Recovery of Vision in Open Globe Injury Patients with Initial No Light Perception Vision

Purpose: To identify clinical characteristics of injured eyes associated with visual recovery in patients with open globe injuries (OGIs) and presenting with no light perception (NLP) vision. Design: Retrospective chart review. Subjects: All patients presenting to Massachusetts Eye and Ear with OGI and NLP vision from January 1999 to March 2022. Methods: Manual data extraction to collect patient demographic characteristics, preoperative, intraoperative, and postoperative characteristics of OGI injury, laceration versus rupture, history of intraocular surgery, time from injury to repair, timing of vitrectomy, lensectomy, choroidal drainage, and silicone oil placement, visual acuity (VA) at last follow-up, and subsequent B-scan ultrasound findings of retinal detachment, choroidal hemorrhage, vitreous hemorrhage, and disorganized intraocular contents. Patients with > 1 week of follow-up and a documented VA at most recent follow-up were included. Exclusion criteria included age < 10 years. Multivariable regression was performed. Main Outcome Measures: Visual acuity recovery defined as light perception or better in patients with OGI and initial NLP vision. Results: One hundred forty-seven eyes with NLP vision after OGI were included. Twenty-five (17%) eyes regained vision at last follow-up. The majority of patients recovered light perception vision (n = 15, 60%) followed by 20/500 or better (n = 5, 20%), hand motions (n = 3, 12%), and counting fingers (n = 2, 8%). Most injuries were zone III (n = 102, 69%) and presented with rupture (n = 127, 86%). The mean time from OGI to surgical repair was 0.85 +/- 1.7 days. B-scan was obtained in 104 (71%) cases. Pars plana vitrectomy was performed in 9 eyes (6%) with NLP at time of vitrectomy. Disorganized intraocular contents on B-scan (odd ratio, 0.170; 95% confidence interval, 0.042- 0.681; P = 0.012) was the only clinical variable significantly associated with visual recovery, corresponding to a lack of visual improvement. Conclusions: Recovery of vision in OGI with NLP vision at presentation cannot be predicted based on presenting clinical features. B-scan findings of disorganized intraocular contents after initial OGI repair was the only factor negatively associated with vision recovery in this patient population. Therefore, all eyes presenting with an OGI and NLP vision should undergo primary repair in hopes of subsequent visual recovery.