Fluoroquinolones First for Bacterial Keratitis
Fluoroquinolones are a good first empiric treatment for patients with bacterial keratitis, according to a recent systematic review and meta-analysis.
Marie-Sophie Hanet, MD, from the Department of Ophthalmology at the University of Louvain, Brussels, and the Scientific Support Unit, CHU Mont-Godinne, Yvoir, Belgium, and colleagues published their findings in the December issue of the Canadian Journal of Ophthalmology.
Fluoroquinolones are readily available and are well-tolerated by patients, the authors write.
"Based on these data, it seems reasonable to consider fluoroquinolones as the initial empiric treatment in most cases of suspected bacterial keratitis, and the use of fortified antibiotics being restricted to eyes unresponsive to initial treatment and to the cases for which a pathogen resistant to fluoroquinolones has been identified," the authors note.
The researchers selected 13 comparative studies for inclusion in the review: 8 prospective randomized trials and 5 nonrandomized studies. Two of the randomized trials were large multicenter trials, involving 28 clinical centers, and the other 6 involved 1 or 2 centers.
Of the randomized studies, 4 case series compared patients treated with fluoroquinolones with either historical cases and/or a group of nonrandomized patients who were treated simultaneously with a standard fortified antibiotic regimen; there was also a single retrospective medical record analysis.
Therapeutic success was defined as healing (complete re-epithelialization) that occurred while receiving the assigned treatment; treatment failure was defined as treatment that had to be changed because the patient's condition worsened or failed to improve.
Most studies used time to healing as an additional efficacy end point, but it was defined differently in the studies reported.
Nonsignificant differences between fluoroquinolones and standard treatment were found in meta-analyses of treatment efficacy, with a trend toward favorability of fluoroquinolone treatment. Using a random-effects model, odds ratios were 1.473 (95% confidence interval [CI], 0.902 - 2.405) for all randomized and nonrandomized studies, 2.374 (95% CI, 1.082 - 5.205) for nonrandomized studies, 1.050 (95% CI, 0.636 - 1.732) for randomized studies, and 1.199 (95% CI, 0.477 - 3.011) for randomized studies that only included patients with a microbiology-confirmed bacterial infection.
Using a fixed-effects model, odds ratios were 1.374 (95% CI, 0.996 - 1.894) for all randomized and nonrandomized studies, 2.192 (95% CI, 1.329 - 3.617) for nonrandomized studies, 0.957 (95% CI, 0.622 - 1.472) for randomized studies, and 1.092 (95% CI, 0.619 - 1.927) for randomized studies that only included patients with a microbiology-confirmed bacterial infection. The researchers found no statistically significant publication bias.
Well-Tolerated by Patients
The 4 largest randomized trials compared local tolerance to topical treatments. In 1 trial, patients reported significantly more severe burning and/or stinging after applying fortified antibiotics than those who used fluoroquinolones ( P < .001).
In another study, patients reported less discomfort after ciprofloxacin use than fortified antibiotics ( P = .012), and 1 study of ofloxacin found a proportion of patients who experienced drug toxicity 5 times greater with use of fortified antibiotics ( P < .001).
One study found signs of local intolerance in only 4 patients, all of whom were treated with fortified antibiotics.
In the 3 large ciprofloxacin treatment series, 13% to 17.6% of patients experienced a "transient white crystalline precipitate in the superficial portion of the corneal epithelial defect" that was determined by liquid chromatography to be ciprofloxacin. In another series, 1 patient among 15 treated with ciprofloxacin experienced a similar effect. This did not happen with the other fluoroquinolones.
A retrospective review of 140 clinical dossiers found 9 (16.7%) major complications (5 perforations and 4 enucleations or eviscerations) among the 54 patients who received fluoroquinolones compared with only 2 (2.4%) major complications (2 enucleations, no perforations) among the 84 patients who received fortified antibiotics.
This observation was confirmed in a larger group during a longer observation period. Patients were older at presentation, but the difference remained significant ( P = .02) after controlling for age, systemic disease, and immunosuppression. This greater perforation risk could be explained by fluoroquinolone-caused alterations in corneal tectonic strength, the authors write, but this effect was not observed in larger randomized studies.
Thomas Steinemann, MD, a professor of ophthalmology at MetroHealth Medical Center and Case Western Reserve University in Cleveland, Ohio, and a clinical correspondent for the American Academy of Ophthalmology, spoke with Medscape Medical News about the study.
"Both randomized and nonrandomized studies confirm that for empiric treatment of serious corneal infections, fluoroquinolones are a logical choice as empiric therapy," Dr. Steinemann said. He added that it is important to note that fluoroquinolones work well in the vast majority of cases but that there will always be some cases (more serious and opportunistic infections) that do not respond to them.
"For the community ophthalmologist and eye care practitioner, empiric therapy with fluoroquinolones probably works well in a community practicing setting in that most practitioners are not going to have access to culture methods in their office and it may not be logistically feasible to send the patient for expensive culturing and also to send the cultures...to a laboratory for testing," Dr. Steinemann explained.
"Fluoroquinolone therapy is very useful and very practical...it's reassuring to know that these commercially available preparations are effective," he noted.
The authors and Dr. Steinemann have disclosed no relevant financial relationships.
Can J Ophthalmol. 2012;47:493-499
沒有留言:
張貼留言