BlueCross and BlueShield of Montana Medical Policy/Codes
Magnetoencephalography (MEG) and Magnetic Source Imaging (MSI)
Chapter: Radiology
Current Effective Date: September 24, 2013
Original Effective Date: October 13, 2011
Publish Date: September 24, 2013
Revised Dates: November 09, 2012; August 28, 2013


Magnetoencephalography (MEG) is a noninvasive functional imaging technique in which the weak magnetic forces associated with the electrical activity of the brain are recorded externally. Using mathematical modeling, the recorded data are then analyzed to provide an estimated location of the electrical activity. This information can be superimposed on an anatomic image of the brain, typically a magnetic resonance imaging (MRI) scan, to produce a functional/anatomic image of the brain, referred to as magnetic source imaging or MSI. The primary advantage of MSI is that while the conductivity and thus the measurement of electrical activity as recorded by the electroencephalogram (EEG) is altered by surrounding brain structures, the magnetic fields are not. Therefore, MSI permits a high-resolution image.

The technique is sophisticated. Detection of the weak magnetic fields depends on gradiometer detection coils coupled to a superconducting quantum interference device (SQUID), which requires a specialized room shielded from other magnetic sources. Mathematical modeling programs based on idealized assumptions are then used to translate the detected signals into functional images. In its early evolution, clinical applications were limited by the use of only 1 detection coil requiring lengthy imaging times, which, because of body movement, were also difficult to coordinate with the MRI. However, more recently, the technique has evolved to multiple detection coils arranged in an array that can provide data more efficiently over a wide extracranial region.

One clinical application is localization of the pre- and postcentral gyri as a guide to surgical planning in patients scheduled to undergo neurosurgery for epilepsy, brain neoplasms, arteriovenous malformations, or other brain disorders. These gyri contain the "eloquent" sensorimotor areas of the brain, the preservation of identified anatomically by MRI, but frequently the anatomy is distorted by underlying disease processes. In addition, the location of the eloquent functions is variable, even among healthy patients. Therefore, localization of the eloquent cortex often requires such intraoperative invasive functional techniques as cortical stimulation with the patient under local anesthesia or somatosensory-evoked responses on electrocorticography (ECoG). While these techniques can be done at the same time as the planned resection, they are cumbersome and can add up to 45 minutes of anesthesia time. Furthermore, sometimes these techniques can be limited by the small surgical field. A preoperative test, which is often used to localize the eloquent hemisphere, is the Wada test. MEG/MSI has been proposed as a substitute for the Wada test.

Another related clinical application is localization of epileptic foci, particularly for screening of surgical candidates and surgical planning. Alternative techniques include MRI, positron emission tomography (PET), or single photon emission computed tomography (SPECT) scanning. Anatomic imaging (i.e., MRI) is effective when epilepsy is associated with a mass lesion, such as a tumor, vascular malformation, or hippocampal atrophy. If an anatomic abnormality is not detected, patients may undergo a PET scan. In a small subset of patients, extended ECoG or stereotactic electroencephalography EEG (SEEG) with implanted electrodes is considered the gold standard for localizing epileptogenic foci. MEG/MSI has principally been investigated as a supplement to or an alternative to invasive monitoring.


Each benefit plan, summary plan description or contract defines which services are covered, which services are excluded, and which services are subject to dollar caps or other limitations, conditions or exclusions.  Members and their providers have the responsibility for consulting the member's benefit plan, summary plan description or contract to determine if there are any exclusions or other benefit limitations applicable to this service or supply.  If there is a discrepancy between a Medical Policy and a member's benefit plan, summary plan description or contract, the benefit plan, summary plan description or contract will govern.


Magnetoencephalography (MEG) and magnetic source imaging (MSI) may be considered medically necessary in the following situations:

  • For the purpose of determining the laterality of language function, as a substitute for the Wada test, in patients undergoing diagnostic workup in preparation for surgery for epilepsy, brain tumors, and other indications requiring brain resection; or
  • As part of the preoperative evaluation of patients with intractable epilepsy (seizures refractory to at least two first-line anticonvulsants), when standard techniques, such as magnetic resonance imaging (MRI) and electroencephalogram (EEG), do not provide satisfactory localization of epileptic lesion(s).

Magnetoencephalography (MEG) and magnetic source imaging (MSI) are considered experimental, investigational and/or unproven for all other indications.


Localization of Seizure Focus

This section is based on a 2008 Blue Cross Blue Shield Association (BCBSA) Technology Evaluation Center (TEC) Special Report reviewing the evidence regarding MEG for localization of epileptic lesions. (1) MEG has been proposed as a method for localizing seizure foci for patients with normal or equivocal magnetic resonance imaging (MRI) and negative video-electroencephalogram (EEG) examinations, so-called “nonlesional” epilepsy. Such patients often undergo MEG, positron emission tomography (PET), or ictal-single photon emission computed tomography (SPECT) tests to attempt to localize the seizure focus. They then often undergo invasive intra-cranial EEG, a surgical procedure in which electrodes are inserted next to the brain. MEG would be considered useful if, when compared to not using MEG, it improved patient outcomes. Such improvement in outcomes would include more patients being rendered seizure-free, use of a less invasive and morbid diagnostic workup, and increased surgical success rates. This is a complicated array of outcomes that has not thoroughly been evaluated in a comprehensive manner.

Ideally, a randomized trial comparing the outcomes of patients who receive MEG as part of their diagnostic workup compared to patients who do not receive MEG could determine whether MEG improves patient outcomes. However, almost all of the studies evaluating MEG have been retrospective, where MEG, other tests, and surgery have been selectively applied to patients. Since patients often drop out of the diagnostic process before having intracranial EEG (IC-EEG), and many patients ultimately do not undergo surgery, most studies of associations between diagnostic tests and between diagnostic tests and outcomes are biased by selection and ascertainment biases. For example, studies that evaluate the correlation between MEG and IC-EEG invariably do not account for the fact that MEG information was sometimes used to deselect a patient from undergoing IC-EEG. In addition, IC-EEG findings only imperfectly correlate with surgical outcomes, meaning that it is an imperfect reference standard.

Numerous studies have shown associations between MEG findings and other noninvasive and invasive methods diagnostic tests, including IC-EEG, and between MEG findings and surgical outcomes. However, such studies do not allow any conclusions regarding whether MEG added incremental information to aid the management of such patients and whether patients’ outcomes were improved as a result of the additional diagnostic information.

A representative study of MEG by Knowlton and colleagues (2) demonstrates many of the problematic issues of evaluating MEG. In this study of 160 patients with nonlesional epilepsy, all had MEG, but only 72 proceeded to IC-EEG. The calculations of diagnostic characteristics of MEG are biased by incomplete ascertainment of the reference standard. However, even examining the diagnostic characteristics of MEG using the 72 patients who underwent IC-EEG, sensitivities and specificities were well below 90%, indicating the likelihood of both false-positive and false-negative studies. Predictive values based on these sensitivities and specificities mean that MEG can neither rule in nor rule out a positive IC-EEG, meaning that MEG cannot be used as a triage test before IC-EEG to avoid the potential morbidity in a subset of patients.

One study more specifically addresses the concept that MEG may improve the yield of IC-EEG, thus allowing more patients to ultimately receive surgery. In a study by Knowlton et al., (3) out of 77 patients who were recommended to have IC-EEG, MEG results modified the placement of electrodes in 18 of the 77 cases. Seven cases out of the 18 had positive intracranial seizure recordings involving the additional electrodes placed because of the MEG results. It was concluded that 4 patients are presumed to have had surgery modified as a result of the effect of MEG on altering the placement of electrodes.

Several studies correlate MEG findings to surgical outcomes. Lau et al. (4) performed a meta-analysis of 17 such studies. In this meta-analysis, sensitivity and specificity have unorthodox definitions. Sensitivity is the proportion of patients cured with surgery in whom the MEG-defined epileptic region was resected, and specificity is the proportion of patients not cured with surgery in whom the MEG-defined epileptic region was not resected. The pooled sensitivity was 0.84, meaning that among the total number of cured patients, 14% occurred despite the MEG-defined region not being resected. Pooled specificity was 0.52, meaning that among 48% of patients not cured, the MEG-localized region was resected. These results are consistent with an association between resection of the MEG-defined region and surgical cure, but that it is an imperfect predictor of surgical success. However, it does not address the question as to whether MEG contributed original information to improve the probability of cure.

Other studies imply a value to MEG, but it is difficult to make firm conclusions regarding its value. In a study by Schneider et al., (5) 14 patients with various findings on MEG, IC-EEG, and interictal SPECT underwent surgery for nonlesional neocortical focal epilepsy. Concordance of IC-EEG and MEG occurred in 5 patients, in whom 4 became seizure free. This concordance of the 2 tests was the best predictor of becoming seizure-free. Although this was prognostic for success, whether this would actually change surgical decision making, such as declining to operate where there is not such concordance, is uncertain. A similar study by Widjaja et al. (6) shows that concordance of MEG findings with the location of surgical resection is correlated with better seizure outcomes. However, the authors admit that MEG is entrenched in clinical practice, and the decision to proceed further in diagnostic and therapeutic endeavors is based on the results of MEG and other tests.

The American Clinical MEG Society released a position statement that supports the routine clinical use of MEG/MSI for pre-surgical evaluation of patients with medically intractable seizures. (7) In this statement, they specifically cite a study by Sutherling et al. (8) as being a “milestone class I study.” Class I evidence usually refers to randomized comparisons of treatment. However, the study by Sutherling et al. is called by its authors a “prospective, blinded crossover-controlled, single-treatment, observational case series.” The study attempts to determine the proportion of patients in whom the diagnostic or treatment strategy was changed as a consequence of MEG. They concluded that the test provided nonredundant information in 33% of patients, changed treatment in 9% of surgical patients, and benefited 21% of patients who had surgery. There was no control group in this study. Benefit of MEG was inferred by assumptions of what might have occurred in the absence of the MEG result. Less than half of the 69 patients went on to receive IC-EEG; thus, there appears to be incomplete accounting for outcomes of all patients in the study. A similar study by De et al. (9) also attempted to determine the number of patients in whom management decisions were altered based on MEG results. They concluded that clinical management was altered in 13% of all patients.

Conclusions. There are no clinical trials demonstrating the utility of MEG in determining location of seizure focus and no high-quality studies of diagnostic accuracy. The available evidence on diagnostic accuracy is limited by ascertainment and selection biases because MEG findings were used to select and deselect patients in the diagnostic pathway thus, making it difficult to determine the role of MEG for the purpose of seizure localization. The evidence supporting the effect of MEG on patient outcomes is indirect and incomplete. Surgical management may be altered in a minority of patients based on MEG, but there is insufficient evidence to conclude that outcomes are improved as a result of these management changes. Trials with a control group are needed to determine whether good outcomes can be attributed to the change in management induced by knowledge of MEG findings.

Localization of Eloquent and Sensorimotor Areas

In a 2003 TEC Assessment of MEG, the evidence for this particular indication concluded that the evidence was insufficient to demonstrate efficacy. (10) At that time, the studies reviewed had relatively weak study methods and very limited numbers of subjects. There are two ways to analyze the potential utility of MEG for this indication. MEG could potentially be a noninvasive substitute for the Wada test, which is a standard method of determining hemispheric dominance for language. The Wada test requires catheterization of the internal carotid arteries, which carries the risk of complications. The determination of the laterality of the language function is important to know to determine the suitability of a patient for surgery and what types of additional functional testing might be needed prior to or during surgery. If MEG provides concordant information with the Wada test, then such information would be obtained in a safe, noninvasive manner.

Several studies have shown high concordance between the Wada test and MEG. In the largest study, by Papanicolaou and co-workers, among 85 patients, there was concordance between the MEG and Wada tests in 74 (87%). (11) In no cases were the tests discordant in a way that the findings were completely opposite. The discordant cases occurred mostly when the Wada test indicated left dominance and the MEG indicated bilateral language function. In an alternative type of analysis, where the test is being used to evaluate the absence or presence of language function in the side in which surgical treatment is being planned, using the Wada procedure as the gold standard, MEG was 98% sensitive and 83% specific. Thus, if the presence of language function in the surgical site requires intraoperative mapping and/or a tailored surgical approach, use of MEG rather than Wada would have “missed” one case where such an approach would be needed, and resulted in 5 cases where such an approach was unnecessary (false- positive MEG). However, it should be noted that the Wada test is not a perfect reference standard, and some discordance may reflect inaccuracy of the reference standard. In another study by Hirata et al., MEG and the Wada test agreed in 19/20 (95%) of cases. (12)

The other potential use of MEG would be for the purpose of mapping the sensorimotor area of the brain, again to avoid such areas in the surgical resection area. Intraoperative mapping just before resection is generally done as the reference standard. Preoperative mapping as potentially done by MEG might aid in determining the suitability of the patient for surgery or for assisting in the planning of other invasive testing. Similar to the situation for localization of epilepsy focus, the literature is problematic in terms of evaluating the comprehensive outcomes of patients due to ascertainment and selection biases. Studies tend to be limited to correlations between MEG and intraoperative mapping. The intraoperative mapping would be performed anyway in most resection patients. Several of the studies evaluated in the 2003 TEC Assessment showed good to high concordance between MEG findings and intraoperative mapping. (10) A technology assessment on functional brain imaging performed by the Ontario Ministry of Health reviewed 10 studies of MEG and invasive functional mapping and showed good to high correspondence between the two tests. (13) However, these studies do not demonstrate that MEG would replace intraoperative mapping or reduce the morbidity of such mapping by allowing a more focused procedure.

Recent studies of the use of MEG in localizing the sensorimotor area provide only indirect evidence of utility. A study by Niranjan et al. (14) reviewed the results of 45 patients in whom MEG was used for localizing somatosensory function. In 32 patients who underwent surgery, surgical access routes were planned to avoid regions identified as somatosensory by MEG. All patients retained somatosensory function. It is unknown to what extent MEG provided unique information not provided by other tests. In a study by Tarapore et al., (15) 24 patients underwent MEG, transcranial magnetic stimulation, and intraoperative direct cortical stimulation to identify the motor cortex. MEG and navigated transcranial magnetic stimulation were both able to identify several areas of motor function, and the median distance between corresponding motor areas was 4.71 mm. When comparing MEG to direct cortical stimulation, the median distance between corresponding motor sites (12.1 mm) was greater than the distance between navigated transcranial magnetic stimulation and direct cortical stimulation (2.13 mm). This study cannot determine whether MEG provided unique information that contributed to better patient outcomes.

Conclusions. There are no clinical trials that demonstrate the utility of using MEG for localization and lateralization of eloquent and sensorimotor regions of the brain. The available evidence consists of studies that correlate results of MEG with the Wada test, which is an alternative method for localization. The evidence generally shows that the concordance between MEG and the Wada test is high. Since MEG is a less invasive alternative to the Wada test, this evidence indicates that it is a reasonable alternative. There is also some evidence that the correlation of MEG with intraoperative mapping of eloquent and sensorimotor regions is high, but the test has not demonstrated sufficient accuracy to replace intraoperative mapping.


The published evidence on magnetoencephalography (MEG) is suboptimal, with no clinical trials demonstrating utility. The literature on diagnostic accuracy has methodologic limitations, primarily selection bias and ascertainment bias. The available studies report that this test has high concordance with the Wada test, which is currently the main alternative for localizing eloquent functions. Management is changed in some patients based on MEG testing, but it has not been demonstrated that these changes in management lead to improved outcomes. Clinical input obtained by the Blue Cross Blue Shield Association in 2011 indicated consensus for use of MEG as a substitute for the Wada test in determining the laterality of language function in patients being considered for surgery to treat epilepsy, brain tumors, and other structural brain lesions. Clinical input also demonstrated consensus on use of MEG as part of the preoperative evaluation of patients with intractable epilepsy when standard techniques, such as MRI, are inconclusive.

Based on the available scientific literature, the results of clinical input, and a strong indirect chain of evidence that outcomes are improved, MEG/MSI may be considered medically necessary as a substitute for the Wada test for the purpose of determining laterality of language function. MEG may also be considered medically necessary as part of the preoperative evaluation of patients with intractable epilepsy when standard techniques such as MRI are inconclusive outcomes.


Disclaimer for coding information on Medical Policies          

Procedure and diagnosis codes on Medical Policy documents are included only as a general reference tool for each policy. They may not be all-inclusive.           

The presence or absence of procedure, service, supply, device or diagnosis codes in a Medical Policy document has no relevance for determination of benefit coverage for members or reimbursement for providers. Only the written coverage position in a medical policy should be used for such determinations.           

Benefit coverage determinations based on written Medical Policy coverage positions must include review of the member’s benefit contract or Summary Plan Description (SPD) for defined coverage vs. non-coverage, benefit exclusions, and benefit limitations such as dollar or duration caps. 

Rationale for Benefit Administration
ICD-9 Codes

191.0, 191.1, 191.2, 191.3, 191.4, 191.5, 191.6, 191.7, 191.8, 191.9, 345.00, 345.01, 345.10, 345.11, 345.2, 345.3, 345.40, 345.41, 345.50, 345.51, 345.60, 345.61, 345.70, 345.71, 345.80, 345.81, 345.90, 345.91, 437.3, 747.81

ICD-10 Codes

C71.0-C71.9, G40.001-G40.419, G40.501-G40.519, G40.801-G40.890, I67.1, Q28.2, Q28.3

Procedural Codes: 95965, 95966, 95967
  1. Special Report: Magnetoencephalography and magnetic source imaging for the purpose of presurgical localization of epileptic lesions—a challenge for technology evaluation. Blue Cross and Blue Shield Association Technology Evaluation Center (TEC) Assessments 2008; Volume 23, Tab 8.
  2. Knowlton RC, Elgavish RA, Limdi NFiI et al. Relative predictive value of intracranial electroencephalography. Ann Neurol 2008; 64(1):25-34.
  3. Knowlton RC, Razdan SN, Limdi N et al. Effect of epilepsy magnetic source imaging on intracranial electrode placement. Ann Neurol 2009; 65(6-Jan):716-23.
  4. Lau M, Yam D, Burneo JG. A systematic review on MEG and its use in the presurgical evaluation of localization-related epilepsy. Epilepsy Res 2008; 79(3-Feb):97-104.
  5. Schneider F, Irene Wang Z, Alexopoulos AV et al. Magnetic source imaging and ictal SPECT in MRI-negative neocortical epilepsies: additional value and comparison with intracranial EEG. Epilepsia 2013; 54(2):359-69.
  6. Widjaja E, Shammas A, Vali R et al. FDG-PET and magnetoencephalography in presurgical workup of children with localization-related nonlesional epilepsy. Epilepsia 2013; 54(4):691-9.
  7. Bagic A, Funke ME, Ebersole J. American Clinical MEG Society (ACMEGS) position statement: the value of magnetoencephalography (MEG)/magnetic source imaging (MSI) in noninvasive presurgical evaluation of patients with medical intractable localization-related epilepsy. J Clin Neurophysiol 2009; 26(4):1-4.
  8. Sutherling WW, Mamelak AN, Thyerlei D et al. Influence of magnetic source imaging for planning intracranial EEG in epilepsy. Neurology 2008; 71(13):990-6.
  9. De TX, Carrette E, Legros B et al. Clinical added value of magnetic source imaging in the presurgical evaluation of refractory focal epilepsy. J Neurol Neurosurg Psychiatry 2012; 83(4):417-23.
  10. Magnetoencephalography and magnetic source imaging: presurgical localization of epileptic lesions and presurgical functional mapping. Chicago, Illinois: Blue Cross Blue Shield Association – Technology Evaluation Center Assessment Program (2003 August) 18(6):1-28.
  11. Papanicolaou AC, Simos PG, Castillo EM et al. Magnetocephalography: a noninvasive alternative to the Wada procedure. J Neurosurg 2004; 100(5-Jan):867-76.
  12. Hirata M, Kato A, Taniguchi M et al. Determination of language dominance with synthetic aperture magnetometry: comparison with the Wada test. Neuroimage 2004; 23(1):46-53.
  13. Ontario Ministry of Health MASMFbiHTPAT, ON: MAS; December 2006. Available online at: Last accessed September 2013.
  14. Niranjan A, Laing EJ, Laghari FJ et al. Preoperative magnetoencephalographic sensory cortex mapping. Stereotact Funct Neurosurg 2013; 91(5):314-22.
  15. Tarapore PE, Tate MC, Findlay AM et al. Preoperative multimodal motor mapping: a comparison of magnetoencephalography imaging, navigated transcranial magnetic stimulation, and direct cortical stimulation. J Neurosurg 2012; 117(2):354-62.
  16. Magnetoencephalography/Magnetic Source Imaging. Chicago, Illinois: Blue Cross Blue Shield Association Medical Policy Reference Manual. (October 2013) Radiology 6.01.21.
October 2011 New Policy with CPT codes: 95965, 95966, 95967, S8035
November 2012 Policy updated with literature review, reference 7 added. No change to policy statements.
September 2013 Policy formatting and language revised.  Policy statement unchanged.  Title changed from "Magnetoencephalography/Magnetic Source Imaging" to "Magnetoencephalography (MEG) and Magnetic Source Imaging (MSI)".
®Registered marks of the Blue Cross and Blue Shield Association, an association of independent Blue Cross and Blue Shield Plans. ®LIVE SMART. LIVE HEALTHY. is a registered mark of BCBSMT, an independent licensee of the Blue Cross and Blue Shield Association, serving the residents and businesses of Montana.
CPT codes, descriptions and material only are copyrighted by the American Medical Association. All Rights Reserved. No fee schedules, basic units, relative values or related listings are included in CPT. The AMA assumes no liability for the data contained herein. Applicable FARS/DFARS Restrictions Apply to Government Use. CPT only © American Medical Association.
Magnetoencephalography (MEG) and Magnetic Source Imaging (MSI)