Intraoperative Electrical Stimulation Mapping and Monitoring
Neurosurgical Use. Whereas presurgical functional mapping serves as a tool to plan surgical treatments of tumors close to eloquent areas, intraoperative electrical stimulation techniques aim to guide resection in an attempt to achieve maximum resection with minimal risk for neurological deficits, and these techniques serve as the gold standard for functional mapping and monitoring.
Technical Details. At present, different techniques for intraoperative stimulation are available: for intraoperative localization of functional areas, so-called mapping, a low-frequency 50- or 60-Hz stimulation technique or a high-frequency train-of-five stimulation using either monopolar or bipolar stimulation, is used for cortical or subcortical functional localization. Furthermore, a sequential repetitive cortical stimulation performed using the train-of-five technique can be used for motor evoked potential (MEP) monitoring to continuously assess the integrity of the CST. Whereas motor mapping can be performed in an asleep or awake setting, for language mapping awake surgery is required.
Current Evidence. The intraoperative mapping and monitoring techniques that are available have been increasingly used in the last decade, and various studies have reported beneficial effects. Although its use was doubted by many surgeons years ago, there is increasing evidence for the high value of intraoperative DCS mapping.
Already in 2005 an analysis by Duffau et al. compared a series of patients with LGG that was resected with intraoperative electrical stimulation with patients in a historic control group in which operations were not performed with the aid of stimulation mapping. The mapping group consisted of more patients with eloquently located LGG, and the rate of severe permanent deficits decreased from 17% to 6.5% in the mapping group while the rate of GTR increased. Furthermore, a recent meta-analysis on 8091 patients strengthened the evidence for stimulation mapping even more. In this analysis, late severe neurological deficits were observed in 3.4% of patients with intraoperative DCS mapping and in 8.2% of patients after resections performed without DCS mapping. Moreover, GTR was 75% with and 58% without stimulation mapping. The use of electrical stimulation mapping has an oncological (by increasing the EOR) and neurological (by reducing the incidence of new neurological deficits inflicted by surgery) benefit for patients with eloquently located tumors.
Motor Mapping and Monitoring
Cortical Monitoring and Mapping. For intraoperative mapping and monitoring of motor function, different methods are available at a cortical or subcortical level to identify and monitor cortical motor areas or the subcortical tracts.
Continuous transcranial or cortical MEP monitoring is performed by stimulation of the motor cortex by using a train-of-five stimulation technique. Resulting MEPs are recorded by extremity electromyography, and the latency and amplitude are evaluated online for any changes during surgery. Certain criteria for a significant MEP change with predictive value concerning motor outcome have been described, as follows: 1) an amplitude reduction of 50% or more; or 2) a necessary ≥ 4-mA increase in stimulation energy to maintain amplitude height. However, the majority uses a 50% or more amplitude decline as a significant warning criterion.
Reversible MEP amplitude declines of 50% or more are generally associated with temporary motor deficits, whereas irreversible MEP declines or an MEP loss predicts a permanent new motor deficit. Recently the reliability of this modality in terms of potentially false-negative events was investigated, proving that these events are mainly due to secondary injury to the motor system (hemorrhages, secondary ischemia) or resections in supplementary motor areas resulting in temporary motor deficits. This study clarified that postoperative events such as hematoma causing deterioration of motor function do not represent false-negative results when MEPs have remained stable during surgery.
The question arises whether MEP monitoring provides warning information resulting in a change of surgical strategy and potential return to baseline MEP signals or whether this monitoring provides predictive information for motor outcome only. In a series reported by Seidel et al., most MEP declines occurred abruptly and were reversible in only 60% of cases. Thus, the warning function of MEP monitoring is limited and the predictive value for motor outcome predominates.
Subcortical Mapping. To guide resections close to the motor pathways, cortical and subcortical mapping of the motor system are frequently used. For subcortical and cortical mapping stimulation, two different techniques are available: 50- or 60-Hz lowfrequency stimulation or a high-frequency train-of-five stimulation as used for motor monitoring. This monitoring can be used as bipolar or monopolar stimulation (anodal cortical stimulation and cathodal subcortical stimulation). Whereas Berman et al. used 60-Hz bipolar stimulation with stimulation intensity ranging from 8 to 12 mA, Ohue et al. used a train-of-five monopolar cathodal stimulation from 5 to 20 mA and Mikuni et al. used 50-Hz bipolar stimulation without reporting any stimulation intensity.
There are a vast number of other studies reporting on bipolar, monopolar anodal, and monopolar cathodal stimulation applied as a train, which mostly concluded in a linear correlation of current and distance to the CST. Concerning this stimulation setup, Szelényi et al. performed a highly cited and crucial study comparing train application with the single-pulse technique as well as bipolar and monopolar stimulation. It was found that the CST is most efficiently identified using a multipulse train technique with a monopolar probe. Additionally, results of a large series of patients who underwent subcortical motor mapping were recently published, comparing the 60-Hz low-frequency technique to the train-of-five stimulation technique for subcortical motor mapping in patients with tumors involving the CST. This study revealed that in most situations high-frequency stimulation is superior to the older 50- or 60-Hz technique in its efficacy in identifying subcortical motor fibers. Train-of-five high-frequency stimulation seems to be the superior technique to stimulate MEPs from subcortical CST. Thereby, the MEP threshold (i.e., the energy necessary to elicit a peripheral MEP response) reflects the distance between the stimulation point and the CST. There have been attempts to provide a direct transfer between stimulation intensity and distance to the CST. These studies reported a linear correlation between the SCS intensity at which an MEP could be elicited and the distance to the CST. Although this correlation is still under discussion, the majority of neurosurgeons presume a linear correlation of 1 mA of stimulation equals approximately 1 mm of distance of the stimulation point to the CST.
Recently we published our own evaluation of the relation of stimulation distance and stimulation energy. These data revealed that the distance-to-energy relationship is not linear and that stimulation points are closer than assumed from the "1 mA resembles 1 mm" rule. In this study we were able to safely resect toward the CST until a threshold of 3 mA, which is approximately a distance of 2 mm. Some other studies also defined electrical safety margins (i.e., at which stimulation intensity at the white matter of the resection cavity the resection should be stopped to avoid injury to the CST) with consecutive surgery-related paresis. Although this safety margin was reported to be 6 mA for some time, a new study on continuous SCS as permanent monitoring of the CST used 1–3 mA in 24 of 67 cases without any new surgery-related permanent paresis.
For many experienced neurosurgeons, SCS is the most reliable method for estimating the proximity to the CST during resection within the white matter. Thus, neurosurgeons are able to perform safer and even more radical tumor resections close to the CST.
Cortical and Subcortical Language Mapping
Cortical Monitoring and Mapping. The meta-analysis by De Witt Hamer et al. included not only motor but also language eloquent tumors and therefore the corresponding DCS mapping. T hus, t here a re sufficient data at hand to demonstrate that it is difficult to operate on patients with left-sided perisylvian LGG without performing any intraoperative awake DCS mapping.
One large series on awake surgery for patients with glioma showed that only 4 of 243 patients (1.6%) suffered from any surgery-related permanent language deficit 6 months after surgery, and reported a GTR rate of 51.6% in patients with LGG.
Moreover, as also shown in recent studies, DCS mapping during awake surgery can provide cortical maps of language function, which showed a high variability within the dominant hemisphere among the patients investigated.
However, language mapping requires awake mapping, which has become a common tool in contemporary neurosurgery. Awake craniotomy is well accepted and failure rates are low. For mapping of language function during awake surgery, various protocols have been published. The most commonly used is presented in this review. Craniotomy should at least expose the tumor and up to 3 cm of surrounding brain surface. One-millimeter bipolar electrodes positioned 5 mm apart are used, starting with a low stimulus of a constant current with 1.5-mA square-wave pulses and increased to a maximum of 6 mA. A generator delivers biphasic trains of 50 or 60 Hz (depending on the electrical currents used in a particular country). The cortex is mapped every 5–10 mm, and positive stimulation sites at which language impairment was caused are marked with sterile numbered tickets (Figs. 4 and 5). Language tasks usually include systematic counting, naming, and reading; repetition and semantic tasks can be used as well, depending on the primary tumor location.
(Enlarge Image)
Figure 4.
This neuronavigational screenshot shows language mapping via DCS during awake surgery by a navigated pointer (green). The cortex is mapped every 5–10 mm (microscope view), and positive stimulation sites at which language impairment was caused are marked with sterile numbered white tickets.
(Enlarge Image)
Figure 5.
This neuronavigational screenshot shows a large insular LGG. During surgery SCS is performed, with a navigated probe (green) used to map the subcortical fibers (yellow and blue). The red dots show cortical motor cortex as identified by DCS. Subcortical stimulation is performed by measuring the amount of electric current necessary to elicit an MEP, and transfers this current into an actual distance to the fiber tract such as the CST (yellow and blue).
Most importantly, continuous electrocorticography can be used to monitor afterdischarge potentials, and therefore eliminate the chance that language is impaired by focal seizures.
Subcortical Mapping. Awake surgery not only allows mapping of cortical language sites by DCS, but also enables mapping and monitoring of subcortical language tracts.
Duffau et al. recently described the hodotopical model of language function. This model (hodotopical means a delocalized and dynamic model of language function) argues that the language network is organized in widespread, corresponding, separated cortico-subcortical subnetworks for syntactic, semantic, and phonological function. This parallel organization makes it possible for language function to recover after impairment of subnetworks due to resection or surgery-related ischemia. Yet this model, which also highly corresponds with clinical experience, makes it even more important not only to map cortical language sites during awake surgery but also to map and monitor subcortical fiber tracts during LGG resection.
Concerning the technical aspect of subcortical mapping of language function, the same 5-mm spaced bipolar electrodes with a biphasic current (pulse frequency 50 or 60 Hz) are used, with a stimulation intensity of 2–6 mA and the same language tasks as for cortical mapping, depending on the targeted subcortical tract.
Parallel to the mapping of cortical language areas and subcortical language tracts, awake surgery also enables continuous monitoring of language function by use of language tasks given by a trained neuropsychologist even during tumor resection.