Investigating hyperostosis frontalis interna: a computed tomography analysis and predictive model for Hershkovitz classification

Background: Axial computed tomography (CT) cross-sections offer an accessible model for assessing diverse pathologies associated with hyperostosis frontalis interna (HFI) based on the Hershkovitz classification. Purpose: To delineate the CT characteristics of HFI, emphasizing a radiological description using the Hershkovitz classification. It investigated whether the Hershkovitz classification can be predicted using density, a variable unexplored in the literature. HFI thickness and the presence of concurrent hyperostosis beyond the frontal bone (HBFB) were evaluated across varying degrees of the Hershkovitz classification. Material and Methods: This retrospective study evaluated characteristics of HFI on CT. Each case was assigned a Hershkovitz classification. Density and thickness of the HFI along with the presence of concurrent HBFB were evaluated and correlated with the Hershkovitz classification. Results: Axial CT of 77 patients with HFI was evaluated. Patient characteristics including sex were uncorrelated with CT measurements (P > 0.25) and Hershkovitz classification (P > 0.06). Increasing HFI thickness was associated with a higher Hershkovitz classification (odds ratio [OR] = 1.863, 95% confidence interval [CI] = 1.452–2.389; P < 0.001), and increasing density of HFI was associated with a lower Hershkovitz classification (OR = 0.995, 95% CI = 0.992–0.998; P = 0.002). Higher Hershkovitz classification was also found to be associated with the presence of concurrent HBFB (OR = 31.694, 95% CI = 6.483–154.938; P < 0.001). Conclusion: In our cohort, Hershkovitz classification on axial CT correlated with increased HFI thickness, presence of HBFB, and lower HFI density.