Statistics from Altmetric.com
Editor,—In recent years, the diode laser utilising 810 nm wavelength has emerged as an increasingly popular and effective tool for treating severe cases of glaucoma which are resistant to other conservative or surgical therapeutic options.1 2 The desired effect (as with other lasers used in this field) is thermal heating and coagulation necrosis of the ciliary epithelium (laser cyclophotocoagulation). However, the laser scleral transmission is increased by the contact method (compared with the non-contact method), allowing for less total energy application while obtaining the same desired effect.3
The side effects of trans-scleral cyclodiode laser range from common ones such as mild iritis4 to rare ones including phthisis bulbi.1 To the best of our knowledge, there has only been one reported case of scleral perforation due to trans-scleral diode cyclophotocoagulation. This was following contact delivery of the laser using the original quartz G-probe (diameter 600 μm) and settings of 2 W for 2 seconds per application.5 The patient in question had scleral thinning following previous cataract surgery. It was thought that the sharp edge of the probe had cut conjunctival vessels causing bleeding and contamination of the probe head. Thin adherent debris was then carbonised allowing the laser tip temperature to rise to 300°C, sufficient to cause scleral perforation. This case report led to the redesigning of the laser probe tip in order to protect the vascular structures from its sharp edges.
We would like to report a second case of scleral perforation following contact trans-scleral cyclodiode treatment. It concerns a 37 year old white man with buphthalmos whose left eye had been enucleated at the age of 13. His other ocular history included right trabeculectomy (age 4 years) and right giant retinal tear repair with vitrectomy, silicone oil, and 360° indirect laser (age 35 years). He had severe scleral thinning through 180° superiorly (see Fig 1), thought to have been caused by his buphthalmos and previous ocular surgery. Despite maximal medical therapy and removal of the silicone oil, his intraocular pressure was poorly controlled (>30 mm Hg). His best corrected visual acuity in the right eye had dropped from 6/36 in 1979 to 6/60 in 1998. This eye was treated with the currently marketed contact G-probe attachment of the Iris medical diode laser (Mountain View, CA, USA) using 14 applications of 2 seconds’ duration and 2 W power each. The eye was transilluminated before application of the laser but ciliary body identification was difficult due to the extensively thinned sclera. There were three audible “pops” from the first 13 applications, some of which were applied to the thinned scleral zones and following the last planned application at the superonasal limbus, a gush of aqueous was seen. Closer microscopic inspection confirmed a round, “punched out” full thickness perforation through conjunctiva, sclera, and choroid immediately posterior to the laser site. This defect was closed with two 10.0 Vicryl (Ethicon) sutures. Following the perforation, the patient developed a large choroidal haemorrhage (confirmed on ultrasound). A week after the laser, the scleral leak recurred and was successfully sealed using three 10.0 nylon (Alcon) sutures. Eight weeks later, his intraocular pressure was well controlled (10 mm Hg) using topical levobunolol and oral acetazolamide and the choroidal haemorrhage had resolved. The scleral wound has remained closed with no further leaks.
Trans-scleral “cyclodiode” photocoagulation has emerged as an effective method of controlling intraocular pressure and pain in refractory glaucoma. With its increasing use, more patients may be at risk of so far rare, but significant, complications. To the best of our knowledge, this report is the first case of scleral perforation using the new contact G-probe attachment.
Scleral perforation appears therefore to be a rare but significant complication of contact trans-scleral cyclodiode treatment. In both this and the previously reported case, pre-existing scleral thinning was the common risk factor. Parma et al 6 looked at the effect of cyclodiode therapy on cadaver eyes. They showed that approximately 40% less energy is needed to achieve ciliary photocoagulation in thin sclera (that is, half to a third full thickness) compared with normal thickness sclera.
Our patient was being treated as part of a standard protocol in use in our unit as part of a prospective trial of cyclodiode therapy. We had previously treated other eyes with thin sclera without incident at the energy levels utilised for the case described above although none has had such extensive thinning as this case. Similarly thinned areas in the same eye had been treated before the event which occurred on the final scheduled application. No conjunctival haemorrhages had been noted before the perforation and the probe tip had not been inspected between applications. It is difficult to know whether the perforation in our case was due to mechanical pressure, carbonisation of debris at the laser tip, or both. We would now warn against treating areas of an eye with severe scleral thinning as there are no known “correction factors” that can be utilised at present. If treatment is absolutely necessary, a lower laser power setting should be used (we would suggest 50%) and minimal pressure applied with the G-probe. Furthermore, care should be taken to ensure that the probe tip is clean before each application in such eyes in order to prevent carbonisation of debris.