EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

How can MRI descriptors be optimally combined to predict idiopathic intracranial hypertension?

George G Bitar iD, Philip Touska, Ata Siddiqui, Joshua Harvey, Ndidi Edi-Osagie, Haziq Chowdhury, James McHugh, Eoin O'Sullivan, Steve Connor iD

British Journal of Radiology · 2025

DOI: 10.1093/bjr/tqaf080 ↗

Kurz gefasst

Fachartikel in der Zeitschrift British Journal of Radiology, 2025. Verfasst von George G Bitar und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, University College London Hospitals NHS Foundation Trust, London, NW1 2BU, United Kingdom. Hauptthemen laut MEDLINE: Magnetic Resonance Imaging, Pseudotumor Cerebri. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: UK Research & Innovation London Medical Imaging and Artificial Intelligence Centre, Cancer Research UK National Cancer Imaging Translational Accelerator, Wellcome Trust.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objectives

To establish how MRI descriptors on standard MRI sequences can be optimally combined to predict idiopathic intracranial hypertension (IIH).

Methods

A retrospective single-institution cross-sectional study evaluated consecutive IIH patients undergoing MRI between 2002 and 2015 and a control group. Six established and 8 exploratory MRI descriptors were evaluated. Two observers independently analysed MRI descriptors on T1w sagittal and T2w axial sequences while blinded to clinical data with consensus obtained. Inter-rater reliability was calculated, and the presence of MRI descriptors was compared between IIH patients and controls (Bonferroni correction, P < 0.004). Forward stepwise logistic regression determined which combination of MRI descriptors best predicted IIH.

Results

Fifty-four IIH patients (mean age 31.2, standard deviation 10.2, 3 men) and 54 control subjects (mean age 31.7, standard deviation 7.1, 3 men) were evaluated. There was excellent inter-rater reliability for 13/14 MRI descriptors. There were 4/6 established and 6/8 exploratory MRI descriptors associated with IIH (P < 0.004). The optimal combination of descriptors to predict IIH was vertical tortuosity of the optic nerve, enlarged optic nerve sheath, globe flattening score ≥ 2, Yuh score ≥ 3, cervical skin folding, and cervical fat thickness ≥ 10.5 mm. The model correctly classified 93.5% of cases (sensitivity 94.4%, specificity 92.6%, area under the receiver operating characteristic curve [AUC] 0.965).

Conclusions

Evaluating a combination of vertical tortuosity of the optic nerve, enlarged optic nerve sheath, globe flattening, Yuh score, cervical skin folding, and cervical fat thickness optimally predicts IIH. Advances in knowledge New MRI features are validated for the diagnosis of IIH and the optimal combination for diagnosis is established. Registration number N006 (local institutional review). The full study protocol can be requested from the corresponding author.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Magnetic Resonance ImagingPseudotumor CerebriHumansMaleFemaleRetrospective StudiesAdultCross-Sectional StudiesReproducibility of ResultsMiddle AgedPredictive Value of TestsYoung Adult

Stichwörter der Autor:innen

case control studycentral nervous systemidiopathic intracranial hypertensionlogistic regressionmagnetic resonance imagingneurology

Autor:innen und Institutionen

  • Department of Radiology, University College London Hospitals NHS Foundation Trust, London, NW1 2BU, United Kingdom.
  • Department of Radiology, Guy's and St Thomas' Hospitals NHS Foundation Trust, London, SE1 9RT, United Kingdom.
  • Department of Radiology, King's College Hospital NHS Foundation Trust, London, SE5 9RS, United Kingdom.
  • Department of Ophthalmology, Moorfields Eye Hospital NHS Foundation Trust, London, EC1V 2PD, United Kingdom.
  • Department of Radiology, Manchester University NHS Foundation Trust, Manchester, M13 9WL, United Kingdom.
  • School of Biomedical Engineering and Imaging Sciences, King's College London, London, SE1 7EH, United Kingdom.

ORCID-Profile: George G Bitar, Steve Connor

Förderung

  • UK Research & Innovation London Medical Imaging and Artificial Intelligence Centre
  • Cancer Research UK National Cancer Imaging Translational Accelerator
  • Wellcome Trust
  • Guy's & St Thomas' Hospitals and King's College London
  • National Institute for Health Research Biomedical Research Centre
  • Engineering and Physical Sciences Research Council Centre for Medical Engineering at King's College London
  • Wellcome
  • Engineering and Physical Sciences Research Council Centre for Medical Engineering at King’s College London

Zitieren

APA

Bitar, G. G., Touska, P., Siddiqui, A., Harvey, J., Edi-Osagie, N., Chowdhury, H., McHugh, J., O'Sullivan, E., & Connor, S. (2025). How can MRI descriptors be optimally combined to predict idiopathic intracranial hypertension?. British Journal of Radiology. https://doi.org/10.1093/bjr/tqaf080

Vancouver

Bitar GG, Touska P, Siddiqui A, Harvey J, Edi-Osagie N, Chowdhury H, et al. How can MRI descriptors be optimally combined to predict idiopathic intracranial hypertension?. British Journal of Radiology. 2025. doi:10.1093/bjr/tqaf080

BibTeX
@article{Bitar2025,
  author = {Bitar, George G and Touska, Philip and Siddiqui, Ata and Harvey, Joshua and Edi-Osagie, Ndidi and Chowdhury, Haziq and McHugh, James and O'Sullivan, Eoin and Connor, Steve},
  title = {How can MRI descriptors be optimally combined to predict idiopathic intracranial hypertension?},
  journal = {British Journal of Radiology},
  year = {2025},
  doi = {10.1093/bjr/tqaf080}
}
Kostenlos starten →