Abdulkadir Eren, Emrah Karatay, Rabia Mercan Sahin, Hasan Sermet
The Evans index, but not the Fazekas score, was associated with papilledema across all adjusted models, an association that persisted after adjustment for demographic and cohort variables, although reader identity and cohort membership could not be fully disentangled in this design. Its association with headache was weaker, driven largely by between-cohort differences. Papilledema was recorded as a binary clinical finding without standardized Frisén-scale grading or optical coherence tomography confirmation. These exploratory findings suggest the Evans index may warrant further prospective study beyond NPH, rather than establishing it as a validated marker of intracranial pressure.
BACKGROUND/AIM: The Evans index and Fazekas score are among the most frequently reported incidental findings on cranial magnetic resonance imaging (MRI), yet their relationship with confirmed headache and papilledema is not well characterized outside the normal pressure hydrocephalus (NPH) literature. In this exploratory, hypothesis-generating study, we evaluated this association in patients undergoing cranial MRI for headache and/or visual disturbance.
MATERIALS AND METHODS: In this retrospective study, cranial MRI of 224 patients (two non-overlapping cohorts, n=100 and n=124) was evaluated by one of two radiologists, who graded the Fazekas score and Evans index. Headache severity (0-3) and papilledema (present/absent) were obtained from clinical records. Associations were assessed with chi-square, Mann-Whitney U, Spearman correlation, and multivariable logistic regression adjusted for age and sex; a sensitivity analysis additionally adjusted for the collinear reader/cohort variable.
RESULTS: Papilledema was present in 47 patients (21.0%). The Evans index was independently associated with papilledema (adjusted OR 2.07/0.05-unit increase, 95% CI 1.27-3.33, p=0.004) and headache (adjusted OR 1.52, 95% CI 1.01-2.25, p=0.047); the Fazekas score showed no independent association with either. The papilledema association was non-significant on univariate testing (p=0.206). After adjusting for reader/cohort, the papilledema association strengthened (adjusted OR 2.76, 95% CI 1.59-4.80, p<0.001), while the headache association was attenuated (adjusted OR 1.37, 95% CI 0.89-2.09, p=0.148).
CONCLUSION: The Evans index, but not the Fazekas score, was associated with papilledema across all adjusted models, an association that persisted after adjustment for demographic and cohort variables, although reader identity and cohort membership could not be fully disentangled in this design. Its association with headache was weaker, driven largely by between-cohort differences. Papilledema was recorded as a binary clinical finding without standardized Frisén-scale grading or optical coherence tomography confirmation. These exploratory findings suggest the Evans index may warrant further prospective study beyond NPH, rather than establishing it as a validated marker of intracranial pressure.