Sir,

We report snailtrack-like corneal changes occurring after subconjunctival injection of 5-flurouracil (5-FU) around a trabeculectomy bleb.

Case report

A 50-year-old male with advanced glaucoma underwent an uncomplicated right-sided trabeculectomy, with mitomycin C augmentation. Two months postoperatively, the right visual acuity (RVA) was 6/6, but the intraocular pressure (IOP) had risen to preoperative levels. Bleb needling was performed, with subconjunctival injection of 5-FU at 2 o’clock and 10 o’clock to the bleb. One week later, RVA was 6/7.5. New snailtrack-like corneal changes were noted in this eye (Figure 1), whereas the left cornea remained entirely normal. Six months after bleb modulation, the patient’s IOP has improved but the unilateral corneal changes persist.

Figure 1
figure 1

Anterior segment photograph of the right eye, showing snailtrack-like corneal changes.

Full size image

Comment

Corneal ‘snailtracks’ (white-grey streaks at the level of the corneal endothelium), may act as markers of endothelial cell damage.1 Subconjunctival injection of 5-FU in proximity to the bleb following glaucoma filtration surgery is commonly employed to sustain good IOP control postoperatively. However, 5-FU has toxic effects on the corneal endothelium in animal studies.2 Uncomplicated subconjunctival injections of 5-FU are unlikely to harm the endothelium, as drug concentrations in the anterior chamber after injection remain low.3 However, case reports describe potential endothelial damage following inadvertent passage of 5-FU into the anterior chamber.4 Although corneal oedema is a recognised manifestation of such toxicity, the appearance of snailtrack-like corneal changes has not previously been reported. We hypothesise that snailtracks seen clinically would appear as dark, excavated lines on the corneal endothelium on specular microscopy, corresponding to linear ruptures of endothelial cells as seen in certain corneal dystrophies. Compromised endothelial cell function may lead to increased propensity to develop corneal oedema on exposure to provoking factors. Our finding leads us to support the view of Khaw et al,5 who in a letter to Eye some years ago, advised caution when injecting 5-FU directly into the bleb. The authors raised concerns about the potential for 5-FU to enter the anterior chamber via a patent sclerostomy. We would also caution against injecting directly above the bleb, where the effects of gravity may encourage downward flow of 5-FU directly into the anterior chamber, and would advise injecting lateral to the bleb instead. Our patient provides a good example of the potential consequences of unintentional intraocular exposure to 5-FU, and reminds clinicians of the possible complications of this technique of bleb revision.