To evaluate the longitudinal changes in multifocal electroretinography (mfERG) in patients receiving hydroxychloroquine and to assess the effects of cumulative hydroxychloroquine dose on mfERG.

Prospective cohort study.

Twenty-four eyes in 12 patients receiving hydroxychloroquine underwent mfERG recordings at baseline and 1 to 2 years later. The first negative (N1) and first positive (P1) response amplitudes and peak latencies were compared with normal controls. Serial changes in the pattern of mfERG abnormalities and in response amplitudes and peak latencies were also compared between eyes in which hydroxychloroquine therapy was continued or stopped. Correlation analyses were performed to assess the effects of a cumulative dose of hydroxychloroquine on mfERG.

At baseline, reductions in N1 and P1 response amplitudes were observed in patients receiving hydroxychloroquine compared with controls. At follow-up, in addition to the reductions in N1 and P1 response amplitudes, increases in P1 peak latencies compared with controls were observed. In patients who stopped hydroxychloroquine therapy, there were significant increases in N1 and P1 response amplitudes at follow-up mfERG.

Patients receiving hydroxychloroquine showed a longitudinal decline in retinal function; patients who stopped hydroxychloroquine therapy showed improvement. Although these data are insufficient to demonstrate the sensitivity of mfERG for evaluating early hydroxychloroquine toxicity, the results suggest that serial mfERG assessment may help detect early retinal changes associated with hydroxychloroquine therapy. Further studies with long-term results will be useful in clarifying the value of mfERG in evaluating early retinal toxicity due to hydroxychloroquine.