lsoprenalin e produced a fall of 15, 20 and 36% in DPTIITTI with corres ponding

increases in stro ke work of l2, 18 and 21%, and of pea k eject ion power of 24, 35 a nd

61%. Adrena line alte red DPTIITTI by +2, - 2, -13 , and -33%, stro ke work by +14,

+21, +60 , and +89%, and peak eject ion power by +21, +33 , +88, and +14 1%.

192 J. COLTART & P. WILLICOMBE

Noradrenaline, however, produced changes in DPTIITTI ofonly -6, -5, -2, and -1%;

correspondingly, stroke work increased 5, 23, 57 and 91% and peak ejection power 2,

27,68, and 113%.

Discussion

Currently many drugs are advocated in the treatment of low cardiac output states

after cardiac surgery. Dobutamine in the present study has been shown to have an

equipotent inotropic effect to isoprenaline with a reduced chronotropic and

peripheral vascular action. The changes in maximal acceleration and stroke work at

a constant left atrial pressure occurred at a lower heart rate with dobutamine than

with isoprenaline and the drug caused a small elevation in systolic blood pressure.

The studies described were performed at a time after surgery when a stable

haemodynamic situation had been achieved . This precaution minimises

uncontrollable changes in the circulatory state which often occur in the immediate

post-bypass period that is, such variables as fluctuations in the myocardial

contractile state, elimination of anaesthetic agents, employment of analgesic agents,

and variable changes in fluid volume. Rastelli & Kirklin (1966) have shown

depressed cardiac function for two to three days following cardiac surgery, and it

seems likely that the present results can be extrapolated back to the immediate postoperative period, with the advantage that any changes observed in the

haemodynamic state were probably induced by the drug.

The cardiovascular elTects of dobutamine wh ich have been found in this study are

not necessarily beneficial. The cardiac output in patients after cardiac pulmonary

bypass would appear to be relatively more dependent on heart rate than stroke

volume. Stroke volume is increased only slightly by inotropic agents, (Jenkins,

Branthwaite & Bradley, 1973; Holloway, Schultz, Stinson & Harrison, 1973) and the

present results confirrn that observation. Since dobutamine has less chronotropic

effect than isoprenaline, the cardiac output for a given inotropic dose will be less.

The second observed effect of dobutamine was that it provoked a modest elevation

of arterial pressure. In low output states, augmentation of aortic flow is more

important than maintenance of a high arterial pressure, provided that perfusion of

vital organs does not fall below a critical value. If the perfusion pressure to the

coronary, renal, and cerebral arteries is adequate, then further elevation of arterial

pressure may be to the detriment offorward flow from the compromised left ventricle

(Cohn, 1973). Pressure work has greater metabolic requirements than volume work

(Sarnoff, Braunwald & Welch, 1958). It would appear that elevation ofblood pressure

is undesirable, especially ifbrought about by constriction ofarterioles in vital organs.

Analysis of our measurements in the myocardial A- V differences for oxygen, lactate,

and pyruvate revealed that similar changes were caused by dobutamine and

isoprenaline. The results could be attributed to both an increase in coronary blood

flow and substrate utilization by dobutarninc, or to identical changes in these

variables with both drugs.

Dobutamine appears to have little if any advantage over isoprenaline in patients

who have undergone cardiac surgery, but the results found in this study should not be

interpreted as necessarily applying to other c1inical situations in wh ich the cardiac

output is reduced, in particular gram-negative septicaemia and myocardial

infarction. There is evidence to suggest that after experimental myocardial infarction

in dogs, dobutamine causes less ischaemic injury than isoprenaline for the same

change in myocardial function (Tuttle et al.. 1973 ; Maroko et al., 1974). This was

attributed to the absence of diastolic hypotension and reduced chronotropic action .

The incidence of arrhythmias was also less than with isoprenaline. Thus in the

assessment ofthe utility or the appropriate use of a particular inotropic agent, it may

be necessary to defmc precisely the c1inical circumstances.

INOT RO PIC DRUGS IN CLIN ICA L PRACTICE

Conclusions

193

The use of a n extractable aort ic c1ec tromagne tic flo w p robe to provide a con ti n uo us

o n-li ne di spl a y of ascend ing aortic flo w a nd cardiac o u tp ut followin g o pen hea rt

s u rge ry is d esc ribed. U ti lizing this eq ui pme n t, the ha emo d yn amic actio ns of

d obutamin e a nd isop re na line a re com pa red in fourt een patients im mediate ly

foll owing card iac surgery. The st udy co nfirrned a n inotropic actio n produced by

d obutamine at a heart rate 10-15% lower th an isoprcnaline, wi t h less peripheral

vascu la r actio n . Arte ria l a nd co rona ry si n us bl o od a na lyses revea led little difference

in the m yocardi al metabol ic act io ns of e ither d rug. Because inotropic d rugs p roduce

onl y relativel y sma ll in crea ses in stroke vo lume in th is gro up o f patients, the ri se in

cardia c o u tp u t ca us ed b y these a gents is m ore dependent on the effe cts upon heart

rate rathe r than im proved m yo cardial con tracti le sta te a nd consequentl y

dobutamine ha s little advantage over iso p re na line in th is si t uatio n .

In otropic a ug men ta tion ca n be ach ieve d with o ut im pairm en t of s ube ndocardia l

perfusion , as demonstrated with noradrenal ine.

Acknowledgements

W e would lik e to acknowledge th e assista nce ofou r co-wo rke rs D r G . R . Lewis, Dr

A . Malcolrn , Mr B. T . Wil liarns, Mr A . Farnswo rth a nd Dr P. Po ol e-Wilson .

Referenccs

Benzing, G ., Stocken, J., Nave, E., Ts uei, Y. G . & ldapl an , S. (1974). Eva luation of left

ventricularcontrac tility. Cardiovas. R es.. 8, 313-322 .

Bourn e. P. B. & Wi lliarn s, B. T . (1 975). A cardiac mon itor co mbining tlow and pressu re

measurement. Bioehern. Eng, 10,453'-455.

Chattergee, K., Parrnl ey, W., Ganz, W., Fo rrester, J., Walensky, P., Crexels, G . & Swan, H.J. C.

(1973). Hem od ynam ic and metabolic responses to vasod ilator therap y in ac ute myocardial

infarction . Circulation, 48, 11 83-1193.

Co hn, J. N. (1973 ). Blood pressure and card iac performan ce. A m. 1. Med., 55, 351-361.

Holloway, E. L., Schultz, C, Stinson , E. & Ha rrison , D. C. (1973). Co mparison of circulatory

effects of dopa min e and isopre nali ne immedia tely following cardiac surgery . Circulation.

(Suppl, IV) VII and VIII, 177, Abst ract.

Jacobs, R. R., Heydon , W. c., Williarns, B. T., Sch midt, U. T. & Sche n k, W. Z. (1970). Cardiac

outpu t in the exercisin g dog. J. Surg. Res.. 10,25- 32.

Je n kins, B. S., Branthwaite , M. A. & Bradley, R. D. (1 973). Cardiac function after op en hea rt

surgery. Cardiovas. Res.. 7,297-30 5.

Jewiu , 0 ., Birkhead, J.. Mitchell, A. & Doller y, C. (1 974). C linical cardiovasc ular

ph armacology ofdobutam ine. A selective inotropic catech ola mine. Lancct, 2, 363-367.

Ma roko, P. R., Swain , J. & Vatner, S. (1 974). Effect of dobutamine on myocardial inju ry after

coronary occl usion. Circulat ion. 49, (Suppl . 111), 189, Abs tract.

No ble, M. I. M., T renchard, D. & G uz. A. (1 966). Left ventricular ejcctio n in con scious dogs,

measurem ent and significance of maximal acce lera tion of blood from the left ventricle.

Circulation R es.. 19,13 9-14 7.

Palmer, R. F. & Lasseter, K. C. (1 975). Sodi um nitroprusside. Ne\\' Eng 1. u.«.292 ,294-297.

Rastelli , G . C. & Kirklin, J. W. (1 966). Hem odynam ic state early after the replacement oft he

mitral valve. Circulation, 34 , 448-461.

Samoff S. J., Braunwald . E. & Welch , J. R. (1958). Hem odynami c determina nts of oxyge n

con sumption with specia l reference to the tension tim e inde x. Am. J. Physiol.. 192,

148-156.

Tuttle, R. R., Pollock. G . D., Todd, G. & T ust, R. (1 973). Dobutam ine: con ta inment of

myocardi al infarct size by a new ino tropic agent. Circulation, 47, (Suppl, IV), 132 ,

Abstra ct.

Vatn er, S. F., McRi ch ie, R. J. & Braunwald . E. (1974). Effects ofdobutam ine on left vent ricular

perforrn an ce, corona ry dynam ics and distribut ion of cardiac output in co nscious dogs. 1.

clin. Invest.. 53 , 1265-12 73.

Wyse, S. D., Gi bson , D. G . & ß ranthwait e, M. A. (19 74). Haem odynarni c effects ofsalbut am ol

in pat icnt s needing circulato ry suppo rt after open heart surgery. Brit. mcdJ; 3, 502-50 3.

Antiplatelet Drugs and

Antithrombotic Chemotherapy

Chairmen:

J. F. MUSTARO, Canada

M. VERSTRAETE, Belgium

THE MECHANISMS OF

ACTION OF DRUGS

THAT MODIFY PLATELET AND

VESSEL WALL FUNCTION

M. A. PACKHAM

Department ofBiochemistry,

University ofToronto.

Toronto, Ontario, Canada,

M5S lAS

J. F. MUSTARD

Department ofPath ology,

McMaster University,

Hamilton, Ontario, Canada,

L8N3Z5

Many drugs that modify platelet or vessel wall functions have been recognized and

are being investigated for their potential usefulness in modifying thromboembolic

events . An understanding of the effects of these drugs involves knowledge of the

factors that initiate thrombosis and ofthe different types of'thrornbosis,

Mechanisms in thrombosis

Thrombus formation in arteries differs from thrombus formation in veins in several

ways. In the initial stages oftheir formation, arterial thrombi are mainly composed of

platelets and fibrin with few red and white blood cells. In veins , where blood flow is

slow, thrombi resemble a blood clot with large numbers ofred blood cells trapped in

a network of fibrin ; the platelet component ofvenous thrombi may only be apparent

as a white head where the thrombus has originated on the vein wall or in a valve

pocket (Mustard & Packharn, 1979). The formation of arterial thrombi may involve

different mechanisms depending upon the initiating factors, the site, the patterns of

blood flow, and whether the thrombi form in response to a single injury or to

repeated injury ofthe vessel wall.

Following the loss ofthe endothelium from anormal blood vessel, platelets adhere

to collagen , basement membrane and the microfibrils in the subendothelium

(Baumgartner, Muggli, Tschopp & Turitto, 1976), release their granule contents

(Hovig , 1963) and form thromboxane A2 (TXA2) from arachidonate liberated from

198 M. A. PACKHAM & J. F. MUSTARD

their membrane phospholipids under the influence of phospholipases that are

stimulated when platelets adhere to collagen (Smith, Ingerman, Kocsis & Silver,

1974; Flower & Blackwell, 1976; Rittenhouse-Simmons, 1979; Bell, Kennerly,

Stanford & Majerus, 1979). At sites where blood flow is laminar, only a monolayer

of adherent platelets forms (Groves, Kinlough-Rathbone, Richardson, Moore &

Mustard, 1979). In regions where blood flow is disturbed, however, TXA2 and

released ADP, possibly augmented by released 5-hydroxytryptamine (5HT), affect

platelets flowing by the injury site so that they adhere to those already adherent to the

wall and a mass of aggregated platelets forms . The coagulation system is act ivated in

a number of ways in and around the aggregate and the thrombin that is generated

causes further platelet aggregation, release ofplatelet granule contents and formation

of TXA2; thrombin also converts fibrinogen to fibrin which stabilizes the platelet

mass (Mustard & Packham, 1979).

The role ofcollagen is less apparent in the response ofthe constituents ofthe blood

to damage of the abnormal surface that forms as a result of repeated vessel wall

injury. Platelet-fibrin thrombi form even in regions where blood flow is laminar, and

fibrin appears to be mainly responsible for the attachment of the thrombi to the

injured vessel wall (Jergensen, Packham, Row sell & Mustard, 1972; Stemerman,

1973; Kinlough-Rathbone, Groves, Jorgensen, Richardson, Moore, Packham &

Mustard, 1980b). Thrombin evidently plays a major part in the initiation and growth

ofsuch thrombi.

Adrenaline is another agent that afTects platelets and vessel walls (Rowsell,

Hegardt, Downie, Mu stard & Murphy, 1966). It enhances the action of other aggregating agents (Ardlie, Glew & Schwartz, 1966), accelerates blood coagulation (Özge,

Rowsell , Downie & Mustard, 1966), shortens platelet survival (Özge & Mustard,

1964), and itselfcauses platelet aggregation (Mitchell & Sharp, 1964). Adrenaline also

afTects vessel wall tone and the metabolism ofthe myocardium (Maling & Highman,

1958; Nahas, Brunson, King & Cavert, 1958). The increase in the concentration of

adrenaline in the blood in response to stress or exercise undoubtedly influ ences the

interactions of platelets with injured vessel walls and with each other (Haft & Fani,

1973).

Mural thrombi are not static structures but may undergo dissolution and reforrnation . Activation of plasminogen to plasmin (Majno & Joris, 1978) results in fibrin

degradation and under the force of blood flow, platelet emboli or platelet-fibrin

emboli may be broken ofTand pass downstream where they impact in the microcirculation (Gunning, Pickering, Robb-Smith & RusselI , 1964: Moore & Mersereau,

1968).

The possibility that prostagiandin b (PGb) has a role in limiting thrombus

formation is under intensive investigation (Moncada & Vane, 1978). Thrombin or

injury stimulates the cells ofthe vessel wall to form PGb and in regions where blood

flow is slow or arrested, PGb may limit the size ofthrombi (Kelton, Hirsh, Carter &

Buchanan, 1978). In large arteries where blood flow is rapid. however, dilution and

the pre ssure that forces materials into the wall probably prevent PGh from

accumulating in sufficient amounts to afTect platelet adherence or thrombus

formation (Dejana, Cazenave, Groves, Kinlough-Rathbone, Richardson, Packham

& Mustard, 1980).

Tests of drugs that inhibit plateIet function

Both in vitro and in vivo testing has provided information about a wide variety of

drugs (Mustard & Packham, 1978). In vitro, drugs are generally first tested in citrated

platelet-rich plasma for their ability to inhibit platelet aggregation in response to

ADP, adrenaline, collagen, arachidonic acid, and other aggregating agents. Although

DRUG ACTION ON PLATELET AND VESSEL WALL FUNCTION 199

many drugs have been shown to inhibit platelet aggregation in vitro, few are suitable

for administration to man, particularly for long-term administration. The drugs that

have been tested in large scale clinical trials - aspirin, sulphinpyrazone,

dipyridamole - were not designed originally as inhibitors of platelet function and the

potential value of sulphinpyrazone and dipyridamole as antithrombotic agents

would not have been recognized by the type of screening systems now being used.

Other in vitro tests ofthe effects ofdrugs on platelet function include examination of

platelet adherence to collagen, the subendothelium or artificial surfaces, and studies

ofthe contribution ofplatelets to the coagulation sequence.

Many experiments have been reported in which drugs have been administered to

experimental animals or man, sampies ofblood removed and platelet function tested

in vitro. This provides information concerning active metabolites, the time du ring

wh ich effective concentrations remain in the plasma, and side effects . A wide variety

of methods of producing thrombosis in experimental an imals has been used to test

th e effect ofdrugs that alter platelet function (Didisheim, 1972; Mustard & Packharn,

1978; Philp, 1979). The article by White & Butler in this section discusses the

effect ofdrugs in vivo on a number ofexperimenta l models ofthrombosis.

Drugs that inhibit platelet function and thrombus formation can be divided into

four main categories: 1. drugs that inhibit thc arachidonate pathway: 2. drugs that

influence platelet cyclic AMP levels ; 3. drugs that inhibit thrombin generation and

the action of thrombin; and 4. drugs that act in less well-defmed ways (propranolol,

clofibrate, ticlopidine, penicill in and related antibiotics, suloctidil, chlorpromazine

and many others).

Drugs that inh ibit the arachidonate pathway

These drugs include inhibitors of phospholipases, cyclo-oxygenase and thromboxane

synthetase. They inhibit platelet aggregation and release that are largely dependent

on TXAl but they do not in hibit ADP-induced aggregation (Packharn, KinloughRathbone, Reimers, Scott & Mustard, 1977), adherence to collagen (Tschopp, 1977;

Cazenave, Kinlough-Rathbone, Packharn & Mustard, 1978a), release ofgranule contents from platelets adherent to collagen (Weiss, Tschopp & Baumgartner, 1975;

Kinlough-Rathbone, Cazenave, Packharn & Mustard, 1980a) or thrombin-induced

aggregation and release (Packham et al., 1977). Thus, thrombus formation in which

these me chanisms playa major part is larg ely unaffected by the drugs that inhibit

TXAl formation .

Inh ibitors 0/ pho spholipases

The first step of the arachidonate pathway is the activation of phospholipases that

catalyze thc removal of arachidonate from membrane phospholipids. The role of

phospholipase Al has been recognized for some time (Flower & Blackwell, 1976;

Marcus, 1978), but recentl y it ha s been suggested that phospholipase C and digl yceride lipase mayaiso act (Rittenhouse-Simmons, 1979; Bell et al., 1979). Inhibitors

of phospholipase Al include the antimalarial agent mepacrine (Blackwell, 1978;

Winocour, Kinlough-Rathbone & Mustard 1979) and some of the steroidal antiinflammatory drugs (Hong & Levine, 1976; Flower, 1978). The latter inhibit platelet

aggregation and release induced by mo st aggregating agents (Pierce, Oshiro &

Nickerson, 1974) and inhibit platelet adherence to collagen and the subendothelium

(Cazenave, Davies, Senyi, Blajchman, Hirsh & Mustard, 1976). However, high

concentrations of the drugs are required for these efTects and they may act in other

ways in addition to inhibiting the freeing ofarachidonate (Mustard & Packham, 1978;

Packharn & Mustard, 1980). Inhibition of membrane bound phospholipase Al from

human platelets by indomethacin but not aspirin has been reported (Jesse & Franson,

1979),

200 M. A. PACKHAM & J. F. MUSTARD

Inhibitors ofthromboxane synthetase

These inhibitors block the conversio n of pro stagiandin H2 to thromboxane A2.

Among these agents are im idazoJe and some der ivati ves of it (Moncada, Bunt ing,

Mull an e, Thorogood, Vane, Raz & Needleman, 1977; Tai & Yuan , 1978), levami sole

(Ho kama, Morita, Abad , Jo yo, Uc hida & Fischer 1978), and 9, ll- azoprosta- 5,

13-dienoic acid (Gorman, 1979). Th ese inhibitors have onl y been tested in vitro.

Cyclo -oxygenase inhib itors

Drugs that inhibit the enzy me cyclo-oxygena se include the non-steroidal ant iinl1ammatory drugs such as aspirin, indomethacin, phenylbutazone, ibuprofen,

fenoprofen, naproxen and man y oth ers as weil as the uricosuri c agent sulphinpyrazon e (Mustard & Packharn, 1978). Aspirin acet ylate s the cyclo- oxygena se and

thus inhibits it irreversibly (Roth & Majerus, 1977). Inh ibition ofplatel et function by

the oth er drugs per sists only as long as effective concentratio ns of the drugs or the ir

active metabolites remain in the circulation (Buchanan, Rosenfeld & Hirsh, 1978;

Butler, Dieterle, Maguire, Pay , Walli s & White, 1980).

A number of experimental approaches have been used in atte mpts to determine

whether or not aspirin and sulphinp yrazone inhibit platelet adh erence to collagen

and the subendothelium. It is now apparent that under phy siologi cal conditions of

110w, protein concentration and haematocrit, these drugs do not inhibit adherence,

although, by preventing thromboxane A2 formation , the y may inhibir the formation

of a platelet aggregate on the adherent platelets (Weiss et al., 197 5; Tschopp, 1977;

Cazenave el al.. 1978a: Kinl ough-Rathbon e et al., 1980a). T he non- steroidal antiinl1a mmatory dr ugs also do not prevent the platelets that are adherent to collagen

from releasing the contents of their storage granules (Weiss et al.. 1975; KinloughRathbone et al., 1980a). Released ADP and 5HT from the dense granules could

contribute to thrombus formation even in the pre sence of cyclo-o xygenase inh ibitors.

In addition, a mitogen that sti mulates smooth mu scle cell proliferation is released

fro m the a -granules (Ross & Vogel, 1978) and its release from adherent platelets is

not pre vented by these drugs. T hus it is not surprising that aspirin does not prevent

smooth mu scle cell prolifera tion in damaged vessel walls of experimental animals

(Clo wes & Karnovsky, 1977; Baumgartner & Studer, 1977; Clopat h, Horsch &

Dieterle, 1980). It should be pointed out, however , that in rats and rabbits (but not in

swine) sulphinpyrazone does lessen the amo unt of prol iferation in response to vessel

injury (Baumgartner & Studer, 1977; Clopath et al., 1980).

By blocking TXA2 formation by platelets, drugs that inhibit cyclo-oxygenase

prevent th e vasoconstriction that TXA2causes (Needleman, Kulkarni & Raz , 1977;

Svensson & Fredholm, 1977). If vessels in the microcir culati on do not constrict,

thromboemboli are less likel y to lodge and persist in them. Th is possibility should be

considered in relation to the beneficial effects of aspirin in tran sient attacks of

cerebral ischaemia (Fields , Lemak, Frankowski & Hard y, 1977; The Canadian

Cooperative Study Group, 1978). TXA2 mayaiso be responsible for coronary spasm

and, indirectly, for ventricular fibrill ation caused by ischaemia (Moschos, Ha ider , De

la Cruz, Lyons & Regan , 1978).

Sever al differences have been noted between the effects of aspirin and sulphinpyrazone on platelets. Aspirin irreversibly acetylates cyclo-oxygenase (Roth &

Majerus, 1977; Burch , Stan ford & Majerus, 1978b) whereas the effect of sulphinpyrazone and its metabolites on thi s enzyme is reversible (Ali & McDonald, 1978).

Low doses of aspirin ha ve a marked effect on platelet responses to aggregating agents

th at act through TXA2formation whereas it is difficult to show an inhibitory effect of

sulphinpyrazone on platelet function when thi s drug is admi nistered to man. Aspirin

has the undesirable side effect of gastrointestina l bleeding in some pat ients but th is is

rarely a problem with sulphinpyrazone (T he Canadian Cooperative Study Group,

197 8; Anturane Reinfarcti on Trial Research G roup, 1978). Sulphinpyrazone

DRUG ACTION ON PLATELET AND VESSEL WALL FUN CTION 201

prolongs shortened platelet survival in man (Smythe, Ogryzlo, Murphy & Mustard,

1965; Genton & Steele, 1977) whereas aspirin does not (Harker & Slichter, 1972).

Aspirin has proved to be of benefit in patients with transient attacks of cerebral

ischaemia in preventing recurrent attacks, stroke and death, sulphinpyrazone ha s not

(Fields et al.. 1977; The Canadian Cooperative Study Group, 1978). Aspirin has not

had a significant beneficial etTect on total mortal ity in a number of trials in patients

who have had a myocardial infarction, although so me trials showed a trend toward

benefit (Elwood, Cochrane, Burr, Sweetnam, Williams, Welsby, Hughes & Renton,

1974; Boston Collaborative Drug Surveillance Group, 1974; Coronary Drug Project

Research Group, 1976; Breddin, Loew, Lechner, Überla & Walter, 1979; Elwood &

Sweetnam, 1979; Aspirin Myocardial Infarction Study Research Group, 1980).

Sulphinpyrazone lessened the incidence of sudden death in post-myocardial

infarction patients in the first six months after the infarction (Anturane Reinfarction

Trial Research Group, 1980).

The rea sons for the ditTerences between aspirin and sulphinpyrazone are not

understood, probably because the knowledge of the mechanisms that these drugs

atTect is still incomplete.

EffeCIS ofcyclo-oxygenase inhibitors on the vessel wall

Drugs that in hibit cyclo-oxygenase in endothelial or smooth muscle cells of blood

vessel walls prevent the formation of PGIz by these cells (Moncada & Vane, 1978).

PGIz is th e most potent inhibitor of platelet aggregation and the platelet release

reaction that has been discovered and it also partially inhibits platelet adherence to

collagen and the subendothelium (Higgs, Moncada, Vane, Caen, Michel & Tobelem,

1978; Cazenave, Dejana, Kinlough-Rathbone, Richardson, Packham & Mustard,

1979; Weis s & Turitto, 1979). Although both aspirin and sulphinpyrazone have been

shown to inhibit the cyclo-oxygenase of the cells of the vessel wall and thus prevent

PGIz formation, sulphinpyrazone's etTect is weak (Burch, Baenziger, Stanford &

Majerus, 1978a ; Gordon & Pearson, 1978). The concentration of aspirin required to

inhibit the cyclo-oxygenase ofthe vessel wall cells has been reported to be higher than

the concentration that inhibits platelet cyclo-oxygenase (Burch et al., 1978a; Kelton

et al., 1978; Masotti, Galanti, Pogges i, Abbate & Neri Serneri, 1979; Villa , Livio &

De Gaetano, 1979).

A1though a contradictory observation has been reported (JatTe & Weks1er, 1979),

many investigators believe that aspirin should be administered at a dose sufficient to

block platelet cyclo-oxygenase but too low to have much etTect on the cyclooxygenase ofthe vessel wall. It has been proposed that this would prevent thrombus

formation that is due to TXAl, but would not interfere with an y inhibitory etTect of

PGIz on platelet adhererence or aggregation . Indeed, it has been suggested that

inhibition of PGIz formation may be thrombogenic and experiments with injured

jugular vein s ofrabbits support this theory (Kelton et al., 1978). It should be pointed

out, however, that patients with rheumatoid arthritis who rece ive large doses of

aspirin over prolonged periods do not have an increased incidence of thromboembolic complications (Davis & Engleman, 1974; Linos, Worthington, O'Fallon ,

Fuster, Whisnant & Kurland, 1978) and that high doses ofaspirin have been reported

to reduce the incidence of venous thromboembolism in patients (mainly female)

undergoing total knee replacement operations (McKenna, Galante , Bachmann,

Wallace, KaushaI & Meredith, 1980).

In contrast to its irreversible etTect on platelets, asp irin's etTect on the cell s ofthe

vessel wall does not persist; it is thought that these nucleated cells can synthesize new

enzyme molecules to replace the acetylated cyclo-oxygenase. The times required for

regeneration of the ability of vessel wall cells to form PGIz have been reported to be

as short as 2 hand as long as 72 h (Kelton et al., 1978 ; Jaffe & Wek s1er,

1979; Czervionke, Sm ith , Fry , Hoak & Haycraft, 1979; Villa et al.. 1979; Buchanan,

202 M. A. PACKHAM &J . F. MUSTARD

Dejana, Mustard & Hirsh, 1979). All investigators agree , however, that the ability to

form PGh does return and this forms the basis for the suggestion that aspirin should

be administered once-daily to inhibit the ability ofthe platelets to form TXA2 but not

prevent PGb formation by the vessel wall.

Sulphinpyrazone, but not aspirin, has been reported to lessen the extent of

endothelial injury caused by the continuous infusion of homocysteine into baboons

(Harker, 1980). Protection ofthe vessel wall may be related to the ability ofsulphinpyrazone to lengthen the shortened platelet survival that has been observed in

patients with atherosclerosis (Genton & Steele , 1977). It seems unlikely that the

protective effect ofsulphinpyrazone on the endothelium is mediated through inhibition of cyclo-oxygenase since aspirin neither protects the endothelium nor prolongs

shortened platelet survival (Harker & Slichter, 1972; Harker, 1980).

Agents that increase cyclic AMP levels

Many platelet functions are inhibited by the agents that raise platelet cyclic AMP

levels (Mustard & Packham, 1978). These agents include the prostaglandins EI, b

and 02 and adenosine. Drugs that inhibit platelet phosphodiesterase (dipyridamole,

papaverine, methylxanthines) and thus prevent the breakdown of cyclic AMP

augment the effect of these prostaglandins (HasIam, Oavidson, Oavies, Lynham &

McClenaghan, 1978; Di Minno, Oe Gaetano & Garattini, 1978) but by themselves,

the phosphodiesterase inhibitors have little or no effect on platelet cyclic AMP

(Haslarn, 1973). PGb is produced by endothelial and smooth muscle cells in response

to injury, thrombin, bradykinin, angiotensin 11 and histamine (Weksler, Marcus &

laffe, 1977b; Weksler, Ley & laffe, 1978; Schrör, Moncada & Vane, 1977; Blumberg,

Oenny, Marshall & Needleman, 1977; Baenziger, Force & Becherer, 1980;

Czervionke et al., 1979). PG02 is formed from arachidonate by platelets (Smith et al.,

1974). Unlike the cyclo-oxygenase inhibitors, the agents that increase platelet cyclic

AMP levels inhibit platelet adherence to collagen and the subendothelium (Higgs et

al., 1978; Cazenave, Packham, Oavies, Kinlough-Rathbone & Mustard, 1978b;

Cazenave et al.. 1979). They also inhibit aggregation and release in response to all

stimuli and, by inhibiting platelet shape change, they prevent the exposure of the

platelet membrane sites involved in the intrinsic pathway of coagulation, thus

lessening the extent to which thrombin is generated at the platelet surface (Mustard &

Packham, 1978; Ehrman, laffe & Weksler, 1~9)

The phosphodiesterase inhibitor that has received most attention is dipyridamole ,

although its vasodilatory effects and ability to reduce blood pressure may account for

some of its beneficial effects in experimental thrombosis (Roberts, Jacobstein,

Cipriano, Alonso, Combes & Gay, 1980). Suggestions have also been raised that it

inhibits thromboxane A2 formation by an effect on thromboxane synthetase (Greenwaid, Wong, Rao, Bianchine & Panganamala, 1978; Horrobin, Ally & Manku, 1978;

Best, McGuire, Jones, Holland, Martin, Preston, Segal & RusselI , 1979). Some

investigators, however, appear to have reached these conclusions without considering

the inhibitory effect of dipyridamole on the contribution that released AOP makes to

arachidonate-induced aggregation. In high concentrations, dipyridamole inhibits

platelet adherence to collagen and the subendothelium (Cazenave et al., 1978b;

Kinlough-Rathbone, Groves, Cazenave, Richardson & Mustard, 1978). Lower

concentrations inhibit platelet aggregation and the release of granule contents

(Niewiarowski, Likasiewicz, Nath & Sha, 1975). Ooses that can be given to man

prolong shortened platelet survival (Harker & Slichter, 1972). In man, however,

dipyridamole has not been found to be beneficial by itself in managing conditions

involving arterial or venous thrombosis although most of the studies had too few

patients to provide data that could be subjected to statistical analysis. Oipyridamole

DRUG ACTION ON PLATELET AND VESSEL WALL FUNCTION 203

has often been used in combination with another inhibitor of platelet function or an

anticoagulant to achieve a dual effect.

Agents that inhibit thrombin

The efTects of thrombin on platelets and on fibrinogen can be prevented with

anticoagulants or heparin. Recently, some interest has revived in the use of oral

anticoagulant therapy in the acute phase of myocardial infarction (Modan, Shani ,

Schor & Modan, 1975; Tonascia, Gordis &Schmerler, 1975; Chalmers, Matta, Smith

& Kunzler, 1977; Szklo, Tonascia, Goldberg & Kennedy, 1979), despite earlier

studies that indicated no efTect on mortality by the administration of these drugs in

this condition (Wasserman, Gutterman, Yoe, Kemp & Richardson, 1966; Report of

the Working Party on Angicoagulant Therapy in Coronary Thrombosis, 1969).

Heparin potentiates the action of antithrombin III and thus inhibits not only

thrombin but also factors Xa, IXa and Xla ofthe coagulation sequence (Rosenberg,

1978). In conditions in which thrombin and fibrin have major roles in thrombus

formation, heparin should be beneficial . Reports ofthe efTects of heparin on platelet

functions are varied, probably because of the difTerences among commercially

available heparins and the heterogeneity ofheparin with respect to its molecular size

and its affinity for antithrombin III (Zucker , 1975; Wessier & Gitel, 1979; Salzman,

Rosenberg, Smith, Lindon & Favreau, 1980). In vitro, heparin prevents the efTect of

thrombin on platelets but may itself cause platelet aggregation and either potentiate

or inhibit the action of other aggregating agents.

Heparin has been tested experimentally in combination with prostaglandins that

stimulate adenylate cyclase. Two aspects of this deserve consideration. First, these

prostaglandins inhibit platelet shape change in response to aggregating agents

(Ehrman et al.. 1979; Kinlough-Rathbone, Packham & Mustard, 1970) and hence

prevent exposure of membrane sites to which factor Xa binds. Since unbound factor

Xa is readily inhibited by the antithrombin III-heparin complex, the agents that

prevent platelet shape change should enhance the inhibitory efTect of heparin on

coagulation. Second, in contrast, heparin has been reported to inhibit the activation

of adenylate cyclase by PGE, or PGh (Saba, Saba, Blackburn, Hartmann & Mason,

1979; Reches, Eldor & Salomon, 1979; Eldor & Weksler, 1979), thus lessening the

efTects of these prostaglandins on platelet functions. Thus a combination of heparin

with one of these prostaglandins might prevent types of thrombosis that are largely

mediated by thrombin formation but might be less beneficial in circumstances in

which thromboxane A2 and ADP have more dominant roles.

Other agents that inhibit plateletfunctions

Many other agents inhibit platelet functions and thrombus formation but in general ,

less is known about their mechanisms ofaction.

Propranolol and other ß-adrenoceptor blocking drugs

Propranolol has many physiological efTects that may be beneficial in patients with

ischaemic heart disease. This drug is a ß-adrenergic receptor blocker which reduces

cardiac work during exercise and thus lessens myocardial oxygen demand (Nickerson

& Collier, 1975; Frohlich, 1980). Although adrenaline-induced platelet aggregation is

thought to be an a -adrenergic response because it is blocked by phentolamine (Mills

& Roberts, 1967), administration of propranolol to patients with angina pectoris

lessens the hypersensitivity oftheir platelets to ADP-induced aggregation (Frishman,

Weksler, Christodoulou, Smithen & Killip, 1974). Propranolol inhibits platelet

aggregation that is media ted by TXA2 possibly through an effect on phospholipase

A2, and diminishes platelet adherence to collagen (Weksler, Gillick & Pink, 1977a;

204 M. A. PACKHAM &J. F. MUSTARD

Vanderhoek & Feinstein, 1979; Hawiger & White, 1978; Vlachakis & Aledort, 1980).

Protective etTects of propranolol or alprenolol on mortality of patients with definite

or suspected myocardial infarction have been reported (Norris, Clarke, Sammel,

Smith & Williams, 1978; Andersen, Bechsgaard, Frederiksen, Hansen, Jürgensen,

Nielsen, Pedersen, Pedersen-Bjergaard & Rasmussen, 1979).