Purchase this article with an account.
A. Gao, P.I. Murray, S. Rauz; A Clinical and Microbiological Review of 303 Consecutive Cases of Suspected Microbial Keratitis in the United Kingdom . Invest. Ophthalmol. Vis. Sci. 2006;47(13):3562.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Clinical outcome of microbial keratitis depends on prompt diagnosis and initiation of appropriate antibiotics. We wished to characterise the clinical and microbiological features of presumed microbial keratitis in our current practice.
A retrospective, non–comparative case note review of consecutive patients having corneal scrapes for suspected microbial keratitis at a single ophthalmic unit over a 29–month period. The incidence, predisposing factors, clinical and microbiological data for each patient episode was analysed.
A total of 288 patients (149 female, 139 male; median age 43 years, range 5–92) and 303 individual patient episodes were identified. Of these, 110 (36.3%) episodes required hospital admission with a median inpatient stay of 7 days (range 1–31). The commonest ocular predisposing factor was contact lens wear (101 cases – 34.2%) with rheumatoid arthritis (32 cases – 10.6%) and diabetes (28 cases – 9.2%) the most frequent systemic associations. Microbial isolates were identified in 114 (37.6%) cases with more isolates detected during the summer months. The median time from scrape to a culture–positive result was 4 days (range 1–144) with 81.6% of the results available within one week. Prior use of antibiotic did not affect the culture–positive rate. The most commonly isolated microbes were Pseudomonas spp. (20.2%), Streptococcus spp. (17.5%), Staphylococcus aureus (16.7%), and coagulase negative Staphylococcus (13.2%). All microbes were sensitive to at least one of a standard panel of antibiotics. The majority of episodes (203 cases – 67%) were commenced on conventional combination therapy (penicillin/cefuroxime and ofloxacin/gentamicin) whereas monotherapy with a fluoroquinolone was instituted in only 31 cases (10.2%). After culture data were available, the antibiotic regimen was altered in 34 episodes (29.8%). In spite of the low antibiotic resistance of the microbial isolates, evisceration was required in six cases. One patient went on to have a penetrating keratoplasty for a spontaneous perforation, and one developed bacterial endophthalmitis.
Our low culture–positive rate and prolonged time to culture result indicates that rapid, more sensitive, diagnostic tests are required for this potentially sight threatening condition.
This PDF is available to Subscribers Only