Harvey L. Edmonds

The role of neuromonitoring in cardiovascular surgery

This review describes the techniques currently used for quantitative neurophysiologic measurement during cardiac surgery and their potential impact on clinical outcome. Electroencephalography (EEG) characterizes cerebrocortical neuronal electrical activity and was part of some of the earliest cardiopulmonary bypass procedures, yet today it is not widespread use. Each of the common misunderstandings regarding a supposed limitation of this technology is explained. Its major genuine shortcoming, a lack of selectivity, may now be overcome with the combined use of additional monitoring modalities. The influence of intracranial hemodynamics on observed EEG changes may be determined continuously and noninvasively with transcranial Doppler (TCD) ultrasound. TCD provides an indication of sudden change in either blood flow or vascular resistance as well as the detection of emboli. In addition, the metabolic status of cortical neurons can be monitored by regional cerebral venous oxygen saturation (rCVOS) using noninvasive transcranial near-infrared spectroscopy. The % rCVOS tends to remain remarkably stable over a wide range of temperatures, perfusion pressures, and anesthetic states. Marked change in either direction signifies a serious imbalance between oxygen delivery and consumption. Measurement of rCVOS does not require blood flow, pulsatile or otherwise, so that it offers the only means of monitoring during circulatory arrest. By characterizing the dynamic interplay among cerebral hemodynamics, metabolism, and electrogenesis, these technologies permit the rapid detection and correction of potentially hazardous conditions.

Evidence for Improved Cerebral Function After Minimally lnvasive Bypass Surgery

Abstract Background: Neurological impairment is a major cause of morbidity after cardiac surgery and may be associated with occurrence of cerebral microemboli generated during cardiopulmonary bypass (CPB). This study evaluates cerebral dysfunction following coronary artery surgery on‐pump and off‐pump. Methods: Neurological outcome was evaluated in 322 patients with a coronary artery bypass graft (CABG). Conventional CPB was used (on‐pump) in 305 patients and in 17 patients no CPB was used (off‐pump). Intraoperatively, a pulsed‐wave transcranial Doppler with a 2‐MHZ probe measured high‐intensity transient signals (HITS) by ultrasonic insonnation of the middle cerebral artery indicating the presence of emboli within the vessel lumen. Transcranial near‐infrared spectroscopy measured cerebral venous oxygen saturation for adequate perfusion. Postoperatively, all patients were subjected to the antisaccadic eye movement (ASEM) test, a sensitive indicator of neurocognitive deficits secondary to frontal lobe dysfunction. Results: While there was no significant difference in O2 saturation, the number of microemboli HITS generated was significantly higher in the on‐pump group than the off‐pump group. In the off‐pump group, 16 (94%) of 17 patients had perfect scores on the ASEM test, while only 108 (35.4%) of 305 patients achieved a perfect score in the on‐pump group (p < 0.01). Furthermore, while all patients in the off‐pump group achieved at least 90%, 28% (86/305) in the on‐pump group scored “zero” on the ASEM test. Conclusion: Cerebral dysfunction as evidenced by ASEM errors is common following coronary bypass on‐pump, but rare with off‐pump bypass surgery. Cerebral microemboli generated during CPB may account for this difference. (J Card Surg 1998;13:27–31

Neurophysiologic monitoring to assure delivery of retrograde cerebral perfusion

BACKGROUND Patients undergoing complex aortic procedures performed with deep hypothermia and circulatory arrest have a significant risk of an adverse neurologic event when the arrest period is prolonged. Retrograde cerebral perfusion appears to improve cerebral protection, although collapsed cortical veins or functional jugular venous valves may restrict flow at the frequently recommended maximum pressure of 25 mm Hg. Therefore, the purpose of this study was to demonstrate the benefit of multimodality neurophysiologic monitoring in assuring delivery of retrograde cerebral perfusion. METHODS Electroencephalography, cerebral blood flow velocity, and regional cerebral venous oxygen saturation were used to quantify the intraoperative neurophysiologic changes accompanying retrograde cerebral perfusion. The magnitude of changes was compared with those previously observed during arrest without retrograde cerebral perfusion. RESULTS Thirty patients underwent complex aortic procedures necessitating circulatory arrest, 22 with retrograde cerebral perfusion. The mean retrograde perfusion pressure of 40 mm Hg (30 to 49 mm Hg, 95% confidence interval) and flow rate of 1.2 L/min (0.9 to 1.6 L/min) necessary to achieve documented retrograde cerebral perfusion was much higher than previously recommended. During both retrograde cerebral perfusion and rewarming, cerebral oximetric monitoring guided adjustments in perfusion parameters to limit the rate of desaturation to 0.4% per minute (0.3% to 0.6%). With retrograde cerebral perfusion there was a rapid (1) recovery of electroencephalographic activity during rewarming (21 minutes [range 16 to 26 minutes]) and (2) return of consciousness after the operation (81% [58% to 95%, 95% confidence interval] awake by 12 hours). There was no transcranial Doppler evidence of cerebral edema with retrograde cerebral perfusion. Two neurologic complications occurred in the 22 patients managed with retrograde cerebral perfusion and one in the eight patients managed with arrest only. CONCLUSIONS Multimodality neurologic monitoring assured optimal brain cooling and bihemispheric delivery of retrograde cerebral perfusion. Necessary retrograde pressure and flow were often higher than values previously reported. Avoidance of profound cerebral venous oxygen desaturation during retrograde cerebral perfusion and rewarming was associated with rapid recovery of neurologic function.

A dual chamber for comparative studies using the brain slice preparation

A dual linear-flow chamber for comparative studies using brain slices is described. Electrophysiological and ultrastructural analysis of rat hippocampal slices incubated in the chamber showed that its two compartments allows performance of reliable paired comparison studies in a highly efficient manner.

Guidelines for intraoperative neuromonitoring using raw (analog or digital waveforms) and quantitative electroencephalography: a position statement by the American Society of Neurophysiological Monitoring

Background contextElectroencephalography (EEG) is one of the oldest and most commonly utilized modalities for intraoperative neuromonitoring. Historically, interest in the EEG patterns associated with anesthesia is as old as the discovery of the EEG itself. The evolution of its intraoperative use was also expanded to include monitoring for assessing cortical perfusion and oxygenation during a variety of vascular, cardiac, and neurosurgical procedures. Furthermore, a number of quantitative or computer-processed algorithms have also been developed to aid in its visual representation and interpretation. The primary clinical outcomes for which modern EEG technology has made significant intraoperative contributions include: (1) recognizing and/or preventing perioperative ischemic insults, and (2) monitoring of brain function for anesthetic drug administration in order to determine depth of anesthesia (and level of consciousness), including the tailoring of drug levels to achieve a predefined neural effect (e.g., burst suppression). While the accelerated development of microprocessor technologies has fostered an extraordinarily rapid growth in the use of intraoperative EEG, there is still no universal adoption of a monitoring technique(s) or of criteria for its neural end-point(s) by anesthesiologists, surgeons, neurologists, and neurophysiologists. One of the most important limitations to routine intraoperative use of EEG may be the lack of standardization of methods, alarm criteria, and recommendations related to its application. Lastly, refinements in technology and signal processing can be expected to advance the usefulness of the intraoperative EEG for both anesthetic and surgical management of patients.ObjectiveThis paper is the position statement of the American Society of Neurophysiological Monitoring. It is the practice guidelines for the intraoperative use of raw (analog and digital) and quantitative EEG.MethodsThe following recommendations are based on trends in the current scientific and clinical literature and meetings, guidelines published by other organizations, expert opinion, and public review by the members of the American Society of Neurophysiological Monitoring. This document may not include all possible methodologies and interpretative criteria, nor do the authors and their sponsor intentionally exclude any new alternatives.ResultsThe use of the techniques reviewed in these guidelines may reduce perioperative neurological morbidity and mortality.ConclusionsThis position paper summarizes commonly used protocols for recording and interpreting the intraoperative use of EEG. Furthermore, the American Society of Neurophysiological Monitoring recognizes this as primarily an educational service.

Computerized monitoring of the EMG and EEG during anesthesia. An evaluation of the anesthesia and brain activity monitor (ABM).

SummaryAn intraoperative evaluation was made of the electroencephalographic (EEG) and electromyographic (EMG) functions of the Anesthesia and Brain activity Monitor (ABM®). This device derives both these measures from a single electrode pair that is typically placed on the mid-forehead and mastoid process. The evaluation consisted of 1) quantifying the zero-crossing frequency (ZXF) of the EEG and mean integrated voltage of both measures (MIVEEG and MIVEMG) that occurred during induction and emergence from general anesthesia in 17 patients and 2) case reports sampled from an additional 41 patients.Alone or combined, variations in these parameters did not consistently accompany changes in the depth or adequacy of anesthesia as determined by standard clinical signs (e.g. heart rate, blood pressure, movement). Interpatient variability in the EEG measures during recovery from anesthesia was so large that neither the absolute value of ZXF nor that of MIVEEG could discriminate between moderate (i.e., maintenance) and light (i.e., emergence) anesthesia. Although MIVEMG uniformly decreased in anesthetized, unparalyzed patients (compared to the pre-operative awake state), noticeable increases during recovery often did not occur until limb movement was observed. Additionally, the common use of neuromuscular blockers made interpretation of low MIVEMG values quite difficult during anesthesia maintenance.However, selected individual case reports illustrated the potential benefit of routine intraoperative, microprocessor-based EEG/EMG monitoring. The single channel EEG/EMG display of the ABM seems sufficient to warn the anesthesiologist of pathologic decreases in cerebral electrical activity. Marked depression of cerebral function is associated with accidental anesthetic overdose, hypoxia or global ischemia. Additionally, the device should be useful for monitoring burst-suppression or isoelectric EEG patterns intentionally produced during barbiturate or isoflurane coma for cerebral protection.

Comparison of neuromuscular blockade in upper facial and hypothenar muscles

Facial and hand muscles are used frequently for monitoring neuromuscular blockade. Therefore, we compared changes in electrically evoked muscle potential magnitude in upper facial and hypothenar muscles after fixed doses of neuromuscular blockers (succinylcholine, 750 µg/kg; pancuronium, 70 µg/kg; vecuronium, 50 µg/kg; and atracurium, 300 µg/kg). Face-hand comparisons were made in both anesthetized (nitrous oxide/narcotic, n=51) and comatose (closed-head injuries, n=5) patients. In 24 anesthetized patients, complete blockade of the hypothenar muscles prevented quantitative comparison. In the remaining 27 patients, the relaxant effect (as determined by the percentage change from prerelaxant baseline muscle potentials) was significantly smaller (P<0.0001) in the upper facial muscles (65±24% versus 92±8%, mean ±SD). All four evoked muscle responses to train-of-four stimulation were detectable in upper facial muscles of the 19 patients receiving non-depolarizing neuromuscular blocking drugs; this pattern was seen in hand muscles of only 7 patients (P<0.001). The neuromuscular blockade in both the hand (49±54%) and the upper facial area (68±28%,P>0.05) of comatose patients was smaller and more variable than that seen during anesthesia. These results illustrate the valuc of quantitative monitoring of neuromuscular function, especially during highly variable and unpredictable drug-induced blockade in the comatose state. We conclude that during narcotic-based anesthesia the upper facial and hand muscles are differentially sensitive to commonly used neuromuscular blockers.

A prospective analysis of intraoperative electromyographic monitoring of posterior cervical screw fixation.

Objective: This is a prospective study of 26 patients undergoing posterior cervical spine instrumentation with lateral mass or pedicle screws to determine the correlation between intraoperative screw stimulation thresholds and the position of posterior cervical lateral mass and pedicle screws. Methods: One hundred forty-seven posterior cervical screws (122 lateral mass screws and 25 C7 pedicle screws) in 26 patients were electrically stimulated intraoperatively and stimulation thresholds recorded. Computed tomography (CT) scans were taken postoperatively and were evaluated independently to assess screw position. Electromyographic (EMG) thresholds and CT data were compared to assess the accuracy of the EMG screw stimulation technique in detecting screw malposition. Results: Intraoperative electrical stimulation was accurate in verifying screw position. A stimulation threshold of 15 mA provided a 99% positive predictive value (89% sensitivity, 87% specificity) that the screw was within the lateral mass or pedicle. Stimulation values of 10-15 mA provided a 13% predictive value (66% sensitivity, 90% specificity) that the screw was within the lateral mass or pedicle. A stimulation value of <10 mA provided a 100% predictive value that the screw was malpositioned (70% sensitivity, 100% specificity). Conclusions: Intraoperative evoked EMG monitoring is a valuable tool in posterior cervical instrumentation using lateral mass and pedicle screws. Stimulation thresholds in this study correlated with screw position. Stimulation values of >15 mA reliably predict acceptable screw position. Values between 10 and 15 mA are generally associated with acceptable screw position, although exploration is recommended. Values below 10 mA are associated with screw malposition and warrant exploration, repositioning, and possible removal.

Quantitative surface electromyography in anesthesia and critical care

SummaryThe frontalis muscle spontaneous (SEMG) and electrically evoked (EEMG) electromyograms were recorded in 4 different clinical settings. Using a standardized isoflurane-based anesthetic protocol, Study 1 examined the SEMG response to both surgical and acoustic stimuli. The acoustic SEMG response was also examined in comatose head-injured patients. Study 2 used the EEMG to compare the extent of vecuronium-induced neuromuscular blockade on the frontalis and hypothenar muscles in both anesthetized and comatose patients. In Study 3 head-injured comatose patients were used to investigate the relationship between SEMG changes and transient elevations in intracranial pressure (ICP). The effect of opiate analgesics on the pain-activated SEMG in conscious post-operative patients was investigated in Study 4.These studies illustrate the following phenomena. First, in conscious, unparalyzed or lightly anesthetized patients, painful (stressful) stimuli are associated with increases in SEMG amplitude. Thus, the SEMG may indicate periods of inadequate analgesia, not only post-operatively (Study 4) but also intra-operatively (Study 1), since we found the frontalis to be relatively insensitive to a non-depolarizing neuromuscular blocker (Study 2). However, the interpretation of intra-operative SEMG changes may be confounded by opiates (Study 4) and perhaps other agents capable of influencing the frontalis through either non-nociceptive central or peripheral mechanisms. Second, the opiate analgesics consistently decreased SEMG amplitude in non-tolerant conscious patients (Study 4). Although this opiate-induced decrease is not necessarily indicative of opiate analgesia, it may provide an objective, quantifiable measure of a central opiate effect. The SEMG is particularly well-suited to determine the precise timecourse of this effect. Third, in deeply anesthetized or comatose patients, unresponsive to either surgical or electrical stimulation, SEMG amplitude may increase in response to elevated ICP or certain sounds (Study 3). The stress (pain) and auditory-evoked SEMGs may thus provide measures of brainstem function that are independent of the level of consciousness.

Rapid recognition and treatment of cerebral air embolism: the role of neuromonitoring

Cerebral air embolism is a rare but serious complication of cardiac operations. Transcranial Doppler ultrasonography is a sensitive monitor of gas bubble entry into the cerebral circulation. Here we describe a case of cerebral air embolism during a Fontan procedure and the role of transcranial Doppler ultrasonography, cerebral oximetry, and electroencephalography (EEG) in its recognition and management.