| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
We investigated the effects of isoflurane-induced burst suppression, monitored with electroencephalography (EEG), on cerebral blood flow velocity (CBFV), cerebral oxygen extraction (COE), and autoregulation in 16 patients undergoing cardiac surgery. The experimental procedure was performed during nonpulsatile cardiopulmonary bypass (CPB) with mild hypothermia (32°C) in fentanyl-anesthestized patients. Middle cerebral artery transcranial Doppler flow velocity, right jugular vein bulb oxygen saturation, and jugular venous pressure (JVP) were continuously measured. Autoregulation was tested during changes in mean arterial blood pressure (MAP) within a range of 4080 mm Hg, induced by sodium nitroprusside and phenylephrine before (control) and during additional isoflurane administration to an EEG burst-suppression level of 69/min. Isoflurane induced a 27% decrease in CBFV (P < 0.05) and a 13% decrease in COE (P < 0.05) compared with control. The slope of the positive relationship between CBFV and cerebral perfusion pressure (CPP = MAP - JVP) was steeper with isoflurane (P < 0.05) compared with control, as was the slope of the negative relationship between CPP and COE (P < 0.05). We conclude that burst-suppression doses of isoflurane decrease CBFV and impair autoregulation of cerebral blood flow during mildly hypothermic CPB. Furthermore, during isoflurane administration, blood flow was in excess relative to oxygen demand, indicating a loss of metabolic autoregulation of flow. IMPLICATIONS: The effects of isoflurane on cerebral blood flow velocity (CBFV) and oxygen extraction (COE) as a function of perfusion pressure were studied. When added to fentanyl anesthesia, isoflurane induced a 27% and 13% decrease in CBFV and COE, respectively. CBFV became more pressure-dependent with isoflurane indicating an impaired autoregulation.
Neurologic dysfunction is a frequent complication after cardiac surgery. A decrease in cognitive function may develop in the immediate postoperative period in 25%-75% of cardiac surgical patients (1). Central nervous system deficit after cardiopulmonary bypass (CPB) is believed to result from either regional or global hypoperfusion or secondary to embolism (2). There is a growing interest in various strategies for reduction of perioperative neurologic morbidity, partly prompted by the fact that advanced age and disease severity of patients undergoing cardiac surgery increase the risk of neurologic and cognitive dysfunction (3). Anesthetics have been proposed as cerebroprotective drugs during CPB, and large doses of thiopental may afford a direct cerebral protective effect after valvular surgery (4), although this finding could not be confirmed in patients undergoing coronary artery bypass surgery (5). Anesthetics were proposed as cerebral protectants because of their ability to reduce cerebral metabolism, but some studies have indicated that metabolic depression is not the only mechanism at work (6). It has thus been suggested that anesthetics such as thiopental may also reduce cerebral embolization during CPB, by reducing cerebral blood flow (CBF) or via mechanisms involving free radical scavenging or cerebral countersteal (2). Isoflurane is a commonly used anesthetic drug during CPB. Previous reports on the effects of isoflurane on cerebral hemodynamics and metabolism during CPB are few and to some extent controversial. Woodcock et al. (7) demonstrated that isoflurane induces an uncoupling of CBF from metabolism and thus disrupts metabolic autoregulation of CBF during hypothermic CPB, a finding that was not confirmed by a more recent report (8). Both of these studies were performed during hypothermic CPB. The aim of the present study was to evaluate the effects of isoflurane on pressure-flow autoregulation, as well as metabolic autoregulation of flow during nonpulsatile CPB and mild hypothermia. We therefore investigated the effects of cortical burst-suppression concentrations of isoflurane on middle cerebral artery transcranial Doppler (TCD) flow velocity (CBFV) and cerebral oxygen extraction (COE) during fentanyl anesthesia at various levels of cerebral perfusion pressure (CPP).
After approval from our local human ethics committee, and after obtaining informed consent, 16 patients undergoing elective aortocoronary bypass surgery, valvular surgery, or combined procedures were included in the study (Table 1). Patients who had diabetes mellitus, uncontrolled hypertension, or cerebrovascular disease were excluded.
The patients were premedicated with flunitrazepam 0.015 mg/kg orally and morphine 0.15 mg/kg and scopolamine 0.006 mg/kg IM. Before the induction of anesthesia, a catheter was placed in the radial or femoral artery for blood sampling and continuous monitoring of mean arterial blood pressure (MAP). Anesthesia was induced with fentanyl 10 µg/kg and thiopental sodium sufficient to abolish the eyelash reflex (23 mg/kg). Neuromuscular block was obtained with a bolus dose of pancuronium 0.1 mg/kg. In the prebypass period, anesthesia was maintained with 50% N2O in oxygen and fentanyl (2.55 µg/kg). During CPB intermittent doses of fentanyl (2.55 µg/kg) were given. After CPB, anesthesia was maintained with isoflurane at an inspiratory concentration of 1.5%2.5%.
Right middle cerebral artery (MCA) flow velocity was monitored beat-to-beat using a 2-MHz pulsed Doppler probe (Multidop X; DWL, Sipplingen, Germany). The right MCA was insonated by the transtemporal approach at a depth of approximately 50 mm using standard criteria (9), and the probe was secured with a headband. The time-mean flow velocity, considered the most physiologic measure of flow velocity, was continously recorded. During anesthesia, before surgery, a 4F, 3-wavelength oximetry catheter (Opticath©; Abbott Laboratories, North Chicago, IL) was placed in the right jugular bulb, using a retrograde internal jugular vein approach, for continuos monitoring of jugular venous pressure (JVP) and jugular bulb saturation (SjO2) by an oximeter (Oximetrix3, Abbott Laboratories). The correct position of the catheter tip was verified by fluoroscopy. The oximeter was calibrated in vitro before insertion. Accurate fiberoptic saturation values were verified by analysis of blood samples drawn from the catheter and measurement of oxygen saturation (ABL 510; Radiometer, Copenhagen, Denmark). The fiberoptic catheter was recalibrated in vivo when there was a discrepancy between fiberoptic and blood oxygen saturation The perfusion system consisted of a hollow fiber, membrane oxygenator, and a Sarns 9000 max pump (Sarns; Ann Arbor, MI). Nonpulsatile perfusion of 2.4 L · min-1 · m-2 was maintained during the experimental procedure. The pump was primed with acetated Ringers solution and mannitol, with the hematocrit maintained between 20%-25%. PaCO2, PaO2, and pH measurements were performed online. PaCO2 was maintained at 3540 mm Hg and was uncorrected for body temperature. PaO2 was maintained at 113150 mm Hg. The study was performed during CPB with stable hypothermia at a nasopharyngeal temperature of 32°C. After the achievement of stable CPB, sodium nitroprusside or phenylephrine was used to induce stepwise increases or decreases in MAP within the range of 8040 mm Hg (first part of the experimental procedure = control). MAP, JVP, CBFV, and SjO2 were continuously monitored. Stepwise changes in MAP induce a cerebral autoregulatory response, which develops during approximately 2030 s (11). Values of the different variables were therefore obtained after a 30-s stabilization period at each blood pressure level. Isoflurane was then administered directly to the CPB circuit by an overflow vaporizer. The initial concentration of isoflurane was 1.5%, which was increased in a stepwise manner, guided by the EEG response, to induce an EEG burst-suppression pattern of 69/min. During isoflurane administration, MAP was again changed in a stepwise manner, using sodium nitroprusside or phenylephrine, recording the same variables as described above. For each patient, the relationship between CPP (MAP - JVP) and CBFV (Fig. 1), as well as the relationship between CPP and COE (COE = (SaO2 - SjO2)/SaO2), before and during isoflurane administration was subjected to a linear regression analysis. The slope (regression coefficient) of the relationship between CPP and CBFV was defined as cerebral pressure-flow autoregulation. CBFV and COE values fitted to each regression line were obtained at CPP values of 30, 40, 50, 60, and 70 mm Hg. These obtained CBFV and COE values were than subsequently evaluated with a two-way analysis of variance for repeated measurements.
The effects of isoflurane on the mean level of CBFV and COE, compared with control, were tested by using a two-way analysis of variance for repeated measurements. The effects of isoflurane on the slopes (regression coefficients) of the relationship between CPP and CBFV, as well as CPP and COE, were compared to control using a paired Students ttest. A P value <0.05 was considered statistically significant. Data are presented as mean ± SEM.
Data on the effects of isoflurane on the relationship between CPP and CBFV and the relationship between CPP and COE are shown in Figures 2 and 3.
Isoflurane decreased mean CBFV from 39.9 ± 2.0 to 29.1 ± 1.6 cm/s (-27%, P < 0.05). The regression coefficients deviated from zero (0.0001<P < 0.05) in 13 and 16 of the 16 patients in the control situation and during isoflurane administration, respectively, indicating a CPP-dependent CBFV. The mean regression coefficient (slope) of the autoregulation curve was more positive with isoflurane compared with control, 0.25 ± 0.04 cm/s/mm Hg and 0.19 ± 0.04 cm/s/mm Hg, respectively (P < 0.05) Isoflurane decreased mean COE from 33.6 ± 1.0 to 29.3 ± 1.2 percent units (-13%, P < 0.05). The slope of the curve relating COE to CPP was more negative with isoflurane compared with control, -0.19 ± 0.02%/mm Hg and -0.14 ± 0.03%/mm Hg, respectively (P < 0.05).
The main findings of the present study were that burst-suppression doses of isoflurane decreased CBFV by 27% and decreased COE by 13% during mild hypothermic CPB. The decrease in COE indicates that CBF was in excess, relative to cerebral oxygen demand, suggesting that isoflurane has a direct intrinsic cerebral vasodilating effect in addition to its effect on cerebral metabolism, thereby disrupting metabolic autoregulation. Furthermore, the mean regression coefficient of the slope relating CBFV to CPP was positive in the control situation, indicating impaired cerebral autoregulation during opioid-based anesthesia and mildly hypothermic CPB. The slopes were even more pronounced when burst-suppression concentrations of isoflurane were added, suggesting that isoflurane might further impair cerebral pressure-flow autoregulation during CPB. From Figure 2 it can be seen that a certain decrease in CPP caused a 30% more pronounced absolute and a 70% more pronounced relative decrease in CBFV with isoflurane compared with opioid-based anesthesia, a finding we believe is clinically relevant.
There is only one previous study addressing the influence of volatile anesthetics on cerebral pressure-flow autoregulation, particularly during CPB. Aladj et al. (12) studied the effects of isoflurane (0.6%1.2%) on pressure-flow cerebral autoregulation during hypothermic (27°C) CPB using the 133Xenon clearance technique. When isoflurane was added to a basal anesthesia of fentanyl and midazolam, CBF decreased by 35%. However, autoregulation appeared preserved during isoflurane anesthesia as a phenylephrine-induced increase in MAP In initial studies, using static methodology (133Xenon clearance technique) it was shown that CBF was independent of MAP during hypothermic CPB (15). However, a more recent study of fentanyl/midazolam-anesthetized patients demonstrated that CBF is pressure-dependent, i.e., the relationship between MAP and CBF has a positive slope, particularly during normothermic CPB but also during hypothemic CPB (16). The presence of a pressure-dependent cerebral perfusion during moderately hypothermic CPB and fentanyl anesthesia was also demonstrated in the present as well as in our previous study (14), using the more dynamic TCD technique. Furthermore, in our studies we have evaluated the CPP-COE relationship, allowing us to establish the relationship between perfusion pressure and cerebral perfusion indirectly, but independently, from the TCD technique. Assuming that cerebral oxygen consumption is constant, a pressure-dependent decrease in CBF would be associated with an increase in COE. A significant negative relationship between COE and CPP was demonstrated both in the present and in our previous study (14), which further supports the finding that CBF is pressure-dependent during CPB. Previous studies on neurosurgical patients have shown that although isoflurane at inhaled concentrations of 0.85%2.3% reduces cerebral oxygen consumption, it maintains CBF (17,18), suggesting an isoflurane-induced disruption of metabolic autoregulation. Data on the effects of isoflurane on CBF and oxygen consumption during hypothermic (25°28°C) CPB are somewhat controversial. Woodcock et al. (7) showed that isoflurane at burst-suppression concentrations (1.1%), when compared with large-dose fentanyl, decreased cerebral metabolic rate with no significant effect on CBF. Newman et al. (8), however, could not demonstrate an isoflurane-induced impairment of metabolic autoregulation of CBF, as indicated by unchanged levels of COE when compared to fentanyl/midazolam. In the present study, isoflurane induced a significant decrease in COE, indicating that isoflurane causes an impairment of metabolic autoregulation of CBF during moderately hypothermic (32°C) CPB.
The use of TCD to measure blood flow velocity in intracranial arteries was first described in 1982 by Aaslid et al. (9). A direct validation of this technique requires correlation with an existing method. Bishop et al. (19) compared TCD with CBF measured by using the IV 133Xenon technique. They showed that the absolute values of CBFV and CBF correlated poorly between the two methods but that the CBFV and CBF response to hypercapnia showed a very good correlation (r = 0.85). These authors therefore concluded that changes in CBFV were reliably correlated with changes in CBF but that the absolute velocity should not be used as an indicator of CBF. Newell et al. (11) compared the relative changes in CBFV and internal carotid artery flow using an electromagnetic flowmeter on the ipsilateral internal carotid artery. A transient decrease in blood pressure induced a cerebral autoregulatory response in CBFV and internal carotid artery flow that were highly correlated (r = 0.99), which suggests that the diameter of the MCA and its flow velocity profile were stable. The validity of TCD has also been evaluated during CPB. Trivedi et al. (20) compared the changes in CBF, using Xenon133 clearance, with changes in CBFV by TCD using both pH-stat and A major limitation of the present study is that we were not able to obtain a cerebral autoregulation response curve after the discontinuation of isoflurane, i.e., we had no postdrug control measurements. This was a result of time constraints because rewarming was started soon after the end of the experimental procedure in all patients. Therefore we cannot completely eliminate the possibility of time-dependent effects on CBFV and COE. Furthermore, in the present study, MAP was varied by infusions of phenylephrine or sodium nitroprusside, which could potentially affect CBF. However, neither of these drugs have any direct effects on CBF during CPB in humans (21,22).
Supported, in part, by a grant from the Swedish Medical Research Council (No. 13156) and the Medical Faculty of Gothenburg (LUA).
Presented, in part, at the 17th Annual Meeting of the European Association of Cardiothoracic Anaesthesiologists, Dublin, Ireland, June 1215, 2002.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|