Diagnostic Accuracy of B-scan Ultrasonography for Posterior Segment Eye Disorders -A Cross-sectional Study

Abstract

Introduction: Ophthalmic Ultrasound (USG) produces real time high resolution images of the eye and orbit. It can categorise, predict location and diagnose posterior segment pathology of the eye very well and is usually used in preoperative evaluation. However, the diagnostic accuracy, in terms of predicting probability of Posterior Segment Disease (PSD) post-USG, has usually not been studied, that too in Indian set-up. Hence, this study was conducted. Aim: To find out the prevalence and pattern of PSD among patients referred to radiology department using B-scan USG, and to find out the diagnostic accuracy of B-scan USG in such cases. Materials and Methods: The study was cross-sectional in nature conducted over a period of two years (2007-09) and conducted in the Department of Radiodiagnosis and Ophthalmology of a tertiary care centre of Eastern India. Patients were referred to the Radiology department for ruling out intraocular pathology using B-scan ophthalmic USG were included in the study, irrespective of any age and gender. Data were captured on an excel sheet and analysed using Stata 12.1. Prevalence and pattern of PSDs were established from the data while diagnostic accuracy was calculated in terms of sensitivity, specificity, predictive values and log Likelihood Ratios (LR). Accuracy of B-scan USG was also estimated and adjusted for verification bias. Results: The mean age of 84 study participants was 37.4±19.5 years, with maximum in between 41-50 years. Males predominated (72.6%). A 50% presented with low vision, and most commonly associated with cataract 45 cases (54%). Prevalence of posterior segment eye disorders was 13.1%. Adjusted sensitivity, adjusted specificity and Negative Predictive Values (NPV) were 100%. Positive Predictive Value (PPV) was 45.33%. Post-test probability was 46.6%, while after adjustment it was infinite. Adjusted accuracy was 100%. Conclusion: Prevalence of posterior segment eye disorders was 13.1% for cases referred to radiology department. The most common PSD that was found was retinal mass. Using B-scan USG for preoperative assessment and confirmation of diagnosis increases the probability of detecting presence or absence of posterior segment pathology. Absence of PSD using this is also very helpful in ruling out disease entirely. It also has a very high sensitivity and hence has a scope to be used even in rural health centres.

Keywords

• eye abnormalities
• predictive value of tests
• prevalence
• sensitivity and specificity