Bruce et al. May. 1985

American Heart Journal

Post TA: Efficacy of diltiazem in effort angina. A multicenter trial. Am J Cardiol 49~567, 1982.

4. Hossack KF, Kannagi T, Day B, Bruce RA: Long-term study of high-dose diltiazem in chronic angina. AM HEART J 107:1215, 1984.

5. Kannagi T, Bruce RA, Hossack KF, Chang K, Kusumi F, Trimble S: An evaluation of the Beckman metabolic cart for measuring ventilation and aerobic requirements during exer- cise. J Cardiac Rehabil 3:38, 1983.

6. Bruce RA: Exercise testing of patients with coronary heart disease. Principles and normal standards of evaluation. Ann Clin Res 3:323, 1971.

7. Detry J-MR, Bruce RA: Effects of nitroglycerin on “maxi- mal” oxygen intake and exercise electrocardiogram in coro- nary heart disease. Circulation 43:155, 1971.

8. Bruce RA, Petersen JB, Kusumi F: Hemodynamic responses to exercise in the upright posture in patients with ischemic heart disease. In Dhalla NH, editor: Myocardial Metabolism. Baltimore, 1973, University Park Press, p. 849.

9. Degre SG, Strappart GM, Sobolski JC, Berkenboom GH, Stoupel EE, Vandermoten PP: Effect of oral sustained- release nitroglycerin on exercise capacity in angina pectoris: Dose-response relation and duration of action during double- blind crossover randomized acute therapy. Am J Cardiol 51:1595, 1983.

10. Pepine CJ, Joyal M, Cremer KF, Hill JA, Feldman RL, Gelman JS: Hemodynamic effects of nitroglycerin combined with diltiazem in patients with coronary artery disease. Am J Med 76:(6A):47, 1984.

Lack of efficacy of nafaxatrom, a novel anti-thrombotic compound, in patients with coronary artery disease

Nafazatrom (Bay G 6575) is a novel antithrombotic compound, which acts by stimulation of prostacyclin as well as by inhibition of lipoxygenase enzymes. To determine its effects on exercise performance in coronary artery disease patients, a double-blind study was conducted. Twenty patlents with coronary artery disease underwent an exercise stress test before and 2 hours after administration of placebo or nafazatrom (1.2 gm). Before the drug admlnistration, there was evidence of enhanced platelet activity, as reflected by elevated resting plasma beta thromboglobulln and thromboxane Ba concentrations. Plasma 6-keto-PGF,, levels were undetectable in most patients. All coagulation tests were in the normal range. None of these parameters changed with exercise. Administration of placebo or nafazatrom before the exercise stress test did not significantly influence any of the coagulation or platelet function parameters or plasma concentrations of thromboxane B2 and 6-keto-PGF,,. This lack of effect was evident both at rest and upon exercise. Compared to placebo, nafazatrom did not significantly increase exercise tolerance tlme or exercise-induced symptoms. In conclusion, nafazatrom did not influence exercise performance in patients with coronary disease. (AM HEART J lOg:1026, 1985.)

Louis Roy, M.D., Jawahar Mehta, M.D., and Paulette Mehta, M.D. Gainesuille, Flu.

Several studies have suggested an important role for platelets and prostaglandins in the genesis and manifestations of ischemic heart disease.ls2 It has been suggested that atheromatous vessels have

From the Department of Medicine, Divisions of Cardiology and Hematol- ogy/Oncology, University of Florida College of Medicine, and the Veterans Administration Medical Center.

Supported by grants from the American Heart Association, Florida Affiliate, and the Miles Laboratories, West Haven, Conn.

Received for publication July 11, 1994; revision received Oct. 23, 1984;

accepted Dec. 3, 1994.

Reprint requests: J. L. Mehta. M.D., Department of Medicine, University of Florida, Box J-277, J. Hillis Miller Health Center, Gainesville, FL 32610.

1026

decreased ability to synthesize prostacyclin.3-5 This vessel wall-generated prostaglandin is a vasodilator and platelet aggregation inhibitor, and counteracts the effects of platelet-generated thromboxane AZ, which is a potent vasoconstrictor and proaggregant prostanoicl. Stress and ischemia have been shown to increase prostacyclin release in normal subjects,6r7 but in patients with exercise-induced angina pecto- ris and coronary artery disease, this increase is only minimal5 On the other hand, thromboxane A2 bio- synthesis is increased in these patients7sa Thus an “imbalance” between prostacyclin and thromboxane A2 occurring during stress may be involved in vaso-

Volume 109

Number 5, Part 1 Nafazatrom and coronary artery disease 1027

constriction and exercise-induced ischemia. Djmm- ic change in vasomotor tone as a mechanistic basis of stasic angina has been postulated by Epstein and Talbot:.

Nafazatrom (3-methyl I- [Z-(2:napthyloxy)-eth- yl] -2-pyrazoline-5-one) is a new compound, which has been demonstrated to decrease thrombus forma- tion in the coronary arteries in experimental ani- ma.ls.‘O Plasma taken from humans 1 hour after administration of 1.2 gm of nafazatrom stimulates prostacyclin release from the rat aorta.” Wong et al.‘* have suggested that nafazatrom inhibits prosta- cyclin degradation resulting in an increase in prosta- cyclin levels. In addition, this compound inhibits tissue and platelet lipoxygenase.13 If the manifesta- tions of coronary artery disease relate to the decrease in prostacyclin-like activity,6 it can be hypothesized that drugs such as nafazatrom would be useful in altering the response to exercise in patients with myocardial ischemia. This preliminary study was designed to examine if administration of nafazatrom would improve exercise performance in patients with documented coronary artery disease.

METHODS

Study patients. Twenty patients (mean age 59 + 2 years; 19 men and one woman) with stable coronary artery disease participated in this study. Nineteen patients had angiographically proved greater than 70% stenosis of at least one major coronary artery (three had one-vessel disease, 10 had two-vessel disease, five had three-vessel disease, and one had left main coronary artery disease). The one remaining patient did not have angiography but had a previously documented myocardial infarction. None of the patients had unstable angina or a history of recent myocardial infarction. All patients stopped platelet-active drugs (aspirin, sulfinpyrazone, dipyridamole, beta-adren- ergic blockers, calcium antagonists) for at least 14 days prior to the study. In most cases, long-acting nitrates were stopped at the same time, except in two patients, who received their last dose of nitrates 36 hours before the study. All patients gave written informed consent before the study, which was approved by the Institutional Review Committee.

Normal values for coagulation and platelet function parameters were obtained from medical house staff and laboratory personnel. None of the normal subjects had taken any platelet-active drug in the previous 2 weeks.

Treadmill exercise stress test. Exercise testing was performed by means of standard Bruce protocol with continuous 12-lead ECG monitoring. Exercise was contin- ued to the point of angina pectoris or appearance of significant ECG changes. The ECG changes were consid- ered significant if there was ST segment depression of at least 1 mm in depth and at least 0.08 seconds in duration for at least 1 minute. The exercise test was performed in all patients on two different days. On the first day, the

exercise stress test was done at 10:00 AM. On the second day, the patients received either placebo or nafazatrom (1.2 gm orally) at 8:00 AM in a double-blind fashion. Blood pressure and heart rate were recorded at 30-minute intervals after the medication (placebo or nafazatrom) was administered, and the exercise test was repeated 2 hours later at the same time as on the first day.

Blood samples were collected before and immediately after exercise. On the second day, blood samples were also collected at 30 minutes after termination of exercise. Placebo and nafazatrom were provided by Miles Labora- tories, West Haven, Connecticut, in identical-appearing packages containing a powder, which was administered orally to the patient with water.

Blood collection. Peripheral venous blood was gently drawn with a No. 19-gauge butterfly needle, without using a tourniquet, to avoid trauma to platelets or blood vessels. The first 2 ml of blood was discarded. The subsequent 2.0 ml was transferred immediately into a test tube kept in ice and containing 0.5 ml EDTA and aspirin (final concentra- tion 4.5 mM and 1 mM, respectively) for measurement of plasma levels of thromboxane B, and 6-keto-PGF,,. Another 9.4 ml of blood was collected in a test tube containing 0.6 ml EDTA, 2-chloroadenosine, and procain- amide and was kept in ice for measurement of plasma beta thromboglobulin and platelet factor 4. Another 9 ml of blood was collected in a test tube containing 1 ml of 3.8% sodium citrate for platelet count and for determination of thrombin time, prothrombin time, and activated partial thromboplastin time. On the first day, blood was also withdrawn for measurement of blood urea nitrogen. Patients with abnormal blood urea nitrogen (>25 mg%) were considered ineligible to receive nafazatrom or place- bo.

Thromboxane B, and 6-keto-PGF,, determinations. Plasma thromboxane B, and 6-keto-PGF,, were measured by radioimmunoassay as described earlier.‘4s’5 Lyophilized standards, 3H standards, and antibodies raised in rabbits were obtained from New England Nuclear, Boston, Mas- sachusetts. With the use of this methodology, the cross reactivity of thromboxane B2 antibody with other prosta- glandins was less than 0.2 % . Cross reactivity of 6-keto- PGF1, antibody with PGF, was 2.7% and with other prostaglandins was less than 0.3 % .

Beta thromboglobulin and platelet factor 4 determina- tions. Plasma beta thromboglobulin and platelet factor 4 were measured by radioimmunoassay. Supplies were obtained from Amersham, Arlington Heights, Illinois and Abbott Laboratories, North Chicago, Illinois respective- ly-

Platelet counts and coagulation parameters. Platelet counts were performed with a Toa Platelet Counter PL-100 (Toa Medical Electronics, Kobe, Japan). Coagula- tion parameters were measured according to standard procedures by means of the Fibro System (Becton, Dick- inson and Co, Cockeysville, Maryland).

Statistical analysis. Analysis of variance and unpaired and paired t tests were used for statistical analysis. For comparison of plasma thromboxane B, and 6-keto-PGF,,

1028 Roy, Mehta, and Mehta May, 1985

American Heart hwnal

RESULTS OF TREADMILL EXERCISE TESTS ON DAY 1 AND DAY 2

Con;rol Nafelzetrom NAFAZATROM GROUP

I I

Control Placebo PLACEBO GROUP

J,’

/’

Fig. 1. Effect of administration of placebo or nafazatrom on blood pressure and heart rate.

0 ’ I I I I Day 1 Day 2 Day 1 Day 2

PLACE80 NAFAZATROM GROUP GROUP

Fig. 2. Exercise duration and blood pressure and heart rate response to exercise in patients given placebo or naiazatrom.

concentrations, Kruskal-Wallis rank-sum test, Mann- Whitney rank-sum test, and Wilcoxon sign-rank test were used. Paired analysis was used for comparison of first day and second day results. Unpaired analysis was used for comparison of patient data with normal control data. Results are expressed as mean -t standard error. Data for thromboxane B, levels are provided as the median because of the wide variation between different subjects. A p value less than 0.05 was considered significant.

RESULTS

Blood pressure and heart tats. There were no differ- ences in blood pressure and heart rates between the patients who received placebo or nafazatrom on the first or second day. Nafasatrom administration did not significantly influence the resting blood pressure or the heart rate (Fig. 1). Peak exercise systolic blood pressure and heart rate with exercise were also similar on both days in both groups of patients (Fig. 2).

Exercise test. On the first day, the patients in the placebo group exercised for 633 * 94 seconds and those in the nafazatrom group for 631 + 49 seconds (p = NS). After administration of placebo on the second day, the exercise duration (mean 622 & 79 seconds was unchanged. The exercise duration in patients who received nafazatrom (mean duration

693 + 46 seconds) was not significantly different from the first day (Fig. 2).

No differences in exercise-induced symptoms and ECG changes were noticed from the first to the second day in patients who received placebo. Simi- larly, in the nafazatrom group four patients who developed typical angina-like chest pain on the first day with exercise also experienced chest pain on the second day at about the same level of exercise. Patients who developed significant. ECG changes on the first day also developed the same ECG changes on the second day after receiving nafazatrom.

Platelet counts and coagulation parameters. Platelet counts increased significantly during exercise on the first day in both groups. Placebo or nafazatrom administration had no effect on resting or exercise- induced platelet counts (Table I). Prothrombin time, thrombin time, and activated partial thrombo- plastin time were in the normal range on the first

Volume 109

Number 5, Part 1 Nafazatrom and coronary artery disease 1029

Table I. Effects of exercise on platelet counts and coagulation parameters

Before exercise

PC PT PTT TT Group (x103/mm3) (set) (set) (see)

Placebo Day 1 325 13.2 24 10.7 +44 +0.1 +1 kO.2

Day 2 264 13.1 23 10.8 *29 kO.1 &l +0.2

Nafazatrom Day 1 314 13.0 25 10.8 -c36 kO.3 +1 -1- 0.4

Day 2 298 13.0 24 10.8 +35 kO.2 +1 kO.3

Peak exercise 30 minutes after exercise

PC PT PTT TT PC PT PTT TT (X10’lmm3) (set) (set) (set) (10-‘/mm3) (secj fsec) (set)

393 13.4 23 9.9 +0.3 -

- - - r44 kO.1 &l 346 13.1 23 10.5 321 13.3 22 10.7

+44 kO.2 +1 LO.1 +37 +0.1 +1 +0.2 408 12.9 24 10.5

+30 kO.3 &I +0.4 - 384 12.9 24 10.6 387 13.0 25 10.9

+44 kO.2 +l LO.2 *40 +0.3 *l +0.3

Abbreviations: PC = platelet count; PT = prothrombin time; PTT = activated partial thromboplastin time; TT = thrombin time.

Table II. Effects of exercise on plasma concentrations of platelet release products

Group

Day 1 Day 2

Plasma P-thromboglobulin Plasma platelet factor 4 Plasma P-thromboglobulin Plasma platelet factor 4 fwlml) Inglml) hglml) Cnglml)

Peak After Peak After Rest Exercise Rest Exercise Rest exercise exercise Rest exercise exercise

Placebo 74 + 8 81 TL 9 23 k 8 18 k 3 65 + 5 68 f 7 68 2 7 16 + 2 17 +- 3 14 + 3 Nafazatrom 68 + 7 70 + 8 15 rt 2 17 f 2 54 a 6 65 rt 6 57 * 7 13 t 3 15 2 2 13 rf: 3

p = not significant for ah results (rest vs exercise; placebo v8 nafazalxom groups).

day in all patients. These parameters were not influenced by placebo or nafazatrom. Exercise stress had no effect on these measurements (Table I).

Plasma beta thromboglobulin. On the first day, mean plasma beta thromboglobulin levels were 74 + 8 rig/ml in the placebo group and 68 + 7 rig/ml in the nafazatrom group (p = NS); both values were significantly (p < 0.01) higher than normal levels (41 + 4 rig/ml). With exercise, there was a slight but insignificant increase in plasma beta thromboglobu- lin levels. Placebo or nafazatrom administration on the second day did not influence resting plasma beta thromboglobulin levels (65 k 5 and 54 -t 6 rig/ml, respectively, p = NS). Immediately after exercise and 30 minutes after exercise, plasma beta thrombo- globulin levels were also not different from the resting levels or from the first day 0, = NS) (Table II).

Plasma platelet factor 4. On the first day, resting plasma platelet factor 4 levels were in the normal range (mean 13 +- 3 rig/ml) in the placebo and nafazatrom groups (23 +- 8 and 15 + 2 rig/ml, respectively, p = NS). These levels did not change with exercise. Placebo or nafazatrom administration did not influence resting levels (16 +- 2 and 13 +- 3 rig/ml, respectively, p = NS). Exercise stress had no

significant effect on platelet factor 4 concentrations whether patients had taken placebo or nafazatrom (Table II).

Plasma thromboxane B,. On the first day, plasma thromboxane B, concentrations were significantly higher in the patient population compared to the normal values (median 138 and 79 pg/ml, p < 0.05).

Twenty-one of 38 normal values were under the reliable range 0100 pg/ml), whereas only 2 of 20 values from the study patients were under 100 pg/ml. There was no significant difference in the two patient subgroups on the first day (placebo group: median 143 pg/ml; nafazatrom group: median 139 pg/ml). With exercise, there was a slight increase in plasma thromboxane B, concentrations (placebo group: median 205 pg/ml; nafazatrom group: median 155 pg/ml), but the increase was not significant. Placebo or nafazatrom administration did not affect resting levels (placebo group: median 142 pg/ml; nafazatrom: median 164 pg/ml) or peak exercise levels (placebo group: median 149 pg/ml; nafaza- trom group: median 157 pg/ml) (Fig. 3).

Plasmad-keto-PGF,,. On the first day, 6-keto- PGF1, levels were under the minimum reliable value (>lOO pg/ml) in 17 of 20 patients (seven in the placebo group and 10 in the nafazatrom group) and

1030 Roy, Mehta, and Mehta

.xB Ig/l

TxB2 wfml

I I I I

Rest Immediate 30 Minutes Immediate 30 Minutes

After-exercise After-exercise

PLACEBO GROUP NAFAZATROM GROUP

Fig. 3. Plasma thromboxane B, (TxB& concentrations at rest, immediately after exercise, and 30 minutes after exercise in patients given placebo or nafazatrom.

in 29 of 34 normal subjects. Exercise stress had no detectable effect on plasma 6-keto-PGF,, levels. Placebo or nafazatrom administration did not result in detectable change in plasma 6-keto-PGF,, levels either at rest or with exercise.

All data were analyzed to identify any correlation with severity of coronary artery disease, exercise duration, and exercise-induced chest pain and ECG changes. There were no significant relationships between platetet function parameters and prosta- glandin metabolites measured. Increases in platelet count with exercise were similar in all patients.

DISCUSSION

Role of platelets and prostaglandins in myocardial

ischemia. Prior to discussing the effects of nafaza- trom in patients with stable coronary artery disease, it is important to discuss the abnormalities in platelet function and thromboxane A,-prostacyclin synthesis in this group of patients. All coagulation parameters were in the normal range at rest and were unaffected by exercise in patients with coro- nary artery disease. Platelet activation as deter- mined by elevated plasma beta thromboglobulin levels was observed in most patients at rest support- ing previous observations. 8, 16* I7 Based on the normal renal function in these patients, it can be presumed that elevated plasma beta thromboglobulin concen- trations relate to platelet activation in stable coro- nary artery disease. l8 Platelet factor 4 levels in plasma, on the other hand, were elevated in some patients, but the mean value was in the normal range. Although Stratton et a1.‘4 and Mathis et aLzo have reported platelet factor 4 levels to be in the

May, 1985

American Heart Journal

normal range, Green et a1.21 demonstrated elevated levels in certain patients with coronary artery dis- ease. These variations may, in part, relate to differ- ent methods employed for preparation of plasma specimens. The issue of in vitro vs in vivo release has also been debated in the recent past. It is certainly possible that some platelet release occurs in vitro during collection of blood. It can, however, be argued that in vitro release occurring in some patients and not in others implies platelet hyperac- tivity in a subset of patients.23 We did not observe any increase in beta thromboglobulin or platelet factor 4 in plasma with exercise. This is in accord with some recent reports.lg. 20* 23

Increased production of thromboxane AZ and decreased production of prostacyclin from athero- sclerotic vessels have been well documented.2~“~s Based on these observations, it has been speculated that an “imbalance” in thromboxane A,-prostacy- clin equilibrium may be a factor in the genesis of myocardial ischemia. 24 A recent study, on the other hand, suggests that urinary excretion of prostacyclin metabolites may not be decreased in patients with atherosclerosis.25 However, the value of urinary metabolites compared to local tissue generation of prostacyclin is debatable.26 It should be recognized that the plasma levels of 6-keto-PGF,,, a stable metabolite of prostacyclin, are usually below the detection limit of most laboratories. Nonetheless, if prostacyclin deficiency is a cause of myocardial ischemia, prostacydin or prostacyclin-stimulating agents may be expected to improve exercise toler- ance in patients with coronary artery disease by modifying thromboxane A,-prostacyclin “balance.”

Nafaratrom and exercise performance in stable angi- na. Nafazatrom, a pyrazoline derivative, has been shown to prevent in vivo thrombus formation in experimental animals. lo It is well tolerated in humans, and plasma obtained from subjects 1 hour after ingestion of 1.2 gm of nafazatrom stimulates prostacyclin-like activity from rat aortic strips.l’ It also potentiates the actions of the phosphodiester- ase inhibitor dipyridamole.” Although the precise mechanism of its action is not known, nafazatrom probably inhibits in vivo degradation of prostacy- clin.” Others have suggested enhanced plasma stabi- lization of prostacyclin as a mode of action of nafazatrom. In the present study, nafazatrom in a single dose of 1.2 gm was well tolerated without any significant change in heart rate and blood pressure. Peak levels in plasma were probably achieved in 2 hours with this dose as shown in a previous study.” We have also established that plasma obtained from our patients who received nafazatrom stimulated prostacyclin release from human arteries.27 Howev-

Volume 109 Number 5, Part 1 Nafazatrom and coronary artery disease 1031

er, nafazatrom administration failed to influence exercise duration or exercise-induced symptoms in our patients. This lack of effect on exercise perfor- mance was associated with absence of any effect of nafazatrom on coagulation parameters or platelet function. Plasma levels of 6-keto-PFG1, remained undetectable. This may relate to either decreased ability of atheromatous coronary vessels to produce prostacyclin or to very low levels of circulating prostacyclin.

Blockade of enzymatic degradation of prostacy- clin would have been expected to be reflected in increased formation of hydrolysis product 6-keto- PGF1,. A recent study2s suggests that enzymatic catabolites of prostacyclin are not present in signifi-

vessel wall prostaglandin equilibrium during exercise- induced stress. AM HEART J 105:895, 1982.

8. deBoer AC, Turpie AGG, Butt RW, Johnston RV, Genton E: Platelet release and thromboxane synthesis in symptomatic coronary artery disease. Circulation 66:327, 1982.

9. Epstein SE, Talbot TL: Dynamic coronary tone in precipita- tion, exacerbation and relief of angina pectoris. Am J Cardiol 48:797, 1981.

10. Shea MJ, Disco11 EM, Romson JL, Pitt B, Lucchesi BR: The beneficial effects of nafazatrom (Bay G 6575) on experimen- tal coronary thrombosis. AM HEART J 107:629,1983.

11. Vermylen K, Chamone DAF, Verstraete M: Stimulation of prostacyclin release from vessel wall by Bay G 6575, an antithrombotic compound. Lancet 1:518, 1979.

12. Wong P Y-K, Chao P H-W, McGiff JC: Nafazatrom (Bay G 6575). an antithrombotic and antimetastatic agent, inhibits 15-hydroxyprostaglandin dehydrogenase. J Pharmacol Exp Ther 223~757, 1982.

13. Busse WD: Effect of nafaxatrom on guaiacol peroxidation and arachidonic metabolism in blood cells and microsomes.

may explain why we did not observe an increase in 6-keto-PGF,, levels in the nafazatrom-treated patients. It is also possible that prolonged therapy, but not a single dose of nafazatrom. mav result in a detectable increase in circulating prostacyclin. In addition, combination with agents such as dipyrida-

cant amounts in human plasma compared to plasma levels of 6-keto-PGF,,. This fact, in association with low sensitivity of 6-keto-PGF,, radioimmunoassay,

mole may be more useful. Conclusions. The absence of any effects of nafaza-

trom on execise tolerance would suggest that prosta- cyclin activity is not a major pathogenic factor in exercise-induced angina pectoris, and any possible stabilization or stimulation of prostacyclin by nafa- zatrom does not improve exercise-induced myocar- dial ischemia. Similar absence of improvement in episodes of myocardial ischemia was shown in patients with variant angina who were given prosta- cyclin.2g

prostaglendin and thromboxane research, vol 5. New York, 1978, Raven Press, p 119.

15. Roy L, Mehta J, Mehta P: Increased plasma concentration of prostacyclin metabolite, 6-keto-PGF,,, in essential hyperten- sion. Influence of therapy with labetalol. Am J Cardiol 51:464, 1983.

16. Mehta J, Mehta P: Comparison of platelet function during exercise in normal subjects and coronary artery disease patients: Potential role of platelet activation in myocardial

Symposium on nafazatrom (Bay G 6575). Port Chester, New

ischemia. AM HEART J 103:49, 1982. 17. Cade JF, Doyle DJ, Chesterman CN, Morgan FJ, Rennie GC:

York, 1982.

Platelet function in coronary artery disease. Effects of coro-

14. Granstrom E, Kindahl H: Radioimmunoassay of prostaglan-

nary surgery and sulfinpyrazone. Circulation 66:29, 1982.

dins and thromboxanes. In Frohlick J. editor: Advances in

18. Dawes J, Smith RC, Pepper DS. The release, distribution, and clearance of human /3-thromboglobulin and platelet factor-4. Thromb Res 12:851, 1978.

19. Stratton JR, Malpass TW, Ritchie JL, Pfeifer MA, Harker LA: Studies of platelet factor-4 and beta thromboglobulin release during exercise: Lack of relationship to myocardial ischemia. Circulation 66:33, 1982.

20. Mathis PC, Wohl W, Wallach SR, Engler RL: Lack of release of platelet factor-4 during exercise induced myocardial ische- mia. N Engl J Med 304:1275, 1981.

We wish to thank Kelly Greetham for her assistance in the preparation of this manuscript.

21. Green LH, Seroppian E, Handin RI: Platelet activation during exercise induced myocardial ischemia. N Engl J Med 302:193, 1980.

REFERENCES

1. Gryglewski RJ: Prostaglandins, platelets, and atherosclerosis. CRC Crit Rev Biochem 7:291, 1980.

2. Mehta J, Mehta P: Role of blood platelets and prostaglandins in coronary artery disease. Am J Cardiol 48:366, 1918.

3. D’Angelo V, Villa S, Myslewiec M, Donati MB, deGaetano G Defective fibrinolytic and prostecyclin-like activity of human atheromatous plaques. Thromb Haemost 39:353, 1978.

4. Sinzinger H, Feigl W, Silberbauer K: Prostacyclin generation in atherosclerotic arteries (letter). Lancet 2:469, 1979.

5. Neri Serneri GG, Masotti G, Poggesi L, Galanti G, Morettini A, Scarti L: Reduced prostacyclin production in patients with different manifestations of ischemic heart disease. Am J Cardiol 49:1146, 1982.

22. Editorial: Platelets, beta-thromboglobulin and diabetes mel- litus. Lancet 1:250, 1978.

23. Tarnay TJ, Davidson AG: Absence of platelet factor-4 release with exercise stress in patients with coronary artery disease (abstr). Circulation 64(Suppl):191, 1981.

24. Conti CR, Feldman RL, Nichols WW, Mehta J: Myocardial ischemia. olatelets and nrostaslandins. In Mehta J. Mehta P,

I . - -

editors: Platelets and prostaglandins in cardiovascular dis: ease. New York, 1981, Futura Publishing Co, p 301.

25. FitzGerald GA, Smith B, Pedersen AK, Brash AR: Increased prostacyclin biosynthesis in patients with severe atheroscle- rosis and platelet activation. N Engl J Med 310:1065, 1984.

26. Majerus PW: Arachidonate metabolism in vascular disorders. J Clin Invest 72:1521, 1983.

27. Mehta P, Mehta J, Roy L: Effects of nafazatrom on platelet TXA, generation and PGI, degradation. Clin Res 30:829A, 1982.

28. Siess W, Dray F: Very low levels of 6-keto-prostaglandin F,, in human plasma. J Lab Clin Med 99:388, 1982.

29. Chierchia S, Patron0 C, Crea F, et al: Effects of intravenous prostacyclin in variant angina. Circulation 65470, 1982.

6. Neri Serneri GG, Masotti G, Poggesi L, Galanti G: Release or prostacyclin into the blood stream and its exhaustion in humans after local blood flow changes (ischemia and venous stasis). Thromb Res 17:197, 1980.

7. Mehta J, Mehta P, Horalek C: The significance of platelet-