Combination regimen may offer best topical keratitis treatment

August 1, 2004

New York-Results of an in vitro susceptibility study suggest a combination regimen consisting of clarithromycin and a fourth-generation fluoroquinolone may offer the best topical treatment for keratitis due to non-tuberculous Mycobacteria, said Mahendra K. Shah, MS.

New York-Results of an in vitro susceptibility study suggest a combination regimen consisting of clarithromycin and a fourth-generation fluoroquinolone may offer the best topical treatment for keratitis due to non-tuberculous Mycobacteria, said Mahendra K. Shah, MS.

Shah, director, microbiology laboratory, New York Eye and Ear Infirmary, New York, and colleagues from the department ofophthalmology determined the minimuminhibitory concentrations (MICs) for seven ocular antibiotics against 16 clinical non-tuberculous Mycobacteria isolates, including M fortuitum (n = 9), M chelonae (n = 4), M abscessus (n = 1), and M gastri (n = 2). Not all of the bacterial isolates were obtained from the eye. Of the nine M fortuitum samples, three were keratitis isolates and six were sputum isolates. The four M chelonae isolates consisted of two keratitis isolates, one vitreous isolate, and one sputum isolate; M abscessus was cultured from keratitis; and both M gastri isolates were from a wound. The antibiotics evaluated included five fluoroquinolones-moxifloxacin (Vigamox, Alcon), gatifloxacin (Zymar, Allergan), levofloxacin (Quixin, Santen), ciprofloxacin (Ciloxan, Alcon), ofloxacin (Ocuflox, Allergan)-along with the macrolide clarithromycin and the aminoglycoside amikacin.

Comparing MIC values among antimicrobial agents for each microbial isolate, the results showed one or both of the fourth-generation fluoroquinolones moxifloxacin and gatifloxacin had the lowest MIC values (range, 0.023 to 0.064 µg/ml) against eight of the nine M fortuitum isolates while clarithromycin consistently had the lowest MIC (0.23 µg/ml) for all four M chelonae isolates.

Amikacin exhibited poor performance against all of the bacteria, and the single M abscessus isolate was resistant to all the antibiotics tested (MIC >32 µg/ml)."The emergence of non-tuberculous Mycobacteria as a cause of subacute keratitis after LASIK has brought renewed attention to these organisms. There are now more than 15 case reports of non-tuberculous mycobacterial post-LASIK infections, and their treatment has been challenging in part because the pathogens are generally resistant to conventional antibiotics. Furthermore, clarithromycin and amikacin, which have been used as the antibiotics of choice, both need to be formulated extemporaneously by a pharmacist, are unstable in solution, and have variable corneal penetration," said Shah.

The fourth-generation fluoroquinolones overcome some of those drawbacks because they are highly soluble, have excellent corneal penetration, and, as this study demonstrates, they have favorable activity against some non-tuberculous mycobacterial species, he explained.

"While in vitro susceptibility data cannot always predict clinical efficacy in biological systems, considering the pharmacokinetics of the fourth-generation fluoroquinolones and their potent activity, moxifloxacin and gatifloxacin would seem to offer a clinical advantage for use in a combination regimen for treating intralamellar corneal infections," he said.

Quantitative MIC values for the various antimicrobial agents were determined using the E-test antimicrobial gradient strips and compared with National Committee of Clinical Laboratory Standards, which, as Shah noted, are used to predict in vivo efficacy based on the safe, achievable concentrations in serum, not tissue.

Study co-author David Ritterband, MD, a cornea and external disease specialist at the New York Eye and Ear Infirmary, acknowledged questions have been raised regarding use of the E-test for evaluating antimicrobial susceptibility of fastidious organisms like non-tuberculous Mycobacteria. Nevertheless, he noted results from several studies evaluating pulmonary isolates show it has promise.

"Use of the E-test appears to be the best initial method to test these ophthalmic pathogens considering that broth microdilutions for the fourth-generation fluoroquinolones are not yet commercially available," said Dr. Ritterband.

He suggested that further study of the fourth-generation fluoroquinolones in treating non-tuberculous mycobacterial keratitis might proceed with a rabbit keratitis model, such as has been used for Staphylococcus or Pseudomonas infections. The method would involve creation of a LASIK flap with inoculation of a known quantity of microbes into the interface and initiation of fluoroquinolone treatment after keratitis develops. OT