CurrentAetnaPolicy 0739

Functional Magnetic Resonance Imaging (fMRI)

This Aetna clinical policy bulletin describes medical necessity and investigational indications for functional MRI (fMRI), primarily for pre-surgical mapping of eloquent cortex in brain tumor, epilepsy, and vascular malformation patients, and lists covered and not-covered diagnosis and procedure codes.

Policy Summary

PayerAetna

PolicyFunctional Magnetic Resonance Imaging (fMRI)

Policy CodePolicy 0739

Change TypeNo material change

Effective Date

Next Review Date

Key ActionDocument appropriate covered ICD-10 diagnosis code(s) corresponding to brain tumor, epilepsy, or vascular malformation when billing CPT 70554 or 70555.

SourceLink

POLICY UPDATE CHANGES

No material clinical or coverage changes in this revision.

non-temporal tumor/epilepsyIndications medically necessary

temporal neocortical epilepsy/tumorsExplicit exclusions

70554,70555Procedure codes

22%–66%Language fMRI sensitivity

91%–98%Language fMRI specificity

Trek Health ingests and normalizes Transparency in Coverage data and payer policy updates to give provider organizations a clear view of how commercial reimbursement behaves across markets, payers, and services. Our platform transforms raw payer disclosures into structured intelligence that supports contract evaluation, payer negotiations, and service line strategy. By combining market benchmarks with ongoing policy visibility, Trek helps teams identify variability, risk, and opportunity in commercial reimbursement. The result is faster insight, stronger negotiating positions, and more informed financial decisions.

Medically Necessary Indications

Aetna considers fMRI medically necessary when used to identify eloquent cortex in pre-surgical evaluation under the following covered conditions (ALL must apply):

Covered indications: Pre-surgical evaluation to identify eloquent cortex in persons with brain tumors (except temporal tumors), epilepsy (except temporal neocortical epilepsy), or vascular malformations.

From policy medical necessity statement; temporal tumors and temporal neocortical epilepsy are excluded.

Experimental / Investigational (Not Covered)

Aetna considers fMRI experimental and investigational (NOT covered) for the following (any of the following are sufficient):

Excluded/Investigational indications: Use to identify eloquent cortex in pre-surgical evaluation of temporal neocortical epilepsy or temporal tumors; and for diagnosis, monitoring, prognosis, or surgical management of other indications including Alzheimer's disease; anger and aggressive behaviors; anxiety disorder; anoxic-ischemic brain injury; attention-deficit hyperactivity disorder; autism spectrum disorder; bipolar disorder; childhood maltreatment; chronic pain (including fibromyalgia); disorders of consciousness (eg, locked-in syndrome, minimally conscious state, coma/vegetative state); emotion-expressive suppression; migraines; multiple sclerosis; obsessive-compulsive disorder; panic disorder; Parkinson's disease; psychosis; psychotic depression; schizophrenia; sleep behavior disorder; stroke/stroke rehabilitation; trauma (eg, head injury) and related conditions listed in the policy.

Non-exhaustive list as provided in policy; ICD-10 examples for not-covered indications are listed in CPT/ICD sections of the policy.

Evidence-based coverage considerations (excerpt)

Evidence-based conclusions and limitations described in the document:

Language mapping limitations: Language fMRI (naming/verb generation) demonstrated poor sensitivity in some frontal and temporo-parietal areas (sensitivity range 22%–66% depending on task and p-value threshold) despite high specificity; combining tasks improved sensitivity. Language fMRI cannot be used alone to make critical surgical decisions in the absence of direct intra-operative brain mapping.p < 0.005 and p < 0.05 examples noted

Supports use of fMRI as adjunctive mapping, not sole determinant; task selection and analysis thresholds affect performance.

Epilepsy lateralization: Concordance of language fMRI with intracarotid amobarbital test (IAT) is variable: overall congruence ~89.3% in right temporal lobe epilepsy (rTLE) and ~72.5% in left TLE (lTLE); fMRI may miss left-hemisphere dominance in some lTLE cases and recognition of bilateral speech is difficult.

fMRI lateralization performance varies by epilepsy subtype and brain region.

Tumor presurgical planning:

Use of fMRI to identify eloquent cortex in the context of temporal neocortical epilepsy or temporal tumors is explicitly excluded and considered experimental/investigational in this policy; such applications do not meet medical necessity and would be denied.

Current reviews (e.g., UpToDate) and guideline summaries do not support routine use of fMRI for prognostication of coma in children; fMRI and other functional neuroimaging modalities are described as not useful for routine evaluation of coma prognosis pending further validation.

European and provincial reviews have concluded that non-conventional MRI techniques — specifically including fMRI, diffusion tensor MRI, magnetization transfer MRI, and MR spectroscopy — are not recommended for routine management of multiple sclerosis due to limited clinical utility.

The evidence base has important limitations across many proposed fMRI applications: prognostic uses after anoxic-ischemic injury remain investigational; studies of language and memory outcome prediction after epilepsy surgery are heterogeneous with generally low or very low quality evidence; and meta-analyses were often not feasible because of small study numbers and high between-study variability.

A meta-analysis of autism spectrum disorder (ASD) fMRI studies had significant methodological drawbacks, including sample overlap and heterogeneity, and its results do not apply to children under 5 years of age because almost none of the included studies enrolled subjects that young.

Outside the specific pre-surgical mapping indications for non-temporal brain tumors, non-temporal epilepsy, and vascular malformations, numerous other clinical uses of fMRI listed in this policy (for example, psychiatric disorders, neurodegenerative diseases, chronic pain, disorders of consciousness, stroke rehabilitation, and others) are considered experimental/investigational and are not covered.

Systematic reviews and guideline assessments (for example, Work Loss Data Institute guidance) have listed fMRI among interventions that were considered but are not recommended for chronic pain management; current evidence does not establish diagnostic or therapeutic utility for routine clinical use in this context.

Multiple reviews and clinical summaries describe fMRI as investigational or of limited utility for routine prognosis in disorders of consciousness and coma. Active fMRI may detect covert awareness in select cases, but performance and interpretation are complex and the modality is not recommended for routine prognostic use in children or adults pending further validation.

Coding

Covered CPT Codes (if selection criteria met)CPTCovered

70554	Magnetic resonance imaging, brain, functional MRI; including test selection and administration of repetitive body part movement and/or visual stimulation, not requiring physician or psychologist administration.
70555	Magnetic resonance imaging, brain, functional MRI; requiring physician or psychologist administration of entire neurofunctional testing.

Covered ICD-10 Codes (examples)ICD-10

C71.0-C71.9	Malignant neoplasm of brain
C79.31-C79.49	Secondary malignant neoplasm of brain and other parts of nervous system
D33.0-D33.2	Benign neoplasm of brain
D43.0-D43.2, D43.4	Neoplasm of uncertain behavior of brain and spinal cord
G40.001-G40.919	Epilepsy and recurrent seizures
Q28.2-Q28.3	Arteriovenous and other malformations of cerebral vessels
R56.1	Post traumatic seizures
R56.9	Unspecified convulsions

ICD-10 Codes Not Covered (examples listed as experimental/investigational)ICD-10Not Covered

F20.0-F20.9	Schizophrenia
F25.0-F25.9	Schizoaffective disorders
F29	Unspecified psychosis not due to a substance or known physiological condition
F30.10-F31.9	Bipolar I disorder
F32.3	Major depressive disorder, single episode, severe with psychotic features
F33.3	Major depressive disorder, recurrent, severe with psychotic symptoms
F41.0-F41.9	Other anxiety disorders
F42.2-F42.9	Obsessive-compulsive disorder
F84.0-F84.9	Pervasive developmental disorders
F90.1-F90.9	Attention-deficit hyperactivity disorder

1–10 of 14

1/2

inv-15: fMRI analysis p-value thresholds

Example thresholds reportedp < 0.005 and p < 0.05 were used as analysis thresholds in language fMRI studies, with sensitivity increasing and specificity decreasing at the less stringent threshold

Effect on sensitivityAt p < 0.005 sensitivity for naming and verb generation tasks was 22% and 36% respectively; combining tasks improved sensitivity to 59%

Effect on specificityAt p < 0.005 specificity was 97% (naming) and 98% (verb generation); increasing threshold to p < 0.05 decreased specificity to 91%

Trade-off summaryRaising the p-value threshold (e.g., to p < 0.05) increases sensitivity (up to 66%) but reduces specificity (to ~91%) compared with p < 0.005

Provider Actions & Billing Rules

Documentation Required

Provider Actions & Billing Rules

Aetna requires providers to document appropriate rationale and coding when requesting or billing for functional MRI (fMRI). Use covered ICD-10 diagnosis codes corresponding to the medically necessary indications (e.g., brain neoplasm, epilepsy excluding temporal neocortical epilepsy, vascular malformations) when submitting claims. Background evidence and systematic reviews cited in this policy support fMRI utility for pre-surgical mapping in selected tumor and epilepsy cases but note study limitations and heterogeneity. The policy excerpt does not specify prior authorization triggers or explicit authorization workflow in this partial document; providers should verify prior authorization requirements with the payer before service delivery.

Provider action: fMRI for temporal neocortical epilepsy or temporal tumors is considered experimental/investigational and will be denied for pre-surgical mapping of eloquent cortex in these conditions.
Provider action: fMRI for listed experimental indications (e.g., Alzheimer’s disease, ADHD, autism spectrum disorder, chronic pain including fibromyalgia, Parkinson’s disease, disorders of consciousness, emotion-expressive suppression, migraines, multiple sclerosis, psychiatric disorders such as schizophrenia, bipolar disorder, psychosis, and other non-covered indications) is considered experimental/investigational and is not covered.
Provider action: Limitations in coma prognosis — current evidence and guidance (including UpToDate reviews) state fMRI is not useful for routine prognostic evaluation of coma in children or adults and is considered investigational for routine prognostic use in disorders of consciousness.
Provider action: This section contains background evidence summaries — studies and systematic reviews cited report heterogeneous methods, small samples, and overlapping cohorts; conclusions stress limited or low-quality evidence in many areas and call for more rigorous, homogeneous studies.
Provider action: Not specified in this partial document — explicit prior authorization rules or denial triggers are not provided here; check payer-specific authorizations for CPT codes 70554 and 70555 before scheduling.
Provider action: Studies and systematic reviews referenced include prospective evaluations of pre-operative fMRI impact on surgical planning, AAN evidence assessments for temporal epilepsy, meta-analytic reviews of Wada and fMRI for language/memory outcomes, and systematic reviews for autism, migraines, and expressive suppression — all noting methodological variability and limits to generalizability.
Provider action: The excerpt contains background and limitations — clinical utility is supported for mapping eloquent cortex in selected non-temporal brain tumors, epilepsy (excluding temporal neocortical epilepsy), and vascular malformations, but many other indications remain experimental.

Definitions

inv-34: Eloquent cortex

DefinitionEloquent cortex: cortical areas that directly control critical functions such as language and sensorimotor function; injury or removal can result in major focal neurological deficits

Typical regionsExamples include Broca's area, Wernicke's area, and primary sensorimotor cortex

Clinical purposeIdentifying eloquent cortex with fMRI is used in pre-surgical planning to avoid postoperative major focal neurological deficits

inv-35: Laterality index (LI)

DefinitionLaterality index (LI): a measure reflecting the inter-hemispheric difference between activated volumes in left and right homologous regions, used to assess language dominance

Calculation contextLIs are calculated from task-based fMRI activation volumes and compared with reference tests (e.g., Wada) to predict language outcomes

Predictive useIn studies, fMRI LIs were predictive of naming decline and correlated with Wada test results (both LI and Wada were predictive of outcome)

inv-36: Language fMRI

Definition and purposeLanguage fMRI uses task paradigms (e.g., naming, verb generation) to lateralize and map language-dominant regions preoperatively

Common tasksNaming and verb generation tasks are commonly used; combining tasks can improve detection of essential language areas

LimitationsSensitivity is variable across regions and tasks (poor sensitivity in some frontal and temporo-parietal areas); fMRI should not be the sole determinant for critical surgical decisions without direct mapping

inv-37: Active fMRI (disorders of consciousness)

DefinitionActive fMRI (disorders of consciousness): task-based ('mental activation') fMRI paradigms used to detect brain activity compatible with (minimal) consciousness in behaviorally unresponsive patients

Comparative performanceIn a validation study, active fMRI had lower sensitivity (45%) and lower congruence (63%) with behavioral CRS-R diagnosis than 18F-FDG PET (sensitivity 93%, congruence 85%)

Clinical roleActive fMRI may assist differential diagnosis in selected cases but appears less accurate than PET and remains investigational pending validation

inv-38: Lateralization and memory/language prediction

Use casefMRI is used to determine hemispheric dominance for language and memory and to attempt to predict post-surgical language or memory outcomes

Evidence strengthEvidence varies by epilepsy subtype; for medial temporal lobe epilepsy fMRI may substitute for IAP for lateralizing language (Level C), but for temporal neocortical epilepsy or temporal tumors evidence is insufficient (Level U)

LimitationsMeta-analyses were often infeasible or yielded very low-quality evidence due to small numbers and high heterogeneity; prediction of outcomes remains uncertain in many subgroups

inv-39: Investigational use in disorders of consciousness

Investigational findingfMRI can detect covert awareness (brain activity compatible with consciousness) in some behaviorally unresponsive patients, but performance and interpretation are complex

Comparative accuracyActive fMRI was less sensitive and had lower congruence with behavioral assessments than 18F-FDG PET in a clinical study

Policy stanceUse of fMRI for disorders of consciousness remains investigational; results require cautious interpretation and further validation

inv-40: Expressive suppression

DefinitionExpressive suppression: the inhibition of emotion-expressive behavior (e.g., suppression of facial expressions) used as an emotion regulation strategy

Neural correlatesSystematic review findings show expressive suppression increases frontoparietal activation (dorsolateral and ventrolateral prefrontal cortices, inferior parietal cortex) and decreases temporo-occipital activation

Evidence noteFindings regarding involvement of amygdala and insula are inconsistent; conceptual and methodological issues limit current conclusions

OpenPayer

Functional Magnetic Resonance Imaging (fMRI)

Coverage Criteria

Coding

Provider Actions & Billing Rules

Background

Definitions