Bacterial conjunctivitis.

CLINICAL QUESTION: What is the best treatment for bacterial conjunctivitis?
RESULTS: Topical antibiotics expedite recovery from bacterial conjunctivitis. The choice of antibiotic usually does not affect outcome. IMPLEMENTATION: Recognition of key distinguishing features of bacterial conjunctivitis Pitfalls that can be recognized in the history and physical examinationChoice of antibioticWhen to refer for specialist treatment.

Bacterial conjunctivitis

Definition: Bacterial conjunctivitis is inflammation of the conjunctiva as a result of bacterial infection.

Etiology: Most commonly Staphylococcus species in adults, and Streptococcus pneumonia and the Gram-negative organisms Haemophilus influenzae and Moraxella catarrhalis in children. Contact lens wearers are at particular risk for Gram-negative infections. such as Pseudomonas aeruginosa. Neisseria gonorrhoeae is primarily a neonatal etiology.

Incidence: One recent study estimates an annual incidence rate of 135 per 10,000 in the US.1

Economics: The same study found the estimated total direct and indirect cost of treating bacterial conjunctivitis in the US to be $589 million annually. Accounting for a 20% variation in annual incidence rate and treatment cost resulted in an estimated cost range of $377 to $857 million per year.

Level of evidence used in this summary: Systematic reviews, meta-analyses, and randomized controlled trials from 1990 to 2010.

Search sources: Ovid MEDLINE, PubMed, Cochrane Library, NHS evidence, Clinical Evidence.

Outcomes: From the patient perspective, the main outcomes are:

Speed of symptomatic resolution

Convenience of treatment

Avoidance of complications.

Consumer summary: Bacterial conjunctivitis is inflammation of the conjunctiva caused by direct contact with infected secretions. The most common organisms are Staphylococcus species, S. pneumonia, H. influenzae, and M. catarrhalis. It presents with conjunctival injection, mucopurulent discharge, and crusty eyelids. The diagnosis is usually clinical. The condition is often self-limiting, but there is good evidence that antibiotics improve remission rates. Most of the current evidence suggests that the choice of topical antibiotics and the treatment regimen do not significantly affect the rate of recovery from infection. Failure to recognize and treat bacterial conjunctivitis may lead to complications, such as keratitis or anterior uveitis.

The evidence

Do any interventions make a difference to the resolution of bacterial conjunctivitis?

The Cochrane systematic review,2 which includes a meta-analysis, concluded that “acute bacterial conjunctivitis is frequently a self-limiting condition, but the use of antibiotics is associated with significantly improved rates of clinical and microbiological remission”. The systematic review by Clinical Evidence3 concludes that topical antibiotics are “ beneficial” in people with culture-positive nongonococcal bacterial conjunctivitis and “likely to be beneficial” when used empirically in people with suspected bacterial conjunctivitis within 1–2 days if symptoms do not resolve on their own. Oral antibiotics, ocular decongestants, warm compresses, and saline were found to be of “unknown effectiveness”.

Most randomized controlled trials (see Table 1) showed that topical antibiotics accelerate bacterial eradication and help resolve the signs and symptoms of bacterial conjunctivitis. However, in two trials,4,5 clinical recovery at seven days after presentation was found to be unaffected by the use of antibiotics, even though one of the two trials4 still found an improvement in microbial cure rate with antibiotics.

Table 1

Randomized controlled trials comparing antibiotics with placebo

Author	Number of patients randomized	Interventions	Outcome measures	Results
Abelson et al4	279	One group received azithromycinOne group received “vehicle”	Clinical resolution and bacterial eradication	Higher rate of microbial and clinical cure with antibiotic.
Everitt et al5	307	Two groups received chloramphenicolOne group received placebo	Symptomatic relief	Antibiotic decreased the duration of symptoms.
Hwang et al6	249	One group received levofloxacinOne group received placebo	Clinical resolution and bacterial eradication	Higher rate of microbial and clinical cure with antibiotic.
Karpecki et al7	269	One group received besifloxacinOne group received “vehicle”	Clinical resolution and bacterial eradication	Higher rate of microbial and clinical cure with antibiotic
Leibowitz8	177	One group received ciprofloxacinOne group received placebo	Culture results	Higher rate of microbial cure with antibiotic.
Lichtenstein and Rinehart9	167	One group received levofloxacinOne group received ofloxacinOne group received placebo	Clinical resolution and bacterial eradication	Higher rate of microbial and clinical cure with antibiotics.
Miller et al10	284	One group received norfloxacinOne group received placebo	Bacterial eradication and clinical resolution	Higher rate of microbial and clinical cure with antibiotic.
Rietveld et al11	181	One group received fusidic acidOne group received placebo	Clinical resolution and bacterial eradication	No difference in clinical recovery rate but higher rate of microbial eradication with antibiotic
Rose et al12	326	One group received chloramphenicolOne group received placebo	Clinical cure by day 7	No significant difference between antibiotic and placebo
Tepedino et al13	957	One group received besifloxacinOne group received “vehicle”	Clinical resolution and bacterial eradication	Higher rate of microbial and clinical cure with antibiotic

Which antibiotics are best for accelerating resolution of bacterial conjunctivitis?

Table 2 lists the antibiotics studied, along with their microbial coverage, mechanism of action, and availability. The systematic review3 concluded that “there is no clear best choice for topical antibiotics – local microbiological resistance patterns, cost, dosing regimens, and other patient factors (such as allergies and compliance) are important considerations in addition to efficacy”. Results from randomized controlled trials (Table 3) are varied, but many found similar clinical and microbiologic efficacy among the topical antibiotics used. Some studies found faster bacterial eradication and/or clinical recovery with fluoroquinolones, azithromycin, or netilimicin compared with the more traditional antibiotics, such as tobramycin or polymyxin B/trimethoprim or gentamicin. Some studies found differences in patient compliance with different antibiotics. Microbiologic resistance patterns can also vary and would affect efficacy rates.

Table 2

Topical antibiotics used to treat bacterial conjunctivitis

Antibiotic	Class	Coverage	Mechanism	Availability
Azithromycin	Macrolide	Broad-spectrum	Baceriostatic	Azasite® 1% (Inspire Pharmaceuticals Inc)
Besifloxacin	Fluoroquinolone	Broad-spectrum	Bactericidal	Besivance® 0.6% (Bausch and Lomb)
Chloramphenicol	Chloramphenicol	Broad-spectrum	Bacteriostatic	Topical drops not marketed in USOptrex Infected Eyes® 0.5% in UK
Ciprofloxacin	Fluoroquinolone	Broad-spectrum	Bactericidal	Ciloxan® 0.3% (Alcon Laboratories Inc)Ointment or drops
Fusidic acid	Protein synthesis inhibitor	Primarily Gram-positive	Bacteriostatic	Not available in USFucithalmic® 1% (Leo Pharma) in Canada and UK
Gatifloxacin	Fluoroquinolone	Broad-spectrum	Bactericidal	Zymar 0.3% (Allergan Inc)
Gentamicin	Aminoglycoside	Primarily Gram-negative	Bactericidal	Generic 0.3% drops
Levofloxacin	Fluoroquinolone	Broad-spectrum	Bactericidal	Iquix® 1.5% (Vistakon Pharmaceuticals)
Lomefloxacin	Fluoroquinolone	Broad-spectrum	Bactericidal	Not available in US
Moxifloxacin	Fluoroquinolone	Broad-spectrum	Bactericidal	Vigamox® 0.5% (Alcon Laboratories Inc)
Neomycin-polymyxin B-gramicidin	Aminoglycoside, polymyxin and gramicidin	Broad-spectrum	Bactericidal	Neosporin® (King Pharmaceuticals Inc)
Netilmicin	Aminoglycoside	Primarily Gram-negative	Bactericidal	Not available in US
Norfloxacin	Fluoroquinolone	Broad-spectrum	Bactericidal	Chibroxin 0.3% (Merck and Co Inc)Not available in US
Ofloxacin	Fluoroquinolone	Broad-spectrum	Bactericidal	Generic 0.3% eye drops
Providone-iodine		Broad-spectrum	Bactericidal	Betadine 5% (Alcon Laboratories Inc)
Rifamycin	Rifamycin	Broad-spectrum	Bactericidal	Not available in US
Tobramycin	Aminoglycoside	Primarily Gram-negative	Bactericidal	Tobrex® 0.3% (Alcon Laboratories Inc) ointment or drops

Table 3

Randomized controlled trials comparing different topical antibiotics

Author	Number of randomized patients	Interventions	Outcome measures	Results
Adenis at al14	131	0.3% ciprofloxacin versus 0.3% norfloxacin	Clinical resolution and bacterial eradication	No difference between the two antibiotics
Adenis et al15	41	0.3% ciprofloxacin versus 1% rifamycin	Clinical resolution and bacterial eradication	Higher clinical cure rate with ciprofloxacin on day 7 (but below statistical significance: P = 0.061), no difference in microbial cure
Bloom et al16	464	Ciprofloxacin versus tobramycin	Clinical resolution and bacterial eradication	No difference between the two antibiotics
Bremond-Gignac et al17	150	1.5% azithromycin versus 0.3% tobramycin	Clinical resolution and bacterial eradication	Greater bacteriologic cure with azithromycin on day 3, no difference in clinical or bacteriologic cure on day 9
Chisari et al18	190	Ciprofloxacin versus norfloxacin	Clinical resolution and bacterial eradication	No difference between the two antibiotics
Cochereau et al19	1043	1.5% azithromycin for 3 days versus 0.3% tobramycin for 7 days	Clinical resolution and bacterial eradication	Higher rate of clinical cure with azithromycin on day 3, no difference in clinical or bacteriologic cure on day 9
Denis et al20	1043	1.5% azithromycin for 3 days versus 0.3% tobramycin for 7 days	Microbiological resolution	No difference between the two groups
Gallenga et al21	99	0.3% lomefloxacin BID versus 0.3% tobramycin QID	Clinical resolution and bacterial eradication	No difference between the two groups
Granet et al22	84 eyes of 56 patients	Polymyxin/trimethoprim QID versus 0.5% moxifloxacin TID	Relief of signs and symptoms	Faster clinical resolution with moxifloxacin
Gwon23	345	0.3% ofloxacin versus 0.3% tobramycin	Clinical resolution and bacterial eradication	Similar efficacy between the two treatments, more rapid symptom relief with ofloxacin
Isenberg et al24	459 total, 124 culture-positive for bacteria	1.25% povidoneiodine versus neomycin-polymyxin B-gramicidin	Clinical resolution	No difference between povidoneiodine and antibiotic
Jackson et al25	484	1% fusidic acid versus 0.3% tobramycin	Clinical resolution, bacterial eradication, compliance, subjective “convenience” of treatment	No difference between clinical or microbial resolution, higher compliance and convenience with fusidic acid among younger patients
Kernt et al26	276	Enhanced-viscosity 0.3% tobramycin BID versus 0.3% tobramycin QID	Clinical resolution	No difference between the two groups
Lichtenstein et al11	167	0.5% levofloxacin versus 0.3% ofloxacin (versus placebo)	Clinical resolution and bacterial eradication	Higher microbial eradication rate with levofloxacin in 2–11-year-old children; no difference between the two antibiotics in other age groups
Malminiemi et al27	45	0.3% lomefloxacin versus 1% fusidic acid	Clinical resolution and bacterial eradication	No difference in clinical recovery but higher rate of bacterial eradication with lomefloxacin after 3–5 days
McDonald et al28	1161	0.6% besifloxacin versus 0.3% moxifloxacin	Clinical resolution and bacterial eradication	No difference between the two groups; higher rate of eye irritation with moxifloxacin
Milazzo et al29	45	0.3% netilmicin versus 0.3% tobramycin	Clinical resolution and bacterial eradication	No difference in clinical resolution, better microbiologic outcome with netilmicin
Miller et al30	246	Norfloxacin versus chloramphenicol	Clinical resolution and bacterial eradication	No difference between the two groups
Normann et al31	456 newborns	1% fusidic acid versus 0.5% chloramphenicol	Clinical resolution and compliance	No difference in efficacy but better compliance with fusidic acid
Papa et al32	209	Netilmicin versus gentamicin	Clinical resolution and bacterial eradication	Greater efficacy rate with netilmicin
Power et al33	?	0.3% ciprofloxacin versus 0.5% chloramphenicol	Clinical resolution and bacterial eradication	No difference between the two groups
Protzko et al34	743	1% azithromycin in DuraSite versus 0.3% tobramycin	Safety, clinical resolution and bacterial eradication	Similar safety and efficacy between the two groups
Robert et al35	1043	1.5% azithromycin versus 0.3% tobramycin	Clinical resolution	No difference between the two groups
Schwab et al36	423	0.5% levofloxacin versus 0.3% ofloxacin	Clinical resolution and bacterial eradication	More rapid microbial resolution with levofloxacin, similar clinical resolution
Tabbara et al37	40	0.3% lomefloxacin versus 0.3% ofloxacin	Clinical resolution	No difference between the two groups
Zhang et al38	132	0.3% levofloxacin versus 0.3% ofloxacin	Clinical resolution and bacterial eradication	No difference between the two groups

Abbreviations: BID, twice daily; TID, three times daily; QID, four times daily.

Which treatment regimen works best for bacterial conjunctivitis?

A few randomized controlled trials (Table 4) have focused on the effect of the treatment regimen, such as dosing, frequency, length of treatment, and route of administration, on efficacy rates. None have found a significant change in cure rate in association with the treatment regimen used.

Table 4

Randomized controlled trials comparing different regimens of treatment

Author	Number of randomized patients	Interventions	Outcome measures	Results
Friedlaender39	50	0.3% ofloxacin BID versus QID	Clinical resolution and bacterial eradication	No difference between the two groups
Szaflik et al40	120	0.5% levofloxacin TID × 5 days versus “standard regimen” (Q2H × 2 days, then Q4H × 3 days)	Clinical resolution and bacterial eradication	No difference between the two groups
Wald et al41	80	Oral cefixime + topical placebo versus topical polymyxinbacitracin + oral placebo	Clinical resolution and bacterial eradication	No difference between the two groups
Yee et al42	104	0.3% gatifloxacin BID versus QID	Clinical resolution, bacterial eradication and safety	No difference between the two groups

Abbreviations: Q2H, two hourly; Q4H, four hourly; BID, twice daily; TID, three times daily; QID, four times daily.

Conclusions

Bacterial conjunctivitis often resolves on its own, but the current evidence suggests that topical antibiotics help accelerate recovery from this self-limiting disease. Topical antibiotics used for treatment of bacterial conjunctivitis have similar efficacy rates. The treatment regimen does not affect recovery from bacterial conjunctivitis. Patients may prefer a simpler regimen.

Contact lens wearers are predisposed to Gram-negative infections, carrying a higher risk of complications, such as bacterial keratitis. Pseudomonas and Acanthamoeba infections in contact lens wearers can lead to serious, sight-threatening complications if not recognized and treated appropriately. The contact lens storage case may be the nidus of the infection.

If there is an associated keratitis or anterior uveitis, referral to a specialist may be recommended

Beware of combination topical antibiotic agents that contain steroids. These should be used with extreme caution and monitored by a specialist.

The practice

Management

Bacterial conjunctivitis can be managed by nonspecialists.

Redness, foreign body sensation and purulent/ mucopurulent discharge are common complaints; there may be itching, chemosis, or conjunctival papillae

Ask about contact lens wear

Assess for corneal involvement and intraocular involvement

Conjunctival swabs can be done for Gram stain, culture, and sensitivity to clarify diagnosis, particularly in more severe or refractory cases

Moderate to severe eye pain, photophobia, or change in visual acuity should raise suspicion for more serious causes.

Uncomplicated cases can be treated with a topical antibiotic such as tobramycin, trimethoprim/polymyxin B, a fluoroquinolone or chloramphenicol four times daily for 5–7 days to accelerate recovery

Patients should be seen every 2–3 days until signs and symptoms are resolved

Failure to respond to topical antibiotics may warrant referral to a specialist.

Change in visual acuity

Evidence of keratitis and/or anterior uveitis on slit-lamp examination

Moderate-to-severe eye pain

Failure to improve or worsening of symptoms in spite of treatment.

Systematic reviews:	2
Meta-analyses:	1
Randomized controlled trials:	10

Systematic reviews:	1
Meta-analyses:	0
Randomized controlled trials:	26

Systematic reviews:	0
Meta-analyses:	0
Randomized controlled trials:	4