Aims To estimate disease progression based on analysis of macular volume measured by spectral-domain optical coherence tomography (SD-OCT) in patients affected by Stargardt macular dystrophy (STGD1) and to evaluate the influence of software errors on these measurements.

Methods 58 eyes of 29 STGD1 patients were included. Numbers and types of algorithm errors were recorded and manually corrected. In a subgroup of 36 eyes of 18 patients with at least two examinations over time, total macular volume (TMV) and volumes of all nine Early Treatment of Diabetic Retinopathy Study (ETDRS) subfields were obtained. Random effects models were used to estimate the rate of change per year for the population, and empirical Bayes slopes were used to estimate yearly decline in TMV for individual eyes.

Results 6958 single B-scans from 190 macular cube scans were analysed. 2360 (33.9%) showed algorithm errors. Mean observation period for follow-up data was 15 months (range 3–40). The median (IQR) change in TMV using the empirical Bayes estimates for the individual eyes was −0.103 (−0.145, −0.059) mm³ per year. The mean (±SD) TMV was 6.321±1.000 mm³ at baseline, and rate of decline was −0.118 mm³ per year (p=0.003). Yearly mean volume change was −0.004 mm³ in the central subfield (mean baseline=0.128 mm³), −0.032 mm³ in the inner (mean baseline=1.484 mm³) and −0.079 mm³ in the outer ETDRS subfields (mean baseline=5.206 mm³).

Conclusions SD-OCT measurements allow monitoring the decline in retinal volume in STGD1; however, they require significant manual correction of software errors.