Prognostic factors associated with the need for surgical treatments in acanthamoeba keratitis.

PURPOSE

The objective of this study was to assess the factors associated with anatomical and visual outcomes in patients presenting with Acanthamoeba keratitis (AK).

METHODS

This is a retrospective noncomparative interventional case series study comprising 44 eyes from 42 patients presenting with AK, treated with topical hexamidine diisethionate and topical polyhexamethylene biguanide, monitored between 2004 and 2008. AK was confirmed by polymerase chain reaction or direct microscopic examination. Correlation between clinical presentation and prognosis was assessed. Anatomical outcome was assessed according to the percentage of eyes requiring at least 1 surgical procedure in addition to topical treatment. Visual outcome was assessed by the best-corrected visual acuity at the end of follow-up.

RESULTS

Polymerase chain reaction results were positive for Acanthamoeba in 40 of the 44 eyes (91%) and in 16 of the 44 eyes (36%) by direct microscopic examination. Confocal microscopy suggested the presence of Acanthamoeba in 12 of 19 eyes (63%). Amniotic membrane transplantation was performed in 8 eyes, penetrating keratoplasty in 4 eyes, and evisceration in 2 eyes. The average follow-up time was 10 months. Surgical treatment was significantly associated (P < 0.05) with time from symptom onset to diagnosis of >30 days, an initial visual acuity of ≤20/200, an infiltrate size of >3 mm, preperforating infiltrates, and corneal neovascularization. The average final visual acuity was 20/48 in eyes that did not require surgical treatment (n = 34) and 20/1702 in eyes that required at least 1 surgical procedure (n = 10; P < 0.0001).

CONCLUSIONS

Late diagnosis, low initial visual acuity, corneal neovascularization, large infiltrates, and preperforated infiltrates were associated with surgical treatment in patients presenting with AK. Surgical intervention was associated with worse visual outcome.