The Diagnostic Utility Of Ultrasound Bio-Microscopy For Identifying Clinical Aqueous Misdirection
Published 2026 - 44th Congress of the ESCRS
Reference: FP29.05 | Type: Free paper | DOI: 10.82333/drqp-3626
Authors: Dr Khaled Abdulaziz Alabduljabbar* 1 , Dr Abdulrahman Al-Gwaiz 2 , Ms Alia Freidi 3 , Mr Abdulrahman Al-Gaeed 4 , Dr Sirisha Senthil 5 , Dr Sushmita Kaushik 6 , Dr Deepak Edward 1 , Dr Rizwan Malik 1
1Glaucoma,King Khaled Eye Specialist Hospital & Research Center,Riyadh,Saudi Arabia, 22. Fellowship and Residency Training Program,King Khaled Eye Specialist Hospital & Research Center,Riyadh,Saudi Arabia, 33. American University of Beirut,Beirut,Lebanon, 44. Diagnostic Imaging Department,King Khaled Eye Specialist Hospital & Research Center,Riyadh,Saudi Arabia, 55. LV Prasad Eye Institute,Hyderabad,India, 6Post Graduate Institute of Medical Education and Research,6. Advanced Eye Center,Chandigarh,India
Purpose
The diagnosis of aqueous misdirection (AqM) often poses a diagnostic dilemma. Although Ultrasound biomicroscopy (UBM) is often to confirm the diagnosis, there are no studies to evaluate the utility of UBM for this purpose. The aim of this study was to estimate the diagnostic sensitivity of UBM for identifying AqM from UBM images of patients with clinical AqM.
Setting
We conducted a case-control study using previously acquired UBM images.
Methods
We conducted a case-control study using previously acquired UBM images. These (30 images for patients with AqM and 58 control images) were evaluated by 2 masked observers without clinical details. Observers classified each image as ‘misdirection, ‘not misdirection’ or ‘unsure’ for images where the diagnosis was in doubt. Cohen’s kappa (ĸ) agreement between observers was assessed. Receiver operator characteristic curves were plotted for both observers and sensitivity at 90% and partial areas under the ROC curve computed after classifying ‘unsure’ images as ‘AqM present’ or ‘AqM absent’.
Results
Eyes in the AqM group were older, with a median age of 76 years (interquartile range [IQR], 65–84), compared with a median age of 58 years (IQR, 28–70) in the control group. Across observers, sensitivity ranged from 60.0% to 80.0%, whereas specificity ranged from 77.6% to 98.3%. The overall observed agreement was 71.6%. Agreement beyond chance was moderate (Cohen’s κ = 0.442; 95% CI, 0.275–0.609; p < .001). Sensitivity ranged from 60.0% to 80.0% across reviewers and classification strategies, whereas specificity ranged from 77.6% to 98.3%. Positive predictive value ranged from 61.8% to 95.7%, and negative predictive value ranged from 81.3% to 89.8%. Overall accuracy ranged from 75.0% to 89.8%.
Conclusions
The diagnostic sensitivity of UBM for identifying eyes with clinical AqM is moderate to good, but not perfect. The main value of this investigation may be in excluding other causes of AC shallowing.