Abstract
Background
Postoperative cognitive dysfunction (POCD) occurs commonly after cardiac surgery. Near-infrared spectroscopy (NIRS) has been used to monitor regional cerebral oxygen saturation (rScO2) in order to minimise the occurrence of POCD by applying dedicated interventions when rScO2 decreases. However, the association between rScO2 intraoperatively and POCD has not been clarified.
Methods
This is a secondary analysis of a randomised trial with physician-blinded NIRS monitoring and cognitive testing at discharge from hospital and at 3 months after surgery. The association between intraoperative rScO2 values and POCD at discharge from hospital and at 3 months after surgery was investigated. The prespecified candidate predictive variable of interest was cumulative time during surgery with rScO2 ≥10% below its preoperative value.
Results
One hundred and fifty-three patients had complete NIRS data and neurocognitive assessments at discharge, and 44 of these patients (29%) had POCD. At 3 months, 148 patients had complete data, and 12 (8%) of these patients had POCD. The median time with rScO2 >10% below preoperative values did not differ for patients with and without POCD at discharge (difference=0.0 min; Hodges-Lehmann 95% confidence interval, −3.11–1.47, P=0.88). Other rScO2 time thresholds that were assessed were also not significantly different between those with and without POCD at discharge. This applied both to absolute rScO2 values and relative changes from preoperative values. Similar results were found in relation to POCD at 3 months.
Conclusions
No significant association was found between intraoperative rScO2 values and POCD. These findings bring into question the rationale for attempting to avoid decreases in rScO2 if the goal is to prevent POCD.
Clinical trial registration
Keywords: cardiac surgery; delayed neurocognitive recovery; neuropsychological tests; postoperative complications; postoperative neurocognitive disoders; spectroscopy, near-infrared
Editor's key points.
-
•
Delayed neurocognitive recovery and longer lasting neurocognitive disorder reportedly occur commonly after cardiac surgery.
-
•
This study, which was a secondary analysis of data from a clinical trial, hypothesised that intraoperative decreases in cerebral oxygen saturation would predict postoperative neurocognitive disorders, previously referred to as postoperative cognitive dysfunction (POCD).
-
•
Based on the criteria used by the investigators to assess cognitive function, there were 44 (29%) patients with POCD (or delayed neurocognitive recovery at the time of hospital discharge) and 12 (8%) patients with POCD (or postoperative neurocognitive disorder) at 3 months after surgery; and there were four patients (2.6%) at discharge and nine patients (6.1%) at 3 months with postoperative neurocognitive improvement.
-
•
Neither the duration nor the extent of intraoperative cerebral oxygen saturation was associated with the occurrence of POCD at discharge (delayed neurocognitive recovery) or at 3 months (postoperative neurocognitive disorder).
-
•
Major limitations of this study include the small number of ‘positive’ outcomes (patients with POCD) and the ongoing imprecision in diagnosing neurocognitive disorders after surgery.
Neurocognitive complications are a major concern after cardiac surgery, and postoperative cognitive dysfunction (POCD) has been observed in 23–81% of these patients.1, 2, 3 Near-infrared spectroscopy (NIRS) has been used to monitor regional cerebral oxygen saturation (rScO2).4, 5 Thus, NIRS might be useful for identifying patients at risk of developing POCD, and to mitigate the cerebral desaturation during cardiac surgery. In previous randomised trials, NIRS has been used in accordance with an intervention algorithm6 to minimise cerebral desaturation with the underlying premise that rScO2 is associated with cognitive deterioration after surgery. Some studies have reported a decrease in the severity, frequency, and accumulated load of cerebral desaturation when applying the intervention algorithm, but the effect on POCD occurrence is inconsistent among studies.5, 7, 8 This may either be related to the fact that the interventions used to prevent cerebral desaturation are not sufficient or alternatively, that the association between rScO2 readings and POCD is weak. In the Perfusion Pressure Cerebral Infarct (PPCI) trial, neurological complications were studied in patients randomised to two distinct BP targets during on-pump cardiac surgery. The trial found that low rScO2 was associated with new cerebral lesions evaluated by diffusion weighted MRI (DWI), but no obvious BP safety threshold could be idenfied.9 In the current study, we aimed to explore associations between blinded NIRS measurements and neuropsychological test data collected in the PPCI trial. We hypothesised that patients with POCD would have longer cumulative time with rScO2>10% below its baseline values.
Methods
Study design and setting
This study is a secondary analysis of the PPCI trial. The primary trial endpoint1 and study protocol10 have been published previously. The PPCI trial was a single-centre, parallel-group, 1:1 randomised trial investigating the importance of specific MAP targets during cardiac surgery in relation to neurologic complications. The trial was approved by the regional ethics committee in the capital region of Denmark (H-3-2013-110, chairperson Michael Bitsch, August 26, 2013) and by the Danish Data Protection Agency (J.no.: 30-0805 and 30-1434) and was registered at clinicaltrials.gov (NCT02185885) on July 7, 2014. Written informed consent was obtained from all participators before inclusion. A decrease in rScO2 was prespecified as a candidate risk factor for POCD, which was a secondary outcome of the PPCI trial. The PPCI trial was conducted at Rigshospitalet, University of Copenhagen, Copenhagen, Denmark. Patients were randomised to either low MAP (LMAP) (40–50 mm Hg) or high MAP (HMAP) (70–80 mm Hg) during cardiopulmonary bypass (CPB). Target MAP was achieved with the use of norepinephrine to a maximum of 0.4 μg kg−1 min−1 during a fixed blood flow of 2.4 L m−2 min−1+10–20%.
Participants
All patients were 18 yr of age or older and undergoing cardiac surgery, either coronary artery bypass grafting, heart valve surgery, or both with the use of CPB. Exclusion criteria are listed in the study protocol,10 and included a history of stroke, history of transient ischemic attack, diagnosis of neurodegenerative disorders, or contraindications to magnetic resonance imaging scans.
Cardiopulmonary bypass and anaesthesia
The management of anaesthesia and CPB has been described in detail previously.1, 10 In short, anaesthesia was induced with fentanyl, propofol, and cisatracurium and maintained with sevoflurane. PaCO2 and pH were managed in accordance with the α-stat strategy. During CPB, sevoflurane was administered (0.5–3.0% concentration) via the bypass circuit.
Signal acquisition and analysis
Regional cerebral oximetry
Regional cerebral oximetry data were collected with self-adhesive sensors (Medtronic/Covidien INVOS Cerebral/Somatic Oximetry Adult Sensors—Somanetics Corporation, Troy, MI, USA) placed bilaterally on the patient's forehead during surgery. The sensor distance between the emitter and detector was 40 mm. The sensors were connected to a Covidien/Medtronic INVOS 5100c Cerebral/Somatic Oximeter monitor (Somanetics Corporation).
The NIRS monitor was operated in a blinded fashion by applying a study mode dedicated specifically to research purposes and had a sampling frequency of approximately 0.166 Hz. Data were stored offline for later analysis. One minute after placement of the sensors and before pre-oxygenation and induction of anaesthesia, preoperative rScO2 values on both the left and right channel were marked as an event. Data files were extracted to Microsoft Excel (Microsoft, Inc., Redmond, WA, USA) with INVOS Analytics Tool, version 1.2.1 (Somanetics Corporation), and exported to dedicated statistical software as described in the statistics section. At each time point, the rScO2 value was taken as the average between the left and right sensors.
Postoperative cognitive dysfunction
A Mini Mental State Examination (MMSE) was conducted after enrolment in the PPCI trial, but before cognitive testing, to make sure patients were capable of understanding test instructions. Patients with an MMSE score ≤24 were not tested any further. Cognitive function was evaluated on the day before surgery and again on the day before discharge from hospital or the 8th postoperative day, whichever came first. The same test battery was repeated 2–4 months after surgery (referred to as 3 months follow-up in the following). The International Study of Postoperative Cognitive Dysfunction (ISPOCD) test battery was used, which included the Visual Verbal Learning test, Concept Shifting test, Stroop Colour Word Interference (SCWI) test, and Letter Digit Coding (LDC) test.11 A detailed description of the assessment of POCD has been published previously and the occurrence of POCD in the PPCI trial has also been published previously.1, 10
Outcome
Postoperative cognitive dysfunction
We used the method described by the ISPOCD group to evaluate POCD.11 Seven variables from four tests were used and the mean learning effect from a control group was subtracted before individual z-scores were calculated based on changes from the baseline test (before surgery). Patients were classified with POCD when two out of seven z-scores for individual tests or the composite z-score were >1.96. This method is therefore based on overall deterioration (the composite z-score) or a severe deterioration in at least two variables, as described previously.11 We also calculated the number of patients in whom the corresponding improvement was found.
Regional cerebral oximetry
For each patient, we calculated the cumulative durations that rScO2 was ≥10% and ≥20% below its preoperative value. We also calculated the cerebral desaturation load (CDL) as the area under the curve according to three rScO2 thresholds over time (minutes). CDL below baseline refers to the CDL below the preoperative value, CDL10 refers to the CDL load below the threshold of 10% below its preoperative value, and CDL20 refers to the CDL load below the threshold of 20% below its preoperative value. Mean rScO2 was calculated for the intraoperative period and subsequently for the dedicated CPB period, defined as the time from full blood flow on CPB until CPB was terminated. Minimum and maximum rScO2 values were defined as the lowest and highest recorded value during surgery. The prespecified main candidate predictive variable of interest was cumulative time during surgery with rScO2 ≥10% below its preoperative value, which was in line with a recently published intervention study.12
Statistical analysis
Statistical analyses were performed using SPSS (version 22.0, IBM Corp., Armonk, NY, USA). Normally distributed data are presented as mean (standard deviation [SD]), whereas non-normally distributed data are presented as median and inter-quartile range (IQR). Groups are compared using the Student's t-test (normally distributed data) or Mann–Whitney U-test (skewed data). Categorical data are presented as numbers and percentages with 95% confidence interval and compared with Pearson's χ2 test or Fischer's exact test. Statistical significance was assessed at the 5% level. We assessed the correlation between rScO2 variables during surgery and selected z-scores for specific cognitive tests (SCWI time and LDC score, and cumulative z-score, which are described to be the most sensitive parts of the POCD test battery)13, 14 at the discharge time point. Correlation was assessed with Spearman's test. The analysis was conducted in three steps: the two treatment groups in the PPCI trial (LMAP and HMAP) separately, and the entire PPCI population. As preoperative rScO2 values decrease with increasing age and as POCD is more common in older adults,15, 16 we repeated all analyses after age stratification with a cut-off age <68 yr or ≥68 yr. No sample size calculation was performed, because this study was a secondary analysis of a randomised trial.
Results
Patient enrolment started on July 8, 2014. The last patient was enrolled on January 6, 2016. The last follow-up was completed in April 2016. Among 197 patients included in the PPCI trial, we obtained complete intraoperative NIRS data and cognitive test data in 153 patients at discharge from hospital. At the 3 month follow-up visit, complete data were obtained for 148 patients. Patients with POCD at discharge were older, and preoperative atrial fibrillation and valve surgery procedures were more common in these patients (Table 1). In addition, administration of norepinephrine was used more frequently during CPB in these patients. Forty-four patients (29%, 95% confidence interval, 22–36%) had POCD at discharge, whereas only 12 had POCD at 3 months (8%, 95% confidence interval, 4–13%). Baseline characteristics and intraoperative data at 3 months can be seen in the online Supplementary Appendix S1, as no difference was found between patients with and without POCD. Improvement in cognitive function was detected in four patients (2.6%) and nine patients (6.1%) at discharge and at 3 months, respectively.
Table 1.
Baseline characteristics and intraoperative data for cardiac surgery patients with and without postoperative cognitive dysfunction (POCD) at discharge—entire Perfusion Pressure Cerebral Infarct (PPCI) trial population. Values are reported as mean (standard deviation) except for age, which is mean with (inter-quartile range) and numbers with percentages. CABG, coronary artery bypass grafting; CPB, cardiopulmonary bypass; EuroSCORE, European system for cardiac operative risk evaluation; LVEF, left ventricular ejection fraction; NE, norepinephrine.
| Patient characteristics | All patients (n=153) | POCD at discharge |
P-value | |
|---|---|---|---|---|
| No (n=109) | Yes (n=44) | |||
| Age (yr) | 66.4 (60.3; 74.8) | 64.7 (57.0; 72.5) | 69.9 (66.0; 76.0) | <0.01 |
| Male sex, n (%) | 138 (90.2) | 99 (90.8) | 39 (88.6) | 0.68 |
| BMI (kg m−2) | 27.1 (3.8) | 27.4 (4.0) | 26.7 (3.4) | 0.28 |
| Comorbidity, before operation, n (%) | ||||
| Obstructive lung disease | 14 (9.2) | 12 (11.0) | 2 (4.5) | 0.35 |
| Hypertension | 130 (85.0) | 94 (86.2) | 36 (81.8) | 0.49 |
| Hypercholesterolemia | 123 (80.4) | 88 (80.7) | 35 (79.5) | 0.87 |
| Smoker, former or current | 109 (71.2) | 79 (72.5) | 30 (68.2) | 0.60 |
| Diabetes | 39 (25.5) | 27 (24.8) | 12 (27.3) | 0.75 |
| Atrial fibrillation before surgery | 22 (14.4) | 9 (8.3) | 13 (29.5) | <0.01 |
| Previous myocardial infarction | 16 (10.5) | 11 (10.1) | 5 (11.4) | 0.78 |
| Recent myocardial infarction (2 weeks before hospitalisation) | 42 (27.5) | 34 (31.2) | 8 (18.2) | 0.10 |
| LVEF (%) | 50.7 (11.3) | 51.4 (10.5) | 48.9 (13.1) | 0.21 |
| Baseline P-creatinine (μmol L−1) | 89.6 (23.5) | 89.6 (25.5) | 90.1 (18.7) | 0.90 |
| Baseline haematocrit (%) | 40.6 (5.4) | 40.7 (5.5) | 39.8 (4.7) | 0.59 |
| EuroSCORE II | 2.4 (2.7) | 2.2 (2.4) | 2.9 (2.0) | 0.10 |
| Surgery | ||||
| CABG only | 84 (55.3) | 67 (61.5) | 17 (39.5) | 0.01 |
| Aortic/mitral valve only | 45 (29.6) | 28 (25.7) | 17 (39.5) | 0.09 |
| Combined CABG and valvular | 22 (14.5) | 13 (11.9) | 9 (20.9) | 0.16 |
| Procedures involving valvular surgery (valve+combined) | 67 (44.1) | 41 (37.6) | 26 (60.5) | 0.01 |
| Intraoperative data | ||||
| Procedure time (min) | 184.6 (45.7) | 181.8 (44.0) | 192.0 (51.1) | 0.22 |
| CPB time (min) | 95.8 (57.0) | 95.0 (64.6) | 99.4 (34.3) | 0.67 |
| Aortic cross clamp time (min) | 62.4 (28.1) | 60.2 (26.9) | 68.6 (30.1) | 0.10 |
| Blood flow during CPB (100% equals 2.4 L m−2 min−1) (%) | 112.1 (4.1) | 112.0 (4.3) | 112.5 (3.5) | 0.46 |
| MAP during CPB (mm Hg) | 55.1 (11.6) | 54.3 (11.9) | 56.4 (11.1) | 0.51 |
| Patients requiring NE during CPB (%) | 101 (66.0) | 65 (59.6) | 36 (81.8) | 0.01 |
Regarding the primary hypothesis of the study, there was no significant difference in the cumulative duration of time ≥10% below the rScO2 preoperative value in patients with and without POCD at discharge. Also, no significant differences were found between the patients with and without POCD at discharge when analysing preoperative rScO2, mean rScO2 during CPB or during surgery, minimum rScO2, time spent below preoperative rScO2 or 20% below preoperative rScO2, or the accumulated CDL (Table 2) for the entire PPCI population. When stratified in the LMAP and HMAP groups, analyses revealed that patients with POCD spent a significantly longer time below preoperative rScO2 in the HMAP group (Table 3). This significant difference was not observed in the LMAP group (Table 4). At 3 months follow-up, there were no significant differences in any of the recorded NIRS variables according to POCD.
Table 2.
Regional cerebral oxygen saturation (rScO2) values for cardiac surgery patients with and without postoperative cognitive dysfunction (POCD) at discharge and at 3 months—entire Perfusion Pressure Cerebral Infarct (PPCI) trial population. Median with (inter-quartile range) and tested with Mann–Whitney U-test. Differences calculated as independent samples Hodges-Lehmann median difference estimates with 95% confidence intervals (95% CI). CDL, cerebral desaturation load (area under these thresholds); CPB, cardiopulmonary bypass
|
Postoperative cognitive dysfunction at discharge |
||||
|
No (n=109) |
Yes (n=44) |
Difference (95% CI) |
P-value |
|
| Preoperative rScO2 | 66.5 (61.0; 71.5) | 64.5 (60.0; 71.0) | 2.0 (−1.0; 4.5) | 0.21 |
| Mean rScO2 during CPB | 67.3 (61.1; 73.3) | 65.3 (61.0; 71.1) | 1.2 (−2.0; 4.3) | 0.47 |
| Mean rScO2 during surgery | 69.0 (64.5; 73.4) | 69.0 (61.4; 71.9) | 1.6 (−1.2; 4.2) | 0.27 |
| Time below preoperative rScO2 (min) | 54.1 (18.2; 122.5) | 75.6 (10.6; 137.1) | 3.6 (−12.8; 29.3) | 0.68 |
| Time below a 10% decrease from preoperative rScO2 (min) | 3.8 (0.2; 28.2) | 5.31 (0.1; 46.4) | 0.0 (−3.1; 1.5) | 0.88 |
| Time below a 20% decrease from preoperative rScO2 (min) | 0.0 (0.0; 0.7) | 0.0 (0.0; 1.1) | 0.0 (0.0; 0.0) | 0.73 |
| CDL preoperative (% × min) | 223.8 (39.4; 493.0) | 267.8 (26.1; 768.1) | 5.3 (−57.0; 134.5) | 0.80 |
| CDL10 (% × min) | 5.3 (0.0; 73.0) | 9.5 (0.0; 105.9) | 0.0 (−4.5; 2.0) | 0.98 |
| CDL20 (% × min) | 0.0 (0.0; 1.5) | 0.0 (0.0; 3.3) | 0.0 (0.0; 0.0) | 0.95 |
| Minimum rScO2 value | 55.0 (49.3; 61.0) | 53.3 (47.1; 58.5) | 2.5 (−0.5; 5.5) | 0.11 |
| Maximum rScO2 value |
82.5 (76.5; 87.8) |
81.8 (77.0; 85.5) |
0.5 (−2.5; 3.5) |
0.68 |
|
Postoperative cognitive dysfunction at 3 months |
||||
|
No (n=136) |
Yes (n=12) |
Difference (95% CI) |
P-value |
|
| Preoperative rScO2 | 66.5 (60.6; 72.0) | 65.0 (58.5; 76.5) | 0.0 (−5.5; 6.5) | 0.97 |
| Mean rScO2 during CPB | 67.2 (61.2; 73.1) | 70.3 (61.5; 72.3) | 1.6 (−4.7; 6.7) | 0.61 |
| Mean rScO2 during surgery | 69.1 (64.5; 73.4) | 69.9 (59.4; 73.6) | 0.2 (−5.4; 4.8) | 0.94 |
| Time below preoperative rScO2 (min) | 59.9 (15.6; 126.3) | 33.2 (17.9; 172.2) | 2.5 (−45.6; 30.1) | 0.80 |
| Time below a 10% decrease from preoperative rScO2 (min) | 4.7 (0.2; 29.6) | 1.0 (0.1; 55.8) | 0.5 (−0.6; 9.8) | 0.53 |
| Time below a 20% decrease from preoperative rScO2 (min) | 0.0 (0.0; 0.8) | 0.0 (0.0; 0.1) | 0.0 (0.0; 0.1) | 0.51 |
| CDL preoperative (% × min) | 236.5 (33.0; 543.3) | 81.0 (41.3; 1046.8) | 22.3 (−87.0; 238.0) | 0.65 |
| CDL10 (% × min) | 7.5 (0.0; 71.5) | 1.8 (0.3; 94.1) | 1.5 (−1.5; 20.0) | 0.42 |
| CDL20 (% × min) | 0.0 (0.0; 1.3) | 0.0 (0.0; 0.4) | 0.0 (0.0; 0.0) | 0.63 |
| Minimum rScO2 value | 55.0 (49.1; 60.5) | 55.3 (48.5; 62.9) | 1.0 (−5.0; 7.0) | 0.70 |
| Maximum rScO2 value | 82.3 (77.0; 87.4) | 83.5 (70.0; 89.1) | 0.0 (−5.5; 6.5) | 0.94 |
Table 3.
Regional cerebral oxygen saturation (rScO2) values for cardiac surgery patients with and without postoperative cognitive dysfunction (POCD) at discharge and at 3 months—high target MAP group. Median with (inter-quartile range) and tested with Mann–Whitney U-test. Differences calculated as independent samples Hodges-Lehmann median difference estimates with 95% confidence intervals (95% CI). CDL, cerebral desaturation load (area under these thresholds); CPB, cardiopulmonary bypass.
|
Postoperative cognitive dysfunction at discharge |
||||
|
No (n=48) |
Yes (n=24) |
Difference (95% CI) |
P-value |
|
| Preoperative rScO2 | 66.0 (61.1; 69.9) | 65.0 (61.5; 71.8) | 0.5 (−3.5; 4.0) | 0.78 |
| Mean rScO2 during CPB | 66.4 (60.6; 71.1) | 64.4 (59.6; 69.1) | 1.8 (−2.3; 5.4) | 0.38 |
| Mean rScO2 during surgery | 68.9 (65.5; 72.8) | 66.6 (60.8; 69.7) | 2.8 (−0.6; 6.2) | 0.12 |
| Time below preoperative rScO2 (min) | 50.0 (10.6; 103.0) | 75.6 (10.6; 137.1) | 41.2 (−9.0; 74.1) | 0.02 |
| Time below a 10% decrease from preoperative rScO2 (min) | 6.7 (0.3; 29.5) | 16.6 (1.3; 65.0) | 5.3 (−0.6; 26.9) | 0.11 |
| Time below a 20% decrease from preoperative rScO2 (min) | 0.1 (0.0; 4.5) | 0.1 (0.0; 6.3) | 0.0 (−0.5; 0.5) | 0.85 |
| CDL preoperative (% × min) | 195.0 (24.8; 486.9) | 392.0 (162.0; 915.5) | 172.3 (2.0; 172.3) | 0.05 |
| CDL10 (% × min) | 10.0 (0.5; 107.0) | 26.8 (1.9; 148.0) | 4.5 (−3.0; 30.0) | 0.24 |
| CDL20 (% × min) | 0.0 (0.0; 6.8) | 0.0 (0.0; 6.1) | 0.0 (0.0; 0.0) | 0.77 |
| Minimum rScO2 value | 53.5 (48.0; 58.9) | 52.5 (46.8; 57.4) | 1.5 (−2.5; 6.0) | 0.44 |
| Maximum rScO2 value |
82.5 (77.5; 86.9) |
81.8 (78.0; 83.9) |
1.0 (−2.0; 4.5) |
0.53 |
|
Postoperative cognitive dysfunction at 3 months |
||||
|
No (n=65) |
Yes (n=4) |
Difference (95% CI) |
P-value |
|
| Preoperative rScO2 | 66.0 (61.0; 71.5) | 65.0 (63.8; 70.8) | 0.0 (−7.5; 6.5) | 0.93 |
| Mean rScO2 during CPB | 65.7 (60.8; 71.0) | 70.5 (69.1; 72.0) | 4.8 (−2.3; 11.0) | 0.12 |
| Mean rScO2 during surgery | 68.8 (64.6; 72.8) | 69.9 (59.36; 73.64) | 1.8 (−3.5; 7.8) | 0.28 |
| Time below preoperative rScO2 (min) | 0.4 (0.1; 1.8) | 31.9 (6.3; 98.7) | 28.9 (−22.8; 98.1) | 0.29 |
| Time below a 10% decrease from preoperative rScO2 (min) | 0.4 (0.1; 1.8) | 12.3 (1.0; 44.1) | 11.7 (−0.1; 49.3) | 0.06 |
| Time below a 20% decrease from preoperative rScO2 (min) | 0.0 (0.3; 0.1) | 0.1 (0.0; 5.9) | 0.1 (−0.1; 7.1) | 0.34 |
| CDL preoperative (% × min) | 80.8 (14.9; 324.8) | 278.0 (62.0; 703.8) | 162.5 (−45.5; 668.5) | 0.20 |
| CDL10 (% × min) | 1.3 (0.3; 2.6) | 21.0 (1.5; 125.5) | 19.3 (0.0; 143.5) | 0.05 |
| CDL20 (% × min) | 0.0 (0.0; 0.0) | 0.0 (0.0; 6.8) | 0.0 (0.0; 7.0) | 0.16 |
| Minimum rScO2 value | 55.3 (51.5; 62.0) | 53.5 (47.8; 58.3) | 3.5 (−4.0; 11.5) | 0.50 |
| Maximum rScO2 value | 84.3 (81.1; 87.4) | 82.5 (77.8; 85.3) | 2.5 (3.0; 2.5) | 0.40 |
Table 4.
Regional cerebral oxygen saturation (rScO2) values for cardiac surgery patients with and without postoperative cognitive dysfunction (POCD) at discharge and at 3 months—lLow target MAP group. Median with (inter-quartile range) and tested with Mann–Whitney U-test. Differences calculated as independent samples Hodges-Lehmann median difference estimates with 95% confidence intervals (95% CI). CDL, cerebral desaturation load (area under these thresholds); CPB, cardiopulmonary bypass.
|
Postoperative cognitive dysfunction at discharge |
||||
|
No (n=61) |
Yes (n=20) |
Difference (95% CI) |
P-value |
|
| Preoperative rScO2 | 67.0 (60.8; 74.3) | 62.3 (56.3; 70.3) | 4.5 (−0.5; 9.0) | 0.07 |
| Mean rScO2 during CPB | 67.9 (61.1; 74.8) | 67.2 (61.2; 75.9) | 0.0 (−5.06; 5.01) | 0.96 |
| Mean rScO2 during surgery | 69.2 (63.3; 74.4) | 70.1 (62.1; 74.5) | 0.1 (−4.19; 4.4) | 0.97 |
| Time below preoperative rScO2 (min) | 65.6 (20.5; 130.3) | 12.3 (4.4; 116.3) | 18.6 (0.0; 56.2) | 0.05 |
| Time below a 10% decrease from preoperative rScO2 (min) | 2.8 (0.2; 26.5) | 0.2 (0.0; 17.3) | 1.1 (0.0; 4.3) | 0.06 |
| Time below a 20% decrease from preoperative rScO2 (min) | 0.0 (0.0; 0.2) | 0.0 (0.0; 0.1) | 0.0 (0.0; 0.0) | 0.91 |
| CDL preoperative (% × min) | 224.5 (42.8; 501.0) | 26.3 (5.9; 428.0) | 55.0 (−0.5; 244.5) | 0.06 |
| CDL10 (% × min) | 3.5 (0.0; 58.8) | 0.0 (0.0; 41.3) | 1.3 (0.0; 5.5) | 0.09 |
| CDL20 (% × min) | 0.0 (0.0; 0.5) | 0.0 (0.0; 0.0) | 0.0 (0.0; 0.0) | 0.58 |
| Minimum rScO2 value | 57.5 (51.0; 62.0) | 54.5 (47.1; 58.9) | 3.0 (−2.0; 7.5) | 0.22 |
| Maximum rScO2 value |
81.0 (75.3; 88.0) |
81.8 (73.4; 90.1) |
0.5 (−4.5; 5.0) |
0.91 |
|
Postoperative cognitive dysfunction at 3 months |
||||
|
No (n=71) |
Yes (n=8) |
Difference (95% CI) |
P-value |
|
| Preoperative rScO2 | 67.0 (60.0; 73.0) | 66.8 (52.9; 77.0) | 0.5 (−8.0; 10.5) | 0.91 |
| Mean rScO2 during CPB | 68.3 (61.2; 74.5) | 66.4 (53.8; 77.8) | 2.0 (−6.2; 10.3) | 0.72 |
| Mean rScO2 during surgery | 70.4 (63.8; 74.7) | 67.7 (56.2; 78.3) | 1.8 (−6.26; 11.0) | 0.61 |
| Time below preoperative rScO2 (min) | 65.6 (12.3; 128.1) | 44.0 (18.1; 199.4) | 11.1 (−53.9; 76.5) | 0.61 |
| Time below a 10% decrease from preoperative rScO2 (min) | 2.3 (0.1; 21.0) | 2.0 (0.1; 81.1) | 0.2 (−2.5; 63.9) | 0.59 |
| Time below a 20% decrease from preoperative rScO2 (min) | 0.0 (0.0; 0.1) | 0.02 (0.0; 00.1) | 0.0 (0.0; 0.0) | 0.86 |
| CDL preoperative (% × min) | 223.0 (22.0; 487.5) | 97.5 (41.3; 1370.5) | 27.0 (−218.0; 832.5) | 0.64 |
| CDL10 (% × min) | 2.5 (0.0; 33.0) | 2.0 (0.4; 130.5) | 0.5 (−4.5; 110.5) | 0.63 |
| CDL20 (% × min) | 0.0 (0.0; 0.0) | 0.0 (0.0; 23.0) | 0.0 (0.0; 0.0) | 0.33 |
| Minimum rScO2 value | 57.5 (51.0; 61.5) | 57.0 (43.8; 63.6) | 0.3 (−8.5; 10.0) | 0.97 |
| Maximum rScO2 value | 82.0 (75.5; 88.0) | 75.8 (68.0; 92.5) | 2.5 (−6.5; 13.5) | 0.72 |
The frequency of POCD was not significantly different between patients who did and did not experience a 10% or 20% decrease in rScO2, both at discharge and at 3 months (Table 5). There was no significant correlation between the z-score of SCWI time, LDC score, or the composite z-score and any rScO2 variables at discharge (Table 6).
Table 5.
Analysis of decrease below 10% and 20% relative to regional cerebral oxygen saturation (rScO2) preoperative (10% and 20% decrease) and postoperative cognitive dysfunction (POCD) at discharge and at 3 months in cardiac surgery patients. 95% CI, 95% confidence interval for odds ratio; OR: odds ratio.
|
POCD at discharge (%) |
||
|
10% decrease |
No |
Yes |
| No | 16 (69.6) | 7 (30.4) |
| Yes | 93 (71.5) | 37 (28.5) |
| Total | 109 | 44 |
| χ2P=0.85, OR to develop POCD with 10% decrease=0.91 (95% CI 0.35; 2.40) | ||
|
POCD at discharge (%) |
||
|
20% decrease |
No |
Yes |
| No | 58 (73.4) | 21 (26.6) |
| Yes | 51 (68.9) | 23 (31.1) |
| Total | 109 | 44 |
| χ2P=0.54, OR to develop POCD with 20% decrease=1.25 (95% CI 0.62; 2.51) | ||
|
POCD at 3 months (%) |
||
|
10% decrease |
No |
Yes |
| No | 22 (95.7) | 1 (4.3) |
| Yes | 114 (91.2) | 11 (8.8) |
| Total | 136 | 12 |
| Fisher's exact P=0.69, OR to develop POCD with 10% decrease=2.12 (95% CI 0.26; 17.29) | ||
|
POCD at 3 months (%) |
||
|
20% decrease |
No |
Yes |
| No | 70 (92.1) | 6 (7.9) |
| Yes | 66 (91.7) | 6 (8.3) |
| Total | 136 | 12 |
| χ2P=0.92, OR to develop POCD with 20% decrease=1.06 (95% CI 0.33; 3.45). | ||
Table 6.
Correlation between regional cerebral oxygen saturation (rScO2) variables and selected cognitive test variable elements for patients with postoperative cognitive dysfunction (POCD) at discharge after cardiac surgery. CDL, cerebral desaturation load (area under these thresholds); CPB, cardiopulmonary bypass; LDC, z-score for Letter Digit Coding; R, correlation coefficient; SCWI, z-score for Stroop Colour Word Interference; z-score, composite z-score.
| SCWI time |
LDC score |
z-score |
||||
|---|---|---|---|---|---|---|
| R | P-value | R | P-value | R | P-value | |
| Preoperative rScO2 | 0.01 | 0.89 | 0.06 | 0.44 | −0.11 | 0.17 |
| Mean rScO2 during CPB | 0.01 | 0.87 | 0.09 | 0.28 | −0.01 | 0.17 |
| Mean rScO2 during surgery | −0.01 | 0.97 | 0.09 | 0.27 | −0.07 | 0.41 |
| Time (min) below preoperative rScO2 | 0.03 | 0.74 | −0.02 | 0.89 | 0.06 | 0.48 |
| Time (min) below a 10% decrease from preoperative rScO2 | −0.01 | 0.90 | −0.04 | 0.67 | −0.02 | 0.81 |
| Time (min) below a 20% decrease from preoperative rScO2 | 0.08 | 0.34 | −0.03 | 0.69 | −0.02 | 0.81 |
| CDL Preoperative (% × min) | 0.06 | 0.46 | −0.06 | 0.49 | −0.03 | 0.76 |
| CDL10 (% × min) | 0.01 | 0.87 | −0.08 | 0.31 | −0.06 | 0.47 |
| CDL20 (% × min) | 0.08 | 0.33 | −0.09 | 0.28 | −0.01 | 0.95 |
| Minimum rScO2 | 0.03 | 0.76 | 0.05 | 0.57 | −0.10 | 0.22 |
| Maximum rScO2 | −0.03 | 0.71 | 0.06 | 0.46 | −0.05 | 0.58 |
Statistical analyses with age-stratification did not change the results (data not shown). There was no difference in oxygenation or ventilation, as judged by the partial pressure of O2 and CO2 and the fraction of delivered O2 during CPB, between patients with and without POCD at discharge. Similarly, there was no difference in choice of anaesthetic agents or haematocrit during CPB.
Discussion
The main finding of the present study was that there was no difference in intraoperative rScO2 variables between patients with and without POCD after cardiac surgery, neither at discharge nor after 3 months.
Previously, randomised studies have assessed whether a dedicated intervention algorithm6 based on rScO2 could ameliorate cerebral oxygen desaturation and potentially decrease the incidence of cognitive deterioration after surgery. These studies mainly report a beneficial effect of the intervention algorithm in relation to cerebral desaturation measured with NIRS, yet conflicting results have been reported in relation to the effect on POCD,17, 18, 19, 20, 21 which has also been summarised in recent meta-analyses.5, 7, 8
Other studies investigated the association between rScO2 and POCD by comparing NIRS values between patients with and without POCD after cardiac surgery2, 3, 22, 23 in order to evaluate NIRS as a screening tool. In general, these studies reported a high variability in the frequency of POCD, varying from 23%3 to 81%.2 Two studies, with 61 and 101 patients, respectively, reported an association between nadir rScO2 values below 50%2 and 35%22 and POCD. The latter study, however, must be cautiously interpreted because of very low rScO2 values. In the same study, an association between POCD and rScO2 below 40% for more than 10 min22 was reported, but similar severe desaturation did not occur in our study. Also worth mentioning is the difference in study methodology for NIRS monitoring between the two studies; one study used an unblinded setup with an intervention algorithm6 to minimise cerebral desaturation2 whereas the other study used blinded NIRS monitoring.22 Three other studies, with 1003, 60,23 and 4724 patients, respectively, used blinded NIRS monitoring and reported no association between any of the analysed rScO2 variables and POCD. These findings are in line with the present study, where we could not demonstrate any significant association between blinded rScO2 values and POCD. This may be because of a true lack of association between rScO2 and POCD, as described above. Having said that, some of the previous studies have observed an association.23
The definition of POCD varies across studies, as some use a restricted number of tests (e.g. MMSE), whereas other studies use extended test batteries designed to test different aspects of cognitive function and incorporate, for instance, the learning effect as in the present study.5, 7, 25 The variation in the test battery of POCD may influence the frequency of this outcome. New recommendations of the nomenclature of cognitive change associated with anaesthesia and surgery have been published recently, recommending a terminology consistent with the Diagnostic and Statistical Manual for Mental Disorders, fifth edition,26 which would make studies more comparable and ease interpretation of the observations. In short, the new nomenclature will define POCD at discharge as ‘delayed neurocognitive recovery’ and POCD at 3 months as ‘postoperative neurocognitive disorder’ (NCD). In addition, the new recommendations propose that subjective complaints are considered, and that activity of daily living are assessed, in order to distinguish between mild and major NCD.
The choice of rScO2 variable(s) and cut-off levels varies between studies, thereby impairing comparison between studies.27 In one study, evaluating the association between rScO2 baseline and 30-day mortality, an association between the European system for cardiac operative risk evaluation (EuroSCORE)28 and rScO2 baseline was observed. This might indicate that NIRS is better for identification of patients at risk, rather than for monitoring intraoperative changes and guiding interventions. The ability of NIRS to identify patients at risk is not an unexpected finding, considering the fact that patient age is one of the variables used for EuroSCORE calculation, and at the same time rScO2 baseline values decrease with increasing age.15 However, it emphasises the importance of deciding which rScO2 variable should be of primary interest. In addition, one should keep in mind that the labels of the NIRS devices used differ between studies. This may cause variations in rScO2 values, which makes it difficult to compare NIRS studies, especially specific reference ranges, as NIRS devices from different manufacturers have been reported to show different rScO2 values and altered relative changes within the same patient.29
The present study shows that some degree of desaturation below rScO2 baseline occurs frequently during cardiac surgery, but the findings do not support the concept that desaturation is an important risk factor for POCD. In other settings with healthy volunteers, hypoxia induced by inhalation of an air mixture containing only oxygen 10% for 40 min is well tolerated and results in a 29% decrease in arterial oxygen saturation and 23% decrease in the venous oxygen saturation.30 These observations suggest that a short-lasting hypoxaemic episode does not result in brain damage in healthy young men, but they also raise the question of how we should define severe cerebral desaturation. In the present study, we tried to explore the relationship between severity and duration of desaturation by presenting time below a certain threshold, degree of desaturation, and the combined overall CDL, which is equivalent to the area under the curve reflecting both time and the severity of desaturation. In clinical practice, time below a certain threshold is easy to manage as it is displayed on screen, but for research purposes, it is possible to extend the analyses by looking at CDL and their relationship with neurological complications compared with the degree or duration separately. In the present study, the CDL analyses did not provide additional information. In a previous paper, we observed that patients with new lesions on DWI had lower rScO2 values compared with patients without lesions, both in terms of time below rScO2 baseline and time below 10% desaturation from the preoperative value, CDL preoperative, CDL10, and minimum rScO2.9 These findings indicate that the NIRS monitoring somehow reflects cerebral damage, whereas there were no differences in severity and duration of desaturation when looking for POCD. Fifty-two patients had POCD either at discharge or at 3 months and new cerebral lesions on DWI, reflecting a markedly higher frequency of DWI lesions in patients with POCD. Thus, even though there might be a statistically significant association between POCD and DWI lesions, no difference was found in any rScO2 variables in the present study evaluating the association between POCD and rScO2. Not all patients with DWI lesions developed POCD, which emphasises that the pathophysiology for POCD is multifactorial and is not necessary caused solely by development of new ischaemic infarcts.
Patients with POCD in the present study were older, had preoperative atrial fibrillation more frequently, and underwent surgical procedures that included the heart valves more often. In general, these characteristics have previously been identified as risk factors for neurological complications after cardiac surgery,31 suggesting that the classification of patients with POCD in the present study is probably valid. We used an extended POCD test battery, which takes learning effects into account, thereby refining the capability of identifying POCD.11, 32, 33 We tested a broad range of rScO2 variables, including absolute values, relative changes, and duration and magnitude of cerebral desaturation to clarify the importance of each individual component. Despite this broad exploratory approach, we found no significant differences in any of the rScO2 variables between patients with and without POCD, apart from one variable in the HMAP group analysis. However, slightly lower, but non-significantly different rScO2 values were observed in patients with POCD when looking at the entire PPCI population. A possible confounder leading to this trend could be the use of norepinephrine. Norepinephrine was given more frequently to patients with POCD and this may have caused a decrease in rScO2,29, 34 as norepinephrine is known to have an impact on NIRS measurements, probably attributable to vasoconstriction in extracranial tissue.35 In addition, we have previously shown, that the HMAP group had more severe cerebral desaturation compared with the LMAP group36 and this may be the reason for the significant finding in the HMAP group in the present study.
There are limitations in the study. Because of the low frequency of POCD at 3 months, the study must be considered inconclusive in relation to the association between rScO2 and POCD at this time point. Only six of the 12 patients with POCD at 3 months had POCD at discharge. This might be because of late onset POCD, problems with POCD classification, statistical noise, or patient trajectories. This limitation is further emphasised by the number of patients with cognitive improvement at 3 months: cognitive change at 3 months according to the criteria used in this study might merely represent variation around the mean or statistical noise. We did not investigate subjective cognitive functioning or quality of life in patients, which is a drawback in relation to evaluating the clinical relevance of POCD.26 Many analyses were conducted, including analyses stratified by the randomisation group in the parent trial. The single statistically significant result could be a chance finding, which can occur when many analyses are conducted. No correction method was applied to counteract the problem of multiple comparisons.
One of the major strengths of the study was that all results were obtained with blinded NIRS monitoring. Furthermore, patients underwent detailed cognitive testing with the ISPOCD test battery, using a systematic evaluation procedure leading to a standardised outcome assessment. In addition, we analysed multiple definitions of cerebral desaturation. To our knowledge, this is the largest study evaluating the association between clinician blinded, intraoperative rScO2 values and POCD in cardiac surgery patients. We did not find any association between intraoperative rScO2 variables and POCD occurrence after cardiac surgery.
Authors' contributions
Concept and design, analysis or interpretation of data, drafting of the manuscript, critical revision of the manuscript, administrative, technical or material support: all authors.
Acquisition of data: FH, AGV, JCN, HBR.
Statistical analysis: FH, LSR, HBR.
Obtaining funding: FH, AGV, JCN, HBR.
Study supervision: AGV, LSR, OBP, JCN, HBR.
Declaration of interest
HBR has received an honorary grant and a travel grant from Orion Pharma A/S.
Funding
The Research Foundations at Rigshospitalet (non-profit, reference number E-22972-04 and E-22972-02), University of Copenhagen, Denmark.
Handling editor: M. Avidan
Editorial decision: 15 March 2019
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.bja.2019.03.045.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
References
- 1.Vedel A.G., Holmgaard F., Rasmussen L.S. High-target vs low-target blood pressure management during cardiopulmonary bypass to prevent cerebral injury in cardiac surgery patients – a randomized controlled trial. Circulation. 2018;137:1770–1780. doi: 10.1161/CIRCULATIONAHA.117.030308. [DOI] [PubMed] [Google Scholar]
- 2.De Tournay-Jett E., Dupuis G., Bherer L., Deschamps A., Cartier R., Denault A. The relationship between cerebral oxygen saturation changes and postoperative cognitive dysfunction in elderly patients after coronary artery bypass graft surgery. J Cardiothorac Vasc Anesth. 2011;25:95–104. doi: 10.1053/j.jvca.2010.03.019. [DOI] [PubMed] [Google Scholar]
- 3.Hong S.W., Shim J.K., Choi Y.S., Kim D.H., Chang B.C., Kwak Y.L. Prediction of cognitive dysfunction and patients’ outcome following valvular heart surgery and the role of cerebral oximetry. Eur J Cardiothorac Surg. 2008;33:560–565. doi: 10.1016/j.ejcts.2008.01.012. [DOI] [PubMed] [Google Scholar]
- 4.Scheeren T.W.L., Kuizenga M.H., Maurer H., Struys M.M.R.F., Heringlake M. Electroencephalography and brain oxygenation monitoring in the perioperative period. Anesth Analg. 2019;128:265–277. doi: 10.1213/ANE.0000000000002812. [DOI] [PubMed] [Google Scholar]
- 5.Yu Y., Zhang K., Zhang L., Zong H., Meng L., Han R. Cerebral near-infrared spectroscopy (NIRS) for perioperative monitoring of brain oxygenation in children and adults. Cochrane Database Syst Rev. 2018;1:CD010947. doi: 10.1002/14651858.CD010947.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Denault A., Deschamps A., Murkin J.M. A proposed algorithm for the intraoperative use of cerebral near-infrared spectroscopy. Semin Cardiothorac Vasc Anesth. 2007;11:274–281. doi: 10.1177/1089253207311685. [DOI] [PubMed] [Google Scholar]
- 7.Serraino G.F., Murphy G.J. Effects of cerebral near-infrared spectroscopy on the outcome of patients undergoing cardiac surgery: a systematic review of randomised trials. BMJ Open. 2017;7:1–13. doi: 10.1136/bmjopen-2017-016613. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zorrilla-Vaca A., Healy R., Grant M.C. Intraoperative cerebral oximetry-based management for optimizing perioperative outcomes: a meta-analysis of randomized controlled trials. Can J Anaesth. 2018;65:529–542. doi: 10.1007/s12630-018-1065-7. [DOI] [PubMed] [Google Scholar]
- 9.Holmgaard F., Vedel A.G., Langkilde A., Lange T., Nilsson J.C., Ravn H.B. Differences in regional cerebral oximetry during cardiac surgery for patients with or without postoperative cerebral ischaemic lesions evaluated by magnetic resonance imaging. Br J Anaesth. 2018;121:1203–1211. doi: 10.1016/j.bja.2018.05.074. [DOI] [PubMed] [Google Scholar]
- 10.Vedel A.G., Holmgaard F., Rasmussen L.S. Perfusion Pressure Cerebral Infarct (PPCI) trial - the importance of mean arterial pressure during cardiopulmonary bypass to prevent cerebral complications after cardiac surgery: study protocol for a randomised controlled trial. Trials. 2016;17:1–11. doi: 10.1186/s13063-016-1373-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Moller J.T., Cluitmans P., Rasmussen L.S. Long-term postoperative cognitive dysfunction in the elderly: ISPOCD1 study. Lancet. 1998;351:857–861. doi: 10.1016/s0140-6736(97)07382-0. [DOI] [PubMed] [Google Scholar]
- 12.Deschamps A., Hall R., Grocott H. Cerebral oximetry monitoring to maintain normal cerebral oxygen saturation during high-risk cardiac surgery. Anesthesiology. 2016;124:1–11. doi: 10.1097/ALN.0000000000001029. [DOI] [PubMed] [Google Scholar]
- 13.Rasmussen L.S., Johnson T., Kuipers H.M. Does anaesthesia cause postoperative cognitive dysfunction? A randomised study of regional versus general anaesthesia in 438 elderly patients. Acta Anaesthesiol Scand. 2003;47:260–266. doi: 10.1034/j.1399-6576.2003.00057.x. [DOI] [PubMed] [Google Scholar]
- 14.Silverstein J., Steinmetz J., Reichenberg A., Harvey P., Rasmussen L.S. Postoperative cognitive dysfunction in older patients with preoperative cognitive impairment. Anesthesiology. 2007;106:431–435. doi: 10.1097/00000542-200703000-00006. [DOI] [PubMed] [Google Scholar]
- 15.Kobayashi K., Kitamura T., Kohira S. Factors associated with a low initial cerebral oxygen saturation value in patients undergoing cardiac surgery. J Artif Organs. 2017;20:110–116. doi: 10.1007/s10047-016-0941-6. [DOI] [PubMed] [Google Scholar]
- 16.Funder K.S., Steinmetz J., Rasmussen L.S. Cognitive dysfunction after cardiovascular surgery. Minerva Anestesiol. 2009;75:329–332. [PubMed] [Google Scholar]
- 17.Casati A., Fanelli G., Pietropaoli P. Continuous monitoring of cerebral oxygen saturation in elderly patients undergoing major abdominal surgery minimizes brain exposure to potential hypoxia. Anesth Analg. 2005;101:740–747. doi: 10.1213/01.ane.0000166974.96219.cd. [DOI] [PubMed] [Google Scholar]
- 18.Zogogiannis I.D., Iatrou C.A., Lazarides M.K. Evaluation of an intraoperative algorithm based on near-infrared refracted spectroscopy monitoring, in the intraoperative decision for shunt placement, in patients undergoing carotid endarterectomy. Middle East J Anesthesiol. 2011;21:367–374. [PubMed] [Google Scholar]
- 19.Vretzakis G., Georgopoulou S., Stamoulis K. Monitoring of brain oxygen saturation (INVOS) in a protocol to direct blood transfusions during cardiac surgery: a prospective randomized clinical trial. J Cardiothorac Surg. 2013;8:145. doi: 10.1186/1749-8090-8-145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Mohandas B., Jagadesh A., Vikram S. Impact of monitoring cerebral oxygen saturation on the outcome of patients undergoing open heart surgery. Ann Card Anaestheisa. 2013;16:102–106. doi: 10.4103/0971-9784.109740. [DOI] [PubMed] [Google Scholar]
- 21.Colak Z., Borojevic M., Bogovic A., Ivancan V., Biocina B., Majeric-Kogler V. Influence of intraoperative cerebral oximetry monitoring on neurocognitive function after coronary artery bypass surgery: a randomized, prospective study. Eur J Cardio-thoracic Surg. 2015;47:447–454. doi: 10.1093/ejcts/ezu193. [DOI] [PubMed] [Google Scholar]
- 22.Yao F.S.F., Tseng C.C.A., Ho C.Y.A., Levin S.K., Illner P. Cerebral oxygen desaturation is associated with early postoperative neuropsychological dysfunction in patients undergoing cardiac surgery. J Cardiothorac Vasc Anesth. 2004;18:552–558. doi: 10.1053/j.jvca.2004.07.007. [DOI] [PubMed] [Google Scholar]
- 23.Kok W.F., Harten AE Van, Koene B.M.J.A. A pilot study of cerebral tissue oxygenation and postoperative cognitive dysfunction among patients undergoing coronary artery bypass grafting randomised to surgery with or without cardiopulmonary bypass. Anaesthesia. 2014;69:613–622. doi: 10.1111/anae.12634. [DOI] [PubMed] [Google Scholar]
- 24.Reents W., Muellges W., Franke D., Babin-Ebell J., Elert O. Cerebral oxygen saturation assessed by near-infrared spectroscopy during coronary artery bypass grafting and early postoperative cognitive function. Ann Thorac Surg. 2002;74:109–114. doi: 10.1016/s0003-4975(02)03618-4. [DOI] [PubMed] [Google Scholar]
- 25.Berger M., Terrando N., Smith S.K., Browndyke J.N., Newman M.F., Mathew J.P. Neurocognitive function after cardiac surgery from phenotypes to mechanisms. Anesthesiology. 2018;129:829–851. doi: 10.1097/ALN.0000000000002194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Evered L., Silbert B., Knopman D.S. Recommendations for the nomenclature of cognitive change associated with anaesthesia and surgery—2018. Br J Anaesth. 2018;121:1005–1012. doi: 10.1016/j.bja.2017.11.087. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Nenna A., Barbato R., Greco S.M. 2017. Near-infrared spectroscopy in adult cardiac surgery : between conflicting results and unexpected uses; pp. 659–661. 14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Heringlake M., Garbers C., Kabler J.H. Preoperative cerebral oxygen saturation and clinical outcomes in cardiac surgery. Anesthesiology. 2011;114:58–69. doi: 10.1097/ALN.0b013e3181fef34e. [DOI] [PubMed] [Google Scholar]
- 29.Davie S.N., Grocott H.P. Impact of extracranial contamination on regional cerebral oxygen saturation. Anesthesiology. 2012;116:834–840. doi: 10.1097/ALN.0b013e31824c00d7. [DOI] [PubMed] [Google Scholar]
- 30.Vestergaard M.B., Larsson H.B.W. Cerebral metabolism and vascular reactivity during breath-hold and hypoxic challenge in freedivers and healthy controls. J Cereb Blood Flow Metab. 2017;0:1–15. doi: 10.1177/0271678X17737909. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Roach G.W., Kanchuger M., Mangona C.M. Adverse cerebral outcomes after coronary bypass surgery. N Engl J Med. 1996;335:1857–1863. doi: 10.1056/NEJM199612193352501. [DOI] [PubMed] [Google Scholar]
- 32.Johnson T., Monk T., Rasmussen L.S. Postoperative cognitive dysfunction in middle-aged patients. Anesthesiology. 2002;96:1351–1357. doi: 10.1097/00000542-200206000-00014. [DOI] [PubMed] [Google Scholar]
- 33.Krenk L., Rasmussen L.S., Siersma V.D., Kehlet H. Short-term practice effects and variability in cognitive testing in a healthy elderly population. Exp Gerontol. 2012;47:432–436. doi: 10.1016/j.exger.2012.03.011. [DOI] [PubMed] [Google Scholar]
- 34.Hagen O.A., Høiseth L.Ø., Roslin A. Impact of norepinephrine on regional cerebral oxygenation during cardiopulmonary bypass. J Cardiothorac Vasc Anesth. 2016;30:291–296. doi: 10.1053/j.jvca.2015.12.015. [DOI] [PubMed] [Google Scholar]
- 35.Sørensen H., Sc M., Secher N.H. Cutaneous vasoconstriction affects near-infrared spectroscopy determined cerebral oxygen saturation during administration of norepinephrine. Anesthesiology. 2012;117:263–270. doi: 10.1097/ALN.0b013e3182605afe. [DOI] [PubMed] [Google Scholar]
- 36.Holmgaard F, Vedel AG, Lange T, Nilsson JC, Ravn HB. Impact of 2 distinct levels of mean arterial pressure on near-infrared spectroscopy during cardiac surgery: secondary outcome from a randomized clinical trial. Anesth Analg. Advance Access published on May 9, 2018, doi:10.1213/ANE.0000000000003418 [DOI] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.