Poster

  • P213

Impact of rater experience on detecting MRI features of idiopathic intracranial hypertension

Beitrag in

Poster session 17

Posterthemen

Mitwirkende

Gabriel Bsteh (Vienna/ AT), Wolfgang Marik (Vienna/ AT), Stefan Macher (Vienna/ AT), Victor Schmidbauer (Vienna/ AT), Nik Krajnc (Vienna/ AT), Philip Pruckner (Vienna/ AT), Christoph Mitsch (Vienna/ AT), Klaus Novak (Vienna/ AT), Christian Wöber (Vienna/ AT), Berthold Pemp (Vienna/ AT)

Abstract

Abstract text (incl. figure legends and references)

Question: In idiopathic intracranial hypertension (IIH), certain MRI features are promising diagnostic markers, but impact of radiologist"s experience on identifying these features correctly is unknown. Therefore, we compared ratings in daily routine by radiologists with unknown awareness of IIH-MRI-features with the ratings of a junior neuroradiologist aware of features but without special IIH training and a senior neuroradiologist with experience in IIH imaging (gold-standard).

Methods: For comparing the 3 settings, we included patients from the Vienna-Idiopathic-Intracranial-Hypertension(VIIH) database with definitive IIH according to Friedman criteria and routine cranial MRI performed for suspected IIH and assessed frequencies of empty sella (ES), optic nerve sheath distension (ONSD), optic nerve tortuosity (ONT), posterior globe flattening (PGF) and transverse sinus stenosis (TSS).

Results: We evaluated MRI scans of 84 IIH patients (88% female, mean age 33.5 years). By gold-standard, 78.6% had ≥1 IIH-MRI-feature and 52.9% had ≥3 features with ONSD most frequent (64.3%) followed by TSS (60.0%), ONT (46.4%), ES (44.0%) and PGF (25.0%). Compared to gold standard, IIH features were described significantly less frequently in routine MRI reports (≥1 feature 64.3%, p=0.04; ≥3 features 15.7%, p<0.001; ONSD 28.6%, p<0.001; TSS 42.9%, p=0.04; ONT 13.1%, p<0.001; PGF 4.8%, p<0.001) except for ES (42.9%, p=0.9). Contrary, rating by a neuroradiologist without special training produced significantly higher frequencies of ≥1 / ≥3 MRI features (95.2%, p=0.001; 72.5%, p=0.017), ONSD (81.0%, p=0.015) and ONT (60.7%, p=0.049), but not ES (47.6%, p=0.6), TSS (68.1%, p=0.3) and PGF (29.8%, p=0.5).

Conclusions: IIH-MRI-features are underestimated in routine MRI reports and partly overestimated by less experienced neuroradiologists, driven by features less well known or methodologically difficile. Reevaluation of MRI scans by an experienced rater improves diagnostic accuracy.

  • © Conventus Congressmanagement & Marketing GmbH