Experimental vitreous replacement with perfluorotributylamine

Perfluorocarbon-perfused vitrectomy: a new method for vitrectomy--a safety and feasibility study

PURPOSE

To evaluate the safety and feasibility of perfluorocarbon-perfused vitrectomy (PCPV) as a technique during vitrectomy for proliferative diabetic retinopathy (PDR) and rhegmatogenous retinal detachment (RRD).

METHODS

In an experimental, prospective, noncomparative and interventional study, 28 eyes of 28 patients were submitted to vitrectomy with the PCPV technique, 18 eyes with PDR and 10 with RRD. In this technique we replaced the traditional balanced saline solution (BSS) in the infusion line (conventional vitrectomy) by perfluorocarbon liquids (PCL). Some patients with PDR were treated with oxygenated PCL. Clinical evaluation, electroretinography (ERG), and endothelial cell count (ECC) were used to assess the safety of this procedure. An arbitrary grading system (grades 0-5) was used to classify the surgery and maneuvers to assess feasibility.

RESULTS

No eyes had a significant decrease in ECC, and the patients in whom we could obtain ERG postoperatively showed no significant differences from preoperative ERG. In the PDR group (using oxygenated and nonoxygenated PCL), 77.8% were graded 5/5; in the RRD group 90% were 5/5. PCPV allowed better visualization of vitreous and intraocular structures, rapid retinal reattachment, less blood in the vitreous cavity, subretinal fluid resolution, blood confinement, retinal stabilization, and easier dissection of epiretinal membranes. In all cases at least one surgical step was eliminated.

CONCLUSIONS

PCPV in humans is a safe and feasible technique. Probably in selected cases the use of PCL offers several advantages over BSS, because of their properties (gravitational forces, immiscibility with bodily fluids, and ability to transport oxygen). Prospective and comparative studies are necessary to establish formal indications and possible contraindications.

Influence on membrane-mediated cell activation by vesicles of silicone oil or perfluorohexyloctane

PURPOSE

This study was conducted to investigate whether macrophage activation through cell membrane attachment might be supported by emulsified tamponade droplets of a certain vesicle size. It has been hypothesized that emulsification of vitreous tamponades might stimulate retinal membrane formation.

METHODS

In this laboratory investigation, similarly sized vesicles of silicone oil and the partially fluorinated alkane perfluorohexyloctane (F6H8) were produced by extrusion through polycarbonate membranes. Human neutrophils were obtained from blood donors. Human monocytes were negatively isolated from mononuclear cells by depletion of other cells. Cell activation status of phagocyting blood neutrophils was measured by a chemiluminescence assay. Fluorescent attached or internalized vesicles were monitored by fluorescent microscopy. The main outcome measures were the altered activation status of monocytes after vesicle incubation and the ability of human macrophages to attach and/or internalize vesicles in vitro.

RESULTS

Extruding silicone oil through a polycarbonate membrane resulted in the production of vesicles that remained stable for at least 2 days. F6H8 vesicles had to be stabilized with an emulsifier, in this case Pluronic PE6800 or Lipoid EPC. The mean vesicle diameter was similar with both components (F6H8: 13.08+/-2.95 microm, silicone oil: 10.05+/-4.6 microm). Neutrophil activation was not influenced by either emulsifier alone or by silicone oil vesicles without emulsifier. Stabilized F6H8 vesicles had a dose-dependent influence on blood neutrophil activation. Only silicone oil vesicles together with Lipoid EPC, not Pluronic PE6800, had a comparable influence on neutrophil activation. Neutrophil activation was influenced neither by 0.125% human serum albumin (HSA) alone nor by vesicles of F6H8 or silicone oil prepared with 0.125% HSA. Monocyte cell membrane attachment of silicone fluid was two times higher than that of F6H8 fluid. F6H8/Pluronic PE6800 vesicles enhanced this process 20-fold, whereas silicone oil vesicles did not enhance cell membrane attachment significantly.

CONCLUSIONS

These in vitro data do not support the hypothesis that emulsification of the tamponades silicone oil or F6H8 in the microenvironment of the eye might easily activate neutrophils or stimulate phagocytosis by monocytes. A prerequisite is the combination of a vesicle shape of the tamponades with specific stabilizing or modifying surfactants. Emulsified tamponades stabilized by artificial surfactants, but not by the naturally occurring protein HSA, favor cell activation by cell membrane attachment.

The effect of specific gravity of perfluorocarbon liquid on the retina after experimental vitreous substitution

PURPOSE

To evaluate the histological changes in the retina after experimental vitreous substitution with various amounts of perfluorophenanthrene (PFPH).

METHODS

Thirty-two rabbit eyes were mechanically vitrectomized and filled up with 0.8 cc or 0.2 cc highly purified PFPH. The substance remained for 1 week (4 eyes), 2 weeks (8 eyes), 4 weeks (10 eyes) or 8 weeks (10 eyes). Eight eyes underwent the same surgical procedure without PFPH exchange to serve as a control group. A histological comparison of corresponding areas in the center and in the periphery of the inferior retina ensued.

RESULTS

After 2 weeks, nuclear drop-downs and irregularities of the outer plexiform layer and of both nuclear layers were observed centrally in the eyes with a 0.8 cc substitution. The changes proceeded to irregularities and cell loss of all retinal layers with focal areas of complete destruction of the retinal architecture after 8 weeks. In contrast, single nuclear drop-downs, wrinkling of the outer nuclear layer and cell loss in the photoreceptor layer were observed in the peripheral retina at the end of the observation period. In those eyes where 0.2 cc PFPH was exchanged nuclear drop-downs were found after 2 weeks, leading to focal thinning of the outer plexiform layer and irregularities of the outer nuclear layer after 4 weeks with an insignificant increase after 8 weeks. At this time these histological alterations were comparable with those that we observed after 2 weeks in the eyes with a 0.8 cc tamponade in a corresponding area.

CONCLUSIONS

Even high purification of PFPH does not prevent retinal damage. The different results in different areas after vitreous substitution with 0.2 cc and with 0.8 cc demonstrate that the high specific gravity of the substance may also play a role in the development of histological changes after extended tamponade.

Perfluorocarbon perfused vitrectomy: animal studies

PURPOSE

To investigate the feasibility and advantages of using perfluorocarbon liquid (PCL) perfusion to remove vitreous during suction-cutting vitrectomy in rabbit and pig eyes.

DESIGN

Experimental study.

METHODS

Balanced salt solution (BSS) was replaced by PCL perfusion during experimental vitrectomy. Oxygenated or nonoxygenated PCL was used in a recycling or a nonrecycling system. Recycling was achieved by two systems: a manual recycling system or a closed-loop system. The experiments in this study consisted of: an in vitro solubility observation, safety and feasibility of vitrectomy in rabbit eyes, effectiveness of vitrectomy with equal vitrectomy time in rabbit eyes, and retinal stability and pigment and blood dispersion in porcine eyes. Toxicity was assessed by a complete ophthalmic examination, endothelial cell count, electroretinography, and histopathology.

RESULTS

Vitreous, blood, and pigments were immiscible in PCL. Manual recycling required less amounts of PCL than nonrecycling (15 vs 25 cc). Oxygenated and nonoxygenated PCL were not toxic. Perfluorocarbon liquid infusion removed more vitreous than balanced salt solution in a 3-minute vitrectomy time using the same settings on the vitrectomy machine. The PCL infusion in porcine eyes stabilized the retina and isolated vitreous cavity from pigment and blood and maintained a clear vitreous cavity.

CONCLUSIONS

These data indicate that perfusion of PCL can be used to remove vitreous with a suction-cutting probe in rabbit and pig eyes. Retinal stability and isolation of the vitreous cavity at the time of vitreous removal along with PCL immiscibility and its specific gravity suggest that PCL has a potential clinical use as an irrigating solution to remove vitreous.

Delivery of midinfrared (6 to 7-microm) laser radiation in a liquid environment using infrared-transmitting optical fibers

Ablation at wavelengths near lambda = 6.45 microm results in tissue ablation with minimal collateral damage (< 40 microm) yet yields a high ablation rate that is useful for human surgery. However, delivery of this wavelength has been limited to that in air and thus to applications in which the target tissue can be readily exposed. The goal of this study is to investigate the potential of a pulsed infrared laser at lambda = 6.45 microm for noncontact ablation in a liquid environment. To this end we investigated fiber delivery in combination with the use of infrared transparent liquids. Transmission characteristics and damage thresholds for two types of fiber materials (silver halide and arsenic sulfide), for high-power pulsed laser radiation were determined using the Mark III free electron laser. Both fibers had comparable bulk losses (0.54 dB/m and 0.62 dB/m, respectively) while the arsenic sulfide fibers showed more coupling losses (37 versus 27%). Damage thresholds were higher in arsenic sulfide fibers than in silver halide fibers (1.12 GW/cm2 versus 0.54 GW/cm2), but both fibers were sufficient to deliver radiant exposures well above the ablation threshold in tissue. Seven different perfluorocarbon liquids (PFCLs), known for their transparency at lambda = 2.94 microm, were investigated and their optical transmission was determined using Fourier transform infrared and direct Beer's law measurements. All of the PFCLs tested had similar values for an absorption coefficient mu(a) at a given wavelength (mu(a) = 0.05 mm(-1) at lambda = 2.94 microm and mu(a) is approximately 3 mm(-1) at lambda = 6.45 microm). Pump-probe imaging showed the ablation sequence (lambda = 6.45 microm) at the fiber tip in a water environment, which revealed a fast expanding and collapsing bubble. In contrast, the volatile PF-5060 showed no fast bubble expansion and collapse, but rather formation of nontransient gas bubbles. Perfluorodecalin did not show any bubble formation at the radiant exposures used. It was shown that using the lambda = 6.45 microm wavelength delivered via fiber optics in combination with perfluorodecalin allows a noncontact laser surgical procedure. Deeper structures, however, are effectively shielded because the radiant exposure of the beam will fall below the ablation threshold owing to the absorption by perfluorodecalin. This may optimize the efficacy and safety of laser-based vitreoretinal surgery.

Our experience with perfluorohexyloctane (F6H8) as a temporary endotamponade in vitreoretinal surgery

PURPOSE

To evaluate the use of F6H8 as a temporary endotamponade for complicated and special cases of retinal detachment instead of silicone oil.

METHODS

We have used F6H8 with 14 patients since February 1999. Eight suffered from rhegmatogenous retinal detachment (RRD) with multiple tears located inferiorly. Three presented inferior traction retinal detachment (TRD) under silicone oil, two suffered from ocular trauma with inferior TRD, and one had idiopathic macular hole. The substance was introduced into the eye after pars plana vitrectomy and membrane peeling if needed, and we tried to introduce as much as possible.

RESULTS

With F6H8 the retina was easily reattached in all cases, like with perfluorocarbon liquids. The postoperative view was very good. F6H8 was removed in all cases after 3-8 weeks. Anatomical success was achieved in 12 out of 14 eyes. Two eyes presented severe PVR. F6H8 entered the anterior chamber in 4 cases, but no corneal complications occurred. In one case there was a marked IOP rise due to an anterior block, treated with superior iridotomy. In two cases retinal detachment (RD) occurred in the upper part and was treated with additional surgery, F6H8 removal and silicone oil injection.

CONCLUSIONS

F6H8 seems to be a promising tamponade agent for special cases of RD.

Perfluorodecalin-induced intravitreal inflammation

OBJECTIVE

To report an unusual case of intravitreal inflammation in a human eye caused by the presence of residual perfluorodecalin in a case of giant retinal tear and retinal detachment.

METHODS

The posterior capsule of the lens, which was infiltrated with deposits, was collected during surgery. The specimen was stained with hematoxylin and eosin, with periodic acid-Schiff, and for melanin. Part of it was examined with electron microscopy. Immunohistochemical staining was performed to demonstrate CD68 antigens, cytokeratin, and glial fibrillary acid protein.

RESULTS

Vacuolated macrophages and retinal pigment epithelial cells infiltrated the posterior capsule. Electron microscopy showed the presence of membrane-lined vacuoles within the macrophages. A monolayer of epithelial cells covered the cellular infiltration.

CONCLUSION

Residual perfluorodecalin can induce an intraocular chronic macrophage response.

Perfluorohexyloctane as a long-term vitreous tamponade in the experimental animal. Experimental perfluorohexyloctane substitution

BACKGROUND

Perfluorocarbon liquids are established tools for intraoperative hydrokinetic retinal manipulation. Because of their high specific gravity, however, they may well cause mechanical damage to the retina when applied as a long-term vitreous substitute. Perfluorohexyloctane, a semi-fluorinated liquid fluorocarbon of low specific gravity (1.35 g/cm3), was developed as a long-term vitreous substitute. The current study was performed to investigate intraocular tolerance to perfluorohexyloctane as a long-term vitreous substitute in the experimental animal.

METHODS

34 vitrectomised eyes of pigmented rabbits underwent intravitreal injection of 1.0-1.2 ml perfluorohexyloctane or balanced salt solution. In 5 eyes the anterior chamber was filled. During the follow-up period of 3 months, the eyes were examined by slit lamp biomicroscopy, by fluorescein angiography and by electroretinography. The eyes were then enucleated and processed for light- and electron microscopy.

RESULTS

Perfluorohexyloctane depicted dispersion beginning between the first and third week. Over a period of 9 weeks no toxic effect on retina, lens and cornea was noticed. At 14 weeks ERG showed a slight decrease in amplitude and early morphological changes in the retina.

CONCLUSION

Our findings suggest that perfluorohexyloctane is tolerated in the rabbit eye for 9 weeks. Since avascular rabbit retina is more susceptible to mechanical and toxic damage than vascularized primate retina, we suggest proceeding from here, and investigating tolerance and quality of tamponade of perfluorohexyl octane in man for up to 9 weeks.

Effects of perfluorophenanthrene on the human corneal endothelium

PURPOSE

To evaluate long term effects of perfluorophenanthrene (C12F24), a perfluorocarbon fluid, in the anterior chamber of the human eye, where residual perfluorophenanthrene was retained in the eye postoperatively.

METHODS

Slit-lamp microscopic and specular microscopic examinations of 4 eyes of 4 patients were performed. All the eyes were operated for complicated retinal detachment and they were aphakic at the end of the operations. Mean follow-up period was 10 months (6-16 months).

RESULTS

Perfluorophenanthrene was seen in the anterior chamber in the first or second day postoperatively as a single drop. In the third week, postoperatively, the first signs of splitting the perfluorophenanthrene (fish egging phenomena) was observed. There was no sign of corneal or anterior segment toxicity with the slit lamp microscopic examinations and the intraocular pressure was within normal limits during the follow-up period in all eyes. However, some structural changes of the corneal endothelium were shown by specular microscopy. Decreased endothelial cell density, a reversed illumination pattern in which the normally dark cellular boundaries appear bright and intracytoplasmic light reflecting bodies were signs of cellular damage at the contact sites whereas no significant changes were seen at non contact sites.

CONCLUSION

Residual perfluorophenanthrene in the anterior chamber does not induce gross corneal damage or ocular inflammation, although structural changes indicating the damage of the corneal endothelium, can be shown by specular microscopy at the contact sites. Corneal endothelial changes seem to arise from a contact-dependent effect of the perfluorophenanthrene.

Outcomes and complications associated with perfluoro-n-octane and perfluoroperhydrophenanthrene in complex retinal detachment repair

OBJECTIVE

To compare rates of perfluorocarbon liquid (PFCL) intraocular retention, anatomic and visual acuity outcomes, and complications associated with intraoperative perfluoro-n-octane (Perfluoron) versus perfluoroperhydrophenanthrene (Vitreon) in retinal detachment repair.

DESIGN

Retrospective noncomparative consecutive case series.

PARTICIPANTS AND METHODS

Records of consecutive patients who underwent retinal detachment repair with intraoperative Perfluoron (n = 78) or Vitreon (n = 84) at the Bascom Palmer Eye Institute between November 1, 1991, and October 31, 1994, were retrospectively reviewed.

MAIN OUTCOME MEASURES

Comparison of PFCL intraocular retention rates, retinal reattachment rates, visual acuity outcomes, and postoperative complication rates between the Perfluoron and Vitreon groups at postoperative day 1, week 1, month 1, month 3, and month 6. To assess the study's generalizability, the Perfluoron data obtained in the current study were compared with results of the Perfluoron Multicenter Clinical Study.

RESULTS

Retained Perfluoron was noted less frequently (P < 0.03) than retained Vitreon at each postoperative visit assessed. At 6 months postoperatively, the cumulative rate of retained PFCL (noted at any of the study postoperative visits) was 7.8% of patients in the Perfluoron group and 38.3% in the Vitreon group (P < 0.001). No significant difference was found between the groups in retinal reattachment rates; at 6 months postoperatively, the retina was attached in 45 of 61 (74%) patients in the Perfluoron group and 36 of 57 (63%) patients in the Vitreon group (P = 0.2). There was a trend (P = 0.055) toward better 6-month visual acuity in the Perfluoron group compared with the Vitreon group. The incidence of corneal abnormality at any study visit was 46.7% in the Perfluoron group and 77.4% in the Vitreon group (P < 0.001). At 6 months postoperatively, the cumulative rate of elevated intraocular pressure (>25 mmHg) was 13 +/- 4% and 37 +/- 5%, respectively (P = 0.004). No significant difference was found between the groups in rates of postoperative hypotony. Results of this study are comparable with those observed in the Perfluoron Multicenter Clinical Study.

CONCLUSIONS

In this study, Perfluoron was retained intraocularly less frequently than Vitreon. Although no significant difference was found between groups in retinal reattachment rates, Perfluoron is associated with slightly better 6-month visual acuity and lower rates of corneal abnormality and elevated intraocular pressure compared with Vitreon.

Perfluorocarbons and semifluorinated alkanes

The introduction of perfluorocarbons (PFCs) and, more recently, semifluorinated alkanes (SFAs) has greatly facilitated vitreoretinal surgery. A distinction is made between the use of these substances as intraoperative tools and internal tamponade agents. This article reviews the physical and chemical properties of PFCs and SFAs and discusses the indications, results, and complications. The effectiveness of these substances as internal tamponade agents is discussed with reference to the specific gravity, contact angle, viscosity and ability to fill model eye chambers and the vitreous cavity. The evidence for the toxicity in animal and human is examined.

The adjunctive use of perfluoroperhydrophenanthrene (Vitreon) in diabetic vitrectomy

OBJECTIVE

To evaluate the utility and efficacy of perfluoroperhydrophenanthrene in the management of retinal detachments secondary to severe proliferative diabetic retinopathy.

PATIENTS AND METHODS

Forty consecutive patients with proliferative diabetic retinopathy and retinal detachments were entered into the study at nine participating clinical centers. Perfluoroperhydrophenanthrene (Vitreon) was used as an adjunct to pars plana vitrectomy and membranectomy.

RESULTS

Preoperative diagnoses included combined traction and rhegmatogenous retinal detachments in 23 eyes (57.5%), traction retinal detachments in 13 eyes (32.5%), and recurrent rhegmatogenous retinal detachments in 4 eyes (10). Vitreous hemorrhage was present in 17 eyes (42.5%). Preoperative visual acuity ranged from light perception or hand motion in 28 eyes (70%) to 5/200 or greater in 12 eyes (30%). Vitreon was primarily used to flatten the retina following relaxing retinotomy in 12 eyes (30%), to displace subretinal fluid in a posterior-to-anterior direction without performing a drainage retinotomy in 15 eyes (37.5%), and to manage intraoperative complications such as iatrogenic tears in 8 (20%) and retinal dialysis in 5 eyes (12.5%). The retina flattened intraoperatively in all cases, facilitating administration of laser photocoagulation. Patients were followed for a minimum of six months (mean 13.2 months). At last follow up, the macula remained attached in 37 eyes (92.5%), including 31 (77.5%) in which the retina was totally attached. The retina remained detached in 3 eyes (7.5%). Visual acuity improved postoperatively in 20 patients (50%), was unchanged in 13 patients (32.5%), and worsened in 7 patients (17.5%).

CONCLUSIONS

Perfluoroperhydrophenanthrene is a useful and effective intraoperative tool for the management of complex retinal detachments secondary to severe proliferative diabetic retinopathy.

Proliferative vitreoretinopathy: an overview

Proliferative vitreoretinopathy (PVR) is still the most common cause of failure of surgery for rhegmatogenous retinal detachment, despite the substantial effort that has been devoted to better understanding and managing this condition during the past 25 years. Basic research has indicated that PVR represents scarring, the end stage of the wound-healing process that occurs after retinal detachment surgery. Medical treatment has been directed toward preventing inflammation, the first phase of the wound healing process, and inhibiting cell proliferation, the second phase. The 1983 Retina Society classification was modified in 1989 by the Silicone Study Group, whose classification differentiates between posterior and anterior forms of PVR and recognizes three patterns of proliferation: diffuse, focal, and subretinal. The anterior form has a worse prognosis than the posterior form, and its treatment requires more complex surgical procedures. In this review, risk factors and pathobiology of PVR are discussed, and management of PVR of various degrees of severity are considered.

Retained perfluorodecalin after retinal detachment surgery

Liquid perfluorocarbons such as perfluorodecalin are widely used as intraoperative vitreous substitutes in certain complicated vitreoretinal conditions. Retained perfluorodecalin postoperatively has been reported to be associated with retinal damage and other complications. We report on a case of retained intraocular perfluorodecalin for eleven postoperative days after retinal detachment surgery with good anatomical and visual outcome after one year follow-up.

Histopathology of rabbit eyes with silicone-fluorosilicone copolymer oil as six months internal retinal tamponade

Silicone-fluorosilicone copolymer oil has low viscosity (175-185 cSt) and is heavier than water (density, 1.16 g cm-3). Short term retinal tolerance (within 2 months) of the silicone-fluorosilicone copolymer oil has been reported to be the same as that of currently used intraocular silicone oil. Ocular response of the purified silicone-fluorosilicone copolymer oil were examined clinically and histopathologically from 2.5 months to 6 months after vitreous cavity injection in rabbit phakic eyes, and compared the oil tolerance with that of purified silicone oil (0.97 g cm-3, 5000 cSt). The effects in anterior chamber also were examined within 4 weeks of the silicone-fluorosilicone copolymer oil injection in different rabbits. Silicone-fluorosilicone copolymer oil recovered from the vitreous cavity at 6 months was analysed for cholesterol and retinol content by high performance liquid chromatography. Because of its low viscosity, silicone-fluorosilicone copolymer oil was easy to inject and remove from the vitreous cavity with a 20-G needle. After the vitreous injection, discrete droplet formation by the silicone-fluorosilicone copolymer oil occurred more easily than by silicone oil. Medullary ray detachment was seen in a silicone oil-, and some silicone-fluorosilicone copolymer oil-injected eyes at 4-6 months. Histopathologically, after 3-6 months disappearance of outer plexiform layer and disorganization of the photoreceptor layer of silicone oil-, and silicone-fluorosilicone copolymer oil-injected eyes were seen in the superior and the inferior retina, respectively. Migration of the photoreceptor cell nuclei to the photoreceptor layer was found in the inferior retina of silicone-fluorosilicone copolymer oil-injected eyes at 5-6 months. Small droplets ingested by mononuclear cells were found in the vitreous cavity or preretina at 4-6 months in silicone-fluorosilicone copolymer oil-injected eyes. After the anterior chamber injection, silicone-fluorosilicone copolymer oil induced endothelial cell damage in the area where the oil contacted continuously. Retinol and cholesterol were identified in silicone-fluorosilicone copolymer oil removed from the vitreous cavity. Silicone-fluorosilicone copolymer oil may be useful as an intraoperative device in retinal detachment surgery and as a short term (up to about 2 months) retinal tamponade but we do not recommend it for long term retinal tamponade.

Perfluorocarbon heavy liquids

Perfluorocarbon heavy liquids (PFCL) facilitate the transfer of intraocular fluids and other surgical manoeuvres in the posterior segment. The indications for their use have extended since first described in the management of giant retinal tears and proliferative vitreoretinopathy. We reviewed our personal experience and experimental evidence, including adverse effects, in order to identify the current role of PFCL in vitreoretinal surgery. High-density perfluorocarbon liquids represent a major advance in vitreoretinal surgery. Clinical experience and the results of experimental investigations have demonstrated their efficacy and safety as an intraoperative device.

Perfluorocarbon liquids in ophthalmology

Perfluorocarbon liquids have been used to facilitate surgery in a wide variety of conditions, including proliferative vitreoretinopathy, giant retinal tears, drainage of suprachoroidal hemorrhages, diabetic traction, retinal detachments with a rhegmatogenous component, dislocated crystalline or intraocular lenses, and retinal detachment associated with choroidal coloboma. The clarity of perfluorocarbon liquids, with a refractive index close to that of water, allows the use of a conventional contact lens for vitreous surgery while the low viscosity facilitates tissue manipulation, injection, and removal. All perfluorocarbon liquids when used as tamponading agents can compress and disorganize the retina. This "toxicity" is a physical effect rather than chemical toxicity and depends upon the amount of perfluorocarbon liquid injected. Perfluorocarbon liquids are not tolerated in the anterior chamber, causing corneal edema within 2-3 days at the site of contact.

The effect of chemical stability and purification of perfluorocarbon liquids in experimental extended-term vitreous substitution

BACKGROUND

To determine the importance of chemical stability and purification of perfluorocarbon liquids (PFCLs) in experimental retinal tolerance, we tested four different substances as long-term vitreous tamponade: purified and nonpurified perfluorodecalin (PFD) and perfluoro-octyl-bromide (PFOB).

METHOD

After mechanical vitrectomy we replaced the vitreous of 65 rabbit eyes. Five groups were formed; four of them received the four PFCLs, while one served as control and received Ringer solution. The eyes were observed clinically every week and examined histologically after 1, 2, 4 and 8 weeks.

RESULTS

After 1 week we observed foam cells and intraretinal macrophages in all eyes with PFCLs. Purified PFD caused retinal lesions in the photoreceptor, ganglion cell and outer nuclear layers after only 2 weeks in the lower part of the eyes. In eyes filled with purified PFOB we observed more pronounced damage of the same nature. Unpurified substances caused severe inflammation and retinal detachment.

CONCLUSION

Our study demonstrates that purification and chemical stability are important factors in retinal tolerance of PFCLs for vitreous replacement. Although purified PFD was tolerated by the rabbit eyes for 1 week, we cannot recommend this substance for short-term clinical use as a vitreous substitute.

Experimental long-term vitreous replacement with purified and nonpurified perfluorodecalin

Highly purified perfluorodecalin and non-purified perfluorodecalin were injected into 34 rabbit eyes after mechanical vitrectomy, to evaluate the retinal tolerance in long-term vitreous replacement and the effect of purification on the retina. Unpurified perfluorodecalin caused severe inflammation and retinal damage and eight of ten eyes had developed retinal detachment after four weeks. In the eyes injected with highly purified perfluorodecalin, we observed retinal changes of the rod and cone, outer nuclear, and ganglion-cell layers after two weeks, which progressed to a thinning of all layers of the retina by the fourth week and to localized areas of retinal atrophy by the eighth week in the lower part of the eyes. Retinal damage in the upper part occurred after four weeks. At the perfluorocarbon-aqueous interface we observed a band of retinal necrosis that was noticed in all eyes four weeks postoperatively regardless of the substance used.

Perfluorooctylbromide (PFOB) as a vitreous substitute in non-human primates

We evaluated the toxicity of perfluorooctylbromide in the primate eye as a short-term postoperative vitreous substitute. Four eyes of 4 African green monkeys underwent complete vitrectomy and vitreous replacement with 1.5-2.0 ml of PFOB. One additional animal received BSS as a control vitreous substitute in one eye. Animals were examined twice weekly for clarity and consistency of the vitreous replacement substance. Anterior segment and lenses remained clear in all eyes, although in the immediate postoperative period one eye became inflamed and had a culture-negative vitritis. The other eyes showed a minimal anticipated postoperative vitreous inflammation. Emulsification of the PFOB began within 3 days of injection and progressed up to 3 weeks, precluding fundus examination and fluorescein angiography after 2 weeks. Eyes were enucleated and light microscopy performed at 2 days, 10 days, 33 days, and 45 days. No toxic effects to the retinal cells were detectable by histological examination, but perivasculitis of retinal vessels was noted at 45 days. Indirect examination was normal up to 10 days; thereafter, the fundus view was obscured by the emulsified PFOB. Because of cellular migration into the vitreous cavity and retinal perivasculitis, observed histologically, PFOB seems most suitable for intraoperative rather than postoperative use.

The use of perfluorophenanthrene in the removal of intravitreal lens fragments

We used perfluorophenanthrene as an aid during pars plana vitrectomy in removing lens fragments dislocated posteriorly into the vitreous cavity. We reviewed the records of all patients (nine patients, nine eyes) referred with dislocated lenses who underwent pars plana vitrectomy with perfluorophenanthrene for removal of the lens material at Wills Eye Hospital from July 1, 1991, through Nov. 30, 1991. The lens was dislocated during cataract extraction in eight eyes. One eye had a history of nonpenetrating trauma and lens dislocation. In one eye a retinal tear was noted intraoperatively and treated. Six eyes had postoperative visual acuity of 20/50 or better. In all eyes the lens was removed without further complications. Perfluorocarbon liquids such as perfluorophenanthrene may enhance current techniques by allowing easier and safer removal of displaced lens fragments through the pars plana.

Retinal detachment and degeneration in response to subretinal perfluorodecalin in rabbit eyes

Perfluorocarbon liquids, including perfluorodecalin (PFD), are useful intraoperative tools in complicated vitreoretinal surgery, such as for giant retinal tears, PVR and intraocular foreign bodies. Due to its high specific gravity (1.91 g/cm3) subretinal complications of PFD may occur. The consequences of subretinal PFD were studied in 23 albino rabbits (28 eyes). Using glass micropipettes (outer diameter of tip 100-120 microns), we injected 50-100 microliters PFD subretinally via the vitreous space. The same volume of BSS was injected into 18 control eyes. Eyes were monitored by indirect ophthalmoscopy and examined by light and electron microscopy at 1, 2 and 3 days, and 1, 2, 4 and 12 weeks. Progressive retinal detachments with newly formed retinal breaks in the inferior quadrants developed in 26 of 28 eyes. As early as 1 day postoperatively, subretinal PFD induced loss of outer and inner segments. Local retinal necrosis occurred in 4 eyes 5-7 days after surgery. A marked vacuole formation in retinal layers and PFD emulsification were regularly seen. The control eyes healed spontaneously. Toxic, mechanical and barrier effects may have caused the retinal damage. We advocate that all PFD be removed from the eye at surgery.

Vitreon, a short-term vitreoretinal tamponade

This investigation of the liquid perfluorocarbon, perfluorophenanthrene (Vitreon), establishes its safety and efficacy as a short-term vitreoretinal tamponade. We utilised Vitreon as an intraoperative tool and postoperative vitreoretinal tamponade in 16 patients. Proliferative vitreoretinopathy (PVR) (six), giant retinal tear (four), rhegmatogenous retinal detachment (three), retinal detachment with keratoprosthesis (two), and submacular and vitreous haemorrhage (one) were successfully repaired. Vitreon was left in the eye and removed 5 days to 4 weeks postoperatively. Complications encountered included proliferative PVR (five), limited peripheral retinal detachment (three), macular pucker (two) cataract (three), hypotony (two), excessive fibrin reaction (one), and elevated intraocular pressure (one). At the latest evaluation, all retinas are attached with a follow-up of 1.25 to 12 months (mean 6.8 months).

Perfluorocarbon heavy liquids in the management of posterior dislocation of the lens nucleus during phakoemulsification

We report a case in which the lens nucleus dislocated into the vitreous cavity through a posterior capsular rupture during phakoemulsification. We performed a vitrectomy and removed the lens nucleus using the perfluorocarbon heavy liquid perfluoro-1,3-dimethylcyclohexane. The management of posterior dislocation of the lens nucleus during cataract surgery is discussed.

Quantitative determination of the partial oxygen pressure in the vitrectomized rabbit eye in vivo using 19F NMR

We have quantitatively determined the PO2 within the vitreous space of the vitrectomized rabbit eye in vivo using 19F NMR and perfluorotributylamine (FTBA). The results of the present work are in good agreement with previous measurements of PO2 within the nonvitrectimized rabbit eye obtained using oxygen microelectrodes. In this study, good precision and accuracy were achieved by: (1) having minimal flow effects present, (2) optimizing the signal-to-noise through the use of neat FTBA and surface coil transmission and detection, and (3) performing an inversion-recovery pulse sequence, with adiabatic pulses, to optimize the dynamic range of the T1 experiment. Possible deleterious effects on the measured T1 due to the vitreal temperature gradient are discussed. To the best of our knowledge the results of this study demonstrate, for the first time, a quantitative determination of intraocular PO2 in vivo using 19F NMR and FTBA.

Vitreon, a new perfluorocarbon

We evaluated a new liquid perfluorocarbon, perfluorophenanthrene (Vitreon). This material has proven to be non-toxic in vitrectomised rabbit eyes for up to six weeks. Present investigation under FDA guidelines establishes both the safety and efficacy of Vitreon in human eyes. We used Vitreon for intraoperative hydrokinetic retinal manipulation in 15 patients. In cases of proliferative vitreoretinopathy (6), rhegmatogenous retinal detachment (5), giant retinal tear (2), retinal dialysis (1), and tractional retinal detachment (1) the retina was successfully reattached. Postoperatively two patients developed proliferative vitreoretinopathy necessitating further surgery, and one patient developed hypotony. Follow-up showed 100% reattachment rate with a mean duration of 6.3 months. Postoperative visual acuity ranges from light perception to 20/30.

Fluorosilicone oil in the treatment of retinal detachment

We evaluated the use of a heavier-than-water fluorinated silicone oil in the treatment of 30 selected cases of complicated retinal detachment from January 1988 to July 1989. Proliferative vitreoretinopathy grade C-2 or greater accounted for 19 cases, proliferative diabetic retinopathy with traction detachment for two cases, giant retinal tears five, ruptured globe with retinal detachment two, massive choroidal effusion with retinal detachment one, and acute retinal necrosis with retinal detachment one. Initial retinal reattachment was achieved in all cases. Complications included redetachment seven (23%), cataract six (75% of phakic patients), raised intraocular pressure four (13%), hypotony four (13%), keratopathy three (10%), uveitis-synechia formation three (10%), phthisis two (3%), choroidal haemorrhage one (3%), and vitreous haemorrhage one (3%). Postoperative visual acuities with at least six months' follow-up range from no light perception to 20/50, with seven patients (23%) 20/400 or better.

Intraoperative perfluorocarbon liquids in the management of proliferative vitreoretinopathy

Three low-viscosity perfluorocarbon liquids were used intraoperatively for hydrokinetic manipulation of the retina during vitreous surgery for retinal detachment with advanced proliferative vitreoretinopathy. All 23 patients had massive proliferative vitreoretinopathy (Grade D, Retina Society classification), and 16 (69.6%) had Grade D-3 with a closed-funnel configuration. In 21 eyes the retina could be flattened intraoperatively by perfluorocarbon liquids without requiring posterior retinotomy for internal drainage of subretinal fluid. The temporary mechanical fixation of the retina provided by this tool facilitated the removal of epiretinal membranes and release of traction. Fifteen eyes (65.2%) maintained long-term retinal reattachment with follow-up of six months or more. These liquids are useful adjuncts in the management of retinal detachment with severe proliferative vitreoretinopathy.

The cisternal anatomy of the vitreous body

The vitreous body of a fresh human eye removed post-mortem was suspended in water after stripping off the corneo-scleral coat, the choroid and the retina. Coloured ink was injected into the vitreous specimen from different points. Individual spaces (cisterns), transitions between cisterns and specialized structures like the bursa premacularis and the canal of Cloquet became visible when partial filling had taken place. SEM-examination of the individual cisternal walls of the bursa premacularis area showed networks of different density and texture and a considerable variation in the length of the individual fibres, some of which extended from the ciliary body (Retzius bundles) to the rim of the Bursa premacularis.