Chloroquine has long been used in the treatment and prevention of malaria 1 2 3 4 , and for a long time, the agent is still being used to treat other diseases. For example, due to its immunosuppressive effects, chloroquine has been used in the treatment of autoimmune disorders, such as rheumatoid arthritis and lupus erythematosus. Chloroquine has also been tested in clinical trials as an antiretroviral agent against HIV 5 and as a radiosensitizer and chemosensitizer for human cancer therapy 6 . However, the side effects of Chloroquine, including ocular toxicities, limit its long-term use 2 3 4 6 .

Herein we describe two patients who developed severe ocular toxicity following long-term chloroquine treatment. The clinical characteristics of their retinopathy were extensively investigated using various techniques, and are reported in this manuscript. We are not trying to provide information which would help in the early screening for chloroquine retinopathy, but to describe the typical manifestations of chloroquine retinopathy with some advanced new technology which have not been applied on this disease before, believing that it would provide a better understanding of chloroquine retinopathy.

Chloroquine toxicity is not uncommon, especially in patients receiving long-term therapy 1 2 3 . While hydroxychloroquine is now frequently prescribed instead of chloroquine because of its higher therapeutic window, its use has also been shown to cause retinal damage 4 . Histological analysis of human and animal retinas with chloroquine toxicity has demonstrated the loss of ganglion cells, photoreceptors, and RPE, which likely result from inhibition of lysosomal phospholipases and protein synthesis. However, the initial retinal damage is thought to occur in ganglion cells 7 .

In contrast to the usual clinical observations, there was no major pigment atrophy ring with a clarified boundary around the central fovea (making the characteristic bulls-eye) in the FFA image from our patients. However, we did observe a marked hyperfluorescence in the macular area, possibly due to RPE atrophy. Although it is frequently used in the clinic, full-field ERG may not be a reliable method for identifying early chloroquine retinopathy 3 . In our cases, because the two patients had severe retinopathy, the FERG amplitudes were non-recordable. Therefore, F-ERG analysis may have limited value in the diagnosis of early chloroquine retinopathy, because the toxicity is localized and a full field analysis may not be sensitive to localized retinal toxocity in early disease.

It has been suggested that mfERG is a more sensitive method that can be used to detect chloroquine retinopathy 9 . For example, in one study, even though clinical examinations found no abnormalities in a patient with a long history of hydroxychloroquine use 3 , mfERG showed an absence of foveal rings and the presence of pericentral isoelectric waveforms. In our current report, the patients had severely damaged visual acuity and center visual field loss, mfERG showed a reduction in focal responses and a delay in the peak latency.

There have been some reports in which OCT was used in the detection of chloroquine and hydroxychloroquine-induced retinopathy 10 . In our cases, not only the "bulls-eye" was shown in the fundus photograph, but also the proliferation-atrophy-proliferation changes of the RPE in the macular area were detected by OCT. Confirming these results, the RNFL thickness measured by GDx (GDx Nerve Fiber Analyzer (NFA) is a relatively new and improved technology that uses scanning laser polarimetry (SLP)) was within the normal range in our patients, this result was different from the conclusions of some other reports. For example, Bonanomi et al 11 reported that mean RNFL thickness measured by GDxFCC from patients using chloroquine were significantly different from the control group in all regions around the optic disc. Possible explanation for this difference might be that the damage of retina in some patients taking chloroquine distributed mainly in the macula area, while the RNFL thickness around the optic disc remained normal.

To the best of our knowledge, this is the first report describing the findings of FFA, OCT, GDxVCC NFA, FERG, mfERG and visual field examinations in the same patients with chloroquine retinopathy. While the fundus pictures from the two patients are similar, there are differences in their macular thickness. Thus, given the differences found using the various methods and instruments, and the well-documented weaknesses of the various techniques, it is evident that multiple analyses are helpful for better understanding of chloroquine retinopathy.

Written informed consent was obtained for publication of these two case reports and accompanying images. Copies of the written consent are available for review by the Editor-in-Chief of this journal.

The authors declare that they have no competing interests.

XM, LY and LH performed ophthalmic examinations, conducted the design of the study and drafted the manuscript. DH participated in the design of the study and drafted the manuscript. HL participated in the electrophysiology examinations and drafted the manuscript. All authors read and approved the final manuscript.