Autologous Serum Eye Drops for Dry Eye: Systematic Review

BACKGROUND Dry eye is a common condition with serious implications worldwide. The unique composition of autologous serum (AS) eye drops has been hypothesized as a possible treatment.OBJECTIVES This study aimed to review the effectiveness and safety of AS.DATA SOURCES We searched five databases and three registries up to September 30, 2022.STUDY ELIGIBILITY We included randomized controlled trials (RCTs) comparing AS with artificial tears, saline, or placebo for participants with dry eye.STUDY APPRAISAL AND SYNTHESIS METHODS We adhered to Cochrane methods for study selection, data extraction, risk-of-bias assessment, and synthesis. We used the Grading of Recommendations Assessment, Development and Evaluation framework to evaluate the certainty of evidence.RESULTS We included six RCTs with 116 participants. Four trials compared AS with artificial tears. We found low-certainty evidence that AS may improve symptoms (0- to 100-point pain scale) after 2 weeks of treatment compared with saline (mean difference, -12.00; 95% confidence interval, -20.16 to -3.84; 1 RCT, 20 participants). Ocular surface outcomes (corneal staining, conjunctival staining, tear breakup time, Schirmer test) were inconclusive. Two trials compared AS with saline. Very low-certainty evidence suggested that Rose Bengal staining (0- to 9-point scale) may be slightly improved after 4 weeks of treatment compared with saline (mean difference, -0.60; 95% confidence interval, -1.11 to -0.09; 35 eyes). None of the trials reported outcomes of corneal topography, conjunctival biopsy, quality of life, economic outcomes, or adverse events.LIMITATIONS We were unable to use all data because of unclear reporting.CONCLUSIONS The effectiveness of AS is uncertain based on current data. Symptoms improved slightly with AS compared with artificial tears for 2 weeks. Staining scores improved slightly with AS compared with saline, with no benefit identified for other measures.IMPLICATIONS OF KEY FINDINGS High-quality, large trials enrolling diverse participants with varying severity are needed. A core outcome set would allow for evidence-based treatment decisions consistent with current knowledge and patient values.