ARTICLE
Community Practice Patterns for Bacterial Corneal Ulcer Evaluation and Treatment Jennifer Park,
B.S.,
Kim M. Lee, B.S., Helen Zhou, B.S., Moriah Rabin, B.A., Kevin Jwo, William B. Burton, Ph.D., and David C. Gritz, M.D., M.P.H.
Objective: To examine current practice patterns in the management of bacterial keratitis among U.S. ophthalmologists and differences in the management and opinions between cornea specialists and non-cornea specialists. Methods: A questionnaire was distributed to randomly selected ophthalmologists in July 2011 using an online survey system. It inquired about the number of patients with corneal ulcers seen monthly, frequency of Gram staining and culturing corneal ulcers, maintenance of diagnostic supplies, opinions on when culturing is necessary for corneal ulcers, treatment preferences for different severities of bacterial corneal ulcers, and opinions regarding relative efficacy of fourth-generation fluoroquinolones and fortified broad-spectrum antibiotics. Results: One thousand seven hundred one surveys were distributed, and 486 (28.6%) surveys were returned. A minority of corneal ulcers was Gram stained (23.7%634.1%, mean6SD) or cultured (35.1%638.0%), but cornea specialists were more likely to perform both. The most popular antibiotic for the treatment of less severe ulcers was moxifloxacin (55.4%), and the most popular treatment of more severe ulcers was fortified broadspectrum antibiotics (62.7%). Cornea specialists were significantly more likely than non-cornea specialists to prescribe fortified antibiotics for more severe corneal ulcers (78.1% vs. 53.7%, P,0.0001). A greater number of cornea specialists stated that fourth-generation fluoroquinolones were less effective than fortified antibiotics for the treatment of more severe corneal ulcers (79.6% of cornea specialists vs. 60.9% of non-cornea specialists, P,0.001). Conclusions: Cornea specialists and non-cornea specialists manage bacterial keratitis differently, with cornea specialists more likely to perform diagnostic testing and prescribe fortified broad-spectrum antibiotics for severe bacterial keratitis. Additional prospective studies demonstrating visual outcomes after differential treatment of bacterial keratitis are needed.
From the Albert Einstein College of Medicine (J.P., K.M.L., M.R., K.J., W.B.B.), Bronx, NY; State University of New York Downstate College of Medicine (H.Z.), Brooklyn, NY; Departments of Ophthalmology and Visual Sciences (K.J., D.C.G.); and Epidemiology and Population Health (D.C.G.), Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, NY. Supported in part from a Research to Prevent Blindness Foundation unrestricted institutional grant award and Medical Student Summer Research Fellowship awards from the Albert Einstein College of Medicine. The authors have no conflicts of interest to disclose. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions this article on the journal’s Web site (www. eyeandcontactlensjournal.com). Address correspondence and reprint requests to David C. Gritz, M.D., M.P.H., Montefiore Medical Center, 111 East 210th Street, Bronx, NY 10467; e-mail: dgritz@montefiore.org Accepted June 10, 2014. DOI: 10.1097/ICL.0000000000000059
12
M.D.,
Key Words: Corneal ulcer—Bacterial keratitis—Fluoroquinolones— Fortified antibiotics—Compounding pharmacy. (Eye & Contact Lens 2015;41: 12–18)
T
he treatment of bacterial corneal ulcers continues to be controversial.1,2 Ophthalmologists have many available treatment options, as well as many differing opinions, on how to treat corneal ulcers. Thus, classic recommendations in corneal ulcer management do not necessarily agree with community practice patterns.2 Corneal ulcers have an annual incidence of 71,000 cases in the United States, or 27.6 per 100,000 people.3–5 With this number currently on the rise,5 it is critical to characterize and evaluate the practice patterns within the United States to reach a consensus on corneal ulcer management. In the past, it was recommended that all corneal ulcers be cultured and Gram stained before initiating treatment.1,6 Despite those recommendations, several studies conducted in the 1990s demonstrated that ophthalmologists did not obtain cultures and Gram stains before initiating therapy and instead took an empirical approach to treat corneal ulcers.7–9 In light of data demonstrating that empiric therapy was almost always adequate for resolution of nonsevere ulcers,10 that culture results usually did not alter patient management, and that many ophthalmologists did not follow the standard recommendations,7 guidelines for the management of bacterial keratitis were revised.6 Current guidelines from the Bacterial Keratitis Preferred Practice Pattern, published by the American Academy of Ophthalmology (AAO) and last updated in March 2013, state that smears and cultures only need to be performed in certain cases, such as corneal ulcers with large and/or deep infiltrates, those that are not responsive to broad-spectrum antibiotics, or those that possess atypical clinical features and/or history.11 These guidelines state that most community-acquired cases of bacterial keratitis can be managed with empiric therapy and do not require smears or cultures.11 At present, most cases of community-acquired bacterial keratitis do in fact receive empiric therapy without cultures or Gram stains.12 There is no universally recommended medication for the empiric treatment of corneal ulcers. This is because no single class of antibiotics covers all of the causative microorganisms of bacterial keratitis and because bacterial resistance patterns are in a state of flux.1,13 In general, monotherapy with a fluoroquinolone is usually sufficient for small or peripheral corneal ulcers, whereas dualfortified broad-spectrum antibiotics are recommended for larger or more severe corneal ulcers.11,13 Eye & Contact Lens Volume 41, Number 1, January 2015
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
Eye & Contact Lens Volume 41, Number 1, January 2015 The advent of topical fluoroquinolone antibiotics in the 1990s14 offered a way to treat corneal ulcers empirically with readily available prescription medication instead of using the more cumbersome fortified topical antibiotics. Fortified antibiotics have a number of shortcomings. For instance, they must be prepared by a pharmacist under sterile conditions close to the time at which the medication is dispensed because of the short shelf-life of fortified antibiotic drops, which last a maximum of 1 month when kept at 4°C but only for 1 week when stored at 24°C.15–17 The introduction of fourth-generation topical fluoroquinolones in 2003,14,18 which offered improved coverage of Gram-positive bacteria compared with the older generations, further encouraged the empiric treatment of corneal ulcers. Many ophthalmologists have since embraced the fourth-generation fluoroquinolones because of their convenience, broad spectrum of activity, and effective ocular penetration, but despite this growing preference for fluoroquinolones, ophthalmologists and cornea specialists are both ambivalent about the superiority of this new therapy.9 The goal of this study was to determine the clinical behavior and opinions of ophthalmologists in the West Coast, East Coast, and Midwest regions of the United States, as well as cornea specialists, to better describe current trends in bacterial keratitis management. In particular, we were interested in discovering opinions regarding current antibiotic therapy options, the accessibility of culture supplies to ophthalmologists, and the degree of compliance with current guidelines. Finally, we attempted to compare differences in behavior and opinions between general ophthalmologists and cornea specialists and among ophthalmologists from different geographic regions.
MATERIALS AND METHODS An electronic survey designed on SurveyMonkey (Palo Alto, CA), an online survey tool, was e-mailed in July 2011 to randomly selected ophthalmologists, whose contact information was obtained from the AAO directory (n=1,769). Included were ophthalmologists from the West Coast (California, n=637), Midwest (Illinois, n=214; Kansas, n=50; Missouri, n=100), and East Coast (New York, n=359; New Jersey, n=148; Connecticut, n=73). Additionally, randomly selected cornea specialists (n=188) on the Proctor Foundation list (n=31 recipients) and all ophthalmologists from the Cornea Society list (n=157) received the survey. The online survey tool automatically excluded e-mail addresses belonging to physicians who had previously opted out of its e-mail services. This produced 1,701 potential respondents (West Coast, n=614; Midwest, n=348; East Coast, n=560; cornea specialists, n=179). This study followed the tenets of the Declarations of Helsinki and was approved by the Institutional Review Board of the Albert Einstein College of Medicine. Recipients of the survey and their responses were kept anonymous. To increase response rates, nonrespondents to the first e-mail invite were then e-mailed a second reminder invitation 1 week later using the online survey tool’s automated reminder system. If recipients remained nonrespondent, a third attempt was made through mailing hard copies of the surveys to them, which recipients were asked to fax over on completion. Additionally, recipients who were initially unable to respond to the online survey and had requested to be sent surveys at a more convenient time were sent surveys accordingly. © 2014 Contact Lens Association of Ophthalmologists
Current Practices in Corneal Ulcer Management The survey featured a total of 19 questions (see Data, Supplemental Digital Content 1, http://links.lww.com/ICL/A20, to view the complete survey). The questionnaire initially asked recipients to characterize their identity with respect to the following features: location of practice, type of specialty, number of years in practice, and whether they treated corneal ulcers. In this study, a corneal ulcer was defined as an epithelial defect with an underlying stromal infiltrate caused by an infectious microorganism. Recipients who indicated that they did not treat ulcers or who immediately referred patients with corneal ulcers to cornea specialists were disqualified from the rest of the survey. The questionnaire asked for estimates on the following: the number of corneal ulcers treated per month, the percentage of ulcers cultured and/or Gram stained, and the percentage of ulcers resistant to initial therapy. Subsequently, the questionnaire inquired about the management of corneal ulcers. These questions included the following: whether recipients had access to a compounding pharmacy and culturing supplies, opinions on which therapeutic steps were necessary before initial treatment, and what drug therapy and dosing frequency would be used for two different clinical scenarios (small/ peripheral corneal ulcers vs. large/central/deep corneal ulcers). Finally, the survey queried recipients about their opinion regarding the efficacy of fourth-generation fluoroquinolones compared with dual broad-spectrum antibiotics for severe corneal ulcers and allowed for optional input on the management of corneal ulcers in free-text format. Responses were analyzed using descriptive statistics performed with SAS 9.3, and t tests and chi-square analyses were performed. P values less than 0.05 were considered statistically significant. The number of respondents (n values) listed in the Results section vary according to the number who answered a given question. Of note, all analyses of regional differences presented in this study include only those respondents who are not cornea specialists. There are two reasons for this: (1) we wanted to examine geographic differences in practice patterns among a population that did not have specialized training in corneal diseases and (2) the number of cornea specialists who were included in this study was too small to determine the presence of significant regional differences.
RESULTS Of the 1,701 surveys that were disseminated, 33 surveys were returned to the sender because of invalid e-mail addresses and 1,182 (69.5%) surveys were not completed because the recipients had retired from practice (1/1,701 or 0.06%) or declined to participate (1,181/1,701 or 69.4%). Of the 1,668 surveys that were sent to valid e-mail recipients, 486 (28.6%) were submitted. Of the 486 submitted surveys, 365 (75.1%) were fully complete and 121 (24.9%) were partially complete. The breakdown of response rates from each geographic region can be found in Table 1. Four respondents did not indicate their location. The differences in response rates of cornea specialists and non-cornea specialists across the 3 regions was not statistically significant (P=0.85).
Survey of Demographics Eighty ophthalmologists stated that they did not treat corneal ulcers in their clinics and were not included in the results. The 406 remaining respondents treated corneal ulcers. One hundred forty13
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
Eye & Contact Lens Volume 41, Number 1, January 2015
J. Park et al. TABLE 1.
Survey Response Rates by Specialty and Region Number (%)
Total responses Cornea specialists Non-cornea specialists General ophthalmologists Whether respondents treat corneal ulcers in practice Yes No Geographic area of practice, all respondents East Coast Midwest West Coast Geographic area of practice, non-cornea specialists East Coast Midwest West Coast
486 145 (29.8) 341 (70.2) 286 (58.8) 406 (83.5) 80 (16.5) 207 (43.0) 126 (26.1) 149 (31.0) 145 (42.8) 87 (25.7) 107 (31.6)
five participants (29.8%) indicated that they were cornea specialists; 341 (70.2%) indicated that they were not cornea specialists (Table 1). The mean number of corneal ulcers treated when pooled was on average 4.2 patients per month, with cornea specialists seeing almost 3 times the number of corneal ulcers per month than non-cornea specialists (7.167.9 vs. 2.563.1 patients per month [mean6SD], P=0.047) (Table 2). Cornea specialists were significantly more likely than non-cornea specialists to culture corneal ulcers (58.1%635.4% of the time vs. 21.9%632.9% of the time, P,0.0001). One physician stated that his/her decision to culture was based on the “1, 2, 3” rule, which classifies corneal ulcers into 2 categories (“rarely sight-threatening” and “potentially sightthreatening”) based on infiltrate size, location, and severity of anterior chamber reaction.19 Results for the frequency of Gram staining were similar to the frequency of culturing, with cornea specialists significantly more likely than non-cornea specialists to perform Gram stain on corneal ulcers (40.7%637.7% of the time vs. 13.9%627.5% of the time, P,0.0001). Finally, therapy-resistant corneal ulcers were uncommon among the surveyed physicians (8.0%612.5%), but cornea specialists saw a significantly greater percentage of initially therapy-resistant ulcers than non-cornea specialists (14.3%617.0% vs. 4.3%66.6%, P,0.0001).
Access to Supplies A total of 81.7% of respondents indicated that they maintained some kind of culturing or Gram staining supply, whereas 18.3% indicated that they did not maintain any kind of supply (Table 3). The most commonly maintained supplies for all ophthalmologists were bacterial culture transport swabs (72.5%), followed by bacterial culture plates, fungal culture media, thioglycollate broth, and Gram stain. Cornea specialists were significantly more likely to own all of the listed supplies than were non-cornea specialists (P,0.0001), with the exception of bacterial culture transport swabs, which were owned by most cornea and non-cornea specialists. Most respondents (91.0%) reported that they had access to a compounding pharmacy, with cornea specialists significantly more likely than non-cornea specialists to have access (98.6% vs. 86.4%, P=0.0004) (Table 3). There were no statistically significant regional differences in terms of access to compounding pharmacies. In terms of opinions regarding the need for culturing corneal ulcers, only 8.6% of respondents stated that culturing corneal 14
TABLE 2. Survey Responses Listing the Frequency of Treating Corneal Ulcers, Testing Performed Before Treating Corneal Ulcers, and Frequency of Corneal Ulcers That Were Initially Therapy-Resistant All Respondents a
Cornea Specialists
4.2 (65.8) 7.1 Average number of corneal ulcer patients seen monthly, n (6SD) 35.1 (638.0) 58.1 Average frequency of culturing corneal ulcers, percentage (6SD) 23.7 (634.1) 40.7 Average frequency of performing Gram stain, percentage (6SD) 8.0 (612.5) 14.3 Reported frequency of corneal ulcers that were initially therapy-resistant, percentage (6SD)
Non-Cornea Specialists a
(67.9)
2.5 (63.1)
a
P 0.047
(635.4) 21.9 (632.9) ,0.0001
(637.7) 13.9 (627.5) ,0.0001
(617.0)
4.3 (66.6)
,0.0001
Percentage of Respondents (Number of Respondents)
Number of corneal ulcer patients seen monthly 0–8 9–17 18+ Frequency of culturing corneal ulcers 0%–33% of the time 34%–67% of the time 68%–100% of the time Frequency of performing Gram stain 0%–33% of the time 34%–67% of the time 68%–100% of the time Frequency of corneal ulcers that were initially therapyresistant 0%–25% 26%–50% 51%–75% 76%–100%
All Respondents (n=378)
Cornea Specialists (n=139)
NonCornea Specialists (n=240)
P
85.7 (324) 10.6 (40) 3.7 (14)
70.5 (98) 21.6 (30) 7.9 (11)
94.6 (226) 4.2 (10) 1.3 (3)
,0.0001 ,0.0001 ,0.0001
60.2 (227) 13.3 (50) 26.5 (100)
31.2 (43) 21.7 (30) 47.1 (65)
77.0 (184) 8.4 (20) 14.6 (35)
,0.0001 ,0.0001 ,0.0001
72.2 (272) 11.4 (43) 16.5 (62)
52.2 (72) 18.1 (25) 29.7 (41)
83.7 (200) 7.5 (18) 8.8 (21)
,0.0001 ,0.0001 ,0.0001
93.1 6.1 0.3 0.5
84.9 12.9 0.7 1.4
97.9 (234) 2.1 (5) 0 (0) 0 (0)
,0.0001 ,0.0001 ,0.0001 ,0.0001
(352) (23) (1) (2)
(118) (18) (1) (2)
a Extreme outlier values were excluded when mean values and standard deviations were calculated for the average number of patients with corneal ulcer seen monthly. Excluded values: 2 cornea specialists who saw 400 patients per month, a non-cornea specialist who saw 300 patients per month, and a non-cornea specialist who saw 120 patients per month.
ulcers is always necessary (Table 4). Compared with non-cornea specialists, a greater number of cornea specialists stated that culturing is necessary for ulcers larger than 1.5·1.5 mm, regardless of the location (50.0% vs. 29.9% of respondents, P=0.0001). In terms of regional differences, ophthalmologists from the Midwest were significantly less likely than East Coast and West Coast physicians to culture corneal ulcers that were unresponsive to antibiotic therapy (P=0.005). Eye & Contact Lens Volume 41, Number 1, January 2015
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
Eye & Contact Lens Volume 41, Number 1, January 2015 TABLE 3.
Current Practices in Corneal Ulcer Management
Maintenance of Diagnostic Testing Supplies and Access to a Compounding Pharmacy Percentage of Respondents (n)
Some type of testing supply Bacterial culture transport swabs Bacterial culture plates (blood and chocolate agar) Fungal culture media Thioglycollate broth Gram stain No supplies
All Respondents (n=373)
Cornea Specialists (n=140)
Non-Cornea Specialists (n=236)
P
81.7 (304)
94.9 (129)
74.2 (175)
,0.0001
72.5 (269)
78.3 (108)
69.1 (161)
0.06
60.6 (226)
88.6 (124)
43.8 (102)
,0.0001
56.0 55.8 43.8 18.3
85.0 85.7 67.1 5.2
38.6 37.8 29.7 25.9
,0.0001 ,0.0001 ,0.0001 ,0.0001
(209) (208) (163) (68)
Access to a Compounding Pharmacy All respondents (n=376) Cornea specialists (n=140) Non-cornea specialists (n=236) East coast (n=100) Midwest (n=55) West Coast (n=80)
(119) (120) (94) (7)
(90) (88) (69) (61)
% of Respondents (n) 91.0 98.6 86.4 82.0 85.5 93.8
(342) (138) (204) (82) (47) (75)
P 0.0004 0.0004
fourth-generation fluoroquinolones (i.e., moxifloxacin and gatifloxacin) were the next most popular treatment choices for more serious ulcers, with 23.6% of all respondents selecting moxifloxacin and 16.0% selecting gatifloxacin. Non-cornea specialists were significantly more likely to choose gatifloxacin for treating large, central, or deep corneal ulcers compared with cornea specialists (21.0% vs. 7.4%, P=0.0005). Of note, cornea specialists were significantly more likely than non-cornea specialists to choose vancomycin (either alone or in combination with other antibiotics) to treat the more severe ulcer (11.0% vs. 0.4%, P,0.0001). In terms of regional differences, West Coast ophthalmologists were significantly more likely to prescribe gatifloxacin for the less severe ulcer (51.3% of West Coast physicians vs. 35.1% of East Coast physicians and 31.5% of Midwest physicians, P=0.03) and significantly less likely to prescribe moxifloxacin (P=0.01) (Table 5). East Coast physicians were significantly more likely to prescribe fortified antibiotics for the less severe ulcer (7.2% of East Coast physicians vs. 0% of Midwest and West Coast physicians, P=0.007). For the more severe ulcer, East Coast physicians were significantly more likely than Midwest or West Coast physicians to prescribe moxifloxacin (35.1% of East Coast respondents vs. 25.5% of Midwest and 16.3% of West Coast respondents, P=0.02).
Fourth-Generation Fluoroquinolones Versus Fortified Antibiotics
Clinical Scenarios The two clinical scenarios that were presented to recipients of the survey consisted of treatment choice for a less serious (small or peripheral ulcer) and a more serious (large, central, or deep) ulcer. For small or peripheral ulcers, 55.4% of all respondents stated that they would treat with moxifloxacin (Table 5). The second most popular treatment choice was gatifloxacin (38.7% of respondents). Significantly more cornea specialists would treat small or peripheral ulcers with 2 fortified broad-spectrum antibiotics compared with non-cornea specialists (8.8% vs. 3.0%, P=0.01). Approximately 62.7% of all respondents would treat the more severe ulcer (large, central, or deep) with 2 fortified broadspectrum antibiotics (i.e., cefazolin and gentamicin), with cornea specialists significantly more likely than non-cornea specialists to prescribe fortified antibiotics (78.1% vs. 53.7%, P,0.0001). The
TABLE 4.
When asked to compare the use of fourth-generation fluoroquinolones with fortified antibiotics for the treatment of more severe corneal ulcers, the majority of respondents (67.8%) believed that fourth-generation fluoroquinolones were less effective compared with fortified antibiotics (Table 5). Approximately 29.2% of all respondents believed that fourth-generation fluoroquinolones were equally as effective as fortified antibiotics. A significantly greater percentage of cornea specialists stated that fourth-generation fluoroquinolones were less effective than fortified broad-spectrum antibiotics, compared with non-cornea specialists (79.6% vs. 60.9%, P=0.001). A significantly greater percentage of non-cornea specialists stated that fourth-generation fluoroquinolones were equally as effective as fortified broad-spectrum antibiotics, compared with cornea specialists (35.6% vs. 18.2%, P=0.001). Only 3.0% of
Opinions on the Need for Culturing Corneal Ulcers Percentage of Respondents (n) Non-Cornea Specialists
When do you believe that culturing is necessary in the treatment of corneal ulcers? (Respondents were able to select multiple answers) Always Ulcers larger than 1.5·1.5 mm, regardless of the location Ulcers larger than 1.5·1.5 mm, within the central 5 mm of the cornea All ulcers that extend to middle to deep stroma Chronic ulcers Ulcers unresponsive to antibiotic therapy Unusual patient history or clinical features Never
© 2014 Contact Lens Association of Ophthalmologists
All Respondents (n=374)
Cornea Specialists (n=138)
Non-Cornea Specialists (n=236)
8.6 (32) 37.4 (139) 52.3 (195)
12.3 (17) 50.0 (69) 53.6 (74)
6.4 (15) 29.9 (70) 51.5 (121)
60.6 57.3 85.8 77.7
65.2 63.8 81.2 79.0
57.9 53.4 88.6 76.9
(226) (213) (321) (289) 0
(90) (88) (112) (109) 0
(136) (125) (209) (180) 0
P
East Coast (n=100)
Midwest (n=55)
West Coast (n=80)
P
5.2 (5) 12.7 (7) 3.8 (3) 0.0001 33.7 (33) 27.3 (15) 26.3 (21) 47.5 (47) 45.5 (25) 61.3 (49) 55.6 (55) 57.1 (56) 92.0 (92) 78.6 (77) 0
52.7 (29) 40.0 (22) 76.4 (42) 72.7 (40) 0
63.8 (51) 57.5 (46) 92.5 (74) 0.005 77.5 (62) 0
15
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
Eye & Contact Lens Volume 41, Number 1, January 2015
J. Park et al. TABLE 5.
Antibiotic Treatment Patterns for Small and Large Corneal Ulcers, Categorized by Whether the Respondent is a Cornea Specialist and by Geographic Region Percentage of Respondents (n) Non-Cornea Specialists All Respondents (n=370)
Small or peripheral corneal ulcer treatment Moxifloxacin Gatifloxacin Ciprofloxacin or ofloxacin Levofloxacin Standard concentrations of nonfluoroquinolone antibiotics (i.e., tobramycin or gentamicin) Two fortified broad-spectrum antibiotics (e.g. cefazolin-gentamicin) Besifloxacin Large, central, or deep corneal ulcer treatment Two fortified broad-spectrum antibiotics (i.e. cefazolin-gentamicin) Moxifloxacin Gatifloxacin Levofloxacin Vancomycin (either alone or in combination with other antibiotic(s)) Ciprofloxacin or ofloxacin Besifloxacin Opinions regarding perceived treatment efficacy: fourthgeneration fluoroquinolones are (less/equally as/more) effective compared with dualfortified broad-spectrum antibiotics Less Equally as More
55.4 38.7 14.1 8.1 7.0
(205) (142) (52) (30) (26)
Cornea Specialists (n=137)
56.2 37.0 9.6 10.9 8.1
(77) (50) (13) (15) (11)
Non-Cornea Specialists (n=233)
54.9 39.7 16.7 6.4 6.4
8.8 (12)
3.0 (7)
3.0 (11)
4.4 (6)
2.1 (5)
62.7 (232)
78.1 (107)
53.7 (125)
23.6 16.0 4.9 4.3
18.4 7.4 3.6 11.0
26.6 21.0 5.6 0.4
(25) (10) (5) (15)
0.01
(62) (49) (13) (1)
,0.0001
1.5 (2) 3.7 (5)
5.2 (12) 2.6 (6)
67.8 (251) 29.2 (108) 3.0 (11)
79.6 (109) 18.2 (25) 2.2 (3)
60.9 (142) 35.6 (83) 3.4 (8)
DISCUSSION This survey study aimed to determine current practices regarding the diagnosis and treatment of patients with bacterial keratitis by ophthalmologists across the United States, as well as to compare the practice patterns of cornea specialists and noncornea specialists. Echoing the findings of previous questionnaire-based studies,7 this study found that Gram stains and cultures of corneal ulcers were performed in a minority of cases (Gram stains were performed on average 23.7%634.1% of the time and cultures performed on average 35.1%638.0% of the time). Similar to the questionnaire study of McLeod et al.,8 our respondents also believed that culturing was necessary only for more severe corneal ulcers, and akin to the survey study of Hsu et al.,9 we also found that cornea specialists were significantly more likely to Gram stain and culture corneal ulcers compared with non-cornea specialists. However, this study was unique because it compared the opinions of both cornea specialists and non-cornea specialists in the treatment choices for two different
,0.0001
0.0005
3.8 (14) 3.0 (11)
respondents believed that fourth-generation fluoroquinolones were more effective than fortified antibiotics.
16
(128) (92) (39) (15) (15)
5.1 (19)
(87) (59) (18) (16)
P
0.001 0.001 0.001
East Coast (n=101)
Midwest (n=56)
61.9 35.1 14.4 10.3 6.2
61.8 31.5 14.6 5.5 10.9
(60) (34) (14) (10) (6)
7.2 (7)
(34) (17) (8) (3) (6)
0 (0)
West Coast (n=80)
41.3 51.3 21.3 2.5 3.8
(33) (41) (17) (2) (3)
0 (0)
3.0 (3)
1.8 (1)
1.3 (1)
46.4 (45)
56.4 (31)
60.0 (48)
35.1 (34) 19.6 (19) 7.2 (7) 0 (0)
25.5 18.2 7.3 1.8
16.3 (13) 25.0 (20) 2.5 (2) 0 (0)
(14) (10) (4) (1)
8.3 (8) 3.0 (3)
3.6 (2) 1.8 (1)
2.5 (2) 2.5 (2)
58.8 (57) 37.1 (36) 4.1 (4)
61.8 (34) 36.4 (20) 1.8 (1)
63.8 (51) 32.5 (26) 3.8 (3)
P
0.01 0.03
0.007
0.02
severities of bacterial keratitis. Indeed, this study found that cornea specialists were significantly more likely than non-cornea specialists to treat more severe corneal ulcers with fortified broad-spectrum antibiotics (78.1% vs. 53.7%, P,0.0001). This study was also notable because it compared regional differences in practices and opinions on bacterial keratitis by sampling a variety of states in three regions of the United States. The cornea specialists in this study saw a greater number of patients with corneal ulcer on a monthly basis compared with noncornea specialists, and cornea specialists were more likely than non-cornea specialists to culture and Gram stain corneal ulcers before treatment (Table 2). Cornea specialists also saw a greater number of initially therapy-resistant ulcers than non-cornea specialists, likely because the more severe and resistant cases of bacterial keratitis are referred to cornea specialists.6 Additionally, cornea specialists were significantly more likely than non-cornea specialists to maintain a number of diagnostic testing supplies, including bacterial culture plates, fungal culture media, thioglycollate broth, and Gram stain supplies. Cornea specialists also were more likely than non-cornea specialists to have access to a compounding pharmacy (98.6% vs. 86.4%, P=0.0004). These results seem logical because cornea specialists see more severe cases of Eye & Contact Lens Volume 41, Number 1, January 2015
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
Eye & Contact Lens Volume 41, Number 1, January 2015 bacterial keratitis and therefore need to perform more diagnostic testing and prescribe more fortified antibiotics. This study also shows that there is a significant difference between cornea specialists and non-cornea specialists in the culturing of corneal ulcers, with cornea specialists more likely than non-cornea specialists to culture ulcers that are larger than 1.5·1.5 mm, regardless of the location (50.0% vs. 29.9%, P=0.0001). This may be due in part to non-cornea specialists immediately referring complicated corneal ulcers to corneal specialists. Several of the physicians indicated in the “comments” section of the survey that there are limits on obtaining Gram stains and fortified antibiotics in a timely manner for patients. Many of the general ophthalmologists commented that for those patients with large and/or deep ulcers, ulcers that do not improve in 24 hours, or for facilities that do not have access to culture media, referrals to cornea specialists are made. The current guidelines published by the AAO in 2013 state that either single-drug therapy with a fluoroquinolone or combination fortified broad-spectrum antibiotics (e.g., cefazolin-tobramycin) may be used as empiric treatment of bacterial keratitis of all severities.11 These guidelines were made in light of three randomized controlled trials that demonstrated that moxifloxacin monotherapy is just as effective for the treatment of bacterial keratitis compared with combination therapy with fortified antibiotics (e.g., cefazolin-tobramycin).20–22 For instance, the randomized controlled trial of Constantinou et al.22 revealed that moxifloxacin, ofloxacin, and fortified cefazolin-tobramycin were all equally effective in the treatment of severe bacterial keratitis, with 34% to 41% of the study participants having ulcers severe enough to cause hypopyon. The randomized controlled trial of Sharma et al.20 showed there was no difference in the number of ulcers that completely healed within 3 months in 224 patients with bacterial keratitis randomized to either moxifloxacin monotherapy or cefazolin-tobramycin–fortified antibiotic therapy. Furthermore, moxifloxacin monotherapy has several practical advantages over dual-fortified antibiotic therapy, including ease of obtaining the medication, increased compliance, and decreased ocular toxicity.20,22 However, the AAO guidelines do state that combination fortified broad-spectrum antibiotic therapy should be considered for severe infections (i.e., central or severe ulcers or infiltrates .2 mm that are threatening the visual axis) and for ulcers unresponsive to initial treatment.11,13 This study demonstrated that more than 56.2% of cornea specialists and 54.9% of general ophthalmologists use moxifloxacin to treat small or peripheral corneal ulcers. These findings are similar to that of Hsu et al.,9 who demonstrated that the fourth-generation fluoroquinolones were the most popular antibiotic choice for bacterial keratitis.9 This is not surprising because moxifloxacin is known to be effective against a broad spectrum of bacteria, including Grampositive and Gram-negative species, and is easy for the patient to obtain. Cornea specialists were significantly more likely than noncornea specialists to treat small or peripheral corneal ulcers with 2 fortified broad-spectrum antibiotics (8.8% vs. 3.0%, P=0.01). This is likely because cornea specialists tend to overtreat rather than undertreat corneal ulcers because they have had exposure to more severe cases of bacterial keratitis and therefore likely want to take precautions to prevent corneal ulcers from deteriorating. With the more severe ulcer (large, central, or deep ulcers), cornea specialists are significantly more likely than general ophthalmologists to treat with 2 fortified broad-spectrum anti© 2014 Contact Lens Association of Ophthalmologists
Current Practices in Corneal Ulcer Management biotics (78.1% of cornea specialists vs. 53.7% of general ophthalmologists, P,0.0001). Non-cornea specialists were significantly more likely than cornea specialists to prescribe gatifloxacin for more severe corneal ulcers (21.0% vs. 7.4%, P=0.0005), perhaps because non-cornea specialists are less likely to have access to compounding pharmacies. Sixteen respondents stated that they would treat the more serious ulcer with vancomycin (either alone or in combination with other antibiotics), with cornea specialists more likely to prescribe vancomycin than non-cornea specialists (11% vs. 0.4%, P,0.0001). This may be because cornea specialists are more aware of the increasing incidence of methicillin-resistant Staphylococcus aureus (MRSA) bacterial keratitis and are thus providing MRSA coverage with vancomycin. Additionally, this study found a difference in moxifloxacin and gatifloxacin prescription rates across the West Coast, Midwest, and East Coast regions of the United States, with a significantly lower rate of moxifloxacin use for small corneal ulcers among the surveyed West Coast ophthalmologists and a significantly higher rate of gatifloxacin use among surveyed West Coast ophthalmologists for small corneal ulcers. It is unclear why the West Coast ophthalmologists who responded to the survey prefer one fourthgeneration fluoroquinolone over the other. Increasing popularity of the fourth-generation fluoroquinolones has contributed to the selection of resistant bacterial strains.23 Knowledge of increasing resistance to fluoroquinolones may be the reason why fortified antibiotics are still perceived as more effective for the treatment of severe corneal ulcers by both the cornea and non-cornea specialists surveyed in this study (Table 5). Specifically, this study showed that 67.8% of respondents believed that fourth-generation fluoroquinolones are less effective than dual-fortified broad-spectrum antibiotics in the treatment of more severe corneal ulcers. Antibiotic resistance is a major concern, with multiple studies indicating that repeated use of antibiotics selects for resistant strains of bacteria.24 For instance, Ray et al.24 demonstrated that organisms from eyes with bacterial keratitis that were pretreated with a fluoroquinolone (second-, third-, or fourth-generation) before presentation had a higher moxifloxacin minimum inhibitory concentration than organisms from eyes with bacterial keratitis that were not exposed to fluoroquinolones before presentation. These studies emphasize the need to avoid indiscriminate use of fluoroquinolones to prevent selection of resistant bacterial strains. A few respondents selected besifloxacin as their initial choice for treating both small and large corneal ulcers (3.0% of all respondents). Besifloxacin is a new fluoroquinolone that has shown greater potency than moxifloxacin and ciprofloxacin against Staphylococci isolated from various ocular sites and will likely become an important player in the battle against organisms resistant to existing fluoroquinolones.23,25 Additionally, besifloxacin is effective against MRSA, unlike most of the other fluoroquinolones, which lack MRSA activity.23,26 Efficacy against MRSA is a critical consideration because the incidence of serious ocular infections caused by MRSA has been increasing.27,28 Additionally, besifloxacin was created specifically as an ocular fluoroquinolone, meaning that it has never been labeled for use in infections in other parts of the body. This reduces the likelihood of organisms resistant to besifloxacin. At present, besifloxacin is only FDA-approved for the treatment of bacterial conjunctivitis, but off-label uses for the treatment of bacterial keratitis in animal models have demonstrated its efficacy.11,29 17
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
Eye & Contact Lens Volume 41, Number 1, January 2015
J. Park et al. In general, judicious use of potent antibiotics is advisable to prevent the development of further antibiotic resistance. Although mandatory culturing of corneal samples and testing of antibiotic susceptibilities in patients with bacterial keratitis is no longer stated in current guidelines, many clinicians recognize the importance of these data on an epidemiologic level. These tests would allow physicians to monitor antibiotic resistance and detect shifting trends in the organisms responsible for bacterial keratitis.2,30 Numerous studies have demonstrated the rapidly evolving pattern of antibiotic resistance,18 which further emphasizes the need to periodically monitor the epidemiology of bacterial corneal ulcers and their susceptibility patterns. Strengths of this study include sampling of ophthalmologists from a wide geographic area and employment of an easy-to-use online survey system that, in theory, encourages a greater number of responses. The benefits of using an online survey system compared with paper survey methods include faster response rates, lower costs, and decreased risk of misplacing the survey. However, one issue with distributing through an Internet-based survey system is that ophthalmologists may be unfamiliar with or may overlook e-mailed surveys, especially if there are no incentives for completing it. Another limitation of the study is selection bias, which is present in any type of study in which participation is optional. Respondents may have been more willing to answer the questionnaire because they followed standards of care, which may have skewed the results toward those that are in line with the AAO guidelines.7 Also, approximately 16% of the surveyed cornea specialists were members of or alumni of the Proctor Foundation, which could contribute to possible selection bias in this study. Despite the ease of answering online surveys, this study had a response rate of 28.6%, which may make the study’s findings less representative of the target population. Additionally, statistical tests were only performed when there seemed to be an obvious difference in the tabulations. Because multiple statistical tests were performed, some of the P-values below 0.05 may be due to chance. Finally, the lack of follow-up information on patients (e.g., ultimate visual acuity, need for penetrating keratoplasty) prevents us from drawing conclusions about the consequences of treating corneal ulcers with different antibiotics. Future studies should investigate patient outcomes for corneal ulcers after treatment with fluoroquinolones versus fortified broadspectrum antibiotics, especially because both fluoroquinolone resistance and the incidence of MRSA bacterial keratitis are on the rise. Trends in antibiotic resistance over time and in different geographic areas should also continue to be monitored to better understand resistance patterns and causative organisms. A study investigating regional differences in bacterial keratitis management patterns among cornea specialists may also be informative. REFERENCES 1. McDonnell PJ. Empirical or culture-guided therapy for microbial keratitis? A plea for data. Arch Ophthalmol 1996;114:84–87. 2. Charukamnoetkanok P, Pineda R II. Controversies in management of bacterial keratitis. Int Ophthalmol Clin 2005;45:199–210. 3. Pepose JS, Wilhelmus KR. Divergent approaches to the management of corneal ulcers. Am J Ophthalmol 1992;114:630–632. 4. Erie JC, Nevitt MP, Hodge DO, et al. Incidence of ulcerative keratitis in a defined population from 1950 through 1988. Arch Ophthalmol 1993;111: 1665–1671.
18
5. Jeng BH, Gritz DC, Kumar AB, et al. Epidemiology of ulcerative keratitis in Northern California. Arch Ophthalmol 2010;128:1022–1028. 6. Rodman RC, Spisak S, Sugar A, et al. The utility of culturing corneal ulcers in a tertiary referral center versus a general ophthalmology clinic. Ophthalmology 1997;104:1897–1901. 7. McDonnell PJ, Nobe J, Gauderman WJ, et al. Community care of corneal ulcers. Am J Ophthalmol 1992;114:531–538. 8. McLeod SD, DeBacker CM, Viana MA. Differential care of corneal ulcers in the community based on apparent severity. Ophthalmology 1996;103: 479–484. 9. Hsu HY, Nacke R, Song JC, et al. Community opinions in the management of corneal ulcers and ophthalmic antibiotics: A survey of 4 states. Eye Contact Lens 2010;36:195–200. 10. McLeod SD, Kolahdouz-Isfahani A, Rostamian K, et al. The role of smears, cultures, and antibiotic sensitivity testing in the management of suspected infectious keratitis. Ophthalmology 1996;103:23–28. 11. American Academy of Ophthalmology Cornea/External Disease Panel. Preferred Practice Pattern Guidelines. Bacterial Keratitis. San Francisco, CA, American Academy of Ophthalmology, 2013. Available at: http:// www.aao.org/ppp. Accessed December 24, 2013. 12. Hanet MS, Jamart J, Chaves AP. Fluoroquinolones or fortified antibiotics for treating bacterial keratitis: Systematic review and meta-analysis of comparative studies. Can J Ophthalmol 2012;47:493–499. 13. Miller D. Pharmacological treatment for infectious corneal ulcers. Expert Opin Pharmacother 2013;14:543–560. 14. Mah FS. Fourth-generation fluoroquinolones: New topical agents in the war on ocular bacterial infections. Curr Opin Ophthalmol 2004;15:316–320. 15. Arici MK, Sumer Z, Guler C, et al. In vitro potency and stability of fortified ophthalmic antibiotics. Aust N Z J Ophthalmol 1999;27:426–430. 16. Prabhasawat P, Chotikavanich S, Leelaporn A. Sterility of non-preservative eye drops. J Med Assoc Thai 2005;88(suppl 9):S6–S10. 17. Gangopadhyay N, Daniell M, Weih L, et al. Fluoroquinolone and fortified antibiotics for treating bacterial corneal ulcers. Br J Ophthalmol 2000;84: 378–384. 18. Afshari NA, Ma JJ, Duncan SM, et al. Trends in resistance to ciprofloxacin, cefazolin, and gentamicin in the treatment of bacterial keratitis. J Ocul Pharmacol Ther 2008;24:217–223. 19. Vital MC, Belloso M, Prager TC, et al. Classifying the severity of corneal ulcers by using the “1, 2, 3” rule. Cornea 2007;26:16–20. 20. Sharma N, Goel M, Bansal S, et al. Evaluation of moxifloxacin 0.5% in treatment of nonperforated bacterial corneal ulcers: A randomized controlled trial. Ophthalmology 2013;120:1173–1178. 21. Shah VM, Tandon R, Satpathy G, et al. Randomized clinical study for comparative evaluation of fourth-generation fluoroquinolones with the combination of fortified antibiotics in the treatment of bacterial corneal ulcers. Cornea 2010;29:751–757. 22. Constantinou M, Daniell M, Snibson GR, et al. Clinical efficacy of moxifloxacin in the treatment of bacterial keratitis: A randomized clinical trial. Ophthalmology 2007;114:1622–1629. 23. Haas W, Pillar CM, Torres M, et al. Monitoring antibiotic resistance in ocular microorganisms: Results from the Antibiotic Resistance Monitoring in Ocular micRorganisms (ARMOR) 2009 surveillance study. Am J Ophthalmol 2011;152:567–574. 24. Ray KJ, Prajna L, Srinivasan M, et al. Fluoroquinolone treatment and susceptibility of isolates from bacterial keratitis. JAMA Ophthalmol 2013;131: 310–313. 25. Miller D, Chang JS, Flynn HW, et al. Comparative in vitro susceptibility of besifloxacin and seven comparators against ciprofloxacin- and methicillinsusceptible/nonsusceptible staphylococci. J Ocul Pharmacol Ther 2013;29: 339–344. 26. Gadaria-Rathod N, Fernandez KB, Kheirkhah A, et al. New insights into infectious keratitis. Int Ophthalmol Clin 2013;53:163–172. 27. Hsiao CH, Chuang CC, Tan HY, et al. Methicillin-resistant Staphylococcus aureus ocular infection: A 10-year hospital-based study. Ophthalmology 2012;119:522–527. 28. Asbell PA, Sahm DF, Shaw M, et al. Increasing prevalence of methicillin resistance in serious ocular infections caused by Staphylococcus aureus in the United States: 2000 to 2005. J Cataract Refract Surg 2008;34:814–818. 29. O’Brien TP. Besifloxacin ophthalmic suspension, 0.6%: A novel topical fluoroquinolone for bacterial conjunctivitis. Adv Ther 2012;29:473–490. 30. Wilhelmus KR, Schlech BA. Clinical and epidemiological advantages of culturing bacterial keratitis. Cornea 2004;23:38–42.
Eye & Contact Lens Volume 41, Number 1, January 2015
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
Community Practice Patterns for Bacterial Corneal Ulcer Evaluation and Treatment Jennifer Park,
B.S.,
Kim M. Lee, B.S., Helen Zhou, B.S., Moriah Rabin, B.A., Kevin Jwo, William B. Burton, Ph.D., and David C. Gritz, M.D., M.P.H.
Objective: To examine current practice patterns in the management of bacterial keratitis among U.S. ophthalmologists and differences in the management and opinions between cornea specialists and non-cornea specialists. Methods: A questionnaire was distributed to randomly selected ophthalmologists in July 2011 using an online survey system. It inquired about the number of patients with corneal ulcers seen monthly, frequency of Gram staining and culturing corneal ulcers, maintenance of diagnostic supplies, opinions on when culturing is necessary for corneal ulcers, treatment preferences for different severities of bacterial corneal ulcers, and opinions regarding relative efficacy of fourth-generation fluoroquinolones and fortified broad-spectrum antibiotics. Results: One thousand seven hundred one surveys were distributed, and 486 (28.6%) surveys were returned. A minority of corneal ulcers was Gram stained (23.7%634.1%, mean6SD) or cultured (35.1%638.0%), but cornea specialists were more likely to perform both. The most popular antibiotic for the treatment of less severe ulcers was moxifloxacin (55.4%), and the most popular treatment of more severe ulcers was fortified broadspectrum antibiotics (62.7%). Cornea specialists were significantly more likely than non-cornea specialists to prescribe fortified antibiotics for more severe corneal ulcers (78.1% vs. 53.7%, P,0.0001). A greater number of cornea specialists stated that fourth-generation fluoroquinolones were less effective than fortified antibiotics for the treatment of more severe corneal ulcers (79.6% of cornea specialists vs. 60.9% of non-cornea specialists, P,0.001). Conclusions: Cornea specialists and non-cornea specialists manage bacterial keratitis differently, with cornea specialists more likely to perform diagnostic testing and prescribe fortified broad-spectrum antibiotics for severe bacterial keratitis. Additional prospective studies demonstrating visual outcomes after differential treatment of bacterial keratitis are needed.
From the Albert Einstein College of Medicine (J.P., K.M.L., M.R., K.J., W.B.B.), Bronx, NY; State University of New York Downstate College of Medicine (H.Z.), Brooklyn, NY; Departments of Ophthalmology and Visual Sciences (K.J., D.C.G.); and Epidemiology and Population Health (D.C.G.), Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, NY. Supported in part from a Research to Prevent Blindness Foundation unrestricted institutional grant award and Medical Student Summer Research Fellowship awards from the Albert Einstein College of Medicine. The authors have no conflicts of interest to disclose. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions this article on the journal’s Web site (www. eyeandcontactlensjournal.com). Address correspondence and reprint requests to David C. Gritz, M.D., M.P.H., Montefiore Medical Center, 111 East 210th Street, Bronx, NY 10467; e-mail: dgritz@montefiore.org Accepted June 10, 2014. DOI: 10.1097/ICL.0000000000000059
12
M.D.,
Key Words: Corneal ulcer—Bacterial keratitis—Fluoroquinolones— Fortified antibiotics—Compounding pharmacy. (Eye & Contact Lens 2015;41: 12–18)
T
he treatment of bacterial corneal ulcers continues to be controversial.1,2 Ophthalmologists have many available treatment options, as well as many differing opinions, on how to treat corneal ulcers. Thus, classic recommendations in corneal ulcer management do not necessarily agree with community practice patterns.2 Corneal ulcers have an annual incidence of 71,000 cases in the United States, or 27.6 per 100,000 people.3–5 With this number currently on the rise,5 it is critical to characterize and evaluate the practice patterns within the United States to reach a consensus on corneal ulcer management. In the past, it was recommended that all corneal ulcers be cultured and Gram stained before initiating treatment.1,6 Despite those recommendations, several studies conducted in the 1990s demonstrated that ophthalmologists did not obtain cultures and Gram stains before initiating therapy and instead took an empirical approach to treat corneal ulcers.7–9 In light of data demonstrating that empiric therapy was almost always adequate for resolution of nonsevere ulcers,10 that culture results usually did not alter patient management, and that many ophthalmologists did not follow the standard recommendations,7 guidelines for the management of bacterial keratitis were revised.6 Current guidelines from the Bacterial Keratitis Preferred Practice Pattern, published by the American Academy of Ophthalmology (AAO) and last updated in March 2013, state that smears and cultures only need to be performed in certain cases, such as corneal ulcers with large and/or deep infiltrates, those that are not responsive to broad-spectrum antibiotics, or those that possess atypical clinical features and/or history.11 These guidelines state that most community-acquired cases of bacterial keratitis can be managed with empiric therapy and do not require smears or cultures.11 At present, most cases of community-acquired bacterial keratitis do in fact receive empiric therapy without cultures or Gram stains.12 There is no universally recommended medication for the empiric treatment of corneal ulcers. This is because no single class of antibiotics covers all of the causative microorganisms of bacterial keratitis and because bacterial resistance patterns are in a state of flux.1,13 In general, monotherapy with a fluoroquinolone is usually sufficient for small or peripheral corneal ulcers, whereas dualfortified broad-spectrum antibiotics are recommended for larger or more severe corneal ulcers.11,13 Eye & Contact Lens Volume 41, Number 1, January 2015
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
Eye & Contact Lens Volume 41, Number 1, January 2015 The advent of topical fluoroquinolone antibiotics in the 1990s14 offered a way to treat corneal ulcers empirically with readily available prescription medication instead of using the more cumbersome fortified topical antibiotics. Fortified antibiotics have a number of shortcomings. For instance, they must be prepared by a pharmacist under sterile conditions close to the time at which the medication is dispensed because of the short shelf-life of fortified antibiotic drops, which last a maximum of 1 month when kept at 4°C but only for 1 week when stored at 24°C.15–17 The introduction of fourth-generation topical fluoroquinolones in 2003,14,18 which offered improved coverage of Gram-positive bacteria compared with the older generations, further encouraged the empiric treatment of corneal ulcers. Many ophthalmologists have since embraced the fourth-generation fluoroquinolones because of their convenience, broad spectrum of activity, and effective ocular penetration, but despite this growing preference for fluoroquinolones, ophthalmologists and cornea specialists are both ambivalent about the superiority of this new therapy.9 The goal of this study was to determine the clinical behavior and opinions of ophthalmologists in the West Coast, East Coast, and Midwest regions of the United States, as well as cornea specialists, to better describe current trends in bacterial keratitis management. In particular, we were interested in discovering opinions regarding current antibiotic therapy options, the accessibility of culture supplies to ophthalmologists, and the degree of compliance with current guidelines. Finally, we attempted to compare differences in behavior and opinions between general ophthalmologists and cornea specialists and among ophthalmologists from different geographic regions.
MATERIALS AND METHODS An electronic survey designed on SurveyMonkey (Palo Alto, CA), an online survey tool, was e-mailed in July 2011 to randomly selected ophthalmologists, whose contact information was obtained from the AAO directory (n=1,769). Included were ophthalmologists from the West Coast (California, n=637), Midwest (Illinois, n=214; Kansas, n=50; Missouri, n=100), and East Coast (New York, n=359; New Jersey, n=148; Connecticut, n=73). Additionally, randomly selected cornea specialists (n=188) on the Proctor Foundation list (n=31 recipients) and all ophthalmologists from the Cornea Society list (n=157) received the survey. The online survey tool automatically excluded e-mail addresses belonging to physicians who had previously opted out of its e-mail services. This produced 1,701 potential respondents (West Coast, n=614; Midwest, n=348; East Coast, n=560; cornea specialists, n=179). This study followed the tenets of the Declarations of Helsinki and was approved by the Institutional Review Board of the Albert Einstein College of Medicine. Recipients of the survey and their responses were kept anonymous. To increase response rates, nonrespondents to the first e-mail invite were then e-mailed a second reminder invitation 1 week later using the online survey tool’s automated reminder system. If recipients remained nonrespondent, a third attempt was made through mailing hard copies of the surveys to them, which recipients were asked to fax over on completion. Additionally, recipients who were initially unable to respond to the online survey and had requested to be sent surveys at a more convenient time were sent surveys accordingly. © 2014 Contact Lens Association of Ophthalmologists
Current Practices in Corneal Ulcer Management The survey featured a total of 19 questions (see Data, Supplemental Digital Content 1, http://links.lww.com/ICL/A20, to view the complete survey). The questionnaire initially asked recipients to characterize their identity with respect to the following features: location of practice, type of specialty, number of years in practice, and whether they treated corneal ulcers. In this study, a corneal ulcer was defined as an epithelial defect with an underlying stromal infiltrate caused by an infectious microorganism. Recipients who indicated that they did not treat ulcers or who immediately referred patients with corneal ulcers to cornea specialists were disqualified from the rest of the survey. The questionnaire asked for estimates on the following: the number of corneal ulcers treated per month, the percentage of ulcers cultured and/or Gram stained, and the percentage of ulcers resistant to initial therapy. Subsequently, the questionnaire inquired about the management of corneal ulcers. These questions included the following: whether recipients had access to a compounding pharmacy and culturing supplies, opinions on which therapeutic steps were necessary before initial treatment, and what drug therapy and dosing frequency would be used for two different clinical scenarios (small/ peripheral corneal ulcers vs. large/central/deep corneal ulcers). Finally, the survey queried recipients about their opinion regarding the efficacy of fourth-generation fluoroquinolones compared with dual broad-spectrum antibiotics for severe corneal ulcers and allowed for optional input on the management of corneal ulcers in free-text format. Responses were analyzed using descriptive statistics performed with SAS 9.3, and t tests and chi-square analyses were performed. P values less than 0.05 were considered statistically significant. The number of respondents (n values) listed in the Results section vary according to the number who answered a given question. Of note, all analyses of regional differences presented in this study include only those respondents who are not cornea specialists. There are two reasons for this: (1) we wanted to examine geographic differences in practice patterns among a population that did not have specialized training in corneal diseases and (2) the number of cornea specialists who were included in this study was too small to determine the presence of significant regional differences.
RESULTS Of the 1,701 surveys that were disseminated, 33 surveys were returned to the sender because of invalid e-mail addresses and 1,182 (69.5%) surveys were not completed because the recipients had retired from practice (1/1,701 or 0.06%) or declined to participate (1,181/1,701 or 69.4%). Of the 1,668 surveys that were sent to valid e-mail recipients, 486 (28.6%) were submitted. Of the 486 submitted surveys, 365 (75.1%) were fully complete and 121 (24.9%) were partially complete. The breakdown of response rates from each geographic region can be found in Table 1. Four respondents did not indicate their location. The differences in response rates of cornea specialists and non-cornea specialists across the 3 regions was not statistically significant (P=0.85).
Survey of Demographics Eighty ophthalmologists stated that they did not treat corneal ulcers in their clinics and were not included in the results. The 406 remaining respondents treated corneal ulcers. One hundred forty13
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
Eye & Contact Lens Volume 41, Number 1, January 2015
J. Park et al. TABLE 1.
Survey Response Rates by Specialty and Region Number (%)
Total responses Cornea specialists Non-cornea specialists General ophthalmologists Whether respondents treat corneal ulcers in practice Yes No Geographic area of practice, all respondents East Coast Midwest West Coast Geographic area of practice, non-cornea specialists East Coast Midwest West Coast
486 145 (29.8) 341 (70.2) 286 (58.8) 406 (83.5) 80 (16.5) 207 (43.0) 126 (26.1) 149 (31.0) 145 (42.8) 87 (25.7) 107 (31.6)
five participants (29.8%) indicated that they were cornea specialists; 341 (70.2%) indicated that they were not cornea specialists (Table 1). The mean number of corneal ulcers treated when pooled was on average 4.2 patients per month, with cornea specialists seeing almost 3 times the number of corneal ulcers per month than non-cornea specialists (7.167.9 vs. 2.563.1 patients per month [mean6SD], P=0.047) (Table 2). Cornea specialists were significantly more likely than non-cornea specialists to culture corneal ulcers (58.1%635.4% of the time vs. 21.9%632.9% of the time, P,0.0001). One physician stated that his/her decision to culture was based on the “1, 2, 3” rule, which classifies corneal ulcers into 2 categories (“rarely sight-threatening” and “potentially sightthreatening”) based on infiltrate size, location, and severity of anterior chamber reaction.19 Results for the frequency of Gram staining were similar to the frequency of culturing, with cornea specialists significantly more likely than non-cornea specialists to perform Gram stain on corneal ulcers (40.7%637.7% of the time vs. 13.9%627.5% of the time, P,0.0001). Finally, therapy-resistant corneal ulcers were uncommon among the surveyed physicians (8.0%612.5%), but cornea specialists saw a significantly greater percentage of initially therapy-resistant ulcers than non-cornea specialists (14.3%617.0% vs. 4.3%66.6%, P,0.0001).
Access to Supplies A total of 81.7% of respondents indicated that they maintained some kind of culturing or Gram staining supply, whereas 18.3% indicated that they did not maintain any kind of supply (Table 3). The most commonly maintained supplies for all ophthalmologists were bacterial culture transport swabs (72.5%), followed by bacterial culture plates, fungal culture media, thioglycollate broth, and Gram stain. Cornea specialists were significantly more likely to own all of the listed supplies than were non-cornea specialists (P,0.0001), with the exception of bacterial culture transport swabs, which were owned by most cornea and non-cornea specialists. Most respondents (91.0%) reported that they had access to a compounding pharmacy, with cornea specialists significantly more likely than non-cornea specialists to have access (98.6% vs. 86.4%, P=0.0004) (Table 3). There were no statistically significant regional differences in terms of access to compounding pharmacies. In terms of opinions regarding the need for culturing corneal ulcers, only 8.6% of respondents stated that culturing corneal 14
TABLE 2. Survey Responses Listing the Frequency of Treating Corneal Ulcers, Testing Performed Before Treating Corneal Ulcers, and Frequency of Corneal Ulcers That Were Initially Therapy-Resistant All Respondents a
Cornea Specialists
4.2 (65.8) 7.1 Average number of corneal ulcer patients seen monthly, n (6SD) 35.1 (638.0) 58.1 Average frequency of culturing corneal ulcers, percentage (6SD) 23.7 (634.1) 40.7 Average frequency of performing Gram stain, percentage (6SD) 8.0 (612.5) 14.3 Reported frequency of corneal ulcers that were initially therapy-resistant, percentage (6SD)
Non-Cornea Specialists a
(67.9)
2.5 (63.1)
a
P 0.047
(635.4) 21.9 (632.9) ,0.0001
(637.7) 13.9 (627.5) ,0.0001
(617.0)
4.3 (66.6)
,0.0001
Percentage of Respondents (Number of Respondents)
Number of corneal ulcer patients seen monthly 0–8 9–17 18+ Frequency of culturing corneal ulcers 0%–33% of the time 34%–67% of the time 68%–100% of the time Frequency of performing Gram stain 0%–33% of the time 34%–67% of the time 68%–100% of the time Frequency of corneal ulcers that were initially therapyresistant 0%–25% 26%–50% 51%–75% 76%–100%
All Respondents (n=378)
Cornea Specialists (n=139)
NonCornea Specialists (n=240)
P
85.7 (324) 10.6 (40) 3.7 (14)
70.5 (98) 21.6 (30) 7.9 (11)
94.6 (226) 4.2 (10) 1.3 (3)
,0.0001 ,0.0001 ,0.0001
60.2 (227) 13.3 (50) 26.5 (100)
31.2 (43) 21.7 (30) 47.1 (65)
77.0 (184) 8.4 (20) 14.6 (35)
,0.0001 ,0.0001 ,0.0001
72.2 (272) 11.4 (43) 16.5 (62)
52.2 (72) 18.1 (25) 29.7 (41)
83.7 (200) 7.5 (18) 8.8 (21)
,0.0001 ,0.0001 ,0.0001
93.1 6.1 0.3 0.5
84.9 12.9 0.7 1.4
97.9 (234) 2.1 (5) 0 (0) 0 (0)
,0.0001 ,0.0001 ,0.0001 ,0.0001
(352) (23) (1) (2)
(118) (18) (1) (2)
a Extreme outlier values were excluded when mean values and standard deviations were calculated for the average number of patients with corneal ulcer seen monthly. Excluded values: 2 cornea specialists who saw 400 patients per month, a non-cornea specialist who saw 300 patients per month, and a non-cornea specialist who saw 120 patients per month.
ulcers is always necessary (Table 4). Compared with non-cornea specialists, a greater number of cornea specialists stated that culturing is necessary for ulcers larger than 1.5·1.5 mm, regardless of the location (50.0% vs. 29.9% of respondents, P=0.0001). In terms of regional differences, ophthalmologists from the Midwest were significantly less likely than East Coast and West Coast physicians to culture corneal ulcers that were unresponsive to antibiotic therapy (P=0.005). Eye & Contact Lens Volume 41, Number 1, January 2015
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
Eye & Contact Lens Volume 41, Number 1, January 2015 TABLE 3.
Current Practices in Corneal Ulcer Management
Maintenance of Diagnostic Testing Supplies and Access to a Compounding Pharmacy Percentage of Respondents (n)
Some type of testing supply Bacterial culture transport swabs Bacterial culture plates (blood and chocolate agar) Fungal culture media Thioglycollate broth Gram stain No supplies
All Respondents (n=373)
Cornea Specialists (n=140)
Non-Cornea Specialists (n=236)
P
81.7 (304)
94.9 (129)
74.2 (175)
,0.0001
72.5 (269)
78.3 (108)
69.1 (161)
0.06
60.6 (226)
88.6 (124)
43.8 (102)
,0.0001
56.0 55.8 43.8 18.3
85.0 85.7 67.1 5.2
38.6 37.8 29.7 25.9
,0.0001 ,0.0001 ,0.0001 ,0.0001
(209) (208) (163) (68)
Access to a Compounding Pharmacy All respondents (n=376) Cornea specialists (n=140) Non-cornea specialists (n=236) East coast (n=100) Midwest (n=55) West Coast (n=80)
(119) (120) (94) (7)
(90) (88) (69) (61)
% of Respondents (n) 91.0 98.6 86.4 82.0 85.5 93.8
(342) (138) (204) (82) (47) (75)
P 0.0004 0.0004
fourth-generation fluoroquinolones (i.e., moxifloxacin and gatifloxacin) were the next most popular treatment choices for more serious ulcers, with 23.6% of all respondents selecting moxifloxacin and 16.0% selecting gatifloxacin. Non-cornea specialists were significantly more likely to choose gatifloxacin for treating large, central, or deep corneal ulcers compared with cornea specialists (21.0% vs. 7.4%, P=0.0005). Of note, cornea specialists were significantly more likely than non-cornea specialists to choose vancomycin (either alone or in combination with other antibiotics) to treat the more severe ulcer (11.0% vs. 0.4%, P,0.0001). In terms of regional differences, West Coast ophthalmologists were significantly more likely to prescribe gatifloxacin for the less severe ulcer (51.3% of West Coast physicians vs. 35.1% of East Coast physicians and 31.5% of Midwest physicians, P=0.03) and significantly less likely to prescribe moxifloxacin (P=0.01) (Table 5). East Coast physicians were significantly more likely to prescribe fortified antibiotics for the less severe ulcer (7.2% of East Coast physicians vs. 0% of Midwest and West Coast physicians, P=0.007). For the more severe ulcer, East Coast physicians were significantly more likely than Midwest or West Coast physicians to prescribe moxifloxacin (35.1% of East Coast respondents vs. 25.5% of Midwest and 16.3% of West Coast respondents, P=0.02).
Fourth-Generation Fluoroquinolones Versus Fortified Antibiotics
Clinical Scenarios The two clinical scenarios that were presented to recipients of the survey consisted of treatment choice for a less serious (small or peripheral ulcer) and a more serious (large, central, or deep) ulcer. For small or peripheral ulcers, 55.4% of all respondents stated that they would treat with moxifloxacin (Table 5). The second most popular treatment choice was gatifloxacin (38.7% of respondents). Significantly more cornea specialists would treat small or peripheral ulcers with 2 fortified broad-spectrum antibiotics compared with non-cornea specialists (8.8% vs. 3.0%, P=0.01). Approximately 62.7% of all respondents would treat the more severe ulcer (large, central, or deep) with 2 fortified broadspectrum antibiotics (i.e., cefazolin and gentamicin), with cornea specialists significantly more likely than non-cornea specialists to prescribe fortified antibiotics (78.1% vs. 53.7%, P,0.0001). The
TABLE 4.
When asked to compare the use of fourth-generation fluoroquinolones with fortified antibiotics for the treatment of more severe corneal ulcers, the majority of respondents (67.8%) believed that fourth-generation fluoroquinolones were less effective compared with fortified antibiotics (Table 5). Approximately 29.2% of all respondents believed that fourth-generation fluoroquinolones were equally as effective as fortified antibiotics. A significantly greater percentage of cornea specialists stated that fourth-generation fluoroquinolones were less effective than fortified broad-spectrum antibiotics, compared with non-cornea specialists (79.6% vs. 60.9%, P=0.001). A significantly greater percentage of non-cornea specialists stated that fourth-generation fluoroquinolones were equally as effective as fortified broad-spectrum antibiotics, compared with cornea specialists (35.6% vs. 18.2%, P=0.001). Only 3.0% of
Opinions on the Need for Culturing Corneal Ulcers Percentage of Respondents (n) Non-Cornea Specialists
When do you believe that culturing is necessary in the treatment of corneal ulcers? (Respondents were able to select multiple answers) Always Ulcers larger than 1.5·1.5 mm, regardless of the location Ulcers larger than 1.5·1.5 mm, within the central 5 mm of the cornea All ulcers that extend to middle to deep stroma Chronic ulcers Ulcers unresponsive to antibiotic therapy Unusual patient history or clinical features Never
© 2014 Contact Lens Association of Ophthalmologists
All Respondents (n=374)
Cornea Specialists (n=138)
Non-Cornea Specialists (n=236)
8.6 (32) 37.4 (139) 52.3 (195)
12.3 (17) 50.0 (69) 53.6 (74)
6.4 (15) 29.9 (70) 51.5 (121)
60.6 57.3 85.8 77.7
65.2 63.8 81.2 79.0
57.9 53.4 88.6 76.9
(226) (213) (321) (289) 0
(90) (88) (112) (109) 0
(136) (125) (209) (180) 0
P
East Coast (n=100)
Midwest (n=55)
West Coast (n=80)
P
5.2 (5) 12.7 (7) 3.8 (3) 0.0001 33.7 (33) 27.3 (15) 26.3 (21) 47.5 (47) 45.5 (25) 61.3 (49) 55.6 (55) 57.1 (56) 92.0 (92) 78.6 (77) 0
52.7 (29) 40.0 (22) 76.4 (42) 72.7 (40) 0
63.8 (51) 57.5 (46) 92.5 (74) 0.005 77.5 (62) 0
15
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
Eye & Contact Lens Volume 41, Number 1, January 2015
J. Park et al. TABLE 5.
Antibiotic Treatment Patterns for Small and Large Corneal Ulcers, Categorized by Whether the Respondent is a Cornea Specialist and by Geographic Region Percentage of Respondents (n) Non-Cornea Specialists All Respondents (n=370)
Small or peripheral corneal ulcer treatment Moxifloxacin Gatifloxacin Ciprofloxacin or ofloxacin Levofloxacin Standard concentrations of nonfluoroquinolone antibiotics (i.e., tobramycin or gentamicin) Two fortified broad-spectrum antibiotics (e.g. cefazolin-gentamicin) Besifloxacin Large, central, or deep corneal ulcer treatment Two fortified broad-spectrum antibiotics (i.e. cefazolin-gentamicin) Moxifloxacin Gatifloxacin Levofloxacin Vancomycin (either alone or in combination with other antibiotic(s)) Ciprofloxacin or ofloxacin Besifloxacin Opinions regarding perceived treatment efficacy: fourthgeneration fluoroquinolones are (less/equally as/more) effective compared with dualfortified broad-spectrum antibiotics Less Equally as More
55.4 38.7 14.1 8.1 7.0
(205) (142) (52) (30) (26)
Cornea Specialists (n=137)
56.2 37.0 9.6 10.9 8.1
(77) (50) (13) (15) (11)
Non-Cornea Specialists (n=233)
54.9 39.7 16.7 6.4 6.4
8.8 (12)
3.0 (7)
3.0 (11)
4.4 (6)
2.1 (5)
62.7 (232)
78.1 (107)
53.7 (125)
23.6 16.0 4.9 4.3
18.4 7.4 3.6 11.0
26.6 21.0 5.6 0.4
(25) (10) (5) (15)
0.01
(62) (49) (13) (1)
,0.0001
1.5 (2) 3.7 (5)
5.2 (12) 2.6 (6)
67.8 (251) 29.2 (108) 3.0 (11)
79.6 (109) 18.2 (25) 2.2 (3)
60.9 (142) 35.6 (83) 3.4 (8)
DISCUSSION This survey study aimed to determine current practices regarding the diagnosis and treatment of patients with bacterial keratitis by ophthalmologists across the United States, as well as to compare the practice patterns of cornea specialists and noncornea specialists. Echoing the findings of previous questionnaire-based studies,7 this study found that Gram stains and cultures of corneal ulcers were performed in a minority of cases (Gram stains were performed on average 23.7%634.1% of the time and cultures performed on average 35.1%638.0% of the time). Similar to the questionnaire study of McLeod et al.,8 our respondents also believed that culturing was necessary only for more severe corneal ulcers, and akin to the survey study of Hsu et al.,9 we also found that cornea specialists were significantly more likely to Gram stain and culture corneal ulcers compared with non-cornea specialists. However, this study was unique because it compared the opinions of both cornea specialists and non-cornea specialists in the treatment choices for two different
,0.0001
0.0005
3.8 (14) 3.0 (11)
respondents believed that fourth-generation fluoroquinolones were more effective than fortified antibiotics.
16
(128) (92) (39) (15) (15)
5.1 (19)
(87) (59) (18) (16)
P
0.001 0.001 0.001
East Coast (n=101)
Midwest (n=56)
61.9 35.1 14.4 10.3 6.2
61.8 31.5 14.6 5.5 10.9
(60) (34) (14) (10) (6)
7.2 (7)
(34) (17) (8) (3) (6)
0 (0)
West Coast (n=80)
41.3 51.3 21.3 2.5 3.8
(33) (41) (17) (2) (3)
0 (0)
3.0 (3)
1.8 (1)
1.3 (1)
46.4 (45)
56.4 (31)
60.0 (48)
35.1 (34) 19.6 (19) 7.2 (7) 0 (0)
25.5 18.2 7.3 1.8
16.3 (13) 25.0 (20) 2.5 (2) 0 (0)
(14) (10) (4) (1)
8.3 (8) 3.0 (3)
3.6 (2) 1.8 (1)
2.5 (2) 2.5 (2)
58.8 (57) 37.1 (36) 4.1 (4)
61.8 (34) 36.4 (20) 1.8 (1)
63.8 (51) 32.5 (26) 3.8 (3)
P
0.01 0.03
0.007
0.02
severities of bacterial keratitis. Indeed, this study found that cornea specialists were significantly more likely than non-cornea specialists to treat more severe corneal ulcers with fortified broad-spectrum antibiotics (78.1% vs. 53.7%, P,0.0001). This study was also notable because it compared regional differences in practices and opinions on bacterial keratitis by sampling a variety of states in three regions of the United States. The cornea specialists in this study saw a greater number of patients with corneal ulcer on a monthly basis compared with noncornea specialists, and cornea specialists were more likely than non-cornea specialists to culture and Gram stain corneal ulcers before treatment (Table 2). Cornea specialists also saw a greater number of initially therapy-resistant ulcers than non-cornea specialists, likely because the more severe and resistant cases of bacterial keratitis are referred to cornea specialists.6 Additionally, cornea specialists were significantly more likely than non-cornea specialists to maintain a number of diagnostic testing supplies, including bacterial culture plates, fungal culture media, thioglycollate broth, and Gram stain supplies. Cornea specialists also were more likely than non-cornea specialists to have access to a compounding pharmacy (98.6% vs. 86.4%, P=0.0004). These results seem logical because cornea specialists see more severe cases of Eye & Contact Lens Volume 41, Number 1, January 2015
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
Eye & Contact Lens Volume 41, Number 1, January 2015 bacterial keratitis and therefore need to perform more diagnostic testing and prescribe more fortified antibiotics. This study also shows that there is a significant difference between cornea specialists and non-cornea specialists in the culturing of corneal ulcers, with cornea specialists more likely than non-cornea specialists to culture ulcers that are larger than 1.5·1.5 mm, regardless of the location (50.0% vs. 29.9%, P=0.0001). This may be due in part to non-cornea specialists immediately referring complicated corneal ulcers to corneal specialists. Several of the physicians indicated in the “comments” section of the survey that there are limits on obtaining Gram stains and fortified antibiotics in a timely manner for patients. Many of the general ophthalmologists commented that for those patients with large and/or deep ulcers, ulcers that do not improve in 24 hours, or for facilities that do not have access to culture media, referrals to cornea specialists are made. The current guidelines published by the AAO in 2013 state that either single-drug therapy with a fluoroquinolone or combination fortified broad-spectrum antibiotics (e.g., cefazolin-tobramycin) may be used as empiric treatment of bacterial keratitis of all severities.11 These guidelines were made in light of three randomized controlled trials that demonstrated that moxifloxacin monotherapy is just as effective for the treatment of bacterial keratitis compared with combination therapy with fortified antibiotics (e.g., cefazolin-tobramycin).20–22 For instance, the randomized controlled trial of Constantinou et al.22 revealed that moxifloxacin, ofloxacin, and fortified cefazolin-tobramycin were all equally effective in the treatment of severe bacterial keratitis, with 34% to 41% of the study participants having ulcers severe enough to cause hypopyon. The randomized controlled trial of Sharma et al.20 showed there was no difference in the number of ulcers that completely healed within 3 months in 224 patients with bacterial keratitis randomized to either moxifloxacin monotherapy or cefazolin-tobramycin–fortified antibiotic therapy. Furthermore, moxifloxacin monotherapy has several practical advantages over dual-fortified antibiotic therapy, including ease of obtaining the medication, increased compliance, and decreased ocular toxicity.20,22 However, the AAO guidelines do state that combination fortified broad-spectrum antibiotic therapy should be considered for severe infections (i.e., central or severe ulcers or infiltrates .2 mm that are threatening the visual axis) and for ulcers unresponsive to initial treatment.11,13 This study demonstrated that more than 56.2% of cornea specialists and 54.9% of general ophthalmologists use moxifloxacin to treat small or peripheral corneal ulcers. These findings are similar to that of Hsu et al.,9 who demonstrated that the fourth-generation fluoroquinolones were the most popular antibiotic choice for bacterial keratitis.9 This is not surprising because moxifloxacin is known to be effective against a broad spectrum of bacteria, including Grampositive and Gram-negative species, and is easy for the patient to obtain. Cornea specialists were significantly more likely than noncornea specialists to treat small or peripheral corneal ulcers with 2 fortified broad-spectrum antibiotics (8.8% vs. 3.0%, P=0.01). This is likely because cornea specialists tend to overtreat rather than undertreat corneal ulcers because they have had exposure to more severe cases of bacterial keratitis and therefore likely want to take precautions to prevent corneal ulcers from deteriorating. With the more severe ulcer (large, central, or deep ulcers), cornea specialists are significantly more likely than general ophthalmologists to treat with 2 fortified broad-spectrum anti© 2014 Contact Lens Association of Ophthalmologists
Current Practices in Corneal Ulcer Management biotics (78.1% of cornea specialists vs. 53.7% of general ophthalmologists, P,0.0001). Non-cornea specialists were significantly more likely than cornea specialists to prescribe gatifloxacin for more severe corneal ulcers (21.0% vs. 7.4%, P=0.0005), perhaps because non-cornea specialists are less likely to have access to compounding pharmacies. Sixteen respondents stated that they would treat the more serious ulcer with vancomycin (either alone or in combination with other antibiotics), with cornea specialists more likely to prescribe vancomycin than non-cornea specialists (11% vs. 0.4%, P,0.0001). This may be because cornea specialists are more aware of the increasing incidence of methicillin-resistant Staphylococcus aureus (MRSA) bacterial keratitis and are thus providing MRSA coverage with vancomycin. Additionally, this study found a difference in moxifloxacin and gatifloxacin prescription rates across the West Coast, Midwest, and East Coast regions of the United States, with a significantly lower rate of moxifloxacin use for small corneal ulcers among the surveyed West Coast ophthalmologists and a significantly higher rate of gatifloxacin use among surveyed West Coast ophthalmologists for small corneal ulcers. It is unclear why the West Coast ophthalmologists who responded to the survey prefer one fourthgeneration fluoroquinolone over the other. Increasing popularity of the fourth-generation fluoroquinolones has contributed to the selection of resistant bacterial strains.23 Knowledge of increasing resistance to fluoroquinolones may be the reason why fortified antibiotics are still perceived as more effective for the treatment of severe corneal ulcers by both the cornea and non-cornea specialists surveyed in this study (Table 5). Specifically, this study showed that 67.8% of respondents believed that fourth-generation fluoroquinolones are less effective than dual-fortified broad-spectrum antibiotics in the treatment of more severe corneal ulcers. Antibiotic resistance is a major concern, with multiple studies indicating that repeated use of antibiotics selects for resistant strains of bacteria.24 For instance, Ray et al.24 demonstrated that organisms from eyes with bacterial keratitis that were pretreated with a fluoroquinolone (second-, third-, or fourth-generation) before presentation had a higher moxifloxacin minimum inhibitory concentration than organisms from eyes with bacterial keratitis that were not exposed to fluoroquinolones before presentation. These studies emphasize the need to avoid indiscriminate use of fluoroquinolones to prevent selection of resistant bacterial strains. A few respondents selected besifloxacin as their initial choice for treating both small and large corneal ulcers (3.0% of all respondents). Besifloxacin is a new fluoroquinolone that has shown greater potency than moxifloxacin and ciprofloxacin against Staphylococci isolated from various ocular sites and will likely become an important player in the battle against organisms resistant to existing fluoroquinolones.23,25 Additionally, besifloxacin is effective against MRSA, unlike most of the other fluoroquinolones, which lack MRSA activity.23,26 Efficacy against MRSA is a critical consideration because the incidence of serious ocular infections caused by MRSA has been increasing.27,28 Additionally, besifloxacin was created specifically as an ocular fluoroquinolone, meaning that it has never been labeled for use in infections in other parts of the body. This reduces the likelihood of organisms resistant to besifloxacin. At present, besifloxacin is only FDA-approved for the treatment of bacterial conjunctivitis, but off-label uses for the treatment of bacterial keratitis in animal models have demonstrated its efficacy.11,29 17
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
Eye & Contact Lens Volume 41, Number 1, January 2015
J. Park et al. In general, judicious use of potent antibiotics is advisable to prevent the development of further antibiotic resistance. Although mandatory culturing of corneal samples and testing of antibiotic susceptibilities in patients with bacterial keratitis is no longer stated in current guidelines, many clinicians recognize the importance of these data on an epidemiologic level. These tests would allow physicians to monitor antibiotic resistance and detect shifting trends in the organisms responsible for bacterial keratitis.2,30 Numerous studies have demonstrated the rapidly evolving pattern of antibiotic resistance,18 which further emphasizes the need to periodically monitor the epidemiology of bacterial corneal ulcers and their susceptibility patterns. Strengths of this study include sampling of ophthalmologists from a wide geographic area and employment of an easy-to-use online survey system that, in theory, encourages a greater number of responses. The benefits of using an online survey system compared with paper survey methods include faster response rates, lower costs, and decreased risk of misplacing the survey. However, one issue with distributing through an Internet-based survey system is that ophthalmologists may be unfamiliar with or may overlook e-mailed surveys, especially if there are no incentives for completing it. Another limitation of the study is selection bias, which is present in any type of study in which participation is optional. Respondents may have been more willing to answer the questionnaire because they followed standards of care, which may have skewed the results toward those that are in line with the AAO guidelines.7 Also, approximately 16% of the surveyed cornea specialists were members of or alumni of the Proctor Foundation, which could contribute to possible selection bias in this study. Despite the ease of answering online surveys, this study had a response rate of 28.6%, which may make the study’s findings less representative of the target population. Additionally, statistical tests were only performed when there seemed to be an obvious difference in the tabulations. Because multiple statistical tests were performed, some of the P-values below 0.05 may be due to chance. Finally, the lack of follow-up information on patients (e.g., ultimate visual acuity, need for penetrating keratoplasty) prevents us from drawing conclusions about the consequences of treating corneal ulcers with different antibiotics. Future studies should investigate patient outcomes for corneal ulcers after treatment with fluoroquinolones versus fortified broadspectrum antibiotics, especially because both fluoroquinolone resistance and the incidence of MRSA bacterial keratitis are on the rise. Trends in antibiotic resistance over time and in different geographic areas should also continue to be monitored to better understand resistance patterns and causative organisms. A study investigating regional differences in bacterial keratitis management patterns among cornea specialists may also be informative. REFERENCES 1. McDonnell PJ. Empirical or culture-guided therapy for microbial keratitis? A plea for data. Arch Ophthalmol 1996;114:84–87. 2. Charukamnoetkanok P, Pineda R II. Controversies in management of bacterial keratitis. Int Ophthalmol Clin 2005;45:199–210. 3. Pepose JS, Wilhelmus KR. Divergent approaches to the management of corneal ulcers. Am J Ophthalmol 1992;114:630–632. 4. Erie JC, Nevitt MP, Hodge DO, et al. Incidence of ulcerative keratitis in a defined population from 1950 through 1988. Arch Ophthalmol 1993;111: 1665–1671.
18
5. Jeng BH, Gritz DC, Kumar AB, et al. Epidemiology of ulcerative keratitis in Northern California. Arch Ophthalmol 2010;128:1022–1028. 6. Rodman RC, Spisak S, Sugar A, et al. The utility of culturing corneal ulcers in a tertiary referral center versus a general ophthalmology clinic. Ophthalmology 1997;104:1897–1901. 7. McDonnell PJ, Nobe J, Gauderman WJ, et al. Community care of corneal ulcers. Am J Ophthalmol 1992;114:531–538. 8. McLeod SD, DeBacker CM, Viana MA. Differential care of corneal ulcers in the community based on apparent severity. Ophthalmology 1996;103: 479–484. 9. Hsu HY, Nacke R, Song JC, et al. Community opinions in the management of corneal ulcers and ophthalmic antibiotics: A survey of 4 states. Eye Contact Lens 2010;36:195–200. 10. McLeod SD, Kolahdouz-Isfahani A, Rostamian K, et al. The role of smears, cultures, and antibiotic sensitivity testing in the management of suspected infectious keratitis. Ophthalmology 1996;103:23–28. 11. American Academy of Ophthalmology Cornea/External Disease Panel. Preferred Practice Pattern Guidelines. Bacterial Keratitis. San Francisco, CA, American Academy of Ophthalmology, 2013. Available at: http:// www.aao.org/ppp. Accessed December 24, 2013. 12. Hanet MS, Jamart J, Chaves AP. Fluoroquinolones or fortified antibiotics for treating bacterial keratitis: Systematic review and meta-analysis of comparative studies. Can J Ophthalmol 2012;47:493–499. 13. Miller D. Pharmacological treatment for infectious corneal ulcers. Expert Opin Pharmacother 2013;14:543–560. 14. Mah FS. Fourth-generation fluoroquinolones: New topical agents in the war on ocular bacterial infections. Curr Opin Ophthalmol 2004;15:316–320. 15. Arici MK, Sumer Z, Guler C, et al. In vitro potency and stability of fortified ophthalmic antibiotics. Aust N Z J Ophthalmol 1999;27:426–430. 16. Prabhasawat P, Chotikavanich S, Leelaporn A. Sterility of non-preservative eye drops. J Med Assoc Thai 2005;88(suppl 9):S6–S10. 17. Gangopadhyay N, Daniell M, Weih L, et al. Fluoroquinolone and fortified antibiotics for treating bacterial corneal ulcers. Br J Ophthalmol 2000;84: 378–384. 18. Afshari NA, Ma JJ, Duncan SM, et al. Trends in resistance to ciprofloxacin, cefazolin, and gentamicin in the treatment of bacterial keratitis. J Ocul Pharmacol Ther 2008;24:217–223. 19. Vital MC, Belloso M, Prager TC, et al. Classifying the severity of corneal ulcers by using the “1, 2, 3” rule. Cornea 2007;26:16–20. 20. Sharma N, Goel M, Bansal S, et al. Evaluation of moxifloxacin 0.5% in treatment of nonperforated bacterial corneal ulcers: A randomized controlled trial. Ophthalmology 2013;120:1173–1178. 21. Shah VM, Tandon R, Satpathy G, et al. Randomized clinical study for comparative evaluation of fourth-generation fluoroquinolones with the combination of fortified antibiotics in the treatment of bacterial corneal ulcers. Cornea 2010;29:751–757. 22. Constantinou M, Daniell M, Snibson GR, et al. Clinical efficacy of moxifloxacin in the treatment of bacterial keratitis: A randomized clinical trial. Ophthalmology 2007;114:1622–1629. 23. Haas W, Pillar CM, Torres M, et al. Monitoring antibiotic resistance in ocular microorganisms: Results from the Antibiotic Resistance Monitoring in Ocular micRorganisms (ARMOR) 2009 surveillance study. Am J Ophthalmol 2011;152:567–574. 24. Ray KJ, Prajna L, Srinivasan M, et al. Fluoroquinolone treatment and susceptibility of isolates from bacterial keratitis. JAMA Ophthalmol 2013;131: 310–313. 25. Miller D, Chang JS, Flynn HW, et al. Comparative in vitro susceptibility of besifloxacin and seven comparators against ciprofloxacin- and methicillinsusceptible/nonsusceptible staphylococci. J Ocul Pharmacol Ther 2013;29: 339–344. 26. Gadaria-Rathod N, Fernandez KB, Kheirkhah A, et al. New insights into infectious keratitis. Int Ophthalmol Clin 2013;53:163–172. 27. Hsiao CH, Chuang CC, Tan HY, et al. Methicillin-resistant Staphylococcus aureus ocular infection: A 10-year hospital-based study. Ophthalmology 2012;119:522–527. 28. Asbell PA, Sahm DF, Shaw M, et al. Increasing prevalence of methicillin resistance in serious ocular infections caused by Staphylococcus aureus in the United States: 2000 to 2005. J Cataract Refract Surg 2008;34:814–818. 29. O’Brien TP. Besifloxacin ophthalmic suspension, 0.6%: A novel topical fluoroquinolone for bacterial conjunctivitis. Adv Ther 2012;29:473–490. 30. Wilhelmus KR, Schlech BA. Clinical and epidemiological advantages of culturing bacterial keratitis. Cornea 2004;23:38–42.
Eye & Contact Lens Volume 41, Number 1, January 2015
Copyright @ Contact Lens Association of Opthalmologists, Inc. Unauthorized reproduction of this article is prohibited.
