- Email: [email protected]
Impact of Norepinephrine on Regional Cerebral Oxygenation During Cardiopulmonary Bypass
Impact of Norepinephrine on Regional Cerebral Oxygenation During Cardiopulmonary Bypass Ove Andreas Hagen, MD,* Lars Øivind Høiseth, MD, PhD,*† André Roslin, MD,* Svein Aslak Landsverk, MD, PhD,* Per Reidar Woldbaek, MD, PhD,‡ Are Hugo Pripp, PhD,§ Rolf Hanoa, MD, PhD,ǁ¶ and Knut Arvid Kirkebøen, MD, PhD*† Objectives: Norepinephrine is used to increase mean arterial pressure during cardiopulmonary bypass. However, it has been suggested that norepinephrine could constrict cerebral arteries, reducing cerebral blood flow. The aim of this study, therefore, was to explore whether there was an association between doses of norepinephrine to maintain mean arterial pressure at E80 mmHg during cardiopulmonary bypass and cerebral oxygen saturation measured using near-infrared spectroscopy. Design: Observational study. Setting: University hospital. Participants: Patients undergoing cardiac surgery (n ¼ 45) using cardiopulmonary bypass. Interventions: Norepinephrine was administered to maintain mean arterial pressure E80 mmHg during cardiopulmonary bypass. Measurements and Main Results: From initiation of cardiopulmonary bypass to removal of the aortic cross-clamp, norepinephrine dose, mean arterial pressure, partial pressure of arterial carbon dioxide, partial pressure of arterial
C
EREBRAL OXIMETERY is being used increasingly in cardiac surgery to monitor cerebral oxygenation. Reduced cerebral tissue oxygen saturation (rSO2) during cardiac surgery has been associated with adverse outcomes,1-3 and interventions based on monitoring of rSO2 during coronary artery bypass grafting (CABG) have been shown to reduce major organ dysfunction.4 Near-infrared spectroscopy (NIRS) is used to monitor rSO2, and details about NIRS have been provided by Scheeren et al.5 There is insufficient evidence to recommend a specific mean arterial pressure (MAP) during cardiopulmonary bypass (CPB), but it has been recommended to maintain MAP 470 mmHg during CPB in high-risk patients.6 Hypoperfusion resulting from impairment of cerebral autoregulation, hypotension, or anemia may lead to cerebral hypoxia. Autoregulation of cerebral blood flow (CBF) ensures delivery of oxygenated blood to the brain and protects the brain from ischemia caused by arterial pressure fluctuations. One approach that aims to improve cerebral oxygenation during CPB is to increase cerebral perfusion pressure by increasing systemic MAP using a vasopressor7 (eg, norepinephrine [NE]). According to departmental practice, the authors routinely administer NE during CPB, aiming for a MAP of 60 to 80 mmHg. However, the question has been raised whether vasopressors may induce cerebral vasoconstriction and thereby actually contribute to reduced cerebral oxygenation.8-10 NE has been shown to reduce rSO2 in healthy volunteers,8 but potential effects of NE on rSO2 during CPB have not been defined. The aim of this study, therefore, was to explore whether NE administered during CPB to maintain MAP at E 80 mmHg was associated with changes in rSO2. The hypothesis was that there would be no such association when controlling for other factors with biologically plausible effects on rSO2.
oxygen, hemoglobin, and pump flow values were averaged over 1 minute, giving a total of 3,460 data points entered as covariates in a linear mixed model for repeated measurements, with cerebral oxygen saturation measured using near-infrared spectroscopy as outcome. There was no statistically significant association between norepinephrine dose to maintain mean arterial pressure and cerebral oxygen saturation (p ¼ 0.46) in this model. Conclusions: Administration of norepinephrine to maintain mean arterial pressure E80 mmHg during cardiopulmonary bypass was not associated with statistically significant changes in cerebral oxygen saturation. These results indicated that norepinephrine could be used to increase mean arterial pressure during cardiopulmonary bypass without reducing cerebral oxygen saturation. & 2016 Elsevier Inc. All rights reserved. KEY WORDS: oximetry, norepinephrine, cardiopulmonary bypass
METHODS
Patients Forty-five patients presenting for elective aortic valve replacement, CABG, or a combination of aortic valve replacement and CABG were included in the study. Exclusion criteria were age o18 or 480 years, patient included in pharmacologic study, vasoactive medication started before measuring baseline rSO2, or any of the following medical conditions known at time of inclusion: frontal lobe pathology, intracranial vascular anomaly, neurologic disease, dementia, history of traumatic brain injury, cerebral insult, transient ischemic attack, or carotid artery stenosis (no study-specific examinations were performed). The study was approved by the Regional Committee for Medical and Health Research Ethics (REK sør-øst, 2010/823), and written informed consent was obtained. The study was registered in ClinicalTrials.gov (NCT01940874).
From the *Department of Anesthesiology, Oslo University Hospital, Oslo, Norway; †Faculty of Medicine, University of Oslo, Oslo, Norway; ‡Department of Cardiothoracic Surgery; §Oslo Centre of Biostatistics and Epidemiology, Research Support Services, and ǁDepartment of Neurosurgery, Oslo University Hospital, Oslo, Norway; and ¶University of Bergen, Bergen, Norway. Results from this study were presented, in part, at the Society of Cardiovascular Anesthesiologists annual meeting in Miami, FL, May 2013. Address reprint requests to Ove Andreas Hagen, MD, Department of Anesthesiology, Oslo Univeristy Hospital, Kirkeveien 166, 0450, Oslo, Norway. E-mail: [email protected] © 2016 Elsevier Inc. All rights reserved. 1053-0770/2601-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2015.12.015
Journal of Cardiothoracic and Vascular Anesthesia, Vol 30, No 2 (April), 2016: pp 291–296
291
292
Anesthetic Procedure Patients were premedicated with intramuscular morphine/ scopolamine. A 20-G arterial cannula was placed in the left radial artery and a 7-Fr triple-lumen catheter in the right internal jugular vein. Baseline rSO2 was measured when patients were breathing room air. Anesthesia was induced with intravenous diazepam, fentanyl, and propofol, and endotracheal intubation was facilitated with intravenous cisatracurium. Patients were mechanically ventilated, and anesthesia was maintained with sevoflurane. Temperature was measured in the urinary bladder. Although core temperature affects cerebral oxygen consumption, temperature measurements were not used in the statistical model due to the poor agreement between bladder and core temperatures. Surgical Procedure and Cardiopulmonary Bypass After cannulation of the aorta and right atrium, CPB primed with Ringer’s acetate was performed in a nonpulsatile manner. CPB flow was set to 2.4 (height [cm] þ weight [kg] – 60)/ 100 L/min, with partial pressure of arterial oxygen (PaO2) 15 to 20 kPa and partial pressure of arterial carbon dioxide (PaCO2) 5.2 to 5.7 kPa. Arterial blood gas analyses (α-stat) were performed intermittently. Temperature was measured in the urinary bladder. CABG procedures were performed with the patient in normothermia (35.0-37.01C), whereas other or combined procedures were performed with the patient in mild hypothermia (32.0-35.01C) or hypothermia (o32.01C). Heparin was administered to maintain activated clotting time 4480 seconds. Packed red blood cells were given routinely if hemoglobin (Hb) was o7 g/dL. Anesthesia was maintained with sevoflurane, 1.5% to 2% adminstered in the CPB circuit. NE (Alaris CC Plus with Guardrails; CareFusion, San Diego, CA) diluted in glucose, 50 mg/mL to 20 mg/mL, was infused in the central venous catheter using a syringe pump (Alaris GH Plus with Guardrails; CareFusion, San Diego, CA). During CPB, NE was administered with the aim of achieving MAP E80 mmHg. If rSO2 was reduced by 420% relative to baseline or to o50% absolute, the following steps were taken to improve rSO2: verification of CPB integrity and function/positioning of the cannulae; and controlling head position, PaCO2, O2, Hb, and CPB flow.11 If the CPB circuit, position of the cannulae, and PaO2 were adequate, PaCO2, Hb, and/or CPB flow were increased. Unless intentionally altered as previously stated, PaO2 and PaCO2 were kept constant by continuous arterial inline monitoring (BMU 40; Maquet, Rastatt, Germany). Signal Acquisition and Analysis Sensors (Adult SomaSensor; Medtronic, Minneapolis, MN) for the oximeter (INVOS 5100C Cerebral/Somatic Oximeter; Medtronic) were placed on the left and right forehead. Values were updated every 7 to 8 seconds and sampled with hemodynamic variables from the patient monitor (Solar 9500; GE Healthcare, Milwaukee, WI) at 1 Hz in a custom-made program in LabVIEW (National Instruments, Austin, TX). Dosage of NE was entered manually into this program. All rSO2 values were relative to the preoperative baseline value,
HAGEN ET AL
and the average of left and right measurements were used for analysis. Data were analyzed from initiation of CPB to removal of the aortic cross-clamp. Offline, data were down-sampled to 1 value every minute by averaging (arithmetic means) after manually removing obviously erroneous data (eg, from flushing the arterial line), providing 3,460 observations. Results from arterial blood gas analyses were entered manually. To obtain continuous values for Hb, PaO2, and PaCO2, these values were interpolated linearly between the blood gas analyses performed. When packed red blood cells were transfused, the Hb value was changed at the time of transfusion (not linearly interpolated). In a subset of 12 patients, arterial blood gas analyses were performed every 10 minutes to increase the resolution of these variables. If CPB flow was altered, flow relative to baseline was entered (eg, a 5% increase giving a value of 1.05). If CPB flow was not altered, it always had a value of 1. To explore autoregulation, as previously reported,12 rSO2 and MAP were averaged over 10 seconds from the raw data, giving a sampling frequency of 0.1 Hz. Pearson correlation coefficients between rSO2 and MAP were calculated from 5 minutes (30 observations) in a moving, overlapping fashion, making each data point contribute to 30 correlation coefficients (except at the ends of the recording). The correlations were binned according to the average MAP in the 5-minute period of the correlation coefficient with bin size 10 mmHg. Lower and upper limits of autoregulation (LLA and ULA, respectively) were defined if there was a transition of bin mean correlation from o0.3 to 40.3 with a reduction or increase in MAP-bin, respectively.12,13 For each bin, MAP was presented by its mean value (eg, 65 mmHg for the 60-70 mmHg-bin). Statistics Values are mean (standard deviation) or median (25th-75th percentiles) unless otherwise stated. The relation between NE dosage and rSO2 was evaluated in a linear mixed regression model with rSO2 as the oucome. The NE dose and other variables considered to have an effect on rSO2 were entered as explanatory variables. These other variables were MAP, PaCO2, PaO2, Hb, and CPB flow. Because the study only sought to describe the possible effect of NE dose on rSO2, the other varaibles (which were entered only to correct for their possible confounding effects) were included in the statistical model regardless of their p values. The model assumed a linear relation with the same slope between all explanatory variables and the outcome, rSO2. The mixed model accounts for dependence of data within each patient (patient being a random effect), and in this model, with random intercept, allowed the regression line of each patient to have different intercepts. Model fit was assessed using the Akaike information criterion (AIC).14 Due to the repeated nature of the data, the AIC was minimized by imposing an autoregressive-1 covariance structure to the residuals. RESULTS
Forty-eight patients were included in the study. Two patients had incomplete data, and 1 underwent surgery in deep hypothermia. Thus, data from 45 patients were analyzed.
293
NOREPINEPHRINE AND REGIONAL CEREBRAL OXYGENATION
Table 1. Patient Characteristics (n ¼ 45) Male, n (%) Age, yr Height, cm Weight, kg Surgical procedures CABG, n (%) Aortic valve replacement Aortic stenosis Aortic insufficiency CABG and aortic valve replacement (aortic stenosis) combined
35 64 176 86 15 20 15 5 10
(78) (33-78) (8.4) (14) (33) (44) (33) (11) (22)
NOTE. Data are mean (standard deviation) or median (25th-75th) percentiles after Kolmogorov-Smirnov test for normality, unless otherwise stated. Abbreviation: CABG, coronary artery bypass grafting.
Surgery was performed using hypothermia in 2 patients, mild hypothermia in 22 patients, and normothermia in 21 patients. All CABG patients underwent surgery in normothermia.15 Ten patients had an absolute rSO2 o50% at any time during CPB. Twenty-one patients had a decrease in rSO2 420% relative to the baseline value at any time during CPB; 8 of these patients had decreases in rSO2 420% relative to the baseline value lasting 420 consecutive minutes. Patient characteristics are presented in Table 1. rSO2 and the explanatory variables for all 3,460 observations (1-minute averages) are presented in Table 2. Scatter plots of rSO2 and MAP versus NE dose and rSO2 versus MAP are presented in Fig 1. When modeling the data, random intercept, random slope, and random intercept-random slope models were attempted, but the best fit (as judged using AIC) was found with a repeated structure with correlated residuals, first-order autoregressive covariance matrix. In the multivariate analysis, there was no significant association between NE dose and rSO2 relative to baseline. There were significant associations between rSO2 and MAP, PaO2, Hb, and CPB flow (see Table 2). In the subset of 12 patients with frequent (every 10 minutes) blood gas analyses, there was no association between rSO2 and NE dose (p ¼ 1.0). By the method of correlation between MAP and rSO2, the authors identified LLA in 26 patients and ULA in 15 patients. Medians (25th-75th percentiles) were LLA 65 (55-65) mmHg and ULA 75 (75-85) mmHg, respectively. All patients were discharged from the cardiothoracic intensive care unit the day after surgery, and there were no perioperative deaths. None of the patients had clinically recognized neurologic complications at discharge. DISCUSSION
The main finding in this study was that there was no significant association between NE dose administered during CPB to maintain MAP at E80 mmHg and rSO2. The lower confidence interval (CI) boundary corresponded to a o0.2% reduction in rSO2 with each 0.1 mg/kg/min NE-increment, which the authors did not believe to be relevant clinically. This study, therefore, indicated that NE could be used to increase
MAP during CPB without significantly affecting rSO2. A direct effect of sympathomimetic agents on cerebral resistance vessels is believed to be practically negligible because vasoactive amines do not cross the blood–brain barrier.16 However, whether sympathetic nerve activity originating from the superior cervical ganglion increases when MAP is augmented pharmacologically has been discussed.17 Because the cerebral vasculature is innervated predominantly from the superior cervical ganglion,18 it seems relevant to investigate and define the effects of sympathomimetic agents on cerebral oxygenation. NE administered in an intact circulation may cause a variety of cardiovascular effects, not only peripheral vasoconstriction. CPB as a model provides the opportunity to study the effect of NE with a fixed “cardiac” output. Sympathomimetic agents are used routinely to increase MAP during surgery. Using NIRS technology, Meng et al19 demonstrated decreased cerebral oxygenation during anesthesia when administering boluses of phenylephrine and unaltered cerebral oxygenation when administering boluses of ephedrine. Brassard et al8 found reduced cerebral oxygenation with increasing doses of NE in healthy volunteers; however, in the study presented here, NE was administered to maintain a fixed MAP, whereas MAP and systemic vascular resistance substantially increased in Brassard et al’s study.8 In a later study, Brassard et al9 found that administration of NE during CPB was associated with reduced rSO2 in diabetic patients but not in nondiabetic patients. Only 4 of the patients in this study were diabetic. In this subgroup, there was a significant relationship between rSO2 and NE dose (p ¼ 0.043). However, this post-hoc analysis on a very low number of patients should be interpreted with caution, and the CIs
Table 2. Summary of Covariates and Outcome and Regression Coefficients Summary Statistics (n ¼ 3,460)
Regression Coefficients
Median (25th-75th Percentiles)
rSO2, % from baseline NE, mg/kg/min MAP, mmHg PaCO2, kPa PaO2, kPa Hb, g/dL CPB flow Intercept
Coefficient (95% CI)
p Value
87 (82-92) 0.15 74 5.4 21.0 8.9 1.0
(0.09-0.25) (64-80) (4.9-5.8) (19.0-23.7) (8.3-10.0) (1.0-1.0)
–0.50 0.76 1.0 0.36 2.11 14 38
(–1.9 to 0.84) 0.46 (0.68-0.83) o0.001 (–0.33 to 2.3) 0.14 (0.072-0.66) 0.014 (1.4-2.8) o0.001 (5.2-23) 0.002 (23-54) 0.424
NOTE. Regional cerebral tissue oxygen saturation (outcome) and variables entered as covariates in the regression model. Summary statistics in the left column. Regression coefficients (95% confidence interval) for variables used as covariates in the middle column with p values (right column). The regression coefficients are the estimated changes in regional cerebral tissue oxygen saturation (%) for a 1-unit increase of each predictor. Note that cardiopulmonary bypass flow is given as a value relative to the initial setting given by the equation presented in the text (ie, a 10% increase would give the value 1.1). Abbreviations: rSO2, regional cerebral tissue oxygen saturation relative to baseline; NE, norepinephrine; MAP, mean arterial pressure; Hb, hemoglobin; CPB, cardiopulmonary bypass.
294
HAGEN ET AL
Fig 1. Scatter plots of norepinephrine and mean arterial pressure versus regional cerebral oxygen saturation and norepinephrine versus mean arterial pressure. Norepinephrine for all observations (1 minute average) in all patients. Abbreviation: rSO2, regional cerebral oxygen saturation.
corresponded with a reduction in rSO2 of 1.9% to 0.03% for a 0.1-mg/kg/min increment in NE dose. The authors of this study aimed for a higher MAP (80 mmHg) than that in the study by Brassard et al9 (460 mmHg). The authors of this study found a modest association between rSO2 and MAP, corresponding to an absolute increase in rSO2 of 7.6% (95% CI 6.8-8.3, p o 0.001) for each 10mmHg MAP increment. This assumed a linear relationship between MAP and CBF. Classically, CBF has been believed to be autoregulated, maintained unaltered when MAP is kept between LLA and ULA. Autoregulation implies that changes in MAP are not associated with changes in CBF. Thus, correlation between the 2 variables are expected to be close to 0. Outside the autoregulatory plateau, CBF is believed to be pressuredependent, reflected by positive correlations between MAP and CBF as MAP varies. This is the basic principle behind a series of studies on autoregulation during CPB in which correlations between MAP and rSO2 over minutes have been used to detect the limits of autoregulation. Using this method, Brady et al found that measurements using transcranial Doppler and cerebral oximetry agreed well.13 The mean nadir temperature in that study was 31.11C, but the same group of authors reported possible impairment of autoregulation in the rewarming phase.20 Autoregulation seemed to be better preserved by using an α-stat management (as in this study) compared with pH-stat,21 but may be impaired by sevoflurane,22 which was used in this study. Paradoxic changes in rSO2 (ie, a reduced rSO2 as MAP was increased and vice versa) with sodium nitroprusside and phenylephrine have been reported in patients with intact autoregulation.23 It should be noted that in the Moerman et al study, MAP was altered intentionally; whereas in this study, a vasocontrictor was given with the aim of maintaining a near-constant MAP. According to the method of correlations,12 the authors ideintified LLA and ULA in this study in 26 and 15 patients, respectively. However, there is a possible caveat when using this method according to the authors’ data. Because the authors aimed to maintain MAP at E80 mmHg, deviations in MAP from this target value will tend to produce both a change in mean MAP from this target value and an increased range of MAP values. Even
under an assumption of no autoregulatory plateau, this phenomenon could tend to increase correlation coefficients in MAP-bins above and below this target value, based on the dependence of the Pearson correlation between 2 variables on the range of values studied.24 Furthermore, each observation was used in 30 correlation coefficients, creating dependence between these values. A correlation coefficient of 0.3 was chosen somewhat arbitrarily as a threshold, as different values have been used. However, this was in accordance with more recent works.23 The overall data did not reveal an association between NE dose and rSO2. Fig 2 illustrates a case in which rSO2 was low despite recommended interventions. After increasing MAP with NE, the anesthesiologist abruptly reduced the NE dose. The data suggested that the increased NE dose was associated with reduced rSO2 despite an increased MAP because the pattern was repeated with a subsequent increase and decrease of NE. Although the authors of the study presented here did not find an overall association between NE dose and rSO2, this may have been the case in individual patients. The observation presented in Fig 2 may be due to chance alone or an unknown confounding factor. The authors of the study presented here did not have data to suggest characteristics of patients with such a response, if present. The NIRS device subtracts a shallow signal from the deeper signal, aiming for a spatial resolution allowing for measurement of deep (brain) tissue only. However, there is a concern regarding contamination from superficial tissues.25-27 Both the authors’ overall findings and the example in Fig 2 thus may be subject to contamination from superficial tissues. Regarding the overall results, it would be expected that, given the near-fixed “cardiac” output during CPB, a change in skin circulation with an increased NE dose would lead to decreased oxygen saturation in superficial tissues. Thus, the fact that the authors did not find reduced rSO2 with increased NE doses indicated that brain oxygen saturation was not reduced, even if contamination from superficial tissues occured. METHODOLOGIC CONSIDERATIONS
The authors did not perform a formal power analysis for sample size determination beause they did not find adequate
295
NOREPINEPHRINE AND REGIONAL CEREBRAL OXYGENATION
Fig 2. Data from a patient in whom regional cerebral oxygen saturation (left and right, grey lines) was reduced considerably with initiation of cardiopulmonary bypass without satisfactory effect of recommended measures. As norepinephrine (dashed line) was reduced, mean arterial pressure (solid line) decreased, whereas regional cerebral oxygen saturation increased. The pattern was repeated when a re-escalation of norepinephrine was attempted, and mean arterial pressure was allowed to drift to 40 to 50 mmHg. Time from initiation of CPB on x-axis. Left yaxis is mean arterial pressure (mmHg) and regional cerebral oxygen saturation relative to baseline (%). Right y-axis is norepinephrine dose.
published data in the planning phase. The authors planned to include 50 patients. However, 2 patients with exclusion criteria had been included (age 480 years), 2 had incomplete data, and 1 underwent surgery in deep hypothermia, leaving 45 patients for analysis. The authors believe that the CI of the outcome indicated that the sample size was sufficiently large. Temperature was measured in the bladder, which may lag considerably behind core temperature during temperature changes. Therefore, although hypothermia reduces cerebral metabolism, temperature was not entered into the statistical model presented. However, entering temperature as measured in the bladder into the model did not significantly change the overall results (data not shown). Baseline measurements were performed with patients breathing room air, giving the possibility of low baselines, especially considering the opioid-containing premedication. Measuring baseline rSO2 with supplemental oxygen has been recommended.11 However, normobaric hyperoxia has been associated with increased rSO2,28 which could lead to false high baselines. Ideally, baseline rSO2 perhaps should be measured after titrating supplemental oxygen to a normal PaO2, but this would be unacceptably cumbersome in clinical practice. Furthermore, baseline rSO2 has been associated with skin pigmentation.29 All the patients in this study were white.
Systematic neuropsychologic testing was not performed; therefore, the authors cannot exclude possible subtle postoperative neurologic disturbances. However, no neurologic complications were detected on routine clinical examination. Moderate cerebral arteriosclerosis probably affected some of the elderly patients. The authors aimed for a fixed MAP during CPB in this study. Therefore, whether increased MAP produces a better cerebral oxygen supply is a topic outside the scope of this study. This study did not answer the question whether maintaing the target MAP of E80 mmHg by using NE affected neurologic outcome and survival compared with a lower MAP. CONCLUSIONS
In this observational study, administration of NE to maintain MAP E80 during CPB was not associated with significant changes in cerebral rSO2. These results indicated that NE could be used to maintain MAP during CPB without reducing rSO2. ACKNOWLEDGMENTS
The authors wish to thank the staff of the departments of cardiothoracic surgery and anesthesia and the cardiothoracic intensive care unit for their assistance in performing this study.
REFERENCES 1. Yao FS, Tseng CC, Ho CY, et al: Cerebral oxygen desaturation is associated with early postoperative neuropsychological dysfunction in patients undergoing cardiac surgery. J Cardiothorac Vasc Anesth 18:552-558, 2004 2. de Tournay-Jette E, Dupuis G, Bherer L, et al: The relationship between cerebral oxygen saturation changes and postoperative cognitive dysfunction in elderly patients after coronary artery bypass graft surgery. J Cardiothorac Vasc Anesth 25:95-104, 2011 3. Slater JP, Guarino T, Stack J, et al: Cerebral oxygen desaturation predicts cognitive decline and longer hospital stay after cardiac surgery. Ann Thorac Surg 87:36-44, 2009 4. Murkin JM, Adams SJ, Novick RJ, et al: Monitoring brain oxygen saturation during coronary bypass surgery: A randomized, prospective study. Anesth Analg 104:51-58, 2007 5. Scheeren TW, Schober P, Schwarte LA: Monitoring tissue oxygenation by near infrared spectroscopy (NIRS): Background and current applications. J Clin Monit Comput 26:279-287, 2012
6. Murphy GS, Hessel EA 2nd, Groom RC: Optimal perfusion during cardiopulmonary bypass: An evidence-based approach. Anesth Analg 108:1394-1417, 2009 7. Miller RD, Eriksson LI: Miller’s anesthesia. Philadelphia: Churchill Livingstone Elsevier, 2010 8. Brassard P, Seifert T, Secher NH: Is cerebral oxygenation negatively affected by infusion of norepinephrine in healthy subjects? Br J Anaesth 102:800-805, 2009 9. Brassard P, Pelletier C, Martin M, et al: Influence of norepinephrine and phenylephrine on frontal lobe oxygenation during cardiopulmonary bypass in patients with diabetes. J Cardiothorac Vasc Anesth 28:608-617, 2014 10. Moerman A, Denys W, De Somer F, et al: Influence of variations in systemic blood flow and pressure on cerebral and systemic oxygen saturation in cardiopulmonary bypass patients. Br J Anaesth 111: 619-626, 2013
296
11. Denault A, Deschamps A, Murkin JM: A proposed algorithm for the intraoperative use of cerebral near-infrared spectroscopy. Semin Cardiothorac Vasc Anesth 11:274-281, 2007 12. Laflam A, Joshi B, Brady K, et al: Shoulder surgery in the beach chair position is associated with diminished cerebral autoregulation but no differences in postoperative cognition or brain injury biomarker levels compared with supine positioning: The anesthesia patient safety foundation beach chair study. Anesth Analg 120:176-185, 2015 13. Brady K, Joshi B, Zweifel C, et al: Real-time continuous monitoring of cerebral blood flow autoregulation using near-infrared spectroscopy in patients undergoing cardiopulmonary bypass. Stroke 41:1951-1956, 2010 14. Sakamoto Y, Ishiguro M, Kitagawa G: Akaike information criterion statistics. Dordrecht. The Netherlands: D. Reidel, 1986 15. Roman PE, Grigore AM: Pro: Hypothermic cardiopulmonary bypass should be used routinely. J Cardiothorac Vasc Anesth 26: 945-948, 2012 16. Olesen J: The effect of intracarotid epinephrine, norepinephrine, and angiotensin on the regional cerebral blood flow in man. Neurology 22:978-987, 1972 17. Cassaglia PA, Griffiths RI, Walker AM: Sympathetic nerve activity in the superior cervical ganglia increases in response to imposed increases in arterial pressure. Am J Physiol Regul Integr Comp Physiol 294:R1255-R1261, 2008 18. Arbab MA, Wiklund L, Svendgaard NA: Origin and distribution of cerebral vascular innervation from superior cervical, trigeminal and spinal ganglia investigated with retrograde and anterograde WGA-HRP tracing in the rat. Neuroscience 19:695-708, 1986 19. Meng L, Cannesson M, Alexander B, et al: Effect of phenylephrine and ephedrine bolus treatment on cerebral oxygenation in anaesthetized patients. Br J Anaesth 107:209-217, 2011 20. Joshi B, Brady K, Lee J, et al: Impaired autoregulation of cerebral blood flow during rewarming from hypothermic
HAGEN ET AL
cardiopulmonary bypass and its potential association with stroke. Anesth Analg 110:321-328, 2010 21. Murkin JM, Farrar JK, Tweed WA, et al: Cerebral autoregulation and flow/metabolism coupling during cardiopulmonary bypass: The influence of PaCO2. Anesth Analg 66:825-832, 1987 22. Reinsfelt B, Westerlind A, Ricksten SE: The effects of sevoflurane on cerebral blood flow autoregulation and flow-metabolism coupling during cardiopulmonary bypass. Acta Anaesthesiol Scand 55: 118-123, 2011 23. Moerman AT, Vanbiervliet VM, Van Wesemael A, et al: Assessment of cerebral autoregulation patterns with near-infrared spectroscopy during pharmacological-induced pressure changes. Anesthesiology 123:327-335, 2015 24. Bland JM, Altman DG: Correlation in restricted ranges of data. BMJ 342:d556, 2011 25. Davie SN, Grocott HP: Impact of extracranial contamination on regional cerebral oxygen saturation: A comparison of three cerebral oximetry technologies. Anesthesiology 116:834-840, 2012 26. Sorensen H, Secher NH, Siebenmann C, et al: Cutaneous vasoconstriction affects near-infrared spectroscopy determined cerebral oxygen saturation during administration of norepinephrine. Anesthesiology 117:263-270, 2012 27. Sorensen H, Rasmussen P, Sato K, et al: External carotid artery flow maintains near infrared spectroscopy-determined frontal lobe oxygenation during ephedrine administration. Br J Anaesth 113: 452-458, 2014 28. McLeod AD, Igielman F, Elwell C, et al: Measuring cerebral oxygenation during normobaric hyperoxia: A comparison of tissue microprobes, near-infrared spectroscopy, and jugular venous oximetry in head injury. Anesth Analg 97:851-856, 2003 29. Sun X, Ellis J, Corso PJ, et al: Skin pigmentation interferes with the clinical measurement of regional cerebral oxygen saturation. Br J Anaesth 114:276-280, 2015
C
EREBRAL OXIMETERY is being used increasingly in cardiac surgery to monitor cerebral oxygenation. Reduced cerebral tissue oxygen saturation (rSO2) during cardiac surgery has been associated with adverse outcomes,1-3 and interventions based on monitoring of rSO2 during coronary artery bypass grafting (CABG) have been shown to reduce major organ dysfunction.4 Near-infrared spectroscopy (NIRS) is used to monitor rSO2, and details about NIRS have been provided by Scheeren et al.5 There is insufficient evidence to recommend a specific mean arterial pressure (MAP) during cardiopulmonary bypass (CPB), but it has been recommended to maintain MAP 470 mmHg during CPB in high-risk patients.6 Hypoperfusion resulting from impairment of cerebral autoregulation, hypotension, or anemia may lead to cerebral hypoxia. Autoregulation of cerebral blood flow (CBF) ensures delivery of oxygenated blood to the brain and protects the brain from ischemia caused by arterial pressure fluctuations. One approach that aims to improve cerebral oxygenation during CPB is to increase cerebral perfusion pressure by increasing systemic MAP using a vasopressor7 (eg, norepinephrine [NE]). According to departmental practice, the authors routinely administer NE during CPB, aiming for a MAP of 60 to 80 mmHg. However, the question has been raised whether vasopressors may induce cerebral vasoconstriction and thereby actually contribute to reduced cerebral oxygenation.8-10 NE has been shown to reduce rSO2 in healthy volunteers,8 but potential effects of NE on rSO2 during CPB have not been defined. The aim of this study, therefore, was to explore whether NE administered during CPB to maintain MAP at E 80 mmHg was associated with changes in rSO2. The hypothesis was that there would be no such association when controlling for other factors with biologically plausible effects on rSO2.
oxygen, hemoglobin, and pump flow values were averaged over 1 minute, giving a total of 3,460 data points entered as covariates in a linear mixed model for repeated measurements, with cerebral oxygen saturation measured using near-infrared spectroscopy as outcome. There was no statistically significant association between norepinephrine dose to maintain mean arterial pressure and cerebral oxygen saturation (p ¼ 0.46) in this model. Conclusions: Administration of norepinephrine to maintain mean arterial pressure E80 mmHg during cardiopulmonary bypass was not associated with statistically significant changes in cerebral oxygen saturation. These results indicated that norepinephrine could be used to increase mean arterial pressure during cardiopulmonary bypass without reducing cerebral oxygen saturation. & 2016 Elsevier Inc. All rights reserved. KEY WORDS: oximetry, norepinephrine, cardiopulmonary bypass
METHODS
Patients Forty-five patients presenting for elective aortic valve replacement, CABG, or a combination of aortic valve replacement and CABG were included in the study. Exclusion criteria were age o18 or 480 years, patient included in pharmacologic study, vasoactive medication started before measuring baseline rSO2, or any of the following medical conditions known at time of inclusion: frontal lobe pathology, intracranial vascular anomaly, neurologic disease, dementia, history of traumatic brain injury, cerebral insult, transient ischemic attack, or carotid artery stenosis (no study-specific examinations were performed). The study was approved by the Regional Committee for Medical and Health Research Ethics (REK sør-øst, 2010/823), and written informed consent was obtained. The study was registered in ClinicalTrials.gov (NCT01940874).
From the *Department of Anesthesiology, Oslo University Hospital, Oslo, Norway; †Faculty of Medicine, University of Oslo, Oslo, Norway; ‡Department of Cardiothoracic Surgery; §Oslo Centre of Biostatistics and Epidemiology, Research Support Services, and ǁDepartment of Neurosurgery, Oslo University Hospital, Oslo, Norway; and ¶University of Bergen, Bergen, Norway. Results from this study were presented, in part, at the Society of Cardiovascular Anesthesiologists annual meeting in Miami, FL, May 2013. Address reprint requests to Ove Andreas Hagen, MD, Department of Anesthesiology, Oslo Univeristy Hospital, Kirkeveien 166, 0450, Oslo, Norway. E-mail: [email protected] © 2016 Elsevier Inc. All rights reserved. 1053-0770/2601-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2015.12.015
Journal of Cardiothoracic and Vascular Anesthesia, Vol 30, No 2 (April), 2016: pp 291–296
291
292
Anesthetic Procedure Patients were premedicated with intramuscular morphine/ scopolamine. A 20-G arterial cannula was placed in the left radial artery and a 7-Fr triple-lumen catheter in the right internal jugular vein. Baseline rSO2 was measured when patients were breathing room air. Anesthesia was induced with intravenous diazepam, fentanyl, and propofol, and endotracheal intubation was facilitated with intravenous cisatracurium. Patients were mechanically ventilated, and anesthesia was maintained with sevoflurane. Temperature was measured in the urinary bladder. Although core temperature affects cerebral oxygen consumption, temperature measurements were not used in the statistical model due to the poor agreement between bladder and core temperatures. Surgical Procedure and Cardiopulmonary Bypass After cannulation of the aorta and right atrium, CPB primed with Ringer’s acetate was performed in a nonpulsatile manner. CPB flow was set to 2.4 (height [cm] þ weight [kg] – 60)/ 100 L/min, with partial pressure of arterial oxygen (PaO2) 15 to 20 kPa and partial pressure of arterial carbon dioxide (PaCO2) 5.2 to 5.7 kPa. Arterial blood gas analyses (α-stat) were performed intermittently. Temperature was measured in the urinary bladder. CABG procedures were performed with the patient in normothermia (35.0-37.01C), whereas other or combined procedures were performed with the patient in mild hypothermia (32.0-35.01C) or hypothermia (o32.01C). Heparin was administered to maintain activated clotting time 4480 seconds. Packed red blood cells were given routinely if hemoglobin (Hb) was o7 g/dL. Anesthesia was maintained with sevoflurane, 1.5% to 2% adminstered in the CPB circuit. NE (Alaris CC Plus with Guardrails; CareFusion, San Diego, CA) diluted in glucose, 50 mg/mL to 20 mg/mL, was infused in the central venous catheter using a syringe pump (Alaris GH Plus with Guardrails; CareFusion, San Diego, CA). During CPB, NE was administered with the aim of achieving MAP E80 mmHg. If rSO2 was reduced by 420% relative to baseline or to o50% absolute, the following steps were taken to improve rSO2: verification of CPB integrity and function/positioning of the cannulae; and controlling head position, PaCO2, O2, Hb, and CPB flow.11 If the CPB circuit, position of the cannulae, and PaO2 were adequate, PaCO2, Hb, and/or CPB flow were increased. Unless intentionally altered as previously stated, PaO2 and PaCO2 were kept constant by continuous arterial inline monitoring (BMU 40; Maquet, Rastatt, Germany). Signal Acquisition and Analysis Sensors (Adult SomaSensor; Medtronic, Minneapolis, MN) for the oximeter (INVOS 5100C Cerebral/Somatic Oximeter; Medtronic) were placed on the left and right forehead. Values were updated every 7 to 8 seconds and sampled with hemodynamic variables from the patient monitor (Solar 9500; GE Healthcare, Milwaukee, WI) at 1 Hz in a custom-made program in LabVIEW (National Instruments, Austin, TX). Dosage of NE was entered manually into this program. All rSO2 values were relative to the preoperative baseline value,
HAGEN ET AL
and the average of left and right measurements were used for analysis. Data were analyzed from initiation of CPB to removal of the aortic cross-clamp. Offline, data were down-sampled to 1 value every minute by averaging (arithmetic means) after manually removing obviously erroneous data (eg, from flushing the arterial line), providing 3,460 observations. Results from arterial blood gas analyses were entered manually. To obtain continuous values for Hb, PaO2, and PaCO2, these values were interpolated linearly between the blood gas analyses performed. When packed red blood cells were transfused, the Hb value was changed at the time of transfusion (not linearly interpolated). In a subset of 12 patients, arterial blood gas analyses were performed every 10 minutes to increase the resolution of these variables. If CPB flow was altered, flow relative to baseline was entered (eg, a 5% increase giving a value of 1.05). If CPB flow was not altered, it always had a value of 1. To explore autoregulation, as previously reported,12 rSO2 and MAP were averaged over 10 seconds from the raw data, giving a sampling frequency of 0.1 Hz. Pearson correlation coefficients between rSO2 and MAP were calculated from 5 minutes (30 observations) in a moving, overlapping fashion, making each data point contribute to 30 correlation coefficients (except at the ends of the recording). The correlations were binned according to the average MAP in the 5-minute period of the correlation coefficient with bin size 10 mmHg. Lower and upper limits of autoregulation (LLA and ULA, respectively) were defined if there was a transition of bin mean correlation from o0.3 to 40.3 with a reduction or increase in MAP-bin, respectively.12,13 For each bin, MAP was presented by its mean value (eg, 65 mmHg for the 60-70 mmHg-bin). Statistics Values are mean (standard deviation) or median (25th-75th percentiles) unless otherwise stated. The relation between NE dosage and rSO2 was evaluated in a linear mixed regression model with rSO2 as the oucome. The NE dose and other variables considered to have an effect on rSO2 were entered as explanatory variables. These other variables were MAP, PaCO2, PaO2, Hb, and CPB flow. Because the study only sought to describe the possible effect of NE dose on rSO2, the other varaibles (which were entered only to correct for their possible confounding effects) were included in the statistical model regardless of their p values. The model assumed a linear relation with the same slope between all explanatory variables and the outcome, rSO2. The mixed model accounts for dependence of data within each patient (patient being a random effect), and in this model, with random intercept, allowed the regression line of each patient to have different intercepts. Model fit was assessed using the Akaike information criterion (AIC).14 Due to the repeated nature of the data, the AIC was minimized by imposing an autoregressive-1 covariance structure to the residuals. RESULTS
Forty-eight patients were included in the study. Two patients had incomplete data, and 1 underwent surgery in deep hypothermia. Thus, data from 45 patients were analyzed.
293
NOREPINEPHRINE AND REGIONAL CEREBRAL OXYGENATION
Table 1. Patient Characteristics (n ¼ 45) Male, n (%) Age, yr Height, cm Weight, kg Surgical procedures CABG, n (%) Aortic valve replacement Aortic stenosis Aortic insufficiency CABG and aortic valve replacement (aortic stenosis) combined
35 64 176 86 15 20 15 5 10
(78) (33-78) (8.4) (14) (33) (44) (33) (11) (22)
NOTE. Data are mean (standard deviation) or median (25th-75th) percentiles after Kolmogorov-Smirnov test for normality, unless otherwise stated. Abbreviation: CABG, coronary artery bypass grafting.
Surgery was performed using hypothermia in 2 patients, mild hypothermia in 22 patients, and normothermia in 21 patients. All CABG patients underwent surgery in normothermia.15 Ten patients had an absolute rSO2 o50% at any time during CPB. Twenty-one patients had a decrease in rSO2 420% relative to the baseline value at any time during CPB; 8 of these patients had decreases in rSO2 420% relative to the baseline value lasting 420 consecutive minutes. Patient characteristics are presented in Table 1. rSO2 and the explanatory variables for all 3,460 observations (1-minute averages) are presented in Table 2. Scatter plots of rSO2 and MAP versus NE dose and rSO2 versus MAP are presented in Fig 1. When modeling the data, random intercept, random slope, and random intercept-random slope models were attempted, but the best fit (as judged using AIC) was found with a repeated structure with correlated residuals, first-order autoregressive covariance matrix. In the multivariate analysis, there was no significant association between NE dose and rSO2 relative to baseline. There were significant associations between rSO2 and MAP, PaO2, Hb, and CPB flow (see Table 2). In the subset of 12 patients with frequent (every 10 minutes) blood gas analyses, there was no association between rSO2 and NE dose (p ¼ 1.0). By the method of correlation between MAP and rSO2, the authors identified LLA in 26 patients and ULA in 15 patients. Medians (25th-75th percentiles) were LLA 65 (55-65) mmHg and ULA 75 (75-85) mmHg, respectively. All patients were discharged from the cardiothoracic intensive care unit the day after surgery, and there were no perioperative deaths. None of the patients had clinically recognized neurologic complications at discharge. DISCUSSION
The main finding in this study was that there was no significant association between NE dose administered during CPB to maintain MAP at E80 mmHg and rSO2. The lower confidence interval (CI) boundary corresponded to a o0.2% reduction in rSO2 with each 0.1 mg/kg/min NE-increment, which the authors did not believe to be relevant clinically. This study, therefore, indicated that NE could be used to increase
MAP during CPB without significantly affecting rSO2. A direct effect of sympathomimetic agents on cerebral resistance vessels is believed to be practically negligible because vasoactive amines do not cross the blood–brain barrier.16 However, whether sympathetic nerve activity originating from the superior cervical ganglion increases when MAP is augmented pharmacologically has been discussed.17 Because the cerebral vasculature is innervated predominantly from the superior cervical ganglion,18 it seems relevant to investigate and define the effects of sympathomimetic agents on cerebral oxygenation. NE administered in an intact circulation may cause a variety of cardiovascular effects, not only peripheral vasoconstriction. CPB as a model provides the opportunity to study the effect of NE with a fixed “cardiac” output. Sympathomimetic agents are used routinely to increase MAP during surgery. Using NIRS technology, Meng et al19 demonstrated decreased cerebral oxygenation during anesthesia when administering boluses of phenylephrine and unaltered cerebral oxygenation when administering boluses of ephedrine. Brassard et al8 found reduced cerebral oxygenation with increasing doses of NE in healthy volunteers; however, in the study presented here, NE was administered to maintain a fixed MAP, whereas MAP and systemic vascular resistance substantially increased in Brassard et al’s study.8 In a later study, Brassard et al9 found that administration of NE during CPB was associated with reduced rSO2 in diabetic patients but not in nondiabetic patients. Only 4 of the patients in this study were diabetic. In this subgroup, there was a significant relationship between rSO2 and NE dose (p ¼ 0.043). However, this post-hoc analysis on a very low number of patients should be interpreted with caution, and the CIs
Table 2. Summary of Covariates and Outcome and Regression Coefficients Summary Statistics (n ¼ 3,460)
Regression Coefficients
Median (25th-75th Percentiles)
rSO2, % from baseline NE, mg/kg/min MAP, mmHg PaCO2, kPa PaO2, kPa Hb, g/dL CPB flow Intercept
Coefficient (95% CI)
p Value
87 (82-92) 0.15 74 5.4 21.0 8.9 1.0
(0.09-0.25) (64-80) (4.9-5.8) (19.0-23.7) (8.3-10.0) (1.0-1.0)
–0.50 0.76 1.0 0.36 2.11 14 38
(–1.9 to 0.84) 0.46 (0.68-0.83) o0.001 (–0.33 to 2.3) 0.14 (0.072-0.66) 0.014 (1.4-2.8) o0.001 (5.2-23) 0.002 (23-54) 0.424
NOTE. Regional cerebral tissue oxygen saturation (outcome) and variables entered as covariates in the regression model. Summary statistics in the left column. Regression coefficients (95% confidence interval) for variables used as covariates in the middle column with p values (right column). The regression coefficients are the estimated changes in regional cerebral tissue oxygen saturation (%) for a 1-unit increase of each predictor. Note that cardiopulmonary bypass flow is given as a value relative to the initial setting given by the equation presented in the text (ie, a 10% increase would give the value 1.1). Abbreviations: rSO2, regional cerebral tissue oxygen saturation relative to baseline; NE, norepinephrine; MAP, mean arterial pressure; Hb, hemoglobin; CPB, cardiopulmonary bypass.
294
HAGEN ET AL
Fig 1. Scatter plots of norepinephrine and mean arterial pressure versus regional cerebral oxygen saturation and norepinephrine versus mean arterial pressure. Norepinephrine for all observations (1 minute average) in all patients. Abbreviation: rSO2, regional cerebral oxygen saturation.
corresponded with a reduction in rSO2 of 1.9% to 0.03% for a 0.1-mg/kg/min increment in NE dose. The authors of this study aimed for a higher MAP (80 mmHg) than that in the study by Brassard et al9 (460 mmHg). The authors of this study found a modest association between rSO2 and MAP, corresponding to an absolute increase in rSO2 of 7.6% (95% CI 6.8-8.3, p o 0.001) for each 10mmHg MAP increment. This assumed a linear relationship between MAP and CBF. Classically, CBF has been believed to be autoregulated, maintained unaltered when MAP is kept between LLA and ULA. Autoregulation implies that changes in MAP are not associated with changes in CBF. Thus, correlation between the 2 variables are expected to be close to 0. Outside the autoregulatory plateau, CBF is believed to be pressuredependent, reflected by positive correlations between MAP and CBF as MAP varies. This is the basic principle behind a series of studies on autoregulation during CPB in which correlations between MAP and rSO2 over minutes have been used to detect the limits of autoregulation. Using this method, Brady et al found that measurements using transcranial Doppler and cerebral oximetry agreed well.13 The mean nadir temperature in that study was 31.11C, but the same group of authors reported possible impairment of autoregulation in the rewarming phase.20 Autoregulation seemed to be better preserved by using an α-stat management (as in this study) compared with pH-stat,21 but may be impaired by sevoflurane,22 which was used in this study. Paradoxic changes in rSO2 (ie, a reduced rSO2 as MAP was increased and vice versa) with sodium nitroprusside and phenylephrine have been reported in patients with intact autoregulation.23 It should be noted that in the Moerman et al study, MAP was altered intentionally; whereas in this study, a vasocontrictor was given with the aim of maintaining a near-constant MAP. According to the method of correlations,12 the authors ideintified LLA and ULA in this study in 26 and 15 patients, respectively. However, there is a possible caveat when using this method according to the authors’ data. Because the authors aimed to maintain MAP at E80 mmHg, deviations in MAP from this target value will tend to produce both a change in mean MAP from this target value and an increased range of MAP values. Even
under an assumption of no autoregulatory plateau, this phenomenon could tend to increase correlation coefficients in MAP-bins above and below this target value, based on the dependence of the Pearson correlation between 2 variables on the range of values studied.24 Furthermore, each observation was used in 30 correlation coefficients, creating dependence between these values. A correlation coefficient of 0.3 was chosen somewhat arbitrarily as a threshold, as different values have been used. However, this was in accordance with more recent works.23 The overall data did not reveal an association between NE dose and rSO2. Fig 2 illustrates a case in which rSO2 was low despite recommended interventions. After increasing MAP with NE, the anesthesiologist abruptly reduced the NE dose. The data suggested that the increased NE dose was associated with reduced rSO2 despite an increased MAP because the pattern was repeated with a subsequent increase and decrease of NE. Although the authors of the study presented here did not find an overall association between NE dose and rSO2, this may have been the case in individual patients. The observation presented in Fig 2 may be due to chance alone or an unknown confounding factor. The authors of the study presented here did not have data to suggest characteristics of patients with such a response, if present. The NIRS device subtracts a shallow signal from the deeper signal, aiming for a spatial resolution allowing for measurement of deep (brain) tissue only. However, there is a concern regarding contamination from superficial tissues.25-27 Both the authors’ overall findings and the example in Fig 2 thus may be subject to contamination from superficial tissues. Regarding the overall results, it would be expected that, given the near-fixed “cardiac” output during CPB, a change in skin circulation with an increased NE dose would lead to decreased oxygen saturation in superficial tissues. Thus, the fact that the authors did not find reduced rSO2 with increased NE doses indicated that brain oxygen saturation was not reduced, even if contamination from superficial tissues occured. METHODOLOGIC CONSIDERATIONS
The authors did not perform a formal power analysis for sample size determination beause they did not find adequate
295
NOREPINEPHRINE AND REGIONAL CEREBRAL OXYGENATION
Fig 2. Data from a patient in whom regional cerebral oxygen saturation (left and right, grey lines) was reduced considerably with initiation of cardiopulmonary bypass without satisfactory effect of recommended measures. As norepinephrine (dashed line) was reduced, mean arterial pressure (solid line) decreased, whereas regional cerebral oxygen saturation increased. The pattern was repeated when a re-escalation of norepinephrine was attempted, and mean arterial pressure was allowed to drift to 40 to 50 mmHg. Time from initiation of CPB on x-axis. Left yaxis is mean arterial pressure (mmHg) and regional cerebral oxygen saturation relative to baseline (%). Right y-axis is norepinephrine dose.
published data in the planning phase. The authors planned to include 50 patients. However, 2 patients with exclusion criteria had been included (age 480 years), 2 had incomplete data, and 1 underwent surgery in deep hypothermia, leaving 45 patients for analysis. The authors believe that the CI of the outcome indicated that the sample size was sufficiently large. Temperature was measured in the bladder, which may lag considerably behind core temperature during temperature changes. Therefore, although hypothermia reduces cerebral metabolism, temperature was not entered into the statistical model presented. However, entering temperature as measured in the bladder into the model did not significantly change the overall results (data not shown). Baseline measurements were performed with patients breathing room air, giving the possibility of low baselines, especially considering the opioid-containing premedication. Measuring baseline rSO2 with supplemental oxygen has been recommended.11 However, normobaric hyperoxia has been associated with increased rSO2,28 which could lead to false high baselines. Ideally, baseline rSO2 perhaps should be measured after titrating supplemental oxygen to a normal PaO2, but this would be unacceptably cumbersome in clinical practice. Furthermore, baseline rSO2 has been associated with skin pigmentation.29 All the patients in this study were white.
Systematic neuropsychologic testing was not performed; therefore, the authors cannot exclude possible subtle postoperative neurologic disturbances. However, no neurologic complications were detected on routine clinical examination. Moderate cerebral arteriosclerosis probably affected some of the elderly patients. The authors aimed for a fixed MAP during CPB in this study. Therefore, whether increased MAP produces a better cerebral oxygen supply is a topic outside the scope of this study. This study did not answer the question whether maintaing the target MAP of E80 mmHg by using NE affected neurologic outcome and survival compared with a lower MAP. CONCLUSIONS
In this observational study, administration of NE to maintain MAP E80 during CPB was not associated with significant changes in cerebral rSO2. These results indicated that NE could be used to maintain MAP during CPB without reducing rSO2. ACKNOWLEDGMENTS
The authors wish to thank the staff of the departments of cardiothoracic surgery and anesthesia and the cardiothoracic intensive care unit for their assistance in performing this study.
REFERENCES 1. Yao FS, Tseng CC, Ho CY, et al: Cerebral oxygen desaturation is associated with early postoperative neuropsychological dysfunction in patients undergoing cardiac surgery. J Cardiothorac Vasc Anesth 18:552-558, 2004 2. de Tournay-Jette E, Dupuis G, Bherer L, et al: The relationship between cerebral oxygen saturation changes and postoperative cognitive dysfunction in elderly patients after coronary artery bypass graft surgery. J Cardiothorac Vasc Anesth 25:95-104, 2011 3. Slater JP, Guarino T, Stack J, et al: Cerebral oxygen desaturation predicts cognitive decline and longer hospital stay after cardiac surgery. Ann Thorac Surg 87:36-44, 2009 4. Murkin JM, Adams SJ, Novick RJ, et al: Monitoring brain oxygen saturation during coronary bypass surgery: A randomized, prospective study. Anesth Analg 104:51-58, 2007 5. Scheeren TW, Schober P, Schwarte LA: Monitoring tissue oxygenation by near infrared spectroscopy (NIRS): Background and current applications. J Clin Monit Comput 26:279-287, 2012
6. Murphy GS, Hessel EA 2nd, Groom RC: Optimal perfusion during cardiopulmonary bypass: An evidence-based approach. Anesth Analg 108:1394-1417, 2009 7. Miller RD, Eriksson LI: Miller’s anesthesia. Philadelphia: Churchill Livingstone Elsevier, 2010 8. Brassard P, Seifert T, Secher NH: Is cerebral oxygenation negatively affected by infusion of norepinephrine in healthy subjects? Br J Anaesth 102:800-805, 2009 9. Brassard P, Pelletier C, Martin M, et al: Influence of norepinephrine and phenylephrine on frontal lobe oxygenation during cardiopulmonary bypass in patients with diabetes. J Cardiothorac Vasc Anesth 28:608-617, 2014 10. Moerman A, Denys W, De Somer F, et al: Influence of variations in systemic blood flow and pressure on cerebral and systemic oxygen saturation in cardiopulmonary bypass patients. Br J Anaesth 111: 619-626, 2013
296
11. Denault A, Deschamps A, Murkin JM: A proposed algorithm for the intraoperative use of cerebral near-infrared spectroscopy. Semin Cardiothorac Vasc Anesth 11:274-281, 2007 12. Laflam A, Joshi B, Brady K, et al: Shoulder surgery in the beach chair position is associated with diminished cerebral autoregulation but no differences in postoperative cognition or brain injury biomarker levels compared with supine positioning: The anesthesia patient safety foundation beach chair study. Anesth Analg 120:176-185, 2015 13. Brady K, Joshi B, Zweifel C, et al: Real-time continuous monitoring of cerebral blood flow autoregulation using near-infrared spectroscopy in patients undergoing cardiopulmonary bypass. Stroke 41:1951-1956, 2010 14. Sakamoto Y, Ishiguro M, Kitagawa G: Akaike information criterion statistics. Dordrecht. The Netherlands: D. Reidel, 1986 15. Roman PE, Grigore AM: Pro: Hypothermic cardiopulmonary bypass should be used routinely. J Cardiothorac Vasc Anesth 26: 945-948, 2012 16. Olesen J: The effect of intracarotid epinephrine, norepinephrine, and angiotensin on the regional cerebral blood flow in man. Neurology 22:978-987, 1972 17. Cassaglia PA, Griffiths RI, Walker AM: Sympathetic nerve activity in the superior cervical ganglia increases in response to imposed increases in arterial pressure. Am J Physiol Regul Integr Comp Physiol 294:R1255-R1261, 2008 18. Arbab MA, Wiklund L, Svendgaard NA: Origin and distribution of cerebral vascular innervation from superior cervical, trigeminal and spinal ganglia investigated with retrograde and anterograde WGA-HRP tracing in the rat. Neuroscience 19:695-708, 1986 19. Meng L, Cannesson M, Alexander B, et al: Effect of phenylephrine and ephedrine bolus treatment on cerebral oxygenation in anaesthetized patients. Br J Anaesth 107:209-217, 2011 20. Joshi B, Brady K, Lee J, et al: Impaired autoregulation of cerebral blood flow during rewarming from hypothermic
HAGEN ET AL
cardiopulmonary bypass and its potential association with stroke. Anesth Analg 110:321-328, 2010 21. Murkin JM, Farrar JK, Tweed WA, et al: Cerebral autoregulation and flow/metabolism coupling during cardiopulmonary bypass: The influence of PaCO2. Anesth Analg 66:825-832, 1987 22. Reinsfelt B, Westerlind A, Ricksten SE: The effects of sevoflurane on cerebral blood flow autoregulation and flow-metabolism coupling during cardiopulmonary bypass. Acta Anaesthesiol Scand 55: 118-123, 2011 23. Moerman AT, Vanbiervliet VM, Van Wesemael A, et al: Assessment of cerebral autoregulation patterns with near-infrared spectroscopy during pharmacological-induced pressure changes. Anesthesiology 123:327-335, 2015 24. Bland JM, Altman DG: Correlation in restricted ranges of data. BMJ 342:d556, 2011 25. Davie SN, Grocott HP: Impact of extracranial contamination on regional cerebral oxygen saturation: A comparison of three cerebral oximetry technologies. Anesthesiology 116:834-840, 2012 26. Sorensen H, Secher NH, Siebenmann C, et al: Cutaneous vasoconstriction affects near-infrared spectroscopy determined cerebral oxygen saturation during administration of norepinephrine. Anesthesiology 117:263-270, 2012 27. Sorensen H, Rasmussen P, Sato K, et al: External carotid artery flow maintains near infrared spectroscopy-determined frontal lobe oxygenation during ephedrine administration. Br J Anaesth 113: 452-458, 2014 28. McLeod AD, Igielman F, Elwell C, et al: Measuring cerebral oxygenation during normobaric hyperoxia: A comparison of tissue microprobes, near-infrared spectroscopy, and jugular venous oximetry in head injury. Anesth Analg 97:851-856, 2003 29. Sun X, Ellis J, Corso PJ, et al: Skin pigmentation interferes with the clinical measurement of regional cerebral oxygen saturation. Br J Anaesth 114:276-280, 2015