Free
Perioperative Medicine  |   April 2016
Cerebral Oximetry Monitoring to Maintain Normal Cerebral Oxygen Saturation during High-risk Cardiac Surgery: A Randomized Controlled Feasibility Trial
Author Notes
  • This article is featured in “This Month in Anesthesiology,” page 1A.
    This article is featured in “This Month in Anesthesiology,” page 1A.×
  • All authors have made substantial contributions to conception and design of the study, as well as acquisition and interpretation of data. A. Deschamps conducted the analysis and wrote the first draft of the manuscript. All authors took part in manuscript revision and have given final approval for the submitted version.
    All authors have made substantial contributions to conception and design of the study, as well as acquisition and interpretation of data. A. Deschamps conducted the analysis and wrote the first draft of the manuscript. All authors took part in manuscript revision and have given final approval for the submitted version.×
  • *Members of Canadian Perioperative Anesthesia Clinical Trials Group are listed in the 1.
    Members of Canadian Perioperative Anesthesia Clinical Trials Group are listed in the 1.×
  • Submitted for publication July 2, 2015. Accepted for publication December 4, 2015.
    Submitted for publication July 2, 2015. Accepted for publication December 4, 2015.×
  • Address correspondence to Dr. Deschamps: Department of Anesthesiology, Montreal Heart Institute, 5000 rue Bélanger, Montreal, Quebec, Canada H1T 1C8. a.deschamps@umontreal.ca. This article may be accessed for personal use at no charge through the Journal Web site, www.anesthesiology.org.
Article Information
Perioperative Medicine / Clinical Science / Cardiovascular Anesthesia / Central and Peripheral Nervous Systems / Neurosurgical Anesthesia / Respiratory System
Perioperative Medicine   |   April 2016
Cerebral Oximetry Monitoring to Maintain Normal Cerebral Oxygen Saturation during High-risk Cardiac Surgery: A Randomized Controlled Feasibility Trial
Anesthesiology 4 2016, Vol.124, 826-836. doi:10.1097/ALN.0000000000001029
Anesthesiology 4 2016, Vol.124, 826-836. doi:10.1097/ALN.0000000000001029
Abstract

Background: Cerebral oxygen desaturation during cardiac surgery has been associated with adverse perioperative outcomes. Before a large multicenter randomized controlled trial (RCT) on the impact of preventing desaturations on perioperative outcomes, the authors undertook a randomized prospective, parallel-arm, multicenter feasibility RCT to determine whether an intervention algorithm could prevent desaturations.

Methods: Eight Canadian sites randomized 201 patients between April 2012 and October 2013. The primary outcome was the success rate of reversing cerebral desaturations below 10% relative to baseline in the intervention group. Anesthesiologists were blinded to the cerebral saturation values in the control group. Intensive care unit personnel were blinded to cerebral saturation values for both groups. Secondary outcomes included the area under the curve of cerebral desaturation load, enrolment rates, and a 30-day follow-up for adverse events.

Results: Cerebral desaturations occurred in 71 (70%) of the 102 intervention group patients and 56 (57%) of the 99 control group patients (P = 0.04). Reversal was successful in 69 (97%) of the intervention group patients. The mean cerebral desaturation load (SD) in the operating room was smaller for intervention group patients compared with control group patients (104 [217] %.min vs. 398 [869] %.min, mean difference, −294; 95% CI, −562 to −26; P = 0.03). This was also true in the intensive care unit (P = 0.02). There were no differences in adverse events between the groups.

Conclusions: Study sites were successful in reversal of desaturation, patient recruitment, randomization, and follow-up in cardiac surgery, supporting the feasibility of conducting a large multicenter RCT.

Abstract

Eight Canadian centers randomized 201 patients. Cerebral desaturation (10% relative reduction from baseline) was common and nearly always reversible. Consequently, saturation was far better preserved in patients randomized to intervention than routine care.

What We Already Know about This Topic
  • Cerebral oxygen desaturation during cardiac surgery is associated with adverse perioperative outcomes

  • The authors, therefore, conducted a randomized feasibility trial to determine whether an intervention algorithm can prevent cerebral desaturation

What This Article Tells Us That Is New
  • Eight Canadian centers randomized 201 patients

  • Cerebral desaturation (10% relative reduction from baseline) was common and nearly always reversible

  • Consequently, saturation was far better preserved in patients randomized to intervention than routine care

NEAR infrared-reflected spectroscopy (NIRS) has been used extensively over the past two decades as a noninvasive continuous monitor of the balance between cerebral oxygen delivery and consumption.1,2  Originally used as a monitor in both adult1  and pediatric3  cardiac surgery, it has also been studied in a wide range of clinical settings including neurology,4  neurosurgery,5  vascular surgery,6  and trauma.7  Multiple reports have outlined the usefulness of NIRS in preventing potentially catastrophic neurologic events that would otherwise have been undetected with other conventional monitoring.1,8 
Cardiac surgery patients with significant decreases in regional cerebral oxygen saturation (rSO2) from baseline values have been shown to be at increased risk of postoperative cognitive decline,9–11  delirium,12  longer intensive care unit (ICU) and hospital length of stay,10,13  and an increase in overall major organ dysfunction.13  These findings have led clinicians to use strategies to reverse decreases in rSO2 during cardiac surgery to enhance both cerebral oxygenation and global tissue perfusion in an attempt to reduce the incidence of postoperative complications.14  However, the few interventional studies that have investigated the impact of preventing intraoperative decreases in rSO2 on postoperative outcomes have had mixed results.10,13,15  In major vascular surgery, preventing cerebral desaturations resulted in shorter recovery room and hospital length of stay.16  In cardiac surgery, there was a decrease in major organ dysfunction and shorter ICU length of stay.13  However, poor compliance with the protocolized interventions limited the interpretation of outcomes in another study.10 
To standardize strategies to reverse decreases in rSO2 during cardiac surgery, we developed a physiologic algorithm8  that we used in a small (n = 48) single-center randomized interventional trial. We showed reversal of desaturations in 93% of patients and an 80% reduction in cerebral desaturation load (CDL, defined as the cumulative area under the curve of desaturation over time for decreases in rSO2 values below 20% relative to baseline) in the group with interventions.17  However, the study was underpowered to examine perioperative outcomes. Accordingly, to determine the impact of reversing decreases in rSO2 on postoperative outcomes, a large multicenter randomized controlled trial (RCT) is needed. However, before such a trial can take place, it is necessary to determine the feasibility in potential participating sites of successfully using the algorithmic approach to prevent decreases in rSO2.
The purpose of this study was to examine the feasibility of undertaking such a trial. Our primary hypothesis was that academic centers participating in the NORMal cerebral Oxygen SATuration (NORMOSAT) in high-risk cardiac surgery trial will have at least an 80% success rate in reversing decreases in rSO2 in patients during high-risk cardiac surgery by implementing the same algorithmic strategy. We also wished to demonstrate that participating centers would also be able to recruit and follow-up a sufficient number of patients so as to make a future large-scale RCT feasible.
Materials and Methods
Study Design
The NORMOSAT study is a prospective, multicenter, parallel-arm, and RCT endorsed by the Canadian Perioperative Anesthesia Clinical Trials group.18  The study was conducted between April 2012 and October 2013 in compliance with the Declaration of Helsinki and the International Conference on Harmonization and on Guidelines for Good Clinical Practice. Institutional Research Ethics Board approval was obtained (see appendix 2) in every participating center, and a written informed consent was obtained from every patient. The trial was registered at clinicaltrials.gov (NCT01432184) in August 2011, with Dr. Deschamps as the principal investigator, and the overall design and patient flow is illustrated in figure 1.
Fig. 1.
Flow chart of the Normal Cerebral Oxygen Saturation study. Two hundred thirty patients were recruited, and 29 refused. Randomization was done on 201 patients, 99 in the control groups and 102 in the intervention group. All 201 patients completed the study. ICU = intensive care unit.
Image Not Available
Fig. 1.
Flow chart of the Normal Cerebral Oxygen Saturation study. Two hundred thirty patients were recruited, and 29 refused. Randomization was done on 201 patients, 99 in the control groups and 102 in the intervention group. All 201 patients completed the study. ICU = intensive care unit.
×
Patients were eligible to participate in the NORMOSAT trial if they were 18 yr or older, had a cumulative EuroScore II19  more than or equal to 10, and/or were undergoing high-risk surgery defined as combined surgery (coronary bypass plus valve replacement or repair), or multiple valve replacement and/or redo surgery. Patients were excluded if they were unable to read French or English, were undergoing off-pump coronary artery bypass surgery, emergency surgery (i.e., less than 6 h after diagnosis) or planned deep hypothermic circulatory arrest, acute endocarditis, or the presence of active delirium or encephalopathy.
Patients were randomly assigned to either a control group or an intervention group. Designated research coordinators from each site logged onto the secure study website of the coordinating central office where randomization was performed in a 1:1 ratio using a computer-generated random number table with permuted random blocks stratified by hospital sites. Research technicians recruited patients the day before the surgery, but to avoid enrolling patients who subsequently did not undergo surgery (e.g., surgical cancellation), randomization took place on the day of surgery just before the induction of anesthesia. Once randomized, patients were followed up and analyzed within the group to which they were allocated regardless of whether they receive the assigned intervention or not (i.e., intention-to-treat principle).
All patients had bilateral NIRS probes applied to the forehead before the induction of anesthesia. Research sites used one of the three Health Canada approved rSO2 monitoring devices for the study: FORE-SIGHT (CAS Medical Systems Inc., USA; 1 site, 9% of patients), EQUANOX Classic 7600 (Nonin Medical Inc., USA; M, 2 sites, 28% of patients), and INVOS 5100C-PB (Covidien, USA; 5 sites, 62% of patients). NIRS values were obtained from the patients while breathing room air as well as on supplemental oxygen (4 l/min O2via nasal prongs). The rSO2 values with supplemental oxygen were used as the baseline values for the study. Once baseline measures were obtained, rSO2 values were recorded continually and saved on the devices’ memory and then exported to Excel files that were then transferred to a secure online databank for analysis by the coordinating center.
Patients allocated to the control group had cerebral oximetry probes applied to the forehead but did not have NIRS values displayed on the monitor (i.e., anesthesiologists were blinded to rSO2 values). Anesthesiologists relied on standard monitoring for the management of these cases.
Patients allocated to the intervention group had NIRS values displayed on the monitor. At an intervention threshold of a 10% decrease in rSO2 value relative to baseline for a duration exceeding 15 s, anesthesiologists used an interventional algorithm (fig. 2) to reverse desaturations.8  The 10% threshold level was chosen so as to intervene as early as possible to maintain rSO2 at baseline values throughout the surgery and avoid an a priori defined clinically significant decreases in rSO2 value below 20% relative to baseline.13,17  Every desaturation occurrence along with its intervention (success or failure) was noted.
Fig. 2.
Physiologic algorithmic approach to the reversal of decreases in cerebral oxygen saturation. Step 1: Verify head position and position of vascular catheters. Step 2: Adjust blood pressure to within 20% of baseline values. Step 3: Increase inspired oxygen fraction to correct low oxygen saturation values. Step 4: Adjust ventilator parameters to increase arterial end tidal carbon dioxide levels within normal ranges (35 to 45 mmHg). Step 5: Consider giving blood transfusions to patients with low hemoglobin level when steps 1 to 4 have not increased cerebral oxygen saturation within 20% of baseline values. Step 6: Consider improving ventricular function in the presence of low output state. Step 7: Rule out hypothermia, seizures, and light anesthesia. Step 8: Consider cerebral edema and increased intracranial pressure. CT Scan = computed tomography scan; Hb = hemoglobin; ICHT = IntraCranial HyperTension; MAP = mean arterial pressure; MRI = magnetic resonance imaging; PaCO2 = arterial carbon dioxide tension; SaO2 = arterial oxygen saturation; SvO2 = mixed venous oxygen saturation.
Image Not Available
Fig. 2.
Physiologic algorithmic approach to the reversal of decreases in cerebral oxygen saturation. Step 1: Verify head position and position of vascular catheters. Step 2: Adjust blood pressure to within 20% of baseline values. Step 3: Increase inspired oxygen fraction to correct low oxygen saturation values. Step 4: Adjust ventilator parameters to increase arterial end tidal carbon dioxide levels within normal ranges (35 to 45 mmHg). Step 5: Consider giving blood transfusions to patients with low hemoglobin level when steps 1 to 4 have not increased cerebral oxygen saturation within 20% of baseline values. Step 6: Consider improving ventricular function in the presence of low output state. Step 7: Rule out hypothermia, seizures, and light anesthesia. Step 8: Consider cerebral edema and increased intracranial pressure. CT Scan = computed tomography scan; Hb = hemoglobin; ICHT = IntraCranial HyperTension; MAP = mean arterial pressure; MRI = magnetic resonance imaging; PaCO2 = arterial carbon dioxide tension; SaO2 = arterial oxygen saturation; SvO2 = mixed venous oxygen saturation.
×
The anesthetic management of patients was in accordance with Canadian guidelines,20  whereas surgical and perfusion techniques including cardiopulmonary bypass as well as the postoperative care were left to the discretion of the clinical caregivers from each individual center. After surgery, all patients were transferred intubated to the ICU with NIRS sensors in place. In the ICU, rSO2 was monitored continuously for 12 h (or until tracheal extubation whichever came first) with the monitor displays hidden from the ICU staff (blinded to group assignment).
Outcome Measures
Daily assessment of patients for postoperative complications was recorded by each site’s research coordinator until patients were discharged from the hospital or for 30 days whichever came first. (Table 1 lists the definitions used for outcome variables.) Discharged patients were contacted by telephone on the 30th postoperative day for a final follow-up interview. The primary endpoint of the study was the success rate of reversing decreases in rSO2 (below 10% of baseline values) in patients undergoing high-risk cardiac surgery with cardiopulmonary bypass.
Table 1.
Definition of Clinical Outcomes
Definition of Clinical Outcomes×
Image Not Available
Table 1.
Definition of Clinical Outcomes
Definition of Clinical Outcomes×
×
The principal secondary endpoints of the NORMOSAT trial included the enrolment rate (defined as the number of patients enrolled per week), the between-group difference CDL (%.min), the between-group difference in major organ morbidity and mortality (MOMM) score, a composite endpoint of stroke, renal failure requiring dialysis, prolonged mechanical ventilation more than 48 h, deep sternal wound infection, reoperation, and death. Other predefined endpoints examined included the rate of 30-day follow-up, defined as the proportion of all randomized patients who participated in the 30-day follow-up interview and the incidence of decreases in rSO2 more than 20% of baseline in the first 12 h in the ICU (or until tracheal extubation).
For the NORMOSAT study to be considered feasible, we determined that every center would need to be able to reverse decreases in rSO2 in at least 80% of patients, and there would need to be at least two patients recruited per month at each site for a total of 200 patients in a 12-month period.
Statistical Analysis
Results are presented as means (SD) for continuous variables and as frequencies (percentages) for categorical variables. Groups were compared using a Student’s t test for continuous variables and Pearson chi-square test for categorical variables. As this was a feasibility trail, no corrections (e.g., Bonferroni) were used for controlling multiplicity of testing of the secondary endpoints. The sample size for our feasibility endpoint was based on convenience, and no a priori calculation was conducted. Statistical analyses were performed using SPSS, version 21.0 (SPSS Inc., USA). Significance level was set at 0.05 for all tests.
Results
Eight centers participated in the study, and 201 patients were recruited. Although recruitment was planned for 1 yr, the total number of patients was recruited over 18 months due to delays in research ethics review board approval (see appendix 2). However, once recruitment began, all centers recruited at the expected recruitment rate. The distribution of patient recruitment between sites and rates of enrolment are shown in table 2. The average enrolment rates across sites varied from 2.1 to 3.3 patients per month. Randomization resulted in 99 patients in the control group and 102 patients in the intervention group. Before study closure, one center included one more patient than required into the intervention group. One patient randomized to the control group was treated as in the intervention group in error. All patients were included for analysis in the group to which they were randomized.
Table 2.
Number of Patients Recruited by Participating Centers, Number of Months Centers Participated, and the Rate of Enrolment
Number of Patients Recruited by Participating Centers, Number of Months Centers Participated, and the Rate of Enrolment×
Image Not Available
Table 2.
Number of Patients Recruited by Participating Centers, Number of Months Centers Participated, and the Rate of Enrolment
Number of Patients Recruited by Participating Centers, Number of Months Centers Participated, and the Rate of Enrolment×
×
Demographic data, medical conditions and perioperative risk estimates, and perioperative data are presented in table 3. The rSO2 data are presented in table 4. Baseline rSO2 values without supplemental oxygen did not differ between the control and intervention groups. With supplemental oxygen, there was an overall statistically significant increase in rSO2 values in both groups (P < 0.001).
Table 3.
Demographic Data, Medical Conditions, Perioperative Risk Estimates, and Perioperative Data
Demographic Data, Medical Conditions, Perioperative Risk Estimates, and Perioperative Data×
Image Not Available
Table 3.
Demographic Data, Medical Conditions, Perioperative Risk Estimates, and Perioperative Data
Demographic Data, Medical Conditions, Perioperative Risk Estimates, and Perioperative Data×
×
Table 4.
Cerebral Desaturation Data in the Operating Room
Cerebral Desaturation Data in the Operating Room×
Image Not Available
Table 4.
Cerebral Desaturation Data in the Operating Room
Cerebral Desaturation Data in the Operating Room×
×
Cerebral desaturation below 10% relative to baseline occurred in 71 of the 102 intervention group patients (69.6%) and 56 of the 99 control group patients (56.6%, P = 0.04). In the intervention group, the total number of interventions was 535, with a mode of 5 interventions per patient (range, 1 to 91, 25 to 75% percentile, 2 to 8). The total number of patients with decreases in rSO2 below 10% was 127, and the total number of desaturations was 979. Of these, 23 patients had more than one desaturation, 12 patients had more than 3 desaturations, and 3 patients had more than 10 desaturations during the course of surgery. Desaturations were successfully reversed in 69 (97%) of the intervention group patients. The success rate of reversing any desaturation below 10% was 63.5% (219 of 345 desaturations).
Interventions at the 10% threshold level appear to have prevented progression of desaturations below 20% relative to baseline (defined as a clinically significant desaturation).17  Only 34 of 71 (48%) patients in the intervention group had desaturations below 20% of baseline as opposed to 46 of 56 (82%) patients in the control group (P < 0.0001). Also, the total number of desaturations below 20% of baseline was significantly less in the intervention group (134) compared with the control group (300, P = 0.012). The mean (SD) CDL in the operating room (OR) was 104 (217) %.min in the intervention group compared with 398 (869) %.min in the control group (P = 0.03; table 4).
The average number of decreases in rSO2 below 20% relative to baseline in the different participating centers was 2.0 ± 1.0 (range, 0.6 to 3.7) desaturations/patient recruited, and an average of 37 ± 16% (range, 13 to 63%) of patient recruited had cerebral desaturations at the 20% level. The number of decrease in rSO2 below 20% with the different NIRS devices was 1.5, 1.7, and 2.3 desaturations/patient recruited with the FORE-SIGHT, EQUANOX Classic 7600, and INVOS 5100C-PB, respectively.
The data for decreases in rSO2 in the ICU are shown in table 5. In total, there were 1,120 episodes of rSO2 below 20% relative to baseline in 66 patients in the ICU. Thirty-four of these patients were in the control group and account for 70% of all desaturations in the ICU. The remaining 32 patients in the intervention group accounted for 30% of all desaturations in the ICU (P = 0.08). All 66 patients who had decreases in rSO2 below 20% of baseline in the ICU also had desaturations below 10% in the OR. Of the patients who had cerebral desaturations below 20% of baseline in the OR, 39 (49%) also had decreases in rSO2 below 20% of baseline in the ICU (23 the control group and 16 in the intervention group, P = 0.79). The CDL in the ICU was 58% smaller in the intervention group, 454 (870) %.min compared with the control group, 1,070 (1,961) %.min (P = 0.02; table 5).
Table 5.
ICU Cerebral Saturation Data
ICU Cerebral Saturation Data×
Image Not Available
Table 5.
ICU Cerebral Saturation Data
ICU Cerebral Saturation Data×
×
Follow-up to 30 days was successful in all 201 patients in the study. The time on mechanical ventilation, ICU length of stay, hospital length of stay, MOMM score, and other adverse events are shown in table 6. No significant differences between groups were found. The impact of interventions to reverse decreases in rSO2 on the components of the composite index, MOMM, is shown in table 7.
Table 6.
Time on Mechanical Ventilation, ICU Length of Stay, Hospital Length of Stay, MOMM Score, and Other Adverse Events at 30 Days Postcardiac Surgery for Patients Who Did Not Have Cerebral Desaturations (No Desaturation) and for Patients in the Control Group and the Intervention Group Who Had Cerebral Desaturations
Time on Mechanical Ventilation, ICU Length of Stay, Hospital Length of Stay, MOMM Score, and Other Adverse Events at 30 Days Postcardiac Surgery for Patients Who Did Not Have Cerebral Desaturations (No Desaturation) and for Patients in the Control Group and the Intervention Group Who Had Cerebral Desaturations×
Image Not Available
Table 6.
Time on Mechanical Ventilation, ICU Length of Stay, Hospital Length of Stay, MOMM Score, and Other Adverse Events at 30 Days Postcardiac Surgery for Patients Who Did Not Have Cerebral Desaturations (No Desaturation) and for Patients in the Control Group and the Intervention Group Who Had Cerebral Desaturations
Time on Mechanical Ventilation, ICU Length of Stay, Hospital Length of Stay, MOMM Score, and Other Adverse Events at 30 Days Postcardiac Surgery for Patients Who Did Not Have Cerebral Desaturations (No Desaturation) and for Patients in the Control Group and the Intervention Group Who Had Cerebral Desaturations×
×
Table 7.
Impact of Interventions to Reverse Cerebral Desaturations on the Components of the Major Organ Morbidity and Mortality Composite Endpoint
Impact of Interventions to Reverse Cerebral Desaturations on the Components of the Major Organ Morbidity and Mortality Composite Endpoint×
Image Not Available
Table 7.
Impact of Interventions to Reverse Cerebral Desaturations on the Components of the Major Organ Morbidity and Mortality Composite Endpoint
Impact of Interventions to Reverse Cerebral Desaturations on the Components of the Major Organ Morbidity and Mortality Composite Endpoint×
×
Discussion
The primary goal of the NORMOSAT trial was to determine whether decreases in rSO2 in cardiac surgery patients could be prevented using a NIRS-guided intervention algorithm in the setting of a multicenter prospective randomized study. Our results confirm the feasibility of reversing decrease in rSO2 below 10% baseline in patients in all sites included in the trial. Although 63% of all patients had decreases in rSO2 below 10% baseline in the trial, only 40% of the patients progressed to desaturations below 20% of baseline mainly because fewer patients continued to desaturate in the intervention group. Only 48% of patients progressed to desaturations below 20% of baseline in the intervention group as opposed to 82% of patients in the control group. These data confirm the feasibility of using early interventions to prevent further cerebral desaturations in this patient population.
This trial also further confirms that interventions to reverse decreases in rSO2 also result in a reduction in the overall severity of desaturation as defined by the CDL. Because prolonged low rSO2 values during cardiac surgery have been associated with significantly higher risks of early postoperative cognitive decline,10  the reduction in CDL seen in the intervention group could potentially lead to a reduction in overall complications in an appropriately powered trial.
In addition to the primary feasibility endpoint, this study also demonstrated the feasibility of conducting a multicenter RCT in several Canadian cardiac anesthesia sites using an algorithm for decision-making and treatment in the OR. Specifically, the transfer of data from research sites to the coordinating center through the NORMOSAT website was easily achieved. Supervision of the application of the algorithm in the OR ensured that all attempts were made to reverse decreases in rSO2, and follow-up of patients for 30 days postsurgery was successful in all centers. The quality of the cerebral saturation data collected did not appear to be influenced by the type of device used, although this needs to be confirmed with more sites using the different devices. Our results suggest that it would be possible to conduct a large-scale randomized clinical trial with a similar design to examine the effect of preventing decreases in rSO2 on meaningful clinical outcomes such as a composite outcome measure consisting of cardiovascular mortality, new-onset neurologic events, atrial fibrillation, and postoperative myocardial infarction. Previous studies have shown some benefits of reversing cerebral saturation during cardiac surgery on outcomes10–15  but not in a high-risk population of patients for whom NIRS is most often used. A large-scale RCT would clarify the benefit of reversing cerebral desaturations in this population of patients.
As previously described,17  a large proportion of patients (75%) also had desaturations in the ICU. Interestingly, patients without desaturations in the OR continued to be free of desaturations in the ICU. Also, there was a significant decrease in the CDL of patients in the intervention group compared with the control group in the ICU. Because no interventions were made in the ICU to reverse decreases in rSO2, these results suggest that reversing cerebral desaturations in the OR might protect patients by reducing the severity of decreases in rSO2 in the ICU. The effect of combining interventions in the OR and in the ICU on the CDL should be the subject of further investigation.
There were several limitations in our study. For example, it was not powered to measure the impact of reversal of decreases in rSO2 in the OR on perioperative adverse events. Patients who had a low baseline rSO2 have been observed to have poor outcome at 1 month and 1 yr after cardiac surgery.21  The same appears to be true for patients who do not increase their baseline rSO2 values with supplemental oxygen.21  We were unable to show an association between increased rSO2 values with supplemental oxygen and outcomes because 100% of our patients had increases in rSO2 values when supplemental oxygen was added in the OR. One possible reason may be the difference in sample size between the studies. We were also unable to find a relationship between postoperative delirium and low baseline rSO2 values or large decreases in intraoperative rSO2 values as others have.12  Our definition of delirium may be at the source of this difference because it was obtained from the charts and not from a systematic screening of the patients in the ICU. For most adverse events measured in this study, the number of events in patients with desaturations was very similar to the number of events in patients who did not have significant decreases in rSO2. Furthermore, successful interventions to reverse decreases in rSO2 had very little impact on reducing the occurrence of adverse events (tables 6 and 7). One reason may the lack of power of the feasibility trial, but we also have to consider the possibility that our patient population may have been so high risk that interventions to reverse decreases in rSO2 have very little impact on outcomes. We preferentially used a high-risk population of patients for two reasons. One, these complex surgeries in high-risk patients tend to have a greater incidence of technical complications resulting in catastrophic events better detected by NIRS than any other available device.1,8  Second, cardiac anesthesiologists tend to reserve the use of NIRS for this patient population believing that the cost–benefit ratio is greater than the use of NIRS in healthier patients.
As for sample size estimates for future outcomes studies, this feasibility trial suggests that 3,080 patients in total would be needed to detect a significant decrease in mortality, 4,638 patients for a decrease in stroke rate, and 1,610 patients for a decrease in the rate of delirium. However, the use of a composite endpoint as a primary outcome, other than MOMM because it was not shown to improve with interventions to reverse cerebral desaturations, could help to reduce the sample size in the large-scale RCT. A definitive RCT would give important information on outcomes and on the numbers needed to treat in this patient population. In parallel, a study could be undertaken to show a benefit of NIRS monitoring in a healthier population, but the number of patients with cerebral desaturations would decrease significantly, thus increasing the total number of patients to be studied.
In conclusion, in this multicenter prospective randomized study, we determined that the use of a physiologic algorithmic approach to management of intraoperative decrease in rSO2 was feasible. There was a high success rate with the algorithmic approach to the reversal of significant decreases in rSO2 in high-risk cardiac surgery patients in all participating centers. Early interventions in the OR had a significant impact on further progression of cerebral oxygen desaturations and on the overall cerebral oxygen desaturation load in these patients. Interventions to reverse decreases in rSO2 in the OR appear to protect patient from desaturations in the ICU. A large multicenter RCT trial would be needed to measure the impact of intraoperative reversal of decreases in rSO2 on clinical outcomes in high-risk cardiac surgery patients, and this study suggests that such a trial is feasible.
Acknowledgments
This study was funded in part by the Canadian Anesthesia Research Foundation (Toronto, Ontario, Canada); the Montreal Heart Institute Foundation (Montreal, Quebec, Canada); and the Anesthesiology Departments of the University of Manitoba (Winnipeg, Manitoba, Canada), Ottawa Heart Institute (Ottawa, Ontario, Canada), McMaster University (Hamilton, Ontario, Canada), University of Calgary (Calgary, Alberta, Canada), University of Alberta (Edmonton, Alberta, Canada), and the University of British Columbia (Vancouver, British Columbia, Canada).
Competing Interests
Dr. Deschamps has received speaking honoraria for educational seminars on the use of cerebral saturation monitoring in cardiac surgery patients sponsored by the companies Nonin Medical Inc., Plymouth, Minnesota, and Covidien Inc. (now a part of Medtronic), Boulder, Colorado. Dr. Denault has received speaking honoraria for educational seminars on the use of cerebral saturation monitoring in cardiac surgery patients sponsored by Covidien Inc. (now a part of Medtronic). NONIN Medical Inc. provided equipment and sensors for one of the centers implicated in the study. The other authors declare no competing interests.
Reproducible Science
Full protocol available from Dr. Deschamps: deschamps.a@gmail.com. Raw data available from Dr. Deschamps: deschamps.a@gmail.com.
References
Taillefer, MC, Denault, AY Cerebral near-infrared spectroscopy in adult heart surgery: Systematic review of its clinical efficacy.. Can J Anaesth. (2005). 52 79–87 [Article] [PubMed]
Edmonds, HLJr, Ganzel, BL, Austin, EHIII Cerebral oximetry for cardiac and vascular surgery.. Semin Cardiothorac Vasc Anesth. (2004). 8 147–66 [Article] [PubMed]
Kurth, CD, Steven, JL, Montenegro, LM, Watzman, HM, Gaynor, JW, Spray, TL, Nicolson, SC Cerebral oxygen saturation before congenital heart surgery.. Ann Thorac Surg. (2001). 72 187–92 [Article] [PubMed]
Sokol, DK, Markand, ON, Daly, EC, Luerssen, TG, Malkoff, MD Near infrared spectroscopy (NIRS) distinguishes seizure types.. Seizure. (2000). 9 323–7 [Article] [PubMed]
Shojima, M, Watanabe, E, Mayanagi, Y Cerebral blood oxygenation after cerebrospinal fluid removal in hydrocephalus measured by near infrared spectroscopy.. Surg Neurol. (2004). 62 312–8; discussion 318 [Article] [PubMed]
Vernieri, F, Tibuzzi, F, Pasqualetti, P, Rosato, N, Passarelli, F, Rossini, PM, Silvestrini, M Transcranial Doppler and near-infrared spectroscopy can evaluate the hemodynamic effect of carotid artery occlusion.. Stroke. (2004). 35 64–70 [Article] [PubMed]
Gracias, VH, Guillamondegui, OD, Stiefel, MF, Wilensky, EM, Bloom, S, Gupta, R, Pryor, JP, Reilly, PM, Leroux, PD, Schwab, CW Cerebral cortical oxygenation: A pilot study.. J Trauma. (2004). 56 469–72; discussion 472–4 [Article] [PubMed]
Denault, A, Deschamps, A, Murkin, JM A proposed algorithm for the intraoperative use of cerebral near-infrared spectroscopy.. Semin Cardiothorac Vasc Anesth. (2007). 11 274–81 [PubMed]
de Tournay-Jetté, 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 [Article] [PubMed]
Slater, JP, Guarino, T, Stack, J, Vinod, K, Bustami, RT, Brown, JMIII, Rodriguez, AL, Magovern, CJ, Zaubler, T, Freundlich, K, Parr, GV Cerebral oxygen desaturation predicts cognitive decline and longer hospital stay after cardiac surgery.. Ann Thorac Surg. (2009). 87 36–44; discussion 44–5 [Article] [PubMed]
Yao, FS, Tseng, CC, Ho, CY, Levin, SK, Illner, P Cerebral oxygen desaturation is associated with early postoperative neuropsychological dysfunction in patients undergoing cardiac surgery.. J Cardiothorac Vasc Anesth. (2004). 18 552–8 [Article] [PubMed]
Schoen, J, Meyerrose, J, Paarmann, H, Heringlake, M, Hueppe, M, Berger, KU Preoperative regional cerebral oxygen saturation is a predictor of postoperative delirium in on-pump cardiac surgery patients: A prospective observational trial.. Crit Care. (2011). 15 R218 [Article] [PubMed]
Murkin, JM, Adams, SJ, Novick, RJ, Quantz, M, Bainbridge, D, Iglesias, I, Cleland, A, Schaefer, B, Irwin, B, Fox, S Monitoring brain oxygen saturation during coronary bypass surgery: A randomized, prospective study.. Anesth Analg. (2007). 104 51–8 [Article] [PubMed]
Murkin, JM Cerebral oximetry: Monitoring the brain as the index organ.. Anesthesiology. (2011). 114 12–3 [Article] [PubMed]
Casati, A, Fanelli, G, Pietropaoli, P, Proietti, R, Tufano, R, Montanini, S, Danelli, G, Nuzzi, M, Mentegazzi, F, Torri, G, Martani, C, Spreafico, E, Fierro, G, Pugliese, F, De Cosmo, G, Aceto, P, Servillo, G, Monaco, F Collaborative Italian Study Group on Anaesthesia in Elderly Patients, Monitoring cerebral oxygen saturation in elderly patients undergoing general abdominal surgery: A prospective cohort study.. Eur J Anaesthesiol. (2007). 24 59–65 [PubMed]
Casati, A, Fanelli, G, Pietropaoli, P, Proietti, R, Tufano, R, Danelli, G, Fierro, G, Fierro, G, De Cosmo, G, Servillo, G Continuous monitoring of cerebral oxygen saturation in elderly patients undergoing major abdominal surgery minimizes brain exposure to potential hypoxia.. Anesth Analg. (2005). 101 740–7 [Article] [PubMed]
Deschamps, A, Lambert, J, Couture, P, Rochon, A, Lebon, JS, Ayoub, C, Cogan, J, Denault, A Reversal of decreases in cerebral saturation in high-risk cardiac surgery.. J Cardiothorac Vasc Anesth. (2013). 27 1260–6 [Article] [PubMed]
Hall, R, Beattie, S, Cheng, D, Choi, P, Denault, AY, Mazer, D, Mutch, WA, Turgeon, AF, Yang, H Can we develop a Canadian Perioperative Anesthesiology Clinical Trials group?. Can J Anaesth. (2010). 57 1051–7 [Article] [PubMed]
Nashef, SA, Roques, F, Sharples, LD, Nilsson, J, Smith, C, Goldstone, AR, Lockowandt, U EuroSCORE II.. Eur J Cardiothorac Surg. (2012). 41 734–44; discussion 744–5 [Article] [PubMed]
Merchant, R, Bosenberg, C, Brown, K, Chartrand, D, Dain, S, Dobson, J, Kurrek, M, LeDez, K, Morgan, P, Penner, M, Shukla, R Guidelines to the Practice of Anesthesia Revised Edition 2010.. Can J Anaesth. (2010). 57 58–87 [Article] [PubMed]
Heringlake, M, Garbers, C, Käbler, JH, Anderson, I, Heinze, H, Schön, J, Berger, KU, Dibbelt, L, Sievers, HH, Hanke, T Preoperative cerebral oxygen saturation and clinical outcomes in cardiac surgery.. Anesthesiology. (2011). 114 58–69 [Article] [PubMed]
Appendix 1: Perioperative Anesthesia Clinical Trials Group of Canada
The members of the group are as follows:
Alain Deschamps, Ph.D., M.D., Montreal Heart Institute, Montreal, Québec
Alan Mutch, M.D., University of Manitoba, Winnipeg, Manitoba
Alexis Turgeon, M.D., Université Laval, Québec, Québec
Andre Denault, M.D., Ph.D., Montreal Heart Institute, Montreal, Québec
Andrea Todd, M.D., University of Saskatchewan, Saskatoon, Saskatchewan
Angela Jerath, M.D., University of Toronto, Toronto, Ontario
Ashraf Fayad, M.D., University of Ottawa, Ottawa, Ontario
Barry Finnegan, M.D., University of Manitoba, Winnipeg, Manitoba
Blaine Kent, M.D., Dalhousie University, Halifax, Nova Scotia
Brent Kennedy, M.D., Northern School of Medicine, Sudbury, Ontario
Brian H. Cuthbertson, M.D., University of Toronto, Toronto, Ontario
Brian Kavanagh, M.D., University of Toronto, Toronto, Ontario
Brian Warriner, University of British Columbia, Vancouver, British Columbia
Charles MacAdams, University of Calgary, Calgary, Alberta
Christian Lehmann, M.D., Dalhousie University, Halifax, Nova Scotia
Christine Fudorow, M.D., University of Manitoba, Winnipeg, Manitoba
Christopher Hudson, M.D., University of Ottawa, Ottawa, Ontario
Colin McCartney, M.D., University of Ottawa, Ottawa, Ontario
Dan McIsaac, M.D., University of Ottawa, Ottawa, Ontario
Daniel Dubois, M.D., University of Ottawa, Ottawa, Ontario
David Campbell, M.D., University of Saskatchewan, Saskatoon, Saskatchewan
David Mazer, M.D., University of Toronto, Toronto, Ontario
David Neilpovitz, M.D., University of Ottawa, Ottawa, Ontario
David Rosen, M.D., University of Ottawa, Ottawa, Ontario
Davy Cheng, M.D., Western University, London, Ontario
Dennis Drapeau, M.D., Dalhousie University, Halifax, Nova Scotia
Derek Dillane, M.D., University of Alberta, Edmonton, Alberta
Diem Tran, M.D., University of Ottawa, Ottawa, Ontario
Dolores Mckeen, M.D., Dalhousie University, Halifax, Nova Scotia
Duminda Wijeysundera, M.D., University of Toronto, Toronto, Ontario
Eric Jacobsohn, M.D., University of Manitoba, Winnipeg, Manitoba
Etienne Couture, M.D., Université Laval, Québec, Québec
Etienne de Medicis, M.D., Université de Sherbrooke, Sherbrooke, Québec
Fahad Alam, M.D., University of Toronto, Toronto, Ontario
Faraj Abdallah, M.D., University of Toronto, Toronto, Ontario
Fiona E. Ralley, M.D., Western University, London, Ontario
Frances Chung, M.D., Western University, London, Ontario
Francois Lellouche, M.D., Ph.D., Université Laval, Québec, Québec
Gary Dobson, M.D., University of Calgary, Calgary, Alberta
Genevieve Germain, M.D., Université Laval, Québec, Québec
George Djaiani, M.D., University of Toronto, Toronto, Ontario
Ian Gilron, M.D., Queens University, Kingston, Ontario
Gregory Hare, M.D., University of Toronto, Toronto, Ontario
Gregory Bryson, M.D., University of Ottawa, Ottawa, Ontario
Hance Clarke, M.D., University of Toronto, Toronto, Ontario
Heather McDonald, M.D., University of Manitoba, Winnipeg, Manitoba
Helen Roman-Smith, M.D., University of Calgary, Calgary, Alberta
Hilary Grocott, M.D., University of Manitoba, Winnipeg, Manitoba
Homer Yang, M.D., University of Ottawa, Ottawa, Ontario
James Douketis, M.D., McMaster University, Hamilton, Ontario
James Paul, M.D., McMaster University, Hamilton, Ontario
Jean Beaubien, M.D., Université Laval, Québec, Québec
Jean Bussières, M.D., Université Laval, Québec, Québec
Jeremy Pridham, M.D., Memorial University, St-John’s, Newfoundland
J. N. Armstrong, M.D., University of Calgary, Calgary, Alberta
Joel Parlow, M.D., Queens University, Kingston, Ontario
John Murkin, M.D., Western University, London, Ontario
Jonathan Gamble, M.D., University of Saskatchewan, Saskatoon, Saskatchewan
Kaylene Duttchen, M.D., University of Calgary, Calgary, Alberta
Keyvan Karkouti, M.D., University of Toronto, Toronto, Ontario
Kim Turner, M.D., Queens University, Kingston, Ontario
Leyla Baghirzada, M.D., University of Calgary, Calgary, Alberta
Linda Szabo, M.D., Western University, London, Ontario
Manoj Lalu, University of Ottawa, Ottawa, Ontario
Marcin Wasowicz, M.D., University of Toronto, Toronto, Ontario
Michael Bautista, M.D., Memorial University, St-John’s, Newfoundland
Michael Jacka, M.D., University of Alberta, Edmonton, Alberta
Michael Murphy, University of Alberta, Edmonton, Alberta
Michael Schmidt, M.D. Ph.D., Dalhousie University, Halifax, Nova Scotia
Michaël Verret, M.D., Université Laval, Québec, Québec
Michel-Antoine Perrault, M.D., Université de Sherbrooke, Sherbrooke, Québec
Nicolas Beaudet, M.D., Université de Sherbrooke, Sherbrooke, Québec
Norman Buckley, M.D., McMaster University, Hamilton, Ontario
Peter Choi, M.D., University of British Columbia, Vancouver, British Columbia
Peter MacDougall, M.D. Ph.D., Dalhousie University, Halifax, Nova Scotia
Philip Jones, M.D., Western University, London, Ontario
Pierre Drolet, M.D., Université de Montréal, Montreal, Québec
Pierre Beaulieu, M.D., Université de Montréal, Montreal, Québec
Ravi Taneja, M.D., Western University, London, Ontario
Rene Martin, M.D., Université de Sherbrooke, Sherbrooke, Québec
Richard Hall, M.D., Dalhousie University, Halifax, Nova Scotia
Ronald George, M.D. Ph.D., Dalhousie University, Halifax, Nova Scotia
Rosa Chun, M.D., University of Calgary, Calgary, Alberta
Sarah McMullen, M.D. Ph.D., Dalhousie University, Halifax, Nova Scotia
Scott Beattie, M.D., University of Toronto, Toronto, Ontario
Sonia Sampson, Memorail University, St-John’s, Newfoundland
Stephen Choi, M.D., University of Toronto, Toronto, Ontario
Stephen Kowalski, M.D., University of Manitoba, Winnipeg, Manitoba
Stuart McCluskey, M.D., University of Toronto, Toronto, Ontario
Summer Syed, M.D., McMaster University, Hamilton, Ontario
Sylvain Boet, M.D., University of Ottawa, Ottawa, Ontario
Tim Ramsay, Ph.D., University of Ottawa, Ottawa, Ontario
Tarit Saha, M.D., Queens University, Kingston, Ontario
Thomas Mutter, M.D., University of Manitoba, Winnipeg, Manitoba
Tumul Chowdhury, M.D., University of Manitoba, Winnipeg, Manitoba
Vishal Uppal, M.D. Ph.D., Dalhousie University, Halifax, Nova Scotia
William Mckay, M.D., University of Saskatchewan, Saskatoon, Saskatchewan
Appendix 2: Research Ethics Review Board
Fig. 1.
Flow chart of the Normal Cerebral Oxygen Saturation study. Two hundred thirty patients were recruited, and 29 refused. Randomization was done on 201 patients, 99 in the control groups and 102 in the intervention group. All 201 patients completed the study. ICU = intensive care unit.
Image Not Available
Fig. 1.
Flow chart of the Normal Cerebral Oxygen Saturation study. Two hundred thirty patients were recruited, and 29 refused. Randomization was done on 201 patients, 99 in the control groups and 102 in the intervention group. All 201 patients completed the study. ICU = intensive care unit.
×
Fig. 2.
Physiologic algorithmic approach to the reversal of decreases in cerebral oxygen saturation. Step 1: Verify head position and position of vascular catheters. Step 2: Adjust blood pressure to within 20% of baseline values. Step 3: Increase inspired oxygen fraction to correct low oxygen saturation values. Step 4: Adjust ventilator parameters to increase arterial end tidal carbon dioxide levels within normal ranges (35 to 45 mmHg). Step 5: Consider giving blood transfusions to patients with low hemoglobin level when steps 1 to 4 have not increased cerebral oxygen saturation within 20% of baseline values. Step 6: Consider improving ventricular function in the presence of low output state. Step 7: Rule out hypothermia, seizures, and light anesthesia. Step 8: Consider cerebral edema and increased intracranial pressure. CT Scan = computed tomography scan; Hb = hemoglobin; ICHT = IntraCranial HyperTension; MAP = mean arterial pressure; MRI = magnetic resonance imaging; PaCO2 = arterial carbon dioxide tension; SaO2 = arterial oxygen saturation; SvO2 = mixed venous oxygen saturation.
Image Not Available
Fig. 2.
Physiologic algorithmic approach to the reversal of decreases in cerebral oxygen saturation. Step 1: Verify head position and position of vascular catheters. Step 2: Adjust blood pressure to within 20% of baseline values. Step 3: Increase inspired oxygen fraction to correct low oxygen saturation values. Step 4: Adjust ventilator parameters to increase arterial end tidal carbon dioxide levels within normal ranges (35 to 45 mmHg). Step 5: Consider giving blood transfusions to patients with low hemoglobin level when steps 1 to 4 have not increased cerebral oxygen saturation within 20% of baseline values. Step 6: Consider improving ventricular function in the presence of low output state. Step 7: Rule out hypothermia, seizures, and light anesthesia. Step 8: Consider cerebral edema and increased intracranial pressure. CT Scan = computed tomography scan; Hb = hemoglobin; ICHT = IntraCranial HyperTension; MAP = mean arterial pressure; MRI = magnetic resonance imaging; PaCO2 = arterial carbon dioxide tension; SaO2 = arterial oxygen saturation; SvO2 = mixed venous oxygen saturation.
×
Table 1.
Definition of Clinical Outcomes
Definition of Clinical Outcomes×
Image Not Available
Table 1.
Definition of Clinical Outcomes
Definition of Clinical Outcomes×
×
Table 2.
Number of Patients Recruited by Participating Centers, Number of Months Centers Participated, and the Rate of Enrolment
Number of Patients Recruited by Participating Centers, Number of Months Centers Participated, and the Rate of Enrolment×
Image Not Available
Table 2.
Number of Patients Recruited by Participating Centers, Number of Months Centers Participated, and the Rate of Enrolment
Number of Patients Recruited by Participating Centers, Number of Months Centers Participated, and the Rate of Enrolment×
×
Table 3.
Demographic Data, Medical Conditions, Perioperative Risk Estimates, and Perioperative Data
Demographic Data, Medical Conditions, Perioperative Risk Estimates, and Perioperative Data×
Image Not Available
Table 3.
Demographic Data, Medical Conditions, Perioperative Risk Estimates, and Perioperative Data
Demographic Data, Medical Conditions, Perioperative Risk Estimates, and Perioperative Data×
×
Table 4.
Cerebral Desaturation Data in the Operating Room
Cerebral Desaturation Data in the Operating Room×
Image Not Available
Table 4.
Cerebral Desaturation Data in the Operating Room
Cerebral Desaturation Data in the Operating Room×
×
Table 5.
ICU Cerebral Saturation Data
ICU Cerebral Saturation Data×
Image Not Available
Table 5.
ICU Cerebral Saturation Data
ICU Cerebral Saturation Data×
×
Table 6.
Time on Mechanical Ventilation, ICU Length of Stay, Hospital Length of Stay, MOMM Score, and Other Adverse Events at 30 Days Postcardiac Surgery for Patients Who Did Not Have Cerebral Desaturations (No Desaturation) and for Patients in the Control Group and the Intervention Group Who Had Cerebral Desaturations
Time on Mechanical Ventilation, ICU Length of Stay, Hospital Length of Stay, MOMM Score, and Other Adverse Events at 30 Days Postcardiac Surgery for Patients Who Did Not Have Cerebral Desaturations (No Desaturation) and for Patients in the Control Group and the Intervention Group Who Had Cerebral Desaturations×
Image Not Available
Table 6.
Time on Mechanical Ventilation, ICU Length of Stay, Hospital Length of Stay, MOMM Score, and Other Adverse Events at 30 Days Postcardiac Surgery for Patients Who Did Not Have Cerebral Desaturations (No Desaturation) and for Patients in the Control Group and the Intervention Group Who Had Cerebral Desaturations
Time on Mechanical Ventilation, ICU Length of Stay, Hospital Length of Stay, MOMM Score, and Other Adverse Events at 30 Days Postcardiac Surgery for Patients Who Did Not Have Cerebral Desaturations (No Desaturation) and for Patients in the Control Group and the Intervention Group Who Had Cerebral Desaturations×
×
Table 7.
Impact of Interventions to Reverse Cerebral Desaturations on the Components of the Major Organ Morbidity and Mortality Composite Endpoint
Impact of Interventions to Reverse Cerebral Desaturations on the Components of the Major Organ Morbidity and Mortality Composite Endpoint×
Image Not Available
Table 7.
Impact of Interventions to Reverse Cerebral Desaturations on the Components of the Major Organ Morbidity and Mortality Composite Endpoint
Impact of Interventions to Reverse Cerebral Desaturations on the Components of the Major Organ Morbidity and Mortality Composite Endpoint×
×