Do Hyperreflective Dots on SD-OCT Predict Response to Macular Edema Treatment?

Journal Highlights

Investigative Ophthalmology & Visual Science
2017;58:5958-5967

Download PDF

Methods to predict therapeutic response may prevent unnecessary treatment and improve outcomes for patients with macular edema. Hwang et al. aimed to determine whether the quantity of hyperreflective dots (HRDs) on spectral-domain optical coherence tomography (SD-OCT) at baseline could indicate treatment response to intravitreal bevacizumab or dexamethasone injections in eyes with macular edema. They found that correlations exist but are different for the 2 therapies.

The authors’ retrospective study included 82 eyes with diabetic macular edema (DME) and 68 eyes with macular edema from retinal vein occlusion (RVO). Patients with treatment-naïve macular edema initially received 3 consecutive bevacizumab injections, and treatment response was documented. Following these injections, nonresponders received dexamethasone. HRDs were counted manually and independently by 2 masked retina specialists. The authors documented treatment response in relation to best-corrected visual acuity (BCVA), number of HRDs, and incidence of outer plexiform layer (OPL) disruption.

Thirty-six eyes with DME (43.9%) and 22 with RVO (32.4%) did not respond to bevacizumab. The number of baseline HRDs in bevacizumab nonresponders (DME, 16.06 ± 6.60; RVO, 14.23 ± 4.09) was significantly greater than in responders (DME, 11.26 ± 3.64, p < .001; RVO, 11.17 ± 4.83, p = .013) and did not decline after bevacizumab treatment. Eyes that responded to dexamethasone but not to bevacizumab had significantly more baseline HRDs than eyes that did not respond to either treatment (19.56 ± 6.75 vs. 11.50 ± 3.78; p = .006). The OPL disruption rate was significantly higher for bevacizumab nonresponders than responders (DME, p < .001; RVO, p = .001). BCVA improved in bevacizumab responders but not in bevacizumab nonresponders.

In summary, the number of HRDs on baseline SD-OCT may indicate whether macular edema will improve with intravitreal bevacizumab or dexamethasone. In bevacizumab responders, the number of HRDs was small. The larger number of HRDs in dexamethasone responders may reflect greater inflammation of the retina. Hence, the latter treatment may be most effective in eyes that exhibit many HRDs and OPL disruptions. Large-scale prospective studies that include automated quantification of HRDs are encouraged.

The original article can be found here.