Anda di halaman 1dari 5

A Single-Blinded Randomized Clinical Trial Comparing Polymyxin

B-Trimethoprim and Moxifloxacin for Treatment


of Acute Conjunctivitis in Children
Lee Williams, DO1,*, Yogangi Malhotra, MD1,5,*, Barbra Murante, NP1, Susan Laverty, RN1, Steve Cook, MD1,
David Topa, MD1,2, Dwight Hardy, PhD3, Hongyue Wang, PhD4, and Francis Gigliotti, MD1,3

Objective To perform a randomized controlled trial comparing moxifloxacin hydrochloride with polymyxin
B-trimethoprim for the treatment of acute conjunctivitis.
Study design Patients ages 1-18 years old with acute conjunctivitis had cultures performed and were random-
ized to receive either moxifloxacin hydrochloride or polymyxin B-trimethoprim ophthalmic solution for 7 days.
Response to treatment was determined by phone query on day 4-6 and by examination with post-treatment con-
junctival culture on day 7-10.
Results One hundred and twenty-four patients were enrolled. Eighty patients (65%) had recognized pathogens
(55 Haemophilus influenzae, 22 Streptococcus pneumoniae, 4 Moraxella catarrhalis) isolated from their conjunctiva.
One hundred fourteen (56/62 moxifloxacin and 58/62 polymyxin B-trimethoprim) completed the 4-6 day evaluation,
with 43/56 (77%) of the moxifloxacin group and 42/58 (72%) of the polymyxin B-trimethoprim group clinically cured
according to parents (noninferiority test P = .04). Eighty-nine (39/56 moxifloxacin and 50/58 polymyxin B-trimeth-
oprim) patients completed the 7-10 day evaluation. Clinical cure was observed in 37/39 (95%) of the moxifloxacin
and 49/51 (96%) of the polymyxin B-trimethoprim treated groups (noninferiority test P # .01). Clinical cure rates for
culture positive and negative conjunctivitis were not different. There was no statistically significant difference in
bacteriologic cure rates between the 2 groups.
Conclusions Polymyxin B-trimethoprim continues to be an effective treatment for acute conjunctivitis with a clin-
ical response rate that does not differ from moxifloxacin. Use of polymyxin B-trimethoprim for the treatment of con-
junctivitis would result in significant cost savings compared with fluoroquinolones. (J Pediatr 2013;162:857-61).

C
onjunctivitis is the most common infectious disease of the eye during childhood.1 The economic impact of bacterial
conjunctivitis is significant. In the US, an estimated 4 million cases occur each year, resulting in direct medical costs
of approximately 500 million dollars.2 Two-thirds of cases occur between the ages of 0 and 19 years and are estimated
to account for 1.5% of ambulatory and emergency room visits.2,3
Conjunctivitis has both infectious and noninfectious causes. In children, bacterial conjunctivitis is the most common type of
acute conjunctivitis, followed by viral, allergic, traumatic, and neoplastic. Haemophilus influenzae, Streptococcus pneumoniae,
and Moraxella catarrhalis are the most common causative bacteria in bacterial conjunctivitis and account for 55%-68% of cases
in children. Interestingly, Staphylococcus aureus is an uncommon pathogen in acute conjunctivitis.4-6
Although conjunctivitis is a self-limited condition, the duration of bacterial conjunctivitis is shortened by topical antibiotic
treatment. In 1984, Gigliotti et al showed that treatment with polymyxin B-bacitracin ointment resulted in significantly more
clinical and bacteriologic cures by 3-5 days than did treatment with the placebo ointment.7 In 1988, Lohr et al compared Poly-
trim (polymyxin B-trimethoprim), gentamicin, and sulfacetamide topical preparations for treatment of conjunctivitis.5 The
clinical response of S pneumoniae infections at 3-6 days was similar for all agents, and polymyxin B-trimethoprim was more
effective for H influenzae infection. Several studies using other antibiotics have shown similar results.
A 2007 Cochrane review assessed the effect of antibiotic therapy in the man-
agement of acute bacterial conjunctivitis and found that clinical cure (relative
1
From the Department of Pediatrics, Golisano Children’s
risk 1.24, 95% CI 1.05-1.45) and microbiologic clearance (relative risk 1.77, Hospital at Strong, University of Rochester School of
8 2
95% CI 1.23-2.54) were accelerated by use of topical antibiotics. No serious out- Medicine and Dentistry, Rochester, NY; Pittsford
3
Pediatrics, Pittsford, NY; Department of Microbiology
comes were observed in either the treatment or the placebo arms. The Cochrane and Immunology, Strong Memorial Hospital, and
4
Department of Biostatistics and Computational Biology,
review concluded that treatment of bacterial conjunctivitis is associated with University of Rochester School of Medicine and
5
Dentistry, Rochester, NY; and Division of Neonatology,
significantly improved rates of clinical and microbiologic cure. Department of Pediatrics, Maria Fareri Children’s
The increasing antibiotic resistance of pathogens associated with acute bacte- Hospital at Westchester Medical Center, Valhalla, NY
*Contributed equally.
rial conjunctivitis has led many authors to propose that topical fluoroquinolones
Y.M. is supported in part by a resident research grant
from the American Academy of Pediatrics. The authors
declare no conflicts of interest.

MIC Minimal inhibitory concentration 0022-3476/$ - see front matter. Copyright ª 2013 Mosby Inc.
All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2012.09.013

857
THE JOURNAL OF PEDIATRICS  www.jpeds.com Vol. 162, No. 4

are superior to older medications such as polymyxin A follow-up phone call was made 4-6 days later to assess clin-
B-trimethoprim.9-12 However, the superiority of fluoroqui- ical response. Parents were asked about the same signs and
nolones generally is based on interpretation of minimal symptoms assessed at enrollment. The patients then returned
inhibitory concentration (MIC) or other in vitro data show- to the office for a follow-up visit at day 7-10. During this visit,
ing increased potency of the fluoroquinolones compared an assessment of physical findings was made by the investiga-
with polymyxin B or trimethoprim against bacteria causing tor and a follow-up conjunctival culture was obtained. Clin-
conjunctivitis. Although such drug data have been validated ical cure required a complete resolution of all signs and
to predict the outcome of systemic therapy of bacterial symptoms of conjunctivitis at the time of evaluation.
infections, their relevance in topical antibiotic therapy is This study was approved by the Research Subject Review
unproven. Board at the University of Rochester School of Medicine
To provide physicians with clinically relevant data regard- and Dentistry. Parent and/or patient provided informed con-
ing the treatment of acute conjunctivitis, we performed a pro- sent or assent to be included in this study.
spective, randomized, single-blind study of moxifloxacin
compared with polymyxin B-trimethoprim for the treatment Study Microbiology
of acute conjunctivitis in children 1-18 years of age. Conjunctival swab specimens were inoculated onto tryptic
soy agar with 5% sheep blood and GC II Agar supplemented
Methods with hemoglobin and Isovitalex (Becton, Dickinson and Co,
Sparks, Maryland) with incubation at 35 C in ambient air
Patients were enrolled from the Golisano Children’s Hospital supplemented with 5% carbon dioxide for 48 hours. S pneu-
Pediatric Ambulatory Clinic (resident based hospital clinic) moniae was identified as alpha-hemolytic catalase-negative
and Pittsford Pediatrics (a suburban pediatric office in Pitts- gram-positive cocci susceptible to desoxycholate or opto-
ford, New York) from the fall of 2007 to the fall of 2010. chin.13 Haemophilus was identified as small gram-negative
Patients were enrolled based on a clinical diagnosis of con- bacilli with characteristic growth on supplemented GC II
junctivitis by the primary physician. The presence of lid Agar, no growth on tryptic soy agar supplemented with
edema, conjunctival erythema, eye discharge, and/or subcon- blood, and requirement for hemin (X factor) and/or nicotin-
junctival hemorrhage was recorded for each patient enrolled. amide adenine dinucleotide (V factor); H influenzae was
Exclusion criteria included a recent history suggestive of typed with type b specific antisera (Becton, Dickinson and
allergies, foreign body or trauma to the eye, and receipt of an- Co).14 M catarrhalis was identified as oxidase-positive, in-
tibiotics during the previous week or during the study. Non- doxyl butyrate-positive, gram-negative diplococci.15
English speaking patients were excluded because interpreter MICs of antimicrobial agents for S pneumoniae and Hae-
services during the visits and phone follow-up could not be mophilus were determined in Muller-Hinton Broth supple-
guaranteed. mented with 2.5% lysed horse blood and haemophilus test
Patients were assigned 1:1 to receive 1 of the 2 antibiotics, medium broth, respectively, with incubation at 35 C in am-
using a randomization table. In the research pharmacy, bient air for 20-24 hours.16 Haemophilus test medium broth
numbered brown paper bags that contained either poly- was supplemented with 0.2 IU/mL thymydine phosphorylase
myxin B- trimethoprim or 0.5% moxifloxacin hydrochloride when testing trimethoprim and sulfamethoxazole. MICs for
ophthalmic solution were prepared along with antibiotic- S pneumonia and H influenzae were interpreted according
specific directions for administration. to systemic breakpoints established by Clinical and Labora-
This was a single-blind study as the study personnel were tory Standards Institute where available.17
unaware of which medication each patient received. Patients
were instructed to follow the written directions contained in Statistical Analyses
the bag of the antibiotic they received. Polymyxin B-trimeth- All statistical analyses were conducted using v. 9.2 of the SAS
oprim was to be applied 4 times a day for 7 days and moxiflox- System for Windows (SAS Institute Inc, Cary, North Caro-
acin was to be applied 3 times a day for 7 days. The patient lina). Patient demographics and clinical characteristic were
and/or parent were asked not to tell the study personnel which summarized and compared between treatment groups using
medication they had been assigned. Patient compliance the 2 sample t test (continuous variables) or c2 test (categor-
to treatment was not formally monitored during this study. ical variables). Clinical and microbiologic outcomes of the
At the initial visit (day 0), the patient was evaluated by the polymyxin B-trimethoprim group were compared with the
study personnel. Purulent discharge was considered to be moxifloxacin group using noninferiority tests of binomial
present if parents reported either discharge or eyelids sticking proportions. Differences in cure rates at 4-6 days, 7-10
together on morning rising. The presence of lid edema, con- days, and associated CIs were estimated.
junctival erythema, or subconjunctival hemorrhage also was
recorded. The severity of the clinical signs was not graded, Results
as a validated scoring system is not available. A bacterial cul-
ture of the affected eye was obtained (if both eyes appeared to One hundred and twenty-four patients were enrolled in this
be infected, conjunctival was obtained only from one eye) and study. Four patients were excluded after enrollment: 2
was sent to the microbiology lab at Strong Memorial Hospital. patients had allergic symptoms, 1 had dacryocystitis, and
858 Williams et al
April 2013 ORIGINAL ARTICLES

1 had otitis media and was started on amoxicillin. A total of


114 patients completed the 4-6 day follow-up (median day Polymyxin B-
trimethoprim
5), 89 patients completed the 7-10 day follow-up (mean Moxifloxacin
42/58
day 9); 9 patients were lost to follow-up before the 4-6 day 4-6 day clinical cure
visit, and 25 were lost at the final follow-up. Patients who 43/56
missed the 4-6 day phone follow-up were eligible for the 7-
10 day final follow-up in the clinic if they returned.
The 2 groups of subjects were found to be similar at enroll-
ment (Table). Cultures from 80 patients (65%) grew 7-10 day clinical
49/51

recognized pathogens from the initial conjunctival sample cure


37/39
(55 H influenzae, 22 S pneumoniae, 4 M catarrhalis
[1 patient’s culture grew both H influenzae and S
pneumoniae]): 35% did not grow a recognizable pathogen. 0 20 40 60 80 100
Sixty-two (50%) patients were randomized to each treatment
Percentage of patients cured by either moxifloxacin or
group. The number of conjunctival cultures positive for polymyxin B-trimethoprim
pathogens was 41/62 (66%) in the moxifloxacin group and
39/62 (62%) in the polymyxin B-trimethoprim group. Of the Figure 1. Clinical outcome of all children enrolled with acute
114 patients who completed the 4-6 day evaluation, 56/62 conjunctivitis.
(90%) were in the moxifloxacin group and 58/62 (93%) were
in the polymyxin B-trimethoprim group. The median time
to evaluation was 5.1 days for the moxifloxacin group and B-trimethoprim group) completed the 4-6 day evaluation. In
4.8 days for those receiving polymyxin B-trimethoprim. At this subgroup, 30/39 (77%) of moxifloxacin treated patients
the 4-6 day follow-up, 43/56 (77%) of the moxifloxacin and 26/36 (72%) of the polymyxin B-trimethoprim treated
group and 42/58 (72%) of the polymyxin B-trimethoprim patients were clinically cured (P = .10) (Figure 2).
group were categorized as clinically cured (Figure 1). The At the final follow-up, a total of 89 patients (39/56 in the
noninferiority test with a margin of 20% showed that the moxifloxacin group and 50/58 in the polymyxin B-trimetho-
clinical cure rate of the polymyxin B-trimethoprim group is prim group) completed the 7-10 day evaluation. The median
not statistically lower than that of moxifloxacin group time for this evaluation was 9 days for both groups. At 7-10
(P = .04, difference: 0.05, 90% CI: 0.20 to 0.11). Cure days, 37/39 (95%) of the entire moxifloxacin group and 49/
rates were not different for the two treatment groups 51 (96%) of the entire polymyxin B-trimethoprim group
(P = .59). No patients reported worsening of symptoms at 4- were clinically cured (noninferiority test P < .01)
6 days. A post-hoc power calculation based on noninferiority (Figure 1). From the culture-positive subgroup, a total of
was performed. Based on previous work demonstrating 60 patients (28/39 in the moxifloxacin group and 32/36 in
a 62% cure rate with conventional therapy at 4 days (range the polymyxin B-trimethoprim group) completed the 7-10
3-5 days), we estimated that a 25% improvement in time to day follow-up. In this subgroup, 26/28 (93%) of the
cure would be clinically significant.7 A sample size of 60 per moxifloxacin group and 30/32 (94%) of the polymyxin B-
group would achieve 85% power to detect a noninferiority
margin difference of 25% at 5% significance level.
In the subgroup of initial culture-positive patients, 75 Polymyxin B-
trimethoprim
(39/41 in the moxifloxacin group and 36/39 in the polymyxin 4-6 day clinical 26/36
Moxifloxacin
cure 30/39

30/32
Table. Characteristics of each study group 7-10 day clinical
cure 26/28
Polymyxin
B-trimethoprim Moxifloxacin P
(n = 62) (n = 62) value 7-10 day 19/31*
bacteriologic cure 22/28
Mean age (mo) 73.4 (SD 50) 70.3 (SD 48) .73
Male % 37 (n = 23) 55 (n = 34) .07
Female % 63 (n = 39) 45 (n = 28)
Signs/symptoms initial visit 0 20 40 60 80 100
Purulent discharge % 97 (n = 60) 98 (n = 61) 1.0
Lid edema % 63 (n = 39) 61 (n = 38) .85 Percentage of patients cured by either moxifloxacin or
Erythema % 85 (n = 53) 82 (n = 51) .81 polymyxin B-trimethoprim
Fever % 10 (n = 6) 5 (n = 3) .49
Organism
S pneumoniae % 19 (n = 10) 19 (n = 12) .81 Figure 2. Clinical and microbiologic outcomes of the sub-
H influenzae % 45 (n = 28) 44 (n = 27) 1.0
M catarrhalis % 3 (n = 2) 3 (n = 2) 1.0 group of children with confirmed bacterial conjunctivitis. *One
No pathogen % 37 (n = 23) 34 (n = 21) .85 patient did not have a follow-up culture done.

A Single-Blinded Randomized Clinical Trial Comparing Polymyxin B-Trimethoprim and Moxifloxacin for Treatment of 859
Acute Conjunctivitis in Children
THE JOURNAL OF PEDIATRICS  www.jpeds.com Vol. 162, No. 4

trimethoprim group patients were considered clinically cured The results of this study differ from those of Granet et al,
at 7-10 days (P < .01) (Figure 2). The bacteriologic cure rate who also compared polymyxin B-trimethoprim and moxi-
was 22/28 (79%) in the moxifloxacin group and 19/31 (61%) floxacin.18 They found an accelerated cure rate in the moxi-
in the polymyxin B-trimethoprim group (P = .52) (Figure 2). floxacin treated children with an 81% clinical cure rate after
Of the 6 patients with bacteriologic failure in the moxifloxacin only 48 hours of treatment. Several aspects of that study
group, 4 had H influenzae and 2 had S pneumonia isolated. make interpretations of their results difficult. Most children
Among the 12 polymyxin B- trimethoprim bacteriologic in the study did not have a recognized pathogen (H influenzae
failures, 5 that had H influenzae, 1 had M catarrhalis, and 6 or S pneumoniae) isolated from the conjunctiva. The high
had S pneumoniae isolated. cure rate after only 48 hours is unexpected, given the inflam-
To determine the potencies of each antibiotic on a micro- matory nature of bacterial components.19,20 The study also is
gram per microgram basis, the MIC90 was determined for inconsistent with other fluoroquinolones trials.21,22 McDo-
moxifloxacin, polymyxin, and trimethoprim to S pneumo- nald et al reported a clinical cure rate for bacterial conjuncti-
niae and H influenzae isolates collected from patients en- vitis treated with moxifloxacin ophthalmic of 59.4% at day
rolled in this study. Moxifloxacin had a MIC90 of 0.06 5.22 This result is consistent with our study’s clinical cure
mg/mL for S pneumoniae and 0.015 mg/mL for H influenzae. rate of 77% by day 4-6. In addition, Hwang et al compared
Polymyxin B had a MIC90 of 64 mg/mL for S pneumoniae levofloxacin ophthalmic with placebo in a study primarily
and 2 mg/mL for H influenzae. Trimethoprim had of adults and found a 27% clinical cure rate 3-5 days after
a MIC90 of 2 mg/mL for S pneumoniae and 16 mg/mL for initiating treatment.21
H influenzae. Many articles report that fluoroquinolones should be used
for first-line treatment of bacterial conjunctivitis because of
Discussion increasing resistance to commonly used antibiotics. In 2007
Lichtenstein et al demonstrated more rapid killing of S au-
To test the hypothesis, based on frequently published state- reus, S pneumoniae, and H influenzae by fluoroquinolones
ments, that fluoroquinolones are the preferred treatment compared with polymyxin B- trimethoprim.12 Block et al
for acute bacterial conjunctivitis, we performed a single- showed that H influenzae and S pneumoniae isolated from
blind, randomized controlled clinical trial of polymyxin B- the eye often were resistant to the beta-lactam agents, poly-
trimethoprim compared with moxifloxacin for the treatment myxin B-bacitracin, and other drugs commonly used to treat
of acute conjunctivitis in children. Our data do not support conjunctivitis and otitis media.23 This study demonstrated
the hypothesis that fluoroquinolones, in this case moxifloxa- that the 2 topical fluoroquinolones tested possessed the high-
cin, are superior to polymyxin B-trimethoprim. est intrinsic activity, as defined by MIC values. However, the
Our study of 120 children, 80 of whom had microbiolog- relationship between MIC and clinical response to topical an-
ically confirmed bacterial conjunctivitis, showed that the 4- tibiotic treatment remains undefined.
to 6-day clinical cure rate after treatment with polymyxin Furthermore, because of the data showing that fluoroqui-
B-trimethoprim or moxifloxacin was not different. nolones have lower MICs, there is the perception that therapy
Seventy-seven percent of all moxifloxacin treated patients with drugs, such as polymyxin B-trimethoprim, is inferior
and 72% of all polymyxin B-trimethoprim treated patients and therefore inappropriate for treatment. However, in re-
were clinically cured at the first evaluation timepoint, and viewing polymyxin B based treatments over the last 30 years
those not cured were all improved. These latter patients this does not seem to be the case. A 1984 study by Gigliotti
were not categorized as cures because complete resolution et al demonstrated a clinical cure rate of 62% at 3-5 days after
of all signs and symptoms was required to be classified as treatment with polymyxin B-bacitracin, and a 1988 study by
a clinical cure. There was no difference in the cure rate for Lohr et al showed a 47% clinical cure rate at 3-6 days after
those patients from whom a pathogen was isolated compared treatment with polymyxin B-trimethoprim.5,7 Given the clin-
with those from whom no pathogen was isolated. As would ical cure rate in this study of 72% on day 4-6 after treatment
be expected, nearly all patients (95% of moxifloxacin; 96% with polymyxin B-trimethoprim, there does not appear to be
of polymyxin B-trimethoprim) had a normal examination any decline in the efficacy of polymyxin B based topical treat-
7-10 days after starting topical antibiotics.7 ment of bacterial conjunctivitis. This supports the need to de-
Bacteriologic cure rates, assessed at the 7-10 day visit, ap- velop a better understanding of how to relate MIC values to
pear to be slightly better for the moxifloxacin group (79%) clinical outcomes when using topical antibiotic therapy.
compared with the group treated with polymyxin B-trimeth- The efficacy of an antibiotic is determined by its antimicro-
oprim (61%). However, this difference was not statistically bial potency and its concentration at the site of infection. There
significant and did not result in more rapid resolution of con- is limited information regarding antibiotic tear concentrations
junctivitis. This suggests that the antimicrobial activity of after topical application in the literature. Data are available
polymyxin B-trimethoprim when applied topically is suffi- for azithromycin, levofloxacin, and moxifloxacin.24-28 Mean
cient, when combined with host defense mechanisms in the tear concentrations of moxifloxacin at 5 minutes following
normal eye, to bring about clinical resolution and that there the first and last topical doses (1 drop every 8 hours for 10
is little added benefit to the increased potency of the fluoro- doses) were 46 and 55.2 mg/mL, respectively.28 Predose
quinolones for the treatment of conjunctivitis. tear concentrations averaged 4 mg/mL. We determined
860 Williams et al
April 2013 ORIGINAL ARTICLES

MIC90s for moxifloxacin, polymyxin B, and trimethoprim 9. Pichichero ME. Bacterial conjunctivitis in children: antibacterial treat-
against S pneumoniae and H influenzae. Moxifloxacin was ment options in an era of increasing drug resistance. Clin Pediatr
2011;50:7-13.
clearly more potent than both polymyxin B and trimethoprim
10. Adebayo A, Parikh JG, McCormick SA, Shah MK, Huerto RS, Yu G, et al.
but as already mentioned, this increased potency did not result Shifting trends in in-vitro antibiotic susceptibilities for common bacteria
in a superior clinical outcome. conjunctival isolates in the last decade at the New York Eye and Ear In-
To estimate the financial implications of these results we firmary. Graefes Arch Clin Exp Ophthalmol 2011;249:111-9.
searched the 2007 National Ambulatory Medical Care Survey 11. Bertino JS. Impact of antibiotic resistance in the management of ocular
infections: the role of current and future antibiotics. Clin Ophthalmol
and National Hospital Ambulatory Medical Care Survey da-
2009;3:507-21.
tabases to estimate the number of annual visits for acute bac- 12. Lichtenstein SJ, Wagner RS, Jamison T, Bell B, Stroman DW. Speed of
terial conjunctivitis in the US. Using International bacterial kill with a fluoroquinolone compared with nonfluoroquino-
Classification of Diseases, 9th Revision codes 372.30, 372.03, lones: clincal implications and a review of kinetics of kill studies. Adv
and 372.00 for patients <18 years old, it was estimated that Ther 2007;24:1098-111.
13. Spellerberg B, Brandt C. Streptococcus. In: Versalovic J, ed. Manual of
there were approximately 3 732 322 cases of bacterial con-
clinical microbiology. 10th ed. Washington, DC: ASM Press; 2011. p.
junctivitis in 2007. These estimates are consistent with the re- 331-49.
sults of Smith and Waycaster who used the 2005 National 14. Ledeboer NA, Doem GV. Haemophilus. In: Versalovic J, ed. Manual of
Ambulatory Medical Care Survey data set searching for Inter- clinical microbiology. 10th ed. Washington, DC: ASM Press; 2011. p.
national Classification of Diseases, 9th Revision codes 372.30, 588-602.
15. Vaneechoutte M, Dijkshoorn L, Nemec A, Kampfer P, Wauters G. Aci-
372.03, and 372.00 and yielding a total 4 016 544 visits.2 Of
netobacter, Chryseobacterium, Moraxella and Other Nonfermentative
these 4 016 544 visits there were 1 356 693 visits by patients Gram-Negative Rods. In: Versalovic J, ed. Manual of clinical microbiol-
<15 years old.2 To estimate the average cost of polymyxin ogy. 10th ed. Washington, DC: ASM Press; 2007. p. 714-38.
B-trimethoprim and moxifloxacin ophthalmic solutions, 9 16. Clinical and Laboratory Standards Institute. Methods for Dilution An-
local pharmacies in Rochester, New York were surveyed. timicrobial Susceptibility Tests for Bacteria that Grow Aerobically;
Approved Standard – 8th ed. Wayne, PA: 2009; CLSI document
The average cost of polymyxin B-trimethoprim was $16/10
M07-A8, CLSI.
mL (range $4-$34) compared with moxifloxacin, which 17. Clinical and Laboratory Standards Institute. Performance Standards for
was $105/3 mL (range $101.88-$111.49). Based on these Antimicrobial Susceptibility Testing: Twenty-First Informational Sup-
rough cost estimates using polymyxin B-tremthiprim rather plement. Wayne, PA: 2011; CLSI document M100-S21, CLSI.
than a fluoroqunilone would save in excess of $300 million/y 18. Granet DB, Dorfman M, Stroman D, Cockrum P. A multicenter com-
parison of polymyxin B sulfate/trimethoprim ophthalmic solution and
based on an estimated 3.7 million cases each year. n
moxifloxacin in the speed of clinical efficacy for the treatment of bacte-
rial conjunctivitis. J Pediatr Ophthalmol Strabismus 2008;45:340-9.
The authors thank Carl T. D’Angio, MD (University of Rochester), for 19. Ginsburg I. Role of lipoteichoic acid in infection and inflammation. Lan-
his help with statistical analysis of the data. cet Infect Dis 2002;3:171-9.
20. Leake ER, Holmes K, Lim DJ, DeMaria TF. Peptidoglycan isolated from
Submitted for publication Apr 24, 2012; last revision received Aug 1, 2012;
nontypeable Haemophilus influenzae induces experimental otitis media
accepted Sep 6, 2012. in the chinchilla. J Infect Dis 1994;170:1532-8.
21. Hwang DG, Schanzlin DJ, Rotberg MH, Foulks G. A phase III, placebo
Reprint requests: Francis Gigliotti, MD, Department of Pediatrics, Golisano
Children’s Hospital at Strong, University of Rochester School of Medicine and
controlled clinical trial of 0.5% levofloxacin ophthalmic solution for the
Dentistry, 601 Elmwood Avenue, Box 690, Rochester, NY 14642. E-mail: treatment of bacterial conjunctivitis. Br J Ophthalmol 2003;87:1004-9.
francis_gigliotti@urmc.rochester.edu 22. McDonald MB, Protzko EE, Brunner LS, Morris TW, Haas W,
Paterno MR, et al. Efficacy and safety of besifloxacin ophthalmic suspen-
sion 0.6% compared with moxifloxacin ophthalmic solution 0.5% for
References treating bacterial conjunctivitis. Ophthalmology 2009;116:1615-23.
23. Block SL, Hedrick J, Tyler R, Smith A, Findlay R, Keegan E, et al. Increas-
1. Fisher RG, Boyce TG. Moffet’s pediatric infectious diseases: A problem- ing bacterial resistance in pediatric acute conjunctivitis (1997-1998).
oriented approach. 4th ed. Philadelphia (PA): Lippincott Williams & Antimicrob Agents Chemother 2000;44:1650-4.
Wilkins; 2005. 24. Chiambaretta F, Garraffo R, Elena PP, Pouliquen P, Delval L, Rigal D,
2. Smith AF, Waycaster C. Estimate of the direct and indirect annual cost of et al. Tear concentrations of azithromycin following topical administra-
bacterial conjunctivitis in the United States. BMC Ophthalmol 2009;9: tion of a single dose of azithromycin 0.5%, 1.0%, and 1.5% eye drops
13-24. (T1225) in healthy volunteers. Eur J Ophthalmol 2008;18:13-20.
3. National Hospital Ambulatory Medical Care Survey (NHAMCS); 2007. 25. Blandizzi C, Malizia T, Batoni G, Ghelardi E, Baschiera F, Bruschini P,
4. Gigliotti F, Williams WT, Hayden FG, Hendley JO, Benjamin J, et al. Distribution of azithromycin in plasma and tonsil tissue after re-
Dickens M, et al. Etiology of acute conjunctivitis in children. J Pediatr peated oral administration of 10 or 20 milligrams per kilogram in pedi-
1981;98:531-6. atric patients. Antimicro Agents Chemother 2002;46:1594-6.
5. Lohr JA, Austin RD, Grossman M, Hayden GF, Knowlton GM, 26. Raizman MB, Rubin JM, Graves AL, Rinehart M. Tear concentrations of
Dudley SM. Comparison of three topical antimicrobials for acute bacte- levofloxacin following topical administration of a single dose of 0.5%
rial conjunctivitis. Pediatr Infect Dis J 1988;7:626-9. levofloxacin ophthalmic solution in healthy volunteers. Clin Ther
6. Weiss A, Brinser JH, Nazar-Stewart V. Acute conjunctivitis in childhood. 2002;24:1439-50.
J Pediatr 1993;122:10-4. 27. Baum H, Bottcher S, Abel R, Gerner HJ, Sonntag HG. Tissues and serum
7. Gigliotti F, Hendley JO, Morgan J, Michaels R, Dickens M, Lohr J. Effi- concentrations of levofloxacin in orthopedic patients. Int J Antimicrob
cacy of topical antibiotic therapy in acute conjunctivitis in children. J Pe- Agents 2001;18:335-40.
diatr 1984;104:623-6. 28. Product Monograph: Vigamox: Moxifloxacin Hydrochloride Ophthal-
8. Sheikh A, Hurwitz B. Antibiotics versus placebo for acute bacterial con- mic Solution, 0.5% w/v as moxifloxacin. Mississauga, Ontario: Alcon
junctivitis. Cochrane Database of Syst Rev 2006;19:CD001211. Canada Inc, L5N 8C7; May 4, 2004.

A Single-Blinded Randomized Clinical Trial Comparing Polymyxin B-Trimethoprim and Moxifloxacin for Treatment of 861
Acute Conjunctivitis in Children