Platelets play a major role in the formation of atherosclerotic plaques and thrombi and in the pathogenesis of ischemic stroke. Aspirin, and increasingly ticlopidine, are the two antiplatelet drugs commonly used in the prevention of stroke (1,2). Their mode of action is different (3). Aspirin irreversibly inhibits the enzyme cyclooxygenase and the formation of prothrombotic thromboxane A2 in the platelets. A drawback is the concomitant inhibition of antithrombotic prostacyclin in the vessel wall (2). The active metabolite and the exact effect of ticlopidine are not yet known. An inhibition of ADP receptors or fibrinogen receptors is discussed (4). Ticlopidine has been demonstrated to be more effective than aspirin in the prevention of stroke (1,2). The antiaggregatory effect of both substances as compared with placebo is well documented. However, to our knowledge, no data are available comparing the effect of aspirin and ticlopidine on platelet aggregation and other coagulatory and hematologic parameters in the same patients.
Ticlopidine has a potent inhibitory effect on ADP- and collagen-induced platelet aggregation (5-9). The effect on epinephrine-induced aggregation was shown to be less pronounced or absent (5,8). Aspirin inhibits platelet aggregation induced by epinephrine and collagen; ADP-induced platelet aggregation is inhibited to a lesser extent (9,10). Fibrinogen has been reported to decrease during long-term ticlopidine treatment (11). The Ticlopidine Aspirin Stroke Study comprised 3,069 patients. Besides the well-known phenomenon of possible neutropenia and increased cholesterol level, no other notable laboratory abnormality has been observed (1). The relative antiaggregatory effect of ticlopidine as compared with aspirin was shown to decrease with higher plasma fibrinogen levels (9).
In the present investigation, 45 patients with cerebrovascular disease were treated for 2 weeks with aspirin and for 2 weeks with ticlopidine in an open, prospective, and randomized cross-over study. The time of 2 weeks was chosen to comprise the mean platelet lifespan of 1 week. Blood samples were taken under each regimen and compared for coagulation and hematologic parameters. We were especially interested in platelet aggregability during treatment with the two medications and in a possible influence of platelet count and fibrinogen level on these parameters during such treatment.
PATIENTS AND METHODS
We examined 45 patients of the department of neurology, all with cerebrovascular disease. They were inpatients admitted for cerebral ischemia and outpatients with varying degrees of carotid stenosis who were examined in regular follow-up visits to our ultrasound laboratory. All these patients had a medical indication for antiplatelet therapy (history of cerebral or retinal ischemia, myocardial infarction, severe carotid stenosis or occlusion, carotid endarterectomy). Most of them had been receiving different doses of aspirin (100-500 mg daily); in the remaining patients, the indication for antiplatelet therapy was determined on their admission to our department. All of them continued antiplatelet therapy after the study. There were 34 men and 11 women, with a mean age of 63.6 years (range 49-88 years). Thirty-two had a history of cerebral or retinal ischemia; 22 had undergone carotid endartherectomy, 9 had a history of myocardial infarction, 3 had angina pectoris, and 19 had intermittent claudication. No patients were investigated in the acute phase. The most recent event was ocular ischemia 19 days before the beginning of the study. Twenty-nine were hypertensive, 14 had increased cholesterol levels, 3 had insulin-dependent diabetes mellitus, 5 had diabetes treated with diet and/or tablets, 15 were smokers, and 25 were exsmokers. Seven other subjects withdrew from the study because of possible side effects (abdominal discomfort 3, nausea 3, diarrhea 1); for 1 it was too inconvenient to come to the follow-up examination, and 1 patient died of heart arrest while still treated with his previous medication before having received the first study medication.
The patients gave informed consent. Previous antiplatelet medication (aspirin) was discontinued, and the patients were randomized to either 300 mg aspirin once daily or 250 mg ticlopidine twice daily for 14 days. While still receiving one of the two medications, they were reexamined on day 14. Venous blood was drawn and placed in tubes containing sodium citrate (twice: 10 ml). For the next 14 days, the patients received the other of the two medications and were reexamined on day 14 at the same time of day. Blood samples were taken again. Twenty-one patients received first aspirin and then ticlopidine, and 24 patients received first ticlopidine and then aspirin. In the patients first randomized to ticlopidine, an additional blood sample containing EDTA was taken for a blood count to detect possible neutropenia.
The following parameters were assessed by investigators blinded to the actual medication: erythrocyte, leukocyte, and platelet count, hematocrit, hemoglobin, mean corpuscular volume (MCV), fibrinogen, prothrombin time (PT), activated partial thromboplastin time (aPTT), thrombin time (TT), and platelet aggregability test (PAT), using ADP, epinephrine, and collagen.
Collection and preparation of plasma
The subjects tested in bed for at least 60 min in a supine position before blood samples were drawn. We used 21-gauge needles (no. 2. 0.8 × 38 mm) for venipuncture of a cubital vein. Eighteen milliliters of blood was drawn atraumatically without using a tourniquet and without suction into plastic tubes (Monovetten closed system, Sarstedt, Nuembrecht, Germany), containing 1 ml sodium citrate (0.11 M, 1 vol plus 9 vol blood). The two blood tubes of each patient were gently inverted for mixing. One of them was centrifuged at 150 g for 10 min at room temperature to prepare platelet-rich plasma (PRP). For platelet poor plasma (PPP); the other tube was centrifuged at 1,500 g for 15 min. The PRP was transferred to a new plastic container, and the platelet count was adjusted to platelets 250-300/nl with PPP from the same sample. Before testing, the PRP was held at room temperature for at least 30 min. Testing was then completed within 3 h after blood collection.
Platelet function test
Aggregation of platelets in PRP was measured photometrically by the increase of light transmission in an aggregometer device (PAP 4, Biodata, Hatboro, PA, U.S.A.). The induction was performed with epinephrine, ADP, and collagen (from calf skin) to a final concentration in the PRP of 2.5 μM, 2 μM, and 0.9 g/L, respectively. The aggregometer blank was prepared by pipetting 0.5 ml PPP in a plastic tube. PRP 450 μl was placed in siliconized plastic tubes in the PAP 4 device and incubated for 2 min at 37°C. A stirring bar and 50 μl of each induction agent (Biodata) were added, and the reaction was started. Within 20 min, the maximum aggregation was measured as maximum transmission in percent of the PPP transmission. Thus, 100% transmission (i.e., transmission of PPP) indicates no inhibition and 0% indicates full inhibition of platelet aggregation.
Blood cell count
Erythrocytes, leukocytes, and platelets were counted and sized in a Sysmex M 2000 device (Toa Medical Electronics, Kobe, Japan) by detecting the difference in conductivity between the particles and the suspending diluent. Hematocrit was measured electrically by integrating cell volume pulses and the volume of the whole blood. MCV was calculated automatically from the red blood cell count and the hematocrit.
Fibrinogen and the global screening tests PT (Dade Innovin, ISI = 1.01, Baxter) and aPTT (Actin-FS, Baxter) were performed on an Electra 1000 C compound (MLA, Baxter, Unterschleissheim, Germany) according to the principle of decrease in scattered light. TT was assessed with the Thrombin Reagenz (Boehringer Mannheim, Germany).
For statistical analysis, we used Wilcoxon tests for all values during treatment with the two medications and linear regression analysis comparing platelet count and fibrinogen levels with platelet aggregability. Statistical significance was declared at the p < 0.05 level.
There were no ischemic events and no hemorrhagic side effects during the study period. The results of the blood parameters during treatment with the two medications are shown in Table 1, including the significance levels from Wilcoxon tests.
Linear regression procedures were performed to test the influence of platelet count and fibrinogen on ADP-, collagen-, and epinephrine-induced platelet aggregability during treatment with the two medications. No relation was observed between fibrinogen and platelet count in a linear regression analysis. Neither fibrinogen nor platelet count had an impact on ADP-, collagen-, or epinephrine-induced aggregability during aspirin treatment. There was a weak relation between fibrinogen level and collagen-induced aggregability during ticlopidine treatment. The equation was as follows: collagen-induced aggregability = 108.9 - 7.13 × fibrinogen (R = 0.322, adjusted R2 = 0.083, p = 0.029). A more significant relation was noted for platelet count and ADP-induced aggregability during ticlopidine treatment. The corresponding equation was as follows: ADP-induced aggregability = -9.41 + 0.132 × platelet count (R = 0.476, adjusted R2 = 0.209, p < 0.001). No influence was noted for fibrinogen and platelet count on epinephrine-induced aggregability during ticlopidine treatment. Notably, there was no effect of fibrinogen level on ADP-induced aggregability during ticlopidine treatment (p = 0.54, adjusted R2 = 0.014).
To our knowledge, studies comparing platelet aggregability during treatment with the two antiplatelet drugs ticlopidine and aspirin in a cross-over study in patients with cerebrovascular disease have not been performed previously. Our study demonstrates that the different modes of action of the two antiplatelet drugs are reflected by different characteristics of platelet aggregability. ADP-induced aggregability was less during aspirin treatment than during ticlopidine treatment. Collagen-induced and epinephrine-induced aggregabilities during aspirin treatment were less than those during ticlopidine treatment. These findings are in concordance with results of previous studies comparing aggregability before and after medication (5,6,8,10). Because ticlopidine is more effective than aspirin in preventing stroke, differences in ADP-induced platelet aggregability particularly may account for the different clinical effects of the two drugs (1,2).
Absolute platelet count was higher during ticlopidine treatment than during aspirin treatment. In an animal study, this effect was also observed (12). Ticlopidine has been shown to increase platelet survival, reflecting less platelet activation and degradation (13), which may explain the higher platelet count observed in our study. There was a marginally significant tendency for ticlopidine, as compared with aspirin, to produce a higher hematocrit and hemoglobin level.
No difference was noted for fibrinogen concentrations, in concordance with the findings of Tohgi and colleagues (9) and in contrast to the findings of Randi and co-workers (11), who reported a decrease in fibrinogen during ticlopidine treatment. No difference in the other clotting parameters was observed, as we had expected, taking into account the action on platelets and not on intrinsic coagulation.
There should be no emphasis on the effect of fibrinogen on collagen-induced aggregability during ticlopidine treatment, as the adjusted R2 value was low (0.083). Collagen-induced aggregability was lower during higher fibrinogen levels. A more significant correlation was observed for the impact of platelet count on ADP-induced aggregability during ticlopidine treatment. Our results suggest that ticlopidine is more effective in inhibiting ADP-induced platelet aggregation when the absolute platelet count is low. No influence of aspirin therapy was observed. Studies involving more patients with considerably increased platelet counts are needed to draw clinical conclusions, i.e., studies in which patients with thrombocytosis are treated with aspirin or higher doses of ticlopidine rather than with 250 mg ticlopidine twice daily.
We could not demonstrate any effect of fibrinogen levels on ADP-induced platelet aggregation during treatment with the two medications. In a previous study an impact of fibrinogen level on ADP-induced platelet aggregation was reported in patients (9). The relative anti-aggregatory effect of ticlopidine was significantly decreased with higher plasma fibrinogen concentrations. No such effect was observed for the aspirin therapy. The study of Tohgi and colleagues (9) compared aggregation before and during treatment in patients who had not previously been treated with antiplatelet drugs. Because almost all our patients were already receiving aspirin and there was a medical indication to treat those patients with an antiplatelet drug, no values for our patients before initiation of treatment were available. The different design of the two studies may account for this incongruity.
Acknowledgment: We thank Dr. S. Brückner for reviewing the manuscript and Mrs. A. Jahn, Mrs. A. Jurat, and Mrs. M. Haerle for excellent technical assistance. We thank Sanofi Winthrop, Munich, Germany (Fr. Dr. I. Kempf) for bibliographical and financial support; Dr. H. J. Friedrich, Department of Medical Statistics and Documentation, for help with the statistical interpretation of the data; and Professor Dr. D. Kömpf.
1. Hass WK, Easton JD, Adams HP, Jr, et al. A randomized trial comparing ticlopidine
hydrochloride with aspirin for the prevention of stroke in high-risk patients. N Engl J Med
2. Weksler BB, Pett SB, Alonso B, et al. Differential inhibition by aspirin of vascular and platelet prostaglandin synthesis in atherosclerotic patients. N Engl J Med
3. Coller B. Platelets and thrombolytic therapy. N Engl J Med
4. McTavish D, Faulds D, Goa KL. Ticlopidine
: an updated review. Drugs
5. Berglund U, von Schenck H, Wallentin L. Effects of ticlopidine
on platelet function in men with stable angina pectoris. Thromb Haemost
6. Knudsen JB, Kjoller E, Skagen K, Gormsen J. The effect of ticlopidine
on platelet functions in acute myocardial infarction. Thromb Haemost
7. Haffrand JP, Bernat A, Delebassée D, Defreyn G, Cazenave JP, Gordon JL. ADP plays a key role in thrombogenesis in rats. Thromb Haemost
8. Cattaneo M, Akkawat B, Lecchi A, Ciminiello C, Capitanio AM, Mannucci PM. Ticlopidine
selectively inhibits human platelet responses to adenosine diphosphate. Thromb Haemost
9. Tohgi H, Takahashi H, Kashiwaya M, Watanabe K. Effect of plasma fibrinogen concentration on the inhibition of platelet aggregation after ticlopidine
compared with aspirin. Stroke
10. Tohgi H, Tamura K, Kimura B, Kimura M, Suzuki H. Individual variation in platelet aggregability and serum thromboxane B2 concentrations after low-dose aspirin. Stroke
11. Randi ML, Fabris F, Crociana ME, Battocchio F, Girolami A. Effects of ticlopidine
on blood fibrinogen and blood viscosity in peripheral atherosclerotic disease. Arzneimittelforschung
12. Boudreaux MK, Jeffers C, Lipscomb D, Spano JS. The effect of ticlopidine
on canine platelets, fibrinogen, and antithrombin. J Vet Pharmacol Ther
13. Wilkinson AR, Hawker RJ, Hawker LM. The influence of antiplatelet drugs on platelet survival after aortic damage or implantation of a dacron arterial prosthesis. Thromb Res