- Email: [email protected]
Variability of piperacillin concentrations in relation to tazobactam concentrations in critically ill patients
ARTICLE IN PRESS International Journal of Antimicrobial Agents ■■ (2016) ■■–■■
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / i j a n t i m i c a g
Short Communication
Variability of piperacillin concentrations in relation to tazobactam concentrations in critically ill patients Johannes Zander a, Gundula Döbbeler b, Dorothea Nagel a, Christina Scharf b, Mikayil Huseyn-Zada a, Jette Jung c, Lorenz Frey b, Michael Vogeser a, Michael Zoller b,* a
Institute of Laboratory Medicine, Hospital of the Ludwig-Maximilians-University of Munich, Marchioninistrasse 15, 81377 Munich, Germany Department of Anesthesiology, Hospital of the Ludwig-Maximilians-University of Munich, Marchioninistrasse 15, 81377 Munich, Germany c Max von Pettenkofer-Institute, Ludwig-Maximilians-University of Munich, Marchioninistrasse 17, 81377 Munich, Germany b
A R T I C L E
I N F O
Article history: Received 15 December 2015 Accepted 18 June 2016 Keywords: Therapeutic drug monitoring Target range Intensive care unit Pharmacokinetics Antibiotics
A B S T R A C T
Therapeutic drug monitoring for critically ill patients receiving piperacillin/tazobactam is described as a useful tool. However, the minimum inhibitory concentration of piperacillin depends on a sufficiently high concentration of tazobactam in case of β-lactamase-producing strains. Therefore, the relationship between piperacillin and tazobactam concentrations was assessed in a heterogeneous group of critically ill patients. Sixty patients with severe infections receiving 4.5 g of piperacillin/tazobactam 2–3 times daily by intermittent infusion were included in this prospective observational study (NCT01793012). Over 4 days, multiple serum samples were obtained to determine the total piperacillin and tazobactam concentrations. The target ranges were defined as trough levels >16 mg/L (>22.5 mg/L) and >4 mg/L (>5.7 mg/ L) for the calculated unbound concentrations (measured total concentrations) of piperacillin and tazobactam, respectively. Despite a high correlation coefficient (r = 0.93) comparing piperacillin and tazobactam trough levels, the piperacillin/tazobactam quotients varied between ca. 1 and 10. From linear regression analysis of piperacillin versus tazobactam values, it follows that a piperacillin trough level of 22.5 mg/L might be associated with tazobactam trough levels ranging from 1.5 mg/L to 10.1 mg/L. A 70 mg/L threshold for total piperacillin trough levels would be necessary to ensure that tazobactam concentrations are >5.7 mg/ L. Because of the observed variability of piperacillin/tazobactam quotients, defining the total piperacillin target range ≥70 mg/L might be useful to ensure that tazobactam concentrations do not fall below 5.7 mg/ L. Further studies are necessary to confirm that the used therapeutic ranges are associated with optimal outcomes in critically ill patients. © 2016 Elsevier B.V. and International Society of Chemotherapy. All rights reserved.
1. Introduction Piperacillin/tazobactam is one of the most frequently used antibiotics in intensive care units (ICUs). Piperacillin concentrations vary substantially following intermittent administration in critically ill patients, which often results in potentially subtherapeutic drug concentrations. Consequently, therapeutic drug monitoring (TDM) of piperacillin has been suggested as a useful tool in ICU patients [1,2]. Quantification of tazobactam is mostly assumed to be dispensable because it was shown several years ago that the pharmacokinetics of piperacillin and tazobactam were similar. In a study by Occhipinti et al, the mean total body clearance and volume of distribution for piperacillin were 182 mL/min/1.73 m2 and 0.152 L/
* Corresponding author. Department of Anesthesiology, Hospital of the LudwigMaximilians-University of Munich, Marchioninistrasse 15, 81377 Munich, Germany. Fax: +49 89 4400 78886. E-mail address: [email protected] (M. Zoller).
kg, respectively, and those for tazobactam were 185 mL/min/ 1.73 m2 and 0.170 L/kg [3]. It might be reasonable to re-assess whether tazobactam quantification is dispensable because individual studies have measured tazobactam in ICU patients and these studies found greater differences in the pharmacokinetics between piperacillin and tazobactam [4,5]. Furthermore, the pharmacokinetic/ pharmacodynamic (PK/PD) underpinnings of β-lactamase inhibitors have been elucidated [6–9] and thresholds for tazobactam concentrations needed to effectively kill single bacterial strains in combination with β-lactam antibiotics have recently been found [6,8,9]. Therefore, it might be interesting to determine whether tazobactam is often underdosed in critically ill patients. This consideration is especially of interest because different studies have proposed to decrease the piperacillin target ranges for isolates with low minimum inhibitory concentrations (MICs) [1,2], and this suggestion has found its way into clinical practice [10]. Subtherapeutic concentrations of β-lactam inhibitors might lead to therapy failure and also promote resistance [11], which could exacerbate the problem of multiresistant strains in ICUs.
http://dx.doi.org/10.1016/j.ijantimicag.2016.06.013 0924-8579/© 2016 Elsevier B.V. and International Society of Chemotherapy. All rights reserved.
Please cite this article in press as: Johannes Zander, et al., Variability of piperacillin concentrations in relation to tazobactam concentrations in critically ill patients, International Journal of Antimicrobial Agents (2016), doi: 10.1016/j.ijantimicag.2016.06.013
ARTICLE IN PRESS J. Zander et al. / International Journal of Antimicrobial Agents ■■ (2016) ■■–■■
2
In this study, we aimed to assess the relationship between piperacillin and tazobactam concentrations in critically ill patients and to evaluate the piperacillin concentrations necessary to ensure tazobactam concentrations >5.7 mg/L. 2. Materials and methods The study design and inclusion and exclusion criteria were recently described in detail [12]. Briefly, 60 consecutive ICU patients with severe infections who met the inclusion criteria were enrolled in this prospective observational study (ClinicalTrials.gov ID: NCT01793012). The study was approved by the Institutional Ethics Committee of Ludwig-Maximilians-University (Munich, Germany) and was performed according to the Declaration of Helsinki. Written informed consent was obtained from each patient. Multiple blood samples (median 29 per patient) were taken over 4 days immediately before (trough level), during and after all piperacillin/ tazobactam administrations on Day 1 and one piperacillin/ tazobactam administration on Days 2–4. The total piperacillin and tazobactam serum concentrations were determined by ultra-highperformance liquid chromatography/tandem mass spectrometry (UHPLC-MS/MS) as recently reported [13]. Several clinical and laboratory parameters were determined. Depending on the point of time of informed consent and organisational aspects, the beginning of the study varied within the first 48 h after the initiation of antibiotic treatment (4.5 g piperacillin/ tazobactam two to three times daily by short-duration infusion). The threshold for the potential therapeutic efficacy of total piperacillin was defined as >22.5 mg/L for trough levels corresponding to a percentage of time above the MIC (%T>MIC) of 100% [14,15] using an assumed MIC of 16 mg/L [15] and taking into consideration an average protein binding fraction of 30% [16]. The threshold of the tazobactam target range was defined as 100%T>5.7 mg/L. Taking an average protein binding fraction of 30% into account [8], this threshold corresponds to a threshold of 4 mg/L free tazobactam. This threshold was selected in accordance with the results of VanScoy et al and Melchers et al [6,8,9] and because antimicrobial susceptibility testing according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) is performed with a fixed concentration of 4 mg/L tazobactam. To describe the relationship between piperacillin and tazobactam concentrations, linear regression analyses and Spearman’s correlation were used. 3. Results Patient characteristics and piperacillin concentrations have already been described elsewhere [12]. Briefly, the 60 patients exhibited high heterogeneity with respect to their clinical and laboratory characteristics. Of the 17 female and 43 male patients, 3 patients were treated with extracorporeal lung assist and 10 were treated with renal replacement therapy (RRT). The median [interquartile range (IQR)] age was 63 years (54–75 years). The median body mass index (BMI) was 27 kg/m2 (IQR, 24–29 kg/m2). The median Acute Physiology and Chronic Health Evaluation (APACHE) score was 24 (IQR, 18–31) and the median creatinine clearance (24-h urine) of non-RRT patients was 60 mL/min (IQR, 24–119 mL/min). Piperacillin and tazobactam trough levels varied interindividually by ≥123-fold [12] and ≥192-fold, respectively, on study Days 1, 2, 3 and 4. On Day 1, potentially inadequate trough levels were observed for piperacillin in 47% of patients [12] and for tazobactam in 50%. Analysis of the trough concentrations of both antimicrobial agents in blood samples revealed a correlation coefficient (r) of logarithmic values of 0.93 (Fig. 1a). A similar correlation coefficient (r = 0.97) was observed when only one trough level (on Day 1) was used for
each patient (Supplementary Fig. S1). Despite these high correlation coefficients, the piperacillin/tazobactam quotients of trough levels showed a relatively high variability, ranging from 1.3 to 10.4, which was independent of the piperacillin concentrations observed (Supplementary Fig. S2a). All piperacillin/tazobactam quotients varied depending on the blood sampling time in relation to the beginning of the last antibiotic infusion (ca. 4.5–10 after 15 min; ca. 1.5–10 after 12 h) (Supplementary Fig. S2b). From the linear regression analysis of piperacillin versus tazobactam values, it follows that a piperacillin trough level of 22.5 mg/L might be associated with tazobactam trough levels ranging from 1.5 mg/L to 10.1 mg/L (Fig. 1b). In this study, 8% of blood samples with piperacillin trough values of >22.5 mg/L had tazobactam trough levels <5.7 mg/ L. Taking the 95% confidence interval (CI) of the regression line of piperacillin versus tazobactam into account, a 70 mg/L threshold for total piperacillin trough levels would be necessary to ensure that tazobactam concentrations are >5.7 mg/L. When using a lower piperacillin target threshold of 2 mg/L, which might be used in studies [1,2] or in clinical practice [10], 41% of the blood samples with adequate piperacillin concentrations would have potentially tazobactam levels <5.7 mg/L (Fig. 2). 4. Discussion Piperacillin/tazobactam is a combination antibiotic commonly used in ICU patients. A number of TDM studies on this substance have recently been published. However, most of these studies have focused on the piperacillin component only. The aim of the current study was to elucidate the pharmacokinetics of the tazobactam portion as well as the correlation between the drug components. A high variability of piperacillin and tazobactam concentrations was observed, leading on study Day 1 to potentially subtherapeutic concentrations in ca. 50% of patients for both antimicrobial agents. Although a high correlation coefficient was observed, the quotients of the piperacillin and tazobactam trough levels varied from ca. 1 to 10. A high variability of piperacillin concentrations has been described in different studies of critically ill patients receiving piperacillin/tazobactam [4,5,14,16]; however, only individual studies also measured tazobactam [4,5]. The high variability observed in the current study led to potentially subtherapeutic levels in 47% of patients for piperacillin and 50% of patients for tazobactam on study Day 1. This finding might contribute to the mortality observed in ICU patients with severe infections. Some studies [17,18] describe that higher concentrations of individual β-lactams correlate with a better outcome. The target ranges used in this study were 100%T>22.5 mg/L for piperacillin and 100%T>5.7 mg/L for tazobactam. The target range for the calculated unbound concentration of piperacillin was defined in accordance with a number of studies in critically ill or immunocompromised patients [15,16]; however, other target ranges were used in some studies [1,2,10]. There is a lack of studies evaluating thresholds for tazobactam concentrations. Strayer et al suggested the area under the concentration–time curve (AUC) as the exposure measure best predicting tazobactam activity [7]. However, an incomplete dose fractionation design was utilised, resulting in an impossibility to clearly discriminate between the different exposure measures [9]. Louie et al suggested that for a β-lactam inhibitor by dose fractionation experiments, ‘time > threshold’ was the pharmacodynamic index linked to cell killing and resistance suppression [11]. This finding is in line with the study of VanScoy et al showing that for tazobactam, 100% of the time above the strain-specific threshold resulted in the best bacterial killing in an in vitro study in combination with a β-lactam antibiotic [8]. Moreover, thresholds as high as 4 mg/L for free tazobactam were detected in their limited number of wild-type strains containing β-lactamases.
Please cite this article in press as: Johannes Zander, et al., Variability of piperacillin concentrations in relation to tazobactam concentrations in critically ill patients, International Journal of Antimicrobial Agents (2016), doi: 10.1016/j.ijantimicag.2016.06.013
ARTICLE IN PRESS J. Zander et al. / International Journal of Antimicrobial Agents ■■ (2016) ■■–■■
3
Fig. 1. Piperacillin trough levels in relation to tazobactam trough levels. Areas I,II: piperacillin levels within their therapeutic range; areas I,III: tazobactam levels below their therapeutic range. The second trough level, if available, is presented per day and per patient. (a) The regression line for piperacillin in relation to tazobactam trough levels is shown (solid line); (b) with corresponding 95% confidence intervals (CIs) (dashed lines). The horizontal and vertical solid lines represent the total piperacillin and tazobactam target thresholds used in this study, respectively (22.5 mg/L for piperacillin and 5.7 mg/L for tazobactam). The red spots represent the points of intersection between the corresponding 95% CI and the used piperacillin target threshold, representing the variability of tazobactam concentrations for a given piperacillin concentration of 22.5 mg/L. The green dashed line represents a concentration of 70 mg/L piperacillin that is necessary to ensure that most tazobactam concentrations are >5.7 mg/L. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
This finding might be a reason to define the threshold for free tazobactam at a minimum of 4 mg/L [6,19]. This fixed concentration is used in the determination of MICs for piperacillin/tazobactam, e.g. by VITEK®2 or Etest (bioMérieux, Lyon, France); here, only the piperacillin concentrations vary. Therefore, strains that are determined to be piperacillin/tazobactam-sensitive by conventional methods might express β-lactamases that mediate resistance to piperacillin/tazobactam under conditions in which the tazobactam concentration is <4 mg/L. It therefore appears to be important that tazobactam concentrations are also above this threshold in vivo to ensure optimal clinical outcomes and to prevent the selection of strains with higher β-lactamase levels. Further studies are needed
to confirm that the used target thresholds are PK/PD parameters associated with optimal outcomes in critically ill patients. A high correlation coefficient of >0.9 between piperacillin and tazobactam trough levels was observed. Despite the high correlation coefficient, the piperacillin/tazobactam quotients still ranged from ca. 1 to 10. This variability has important implications for the predictability of tazobactam concentrations if only piperacillin concentrations are measured. We showed that, for example, a piperacillin trough level of 22.5 mg/L might be related to either potentially therapeutic or subtherapeutic tazobactam concentrations (Fig. 1b). For unknown causative strains, a frequently proposed target range would be 100%T>16 mg/L for unbound piperacillin (or 100%T>22.5 mg/L for total
Please cite this article in press as: Johannes Zander, et al., Variability of piperacillin concentrations in relation to tazobactam concentrations in critically ill patients, International Journal of Antimicrobial Agents (2016), doi: 10.1016/j.ijantimicag.2016.06.013
ARTICLE IN PRESS 4
J. Zander et al. / International Journal of Antimicrobial Agents ■■ (2016) ■■–■■
Fig. 2. Piperacillin trough levels in relation to tazobactam trough levels after an adaption of the piperacillin target threshold to 2 mg/L. The horizontal solid line represents a piperacillin target threshold of 2 mg/L, and the vertical line represents a tazobactam target threshold of 5.7 mg/L. Areas I,II: piperacillin levels within their therapeutic range; areas I,III: tazobactam levels below their therapeutic range. The second trough level, if available, is presented per day and per patient.
piperacillin). In this case, we observed 8% potentially subtherapeutic tazobactam concentrations. Moreover, a reduction in piperacillin target thresholds in cases of known causative strains with lower MICs, as proposed in different studies [1,2] and as applied in clinical practice [10], would greatly increase this percentage (Fig. 2). This practice might have a negative effect on patient outcomes and promote resistance development. According to the corresponding 95% CI in this study, only piperacillin concentrations ≥70 mg/L would mostly ensure that tazobactam concentrations would be within the defined therapeutic range (>5.7 mg/L) (Fig. 1b). Similar higher target ranges are occasionally used in the literature [1,2] and it is tempting to speculate that these piperacillin concentrations themselves might also increase efficacy [2] and reduce the development of resistance [20] without being within the toxic range [1]. This study investigated a highly heterogeneous patient group that might reflect the typical ICU population and the range of piperacillin/ tazobactam quotients commonly observed in this setting. However, there are patient subgroups that were not addressed. Therefore, there might be an even higher variation in the piperacillin/tazobactam quotients, and an even higher piperacillin target range might be preferable in cases where tazobactam is not measured. Moreover, it might be a limitation of this study that patients received 4.5 g of piperacillin/tazobactam only two to three times as short-duration infusions. Other regimens, which might also influence piperacillin/ tazobactam quotients, were not used. Furthermore, the unbound fraction of piperacillin and tazobactam was not measured. Finally, pharmacokinetic modelling was not used to describe piperacillin or tazobactam concentrations. Imprecise individual predictions might occasionally occur via pharmacokinetic modelling if the individual estimated parameter values substantially vary from the typical population, which might be a problem in a highly heterogeneous patient group, as in this study. In conclusion, although there was a high correlation coefficient when comparing piperacillin and tazobactam trough levels in critically ill patients, a variability of piperacillin/tazobactam quotients was found that might be important for therapy. Piperacillin trough levels within the therapeutic range might be associated with subtherapeutic or therapeutic tazobactam trough levels. If physi-
cians want to ensure that free tazobactam trough levels are >4 mg/ L, the current data indicate that total piperacillin trough levels ≥70 mg/L should be reached in critically ill patients. Further studies should verify these observations and determine whether similar recommendations are also applicable to patients with extended/ continuous piperacillin/tazobactam infusions to avoid tazobactam levels being consistently below the desired target range. Funding: This work was supported by a Mérieux Research Grant (Institut Mérieux, Lyon, France). The design, collection, analysis and interpretation of data as well as the writing and publication of the manuscript were undertaken by the authors without the participation or influence of the funding source. Competing interests: None declared. Ethical approval: The study protocol (NCT01793012) for this prospective observational study was approved by the Institutional Ethics Committee of Ludwig-Maximilians-University (Munich, Germany) [No. 428-12] and was performed according to the Declaration of Helsinki. Written informed consent was obtained from each patient. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ijantimicag.2016.06.013. References [1] Blondiaux N, Wallet F, Favory R, Onimus T, Nseir S, Courcol RJ, et al. Daily serum piperacillin monitoring is advisable in critically ill patients. Int J Antimicrob Agents 2010;35:500–3. [2] De Waele JJ, Carrette S, Carlier M, Stove V, Boelens J, Claeys G, et al. Therapeutic drug monitoring-based dose optimisation of piperacillin and meropenem: a randomised controlled trial. Intensive Care Med 2014;40:380–7. [3] Occhipinti DJ, Pendland SL, Schoonover LL, Rypins EB, Danziger LH, Rodvold KA. Pharmacokinetics and pharmacodynamics of two multiple-dose piperacillin–tazobactam regimens. Antimicrob Agents Chemother 1997;41:2511–17. [4] Asín-Prieto E, Rodríguez-Gascón A, Trocóniz IF, Soraluce A, Maynar J, Sánchez-Izquierdo JÁ, et al. Population pharmacokinetics of piperacillin and tazobactam in critically ill patients undergoing continuous renal replacement therapy: application to pharmacokinetic/pharmacodynamic analysis. J Antimicrob Chemother 2014;69:180–9.
Please cite this article in press as: Johannes Zander, et al., Variability of piperacillin concentrations in relation to tazobactam concentrations in critically ill patients, International Journal of Antimicrobial Agents (2016), doi: 10.1016/j.ijantimicag.2016.06.013
ARTICLE IN PRESS J. Zander et al. / International Journal of Antimicrobial Agents ■■ (2016) ■■–■■
[5] Varghese JM, Jarrett P, Boots RJ, Kirkpatrick CM, Lipman J, Roberts JA. Pharmacokinetics of piperacillin and tazobactam in plasma and subcutaneous interstitial fluid in critically ill patients receiving continuous venovenous haemodiafiltration. Int J Antimicrob Agents 2014;43:343–8. [6] Melchers MJ, van Mil AC, Mouton JW. In vitro activity of ceftolozane alone and in combination with tazobactam against extended-spectrum-β-lactamaseharboring Enterobacteriaceae. Antimicrob Agents Chemother 2015;59:4521–5. [7] Strayer AH, Gilbert DH, Pivarnik P, Medeiros AA, Zinner SH, Dudley MN. Pharmacodynamics of piperacillin alone and in combination with tazobactam against piperacillin-resistant and -susceptible organisms in an in vitro model of infection. Antimicrob Agents Chemother 1994;38:2351–6. [8] VanScoy B, Mendes RE, McCauley J, Bhavnani SM, Bulik CC, Okusanya OO, et al. Pharmacological basis of β-lactamase inhibitor therapeutics: tazobactam in combination with ceftolozane. Antimicrob Agents Chemother 2013;57:5924–30. [9] VanScoy B, Mendes RE, Nicasio AM, Castanheira M, Bulik CC, Okusanya OO, et al. Pharmacokinetics–pharmacodynamics of tazobactam in combination with ceftolozane in an in vitro infection model. Antimicrob Agents Chemother 2013;57:2809–14. [10] Wong G, Brinkman A, Benefield RJ, Carlier M, De Waele JJ, El Helali N, et al. An international, multicentre survey of β-lactam antibiotic therapeutic drug monitoring practice in intensive care units. J Antimicrob Chemother 2014;69:1416–23. [11] Louie A, Castanheira M, Liu W, Grasso C, Jones RN, Williams G, et al. Pharmacodynamics of β-lactamase inhibition by NXL104 in combination with ceftaroline: examining organisms with multiple types of β-lactamases. Antimicrob Agents Chemother 2012;56:258–70. [12] Zander J, Döbbeler G, Nagel D, Maier B, Scharf C, Huseyn-Zada M, et al. Piperacillin concentration in relation to therapeutic range in critically ill patients—a prospective observational study. Crit Care 2016;20:79.
5
[13] Zander J, Maier B, Suhr A, Zoller M, Frey L, Teupser D, et al. Quantification of piperacillin, tazobactam, cefepime, meropenem, ciprofloxacin and linezolid in serum using an isotope dilution UHPLC-MS/MS method with semi-automated sample preparation. Clin Chem Lab Med 2015;53:781–91. [14] Roberts JA, Paul SK, Akova M, Bassetti M, De Waele JJ, Dimopoulos G, et al. DALI: Defining Antibiotic Levels in Intensive care unit patients: are current β-lactam antibiotic doses sufficient for critically ill patients? Clin Infect Dis 2014;58:1072–83. [15] Udy AA, Lipman J, Jarrett P, Klein K, Wallis SC, Patel K, et al. Are standard doses of piperacillin sufficient for critically ill patients with augmented creatinine clearance? Crit Care 2015;19:28. [16] Sime FB, Roberts MS, Warner MS, Hahn U, Robertson TA, Yeend S, et al. Altered pharmacokinetics of piperacillin in febrile neutropenic patients with hematological malignancy. Antimicrob Agents Chemother 2014;58:3533–7. [17] Ariano RE, Nyhlén A, Donnelly JP, Sitar DS, Harding GK, Zelenitsky SA. Pharmacokinetics and pharmacodynamics of meropenem in febrile neutropenic patients with bacteremia. Ann Pharmacother 2005;39:32–8. [18] McKinnon PS, Paladino JA, Schentag JJ. Evaluation of area under the inhibitory curve (AUIC) and time above the minimum inhibitory concentration (T>MIC) as predictors of outcome for cefepime and ceftazidime in serious bacterial infections. Int J Antimicrob Agents 2008;31:345–51. [19] Mueller SC, Majcher-Peszynska J, Hickstein H, Francke A, Pertschy A, Schulz M, et al. Pharmacokinetics of piperacillin–tazobactam in anuric intensive care patients during continuous venovenous hemodialysis. Antimicrob Agents Chemother 2002;46:1557–60. [20] Felton TW, Goodwin J, O’Connor L, Sharp A, Gregson L, Livermore J, et al. Impact of bolus dosing versus continuous infusion of piperacillin and tazobactam on the development of antimicrobial resistance in Pseudomonas aeruginosa. Antimicrob Agents Chemother 2013;57:5811–19.
Please cite this article in press as: Johannes Zander, et al., Variability of piperacillin concentrations in relation to tazobactam concentrations in critically ill patients, International Journal of Antimicrobial Agents (2016), doi: 10.1016/j.ijantimicag.2016.06.013
Contents lists available at ScienceDirect
International Journal of Antimicrobial Agents j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / i j a n t i m i c a g
Short Communication
Variability of piperacillin concentrations in relation to tazobactam concentrations in critically ill patients Johannes Zander a, Gundula Döbbeler b, Dorothea Nagel a, Christina Scharf b, Mikayil Huseyn-Zada a, Jette Jung c, Lorenz Frey b, Michael Vogeser a, Michael Zoller b,* a
Institute of Laboratory Medicine, Hospital of the Ludwig-Maximilians-University of Munich, Marchioninistrasse 15, 81377 Munich, Germany Department of Anesthesiology, Hospital of the Ludwig-Maximilians-University of Munich, Marchioninistrasse 15, 81377 Munich, Germany c Max von Pettenkofer-Institute, Ludwig-Maximilians-University of Munich, Marchioninistrasse 17, 81377 Munich, Germany b
A R T I C L E
I N F O
Article history: Received 15 December 2015 Accepted 18 June 2016 Keywords: Therapeutic drug monitoring Target range Intensive care unit Pharmacokinetics Antibiotics
A B S T R A C T
Therapeutic drug monitoring for critically ill patients receiving piperacillin/tazobactam is described as a useful tool. However, the minimum inhibitory concentration of piperacillin depends on a sufficiently high concentration of tazobactam in case of β-lactamase-producing strains. Therefore, the relationship between piperacillin and tazobactam concentrations was assessed in a heterogeneous group of critically ill patients. Sixty patients with severe infections receiving 4.5 g of piperacillin/tazobactam 2–3 times daily by intermittent infusion were included in this prospective observational study (NCT01793012). Over 4 days, multiple serum samples were obtained to determine the total piperacillin and tazobactam concentrations. The target ranges were defined as trough levels >16 mg/L (>22.5 mg/L) and >4 mg/L (>5.7 mg/ L) for the calculated unbound concentrations (measured total concentrations) of piperacillin and tazobactam, respectively. Despite a high correlation coefficient (r = 0.93) comparing piperacillin and tazobactam trough levels, the piperacillin/tazobactam quotients varied between ca. 1 and 10. From linear regression analysis of piperacillin versus tazobactam values, it follows that a piperacillin trough level of 22.5 mg/L might be associated with tazobactam trough levels ranging from 1.5 mg/L to 10.1 mg/L. A 70 mg/L threshold for total piperacillin trough levels would be necessary to ensure that tazobactam concentrations are >5.7 mg/ L. Because of the observed variability of piperacillin/tazobactam quotients, defining the total piperacillin target range ≥70 mg/L might be useful to ensure that tazobactam concentrations do not fall below 5.7 mg/ L. Further studies are necessary to confirm that the used therapeutic ranges are associated with optimal outcomes in critically ill patients. © 2016 Elsevier B.V. and International Society of Chemotherapy. All rights reserved.
1. Introduction Piperacillin/tazobactam is one of the most frequently used antibiotics in intensive care units (ICUs). Piperacillin concentrations vary substantially following intermittent administration in critically ill patients, which often results in potentially subtherapeutic drug concentrations. Consequently, therapeutic drug monitoring (TDM) of piperacillin has been suggested as a useful tool in ICU patients [1,2]. Quantification of tazobactam is mostly assumed to be dispensable because it was shown several years ago that the pharmacokinetics of piperacillin and tazobactam were similar. In a study by Occhipinti et al, the mean total body clearance and volume of distribution for piperacillin were 182 mL/min/1.73 m2 and 0.152 L/
* Corresponding author. Department of Anesthesiology, Hospital of the LudwigMaximilians-University of Munich, Marchioninistrasse 15, 81377 Munich, Germany. Fax: +49 89 4400 78886. E-mail address: [email protected] (M. Zoller).
kg, respectively, and those for tazobactam were 185 mL/min/ 1.73 m2 and 0.170 L/kg [3]. It might be reasonable to re-assess whether tazobactam quantification is dispensable because individual studies have measured tazobactam in ICU patients and these studies found greater differences in the pharmacokinetics between piperacillin and tazobactam [4,5]. Furthermore, the pharmacokinetic/ pharmacodynamic (PK/PD) underpinnings of β-lactamase inhibitors have been elucidated [6–9] and thresholds for tazobactam concentrations needed to effectively kill single bacterial strains in combination with β-lactam antibiotics have recently been found [6,8,9]. Therefore, it might be interesting to determine whether tazobactam is often underdosed in critically ill patients. This consideration is especially of interest because different studies have proposed to decrease the piperacillin target ranges for isolates with low minimum inhibitory concentrations (MICs) [1,2], and this suggestion has found its way into clinical practice [10]. Subtherapeutic concentrations of β-lactam inhibitors might lead to therapy failure and also promote resistance [11], which could exacerbate the problem of multiresistant strains in ICUs.
http://dx.doi.org/10.1016/j.ijantimicag.2016.06.013 0924-8579/© 2016 Elsevier B.V. and International Society of Chemotherapy. All rights reserved.
Please cite this article in press as: Johannes Zander, et al., Variability of piperacillin concentrations in relation to tazobactam concentrations in critically ill patients, International Journal of Antimicrobial Agents (2016), doi: 10.1016/j.ijantimicag.2016.06.013
ARTICLE IN PRESS J. Zander et al. / International Journal of Antimicrobial Agents ■■ (2016) ■■–■■
2
In this study, we aimed to assess the relationship between piperacillin and tazobactam concentrations in critically ill patients and to evaluate the piperacillin concentrations necessary to ensure tazobactam concentrations >5.7 mg/L. 2. Materials and methods The study design and inclusion and exclusion criteria were recently described in detail [12]. Briefly, 60 consecutive ICU patients with severe infections who met the inclusion criteria were enrolled in this prospective observational study (ClinicalTrials.gov ID: NCT01793012). The study was approved by the Institutional Ethics Committee of Ludwig-Maximilians-University (Munich, Germany) and was performed according to the Declaration of Helsinki. Written informed consent was obtained from each patient. Multiple blood samples (median 29 per patient) were taken over 4 days immediately before (trough level), during and after all piperacillin/ tazobactam administrations on Day 1 and one piperacillin/ tazobactam administration on Days 2–4. The total piperacillin and tazobactam serum concentrations were determined by ultra-highperformance liquid chromatography/tandem mass spectrometry (UHPLC-MS/MS) as recently reported [13]. Several clinical and laboratory parameters were determined. Depending on the point of time of informed consent and organisational aspects, the beginning of the study varied within the first 48 h after the initiation of antibiotic treatment (4.5 g piperacillin/ tazobactam two to three times daily by short-duration infusion). The threshold for the potential therapeutic efficacy of total piperacillin was defined as >22.5 mg/L for trough levels corresponding to a percentage of time above the MIC (%T>MIC) of 100% [14,15] using an assumed MIC of 16 mg/L [15] and taking into consideration an average protein binding fraction of 30% [16]. The threshold of the tazobactam target range was defined as 100%T>5.7 mg/L. Taking an average protein binding fraction of 30% into account [8], this threshold corresponds to a threshold of 4 mg/L free tazobactam. This threshold was selected in accordance with the results of VanScoy et al and Melchers et al [6,8,9] and because antimicrobial susceptibility testing according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) is performed with a fixed concentration of 4 mg/L tazobactam. To describe the relationship between piperacillin and tazobactam concentrations, linear regression analyses and Spearman’s correlation were used. 3. Results Patient characteristics and piperacillin concentrations have already been described elsewhere [12]. Briefly, the 60 patients exhibited high heterogeneity with respect to their clinical and laboratory characteristics. Of the 17 female and 43 male patients, 3 patients were treated with extracorporeal lung assist and 10 were treated with renal replacement therapy (RRT). The median [interquartile range (IQR)] age was 63 years (54–75 years). The median body mass index (BMI) was 27 kg/m2 (IQR, 24–29 kg/m2). The median Acute Physiology and Chronic Health Evaluation (APACHE) score was 24 (IQR, 18–31) and the median creatinine clearance (24-h urine) of non-RRT patients was 60 mL/min (IQR, 24–119 mL/min). Piperacillin and tazobactam trough levels varied interindividually by ≥123-fold [12] and ≥192-fold, respectively, on study Days 1, 2, 3 and 4. On Day 1, potentially inadequate trough levels were observed for piperacillin in 47% of patients [12] and for tazobactam in 50%. Analysis of the trough concentrations of both antimicrobial agents in blood samples revealed a correlation coefficient (r) of logarithmic values of 0.93 (Fig. 1a). A similar correlation coefficient (r = 0.97) was observed when only one trough level (on Day 1) was used for
each patient (Supplementary Fig. S1). Despite these high correlation coefficients, the piperacillin/tazobactam quotients of trough levels showed a relatively high variability, ranging from 1.3 to 10.4, which was independent of the piperacillin concentrations observed (Supplementary Fig. S2a). All piperacillin/tazobactam quotients varied depending on the blood sampling time in relation to the beginning of the last antibiotic infusion (ca. 4.5–10 after 15 min; ca. 1.5–10 after 12 h) (Supplementary Fig. S2b). From the linear regression analysis of piperacillin versus tazobactam values, it follows that a piperacillin trough level of 22.5 mg/L might be associated with tazobactam trough levels ranging from 1.5 mg/L to 10.1 mg/L (Fig. 1b). In this study, 8% of blood samples with piperacillin trough values of >22.5 mg/L had tazobactam trough levels <5.7 mg/ L. Taking the 95% confidence interval (CI) of the regression line of piperacillin versus tazobactam into account, a 70 mg/L threshold for total piperacillin trough levels would be necessary to ensure that tazobactam concentrations are >5.7 mg/L. When using a lower piperacillin target threshold of 2 mg/L, which might be used in studies [1,2] or in clinical practice [10], 41% of the blood samples with adequate piperacillin concentrations would have potentially tazobactam levels <5.7 mg/L (Fig. 2). 4. Discussion Piperacillin/tazobactam is a combination antibiotic commonly used in ICU patients. A number of TDM studies on this substance have recently been published. However, most of these studies have focused on the piperacillin component only. The aim of the current study was to elucidate the pharmacokinetics of the tazobactam portion as well as the correlation between the drug components. A high variability of piperacillin and tazobactam concentrations was observed, leading on study Day 1 to potentially subtherapeutic concentrations in ca. 50% of patients for both antimicrobial agents. Although a high correlation coefficient was observed, the quotients of the piperacillin and tazobactam trough levels varied from ca. 1 to 10. A high variability of piperacillin concentrations has been described in different studies of critically ill patients receiving piperacillin/tazobactam [4,5,14,16]; however, only individual studies also measured tazobactam [4,5]. The high variability observed in the current study led to potentially subtherapeutic levels in 47% of patients for piperacillin and 50% of patients for tazobactam on study Day 1. This finding might contribute to the mortality observed in ICU patients with severe infections. Some studies [17,18] describe that higher concentrations of individual β-lactams correlate with a better outcome. The target ranges used in this study were 100%T>22.5 mg/L for piperacillin and 100%T>5.7 mg/L for tazobactam. The target range for the calculated unbound concentration of piperacillin was defined in accordance with a number of studies in critically ill or immunocompromised patients [15,16]; however, other target ranges were used in some studies [1,2,10]. There is a lack of studies evaluating thresholds for tazobactam concentrations. Strayer et al suggested the area under the concentration–time curve (AUC) as the exposure measure best predicting tazobactam activity [7]. However, an incomplete dose fractionation design was utilised, resulting in an impossibility to clearly discriminate between the different exposure measures [9]. Louie et al suggested that for a β-lactam inhibitor by dose fractionation experiments, ‘time > threshold’ was the pharmacodynamic index linked to cell killing and resistance suppression [11]. This finding is in line with the study of VanScoy et al showing that for tazobactam, 100% of the time above the strain-specific threshold resulted in the best bacterial killing in an in vitro study in combination with a β-lactam antibiotic [8]. Moreover, thresholds as high as 4 mg/L for free tazobactam were detected in their limited number of wild-type strains containing β-lactamases.
Please cite this article in press as: Johannes Zander, et al., Variability of piperacillin concentrations in relation to tazobactam concentrations in critically ill patients, International Journal of Antimicrobial Agents (2016), doi: 10.1016/j.ijantimicag.2016.06.013
ARTICLE IN PRESS J. Zander et al. / International Journal of Antimicrobial Agents ■■ (2016) ■■–■■
3
Fig. 1. Piperacillin trough levels in relation to tazobactam trough levels. Areas I,II: piperacillin levels within their therapeutic range; areas I,III: tazobactam levels below their therapeutic range. The second trough level, if available, is presented per day and per patient. (a) The regression line for piperacillin in relation to tazobactam trough levels is shown (solid line); (b) with corresponding 95% confidence intervals (CIs) (dashed lines). The horizontal and vertical solid lines represent the total piperacillin and tazobactam target thresholds used in this study, respectively (22.5 mg/L for piperacillin and 5.7 mg/L for tazobactam). The red spots represent the points of intersection between the corresponding 95% CI and the used piperacillin target threshold, representing the variability of tazobactam concentrations for a given piperacillin concentration of 22.5 mg/L. The green dashed line represents a concentration of 70 mg/L piperacillin that is necessary to ensure that most tazobactam concentrations are >5.7 mg/L. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
This finding might be a reason to define the threshold for free tazobactam at a minimum of 4 mg/L [6,19]. This fixed concentration is used in the determination of MICs for piperacillin/tazobactam, e.g. by VITEK®2 or Etest (bioMérieux, Lyon, France); here, only the piperacillin concentrations vary. Therefore, strains that are determined to be piperacillin/tazobactam-sensitive by conventional methods might express β-lactamases that mediate resistance to piperacillin/tazobactam under conditions in which the tazobactam concentration is <4 mg/L. It therefore appears to be important that tazobactam concentrations are also above this threshold in vivo to ensure optimal clinical outcomes and to prevent the selection of strains with higher β-lactamase levels. Further studies are needed
to confirm that the used target thresholds are PK/PD parameters associated with optimal outcomes in critically ill patients. A high correlation coefficient of >0.9 between piperacillin and tazobactam trough levels was observed. Despite the high correlation coefficient, the piperacillin/tazobactam quotients still ranged from ca. 1 to 10. This variability has important implications for the predictability of tazobactam concentrations if only piperacillin concentrations are measured. We showed that, for example, a piperacillin trough level of 22.5 mg/L might be related to either potentially therapeutic or subtherapeutic tazobactam concentrations (Fig. 1b). For unknown causative strains, a frequently proposed target range would be 100%T>16 mg/L for unbound piperacillin (or 100%T>22.5 mg/L for total
Please cite this article in press as: Johannes Zander, et al., Variability of piperacillin concentrations in relation to tazobactam concentrations in critically ill patients, International Journal of Antimicrobial Agents (2016), doi: 10.1016/j.ijantimicag.2016.06.013
ARTICLE IN PRESS 4
J. Zander et al. / International Journal of Antimicrobial Agents ■■ (2016) ■■–■■
Fig. 2. Piperacillin trough levels in relation to tazobactam trough levels after an adaption of the piperacillin target threshold to 2 mg/L. The horizontal solid line represents a piperacillin target threshold of 2 mg/L, and the vertical line represents a tazobactam target threshold of 5.7 mg/L. Areas I,II: piperacillin levels within their therapeutic range; areas I,III: tazobactam levels below their therapeutic range. The second trough level, if available, is presented per day and per patient.
piperacillin). In this case, we observed 8% potentially subtherapeutic tazobactam concentrations. Moreover, a reduction in piperacillin target thresholds in cases of known causative strains with lower MICs, as proposed in different studies [1,2] and as applied in clinical practice [10], would greatly increase this percentage (Fig. 2). This practice might have a negative effect on patient outcomes and promote resistance development. According to the corresponding 95% CI in this study, only piperacillin concentrations ≥70 mg/L would mostly ensure that tazobactam concentrations would be within the defined therapeutic range (>5.7 mg/L) (Fig. 1b). Similar higher target ranges are occasionally used in the literature [1,2] and it is tempting to speculate that these piperacillin concentrations themselves might also increase efficacy [2] and reduce the development of resistance [20] without being within the toxic range [1]. This study investigated a highly heterogeneous patient group that might reflect the typical ICU population and the range of piperacillin/ tazobactam quotients commonly observed in this setting. However, there are patient subgroups that were not addressed. Therefore, there might be an even higher variation in the piperacillin/tazobactam quotients, and an even higher piperacillin target range might be preferable in cases where tazobactam is not measured. Moreover, it might be a limitation of this study that patients received 4.5 g of piperacillin/tazobactam only two to three times as short-duration infusions. Other regimens, which might also influence piperacillin/ tazobactam quotients, were not used. Furthermore, the unbound fraction of piperacillin and tazobactam was not measured. Finally, pharmacokinetic modelling was not used to describe piperacillin or tazobactam concentrations. Imprecise individual predictions might occasionally occur via pharmacokinetic modelling if the individual estimated parameter values substantially vary from the typical population, which might be a problem in a highly heterogeneous patient group, as in this study. In conclusion, although there was a high correlation coefficient when comparing piperacillin and tazobactam trough levels in critically ill patients, a variability of piperacillin/tazobactam quotients was found that might be important for therapy. Piperacillin trough levels within the therapeutic range might be associated with subtherapeutic or therapeutic tazobactam trough levels. If physi-
cians want to ensure that free tazobactam trough levels are >4 mg/ L, the current data indicate that total piperacillin trough levels ≥70 mg/L should be reached in critically ill patients. Further studies should verify these observations and determine whether similar recommendations are also applicable to patients with extended/ continuous piperacillin/tazobactam infusions to avoid tazobactam levels being consistently below the desired target range. Funding: This work was supported by a Mérieux Research Grant (Institut Mérieux, Lyon, France). The design, collection, analysis and interpretation of data as well as the writing and publication of the manuscript were undertaken by the authors without the participation or influence of the funding source. Competing interests: None declared. Ethical approval: The study protocol (NCT01793012) for this prospective observational study was approved by the Institutional Ethics Committee of Ludwig-Maximilians-University (Munich, Germany) [No. 428-12] and was performed according to the Declaration of Helsinki. Written informed consent was obtained from each patient. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ijantimicag.2016.06.013. References [1] Blondiaux N, Wallet F, Favory R, Onimus T, Nseir S, Courcol RJ, et al. Daily serum piperacillin monitoring is advisable in critically ill patients. Int J Antimicrob Agents 2010;35:500–3. [2] De Waele JJ, Carrette S, Carlier M, Stove V, Boelens J, Claeys G, et al. Therapeutic drug monitoring-based dose optimisation of piperacillin and meropenem: a randomised controlled trial. Intensive Care Med 2014;40:380–7. [3] Occhipinti DJ, Pendland SL, Schoonover LL, Rypins EB, Danziger LH, Rodvold KA. Pharmacokinetics and pharmacodynamics of two multiple-dose piperacillin–tazobactam regimens. Antimicrob Agents Chemother 1997;41:2511–17. [4] Asín-Prieto E, Rodríguez-Gascón A, Trocóniz IF, Soraluce A, Maynar J, Sánchez-Izquierdo JÁ, et al. Population pharmacokinetics of piperacillin and tazobactam in critically ill patients undergoing continuous renal replacement therapy: application to pharmacokinetic/pharmacodynamic analysis. J Antimicrob Chemother 2014;69:180–9.
Please cite this article in press as: Johannes Zander, et al., Variability of piperacillin concentrations in relation to tazobactam concentrations in critically ill patients, International Journal of Antimicrobial Agents (2016), doi: 10.1016/j.ijantimicag.2016.06.013
ARTICLE IN PRESS J. Zander et al. / International Journal of Antimicrobial Agents ■■ (2016) ■■–■■
[5] Varghese JM, Jarrett P, Boots RJ, Kirkpatrick CM, Lipman J, Roberts JA. Pharmacokinetics of piperacillin and tazobactam in plasma and subcutaneous interstitial fluid in critically ill patients receiving continuous venovenous haemodiafiltration. Int J Antimicrob Agents 2014;43:343–8. [6] Melchers MJ, van Mil AC, Mouton JW. In vitro activity of ceftolozane alone and in combination with tazobactam against extended-spectrum-β-lactamaseharboring Enterobacteriaceae. Antimicrob Agents Chemother 2015;59:4521–5. [7] Strayer AH, Gilbert DH, Pivarnik P, Medeiros AA, Zinner SH, Dudley MN. Pharmacodynamics of piperacillin alone and in combination with tazobactam against piperacillin-resistant and -susceptible organisms in an in vitro model of infection. Antimicrob Agents Chemother 1994;38:2351–6. [8] VanScoy B, Mendes RE, McCauley J, Bhavnani SM, Bulik CC, Okusanya OO, et al. Pharmacological basis of β-lactamase inhibitor therapeutics: tazobactam in combination with ceftolozane. Antimicrob Agents Chemother 2013;57:5924–30. [9] VanScoy B, Mendes RE, Nicasio AM, Castanheira M, Bulik CC, Okusanya OO, et al. Pharmacokinetics–pharmacodynamics of tazobactam in combination with ceftolozane in an in vitro infection model. Antimicrob Agents Chemother 2013;57:2809–14. [10] Wong G, Brinkman A, Benefield RJ, Carlier M, De Waele JJ, El Helali N, et al. An international, multicentre survey of β-lactam antibiotic therapeutic drug monitoring practice in intensive care units. J Antimicrob Chemother 2014;69:1416–23. [11] Louie A, Castanheira M, Liu W, Grasso C, Jones RN, Williams G, et al. Pharmacodynamics of β-lactamase inhibition by NXL104 in combination with ceftaroline: examining organisms with multiple types of β-lactamases. Antimicrob Agents Chemother 2012;56:258–70. [12] Zander J, Döbbeler G, Nagel D, Maier B, Scharf C, Huseyn-Zada M, et al. Piperacillin concentration in relation to therapeutic range in critically ill patients—a prospective observational study. Crit Care 2016;20:79.
5
[13] Zander J, Maier B, Suhr A, Zoller M, Frey L, Teupser D, et al. Quantification of piperacillin, tazobactam, cefepime, meropenem, ciprofloxacin and linezolid in serum using an isotope dilution UHPLC-MS/MS method with semi-automated sample preparation. Clin Chem Lab Med 2015;53:781–91. [14] Roberts JA, Paul SK, Akova M, Bassetti M, De Waele JJ, Dimopoulos G, et al. DALI: Defining Antibiotic Levels in Intensive care unit patients: are current β-lactam antibiotic doses sufficient for critically ill patients? Clin Infect Dis 2014;58:1072–83. [15] Udy AA, Lipman J, Jarrett P, Klein K, Wallis SC, Patel K, et al. Are standard doses of piperacillin sufficient for critically ill patients with augmented creatinine clearance? Crit Care 2015;19:28. [16] Sime FB, Roberts MS, Warner MS, Hahn U, Robertson TA, Yeend S, et al. Altered pharmacokinetics of piperacillin in febrile neutropenic patients with hematological malignancy. Antimicrob Agents Chemother 2014;58:3533–7. [17] Ariano RE, Nyhlén A, Donnelly JP, Sitar DS, Harding GK, Zelenitsky SA. Pharmacokinetics and pharmacodynamics of meropenem in febrile neutropenic patients with bacteremia. Ann Pharmacother 2005;39:32–8. [18] McKinnon PS, Paladino JA, Schentag JJ. Evaluation of area under the inhibitory curve (AUIC) and time above the minimum inhibitory concentration (T>MIC) as predictors of outcome for cefepime and ceftazidime in serious bacterial infections. Int J Antimicrob Agents 2008;31:345–51. [19] Mueller SC, Majcher-Peszynska J, Hickstein H, Francke A, Pertschy A, Schulz M, et al. Pharmacokinetics of piperacillin–tazobactam in anuric intensive care patients during continuous venovenous hemodialysis. Antimicrob Agents Chemother 2002;46:1557–60. [20] Felton TW, Goodwin J, O’Connor L, Sharp A, Gregson L, Livermore J, et al. Impact of bolus dosing versus continuous infusion of piperacillin and tazobactam on the development of antimicrobial resistance in Pseudomonas aeruginosa. Antimicrob Agents Chemother 2013;57:5811–19.
Please cite this article in press as: Johannes Zander, et al., Variability of piperacillin concentrations in relation to tazobactam concentrations in critically ill patients, International Journal of Antimicrobial Agents (2016), doi: 10.1016/j.ijantimicag.2016.06.013