Abstract
Purpose: :
To identify factors influencing the likelihood of isolating causal organisms in cases of infectious keratitis.
Methods: :
The Saint Louis University (SLU) Hospital microbiology records for all specimens submitted from the ophthalmology department at SLU were identified for the period 1999 to 2003, as well as records identified by ICD–9 code. The corresponding patient records were then reviewed to confirm the nature of the specimens submitted. Non–viral, infectious keratitis cases were then studied to determine the ulcer size, history of antibiotic treatment prior to culture, and time from onset of symptoms to presentation to the department of ophthalmology at SLU. The influence of each of these factors on the growth rate of the specimens was examined.
Results: :
352 specimens were received at the microbiology laboratory from the department of ophthalmology from Jan. 1, 1999 to Dec. 31, 2003. Of these specimens, 92 (27%), 83 from Saint Louis University Hospital and 9 from outside laboratories, were from patients with the clinical diagnosis of non–viral, infectious corneal ulcers. The growth rate was 72% (66 specimens); however, 28 specimens were considered contaminants, resulting in a positive, non–contaminant growth rate of 41% (95% CI 32 – 52%). The positive, non–contaminant growth rate for large ulcers (>2 mm) was 76% (N=29, 95% CI 58 – 88%), compared to that of small ulcers (≤2 mm) which was 24.2% (N=33, 95% CI 13 – 41%). Ulcer size was associated with growth rate (Yates chi square=14.47, p–value=0.0001). Cases with antibiotic treatment prior to culture had a positive growth rate of 30% (N=43, 95% CI 19 – 45%) compared to cases without treatment prior to culture with a growth rate of 50% (N=46, 95% CI 36 – 64%). Prior antibiotic treatment was not associated with positive growth rate (Yates chi square=2.83, p–value=0.093). The mean time to presentation was not different between cases in which a causal organism was identified (8.5 days, 95% CI 3.8 to 13.2 days) and those in which there was no causal organism identified (9.1 days, 95% CI 4.94 to 13.3 days, p=0.853).
Conclusions: :
Treatment of infectious keratitis is often guided by the microbiological identification of causal organisms. Large ulcer size was associated with a greatly increased growth rate compared to small ulcers. A history of pre–treatment did not show a statistically significantly decrease in growth rate, and time to presentation showed no relationship to growth rate. Obtaining a specimen is most likely to identify a causal organism in cases with large ulcers regardless of prior treatment with antibiotics and time to presentation.
Keywords: keratitis • cornea: clinical science • clinical laboratory testing