Abstract
Abstract: :
Purpose: To optimize stimulus parameters for the detection of relative afferent pupillary defect (RAPD) in patients with open–angle glaucoma. Methods: We have constructed a pupillometer for measuring RAPD. It consists of two digital high–resolution video cameras (60 frames/sec), one for each examined eye, two white diode lamps allowing alternating illumination and a background infrared illumination device. Pupil area was measured in each video frame using computerized digital image analysis. The sum of right and left pupil area (Atot)was calculated. The amplitude of the pupil response (max Atot – min Atot) was computed for each stimulus cycle. Pupil area ratio was defined as the as the ratio between the mean amplitudes for right and left eye stimulation, respectively. A ratio <1 denoted a right RAPD and a ratio >1 a left RAPD. Thirty–four subjects (18 with asymmetric or unilateral open–angle glaucoma and 16 who were healthy) underwent standard clinical examination, including perimetry (Humphrey Field Analyzer II, SITA 24–2). RAPD was measured using a stimulus intensity of 1000 cd/m2. Different sets of stimulus–pause combinations were studied (0.5s – 1s, 1s – 0.5s, 1s –1s, 1.5s – 1s, and 1.5s – 1.5s). Receiver Operating Characteristic curves were calculated for each stimulus set and the areas under the curves were compared. Results: The area under the curve ranged between 0.748 and 0.894 and differed significantly from reference line (p<0.005 for all stimulus sets). The area was largest for the 1s – 1s stimulus–pause combination. Conclusions: Alternating light stimulation consisting of one second stimulation followed by one second pause was the best stimulus pattern for detecting RAPD in subjects with glaucoma. Presence of RAPD can differentiate between subjects with asymmetric glaucoma and normals with high sensitivity and specificity.
Keywords: pupillary reflex • neuro-ophthalmology: optic nerve • clinical (human) or epidemiologic studies: systems/equipment/techniques