AIM To evaluate the feasibility of cellular photoablation using fluorescence generated photoreaction products as a method to control postoperative fibrosis.
METHODS The fluorescent probe, 2′,7′- bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein, acetoxymethyl ester (BCECF-AM) is a cell membrane permeable compound rendered membrane impermeable and fluorescent upon cleavage by intracellular esterases. Rabbits (ChBB:CH; n=20) received a unilateral subconjunctival injection of BCECF-AM (40, 70, 80, or 100 μg) 30 minutes before surgery followed by intraoperative illumination with diffuse blue light (450–490 nm; 51.9×103cd/m2) for 10 minutes. Controls received either the probe or illumination. Antifibrotic efficacy was established by clinical response and histological examination. Clinical response was assessed by comparing intraocular pressure (IOP) between the treated experimental eye and the fellow eye, which served as control. Success was defined by >20% difference in IOP.
RESULTS IOP was significantly decreased in all groups within 4 days postoperatively. In control groups IOP rose within 10 days to normal levels. This was similar in the group receiving 40 μg of BCECF-AM. In the other groups (subconjunctival injection of 70–100 μg BCECF-AM) IOP was significantly (p < 0.02) decreased for 2–3 weeks. Clinical and histological examination revealed no toxic damage to adjacent tissues.
CONCLUSIONS Cellular photoablation in contrast with chemotherapeutic agents acts on cells that have incorporated BCECF-AM and have been exposed to light at the appropriate wavelength. Though safety and reliability demand further studies this method might be an useful therapeutic approach to control postoperative fibrosis in humans undergoing filtration surgery.
- filtration surgery
- cellular photoablation
Statistics from Altmetric.com
Trabeculectomy is the most frequently applied surgical method to reduce intraocular pressure in patients with glaucoma. The wound healing process within the filtration area, however, is a cause of surgery failure in 30% of the cases.1 Characteristically fibroblasts from Tenon's capsule and episclera lead to a fibroproliferative response involving and closing the created fistula.2-12 To increase the success rates of filtering surgery several substances, such as corticosteroids13-15and antimetabolites (mitomycin-C and 5-fluorouracil)4 16-34 have been used as antifibrotic therapy. Despite their positive effect on filtering surgery success rates, diffusion into adjacent ocular tissues also leads to toxic effects on other than the targeted cells.35-37Photodynamic therapy might be an alternative way to solve this problem.9 38 Photosensitisers can be used as mediators of light induced cell toxicity. They seem to act via the formation of reactive oxygen intermediates and free radicals.39-42Selective activation of the photosensitiser by light application at the appropriate wavelength limits the drug effect on the targeted area. Several photosensitisers are currently under investigation in ophthalmology.43-47 Cellular photoablation can be mediated by BCECF-AM (2′,7′-bis- (2-carboxyethyl)-5-(and-6)-carboxyfluorescein, acetoxymethyl ester). The photoablative effect of this carboxyfluorescein derivate has been shown before40 47 and could be confirmed in our previously conducted in vitro experiments. BCECF-AM is a cell membrane permeable compound rendered membrane impermeable and fluorescent upon cleavage by intracellular esterases (Fig 1A). Exposure of cells that have incorporated BCECF-AM to light at the appropriate wavelength leads to cellular photoablation (Fig 1B, C).
The objective of this pilot study was to investigate the impact of cellular photoablation mediated by BCECF-AM on fibrosis after filtration surgery in a rabbit model.
ANIMALS AND GROUPS
All experiments were performed with chinchilla bastard rabbits (ChBB:CH), 3–6 months old and weighing 1.5–2.5 kg. Animals were obtained from Thomae (Biberach) and maintained in a common room of the vivarium at the university eye clinic of Cologne where an overhead fluorescent light provided 12 hour cycles of light and dark. Surgery and clinical examinations were performed under anaesthesia induced by intramuscular injections of 1.5 ml ketamine (Ketanest, Parke Davis, Berlin, Germany) and 0.7 ml xylazine (Rompun, Bayer, Leverkusen, Germany). Before enucleations of treated eyes animals were killed by lateral ear vein injection of 1 ml T-61 (Hoechst, Frankfurt, Germany). All experimental procedures conformed to the ARVO resolution on the use of animals in research and to institutional guidelines.
Surgical procedure was performed on 18 rabbits subdivided into the following six groups: group A and D served as controls; group A (n=3) received no photosensitiser, but intraoperative illumination with diffuse blue light (10 minutes); group D (n=3) received 30 minutes before surgery a subconjunctival injection of the photosensitiser (80 μg), but no intraoperative illumination; groups B, C, E, and F (n=3 per group) received before surgery different concentrations of the photosensitiser (40–100 μg) and intraoperative illumination with diffuse blue light (10 minutes). Two animals served as controls without surgery. Both animals were given a subconjunctival injection of photosensitiser (80 μg). Thirty minutes later one of these animals received a complete illumination of the anterior part of the eye for 10 minutes. The other animal received no illumination (Table 1). In all groups the fellow eye served as additional control.
PHOTOSENSITISER AND PHOTOACTIVATION
The fluorescent probe, 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein, acetoxymethyl ester (BCECF-AM) was provided by Sigma-Aldrich (Steinheim) and diluted in balanced salt solution containing DMSO to experimental concentration. Animals received 30 minutes before surgery a unilateral (right eye) subconjunctival injection (100 μl) of a BCECF-AM (40–100 μg) solution in the superior quadrant (Fig 2). After photosensitiser injection rabbits remained in unchanged lighting conditions. Illumination with diffuse blue light (450–490 nm; 51.9×103 cd/m2, 10 minutes) was performed after peritomy and removal of superfluous, non-incorporated BCECF-AM by BSS, before filtration surgery. Photoactivating light was delivered by a portable slit lamp (Zeiss) equipped with a blue filter fixed at a distance of 10 cm from the target. The light was focused onto an area of 5.0 mm diameter (ø) comprising conjunctiva and episcleral tissue and sparing the cornea that was partly covered by the reverted conjunctival peritomy flap. The illuminated area encompassed the consequently produced sclerostomy (ø 1.5 mm). Since the illuminated area was well focused, no mask was used to minimise light absorption to adjacent tissues.
The surgical procedure requiring 5–10 minutes was performed with the aid of a surgical microscope on previously anaesthetised animals. Additionally, local anaesthesia with oxybuprocaine drops (Novesine 0.4%, Ciba Vision) was applied. After placement of a lid speculum, a peritomy at 3.0–4.0 mm from the limbus was created in the superior quadrant of the right eye. To prevent intraocular entry of the fluorescent probe, filtration surgery was preceded by removal of non-incorporated BCECF-AM. Illumination of the surgical area was performed before filtration surgery. The entry into the anterior chamber was created with a trephine (ø 1.5 mm) posterior to the limbal insertion of the conjunctiva. After the excision of the trephined block of tissue, a mid-peripheral iridectomy was performed. Tenon's capsule and conjunctiva were closed with a single 8-0 Vicryl suture. Postoperatively the eyes were treated with an antibiotic for 5 consecutive days. No steroids were applied.
Clinical examination was performed in order to evaluate both the general appearance of the treated eyes and to measure the intraocular pressure (IOP) of experimental and control eyes. On designated days, rabbits were anaesthetised and their eyes examined. In order to measure the IOP, the eyes were anaesthetised with proxymetacaine (proparacaine). Tonometry was performed with an applanation tonometer (Tono-Pen XL, Mentor, USA) before experimentation and postoperatively. To exclude interindividual and cyclic variations IOP was compared between the manipulated experimental eye and the non-manipulated fellow eye. The measurements were performed as triplicates. The difference in measured IOP was expressed as right eye/left eye ratio. Preoperative IOP ratios (R/L) ranged from 0.85 to 1.2. Success was defined by >20% difference in IOP reflected as IOP ratio <0.8. A two tailed Student'st test was performed and significance assumed to exist if p <0.05.
On designated days eyes were enucleated and immediately fixed in 10% buffered formaldehyde for at least 24 hours. Consequently eyes were examined with a dissecting microscope (Zeiss) and a ring in the sagittal axis comprising the relevant area excised. Tissue samples were dehydrated and embedded in paraffin, 5 μm serial sections cut, rehydrated, stained with haematoxylin and eosin, and cover slipped.
Photographs were produced with a 35 mm camera system (Nikon). Micrographs of sections were taken with a 35 mm camera system attached to a Zeiss research microscope.
CLINICAL APPEARANCE AND INTRAOCULAR PRESSURE
The fate of the experimental and control eyes was evaluated clinically daily within the first week and twice a week thereafter.
Inflammatory response in surgical eyes such as conjunctival hyperaemia was mild in all groups and lasted for 1 week (Fig 3). There was no remarkable difference between injected (BCECF-AM) and non-injected eyes. No endophthalmitis was observed. Most of the eyes displayed a slight corneal haze near the surgical area which usually disappeared within 2 weeks. This occurred also independently of BCECF-AM injection or illumination and was probably due to surgical trauma.
Inflammatory response in control eyes having received subconjunctival injection of BCECF-AM without surgical intervention was absent or mild when blue light was applied. Light application did not cause clinically detectable tissue damage (Fig 4). One week postoperatively the conjunctiva still displayed a yellowish colour at the injection site (Fig 4B). This has been also observed in vitro and is probably due to the decay of the BCECF-AM probe.
Wound healing could be clinically observed as closure of the filtration area by ingrowing fibrotic tissue. This process was more prominent in the groups without photodynamic therapy (Fig 3). After a delay of 1 week it also progressed in the treated groups.
Intraocular pressure was measured as the mean of triplicates on designated days. Preoperative intraocular pressure ranged between 10–18 mm Hg. Preoperative IOP ratios (R/L) ranged from 0.85 to 1.2 and were drastically reduced in all groups within the second postoperative day displaying a mean value of 0.6 (Figs 5 and 6). In group A receiving intraoperative illumination without photosensitiser IOP ratios increased to preoperative values within 1 week. Group B treated with 40 μg BCECF-AM showed a slight delay, but no significant difference from group A. Group C receiving 80 μg BCECF-AM preoperatively retained a significant (p <0.02) IOP decrease within 3 weeks postoperatively but consequently acquired preoperative values (Fig 5). To exclude an antifibrotic effect by the probe itself in the second set of experiments (Fig 6) Animals in group D received a subconjunctival injection of photosensitiser (80 μg), but no intraoperative illumination. Similar to group A filtering surgery failure occurred within 1 week and IOP ratios increased to preoperative values. A reduced dose of BCECF-AM (70 μg) was still effective and prolonged the surgical effect in group E for 2 weeks. To examine if a higher concentration could have a better impact on filtering surgery survival, animals in group F received a subconjunctival injection 100 μg BCECF-AM. Though IOP was significantly (p <0.02) decreased until day 14, it increased thereafter and failure was assessed after the third week.
All rabbits in groups A–C were sacrificed on the 38th postoperative day and the treated eyes enucleated for histological examination. In the second set of experiments (groups D–F) treated eyes were enucleated on designated days. The tissues were stained with haematoxylin and eosin for light microscopy.
In the first set of experiments eyes were enucleated (38th postoperative day) when both control and experimental groups had already reached preoperative IOP ratios. Though fibrotic tissue ingrowth at the filtering area of group C appeared hypocellular compared with groups A and B, the sclerostomy site was closed in all groups and obviously responsible for IOP increase.
In the second set of experiments control eyes receiving subconjunctival injection of BCECF-AM without intraoperative illumination were enucleated on postoperative day 10, when surgery had failed and IOP was increased. Histological analysis of these tissues disclosed a sclerostomy site that was obstructed by hypercellular fibrotic tissue (Fig 7A). To evaluate the wound healing process in an experimental eye with functioning filtering area, a rabbit in group F was sacrificed on postoperative day 10, when the IOP ratio was 0.6. Histological profiles revealed a mild lamellar ingrowth of hypocellular fibrotic tissue within the filtering zone (Fig 7B). Though a complete open fistula could not be found, it is conceivable that this thin lamella might allow filtration of aqueous humour.
Photodynamic therapy is an alternative method for the treatment of localised pathologies such as skin cancers.43 Specific activation of pertinent drugs at the targeted area should avoid systemic side effects. Photodynamic therapy has also been evaluated for some distinct ophthalmological diseases such as ocular tumours,20 44 choroidal45 and corneal46 neovascularisation, proliferative vitreoretinal disorders,47 and postoperative fibrosis in glaucoma surgery.9 A recent study has investigated the feasibility of photodynamic therapy in a rabbit model of filtration surgery. Using ethyl etiopurin, a photosensitiser traditionally delivered by intravenous injection, Hill et al showed that subconjunctival delivery could have an impact on filtering bleb survival.38
The aim of this pilot study was to evaluate photodynamic therapy for antifibrosis using BCECF-AM. This carboxyfluorescein derivative is a cell membrane permeable compound rendered membrane impermeable and fluorescent upon cleavage by intracellular esterases.48-50 Exposure of cells that have incorporated BCECF-AM to light at the appropriate wavelength leads to cellular photoablation (Fig 1). The light induced cytotoxic ability of BCECF-AM has been shown before in in vitro studies.40 47
In the present pilot study we tested BCECF-AM in a rabbit model of filtration surgery. Because of the aggressive wound healing response in rabbits, this animal model is believed to be equivalent to high risk eyes in humans and surgical failure results within 1 week. The use of antimetabolites, such as 5-fluorouracil and mitomycin C, will prolong bleb survival. However, the effect is not permanent and depending on drug concentration it will subside within a few weeks.3 21
Our experiments show that filtration surgery success can be prolonged when cellular photoablation is applied. As demonstrated by groups A and D wound closure occurs and IOP will increase within 1 week when one of the components (photosensitiser, illumination with blue light) necessary for photoablation is missing. This time for surgical failure is similar to previous reports and demonstrates that non-activated photosensitiser or illumination has no detectable impact on wound closure.
In group B receiving 40 μg of BCECF-AM photoablative therapy seemed to be ineffective, since surgical failure occurred, such as in control groups A and D, within 10 days (Figs 5 and 6). Using BCECF-AM at a concentration of 70–100 μg filtration surgery success could be prolonged for about 3 weeks. This was expressed both in lowered IOP levels (Figs 5 and 6) and reduced fibrosis at the sclerostomy site (Figs 5 and 7).
Though some authors claim the creation of filtration blebs as evidence of successful surgery, in our hands blebs were not always associated with reduced IOP. This is probably due to encapsulation. Such a case is demonstrated in Figure 3A, which displays a rabbit eye from group D. Though a filtering bleb is still present on the 10th postoperative day, IOP had already risen to normal levels (Fig 6). Filtration blebs were, therefore, not evaluated as markers of succesful treatment.
BCECF-AM will be incorporated by all cells including proliferating and non-proliferating cells. Both clinical and light microscopic examination revealed no detectable damage in adjacent non-illuminated tissues. However, ultrastructure should be evaluated in further studies. BCECF-AM is like other drugs and may diffuse into other tissues. However, because illumination is achieved with diffuse light, which can activate the dye only at the surface of the tissue irradiated, it is not thought that decreased IOP results from a photoablative effect on other tissues—for example, the ciliary body. Though it might have been expected, subconjunctival injection of BCECF-AM (80 μg) followed by illumination did not cause a conjunctival defect. A possible explanation for this result might be an insufficient transmission of the diffuse blue light through the intact conjunctiva and/or the insufficient penetration of the probe into the superficial areas. In evaluating in vitro studies cellular photoablation could be only detected in cells that had incorporated BCECF-AM and were exposed to light at the appropriate wavelength (450–490 nm) irradiated by a microscope. Since daylight includes the exciting wavelength, we evaluated the effect of daylight on scleral fibroblasts that had incorporated BCECF-AM and could exclude a lethal effect (results not shown). This result might be because of the lesser light intensity than that from a microscope. Furthermore, in the animal model and clinical practice, the upper lid will protect the injected area.
A more intense light exposure will occur during surgery because of the operating microscope. In group D, however, this circumstance did not seem to have an effect on the wound healing process (Fig 6). In practice, a photoablative effect induced by the operating microscope would be beneficial when limited to the filtration area, but critical in adjacent tissues such as the cornea. Though the results are promising in this pilot study, further studies are necessary to optimise the effect by adjusting drug concentration and irradiation time.
Cellular photoablation might be an alternative to chemotherapeutic agents as a therapeutic approach to control postoperative fibrosis in glaucomatous eyes. Safety and reliability need to be studied in the future.
This study was supported by the Deutsche Forschungsgemeinschaft (HE 840/7-1); Retinovit-Foundation.
Proprietary interest: None.
This work was presented in part at the Association of Research in Vision and Ophthalmology Annual Meeting, May 1998, Ft Lauderdale, Florida, USA.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.