Pulvinar signal abnormalities on MRI are associated with atraumatic hip pain.

Abstract

To investigate MRI abnormalities of the pulvinar and ligamentum teres (LT) in patients with atraumatic hip pain, as well as to describe a grading system for pulvinar signal abnormalities and test its reproducibility. This retrospective study included 128 patients with atraumatic hip pain and no reported structural abnormalities other than those of the pulvinar and LT on hip MRI, along with 64 asymptomatic controls. MR images were evaluated by two readers and included grading of pulvinar signal abnormalities (normal, grade 1: < 50% fat replacement, grade 2: > 50% fat replacement), LT signal abnormality, LT thickening, and LT tear. Ordinal variables were dichotomized for further analysis. Group differences were analyzed using logistic regression models. Inter-reader agreement was assessed using kappa statistics. Pulvinar signal abnormalities were significantly associated with a higher odds ratio (OR) of hip pain (OR, 4.28; 95% CI, 1.81-11.49; p = 0.002). No significant group differences were found for LT signal increase, LT thickening, or LT tear (p-value range = 0.43-0.71). Inter-reader and intra-reader agreement for pulvinar signal abnormalities grading were almost perfect (κ = 0.85 [95% CI, 0.76-0.93] and κ = 0.92 [95% CI, 0.64-1.00]). Patients with atraumatic hip pain undergoing MRI demonstrated a significantly higher prevalence of pulvinar signal abnormalities in the absence of other relevant structural pathology. These results underscore the importance of systematically evaluating the pulvinar during hip MRI interpretation. Furthermore, a grading system for pulvinar signal abnormalities was introduced.