MRI interpretation errors in adult patients with Medically Refractory Epilepsy

Aoife M Haughey; Nadav Gasner; Timo Krings

doi:10.3174/ajnr.A8681

Compose eLetter

Title *

Author Information

Contributors
First Name and Middle Initial * First or given name, e.g. 'Peter'. Last Name * Your last, or family, name, e.g. 'MacMoody'. Email Address * Your email address, e.g. higgs-boson@gmail.com Occupation * Your role and/or occupation, e.g. 'Orthopedic Surgeon'. Affiliation * Your organization or institution (if applicable), e.g. 'Royal Free Hospital'.

Statement of Competing Interests

Competing interests? *

Yes

Please describe the competing interests

Vertical Tabs

Jump to comment:

Commentary on "MRI interpretation errors in adult patients with Medically Refractory Epilepsy"
Timo Krings and Aoife Haughey

Published on: 18 April 2025
RE: MRI in Medically Refractory Epilepsy: Not without Postprocessing
Horst Urbach, Marcel Heers and Theo Demerath

Published on: 07 April 2025

Published on: (18 April 2025)
Page navigation anchor for Commentary on "MRI interpretation errors in adult patients with Medically Refractory Epilepsy"
Commentary on "MRI interpretation errors in adult patients with Medically Refractory Epilepsy"
Timo Krings, Neurointerventional Radiologist, Lahey Clinic UMass
Other Contributors:
Aoife Haughey, Diagnostic Neuroradiologist
We sincerely thank our colleagues for their thoughtful and detailed commentary on our recent study, and we appreciate their recognition of the importance of dedicated imaging protocols, postprocessing and integrating clinical data into MRI interpretation for patients with drug-resistant focal epilepsy.
We note with interest the comparison to their 1993–1998 cohort, which revealed a similar rate of interpretation errors (33%) as observed in our study (31.5%). Despite the passage of time and advances in MRI technology, this similarity underscores the persistent diagnostic challenges in epilepsy imaging and highlights that improved protocols must be matched by refined interpretation strategies.
As the authors rightly point out, the landscape of commonly missed lesions has shifted. Whereas hippocampal sclerosis (HS) was the most frequent missed finding in earlier studies, we observed a markedly lower rate of missed mesial temporal sclerosis (MTS) in our cohort—only 6%. This likely reflects the growing routine use of high-resolution coronal T2-weighted and FLAIR imaging acquired perpendicular to the hippocampal axis, which has improved sensitivity for MTS detection.
The evolving discussion around “no hippocampal sclerosis/gliosis only” lesions—without clear neuronal loss—is well taken. As we agree, such findings remain a diagnostic gray zone, both radiologically and histopathologically. Continued refinement of MRI criteria and correlation with histopathologi...
Show More

We sincerely thank our colleagues for their thoughtful and detailed commentary on our recent study, and we appreciate their recognition of the importance of dedicated imaging protocols, postprocessing and integrating clinical data into MRI interpretation for patients with drug-resistant focal epilepsy.
We note with interest the comparison to their 1993–1998 cohort, which revealed a similar rate of interpretation errors (33%) as observed in our study (31.5%). Despite the passage of time and advances in MRI technology, this similarity underscores the persistent diagnostic challenges in epilepsy imaging and highlights that improved protocols must be matched by refined interpretation strategies.
As the authors rightly point out, the landscape of commonly missed lesions has shifted. Whereas hippocampal sclerosis (HS) was the most frequent missed finding in earlier studies, we observed a markedly lower rate of missed mesial temporal sclerosis (MTS) in our cohort—only 6%. This likely reflects the growing routine use of high-resolution coronal T2-weighted and FLAIR imaging acquired perpendicular to the hippocampal axis, which has improved sensitivity for MTS detection.
The evolving discussion around “no hippocampal sclerosis/gliosis only” lesions—without clear neuronal loss—is well taken. As we agree, such findings remain a diagnostic gray zone, both radiologically and histopathologically. Continued refinement of MRI criteria and correlation with histopathological subtypes will be critical in better characterizing these ambiguous entities.
We also share the authors’ concern regarding focal cortical dysplasias (FCD), which have become the predominant category of overlooked lesions in our cohort. The identification of FCD remains one of the most challenging aspects of epilepsy imaging—even with optimized sequences and experienced readers. We fully endorse the necessity of post-processing approaches, such as surface-based morphometry and voxel-based analyses, with the caveat that these may be difficult to implement in clinical routines.
We appreciate the authors’ observations on amygdala enlargement, which can be related to prior seizures but also has been shown to be a primary epileptogenic lesion. Regression of this lesion over time as well as its occurence in temporal proximity to a prior seizure are the key indicators to differentiate the secondary enlargement from the primary epileptogenic lesions, and we echo the call for vigilance in interpreting such findings in context.
Finally, the debated significance of malrotated hippocampi in epilepsy is an important reminder of the nuanced nature of radiological interpretation. As with any morphologic variant, correlation with clinical, electrographic, and surgical data remains essential.
In conclusion, we wholeheartedly agree with the commentary's emphasis on the need for a comprehensive diagnostic approach. Our findings, alongside theirs, reinforce the importance of epilepsy-dedicated imaging protocols, expert review, and interdisciplinary case discussions. Crucially, they also support the integration of quantitative post-processing methods as a vital adjunct in identifying subtle cortical malformations.
Show Less

Competing Interests: None declared.
Published on: (7 April 2025)
Page navigation anchor for RE: MRI in Medically Refractory Epilepsy: Not without Postprocessing
RE: MRI in Medically Refractory Epilepsy: Not without Postprocessing
Horst Urbach, Neuroradiologist, University Medical Center Freiburg, Dept. of Neuroradiology
Other Contributors:
Marcel Heers, Epileptologist

Theo Demerath, Neuroradiologist
We appreciate the work by Haughey and co-workers emphasizing the need to integrate clinical information into the interpretation of MRI findings in patients with drug-resistant focal epilepsies. Haughey and co-workers report on 148 initial interpretation errors in a cohort of 438 patients, in whom the epileptogenic lesion was identified after acquisition of an Epilepsy-dedicated 3 Tesla MRI protocol and discussion in an interdisciplinary epilepsy conference [1]. The rate of 31.5% is comparable to our 1993-1998 study, in which 37 interpretation errors occurred in 112 patients (33%) using a similar approach [2].
Some things however have changed over the years: In the 1993-1998 study, hippocampal sclerosis was the most common overlooked lesion (47/112 =42%).
In the Haughey study, only 13 of 233 (6%) mesial temporal sclerosis (≈ hippocampal sclerosis) were missed [1]. Hippocampal sclerosis can be clearly identified on coronal FLAIR and T2-weighted 3 Tesla images which are angulated perpendicular to the long axis of the hippocampus, and we did not encounter an overlooked, histopathologically proven hippocampal sclerosis on MRI during the past 10 years. An ongoing matter of discussion is the lesion “no hippocampal sclerosis/ gliosis only” which has no neuronal loss but is poorly defined, both on MRI and histopathology [3-6].
Most overlooked lesions are now FCD, which were missed in 19/56 (33.9%) patients and some are probably “hidden” in the 300 MRI negative cas...
Show More

We appreciate the work by Haughey and co-workers emphasizing the need to integrate clinical information into the interpretation of MRI findings in patients with drug-resistant focal epilepsies. Haughey and co-workers report on 148 initial interpretation errors in a cohort of 438 patients, in whom the epileptogenic lesion was identified after acquisition of an Epilepsy-dedicated 3 Tesla MRI protocol and discussion in an interdisciplinary epilepsy conference [1]. The rate of 31.5% is comparable to our 1993-1998 study, in which 37 interpretation errors occurred in 112 patients (33%) using a similar approach [2].
Some things however have changed over the years: In the 1993-1998 study, hippocampal sclerosis was the most common overlooked lesion (47/112 =42%).
In the Haughey study, only 13 of 233 (6%) mesial temporal sclerosis (≈ hippocampal sclerosis) were missed [1]. Hippocampal sclerosis can be clearly identified on coronal FLAIR and T2-weighted 3 Tesla images which are angulated perpendicular to the long axis of the hippocampus, and we did not encounter an overlooked, histopathologically proven hippocampal sclerosis on MRI during the past 10 years. An ongoing matter of discussion is the lesion “no hippocampal sclerosis/ gliosis only” which has no neuronal loss but is poorly defined, both on MRI and histopathology [3-6].
Most overlooked lesions are now FCD, which were missed in 19/56 (33.9%) patients and some are probably “hidden” in the 300 MRI negative cases [1]. FCD were rarely identified in the 1993-1998 study, which is due to the fact that only 2D FLAIR sequences with a poor signal to noise ratio could be acquired at this time. Despite the acquisition of 3D FLAIR and other sequences highlighting the subcortical abnormalities of (type II) FCD [7-8], subtle FCD are still overlooked even by experts (Fig. 1-2). It is therefore mandatory to make these lesions visible using post-processing tools. Several post-processing tools (surface- and voxel-based) are available and have clearly increased the number of FCD [9-13]. Even FCD that are too small to be categorized on histopathology but are clearly visible on MRI can be found this way [12]. We want to stress two other issues:
1) Enlargement of the ipsilateral amygdala is detected but the FCD nearby is overlooked (Fig. 2). In these cases, amygdala enlargement is not the epileptogenic lesion but likely a consequence of focal seizures. Normalization of amygdala enlargement following the resection of an epileptogenic lesion nearby also underlines this assumption.
2) Whether malrotated hippocampi are epileptogenic lesions is at least debated as malrotated hippocampi are also found in healthy patients (15).
In conclusion, a high number of overlooked FCD does not only require Epilepsy-dedicated 3 Tesla protocols, expert’s readings and interdisciplinary case conferences but also postprocessing to highlight the subtle features of these lesions [16].

1) Haughey AM, Gasner N, Krings T. MRI interpretation errors in adult patients with Medically Refractory Epilepsy. AJNR Am J Neuroradiol published online 5 February 2025
2) Von Oertzen J, Urbach H, Jungblut S, Kurthen M, Reuber M, Fernandez G, Elger CE. Standard MRI is inadequate for patients with refractory focal epilepsy. J Neurol Neurosurg Psychiatry 2002;73:643-647
3) Thom M, Zhou J, Martinian L, Sisodiya S (2005) Quantitative postmortem study of the hippocampus in chronic epilepsy: seizures do not inevitably cause neuronal loss. Brain 128:1344–1357
4) Blumcke I, Thom M, Aronica E, et al. The clinico-pathological spectrum of Focal Cortical Dysplasias: a consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission. Epilepsia 2011; 52:158–174
5) Urbach H, Huppertz HJ, Schwarzwald R, Becker AJ, Wagner J, Bahri MD, Tschampa HJ. Is the type and extent of hippocampal sclerosis measurable on high-resolution MRI? Neuroradiology. 2014;56:731-5.
6) Hattingen E, Enkirch SJ, Jurcoane A, Kruse M, Delev D, Grote A, Becker A. Hippocampal “gliosis only” on MR imaging represents a distinct entity in epilepsy patients. Neuroradiology (2018) 60:161–168
7) Chen X, Qian T, Kober T, et al. Gray-matter-specific MR imaging improves the detection of epileptogenic zones in focal cortical dysplasia: A new sequence called fluid and white matter suppression (FLAWS). NeuroImage Clin 2018;20:388–97.
8) Middlebrooks EH, Kin C, Westerhold E. Improved Detection of Focal Cortical Dysplasia Using a Novel 3D Imaging Sequence: Edge-Enhancing Gradient Echo (3D-EDGE) MRI. NeuroImage: Clinical 28 (2020) 102449
9) David B, Kröll-Seger J, Schuch F, Wagner J, Wellmer J, Woermann F, Oehl B, Van Paesschen W, Breyer T, Becker A, Vatter H, Hattingen E, Urbach H, Weber B, Surges R, Elger CE, Huppertz HJ, Rüber T (2021) External validation of automated focal cortical dysplasia detection using morphometric analysis. Epilepsia 62:1005-1021.
10) Gill RS, Lee HM, Chaldairou B, Hong SJ, Barba C, Deleo F, D’Incerti L, Coelho VCM, Lenge M, Semmelroch M, Schrader D, Bartolomei F, Guye M, Schulze-Bonhage A, Urbach H, Cho KH, Cendes F, Guerrini R, Jackson G, Hogan RE, Bernasconi N, Bernasconi A. Automated Detection of Focal Cortical Dysplasia using deep learning: A multicenter validation study, Neurology. 2021;97:e1571-e1582
11) Spitzer H, Ripart M, Whitaker K, D'Arco F, Mankad K, Chen AA, Napolitano A, De Palma L, De Benedictis A, Foldes S, Humphreys Z, Zhang K, Hu W, Mo J, Likeman M, Davies S, Güttler C, Lenge M, Cohen NT, Tang Y, Wang S, Chari A, Tisdall M, Bargallo N, Conde-Blanco E, Pariente JC, Pascual-Diaz S, Delgado-Martínez I, Pérez-Enríquez C, Lagorio I, Abela E, Mullatti N, O'Muircheartaigh J, Vecchiato K, Liu Y, Caligiuri ME, Sinclair B, Vivash L, Willard A, Kandasamy J, McLellan A, Sokol D, Semmelroch M, Kloster AG, Opheim G, Ribeiro L, Yasuda C, Rossi-Espagnet C, Hamandi K, Tietze A, Barba C, Guerrini R, Gaillard WD, You X, Wang I, González-Ortiz S, Severino M, Striano P, Tortora D, Kälviäinen R, Gambardella A, Labate A, Desmond P, Lui E, O'Brien T, Shetty J, Jackson G, Duncan JS, Winston GP, Pinborg LH, Cendes F, Theis FJ, Shinohara RT, Cross JH, Baldeweg T, Adler S, Wagstyl K. Interpretable surface-based detection of focal cortical dysplasias: a Multi-centre Epilepsy Lesion Detection study Brain. 2022;145:3859-3871
12) Urbach H, Scheiwe C, Shah MJ, Nakagawa JM, Heers M, San Antonio-Arce MV, Altenmueller D, Schulze-Bonhage A, Huppertz HJ, Demerath T, Doostkam S. Diagnostic accuracy of epilepsy-dedicated MRI with post-processing. Clin Neuroradiol. 2023;33:709-719
13) Urbach H, Heers M, Altenmueller DM, Schulze-Bonhage A, Staack AM, Bast T, Reisert M, Schwarzwald R, Kaller CP, Huppertz HJ, Demerath T. “Within a minute” detection of focal cortical dysplasia (FCD). Neuroradiology. 2022;64:715-726
14) Shakhatreh L, Sinclair B, McLean C, et al. Amygdala enlargement in temporal lobe epilepsy: Histopathology and surgical outcomes. Epilepsia 2024;
15) Fu TY, Ho CR, Lin CH, Lu YT, Lin WC, Tsai MH. Hippocampal Malrotation: A Genetic Developmental Anomaly Related to Epilepsy? Brain Sci. 2021;11:463
16) Bernasconi A, Cendes F, Theodore WH, Gill RS, Koepp MJ, Hogan RE, et al. (2019). Recommendations for the Use of Structural Magnetic Resonance Imaging in the Care of Patients with Epilepsy: A Consensus Report from the International League against Epilepsy Neuroimaging Task Force. Epilepsia 60, 1054–1068
Show Less

Competing Interests: None declared.