JAC Advance Access originally published online on January 31, 2008
Journal of Antimicrobial Chemotherapy 2008 61(4):962-964; doi:10.1093/jac/dkn009
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research letters |
Potential for underdosing and emergence of resistance in Acinetobacter baumannii during treatment with colistin
Department of Microbiology, University Hospital Birmingham NHS Foundation Trust, Queen Elizabeth Hospital, Birmingham B15 2TH, UK
* Correspondence address. Department of Microbiology, University Hospitals Coventry and Warwickshire NHS Trust, Walsgrave Hospital, Coventry CV2 2DX, UK. E-mail: m.david{at}nhs.net
Keywords: A. baumannii , antibiotic resistance , therapeutic drug monitoring
Sir,
Increasing antibiotic resistance in Gram-negative bacteria and continuous narrowing of antimicrobial therapeutic choices are worrying trends. Colistin remains sometimes the only option in the treatment of Acinetobacter baumannii strains with in vitro resistance to most antibiotics including carbapenems, aminoglycosides, fluoroquinolones and glycylcyclines.
In particular, we wish to highlight the potential for underdosing and the associated risk of emergence of resistance, especially when managing certain groups of patients such as those with extensive burns, when an abnormal creatinine clearance as well as features of immunosuppression may be intrinsically present. We illustrate this with our experience of treating infections caused by such multidrug-resistant A. baumannii with the parenteral preparation of colistin, colistimethate sodium (CMS).
A previously healthy 22-year-old lady sustained 80% burns requiring admission to the intensive care unit. During her prolonged hospital stay, the wounds became colonized with an A. baumannii strain susceptible only to colistin. The isolate’s resistance to gentamicin, amikacin, tobramycin, imipenem, meropenem, piperacillin/tazobactam, sulbactam and ciprofloxacin was confirmed by the Antibiotic Resistance Monitoring and Reference Laboratory (ARMRL), Health Protection Agency Centre for Infections, Colindale, London, UK. The MIC of colistin as determined by agar dilution was 
0.5 mg/L, well within the susceptible range. When the patient became septicaemic and A. baumannii was isolated from blood culture, she was prescribed initially 1 million units of CMS (Colomycin®) three times a day; a suboptimal clinical response with persistence of pyrexia, on a background of normal renal function, prompted an increase in the dose to 2 million units three times a day. However, because of new onset of seizures of unknown aetiology, thought to possibly represent neurotoxic side effects, 24 h later the CMS dose was reduced back to 1 million units three times a day.
After an initial clinical improvement and resolution of the acute infective episode, a blood culture collected during a recurrent septic event again grew A. baumannii and thus CMS (Colomycin®) was recommenced at gradually increasing doses from 1 million units twice a day to 1.5 million units three times a day. Pre- and post-dose drug levels measured by bioassay at the Regional Antimicrobial Reference Laboratory (RARL), Bristol, were 3.7 and 5.4 mg/L, respectively (target levels: pre-dose 2–6 mg/L, post-dose 5–15 mg/L).1 Despite therapy, the patient’s condition deteriorated and she died from sepsis-related multiorgan failure.
When tested by ARMRL, the MIC of colistin for the second isolate was found to have increased to the breakpoint level of 4 mg/L. It is known that clinical efficacy usually reduces as the MIC approaches the breakpoint and that isolates with such an MIC may produce subpopulations with an MIC just over the breakpoint. In our case, suboptimal colistin serum concentrations may have contributed to the selection of a population with higher MIC.
We have treated with colistin since then several patients (including other burns patients), using a starting dose of 2 million units three times a day if the baseline urea and electrolytes (U&Es) are normal. Pre- and post-dose levels would normally be checked 3–5 days after commencement of therapy; if colistin concentrations are outside the therapeutic range, taking also into consideration the overall clinical picture, doses are adjusted accordingly.
It is our belief that therapeutic drug monitoring may have a role in managing colistin therapy. The British National Formulary (BNF) recommends serum colistin concentration monitoring especially in renal impairment, cystic fibrosis and neonates; the target ‘peak’ plasma concentration (
30 min after intravenous injection or infusion) is 10–15 mg/L (125–200 U/mL),2 a range with a lower limit even higher than that proposed by the RARL. However, measuring by bioassay the serum level of the active component (colistin) while it co-circulates with its intravenously administered prodrug (CMS) is not straightforward because of uncertainties regarding the rate of both in vivo and in vitro hydrolysis of CMS to colistin. Employing HPLC as an alternative method for determining the concentration of these components may partially but not completely resolve these issues.3
Furthermore, there is little information on the pharmacokinetics of colistin at infection sites after intravenous administration4 and in particular there are no such studies in burns patients. It is generally acknowledged that in this group, an apparently ‘normal’ renal function may in fact conceal an elevated glomerular filtration rate (GFR) which potentially may reflect an accelerated clearance of certain antibiotics. Under these circumstances, it is advisable that such groups of antibiotics are used at least at their maximum recommended dose; preferably drug levels should be monitored.
The ideal dose of intravenous colistin is not clearly established. Although the BNF recommends 1–2 million units every 8 h for adults over 60 kg,2 some authors advise up to 3 million units every 8 h for life-threatening infections.5 In our experience, there is a general reluctance among clinicians and Medical Microbiologists to use high doses of colistin because of perceived risk of significant toxicity. However, in a recent review, Falagas and Kasiakou6 concluded that the incidence of nephrotoxicity in recently published experience with polymixins in general is less common and severe compared with older reports. Furthermore, the neurotoxic side effects are much less common than the nephrotoxic ones, they are usually mild and resolve shortly after therapy is discontinued.6
In conclusion, in burns patients we advocate prescribing intravenous colistin at the maximum recommended dose in order to ensure best chance of therapeutic success and prevention of resistance. It is of course mandatory that renal function is taken into account and closely monitored, preferably by measurement of more sensitive parameters such as the GFR. In addition, therapeutic drug monitoring is valuable, especially in critically ill patients, as well as in those in other categories of patients with abnormally high or low drug clearance. Further studies are needed on the pharmacokinetics of CMS and colistin, particularly in burns patients.
 |
Transparency declarations |
|---|
 Top Transparency declarations References |
|---|
None to declare.
|
References |
|---|
|
TopTransparency declarationsReferences |
|---|
1 Antimicrobial Reference Laboratory. Antimicrobial Assay Service. Guidelines for Users (2007) 13th edn.
2 British National Formulary issue no. 53 March 2007. British Medical Association and Royal Pharmaceutical Society of Great Britain. London, UK. (ISBN 085369-7312).
3
Bergen PJ, Li J, Rayner CR, et al. Colistin methanesulfonate is an inactive prodrug of colistin against Pseudomonas aeruginosa. Antimicrob Agents Chemother (2006) 50:1953–8.
4 Li J, Nation RL, Turnidge JD, et al. Colistin: the re-emerging antibiotic for multidrug-resistant Gram-negative bacterial infections. Lancet Infect Dis (2006) 6:589–601.[CrossRef][Web of Science][Medline]
5 Falagas ME, Kasiakou SK. Colistin: the revival of polymixins for the management of multidrug-resistant Gram-negative bacterial infections. Clin Infect Dis (2005) 40:1333–41.[CrossRef][Web of Science][Medline]
6 Falagas ME, Kasiakou SK. Toxicity of polymixins: a systematic review of the evidence from old and recent studies. Crit Care (2006) 10:R27.[CrossRef][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||