Share this article on:

Determinants of cerebral oximetry in patients undergoing off-pump coronary artery bypass grafting: an observational study

Moritz, Stefan; Rochon, Justine; Völkel, Sabine; Hilker, Michael; Hobbhahn, Jonny; Graf, Bernhard M; Arlt, Matthias

European Journal of Anaesthesiology (EJA): June 2010 - Volume 27 - Issue 6 - p 542–549
doi: 10.1097/EJA.0b013e3283343ef8

Background and objective The present study investigated whether positioning of the heart during off-pump coronary artery bypass grafting leads to changes in regional cerebral oxygen saturation measured by near infrared spectroscopy.

Methods In 35 consecutive patients undergoing elective off-pump coronary artery bypass grafting, cerebral oxygen saturation was measured by near infrared spectroscopy. A 20% decrease from baseline values was considered to be cerebral hypoperfusion. Mean arterial pressure, cardiac index, arterial carbon dioxide partial pressure, central venous pressure, haemoglobin concentration, and body temperature were determined simultaneously. Each value was recorded at the following time points: at baseline; after positioning the heart for distal anastomoses of the left anterior descending artery, the circumflex, and the right coronary artery; after repositioning the heart; and after chest closure. Significant changes in each parameter from baseline were detected using repeated-measures analysis of variance. Determinants of cerebral oxygen saturation during surgery were analysed by means of generalized estimation equations.

Results Cerebral oxygen saturation remained constant during grafting of the left anterior descending artery and decreased significantly during grafting of the circumflex (P = 0.001) and the right coronary artery (P < 0.001). Cerebral hypoperfusion was found in seven (20%) patients. Changes in haemoglobin concentration (P = 0.014), arterial CO2 partial pressure (P = 0.001), cardiac output (P < 0.001), and central venous pressure (P = 0.014) determined alterations in cerebral oxygen saturation.

Conclusion Positioning of the heart during off-pump coronary artery bypass grafting leads to a decrease in regional cerebral oxygen saturation. This decrease is associated with changes in cardiac output, haemoglobin concentration, arterial CO2 partial pressure, and central venous pressure.

From the Department of Anaesthesiology (SM, SV, JH, BMG, MA), Department of Cardiovascular Surgery (MH) and Center for Clinical Studies (JR), University Hospital Regensburg, Regensburg, Germany

Received 1 July, 2009

Revised 12 October, 2009

Accepted 13 October, 2009

Correspondence to Dr Stefan Moritz, Department of Anaesthesiology, University Hospital Regensburg, Franz-Josef-Strauss-Allee 11, 93053 Regensburg, Germany Tel: +49 941 944 7801; fax: +49 941 944 7802; e-mail:

The present study was presented in part at the Annual Meeting of the American Society of Anesthesiologists, San Francisco, California, 13–17 October 2007.

Back to Top | Article Outline


Coronary artery bypass grafting (CABG) has been shown to reduce cardiac morbidity and mortality. However, benefits are limited by the association of CABG with central nervous system complications such as stroke or postoperative cognitive dysfunction (POCD). The incidence of adverse cerebral outcomes (i.e. fatal or nonfatal stroke, coma, stupor, seizures, memory deficits, etc.) after CABG ranges from 1.25 to 6.1%.1–3 As cardiopulmonary bypass is suggested to play a major role in the development of perioperative neurological complications, many efforts have been made to minimize the deleterious side effects of cardiopulmonary bypass. In its extreme, surgical coronary revascularization is performed without cardiopulmonary bypass (off-pump). Although decreasing systemic inflammation4 and the embolic load,5 meta-analyses could not demonstrate improved neurological outcome following off-pump coronary artery bypass grafting (OPCABG).6,7 This led to the suggestion that lifting the heart during the OPCAB procedure can lead to haemodynamic impairment that may cause cerebral hypoperfusion.8

Cerebral oximetry is a noninvasive technique that uses near infrared spectroscopy to monitor regional cerebral oxygen saturation (rSO2). Although this monitoring has been demonstrated to be clinically useful during carotid surgery and on-pump cardiac surgery, data on its use during OPCAB are not available.

The primary goal of this prospective clinical trial was to investigate rSO2 in patients undergoing OPCABG. We hypothesized that the positioning of the heart for distal coronary anastomoses during OPCAB results in a decrease in rSO2. The second objective was to identify systemic determinants that are associated with changes in rSO2.

Back to Top | Article Outline

Patients and methods


After approval by the local ethics committee (Ethics Committee of the University of Regensburg, Medical Centre, Regensburg, Germany) and obtaining written informed consent, 40 consecutive patients undergoing elective OPCABG were enrolled within a 7-month period. Exclusion criteria were haemodynamic instability, persistent angina, recent myocardial infarction (<1 month), reoperation, a history of any neurological disorder (i.e. stroke, seizure, head injury, etc.), moderate-to-severe carotid artery stenosis, and psychiatric illness.

Back to Top | Article Outline

Anaesthesia and haemodynamic monitoring

Each patient received 20 mg dipotassium clorazepate the preceding evening and on the morning of the intervention. Except for angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers, antihypertensive drugs were continued on the day of surgery. Before inducing anaesthesia, we installed routine monitoring (five-lead electrocardiogram with continuous ST-segment analysis, peripheral oxygen saturation, and invasive arterial blood pressure monitoring). Anaesthesia was induced with 5 μg kg−1 fentanyl, 0.2 mg kg−1 etomidate, and 0.1 mg kg−1 pancuronium. Ventilation was adjusted to normocapnia (paCO2 37–43 mmHg) with an air–oxygen mixture (FiO2 = 0.6). Anaesthesia was maintained with 0.6 minimum alveolar concentration (MAC) sevoflurane. Bolus doses of fentanyl were applied as required. Mean arterial pressure (MAP) was kept at the patient's preoperative level by adjusting the volume balance or by continuous administration of norepinephrine, or both. For adjusting MAP, immediately after positioning the heart for distal anastomoses, bolus doses of norepinephrine were administered when required. Stable conditions for data collection (see below) were assumed when the MAP remained constant without administering bolus doses of norepinephrine. After inducing anaesthesia, we inserted central venous and pulmonary artery catheters. Cardiac output was measured by thermodilution, and the cardiac index (CI) was calculated. Zero reference levels for central venous pressure (CVP) and MAP monitoring were adjusted to the right atrium level. Arterial blood samples were analysed for arterial oxygen saturation, arterial partial pressure of carbon dioxide (paCO2), and haemoglobin (Hb) concentration. All parameters were determined simultaneously at the time points described below.

Each patient received tranexamic acid (10 mg kg−1 bolus and 2 mg kg−1 h−1). Heparin was administered at a dosage of 150 IE kg−1 to achieve an activated clotting time of more than 250 s. After completion of the final anastomosis, heparin was antagonized with protamine sulphate at a 1: 0.5 dosage followed by antagonization according to the activated clotting time.

Back to Top | Article Outline

Surgical procedure

All patients underwent standard median sternotomy. Deeply placed pericardial sutures were used to facilitate grafting of the posterolateral wall. We used a vacuum-assisted heart manipulator (Xpose; Guidant Corp., Cupertino, California, USA) and a mechanical stabilizer (Ultima; Guidant Corp.). The target vessel was exposed and ensnared with a soft silicon loop at the proximal site of the anastomoses. All anastomoses were performed with an intracoronary shunt (Axius coronary shunts; Guidant Corp.) to ensure distal coronary perfusion. When appropriate, the first anastomosis was the left internal mammary artery to the left anterior descending artery (LAD). Central aortovenous anastomoses were established with a clamp-free technique (Heart-string; Guidant Corp.). In case of bradycardia, atrial pacing was established.

All patients remained in the supine position during the whole procedure. Legs were moved upwards or downwards to modulate venous return. For grafting lateral and posterior walls, the table was moderately tilted towards the right side.

Back to Top | Article Outline

Cerebral oximetry

rSO2 was measured by using an INVOS 5100b cerebral oximeter (Somanetics, Troy, Michigan, USA). The light-emitting diode emits near infrared light of two different wavelengths to measure the ratio of oxyhaemoglobin to total Hb. The sensor is composed of one light-emitting diode and two detectors located 30 and 40 mm away from the diode, allowing the removal of the extracranial contribution of scattered light. Technical details of this device are described elsewhere.9,10 The probe was positioned over the right frontal area of the brain to monitor regional cerebral oxygenation in the watershed area of the middle and anterior cerebral artery. For each time point, the percentage change of rSO2 compared with baseline (i.e. the value after skin incision) was calculated. To avoid confusion between absolute and relative changes, which are both given as percentage values throughout the article, absolute changes are given in ‘units’ and relative changes as ‘%’. A decrease of more than 20% from baseline values was considered to be an indicator of hypoperfusion.11–14

Back to Top | Article Outline

Data collection

All data were simultaneously recorded at the following time points: after skin incision (baseline; n = 35); after obtaining stable conditions during the positioning of the heart for grafting the LAD (n = 32), the circumflex (n = 25), and the right coronary artery (RCA; n = 20); after repositioning the heart (n = 35); and after chest closure (n = 35). In patients receiving two grafts in nearby regions with the same positioning of the heart, only those values obtained during grafting of the first vessel were included in the analysis.

Back to Top | Article Outline


The primary outcome measure was rSO2. Sample size was estimated for a repeated-measures design with one within-subjects factor to detect a 10% decrease from baseline rSO2. Each subject was measured six times. A constant SD of 10 units over time and a first order autocorrelation (r = 0.5) were assumed. A total of 34 patients were required to achieve at least 90% power for detection of change in rSO2 over time using a Geisser–Greenhouse corrected F-test with a 5% significance level. Allowing for a 15% dropout, the sample size was adjusted to 40 patients.

Normality assumption was graphically checked using QQ plots for continuous variables. Continuous data were then summarized by using means with SD. Categorical data were expressed as frequency counts and percentages. Changes in rSO2, MAP, CVP, paCO2, Hb, and body temperature over time were evaluated with a one-way repeated-measures analysis of variance (ANOVA). Post-hoc comparisons were conducted using the Bonferroni adjustment for multiple comparisons.

Owing to the fact that not all patients required grafting of each coronary artery (LAD, circumflex, and RCA), there were missing values. As repeated-measures ANOVA requires complete data sets, these missing values were imputed using an expectation-maximization algorithm. Data imputation was done only for repeated-measures ANOVA.

Generalized estimation equations (GEEs), which take into account the dependence between repeated measures within patients, were used to investigate which of the systemic parameters are associated with changes in regional oxygen saturation.15,16 A population-averaged model with autoregressive correlation structure and the following covariates was fitted: MAP, CVP, CI, Hb concentration, arterial carbon dioxide partial pressure, and body temperature. In addition, linear and quadratic time effects were included as controlling variables, as data were expected to follow a quadratic time trend. Estimates (β) and corresponding 95% confidence intervals were calculated and considered statistically significant if the confidence interval excluded 0. The significance level was set to 0.05 (two-sided). However, because this was an exploratory analysis, P values are descriptive in nature. Statistical analyses were carried out with SPSS version 15.0 (SPSS Sciences, Chicago, Illinois, USA) and GEE analyses by using the procedure PROC GENMOD in SAS version 9.1 (SAS Institute, Cary, North Carolina, USA).

Back to Top | Article Outline


Two patients were excluded because it was not possible to correctly position the pulmonary artery catheter, another patient because of conversion to an on-pump procedure, and two patients because of incomplete data acquisition. Thus, final analysis was conducted in 35 patients. The demographic data of the patients included are shown in Table 1. One patient died 7 days after surgery because of therapy-resistant ventricular fibrillation.

Table 1

Table 1

Figure 1 shows the individual courses of cerebral oxygen saturation (Fig. 1a) and the corresponding mean values for each time point (Fig. 1b). Repeated-measures ANOVA yielded a significant within-subjects effect (P < 0.001) and showed that cerebral oxygen saturation remained constant during LAD anastomosis but was significantly reduced during circumflex and RCA anastomoses. After repositioning the heart, rSO2 increased again (P < 0.001) but remained below baseline values after chest closure (P = 0.001; Fig. 1b). A decrease of more than 20% from baseline was observed in seven patients (20%) at one or more time points.

Fig. 1

Fig. 1

Figure 2 shows the individual courses of haemodynamic parameters, Hb concentration, paCO2, and body temperature. The corresponding mean values are shown in Fig. 3. Significant within-subjects effects were detected for MAP (P = 0.017), CI (P < 0.001), CVP (P < 0.001), Hb concentration (P < 0.001), and body temperature (P < 0.001) but not for paCO2 (P = 0.134). Hb concentration and temperature decreased continuously. Decreases in CI and increases in CVP were observed during the positioning of the heart, especially during grafting the circumflex and the RCA (Fig. 3).

Fig. 2

Fig. 2

Fig. 3

Fig. 3

Generalized estimation equations showed CI (P < 0.001), Hb concentration (P = 0.014), paCO2 (P = 0.001), and CVP (P = 0.014) to significantly influence rSO2. In contrast, neither body temperature (P = 0.799) nor MAP (P = 0.223) had an effect on rSO2. The effects of CI, Hb concentration, and paCO2 were greater than zero, indicating decreased rSO2 with a decrease in these parameters after controlling for time and other covariates. The effect of CVP was negative; thus, rSO2 decreases with increasing CVP. The time effects showed a negative linear and a positive quadratic influence, but they did not reach statistical significance (P = 0.151 for linear time effect and P = 0.369 for quadratic time effect). Table 2 shows estimates, empirical standard errors, and the corresponding 95% confidence intervals from the fitted model.

Table 2

Table 2

Back to Top | Article Outline


OPCAB is proposed to improve neurological outcome in patients undergoing CABG. Clinical studies have shown that the avoidance of a cardiopulmonary bypass lowers the embolic load detected by transcranial Doppler sonography and weakens the systemic inflammatory response.4,5 Until now, however, meta-analyses were not able to demonstrate any improved neurological outcome following OPCAB.6,7 In addition, the final results of the Octopus trial showed a similar incidence of POCD in patients undergoing both on-pump and off-pump CABG. This raised the question of whether tilting the heart during off-pump procedure compromises cerebral perfusion because of haemodynamic impairment.

In the present study, the positioning of the heart for distal anastomoses at the lateral and posterior walls led to significant decreases in cerebral oxygen saturation. A decrease of more than 20% from baseline was detected in 20% of the patients investigated. Studies on cerebral blood flow (CBF) during OPCABG are rather limited and have partly yielded conflicting results. Talpahewa et al. 17 also used near infrared spectroscopy and reported that particularly the positioning of the heart for circumflex and RCA anastomoses provoked a decrease in cerebral oxygen saturation. This decrease was accompanied by a decrease in oxyhaemoglobin concentration and an increase in deoxyhaemoglobin concentration. In addition, these authors found an increase in cerebral blood volume during the positioning. Thus, Talpahewa et al. concluded that the increase in intracranial pressure and CVP caused by the Trendelenburg positioning and right lateral tilting of the table results in reduced CBF. Yet, these authors used a different technique (NIRO 300; Hamamatsu Photonics KK, Hamamatsu City, Japan) with only one detector, which does not allow the removal of extracranial contributions. Therefore, the influence of nonbrain sources (i.e. scalp tissue) on these findings remains unclear. Kim et al. 18 found significant changes in jugular bulb venous oxygen saturation during the positioning of the heart for distal anastomoses. However, the values remained within normal limits (i.e. >55%) during the entire procedure. In contrast, Diephuis et al. 19 described one or more intraoperative episodes of jugular bulb venous desaturation (i.e. ≤50%) in 50% of the patients investigated, but these episodes were independent of tilting of the heart. The different types of monitoring used might explain these differing results. Both jugular bulb venous monitoring and near infrared spectroscopy reflect the balance of cerebral oxygen supply and demand. However, jugular bulb venous monitoring reflects this ratio for the whole brain hemisphere, whereas near infrared spectroscopy monitors the watershed area of the middle and anterior cerebral artery, an area known for having the highest risk of cerebral hypoperfusion.20,21 In our previous studies on patients undergoing carotid endarterectomy, we correlated the findings of near infrared spectroscopy and jugular bulb venous monitoring to clinical signs of cerebral ischaemia. We could demonstrate that the interpretation of jugular bulb venous oxygen saturation can be misleading as it often increases up to supranormal values at the onset of cerebral ischaemia.22 In contrast, near infrared spectroscopy showed a continuous decrease and correlated well with neurological deterioration.9,11,12 However, carotid endarterectomy provokes regional changes in CBF, whereas OPCAB should lead to global changes in CBF. In addition, these studies were conducted on awake patients. Therefore, the transferability of these data to patients undergoing OPCAB remains uncertain.

As determinants of changes in regional cerebral oxygenation, we identified alterations in CI, CVP, Hb concentration, and paCO2. Whereas the Hb concentration showed a continuous decrease during the whole procedure, CVP and CI were especially altered during positioning of the heart for distal anastomoses at the lateral and posterior wall (Fig. 3). Changes in paCO2 occurred (Fig. 2), but could not be attributed to a specific time point (Fig. 3). Therefore, our data suggest that, during circumflex and RCA anastomoses, changes in CI and CVP predominantly provoke the decrease in rSO2, whereas the lower value after chest closure might be the result of the lowered Hb concentration. The influence of Hb concentration and paCO2 on cerebral oxygen supply is undisputed, but little attention is usually paid to the role of cardiac output and CVP in the regulation of CBF.21 Patients with a low cardiac output are generally considered to have normal CBF because of a redistribution of blood flow towards the heart and the brain and away from peripheral tissues.23

However, little is known about this mechanism in patients undergoing general anaesthesia. As general anaesthesia with volatile anaesthetics and opioids shows direct vasodilatory effects and reduces sympathoexcitation, an impairment of this redistribution seems likely in anaesthetized patients.24–27 Further studies are required to clarify this assumption.

In addition, a growing number of clinical studies suggest a significant influence of cardiac output on CBF, even in nonanaesthetized patients. In healthy volunteers, a linear relationship between changes in cardiac output and CBF has been shown that was independent of cerebral autoregulation.28 In healthy individuals, a β-blockade-induced reduction in cardiac output resulted in a diminished increase in CBF from rest to dynamic exercise despite an increase in MAP. Patients with chronic heart failure and atrial fibrillation exhibited an attenuated ability to elevate cerebral perfusion during exercise because of their impaired ability to increase cardiac output.29 In patients with congestive heart failure, CBF could be decreased by about 30% in response to the reduction in cardiac output.30 After heart transplantation, CBF was restored towards normal levels in these patients.31 Collectively, these findings and our data suggest that cardiac output might influence CBF independent of cerebral autoregulation.

CVP is also assumed to determine cerebral perfusion during OPCABG. Murkin8 suggested that the increase in CVP due to right ventricular inflow obstruction is most likely to determine changes in CBF. In the present study, changes in CVP did in fact influence cerebral oxygen saturation. The negative sign of β means that rSO2 decreases with increasing CVP values. However, not only CBF might determine rSO2. Near infrared spectroscopy measures oxygen saturation from arterial, venous, and capillary blood and presumes a fixed ratio of these three compartments. Any increase in cerebral venous blood volume (for instance, due to venous congestion) alters this ratio and leads to a decrease in rSO2, even when venous oxygen saturation remains constant. Therefore, further studies are needed to clarify whether the observed changes in rSO2 are due to changes in CBF or whether they are only the ‘technical’ consequences of an expanding venous compartment.

Some limitations of the present investigation need to be outlined. First, we used cerebral oximetry by near infrared spectroscopy to estimate changes in cerebral perfusion. Although increasingly used in cardiac surgery, this monitoring modality is still being discussed controversially. The regional nature, the high intersubject variability, and the unclear contribution of nonbrain sources represent only some of the limitations. A detailed summary of limitations has been recently published by Davies and Janelle.32

In OPCAB, especially venous congestion (e.g. due to a steep Trendelenburg position, torsion of the right ventricular inflow tract, etc.) can lead to alterations in the scalp thickness and therefore altering the transmission path length. Therefore, as already mentioned above, it remains unclear whether the observed changes in rSO2 do really reflect alterations in CBF.

Second, we calculated the relative changes in cerebral oxygenation compared with the values obtained after anaesthesia induction. This presumes that cerebral perfusion is well maintained after anaesthesia induction, which is not necessarily the case. In addition, the higher oxygen concentration (FiO2 = 0.6) after anaesthesia induction may lead to increased rSO2.33 This increase might have affected the incidence of cerebral hypoperfusion in our study, but the effect should be negligible for determining the systemic factors that influence rSO2. However, further studies should determine the baseline values both in awake patients and with room air.

Third, we only carried out right-sided monitoring, but tilting the patient to one side might lead to interhemispheric differences. This issue has to be investigated in further studies.

Fourth, we used a pulmonary artery catheter to determine CI. However, especially during tilting the heart, tricuspid regurgitation might affect the accuracy of thermodilution measurements. To confirm our results, we constructed an additional model and replaced the CI and Hb concentration by the mixed venous oxygen saturation (SvO2), as the latter should be independent of tricuspid regurgitation. Even SvO2 was correlated to rSO2 (P < 0.001; data not shown). However, to outline the different effects of CI and Hb, these parameters were included in our final model instead of SvO2.

Fifth, the primary endpoint of the present study was cerebral oxygen saturation and not clinical outcome. Therefore, we cannot finally conclude that cerebral oxygen desaturation during off-pump surgery increases the risk of poor neurological outcome. However, earlier studies demonstrated that a decrease of more than 20% of baseline value is associated with clinical signs of cerebral ischaemia and POCD.12–14

Finally, the present study was observational in nature and therefore further studies are required to clarify whether treatment of cerebral desaturations results in improved neurocognitive outcome. In patients undergoing on-pump cardiac surgery treatment of cerebral desaturations has already been shown to result in lower major organ morbidity and mortality, but data on the neurological outcome are also missing.34

In conclusion, the present investigation shows that, in patients undergoing OPCABG, tilting the heart for distal anastomoses at the lateral and posterior walls is associated with a decrease in rSO2. Changes in CI, CVP, Hb concentration, and paCO2, but not in MAP and body temperature, were identified to be associated with changes in rSO2. Further studies are required to determine whether the observed changes reflect changes in CBF and whether treatment or avoidance of these desaturations results in improved neurological outcome.

Back to Top | Article Outline


We thank the staff and the participants of our study. We offer special thanks to Ms Monika Schoell (Centre of Clinical Studies, University Hospital Regensburg, Regensburg, Germany) for the linguistic revision of our article.

The present work should be attributed to the department of Anaesthesiology, University of Regensburg, Germany.

The support was provided solely from institutional and/or departmental sources.

Back to Top | Article Outline


1 Cleveland JC Jr, Shroyer AL, Chen AY, et al. Off-pump coronary artery bypass grafting decreases risk-adjusted mortality and morbidity. Ann Thorac Surg 2001; 72:1282–1288, discussion 1288–1289.
2 Hogue CW Jr, Murphy SF, Schechtman KB, Davila-Roman VG. Risk factors for early or delayed stroke after cardiac surgery. Circulation 1999; 100:642–647.
3 Roach GW, Kanchuger M, Mangano CM, et al. Adverse cerebral outcomes after coronary bypass surgery. Multicenter Study of Perioperative Ischemia Research Group and the Ischemia Research and Education Foundation Investigators. N Engl J Med 1996; 335:1857–1863.
4 Raja SG, Berg GA. Impact of off-pump coronary artery bypass surgery on systemic inflammation: current best available evidence. J Card Surg 2007; 22:445–455.
5 Ascione R, Ghosh A, Reeves BC, et al. Retinal and cerebral microembolization during coronary artery bypass surgery: a randomized, controlled trial. Circulation 2005; 112:3833–3838.
6 Takagi H, Tanabashi T, Kawai N, Umemoto T. Off-pump surgery does not reduce stroke, compared with results of on-pump coronary artery bypass grafting: a meta-analysis of randomized clinical trials. J Thorac Cardiovasc Surg 2007; 134:1059–1060.
7 Takagi H, Tanabashi T, Kawai N, Umemoto T. Cognitive decline after off-pump versus on-pump coronary artery bypass graft surgery: meta-analysis of randomized controlled trials. J Thorac Cardiovasc Surg 2007; 134:512–513.
8 Murkin JM. Hemodynamic changes during cardiac manipulation in off-CPB surgery: relevance in brain perfusion. Heart Surg Forum 2002; 5:221–224.
9 Samra SK, Dorje P, Zelenock GB, Stanley JC. Cerebral oximetry in patients undergoing carotid endarterectomy under regional anesthesia. Stroke 1996; 27:49–55.
10 Williams IM, Vohra R, Farrell A, et al. Cerebral oxygen saturation, transcranial Doppler ultrasonography and stump pressure in carotid surgery. Br J Surg 1994; 81:960–964.
11 Moritz S, Kasprzak P, Arlt M, et al. Accuracy of cerebral monitoring in detecting cerebral ischemia during carotid endarterectomy: a comparison of transcranial Doppler sonography, near-infrared spectroscopy, stump pressure, and somatosensory evoked potentials. Anesthesiology 2007; 107:563–569.
12 Samra SK, Dy EA, Welch K, et al. Evaluation of a cerebral oximeter as a monitor of cerebral ischemia during carotid endarterectomy. Anesthesiology 2000; 93:964–970.
13 Monk T, Weldon C, Weldon JE, van der Aa BS. Cerebral oxygen desaturations are associated with postoperative cognitive dysfunction in elderly patients. Anesthesiology 2002; 96:A40.
14 Slater JP, Guarino T, Stack J, et al. Cerebral oxygen desaturation predicts cognitive decline and longer hospital stay after cardiac surgery. Ann Thorac Surg 2009; 87:36–44, discussion 44–45.
15 Liang KY, Zeger SL. Regression analysis for correlated data. Annu Rev Public Health 1993; 14:43–68.
16 Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986; 42:121–130.
17 Talpahewa SP, Ascione R, Angelini GD, Lovell AT. Cerebral cortical oxygenation changes during OPCAB surgery. Ann Thorac Surg 2003; 76:1516–1522, discussion 1522.
18 Kim JY, Kwak YL, Oh YJ, et al. Changes in jugular bulb oxygen saturation during off-pump coronary artery bypass graft surgery. Acta Anaesthesiol Scand 2005; 49:956–961.
19 Diephuis JC, Moons KG, Nierich AN, et al. Jugular bulb desaturation during coronary artery surgery: a comparison of off-pump and on-pump procedures. Br J Anaesth 2005; 94:715–720.
20 Bladin CF, Chambers BR. Clinical features, pathogenesis, and computed tomographic characteristics of internal watershed infarction. Stroke 1993; 24:1925–1932.
21 Suter OC, Sunthorn T, Kraftsik R, et al. Cerebral hypoperfusion generates cortical watershed microinfarcts in Alzheimer disease. Stroke 2002; 33:1986–1992.
22 Moritz S, Kasprzak P, Woertgen C, et al. The accuracy of jugular bulb venous monitoring in detecting cerebral ischemia in awake patients undergoing carotid endarterectomy. J Neurosurg Anesthesiol 2008; 20:8–14.
23 Saxena PR, Schoemaker RG. Organ blood flow protection in hypertension and congestive heart failure. Am J Med 1993; 94:4S–12S.
24 Akata T, Izumi K, Nakashima M. The action of sevoflurane on vascular smooth muscle of isolated mesenteric resistance arteries (part 2): mechanisms of endothelium-independent vasorelaxation. Anesthesiology 2000; 92:1441–1453.
25 Izumi K, Akata T, Takahashi S. The action of sevoflurane on vascular smooth muscle of isolated mesenteric resistance arteries (part 1): role of endothelium. Anesthesiology 2000; 92:1426–1440.
26 Katoh T, Kobayashi S, Suzuki A, et al. The effect of fentanyl on sevoflurane requirements for somatic and sympathetic responses to surgical incision. Anesthesiology 1999; 90:398–405.
27 Matta BF, Heath KJ, Tipping K, Summors AC. Direct cerebral vasodilatory effects of sevoflurane and isoflurane. Anesthesiology 1999; 91:677–680.
28 Ogoh S, Brothers RM, Barnes Q, et al. The effect of changes in cardiac output on middle cerebral artery mean blood velocity at rest and during exercise. J Physiol 2005; 569:697–704.
29 Ide K, Gullov AL, Pott F, et al. Middle cerebral artery blood velocity during exercise in patients with atrial fibrillation. Clin Physiol 1999; 19:284–289.
30 Choi BR, Kim JS, Yang YJ, et al. Factors associated with decreased cerebral blood flow in congestive heart failure secondary to idiopathic dilated cardiomyopathy. Am J Cardiol 2006; 97:1365–1369.
31 Gruhn N, Larsen FS, Boesgaard S, et al. Cerebral blood flow in patients with chronic heart failure before and after heart transplantation. Stroke 2001; 32:2530–2533.
32 Davies LK, Janelle GM. Con: all cardiac surgical patients should not have intraoperative cerebral oxygenation monitoring. J Cardiothorac Vasc Anesth 2006; 20:450–455.
33 Stoneham MD, Lodi O, de Beer TC, Sear JW. Increased oxygen administration improves cerebral oxygenation in patients undergoing awake carotid surgery. Anesth Analg 2008; 107:1670–1675.
34 Murkin JM, Adams SJ, Novick RJ, et al. Monitoring brain oxygen saturation during coronary bypass surgery: a randomized, prospective study. Anesth Analg 2007; 104:51–58.

intraoperative monitoring; off-pump coronary artery bypass; near-infrared spectroscopy

© 2010 European Society of Anaesthesiology