Search

Purpose: To analyze and provide an overview of the incidence, management, and prevention of conjunctival erosion in Argus II clinical trial subjects and postapproval patients., Methods: This retrospective analysis followed the results of 274 patients treated with the Argus II Retinal Prosthesis System between June 2007 and November 2017, including 30 subjects from the US and European clinical trials, and 244 patients in the postapproval phase. Results were gathered for incidence of a serious adverse event, incidence of conjunctival erosion, occurrence sites, rates of erosion, and erosion timing., Results: Overall, 60% of subjects in the clinical trial subjects versus 83% of patients in the postapproval phase did not experience device- or surgery-related serious adverse events. In the postapproval phase, conjunctival erosion had an incidence rate of 6.2% over 5 years and 11 months. In 55% of conjunctival erosion cases, erosion occurred in the inferotemporal quadrant, 25% in the superotemporal quadrant, and 20% in both. Sixty percent of the erosion events occurred in the first 15 months after implantation, and 85% within the first 2.5 years., Conclusion: Reducing occurrence of conjunctival erosion in patients with the Argus II Retinal Prosthesis requires identification and minimization of risk factors before and during implantation. Implementing inverted sutures at the implant tabs, use of graft material at these locations as well as Mersilene rather than nylon sutures, and accurate Tenon's and conjunctiva closure are recommended for consideration in all patients.

Purpose: We describe a novel method to oxygenate intraocular irrigation solutions involving an in-line oxygenator., Methods: Either lactated Ringer (LR) solution or balanced salt solution (BSS) was oxygenated with the FE390 Stainless In-line Oxygenation Assembly (Beer, Beer, and More Beer, Concord, CA). After running a 100-mL of solution through the in-line oxygenator, oxygen saturation was measured with a dissolved oxygen meter. A control experiment involving the oxygenator without oxygen perfusion was performed. Paired t-tests were used to compare oxygen saturation changes before and after oxygenation., Results: In comparison with the original BSS, there was an increase in oxygen saturation of 162 +/- 47% (n = 40; P < 0.05). Without oxygenation perfusion, there was only a 24 +/- 14% (n = 40) increase in oxygen levels in BSS, which was significantly lower than that in treated BSS (P < 0.05). In comparison with the original LR solution, there was an increase in oxygen saturation of 208 +/- 21% (n = 40; P < 0.05). Without oxygenation perfusion, there was only a 21 +/- 9% (n = 40) gain in oxygen saturation in the control LR solution, which was statistically lower as well (P < 0.05)., Conclusion: The in-line oxygenator is an efficient tool for oxygenating BSS and LR solution. It represents a potential efficient and convenient method to oxygenate irrigating solutions for vitreoretinal surgeries.

Purpose: To investigate the effect of oxygenated intraocular irrigating solutions on electroretinograms (ERGs) for postvitrectomy rabbits., Methods: Eight groups of five rabbits each underwent pars plana vitrectomy on the right eye; the left eye of each rabbit served as control. The intraocular irrigating solutions were balanced salt solution (BSS), BSS-plus, BSS + oxygen (BSS + O2), BSS-plus + O2, and combinations of each with the addition of endoillumination (L). All animals were evaluated by single-flash scotopic electroretinography on the operated and nonoperated eyes before surgery and at 1 hour, 1 day, 1 week, and 1 month after surgery by an unmasked observer. The main outcome measures were dark-adapted b-wave amplitudes of operated eye/nonoperated eye ratios. The results for the different groups were compared by analysis of variance., Results: Mean dark-adapted b-wave amplitudes of operated eye/nonoperated eye ratios +/- SD for BSS-plus and BSS-plus + O2 before surgery and 1 hour, 1 day, 1 week, and 1 month after surgery were 1.01 +/- 0.09, 0.50 +/- 0.11, 0.95 +/- 0.11, 0.97 +/- 0.11, and 0.99 +/- 0.08 and 0.98 +/- 0.08, 0.59 +/- 0.10, 0.92 +/- 0.06, 0.97 +/- 0.08, and 0.97 +/- 0.10, respectively. At both 1 hour and 1 day after surgery, rabbits treated with BSS-plus + O2 had an earlier b-wave return to baseline findings, but the differences were not statistically significant (P > 0.05). Mean dark-adapted b-wave amplitudes of operated eye/nonoperated eye ratios +/- SD for BSS and BSS + O2 before surgery and 1 hour, 1 day, 1 week, and 1 month after surgery were 1.02 +/- 0.10, 0.47 +/- 0.09, 0.77 +/- 0.07, 0.89 +/- 0.07, and 0.89 +/- 0.07 and 1.02 +/- 0.06, 0.62 +/- 0.16, 0.94 +/- 0.09, 0.97 +/- 0.08, and 0.96 +/- 0.06, respectively. One hour and 1 day after surgery, ERG readings for rabbits treated with BSS + O2 exhibited a statistically significantly earlier return of ERG voltage to baseline values compared with both BSS and BSS + L (P = 0.05 and P = 0.02, respectively). One day after surgery, rabbits treated with BSS alone had the lowest ERG ratios. One week and 1 month after surgery, for all solutions tested other than BSS, ERG values were within normal limits., Conclusion: The use of oxygenated solutions appears to affect ERG readings after pars plana vitrectomy. Further studies to evaluate the role of oxygenated solutions in different vitreoretinal surgical procedures are warranted.

Purpose: To evaluate a new technique, 25-gauge transvitreal limited arteriovenous-crossing manipulation without vitrectomy (LAM), for the treatment of branch retinal vein occlusion (BRVO) complicated by macular hemorrhage and/or macular edema recalcitrant to grid laser photocoagulation., Methods: Twelve eyes of 12 patients underwent LAM for BRVO performed by a single surgeon (M.S.H.) using the 25-gauge nitinol flexible-extendable blunt pick. The presence or absence of intraoperative reperfusion visualization, pre- and postoperative visual acuity, macular thickness as measured by optical coherence tomography, intraocular pressure, and lens status were evaluated., Results: Restoration of blood flow was noted in all patients and was based on intraoperative reestablishment of a red column of erythrocytes through the previously closed vessel. Mean visual acuity improved from 20/200 (logarithm of the minimal angle of resolution [LogMAR] +/- SD, 1.00 +/- 0.32) preoperatively to 20/70 (LogMAR +/- SD, 0.56 +/- 0.28) (P = 0.0003) at the final visit. Eleven (92%) of 12 eyes had >or=2 lines of visual improvement. Five eyes (45%) had final visual acuity of 20/50 or better. Mean macular thickness +/- SD improved from 401.0 +/- 73.2 to 178.7 +/- 19.6 microm (P < 0.0001) at the final visit. No statistically significant difference was noted in cataract progression or intraocular pressure. Mean follow-up +/- SD was 49.9 +/- 19.6 weeks. All patients were observed for at least 12 weeks., Conclusion: LAM may achieve outcomes comparable with those of arteriovenous adventitial sheathotomy for complicated BRVO.

Objective: To demonstrate the feasibility and potential applicability of a virtual reality simulator for vitreoretinal surgery as a training and/or assessment tool., Methods: The subjects of this study included medical students, ophthalmologic residents, and trained vitreoretinal surgeons. There were three study groups. Group I comprised 22 subjects who performed a navigation task. The time to complete the task was recorded. The relationship between the completion time, experience, and stereopsis was evaluated. Group II included 6 subjects who consecutively performed the navigation task to evaluate their learning curve. Group III included 16 subjects who performed the membrane peeling task. The number of retinal contacts and the completion time were recorded. The relationship between experience and stereopsis with the number of contacts and the completion time were evaluated., Results: The average completion time in Group I for students, residents, and trained surgeons was 121.6, 92.5, and 70.6 seconds. There was a significant difference between students and trained surgeons (P = 0.004). In Group II, there was a significant decrease in the completion time with training (P = 0.001). In Group III, the average completion time for students, residents, and trained surgeons was 197, 144, and 118.2 seconds; the respective number of retinal contacts was 14, 8, and 3. There was a significant difference between students and residents (P = 0.05) and between residents and trained surgeons (P = 0.003) for the average completion time in Group III. There was a significant difference between students and trained surgeons (P = 0.003) for the number of contacts per average time and between students and residents (P = 0.05). There was a significant inverse correlation between stereopsis vision score and completion time in Group I and number of contacts per average time (P = 0.0004 and P = 0.01, respectively)., Conclusions: This study demonstrates potential applications of a vitreoretinal surgical simulator as a training and skills assessment tool for novice, inexperienced, and trained surgeons. A simulator can be used to teach specific techniques and train surgeons.

Purpose: To evaluate the effects of indocyanine green (ICG) injection on the retinal surface and into the subretinal space of rabbit eyes., Methods: Twenty-two Dutch-belted rabbits underwent two-port vitrectomy followed by injection of ICG (5 mg/mL) on the retinal surface and into the subretinal space. Balanced salt solution (BSS) was also injected subretinally. The locations where ICG was delivered (both epiretinal and subretinal) were exposed to light from an endoilluminator for 7 minutes. The animals were examined at 1, 7, and 14 days after surgery. The eyes were studied by fluorescein angiography as well as light and electron microscopy., Results: No damage was observed after epiretinal ICG injection, but subretinal ICG injection resulted in damage to the outer nuclear layer, photoreceptor inner and outer segments, and retinal pigment epithelium. This damage was more severe with longer follow-up. Control experiments without ICG, in which balanced salt solution was injected into the subretinal space or light was delivered on the epiretinal surface, demonstrated only damage to the photoreceptor outer segments., Conclusion: Subretinal delivery of ICG (5 mg/mL) in rabbits induces retinal pigment epithelium, photoreceptor inner and outer segment, and outer nuclear layer damage. These mechanisms of damage may explain the retinal pigment epithelium changes that are sometimes seen after ICG-assisted internal limiting membrane peeling in humans.

Purpose: To report the clinical, electrophysiologic, and histologic findings of different concentrations of indocyanine green (ICG) injected into the vitreous cavity of rabbit eyes., Methods: Forty-two rabbits underwent intravitreal injection of 0.1 mL of ICG in three different concentrations: 0.5 mg/mL (250 mOsm), 5 mg/mL (270 mOsm), and 25 mg/mL (170 mOsm). Fellow eyes were injected with 0.1 mL of balanced salt solution. Biomicroscopy, ophthalmoscopy, electroretinography, fluorescein angiography, and histologic evaluation were performed., Results: Eyes injected with 0.5 mg/mL of ICG showed b-wave latency delay on the first day after injection. Eyes injected with 5 mg/mL of ICG showed b-wave latency delay and decreased b-wave amplitude on the first and seventh days after injection; delayed a-wave latency on the first day after injection was also observed. Eyes injected with 25 mg/mL of ICG showed b- and a-wave amplitude and latency abnormalities during the entire follow-up. Direct correlation of increasing retinal edema proportional to the progressively increasing ICG concentrations was shown on histologic evaluation., Conclusion: Intravitreal ICG injection in rabbit eyes may impair retinal function and morphology proportional to the progressively increasing ICG dosages.

Objective: To develop and characterize an infusion cannula that reduces the intraocular jet forces created during fluid-air exchange., Methods: A new infusion cannula was manufactured by adding a baffle 1.0 mm perpendicular to the distal opening of a 20-gauge infusion. The forces generated by the modified and standard cannulas were measured at a constant air infusion pressure of 50 mmHg by using a force sensor coupled to a schematic eye with eight radial positions from 0 degrees to 180 degrees along the circumference. Five different orientations relative to the baffle support of the modified cannula were evaluated at each position to assess the turbulent flow., Results: The standard cannula produced a maximum force of 0.14 mmHg at 180 degrees from the sensor. The modified cannula generated a maximum pressure of 0.017 mmHg at 67.5 degrees from the sensor. Overall, the force from the modified cannula was an order of magnitude less than the standard infusion cannula (P < 0.01)., Conclusions: The modifications of the new infusion cannula are passive with respect to its surgical performance. However, the additional baffle transforms laminar air currents into turbulent ones and allows dispersion of the jet forces, which should reduce barotrauma and desiccation of the retina during fluid-air exchange.

Purpose: To compare the efficacy of one intraoperative dose of intravenous ketorolac tromethamine to saline placebo in controlling postoperative eye pain, nausea, and sedation following vitreoretinal surgery., Study Design: Prospective, randomized, placebo-controlled clinical trial., Methods: One hundred fourteen vitreoretinal surgical patients were randomized over a 7-month period to a single intraoperative intravenous dose of ketorolac tromethamine or placebo. All patients underwent surgery with intravenous sedation and retrobulbar anesthetic injection and had access to a standard postoperative analgesic regimen. After 2 and 5 hours, eye pain, nausea, and sedation levels were recorded using a standard visual analog scale. Analgesic requirements were tabulated., Results: Patients given ketorolac were significantly less likely to have postoperative eye pain than were placebo-treated patients (P = 0.0043). Fewer ketorolac patients required postoperative analgesia than placebo patients did (P < 0.0001). Four ketorolac patients (7%) required a narcotic compared to 11 placebo patients (20%) (P < 0.0001). Fewer ketorolac patients experienced nausea and sedation than placebo patients did, but this difference was not statistically significant., Conclusion: Ketorolac tromethamine is an effective nonnarcotic choice for postoperative pain control following vitreoretinal surgery. Pain control is particularly important in this group of patients, whose compliance with postoperative procedures and positioning may significantly affect surgical outcome.

Purpose: To report three cases of choroidal neovascularization (CNV) that occurred following surgical disruption of Bruch membrane during macular surgery., Methods: A retrospective case series was compiled. Macular translocation surgery with punctate retinotomy was performed in two patients for subfoveal CNV, and pars plana vitrectomy with epiretinal membrane (ERM) peeling was performed in one patient for idiopathic ERM., Results: CNV developed at a site of subretinal cannulation in two cases of macular translocation, and at a site of inadvertent injury to Bruch membrane with a vitreoretinal pick in the one case that underwent ERM peeling., Conclusions: These cases underscore the need to monitor breaks in Bruch membrane for CNV following macular surgery.

Purpose: To evaluate the extent of neural cell death in eyes with geographic atrophy (GA)., Methods: Ten eyes with GA and five age-matched control eyes were selected for morphometric analysis. The nuclei of the ganglion cell, inner nuclear, and outer nuclear layers were counted in contiguous 100-microm segments from 1,500 microm nasal to 1,500 microm temporal to the fovea., Results: The outer nuclear layer was most severely attenuated in eyes with GA, demonstrating a 76.9% reduction relative to control eyes (P < 0.0001). A significant loss of ganglion cells (by 30.7%) was also observed (P = 0.0008). There was no significant difference in the inner nuclear layer cells (P = 0.30). Among the GA eyes, the nuclei in all three layers were significantly reduced in segments in which the retinal pigment epithelium was completely absent (P

Purpose: To evaluate the extent of neural cell death in eyes with disciform age-related macular degeneration., Methods: Six eyes with disciform degeneration at various stages and five age-matched control eyes were selected for morphometric analysis using digitized light microscopic images. Disciform scars were classified as subneurosensory retinal, subretinal pigment epithelial, or combined lesions. The nuclei of the ganglion cell, inner nuclear, and outer nuclear layers were counted in contiguous 100 microm segments spanning a distance from 1,500 microm nasal to 1,500 microm temporal to the fovea., Results: The outer nuclear layer was most severely attenuated in eyes with disciform scars, demonstrating a 69.4% reduction in cell number relative to control eyes. A loss in retinal ganglion cells (by 7.3%) and an increase in inner nuclear layer cells (by 10%) were observed, but these changes were not significant. Photoreceptor loss was most pronounced when the disciform scar was not covered by the retinal pigment epithelium., Conclusion: The nuclei of the outer nuclear layer are significantly attenuated in eyes with disciform age-related macular degeneration, while the ganglion cell and inner nuclear layers are relatively preserved. These findings suggest that replacement of outer nuclear function, by either retinal transplantation or implantation of the intraocular retinal prosthesis, might be a feasible therapeutic option for patients with this condition.

Purpose: To assess subjective levels of eye pain, nausea, and sedation following vitreoretinal surgery performed with intravenous sedation and retrobulbar anesthesia., Methods: One hundred eighty-five consecutive patients who underwent vitreoretinal surgery and had access to a standard postoperative analgesic regimen prospectively quantified levels of eye pain, nausea, and sedation 2 hours and 5 hours after surgery by using a standard visual analog scale. Analgesic requests were recorded. Responses were statistically analyzed., Results: Fifty-six percent of patients had some eye pain after vitreoretinal surgery; 48% of patients requested an analgesic within 5 hours after surgery. Twenty-seven percent of patients required narcotic analgesia. There was a significant relationship between the presence of eye pain and surgery duration of >2 hours (P < 0.02). Sixteen percent of patients had postoperative nausea, which more likely occurred in those who received a narcotic analgesic (P < 0.02). Eighty percent of patients had postoperative sedation, which more likely occurred in those who received a narcotic analgesic (P < 0.02)., Conclusion: One half of individuals undergoing vitreoretinal surgery, especially those who have lengthy procedures (>2 hours), will request pain medication within 5 hours after surgery; one half of these patients will need narcotic analgesia for pain control. Narcotics may result in nausea and sedation.

Background/purpose: A safe, effective adhesive could be useful in the management of retinal holes or tears and selected complicated retinal detachments, as well as for attaching a small electronic device (retinal prosthesis) to the retina. In this study, we examined nine commercially available compounds for their suitability as intraocular adhesives., Methods: The following materials were studied: commercial fibrin sealant, autologous fibrin, Cell-Tak, three photocurable glues, and three different polyethylene glycol hydrogels. An electronic strain gauge measured the adherence forces between different glues and the retina. The stability of hydrogels at body temperature and the impermeability of the hydrogel adhesive to dextran blue were examined. Long-term biocompatibility testing of the most promising glues in terms of adhesive force, consistency, and short-term safety (hydrogels) were done in rabbits. Funduscopy, electroretinogram, and histology of the retina were performed., Results: Hydrogels had 2 to 39 times more adhesive force (measured in mN) than the other glues tested. They liquefied at body temperature after 3 days to a few months. Hydrogels were impermeable to dextran blue. One type of hydrogel proved to be nontoxic to the retina., Conclusions: Hydrogels proved to be superior for intraocular use in terms of consistency, adhesiveness, stability, impermeability, and safety.

Purpose: To report postsurgical findings in patients with Marfan syndrome and retinal detachment (RD)., Methods: The authors identified and retrospectively reviewed the charts of one cohort of 12 patients (15 eyes) with Marfan syndrome and RD who were operated on at the Wilmer Institute and a second cohort of 16 such patients (24 eyes) who were operated several years earlier and elsewhere., Results: First cohort--Final visual acuity (VA) was 20/80 or better and the retina was flat in all five phakic eyes (100%). The RD occurred after the eye had undergone lens removal in 10 eyes, 6 of which (60%) had a final VA of 20/80 or better, and 9 of which (90%) had a final VA of 5/200 or better (P = 0.20). Second cohort--The final VA was 20/80 or better and the retina was flat in 6 of the 7 phakic eyes (86%). Among 17 aphakic or pseudophakic eyes, only 5 (29%) had a flat retina and VA of 20/80 or better, whereas 12 (71%) had no light perception (P = 0.03)., Conclusions: The results of RD operations done in the past in Marfan patients were worse when the eye was aphakic. In most cases operated more recently, the prognosis for successful RD repair was good regardless of whether the eye was phakic.

Purpose: To test the hypothesis that healthy fetal retinal pigment epithelium (RPE) can rescue the remaining viable RPE and choriocapillaries and thereby the photoreceptors in non-neovascular age-related macular degeneration (ARMD) (geographic atrophy [GA])., Methods: A 65-year-old legally blind woman with non-neovascular ARMD underwent fetal RPE transplantation. Best-corrected visual acuity testing, detailed fundus examination, fundus photography, fluorescein angiography, scanning laser ophthalmoscope macular perimetry, and humoral and cellular immune response testing were performed. A suspension of RPE was infused into the subretinal space through a retinotomy along the superotemporal arcade at the edge of the area of GA. The patient did not take systemic immunosuppressants., Results: The patient's vision remained unchanged for 5 months after the surgery. Fluorescein angiography after transplantation showed leakage and staining at the level of the outer retina. There was progressive subretinal fibrosis in the area of the transplant. Immune response studies showed a weakly positive mixed lymphocyte response against phosducin and rhodopsin., Conclusion: Although it is surgically feasible to transplant fetal RPE to the subretinal space of patients with GA, such an allogenic RPE transplant without immunosuppression leads to leakage on fluorescein angiography and eventual fibrosis. A very weak immune response against proteins associated with photoreceptors is also of concern.

Purpose: To evaluate the safety and efficacy of transscleral diode laser for retinopexy in rhegmatogenous retinal detachment surgery in a multicenter trial., Methods: Seventy-two patients with primary rhegmatogenous retinal detachments were enrolled. No patient with chronic detachment, a retinal break greater than 90 degrees, history of uveitis or infectious retinopathy, or proliferative vitreoretinopathy was enrolled., Results: Information from follow-up of 6 months or longer was available on 65 eyes. Retinas were attached at 6 months with a single operation in 58 (89%) of these eyes. Complications included apparent pinpoint breaks in Bruch's membrane in 15 eyes, scleral-thermal effect in 14 eyes, and limited hemorrhage, which was intraretinal in 10 eyes and extended into the vitreous in 3 eyes. In one case, hemorrhage was judged perhaps to have contributed to initial surgical failure. The other complications had no known adverse effects. Complications were significantly associated with the physician's experience in using transscleral laser retinopexy., Conclusions: In this multicenter study, transscleral diode laser retinopexy served as a safe and effective means of creating chorioretinal adhesion during retinal reattachment surgery. Minor complications were minimized by increasing experience with the technique.

Background: Cryptococcal endophthalmitis is a rare disorder, almost invariably diagnosed after enucleation or at postmortem examination. There are therefore few guidelines as to its identification or treatment., Methods: A case of culture-positive cryptococcal endophthalmitis in a patient with chronic uveitis was diagnosed by vitreous biopsy at the time of retinal detachment repair. The patient was treated with oral fluconazole for 5 months. All reported cases of cryptococcal endophthalmitis were reviewed and compared., Results: After oral fluconazole therapy, the patient was culture negative on repeat tap. Despite conversion to culture-negative status, however, visual acuity declined to hand motions because of hyphema and hypotony. The organism was successfully identified as a non-neoformans species, Cryptococcus laurentii, previously unreported as an ocular pathogen., Conclusion: This unique case demonstrates that cryptococcal disease can be diagnosed antemortem by vitreous biopsy, and should be added to the differential diagnosis in cases of chronic smoldering uveitis. A non-neoformans organism is also identified for the first time as a cause of ocular cryptococcosis. Fluconazole, used here for the only time of which we are aware to treat cryptococcal endophthalmitis, produced successful conversion to culture negativity and resolution of the uveitis.

The major cause of failure in retinal detachment surgery is proliferative vitreoretinopathy. The surgical treatment of proliferative vitreoretinopathy has improved dramatically over the past ten years. Still failure is too common and the need for reoperation remains high. We believe this is often due to anterior vitreous traction. We report our present "conservative" surgical approach to treating such patients using newly developed fiberoptic instrumentation. Using this approach, we have achieved total retinal reattachment in 42 of 48 patients with C3 or greater proliferative vitreoretinopathy with six months or greater follow up.

Penetrating injuries are a leading cause of unilateral visual loss in young patients. We reviewed the records of 197 patients aged 18 years or younger who underwent primary repair of a penetrating ocular injury at the Wilmer Ophthalmological Institute from January 1970 through December 1981. The injury was caused by sharp objects in 49% of cases, missiles in 35%, and blunt trauma in 14%. Of 159 patients with at least 6 months follow-up, 110 (69%) achieved final vision of 5/200 or better, and 77 patients (48%) achieved final visual acuity of 20/50 or better. The prognosis after a penetrating injury is strongly influenced by the nature of the injury and the extent of initial damage. Several factors were found to correlate with an unfavorable visual outcome, including: initial preoperative visual acuity of worse than 5/200, injuries due to blunt trauma, wounds involving the sclera, double penetrating injuries, dense vitreous hemorrhage, and wounds associated with an intraocular "BB" pellet.

The use of silicone oil in the management of intractable retinal detachment associated with proliferative vitreoretinopathy is increasing. Based on clinical experience in over 240 cases, the authors describe their current surgical techniques involving silicone oil, as well as the avoidance and management of associated intraoperative complications.

Vitreous seeding by a peripapillary retinal astrocytic hamartoma in a patient with tuberous sclerosis was supported by vitreous biopsy and the use of millipore filter and a modified Papanicolaou staining technique. Vitreous seeding may be associated with vitreous inflammation and hemorrhage.

The surgical results and postoperative complications associated with silicone oil injection in the treatment of complicated retinal detachment were evaluated in a series of 164 eyes followed for a minimum of 6 months. The incidence of recurrent retinal detachment, corneal decompensation, cataract, glaucoma, and hypotony were noted to be high, and the final visual results were often disappointing. The avoidance and treatment of these complications are discussed, and removal of the silicone oil after sustained anatomic retinal reattachment has been attained is recommended.