Comment on: MIC-based dose adjustment: facts and fables

Comment on: MIC-based dose adjustment: facts and fables Sir, We read with interest the article by Mouton et al.1 entitled ‘MIC-based dose adjustment: facts and fables’. The authors of that paper highlight the important issue of MIC variability when considering an MIC measurement for drug dosage adjustment. The authors state that the accuracy of individual MIC measurements is poor and that such measurements should not be used as guidance for dose adjustment. They suggest that the epidemiological cut-off (ECOFF) value should be used for dosing purposes when the measured MIC is within the WT range and ≤ECOFF. We certainly agree that the variability and uncertainty of measured MICs should not be ignored in the interpretation for drug dosing. However, the evidence and statistical consistency of the approach suggested by Mouton et al.1 for MIC interpretation are questionable. This approach appears overly conservative and may not be appropriate in all clinical settings. As pointed out by Mouton et al.,1 any variability in measured MICs has several components, including strain and assay, the latter including inter-laboratory and intra-laboratory variability. Mouton et al.1 state that when MICs are determined several times in different laboratories, the overall assay variability is greater than the between-strain variability, and this jeopardizes any interpretation of the result in terms of susceptibility. The authors provide a single study, presented in abstract form, to support this important statement.2 This work reported the reproducibility of linezolid MICs determined by eight Dutch laboratories with Etest. However, the results presented are somewhat limited. Apparently, no variance analysis was performed, and the magnitude of between-strain and assay variability is not clearly reported. Indeed, considerable assay variability in MIC determination is not a general fact. This has been well shown by Hoogkamp-Korstanje et al.,3 who performed a multicentre evaluation (n = 7 laboratories) of the MICs of four antimicrobial agents for a total of 141 strains from three bacterial species (Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis), comparing Etest with broth dilution as the reference method. For M. catarrhalis with ciprofloxacin and sparfloxacin, the Etest results showed excellent correlation with those from the reference method, with all measured MICs (n = 45) within ±1 2-fold dilution of the reference value. Considering that the range of MICs determined by broth dilution was 0.004–4 mg/L for ciprofloxacin (10 2-fold dilutions), these results indicate low assay variability of Etest compared with between-strain variability in this example. By contrast, the Etest results were much more variable and correlated poorly with the reference method for clarithromycin. Hence, the performance of MIC assays is quite variable, but not always poor. In another multicentre study (n = 17 laboratories in the Netherlands), van de Kassteele et al.4 also showed that the fold differences between laboratories for Neisseria gonorrhoeae mean that MICs were low compared with differences between strains. In this study, the authors used an interesting statistical approach to analyse MIC data. A linear mixed-effects regression model was used. The strain and antimicrobial agent were modelled as fixed effects, and the laboratory as a random effect. The model was fitted to MIC data and parameters were estimated by a Bayesian approach. We think that such statistical approaches should be encouraged in the analysis of MIC data as they permit separation and quantification of the various sources of variability in MIC measurement. In addition, confidence intervals can be calculated with these Bayesian approaches. A similar mixed-effects model was used by Wexler et al.5 to estimate the contribution of various components, including the bacterial isolate, the test, the reader and the replicate, in the determination of the MIC of cefoxitin for Bacteroides fragilis on agar medium. Interestingly, although the influences of isolate, test and reader were all significant for the 20 isolates, the replicate did not significantly contribute to the MIC variance. The model was used to calculate the probability of observing a specific MIC for various true MIC values. For example, for a true MIC of 24 mg/L, the calculated probabilities of reading 4, 8, 16, 32 and 64 mg/L were 0%, 0.5%, 15.6%, 59.1% and 23.8%, respectively. Conversely, those results showed that the lower the measured MIC, the lower the probability that the true MIC is a specific higher value, even considering the assay variability. Clearly, this finding does not support the suggestion of Mouton et al.1 to use the ECOFF for a measured MIC within the WT range, whatever the measurement is. Using the ECOFF value for drug dosing sounds reasonable when the imprecision of the MIC assay is large compared with the MIC WT range or a range of interest. An illustrative example is the vancomycin MIC for Staphylococcus aureus.6,7 When the assay imprecision is low compared with the range of possible MICs, using the ECOFF for dose adjustment when the measured MIC is low may lead to unnecessarily high dosage requirements and potentially toxic drug concentrations.8 To conclude, Mouton et al.1 are right to point out the importance of variability in MIC measurement. An individual MIC measurement should not be considered a true value, but rather an estimate of the susceptibility, the accuracy of which may be quite variable. Assay variability is undoubtedly an important component of the overall variability in MIC determination, and it depends on many parameters. However, poor performance of MIC assays is not a general characteristic, and the assay variability can be much lower than the between-strain variability in some cases. In addition, for a given measured MIC, the probability that the true MIC is equal to the ECOFF is not uniform over the MIC WT range, and therefore using the ECOFF as a reference for drug dose adjustment is not sound in all situations. The use of a statistical and quantitative modelling approach should be encouraged in laboratories to analyse data and assess the variability and reliability of MIC measurement. In addition, information on MIC assay accuracy should be provided along with the individual MIC results to better inform clinical decisions. Transparency declarations All authors: none to declare. References 1 Mouton JW, Muller AE, Canton R et al.   MIC-based dose adjustment: facts and fables. J Antimicrob Chemother  2018; 73: 564– 8. Google Scholar CrossRef Search ADS   2 Voss A, Mouton JW, Elzakker EP et al.   Linezolid susceptibility of glycopeptide-intermediately susceptible Staphylococcus aureus (GISA)—the Dutch experience. In: Abstracts of the Thirteenth European Congress of Clinical Microbiology and Infectious Diseases, Glasgow, UK, 2003. Abstract P1309. ESCMID, Basel, Switzerland. 3 Hoogkamp-Korstanje JA, Dirks-Go SI, Kabel P et al.   Multicentre in-vitro evaluation of the susceptibility of Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis to ciprofloxacin, clarithromycin, co-amoxiclav and sparfloxacin. J Antimicrob Chemother  1997; 39: 411– 4. Google Scholar CrossRef Search ADS PubMed  4 van de Kassteele J, van Santen-Verheuvel MG, Koedijk FD et al.   New statistical technique for analyzing MIC-based susceptibility data. Antimicrob Agents Chemother  2012; 56: 1557– 63. Google Scholar CrossRef Search ADS PubMed  5 Wexler HM, Lavin PT, Molitoris E et al.   Statistical analysis of the effects of trial, reader, and replicates on MIC determination for cefoxitin. Antimicrob Agents Chemother  1990; 34: 2246– 9. Google Scholar CrossRef Search ADS PubMed  6 Falcon R, Madrid S, Tormo N et al.   Intra- and interinstitutional evaluation of an Etest for vancomycin minimum inhibitory concentration measurement in Staphylococcus aureus blood isolates. Clin Infect Dis  2015; 61: 1490– 2. Google Scholar CrossRef Search ADS PubMed  7 Charlton CL, Hindler JA, Turnidge J et al.   Precision of vancomycin and daptomycin MICs for methicillin-resistant Staphylococcus aureus and effect of subculture and storage. J Clin Microbiol  2014; 52: 3898– 905. Google Scholar CrossRef Search ADS PubMed  8 Imani S, Buscher H, Marriott D et al.   Too much of a good thing: a retrospective study of β-lactam concentration–toxicity relationships. J Antimicrob Chemother  2017; 72: 2891– 7. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Antimicrobial Chemotherapy Oxford University Press

Comment on: MIC-based dose adjustment: facts and fables

Loading next page...
 
/lp/ou_press/comment-on-mic-based-dose-adjustment-facts-and-fables-USTDGO0TNZ
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please email: journals.permissions@oup.com.
ISSN
0305-7453
eISSN
1460-2091
D.O.I.
10.1093/jac/dky131
Publisher site
See Article on Publisher Site

Abstract

Sir, We read with interest the article by Mouton et al.1 entitled ‘MIC-based dose adjustment: facts and fables’. The authors of that paper highlight the important issue of MIC variability when considering an MIC measurement for drug dosage adjustment. The authors state that the accuracy of individual MIC measurements is poor and that such measurements should not be used as guidance for dose adjustment. They suggest that the epidemiological cut-off (ECOFF) value should be used for dosing purposes when the measured MIC is within the WT range and ≤ECOFF. We certainly agree that the variability and uncertainty of measured MICs should not be ignored in the interpretation for drug dosing. However, the evidence and statistical consistency of the approach suggested by Mouton et al.1 for MIC interpretation are questionable. This approach appears overly conservative and may not be appropriate in all clinical settings. As pointed out by Mouton et al.,1 any variability in measured MICs has several components, including strain and assay, the latter including inter-laboratory and intra-laboratory variability. Mouton et al.1 state that when MICs are determined several times in different laboratories, the overall assay variability is greater than the between-strain variability, and this jeopardizes any interpretation of the result in terms of susceptibility. The authors provide a single study, presented in abstract form, to support this important statement.2 This work reported the reproducibility of linezolid MICs determined by eight Dutch laboratories with Etest. However, the results presented are somewhat limited. Apparently, no variance analysis was performed, and the magnitude of between-strain and assay variability is not clearly reported. Indeed, considerable assay variability in MIC determination is not a general fact. This has been well shown by Hoogkamp-Korstanje et al.,3 who performed a multicentre evaluation (n = 7 laboratories) of the MICs of four antimicrobial agents for a total of 141 strains from three bacterial species (Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis), comparing Etest with broth dilution as the reference method. For M. catarrhalis with ciprofloxacin and sparfloxacin, the Etest results showed excellent correlation with those from the reference method, with all measured MICs (n = 45) within ±1 2-fold dilution of the reference value. Considering that the range of MICs determined by broth dilution was 0.004–4 mg/L for ciprofloxacin (10 2-fold dilutions), these results indicate low assay variability of Etest compared with between-strain variability in this example. By contrast, the Etest results were much more variable and correlated poorly with the reference method for clarithromycin. Hence, the performance of MIC assays is quite variable, but not always poor. In another multicentre study (n = 17 laboratories in the Netherlands), van de Kassteele et al.4 also showed that the fold differences between laboratories for Neisseria gonorrhoeae mean that MICs were low compared with differences between strains. In this study, the authors used an interesting statistical approach to analyse MIC data. A linear mixed-effects regression model was used. The strain and antimicrobial agent were modelled as fixed effects, and the laboratory as a random effect. The model was fitted to MIC data and parameters were estimated by a Bayesian approach. We think that such statistical approaches should be encouraged in the analysis of MIC data as they permit separation and quantification of the various sources of variability in MIC measurement. In addition, confidence intervals can be calculated with these Bayesian approaches. A similar mixed-effects model was used by Wexler et al.5 to estimate the contribution of various components, including the bacterial isolate, the test, the reader and the replicate, in the determination of the MIC of cefoxitin for Bacteroides fragilis on agar medium. Interestingly, although the influences of isolate, test and reader were all significant for the 20 isolates, the replicate did not significantly contribute to the MIC variance. The model was used to calculate the probability of observing a specific MIC for various true MIC values. For example, for a true MIC of 24 mg/L, the calculated probabilities of reading 4, 8, 16, 32 and 64 mg/L were 0%, 0.5%, 15.6%, 59.1% and 23.8%, respectively. Conversely, those results showed that the lower the measured MIC, the lower the probability that the true MIC is a specific higher value, even considering the assay variability. Clearly, this finding does not support the suggestion of Mouton et al.1 to use the ECOFF for a measured MIC within the WT range, whatever the measurement is. Using the ECOFF value for drug dosing sounds reasonable when the imprecision of the MIC assay is large compared with the MIC WT range or a range of interest. An illustrative example is the vancomycin MIC for Staphylococcus aureus.6,7 When the assay imprecision is low compared with the range of possible MICs, using the ECOFF for dose adjustment when the measured MIC is low may lead to unnecessarily high dosage requirements and potentially toxic drug concentrations.8 To conclude, Mouton et al.1 are right to point out the importance of variability in MIC measurement. An individual MIC measurement should not be considered a true value, but rather an estimate of the susceptibility, the accuracy of which may be quite variable. Assay variability is undoubtedly an important component of the overall variability in MIC determination, and it depends on many parameters. However, poor performance of MIC assays is not a general characteristic, and the assay variability can be much lower than the between-strain variability in some cases. In addition, for a given measured MIC, the probability that the true MIC is equal to the ECOFF is not uniform over the MIC WT range, and therefore using the ECOFF as a reference for drug dose adjustment is not sound in all situations. The use of a statistical and quantitative modelling approach should be encouraged in laboratories to analyse data and assess the variability and reliability of MIC measurement. In addition, information on MIC assay accuracy should be provided along with the individual MIC results to better inform clinical decisions. Transparency declarations All authors: none to declare. References 1 Mouton JW, Muller AE, Canton R et al.   MIC-based dose adjustment: facts and fables. J Antimicrob Chemother  2018; 73: 564– 8. Google Scholar CrossRef Search ADS   2 Voss A, Mouton JW, Elzakker EP et al.   Linezolid susceptibility of glycopeptide-intermediately susceptible Staphylococcus aureus (GISA)—the Dutch experience. In: Abstracts of the Thirteenth European Congress of Clinical Microbiology and Infectious Diseases, Glasgow, UK, 2003. Abstract P1309. ESCMID, Basel, Switzerland. 3 Hoogkamp-Korstanje JA, Dirks-Go SI, Kabel P et al.   Multicentre in-vitro evaluation of the susceptibility of Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis to ciprofloxacin, clarithromycin, co-amoxiclav and sparfloxacin. J Antimicrob Chemother  1997; 39: 411– 4. Google Scholar CrossRef Search ADS PubMed  4 van de Kassteele J, van Santen-Verheuvel MG, Koedijk FD et al.   New statistical technique for analyzing MIC-based susceptibility data. Antimicrob Agents Chemother  2012; 56: 1557– 63. Google Scholar CrossRef Search ADS PubMed  5 Wexler HM, Lavin PT, Molitoris E et al.   Statistical analysis of the effects of trial, reader, and replicates on MIC determination for cefoxitin. Antimicrob Agents Chemother  1990; 34: 2246– 9. Google Scholar CrossRef Search ADS PubMed  6 Falcon R, Madrid S, Tormo N et al.   Intra- and interinstitutional evaluation of an Etest for vancomycin minimum inhibitory concentration measurement in Staphylococcus aureus blood isolates. Clin Infect Dis  2015; 61: 1490– 2. Google Scholar CrossRef Search ADS PubMed  7 Charlton CL, Hindler JA, Turnidge J et al.   Precision of vancomycin and daptomycin MICs for methicillin-resistant Staphylococcus aureus and effect of subculture and storage. J Clin Microbiol  2014; 52: 3898– 905. Google Scholar CrossRef Search ADS PubMed  8 Imani S, Buscher H, Marriott D et al.   Too much of a good thing: a retrospective study of β-lactam concentration–toxicity relationships. J Antimicrob Chemother  2017; 72: 2891– 7. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Journal of Antimicrobial ChemotherapyOxford University Press

Published: Apr 11, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off