Plasma Antioxidant Consumption Associated With Ischemia/Reperfusion During Carotid Endarterectomy
Background and Purpose Thirteen patients undergoing carotid endarterectomy were studied to correlate changes in jugular bulb venous oxygen saturation (Sjo2) with indices of free radical production during cerebral ischemia/reperfusion. Levels of oxidant species were also monitored in arterial samples to determine any change across the cerebral circulation.
Methods Blood was sampled from a venous catheter inserted in the ipsilateral jugular bulb and from an arterial catheter. Co-oximetry measurements were made to determine jugular bulb venous oxygenation saturation. To monitor changes in oxidant stress, a colorimetric assay was used to determine plasma antioxidant potential, and electron paramagnetic resonance spectroscopy was used to quantify free radical–spin trap adducts formed in a blood sample treated with the spin trap α-tert-butyl phenyl nitrone (PBN).
Results Sjo2 decreased significantly from 68±11% to 61±10% (P<.05) during clamping of the internal carotid artery and returned to baseline (65±11%) when the carotid clamp was removed. Jugular venous plasma antioxidant potential decreased significantly from 32.76±5.42% inhibition to 28.02±6.77% inhibition (P<.05). There was no concomitant change in arterial plasma antioxidant potential values, indicating a decrease in antioxidant capacity across the cerebral circulation. However, analysis of spin trap–free radical adducts did not provide conclusive evidence for free radical production.
Conclusions These results provide supportive evidence for oxidant production during cerebral ischemia/reperfusion in a clinical setting.
The brain is at special risk from oxidant-mediated injury because of its large iron stores,1 high levels of polyunsaturated lipids, and poor antioxidant defenses.2 Consequently, oxidant mechanisms have been implicated in cerebral ischemia/reperfusion injury,3 and such injury in animal models has been shown to be prevented or ameliorated by antioxidant therapy.4 5 6 7 8 However, the demonstration of oxidant injury in clinical cerebral ischemia has been more difficult, and there is little evidence to support a role for free radical–mediated injury in human disease.
Carotid endarterectomy represents a discrete episode of focal cerebral ischemia followed by reperfusion. This paradigm of cerebral ischemia/reperfusion injury has both advantages and disadvantages as a model of pathophysiology of ischemic neuronal injury. The major disadvantage, in experimental terms, is the lack of severity of the insult, since the short period of ischemia, presence of collateral circulation, and use of shunts during surgery all combine to reduce the extent of injury, and clinically significant neuronal injury is rare.9 However, major advantages of the model include its clinical substrate, controllability, and ability to estimate the severity of the ischemic insult with multimodality monitoring, based on transcranial Doppler, jugular venous saturation, and electrophysiological detection of the consequences of cerebral ischemia.10 Furthermore, the ability to sample arterial and jugular bulb venous blood provides us with a unique opportunity to study changes in levels of biologically significant molecules across the cerebral circulation.
We have used carotid endarterectomy as a model for studying oxidant production identified by PBN spin-trapping techniques11 and changes in plasma antioxidant levels12 during an episode of cerebral ischemia/reperfusion. Changes in these indices of oxidant production were correlated with the severity of the ischemic episode, which was measured by jugular bulb oximetry.
Subjects and Methods
This study was approved by the Addenbrooke's Hospital Ethics Committee. Thirteen patients undergoing unilateral carotid endarterectomy for occlusive vascular disease gave informed consent to be included in this study. They presented with a history of transient ischemic attacks, amaurosis fugax, or a single nondebilitating stroke. All were shown to have a significant focal stenosis at the bifurcation of the carotid artery as ascertained by conventional digital subtraction angiography. The median age of the patients was 64 years (range, 54 to 76 years), and the group included 11 men and two women.
Anesthesia and Surgery
Patients received 10 to 20 mg of temazepam as premedication. Anesthesia was induced with fentanyl 2 mg/kg and thiopental 3 to 4 mg/kg. Neuromuscular blockade was achieved by the administration of vecuronium 0.1 mg/kg, with additional supplements as required. Anesthesia was maintained with nitrous oxide and oxygen (Fio2, 0.35 to 0.5) supplemented with isoflurane and additional doses of fentanyl, with controlled ventilation to a Paco2 level of 4 to 4.5 kPa. Before surgery, patients were given 10 to 20 mg of temazepam. Routine intraoperative monitoring included direct arterial pressure monitoring, measurement of middle cerebral artery flow velocity by transcranial Doppler ultrasound (Neuroguard, Medasonics Inc), and processed electroencephalographic monitoring (CFM, Lectromed Ltd). After exposure of the carotid artery, a 5F fiberoptic catheter (Abbott) was inserted into the internal jugular vein under direct vision and advanced until its tip lay in the jugular bulb, above the confluence of the common facial vein. Sjo2 was continuously measured with an Abbott Oximetrix system, with discrete samples obtained for calibration and study purposes. Our criteria for shunt placement have been described previously.10 In any event, none of the subjects in this study required a shunt.
Blood was sampled from the venous catheter inserted into the ipsilateral jugular bulb for monitoring of Sjo2, PAP, and spin-trap (PBN) adduct concentrations (see below) in all 13 patients. In a subgroup of these patients (n=7), blood was simultaneously sampled from an arterial catheter to ensure that any changes in antioxidant capacity or free radical concentration were occurring across the cerebral circulation. Samples were taken at the following time points for Sjo2 (analyzed by CO oximetry): (1) immediately before clamping of the external carotid artery; (2) 5 minutes after clamping of the external carotid artery; (3) immediately before declamping of the external carotid artery; (4) immediately after declamping of the external and internal carotid arteries; and (5) 5 minutes after declamping of the external and internal carotid arteries.
At time points 1, 3, and 4, samples of blood were also taken for analysis of PAP and PBN adduct concentrations. For analysis of PAP, 2 mL of blood was sampled into a heparinized syringe. For analysis of spin trap–free radical adducts, an additional 4 mL of blood was sampled into heparinized syringes containing 2 mL PBN (80 mmol/L), giving a final blood PBN concentration of 26.7 mmol/L. All samples were immediately centrifuged at 8000 rpm for 5 minutes, and the plasma layer was aspirated and stored in liquid nitrogen until transfer to −70°C for long-term storage. All samples were analyzed within 1 month of collection.
Myoglobin type III and Sephadex G-15-120 (40 to 120 μm) were purchased from Sigma; ABTS, hydrogen peroxide, potassium ferricyanide, Trolox (Hoffmann–La Roche), and the spin trap PBN were purchased from Aldrich. All chemicals were the highest purity available.
This method12 is based on the interaction of the phenothiazine compound ABTS with ferryl myoglobin, produced through the reaction of metmyoglobin with hydrogen peroxide. This leads to the production of the ABTS•+ radical, which can be measured spectrophotometrically at 650, 734, and 820 nm. In the presence of antioxidants the absorbance of this radical cation is quenched to an extent directly related to the concentration of antioxidants present in the added fluid.
Briefly, ABTS (300 μL, 500 μmol/L), metmyoglobin (36 μL, 70 μmol/L), and 497 μL of buffer (of which 8.4 μL was replaced when a sample was being investigated) were mixed with 167 μL of H2O2 (450 μmol/L) to initiate the reaction. Six minutes after addition of the hydrogen peroxide, the absorbance of the solution was read at 734 nm, along with a buffer blank that did not contain the antioxidant solution or fluid to be tested. The extent to which the antioxidants in the sample solution inhibited the reaction was calibrated with the use of a standard antioxidant solution, Trolox (2.5 mmol/L). Samples were run in triplicate to ensure intra-assay precision.
PBN Adduct Analysis
Plasma/PBN samples were extracted into toluene (3 mL sample: 0.225 mL toluene) to concentrate lipophilic PBN spin adducts and to maximize the signal. The toluene/plasma/PBN sample was centrifuged at 8000 rpm for 5 minutes, and the toluene layer was aspirated. We analyzed this on a Bruker EMS 104 electron paramagnetic resonance spectrometer at 24°C using a quartz electron paramagnetic resonance tube, nonsaturating microwave power of 15.8 mW, and a magnetic modulation of 50 kHz. A PBN adduct was identified by the presence of a triplet spectrum. Since the spectrophotometer was set at maximal sensitivity rather than resolution, fine structure was not identified. PBN adduct intensities were calculated from the spectral peak height and expressed in arbitrary units.
Nonparametric statistics were used to compare changes in PAP and PBN adduct concentration during surgery (Wilcoxon rank sum test) with the use of Stat View 4 (Abacus).
Fig 1⇓ shows the change in Sjo2 during the surgical procedure. A significant decrease was seen during arterial clamping (time 2; P<.05). This continued to fall with time (time 3; P<.05) compared with the preclamp value. The saturation recovered to almost preclamp levels after removal of clamps.
Baseline jugular bulb venous PAP was 32.76±5.42; this decreased to 32.02±6.02 during ischemia and showed a further decrease to 28.02±6.77 on reperfusion (P<.01). Fig 2⇓ shows the decrease in jugular venous PAP seen in each patient individually throughout the procedure. In a subgroup of these patients (n=7), arterial samples were analyzed to ensure that the decrease in PAP was due to consumption across the cerebral circulation by oxidant production rather than a dilution effect. There was no concomitant reduction in arterial PAP during ischemia (30.26±9.93) and on reperfusion (29.10±10.5) compared with baseline (30.14±8.44) (Fig 3⇓). This resulted in a significant increase in the arteriojugular venous difference during reperfusion compared with baseline (P<.05) in all patients with the exception of one (patient 13) who had an unusually low baseline PAP (Fig 3⇓). All PAP values are expressed in terms of percent inhibition.
PBN Adduct Concentration
No statistically significant increase in concentration of PBN adducts was seen in either jugular venous or arterial blood samples (Fig 4⇓).
Free radical species generated during ischemia/reperfusion may attack cerebral tissue, leading to tissue damage if antioxidant defenses are overwhelmed.13 Direct evidence of oxidant-mediated tissue injury would be the ideal method by which to demonstrate this phenomenon. However, in this type of surgery, brain tissue is not available for analysis. An alternative measure of free radical production is to examine changes in PBN–free radical adduct concentration. Spin traps have been used in vivo to demonstrate free radical production during cerebral ischemia/reperfusion injury in animal studies.14 Ideally, spin traps such as PBN should be perfused into the systemic circulation, allowing direct trapping of the radical at the point of production,15 but PBN may be toxic at the concentrations needed. Spin traps have been shown to have an effect on the circulatory system, including altered contractile strength and heart rate, cardiac arrhythmia, and vasodilation.16 17 Hence, no spin trap has yet received approval for use in humans, resulting in ex vivo spin trapping techniques as the only viable option. MR spectroscopy and diffusion-weighted MRI18 19 may provide an early indication of changes in energy status and the development of cytotoxic edema after cerebral ischemia/reperfusion injury but cannot provide direct in vivo evidence of free radical generation. The use of proton electron double resonance imaging may be useful in principle, but current implementations do not possess the sensitivity needed to image biological concentrations of free radicals in vivo.20
In this study, no change in PBN adduct concentration was detected during ischemia or on reperfusion. This may be due either to experimental factors or to the possibility that no significant oxidant insult occurred. Experimental factors that may have hindered free radical detection include the fact that radical species are produced distant to the sampling point and, because of their high reactivity, may have reacted with blood, tissue, or the catheter before sampling. A second difficulty with ex vivo spin trapping is that any species detected may be generated as a result of interaction with the plastic of the catheter or syringe (possibly because of the products of gamma radiation used for sterilization) and hence does not provide an accurate measurement of the in vivo situation.21
An alternative method is to obtain evidence of free radical production from indirect markers of oxidative injury such as changes in PAP. Previously, we have shown that such changes may be seen in clinical situations in which oxidant injury has been implicated.22 23 The antioxidant levels across the cerebral circulation decreased during ischemia with a further significant decrease on reperfusion. This is consistent with the hypothesis that cellular changes during ischemia produced free radical species on reperfusion that consumed antioxidants in plasma. The marked changes during reperfusion rather than ischemia also suggest that the reduction in PAP was not merely due to a dilution effect caused by contamination from extracranial blood such as facial blood. The efficient consumption of free radicals by plasma antioxidants may explain both the decrease in total PAP seen and the lack of increase in PBN adduct concentration.
The difference seen in PAP occurred during a period of relative rather than complete ischemia. The reductions seen were also relatively small and hence may not have real clinical implications since these patients do not generally suffer from major stroke or other ischemic attack after surgery.9 However, the results do lend support to the theory that antioxidants are consumed during ischemia/reperfusion injury, presumably as a result of free radical generation.
We believe that these results have two important implications. First, they provide supportive evidence for oxidant production during cerebral ischemia/reperfusion in a clinical setting. Second, the clinical scenario of carotid endarterectomy may provide a useful paradigm for other clinical studies of cerebral ischemia/reperfusion.
Selected Abbreviations and Acronyms
|ABTS||=||2,2′-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) (diammonium salt)|
|PAP||=||plasma antioxidant potential|
|PBN||=||α-tert-butyl phenyl nitrone|
|Sjo2||=||jugular bulb venous oxygen saturation|
This study was supported in part by a generous grant from the Critical Care Trust, Leeds (Dr Menon).
- Received May 21, 1996.
- Revision received July 2, 1996.
- Accepted July 4, 1996.
- Copyright © 1996 by American Heart Association
Cohen G. Oxygen radicals and Parkinson's disease. In: Halliwell B, ed. Oxygen Radicals and Tissue Injury. Bethesda, Md: FASEB; 1988:130-135.
Abe K, Yuki S, Kogure K. Strong attenuation of ischemic and postischemic brain edema in rats by a novel radical scavenger. Stroke. 1988:19:480-485.
Roda JM, Carceller F, Pajares R, DiezTejedor E. Prevention of cerebral ischaemic reperfusion injury by intra-arterial administration of superoxide dismutase in the rat. Neurol Res. 1991:13;160-163.
Sakamoto A, Ohnishi ST, Ohnishi T, Ogawa R. Protective effect of a new antioxidant on the rat brain exposed to ischaemia-reperfusion injury: Inhibition of free radical formation and lipid peroxidation. Free Radic Biol Med. 1991:11;385-391.
Yu MJ, McCowan JR, Smalstig EB, Bennett DR, Roush ME, Clemens JA, Kontos HA. A phenothiazine derivative reduces rat brain damage after global or focal ischemia. Stroke. 1992:23;1287-1291.
Santiago LA, Osato JA, Ogawa N, Mori A. Antioxidant protection of bio-normalizer in cerebral ischaemia-reperfusion injury in the gerbil. Neuroreport. 1993:4;1031-1034.
Mason RP, Knecht KT. In vivo detection of radical adducts by electron spin resonance. In: Packer L, ed. Methods in Enzymology. New York, NY: Academic Press, Inc; 1994:112-117.
Watson BD, Ginsberg MD. Mechanisms of lipid peroxidation potentiated by ischaemia in brain. In: Halliwell B, eds. Oxygen Radicals and Tissue Injury. Bethesda, Md; FASEB; 1988:81-91.
Traystman RJ, Kirsch JR, Koehler RC. Oxygen radical mechanisms of brain injury following ischaemia and reperfusion. J Appl Physiol. 1991;71:1185-1195.
Bolli R, Patel BS, Jeroudi MO, Lai EK, McCay PB. Demonstration of free radical generation in ‘stunned’ myocardium of intact dogs with the use of the spin trap α-phenyl N-tert-butyl nitrone. J Clin Invest. 1988;82:476-485.
Graham GD, Kalvach P, Blamire AM, Brass LM, Fayad PB, Prichard JW. Clinical correlates of proton magnetic resonance spectroscopy findings after acute cerebral infarction. Stroke. 1995;26:225-229.
Lurie DJ. Progress toward whole-body proton-electron double-resonance imaging of free radicals. Magn Reson Materials Phys Biol Med.. 1994;2:267-271.
Bacon PJ, Pickworth AJ, Jones JG, Menon DK. Plasma antioxidant activity in smokers. Br J Anaesth.. 1994;73:721P-722P.