In this trial 1629 patients who had sutTered a myocardial infarction were allocated at
random to sulphinpyrazone 200 mg, four times daily, or to placebo, for 12 to 24
months. Entry to the trial was 25 to 35 days after myocardial infarction. Before the
trial began it was decided to restriet the analysis to so called eligible patients and
analysable events. In other words, the designers of the study decided to restriet
analysis to patients sutTering from their disease , dying their kind of death, while
under the influence oftheir drug. It was dec ided that 71 patients were non-eligible and
43 deaths were regarded as non-analysable.
Deaths from all causes, and cardiac deaths were reduced at up to 24 months,
although conventional levels of significance were not achieved. Although deaths
from myocardial infarction were almost identical in the treatment and placebo
groups, there was a marked and significant reduction in sudden deaths which took
piace almost entirely during the first six months of treatment. It is of course by no
means c1ear that a reduction in sudden deaths is due to an etTecton blood platelets or
the vessel wall , and other mechanisms, for example, an anti-arrhythmic effect, may
If the da ta from the Anturane Reinfarction Trial (1980) (ART) are analysed on a
'intention to treat' bas is, includingall patients and all deaths, there were 74 deaths in
813 sulphinpyrazone treated patients and 89 deaths in 816 placebo patients.
( X2 = 1.2052 , P> 0.05) (Dornenet, 1980, personal communication).
The design of the Anturane Reinfarction Trial study has been the subject ofsome
criticism (e.g. MitcheII, 1980), and in addition at the time of writ ing (July 1980) an
unfortunate difficulty has ari sen with the interpretation of the data. It is reported
(Science, 1980) that in the Un ited States the Food and Drug Administration has
announced that Anturane cannot be labelIed and advertised for the prevention of
death in the critical months following a heart attack. Although the classification of
death was catried out blind by non-involved experts, the Food and Drug
Administration's non-blind analysis of our Anturane Reinfarction Trial indicates
that in many cases the cause of death was misclassified or vaguely defmed. In
particular, the Food and Drug Administration states that many ofthe deaths called
'sudden death' in the control group ofpatients tak ing an inert substance were really
heart attack deaths or deaths due to other causes. The Food and Drug Administration
is quoted as feeling that although the Anturane Reinfarction Trial study provides
suggestive evidence that sulphinpyrazone may be etTective in preventing sudden
death, the study does not provide the quality of scientific evidence required by
American law to approve the drug for use after myocardial infarction. The
manufactures ofsulphinpyrazone (Ciba-Geigy) are quoted as stating that they remain
confident that the drug will be found to be useful in the prevention ofsudden death
following a heart attack and that the ditTerences between the company and the Food
and Drug Administration will be resolved.
the defmition and identification of sudden death, particularly in post-hoc analysis.
Everyone working in the field very much hopes that the problem will soon be
resolved by the joint re-examination of the data being carried out by the trial policy
Committee and the Food and Drug Administration at the time ofwriting.
Clinical trials of transient cerebral ischaemic attacks present many difficulties,
including poor understanding of the pathogenesis, variability of the natural history
and 'soft' end -points. Two trials must be mentioned: Fields, Lemak, Frankowski &
Hardy (1977) compared the efTects of aspirin and placebo in transient ischaemic
attacks and found that if the incidence of death, cerebral infarction and continuing
transient ischaemic attacks were grouped together, aspirin appeared to exercise a
significant benefit, but there was no significant change in the ind ividual end-points
when considered in isolation. In the Canadian Cooperat ive Study (1978) patients
with transient ischaemic attacks were randomly allocated to placebo or to aspirin or
to asp irin plus sulphinpyrazone. In thi s trial, aspirin reduced stroke and death in men
but not in women, and sulphinpyrazone had no signficant effect. In men there was
pathogenesis of transient ischaemic attacks is not fully understood, and it is by no
means certain that platelet-fibrin emboli are consistently implicated.
Where do we stand at the present time in relation to aspirin, sulphinpyrazone and
dipyridamole? I think that at the moment the answer must be that we cannot be
confident of the place of these agents in the modification of thrombotic arterial
disease. In several trials, aspirin appeared to be beneficial, although, apart from the
PARIS study, statistical significance was not achieved. In the rnajor negative study,
the AMIS study, although large numbers ofpatients were involved there is ofcourse a
real possibility that by chance a genuine effect of aspirin ma y have been missed. In a
recent leader in the Lancet (1980) it has been suggested that five trials involving
aspirin quoted here , plus the trial from Breddin, Loew, Lechner, Uberla & Walter
(1980), can be legitimately grouped together. I must say I have some personal
reservations about the validity of this approach, which if not adding chalk and
cheese, is certainly adding camembert and cheddar. However, ifthe pooled data from
the six trials are analysed, aspirin appeared to be significantly efTective in reducing
cardiovascular mortality after myocardial infarction. In the PARIS trial, the aspirin
dipyridamole combination appeared to be somewhat more efTecti ve than aspirin
alone, but the da ta require confirrnation.
Although in the study of sulphinpyrazone after myocardial infarction, the
Anturane Reinfarction Trial (1980), there was evidence ofreduction in sudden death
after the Irrst six months, this ma y not be a platelet or platelet-vessel wall effect, and
interpretation of the da ta in terms of general therapeutic advice must await the
reappraisal at present being carried out jointly by the Food and Drug Administration
Despite all the efTorts which have been made up to the present time, the practising
physician cannot at the moment be given firm guidelines. As always, each individual
clinician must make his own mind on the basis of the existing evidence, but my
advice today would be not to drift in to premature acceptance of the value of
antiplatelet intervention after myocardial infarction until more information is
The anticoagulant story is a warning; on the basis of early studies the generally
accepted wisdom of the medical profession was to give anticoagulants to patients
What are needed now are furt her large scale weil designed, randomised,
prospective, controlled trials ofwhat in shorthand we can call antiplatelet drugs after
myocardial infarction. In the light ofthe PARIS (1980) and ART (1980) studies, trials
with entry as early as possible after myocardial infarction would seem desirable.
Several such trials are now impending, and until the results are available, a mood of
optmustic scepticism is I th ink appropriate. However, the present conflicts of
evidence, ambiguities and uncertainties are most disappointing when so much
thought and efTort have already been invested in controlled trials.
As regards the cerebral circulation, there is suggestive evidence of a curious
sex-associated benefit from aspirin, and despite the difficulties of the soft end-point,
further large scale trial s are aga in needed
One final word: large scale trials with aspirin, sulphinpyrazone and dipyridamole
have only been possible at the present time because all these agents have been in use
in other contexts for many years, and their potential toxicity, or lack of it, is weil
defmed. The new generation of potential anti-thrombotics, for example, specifrc
thromboxane synthetase inhibitors or synthetic PGIz analogues, must still be many
years away from large scale trial, and I suspect for some years we must content
ourselves with c1 inical validation ofthe drugs which are already on the market.
Anturane Reinfarction Trial Research Group (l980).Sulfmpyrazone in the prevention of
sudden death after myocardial infarction. New Eng. J. Med., 302,250-256.
Aspirin Myocardial Infarction Study Research Group (1980). A randomised controlled trial of
myocardial infarction: a comparison of acetylsalicylic acid, phenprocoumon and placebo.
Canadian Cooperative Study Group (1978). A randomised trial ofaspirin and sulfinpyrazone in
threatened stroke. New Eng. J. Med., 299,53-59.
Coronary Drug Project Research Group (1976). Aspirin in coronary heart disease. J. clin. Dis..
Elwood, P. C, Cochrane, A. C, Burr, M. L., Sweetnam, P. M., WilIiams, G., Melsky, E.,
Hughes, J. S. & Renton , R. (1974). A randomised controlled trial ofacetyl salicylic acid in
the secondary prevention of'mortality from myocardial infarction. Lancet, 1,436-440.
Elwood, P. C & Sweetnam, P. M. (1979). Aspirin and secondary mortality after myocardial
infarction. Lancet, 2, 1313-1315.
Fields, W. S., Lemak, N. A., Frankowski, R. F. & Hardy, R. J. (1977). Controlled trial of
aspirin in cerebral ischaemia. Stroke, 8,301-316 .
Jick, H. & Miettinen, O. S. (1976). Regular aspirin use and myocardial infarction. Bril. med. J..
Lancet,(1980). Leadingarticle, Aspirin after myocardial infarction. Lancet. 1, 1172-1173.
MitcheII, J. R. A. (1980). Secondary prevention of myocardial infarction, Brit. med. J.. 280,
O'Grady,J. & Moncada,S. (1978) Aspirin: a paradoxical effecton bleeding time. Lancet, 2,780.
Persantine-aspirin Reinfarction Study Research Group (1980). Persantine and aspirin in
coronary heart disease. Circulation, in press.
Science (1980). FDA says no to Anturane . Science, 208, 1130-1132.
Centre de Recherche Mer reli International.
16. rue d 'Ankara, 67084 Stra sbourg Cedex , France
Pickering (1978) has suggested that some physicians treat their pat ients as
individual s, while others treat the labels which they have fixed to their patients . The
latter approach is peculiarly inappropriate for patients who have been labelIed
hypertensive. All present defmitions of idiop athic hypertension are based upon
arbitrary deviations from supposedly normal ranges of a single, highly variable
haemodynamic parameter which is the outcome of an intricate interplay of a host of
cardiovascular functions. Thi s type of defmition ma y have been inevitable in view of
the ease with which arterial pressure can be estimated, but as a description of disease
and guide to therapy it leaves much to be desired .
Patients with idiopathic hypertension difTer in many factors that can influence
the ir respon se to antihypertensive drugs (Koch-Weser, 1973 ). They obviously difTer
in the severity oftheir hypertension and in the amount by which elevation ofsystolic
pressure exceeds that of diastolic pressure . They can be of any age and have all
degrees ofd isturbance offunction of vital organs , which may or may not be related to
their hypertension. They vary greatl y in the extent to which abnor malities of
resistance and capacitanc e vessels in specific vascular beds, of total peripheral
hypertension. There are also large indi vidual difTerences among hypertensive
pat ients in the patterns of absorption, distribution and elimination of man y
antihypertensive drugs and , accordingly, in the relation between drug dosage and
concentration at the sites ofaction .
When one considers all these var iables, it is hardly surprising that hypertensive
subjects are notoriously heterogeneous in their response to various antihypertensive
drugs. Drugs that efTectively restor e normotension in some hypertensives fail almost
completely in others or reduce blood pressure onl y at the cost of intolerable side
effects. This fact makes the 'relative effectiveness' of antihypertensive drugs as
determined by comparative studies in groups ofpatients oflimited value for selecting
the best drug or drugs for a given pat ient. More importantl y, it certainly contributes
to the unhappy fact that hypertension is often not weil treated. Community sur veys
weil tolerated blood pressure reduction (Do yle, 1980).
Several schematized approaches to the treatment of high blood pressure (for
example, Joint National Committee, 1977) have been put forth in recent years to
guide physicians in antihypertensive therapy. Such approaches outline ranges ofdrug
dosage and the progression from a single drug to two or more . They are quite
adequate to achieve satisfactory blood pressure control in most hypertcnsives and
effectively preclude selection of inappropriate drug combinations. However, they
make little or no atternpt to tailor therapy to patient characteristics and their
recommendations are necessarily arbitrary and often arguable. Thus, it is difficult to
accept a thiazide as the initial drug of choice for all patients or to consider
propranolol, methyldopa, reserpine and clonidine as equivalcnt drugs for every
patient (Joint National Committee, 1977). It seems doubtful that any predetermined
therapeutic scheme , no matter how elaborate, can be sufficiently responsive to the
widely different therapeutic needs of all hypertensive individuals. It cannot possibly
'fit the drug to the patient'.
Drug therapy ofprimary hypertension can be considered fully successful only when it
reduce s arterial pressure into the normal range , maintains it there under all
conditions ofposture and activ ity, and does not cause haemodynamic abnormalities
or other undesirable side effects. To achieve this goal, particularly in patients with
marked blood pressure elevation, individualization of drug therapy is inescapable.
Many physicians today use this approach to some extent. They adm inister their
favourite drugs in some sequcnce until they chance upon a drug or combination of
drugs that produce a reduction in arterial pressure which they judge acceptable. To
be sure, control of mcan arterial pressure at that poin t is often far from optimal and
may be qu ite unsat isfactory in the supine posit ion or during certain activities. Drug
side effects may be distressing to the point of making long-term patient compliance
Trial and crror individualization of therapy is successful in proportion to the skilI,
patience and perseverance ofthe physician and to his ability to motivate the patient.
When drug dosages are carefully titrated and when drugs are logically combined so
that their antihypertensive actions are complementary and additive but their side
effects are not, the rate of eventual therapeutic success is quite high. Many patients
with moderate or severe hypertension are 'refractory' to some antihypertensive
drugs, but it is very rare that diastolic pressure cannot be controlled satisfactorily by a
painstakingly individualized regimen . Systolic pressure elevation when isolated or
grossly disproportionate represents a much more difficult and as yet unanswered
therapeutic challenge (Koch-Weser, 1973; Koch-Weser, 1979; Tarazi, Magrini &
Dustan, 1975; Simon , Safar, Levenson , Kheder & Levy, 1979).
Successful antihypertensive therapy is today generally defined by how nearly
normotensive a patient is rendered. Th is definition may be inappropriately
undemanding. While lowering of arterial pressure, if possible into the normal range,
blood pressure when his overall haemodynamic status remains grossly abnormal.
ß -adrenoceptor antagonists (Lund-Johansen, 1977).
Most patients with established idiopathic hypertension have an abnormally
elevated total peripheral resistance while their cardiac output is in the normal range
Comments
Post a Comment
اكتب تعليق حول الموضوع