Cerebral tissue oxygen saturation values in volunteers and patients in the lateral decubitus and beach chair positions: a prospective observational study
- 1.1k Downloads
The objective of this study was to describe changes in cerebral tissue oxygen saturation (SctO2) due to changes in body position in healthy volunteers and in patients undergoing surgery under general anesthesia in the beach chair position (BCP) and lateral decubitus position (LDP).
In this prospective observational study, SctO2 was measured in 85 awake volunteers serially positioned every 15 min, beginning with the supine position (SP) and followed by the beach chair, supine, and lateral decubitus positions. Cerebral tissue oxygen saturation was also measured supine and in either the BCP or the LDP in 195 patients (according to surgical preference) undergoing elective arthroscopic shoulder surgery. We measured the lowest stable SctO2 values in each position as well as changes in blood pressure and heart rate.
In healthy volunteers, the median (interquartile range [IQR]) lowest stable SctO2 value in the SP was 69 [66-71] %. A change in position to the BCP caused a small but statistically significant decrease in the median [IQR] lowest SctO2 value to 67 [65-70] % (P = 0.028 compared with baseline). This decrease was associated with an increase in median [IQR] arterial pressure from 83 [78-88] mmHg in the SP to 85 [81-93] mmHg in the BCP (P < 0.001 compared with baseline). In patients undergoing surgery in the BCP, the median [IQR] lowest stable SctO2 value was 55 [51-59] %, which was significantly lower (P < 0.001) than the median [IQR] lowest SctO2 value in patients in the LDP (66 [62-69] %). More patients in the BCP group (57%) showed SctO2 values ≤ 55% and/or a decrease of ≥ 20% from baseline (57%) compared with the LDP group (5% and 6%, respectively; P < 0.001 for each comparison).
More than 55% of patients undergoing arthroscopic shoulder surgery in the BCP experience cerebral desaturation events. In volunteers without anesthesia, no desaturation events were observed. The clinical importance of these findings needs further investigation.
KeywordsMean Arterial Pressure Lateral Decubitus Position Beach Chair Position Control Hypotension Cerebral Oxygen Saturation
Valeurs de saturation en oxygène cérébral tissulaire de volontaires sains et de patients en décubitus latéral et en position assise: une étude observationnelle prospective
L’objectif de cette étude était de décrire les changements de la saturation en oxygène cérébral tissulaire (SctO2) provoqués par les changements de position du corps chez des volontaires sains et des patients subissant une chirurgie sous anesthésie générale en position assise (PAs - beach chair position) et en décubitus latéral (DL).
Dans cette étude observationnelle prospective, on a mesuré la SctO2 de 85 volontaires sains éveillés repositionnés en série toutes les 15 minutes, en commençant en position allongée (PAl) puis en position assise, puis allongée à nouveau, et finalement en décubitus latéral. La saturation en oxygène cérébral tissulaire a également été mesurée en position allongée et en PAs ou en DL chez 195 patients (selon la préférence chirurgicale) subissant une arthroscopie non urgente de l’épaule. Nous avons mesuré les valeurs stables de SctO2 les plus basses dans chaque position ainsi que les changements de tension artérielle et de fréquence cardiaque.
Chez les volontaires sains, la valeur stable moyenne (écart interquartile [ÉIQ]) la plus basse de SctO2 en PAl était 69 [66-71] %. Un changement en PAs a provoqué une réduction, faible mais significative d’un point de vue statistique, de la valeur médiane [ÉIQ] la plus basse de SctO2 à 67 [65-70] % (P = 0,028 comparativement aux valeurs de base). Cette réduction a été associée à une augmentation de la tension artérielle médiane [ÉIQ] de 83 [78-88] mmHg en PAl à 85 [81-93] mmHg en PAs (P < 0,001 comparativement aux valeurs de base). Chez les patients subissant une chirurgie en PAs, la valeur médiane [ÉIQ] stable la plus basse de SctO2 était de 55 [51-59], soit significativement plus basse (P < 0,001) que la valeur médiane [ÉIQ] stable la plus basse telle que mesurée chez les patients en DL (66 [62-69] %). La plupart des patients du groupe PAs (57 %) ont montré des valeurs de SctO2 ≤ 55 % et/ou une réduction ≥ 20 % par rapport aux valeurs de base (57 %) comparativement au groupe DL (5 % et 6 %, respectivement; P < 0,001 pour chaque comparaison).
Plus de 55 % des patients subissant une arthroscopie de l’épaule dans le groupe PAs ont subi des épisodes de désaturation cérébrale. Chez les volontaires sains n’ayant pas subi d’anesthésie, aucun épisode de désaturation n’a été observé. L’importance clinique de ces résultats nécessite des recherches plus approfondies.
Surgery in the beach chair position (BCP) has repeatedly been associated with significant hemodynamic changes that have the potential to compromise the cerebral circulation and place a patient at risk for cerebral ischemia.1,2 Moreover, for many surgical procedures (such as endoscopic shoulder surgery), optimal surgical visualization may require the use of controlled hypotension. Combining the BCP with controlled hypotension could increase the threat of inadequate cerebral perfusion. These assumptions are supported by case reports on adverse neurological sequelae after shoulder surgery in the BCP.3,4
Cerebral oximetry near-infrared spectroscopy (NIRS) permits the detection of changes in cerebral tissue oxygen saturation (SctO2). This technology has been extensively used to provide a noninvasive real-time indicator of cerebral hypoperfusion5 - mainly during cardiac surgery. Not surprisingly, several groups have reported their experience on the use of cerebral oximetry during shoulder surgery in the BCP, and every group revealed significant decreases in SctO2 during the procedure.6-13 Nevertheless, many questions remain about what constitutes a “normal” SctO2 value and what SctO2 values are acceptable during general anesthesia.
The first objective of this study was to measure the changes in SctO2 values in healthy volunteers after positioning in the supine, BCP, and lateral decubitus positions (LDP). Our second objective was to measure the changes in SctO2 values in patients after position changes during shoulder surgery under general anesthesia.
Cerebral tissue oxygen saturation monitoring
Bilateral SctO2 was measured using the FORE-SIGHTTM technology (CAS medical systems Inc, Branford, CT, USA) in both volunteers and patients. The FORE-SIGHT cerebral oximeter is a noninvasive device that uses four wavelengths of laser light to determine absolute cerebral oxygen saturation. Cerebral tissue oxygen saturation represents the oxygen saturation level in the microvasculature of brain tissue, which contains a mixture of arterial and venous blood (estimated to be 30/70%). Sensors were applied to each frontotemporal area and covered to prevent external light interference. For each body position, the lowest stable (i.e., displayed for at least ten seconds) SctO2 value (right or left side) was recorded and used for statistical analysis. In the case of shoulder surgery, the attending anesthesiologist was blinded to the SctO2 data, and therefore, any changes in SctO2 could not have resulted in any changes in therapy such as blood pressure management.
The study protocol was approved by the local committee (Ziekenhuis Oost-Limburg, Genk, Belgium) for medical ethics, and written informed consent was obtained from all participants (patients and volunteers).
Volunteers were healthy (American Society of Anesthesiologists physical status I) males and females 18-30 yr of age. Monitoring of baseline bilateral SctO2 values was started in the supine position. The position was then changed to the beach chair, supine, and the lateral decubitus positions for 15 min each. Apart from the bilateral SctO2 monitoring, heart rate (HR), pulse oximetry, and noninvasive blood pressure measurements were taken every five minutes.
Patients scheduled to undergo elective arthroscopic shoulder surgery under general anesthesia in the BCP or LDP were enrolled in the study. Exclusion criteria were pre-existing cerebrovascular disease, peripheral vascular disease, and age < 18 yr. Monitoring of baseline bilateral SctO2 values was started in the supine position and monitoring continued with the change to either the BCP or LDP. Allocation to either the BCP or the LDP group was determined by surgical preference.
Anesthetic management was performed in both patient groups by the same team of anesthesiologists. Thirty minutes prior to surgery, all patients received an interscalene plexus block with a single injection of a maximum of 30 mL of 0.5% bupivacaine. All surgical procedures were performed under general anesthesia (propofol/rocuronium/remifentanil; F i O2 30%). Standard monitoring (noninvasive blood pressure measurement on the contralateral arm, electrocardiography, peripheral oxygen saturation, and end-tidal capnography) was applied. The management of anesthesia and, when requested by the surgeon, controlled hypotension was left to the discretion of the attending anesthesiologist who was blinded to the cerebral oximetry values. In all patients, severe arterial hypotension [i.e., mean arterial pressure (MAP) < 60 mmHg] was countered by reducing the remifentanil or propofol infusions or by administering boluses of colloids, phenylephrine, or ephedrine.
Statistical analysis was performed using SPSS® V19.0 (SPSS Inc, Chicago, IL, USA). Discrete data were compared using the χ2 test. The normality of variable distributions was tested using the Kolmogorov-Smirnov test. Normally distributed continuous data were compared using the unpaired Student’s t test. Hemodynamic and cerebral oxygen saturation values were compared between groups using the Mann-Whitney U test. To compare values within the same group, the paired Wilcoxon test was used. Possible correlations between SctO2 and MAP were analyzed using the Pearson correlation test. The results are represented as median (interquartile range [IQR]), mean [standard deviation (SD)], or percent (%) as indicated. All reported P values are two sided.
Age, yr (SD)
ASA status, n (I/II/III)
Hypertension, n (%)
Results of healthy volunteers
There were 236 patients recruited for the study; however, 41 of the participants were excluded from further analyses because SctO2 and/or hemodynamic data were incompletely recorded or stored. Of the 195 remaining patients, 101 and 94 patients underwent surgery in the BCP and LDP, respectively. The two patient groups were comparable regarding sex, American Society of Anesthesiologists status, pre-existing hypertension, and preoperative hemoglobin levels. The age of patients in the BCP was significantly higher than the age of patients in the LDP (Table 1).
In patients undergoing elective arthroscopic shoulder surgery, the median [IQR] lowest stable SctO2 value in the supine position before induction of anesthesia (during pre-oxygenation) was used as baseline (72 [70-75] % and 73 [71-76] % for the BCP and LDP groups, respectively; P = 0.18).
Results of patients in beach chair and lateral decubitus position
After induction of anesthesia and change in position, MAP (measured at the level of the heart) decreased in each patient group, but this decrease was not significantly different between the two groups (P = 0.097). Nevertheless, assuming a decrease of 0.77 mmHg for every centimetre gradient14 and an approximate gradient of 22 cm (30° angle) between the external auditory meatus and the site of the blood pressure cuff, the median [IQR] decrease in MAP measured at the level of the brain was significantly more important in the BCP group than in the LDP group (61 [57-68] mmHg vs 43 [31-46] mmHg, respectively; P < 0.001).
For the patients in the BCP group, a correlation (R2 = 0.4) was observed between the decrease in MAP (measured at the level of the brain) and the decrease in SctO2.
The median [IQR] end-tidal CO2 (EtCO2) at the moment of the lowest stable SctO2 value was 31 [29-34] mmHg in the BCP group and 31 [28-34] mmHg in the LDP group (P = 0.34).
In this study, a very small decrease in SctO2 was observed in healthy volunteers when body position changed from the supine to the BCP. Only a very limited number of studies have been published on the impact of body positioning on SctO2 in healthy volunteers.15-18 Moreover, in these studies, the numbers of healthy volunteers studied were rather small (minimum, n = 5; maximum, n = 28) and different types of cerebral oximeters were used (NIRO™,15 Oxymon™,16 INVOS™.17,18 Generally, a decrease (although not significant in all studies) in SctO2 was reported when changing from the supine to the upright position, which is confirmed by the results from our current study.
In awake subjects, upright positioning activates the sympathetic nervous system, resulting in an increase in blood pressure.1 In our group of healthy volunteers, MAP increased from 83-85 mmHg when changing from the supine to the upright position (Table 2). This small increase in MAP was accompanied by a small decrease in SctO2.
In patients under general anesthesia, activation of the sympathetic nervous system to maintain cerebral perfusion in the sitting position is often blunted or absent.19 With 57% of our patients under general anesthesia revealing SctO2 values < 55% and a decrease of ≥ 20% from baseline in the BCP, we confirm previous results on the frequent occurrence of cerebral tissue oxygen desaturation in patients undergoing arthroscopic shoulder surgery in the BCP.6-11
The combination of general anesthesia and controlled hypotension resulted in a median decrease in blood pressure of 45 mmHg in both of our patient groups. Brachial noninvasive blood pressure monitoring, as in our study, overestimates the pressure in the elevated brain.13,14 When we apply the hydrostatic gradient14 to the blood pressure values of our patients, MAP would be approximately 42 mmHg in the brain of patients in the BCP compared with 59 mmHg measured at the level of the heart. The MAP at the top of the cortex could add another 6-9 mmHg to this gradient.20 This low cerebral MAP could help explain the lower SctO2 values in patients in the BCP compared with patients in the LDP. Nevertheless, the calculated R2 of 0.4 between a decrease in MAP and a decrease in SctO2 suggests that some patients with higher MAPs may have lower SctO2 values and vice versa, so factors other than MAP must contribute to SctO2.
To maintain arterial pressure in patients undergoing shoulder surgery in the BCP, phenylephrine and/or ephedrine are most commonly used. Meng et al. observed that SctO2 did not change when ephedrine was used but decreased (together with cardiac output) after the use of phenylephrine.21 Ogoh et al. recently showed that vasoconstriction in the extracranial vasculature could be responsible for this decrease in SctO2.22 To monitor the effect of vasopressor treatment on cerebral oxygenation, NIRS monitoring could be of further value.
In addition to the effects of the sitting position and hypotension in patients, ventilator strategies, such as hyperventilation, can influence cerebral blood flow and thus cerebral oxygenation. Recently, Murphy et al. observed a lower incidence of desaturation events and higher cerebral oxygenation values in a patient group with an EtCO2 of 40-42 mmHg compared with a group with an EtCO2 of 30-32 mmHg.7 We report EtCO2 values of 31 mmHg in both of our patient groups, which corresponds with the standard practice group of Murphy et al. with lower SctO2 values.
Finally, NIRS technology measures saturation in the arterial, venous, and capillary compartments. Change in body position may alter the ratio of these compartments in the cerebral circulation, and therefore, reductions in SctO2 may not necessarily reflect an actual change in brain oxygenation.
There are several limitations to the present investigation. Near-infrared spectroscopy measurements are limited to the prefrontal cortex; therefore, we were not able to measure oxygenation in other parts of the brain. Near-infrared spectroscopy monitors are provided with an algorithm to remove the signal of extracerebral tissues from the actual value. Nevertheless, this could still contribute significantly to the signal.23 Patients were not randomized but allocated to either the BCP or the LDP according to surgical preference, and this preference may introduce bias. Finally, cerebral tissue oxygenation is a surrogate marker for clinically important outcomes such as adverse neurological events. We have not attempted to associate SctO2 values with clinically important outcomes. It is still unknown whether low SctO2 values presage worse postoperative neurological outcomes.
More than 55% of the patients undergoing arthroscopic shoulder surgery in the BCP experienced cerebral desaturation events. In contrast, SctO2 in awake volunteers decreased only minimally. Further research should be performed to elucidate the clinical importance of cerebral desaturation events in patients undergoing shoulder surgery under general anesthesia in the beach chair position.
The authors sincerely thank the nursing and medical staff of the Department of Anesthesiology and the Department of Orthopedic Surgery.
Conflicts of interest
None to declare.
Sources of support
This manuscript is part of the Limburg Clinical Research Program (LCRP), UHasselt-ZOL-Jessa, supported by the Limburg Sterk Merk foundation, Hasselt University, Ziekenhuis Oost-Limburg and Jessa Hospital.
- 20.Cullen DJ, Kirby RR. Beach chair position may decrease cerebral perfusion - catastrophic outcomes have occured. APSF Newsletter 2007; 22: 27.Google Scholar
- 22.Ogoh S, Sato K, Okazaki K, et al. A decrease in spatially resolved near-infrared spectroscopy-determined frontal lobe tissue oxygenation by phenylephrine reflects reduced skin blood flow. Anesth Analg 2014; 118: 829-9.Google Scholar