Predictors of chloroquine and derivates treatment involving ocular toxicity: Results from a cohort

Dos-Santos, R.; Fernández Fernández, D.; González Fernández, I.; Sanchez-Wonenburger, M.; Castro-Santamaria, P.; Souto Vilas, A.; Perez-Pampín, E.; Mera Varela, A.

Background: Chloroquine (CQ) and hydroxychloroquine (HCQ) have been employed in a huge range of indications, from autoimmune diseases (such as rheumatoid arthritis [RA], cutaneous lupus or systemic erythematosus lupus [SLE]) to infectious ones (as malaria or helminthiasis).1 A newer purpose came upon the new coronavirus disease 2019 (COVID-19), where they seem to be effective modulating immune response. Controversial results have been published from clinical and observational data concerning its effectiveness.2 Ocular toxicity have been described as a serious adverse event of these antimalarial drugs and screening protocols have been displayed for its prevention.3 Objectives: To evaluate CQ/HCQ ocular toxicity and to identify potential predictors of appearance. Also to asses screening protocols compliance. Methods: Demographic, diagnostic and treatment data were collected from patients under CQ or HCQ treatment in the Clinical University Hospital in Santiago De Compostela (Spain) during the first wave of the COVID-19 pandemic (January to April 2020). Univariable logistic regression was performed to identify potential predictors of maculopathy. Variables with p<0.20 were selected for multivariable testing. Stata 15.1 was used to perform statistical analysis. Results: 503 patients taking CQ/HCQ were identified. 495 were women. Most frequent diagnosis were SLE (48.28%), cutaneous lupus (22.85%) and rheumatoid arthritis (RA, 12.54%). Mean age at diagnosis was 44.99 years (SD 17,88). 93.33% of patients were under treatment with HCQ. Mean age at beginning of CQ/HCQ treatment was 48.10 years (SD 17.79) and mean time between diagnosis and CQ/ HCQ onset was 2.03 years (SD 5.50). Mean maximum HCQ dosage per patient was 3.83 mg/kg (SD 1,59;252.57 mg per day, SD 89.98) and CQ was 3.24 mg/kg (SD 1,91;219.49 mg per day, SD 103.90). Mean time under CQ/HCQ treatment was 6.39 years (SD 5.63). 20 patients developed maculopathy. Mean time between CQ/HCQ onset and maculopathy appearance was 2.67 years (SD 3.10). Only 25 patients did not complete ophthalmologic exams for maculopathy screening. After univariable analysis, higher age at diagnosis and age at beginning of CQ/ HCQ treatment were identified as potential predictors of maculopathy (p<0.20). After multivariable analysis, both higher age at diagnosis and higher age at CQ/ HCQ onset were identified as predictors for suffering maculopathy under treatment with CQ/HCQ (OR 1.06 [CI95% 1.03-1.10] p=0.000 and OR 1.09 [CI95% 1.02-1.16] p=0.008, respectively). Conclusion: Ocular toxicity remains as one of the most harmful and disabling adverse events in patients under CQ/HCQ treatment. Higher age at diagnosis and higher age at beginning of treatment appear to be risk factors for maculopathy appearance. Screening protocols are well-assumed by patients and seemed to be helpful for preventing and early identifying events. CQ/HCQ usage in COVID-19 patients should be individualized, specially in older patients, and protocols involving ocular toxicity should be implemented in the follow-up of this population.