Affiliations: [a] Department of Neurology, Hallym University Sacred Heart Hospital, Hallym University College of Medicine, Anyang, South Korea | [b] Department of Neurology, Hallym University Dongtan Sacred Heart Hospital, Hallym | [c] ILSONG Institute of Life Science, Hallym University, Anyang, Republic of Korea | [d] Department of Occupational and Environmental Medicine, Hallym University Sacred Heart Hospital, Hallym University College of Medicine, Anyang, South Korea | [e] Hallym Institute of Translational Genomics and Bioinformatics, Hallym University Medical Center, Anyang, Republic of Korea
Correspondence:
[*]
Correspondence to: Yun Joong Kim, M.D., Ph.D., Rm #607, ILSONG Bldg, ILSONG Institute of Life Science, Hallym University, 1605-4 Gwanyang-dong, Dongan-gu, Anyang-si, Gyeonggi-do, 431-061, Korea. Tel.: +82 31 380 1666; Fax: +82 31 388 3427; yunkim@hallym.ac.kr
Abstract:
Background: Although the hummingbird sign (HBS) is a distinctive feature of progressive supranuclear palsy (PSP) VS. idiopathic Parkinson’s disease (IPD) and other parkinsonian disorders, there are no consensus criteria for its identification.
Objective: To develop consensus criteria for the HBS to enable accurate communication between researchers and physicians, we developed a new radiologic rating scale for the HBS (the HBS-RS), and the inter-rater reliability (IRR) and diagnostic validity of the HBS-RS were evaluated.
Methods: Two raters blinded to the clinical diagnoses reviewed T1 midsagittal magnetic resonance images of 133 patients with IPD (n = 93) or PSP (n = 40). The existence of the HBS was assessed in two steps that were separated by two weeks; the first was based on their own experience and the second was according to the HBS-RS. The HBS-RS comprises 4 items (contour of the third ventricle floor, shape of the beak, shape of the hummingbird head, and midbrain atrophy), with weighted scores from 0 to 2.
Results: The IRR of individual items in the HBS-RS and of the composite scores showed moderate-to-good agreement (Cohen’s kappa [κ], 0.479–0.766) and were observed to be highest for the contour of the third ventricle floor. The sensitivities and specificities varied depending on the cut-off for each item or for the composite scores. The sensitivities for each item were high (85.0–92.5) at a low cut-off (0 VS. 1 or 2). The specificities reached more than 80% when the composite scores of the HBS-RS were used. Receiver operating characteristic curves for the total HBS-RS scores showed fair diagnostic accuracy for PSP (AUC, 0.76 and 0.73).
Conclusions: The HBS-RS is a simple and measurable visual assessment tool to identify the HBS, with adjustable diagnostic validity for PSP. The results suggest that the HBS-RS may be used for objective measurements of the HBS in research and in the clinic.
