Population-based cohort study of microbial keratitis in Scotland: incidence and features

Abstract

Objective: To investigate the incidence and features of bacterial, fungal and protozoal keratitis in Scotland. Design: Prospective, population-based cohort study of all persons who developed culture proven microbial keratitis over an 8 month period. Setting: West of Scotland, UK. Subjects: Approximately 3 000 000 population. Main outcome measures: Incidence and risk factors for microbial keratitis. Methods: All patients were included who had presumed microbial keratitis from which bacteria, fungi or Acanthamoeba was isolated from the corneal scraping by the hospital laboratory using a standardised protocol. In addition, contact lens wearing patients with pathognomonic features of Acanthamoeba keratitis, who yielded a negative culture result when referred on chlorhexidine therapy, were included if Acanthamoeba could be cultured from their lens storage case. Results: The overall annual incidence of culture-proven microbial keratitis was 0.26 per 10 000 with a rate of 1.8 per 10 000 for contact lens wearers (all types, soft and rigid). Based on a previous pilot study of ‘presumed’ microbial keratitis in Glasgow, it was possible to estimate the incidence of expected ‘presumed’ microbial keratitis as 0.36 per 10 000 overall and 2.44 per 10 000 for contact lens wearers (all types). The incidence for Acanthamoeba keratitis was 1.49 per 10 000 soft contact lens wearers; this infection was not detected in the absence of contact lens wear nor with use of gas permeable or rigid contact lenses. Conclusions: ‘Presumed’ microbial keratitis from all causes, in the adult population, was approximately three times less common in the West of Scotland (0.36 per 10 000) than would be expected from a comparable retrospective study from Minnesota, USA for the years 1980–1988 (1.1 per 10 000). It was rare (approximately one case expected in 2 million per year) in the absence of pre-existing corneal disease, cosmetic contact lens wear or trauma. Ocular surface disease was the underlying cause predisposing to infection in 58% of cases, with an incidence of ‘presumed’ keratitis of 0.21 per 10 000 population; the highest incidence was found in the elderly population. Contact lens wear was responsible for 38% of cases, emphasising the importance of preventive hygiene and effective disinfection in this group. The estimated incidence of ‘presumed’ microbial keratitis in the West of Scotland associated with cosmetic wear (daily and extended use) of soft contact lenses was significantly less (P< 0.05) than that expected from a prospective study in New England, America in 1985 (266 per 10 000, rather than 8.05 per 10 000). However, the estimated incidence for presumed microbial keratitis for the West of Scotland asssociated with wearing soft contact lenses for cosmetic purposes in the daily wear modality (266 per 10 000) was less, but not significantly less, than that found in the prospective American study (4.20 per 10 000). The daily wear mode for contact lenses is almost universal in the West of Scotland, where extended wear has never been recommended. Extended wear has been shown in the USA to be associated with an incidence of presumed microbial keratitis between five and ten times higher than that associated with daily wear. This explains the lower incidence we have observed and a difference with the US study for overall infection rates but not when associated with daily wear alone. The incidence of proven Acanthamoeba keratitis found in the Scottish study among wearers of soft contact lenses for daily wear cosmetic purposes was exceptionally high at 1.49 per 10 000.