| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the Department of Pharmacology, Erasmus University Rotterdam, and
Department of Neurology (M.D.F.), Leiden University Medical Centre, the
Netherlands. Dr Bax is now with the Department of Cardiology and Internal
Medicine, Eemland Hospital, Amersfoort, the Netherlands.
Correspondence to P.R. Saxena, MD, PhD, Department of Pharmacology, Erasmus University Rotterdam, Dr Molewaterplein 50, PO Box 1738, 3000 DR Rotterdam, Netherlands. E-mail saxena{at}farma.fgg.eur.nl
Methods and Results—Concentration-response curves to the
antimigraine drugs were constructed in human isolated coronary
artery segments to obtain the maximum contractile response
(Emax) and the concentration eliciting 50% of
Emax (EC50). The EC50 values were
related to maximum plasma concentrations (Cmax) reported in
patients, obtaining Cmax/EC50 ratios as an
index of coronary vasoconstriction occurring in the clinical
setting. Furthermore, we studied the duration of contractile responses
after washout of the acutely acting antimigraine drugs to assess their
disappearance from the receptor biophase. Compared with sumatriptan,
all drugs were more potent (lower EC50 values) in
contracting the coronary artery but had similar efficacies
(Emax <25% of K+-induced contraction). The
Cmax of avitriptan was 7- to 11-fold higher than its
EC50 value, whereas those of the other drugs were <40% of
their respective EC50 values. The contractile responses to
ergotamine and dihydroergotamine persisted even after repeated
washings, but those to the other drugs declined rapidly after
washing.
Conclusions—All current and prospective antimigraine drugs
contract the human coronary artery in vitro, but in view of low
efficacy, these drugs are unlikely to cause myocardial ischemia
at therapeutic plasma concentrations in healthy subjects. In patients
with coronary artery disease, however, these drugs must remain
contraindicated. The sustained contraction by ergotamine and
dihydroergotamine seems to be an important disadvantage compared with
sumatriptan-like drugs.
The present study deals with 2 major issues in clinical
practice: (1) Do new antimigraine compounds cause less coronary
artery constriction than sumatriptan? and (2) Is sumatriptan better
than ergot derivatives in this respect? Obviously, these questions
cannot be easily answered by clinical trials. We therefore employed a
pharmacological approach using the human isolated coronary
artery to determine the potency (sensitivity) and efficacy (magnitude)
of the contractile responses to sumatriptan and other current
(ergotamine, dihydroergotamine, methysergide, and its active metabolite
methylergometrine16) as well as new
(naratriptan,17
zolmitriptan,18 19
rizatriptan,20 and
avitriptan21) antimigraine drugs. Results were
related to the respective Cmax reported in
patients.
Because sumatriptan-induced contractions of coronary arteries
show substantial variability both within and between
studies,22 23 we used a "parallel"
experimental design involving segments from the same coronary
artery. These coronary arteries were obtained from organ donors
who died of causes unrelated to cardiac diseases and therefore may
potentially represent the population treated with antimigraine
drugs.
Experimental Protocol
Concentration-Response Curves and Relation With Clinical
Plasma Concentrations
Duration of Action
Analysis of Data
Concentration-Response Curves and Relation With Substance P
Response and Clinical Plasma Concentrations
In the 9 hearts in which concentration-response curves to the
antimigraine agents were constructed, the correlation between the
Emax of the drugs and the coronary artery
relaxation to substance P (1 nmol/L) after precontraction with
PGF2
Figure 2
Duration of Action
With respect to the antimigraine compounds, the results of our study
show that all drugs, but in particular ergotamine, dihydroergotamine,
and methylergometrine (metabolite of
methysergide16), were more potent (lower
EC50 values) than sumatriptan in contracting the
human isolated coronary artery. Although the
Emax of ergotamine tended to be somewhat higher
than that of sumatriptan, we observed no statistically significant
differences between Emax values of sumatriptan
and any of the other antimigraine drugs. Our findings are in agreement
with a recent report on zolmitriptan,19 but they
appear to be at variance with earlier studies claiming that rizatriptan
is only half as effective as sumatriptan in contracting the human
isolated coronary artery.27 28
Admittedly, data obtained in human post mortem tissues can vary between
experiments because of a variety of factors, which are difficult to
discern. However, the coronary arteries used in these
studies27 28 were
endothelium-denuded and obtained from hearts from
patients undergoing cardiac transplantation. In contrast, the
coronary arteries in the present experiments were obtained
from organ donors who died of noncardiac causes, and the
endothelium was not removed. Thus, our population may
better reflect the population likely to be treated with
5-HT1B/1D receptor agonists.
Coronary Artery Contraction and Relaxation to
Substance P
Coronary Artery Contraction at Therapeutic Plasma
Concentrations
Additional Factors Involved in Coronary Artery Constriction
in Patients
In conclusion, all current and prospective antimigraine drugs
investigated contract the human coronary artery in vitro. Apart
from avitriptan, therapeutic plasma concentrations of the drugs do not
reach levels likely to cause myocardial ischemia in individuals
with normal coronary circulation. However, like sumatriptan,
all antimigraine compounds investigated, including the newer drugs,
must remain contraindicated in patients with coronary artery
disease. The sustained coronary artery contraction induced by
ergotamine and dihydroergotamine is an important disadvantage compared
with the sumatriptan-like drugs.
Received November 11, 1997;
revision received February 11, 1998;
accepted February 25, 1998.
© 1998 American Heart Association, Inc.
Clinical Investigation and Reports
Coronary Side-Effect Potential of Current and Prospective Antimigraine Drugs
Abstract
Top
Abstract
Introduction
Methods
Results
Discussion
References
Background—The antimigraine drugs
ergotamine and sumatriptan may cause angina-like symptoms, possibly
resulting from coronary artery constriction. We compared the
coronary vasoconstrictor potential of a number of current and
prospective antimigraine drugs (ergotamine, dihydroergotamine,
methysergide and its metabolite methylergometrine, sumatriptan,
naratriptan, zolmitriptan, rizatriptan, avitriptan).
Key Words: coronary disease • vasoconstriction • migraine • pharmacology • antimigraine drugs
Introduction
Top
Abstract
Introduction
Methods
Results
Discussion
References
Sumatriptan, a
5-HT derivative with agonist activity at
5-HT1B/1D receptors, is highly effective in
aborting attacks of migraine and cluster headache. The drug is
generally well tolerated. However, up to 15% of patients
consistently report chest symptoms, including chest pressure,
tightness, and pain, often mimicking angina
pectoris.1 2 3 Although extracardiac mechanisms
have been invoked,4 chest symptoms may well be
caused by coronary vasoconstriction, which has been observed
after sumatriptan both in vivo5 and in
vitro.6 7 8 In some cases, the use of sumatriptan,
like that of ergotamine,9 10 11 was even associated
with myocardial infarction12 13 and cardiac
arrest.14 "Second-generation"
sumatriptan-like antimigraine drugs are all aimed at, in addition to
achieving high efficacy and long duration of action, avoiding
coronary vasoconstrictor activity.15
Methods
Top
Abstract
Introduction
Methods
Results
Discussion
References
Preparation of Tissue
Right epicardial coronary arteries were obtained from 14
"heart beating" organ donors (7 male, 7 female; age, 7 to 61 years)
who died of noncardiac disorders (11 of cerebrovascular accident, 3 of
head trauma) <24 hours before the tissue was taken to the laboratory.
Hearts were provided by the Rotterdam Heart Valve Bank (Bio Implant
Services/Eurotransplant Foundation) after removal of the aortic and
pulmonary valves for transplantation purposes. The study was
approved by the joint Ethical Committee of the Erasmus University
Rotterdam and the University Hospital Rotterdam
"Dijkzigt." The hearts were stored at 0°C to 4°C in a
sterile organ-protecting solution immediately after circulatory arrest.
After arrival in the laboratory, the right coronary artery was
removed and placed in a cold, oxygenated Krebs bicarbonate
solution of the following composition (mmol/L): NaCl 118, KCl 4.7,
CaCl2 2.5, MgSO4 1.2,
KH2PO4 1.2,
NaHCO2 25, and glucose 8.3; pH 7.4. Vessels were
cut into rings 
4 mm long, suspended on stainless steel hooks in
15-mL organ baths containing Krebs bicarbonate solution, aerated with
95% O2/5% CO2, and
maintained at 37°C. Vessel segments containing distinct,
macroscopically visible atherosclerotic lesions were not used.
After equilibration for at least 30 minutes and a wash every 15
minutes, changes in tissue force were recorded with a Harvard
isometric transducer. The vessel segments, stretched to a stable force
of 15 mN, were exposed to 30 mmol/L K+
twice, and the functional integrity of the endothelium
was verified by observation of relaxation to substance P (1 nmol/L)
after precontraction with PGF2
(1
µmol/L). The tissue was washed and then exposed to 100 mmol/L
K+. The data obtained with
PGF2, substance P, and 100 mmol/L
K+ were averaged for each coronary
artery. Subsequently, the vessel segments were washed again and, after
a 30-minute equilibration period, 2 series of experiments were
performed.
In the first series of 9 experiments, a concentration-response
curve was constructed with the different compounds (ergotamine,
dihydroergotamine, methysergide, methylergometrine, sumatriptan,
naratriptan, rizatriptan, avitriptan, and zolmitriptan, as well as 5-HT
used as a marker). In some cases, concentration-response curves to
sumatriptan were obtained in duplicate, which were averaged and
regarded as one curve in further analysis. As described earlier
in detail,23 contractile responses were expressed
as a percentage of the contraction induced by 100 mmol/L
K+ in the respective segments, and the data were
analyzed to obtain, in each case, values of
Emax and EC50. The
Emax and EC50 values
represent the efficacy and potency, respectively, of a drug in
eliciting a response (in this case, coronary artery
contraction). Thus, the lower the EC50 of a drug,
the more likely it is to cause coronary vasoconstriction at
lower plasma concentrations; the Emax is
obviously only of importance when a drug is present in high enough
concentrations, as dictated by its potency. To assess the capacity of
various agonists to contract the human coronary artery during
clinical use in migraine, we calculated the ratio between the reported
Cmax after administration of clinically effective
doses (Table
) and the EC50 value
of the compounds in contracting the human isolated coronary
artery. Thus, a high
Cmax/EC50 ratio indicates
that the plasma concentration of the drug is high enough to contract
the human coronary artery in the clinical situation. The
magnitude of this contraction will be dictated by the
Emax of the drug.
View this table:
[in a new window]
Table 1. Emax (Expressed as % of Effect Caused by 100
mmol/L K+) and EC50 (nmol/L) Values in Human
Isolated Coronary Artery of 5-HT and Antimigraine Drugs, Together With
Their Therapeutic Doses and Cmax in
Patients
In a second series of 5 experiments, the durations of
action of the acutely acting antimigraine drugs (ergotamine,
dihydroergotamine, sumatriptan, naratriptan, rizatriptan, avitriptan,
and zolmitriptan) were compared. For this purpose, contractions of
coronary artery segments were elicited with a single
concentration of these drugs (2 times EC50, as
determined in the first series of experiments), and the time to reach a
stable contraction was noted. The segments were then washed twice every
15 minutes, and contractions remaining after each wash were noted for a
total period of 90 minutes.
Differences between EC50 and
Emax values of sumatriptan and other compounds as
well as between contractions remaining after each wash (every 15
minutes) were evaluated with Duncan's new multiple range test, once an
ANOVA (randomized block design) had revealed that the samples
represented different populations. The
Emax of the compounds tested was correlated with
the relaxant response to substance P obtained in the individual
coronary arteries (Pearson's correlation coefficient). Values
of P
0.05 were considered to indicate significant
differences. All data in the text and illustrations are
presented as mean±SEM.
Results
Top
Abstract
Introduction
Methods
Results
Discussion
References
Basic Properties of the Preparations: Effects of Substance P
and Potassium
All coronary artery segments, obtained from 14 hearts,
relaxed after substance P (1 nmol/L), the response amounting to 61±8%
of the precontraction (35±2 mN) to PGF2
(1 µmol/L). Contraction to 100 mmol/L KCl was 47±3 mN.
Concentration-response curves obtained in 9 coronary
arteries with the various compounds investigated are shown in Figure 1, and the derived values of
EC50 (nmol/L) and Emax (%
of contraction elicited by 100 mmol/L K+)
are shown in the Table
. As reported earlier,22 23
the contractile effect of sumatriptan on the isolated human
coronary artery showed a considerable variability; the
EC50 and Emax values ranged
from 117 to 2042 nmol/L and 2.3% to 27.0% of the response to 100
mmol/L K+, respectively. The
EC50 values of all compounds, in particular
ergotamine, dihydroergotamine, and methylergometrine, were
significantly lower than that of sumatriptan. The
Emax of 5-HT was significantly higher, but those
of the other compounds did not differ significantly from that of
sumatriptan. However, it may be noted that the
Emax of other triptan derivatives is about half
that of ergotamine.
View larger version (17K):
[in a new window]
Figure 1. Concentration-response (expressed as % of
response to 100 mmol/L K+) curves in human isolated
coronary arteries (n=9) obtained with current (left:
ergotamine, 
; dihydroergotamine, 
; methysergide, 
; and its
metabolite methylergometrine, 
) and potential (right: zolmitriptan,
; rizatriptan, 
; naratriptan, 
; and avitriptan, 
)
antimigraine drugs compared with sumatriptan (
). Data are
mean±SEM. (1 µmol/L) was assessed.
Pearson's correlation coefficient did not yield a significant
P value with any of the drugs (data not shown). depicts the ratio between
the reported plasma Cmax obtained after
administration of a clinically effective dose (see Table) and the
EC50 value of the compounds in contracting the
human isolated coronary artery. The data show that, compared
with that of 100 mg PO sumatriptan, the
Cmax/EC50 ratios of
avitriptan (75 and 150 mg PO) were higher, those of ergotamine (2 mg
PO) and zolmitriptan (2.5 and 5 mg PO) were lower, and those of the
other compounds (naratriptan, rizatriptan, and methysergide measured as
its active metabolite methylergometrine) were in the same range. The
ranges of Cmax/EC50 ratios
obtained with parenteral ergotamine (0.5 mg IM) and dihydroergotamine
(1 mg SC) were not much different from that with parenteral sumatriptan
(6 mg SC).
View larger version (32K):
[in a new window]
Figure 2. Relationship between reported Cmax
concentration in patients and EC50 values of various
antimigraine compounds in contracting human isolated coronary
artery. Clinical doses and mode of administration associated with
Cmax values (see Table for references) are indicated in
each case, except methylergometrine (measured after methysergide, 2 mg
PO). When lower and upper values are mentioned, they indicate reported
range. Note that a Cmax/EC50 ratio of 1
indicates that, if same conditions were applicable in patients as in
present in vitro experiments, drug would elicit 50% of its maximum
contraction of coronary artery. In view of use of
Cmax data from literature,
Cmax/EC50 ratios of different drugs
were not subjected to statistical analysis.
Contractions to a single concentration (2 times
EC50) of the acutely acting antimigraine drugs
were elicited in 5 coronary arteries. The peak stable
contraction (% of 100 mmol/L K+) and the
time required to reach the peak contraction with the different
compounds were as follows: sumatriptan 15±4% (3±2 minutes),
naratriptan 10±3% (3±1 minutes), zolmitriptan 21±15% (4±7
minutes), rizatriptan 22±16% (4±2 minutes), avitriptan 17±12%
(5±1 minutes), ergotamine 26±7% (24±13 minutes), and
dihydroergotamine 27±11% (18±10 minutes). The effects of repeated
washings on the peak contractions elicited by each drug are
presented in Figure 3. The data
show that the contractile responses to ergotamine and dihydroergotamine
were sustained over the 90-minute period (P0.05 versus
sumatriptan), whereas those to sumatriptan, naratriptan, rizatriptan,
avitriptan, and zolmitriptan nearly completely disappeared after the
second wash 30 minutes later.
View larger version (18K):
[in a new window]
Figure 3. Effect of repeated washings (3 times every 15
minutes) on contractions of human coronary arteries (n=5)
elicited by acutely acting antimigraine drugs (ergotamine, ;
dihydroergotamine, ; sumatriptan, ; zolmitriptan, ;
rizatriptan, ; naratriptan, ; and avitriptan, ). All drugs
were administered once at a concentration 2 times their
EC50 (see Table). Data are mean±SEM.
Discussion
Top
Abstract
Introduction
Methods
Results
Discussion
References
Human Coronary Artery Contraction In Vitro
As reported earlier in vitro6 8 24 and in
vivo,25 26 5-HT induced contraction of the human
isolated coronary artery. 5-HT was more efficacious (higher
Emax value) than sumatriptan, because of a more
prominent action mediated via 5-HT2
receptors.6 8 24 26
It is suggested that relaxation to substance P is a measure of the
functional integrity of the endothelium and could be
related to underlying coronary artery
disease.29 30 Because there was no significant
correlation between the Emax of the compounds and
relaxant response to substance P in the coronary arteries used
in the present investigation, it would appear that the contractile
effect of the antimigraine drugs is not increased by underlying
coronary artery disease. We concede that the present study,
with few data points at the outer sides of the range of the relaxant
response to substance P, may not be particularly suitable for such an
analysis. However, even in a larger analysis (78 donor
hearts), we did not find an inverse but rather in fact a positive
correlation between the Emax of sumatriptan and
the magnitude of substance P relaxation in the human coronary
artery.31
We calculated the ratio between the clinically effective plasma
Cmax and EC50 values of the
different antimigraine compounds to estimate the degree of
coronary vasoconstriction to be expected during therapeutic use
(see Figure 2). Except for avitriptan, the plasma
Cmax values of all drugs tested remain <40%,
and in the case of zolmitriptan, ergotamine (2 mg PO), and methysergide
(but note, not its metabolite methylergometrine), even <10% of their
EC50 values. Thus, with the exception perhaps of
avitriptan, therapeutic doses of the antimigraine drugs investigated
will cause little coronary artery constriction in vivo. It was
recently shown with PET that sumatriptan did not affect myocardial
perfusion in healthy migraineurs at therapeutic plasma
concentrations.32 Even in the case of avitriptan,
the maximum coronary constriction (Emax
8% of 100 mmol/L K+ response) will be
barely perceptible and is unlikely to affect coronary artery
blood flow, which remains unchanged until the arterial
lumen is compromised by >80%.33 In contrast, in
patients with preexisting coronary artery lesions who have only
a limited coronary reserve,33 even a
small coronary artery contraction that may occur with plasma
concentrations encountered during clinical use could be enough to cause
myocardial ischemia. A similar phenomenon may also be observed
in patients with "variant" angina pectoris, who have increased
coronary artery sensitivity to 5-HT.26 It
has been reported that the contractile effect of sumatriptan on the
human isolated coronary artery is potentiated by
thromboxane A223 34
and is inhibited by aspirin as well as the thromboxane
receptor antagonist SQ30741.23 Thus,
exaggerated production of such substances locally may augment
coronary artery contractions to similar antimigraine drugs in
vivo.
Other factors involved in the contraction of coronary
arteries in vivo are slow diffusion from the receptor biophase, plasma
protein binding, and the formation of active metabolites. Slow
diffusion from the receptor biophase, which has been reported for ergot
derivatives,35 is in accordance with our findings
concerning the sustained response to both ergotamine and
dihydroergotamine despite repeated washings (Figure 3). In the clinical
situation, it is also known that the effects of ergotamine and
dihydroergotamine sustain much longer than is to be expected from their
plasma concentration profiles.35 36 37 Indeed,
substernal chest pain and discomfort suggestive of myocardial
infarction have repeatedly been described with ergot preparations,
particularly ergotamine.10 38
Dihydroergotamine,39
methysergide,16 and possibly
ergotamine38 form active metabolites, which, as
is clearly the case with methysergide (see results with
methylergometrine), may also cause coronary artery
constriction. Sumatriptan does not have active
metabolites,40 but zolmitriptan forms an
N-desmethyl derivative, which is approximately
twice as potent as the parent compound in causing
vasoconstriction.41 We do not know whether active
metabolites are formed by avitriptan, naratriptan, or rizatriptan.
However, in view of its structural similarity to zolmitriptan,
rizatriptan may be expected to form an N-desmethyl
derivative, which may be pharmacologically active. Except for
sumatriptan (14% to 21% plasma protein
binding40), we do not know the extent of plasma
protein binding of the other drugs.
Selected Abbreviations and Acronyms
Cmax
=
maximum plasma concentrations
EC50
=
concentration eliciting 50% of its own Emax
Emax
=
maximum effect, maximum contraction
5-HT
=
5-hydroxytryptamine
PGF2
=
prostaglandin F2
Acknowledgments
This study was supported by the Netherlands Heart Foundation,
grant 93.146. The authors wish to express their gratitude to the
Rotterdam Heart Valve Bank (Bio Implant Services/Eurotransplant
Foundation) for supplying us hearts.
References
Top
Abstract
Introduction
Methods
Results
Discussion
References
This article has been cited by other articles:
 |
|
 |
|
  A. Bhattacharya, K. W. Schenck, Y.-C. Xu, L. Nisenbaum, E. Galbreath, and M. L. Cohen 5-Hydroxytryptamine1B Receptor-Mediated Contraction of Rabbit Saphenous Vein and Basilar Artery: Role of Vascular Endothelium J. Pharmacol. Exp. Ther., May 1, 2004; 309(2): 825 - 832. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Tepper, A. M. Rapoport, and F. D. Sheftell Mechanisms of Action of the 5-HT1B/1D Receptor Agonists Arch Neurol, July 1, 2002; 59(7): 1084 - 1088. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. P. A. Humphrey, W. Fenvik, R. Salonen, P. R. Saxena, R. W. M. van den Broek, and A. MaassenVanDenBrink Craniovascular selectivity of eletriptan and sumatriptan in human isolated blood vessels Neurology, July 10, 2001; 57(1): 162 - 163. [Full Text] [PDF]
|
|
|
|
|
|
E. L. H. Spierings, B. Gomez-Mancilla, D. E. Grosz, C. R. Rowland, F. S. Whaley, and K. J. Jirgens Oral Almotriptan vs Oral Sumatriptan in the Abortive Treatment of Migraine: A Double-blind, Randomized, Parallel-Group, Optimum-Dose Comparison Arch Neurol, June 1, 2001; 58(6): 944 - 950. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. VanDenBrink, R. W. M. van den Broek, R. de Vries, A. J. J. C. Bogers, C. J. J. Avezaat, and P. R. Saxena Craniovascular selectivity of eletriptan and sumatriptan in human isolated blood vessels Neurology, November 28, 2000; 55(10): 1524 - 1530. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Gnecchi-Ruscone, X. Bernard, P. Pierre, D. Anderson, N. Legg, H. Enahoro, P. D. O. Winter, A. Crisp, J. A. Melin, and P. G. Camici Effect of naratriptan on myocardial blood flow and coronary vasodilator reserve in migraineurs Neurology, July 12, 2000; 55(1): 95 - 99. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Bahra, M. J. Gawel, J.-E. Hardebo, D. Millson, S. A. Breen, and P. J. Goadsby Oral zolmitriptan is effective in the acute treatment of cluster headache Neurology, May 9, 2000; 54(9): 1832 - 1839. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Tfelt-Hansen, P. R. Saxena, C. Dahlof, J. Pascual, M. Lainez, P. Henry, H.-C. Diener, J. Schoenen, M. D. Ferrari, and P. J. Goadsby Ergotamine in the acute treatment of migraine: A review and European consensus Brain, January 1, 2000; 123(1): 9 - 18. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. MaassenVanDenBrink, R. de Vries, P. R. Saxena, M. A.D.H. Schalekamp, and A.H.J. Danser Vasoconstriction by in situ formed angiotensin II: role of ACE and chymase Cardiovasc Res, November 1, 1999; 44(2): 407 - 415. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. W. John, P. J. Pauwels, M. Perez, S. Halazy, B. Le Grand, Y. Verscheure, J.-P. Valentin, C. Palmier, T. Wurch, P. Chopin, et al. F 11356, a Novel 5-Hydroxytryptamine (5-HT) Derivative with Potent, Selective, and Unique High Intrinsic Activity at 5-HT1B/1D Receptors in Models Relevant to Migraine J. Pharmacol. Exp. Ther., July 1, 1999; 290(1): 83 - 95. [Abstract] [Full Text]
|
|
|
|
 |
|
 Coronary Vasoconstriction with Antimigraine Drugs Journal Watch (General), July 24, 1998; 1998(724): 5 - 5. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||