Ashraf Ali Tayyab, Tehseen-Un-Nabi Sahi, Muhammad Zafar Iqbal Muhammad Zafar Ullah, Shafqat Rasool

Prof. Dr. M. Lateef Chaudhry

Correspondence to:
M. Saleem
Deptt. of Microbiology
Akhtar Eye Hospital
Postgraduate Teaching Hospital of Ocular Disease
FL-1 (4/C) Block-05 Gulshan-e-Iqbal
Karachi

Aims: In this study, we compare the susceptibility of the corneal bacterial isolates on ten different antibiotics from various groups. Using in vitro bacterial sensitivity data, the authors attempt to (1) confirm the most common susceptible antibiotic against the various corneal bacterial isolates. (2) Significance and choice of board-spectrum antibiotics as a first line therapy. (3) Whether fluoroquinolones antibiotics (Ciprofloxacin) is superior to aminoglycosides and other antibiotics.
Methods: Antibiotic susceptibilities were determined for 61 isolates from patients with bacterial keratitis. Results were analysed for each antibiotic individually.
Results: We found fluoroquinolones (Ciprofloxacin) are effective to all the corneal bacterial isolates. The efficacy of Ciprofloxacin against individual species were noted as 100% Streptococcus species, 100% Pseudomonas aeruginosa, 100% other gram-negative bacteria, 95.8% Staphylococcus coagulase-positive and 85.7% other gram-positive bacteria.

The over-all susceptibilities of ten individual antibiotic against both Gram-positive and Gram negative were 95.0% Ciprofloxacin, 90.1% Amikacin, 80.3% Tobramycin, 77% Chloramphenicol, 73.8% Gentramicin, 70.4% Vancomycin, 63.9% Neomycin, 49.1% Cotrimoxazole, 47.5% Polymyxin B and 31.1% Cefixime.
We compared the susceptibilities of fluoroquinolones (Ciprofloxacin) to aminoglycosides (Amikacin, Gentamicin, Neomycin, and Tobramycin) against all the isolates. All the corneal bacterial isolates showed significantly better susceptibility to fluoroquinolones (Ciprofloxacin) than to aminoglycosides antibiotics.
Among the aminoglycosides, 90.1% Amikacin showed excellent efficacy, 80.3% Tobramycin, 73.8% Gentamicin and 63.9% Neomycin against all the corneal bacterial isolates.
Conclusions: Consequently, fluoroquinolones (Ciprofloxacin) is apropos antibiotic to be used as single therapeutic agent in bacterial keratitis. Fluoroquinolones is superior to aminoglycosides in bacterial keratitis in this study.

In mid nineteenth century discovery of Sulfa drug and Penicillin revolutionized the treatment of bacterial infection1.
In bacterial ocular infections it is essential to decide an optimal antibiotic as the distribution of various ophthalmic pathogens and their sensitivity to antibiotic treatment vary from one geographical region to another2.
Bacterial keratitis is a potentially sight-threatening disorder that is commonly observed by ophthalmologist. Bacterial keratitis can result in permanent loss of vision if not treated promptly and appropriately. The cornerstone of successful treatment is effective topical antimicrobial therapy. Empiric therapy must be initiated with board-spectrum regimen until culture results confirm the identity and antibiotic susceptibility of the causative organism3.
The successful treatment of bacterial infection depends on identification of the causative organism and its antibiotic susceptibility. Local antibiotic sensitivity differs in various parts of the world4 according to many reports from USA, UK, Europe, Mid-East and some countries of Asia2,5,6. Thus, it was important for us to find out, whether suggested antibiotic regimens based on information from other regions are efficient for treating ocular infection in Pakistan.

MATERIALS AND METHODS
Between June 2002 and September 2003, 61 consecutive ophthalmic bacteriological cultures from patients with corneal ulcers where obtained in the OPD of Akhter Eye Hospital Karachi, and the cultures were analyzed in the laboratory of Microbiology. All suspected infectious corneal ulcers were scraped for laboratory studies before treatment was initiated. Surgical blade No. 11 (Feather safety Razor Co. Japan) was used to scrape the leading edges of the filtrate as firmly as was judged to be safe.
For culture test, swab (Commercially available Stuart transport medium, Oxide, UK) was taken from the affected area. Another scraping was smeared on a slide for microscopic examination by gram stain.
Laboratory data were documented including gram-staining result, bacterial and fungal cultures, and susceptibility testing. Patients suspected of having fungal infection were excluded from study. Patients suspected of having bacterial infection but on culture results were found no growth were excluding from this study too.
Bacterial isolates were tested for susceptibility to various groups of antibiotics, fluoroquinolones (Ciprofloxacin), aminoglycosides (Amikacin, Gentamicin, Neomycin, Tobramycin) cotrimoxazole, Polymyxin B, Vancomycin, and Cefixime with different minimum inhibition concentration (MIC), using the standard Bauer-Kirby Disk diffusion methods7. Each corneal bacterial isolate was reported susceptible, intermediate and resistant. All tests were interpreted following the guidelines established by the National Committee on Clinical Laboratory Standard (NCCLS)8.

RESULTS
Sixty-one various bacterial pathogens were isolated from 61 patients; 40 (65.6%) were gram-positive 21 (34.4%) gram-negative. The susceptibility of the pathogens was tested, against various groups of antibiotics. The mode of action and group of antibiotics, used in susceptibility test, is given in
table 1.
Out of the 61 bacterial isolates, 24 (39.3%) were confirmed as Staphylococcus coagulase-positive bacteria which showed 95.8% sensitivity to Ciprofloxacin and 95.5% to amikacin while showed lower sensitivity to 33.3% Cefixime. Ciprofloxacin was also found sensitive to 8 (100%) Pseudomonas aeruginosa, 8 (100%) Streptococcus spices, 13 (100%) Gram negative organism and 8 (87.5%) other gram-positive. The susceptibilities of Staphylococcus coagulase-positive bacteria to other antibiotics were noted as 23 (95.8%) Amikacin, 21 (87.5%) Vancomycin and Tobramycin, 19 (79.1%) Chloramphenicol and gentamicin, 15 (62.5%) Neomycin, 16 (66.7%) Cotrimoxazole and Polymyxin B while 8 (33.3%) Cefixime showed the lowest susceptibility receptively Pseudomonas aeruginosa was showed individually. Susceptibilities to other antibiotics were noted as 7 (87.5%) Tobramycin and amikacin, 6 (75%) Chloramphenicol, Gentamicin, 5 (62.5%) Neomycin 3 (37.3%) Cotrimoxazole and Polymyxin B, Cefixime, one While 1 (12.5%) Vancomycin showed the similar susceptibility receptively.
The susceptibilities of Streptococcus spices, was noted against 8 (100%) Chloramphenicol, 7 (87.5%) Amikacin, Gentamicin and Neomycin, 6 (75%) Vancomycin and Tobramycin, 5 (62.1%) Cotrimoxazole while 1 (12.5% Polymyxin B and Cefixime showed the similar susceptibility receptively.
The over-all susceptibilities of ten individual antibiotics against both Gram-positive and Gram-negative were 95% Ciprofloxacin, 90.1% Amikacin, 80.3% Tobramycin, 77% Chloromphenicol, 73.8% Gentamicin, 70.4% Vancomycin, 63.9% Neomycin, 49.1% Cotrimoxazole, 47.5% Polymixin B and 31.1% Cefixime.
All the corneal bacterial isolates showed significantly better susceptibility to fluoroquinolones (Ciprofloxacin) than aminoglycosides antibiotics. Among the aminoglycosides, 90.1% Amikacin showed excellent susceptibility than 80.3% Tobramycin, 73.8% Gentamicin and 63.9% Neomycin against all the isolates.
The spectrum of susceptibilities and MIC of antibiotics disc and all bacterial isolates are given in table 2.

DISCUSSION
Antibiotic agents have been used for years by the ophthalmologist in ocular infection. In the past few years, new commercially prepared antibiotics agent for topical ocular use have become available. Because treatment of bacterial keratitis typically is initiated before the result from a laboratory determination of the causative organism are obtained, it is important to use a board spectrum antibiotics as a first line therapy.
Topical fluoroquinolones has been available since 1990 and represent the newest class of ophthalmic antibiotics. An in-vitro antibiotic susceptibility study previously demonstrated that fluoroquinolones (Ciprofloxacin, Ofloxacin) were more active against common ocular pthogens than were Aminoglycosides (Tobramycin, Neomycin, and Gentamicin), Chloramphenicol, Teracycline, and Erythromycin9. In study by Jensen and Felix,9 all three fluoroquinolones tested had similar efficacy against gram-negative bacteria. Ofloxacin demonstrated similar efficacy against gram-negative than gram-positive pathogens. In our study we also found Ciprofloxacin highly effective against gram-negative and gram-positive organism.
Osato et al10 found that among 419 ocular bacterial isolates comprising 55 species, Ofloxacin had low MIC 90 for both gram-negative and gram-positive organism in comparison to other five antibiotics: such as Norfloxacin Gentamicin, Tobramycin, Chloram-phenicol and Polymyxin B. They also found that the incidence of resistant to be lowest for ciprofloxacin (7.3%) and Ofloxacin (3.7%). In our study we found that only 4.2% Staphylococcus coagulase positive showed resistance to Ciprofloxacin.
Cutarelli et at11 studied 96 bacterial isolates, which unlike the studies cited above, were from extra-ocular as well as intraocular sources. The authors found Ciprofloxacin the most effective antibiotic tested with MIC 90 of 1 milligram per liter (mg/l), followed by Ofloxacin 2 mg/l.
Gauze et al12 found Ciprofloxacin therapy better in comparison to two other conventional antibiotics, against Pseudomonas aerugnosa in keratitis. Similarly in our study we found 100% Ciprofloxacin, 87.5% Tobramycin and 37.3% Polymyxin B susceptible against Pseudomonas aeruginosa.
Another study of bacterial keratitis Bower et al3 used fluoroquinolones individually and in combination with cefazolin. Bower and colleagues noted susceptibilities of Ciprofloxacin to 100% Pseudomonas aeruginosa, 100% other gram negative bacteria, 95.3% Staphylococcus coagulase-positive, 85.7% other gram-positive bacteria and 42.8% Streptococcus species. Out data is comparable to that of Bower and colleagues noted above.
In bacterial corneal ulcer study, Hyndiuk and coworker13 found Pseudomononas aeruginosa susceptible to 100% Tobramycin+cefazolin and 92.8% Ciprofloxacin. Staphylococcus coagulase-positive 85.7% to Ciprofloxacin and 66.7% were Susceptible to Tramycin+cefazolin respectively. Although our data are not exactly comparable with Hyndiuk and coworker noted above, our results are not dissimilar.
Han et al14 noted Staphylococcus coagulase-positive were susceptible to 100% Vancomycin, 57.5% Ciprofloxacin, 49.3%) Amikacin, while other gram-positive & gram-negative organism showed 94.9% Ciprofloxacin, 89.5% Amikacin. Although his report is not strictly similar to our data and nor analog to our data. Davison et al15 study the susceptibilities of isolated from bulbar conjunctival of 10 patients, using Tobramycin, Gentamicin and other antibiotics but did not noted the susceptibilities of each antibiotics in percentage. Brain et al6 noted Ciprofloxacin is highly active in vitro against Pseudomonas aeruginosa as we found.
Ciprofloxacin is a fluoroquinolones antibiotic with broad-spectrum activity against most aerobic gram-positive and gram-negative bacteria. Ciprofloxacin usually is considered to have the greatest overall antibacterial activity in fluoroquinolones group.12 Ciprofloxacin received US. Food and Drug Administration approval for topical treatment of bacterial corneal ulcers in December 199016.
Currently, Ciprofloxacin is available in solution and Ointment dosage marketed in Pakistan for topical ocular treatment of corneal ulcers. There are four major advantages of Ciprofloxacin and other fluoroquinolones. Their is excellent ocular penetration, demonstrated board-spectrum efficacy, safety profile in ocular infections, and distinct mode of resistance acquisition Chromosomal mutation rather than plasmid-mediated.13
In developing countries particularly in Pakistan, the access of ophthalmologist is difficult to Laboratory on account of scarcity of Microbiological Laboratories. This study will help for the local ophthalmologists, treating bacterial keratitis. This Uni-center study will also give a baseline for further investigations.
In summary, Bacterial keratitis is most frequently caused by Staphylococcus coagulase-positive. Pseudomonas aeruginosa, Streptococcus species and few other gram-negative organisms.3,15 Although Staphylococcus coagulase-positive, Pseudomonas aeruginosa and Streptococcus species were the leading organism isolated from corneal culture in our study. The organism most frequently isolated from corneal ulcer was Staphylococcus coagulase-positive. We found Ciprofloxacin was effective against all the corneal bacterial isolates. Fluoroquinolones (Ciprofloxacin is superior to aminoglycosides (Amikacin, Gentamicin, Neomycin, and Tobramycin) in this study.

Table 1: Groups of Antibiotics and Antibiotic Disc Used in Susceptibility Tests

Table 2: In-vitro Susceptibity of Corneal Bacterial Isolates to Individual Antibiotic

Among Aminoglyocosides group Amikacin showed the excellent susceptibility to gram-positive and gram-negative organism. Our report therefore supports the tendency to use fluoroquinolones (Ciprofloxacin) as initial single drug therapy or first line therapy for bacterial corneal ulcer.4,12,17 Ciprofloxacin have board-spectrum antimicrobial activity, good ocular penetration, low resistance, and high patients’ acceptance with few adverse effects18.

Authors Affiliations
Akhtar Jamal Khan
Medical Director
Akhtar Eye Hospital
Karachi
M Saleem
Department of Microbiology
Akhtar Eye Hospital
Karachi

REFERENCES
1. Pelecxzar JM, Chan SCE and Krieg RN. “Microbiology” 5th edition, Internation student Edition USA 510-1; 1986.
2. Mezer E, Gelfand AY, Lotan R, Tanir A and Miller B. Bacteriological profile of ophthalmic infection in an Israeli hospital. Eur J. Ophthalmol 1999; 9:120-4.
3. Bower SK, Kawalski PR and Gordon JY. Fluoroquinolones in treatment of bacterial keratitis, Brief report. Am J Ophthalmol 1996;121:713-5.
4. Allan BDS, Dort JKG. Strategies for management of Microbial keratitis. Br J Ophthalmol 1995;79:777-86.
5. Hartikainen J, Lektonen OP, Saari KM. Bacteriology of bacterial ducts obstruction in Adults. Br. J Ophthalmol1997; 82: 653-8.
6. Brian PT, Swausch RM, Dick DJ and Gottsch DJ. Topical ciprofloxacin treatment of Pseudomonas keratitis in Rabbits, Arch Ophthalmol 1988;106: 1444-6.
7. Baron JE, Frieglod MS. Bialy & Scott diagnostic microbiology ed. 8th CV. Mosby company 181-3; 1990.
8. National Committee for Clinical Laboratory Standards (NLCCS). Performance standards for antimicrobial susceptibility testing. Villanova, PA: 1992 (Document M100-54 Vol. 12 No. 20.
9. Jensen HG, Flix C. In-vitro susceptibilities of ocular isolates in north and South America. In-vitro antibiotic testing group. Cornea 1998;17: 78-9.
10. Osato MS, Jensan HG, Trousdale MD and Basso JA. The comparative in in-vitro activity of Ofloxacin, and selected ophthalmic antimicrobial agent ocular bacterial isolates. AM J Ophthalmol 1998;108: 380-6.
11. Cutarella PE, Lass JH, Lazarus HM. Topical fluoroquinolones antimicrobial activity and in-vitro corneal eqithelical toxicity. Current Eye Research, 1991;10: 557-63.
12. Guzek PJ, Chico D, Kettering DJ, Weasels FI, Appricot MR. Comparison of Topical Ciprofloxacin to conventional antibiotic Therapy in the treatment of exaperimental Pseudomonasaeruginosa keratitis. Cornea 1994;13(6): 500-4.
13. Hydiuk AR, Caldwell RD, Rosenwasser, Santos IC, Katz RH, Badrinath SS, Reddy KM, Advenis PJ and Klaus V. The ciprofloxacin bacterial keratitis group, comparison of Ciprofloxacin ophthalmic solution 0.3% to foritified Tobramycin-Cefazolin and treating bacterial corneal ulcer.Ophthalmol 1996;103:1854-63.
14. Han PD, Wisniewk RS, Wilson AL, Barza M, Vine KA, Doft HB and Kelsey FS. Spectrum of susceptibilities of microbiologic isolates in the endophthalmitis vitrectomy study. Am J Ophthalmol 1996;122:1-17.
15. Davison CR, Tuft JS, Dart GK. Conjuntival necrosis after administration of topical fortified aminoglycosides. Am J Ophthalmol 1991;111: 690-3.
16. Synder EM, Katz RH. Ciprofloxacin-resistant bacterial keratitis, Am J Ophthalmol 1992;114:336-8.
17. Alxendrak G, Alonos CE and Miller D. Shifting trends in bacterial keratitis in the south Floridsa and emergingh resistance to fluoroquinolones, Ophthalmol 2000;107: 1497-1502.
18. Neu HC. Microbiologic aspect of fluoroquinolones. Am J Ophthalmol 1991;15s-24s: 112.

Correspondence to:
Mr Q. Mansoor
Block-L, Flat-4, Staff Village
Royal Preston Hospital
Preston, Lancashire
UK

It is well established that preoperative biometry enables more accurate selection of intraocular lens power for a desired postoperative refraction1. Biometry involves the keratometeric measurement of curvature of cornea and also the measurement of axial length. A combination of these two measurements is incorporated in to a formula for calculation of power of intraocular lens2. Axial length measurements have been identified as being the major source of error in intraocular lens power calculations and less accurate than keratometry3, 4.

There are two methods of biometry currently in practice, one is acoustic biometry and the other one is called optical biometry. In acoustic biometry ultrasonic waves follow the optical axis of the eye. These waves are reflected back from internal limiting membrane to the probe are amplified and displayed on the screen5. In optical biometry, partial coherence laser interferometer measures the axial length along visual axis. The light from diode laser passes an interferometer that splits the beams into two parallel beams. One of the beams is retarded by a path difference of twice the plate spacing, which reflect that beam. Both of the beams illuminate the eye through the beam splitter. They serve as a measuring beams as well as fixation target. The time needed for measurement of axial length is about 0.5 sec. The optical biometry is a fast, non contact, highly sophisticated technique of measuring axial length with a resolution of 0.1mm. The PCLI measures from anterior pole of the eye to retinal pigment epithelium in contrast to acoustic biometry that measures from anterior pole to internal limiting membrane. The distance between internal limiting membrane and retinal pigment epithelium ranges 150-350?m.
We did a prospective comparison of accuracy of partial coherence laser interferometry (optical biometry) and ultrasound A- scan (Acoustic biometry) in measuring axial length in order to determine the relative accuracy of the two methods for routine cataract surgery refractive outcomes.

METHODS
Fifty eyes of 50 patients were analysed. These patients were divided into two groups, each containing 25 patients. Preoperative assessment of Group (I) was done by Zeis IOL Master (partial coherence laser interferometry) and patients in Group (II) were assessed by Ocuscan (ultrasound A- scan) with standard keratometry (automated). Ultrasonic axial length measurements were done by using applanation method. In all patients SRK/T formula was used to calculate IOL power and A- constant was 118.0.

The cataract extraction and intraocular lens implantation was performed by phacoemulsification with small incision (2.8-3.2mm). In all patients surgery was uneventful. Results were analysed 6 weeks postoperatively.

RESULTS
The results were based on the difference between planned postoperative refraction and final postoperative refraction (6 weeks post-op). The spherical equivalent of postoperative refraction was used in all patients for the refractive comparison. Mean keratometric value in patients where axil length was measured by partial coherence laser interferometer (PCLI) was 44.49D where as in cases where biometery was done with ocuscan was 43.72D. Axial length range was 20.8mm-26.6mm almost symmetrical for these two different groups as shown in figure I.

Fig. 1: Axial length, comparing Ocuscan and Zeiss Master

Fig. 2: Comparison of Expected and Actual post-op refraction

of  patients with various Axial length by Ocuscan.

Fig. 3: Comparison of Expected and Actual post-op refraction

of patients with various Axial length by Zeis IOL Master

The average difference between the expected refraction and the actual 6 week post operative refraction in patients in whom biometry was done by partial coherence laser interferometry (PCLI) was 0.66D and in patients who were pre-assessed with acoustic biometry the average difference was 0.73D. The difference between the expected refraction and the actual 6-week postoperative refraction between PCLI and Ocuscan was 0.07D.

DISCUSSION
Cataract extraction is one of the most frequently and successfully performed ophthalmic procedures. Over the last fifty years, the main objective of cataract extraction has been transformed from merely improving visual acuity to that of improving the quality of vision, hence to improve quality of life. While the techniques of cataract operation are constantly improving, the desire and demand of patients and surgeons for high predictability in refractive results has increased dramatically5.

Axial length measurements have been identified as major source of error in intraocular lens power calculation4. The survey has shown that 95.5% of axial length measurements showed a standard deviation of less than 0.3mm3. Axial length measurements also differ with axis and five-degree off-axis shift is suggested to cause 0.3mm variation in axial length. This disparity increases if there is a posterior staphyloma in which visual axis move further away from optical axis3, 4,,6,7

The different studies suggest that there is a mean difference in axial length of 0.2mm5,7,8. Axial lengths measured by applanation technique in acoustic biometry further increase this difference by applanating the cornea. As we know that 0.1mm difference in axial length measurement produces 0.25D refractive error, so the potential difference in axial length measurement by these two different equipments may cause noticeable difference between planned and final postoperative refraction in cataract surgery4,5,7,9.

CONCLUSION
Our study shows that the IOL power calculation by optical biometry improves the predictive value of postoperative refraction. This is because it utilizes the principal of partial coherence laser interferometer that gives more accurate measurement of axial length that is near to original axial length of the eye. Optical biometry is operator independent, fast and highly reproducible as it contains database. In view of axial length measurement that is major source of error, optical biometry remarkably reduces the chances of this error. However, in patients with dense cataract, partial coherence laser interferometer does not work very well so that in these cases acoustic biometry is still needed. However, ultrasonic biometry in experienced hands gives reliable and predictive measurement of axial length and works equally well in dense cataracts.

Authors affiliations
Q. Mansoor
Block-L , Flat-4, Staff Village
Royal Preston Hospital
Preston, Lancashire
UK
S. A. Hussain
Royal Preston Hospital
Preston, Lancashire, UK
Miss W Hameed
Royal Preston Hospital
Preston, Lancashire, UK

REFERENCES
1. Thomson SM, Mahon RV. A comparison of postoperative results with and without intraocular lens power calculation. Br J Ophthalmol, 1986; 70:22-5.
2. Sanders DR, Retzlaff JA, Kraff MC et al. Comparison of SRK/T formula and other theoretical and regression formulae. J Cat Refract Surg 1990; 16:341-3.
3. Longstaff S. Factors affecting IOL power calculation. Trans.Ophthalmol.Soc.UK 1986, 105:642-6.
4. Richards SC, Oslen RJ. Factors associated with poor predictability by IOL calculation formulas. Arch Ophthalmol 1985; Apr; 103(4):515-8.
5. Rajan MS, Keilhorn I. Partial laser coherent interferometer and conventional ultrasound Biometry in IOL power calculation. Eye 2002 Sep; 16(5): 552-6.
6. Packer M, Fine IH, Hoffman RS et al. Immersion A- scan compared with partial coherence interferometry: Outcome analysis. J Cataract Refract Surg 2002; 28(2):239-42.
7. Kiss B, Findl O, Menapace R, Wirtitsch M et al. Refractive outcome of cataract surgery using partial coherence interferometry and ultrasound biometry: clinical feasibility study of a commercial prototype II. J Cataract Refract Surg. 2002; 28(2):230-4.
8. Drexler W, Findl O, Menapace R et al. Partial coherence interferometry: a novel approach to biometry in cataract surgery. Am J Ophthalmol 1998; 126(4):524-34.
9. Vogel A, Dick HB, Krummenauer F. Reproducibility of optical biometry using partial coherence interferometry : intraobserver and interobserver reliability. J Cataract Refract Surg. 2001; 27(12):1961-8.