Potential rare adverse effects of fluoroquinolone antibiotics, such as ciprofloxacin and levofloxacin, have received much attention over the past decade. More than 10 years ago, studies first suggested an association with Achilles tendon rupture; more recently, a study published in JAMA last year by researchers in Canada raised concerns about a possible link between fluoroquinolone use and retinal detachment, which can lead to irreversible vision loss.
Now, however, a study released in JAMA today conducted by researchers in Denmark found no association in the general Danish population between fluoroquinolone use and retinal detachment. Researchers in Denmark looked at national registry data for more than 5 million Danish adults, which included nearly 750 000 episodes of fluoroquinolone use from 1997 to 2011. Fluoroquinolone users were separated into current, recent, past, and distant users, and the rates of retinal detachment in each of these groups were compared with the rate in the nonuser (control) group. Results showed the groups did not demonstrate significant differences in incidence rates of retinal detachment.
Lead author Björn Pasternak, MD, PhD, of the Statens Serum Institut in Copenhagen, discussed these results with news@JAMA.
news@JAMA: Why were you interested in conducting this study?
Dr Pasternak: Our study was prompted by a Canadian study published in 2012 that suggested an alarming effect of fluoroquinolone use on retinal detachment: a 4.5-fold increased risk with fluoroquinolone use. Because fluoroquinolones are a commonly used class of antibiotics, these findings translated into quite a significant number of extra retinal detachments, if the association was true. This, in turn, would have implications for antibiotic treatment guidelines. Thus, we felt that this was a topic that needed to be examined further, and we had access to high-quality data that were suited to do so.
news@JAMA: In contrast to the Canadian study, your study found no statistically significant link between fluoroquinolone use and the risk of retinal detachment. What factors could account for these differing results, and why should physicians have more confidence in your findings compared with the Canadian study’s findings?
Dr Pasternak: I think the key difference between the two studies is the study populations. Our study population was derived from the entire country of Denmark, whereas the Canadian study was based on patients who had visited an ophthalmologist. The Canadian study therefore included patients at higher baseline risk for retinal detachment: the background incidence of retinal detachment in that study was more than 10-fold higher than in other population-based studies. Our study, on the other hand, showed a background incidence comparable to other population-based studies. Another difference is that we were able to account for a number of known risk factors for retinal detachment that the Canadian study did not. Imbalances in unmeasured risk factors for retinal detachment in a population at high baseline risk can strongly bias results.
news@JAMA: Does this mean that there still may be an increased risk of retinal detachment with fluoroquinolone use in a specific subset of individuals who have a higher baseline risk?
Dr Pasternak: We cannot exclude this possibility. But even if fluoroquinolone use is associated with retinal detachment in a high-risk population, the point is that it has no impact on the level of a general population. The findings from our population-based study, when seen in absolute numbers, indicate that even though there might be an association, this has minimal, if any, influence on everyday clinical practice.
news@JAMA: So physicians do not necessarily need to change their current practice when considering prescribing a fluoroquinolone to a patient?
Dr Pasternak: As researchers, I don’t want to say that we are in a position to directly guide treatment decisions. But personally, I continue to prescribe fluoroquinolones with confidence in their safety and I would have no hesitations about taking a fluoroquinolone myself.
news@JAMA: If a patient who is about to be prescribed a fluoroquinolone says he or she has “heard something” about it increasing the risk of retinal detachment, how would you respond?
Dr Pasternak: As a prescribing physician, I would say that there are research findings out there that show opposite results, but there are no official recommendations that indicate that we should change our current practice of how we use fluoroquinolones. I would also add that it appears highly unlikely that someone without known risk factors for retinal detachment would develop this condition within a short period after starting on a course of fluoroquinolones.