March 14, 2023
A 67-year-old man was referred for a recent injury to his left eye. He was working in his shed drilling metal and felt a sharp pain in his left eye. He went to his OD who noted a corneal metallic foreign body with mild ocular inflammation. He removed the foreign body and put him on topical antibiotics. He wasn’t sure if he removed the entire piece of metal and referred him to our corneal service for evaluation. The corneal specialist felt that the referring doctor had done a successful removal and no additional metal was noted in the cornea. VA during that visit was OD: 20/20 and OS: 20/30. Because the eye was inflamed, the patient was asked to see me to determine the cause of inflammation. He set up an appointment for the following day, two days after the original injury.
On examination VA in OS was 20/70. IOP was 8. The SLE examination was significant for fibrinous aqueous humor with an early hypopyon. The cornea was edematous but no foreign body was noted. There was an early cataract but there was no obvious foreign body tract through the iris or the lens. DFE was significant for moderate vitreous haze. There were numerous intraretinal hemorrhages for 360 degrees as well as inferior pre-retinal layered inflammatory material. Details of the retina were hazy, but no obvious intraocular foreign body was noted. A bscan was performed which showed moderate vitreous haze, and a possible refractile object inferiorly. The working diagnosis at this point was a probable intraocular metallic foreign body with definite endophthalmitis. Options at this point include urgent vitrectomy with culture and injection of antibiotics vs in-office “tap and inject” procedure with an urgent CT scan of the orbits to look for a definitive foreign body. Our thought process at the time was that because we were unable to definitively identify a foreign body, and since no tract was noted, we would start with the office “tap and inject” procedure and obtain a CT scan (MRI cannot be performed due to the possibility of a metallic object). Due to the fact that eyes with endophthalmitis can decompensate quickly, we did the procedure immediately.
During a “tap and inject” procedure, a small sample of vitreous is aspirated via the pars plana with a 3ml syringe and 25g needle. Injections of intravitreal vancomycin, cephazolin, and dexamethasone are performed. The procedure went well. The patient had a rapid recovery. The CT scan showed no evidence of a metallic foreign body. The sample interestingly never disclosed an obvious bacterial source of the infection. The vitreous gradually cleared, the retinal hemorrhages resolved, and acuity returned to 20/25. At this point we thought we were done – we were not.
The patient returned with increasing cataracts a month later. The vision was reduced to OD: 20/20 and OS: 20/30. He was bothered by the glare and agreed to proceed with CE OS. Surgery was uneventful with initial VA 20/25. Vision proceeded to drop to 20/50 three weeks after CE due to extensive CME (see OCT). He was placed on topical pred and nsaid qid each, with a prompt resolution of the CME and improvement to 20/30. Again, we thought we were done – we were not.
Three months later the patient noted a nasal shadow in his vision. Acuity remained 20/30, but the examination was significant for a moderately bullous temporal retinal detachment with the macula on (see photo). There were several small tears near the ora serrata temporally. It is unclear if these were related to the initial infection, but were likely unrelated. The patient underwent a vitrectomy with retinal reattachment. The surgery was uneventful and the retina flattened (see post-op photo). Again, we thought we were done, but we were not. Despite the full retinal reattachment and the macular never detaching, the vision was only 20/70. There was a moderate PCO but also a new growth of epiretinal membrane with recurrence of macular traction and edema (see OCT). We advised YAG capsulotomy to maximize the vision possibly followed by ERM peel if the vision did not improve. The YAG, unfortunately, did not improve the vision, and the patient was told he would need additional surgery with ERM peel. At this point, he became so frustrated that he left and never came back, so his final outcome is unknown. However, this case illustrates an interesting assortment of retinal issues that can occur in one patient and underscores the need for rapid intervention if an infection is suspected.