In patients with prior stroke or transient ischemic attack, treatment with dipyridamole substantially reduced stroke recurrence, with a beneficial effect comparable to and additive with that induced by aspirin (1). The mechanisms of dipyridamole-induced neuroprotection are probably complex and multifactorial (2), including the antiplatelet effect (3,4), a possible prolonged angiogenetic effect (5,6), an indirect adenosine-mediated neuroprotection, and an antioxidant effect (7). According to this "radical theory," ischemic tissue injury associated with ischemia-reperfusion determines an increased oxidative stress, which can contribute significantly to worsening of tissue injury (8). The radical mechanism might be effectively counteracted by dipyridamole, which shows direct powerful antioxidant properties in vitro (9-11).
We used carotid cross-clamping in carotid endarterectomy as a clinical model to assess the interaction between cerebral hypoperfusion, oxidative stress, and dipyridamole administration. In this clinical model, the acute brain hypoperfusion and reperfusion is known to provoke a significant increase in oxidative stress in the brain (12). We therefore serially assessed vitamin E plasma concentrations in the jugular vein ipsilateral to carotid cross-clamping before, during, and after carotid endarterectomy in 21 patients. In 12 of them, lipoperoxides also were assayed to have an additional index of oxidative stress different from and complementary to vitamin E.
Our study hypothesis was that cerebral ischemic oxidative stress, evaluated through serial assessment of plasma vitamin E and lipoperoxide concentration in the jugular vein ipsilateral to the clamped carotid, can be significantly decreased by pretreatment with oral dipyridamole in patients undergoing carotid endarterectomy.
Carotid endarterectomy patients (age, 65 ± 10 years; 11 men and 10 women) were studied in the Neurosurgery Clinic of the University of Pisa. The study was approved by the Human Research Committee of the Institute of Clinical Physiology. All patients gave informed consent to be included in this study. Patients were randomly allocated in the two groups (I, treated; II, untreated). Treated patients received the slow-release preparation of dipyridamole, 200 mg p.o. (the same dosage used in the ESPS-2 study) (1), 3-4 h before surgery; untreated patients received a corresponding placebo.
All patients had symptomatic severe carotid stenosis (>70% diameter reduction) and underwent carotid endarterectomy under general anesthesia, according to the North American Carotid Endarterectomy Trial criteria (13). None of the patients was taking multivitamin preparations or over-the-counter herbal medications.
Anesthesia was induced with thiopentone (4 mg/kg, i.v.) and succinylcholine (1 mg/kg). After intubation, muscle relaxation was obtained by administration of pancuronium. In the preclamp time, the patients were ventilated with 60% nitrous oxide in oxygen and 0.6-1.2% isofurane concentration to maintain an adequate level of anesthesia. During carotid cross-clamping, isofurane concentration was lowered to 0.2-0.6% values to allow a higher systemic arterial blood pressure value (>160 mm Hg). Opioid was never used. Ventilation was mechanically controlled to maintain mild hypocapnia with PaCO2 ranging from 32 to 35 mm Hg.
Somatosensory potentials and mean middle cerebral artery velocity were continuously recorded, as previously described (14,15), and intermittently analyzed at the time points corresponding to blood sampling from jugular vein: at baseline, at 15 and 30 min of clamping, at 2 and 10 min after declamping. According to Halsey's criteria (15), ischemia was classified as severe if the mean cerebral artery velocity during the first minute after clamping was 0-15% (grade 1), mild if it was 16-40% (grade 2), and absent if it was >40% (grade 3) of preclamp values.
All collected samples were coded and analyzed in the Biochemistry laboratory of the Institute of Clinical Physiology of the National Research Council by observers blinded to the study changes and patient condition.
Blood sample processing
Blood samples were collected from the jugular vein: before carotid clamp (basal), at 15 and 30 min during carotid clamp, and at 2 and 10 min after declamping. With lithium heparin as anticlotting agent, the samples were immediately cooled in ice, centrifuged at 1,100 g for 10 min; plasma was frozen in liquid nitrogen and stored at −80°C. All plasma samples were routinely analyzed to assess total cholesterol concentration (mg/dl).
Biochemical parameters of oxidative damage
Vitamin E assay. After addition of 50 μl of 50 mM ethanol solution of butylated hydroxytoluene to 60 μl of plasma, vitamin E was extracted according to the procedure described by Lang et al. (16). Plasma aliquots were mixed with 1 ml of 100 mM sodium dodecylsulfate solution in bidistilled water, 2 ml of ethanol/isopropanol (95:5 vol/vol), 2 ml of n-hexane, and mixed with a vortex mixer for 2 min. The hexane phase was separated by centrifugation; 1.5-ml aliquots were evaporated under nitrogen flux and resuspended in 700 μl of methanol. Determinations by high performance liquid chromatography (HPLC) were performed using a 5-μm ODS C-18 reverse phase column (Beckman) with 1 ml/min of methanol as eluent and a fluorimeter operating at 286 nm excitation and 330 nm emission wavelengths.
Because the protection of lipoprotein lipids from oxidation depends on the relative concentration of α-tocopherol to total lipid, plasma vitamin E levels were expressed both in absolute concentration (μM) and as the ratio of α-tocopherol/total cholesterol (mmol/mol), as described (17), and they were plotted as percentage of the basal value measured before surgery.
Plasma lipoperoxide assay. To measure plasma hydroperoxides, the analytic method patented by Diacron s.r.l. (d-ROMs test) was used (18). The principle of this method is based on the capacity of transition metals, when changed from the chelated state into transport and deposit proteins, as generally found in plasma, to catalyze reactions of formation of reactive oxygen species following Fenton's reaction or on radical propagation during lipid peroxidation. The oxyradical species produced, whose quantity is directly proportional to the quantity of plasma peroxides, were trapped by an alchylamine, a phenolic compound able to react forming a colored stable radical detectable by photometry at 505 nm.
The results were expressed in arbitrary units (a.u.). The value of 1 a.u. corresponded to a concentration of 0.08 mg/dl of hydrogen peroxide. The normal range of plasma hydroperoxides has been determined using >4,500 clinically asymptomatic subjects (18,19), and it has been estimated between 250 and 320 a.u.
The d-ROMs test kit was kindly provided by DIACRON as a gift.
Data are expressed as mean ± SEM. Variables were compared using analysis of variance (ANOVA) for repeated measurements both between and within groups, using Bonferroni/Dunn test as a post hoc test. A probability (p) value <0.05 was considered to be statistically significant.
The two groups (treated and untreated) were similar for age, gender distribution, and concomitant medications (Table 1). The average time of carotid cross-clamping was 49.7 ± 4.0 min for group I and 51.4 ± 2.8 min for group II.
By somatosensory evoked potentials, central conduction time was similar in the two groups at baseline (group I, 6.3 ± 0.1 vs. group II, 6.3 ± 0.2 ms; p = NS) at 15 min after clamping (group I, 6.7 ± 0.1 vs. group II, 6.5 ± 0.3 ms; p = NS), at 2 min after declamping (group I, 6.6 ± 0.1 vs. group II, 6.4 ± 0.2 ms; p = NS) and at 10 min after declamping (group I, 6.47 ± 0.2 vs. group II, 6.6 ± 0.3 ms; p = NS).
Transcranial Doppler findings
Carotid cross-clamping induced a decrease in middle cerebral artery velocity, with an average reduction of 23.7% ± 5.8 in group I and 33.1% ± 7.1 in group II over baseline value, indicating a condition of mild ischemia, according to Halsey's criteria in both groups (15).
By transcranial Doppler, mean velocity in middle cerebral artery was similar in the two groups. Mean velocity in the middle cerebral artery was 450 ± 40 mm/s (group I) and 490 ± 40 mm/s (group II) at baseline (I vs. II; p = NS); 340 ± 40 mm/s (group I) and 330 ± 60 mm/s (group II) at 15 min after clamping (p < 0.05 vs. baseline for both groups); 478 ± 58 mm/s (group I) and 511 ± 40 mm/s (group II) at 2 min after declamping (p < 0.05 vs. baseline; p = NS vs. clamping); 474 ± 42 mm/s (group I) and 511 ± 33 mm/s (group II) at 10 min after declamping (p < 0.5 vs. baseline; p = NS vs. clamping).
The mean plasma α-tocopherol and cholesterol concentrations before surgery were 19.2 ± 2.1 μM and 198 ± 15 mg/dl for group I, and 23.2 ± 1.9 μM and 224 ± 16 mg/dl for group II. The cerebral hypoperfusion, secondary to carotid cross-clamping, induced a significant decrease of plasma α-tocopherol content in the patients of group II (Fig. 1). Vitamin E content before clamping was 3.71 ± 0.22 mmol/mol of cholesterol (100%), decreased to 3.39 ± 0.2 mmol/mol (91.37% of baseline) 15 min after carotid cross-clamping (p < 0.01 vs. baseline) and 3.29 ± 0.17 mmol/mol (88.68% of baseline) at 10 min after reperfusion (p < 0.01 vs. baseline). Conversely, in patients of group I, a slight, insignificant decrease of plasma vitamin E was measured during the entire intervention period [Fig. 1; 3.48 ± 0.4 mmol/mol of cholesterol (100%) at baseline versus 3.43 ± 0.41 mmol/mol (98.56%) at 10 min after reperfusion; p = NS]. Analysis of variance for repeated measurements comparing the two groups revealed a significant effect of vitamin E levels. The post hoc Bonferroni/Dunn test indicated a decrease in plasma tocopherol of placebo group; p < 0.001.
Plasma lipoperoxide amounts measured at baseline were in the reference range (group I, 302 ± 8.5 a.u.; group II, 313 ± 4.9 a.u.). In group II, lipoperoxide concentration increased after 15 min of clamping to an interval of values indicating a condition of slight oxidative stress (335 ± 6.1 a.u.; p < 0.05 vs. rest), at 30 min of clamping reached a level of oxidative stress (352 ± 9.2 a.u.; p < 0.01 vs. rest), and decayed very slowly after 10 min of reperfusion (336 ± 4.5 a.u.; p < 0.05 vs. rest). Lipoperoxide production in patients of group I never exceeded the reference range during the carotid endarterectomy intervention (Fig. 2). Analysis of variance for repeated measurements to compare the two groups showed a significant effect of plasma lipoperoxides. The Bonferroni/Dunn test indicated an increased lipoperoxide level in the control group; p < 0.01.
Indices of oxidative stress, such as vitamin E consumption and increase of lipoperoxides, briefly increase in the jugular vein ipsilateral to carotid cross-clamping during carotid endarterectomy. It is generally accepted that reperfusion poses an increased oxidative stress, whereas the reduction of antioxidant systems during ischemia may appear paradoxic. However, we have previously reported in the rat (20) and in humans (21) that ischemia per se reduces cellular antioxidant defenses. The lowering of vitamin E during ischemia to a level comparable to that achieved after postischemic reperfusion may be explained by oxidation of α-tocopherol by free radicals produced in the globally ischemic but not totally anoxic tissue.
The reduction of plasma vitamin E and the increase of plasma lipoperoxides that we observed were relatively small, so they may not have real clinical implications because patients did not generally have major stroke or other ischemic attacks after surgery. Carotid endarterectomy is a discrete episode of focal cerebral hypoperfusion followed by reperfusion. However, it is not so severe as in other experimental animal models (8,22) because the short period of ischemia and presence of collateral circulation combine to reduce the extent of injury. The major advantage of this model is its clinical substrate to study changes in levels of biologically significant molecules during current surgical procedures. Direct evidence of oxidative-mediated cellular damage would be the optimal methods to demonstrate this phenomenon. Unfortunately, in this type of surgery, brain tissue was not available for analysis. Our results do give support to the evidence that antioxidants (α-tocopherol) are consumed during hypoperfusion-reperfusion, probably as a result of free radical generation. Recently it was demonstrated that plasma vitamin E favorably affects endothelium-dependent vasodilator function in patients with coronary atherosclerosis (23).
This increase in circulatory indices of cerebral oxidative stress is blunted by pretreatment with oral dipyridamole.
It has been demonstrated that dipyridamole behaves in vitro as an antioxidant twice as effective as α-tocopherol in inhibiting lipid peroxidation of methyl linoleate (9), and in the oxidation of low-density lipoprotein (LDL) induced chemically and by endothelial cells (10). Moreover, we observed that dipyridamole prevents lipid peroxidation and endogenous antioxidant consumption in a ex vivo model of oxidative stress chemically induced in red blood cell suspensions (24).
The findings reported here could be consistent with an in vivo antioxidant effect of dipyridamole, which in theory may contribute to its documented therapeutic virtue as a neuroprotective agent (2,7).
Dipyridamole has a documented effect in secondary prevention of stroke. The mechanisms of this powerful protective effect are, however, still elusive and incompletely understood.
The antiplatelet effect of dipyridamole (3,4) and its adenosine-mediated neuroprotective effect can be important. However, at least in theory, the antioxidant effect of dipyridamole per se might be important in attenuating tissue damage-which can be triggered, amplified, and propagated by the radical burst of ischemia and reperfusion. Obviously, as has been repeatedly emphasized, "demonstration of increased free radical damage is not the same as proof that it is important" (25). Nevertheless, it is highly likely that oxidative stress is an important variable in the complex equation linking ischemia in a clinical model to tissue damage. Our documentation of an antioxidant effect of oral dipyridamole, at concentrations previously shown to be neuroprotective, opens a further window on the many actions of this drug (1,2).
1. Diner HC. Dipyridamole
and acetylsalicylic acid in the secondary prevention of stroke. J Neurol Sci
2. Picano E, Abbracchio MP. European Stroke Prevention Study-2 results: serendipitous demonstration of neuroprotection induced by endogenous adenosine accumulation? TIPS
3. Fitzgerald GA. Dipyridamole
. N Engl J Med
4. Rajah SM, Penny AF, Crow MJ, Pepper MD, Watson DA. The interaction of varying doses of dipyridamole
and acetylsalcylic acid on the inhibition of platelet functions and their effect on bleeding time. Br J Clin Pharmacol
5. Picano E, Michelassi C. Chronic oral dipyridamole
as a "novel" antianginal drug: the collateral hypothesis. Cardiovasc Res
6. Neary JT, Rathbone MP, Cattabeni F, Abbracchio MP. Trophic actions of extracellular nucleotides and nucleosides on glial and neural cells. Trends Neurosci
7. Farinelli SE, Greene LA, Friedman WJ. Neuroprotective actions of dipyridamole
on cultured CNS neurons. J Neurosci
8. Siesjo BK, Agardh CD, Bengtsson F. Free radical and brain damage. Cerebrovasc Brain Metab Rev
9. Iuliano L, Pedersen J, Rotilio G, Ferro D, Violi F. A potent chain-breaking antioxidant
activity of the cardiovascular drug dipyridamole
. Free Radic Biol Med
10. Iuliano L, Colevita AR, Camastra C, et al. Protection of low density lipoprotein oxidation at chemical and cellular levels by antioxidant
. Br J Pharmacol
11. Pedulli GF, Lucarini M, Marchesi E, et al. Medium effects on the antioxidant
activity of dipyridamole
. Free Radic Biol Med
12. Bacon PJ, Love SA, Gupta AK, Kirkpatrick PJ, Meton DK. Plasma antioxidant
consumption associated with ischemia/reperfusion
during carotid endarterectomy. Stroke
13. North American Symptomatic Carotid Endarterectomy Trial. Methods, patients characteristics, and progress. Stroke
14. Fiori L, Parenti G, Marconi F. Combined transcranial Doppler and electrophysiologic monitoring for carotid endarterectomy. J Neurosurg Anesthesiol
15. Halsey JH, for the International Transcranial Doppler Collaborators. Risk and benefits of shunting in carotid endarterectomy. Stroke
16. Lang JK, Gohill R, Packer L. Simultaneous determination of tocopherols, ubiquinols and ubiquinones, in blood plasma, tissues and subcellular fractions. Anal Biochem
17. Thurnham DI, Davies JA, Crump BJ, Situnayake RD, Davis M. The use of different lipids to express serum tocopherol: lipid ratios for the measurement of vitamin E status. Ann Clin Biochem
18. Cesarone MR, Belcaro G, Carratelli M, et al. A simple test to monitor oxidative stress. Int Angiol
19. Belcaro G, Carratelli M, Cesarone MR, Cornelli U, Incandela L, Nicoladies A. Oxygen free radicals
: a method, a product, an application. SFRR Summer Meeting, Abano Terme, Italy, 1997:183-4.
20. Barsacchi R, Coassin M, Maiorino M, Pelosi G, Simonelli C, Ursini F. Increased ultra weak chemiluminescence emission from rat heart at postischemic reoxygenation: protective role of vitamin E. Free Radic Biol Med
21. Barsacchi R, Pelosi G, Maffei S, et al. Myocardial vitamin E is consumed during cardiopulmonary bypass: indirect evidence of free radical generation in human ischemic heart. Int J Cardiol
22. Sakamoto A, Ohnishi ST, Ohnishi T, Ogawa R. Protective effect of a new antioxidant
on the brain exposed to ischemia-reperfusion
injury: inhibition of free radical formation and lipid peroxidation. Free Radic Biol Med
23. Kinlay S, Fang JC, Hikita H, et al. Plasma α-tocopherol and coronary endothelium-dependent vasodilator function. Circulation
24. Kusmic C, Picano E, Busceti CL, Petersen C, Barsacchi R. The antioxidant
spares the vitamin E and thiols in red blood cells after oxidative stress. Cardiovasc Res
25. Halliwell B. Free radicals
and human disease: curiosity, cause or consequence? Lancet