Safety and efficacy of ultrasound-guided pars plana vitrectomy: a prospective, single-arm study

Purpose To estimate the amount of eliminated vitreous after ultrasound-guided pars plana vitrectomy and complete the safety profile of the technique. Methods We enrolled patients with vitreous hemorrhage and clear media who had an ultrasound-guided pars plana vitrectomy. The amount of eliminated vitreous was calculated by digital analysis (ImageJ) of before and after photos, obtained intraoperatively. All cases were completed under direct visualization for safety evaluation. All patients had a comprehensive ophthalmological examination at baseline and at day 1, and 1 month follow-up. Adverse events were reported in proportions ± 95%CI. A learning curve was plotted via the formula y = ax^b. Interobserver agreement was assessed with a Cohen-Kappa test. Results We enrolled 62 patients. Mean age: 64.3 ± 13.3 years. Digital image analysis showed that there was a significant reduction in the number of pixels representing vitreous hemorrhage (≈ 70%, p < 0.01). The minimum number of cases needed to achieve a 70% or greater elimination of vitreous was 18. The proportion of potentially related adverse events to the surgical technique was 1.6%, 95%CI: 0.04–8.66). Conclusion Ultrasound-guided pars plana vitrectomy is feasible and has an acceptable safety profile for cases with minimal to no visibility of the posterior pole.