Scleral Cross-linking Using Riboflavin and Ultraviolet-A Radiation for Prevention of Axial Myopia in a Rabbit Model.

Myopic individuals, especially those with severe myopia, are at higher-than-normal risk of cataract, glaucoma, retinal detachment and chorioretinal abnormalities. In addition, pathological myopia is a common irreversible cause of visual impairment and blindness. Our study demonstrates the effect of scleral crosslinking using riboflavin and ultraviolet-A radiation on the development of axial myopia in a rabbit model. The axial length of the eyeball was measured by A-scan ultrasound in New Zealand white rabbits aged 13 days (male and female). The eye then underwent 360° conjunctival peritomy with scleral crosslinking, followed by tarsorrhaphy. Axial elongation was induced in 13 day-old New Zealand rabbits by suturing their right eye eyelids (tarsorrhaphy). The eyes were divided into quadrants, and every quadrant had two scleral irradiation zones, each with an area of 0.2 cm² and a radius of 4 mm. Crosslinking was performed by dropping 0.1% dextran-free riboflavin-5-phosphate onto the irradiation zones 20 sec before ultraviolet-A irradiation and every 20 sec during the 200 sec irradiation time. UVA radiation (370 nm) was applied perpendicular to the sclera at 57 mW/cm² (total UVA light dose, 57 J/cm²). Tarsorrhaphies were removed on day 55, followed by repeated axial length measurements. This study demonstrates that scleral crosslinking with riboflavin and ultraviolet-A radiation effectively prevents occlusion-induced axial elongation in a rabbit model.

Scleral cross-linking using riboflavin and ultraviolet-a radiation for prevention of progressive myopia in a rabbit model.

Our study demonstrates the effect of scleral cross-linking using riboflavin and ultraviolet-A radiation on the development of axial myopia in a rabbit model. Axial length of the eyeball was measured by A-scan ultrasound in 22 New Zealand white rabbits aged 13 days. The right eyes then underwent 360-degree conjunctival peritomy with (experimental group, n = 11) or without (control group, n = 11) scleral cross-linking, followed by tarsorrhaphy. The left eyes served as a control eye. In the experimental group, the right eyeballs were divided into quadrants, and every quadrant had either 2 (n = 8) or 6 (n = 3) scleral irradiation zones, each with an area of 0.2 cm² and radius of 4 mm. Cross-linking was performed by dropping 0.1% dextran-free riboflavin-5-phosphate onto the irradiation zones at 20 s before ultraviolet-A irradiation and every 20 s during the 200-s irradiation time. UVA radiation (370 nm) was applied perpendicular to the sclera at 57 mW/cm² (total UVA light dose, 57 J/cm²). Tarsorrhaphies were removed on day 55, followed by repeated axial-length measurement. In the control group, mean axial length in the right eyes increased from 10.50 ± 0.67 mm at baseline to 15.69 ± 0.39 mm 55 days later, for a mean change of 5.19 ± 0.85 mm. In the experimental group, corresponding values were 10.68 ± 0.74 mm and 14.29 ± 0.3 mm, for a mean change of 3.61 ± 0.76 mm. The between-group difference in the change in mean axial length was statistically significant (p < 0.001, Mann-Whitney nonparametric test). The present manuscript demonstrates that scleral cross-linking with riboflavin and ultraviolet-A radiation effectively prevents occlusion-induced axial elongation in a rabbit model.

Novel 'heavy' dyes for retinal membrane staining during macular surgery: multicenter clinical assessment.

PURPOSE: To evaluate the feasibility of two novel 'heavy' dye solutions for staining the internal limiting membrane (ILM) and epiretinal membranes (ERMs), without the need for a prior fluid-air exchange, during macular surgery. METHODS: In this prospective nonrandomized multicenter cohort study, the high molecular weight dyes ILM-Blue™ [0.025% brilliant blue G, 4% polyethylene glycol (PEG)] and MembraneBlue-Dual™ (0.15% trypan blue, 0.025% brilliant blue G, 4% PEG) were randomly used in vitrectomy surgeries for macular disease in 127 eyes of 127 patients. Dye enhanced membrane visualization of the ILM and ERMs, 'ease of membrane peeling', visually detectable perioperative retinal damage, postoperative best-corrected visual acuity (BCVA), dye remnants and other unexpected clinical events were documented by 21 surgeons. RESULTS: All surgeries were uneventful, and a clear bluish staining, facilitating the identification, delineation and removal of the ILM and ERMs, was reported in all but five cases. None of the surgeries required a fluid-air exchange to assist the dye application. BCVA at 1 month after surgery improved in 83% of the eyes in the MembraneBlue-Dual™ group and in 88% in the ILM-Blue™ group. No dye remnants were detected by ophthalmoscopy, and no retinal adverse effects related to the surgery or use of the dyes were observed. CONCLUSION: The 'heavy' dye solutions ILM-Blue™ and MembraneBlue-Dual™ can be injected into a fluid-filled vitreous cavity and may facilitate staining and removal of the ILM and/or ERMs in macular surgery without an additional fluid-air exchange.

The use of ophthalmic viscosurgical devices during donor's corneal harvesting.

PURPOSE OF THE STUDY: To evaluate the protective effect of Biolon (sodium-hyaluronate 1%) use on corneal endothelium, during donor's corneal harvesting. MATERIALS AND METHODS: A prospective, randomized, double blind, comparative study was performed on 120 corneas donated from 60 donors. One cornea from each donor (n = 60) was harvested using intracameral injection of ophthalmic viscosurgical device (OVD), and the fellow cornea with no use of OVD (n = 60, control group). Endothelial cell density (ECD) values were obtained at the center of each corneal graft with a specular microscope one day after harvesting. RESULTS: ECD in the OVD group were 2839 ± 412.5 cells/mm(2) and in the control group 2748 ± 429.7 cells/mm(2). (p = 0.03). Distribution curve of the difference in ECD between the OVD and no OVD eyes showed that 47% of the donors had > 100 cells/mm2 difference, and 32% of donors had >200 cells/mm(2) difference in favor of the OVD group. CONCLUSIONS: Intracameral OVD injection during corneal graft harvesting may protect the endothelium from the mechanical damage induced during the procedure. Further investigation is needed before routine the use of intracameral OVD should be considered in this setting.

Intravitreal bevacizumab treatment for exudative age-related macular degeneration with good visual acuity.

PURPOSE: To investigate the effect of intravitreal bevacizumab on the visual and anatomic outcome of patients with exudative age-related macular degeneration presenting with good visual acuity (VA). METHODS: A file review was performed for all consecutive patients with newly diagnosed exudative age-related macular degeneration and initial VA of ≥ 20/40 treated in 2005 to 2010 and followed for at least 6 months. Treatment consisted of 3 loading doses of intravitreal bevacizumab every 6 weeks and was repeated when fluid or hemorrhage was present. RESULTS: The cohort included 130 patients (150 eyes). Mean follow-up was 20.2 ± 13.2 months, and mean number of injections was 11.3 ± 6.2. At the last examination, VA was stable or improved in 106 eyes (70.7%); 11 eyes (7.3%) lost ≥ 3 lines. Mean logarithm of the minimum angle of resolution VA measured 0.22 ± 0.1 (0-0.3) at presentation and 0.22 ± 0.2 (0-1.3) at the last visit. Corresponding values for central macular thickness were 267 ± 75 µm (137-562) and 226 ± 75 µm (75-568) (P = 0.14). The most frequent complication (18 eyes, 12%) was corneal epithelial defects. CONCLUSION: Prompt intravitreal bevacizumab treatment for newly diagnosed exudative age-related macular degeneration in patients with good initial best-corrected visual acuity is associated with sustained or improved vision and a good safety profile. Attempts should be made to expedite the access of these patients to treatment, regardless of initial VA.

Intravitreal bevacizumab treatment for macular edema due to branch retinal vein occlusion in a clinical setting.

PURPOSE: To report the functional and anatomical outcome of intravitreal bevacizumab (IVB) treatment for macular edema due to branch retinal vein occlusion (BRVO) in a clinical setting. METHODS: The files of 45 patients treated with IVB for BRVO-induced macular edema at a tertiary medical center in 2007-2010 were reviewed. All received three loading doses (1.25 mg) and were followed every 6 weeks. Treatment was repeated for persistent or recurrent edema. If the edema did not resolve after 4-6 injections, grid laser photocoagulation was performed. RESULTS: Mean patient age was 70.7 years (SD 8.5); mean follow-up time, 18.8 months (SD 8.3); mean number of injections, 8.8 (SD 3.8). Fourteen patients (33%) received grid laser treatment before bevacizumab and 23 (51%) after. Mean logMAR visual acuity (VA) was 0.63 (SD 0.43) before treatment (Snellen, 20/140) and 0.4 (SD 0.43) (Snellen, 20/70) after (p < 0.0005). Corresponding central macular thickness (CMT) values were 382.2 microns (SD 155.6) and 320.5 microns (SD 172.8) (p= 0.028). Positive correlations were found between initial VA and initial and final CMT (p = 0.004) and between gain in VA and reduction in CMT (p = 0.03). There was no statistically significant difference in mean initial or final VA and CMT between patients who received grid laser treatment before or during the study and those who did not. CONCLUSIONS: IVB treatment improves visual function and reduces CMT in patients with BRVO-induced macular edema.

Intravitreous bevacizumab treatment for macular edema due to central retinal vein occlusion.

PURPOSE: To investigate the visual and anatomical effects of intravitreal bevacizumab treatment of macular edema due to central retinal vein occlusion (CRVO). METHODS: Data were collected by medical chart review for 35 consecutive patients (35 eyes) with CRVO-induced macular edema treated with intravitreal bevacizumab in 2007-2010 and followed for at least 6 months. All patients received 3-4 loading doses (1.25 mg) with follow-up every 6-8 weeks and repeated injections as necessary. RESULTS: Mean patient age was 65.5 years (SD 13.5); mean follow-up time, 17.7 months (SD 10.8); mean number of injections, 9.3 (SD 5). Mean logMAR visual acuity (VA) improved from 0.9 (SD 0.49) at baseline to 0.7 (SD 0.5) at the last visit; corresponding Snellen values were 6/98 and 6/15 (p = 0.009). Four patients (11%) lost ≥3 lines, and 13 patients (37%) gained ≥3 lines. There was a positive correlation between initial and final VA (p < 0.0005). Central macular thickness (CMT) measured 489.5 microns (SD 175) at baseline and 395 microns (SD 223) at the last visit (p = 0.24). VA gain was positively correlated with CMT reduction (p < 0.0001). CONCLUSIONS: Intravitreal bevacizumab treatment of CRVO-induced macular edema improves vision, especially in patients with good initial VA.

Consecutive appearance of corneal subepithelial infiltrates after intravitreous bevacizumab injections.

PURPOSE: To report a case of corneal subepithelial infiltrates appearing after intravitreous bevacizumab injections. METHODS: A review of the patient's history and clinical examination findings in a patient who had epidemic keratoconjunctivitis more then 20 years before treatment with bevacizumab for age-related macular degeneration. RESULTS: After the third and fourth bevacizumab injections, the patient presented with unilateral corneal subepithelial infiltrates. The infiltrates were accompanied by mild anterior chamber reaction and resolved with topical steroid treatment. CONCLUSIONS: Treatment with intravitreous bevacizumab may precipitate an immune response leading to the appearance of corneal subepithelial infiltrates.

Outcomes and complications of rhegmatogenous retinal detachment repair with selective sutureless 25-gauge pars plana vitrectomy.

PURPOSE: To evaluate the outcomes and complications associated with the repair of rhegmatogenous retinal detachment (RRD) using 25-gauge pars plana vitrectomy (PPV) with selective sclerotomy suturing. DESIGN: Retrospective case series. METHODS: Analysis of consecutive 25-gauge pars plana vitrectomy cases performed for treating RRD was conducted. Reviewed parameters included demographics, ophthalmic history, results of ocular examinations, and intraoperative as well as postoperative complications. Analysis of the surgery digital video disc recordings revealed complications such as iatrogenic retinal breaks and retinal or subretinal tugging by the soft-tip cannula. RESULTS: Forty-two eyes with RRD were evaluated with a follow-up of at least 3 months. The preoperative best-corrected visual acuity (BCVA) of 20 eyes with macula-on RRD ranged between 20/20 and 20/40 (mean, 20/30). Twenty-two eyes with macula-off RRD had preoperative BCVA ranging between 20/70 and hand movements (mean, 20/400). The single-surgery success rate was 97.4%. The final BCVA of the macula-on eyes ranged between 20/20 and 20/40 (mean, 20/30). In the eyes with macula-off RRD, the postoperative BCVA ranged between 20/30 and 20/400 (mean, 20/73). Sutures were placed on at least 1 scleral wound because of intraoperative gas leakage in 36.4% of the eyes. On the first postoperative day, all the eyes receiving gas tamponade had a 100% fill. None of the eyes in the study had postoperative hypotony, ciliochoroidal effusion, or choroidal hemorrhage. No cases of postoperative subconjunctival gas or oil leakage were noted. CONCLUSIONS: Modification of the standard 25-gauge sutureless technique with selective scleral wound suturing may contribute to preventing wound leakage and possible postoperative complications of hypotony or partial tamponade.

Intravitreal bevacizumab as an adjunct treatment for neovascular glaucoma.

PURPOSE: To assess the effect of adjunctive intravitreal bevacizumab treatment on neovascular glaucoma (NVG). METHODS: The medical records of all consecutive patients with NVG treated with intravitreal bevacizumab at our center from May 2006 to February 2008 were reviewed. The data collected included background features, findings on full ophthalmologic examination (including visual acuity, gonioscopy, and intraocular pressure), glaucoma medications prescribed, and additional procedures for glaucoma performed before and after bevacizumab injection.The interval between the diagnosis of NVG and intravitreal bevacizumab treatment was calculated. RESULTS: Eighteen patients (6 male, 12 female; mean age 63-/+13.2 years) met the study criteria. Causes of NVG were proliferative diabetic retinopathy (n=14), central retinal vein occlusion (n=2), occlusive vasculitis (n=1), and panuveitis (n=1). The mean duration of followup was 52 (-/+12) weeks. Mean intraocular pressure decreased from 32.3 (-/+4.99) to 18 (-/+6.1) mmHg (p<0.0001) and mean number of glaucoma medications decreased from 3.16 (-/+1.2) to 2.55 (-/+1.46) (p=0.1938). An interval of less than 6 months between the start of bevacizumab treatment and diagnosis was associated with better final visual acuity than delayed treatment (0.82-/+0.4 logMAR vs 1.88-/+1.1 logMAR, p=0.002) and a better regression of iris neovascularization (22% vs 89%; p=0.015). CONCLUSIONS: Intravitreal bevacizumab is beneficial for the treatment of anterior segment neovascularization and NVG when used as an adjunct, making the administration of additional treatment for the underlying cause possible. Bevacizumab should be instituted promptly after diagnosis, before irreversible anatomic and functional damage occurs.

20-, 23-, and 25-gauge vitreous cutters: performance and characteristics evaluation.

PURPOSE: The flow rate of nine vitreous cutters was tested in two different viscosity environments. Physical features of each probe, such as flexibility and internal lumen diameter, were measured. METHODS: For each probe, a total of 26 aspiration tests were performed in two different liquids. The stiffness of the shaft of each probe was calculated by measuring the displacement of the tip of the probe under a known constant force. All the probes were dissected and the internal lumen diameter was measured. RESULTS: Significant differences in flow rates were observed among the three separate conditions, among the three different gauge groups (25-, 23-, and 20-gauge vitreous cutter), and also within each gauge group. Internal lumen diameters and probe stiffness characteristics are presented. CONCLUSION: Vacuum, internal lumen diameter, and duty cycle seem to be the critical features governing actual flow rate. Knowledge of flow rate and stiffness of the various probes may aid surgeons in making an educated choice and may improve patient care.

Feasibility study of intraocular robotic surgery with the da Vinci surgical system.

PURPOSE: To assess the feasibility of performing intraocular robotic surgery with the da Vinci Surgical System (Intuitive Surgical, Sunnyvale, CA). METHODS: Using modified robotic instruments, 25-gauge pars plana vitrectomy, intraocular foreign body removal, and anterior capsulorhexis were performed with the da Vinci system on porcine eyes. We assessed the surgical system's ability to provide the control, dexterity, maneuverability, and visualization necessary for intraocular surgery. RESULTS: Control of the robotic wristlike instruments allowed for full range of movement. The dexterity of the robotic arms was excellent, with steady instrument motion. Controlling the robotic arms was not as intuitive as moving the wrist. A high stable point of rotation induced motion-related stress at the site of instrument insertion. Visualization of the external operative field during intraocular procedures required camera realignment, and absent retroillumination made anterior segment surgery hard to perform. CONCLUSIONS: The da Vinci Surgical System provided adequate dexterity for performing delicate intraocular manipulations. In the current design, the kinematics of the robotic arms was found to be insufficient for standard intraocular surgery. The system's endoscope did not did not yield the same detail acquired by an ophthalmic microscope.

Association of systemic risk factors with the progression of laser-treated retinopathy of prematurity to retinal detachment.

PURPOSE: To identify systemic risk factors associated with treatment failure of diode laser photocoagulation for threshold or prethreshold retinopathy of prematurity (ROP). METHODS: Consecutive infants treated with laser at our institution and infants referred from outside centers with progression to stage 4 or 5 ROP after laser treatment were included in this retrospective study. Two groups were identified: those with regression of ROP after laser (LT group; 196 eyes of 98 infants) and those with progression to stage 4 or 5 ROP after laser (ST group; 134 eyes of 80 infants). RESULTS: Using a multiple logistic regression model, there was a statistically significant association for progression of treated ROP to retinal detachment (RD) with history of sepsis (P < 0.001), oxygen therapy (P = 0.003), mechanical ventilation (P = 0.003), respiratory distress syndrome (RDS; P = 0.005), and patent ductus arteriosus (PDA; P = 0.042). On the other hand, jaundice (P = 0.563), anemia (P = 0.599), thrombocytopenia (P = 0.689), and blood transfusion (P = 0.878) did not show a statistically significant association with failure of laser treatment and progression of ROP to stage 4 or 5 disease. CONCLUSIONS: Premature infants with a history sepsis, oxygen therapy, mechanical ventilation, RDS, and PDA may be at a higher risk for progression of ROP despite diode laser treatment.

Surgical and functional results of augmented superior oblique muscle z-tenotomy in patients with superior oblique overaction and Brown's syndrome.

BACKGROUND AND OBJECTIVE: To compare the effectiveness of augmented superior oblique Z-tenotomy (SOZT) with fixed standard SOZT in canceling preoperative superior oblique overaction associated with A pattern anisotropia or V pattern in Brown's syndrome. PATIENTS AND METHODS: Sixteen consecutive patients with superior oblique overaction or Brown's syndrome were treated by removal of a triangular piece of the superior oblique tendon near its insertion (augmented SOZT). Outcome was compared with 20 consecutive historical controls after standard SOZT. RESULTS: The decrease in superior oblique overaction in the right and left eyes and fundus intorsion and the collapse of A pattern anisotropia were more significant for patients with superior oblique overaction (P = .003, P = .007, P = .05, P = .0015, respectively) and patients with Brown's syndrome (P = .025, P = .03, and P = .05, respectively). No study patient with superior oblique overaction and A pattern anisotropia required reoperation compared with 5 of 14 controls (37.5%); rates for patients with Brown's syndrome were 0 for the study group and 3 of 6 (50%) for the control group. CONCLUSIONS: Augmented SOZT is superior to standard SOZT for correcting superior oblique overaction, intorsion, A or V pattern, and stereopsis. It is not associated with complications or reoperation. The size of the Z-tenotomy can be modified according to the intraoperative assessment to achieve symmetric results.

Treatment considerations for primary uveal melanoma with choroidal metastasis to the fellow eye.

We report two cases of primary uveal melanoma with metastatic involvement of the contralateral eye. Two female patients presented with primary choroidal melanoma. In the first case, primary enucleation of the affected eye was performed. Two years later, systemic tumor spread with contralateral choroidal melanoma was detected. A decision for observation of the ocular metastasis was made. In the second case, systemic tumor spread was already evident at time of initial diagnosis of the ocular melanoma. Six months later, a choroidal metastasis was detected in the fellow eye. Again, observation was recommended. In conclusion, systemic spread of primary choroidal melanoma may include a choroidal metastasis to the contralateral eye. Observation of the second affected eye may be prescribed.

Peripheral retinopathy and maculopathy in high-dose tamoxifen therapy.

PURPOSE: To describe the clinical, angiographic, and optical coherence tomography (OCT) features of high-dose tamoxifen retinopathy in three male patients. DESIGN: Observational case series. METHODS: A review of history, clinical examination, and findings on fluorescein angiography (FA) and optical coherence tomography (OCT) was conducted. RESULTS: Three male patients receiving high-dose tamoxifen therapy sought treatment for vision loss and a crystalline maculopathy. Crystalline deposits were noted in the peripheral retina of two patients. All the patients showed macular leakage by FA, but cystoid macular edema (CME) on OCT was detected in two patients. Inner retinal hyperreflective deposits were identified by OCT in all the patients. CONCLUSIONS: High-dose tamoxifen therapy may result in peripheral crystalline retinopathy in addition to perifoveal opacities. Angiographic evidence of macular edema may not unanimously correlate with presence of CME on OCT in these cases.

Familial exudative vitreoretinopathy and macular hole exhibited in same individual.

This observational case report is designed to report the first known occurrence of Familial Exudative Vitreoretinopathy (FEVR) and macular hole in the same individual. Clinical exams and fluorescein angiography were used to evaluate patient. Indirect laser panretinal photocoagulation was used to treat the right eye. A nine-year old male was diagnosed with familial exudative vitreoretinopathy in both eyes, as well as a full-thickness macular hole in his right eye. Medical histories indicated that the macular hole was not caused by trauma. Indirect laser panretinal photocoagulation was performed to treat an exudative process caused by FEVR in the right eye, and the exudative retinal detachment regressed. A vitrectomy was later also performed to treat traction retinal detachment as well as macular hole in the right eye. Our conclusion is that macular hole can be associated with familial exudative vitreoretinopathy.

Intravitreous vascular endothelial growth factor (VEGF) inhibitor therapy for tamoxifen induced macular edema.

A 59-year-old male patient with tamoxifen induced macular edema in both eyes was treated with intravitreous sodium pegaptanib. Follow-up clinical examination, OCT, and FA demonstrated reduced edema and leakage with improvement in visual acuity.

Risk for eye splash injury during administration of intraocular injections: a study of retina specialists and fellows.

OBJECTIVE: To evaluate the use of eye protection and frequency of eye splash events during intraocular injections as well as infection risk awareness among retina specialists and fellows in training. METHODS: In a prospective survey of practicing retina specialists and retina fellows, frequency of use and type of eye protection employed during intraocular injections, frequency of eye splash occurrences, description of the eye splash event, number of procedures performed, and awareness of transconjunctival infection risk were investigated. RESULTS: Sixty-four ophthalmologists responded to the questionnaire: 40 retina fellows and 24 retina specialists. The response rate was 100%. Twenty-five percent of the fellows and 33.3% of the specialists reported using eye protection, including corrective glasses, during all intraocular injections. Two of the retina fellows and none of the specialists used special forms of eye protection. Retina fellows had a mean +/- SD of 2.1 +/- 1.3 years experience and the specialists had a mean +/- SD of 10.4 +/- 6.7 years experience in performing intraocular injections. The mean number of injections +/- SD performed by the fellows and specialists was 23 +/- 14.6 and 35 +/- 11.9 per month, respectively. Twelve conjunctival or corneal splash occurrences were reported by six fellows and two retina specialists. Eleven splash events occurred due to reflux of fluid during administration of subconjunctival anesthetic injection, and one event occurred during an anterior chamber tap. Splash events were significantly more likely to occur during procedures performed by fellows, with a relative risk of 8.4 for unprotected procedures (P< 0.001, Fisher exact test). Most (87.5%) of the participants were aware of the risk for transconjunctival viral infection. CONCLUSION: Special eye protection is seldom used during administration of intraocular injections. Although the risk for eye splash during administration of subconjunctival anesthetic before intraocular injections is relatively small, protective measures may be considered when treating high-risk patients.

The slower the better: on the instability of gas jets in a model of pneumatic retinopexy.

PURPOSE: To investigate the effect of injection technique parameters on the formation of multiple gas bubbles in a porcine eye model for pneumatic retinopexy. METHODS: Three hundred twenty-four adult porcine eyes were injected with 0.4 mL of C3F8 with variations in the depth of injection, speed of injection, and size of needle bore. The number of gas bubbles in the eye was assessed with indirect ophthalmoscopy. RESULTS: Shallow injections resulted in a higher incidence of a single bubble than did deep injections (P < 0.001; Fisher exact and Wilcoxon rank sum tests). Slow injections were significantly advantageous in producing a single gas bubble during shallow as well as during deep injections (P < 0.001, Fisher exact and Wilcoxon rank sum tests). With a shallow needle insertion, the slow speed of injection produced a single bubble in 75.9% of the eyes, whereas moderately brisk injections resulted in one bubble in 50.9% of the eyes. During deep needle insertion, 44.4% of the eyes had one bubble if the gas was injected slowly and only 11.1% had a single bubble with moderately brisk gas injections. The bore of the needle did not significantly change the number of bubbles during deep or shallow injections. CONCLUSIONS: The factors that were found to be important in reducing the formation of multiple gas bubbles in the eye were shallow depth of injection and slow speed of gas delivery.