Stargardt Macular Dystrophy: Changes in Fixation When Asked to Look Straight Ahead

To examine whether individuals with Stargardt disease macular dystrophy (STGD) change the location of fixation with instruction.

Case-control study.

Thirteen normally sighted and 37 STGD participants.

Using an Optos scanning laser ophthalmoscopy/OCT microperimeter (Optos plc, Dunfermline, UK), fixation was measured under 2 different instructions: "look at the cross" (LC) and "look straight ahead, even if you do not see the cross" (LS). Visual acuity, contrast sensitivity, disease duration, and age at disease onset were obtained from medical records.

Change in fixation with instruction.

Mean age of the STGD participants was 39.2 years, and 24 were women. Mean acuity was 1.01±0.29 logarithm of the minimum angle of resolution (logMAR), and mean contrast sensitivity was 1.16±0.41 log. The largest number of fixations under the LC condition were in the superior retina. Patients with STGD were divided into 3 groups, depending on the change in fixation locus when asked to look straight ahead: those having fixation closer the fovea, those with no change in the location of fixation, and those looking farther away from the fovea. Fifty-one eyes of 32 participants had fixations closer to the fovea when asked to look straight ahead (average change, -6.3°), whereas 13 eyes of 11 participants did not change fixation. There were no significant differences between groups in age, visual acuity, contrast sensitivity, bivariate contour ellipse area, and age at disease onset.

Despite having eccentric fixation, most STGD participants did not have a complete directional re-referencing from the fovea to the eccentric location, and moved fixation when asked to look straight ahead. This finding emphasizes that reliable assessment of visual function during evaluations of disease progression or in therapeutic intervention trials requires consistent instructions and monitoring of fixation. Otherwise, a patient's interpretation of fixation instruction may confound the results.