METHODS

Patients

Thirty patients with stable stage II PVD, who also were smokers, were recruited from those attending the angiology section of the clinical and experimental medicine department, Perugia General Hospital, Italy.

PVD had originally been confirmed by ultrasound screening of the femoral tracts by high resolution ultrasonography Biosound II s.a. (Esaote Biosound, Florence, Italy) equipped with an 8 Mhz probe, combined with Doppler velocimetry (including the treadmill test), and spectral analysis.

Inclusion criteria were male sex, ankle/arm pressure ratio < 0.80 at rest, stable maximum walking distance of 200 to 500 m, and a maximum walking time (MWT) of 180 to 300 seconds at the last two monthly checkup (reconfirmed immediately before enrolment in our study) during a standard treadmill test.

Exclusion criteria were rest pain, diabetes mellitus, hyperlipidaemia (serum cholesterol, > 240 mg%; serum triglycerides, > 160 mg%), and hypertension (systolic blood pressure, >160 mm Hg; diastolic blood pressure, > 90 mm Hg), vascular disease in other regions, any other serious disease, recent infection, history of vascular surgery and/or acute thromboembolism, and chronic alcoholism.

Study plan

One month before randomisation, all 30 patients started a normocaloric diet and began an outpatient physical training programme¹⁶; one week later they suspended antiplatelet treatment.

Table 1 reports the clinical details of the 30 patients.

Informed consent was obtained from all patients, and our study was approved by the Umbria hospital board ethics committee.

The patients were randomised to two treatment groups in a double blind, placebo controlled design by a predetermined computer code to ensure homogenous distribution of inclusion criteria.

Table 2 illustrates the double blinded, placebo controlled study design.

Biological safety parameters included a clinical checkup, blood pressure measurement, and a complete haematological profile before and after Clopidogrel administration.

Treadmill test model

Each exercise test was composed of two treadmill tests.

TcPO₂ was monitored constantly during the two treadmill tests (10% slope, 3 km/hour) and the recovery period between the two walks. A TcPO₂ monitor (Ts 3300; Kontron Instruments, Basel, Switzerland) was used, with one electrode being positioned 10 cm below the knee on the anterior tibial muscle. The second electrode was placed on the chest in the subclavicle area for control readings.

When walking was stopped because of pain, the time of recovery to half the original TcPO₂ value was monitored. It was calculated according to the following formula: the difference between the basal TcPO₂ value and the minimum was divided by two and the results were added to the minimum value. Walking was immediately resumed until calf pain became prohibitive and the walking time was monitored.

Blood measurements

Blood samples were drawn from an antecubital vein at baseline and after the second treadmill test. The following parameters were determined:

• Whole blood viscosity at corrected haematocrit as described previously,^17,18 with an intra-assay coefficient of variation of 1.1%.

• Leucocyte activation, measured as pseudopodia formation (morphological assay of white blood cell activation), as described elsewhere.¹⁹

• The leucocyte filterability rate after separation and suspension in phosphate buffered saline.

• Plasma rich in white blood cells but relatively few red blood cells and platelets was obtained according to the method of Mikita et al.²⁰

• The erythrocyte filterability rate after erythrocytes were separated as described elsewhere.²¹ All filtrations were carried out at room temperature (25 ± 1°C). The methodology of filtration has been described in detail elsewhere.²² The filterability rate was expressed as the final filtration pressure generated by the cell suspension relative to the buffer, previously filtered through the same filter (a 5 μm pore filter; nuclepore, batch 54B5E2; Costar Corporation, Cambridge, Massachusetts, USA). The reliability of the filtration procedures was confirmed by the coefficients of variation from five successive filtrations of each suspension sample, which were 2.9% for erythrocytes and 3.1% for leucocytes.

• Platelet activation evaluated as concentrations of soluble P-selectin, measured by a standard enzyme linked immunosorbent assay (Takara Shuzo Ltd, Japan), which showed intra-assay and interassay coefficients of variation of 2.5% and 5.4%, respectively. The minimum sensitivity for the assay was 1.3 ng/ml.

• Total platelet and leucocyte counts assayed by an automatic analyser.

• Plasma concentrations of fibrinogen were assayed by radial immunodiffusion on Partigen plates (NOR Partigen plates, Behringwerke, Germany).

• The haematocrit was measured by microcentrifugation at 12.500 ×g for five minutes and results were expressed as the percentage of erythrocytes in the total blood volume.

Statistical analysis

Comparisons between groups were performed using the Student’s t test for parametric variables and Wilcoxon rank sum test for non-parametric variables. Repeated measures analysis of variance was used for testing intragroup differences at different times. Correlations were tested with Pearson’s and Spearman’s rank tests. The data were analysed longitudinally (comparing baseline values with those at the MWT and at the TcPO₂ half recovery times and comparing data before and after treatment) and cross sectionally (comparing data in the Clopidogrel group with data in the placebo group). Only p values < 0.05 were taken as significant and are reported here.

RESULTS

The clinical details of 30 male subjects with peripheral arterial occlusive disease (PAD) are reported in table 1. Age, cardiovascular risk factors, and mean Windsor index were evenly distributed in the two groups of patients with PVD.

Table 2 illustrates the double blinded, placebo controlled study design. Briefly, 15 patients with PAD, assigned to antiplatelet drug treatment, took 300 mg of Clopidogrel after a repeated treadmill test on day 1 and 75 mg on days 2 to 4, completing their last treadmill test on day 5. Fifteen matched subjects completed the same study protocol by taking placebo instead of Clopidogrel (table 2). Blood samples were collected for all patients before and after the treadmill tests.

Table 3 shows the haemorheological behaviour of all the participants during the repeated treadmill exercise tests. At baseline (first day), all markers of platelet and leucocyte activation, in addition to low shear rate WBV, rose significantly after the treadmill exercise both in subjects assigned to Clopidogrel and in control patients. Thus, for instance, in the placebo group, 51% and 11% increases in WBC activation and P-selectin levels, respectively, were accompanied by a significant 11% increase in WBV after exercise. Similar changes were seen in those patients with PAD who were assigned to Clopidogrel (table 3).

On day 2, similar haemorheological post exercise patterns to those seen on day 1 were documented in the placebo and Clopidogrel groups, except for the soluble P-selectin plasma values in the Clopidogrel group, which did not change after repeated exercise (1.6% decrease after exercise; p = NS (not significant)). Conversely, despite the use of Clopidogrel, the exercise test still induced increases in WBC activation (from 10.7% (SD, 1.2%) to 16.1% (SD, 2.5%); p < 0.01) and WBV (from 26.6 cP (1.0) to 29.5 cP (1.1); p < 0.01).

On day 5, in the Clopidogrel group, soluble P-selectin plasma values continued to be unaffected by exercise (1.6% reduction after exercise; p = NS). Interestingly, markers of leucocyte activation and WBV remained stable after the treadmill test; thus, post exercise variations in leucocyte activation (4.5% increase), leucocyte filterability (0% variation), and WBV (0% variation) were not significantly different (p = NS for all). In the placebo group, changes after exercise in the haemorheological profile were not significantly different from those seen on day 1 and day 2 (increases in WBV, WBC activation and filterability, soluble P-selectin values, platelet and WBC counts; table 3).

When analysing the longitudinal behaviour of blood parameters (day 5 v day 1), we found that post exercise WBV (−10%; p < 0.01), WBC activation (−27%; p < 0.01), and filterability of unfractionated leucocytes (−20%; p < 0.01), in addition to plasma soluble P-selectin values (−13%; p < 0.01), were all reduced on day 5 compared with day 1 in the Clopidogrel group.

Table 4 reports TcPO₂ half recovery times and MWT in those patients assigned to either Clopidogrel or placebo during the study programme. The TcPO₂ half recovery times and MWT were not dissimilar between the two groups on days 1 and 2. However, a significant difference between the groups emerged on day 5, with TcPO₂ half recovery times being lower (Clopidogrel group, 183 seconds (SD, 13) v placebo group: 221 seconds (SD, 17); p < 0.01) and MWT higher (Clopidogrel group, 320 seconds (SD, 62) v placebo group 241 seconds (SD, 45); p < 0.01) in patients with PAD taking Clopidogrel. Both the TcPO₂ half recovery times and MWT were significantly ameliorated by Clopidogrel treatment on day 5 (183 seconds (SD, 13) and 320 seconds (SD, 62), respectively) compared with day 1 (222 seconds (SD, 14) and 236 seconds (SD, 44), respectively). No variation of these functional parameters was present in patients with PAD assigned to placebo.

Thus, improvement of indicators of vascular perfusion—namely, TcPO₂ half recovery time and MWT—parallelled the improvement in the haemorheological profile only in patients taking Clopidogrel for the duration of the study (four consecutive days).

At no stage of the study were significant changes seen in fibrinogen, haematocrit, or the erythrocyte filterability rate (results not shown). No patients dropped out of our study and no side effects were reported to, or observed by, the physician.

DISCUSSION

Our experimental data showed that repeated exercise in patients with stage II PVD was associated with platelet activation, as documented by post exercise increased soluble P-selectin values and platelet count, in addition to white blood cell activation and increased blood viscosity. This is in agreement with our previous report and those of others reporting either the activation of leucocytes or platelets after repeated treadmill tests.^13,14 Because exercise did not produce changes in other parameters that are known to contribute to blood viscosity (that is, fibrinogen, red blood cell deformability, and haematocrit), changes in leucocytes and platelet behaviour might have had an influence on the observed post exercise increase in blood viscosity. This is supported by our results showing the lack of an increase in post exercise low shear rate blood viscosity when both post exercise platelet and white blood cell activation were suppressed in subjects taking Clopidogrel for the entire duration of our study. In our present study, we monitored both platelet and white blood cell characteristics, in addition to blood viscosity at low and high shear rate. Moreover, by using an antiplatelet agent like Clopidogrel, we investigated the effects of platelet inhibition on microvascular blood viscosity. Our results showed that an interaction between leucocytes and platelets occurred in patients with PVD at the onset of calf pain. In particular, our data indicated that a significant post exercise reduction in blood viscosity at low shear rate was reached only when platelet activation was inhibited and white blood cell rheology improved. Indeed, the observed post exercise increase of low shear rate blood viscosity on days 1 and 2 was not seen in the Clopidogrel group on day 5, when post exercise soluble P-selectin concentrations, WBC activation, and filterability remained unchanged after the treadmill test. Thus, the observed antiplatelet effects of Clopidogrel might be accompanied by effects on WBC activation, probably as a consequence of an interaction between platelets and white blood cells.

“A significant post exercise reduction in blood viscosity at low shear rate was reached only when platelet activation was inhibited and white blood cell rheology improved”

Clopidogrel had no effect on plasma fibrinogen concentrations or red blood cell deformability, which remained substantially unaltered throughout the study. This discrepancy with other reports^21,22 can be explained by our short term administration of the drug (four days). A previous study showed that ticlopidine derivatives (including Clopidogrel) improved red blood cell deformability by reducing erythrocyte membrane microviscosity.²¹ This effect seems to depend on lipid–lipid and cholesterol–phospholipid bond variations and appears to be common to cell substances which, like Clopidogrel, are amphiphiles.²³ A recent study of ours²⁴ showed that Clopidogrel improved red blood cell deformability after three weeks in subjects with subclinical atherosclerosis. Interestingly, in the same study we found that whole blood viscosity improved after seven days of treatment with 75 mg Clopidogrel daily.²⁴

Another interesting point to emerge from our study is that, although platelet activation was inhibited by the loading dose of Clopidogrel, on day 2 the single large dose had no significant effect on MWT, TcPO₂ half recovery time, or other functional measurements. Half recovery time improved only when white blood cell rheology improved and platelet activation was inhibited. Thus, a concerted platelet–leucocyte action is undoubtedly present at the onset of calf pain, and might have an important influence on microvascular perfusion. Interestingly, the shortening in half recovery time correlated with the improvement in low shear rate blood viscosity (results not shown).

In conclusion, our study provides evidence that platelet–leucocyte adhesive interactions occur at the onset of calf pain in claudicants and that the suppression of the post exercise activation of both cell populations does not allow low shear rate whole blood viscosity to increase after exercise, thereby improving the half recovery time after the treadmill test. Clopidogrel, by inhibiting platelet activation and aggregation, might also blocks the vicious cycle of leucocyte entrapment, plugging, and activation, thus improving the functioning of the microcirculation.

Clinical trials should be started to determine whether Clopidogrel has a place in the pharmacological management of peripheral arterial occlusive disease.