To the Editor:

We thank the authors for their interest in our paper [1]. We agree that the retention rate would drop to 24.3% from 37.9% if the deceased cohort was included, however their data cannot be included in complete case analysis. Last observation carried forward is not suitable for imputation of missing data in studies of long follow-up [2]. In addition, we confirm that there were no differences in final mean visual acuity of deceased patients versus those lost to follow-up per year.

It would be useful to report the number of patients in whom treatment was rendered futile due to permanent structural damage to the macula. However, decision to stop treatment is based on the discretion of treating clinician and is not systematically recorded in electronic medical records for retrospective analysis.