Do Vitreous Floaters Predict Retinal Detachment? Retrospective Cohort Study in Primary Care

PURPOSE Vitreous floaters are often considered harmless. However, floaters might be the first and only alarm symptom for a retinal tear or early retinal detachment (RD). Data from ophthalmology practices show that experiencing new-onset floaters is more strongly associated with retinal tears than experiencing flashes, but these associations have not been studied in primary care. We aimed to calculate the absolute risk (AR) and relative risk (RR) of floaters for RD in primary care. METHODS We performed a retrospective cohort study of patients (aged ≥18 years) with new-onset floaters or flashes during the period 2012 to 2021 among 7 family practices in the Netherlands. Based on electronic health record reviews, cases (n = 1,181) were categorized into floaters, flashes and floaters, and flashes. The duration (acute, defined as ≤14 days) and number of floaters (many, defined as ≥10 floaters or a cloud/haze/curtain) were used as subgroups. We calculated the RR for flashes alone as the reference group. RESULTS The incidence (1,000 patient-years) of floaters and flashes was 5.5 and 2.7, respectively. The AR of RD was 6.1% for floaters alone, 4.7% for flashes alone, and 8.4% for floaters and flashes. Both acute and many floaters increased AR. The RRs for acute floaters and flashes (2.39; 95% CI, 1.11-5.15), many floaters (4.20; 95% CI, 1.87-9.40), and many floaters and flashes (6.20; 95% CI, 2.47-15.55) were significantly increased compared with flashes alone. CONCLUSIONS Data from primary care confirm that new-onset floaters confer increased risk of RD. Currently, family physicians use vision loss and flashes as important alarm symptoms for RD. Our primary care data revealed that floaters confer a greater risk of RD than flashes. This enables family physicians to make an evidence-based risk assessment for patients with floaters or flashes.