AIMS: To characterise and correlate crystalline precipitations implicated in non-healing corneal ulceration in two patients with a previous history of acanthamoeba keratitis. MATERIALS AND METHODS: Persistence of acanthamoeba and secondary bacterial infection was excluded with negative corneal scrapes. Confocal microscopy identified crystal-like deposits within the corneal stroma. To investigate possible precipitating combinations, all concurrent treatments at the time of presentation were mixed in wells, with observation of precipitate formation. Precipitates were observed with phase-contrast microscopy, and subsequently characterised via crystallography techniques and electrospray ionisation mass spectrometry. RESULTS: Combinations of dexamethasone 0.1% minims and chlorhexidine gluconate 0.2% formed an amorphous material characterised by electrospray ionisation mass spectrometry as an insoluble chlorhexidine salt. Combinations of chloramphenicol drops and timolol 0.5% formed a crystal identified via X-ray crystallography as santite (K(B₅O₆(OH)₄).(H₂O)₂). This is a borate mineral identified in nature, arising from thermal springs, but never reported in biological tissues. Clinical improvement was observed following the cessation of the implicated precipitating combinations. CONCLUSION: Our observations suggest iatrogenic precipitate formation, with a potential deleterious effect upon healing. The substrates for these precipitates include several frequently prescribed topical ophthalmic treatments. These findings shed new light on the aetiopathogenesis of non-healing corneal ulceration, and have broad implications on topical prescribing for this challenging condition.
AIMS: To characterise and correlate crystalline precipitations implicated in non-healing corneal ulceration in two patients with a previous history of acanthamoeba keratitis. MATERIALS AND METHODS: Persistence of acanthamoeba and secondary bacterial infection was excluded with negative corneal scrapes. Confocal microscopy identified crystal-like deposits within the corneal stroma. To investigate possible precipitating combinations, all concurrent treatments at the time of presentation were mixed in wells, with observation of precipitate formation. Precipitates were observed with phase-contrast microscopy, and subsequently characterised via crystallography techniques and electrospray ionisation mass spectrometry. RESULTS: Combinations of dexamethasone 0.1% minims and chlorhexidine gluconate 0.2% formed an amorphous material characterised by electrospray ionisation mass spectrometry as an insoluble chlorhexidine salt. Combinations of chloramphenicol drops and timolol 0.5% formed a crystal identified via X-ray crystallography as santite (K(B₅O₆(OH)₄).(H₂O)₂). This is a borate mineral identified in nature, arising from thermal springs, but never reported in biological tissues. Clinical improvement was observed following the cessation of the implicated precipitating combinations. CONCLUSION: Our observations suggest iatrogenic precipitate formation, with a potential deleterious effect upon healing. The substrates for these precipitates include several frequently prescribed topical ophthalmic treatments. These findings shed new light on the aetiopathogenesis of non-healing corneal ulceration, and have broad implications on topical prescribing for this challenging condition.
Authors: A Castillo; J M Benitez del Castillo; N Toledano; D Diaz-Valle; O Sayagues; J Garcia-Sanchez Journal: Cornea Date: 1997-07 Impact factor: 2.651
Authors: W Bernauer; M A Thiel; M Kurrer; A Heiligenhaus; K M Rentsch; A Schmitt; C Heinz; A Yanar Journal: Br J Ophthalmol Date: 2006-03 Impact factor: 4.638