Battling Bacteria
How to choose the most effective drugs to treat infection and avoid complications.

MARK T. DUNBAR, O.D., F.A.A.O. Miami

Therapeutic laws in most states now allow you to treat a majority of ocular infections that present in your practice. Bacterial infections are perhaps one of the most commonly seen. Fortunately, you have at your disposal an armament of very effective drugs. Depending on the type of infection, naturally, it’s important to choose the most effective drug or drugs to treat the infection to prevent vision loss or other secondary complications. Here, I will discuss the treatment options we have for bacterial and corneal infections.
Bacterial infection
Bacterial conjunctivitis is one of the most common reasons for emergent office visits in both pediatric- and primary-care practices. It’s important to treat these cases early to shorten the disease course, prevent spread of infection and reduce patient discomfort, so these patients may resume their daily activities as soon as possible. A number of topical drug therapies are available to treat bacterial conjunctivitis. Because the causative organism can differ by age of the patient, it’s important to choose a drug that has a broad spectrum of activity against both gram-positive and gram-negative organisms. In newborns, Staphylococcus aureus tends to be the most common pathogen, whereas in older children, you’re more likely to see a mix between Streptococcus pneumoniae and Haemophilus influenzae. In adults, Staphylococcus species tend to be the most common.1 Culturing isn’t routinely done with bacterial conjunctivitis, as most cases respond well to antibiotic therapy. The most commonly prescribed drugs include the aminoglycosides (tobramycin and gentamycin), combination therapy with trimethoprim sulfate and polymyxin B sulfate (Polytrim, Allergan) and the fluoroquinolones. The aminoglycosides have excellent coverage against gram-negative bacteria but don’t work as well against gram-positive organisms, with little effect on Streptococci organisms.
Bacterial infection treatment options
Trimethoprim sulfate and polymyxin B sulfate has a broad spectrum of coverage and is safe for use in children as young as two months of age. Even though it’s been available since the early 1990s, there has been very little resistance to this drug. This is partially due to the fact that it’s not available in an oral form. This would increase exposure and therefore, increase resistance. In young children, trimethoprim sulfate and polymyxin B sulfate is a very good choice for bacterial conjunctivitis associated with otitis media because of its broad spectrum of activity and minimal resistance. The fourth-generation fluoroquinolones, gatifloxacin (Zymar, Allergan) and moxifloxacin (Vigamox, Alcon), may represent the best choice for treating bacterial conjunctivitis. Their release in 2003 represented an important breakthrough for treating bacterial infections. The older generation fluoroquinolones (Ciloxan [ciprofloxacin, Alcon], oflaxacin [Oculflox, Allergan], norfloxacin [Noroxin, Merck], and levofloxacin [Iquix, Vistakon]) were very potent against gram-negative organisms. However, they were very vulnerable against grampositive infections. As a result, increasing resistance to the oldergeneration fluoroquinolones developed at an alarming rate throughout the 1990s and into the current millennium with one study demonstrating 100% resistance to S. aureus endophthalmitis isolates.2 The fourth-generation fluoroquinolones seemed to address many of the weaknesses of the older generation fluoroquinolones. The addition of an 8- methoxy group on the basic ring structure gave them much greater potency against gram-positive organisms and at the same time, addressed the concern of increasing resistance. Where the older generation fluoroquinolones only needed one mutation within the DeoxyriboNucleic Acid (DNA) of the bacteria for resistance to develop, the fourth-generation fluoroquinolones attack the bacteria in two places, causing lethal breaks within the bacteria’s DNA. Because of this, the bacteria would need to undergo two separate mutations for resistance to develop, which was thought to be virtually impossible. However, if the bacteria already developed resistance to the older generation fluoroquinolones, it became much easier for it to become resistant to the fourth-generation fluoroquinolones by developing a secondstep mutation. So, the recommendation among treating eye doctors was to use the fourth-generation fluoroquinolones as a first-line treatment option for ocular infections. Doing so effectively wipes out bacteria, removing any chance for resistance. Gatifloxacin and moxifloxacin are FDAapproved for treating bacterial conjunctivitis and both show great efficacy against various strains of bacterial pathogens. In one in vitro study of 498 strains of ocular bacterial isolates, gatifloxacin was effective against 99% of the gram-positive strains and 92% of gram-negative strains.3 These drugs have proven safe for those as young as one year of age. Both drugs have been shown to be effective with b.i.d. dosing, however, you should base dosing on presentation and severity. The standard for most typical bacterial conjunctivitis is q.i.d. dosing for five to seven days. After the recommended treatment time, instruct patients to completely stop the antibiotic without a taper. Unnecessary tapering and chronic misuse has been linked to increasing resistance.
Corneal infections
Can the fourth-generation fluoroquinolones be used for treating corneal infections? This is a legitimate question, as this class hasn’t been approved for treating microbial keratitis. Minimum inhibitory concentration (MIC90) data suggest that these antibiotics are just as potent compared with the older generation fluoroquinolones, which have been used to treat microbial keratitis almost since their inception in the early 1990s. Against gramnegative bacterial strains, MIC90 data for the fourth-generation fluoroquinolones are equal to, if not better than, their predecessors, which were considered the benchmark for treating pseudomonas. Specifically against P. aeruginosa, MIC90s of gatifloxacin are equal to that of ciloxin and are four to six times lower than moxifloxacin. With regards to gram-positive infections, the newer generation fluoroquinolones are clearly better with much lower MIC90s.4 (See “What does MIC data tell us?” page 42.) One study compared the MIC90 data of various strains of gram-positive bacteria with both the pure molecule and commercial preparation of gatifloxacin and moxifloxacin. The MIC90 of the pure molecule was fairly equal between the two drugs, with moxifloxacin having a slight edge against various strains of Staphylococci and Streptococci organisms. Even against methicillinresistant S. aureus (MRSA) isolates, against which neither molecule has been shown to have good susceptibility, the commercial preparation of gatifloxacin with benzalkonium chloride (BAK) was very effective compared with moxifloxacin.5 This is significant because one of the growing trends in healthcare is the increasing prevalence of MRSA infections outside of what is typically seen in a hospital. The rise of MRSA infections is very concerning, as almost all the commercially available antibiotics have shown poor susceptibility to MRSA. The fact that commercially available Zymar with 0.005% BAK, compared with the molecules gatifloxacin or moxifloxacin without BAK, shows fairly good susceptibility against MRSA would suggest that Zymar would be the preference for treating suspicious bacterial corneal infections. In another study, regardless of the antibiotic used, the addition of 0.005% BAK to either of the fourth-generation fluoroquinolones significantly reduced the time to kill with various Staphylococci isolates from four hours to less than one hour.6 These results demonstrate that BAK can have an additive effect.
Corneal infection treatment options
Direct comparisons between the older- and newer-generation fluoroquinolones have also shown that fourth-generation fluoroquinolones are very affective against corneal infections. Researchers in one study randomized 104 corneal ulcer patients to treatment with either gatifloxacin or ciprofloxacin. They reported a higher overall complete healing with gatifloxacin (95%) than with ciprofloxacin (81%). Of note: a vast majority of the patients in this study presented with gram-positive infections.7 In another similar study, 129 corneal ulcer patients were randomized to treatment with either gatifloxacin or ciprofloxacin and reported similar results with a significantly higher percentage of the gatifloxacin patients showing more rapid complete healing than the ciprofloxacin group. Also of interest, 93% of all organisms isolated were susceptible to gatifloxacin vs. only 38% susceptible to ciprofloxacin.8 The authors recommended a dosing schedule for microbial keratitis on the first day of presentation as one drop every five minutes for the first 30 minutes, followed by one drop every 30 minutes while awake. Upon bedtime, they recommended one drop q2h. From day two to six, the recommended dosing schedule is one drop every q2h while awake. From day seven to 21, the recommended dosing is one drop q.i.d., while awake. This dosing schedule is consistent with that of the older fluoroquinolones and may vary depending on the ulcer’s location and severity. For instance, peripheral ulcers may not need such aggressive dosing. (See “Dosing schedule for microbial keratitis using fourth-generation fluoroquinolones,” page 45.)
Corneal infection management
At least initially, you should see patients who have a corneal infection daily. As the treating clinician, you should expect to see a steady improvement, that is, the size of the epithelial defect should get smaller and smaller with resolution of the infiltrate. Individual factors and the type of organism may dictate the speed at which the ulcer will heal. Among some corneal specialists, particularly in academic referral centers, there is some reluctance to use single drug therapy for some of the more severe cases of microbial keratitis. Rather, these specialists prefer to use combination therapies; either with a fourth-generation fluoroquinolone and a fortified antibiotic, or two different fortified antibiotics. This is mainly due to the issues with resistance, as previously described and the rise in MRSA. When looking at these trends, it’s important to take into consideration how these data are generated. Many of the patients we see in referral centers are often referred because they’ve failed on whatever therapy (or therapies) their eye doctor initiated. The fact is, there is no accurate data on the number of patients who are successfully treated by their primary-care eye doctor. The reality is that outside these referral practices, most of you are effectively treating with single therapy, typically with a fourth-generation fluoroquinolone, so the patient never makes it beyond the primary-care physician. The likelihood is that only a small minority of patients fail to respond to monotherapy, and these are the patients who are likely to have more resistant organisms. For these reasons, the clinicians at referral centers culture to determine the causative organism and which drug is most sensitive. Healthcare professionals always recommend culturing, particularly for more central lesions. Clearly, not all treating physicians culture their patients. If you choose not to culture or lack the ability to culture, exert more caution in following recovering patients, and follow-up sooner. Immediately refer lesions that don’t respond to treatment to a corneal specialist. The lack of a response to treatment may be due to the fact that the bacteria aren’t susceptible to the antibiotic. This can result from resistant organisms, an MRSA infection, or it could be because the patient has an organism outside the coverage of that particular antibiotic. It’s even possible that the patient doesn’t have a bacterial infection, but rather a completely different type of infection, such as a fungus or a viral infection. Even consider acanthamoeba in these instances. As the corneal ulcer begins to heal, trying to prevent or limit scar formation becomes an important battle, as that is the most common cause of vision loss in these patients. As a result, you often prescribe 48 to 72 hours of topical steroids after initiating antibiotic treatment. There has always been some controversy as to the best time to initiate steroid treatment. Though these drugs are superb in treating the inflammatory response generated by the body’s immune system, they can slow or delay healing. The other danger with steroids, of course, is that certain infections, such as fungi, can be made much worse with the application of topical steroids. For these reasons, prescribe topical steroids with caution, and after proven efficacy with antibiotic treatment. More recently, clinicians have begun to use oral doxycycline (Vibramycin, Pfizer) 100mg b.i.d. concomitantly with topical antibiotic therapy. This isn’t because of the antibiotic properties of doxycycline, which is a potent macrolide, but rather due to its very good anti-inflammatory properties. Doxycycline, has been shown to be very effective in inhibiting metalloproteinase-9 activity, a key player in the inflammatory cascade. Doxycycline also has been shown to help decrease inflammation and prevent perforation in more severe cases of microbial keratitis. You play a key role in treating patients who present to your practices with bacterial infections. The fourth-generation fluoroquinolones represent an important advancement in antibiotic therapy, so use them as a first line for most cases of bacterial conjunctivitis and all cases Resistance to antibiotic therapy will always be a concern among treating eye doctors, and you should suspect it in cases that don’t respond appropriately. Culturing is always recommended prior to initiating treatment. However, fortunately, most patients do respond to their medications despite the fact that a culture isn’t always performed. Finally, refer the patient to a corneal specialist if he or she does not respond appropriately or if you feel the patient has an infection that is beyond your comfort level.

What does MIC data tell us? Minimum Inhibitory Concentration (MIC) data can be very helpful in predicting how effective an antibiotic is against various strains of bacteria, but it doesn’t always give the true picture. One of the limitations of MIC90 data are that commercial antibiotic preparations are not used to determine the MIC against a particular bacterial pathogen but rather the pure molecule is used, without any of its commercial ingredients. The commercial preparations contain a mix of ingredients.

Dosing Schedule For Microbial Keratitis Using Fourth-Generation Fluoroquinolones

DAY 0 (BASELINE)
1 drop every 5 min. for the first 30 min.
1 drop every 30 min. while awake
1 drop q2h after bedtime
DAY 1
1 drop q1h while awake
1 drop q4h after bedtime
DAY 2 – DAY 6
1 drop q2h while awake
DAY 7 – DAY 21
1 drop q.i.d. while awake
DAY 22-28 (IF APPROPRIATE)
1 drop q.i.d. while awake

1. Mah F. Bacterial conjunctivitis in
pediatrics and primary care. Pediatr Clin
North Am. 2006 May;53 Suppl 1:7-10;
quiz 11, 13-5.
2. Kowalski RP, Karenchak LM, Romanowski
EG. Infectious disease: changing
antibiotic susceptibility. Ophthalmol
Clin North Am. 2003 Mar;16(1):1-9.
3. Kaliamurthy J, Nelson Jesudasan
CA, Geraldine P, et al. Comparison of in
vitro susceptibilities of ocular bacterial
isolates to gatifloxacin and other topical
antibiotics. Ophthalmic Res. 2005 May-
Jun;37(3):117-22.
4. Kowalski RP, Dhaliwal DK,
Karenchak LM, et al. Gatifloxacin and
moxifloxacin: An in vitro susceptibility
comparison to levofloxacin, ciprofloxacin
and ofloxacin using bacterial keratitis
isolates. Am J Ophthalmol. 2003
Sep;136(3):500-5.
5. Callegan MC, Novosad BD. Efficacy
of fourth-generation fluoroquinolones
against gram-positive species
commonly involved in ocular infections.
Presented at: Annual Meeting of the Association
for Research in Vision and
Ophthalmology (ARVO); April 30-May
4, 2006: Fort Lauderdale, Fla.
6. Kowalsk RP, Kowalski BR,
Thompson PP, et al. 0.005% Benzalkonium
chloride (BAK): is it effective in
eliminating Staphylococcus aureus and
Pseudomonas aeruginosa? Presented at
Ocular Microbiology and Immunology
Group (OMIG) meeting; October 15,
2005: Chicago.
7. Parmar P, Salman A, Kalavathy
CM, et al. Comparison of topical gatifloxacin
0.3% and ciprofloxacin 0.3%
for the treatment of bacterial keratitis.
Am J Ophthalmol. 2006 Feb;141(2):
282-6.
8. Prajna V, Vajpayee R, Trocme S,
Davitt III WF, et al. Safety and efficacy
of gatifloxacin 0.3% as compared with
ciprofloxacin 0.3% for the treatment of
acute bacterial corneal ulcers. Presented
at the Association for Research in Vision
and Ophthalmology (ARVO) meeting;
May 1, 2006: Fort Lauderdale.

Dr. Dunbar is the Director of Optometric Services and Optometry Residency Supervisor at the University of Miami’s Bascom Palmer Eye Institute. Dr. Dunbar can be reached at mdunbar@ med.miami.edu.