Readily available index provides practical landmarks for glaucoma patients

Sep 12 2012

By Sarah Guy, medwireNews Reporter

The mean deviation (MD) index, a widely understood metric for summarizing damage to an individuals' visual field (VF), could help physicians estimate practical landmarks in patients with glaucoma such as fitness to drive, report UK researchers.

The results show that glaucoma patients' best eye MD (BEMD) can predict the likelihood of failing the Esterman test - the test currently used in the UK to prove fitness to drive - with an accuracy of 96.2% in receiver operating characteristic (ROC) analysis.

Furthermore, the team found that a BEMD threshold of -7.0 dB or lower could serve as a promising clinical tool since it had 95% and 82% sensitivity and specificity, respectively, to predict the probability of failing the Esterman test.

"For now, results from this study can potentially aid clinicians in making treatment decisions that will affect the QoL [quality of life] of a patient, as well as serving to remind the clinician to refer patients with binocular VF damage to undergo the Esterman test," suggest David Crabb (City University, London) and colleagues.

The study involved 5208 VF measurements from 2604 individuals with glaucoma who had MDs larger than -2.07 dB and were therefore likely to be referred for an Esterman test.

Once ROC analysis revealed that BEMD was superior to worst eye MD at discriminating those patients likely to fail the Esterman test, Crabb and team sought to identify a useful MD threshold to be used in clinical practice.

They report that with a -7.0 dB or lower threshold, the probability of failure, ie, the number of patients who are unfit to drive, was low, at 46%. Nevertheless, they add, less than 1% of patients with a BEMD above this would fail the Esterman test, giving a negative predictive value of 99%.

By contrast, expanding the threshold to BEMD -10 dB or less provided less diagnostic coverage than the -7 dB threshold, with a respective sensitivity and specificity of 80% and 94%. However, it increased the probability of failure to 70% of the cohort, "thereby offering a good compromise between coverage and accurate diagnosis," write Crabb et al in the British Journal of Ophthalmology.

The team suggests that the more conservative threshold could serve as a reminder to refer a patient for an Esterman test, while the -10 dB threshold could be useful "as a clinical trial endpoint for 'visually disabling' VF damage."

Licensed from medwireNews with permission from Springer Healthcare Ltd. ©Springer Healthcare Ltd. All rights reserved. Neither of these parties endorse or recommend any commercial products, services, or equipment.