7 Tesla MRI in Multiple Sclerosis: Insights From Its Use in Clinical Routine

ABSTRACT

Background

7 Tesla (7 T) magnetic resonance imaging (MRI) offers higher spatial resolution and signal‐to‐noise ratio, enhancing visualization of multiple sclerosis (MS) lesions, including cortical and deep gray matter lesions. It improves detection of MS biomarkers like paramagnetic rim lesions (PRLs) and central vein sign (CVS). Costs have impacted its adoption and experience in clinical practice.

Objectives

To present real‐life data on the routine clinical use of 7 T MRI and its impact on patient management from a single‐center perspective.

Methods

This retrospective study, approved by the local ethics committee (KEK Bern No 2020–02902), analyzed referrals for 7 T MRI (06/2020–06/2024) at University Hospital Bern for suspected CNS inflammatory disorders. Imaging reports were compared to clinical data from medical records. Statistical analysis evaluated the diagnostic value of 7 T MRI, focusing on sensitivity, specificity, Negative Predictive Value (NPV), and Positive Predictive Value (PPV). Exclusions included contraindications for 7 T MRI, incomplete medical records, or non‐CNS conditions.

Findings

61 patients underwent 7 T MRI, enabling lesion reclassification and MS diagnosis in 19/47 patients with indefinite diagnosis despite extensive diagnostic workup with adequate 3 T MRI. In 14 MS patients, it clarified diagnostic uncertainties, leading to diagnosis revision in 1/14 patients and informed treatment decisions in 4/14 (including treatment escalation (3/14) and discontinuation (1/14)). 7 T MRI showed 89.5% sensitivity and 78.6% specificity for MS (PPV 73.9%, NPV 91.7%). MS patients were more likely to exhibit CVS and PRLs compared to non‐MS patients (p < 0.05).

Interpretation

7 T MRI enhances MS diagnosis certainty in diagnostically challenging cases, potentially impacting clinical practice.

Insights from the impact of 7T MRI in routine clinical care in patients with not yet clearly defined neuro‐inflammatory disorders of the central nervous system or in MS patients with uncertainties during routine clinical care.

1. Introduction

The first investigative ultrahigh‐field (UHF) 7 Tesla (7 T) magnetic resonance imaging (MRI) studies in patients with multiple sclerosis (MS) were published in 2008 [1, 2, 3]. The higher field strength of 7 T MRI enables dedicated sequence selection and provides superior spatial resolution and signal‐to‐noise ratio compared to conventional 1.5 and 3 T systems [4, 5]. These technical advantages allowed, for the first time, the visualization of findings characteristic of MS pathology, previously limited to histopathological diagnostics [1, 2, 3]. These included visualization of smaller and more subtle lesions, such as those in the cortical and deep gray matter structures [4, 6, 7], which may be missed at lower field strengths. Beyond lesion detection, 7 T MRI can improve lesion characterization by depicting changes in lesion morphology, iron deposition, and processes associated with microstructural integrity [6, 8]. The increased sensitivity to susceptibility effects, enabled the detection of lesion distribution along a central vein, termed “central vein sign,” CVS [1, 2]; first described by Dawson in 1916 [9], and evidence of iron depositions in MS lesions [3, 8, 10], which had previously been identified in histology as iron‐rich macrophages/microglia [11]. These iron depositions in MS lesions reflect a dynamic process that is more typical of MS than, for instance, neuromyelitis optica (NMO) [12]. Over time, iron‐laden MS lesions can evolve from a nodular to a ring pattern (paramagnetic rim lesions, PRLs) or the other way around and these changes can be better depicted using 7 T cerebral MRI (MRI) [13].

While these findings were subsequently translated to lower‐field MRI [14, 15], studies comparing 7 T with 3 T MRI have consistently demonstrated the superiority of 7 T in the detection of cortical lesions, CVS, and PRLs [16, 17, 18]. These findings can have an impact on the clinical management of MS patients. Cortical lesions appear to play a crucial role in disease progression, contributing to cognitive impairment and disability, and have been postulated as a diagnostic and prognostic imaging biomarker [6, 19, 20]. Moreover, PRLs, which appear as rings of iron‐laden microglia around chronic lesions, have been associated with persistent and slowly progressive inflammation and may serve as a biomarker for smoldering disease progression or progression independent of relapse activity (PIRA) [20]. Such imaging biomarkers can assist in differentiating MS subtypes, predicting clinical outcomes or treatment response. Therefore, the clinical application of 7 T MRI in MS has the potential to enable earlier and more accurate diagnosis, as well as improve disease understanding and monitoring [21, 22].

Despite its capabilities, clinical implementation of 7 T MRI has been limited. Although FDA approval of a commercial 7 T scanner in 2017 [23] permitted its use in clinical practice, access remains restricted due to high investment and maintenance costs, as well as availability and specific safety considerations. To date, in MS, 7 T MRI has not been widely integrated into routine diagnostic workflows, and its impact on clinical management remains uncertain, despite more accurate detection of MS‐related pathologies. Nonetheless, its role as a valuable clinical tool for MS is increasing. Recently, the North American Imaging in Multiple Sclerosis (NAIMS) Cooperative published a statement on the future use of 7 T MRI in MS, but focused primarily on the use of 7 T MRI in MS research [24]. A recent publication tried to address the use of 7 T MRI in clinical routine from single‐center experience, but focused primarily on a technical perspective, including detailed protocols and sequence parameters [25].

In MS and related disorders, two prominent settings for the clinical use of 7 T MRI are (i) initial diagnostic workup in unclear, diagnostically challenging cases when MS is suspected and (ii) aspects suggestive of differential diagnoses in patients with diagnosed MS, like relevant comorbidities, unclear progression despite adequate immunotherapy/rehabilitative treatments, and clinical/imaging ambiguities despite adequate diagnostic workup with conventional 1.5 or 3 T MRI.

2. Objectives

The objective of our study is to present data from the clinical routine, as opposed to a research environment and patients enrolled in imaging studies, and to provide an explorative assessment of potential implications of 7 T MRI on patient management in these clinical scenarios: (i) initial diagnostic with unclear diagnostically challenging cases; (ii) diagnosed MS with unclear progression, comorbidities suggestive of an alternative diagnosis, or clinical/imaging ambiguities despite adequate diagnostic workup.

3. Methods

3.1. Ethics Statement

This retrospective analysis was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki and was approved by the local ethics committee (KEK Bern No 2020–02902).

3.2. Artificial Intelligence Generated Content (AIGC) Statement

No content created by generative artificial intelligence was used for this manuscript.

3.3. Sample Identification

Referrals for 7 T MRI at the University Hospital of Bern (Bern, Switzerland) between June 2020 and June 2024 were retrospectively screened, including those from both internal and external neurologists, to assess inflammatory disorders affecting the central nervous system (CNS). Only patients with at least one complete neurological report before and after the 7 T MRI were included. As per clinical standard procedures, all patients had already undergone at least one 3 T MRI with state‐of‐the‐art MS protocol before the 7 T MRI. All data were collected during clinical routine and retrospectively extracted from the electronic medical records (EMRs). Diagnostic criteria for MS and clinically isolated syndrome (CIS) were based on the 2017 McDonald Criteria [26], as all data were acquired before the 2024 update [27].

A total of 78 referrals for 7 T MRI due to suspected inflammatory disorders were identified (selection process illustrated in Figure 1). Reasons for exclusion of cases included inability to undergo 7 T MRI due to contraindications (n = 1; prior neurosurgical intervention), explicit withdrawal of general research consent (n = 6), referral for workup of non‐CNS inflammatory disorders (n = 1, orbital myositis; n = 1, cranial polyradiculitis), and insufficient clinical or imaging data in EMRs (“lost to follow‐up”; n = 8).

FIGURE 1.

Selection process of the patients scheduled for ultrahigh‐ field (7 T) magnetic resonance imaging (MRI).

3.4. MRI Protocol and Analysis

Clinical examination at 7 T was performed on a MAGNETOM Terra scanner (Siemens Healthineers AG, Forchheim, Germany) with a 1Tx/32Rx head coil (Nova Medical, Wilmington, MA). For routine acquisition at 3 T at our institution, we used either a MAGNETOM Vida scanner (Siemens Healthineers AG, Forchheim, Germany), a MAGNETOM Skyra Fit scanner (Siemens Healthineers AG, Forchheim, Germany), or a MAGNETOM Prisma scanner (Siemens Healthineers AG, Forchheim, Germany). As per international recommendations [28], whenever possible, follow‐up 3 T MRIs were performed using the same scanner. Standardized MS protocols for 7 T and 3 T are described in detail in Supplementary Tables 1 and 2.

Routine clinical MRI interpretation was performed by two readers and documented in the report and EMR. 7 T MRI interpretation was performed by two neuroradiologists experienced in 7 T MRI readings, following standard clinical practice and in alignment with patient‐oriented services. All findings were cross‐referenced in all three planes and across multiple contrasts (T1, T2, FLAIR, SWI) according to best radiological practice. All data were collected during clinical routine and retrospectively extracted from the EMRs. Secondary interpretation of the imaging was waived to maintain a realistic clinical environment for this study. During the image interpretation of MS‐typical localizations in the brain (periventricular, infratentorial, cortical/juxtacortical), the presence of central vein signs (CVS) and paramagnetic rim lesions (PRLs) were systematically analyzed.

3.5. Statistical Analysis

Chi‐Square tests were used to analyze associations between sex and diagnostic group (initial vs. follow‐up diagnostics) and to assess the relationship between CVS/PRLs and MS diagnosis in newly diagnosed patients (initial diagnostics group). To evaluate the diagnostic value of 7 T MRI statistical metrics such as sensitivity, specificity, Negative Predictive Value (NPV), and Positive Predictive Value (PPV), the most recent clinical diagnosis, based on the 2017 McDonald Criteria [26], was used as the gold standard. Data processing and analysis were conducted in R 4.3 and jamovi 2.5 with the consort [29] and networkD3 [30] packages.

4. Results

4.1. Demographic Characteristics

Sixty‐one patients with a suspected or confirmed inflammatory CNS disorder on 3 T MRI underwent 7 T head MRI for further evaluation and were included in this study's analysis. Patient demographics are summarized in Table 1. The female:male ratio was 2.4:1 with a non‐significantly higher ratio among the “early diagnostic” patients (2.9:1; p = 0.35). Clinical information, cerebrospinal fluid (CSF) findings, and details on disease modifying treatments (DMTs) are provided in the supplementary tables (Supplementary Tables 3 and 4).

TABLE 1.

Demographic characteristics of included patients.

	n	Initial diagnostics n = 47	Follow‐up diagnostics n = 14
Age (Years) †	61	39.7 (17–65)	43.8 (26–68)
Sex‡	61
Female		35 (74.5%)	8 (57.1%)
Male		12 (25.5%)	6 (42.9%)

^† Mean (Min ‐ Max), ^‡ n (%).

4.2. 7 T MRI Early in Diagnostic Workup (Initial Diagnostics Group)

Forty‐seven cases (77%) had diagnostic ambiguity or lacked a clear, definitive diagnosis despite adequate 3 T MRI (Figure 2, left panel) and were therefore referred for 7 T MRI early in the diagnostic workup for further evaluation (“initial diagnostics”; details in Supplementary Table 3).

FIGURE 2.

Initial diagnostics group: Patients were categorized according to the initial syndromal diagnosis (left panel; details in Table S3). 7 T MRI findings (middle panel) were categorized according to interpretation between findings suggestive or not suggestive of Multiple Sclerosis (MS vs. rad. not MS). Final category (right panel) was differentiated between MS and not MS according to final diagnosis in the last follow‐up after 7 T MRI. Abbreviations: CI‐CNS = chronic inflammatory disorder of the central nervous system; CIS = clinically isolated syndrome; MS = Multiple Sclerosis; RIS = radiologically isolated syndrome.

Twenty‐two patients (47%) had imaging findings compatible with MS (Figure 2, middle panel). Overall, 19 patients (40.4%) were diagnosed with MS during clinical follow‐up (mean: 15 months after 7 T MRI; range: 0–44 months), including three (6.4%) with a progressive form of the disease. The remaining patients exhibited a chronic inflammatory profile on CSF analysis (Supplementary Table 3), but none had objective findings, such as MRI lesions or pathological electrophysiological results, consistent with their reported symptoms.

The remaining 25 patients (53%) did not have imaging findings indicative of MS. Of these, eight (32%) had a history of myelitis but no cerebral lesions, and nine (36%) had white matter hyperintensities of an unspecific or presumed vascular etiology. Five patients (20%) had lesions in only one of the four MS‐typical locations according to the 2017 McDonald criteria, without any additional suggestive findings such as CVS or PRLs.

In relation to the initial diagnosis, MS was diagnosed in 13 of 19 patients (68.4%) with an indefinite chronic inflammatory CNS disorder (CI‐CNS), five of seven (71.4%) with CIS, and one of six (16.7%) with a history of myelitis.

Overall, 7 T MRI demonstrated a sensitivity of 89.5% (95% CI: 65.5%–98.2%) and a specificity of 78.6% (95% CI: 58.5%–91.0%) for the diagnosis of MS. PPV was 73.9% (95% CI: 51.3%–88.9%) and NPV was 91.7% (95% CI: 71.5%–98.5%). Compared to non‐MS patients (Figure 2, right panel), those newly diagnosed with MS were more likely to have CVS (n = 12 vs. n = 5; p < 0.01; exemplary depiction of a CVS in Supplementary Figures 1 and 6) and PRLs (n = 8 vs. n = 2; p < 0.01; exemplary depiction of PRLs in Supplementary Figures 6 and 7).

Further, we analyzed the performance in the detection of demyelinating lesions according to the classical localizations in MS (Figure 3). Juxtacortical/cortical (J/C) lesions (Figure 3, left image; exemplary depiction of a J/C lesions in Figures S2–S5) were detected in 17 patients (36.2%) on 7 T MRI and in 14 (29.8%) on previous 3 T MRI. Compared to 3 T MRI, 7 T MRI enabled the detection of J/Cions in seven additional patients (14.9%) but also led to the reclassification of lesion localization in four patients (8.5%). Periventricular (PV) lesions (Figure 3, middle image) were identified in 27 patients (57.5%) on both 7 T and 3 T MRI. However, 7 T MRI prompted a revision of 3 T MRI findings in ten patients (21.3%), with five no longer meeting the criteria for PV lesions and five additional patients newly identified with PV lesions. Infratentorial (IF) lesions (Figure 3, right image) were detected in 15 patients (31.9%) on 7 T MRI and in eight (17.0%) on 3 T MRI. 7 T MRI allowed the identification of IF lesions in ten additional patients (21.3%), and in three patients (6.4%), initially presumed IF lesions were not confirmed on 7 T MRI.

FIGURE 3.

Lesion localization in patients from the “initial diagnostics” group for MS‐typical regions. On the left side of each graphical column are the findings according to 3 T MRI and on the right according to 7 T MRI. The upper and lower segments represent the number of patients with or without lesions in the described localizations, respectively. The curved flows (lighter colors) in the middle represent patients with revised findings on 7 T MRI with evidence of the presence or absence of lesions in the corresponding localizations, respectively. Details in text. Abbreviations: IF = infratentorial; J/C = juxtacortical/cortical; MS = Multiple Sclerosis; PV = periventricular.

4.3. 7 T MRI for Clarification of Uncertainties During Follow‐Up (Follow‐Up Diagnostics Group)

Fourteen (23%) patients with an established MS diagnosis (mean disease duration: 5.9 years; range: 0–14 years; Figure 4, left‐side panel) underwent 7 T MRI for detailed imaging follow‐up to resolve clinical ambiguities despite adequate 3 T MRI. The most common reason was reevaluation (ten patients, 64.3%) due to clinical uncertainty regarding their diagnosis: four (28.6%) only marginally met the imaging criteria for MS on previous 3 T MRI; three (21.4%) had blood test abnormalities, clinical signs, or comorbidities suggestive of an alternative diagnosis (e.g., headaches with elevated inflammatory markers, antiphospholipid antibodies, inflammatory bowel disease, and microangiopathic brain lesions), and two (14.3%) had tumefactive lesions despite adherence to DMT (sphingosine‐1‐analogon, B‐cell depletion—BCDT). The remaining four patients (28.6%) exhibited significant clinical progression despite BCDT. 7 T MRI influenced the clinical management of five patients (35.7%), leading to DMT escalation in three (21.4%), treatment de‐escalation in one (7.1%), and a revised diagnosis in one patient (7.1%; Figure 4, right‐side panel).

FIGURE 4.

Follow‐up diagnostics group: Patients with a diagnosis of Multiple Sclerosis (MS) were referred to 7 T MRI for detailed imaging due to uncertainties during clinical follow‐ups (details in text). Imaging findings on 7 T MRI (middle panel) were classified between radiological findings typical of MS (imaging suggestive of MS) and not specific to MS (imaging not suggestive of MS), additionally one patient (middle flow of the Sankey diagram) was deemed to have typical findings of MS but also findings compatible with a comorbidity (antiphospholipid syndrome, “coagulopathy”).

5. Discussion

UHF MRI has enabled superior visualization of findings characteristic of MS pathology, leading to a better understanding of the disease's pathophysiology [1, 2, 3]. Despite the technical advantages of UHF MRI compared to conventional MRI [4, 5], clinical implementation has remained limited due to various challenges. These include the need for specialized training, high costs, and the integration of new advanced imaging protocols into existing workflows [31]. Moreover, the clinical benefit of implementing 7 T MRI in routine practice has remained unclear [32].

Our study highlights the potential use and clinical implications of 7 T MRI findings in routine clinical practice for neuroimmunological disorders, viewed from an interdisciplinary perspective at our institution. This may be of utility to both radiologists and neurologists. As demonstrated in our study, the enhanced imaging capabilities of 7 T MRI may provide critical insights in patients with white matter lesions or suspected neuroimmunological disorders, particularly by enabling earlier MS diagnosis and differentiation between MS and its mimics. The latter is especially important, as it can prevent inappropriate management and significant patient distress [33]. Furthermore, in patients already diagnosed with MS but presenting with clinical ambiguities, 7 T MRI can help clarify these ambiguities, assess disease progression, and monitor treatment response, thereby informing diagnostic decision‐making and improving patient management. By facilitating more accurate detection and characterization of lesions, 7 T MRI could play a pivotal role in reducing diagnostic uncertainty in conditions like MS and other demyelinating diseases.

The clinical implications of 7 T MRI hold the potential to enhance diagnostic accuracy in specific cases of multiple sclerosis. Iron‐sensitive imaging at 7 T MRI excels in visualizing the central vein sign and paramagnetic rim lesions, both of which are relevant for MS diagnosis. Dedicated sequences, such as MP2RAGE, have demonstrated superior detection of juxtacortical and cortical lesions at 7 T, but have also significant potential when used at lower field strengths [17, 34, 35, 36]. Current diagnostic criteria are based on 1.5 T and 3 T MRI, but a potential approach could involve applying the same criteria to 7 T MRI.

Challenging cases, particularly individuals with comorbidities, may benefit from the use of 7 T MRI. One such group includes individuals aged ≥ 50 years who are being evaluated for a potential MS diagnosis. MRI in this age group often reveals lesions deemed of vascular origin or findings indicative of manifest vascular disease. Consequently, the 2024 revisions of the McDonald criteria underscore the importance of additional diagnostic features, like the number of CVS [27]. In this context, 7 T MRI enhances diagnostic precision by providing superior visualization of lesions, allowing a better differentiation between vascular etiology and MS. This leads to more accurate diagnoses, informed clinical management, and ultimately, better patient outcomes, as it enables clinicians to tailor treatment plans that address the specific underlying pathologies. In our cohort, one of four individuals aged ≥ 50 years from the follow‐up diagnostics group had an increase of lesion load despite BCDT. On 7 T MRI, the lesion load was primarily attributed to microangiopathy (Figure 4, light green flow) leading to the discontinuation of BCDT.

While our study provides valuable insights into the potential clinical utility of 7 T MRI outside of clinical trials or scientific environments, our study has limitations. The main limitation is the retrospective design of our study and potential bias. Despite screening all referrals for 7 T MRI for inflammatory disorders affecting the central nervous system (CNS), many patients with challenging cases both at our institution and from external neurologists are not systematically referred for 7 T MRI.

The decision not to perform a systematic secondary radiological interpretation of imaging findings did not allow for the comparison of the diagnostic accuracy of 7 T versus 3 T MRI for detecting PRLs and CVS; yet, it was not the scope of our work designed primarily to report on clinical practice. Moreover, already a robust body of evidence confirmed the superiority of 7 T MRI in radiological outcomes—the detection and visualization of these findings [24, 37, 38]. This may have implications for the broader adoption of 7 T MRI in clinical practice, particularly as upcoming revisions to the McDonald criteria are expected to incorporate the quantification of CVS lesions and the presence of PRLs [27].

Although standard imaging remains sufficient for most clinical MS cases, the increasing availability of 7 T MRI, particularly in specialized academic centers, necessitates the establishment of clear workflows and definitions for patient selection. Specifically, 7 T may be most beneficial in cases where the initial diagnosis remains uncertain despite adequate 3 T imaging. In conclusion, 7 T MRI should be further explored for its diagnostic utility in unclear or challenging MS cases and unexplained clinical progression. Despite the limitations, our study generates hypotheses for future research, which should focus on patient‐centered outcomes in clinical settings or studies designed to mimic real‐world scenarios.

In conclusion, the integration of 7 T MRI into clinical practice for neuroimmunological disorders, particularly MS, presents significant opportunities for enhancing diagnostic accuracy and patient management.

As the field of neuroimaging continues to evolve, the establishment of standardized workflows and clear criteria for patient selection will be essential for maximizing the clinical benefits of 7 T MRI.

Author Contributions

Conceived and designed the analysis: A.L.B., F.M., P.R. Collected the data: A.L.B., F.M., R.W., R.P. Contributed data or analysis tools: A.L.B., A.C., R.W., R.H., H.H., P.R. Performed the analysis: A.L.B., F.M., P.R. Wrote the paper: A.L.B., R.R., A.C., F.W., G.B., R.H., H.H., P.R.

Disclosure

Patient and Public Involvement: Patients and/or the public were not involved in the design, conduct, reporting, or dissemination plans of this research.

Ethics Statement

Use of clinical and imaging data was approved by the local ethics committee (KEK Bern No 2020–02902).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: Supporting Information

León Betancourt A., Messmer F., Chan A., et al., “7 Tesla MRI in Multiple Sclerosis: Insights From Its Use in Clinical Routine,” European Journal of Neurology 32, no. 8 (2025): e70330, 10.1111/ene.70330.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.