The INVasiveness in Pituitary ADEnomas Score: Development and Validation of a Novel Radiographic Grading Scale for Pituitary Adenoma Invasion

Le résumé fourni par la source

BACKGROUND AND OBJECTIVES: A comprehensive, linear, and reliable radiographic scale for pituitary adenoma invasiveness that accounts for tumor invasion into multiple anatomic structures is needed to correlate with surgical outcomes and genetic and molecular signatures. In this study, we present the INVasiveness in pituitary ADEnomas (INVADE) score and evaluate its inter-rater and intra-rater reliability. We evaluated the INVADE score to predict the extent of resection (EOR) and functional status. METHODS: Six neurosurgeons or neuroradiology attending physician raters each individually scored 40 unique MRI and computed tomography scans of biopsy-proven pituitary adenomas. Inter-rater reliability was calculated for the total score and for each component of the score. After a 4-week washout period, 3 raters rescored each set of MRI and computed tomography scans and intra-rater reliability was evaluated. Inter-rater and intra-rater reliability values were determined using weighted Cohen's κ coefficients. Tumors were dichotomized into high (3-6) and low (0-2) INVADE scores, and the EOR and functional status were compared. RESULTS: Overall, the inter-rater reliability of the INVADE score was strong (κ = 0.725; 95% CI, 0.612-0.824). Regarding the individual components of the score, the reliability for arachnoid, cavernous sinus, and sellar face/floor invasion was strong, whereas the reliability for brain parenchyma, clivus, and lateral sphenoid bone invasion was moderate. Intra-rater reliability was nearly perfect (κ = 0.836; 95% CI, 0.713-0.909). Tumors with high (3-5) INVADE scores had significantly lower volumetric EOR (79.1% vs 96.0%, P < .001) and were more likely functional (P = .020) and outperformed the Knosp score. CONCLUSION: We present a novel scale to assess global radiographic invasiveness in pituitary adenomas. The INVADE score demonstrated strong inter-rater reliability and nearly perfect intra-rater reliability. High INVADE scores correlated with significantly lower volumetric EOR. Our results suggest that the INVADE score may be useful for future clinical and research applications.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.