RESUMEN
PURPOSE: To study the role of serum biomarkers as prognostic factors for qualitative and quantitative response to anti-vascular endothelial growth factor injections for diabetic macular edema (DME). METHODS: Sixty-seven eyes with DME were treated with intravitreal bevacizumab during a 12-month follow-up period. All cases underwent a baseline workup consisting of 12 inflammatory, metabolic and prothrombotic factors. The following outcomes were evaluated at 3-month intervals until 1 year of follow-up: visual acuity, central subfield thickness (CST), macular volume (MV), % of change from baseline in CST, occurrence of a CST change < 10%, a CST change >20%, and a CST <330 µm, achieving an improvement ≥2 lines of visual acuity, achieving visual acuity ≥20/40. RESULTS: A significant improvement in CST and visual acuity was seen from third month onwards. Twenty-eight (48.1%) cases were classified as "early responders," 24 (35.8%) as "late responders", and 15 (22.4%) as "poor responders." Serum vascular endothelial growth factor-A levels were significantly lower in "poor responders" (P = 0.006). C-reactive protein (hsCRP) was associated with a limited anatomic response (<10% CST change) (P = 0.002, OR = 1.845, cutoff value of hsCRP = 1.84 mg/L). hsCRP was also negatively associated with obtaining a final CST <330 µm (P = 0.04, r2 = 0.112, OR = 0.643). Baseline visual acuity was significantly associated with 12th month visual acuity (P < 0.001, r2 = 0.602) and also with an improvement ≥2 visual acuity lines (P = 0.009, OR = 20.54). CONCLUSION: Increased high-sensitivity C-reactive protein was associated with limited anatomic response to anti-vascular endothelial growth factor treatment and persistent DME. Poor responders had significantly lower values of serum vascular endothelial growth factor-A, suggesting an alternative pathogenic pathway for persisting DME.