LABORATORY SCIENCES. 193-nm Excimer Laser Induced Fluorescence Detection of Fluoroquinolones in Rabbit Corneas

Transcription

1 LABORATORY SCIENCES 193- Excimer Laser Induced Fluorescence Detection of Fluoroquinolones in Rabbit Corneas Roy S. Chuck, MD, PhD; Ramez E. N. Shehada, PhD; Mehran Taban, MD; Tulaya Tungsiripat, MD; Paula M. Sweet, MT; Hebah N. Mansour, MD; Warren S. Grundfest, MD; Peter J. McDonnell, MD Objective: To measure the 193- excimer laser induced fluorescence of fluoroquinolone-treated cadaver rabbit corneas. Methods: Prior to ablation with a commercially available ophthalmic excimer laser (Nidek EC-5; Nidek Technologies, Pasadena, Calif), 35 cadaver rabbit corneas were treated with topical sterile balanced salt solution,.3% tobramycin sulfate, or the fluoroquinolones.3% ofloxacin,.5% levofloxacin,.3% ciprofloxacin hydrochloride, or.3% gatifloxacin. The fluorescence generated from each ablated corneal layer was measured and used to identify the presence of antibiotic. This was achieved by training a partial least-squares model to discriminate between the fluorescence spectra of antibiotic-treated and antibiotic-free (healthy) cornea. Antibiotic concentrations down to.6 µg/ml were detected with high accuracy. Assuming a constant ablation rate of.3 per laser pulse, the number of corneal layers ablated to reach antibiotic-free cornea is used to calculate the penetration depth of the antibiotic. Results: The mean ± SD penetration to a detectable depth was as follows:.3% ofloxacin, 7.1±3. ;.5% levofloxacin, 6.7±1.4 ;.3% ciprofloxacin, 1.2±.6 ; and.3% gatifloxacin, 7.±1.9. The penetration depth of.3% tobramycin could not be determined because its fluorescence spectrum overlapped with that of the native cornea. Conclusions: Topical administration of fluoroquinolonecontaining solutions results in measurable differences in laser-induced corneal fluorescence. Under these experimental conditions,.3% ofloxacin,.5% levofloxacin, and.3% gatifloxacin all appear to penetrate the epithelium significantly more than.3% ciprofloxacin (P.2). Clinical Relevance: Monitoring of laser-induced fluorescence may be helpful in determining the penetration depths and concentrations of topically applied fluoroquinolones within the cornea. Arch Ophthalmol. 24;122: Author Affiliations: Departments of Ophthalmology (Drs Chuck, Taban, Tungsiripat, and McDonnell and Ms Sweet) and Biomedical Engineering (Dr Chuck), University of California, Irvine; the Department of Biomedical Engineering and University of Southern California, Los Angeles (Dr Shehada); Medical Technology Laboratories, La Mirada, Calif (Dr Mansour); and Departments of Surgery, Electrical Engineering, and Bioengineering, University of California, Los Angeles (Dr Grundfest). Drs Chuck and McDonnell are now with the Wilmer Ophthalmological Institute, The Johns Hopkins University, Baltimore, Md. Financial Disclosure: None. WITH THE INCREASING popularity of such techniques as laser in situ keratomileusis (LASIK) and photorefractive keratectomy, excimer laser refractive surgery has emerged as one of the most common procedures performed in the United States. In 2 alone, more than 1.5 million of these procedures were performed in the United States, 1 and they are predicted to surpass surgical procedures for cataract as the most common eye operations conducted within the next several years. Over the years, numerous aspects of excimer laser technology have been scrutinized, including safety and the possible health adverse effects of laser application. One of the earliest concerns regarding the use of photorefractive keratectomy was initially raised by Loree et al 2 about the potential damage to surrounding ocular tissues. This group examined laser-induced fluorescence produced by the ablation of porcine corneas at 193- (argonfluoride), 248- (krypton-fluoride), and 38- (xenon-chloride) wavelengths using an intensified diode array detector system. They reported a potentially increased risk of secondary cataract formation from the low-energy UV light fluorescence produced during ablation of the cornea. Muller-Stolzenburg et al 3 also explored this issue in bovine corneas by placing a quartz fiber into the anterior chamber of the treated eye; they also expressed a similar concern. However, 2 subsequent and more complete analyses by Ediger 4 and Tuft et al 5 using photodiode systems independently showed a total UV energy dispersion that suggested a low risk of cataract from the laser-induced fluorescence. Unlike the slower photodiode systems, using an intensified charge-coupled detector system, we monitored real-time, 193- laser-induced fluorescence spec- (REPRINTED) ARCH OPHTHALMOL / VOL 122, NOV Downloaded From: on 11/4/ American Medical Association. All rights reserved.

2 Off-Axis Parabolic Mirror Mirror Holder Laser Aperture Corneal Fluorescence Metal Holder XYZ Translator Figure 1. Experimental setup. Laser Beam Spectrograph Optical Table Computer XYZ Translator tra during transition from epithelium to stroma at clinically relevant laser fluences. We were able to corroborate these later studies by reporting qualitatively similar laserinduced fluorescence spectra. 6 In related work, using a pulsed 193- excimer laser at ablative energies on cadaver human and rabbit corneas, we determined the UV and visible fluorescence spectra as ablation progressed through the entirety of the cornea 7,8 with the ultimate goal of developing a fluorescence-guided spectrometer to provide an objective means to control excimer laser ablation. To our knowledge, however, a key question that has not been addressed to date is whether the topical application of fluorescent medications to the cornea might alter the outcomes of these studies. Use of such agents prior to excimer laser ablation might alter the pattern and intensity of fluorescence and related potential phototoxic reactions. Among the most common ophthalmic drops used preoperatively and postoperatively are fluoroquinolone antibiotic solutions. 9 In fact, over the past decade, fluoroquinolones have become the antibiotic class of choice for most topical ophthalmic antibiotic indications. The fluoroquinolones, derivatives of nalidixic acid, 1 are a fluorescent class of drug, and, thus, could potentially be detected by fluorescence spectroscopy. 11 Another issue in photorefractive surgery that has been examined throughout the past dozen years is the risk of postoperative infection. Although a rare phenomenon (.25%-1.2%), 12,13 infectious keratitis is a feared complication that is increasingly associated with unusual pathogens and pathogens resistant to the commonly used antiinfective agents. 9,14-2 The emergence of resistant strains of Staphylococcus aureus, the most common cause of infectious keratitis, to the familiar fluoroquinolones has not helped alleviate these concerns. Although many techniques have been used to assess the efficacy of antiinfective agents, drawbacks exist. In vitro tests fail to consider many factors such as drug penetrance into corneal tissue and the intrinsic anatomy of the eye. There have been numerous reports that have measured the antibiotic level in aqueous humor subsequent to aspiration Three groups have determined the intracorneal concentration of drug only after excision of corneal samples from patients undergoing penetrating keratoplasty Donnenfeld et al 28 enzymatically digested corneal samples. In articles by McDermott et al 29 and Diamond et al, 3 the specimen corneas were mechanically pulverized and analyzed spectroscopically after highperformance liquid chromatography for antibiotic concentration. To our knowledge, there have been no reported studies assessing the depth of penetration of fluoroquinolone antibiotics in corneal tissue as determined by real-time detection of excimer laser induced fluorescence during corneal ablation. This information is critical because it is the intracorneal level of antibiotic therapy that is thought to be the most important for optimal effectiveness against unforgiving pathogens involved in infectious keratitis. The purpose of this investigation was to measure the 193- excimer laser induced fluorescence spectrum of fluoroquinolone-treated cadaver rabbit cornea during excimer laser ablation. We sought to assess whether fluorescence detection of fluoroquinolone antibiotics in the cornea was possible on a shot-to-shot depth-dependent basis. METHODS EXCIMER LASER CORNEAL FLUORESCENCE ABLATIVE SPECTROSCOPY A schematic diagram of the experimental spectroscopy setup is shown in Figure 1. The main component is the ArF excimer laser corneal surgery system (model EC-5; Nidek Technologies Inc, Pasadena, Calif) operating at 193. The laser was operated in manual mode at 4 Hz, a spot size of mm, and an average energy output of 16.4 mj/pulse. The laser-induced fluorescence generated from the ablated corneal layers is reflected by a 2-in off-axis parabolic mirror (AlSiO-coated, model A ; Janos Technology Inc, Townsend, Vt) and focused in the entrance slit of the spectrograph (model SD2; Ocean Optics, Dunedin, Fla). The long-pass filter commonly used to eliminate the excitation radiation in laser-induced fluorescence spectroscopy was not needed because the mirror s coating did not reflect the 193- radiation scattered from the surface of the cornea. The spectrometer was set to automatically measure and save the fluorescence spectrum. Because of the extremely high absorption of the 193 in the corneal tissue, the fluorescence signal is known to arise from the ablated corneal layer. 4,5 Assuming a linear rate of removal of tissue with each laser pulse, a constant ablation rate of.3 /pulse was experimentally determined. EXPERIMENTAL PROTOCOL Fresh New Zealand white rabbit heads were obtained from a local slaughterhouse and kept at 4 C until the eyes were enucleated within 7 hours of death to maximize preservation of the corneal epithelium. The globes were stored in a moist chamber at 4 C for approximately 4 hours until time of use. A total of 35 globes were used in this study. An aminoglycoside (tobramycin sulfate) and 4 different fluoroquinolones (ofloxacin, levofloxacin, ciprofloxacin hydrochloride, and gatifloxacin) were tested, as these are some of the most common agents used for surgical prophylaxis. The applied solution concentrations were as follows:.3% tobramycin (Tobrex; Alcon Laboratories Inc, Fort Worth, Tex),.3% ofloxacin (Ocuflox; Allergan Inc, Irvine, Calif),.5% levofloxacin (Quixin; Santen USA Inc, Napa, Calif),.3% ciprofloxacin (REPRINTED) ARCH OPHTHALMOL / VOL 122, NOV Downloaded From: on 11/4/ American Medical Association. All rights reserved.

3 (Ciloxan; Alcon Laboratories Inc), and.3% gatifloxacin (Zymar; Allergan Inc). Sterile balanced salt solution (BSS; Alcon Laboratories Inc) was used as a negative control. Four drops of BSS (n=5), tobramycin sulfate (n=5), ofloxacin (n=5), levofloxacin (n=3), ciprofloxacin hydrochloride (n=3), or gatifloxacin (n=4) were applied topically to separate corneas and allowed to rest for 1 minutes. Three drops of the appropriate solution were then administered a second time and globes were allowed to rest for an additional 5 minutes. To wash away excess antibiotic, especially at the corneal surface, each globe was rinsed with approximately 2 ml of lactated Ringer solution (Baxter Healthcare Corp, Deerfield, Ill) subsequent to antibiotic administration and prior to laser ablation. After washing, the globes were mounted on the laser platform and the corneas were ablated to perforation and the associated fluorescence spectra were measured. In a separate preparation, each of the above antibiotics was diluted to.6 µg/ml and topically applied to the corneas of another set of globes (n=2 for each antibiotic, 1 total) at room temperature and allowed to rest for 5 minutes. Three drops of the appropriate solution were then administered a second time and globes were allowed to rest for an additional 3 seconds. Accordingly, the anterior ( 1 ) of the antibiotic-treated cornea would have an antibiotic concentration equal to the known concentration of the topically applied antibiotic solution of.6 µg/ml. The globes were immediately mounted on the laser platform and the corneas were ablated for about 5 seconds; the resulting fluorescence spectra were measured. Assuming an ablation rate of.3 /pulse, the first 2 fluorescence spectra generated by the first 2 ablative pulses would be arising from the anterior.6- corneal layer and have the characteristic spectral features of corneal tissue saturated with.6 µg/ml of the specific antibiotic. PURE ANTIBIOTIC LASER-INDUCED FLUORESCENCE SPECTROSCOPY An aliquot (ie, 5 drops) of each antibiotic solution and the BSS was placed in an aluminum container at the focal point of the laser. The solution was irradiated with the excimer laser and the generated autofluorescence was measured. DETECTION OF ANTIBIOTIC PENETRATION The measured fluorescence spectra (25-65 ) were corrected for the dark current of the detector and the background light. Each spectrum was smoothed using a 5-point moving average window and resampled every 5 to reduce the number of spectral points from 379 to 76. DEVELOPMENT AND TRAINING OF THE MODEL For each antibiotic tested, a partial least-squares (PLS) model was developed for the discrimination between healthy and antibiotic-treated corneal tissue. The model is basically a numerical matrix that when multiplied by the fluorescence spectrum of an ablated corneal layer would yield a numerical value that is indicative of the antibiotic concentration in that layer. Detailed description of the PLS technique is available elsewhere The training inputs of the model were the first 2 fluorescence spectra acquired from the anterior 6 of healthy cornea (antibiotic free), and the first 2 fluorescence spectra acquired from the anterior.6 of the corneas treated with the diluted antibiotic (ie,.6 µg/ml). The training outputs of the model were 1 for the antibiotic-free cornea and 1 for antibiotic present with a concentration of.6 µg/ml and above. The PLS modeling algorithm is implemented using the scientific programming language MATLAB (version 5.3; The MathWorks Inc, Natick, Mass). The model s accuracy in detecting the antibiotics in the cornea was determined using the method of cross validation. In the latter, one input-output pair is excluded from the input-output data set used in training the model. The trained model is then used to predict the output (ie, antibiotic present/ absent) for the excluded input (ie, spectrum) and the result is compared with the actual output. This process is repeated for each input-output pair in the training data set and the model s prospective accuracy is calculated by: % Accuracy=NumberofCorrectlyPredictedOutput/ NumberofInput-OutputPairsintheDataSet Because the tested input-output pair was not used in training the model, this prospective accuracy is unbiased and represents the accuracy with which the model would detect the antibiotic s presence or absence in the cornea. A separate model is developed for each antibiotic to be tested for corneal penetration. DETERMINATION OF THE PENETRATION DEPTH Starting with the fluorescence spectrum produced by the ablation of the topmost corneal layer, the model processes each subsequent spectrum to estimate the presence or absence of antibiotic in the corresponding corneal layer. As described earlier, the model is trained to detect antibiotics with concentrations of.6 µg/ml or greater. The sequential number of the first spectrum indicating antibiotic-free cornea is multiplied by the assumed ablation rate (ie,.3 /pulse or spectrum) to calculate the maximum penetration depth of the antibiotic. RESULTS FLUORESCENCE SPECTRA OF PURE ANTIBIOTICS The fluorescence spectra of.3% ofloxacin,.5% levofloxacin,.3% gatifloxacin, and.3% ciprofloxacin are shown in Figure 2. Each antibiotic exhibited a characteristic fluorescence spectrum with peaks between 4 and 5. Levofloxacin produced the highest fluorescence intensity followed by ciprofloxacin, gatifloxacin, and ofloxacin, respectively. The spectral peak of ofloxacin, levofloxacin, and gatifloxacin occurred at 483, while ciprofloxacin exhibited double peaks at 433 and 483. The fluorescence spectrum of.3% tobramycin is shown in Figure 2E. In contrast to the other 4 antibiotics, tobramycin exhibited its peak fluorescence at about 3. The fluorescence spectrum of tobramycin coincided with the autofluorescence spectrum of the corneal structural proteins (ie, elastin and collagen), which prevented its spectroscopic detection in the corneal tissue. FLUORESCENCE SPECTRA OF ANTIBIOTIC-TREATED CORNEAS The fluorescence spectra of corneal ablation are plotted vs ablation depth in the 3-dimensional plots shown in Figure 3. The corneas treated with.3% tobramycin exhibited a single-peak fluorescence spectrum that is similar to that of healthy cornea. On the other hand, the corneas treated with.3% ofloxacin,.5% levofloxacin,.3% gatifloxacin, and.3% ciprofloxacin exhibited a charac- (REPRINTED) ARCH OPHTHALMOL / VOL 122, NOV Downloaded From: on 11/4/ American Medical Association. All rights reserved.

4 A Ofloxacin B Levofloxacin C Gatifloxacin D Ciprofloxacin Hydrochloride E Tobramycin Sulfate Figure 2. The fluorescence spectra of 4 fluoroquinolones and tobramycin sulfate to test penetration depth in rabbit cornea. The following strengths were used:.3% ofloxacin (A),.5% levofloxacin (B),.3% gatifloxacin (C),.3% ciprofloxacin hydrochloride (D), and.3% tobramycin sulfate (E). Notice how the peak fluorescence of tobramycin is close to 3, in contrast with the rest of the fluoroquinolones tested. AU indicates absorbance units. teristic double-peak fluorescence spectrum. The first peak at about 3 coincides with the natural fluorescence of the structural proteins of the corneal tissue while the second peak at about 483 coincides with the characteristic fluorescence of the antibiotic. The fluorescence spectra of the diluted antibiotics in BSS are shown in Figure 4. The results indicate that even at this low concentration (.6 µg/ml), each antibiotic generates more than 4 times the fluorescence produced by BSS at the antibiotic s maximum emission wavelength. ANTIBIOTIC DETECTION IN THE CORNEA USING PLS MODELING The PLS model was able to detect the presence of antibiotics in concentrations of.6 µg/ml and above with a prospective accuracy of 9%. The maximum penetration depth of the antibiotic into the corneal tissue is given in the Table. The mean ± SD penetration to a detectable depth was as follows:.3% ofloxacin, 7.1±3. ;.5% levofloxacin, 6.7±1.4 ;.3% ciprofloxacin, 1.2±.6 ; and.3% gatifloxacin, 7.±1.9. The prospective accuracy of the above measurements is about 9% with a detection sensitivity of.6 µg/ml. Under these experimental conditions, none of the fluoroquinolone antibiotics tested appear to have diffused significantly past the intact epithelium. Ofloxacin, levofloxacin, and gatifloxacin all seem to penetrate the epithelium significantly more than ciprofloxacin (P.2). Tobramycin absorption could not be detected because of the model s inability to discriminate between its fluorescence spectrum and that of the cornea due to their close resemblance. COMMENT We have determined that applied fluoroquinolone antibiotic preoperatively prior to laser refractive surgery can certainly alter the excimer laser induced fluorescence patterns observed during surgery. Most of the doped fluorescence occurs between 4 to 5. It is not clear that observed intensities at these wavelengths in vitro translate into toxic irradiation in the clinical setting. Prospective human studies using clinical drug concentrations will need to be performed to answer this question. In this pilot study, we observed a detection sensitivity of better than.6 µg/ml. In future experiments, we will need to determine the lower limits of detection in titration studies. However, even in these preliminary studies, the lowest concentration of fluoroquinolone antibiotics tested,.6 µg/ml, is well below the minimum inhibitory concentrations of many ocular bacterial isolates. 34 Anticipated improvements in our fluorescence collection and detection systems should result in even better detection limits than our current arrangement. Thus, we (REPRINTED) ARCH OPHTHALMOL / VOL 122, NOV Downloaded From: on 11/4/ American Medical Association. All rights reserved.

5 A Healthy Cornea B.3% Tobramycin Sulfate C.3% Ofloxacin Ablation Depth, Ablation Depth, Ablation Depth, D.5% Levofloxacin E.3% Gatifloxacin F.3% Ciprofloxacin Hydrochloride Ablation Depth, Ablation Depth, Ablation Depth, Figure 3. The 3-dimensional representation of mean fluorescence spectra vs the ablation depth in rabbit corneas treated with 4 fluoroquinolones and.3% tobramycin sulfate. The treatments were applied as follows: A, balanced salt solution (n=5); B,.3% tobramycin (n=5); C,.3% ofloxacin (n=5); D,.5% levofloxacin (n=3); E,.3% gatifloxacin (n=4); and F,.3% ciprofloxacin hydrochloride (n=3). AU indicates absorbance units. should be able to detect these antibiotics in the cornea at the level of their minimum inhibitory concentrations for many more bacterial species. Moreover, different than most current techniques used to determine the total corneal antibiotic depot, we should be able to determine these values in a depth-dependent manner in vivo. Knowledge of the antibiotic concentration gradients will be invaluable during lamellar corneal surgery, for example, LASIK and treatment of corneal infections at specific depths. If fluoroquinolones do not penetrate much past the epithelium immediately after application to intact epithelium, the epithelial levels of fluoroquinolone would increase the fluorescence during transepithelial ablation. This should not be a concern with LASIK. Under our in vitro experimental conditions, the current topical fluoroquinolone antibiotics did not penetrate very deeply past the corneal surface. Thus, repeated and frequent application might be needed to accumulate enough of the drug in the deeper layers of the cornea to reach therapeutic concentrations. Live animal and human trials are needed to determine antibiotic penetration in vivo under normal conditions with intact epithelium, Bowman layer, and anterior stroma. Possible explanations for greater penetration of antibiotics detected in living corneas include the repeated administration in many in vivo studies and coapplication of epithelium-damaging drugs such as topical anesthetics, as well as the intraoperative dehydration that occurs when the eyelids are opened with an eyelid speculum under the operating microscope. The resulting dehydration will cause the cornea to imbibe an antibioticcontaining solution. Because we did not first desiccate our corneas under the operating microscope, our experimental situation did not completely mimic the clinical enviroent. There are 2 major factors that enhance the statistical reliability (significance) of our small data set in this pilot study. First, the measured penetration depths for each drug seemed to be significantly close with narrow standard deviations. Second, the number of corneal samples does not affect the accuracy of the penetration-depth measurement because this measurement is determined from the hundreds of fluorescence spectra acquired during the corneal ablation process. The cross-validated accuracy of the penetration-depth measurements exceeded 9%. Initially, 3 globes were tested for each antibiotic. Based on gross visual inspection of the initial unprocessed data, a few extra globes were added to validate that the fluoroquinolones did not penetrate past the epithelium (Table). The final spectral analysis of the large data sets for each globe proved that all of the measured spectra were valid. If we truly only achieve therapeutic levels of fluoroquinolone antibiotic in the epithelium after topical drug administration shortly before surgery, then we may need to reconsider our standards of preoperative- and postoperative-dosing regimens. Methods to increase the corneal drug depot include increasing applied topical drug concentrations and frequency of dosing, and application only after creation of a surgical wound when an in- (REPRINTED) ARCH OPHTHALMOL / VOL 122, NOV Downloaded From: on 11/4/ American Medical Association. All rights reserved.

6 A.3% Ofloxacin B.5% Levofloxacin C.3% Gatifloxacin D.3% Ciprofloxacin Hydrochloride E BSS Figure 4. Fluorescence spectra of.3% ofloxacin (A),.5% levofloxacin (B),.3% gatifloxacin (C), and.3% ciprofloxacin hydrochloride (D) diluted with balanced salt solution (BSS), each with final concentrations of.6 µg/ml; E, Fluorescence spectra of BSS alone. AU indicates absorbance units. Table. Penetration Depth of Each Antibiotic Into Rabbit Corneal Tissue Antibiotic Penetration Depth, Case No..3% Ofloxacin.3% Gatifloxacin.5% Levofloxacin.3% Ciprofloxacin Hydrochloride NA NA NA NA NA Mean ± SD 7.1 ± ± ± ±.6 Prospective accuracy, %* Abbreviation: NA, not applicable. *Projected values are for antibiotic concentrations of.6 µg/ml or greater. tact epithelial barrier is no longer present. Among the antibiotics tested, only.5% levofloxacin was not used near its maximum solubility concentration. Finally, with respect to photorefractive keratectomy, if the therapeutic antibiotic depot does reside for the most part in the epithelium, then preoperative antibiotic administration provides little advantage, as the epithelium is ablated away during the initial stages of surgery. During this ablation of doped epithelium, one must also be wary of the generated secondary fluorescence derived from the fluoroquinolone dopant. In LASIK, however, the epithelium, and, thus, possibly the drug depot, remains intact for the most part. Because infections after anterior segment surgery are typically caused by the patient s own resident flora that may be recovered from the conjunctivae, eyelids, and nasopharynx, it would still be helpful in any case to apply these drops preoperatively, as they may be effective at reducing conjunctival flora. Submitted for Publication: June 5, 23; final revision received March 9, 24; accepted April 29, 24. Correspondence: Roy S. Chuck, MD, PhD, Wilmer Ophthalmological Institute, John Hopkins University, Jefferson Building, 6 N Wolfe St, Baltimore, MD 2/ (rchuck1@jhmi.edu). (REPRINTED) ARCH OPHTHALMOL / VOL 122, NOV Downloaded From: on 11/4/ American Medical Association. All rights reserved.

7 Funding/Support: This study was supported by grants from Nidek Technologies Inc, Pasadena, Calif (Drs Chuck and Shehada) and grant EY412-1A1 from the National Eye Institute, National Institutes of Health, Bethesda, Md (Dr Chuck). REFERENCES 1. McDonnell PJ. Emergence of refractive surgery. Arch Ophthalmol. 2;118: Loree TR, Johnston TM, Birmingham BS, McCord RC. Fluorescence spectra of corneal tissue under excimer laser irradiation. Proc SPIE. 1988;98:6. 3. Muller-Stolzenburg NW, Muller G, Buchwald HJ, Schrunder S. UV exposure of the lens during 193- excimer laser corneal surgery. Arch Ophthalmol. 199; 18: Ediger MN. Excimer-laser-induced fluorescence of rabbit cornea: radiometric measurement through the cornea. Lasers Surg Med. 1991;11: Tuft S, Al-Dhahir R, Dyer P, Zhu Z. Characterization of the fluorescence spectra produced by excimer laser irradiation of the cornea. Invest Ophthalmol Vis Sci. 199;31: Phillips AF, McDonnell PJ. Laser-induced fluorescence during photorefractive keratectomy: a method for controlling epithelial removal. Am J Ophthalmol. 1997; 123: Chuck RS, Arnoldussen ME, Cohen D, et al. 193 excimer laser-induced fluorescence from cadaveric rabbit corneas. Revista Oftalmoigical Venezuela. 21; 57: Cohen D, Chuck RS, Bearman G, McDonnell PJ, Grundfest WS. Ablation spectra of the human cornea. J Biomed Opt. 21;6: Goldstein MH, Kowalski RP, Gordon YJ. Emerging fluoroquinolone resistance in bacterial keratitis: a 5-year review. Ophthalmology. 1999;16: Neu HC. Chemical evolution of the fluoroquinolone antimicrobial agents. Am J Med. 1989;87(suppl 6C):2S-9S. 11. Liang H, Kays MB, Sowinski KM. Separation of levofloxacin, ciprofloxacin, gatifloxacin, moxifloxacin, trovafloxacin and cinoxacin by high-performance liquid chromatography: application to levofloxacin determination in human plasma. J Chromatogr B Analyt Technol Biomed Life Sci. 22;772: Jain S, Azar DT. Eye infections after refractive keratotomy. J Refract Surg. 1996; 12: Al-Rajhi AA, Wagoner MD, Badr IA, al-saif A, Mahmood M. Bacterial keratitis following phototherapeutic keratectomy. J Refract Surg. 1996;12: Forster W, Becker K, Hungermann D, Busse H. Methicillin-resistant Staphylococcus aureus keratitis after excimer laser photorefractive keratectomy. J Cataract Refract Surg. 22;28: Kaldawy RM, Sutphin JE, Wagoner MD. Acanthamoeba keratitis after photorefractive keratectomy. J Cataract Refract Surg. 22;28: Heidemann DG, Clune M, Dunn SP, Chow CY. Infectious keratitis after photorefractive keratectomy in a comanaged setting. J Cataract Refract Surg. 2; 26: Dunphy D, Andrews D, Seamone C, Ramsey M. Fungal keratitis following excimer laser photorefractive keratectomy. Can J Ophthalmol. 1999;34: Brancato R, Carones F, Venturi E, Cavallero A, Gesu G. Mycobacterium chelonae keratitis after excimer laser photorefractive keratectomy. Arch Ophthalmol. 1997; 115: Wee WR, Kim JY, Choi YS, Lee JH. Bacterial keratitis after photorefractive keratectomy in a young, healthy man. J Cataract Refract Surg. 1997;23: Amayem A, Ali AT, Waring GO III, Ibrahim O. Bacterial keratitis after photorefractive keratectomy. J Refract Surg. 1996;12: Kalayci D, Basci N, Akpek E, Hasiripi H, Bozkurt A. Penetration of topical.3% ciprofloxacin into human aqueous humor. Ophthalmic Surg Lasers. 1996;27: Hanioglu-Kargi S, Basci N, Soysal H, et al. The penetration of ofloxacin into human aqueous humor given by various routes. Eur J Ophthalmol. 1998;8: Basci NE, Bozkurt A, Kalayci D, Kayaalp SO. Rapid liquid chromatographic assay of ciprofloxacin in human aqueous humor. J Pharm Biomed Anal. 1996; 14: Basci NE, Hanioglu-Kargi S, Soysal H, Bozkurt A, Kayaalp SO. Determination of ofloxacin in human aqueous humour by high-performance liquid chromatography with fluorescence detection. J Pharm Biomed Anal. 1997;15: Cekic O, Batman C, Totan Y, et al. Penetration of ofloxacin and ciprofloxacin in aqueous humor after topical administration. Ophthalmic Surg Lasers. 1999; 3: Beck R, van Keyserlingk J, Fischer U, Guthoff R, Drewelow B. Penetration of ciprofloxacin, norfloxacin, and ofloxacin into the aqueous humor using different topical application modes. Graefes Arch Clin Exp Ophthalmol. 1999;237: Cantor LB, Donnenfeld E, Katz LJ, et al. Penetration of ofloxacin and ciprofloxacin into the aqueous humor of eyes with functioning filtering blebs. Arch Ophthalmol. 21;119: Donnenfeld ED, Perry HD, Snyder RW, et al. Intracorneal, aqueous humor, and vitreous humor penetration of topical and oral ofloxacin. Arch Ophthalmol. 1997; 115: McDermott ML, Tran TD, Cowden JW, Bugge CJL. Corneal stromal penetration of topical ciprofloxacin in humans. Ophthalmology. 1993;1: Diamond JP, White L, Leeming JP, Bing Hoh H, Easty DL. Topical.3% ciprofloxacin, norfloxacin, and ofloxacin in treatment of bacterial keratitis: a new method for comparative evaluation of ocular drug penetration. Br J Ophthalmol. 1995; 79: Kowalski BR, ed. Chemometrics, Mathematics and Statistics in Chemistry. Hingham, Mass: D Kluwer Academic Publishers; Geladi P, Kowalski BR. Partial least-squares regression: a tutorial. Anal Chim Acta. 1986;185: Wold H. Research Papers in Statistic. In: David FN, ed. New York, NY: John Wiley & Sons Inc; 1966: Mather R, Karenchak LM, Romanowski EG, Kowalski RP. Fourth-generation fluoroquinolones: new weapons in the arsenal of ophthalmic antibiotics. Am J Ophthalmol. 22;133: (REPRINTED) ARCH OPHTHALMOL / VOL 122, NOV Downloaded From: on 11/4/ American Medical Association. All rights reserved.