NSRG-13. INTER-RATER RELIABILITY OF A METHOD FOR DETERMINING THE PRE-OPERATIVE HYPOTHALAMIC INVOLVEMENT OF PEDIATRIC CRANIOPHARYNGIOMA. Issue 2 (22nd June 2018)

• Record Type: Journal Article
• Authors: Whelan, Ros; Prince, Eric; Mirsky, David; Naftel, Robert; Bhatia, Aashim; Pettorini, Benedetta; Avula, Shivram; Staulcup, Susan; Cox-Martin, Matthew; Hankinson, Todd
• Is Part Of: Neuro-oncology. Volume 20:Issue 2(2018)supplement 2
• Page Start: i148
• Page End: i148
• Publication Date: 2018-06-22
• Subjects: Brain Neoplasms -- Periodicals; Brain -- Tumors -- Periodicals; Brain -- Cancer -- Periodicals; Nervous system -- Cancer -- Periodicals; 616.99481
• DOI: 10.1093/neuonc/noy059.535
• Languages: English
• ISSNs: 1522-8517

Abstract

Abstract: INTRODUCTION: Pediatric craniopharyngioma confers significant morbidity, with injury to the hypothalamus representing a particular challenge. This insight led to reconsideration of goals of surgery for certain patients. Puget et al (2007), proposed a pre- and post-operative grading system based on the degree of hypothalamic invasion on MRI. Their system defined three grades: 0, 1, and 2 where 0 represents no involvement of the hypothalamus, 2 represents hypothalamic involvement and 1 is defined as tumor abutting the hypothalamus. In a prospective cohort, they found that use of the system decreased morbidity. This system, however, has not been externally validated. We present the results of an inter-rater reliability study to predict the feasibility of this approach in the treatment of pediatric craniopharyngioma. METHODS: Six experts graded twenty-nine pre- and postoperative MRI scans according to the methods of Puget and colleagues. Inter-rater reliability was calculated using Fleiss K and Krippendorf alpha values. RESULTS: Inter-rater reliability in the pre-operative context was in the range of 0.57-0.58 (K and alpha values respectively), demonstrating only moderate agreement. Inter-rater reliability in the post-operative context was in the range of 0.39 for both methods of statistical evaluation. CONCLUSION: Inter-rater reliability for the system as defined is moderate. It may be possible that slight refinements of the MRI grading system, such as collapsing the 3 grades into 2, will allow for improvement in reliability.