An investigation of the hospital charges related to the treatment of endophthalmitis in the Endophthalmitis Vitrectomy Study.

PURPOSE: The purpose of the study is to assess the hospital charges associated with the treatment of endophthalmitis using a sample of patients from the Endophthalmitis Vitrectomy Study (EVS). METHODS: The Endophthalmitis Vitrectomy Study was a multicenter, randomized clinical trial with a two-by-two factorial design to compare immediate pars plana vitrectomy to tap-biopsy and to compare the use of systemic antibiotics (intravenous) to no intravenous antibiotics in the management of postoperative endophthalmitis. Hospital charge data were collected retrospectively from 129 patients from the 4 clinical centers participating in this ancillary study. This represents 31% of the total Endophthalmitis Vitrectomy Study population. An analysis of variance was used to compare hospital charges across center and treatment. A charge-effectiveness analysis compared measures the effectiveness across treatment groups. The annual savings of hospital charges in the United States was estimated for a range of annual incidence rates of endophthalmitis. RESULTS: The use of intravenous antibiotics significantly increased hospital charges. Patients undergoing vitrectomy had significantly higher hospital charges than did patients undergoing tap-biopsy. The most charge-effective treatment for patients presenting with light perception only vision was immediate vitrectomy, whereas the most charge-effective treatment for patients presenting with better vision was tap-biopsy. Factors other than treatment independently associated with hospital charges were female sex, history of diabetes, symptom of red eye, and baseline vision of light perception only. CONCLUSIONS: Assuming the results of the Endophthalmitis Vitrectomy Study were used as a guide for the treatment of endophthalmitis, the estimated annual nationwide reduction of hospital charges would be between $7.6 million and $40.0 million.

Management of giant retinal tears in dogs.

Sixteen dogs with giant retinal tears resulting in total retinal detachment (360 degrees) underwent unilateral retinal reattachment surgery. In all dogs, complete vitrectomy was performed, and in most, perfluorooctane liquid was used to unroll and flatten the detached retina. Stainless steel tacks were used to anchor the retina to the choroid and sclera. Silicone oil was injected into the vitreous cavity to provide for permanent retinal tamponade. Endophotocoagulation or cryosurgery was used in several dogs to induce formation of permanent chorioretinal adhesions. Retinal attachment was successful in 14 of 16 dogs, and 13 of 14 dogs in which reattachment was successful retained some degree of vision. Complications included formation of cataracts and corneal erosions, development of glaucoma, and dislodgement of tacks.

Endogenous Candida endophthalmitis after intravenous anesthesia with propofol.

Propofol is an intravenous anesthetic agent that was approved by the US Food and Drug Administration in October 1989. It has gained rapid acceptance and is in widespread use. In June 1990, the Centers for Disease Control reported four clusters of postsurgical infections associated with the use of propofol. We describe one of those clusters, consisting of four cases of endogenous Candida albicans endophthalmitis. These infections occurred in nonimmunocompromised patients after they had undergone nonophthalmologic surgery in which propofol had been used as anesthesia. An investigation by the Centers for Disease Control concluded that the infections in these patients were due to extrinsic contamination of propofol during preparation for use at the hospital. Ophthalmologists should be aware of this new potential source of endogenous endophthalmitis.

Iatrogenic choroidal neovascularization after krypton red laser photocoagulation.

A major cause for failure of krypton red laser photocoagulation in patients with exudative age-related macular degeneration has been the development of recurrent choroidal neovascularization adjacent to the previously treated areas. After reviewing the possible causes of recurrences, it is apparent that a certain number are iatrogenic, ie, induced by krypton red laser causing disruption and damage to the pigment epithelium-Bruch's membrane-choroidal complex. We describe three separate episodes in two patients of iatrogenic recurrent choroidal neovascularization after krypton red laser photocoagulation.

Macular serpiginous choroiditis.

The authors describe in seven eyes of four patients a form of serpiginous choroiditis beginning in the macula without initial peripapillary activity. Eyes with macular serpiginous choroiditis often had a poor visual prognosis and sometimes developed subretinal neovascularization. Fluorescein angiography of the acute lesions showed hypofluorescence; some cases were thus initially diagnosed as having choroidal ischemia. Fluorescein angiography in one eye, however, suggested that at least some of the hypofluorescence seen acutely in eyes with serpiginous choroiditis is secondary to blocked fluorescence from "opaque" retinal pigment epithelium.

Intrusion of retinal tacks.

We examined nine patients in whom retinal tacks intruded into the eye and lodged in the subretinal space, preretinal space, vitreous cavity, or anterior chamber. Complications included retinal pigment epithelium atrophy; retinal phlebitis; vitreous hemorrhage; focal corneal, iris, and retinal injury; and corneal edema. The intrusion of the retinal tacks did not apparently cause, but was associated with retinal redetachment in five patients. Factors associated with intrusion of the retinal tacks included absence of a barb at the end of the tack to anchor it to the sclera, absence of a groove in the tack, a short shaft, incomplete penetration of the retina, choroid, and sclera by the tack, self-inflicted trauma to the eye, placing a scleral buckle after inserting the tacks, and reproliferation of periretinal membranes. In four patients the intruded tacks did not cause any complications. In four patients the intruded tacks were removed without complications and in the remaining five patients, they were left in the eye.