Statistics from Altmetric.com
Editor,—We describe two cases of recurrent corneal epithelial breakdown following indirect diode laser panretinal photocoagulation (PRP) in diabetic patients. The particular nature of diode laser burns may have contributed to this event.
A 21 year old insulin dependent diabetic woman underwent bilateral indirect diode laser PRP under general anaesthesia for proliferative diabetic retinopathy (PDR). She presented 4 weeks later with a 3 week history of a painful left eye. Visual acuity was 6/12 and a 3 mm × 2 mm infiltrative corneal ulcer was noted. There was total corneal anaesthesia bilaterally and both pupils were mid dilated and non-reactive to light and accommodation. Microbiology was negative and the ulcer gradually healed with intensive topical antibiotic treatment. Over the next 3 months corneal anaesthesia persisted in the left eye, but resolved in the right. She had two further episodes of corneal epithelial breakdown on the left which were successfully managed with topical lubricants. One year after laser treatment visual acuity was 6/6 in the right eye and 6/12 in the left, with residual stromal scarring on the left (Fig 1). Both pupils remained mid dilated and non-reactive.
A 26 year old insulin dependent diabetic woman underwent indirect diode PRP to the left eye under general anaesthesia for PDR. One week later she presented with a dilated and non-reactive left pupil, complained of not being able to focus, and had an anaesthetic cornea. Subsequently she underwent two further treatment sessions to both eyes for persistent PDR. She then developed a large epithelial defect of the left cornea, total corneal anaesthesia bilaterally, and semidilated pupils non-reactive to light and accommodation. The epithelial defect was managed with topical antibiotics and lubricants with slow resolution. She went on to have three further episodes of epithelial breakdown on the left and two on the right. Topical lubrication and padding proved to be the most effective treatment. Eighteen months after initial laser treatment, corneal sensation had recovered, visual acuity was 6/18 in both eyes, and there was mild central stromal scarring bilaterally.
Corneal sensation and innervation of the pupil and accommodation is supplied by the long and short posterior ciliary nerves which pierce the eye posteriorly and run forward in the suprachoroidal space. Disruption of this anatomical pathway could therefore lead to the problems experienced by the patients described above. Indeed, pupillary abnormalities and a temporary reduction in corneal sensitivity have been noted following argon, xenon, and krypton laser PRP.1 2 However, to our knowledge, such profound corneal abnormalities with ulceration and scarring have not been described previously.
The histopathological damage inflicted by laser burns has been studied in both the animal and human eye.3-5 Clinically burns can be classified as follows: grade 1 consists of a greyish retinal discoloration only; grade 2 shows whitish discoloration surrounded by a greyish periphery; and finally grade 3 burns have a distinct white centre, representing the highest intensity burn.
High intensity laser burns have been shown to damage choroidal nerves in humans and animals.2-6 The thermal damage profile of a retinal burn caused by the diode laser (wavelength 810 nm) extends deeper into the choroid than that from an argon blue green laser (488–514.5 nm).3 5 This is due to the lower absorption within melanin of the longer diode laser wavelength. Hence there is potentially a greater risk of choroidal nerve damage from the diode laser, particularly with an excessively intense burn. The burn size associated with indirect laser tends to be larger than that achieved with direct laser, particularly if the working distance increases during PRP, thus causing thermal damage to a greater area of retina. Indeed, the post PRP fundus photographs of both patients show large burns tending towards confluence (Fig 2).
Diabetics with and without retinopathy have significantly increased epithelial fragility when compared with normal controls.7 Diabetics also have reduced corneal sensitivity when compared with normal controls.8 However, before laser treatment, neither of our two patients had ever had any symptoms or signs of corneal disease. All the corneal problems started post laser and were associated with a total lack of corneal sensation. As the corneal sensation recovered over the months the frequency of the epithelial erosions diminished.
The optics of the diode laser beam mean that it is possible to clip the edges of the iris in the laser beam during treatment.9 10It is possible that this could account for some of the pupillary abnormalities post PRP, but taking into account the corneal anaesthesia and reduced accommodation, it would seem more likely to be a result of damage to a common denominator.
We hypothesise that the large area involved in performing an indirect PRP, and the depth of the burn achieved with diode laser, damaged a large number of choroidal nerves causing the pupillary and corneal abnormalities.
We propose the following measures to avoid choroidal nerve damage when performing indirect diode laser PRP. Grade 1 to 2 burns should be discretely placed so as to prevent confluence certainly avoiding high intensity grade 3 burns. Under peribulbar and general anaesthesia, it is particularly important to use less intense burns as the painful feedback from choroidal nerve damage is absent. Care should be taken with the working distance of the laser indirect to avoid unduly large burns. Particular caution should be exercised when treating in the horizontal retinal meridia, the typical location of the long ciliary nerves. Finally, the patient should be warned of possible effects on the anterior segment which may occur after indirect diode laser PRP.