DOSE EXCRETED RS SRLICYL PHENOLIC GLUCURONIDE
Figure 6. Frequency distribution of the recovery of salicyl phenolic glucuronide in the 0-12 h urine of
129 volunteers given an oral dose of 900 mg aspirin.
For personal use only.

Table 1. Urinary excretion of salicylate and metabolites in the 0-12 h urine of 129 healthy volunteers
each given 900mg aspirin orally.

yo Dose excreted
Number of subjects in
Metabolite which metabolite
detected Mean fS.D. Range

Salicylic acid 129 8.3 5.9 I .3-31.1

Gentisic acid 117 1.Of 0.4 0.42.3
Salicyluric acid 129 45,Ok 10.6 19.8-65 '3
Salicyluric acid
phenolic glucuronide 114 3.2+ 1.3 0.5-63
Salicyl acyl
glucuronide 115 +
5.6 2.4 1+12.3
Salicyl phenolic
glucuronide? 129 7.2k7.6 0-37.9
Total salicylate 129 68.1k11.9 42.3-104

?Values for salicyl phenolic glucuronide were calculated as outlined in the text.

salicyl acyl glucuronide (R(s)-0.338, P<O.OOl), which may be due to the more
facile hydrolysis of ester glucuronides under mild alkaline conditions (Caldwell et al.
1983). NQother statistically significant relationships were revealed by this analysis of
the data.

Gender. T h e excretion of salicylate and its metabolites in the two sexes was
compared by the unpaired Student t test, and the data are presented in table 2.
Females excreted significantly more gentisic acid than males, while the reverse was
the case for salicyluric acid. T h e total recovery of salicylate and metabolites was
greater in males, presumably due to the increased excretion of salicylurate. No
 246 A. J . Hutt et al.

Table 2. Sex differences in the urinary excretion of salicylate and metabolites in the G12 h urine of 129
healthy volunteers given 900 mg aspirin orally.

% Dose excreted in
(mean+S.D.)

Metabolite Males? Females$ P§

Salicylic acid 8.1 f6.5 8.4k5.3 n.s.
Gentisic acid 0.9 k 0.3 1 . 1 k0.4 <0.001
Salicyluric acid 46.9f 10.6 43.0k 10.2 0.05>P>0.02
Salicyluric acid
phenolic glucuronide 3.4k1.4 2.9 k 0.2 0.05 >P > 0.02
Salicyl acyl glucuronide 5.2k2.2 6.0 k 2.5 n.s.
Xenobiotica Downloaded from informahealthcare.com by UB der LMU Muenchen on 06/18/14

Salicyl phenolic
glucuronide 11 +
8.0 8.2 6.3k 6.8 ns.
Total salicylate 70.2k11.6 65.9+12.0 0.05 > P> 0.02

+
t N = 6 7 ; mean age, 21.3 3.4 years; mean weight, 73.0f9.0kg; mean urine val., 843 +458ml; mean
pH value, 6.5 f0.7.
$ N = 6 2 ; mean age, 21.0f3.6 years; mean weight, 57.1 k 6 . 1 kg; mean urine vol., 813k376ml; mean
pH value, 6.3 kO.5.
§Unpaired t test, level of significance P=O.O5.
I/Salicyl phenolic glucuronide values were calculated as outlined in the text.
For personal use only.

statistically significant differences at the P= 0.05 level were observed between the
sexes in the excretion of salicylic acid or its glucuronides.
Competing metabolic options. The major routes of salicylic acid metabolism are
conjugation with glycine at the -COOH group, and glucuronidation at either the
-COOH or -OH groups. The formation of the two glucuronides appears to
represent the principal metabolic alternative to glycine conjugation, which nearly
always predominates. In only five of the 129 subjects in this study did the sum of the
two glucuronides exceed the percentage dose eliminated as salicyluric acid. T h e sum
of both the salicyl glucuronides ranges from 0.8 to 41.6% (mean 12.3%) of the dose,
and in only one individual was neither detected. It must be stressed that this subject
did excrete 5.5yoof the administered dose as the glucuronide of salicyluric acid.
Spearman rank correlation revealed an inverse relationship between salicyluric acid
elimination and the excretion of the two glucuronides (R(s)-0.389, P<O.OOl). T h e
excretion of salicyluric acid was related to the total recovery of salicylate and
metabolites (R(s)0.675, P<O.OOl),which is to be expected since salicyluric acid is
nearly always the major excretion product. However, the sum of the two glu-
curonides was unrelated to the total elimination of salicylate.

Discussion
The findings presented in this paper show that the excretion of salicylate and its
metabolites following the administration of aspirin to a panel of 129 healthy
volunteers is highly variable. This confirms both expectations from the literature
based on small groups of subjects (for references, see Introduction) and a previous
report from this laboratory based on a separate group of 85 volunteers (Caldwell,
O’Gorman and Smith 1980). The variability within the population, like that in our
previous preliminary report (Caldwell et al. 1980), was normally distributed.
As would be expected from the literature, the major metabolite was the glycine
conjugate, salicyluric acid, which was excreted principally free but also to a small
 Population study of aspirin metabolism 247

extent as its phenolic glucuronide. The formation of this glucuronide as a metabolite

of aspirin was demonstrated by Hutt et al. (1982).
T h e urinary excretion of salicyluric acid would appear to offer an index of the
glycine-conjugative capacity of a subject, as its elimination is relatively reproducible
within individuals (Caldwell et al. 1980). Thus, when administered to volunteers, it
is not cleaved to salicylic acid (Bochner et al. 1981) and has been reported to be
rapidly and extensively excreted unchanged (Bedford, Cummings and Martin 1965,
Levy and Amsel 1966, Levy, Amsel and Elliott 1969, Bochner et al. 1981).
In order of quantitative significance, the next most important metabolites are the
two glucuronides of salicylic acid. T h e elimination of salicyl acyl glucuronide was
normally distributed in the population, while that of salicyl phenolic glucuronide
Xenobiotica Downloaded from informahealthcare.com by UB der LMU Muenchen on 06/18/14

showed a skewed normal distribution. However, the distribution of the sum of these
two glucuronides did not follow a normal distribution by probit analysis (data not
shown). T h e significance of this is unclear. T h e total formation of both these
glucuronides (acyl plus phenolic) was inversely related to the elimination of
salicyluric acid, indicating that these two reactions are alternative metabolic options
for the salicylic acid molecule. T h e formation of the acyl glucuronide was always less
than that of the glycine conjugate, both metabolites involving conjugation of the
-COOH group. T h e glucuronides are formed by the UDP-glucuronyl transfer-
ases located in the endoplasmic reticulum (Kasper and Henton 1980) whereas the
glycine-conjugating mechanisms are found in the mitochondria (Killenberg and
For personal use only.

Webster 1980), and the choice between these two alternative options may involve
aspects of the subcellular distribution of salicylate (see Dixon, Caldwell and Smith
1977).
T h e two glucuronides of salicylate are also noteworthy from the analytical
viewpoint. Authentic standards of these compounds are not available, and the
majority of the work reported in the literature has not attempted to differentiate
between these metabolites of salicylate. T h e present report would appear to be the
first to do so since the work of Levy some 1 2 years ago (Levy, Tsuchiya and Amsel
1972).
Acyl glucuronides are known to be unstable, undergoing both hydrolysis to the
aglycone and glucuronic acid and the interesting phenomenon of acyl migration, in
which the biosynthetic 1-0-acyl glucopyranosiduronates undergo intramolecular
rearrangement in mild (pH 7 and above) alkaline conditions to yield the correspond-
ing 8-glucuronidase-resistant2-, 3- and 4-glucuronic acid esters (see Heirwegh and
Compernolle 1979, Sinclair and Caldwell1982, Caldwell et al. 1983). T h e use here of
hydroxamic acid formation for the analysis of the ester glucuronide, rather than
reliance upon enzymic hydrolysis, obviates analytical problems which might be
caused by the migration of the salicyl moiety around the sugar ring. Care was also
taken to prevent errors due to the spontaneous hydrolysis of the acyl glucuronide by
either analysing the sample immediately upon collection, or by freezing at - 20°C.
The phenolic glucuronide was estimated by difference, following the hydrolysis of
metabolites with concentrated acid, an approach validated by our previous study of
the fate of I4C-aspirin in man (Hutt et al. 1982).
The oxidation product gentisic acid was found to be a minor metabolite of
aspirin, in agreement with the results of Levy et al. (1972) and Bochner et al. (1981).
T h e excretion of salicylic acid by these volunteers was highly variable, and in this
case the variability was related to the urinary p H and volume. It is important to
appreciate that the metabolic fate of that fraction of the salicylate dose which does
 A. J. Hutt et al.

undergo biotransformation is independent of the percentage eliminated as free

salicylate. Thus, although the amount of free salicylate present is highly variable,
this variation is not consistently related to a relative inability to form one or other
metabolite(s), and the variation in salicylate excretion appears to arise more from
physicochemical factors involving urine volume and p H than from differences on the
activities of the various metabolic pathways.
T h e results of this study reveal the occurrence of considerable inter-individual
variation in the elimination of salicylic acid and its metabolites by volunteers given
900 mg of aspirin, and it is of interest to consider possible origins of this variability. It
is clear that, with the exception of salicylic acid excretion, urine p H and volume have
little meaningful impact on the pattern of metabolites excreted.
Xenobiotica Downloaded from informahealthcare.com by UB der LMU Muenchen on 06/18/14

The influence of gender upon the fate of aspirin is the subject of controversy in
the literature (see literature cited in the Introduction). This study shows the existence
in this panel of small but statistically significant differences between the sexes in the
elimination of gentisic acid, salicyluric acid and its phenolic glucuronide, and total
salicylate (see table 2). These differences just may be relevant to the higher incidence
of aspirin intolerance in females (Reynolds 1982).
It is likely that the explanation for the observed inter-individual differences in
aspirin metabolism lies in differences in the relative activities of the major metabolic
pathways open to the salicylic acid molecule, which in turn presumably arise from
differences in the activities of the enzymes catalysing the reactions. This is hard to
For personal use only.

assess on the basis of the present data, as the various routes of metabolism exhibit
different kinetic characteristics: thus, at the present dose level (a reasonable one in
terms of the intermittent use of aspirin as an analgesic/anti-inflammatory), studies in
very small numbers of subjects have shown that the glycine conjugation and phenolic
glucuronidation follow zero-order (i.e., capacity-limited) kinetics, while the other
pathways follow first-order kinetics (see Levy 1979, Bochner et al. 1981). It appears
from the present data that the activities of the individual metabolic pathways vary
independently, and that the extensive elimination of one product does not necessari-
ly result in a low excretion of others. It is apparent that the percentage dose
eliminated as salicylic acid has little impact on the pattern of disposition of that
fraction of the dose undergoing metabolism. However, there was a poor
(R(s)- 0.39) but significant ( P <0.001) inverse relationship between the elimin-
ation of salicyluric acid and that of salicyl glucuronides. No other rela-
tionships emerged upon rank correlation of the excretion of salicylic acid with
that of the various metabolites. T h e present paper has defined the nature and extent
of the inter-individual differences in the handling of aspirin by a group of 129 young
healthy volunteers, and the kinetic aspects of this are addressed in a subsequent
paper (Hutt et al. to be submitted).

Acknowledgements
This work was supported by a grant from the Sir Halley Stewart Trust. We are
grateful to D r N. S. Oates for help with statistical analysis.

References
ALPEN,E. L., MANDEL, H. G., RODWELL, V. W., and SMITH, P. K., 1951, J. Pharmac. exp. Ther., 102,
150.
ANGELO-KHATTER, M. M., AL-KHARO, A,, and AL-NASSER,K. E. A,, 1981, Eighth International
Congress on Pharmacology, Tokyo, Abstract no. 1659.
BEDFORD,C . , CUMMINGS,A. J., and MARTIN,B. K., 1965, BY.J. Pharrnac., 24,418.
 Population study of aspirin metabolism 249

BENNETT,P. N., NOTARIANNI, L. J., OLDHAM, H. G., and OGUNBONA, F. A., 982, Eighth European
Workshop on Drug Metabolism, Liege, Belgium, Abstract no. 292.
BOCHNER, F., GRAHAM, G. G., CHAM,B. E., IMHOFF, D. M., and HAAVISTO, I. M 1981, Clin. Pharmac.
Ther., 30,266.
CALDWELL, J., HUTT,A. J., MARSH,M. V., and SINCLAIR, K. A,, 1983. In Drug Metabolite Isolation and
Determination, edited by E. Reid and J. P. Leppard (New York: Plenum Press), p. 161.
CALDWELL, J., O'GORMAN, J., and SMITH,R. L., 1980, BT.J. d i n . Pharmac., 9, 114P.
DAY,R. D., SHEN,D. D., and AZARNOFF, D. L., 1983, Clin.Pharmacokin., 8 , 2 6 3 .
DIXON,P. A. F., CALDWELL, J., and SMITH,R. L., 1977, Xenobiotica, 7, 727.
EVANS,D. A. P., and CLARKE, C. A,, 1961, B T . med. Bull., 17,234.
FINNEY, D. J., 1962, Probit Analysis (Cambridge: Cambridge University Press).
FURST, D. E., GUPTA,N., and PAULUS, H. E., 1977, J. clin. Invest., 60, 32.
GOEHL,T. J., DEWOODY, C. T., and SUNCLARESEN, G. H., 1981, Clin. Chem., 27,776.
GRAHAM,G. G., CHAMPION, G. D., DAY,R. O., and PAULL,P. D., 1977, Clin. Pharmac. Ther., 22,410.
Xenobiotica Downloaded from informahealthcare.com by UB der LMU Muenchen on 06/18/14

HEIRWEGH, K. P. H., and COMPERNOLLE, F., 1979, Biochem. Pharmac., 28, 2109.
HUTT,A. J., CALDWELL, J., and SMITH,R. L., 1982, Xenobiotica, 12,601.
KAPP,E. M., and COBURN, A. F., 1942, J. biol. Chem., 145,549.
KASPER,C. B., and HENTON,D., 1980. In Enzymatic BasisojDetoxication, edited by W. B. Jakoby (New
York: Academic Press), vol. 11, p. 3.
KILLENBERG, P. G., and WEBSTER, JR. L. T., 1980. In Enzymatic Basic OjDetoxication, edited by W. B.
Jakoby (New York: Academic Press), vol. 11, p. 141.
LEVY,G., 1979, Drug Metab. Rev., 9, 3.
LEVY,G., and AMSEL,L. P., 1966, Biochem. Pharmac., 15,1033.
LEVY,G . ,AMSEL,L. P., and ELLIOT,H. C., 1969, J . Pharm. Sci., 58, 827.
LEVY,G., TSUCHIYA, T., and AMSEL,L. P., 1972, Clin. Pharmac. Ther., 13,258.
MENGUY, R., DESBAILLETS, L., MASTERS,Y. F., and OKABE,S., 1972, Nature, 239,102.
MIASKEWICZ, S. L., SHIVELY, C. A,, and VESELL,E. S., 1982, Clin. Pharmac. Ther., 31,30.
For personal use only.

MONTGOMERY, P. R., and SITAR,D. S., 1981, Gerontology, 27, 329.

NOTARIANNI, L. J., OGUNBONA, F. A,, OLDHAM,H. G., PATEL, D. K., BENNETT,P. N., and
HUMPHRIES, S.J., 1983, B Y . J . clin. Pharmac., 15, 587P.
RAINSFORD, K. D., FORD, N. L. V., BROOKS, P. M., and WATSON,H. M., 1980, EUT.J . Clin. Invest., 10,
413.
REYNOLDS, J. E. F. (editor), 1982, Martindale. The Extra Pharmacopoiea, (London: The Pharmaceutical
Press), p. 234.
SCHACHTER, D., 1957a, Science, 126, 507.
SCHACHTER, D., 1957b, J. clin. Invest., 36, 297.
SINCLAIR, K. A., and CALDWELL, J., 1982, Biochem. Pharmac., 31,953.