Skip to main content
PMC home page
The BMJ logoLink to The BMJ
. 2002 Jun 8;324(7350):1394. doi: 10.1136/bmj.324.7350.1394/a

Fusidic acid cream for impetigo

Fusidic acid should be used with restraint

Erwin M Brown 1,2, Richard Wise 1,2
PMCID: PMC1123338  PMID: 12052817

Editor—Koning et al report the results of a clinical trial that showed the efficacy of topical fusidic acid as treatment of patients with impetigo.1 This agent has been recommended by the Dutch College of General Practitioners as the treatment of choice in patients with this infection. Koning et al observed that none of the pretreatment isolates of Staphylococcus aureus was resistant to fusidic acid and concluded that many years of use of topical fusidic acid has not resulted in appreciable resistance in staphylococci in the general population.

These findings illustrate one of the problems surrounding antimicrobial resistance—namely, that patterns of resistance in one country cannot be extrapolated to those in another. Specifically, data for resistance rates to fusidic acid among S aureus isolates in the United Kingdom differ markedly from those in the Netherlands. In a survey of 28 centres in the United Kingdom the incidence of resistance to fusidic acid among S aureus isolates from the community (excluding strains of methicillin resistant S aureus, which, by their clonal nature, might distort the data) increased from 8.1% in 1995 to 17.3% in 2001 (figure) (R Wise, unpublished data).2 A similar study carried out in Bristol showed an approximately twofold increase in resistance rates (from 6% to 11.5%) among methicillin susceptible S aureus strains isolated between 1998 and 2001.3

The figure also shows that between 1995 and 2001 the number of prescriptions of fusidic acid in the United Kingdom (expressed as total units dispensed and accounted for almost entirely by the topical formulation) nearly doubled (data supplied by Leo Pharmaceuticals).

We cannot explain why the Dutch experience does not mirror our own, although Koning et al have not specified the technique they used to determine the susceptibilities of their isolates, nor have they provided information about the susceptibilities to fusidic acid of any isolates after treatment. A further confounding factor could be the small number of isolates tested (67 strains); evaluating a larger, and therefore more representative, sample might yield a different pattern of resistance.

We do not dispute the efficacy of topical fusidic acid as treatment of patients with impetigo and other superficial skin infections. But fusidic acid is a very valuable drug that is also administered systemically in combination with another antistaphylococcal antibiotic, usually flucloxacillin, as treatment of patients with severe staphylococcal infections. Rates of resistance to this agent among S aureus isolates are increasing in the United Kingdom, directly in line with usage, and we are concerned that further increases in the prescribing of topical fusidic acid will result in even higher levels of resistance. The price of continuing to administer the drug in this way will, in the long term, be the loss of the therapeutic efficacies of both the topical and systemic formulations, and we urge restraint, particularly among general practitioners and dermatologists.

Figure.

Figure

Open in a new tab

Annual rates of resistance to fusidic acid among isolates of Staphylococcus aureus, with numbers of prescriptions for fusidic acid

References

  • 1.Koning S, van Suijlekom-Smit LWA, Nouwen JL, Verduin CM, Bernsen RMD, Oranje AP, et al. Fusidic acid cream in the treatment of impetigo in general practice: double blind randomised placebo controlled trial. BMJ. 2001;324:203–206. doi: 10.1136/bmj.324.7331.203. . (26 January.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Andrews J, Ashby J, Jevons G, Marshall T, Lines N, Wise R. A comparison of antimicrobial resistance rates in Gram-positive pathogens isolated in the UK from October 1996 to January 1997 and October 1997 to January 1998. J Antimicrob Chemother. 2000;45:285–293. doi: 10.1093/jac/45.3.285. [DOI] [PubMed] [Google Scholar]
  • 3. Brown EM, Thomas P. Fusidic acid resistance in Staphylococcus aureus isolates. Lancet 2002;359 (in press). [DOI] [PubMed]
BMJ. 2002 Jun 8;324(7350):1394.

Judicious use is advisable

Olajumoke Sule 1,2,3, Nick Brown 1,2,3, Derek F J Brown 1,2,3, Nigel Burrows 1,2,3

Editor—Koning et al compared topical fusidic acid cream with placebo in 160 children with impetigo and found no resistance to fusidic acid in Staphylococcus aureus isolates from 135 children.1-1 We believe that resistance to fusidic acid is underrecognised and widespread use of topical treatment may lead to an increase in resistance.

Resistance rates to fusidic acid of 1.8-9.8% have been reported in the United Kingdom in the past decade, higher rates being associated with chronic skin infections.1-2–1-4 We reviewed resistance to fusidic acid for all S aureus isolates obtained from blood and wound cultures at this hospital during 2000 and 2001. Susceptibility was determined by the disc method of the British Society for Antimicrobial Chemotherapy.1-5

Higher rates were seen among isolates from dermatology outpatients (33.7%) and patients from the community (18.1%) compared with isolates from blood cultures (4.6%) and hospital inpatients (6.9%) (table). These differences may partly explain the variation in reported rates of resistance in different studies. We found no significant difference between resistance rates in 2000 and 2001.

Table.

Fusidic acid resistance in Staphylococcus aureus isolates in Addenbrooke's Hospital, 2000-1

Source 2000
2001
Total No (%) resistant
No of
isolates		 No (%) resistant No of isolates No (%) resistant
Dermatology outpatients (wound)  332 112 (33.7)  311 105 (33.8) 217 (33.7)
Community patients (wound) 1847 327 (17.7) 2038 377 (18.5) 704 (18.1)
Hospital inpatients:
 Wound 2157 164 (7.6) 2397 151 (6.3) 315 (6.9)
 Blood  278 10 (3.6)  327 18 (5.5) 28 (4.6)
Total 4614 613 (12.8) 5073 651 (12.8) 1264 (13)   
Open in a new tab

Resistance varied with age among dermatology outpatients (59% in 114 patients ⩽20 years old, 32% in 155 patients aged 21-49, and 21% in 101 patients ⩾50). The increased resistance in younger patients is probably related to their greater likelihood of being treated with topical fusidic acid for chronically infected eczema.

Koning et al described 10 isolates (7%) as being of intermediate susceptibility, although they did not define intermediate or state the method of susceptibility testing. The disc susceptibility method does not distinguish between intermediate and resistant categories, all isolates with a minimum inhibitory concentration ⩾2 mg/l being reported resistant. We determined minimum inhibitory concentrations for fusidic acid by an agar dilution method for 92 consecutive isolates resistant to fusidic acid by the disc method. Eighty six (94%) of the isolates had values in the range 2-16 mg/l and might have been reported as either resistant or intermediate by other methods. Six (7%) had higher values (32 to >128 mg/l). Given the high concentrations of fusidic acid achieved in topical treatment, the clinical importance of isolates with lower levels of resistance is uncertain. We know of no data relating degree of resistance to treatment outcome.

In combination with other antibiotics, fusidic acid remains a useful systemic antibiotic for the treatment of severe staphylococcal infections, including those caused by methicillin resistant strains. Judicious use of topical fusidic acid is advisable, particularly in dermatology practice and the community.

References

  • 1-1.Koning S, van Suijlekom-Smit LWA, Nouwen JL, Verduin CM, Bernsen RMD, Oranje AP, et al. Fusidic acid cream in the treatment of impetigo in general practice: double blind randomised placebo controlled trial. BMJ. 2001;324:203–206. doi: 10.1136/bmj.324.7331.203. . (26 January.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 1-2.Speller DCE, Johnson AP, James D, Marples RR, Charlett A, George RC. Resistance to methicillin and other antibiotics in isolates of Staphylococcus aureus from blood and cerebrospinal fluid, England and Wales, 1989-95. Lancet. 1997;350:323–325. doi: 10.1016/s0140-6736(97)12148-1. [DOI] [PubMed] [Google Scholar]
  • 1-3.Andrews J, Ashby J, Jevons G, Marshall T, Lines N, Wise R. A comparison of antimicrobial resistance rates in Gram-positive pathogens isolated in the UK from October 1996 to January 1997 and October 1997 to January 1998. J Antimicrob Chemother. 2000;45:285–293. doi: 10.1093/jac/45.3.285. [DOI] [PubMed] [Google Scholar]
  • 1-4.Ravenscroft JC, Layton A, Barnham M. Observations on high levels of fusidic acid resistant Staphylococcus aureus in Harrogate, North Yorkshire, UK. Clin Exper Dermatol. 2000;25:327–330. doi: 10.1046/j.1365-2230.2000.00654.x. [DOI] [PubMed] [Google Scholar]
  • 1-5.Andrews J.for the BSAC Working Party on Susceptibility Testing. BSAC standardized disc susceptibility testing method J Antimicrob Chemother 200148(suppl S1)43–57. [DOI] [PubMed] [Google Scholar]
BMJ. 2002 Jun 8;324(7350):1394.

Emergence of resistance to fusidic acid limits its use

Vivienne C Weston 1,2,3,4, Timothy C J Boswell 1,2,3,4, Roger G Finch 1,2,3,4, William Perkins 1,2,3,4

Editor—Koning et al compared the efficacy of topical fusidic acid cream and povidone iodine shampoo with the combination of placebo and povidone iodine shampoo in the treatment of impetigo.2-1 As we have found in Nottingham, Staphylococcus aureus rather than S pyogenes is the main pathogen isolated, but whereas no resistance to fusidic acid was found in the Dutch study, this is not our current experience.

Table.

Percentages of Staphylococcus aureus isolates resistant to fusidic acid in Nottingham, 1998-2001

Year Hospital isolates Community isolates
1998 10.0 10.9
1999 10.7 14.7
2000  9.0 14.6
2001*  9.7 13.9
Open in a new tab
2-150

Up to 31 August. 

In Nottingham the topical use of fusidic acid cream alone or in combination with topical steroids for impetigo and eczema has increased, but this has been associated with a rise in the rate of resistance to fusidic acid in S aureus isolates, particularly from cases of impetigo. Over the past three years we have consistently found significantly higher rates of fusidic acid resistance among community isolates of S aureus compared with hospital strains (table), in complete contrast to the situation with methicillin resistant S aureus. Resistance has been associated with failure of treatment and subsequent clusters of cases. This has led us to change our advice for the treatment of impetigo, for which we now locally recommend oral antistaphylococcal agents for all but very mild cases.

We caution against excessive use of topical fusidic acid with active monitoring for the emergence of resistance in the Netherlands. In the United Kingdom topical fusidic acid can no longer be relied on for the treatment and reduction of infectivity of impetigo.

References

  • 2-1.Koning S, van Suijlekom-Smit LWA, Nouwen JL, Verduin CM, Bernsen RMD, Oranje AP, et al. Fusidic acid cream in the treatment of impetigo in general practice: double blind randomised placebo controlled trial. BMJ. 2001;324:203–206. doi: 10.1136/bmj.324.7331.203. . (26 January.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2002 Jun 8;324(7350):1394.

Problem may be clinically important

Bethan Stoddart 1,2, Timothy Collyns 1,2, Miles Denton 1,2

Editor—Koning et al assert that topical fusidic acid is effective when used in combination with povidone iodine for the treatment of non-bullous impetigo.3-1 They isolated no Staphylococcus aureus strain classified as resistant to fusidic acid up to 28 days after treatment. They advocated the use of topical fusidic acid as the first choice drug for the treatment of impetigo in general practice.

Although the incidence of resistance to fusidic acid resistance among S aureus generally remains low, it has long been recognised that the use of fusidic acid may lead to the emergence of resistance to this agent in previously susceptible strains.3-2,3-3 A recent investigation into fusidic acid resistant S aureus isolates in Harrogate, North Yorkshire, postulated a link with topical use of fusidic acid.3-4

We analysed the sensitivity pattern of all 5093 isolates of S aureus cultured from swabs labelled “wound” or “skin” over one year (29 January 2001 to 28 January 2002) in the microbiology department at Leeds General Infirmary. The swabs were taken from patients admitted to hospital and patients seen in the outpatient and accident and emergency departments. Susceptibility to fusidic acid was determined on the basis of disc diffusion testing. We found a significant association between resistance to fusidic acid and swabs taken in the dermatology ward and clinic, which was found neither for other units nor for the other antibiotics analysed (table). The proportion of strains resistant to other antibiotics was smaller among dermatology patients than for patients analysed overall.

Table.

Patients with one or more skin or wound swabs positive for Staphylococcus aureus. Values are numbers (percentages) of patients with resistance to various antibiotics

Fusidic acid Methicillin Erythromyci n Gentamicin Penicillin
Total (n=1709) 212 (12.4) 512 (30.0) 563 (32.9) 12 (0.7) 1545 (90.4)
Dermatology ward or clinic (n=220) 80 (36)* 16 (7) 49 (22) 2 (1) 193 (88)
Accident and emergency (n=143) 21 (15) 8 (6) 25 (18) 0 115 (80)
Children under 12 years 65 (29)* 7 (3) 34 (15) 0 204 (90)
 Eczema or impetigo (n=61) 39 (64)* 1 (2) 16 (26) 0 59 (97)
 Impetigo (n=10) 9 (90) 0 1 (10) 0 9 (90)
 Accident and emergency (n=25) 11 (44)* 0 3 (12) 0 20 (80)
Open in a new tab
3-150

P<0.001 in χ2 test. 

The paediatric patients were not exempt from this finding. Patients younger than 12 years were significantly more likely to carry fusidic acid resistant S aureus, even in the primary care setting of accident and emergency. This result was particularly marked in patients whose clinical details included infected eczema or impetigo and swabs taken in the dermatology clinic (data not shown).

The pharmacy department at Leeds General Infirmary confirmed that the dermatologists are the only clinicians regularly using topical fusidic acid. High rates of resistance to fusidic acid have been seen in other dermatology wards, where patient to patient transmission has been implicated.3-2,3-4 But in this analysis, as both inpatients and outpatients of all ages were included, extensive cross infection with a limited number of strains resistant to fusidic acid seems unlikely.

Acute treatment of a first episode of impetigo in the absence of chronic skin disease may be successful, but it would be prudent to be wary of relying on topical fusidic acid for the treatment of impetigo in other settings, at least without microbiological confirmation of susceptibility.

References

  • 3-1.Koning S, van Suijlekom-Smit LWA, Nouwen JL, Verduin CM, Bernsen RMD, Oranje AP, et al. Fusidic acid cream in the treatment of impetigo in general practice: double blind randomised placebo controlled trial. BMJ. 2001;324:203–206. doi: 10.1136/bmj.324.7331.203. . (26 January.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3-2.Turnidge J, Collignon P. Resistance to fusidic acid. Int J Antimicrob Agents. 1999;12(suppl 2):S35–S44. doi: 10.1016/s0924-8579(98)00072-7. [DOI] [PubMed] [Google Scholar]
  • 3-3.Lowbury EJL, Cason JS, Jackson D, Miller RWS. Fucidin for staphyloccocal infection of burns. Lancet. 1962;ii:478–480. doi: 10.1016/s0140-6736(62)90339-2. [DOI] [PubMed] [Google Scholar]
  • 3-4.Ravenscroft JC, Layton A, Barnham M. Observations on high levels of fusidic acid resistant Staphylococcus aureus in Harrogate, North Yorkshire, UK. Clin Exper Dermatol. 2000;25:327–330. doi: 10.1046/j.1365-2230.2000.00654.x. [DOI] [PubMed] [Google Scholar]
BMJ. 2002 Jun 8;324(7350):1394.

Resistance trends must be monitored

Paul Zadik 1,2, Nicola Young 1,2

Editor—The study by Koning et al shows the effectiveness of topical fusidic acid for impetigo in a country where resistance of Staphylococcus aureus to this agent is still low.4-1 Resistance arises readily, however, and we wish to bring your attention to the trend we are seeing.4-2,4-3 Resistance rates for isolates of S aureus from both hospital and community and corrected for duplicates have risen from 5% in 1995 to 17% in 2001. The resistance rate was 9% in 1996, 8% in 1997, 11% in 1998, 15% in 1999, and 15% in 2000.

The usefulness of topical fusidic acid is threatened by this development, which may itself be a result of the widespread use of these preparations. The usefulness of systemic fusidic acid, crucial for treating bone infections and against methicillin resistant strains, is also threatened. Topical preparations of fusidic acid must be used responsibly, and resistance trends must be monitored.

References

  • 4-1.Koning S, van Suijlekom-Smit LWA, Nouwen JL, Verduin CM, Bernsen RMD, Oranje AP, et al. Fusidic acid cream in the treatment of impetigo in general practice: double blind randomised placebo controlled trial. BMJ. 2001;324:203–206. doi: 10.1136/bmj.324.7331.203. . (26 January.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4-2.Hilson GRF. In-vitro studies of a new antibiotic (Fucidin) Lancet. 1962;i:932–933. doi: 10.1016/s0140-6736(62)91970-0. [DOI] [PubMed] [Google Scholar]
  • 4-3.Lowbury EJL, Cason JS, Jackson DM, Miller RWS. Fucidin for staphylococcal infection of burns. Lancet. 1962;ii:478–480. doi: 10.1016/s0140-6736(62)90339-2. [DOI] [PubMed] [Google Scholar]
BMJ. 2002 Jun 8;324(7350):1394.

Findings cannot be extrapolated

Susannah E Owen 1,2, John S Cheesbrough 1,2

Editor—Koning et al found that fusidic acid was more effective than placebo for the treatment of staphylococcal impetigo in a sample drawn from Dutch general practitioners.5-1 They went on to recommend topical fusidic acid as first line treatment for impetigo.

Our experience in Lancashire does not support the use of fusidic acid for this indication. We investigated the local epidemiology of impetigo after an outbreak that was first noted in two primary schools, in the summer of 2000. In a general practice population of 10 000, 67 patients presented with impetigo over a six month period in 2000, compared with only six over the same six month period in 1997. Between September 2000 and October 2001, Staphylococcus aureus was isolated from 46 patients with impetigo, of whom only two had received fusidic acid before sampling. Resistance to fusidic acid was present in 17 (37%) of these isolates.

This high level of resistance to fusidic acid contrasts sharply with the Dutch study, in which no resistance to fusidic acid was detected among S aureus isolated from pretreatment skin swabs. Detailed typing of a selection of 13 impetigo related isolates, with varying fusidic acid sensitivity, from seven Preston practices showed lysis by group 2 phages, a pattern typically associated with skin infection. The isolates were also indistinguishable by genotyping, and all possessed the exfoliative toxin a and b genes.

Resistance to fusidic acid among all isolates of S aureus from specimens submitted by general practitioners to the Preston Public Health Laboratory in 2001 was 11%. The higher rate of resistance to fusidic acid among impetigo related isolates suggests that this resistance has conferred a selective advantage on a strain with the potential to cause impetigo. This may in part account for the large increase in the incidence of impetigo recently observed. Fusidic acid is used widely in the community, not only in the treatment of impetigo but also, in combination with steroids, in the treatment of eczema. Our policy has therefore been to recommend the use of oral antibiotics such as flucloxacillin for first line treatment of impetigo and to reserve topical mupirocin for patients with just one or two lesions. While recognising the importance of the study by Koning et al of a very common condition, we should be cautious about extrapolating their recommendations to the United Kingdom, where antibiotic use is more intensive and antibiotic resistance encountered more often in S aureus.5-2,5-3

Footnotes

We thank Hazel Aucken, Laboratory of Hospital Infection, Central Public Health Laboratory, for her help.

References

  • 5-1.Koning S, van Suijlekom-Smit LWA, Nouwen JL, Verduin CM, Bernsen RMD, Oranje AP, et al. Fusidic acid cream in the treatment of impetigo in general practice: double blind randomised placebo controlled trial. BMJ. 2001;324:203–206. doi: 10.1136/bmj.324.7331.203. . (26 January.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5-2.Cars O, Möllstad S, Melander A. Variation in antibiotic use in the European Union. Lancet. 2001;357:1851–1853. doi: 10.1016/S0140-6736(00)04972-2. [DOI] [PubMed] [Google Scholar]
  • 5-3.De Neeling AJ, van Leeuwen WJ, Schouls LM, Scot CS, van Veen-Rutgers A, Beunders A, et al. Reistance of staphylococci in the Netherlands: surveillance by an electronic network during 1989-1995. J Antimicrob Chemother. 1998;41:93–101. doi: 10.1093/jac/41.1.93. [DOI] [PubMed] [Google Scholar]
BMJ. 2002 Jun 8;324(7350):1394.

Authors' reply

S Koning 1,2,3,4,5,6, L W A van Suijlekom-Smit 1,2,3,4,5,6, J L Nouwen 1,2,3,4,5,6, C M Verduin 1,2,3,4,5,6, R M D Bernsen 1,2,3,4,5,6, A P Oranje 1,2,3,4,5,6, S Thomas 1,2,3,4,5,6, J C van der Wouden 1,2,3,4,5,6

Editor—Guidelines for treating impetigo may vary between countries or regions. The wish to reserve certain antibiotics for specific vital conditions, and local resistance rates, may vary and lead to different recommendations. The letters commenting on our article focus on the question of whether the use of fusidic acid cream should be promoted in the United Kingdom in light of the higher reported staphylococcal resistance rates to fusidic acid in the United Kingdom and the risk of rising resistance rates because of its use.

The main purpose of our study was to assess the effectiveness of fusidic acid cream in impetigo compared with disinfection with povidone iodine alone. The large treatment effect we found may partly be explained by the low resistance rate to fusidic acid, which was an unexpected co-finding. We used Vitek II equipment to test the susceptibility of isolates, using the guidelines of the national committee for clinical laboratory standards to categorise strains as resistant, intermediately sensitive, or sensitive, as described in the website version of our article.6-1

Brown et al are not correct in saying that we tested just 67 strains; we tested 135 strains. The low resistance rate in our study is probably related to the low antibiotic use in general by outpatients in the Netherlands and the fact that systemic fusidic acid is scarcely used in the Netherlands.6-2 It is not used as the first or second choice systemic antibiotic for complicated staphylococcal infections. But topical fusidic acid has been the first choice antibiotic in the Dutch general practitioners' guideline for several years now, and its short term use has apparently not led to a significant resistance rate in Staphylococcus aureus causing impetigo in the Netherlands.

Resistance rates may vary between countries. One should be cautious, however, in comparing resistance rates from different samples. High staphylococcal resistance rates from dermatology ward patients may be due to chronic use of fusidic acid cream in patients with infected eczema. Furthermore, even a laboratory sample from patients seen in general practice cannot easily be compared with our sample. We took swabs from all patients with impetigo who consulted their general practitioner, whereas general practitioners do not take bacterial swabs as a routine. A laboratory sample from a general practice is therefore a selected sample of worse or treatment resistant cases. Data from seven Dutch public health laboratories, reflecting both hospital and primary care patients, show a resistance rate to fusidic acid in staphylococci that has slowly risen since 1989-95 and was 4.6% in 2001 (personal communication, A J de Neeling).6-3 This shows the same trend as the data mentioned in the letters, but at a lower level.

The high resistance rate to fusidic acid in S aureus isolates from primary care patients in the impetigo outbreak in Preston reported by Cheesbrough may not be considered representative for all S aureus causing impetigo, as the strains from this specific outbreak are likely to be clonally related. It does, however, show the importance of bacteriological determination and assessment of resistance, also in general practice and especially in case of an epidemic. Moreover, regional and national monitoring of trends in antibiotic drug resistance remains very important.

Excessive and chronic use of an antibiotic will lead to an increase in resistance. But, as Brown et al apparently agree, excessive use of other local treatments such as mupirocin or oral antistaphylococcal agents such as flucloxacillin will similarly promote resistance to the concerned antibiotic.6-3 We therefore recommend a responsible use of fusidic acid cream, as is true for any other topical or oral antibiotic, following current regional policies.

References

  • 6-1.Cars O, Möllstad S, Melander A. Variation in antibiotic use in the European Union. Lancet. 2001;357:1851–1853. doi: 10.1016/S0140-6736(00)04972-2. [DOI] [PubMed] [Google Scholar]
  • 6-2.De Neeling AJ, van Leeuwen WJ, Schouls LM, Schot CS, van Veen-Rutgers A, Beunders AJ, et al. Resistance of staphylococci in the Netherlands: surveillance by an electronic network during 1989-1995. J Antimicrob Chemother. 1998;41:93–101. doi: 10.1093/jac/41.1.93. [DOI] [PubMed] [Google Scholar]
  • 6-3.Brown EM, Thomas P. Fusidic acid resistance in Staphylococcus aureus isolates. Lancet. 2002;359:803. doi: 10.1016/S0140-6736(02)07869-8. [DOI] [PubMed] [Google Scholar]

Articles from BMJ : British Medical Journal are provided here courtesy of BMJ Publishing Group

RESOURCES