- Hydroxychloroquine (HCQ) is used to treat many rheumatologic conditions, but retinopathy and vision loss can occur with long-term dosing.
- For HCQ-related retinopathy, solid estimates of prevalence, incidence over time, and risk factors for above-average incidence have been largely lacking.
- This large meta-analysis estimated 15-year cumulative risk at 5.6%, with female sex, daily HCQ doses higher than 5 mg/kg, and chronic kidney disease as major risk factors.
About one in 18 patients taking hydroxychloroquine (HCQ) for 15 years develop retinal damage, a new meta-analysis indicated.
With a total of some 20,000 patients in 19 included datasets, the 15-year cumulative incidence of retinopathy as detected with spectral-domain optical coherence tomography (SD-OCT) was 5.6%, according to J. Antonio Avina-Zubieta, MD, PhD, of the University of British Columbia in Vancouver, and colleagues.
Across all patients, the overall prevalence of retinopathy was 5.1% (95% CI 3.9-6.5), the group reported in Arthritis Care & Research.
But higher dosing and certain other modifiable factors presented greater risk, the researchers noted, findings which "underscore the importance of appropriate dosing and adherence to established screening guidelines."
HCQ is commonly prescribed for a range of rheumatological conditions including systemic lupus erythematosus, Sjögren's disease, rheumatoid arthritis, and antiphospholipid syndrome. In many of these, HCQ therapy is recommended to be continued indefinitely, which can add up to decades for diseases such as lupus with young-adult onset.
Retinopathy and permanent vision loss have long been recognized as risks with chronic HCQ, but the details of just how often they develop, and which patients may face extra risk, are still being worked out. Avina-Zubieta and colleagues observed that many past studies were too small or had too little follow-up to provide reliable estimates of the risk. "These evidence gaps may limit patient counseling and optimization of long-term HCQ monitoring," they wrote.
And because most of the previous estimates were fairly low, it would be impossible for a single study to do much better without enrolling a very large cohort to be followed for many years. That set meta-analysis as a more pragmatic approach.
Avina-Zubieta's group reviewed several hundred published studies, settling on 18 with well-characterized retrospective or prospective cohorts and one case-control study. These studies had at least 1 year of follow-up with SD-OCT screening (the recommended method since 2011) as well as detection rates, data on dosing, and patient characteristics that could be examined for their impact on retinopathy risk.
Not every study included in the analysis met all of these criteria, however. Twelve had sufficient data to estimate overall prevalence, and only three followed patients for the full 15 years. (Those three, however, had over 5,000 participants in all.)
Pooled data, after weighting for sample size, yielded the following estimates for cumulative incidence of retinopathy:
- 5 years: 0.1% (95% CI 0-0.5)
- 10 years: 2.6% (95% CI 1.6-4.1)
- 15 years: 5.6% (95% CI 3.2-9.6)
But these risk levels don't apply to every patient, the researchers emphasized. Some individual factors were found to increase the risk considerably.
Most notably, patients taking an average daily dosing above 5 mg/kg were 4.32 times as likely to develop retinopathy as those on lower doses. That's concerning, insofar as doses of 200-400 mg/day -- or 6.5 mg/kg/day in some dosing schemes -- are recommended to optimize disease control in lupus. (One recent study called for a minimum of 400 mg/day.)
Chronic kidney disease was also found to increase retinopathy risk, approximately doubling it (HR 1.94, 95% CI 1.27-2.96). Another substantial risk factor was female sex (HR 3.78, 95% CI 1.90-7.48). Persons of Asian ethnicity were at modestly higher risk (HR 1.67, 95% CI 1.07-2.62).
It's important to note that these estimates were not very certain, as the confidence intervals suggest, Avina-Zubieta and colleagues acknowledged. They wrote that the certainty of evidence behind the 5- and 10-year cumulative incidence figures was low, and it was very low for the 15-year estimate, the overall prevalence, and the extra risks associated with ethnicity and sex.
Other limitations included lack of consistent data on age at HCQ initiation, other differences between studies in designs and data collection, and lack of detail on dosing.
Nevertheless, with this being "the first meta-analysis to estimate the pooled prevalence, cumulative incidence, and risk factors ... focusing on studies that used SD-OCT" to screen for retinopathy, the findings can help guide patient management and counseling, the researchers argued.
Disclosures
The study was funded by the Canadian Institutes for Health Research.
Authors' conflict of interest disclosures were not available.
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.