The inhibition by clonidine of peripheral noradrenergic neurotransmission is a

frequency-dependent phenomenon: clonidine being more effective at low and

intermediate frequencies ofnerve stimulation (Kobinger, 1967; Armstrong & Boura,

1973). In fact at high frequencies of nerve stimulation, clonidine appears to act as a

partial agonist at presynaptic a-adrenoceptors and it either fails to modify, or even

increases, the stimulation-evoked release of noradrenaline (Medgett , McCulloch &

Rand,I978).

Stimulation by clonidine of a 2-adrenoceptors in the central nervous system

appears to be the rnajor mechanism through which this drug produces hypotension

and bradycardia (Kobinger, 1978). Since these central effects ofclonidine persist after

depletion ofthe endogenous noradrenaline stores by pretreatment with reserpine and

a -methyl-p-tyrosine, it is generally believed that central postsynaptic

a -adrenoceptors (Figure la) are the target of the cardiovascular inhibitory effects of

clonidine (Kobinger, 1978).

Although the exact localisation of the central a s-adrenoceptors that mediate the

cardiovascular effects of clonidine is still the subject of somecontroversy, the

available evidence indicates that these a2-adrenoceptors are located within the

pontomedullary region. The possible role ofthe area ofthe nucleus tractus solitarii in

the pharmacological effects of clonidine remains an open question. Clonidine

administered systemically reduces the turnover of noradrenaline and adrenaline in

several areas of the rat brain (Scatton, Pelayo, Dubocovich, Langer & 8artholini,

1979). These effects of clonidine are antagonized by low doses of yohimbine and are

resistant to blockade by prazosin, indicating that the adrenoceptors involved in this

effect are of the (12-subtype. It is possible that the decrease in turnover of

noradrenaline and adrenaline produced by clonidine is linked to the presynaptic

inhibition by the (1 radrenoceptor agonist of the release of noradrenaline (Pelayo et

al., 1980) and adrenaline (Scatton et al., 1979). This is shown schematically in Figure

Ib).

Receptor binding studies and the anatomicallocalisation

ofthe a2-adrenoceptor

In spite of the fact that clonidine can decrease noradrenergic neurotransmission by

acting on presynaptic inhibitory a radrenoceptors in the periphery and in the central

nervous system, the specific binding of[3H]-clonidine in the central nervous systern is

unaffected by chemical sympathectomy with 6-hydroxydopamine (U'Prichard,

8echtel, Rouot & Snyder, 1979). Such results suggest that the majority of the (12-

adrenoceptors in the central nervous system are located postsynaptically. However,

one should be cautious when agonist binding studies are used to establish changes in

receptor number and these studies should ideally be carried out with an antagonist

ligand like [3H]-yohimbine which has recently become available in a high specific

activity form.

160 S. Z. LANGER & R. MASSINGHAM

Recently, it was reported that after short term surgical sympathetic denervation of

the rat submaxillary gland, there is a significant increase in both apparent affmity and

in maximal binding of [3H]-c1onidine (Pimoule, Briley & Langer, 1980). As early as

24 or 48 h after surgical denervation, the affmity for [3H]-c1onidine binding is

increased threefold and the number of sites by about 60% (Pimoule et al.. 1980).

These early changes in [3H]-c1onidine binding after denervation complicates the

a

CNS

PERIPHERY

HEART

Inhibitory

neurone

Ganglion

VSM

Figures la and b Diagrammatic representations of the possible central and peripheral sites of

action of clonidine which could be involved in the drug's hypotensive, bradycardic and sideetfect profile,

Abbreviations used:

a l : a ,-adrenoceptor

a, : aradrenoceptor

ßI : ßI-adrenoceptor

\rSM : vascular smooth muscle

NA: noradrenaline

ACh : acetylcholine

CNS : central nervous system

ganglion: sympathetic ganglion.

a -ADRENOCEPTORS AND CLONIDINE 161

interpretation of experiments using surgical or chemical sympathetic denervation to

establish the pre- or postsynaptic localisation of uradrenoceptors. As suggested

earlier, such experiments should ideally be carried out with an antagonist ligand and

with a careful study ofthe time course ofthe efTects after denervation. The only study

in which chemical sympathectomy with 6-hydroxydopamine was shown to reduce

the maximal binding of an a-adrenoceptor ligand was carried out in the rat heart,

using (lH]-dihydroergocryptine as an antagonist ligand for a 2-adrenoceptors (Story,

Briley & Langer, 1979) .

b

CNS

PERIPHERY

HEART VSM

Excilalory

neurone

Ganglion

Stimulation of a 2-adrenoceptors within the CNS appears to be responsible for both the drug's

hypotensive and sedative properties. Clonidine may reduce sympathetic drive by activating

postsynaptic a 2-adrenoceptors located on an inhibitory neurone (Figure la) or by stimulating

presynaptic inhibitory a 2-adrenoceptors on nerves having an input to an excitatory neurone

(Figure Ib). In addition, stimulation of peripheral presynaptic a radrenoceptors on sympathetic

nerve terminals mayaiso contribute to the bradycardic action ofthis drug. Finally, stimulation

of a 2-adrenoceptors located on parasympathetic neurones in the CNS (Figure Ib) or in the

periphery may be involved in the dry mouth syndrome produced by this drug in man. The role

of ganglionic a2-adrenoceptors in the haemodynamicchanges produced by clonidine remain to

be clarified, (Seetext for further explanations .)

162 S. Z. LANGER & R. MASSINGHAM

Side effects of clonidine under acute or chronic administration

Sedation and dry mouth are the most prominent and important side efTects observed

during the administration of clonidine to man. lt is possible that the sedation is

linked to the decreased central turnover of noradrenaline producedby tJiis drug.

Furthermore, many pharmacological studies suggest that aradrenoceptor

stimulation is involved in the sedative action of clonidine. Studies in chicks have

demonstrated that the sleep-inducing action of this drug is antagonized by phentolamine, a non-selective a-adrenoceptor antagonist, but not by prazosin, a selective

a,-adrenoceptor blocking agent (Cavero & Roach, 1978). It is also possible that

stimulation ofhistamine Hy-receptors by clonidine and other imidazoline derivatives

could be related to the sedative efTectsof these agents (McCulloch, Medgett, Rand &

Story, 1980) but it should be stressed that stimulation of Hy-receptors by clonidine

and close analogues is not causally related to the hypotensive action of these drugs

(McCulloch et al.. 1980).

Inhibition by clonidine of acetylcholine release in the periphery and in the central

nervous system through the activation of presynaptic aradrenoceptors may be

involved in the decrease in salivary secretion which results in dry mouth, a comrnon

side efTect with clonidine (Figure Ib). Indeed, in cats, clonidine inhibits submaxillary

salivation evoked by peripheral parasympathetic nerve stimulation by reducing

cholinergic transmission through activation ofpresynaptic aradrenoceptors (Green,

Wilson & Yates, 1979).

The abrupt withdrawal of clonidine after chronic administration during antihypertensive therapy can precipitate a hypertensive crisis, with tachycardia and

increased levels of catecholamines in the blood and urine. The mechanism

underlying this hypertension following withdrawal after chronic administration of

clonidine remains an open question, but it is tempting to speculate as folIows :

Chronic administration of clonidine produces a subsensitivity of a 2-adrenoceptors

as a result of a long lasting activation. At the same time, the rcduction in noradrenergic neurotransmission produced by clonidine acting on central and peripheral ar

adrenoceptors results in the deve!opment of supersensitivity at the level of the

postsynaptic a ,-adrenoceptors and ßI-adrenoceptors present in vascular smooth

muscle and cardiac muscle respectively (see Figure I). Consequently, when the

administration of clonidine is interrupted, there is an increase in noradrenaline

release with hypertension and tachycardia due to a subsensitivity of the inhibitory

presynaptic a radrenoceptors and a supersensitivity of postsynaptic a 1- and

ß,-adrenoceptors (Figure la and b).

Conclusions

The most important mechanism involved in the antihypertensive and bradycardic

efTects of clonidine appears to involve a stimulation of central a radrenoceptors.

These a-adrenoceptors probably have a postsynaptic location at the level of the

pontomedullary region within the central nervous system.

Stimulation of presynaptic aradrenoceptors on peripheral noradrenergic nerve

endings may, however, contribute to the decrease in sympathetic tone elicited by

c1onidine. At the level of the heart it appears that inhibition of noradrenaline release

through the activation of pre synaptic inhibitory a radrenoceptors can contribute

significantly to the bradycardia produced by c1onidine.

The therapeutic efTectiveness of clonidine in the treatment of the abstinence

syndrome, triggered by acute opiate withdrawal, appears to be related to an

inhibition of hyperactive centra1 noradrenergic neurones through the activation of

inhibitory a2-adrenoceptors (Gold, Redmond & Kleber, 1978).

Although the presence of a 2-adrenoceptors in sympathetic ganglia mediating

a -ADRENOCEPTORS AND CLO NIDINE 163

hyperpolarization ha s been recently reported, (Brown & Ca ulfield, 1979) (see Figure

1), there is as yet no evidence for the involveme nt of these receptors in the cardiova scular effects ofclonidinc.

References

Armstrong, J. M. & Boura, A. L. A. (1973). Effects ofclonidine and guanethidine on peripheral

sympathetic nerve function in the pithed rat. Brit. J. Pharrnac.. 47, 850-852.

Brown, D. A. & Caulfield , M. P. (1 979). Hyperpolarizing a 2-adrenoceptors in rat sympathetic

ganglia. Brit. J. Pharrnac., 65,435-446.

Cavero, I. & Roach , A. G. (1 978). Th e effects of prazosin on the clonidine-induced hypotension

and bradycardia in rats and sedation in chicks. Brit. J. Pharmac., 62, 468P .

Cavero, I. & Roach, A. G. (19 80). The effects of clonidine on canine cardiac neuroeffector

structures controlling heart rate. Brit. J. Pharmac., in press.

Gold , M. S., Redmond, D. E. & Kleber, H. D. (1978). Clonidine blocks acute opiate-withdrawal

symptoms. Lancet, 2, 599-{i02.

Green, G. J., Wilson, M. & Yates, M. S. (1979). The mechanism of the clonidine-induc ed

reduction in periph eral parasympatheti c submaxillary salivation. Eur. J. Pharrnac. , 56,

331-345 .

Kobinger, W. (1967). Uber den Wirkun gsmechanismus einer neuen antihypertensiven Substanz

mit Imidazolinstructure. Naunyn -Sch m iedeberg's Arch. Pharmac., 258,48-5 8.

Kobinger W. (1978). Central a -adrenergic systems as targets for ant ihypertensive drugs. Rev.

Physio/. Biochem. Pharrnac.. 81,40-100.

Langer, S. Z. (1974). Presynapt ic regulation ofcatecholamine release. Biochem. Pharrnac.. 23,

1793-1800.

Langer, S. Z. (1977). Presynaptic receptors and their role in the regulation of transmitter release.

Sixth Gaddum Memorial Lecture. Brit, J. Pharmac.. 60,481-497.

Langer, S. Z. (1979). Presynaptic recepto rs and the regulation of transmitter release in the

peripheral and central nervous system: physiological and pharmacological significance. In

Catecholamines: basic and clinicalfrontiers, Vol. I, ed. Usdin, E. pp. 387- 398. New York:

Pergamon Press.

Langer, S. Z. (1 980). Presynaptic regulation ofthe release of catechola mines. Pharmac. Rev., in

press.

Langer, S. Z., Cavero, I. & Massingham , R. (1980). Recent developments in noradrenergic

neurotransmission and its relevance to the mechanism of action of certain antihypertensive

agents. Hypertension. in press.

Langer, S. Z. & Dubocovich, M. L. (1 980). Cocaine and amphetamine antagonize the decrease

of noradrenergic neurot ransmission elicited by oxymetazoline but potentiate the

inhibition by alpha-methylnorepinephrine in the perfused cat spleen. J. Pharmac. exp.

Thet, in press.

Langer, S. Z. & Luchelli-Fort is, M. A. (1977). Subsensitivity of the presynaptic a-adrenoceptors

after short term surgical denervation of the cat nictitating membrane. 1. Pharrnac. Ther..

202, 6IQ-{)2 I.

McCulloch, M. W., Medgett, I. c., Rand , M. J. & Story, D. F. (1 980). Structure-activity

relation ship of imidazoline derivatives related to clonidin e at histamine Hj-receptors in

guinea-pig isolated atria . Brit, J. Pharma c., 69, 397-406.

Medgett, I. c., McCulloch , M. W. & Rand , M. J. (19 78). Partial agonist action of clonidine on

prejunctional and postjun ctional a -adrenoceptors. Naunyn-Schmiedeberg's Arch.

Pharrnac.. 304,215-221.

Pelayo, F., Dubocovich, M. L. & Langer, S. Z. (1980). Inhibition of neuronal uptake reduces the

presynapt ic effects of clonidine but not of a -meth ylnoradrenaline on the stimulationevoked release of[ JHj-noradrenaline from rat occipital cortex slices. Eur. J. Pharrnac.. 64,

143-155.

Pimoule, c., Briley, M. S. & Langer, S. Z. (1980). Short term surgical denervation increases

(lH]-clonid ine binding in rat salivary gland. Eur. J. Pharrnac.. 63, 85- 87.

Scatton, B., Pelayo. F., Dubocovich, M. L., Langer, S. Z. & Barthol ini, G. (1979). Effect of

clonidine on utilizat ion and potassium-evoked release of adrenaline in brain slices. Brain

Res., 176, 197- 201.

164 S. Z. LANGER & R. MASSINGHAM

Starke, K. (1977). Regulation of noradrenaline release by presynaptic receptor systems. Rev.

Physial. Biachem. Pharmac.. 77,3-124.

Story, D. F., Briley, M. S. & Langer, S. Z. (1979). The cffects of chemical sympathectomy with

6-hydroxydopamine on rz-adrenoceptor and muscarinic cholinoceptor binding in rat heart

ventricle . Eur . J. Pharmac.. 57,423-426.

U'Prichard, D. c., Bechtel, W. P., Rouot, B. M. & Snyder, S. H. (1979). Multiple apparcnt

a-noradrenergic receptor binding sites in rat brain : effect of 6-hydroxydopamine. Mol.

Pharmac.. 16,47-60.

THE EFFECT OF ORAL

CONTRACEPTIVES ON

PLATELET NORADRENALINE AND

5-HYDROXYTRYPTAMINE RECEPTORS

AND AGGREGATION

J. R. PETERS, J. M. ELLIOTT & D. G. GRAHAME-SMITH

AI RC Unit and University Departm ent ofClinical Pharm acology,

Radcliffe Inftrma ry. Woodstock Road. Oxfo rd. OX2 6HE, UK

Introduction

No radrenaline (NA) and 5-h ydrox ytrypt amine (5HT) cau se platelet aggregation in

vitro, and the y enhance aggregation produced by ADP (Mill s & Roberts, 1967),

collagen and thrombin (Yu & Latour, 1977). There is evidence that platelet

membranes possess receptor sites for NA and 5HT. This evidence comes both from

funct ional studies on platelet aggregation and upt ake , and from results ofradioligand

binding techniques which are being used increasingly to demonstrate receptor sites

for horrnones, neurotran smitters and drugs in vari ous tissues (Te ll, Haour & Saez,

1978). The latter technique may enable characteristics ofthe mol ecular site ofaction

ofa hormon e or drug to be examined in some detail and provides a new method for

studying aspects of hormone and dru g action clinicall y. Alt eration s in receptor

characteristics have been not ed in a number of human disorders (Bar & Roth, 1977;

Kaywin , McDonough, Insel & Shatt il, 1978).

Radioligand binding methods have been used to demonstrate 5HT and NA

receptors on human platelets (Boullin & Elliott, 1979; Boullin, Glenton, Molyneux,

Peters & Roach , 1977). During the study of normal volunteers it became apparent

that men and post-menopausal women had a smaller and less variable number of

platel et 5HT and NA binding sites than young wom en , and enquiry seemed to

indicate that the young women with the greatest variation were taking the contraceptive pil!. Animal studies (Elliott, Peters & Grahame-Smith , 1980) have indicated

that oestrogens alter the characteristics of 5HT and NA platelet receptors. The aims

of this study therefore were to investigat e the influence of the cyclic al use of the

combined contraceptive pill as compared with normal men struation upon platelet

NA and 5HT receptors, and to see whether an y chan ges in receptor characteristics

were accompan ied by alteration in platelet aggregation and 5HT uptake.

166 J. R. PETERS. J. M. ELLIOTT & D. G. GRAHAME-SMITH

Methods

Subjects

Two groups ofhealthy female volunteers were studied. The treated group (n = 15, age

range 22-30) were all taking combined oral contraceptive preparations containing

30 \.Jg of ethinyloestradiol and 150 or 250 \.Jg oflevonorgestrel. Duration oftreatment

2-24 cycles . The control group (n = 8) were age matched, never having taken oral

contraceptives and with a predictable menstrual cycle. No subject in either group was

taking any other drugs and all were non-smokers. Two sampies were obtained from

the treated group. The first on, or not more than 48 h prior to day 21, that is, the

last day oftreatment in the combined pill cycle , and the second on, or not more than

36 h prior to the beginning of the next cycle of treatment, that is, seven days later.

The control group were also sampled on two occasions. Firstly, du ring the luteal

phase between one and five days prior to menstruation, and secondly between five

and eight days following the onset of menstruation. Thus in both groups paired

sampIes were taken at the zenith and nadir of oestrogen level (endogenous in control

group, exogenous in oral contraceptive treated group). This variation was confirmed

for the control group by measurement of plasma oestradiol and progesterone levels

on both sampling occasions.

On each of the study days in both groups the subjects were bled via venous

cannulae to provide 60 ml ofblood into 1% EDTA (9:1) for receptor assays and 5HT

uptake, and 10 ml ofblood into 3.8% (w/v) trisodium citrate for aggregation studies.

a -Adrenergic receptor binding assay

a -Adrenergic receptors on intact platelets were measured using the tritiated

antagonist [3H]-dihydroergocryptine ([3H]-DHE) according to the method ofBoullin

& Elliott (1979). Platelet rich plasma (PRP) was prepared by centrifugation of whole

blood EDTA at 180 g for 15 min at 20·C . Platelets were separated from plasma by

centrifugation at 1700 g for 5 min at IO·C then resuspended in 0.1% EDTA/150 mM

NaCI at a density of approximatcly 0.8 x 108 cells ml:", checked using a Coulter

counter. Aliquots of platelet suspension wcre incubated with [3H]-DHE (range

1-15 nM) in the presence or absence of phentolamine (5 IJM). Incubates, total volume

1 ml, were maintained at 37"C for 20 min with shaking and tcrminated by centrifugation at 6500 g for 1 min. The platelet pellet was washed three times in 0.1%

EDTAthen sonicated and counted by liquid scintillation. Non-specific binding was

defined as that observed in the presence of 5 IJM phentolamine. Specific bound

[3H]-DHE was therefore calculated as the ditTerence between that bound in the

absence and presence of 5 IJM phentolaminc. Frec [3H]-DHE concentration was

determined by sampling the supernatant ofthe incubation after centrifugation.

5HT receptor binding assay

Binding characteristics of human platelet 5HT receptors were identified using

tritiated 5-hydroxytryptamine ([3H]-5HT) in a centrifugation separation assay

(Boullin et al., 1977; Peters & Grahame-Smith, 1980). PRP was incubated with

[3H]-5HT (range 2-10 nM) for 2 min at 2·C in a volume of 225 IJI , then 25 IJI of

incubation butTer (1%EDTA/150 mv NaCI) or 5HT (10-9 or IO-4M) in identical buffer

was added. After a further two minutes the incubations were terminated by centrifugation at 2500 g for Imin at O"C. Pellets were sonicated and counted as above.

Values for binding were corrected for cell density of PRP using a Coulter Counter.

Specific binding to the high affinity site A was taken as the ditTcrence between total

binding and that remaining after the addition of 1O-9M unlabelled 5HT. The specific

binding to the lower affmity site B was taken as the further decrease in binding seen in

PLATE LET RECEPTORS AND ORAL CON TRACEPTIVES 167

those tubes receiving 10-4M unlabelled 5HT, compared with those receiving 10-9M

unlabelled 5HT in parallel incubations (Peters & Grahamc-Smith, 1980).

5HTuptake

Specific 5HT binding is therefore defmed as 'displaceable'. Non-specific or 'nondisplaceable' binding of 5HT as measured at 2°C during the binding assay can be

resolved into an active component which is inhibited by metabolic and 5HT uptake

inhibitors and a passive component not inhibited by these agents (Peters & GrahameSmith,1980), Stahl & Meltzer, 1978). The passive uptake at each concentration was

determined in separate experiments using no rmal volunteers and represents passive

difTusion plus radioactive ligand trapped between cells in the pellet.

The active component of [3H]-5HT uptake was therefore measured as the

difTerence between the amount of 5HT non-specifically bound in the presence and

absence ofmetabolic inhibitors, and an uptake curve drawn for each subject on both

sampling occasions. Comparative inhibitor studies suggest that this measurement of

uptake at 2°C is the same process as that occuring at more physiological temperatures

(Sneddon , 1969) and that it is functionally related to the lower affmity 5HT site

described here (Drummond & Gordon, 1975a).

Platelet aggregation

PRP was obtained as described from citrated whole blood in all subjects ofboth oral

contraceptive and control groups, on the same occasions as receptor binding assays

were performed. The aggregation response to NA (final concentration 2 ~M) 5HT

(20 ~M) or ADP (20 ~M) was measured (Boullin, Green & Price, 1972) using a light

transmission method. The response to ADP was taken as the reference aggregation

response both in terms of rate of change in mV min-I and total change in optical

density (60D) in 2 min , measured using an optical planimeter. These same two

functions were measured for both NA and 5HT and expressed as percentages of the

reference ADP response for that subject on that occasion .

Statistical evaluation

Receptor binding:

Data from each group was analysed in the form of a Scatchard plot by linear

regression analysis according to the method ofleast squares. The standard error ofthe

receptor capacity was calculated from the sampIe standard error ofthe regression line

at the point of intercept of the regression line with the 'bound' axis. Similarly the

standard error of the receptor affmity was calculated from the sample standard

deviation of the regression coefficient (Snedecor & Cochran, 1967). Comparisons

between graups of receptor affmity and capacity were made by means of Student's

unpaired z-test.

5HT uptake:

Absolute values of[3H]-5HT taken up at various free concentrations were compared

using the Student's paired r-test for within graup subjects (21 v 28 days or oral contraceptive group and luteal v post-menstrual phase of contral group), and by the

unpaired r-test when comparing controls with groups treated with oral contraceptives.

Platelet aggregation:

Results were analysed 'within group' by Student's paired z-test and between control

and oral contraceptive treated group by unpaired Student's t-test - both two-tailed.

168 J. R. PETERS.J. M. ELLIOTT & D. G. GRAHAME-SMITH

R esu lts

a -Adrenergic receptor

The characterisation of [JH]- DHE binding to the pl atelet indicates a single binding

site . Val ue s for the afTmity or eq uilibrium dissociation constant expressed as n M

[lH ]-DHE, and for capacity of the site expressed as fmo l [lH]-DHE bound 108

platelets-I

, were obtained by Scatchard analysi s (Figure I).

80

I O. y 21

0;

0;

Control

60 0

Ö O. y 28

E

24 x

0 40

18

c

<i 0

.. 12 s .- 20

5 ' 10

Fre . lJH}DHE [ nM]

Figure 1 Platelet a-adrenergic receptor. Specific bind ing of[3H]-DHE to human platelets as a

function of[3H]- DHE concentrations. • • oral contraceptive group day 21, 0------0 oral

contraceptive group day 28, 6. - - - - 6. control group. Inset shows Scatchard analysi s of same

data.

Control group:

The untreated group showed no difTerence between sampies obtained during the

luteal and post-menstrual phases for either afTmit y or capacity (Table 1), and the data

from both phases were pooled to form a single control group, for purposes of

comparison with the efTects oforal contraceptives .

Ta ble 1 Equilibrium dissociation constant s J(,j and capac ities (c) of a-adrenergic and 5HT

receptor sites on platelets of women taking oral contraceptives and controls"

a-adrenergic receptor 5HT receptor

SiteA Site B

crfmol lü' ctfmol lü! ctfrnol Iü'

Kd plate lets'") J(,j platelets') J(,j platelets")

Lutea l 4.0 ±0.4 83.7 ±3.9 1.2±0.9 2.7 ±1.7 12.6± 1.7 94.6±9.1

Control N.S. N.S. N.S. N.S. N.S. N.S. group

n=8 Post- 3.4 ±0.4 81.0 ±4.6 2.1 ± 0.6 3.6 ± 1.1 13.3 ± 2.4 89.8± 11.6

menstrual

Oral Day21 5.1 ±0.6 94.6 ±6.2 2.9±0.7 6.1 ±0.6 13.6 ± 1.6 113.8± 10.1

Contracepti ve P < 0.001 P < 0.001 N.S. P < 0.001 N.S. P < 0.005

group

n=15 Day28 3.3±0.2 66.7 ±2.0 1.9±0.7 3.6 ±0.7 14.8 ± 1.2 87.1 ±5.3

* Values are means ± s.e. mean . Significance values as shown N.S. = not significant

PLATEL ET RECEPTORS AND OR AL CONTRACEPTIVES 169

Oral contraceptive group:

On da y 21, the oral contraceptive group showed a significantly higher capacity but a

decreased affmity than on da y 28 (Figure 1 and Table I). The capacity of the control

group lay between the values for da y 21 and day 28 in women on the pill , and was

significantly different from both, (Control v day 21, P < 0.0 05; control v day 28,

P < 0.001).