Optimal platelet inhibition is a key therapeutic goal for patients undergoing percutaneous coronary intervention (PCI). The combination of aspirin plus clopidogrel currently represents the standard of care in this setting (1- 3). Clopidogrel loading affords more rapid and more potent platelet aggregation inhibition than the chronically administered dose (4). This better inhibition leads to clinical benefits in the context of PCI (5- 6) and is therefore currently recommended before interventions (1- 3,7- 8). Single loading doses up to 900 mg before PCI have failed to show significantly greater platelet inhibition than the 600 mg loading dose (7- 8). This limitation may be due to limited clopidogrel absorption (8), but time-dependent, cumulative efficacy also has been suggested (9). Clopidogrel loading doses of 600 mg at least 2 h before and 300 mg given at least 15 h before catheterization have been independently proposed as optimal dosages but have not been directly compared. In addition, the effects of 2 separate loading doses of 600 mg, given 18 to 24 h apart, have not been described. Therefore, the optimal clopidogrel loading dose remains unknown. The main objective of this randomized study was to compare the degree of platelet inhibition achieved by 3 different clopidogrel loading regimens, including a double 600-mg loading dose in patients undergoing elective angiography and PCI when appropriate. Our hypothesis was that a larger loading dose and optimal treatment duration would accelerate and potentiate the antiaggregatory effects of clopidogrel.

The protocol was approved by the institutional ethics committee, and patients gave written informed consent. Patients with suspected or documented coronary artery disease admitted to our hospital for elective coronary angiography and PCI when appropriate were considered eligible. Patients with unstable angina, acute or recent (<14 days) myocardial infarction; who were receiving glycoprotein IIb/IIIa inhibitors; suffered a stroke within 3 months; exhibited malignancies, active bleeding, or bleeding diathesis; underwent oral anticoagulation with a coumarin derivate or prothrombin time >1.5 times control; had recent treatment (<30 days) with a GP IIb/IIIa antagonist or thienopyridine; a platelet count <100,000/mm3; a serum creatinine >180 mmol/l; with severe liver disease resulting in abnormal bilirubin levels; with a known allergic reaction to thienopyridines (clopidogrel or ticlopidine); or had concomitant investigational drug use within 1 month were excluded.

Patients were randomly assigned to 1 of 3 loading dose regimens according to an automated randomization list (1:1:1 ratio) provided by the Montreal Heart Institute Coordinating Center (Montreal, Canada) in sealed envelopes: Group A, clopidogrel 300 mg the day before (≥15 h) + 75 mg the morning of the interventional procedure; Group B, clopidogrel 600 mg the morning of (≥2 h before) the interventional procedure; and Group C, clopidogrel 600 mg the day before (≥15 h) and 600 mg the morning of (≥2 h before) the interventional procedure. Clopidogrel loading doses were administered in an open label fashion by research nurses, but technicians performing aggregation studies remained completely blinded to group assignment and did not have any direct contact with patients. All stented patients received 75 mg of clopidogrel daily for at least 30 days as clinically indicated after implantation. Blood was collected at the time of randomization (baseline), immediately before coronary angiography, and the next morning (12 to 24 h) when a PCI was performed (post-PCI). All patients were examined for complications before hospital discharge and contacted by telephone at 1 month to assess for adverse events. Diagnostic and interventional procedures were performed according to standard clinical practice.

Blood was collected from the forearm into 2 4.5-ml Vacutainer tubes containing 0.5 ml of sodium citrate 3.2% (Becton Dickinson, Mississauga, Ontario, Canada). Platelet counts were measured with a Hematology system ADVIA60 (Bayer Inc., Toronto, Ontario, Canada). Platelet-rich plasma (PRP) was prepared by centrifuging blood for 10 min at 130 g. The remaining blood was further centrifuged for 15 min at 1,800 g to prepare platelet-poor plasma (PPP). Platelet-rich plasma was adjusted to platelet counts of 250,000 platelets/μl by adding PPP as needed. Platelet aggregation was performed in stirred (135 g) diluted PRP using a light transmission aggregometer (Model 570VS, Chrono-Log Corporation, Havertown, Pennsylvania). The PRP was used to set 0 light transmission and PPP to set 100 light transmission. Platelet aggregation was systematically measured after the addition of ADP at concentrations of 5 and 20 μmol/l (Sigma Chemical Co., Oakville, Ontario, Canada). Peak aggregation (Agg_peak) was measured at 1 to 2 min after the addition of the agonist, and late aggregation (Agg_6min) was measured at 6 min, accounting then for disaggregation (10). Percent inhibition of platelet aggregation was calculated as follows: (intensity of aggregation on treatment − intensity of aggregation at baseline)/(intensity of aggregation at baseline).

The primary end point of the study was the percent inhibition of Agg_peak at the time of angiography. Secondary end points included the percent inhibition of Agg_6min; the percent inhibition of Agg_peak post-PCI; the prevalence of nonresponders (resistance) to clopidogrel using 3 definitions (<10%, <20%, and <40% decrease in Agg_peak, respectively) (11- 14); the occurrence of death, myocardial infarction, or target vessel revascularization up to 30 days after procedure; and the occurrence of any of the following vascular or hemorrhagic complications: 1) major bleeding (defined as intracranial or clinically relevant bleeding associated with a decrease in hemoglobin of >5 g/dl); 2) minor bleeding (clinically overt hemorrhage associated with a fall in hemoglobin ≤5 g/dl); 3) access-site complications (hematoma, pseudoaneurysm, or arteriovenous fistula); and 4) thrombocytopenia with platelet count <70,000/mm³ or side effect requiring interruption of clopidogrel.

The primary assumption was that Group C regimen (600-mg double loading dose) would increase inhibition of platelet aggregation by at least 15% over that of Groups A and B. The study had 80% power to evaluate this difference in the primary end point with an alpha-level of 0.05 and 50 patients in each group, assuming a standard deviation of 25%. Continuous variables are expressed as mean ± SD, and their differences were tested with a one-way analysis of variance. Categorical variables are expressed as frequencies and percentage; the chi-square test was used unless otherwise indicated. The Fisher exact test was used when the row total was <10. A 1-way analysis of variance was used to compare the inhibition of platelet aggregation among the 3 groups. All pairwise comparisons were performed only if the global F test was significant at the 0.05 level. No adjustments were made to control for the family-wise error rate. A repeated measure analysis of covariance model, including baseline value, time (baseline, angiography, post-PCI), group (Group A, Group B, Group C), and the interaction term (time*group), was performed to determine the post-PCI inhibition of Agg_peak. If the interaction term was significant at the 0.05 level, then comparisons between groups at each time point were done. To quantify the changes between the time points, estimates were calculated to compare the groups on the absolute changes between the 3 time points. The changes between the time points are represented as percentages for clinical interpretation although the p values are calculated from the absolute changes. All analyses were done with SAS version 9.1 (SAS Institute Inc., Cary, North Carolina) and conducted at the 0.05 significance level.

The baseline characteristics of patients were well matched in the 3 study groups, except for diabetes which was more frequent in Groups A and B (Table 1). Adjusting statistically for diabetes did not alter any of the results.

Percent inhibition of Agg_peak at the time of angiography when platelets were stimulated with 5 or 20 μmol/l ADP is illustrated in (Figure 1). Inhibition of platelet aggregation was consistently better in the clopidogrel 600 mg double bolus group as compared with the other 2 regimens. Percent inhibition of Agg_peak was 31.4% in Group A, 29.0% in Group B, and 49.5% in Group C (p < 0.0001) when platelets were stimulated with 5 μmol/l ADP and 22.4%, 22.3%, and 39.8%, respectively, with 20 μmol/l ADP (p < 0.0001). The Agg_peak values did not differ at baseline (p = 0.29 with 5 μmol/l ADP and p = 0.24 with 20 μmol/l ADP) but were significantly different on treatment (p < 0.0001 with both 5 and 20 μmol/l ADP) (Table 2). Comparisons between Group C and other groups were highly significant (p = 0.0002 vs. Group A, p < 0.0001 vs. Group B), and those between Groups A and B were not (p = 0.62).

Percent inhibition of Agg_6min results are illustrated in (Figure 2). Late platelet aggregation inhibition was also consistently stronger in the clopidogrel 600 mg double bolus group as compared with the other 2 regimens. Percent inhibition of Agg_6min was 54.1% in Group A, 57.7% in Group B, and 81.1% in Group C when platelets were stimulated with 5 μmol/l ADP (p < 0.0001) and 36.7%, 40.5%, and 68.0%, respectively, with 20 μmol/l ADP (p < 0.0001). Agg_6min values did not differ at baseline (p = 0.09 with 5 μmol/l ADP and p = 0.12 with 20 μmol/l ADP) but were significantly different on treatment (p < 0.0001) (Table 3). Comparisons between Group C and other groups were highly significant (p < 0.0001 vs. Group A, p < 0.0001 vs. Group B), and those between groups, A and B were not (p ≥ 0.50). The more potent platelet inhibition observed at the time of angiography in Group C was maintained after PCI (n = 36). Percent inhibition of post-PCI Agg_peak (20 μmol/l ADP) was 21.0% in Group A, 28.4% in Group B, and 59.8% in Group C (p < 0.0001 Group C vs. A; p = 0.0005 Group C vs. B; p = 0.25 Group A vs. B).

Interestingly, there were no nonresponders to clopidogrel in Group C as compared with the other groups using <10% decrease in Agg_peak in response to 5 μmol/l ADP (9 in Group A, 8 in Group B and 0 in Group C, p = 0.005). When using <20% or <40% decrease as alternative definitions for nonresponsiveness, there remained a highly significant advantage of the double bolus regimen (Group C) as compared with the other 2 groups (p = 0.001 and p = 0.0002, respectively) (Figure 3A). Similarly striking results were seen when 20 μmol/l ADP was used (Figure 3B).

There was no death, rehospitalization for myocardial infarction, or repeat target vessel revascularization up to 30 days. Post-PCI cardiac troponin T elevations ≥0.1 μg/l were infrequent and equally distributed (2 [22%] in Group A, 2 [18%] in Group B and 3 [21%] in Group C, Fisher exact p = 1.0). There was no episode of major bleeding during 1 month follow-up. The incidence of minor bleeding was low and similar in the 3 groups, although numerically more frequent in Group B (3 in Group A, 8 in Group B, and 4 in Group C, p = 0.19) and mainly related to puncture sites (e.g., ecchymosis, oozing, hematoma). Clopidogrel up to 600 mg twice within 18 to 24 h was otherwise well tolerated. No patient developed clinically evident allergy to clopidogrel.

To our knowledge, this is the largest study to date comparing clopidogrel loading regimens and the first to report the effects of a double bolus loading strategy before an interventional procedure. Although other studies have compared increasing loading doses (up to 900 mg) given as single boluses only (7- 8), the present study was specifically designed to integrate the concepts of larger doses to accelerate and potentiate the antiaggregratory effects of clopidogrel and optimal treatment duration to benefit from time-dependent, cumulative effects. The most striking finding in this study is that a double bolus of 600 mg of clopidogrel achieves more potent platelet inhibition than either clopidogrel 600-mg single bolus or 300-mg single bolus ≥15 h before plus 75 mg the morning of an interventional procedure. Another important finding is that clopidogrel 300-mg single bolus given ≥15 h before plus 75 mg the morning of an interventional procedure provides equivalent platelet inhibition as a larger 600-mg single bolus given the morning of the procedure (≥2 h).

The most likely explanation for the better platelet inhibition achieved with a double bolus of clopidogrel lies in the higher plasma concentrations of clopidogrel and its active metabolite. von Beckerath et al. (8) have demonstrated that loading with a single bolus of clopidogrel 600 mg resulted in greater plasma concentrations of the active metabolite of clopidogrel as compared with loading with a single bolus of 300 mg. In the same study, loading with clopidogrel 900-mg single bolus failed to increase further the plasma concentrations of the active metabolite of clopidogrel and its unchanged form, strongly suggesting that intestinal absorption is the limiting step when single doses exceeding 600 mg are administered. A strategy of 2 separate clopidogrel 600 mg doses at 18- to 24-h interval would likely circumvent this limiting step and allow more absorption of the drug. Given that clopidogrel is an inactive prodrug that requires to be metabolized to an active compound and provided that the metabolism of clopidogrel does not become saturated, the double bolus strategy can be expected to achieve better inhibition of platelet aggregation, as was the case in this study. More specifically, a greater concentration of the active metabolite could lead to a more complete saturation of the P2Y12 receptors, which appears very partial at standard clopidogrel doses (15). In addition, if the circulating levels are high enough, it is not completely impossible that other platelet receptors be blocked concomitantly (e.g., P2Y1), further inhibiting aggregation.

Interestingly, our results highlight a time-dependent catch-up phenomenon after a 300-mg loading dose as compared with the 600-mg loading dose. Previous studies have convincingly demonstrated a more rapid and somewhat more potent inhibition of platelet aggregation when 600 mg of clopidogrel was compared with 300 mg during the first 24 h (12,16). However, these studies have not described the peak level of inhibition achieved when loading doses were administered at different intervals before intervention. Groups A and B in the current study did not differ in terms of the intensity of the inhibition of platelet aggregation at the time of the angiography despite a loading dose twice as large. The main difference between the respective regimens of both groups (in addition to the dosages) was the timing of the administration of the loading dose (>15 h vs. >2 h). It is possible that although the peak level of active clopidogrel metabolite reached after 300-mg was lower than with 600 mg, the length of exposure might have compensated if the clearance of the active metabolite is slower than previously described (8). The effect of the additional 75-mg dose the morning of the intervention might also have played a role to potentiate to the inhibitory effect of the 300-mg loading dose on the existing platelets or by inhibiting the platelets produced by the bone marrow in the interim. Therefore, a more modest loading dose given early enough (e.g., ≥15 h) could be equivalent to a larger loading dose given shortly before PCI.

The importance of nonresponsiveness (resistance) to clopidogrel may be underestimated clinically. Indeed, an increased level of inhibition in patients that would not have otherwise achieved an adequate response with standard dosing may have a larger impact than a further increase in those with an already adequate response. Many data now suggest that a poor response to clopidogrel is associated with increased event rates (13- 14,17- 21). Recently, Patti et al. (22) demonstrated that pretreatment with a 600-mg loading dose of clopidogrel reduced periprocedural myocardial infarctions by one-half in patients undergoing PCI when compared with a 300-mg loading dose. Buonamici et al. (23) identified nonresponsiveness to clopidogrel in vitro as a strong independent predictor of stent thrombosis in patients receiving drug-eluting stents. Furthermore, Bliden et al. (24) showed that patients undergoing nonemergent PCI who exhibit high ADP-induced platelet aggregation while being treated with clopidogrel are at increased risk for post-procedural ischemic events. The absence of nonresponders to clopidogrel when using the 600-mg double loading strategy in the present study suggests that this may represent a clinically relevant finding. Because we identified a clopidogrel loading regimen that achieves stronger platelet inhibition and decreases the proportion of nonresponders without apparently increasing bleeding, it is indeed reasonable to expect reductions in clinical events. However, larger studies are needed to confirm the clinical impact of this double loading dose. The hypothesis that stronger platelet inhibition is associated with better outcomes is now being tested with P2Y12 receptor blockers, both thienopyridines and nonthienopyridines. If this hypothesis is demonstrated efficacious and safe, the double loading dose as used in this study would become of even greater interest given the well documented efficacy of clopidogrel in large clinical trials. Until then, the double loading dose approach could be used in the large number of patients referred for coronary angiography and ad hoc PCI from other hospitals or as outpatient.

There are limitations to this study. First, open-label administration of clopidogrel was used. Because the primary end point was not clinical and all laboratory analyses were performed with researchers blinded to treatment assignments, no impact on the primary end point is expected. Second, the study was underpowered to detect meaningful differences in clinical events.

A clopidogrel 600-mg double loading dose (first dose given more than 15 h before and second dose given the morning of an interventional procedure) achieves greater platelet inhibition than conventional single loading doses. Clopidogrel administered as a 300-mg single bolus given ≥15 h prior to plus 75 mg on the morning of an intervention achieves equivalent platelet inhibition as a 600-mg single bolus loading dose given on the morning of the procedure.