Over the counter chloramphenicol eye dropsBMJ 2010; 340 doi: https://doi.org/10.1136/bmj.c1016 (Published 26 February 2010) Cite this as: BMJ 2010;340:c1016
- Geoff Scott, consultant microbiologist, University College London Hospitals
After much agonising, in 2005 the Medicines and Healthcare Regulatory Agency (MHRA) in the United Kingdom decided to allow pharmacists to dispense chloramphenicol eye drops and ointment without a doctor’s prescription. Pharmacists used to be able to sell “Golden Eye Ointment,” which contained a sulphonamide, until the sulphonamide component was withdrawn some years ago by the Committee on Safety of Medicines because its clinical activity had not been proved. Chloramphenicol preparations retail at £5.50 (€6.3; 8.5) to £6 each, a little less than the current cost of a prescription.
A recent study shows that the number of units of chloramphenicol eye drops dispensed between 2004 and 2007 increased dramatically, from 2.3 million to 3.4 million a year.1 This rise of 48% coincided with a fall in medical prescriptions of 16%. Might this have led to an increase of resistance in common flora that can cause invasive disease, and if so, does it matter? Eye preparations pass down the nasolacrimal duct, where they will encounter Staphylococcus aureus, and further back into the nasopharynx, where they will meet Streptococcus pneumoniae and Haemophilus influenzae. Chloramphenicol will also be swallowed so that gut flora may be exposed to very low concentrations of the antibiotic.
In the UK, the widespread use of systemic chloramphenicol gradually faded during the 1980s to be replaced by cephalosporins, quinolones, and macrolides as first line agents in the management of, for example, invasive infections of the lung and central nervous system and enteric fever. Resistance to chloramphenicol in imported Salmonella typhi and S paratyphi is relatively common, but chloramphenicol is probably slightly less active than quinolones and azithromycin even against sensitive strains of these organisms. So very little of this antibiotic is used systemically in the UK now. Nevertheless, perpetuation of chloramphenicol resistance in—for example, Escherichia coli in the urine—results from genes that encode chloramphenicol resistance being linked on plasmids or integrons to those that encode resistance to trimethoprim.
Bacteria do not have simple mechanisms of mutation to resistance to chloramphenicol but have to acquire “new” genetic information. Resistance is generally mediated by the gene encoding chloramphenicol acetyl transferase (catA), which is usually borne on a plasmid. This gene may also be included in integrons containing several resistance genes, which can also be transferred from one bacteria to another. Other possible mechanisms include change in permeability of bacteria or even active export of chloramphenicol, genes for which could be upregulated by exposure to the drug.2
It is difficult to measure antimicrobial resistance in any meaningful way. Clinical isolates are by their very nature highly selected and do not reflect changes in normal flora. Changes in resistance to chloramphenicol in clinical isolates are unlikely to be noticed in the UK because the drug is rarely prescribed. In many laboratories, isolates may not even be tested. Prospective microbiological studies of normal nasal and pharyngeal flora before and after using chloramphenicol would indicate the true risk of selecting resistance but would not be funded.
Specimens for culture and virology are rarely received in the laboratory given the large number of patients with transient conjunctivitis. And neither would pharmacists have the inclination or ability to send such specimens to the laboratory. Furthermore, a recent randomised trial in 326 children found that chloramphenicol is no better than placebo in terms of clinical cure at seven days (83% in those taking placebo and 86% in those taking chloramphenicol; risk difference 3.8%, 95% CI −4% to 12%).3 The reason for the failure of topical chloramphenicol in this acute bacterial infection is not clear—most infections are associated with susceptible virulent bacteria, and the recoverable bacterial load does fall with chloramphenicol treatment. Perhaps it is because the drug fails to get to where the bacteria are dividing. It is more likely, however, that the inflammatory process is self limiting, and that killing the bacteria has no effect on this process. The relatively few cases of viral (for example, adenovirus) infection would not be expected to respond anyway.
Oral chloramphenicol, although cheap to make and used extensively worldwide, is now expensive in the UK—it is one of the most expensive antimicrobials in the formulary—so it is unlikely to be used extensively in the foreseeable future. Intravenous chloramphenicol is still useful for managing occasional intracranial infections or ventilator associated pneumonia caused by coliforms and staphylococci that are resistant to other antibiotics. Chloramphenicol is also used in certain practices in agriculture, such as the farming of prawns.
The MHRA made an error in allowing the sale of chloramphenicol, not because of the dangers of unwanted effects or of selecting for resistance, which are minimal, but because the treatment is ineffective. Soothing eye preparations are all that is needed, and people who do not respond to this management after about three days should see a doctor to have specimens taken so that appropriate treatment can be given.
The prescription of antibiotics by non-medically qualified personnel raises strong bipolar opinions. Some people thought that this move was the thin end of an unacceptable wedge, and that pharmacists would soon be able to sell common oral antibiotics without prescription. Ideally, that would lead to a reduction in overall use of antibiotics, but the case of chloramphenicol eye drops suggests that an increase would be more likely.
Cite this as: BMJ 2010;340:c1016
Competing interests: The author has completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declares: (1) No financial support for the submitted work from anyone other than his employer; (2) No financial relationships with commercial entities that might have an interest in the submitted work; (3) No spouse, partner, or children with relationships with commercial entities that might have an interest in the submitted work; (4) No non-financial interests that may be relevant to the submitted work.
Provenance and peer review: Commissioned; not externally peer reviewed.