EEG changes generally occur when CBF decreases to 20cc/100g/min [except under isoflurane, where CBF can fall to 8-10cc/100g/min]. Normal brain perfusion is approximately 50 cc/100g/min, and cellular damage occurs at 12cc/100g/min, thus EEG can theoretically warn of impending perfusion deficits. That said, the EEG is a global measure and can be affected by a variety of variables
Evoked potentials can be sensory (SSEPs, BEPs), motor, or EMG. SEPs are the response to peripheral (SSEP) /cranial nerve (BEP) stimulation, recorded at the scalp. Because the S/N ratio is so low, hundreds of signals have to be averaged out to generate meaningful information. Damage to any element of the neuronal chain causes prolonged latency and decreased signal amplitude.
Somatosensory Evoked Potentials
SSEPs are generally used in spine surgery, and BEPs are generally used in posterior fossa procedures. Volatile anesthetics prolong the latency and reduce the amplitude of SSEPs, thus these procedures are usually done under TIVA or with mixed modal anesthesia. BEPs are affected, but less-so
Motor Evoked Potentials
MEPs may be more reliable than SSEPs for spine procedures [Dunn PF. Clinical Anesthesia Procedures of the Massachusetts General Hospital, 7th ed. LWW (Philadelphia) p. 448, 2007], as they assess activity of the ventral spinal cord, which is traditionally considered to be more “at risk” than the dorsal columns predominantly assessed by SSEPs. Transcranial electrical stimulation is required to produce a motor impulse, and the response can be measured in the muscle of interest or over the spinal cord distal to the site of surgery. Volatile anesthetics can completely obliterate MEPs (less-so if measured at the spinal cord), thus TIVA is indicated in these cases. NMBDs should be avoided if the response to MEPs is measured at the muscle itself (as opposed to on the cord)
EMG (stimulation of a motor nerve with subsequent measured muscle response) is used for CPA tumors, and if used NMBDs should be avoided
Data on Outcomes
There are no high quality studies showing improved outcome with electrophysiologic monitoring, likely because of the relatively high false positive rate. Such studies are unlikely to be done, however, as practitioners are (understandably) reluctant to ignore the potential information that can be gleaned from these studies.