| Background: Of significant public health importance is the need to detect and treat people with glaucoma to prevent progression of visual loss. Given the high prevalence of undiagnosed glaucoma, research focusing on the development of potential screening tools is important. Glaucoma is characterized by the loss of retinal ganglion cells and may result in an abnormal pupillary response to light. This response can be measured with high-resolution pupillography and provides a potential method for detection of glaucoma. Compared to existing techniques, pupillography has the advantage of being low cost, does not require patient response, and can be operated by trained technicians. We conducted a prospective case-control study to assess the accuracy of detecting glaucoma using pupillography.;Methods: A total of 167 patients with glaucoma (mean age 67+/-11, 50% female) and 78 controls (mean age 60+/-10, 69% female) were recruited from April 2011 to June 2012 in a clinical setting. Using a pupillometer, we recorded pupillary light reflex (PLR) to stimulations of varied colors, intensities and patterns. The pupil waveform was divided into six pupil metrics, including amplitude, latency, maximum contraction and dilation velocities, and time to maximum contraction and dilation of the pupillary response. We combined three approaches to increase the sensitivity of this technology. The three approaches are (1) comparing pupil responses between the two eyes, (2) comparing responses to stimuli between superonasal and inferonasal visual field within each eye, and (3) calculating the absolute pupil response of each individual eye. Associative models were developed using forward stepwise selection and validated with five-fold cross validation. We further assessed the face validity of pupillography by comparing PLR to two standard tests of glaucoma diagnosis: visual field (VF) testing and retinal nerve fiber layer (RNFL) thickness measurement.;Results: Between-eye asymmetry of PLR was correlated with differences in visual field mean deviation (R=0.82) and differences in RNFL thickness between the two eyes (R=0.66). Greater VF damage and a thinner RNFL for each individual eye were associated with a smaller response amplitude, slower velocity, and longer time to peak constriction and dilation (all p<0.001 after adjusting for age and gender). Persons with glaucoma had a more asymmetric pupil response between the two eyes (p<0.001), between superonasal and inferonasal visual field within the same eye (p=0.014), and also had a smaller and slower pupil response as compared to controls (p<0.001). An associative model including age and these three components resulted in an AUROC of 0.87 (95% CI 0.82-0.91) with 79% sensitivity and 80% specificity in detecting glaucoma. This result remained robust after cross-validation.;Conclusion: Using pupillography, we were able to differentiate persons with glaucoma from those with normal eye exams. Our findings also indicate that there is an overall correlation between the pupil deficits and both structural (RNFL thickness) and functional (VF) glaucomatous changes. With refinement, pupil testing may provide a simple and relatively inexpensive approach for glaucoma screening in community settings. |
