BACKGROUND: Dimensions of the bony cochlear nerve canal (BCNC) have been proposed as a potential diagnostic standard for BCNC hypoplasia, but the standard remains inconsistent. Some studies have revealed that computed tomography virtual endoscopy (CTVE) with variation of reconstruction threshold is good at identifying labyrinthine fistulas or semi-circular canal dehiscence. PURPOSE: To examine diagnostic accuracy of VE in the evaluation of BCNC hypoplasia on basis of opening threshold. MATERIAL AND METHODS: Twenty ears diagnosed with BCNC hypoplasia were included in this study. One hundred ears served as controls. The opening thresholds were measured by two reviewers. Intraclass correlation coefficients (ICC) were reported. The width of BCNC was obtained on transverse CT images. Association between the width of BCNC and opening threshold was assessed by using Spearman correlation coefficients. ROC curves were applied to assess the best cut-off value of opening threshold. RESULTS: The opening threshold of BCNC was significantly higher in diseased ears (926 ± 308 Hounsfield units [HU]) than that in control ears (287 ± 72 HU). Inter-observer agreement was very good. The mean width of BCNC for diseased ears and control ears were 0.83 ± 0.36 mm and 2.21 ± 0.35 mm, respectively. Significant negative linear correlation was found between opening thresholds and width of BCNC (Spearman coefficient, -0.768; P < 0.001). The area under the ROC curve was 0.981. The cut-off opening thresholds of 408 for differentiation between the two groups provided the best combination of sensitivity (85%) and specificity (94%). CONCLUSION: The VE with opening threshold may be an effective optional tool for prediction of hypoplasia of BCNC.
BACKGROUND: Dimensions of the bony cochlear nerve canal (BCNC) have been proposed as a potential diagnostic standard for BCNC hypoplasia, but the standard remains inconsistent. Some studies have revealed that computed tomography virtual endoscopy (CTVE) with variation of reconstruction threshold is good at identifying labyrinthine fistulas or semi-circular canal dehiscence. PURPOSE: To examine diagnostic accuracy of VE in the evaluation of BCNC hypoplasia on basis of opening threshold. MATERIAL AND METHODS: Twenty ears diagnosed with BCNC hypoplasia were included in this study. One hundred ears served as controls. The opening thresholds were measured by two reviewers. Intraclass correlation coefficients (ICC) were reported. The width of BCNC was obtained on transverse CT images. Association between the width of BCNC and opening threshold was assessed by using Spearman correlation coefficients. ROC curves were applied to assess the best cut-off value of opening threshold. RESULTS: The opening threshold of BCNC was significantly higher in diseased ears (926 ± 308 Hounsfield units [HU]) than that in control ears (287 ± 72 HU). Inter-observer agreement was very good. The mean width of BCNC for diseased ears and control ears were 0.83 ± 0.36 mm and 2.21 ± 0.35 mm, respectively. Significant negative linear correlation was found between opening thresholds and width of BCNC (Spearman coefficient, -0.768; P < 0.001). The area under the ROC curve was 0.981. The cut-off opening thresholds of 408 for differentiation between the two groups provided the best combination of sensitivity (85%) and specificity (94%). CONCLUSION: The VE with opening threshold may be an effective optional tool for prediction of hypoplasia of BCNC.