Computer Reduces Torsion Measurement Bias

Laura B. Bruck
Managing News Editor

Published in Ophthalmology Times

Kiawah ISLAND, SC-- A computer-automated test may prove valuable in the diagnosis and treatment of cyclovertical strabismus by providing measurements of torsion in all diagnostic positions of gaze independent of examiner bias, reported H. Jay Wisnicki, MD, here at the 15th annual meeting of the American Association for Pediatric Ophthalmology and Strabismus.

Ocular torsion measurements are extremely useful in both the diagnosis and treatment of cyclovertical strabismus. But "obtaining precise reproducible measurements of cyclodeviations in the positions of gaze is both difficult and time consuming when traditional methods are employed," said Dr. Wisnicki, director of ophthalmology, Beth Israel Medical Center, New York.

While methods such as Hess screen and Lancaster red-green testing can measure deviations in various positions, the examiner's subjective recording of torsion introduces an element of bias.

"But by automating both the administration of the test as well as the documentation of the measurements, the red-green test can be used to determine torsion rapidly and objectively in all nine positions of gaze." He reported reproducibility of nearly 2' after obtaining over 50 measurements with the automated system.

The red-green test is traditionally administered with a hand held projector, said Dr. Wisnicki, who is also assistant professor of ophthalmology, Mt. Sinai School of Medicine. In the automated system, however, the test is administered by an Apple MacIntosh II computer equipped with a high-resolution color monitor that displays the visual targets to the patient.

Images Separated

Seated 1 meter from the color screen in a darkened room, the patient uses video-game-style buttons to control the horizontal and vertical positions as well as the tilt of one red and one green line displayed on the monitor, he explained. Red-green glasses are worn to separate the images to each eye.

The patient is then asked to superimpose the two images. The computer calculates the angle between the lines as an objective recording of the subjective torsion and also stores horizontal and vertical misalignments. The measurements are repeated for the nine diagnostic positions of gaze.

Some Drawbacks

Like the traditional test, the automated version is not able to provide precise control of head position, and the patient's head must be moved to obtain measurements in different positions of gaze. The results, however, compare favorably with those obtained by traditional methods, said Dr. Wisnicki.

He cited the case of a 31-year-old patient with a 16-month history of diplopia following a motor vehicle accident. Despite excellent corrected acuity, the patient complained of difficulty reading that was especially evident on downward left gaze.

Torsion testing revealed a slightly underacting right superior oblique and overacting right inferior oblique. Clinically, the patient demonstrated cyclodeviation. Dr. Wisnicki noted that deviations were recorded as a 6 prism diopter right hypertropia that increased to 10 prism D with downward left gaze and minimal hypertropia on head tilt. Fundus excyclotorsion was evident on indirect ophthalmoscopy.

Computer testing revealed 9' of excyclotorsion in the primary position, which increased to 18' on downward left gaze. The subsequent diagnosis of bilateral superior oblique paresis enabled successful surgical correction, and a second computerized test performed postoperatively showed minimal cyclodeviation in the various positions of gaze. At 6 months, reported Dr. Wisnicki, the patient is essentially asymptomatic, with minimal right hyperphoria that is easily controlled.

Coauthor in this work was Bari M. Brandt, MD, resident in ophthalmology, Mt. Sinai School of Medicine, New York.

OPHTHALMOLOGY TIMES / JULY 15,1989
Reprinted from OPHTHALMOLOGY TIMES, Vol. 14, No. 14, pp. 1 & 26, July 15, 1989
Copyright 1989 by Edgell Communications, Inc.