Abnormally Low Regional Cerebral Oxygen Saturation after Induction of Anesthesia without Neurological Abnormality: A Case Report

ABSTRACT

Measurement of regional cerebral oxygen saturation (rSO2) using near-infrared spectroscopy (NIRS) in cardiac surgery is known to be useful in reducing postoperative neurological complications. We here present a case of a 71-year-old man in whom severe decrease in unilateral rSO2 was observed after induction of general anesthesia for percutaneous mitral valve clipping, although no neurological abnormalities were found. NIRS does not always predict postoperative neurological complications.

INTRODUCTION

Neurological abnormalities are a major concern after cardiac surgery, and continuous measurement of regional cerebral oxygen saturation (rSO2) using near-infrared spectroscopy (NIRS) is reported to be useful in preventing such neurological abnormalities. However, NIRS measurements may be affected by various factors and do not necessarily reflect brain ischemia. We herein report a case in which NIRS detected a unilateral decrease in rSO2 after induction of general anesthesia, although no postoperative neurological abnormalities were observed. In the present case, we used O3 Regional Oximetry (Masimo Corporation, Tokyo, Japan) with disposable sensors to obtain data from the patient.

CASE REPORT

A 71-year-old male (height 165 cm, weight 56 kg) with no history of stroke was scheduled to undergo percutaneous mitral valve clipping for severe mitral regurgitation. Preoperative transthoracic echocardiography showed low cardiac function with a left ventricular ejection fraction of 23%, but no intracardiac thrombus, and carotid artery ultrasound showed plaque but no stenosis in the right and left carotid arteries.

Prior to induction of general anesthesia, an arterial pressure line was inserted into the right radial artery, O3 Regional Oximetry sensors were affixed to the right and left foreheads, and SedLine sensors (Masimo Corporation, Tokyo, Japan) was affixed to the center of foreheads to measure sedation level. Before induction of general anesthesia, blood pressure was 120/50 mmHg, and rSO2 on the right was 60%, while rSO2 on the left was 45%, showing a left-right difference. General anesthesia was induced with midazolam 2 mg, fentanyl 100 μg, rocuronium 40 mg, and remifentanil 0.6 mg/h continuous intravenous injection [Figure 1]. After tracheal intubation, a transesophageal echocardiographic probe was inserted, and anesthesia was maintained with remifentanil 0.6 mg/h, desflurane 3.5%, inhaled oxygen level 40%, dobutamine 3 μg/kg/min, and noradrenaline 0.02 μg/kg/min. After induction of general anesthesia, a central venous catheter was inserted through the right internal jugular vein. Positive end-expiratory pressure was set at 5 cm H₂O and end-tidal carbon dioxide tension remained at 40 to 45 mmHg. When the blood pressure dropped to 80/40 mmHg, the right rSO2 was in the about 70%, which was higher than before induction of anesthesia, but the left rSO2 was very low at 5%. An abnormal sensor connection was suspected, and the left frontal sensor was therefore reconnected; however, the left rSO2 remained low. Next, a sensor abnormality was suspected, and the left frontal sensor was replaced with a new one, but the left rSO2 was very low at 8%. The removed sensor was applied to the left occipital region, and the rSO2 was in the 70%, which was almost the same as the right rSO2, indicating that the sensor was responding normally. Ultrasound of the left carotid artery showed plaque but normal blood flow. Since the left rSO2 was very low, we suspected localized hypoperfusion in the left frontal region and decided to awaken the patient from general anesthesia temporarily. The patient was awakened approximately 10 minutes after administration of 110 mg of sugammadex and termination of desflurane to increase the inhaled oxygen level to 60%. Upon awakening, blood pressure rose and the left rSO2 rose to 55%. When the patient’s name was spoken while being intubated, he nodded in response to the call and moved his left and right upper and lower limbs as instructed. After confirming that there were no major neurological abnormalities, anesthesia was induced again with 2 mg of midazolam. Thereafter, the patient’s rSO2 maintained at 80% on the right and 20-30% on the left with 60% inspired oxygen, and systolic blood pressure of 80-100 mmHg. The surgery was then completed without any complications. No neurological abnormalities were observed after the patient awoke from the anesthesia after surgery. Noradrenaline was terminated as blood pressure increased with anesthetic arousal, and rSO2 was similar to that before induction of anesthesia.

Figure 1.

Anesthetic chart HR = heart rate, SpO2 (Root) = saturation of percutaneuous oxygen, Trect = rectal temperature, NIBP = non-invasive blood pressure, ABP = arterial blood pressure, ETCO2 = end tidal carbon dioxide, rSO2L (Root) = rSO2 in the left frontal area, rSO2R (Root) = rSO2 in the right frontal area, PSI (Root) = EEG signal processing parameters

DISCUSSION

The principle of NIRS is to irradiate near-infrared light and measure oxygenation and deoxygenation of tissues based on differences in tissue absorbance. Because of its characteristics, NIRS is commonly used to detect cerebral ischemia using sensors attached to the forehead area in neurosurgery and cardiovascular surgery and has been reported to be associated with a decrease in the incidence of postoperative cognitive dysfunction, especially in cardiac surgery.[1] However, NIRS does not assess the oxygen saturation of the entire brain, but only monitors the oxygen saturation of the are local to which the sensor is affixed. Furthermore, NIRS values are reported to be affected by hemoglobin concentration, skull thickness, area of cerebrospinal fluid layer area, pigmentation, and carbon dioxide concentration,[2-4] which are thought to affect the attenuation of near-infrared light. Although there is no absolute value indicating brain hypoxia, rSO2 can monitor its trends,[5] and is reported to decrease to 65-80% of baseline in postoperative stroke patients.[6]

In the present case, a left-right difference in rSO2 values was observed before the induction of anesthesia, but only the left value dropped significantly after the induction of anesthesia, leading us to suspect a stroke in the left frontal lobe. Furthermore, the values did not change even after the sensor was replaced with a new one, and when the removed sensor was affixed to a different position, the rSO2 value was increased, making it unlikely that the sensor was defective, leading us to suspect a decrease in regional blood flow in the left frontal lobe. However, when the patient was awakened from anesthesia before surgery, there was no paralysis, and cerebral ischemia was ruled out. In addition, preoperative head CT showed no atrophy of the frontal lobe. It is possible that even more detailed information about the cerebral vessels and carotid arteries could have been obtained if the CT angiography had been taken after the surgery. The cause of the severe decrease in rSO2 in this case is not clear, but it is thought that there was a change in blood flow that, while it did not affect brain function, it did affect NIRS. There are two possible causes: lowered blood pressure by induction of anesthesia and reduced blood flow to the scalp by noradrenaline used to raise blood pressure. Noradrenaline, a vasopressor, has been noted to decrease rSO2 values by decreasing oxygen saturation in the skin.[7] In the present case, there was originally a left-right difference in skin blood flow, and it is possible that induction of anesthesia and use of noradrenaline affected rSO2.

McAvoy et al. reported that brain oximetry was insufficient to detect cerebral ischemia, and questioned its usefulness.[8] The measurement depth of NIRS is typically around 30-40 mm, and the measured sensitivity and specificity for cerebral tissue oxygenation detecting intracranial flow changes were 87.5% and 100%, respectively, and those for extracranial flow changes were both 0%.[9] However, some reports have suggested that the scalp and skull contribute about 5-9% to the spectroscopic signal.[4] Furthermore, it is also possible that positive pressure ventilation and hypocapnia due to ventilation may have contributed to the local cerebral blood flow.[4] In the present case, although the rSO2 was low, the patient had no clinical problems; however, the appearance of postoperative sequelae despite a high rSO2 is problematic.

Relying solely on rSO2 for detection of cerebral blood flow is cautious, as NIRS is susceptible to a variety of influences, and low control values may be susceptible to changes in blood flow that have no effect on brain function. In addition, since there are generally large hemodynamic changes immediately after the induction of anesthesia, changes in rSO2 are likely to occur at this time point. There are several problems with awakening patients from anesthesia in the middle of surgery to confirm abnormalities in rSO2, and it is not applicable to all cases. However, in the present case, since the abnormal values were found before the start of surgery, it was acceptable to awaken the patient from anesthesia to check “consciousness,” which is the most important monitoring item of brain function. However, considering that rSO2 increased along with the rise in blood pressure upon awakening from anesthesia, one option might have been to administer catecholamine more intraoperatively with the aim of increasing cardiac output. In addition, transcranial doppler may have been useful in noninvasively assessing cerebrovascular function in real time.[10] Further study is needed to determine the best course of action.

CONCLUSION

When NIRS is used for intraoperative rSO2 measurement, the measurement should also be performed before induction of anesthesia, and when the value is low, attention should be paid to its further decrease after induction of anesthesia. NIRS does not necessarily reflect ischemic symptoms in the brain and may be affected by a variety of factors, so other parameters that influence it should also be considered, rather than relying solely on the NIRS value.

Declaration of patient consent

Written consent for the presentation of this case was obtained from the patient.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.