ABSTRACT Purpose To evaluate the efficacy of atropine penalization after non-response to patch therapy in amblyopic children and investigate the factors associated with treatment success. Patients and Methods In this retrospective study, 26 children with amblyopia who were non-responders to patch therapy who were then switched to 1% atropine eye drops in the sound eye for a minimum follow-up of one year were included. All patients underwent detailed eye examinations, including optical coherence tomography and fundus autofluorescence (FAF) imaging. Response to treatment was defined as a two-line improvement in best-corrected visual acuity (BCVA) in the amblyopic eye, and patients were divided into two groups: the responder group and the non-responder group. Demographic and clinical parameters were compared between the two groups. The average central macular thickness and FAF were analyzed. Results Sixteen of 26 patients (61.5%) showed treatment response. The mean age of the patients was 10.62 ± 3.42 (5–17) years. There was no difference between the groups in age, age at start of patch therapy, sex, follow-up period, refractive errors, type of amblyopia, reason for patch therapy non-response, or mean effective patching time per day. In the responder group, the LogMAR values of pretreatment BCVA, BCVA after optical correction, and BCVA after occlusion were significantly higher, but BCVA after atropine treatment showed no difference. FAF images of all patients were normal, and the mean central macular thickness did not significantly differ between the groups. Conclusions Atropine penalization can improve BCVA in children with amblyopia who are non-responders to patch therapy. Atropine penalization may be more successful in children with poor BCVA at the start of atropine penalization in the amblyopic eye. The results of FAF imaging and mean central macular thickness were not associated with treatment outcomes.