Aim To compare standard short-wavelength fundus autofluorescence (SW-FAF) and near infrared-wavelength fundus autofluorescence (NIR-FAF) in detecting geographic atrophy (GA) secondary to age-related macular degeneration, and its retinal sensitivity impairment.

Methods Twenty-five consecutive patients (36 eyes) affected by GA were studied by means of fundus autofluorescence imaging, using both SW-FAF (excitation: 488 nm, emission >500 nm) and NIR-FAF (excitation: 787 nm, emission >800 nm). All patients underwent microperimetry to assess fixation characteristics and retinal sensitivity.

Results In the extrafoveal region, the total hypoautofluorescent (hypo-FAF) area was significantly wider with NIR-FAF than with SW-FAF (8.03±6.68 mm² vs 7.37±6.34 mm² respectively; p=0.005). In the foveal area, the total hypo-FAF area was smaller with NIR-FAF than with SW-FAF (0.19±0.03 mm² versus 0.42±0.12 mm² respectively; p=0.008). Foveal sparing was larger at NIR-FAF compared with SW-FAF (p=0.021). In nine cases (25%) the site of fixation was hypoautofluorescent on SW-FAF, but normal on NIR-FAF with preserved retinal sensitivity.

Conclusions Standard SW-FAF may overestimate GA in the foveal area, correctly detected by NIR-FAF. In the extrafoveal area, SW-FAF may underestimate GA. Standard SW-FAF should be integrated with NIR FAF when detecting and following GA to avoid inconsistent results and misinterpretation, from both a morphological and functional perspective. Microperimetry helps to quantify retinal sensitivity in GA.