Falsely Increased Bispectral Index Values in a Series of Patients Undergoing Cardiac Surgery Using Forced-Air-Warming Therapy of the Head

Monitoring the bispectral index (BIS) is an important part of fast-tracking cardiac anesthesia in our hospital setting. Case reports of falsely increased BIS during hypothermia (1), paradoxical changes during nitrous oxide administration (2), or falsely increased BIS values caused by interferences with pacemakers (3) or forced-air-warming therapy (4) are published examples of instances when the BIS monitor is inaccurate. The second generation of BIS monitors (Aspect A 2000; Aspect Medical Systems, Newton, MA) promised to reduce interference with other electric devices. We present several cases of falsely increased BIS values that we determined were caused by interference with a forced-air-warming blanket around the patient’s head. Case Report All patients (five men, aged 51–73 yr) had an epidural thoracic catheter inserted at the T4 level upon arrival in the operating room and were scheduled for ultra-fast-track anesthesia (immediate extubation in the operating room). Anesthesia was started with propofol 1–1.5 mg/kg and fentanyl 5–7 μg/kg; intubation was facilitated with succinylcholine 1 mg/kg. Anesthesia was maintained with sevoflurane in a breathing gas of 30% oxygen and air, with the end-tidal concentrations titrated to maintain a BIS between 40 and 50. Analgesia was provided by using fentanyl in increments of 1–2 μg/kg (a bolus of 6 mL of bupivacaine 0.25% and infusion of bupivacaine 0.125% 4–10 mL/h were commenced after completion of the coronary grafts), and neuromuscular blockade was provided by rocuronium in increments of 0.2 mg/kg at the discretion of the anesthesiologist. Active temperature control consisted of Bair Hugger® blankets for the head and lower body and increased room temperature of at least 24°C. After the induction of anesthesia, insertion of a right femoral artery catheter, and cannulation of either the right subclavian or internal jugular vein, an original BIS sensor was applied to monitor BIS during surgery. Initial BIS values in all patients were <45. After an initial BIS reading, one Bair Hugger warming device was attached to the head-warming blanket, allowing the creation of a continuous warm-air flow around the head. Within 5 min of commencement of this forced-air flow, the BIS in all five patients increased significantly and showed values of consistently >70 in two patients and values of >90 in three patients throughout surgery (Table 1).Table 1: BIS Values Before, with, and at Pausing of Forced Warm Air to the Head for Each Patient, the Concentration of Sevoflurane, and the Total Fentanyl Dose up to This PointThere was no clinical sign (increased blood pressure, tachycardia, or physical signs such as sweating) suggesting insufficient depth of anesthesia in any patient. The A 2000 monitor showed an optimal signal quality in all patients, with no signal indicating activity of facial muscles. The electroencephalographic (EEG) raw signal showed fast-moving waves of high amplitude as expected in the awake state. To establish the source of the problem, the forced-warm-air flow was switched off, which reduced the BIS in all patients within 15 s to values corresponding to the clinically judged sufficient depth of anesthesia (values between 35 and 55). There did not seem to be an electric interference, because the Bair Hugger heating device switched on alone did not cause any interference. The anesthesiologist continued active temperature control with the Bair Hugger heating blankets and forced-air warming therapy; BIS monitoring was continued and indexes checked intermittently by interrupting the forced-air flow. At all these moments, the high BIS values decreased immediately to levels between 35 and 55, and they increased again to values >70 when forced-air flow returned. Discussion During the last decade, BIS monitoring has become a helpful part of anesthetic monitoring, especially in cardiac anesthesia, in which titration of anesthesia is important, clinical signs are difficult to interpret, and intraoperative awareness is a concern. However important its use might be in cardiac anesthesia, the influence of artifacts and interferences with other electrical devices, such as pacemakers or electrocautery devices, are relevant and make the interpretation of BIS more difficult. The new Aspect A 2000 monitor provides methods to identify artifacts, including a bar of signal quality, a bar that reflects spontaneous activity of facial muscles, and the screening of the original raw EEG. Disturbance signals should therefore be easily recognized and the corresponding BIS interpreted with care. However, if signal quality is indexed as optimal, if there is no electromyelogram signal, and there are no visible artifacts at the raw EEG tracing, the anesthesiologist generally believes that the BIS is valid and acts accordingly. Our experience therefore is disconcerting: all these criteria were met, and there was no indication for any interference. The only reasons the anesthesiologist questioned the BIS were the extraordinarily high indices that did not match the clinical assessment of the depth of anesthesia and a case report published in 2000, which stated problems with the concomitant use of upper-body Bair Hugger warming blankets and increased BIS values in one patient (4). However, the problem of reliability and credibility of BIS values was particularly important in our series of patients when BIS values were in the range of 60–70. It was further complicated by the fact that with the same forced-air-warming method, five other patients showed adequately low BIS values that did not change when air flow was stopped. This means that out of 10 consecutive patients in whom we used the same forced-air-flow warming method around the head, half of these patients showed falsely increased BIS values without any sign of artifact recognition. Falsely increased BIS values caused by forced-air flow are not a singular event. Forced-air flow can severely impair the interpretation of BIS and its use to titrate anesthesia. The only way to reliably separate those falsely increased BIS values from BIS values indicating insufficient depth of anesthesia is to stop warm-air flow and control the BIS. In all our patients this maneuver showed significantly lower levels corresponding to clinical criteria of sufficient anesthesia. The reason(s) that warm air flow falsely increases BIS are not clear. Guignard and Chauvin (4) speculated that air circulation generated vibration of the head wires. However, they did not see any artifact on the raw EEG trace. Another explanation might have been the plastic cover flapping and touching the electrodes, thus interfering with the readings. It is interesting to note that we used the new sensor, which differs completely from the older wires used by Guignard and Chauvin. We checked the impedances in all our patients with increased BIS values with and without warm air flow and could not find any differences. Additionally, circulating air does not seem to cause interferences. Very high frequency artifacts, such as those created by electrocautery, are readily visible in raw EEG, although artifacts created by warm-air flow are not detectable. We suggest the theory that forced-air flow might cause artifacts of the same frequency as EEG waves during light anesthesia or the awake state. How forced-air flow vibrations are transferred into these recordings cannot easily be explained without additional controlled experiments. We recommend additional caution in interpretation of BIS readings whenever the monitoring sensor is in close proximity to a forced-warm-air therapy blanket. Temporary interruption of the warm-air flow may even be required to ascertain accurate BIS readings.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.