Topical fluoroquinolones may be used as antibacterial treatment and have been associated with corneal deposits [
A 59-year-old man opted to be submitted to a LASEK procedure in December 2008. His best-corrected visual acuity (BCVA) in the right eye (RE) was 20/25 with +3,75 sph = +0,50 cyl at 30° and in the left eye (LE) was 20/25 with +3.25 sph. His maculas showed some alteration in the retinal pigmented epithelium (RPE). The LASEK procedure was carried out bilaterally according to Dr. Camellin guidelines, applying Mytomicin C (MMC) 0.02% for two minutes at the end of the the laser-ablation [
At the end, lidofilcon contact lenses (Actifresh 400 – Hydron – Hamble, UK) were placed to facilitate epithelial adhesion.
Ciprofloxacin 0.3% four times/day (Oftacilox drops – Alcon Cusi – El Masnou, Spain) was started two days earlier, while the postsurgical treatment resulted in Ciprofloxacin 0.3% (as above), Diclofenac 0.1% four times/day (Diclofenaco Lepori monodosis – Angelini – Barcelona, Spain), Polividone 5% ad libitum (Oculotect monodosis – Novartis – Barcelona, Spain).
Debris from the deposit was analyzed with dark field microscopy and placed on a blood-agar plate seeded with a susceptible stain of
The follow-up during the three days after LASEK was regular, so the patient was appointed to remove contact lenses (CL) after three days.
At the scheduled visit, his CLs were not “in situ”, although he could not explain when or how he had lost them.
His LE showed a microcistic “caramel-like” opacity of subepithelial, paraoptic, inferior corneal stroma with epithelial discontinuities, without any sign of ocular or periocular inflammation (Figures
Shows the microcistic “caramel-like” opacity of subepithelial, paraoptic, inferior corneal stroma with epithelial discontinuities as appeared in the beginning (a, b) and after discontinuation of ciprofloxacin (c).
His RE showed only minor epithelial defects.
Administration of ciprofloxacin and diclofenac in his LE was stopped and substituted with tobramycin 0.3% (Tobrex – Alcon Cusi – El Masnou, Spain) and dexamethasone 0.1% (Dexafree monodosis – Thea – Barcelona, Spain) six times/day.
On the following day, RE too showed initial opacities of the inferior cornea and was moved to the same treatment of LE, which was gently scraped to analyze the debris as indicated in Section
After four days the gross opacity in LE disappeared, although some minor subepithelial deposits and punctiform desepithelisation still persisted.
RE showed no more peculiar findings.
Tobramycin was then stopped bilaterally and Dexamethasone was tapered in 2 weeks, switching to Fluorometolone 0.1% (FML drops – Allergan – Westport, Ireland) three times/day, tapered in 6 weeks bilaterally.
During the whole period the IOP did not show any hypertensive spike.
After one month, LE opacity was very similar to a vertex keratopathy (Figure
Deposits were confirmed to be ciprofloxacin by dark field microscopy.
The analysis of the images obtained with this technique revealed the precipitates to be polydisperse crystalline needles of 183
The excised precipitate resulted in a zone of inhibition that measured 22.0 mm on ATCC 29213 plate; the control disk measured 17.0 mm after 24 hours.
Ciprofloxacin, a second-generation fluoroquinolone, is extensively used in bacterial keratitis, owing to its ease of availability, broad spectrum of activity, and lack of toxicity. Although there are many reports of crystalline corneal deposits occurring with the use of topical ciprofloxacin 0.3% and norfloxacin 0.3% [
In “in vitro” tear model, drug concentration of ciprofloxacin determined a rapid precipitation that starts at
Scuderi et al. in 2003 [
Moreover gatifloxacin, a fourth-generation fluoroquinolone, has been reported to cause intrastromal crystalline deposits with a compromised corneal epithelium, in a similar manner to ciprofloxacin [
Although the use of ciprofloxacin alone seems to cause physicochemical changes in the tear film, combination therapy and other factors probably contribute to alter the pH thus causing or accelerating the precipitation process.
Such picture could be supported by our case, where LE was more affected than RE but further studies are needed to understand if CL loss, presurgical antibiotic treatment, LASEK technique itself, or the application of MMC could have affected the re-epithelialisation in this case.
Accumulation of large numbers of dead cells and an altered general morphology may be further factors influencing the formation of deposits.
With reference to the case here described, at the time of presentation to the corneal service, this patient had been treated with persistent, preserved and unpreserved topical drops.
Some of these topical medications (Oftacilox and Tobrex) contain benzalkonium chloride as preservative, which is known to disrupt cell walls by emulsifying membrane lipids, while decreasing epithelial microvilli, corneal wetting and inhibiting cell motility and surface healing [
It is probable that the pH balance had been altered owing to the interaction of multiple medications which, in the presence of a compromised ocular surface and of a nonhealing epithelial defect (due to the loss of the contact lenses), could result in ciprofloxacin precipitation and relative appearance of corneal deposits.
Discontinuation of ciprofloxacin and modification of topical therapy should aid the resolution of these deposits in most cases, as well as the promotion of epithelialisation, and the reduction of corneal and conjunctival toxicity.
If deposits persist and prevent epithelialisation, they should be debrided to permit the cornea to epithelialise, as in this case, without any adverse outcome, although some deposits persisted subepithelially despite prolonged topical cortisone therapy.
In any case, for the future it is hoped that more topical drugs are going to be available as unpreserved and/or with higher pH stability, in order to reduce at the lowest level any possible complication after local treatment, especially in refractive surgery where patients are very demanding for fast outcomes.