Journal of Critical Care xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review☆ Erlangga Yusuf, MD, PhD a,⁎, Herbert Spapen, MD, PhD b, Denis Piérard, MD, PhD a a b
Department of Medical Microbiology and Infection Control, Universitair Ziekenhuis Brussel, Brussels, Belgium Intensive Care Unit, Universitair Ziekenhuis Brussel, Brussels, Belgium
a r t i c l e
i n f o
Keywords: Piperacillin/tazobactam Infection Intensive care Prolonged infusion Pharmacokinetic Pharmacodynamic
a b s t r a c t Purpose: The purpose of this study is to review the rationale of prolonged (ie, extended or continuous) infusion of piperacillin/tazobactam (PIP/TAZ) in critically ill patients and to perform a systematic review that compare the effectiveness of prolonged infusion with intermittent bolus of PIP/TAZ. Materials and methods: A search of Medline, Web of Science, Embase, and Cochrane databases was conducted up to April 2014. For systematic review, studies comparing the effectiveness of prolonged and bolus administration of PIP/TAZ were included. The level of evidence is determined using best-evidence synthesis, which consisted of 5 possible levels of evidence: strong, moderate, limited, conflicting, or no evidence. Results: The pharmacokinetic/pharmacodynamic studies that account for an eventual benefit of prolonged PIP/TAZ infusion were reviewed. In the systematic review, 1 randomized controlled trial was identified that showed higher “cure” in the prolonged than in the intermittent infusion group, yet the chosen clinical outcome in this study, decline in mean Acute Physiology and Chronic Health Evaluation II score is controversial. Of 6 retrospective cohort studies, 4 showed either less mortality, a higher clinical cure rate, or shorter length of hospital stay with prolonged PIP/TAZ treatment. The level of evidence supporting a better clinical outcome with prolonged infusion of PIP/TAZ is moderate. Conclusion: Pharmacokinetic/pharmacodynamic studies provide a robust rationale to prefer prolonged above intermittent infusion of PIP/TAZ. However, although some studies suggest a better outcome in critically ill patients receiving prolonged infusion, the level of evidence is moderate. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Piperacillin/tazobactam (PIP/TAZ) is a broad spectrum combination antibiotic commonly used to treat severe infections in the intensive care unit (ICU) [1]. It consists of 2 active components: piperacillin with high-antimicrobial activity and tazobactam, a βlactamase inhibitor with limited antimicrobial activity [2]. Tazobactam inactivates β-lactamase enzymes produced by bacteria [2], thereby restoring their susceptibility [3]. Despite this interesting microbiological profile, optimization of pharmacokinetic/pharmacodynamic (PK/PD) behavior of PIP/TAZ remains needed to improve treatment outcome and to prevent selection and spread of resistant strains [4-7]. Pharmacokinetics provides information about the movement of a drug from its administration site to the site of action and its elimination from the body. Pharmacodynamics for a given antibiotic refers to its ability to kill or inhibit the growth of
☆ Conflict of interest: D. Piérard received sponsoring for a clinical study from Pfizer, Bayer, and Astra-Zeneca and was member of an advisory board organized by Astra-Zeneca. ⁎ Corresponding author at: Department of Medical Microbiology and Infection Control, Universitair Ziekenhuis Brussel, Laarbeeklaan 101, 1090 Jette, Belgium. Tel.: + 32 2 477 50 00. E-mail address: [email protected] (E. Yusuf).
microorganisms. A key PD feature of an antimicrobial is the minimum inhibitory concentration (MIC). Minimum inhibitory concentration is the term that is used to express the lowest concentration of an antibiotic that inhibits bacterial growth. Pharmacokinetic/pharmacodynamic study combines pharmacokinetic and pharmacodynamics features to predict the probability of successful antibiotic treatment [8]. From a PK/PD perspective, various antibiotic classes show different bacterial kill characteristics. Preclinical studies have defined PIP/TAZ as time-dependent antibiotic; the time during which the free (unbound) antibiotic concentration is maintained above the MIC (fTNMIC), and not the magnitude of its concentration, is the determining factor for bacterial killing [9,10]. Piperacillin/tazobactam has a short half-life (between 0.8 and 1.1 hour) [11] and is given intravenously because of its poor oral absorption [2], generally in bolus dose (ie, infused over 20-60 minutes every 6 or 8 hours) [6,12]. Arguably, giving PIP/TAZ PD features to produce sustained/PD features to produce sustained fTNMIC [13]. For this reason, continuous administration has been proposed as a valuable alternative. However, ICU physicians often argue against this strategy because it requires a dedicated venous access and may cause unwarranted incompatibility with other intravenous therapy. Another proposed method is extended administration, which can be defined as an infusion time beyond 1 hour [14]. Several reviews have
http://dx.doi.org/10.1016/j.jcrc.2014.07.033 0883-9441/© 2014 Elsevier Inc. All rights reserved.
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
2
E. Yusuf et al. / Journal of Critical Care xxx (2014) xxx–xxx
compared continuous with intermittent administration of β-lactam antibiotics [15-18], but none has specifically focused on PIP/TAZ in an ICU setting. Critically ill patients in particular might benefit from prolonged PIP/TAZ infusion [19], for the reasons that will be explained below. We also noticed that the published reviews often did not discuss other types of prolonged infusion than continuous infusion, and they discussed only PK/PD characteristics [16] or only clinical outcome [15]. Furthermore, new studies have been published since the last review [20-22]. Therefore, a systematic review with updated information on PK/PD and clinical outcome of PIP/TAZ use in ICU setting is needed. The present article has 2 aims. First, to assess the rationale of prolonged (ie, extended or continuous) infusion of PIP/TAZ in critically ill patients. Second, to perform a qualitative systematic review comparing the effectiveness of prolonged vs intermittent bolus of PIP/TAZ. 2. Rationale of prolonged infusion of PIP/TAZ in critically ill patients Intensive care unit patients differ from other hospitalized patients in terms of pathophysiology and disease severity. Both factors will affect drug metabolism and PK/PD behavior. For example, the microvascular endothelium becomes highly permeable during sepsis [23]. This will augment the distribution volume (a theoretical volume that relates the plasma concentration of a drug to the administered dose) of hydrophilic drugs such as PIP/TAZ [19]. Volume resuscitation during the early stage of severe sepsis also highly increases cardiac output, thereby enhancing renal and hepatic blood flow [24]. This will significantly affect PIP/TAZ metabolism and excretion rate because piperacillin is mainly (50%-60%) excreted by the kidney and partly in bile [2]. Increased distribution volume and clearance will finally result in lower plasma piperacillin concentrations. This is confirmed by studies showing maximum plasma piperacillin concentrations of 380 mg/L in healthy volunteers [10] and 231 mg/L [11] in ICU patients after intermittent infusion of 4 g piperacillin [10,11]. Augmented renal clearance (ie, elevated renal elimination resulting in subtherapeutic plasma concentrations) of antibiotics is increasingly reported in critically ill patients [25,26]. This phenomenon is probably related to the innate immune response to infection and inflammation (with its associated systemic and hemodynamic consequences) but also to fluid loading and use of vasoactive medications. As a result, cardiac output and renal blood flow increase, which subsequently lead to enhanced glomerular filtration. Increased filtration induces substantial drug elimination and causes subtherapeutic antibiotic plasma levels [25]. Therefore, alternative antibiotic dosing regimens (ie, other than “traditional” intermittent bolus infusion) must be considered when augmented renal clearance is present to offer better treatment options and to reduce the risk of resistance development. As mentioned above, the fTNMIC is an important determinant of bacterial killing when β-lactam antibiotics are used. Animal studies showed that TNMIC between 40% and 70% of the dosing interval is required [11]. Rafati et al [27] studied TNMIC in 40 septic critically ill patients. The TNMIC (many studies mentioned TNMIC instead of fTNMIC which does not specifically mention the measurement of free antibiotic concentration) of continuous infusion of PIP/TAZ (2 g/0.25 g loading dose over 30 minutes followed by 8 g/1 g daily) was higher than the TNMIC after intermittent bolus of PIP/TAZ (3 g/0.375 over 30 minutes every 6 hours): 100% vs 62% and 65% vs 39% for an MIC of 16 mg/L and 32 mg/L, respectively. Noteworthy, these MICs values are high. The European Committee on Antimicrobial Susceptibility Testing (EUCAST) considers Pseudomonas aeruginosa with an MIC higher as 16 mg/L as resistant (EUCAST breakpoint table, version 3.1). Dulhunty et al [28] showed that patients under continuous infusion of PIP/TAZ more often reached plasma antibiotic concentrations above the MIC than patients receiving intermittent bolus (9 [75%] of 12 vs 4 [36%] of 11, respectively). In this study, the 24-hour dose was chosen at the clinician's discretion. Two studies using a Monte Carlo simulation to calculate probability of target attainment (ie, the
probability that a specific value of a PD index [eg, 50% fTNMIC] is achieved at a certain MIC) also confirmed the benefit of prolonged infusion [29,30]. Using MIC distributions from Canadian ICU surveillance data, Zelenitsky et al [29] showed that the cumulative target attainment, determined by integrating each probability of target attainment with the corresponding pathogen and MIC distributions from Canadian ICU surveillance data of 50% fTNMIC in extended infusion (3-hour infusion of 3 g/0.375 g PIP/TAZ every 6 hours) was higher than when an intermittent bolus (30-minute infusion of an equivalent PIP/TAZ dose at similar frequency) was provided (0.84 vs 0.79) . This difference became even more significant in favor of extended infusion at the 100% fTNMIC (0.63 vs 0.36). For an MIC of 1 mg/L, Roberts et al [30] demonstrated a 50% fTNMIC for continuous (8 g/1 g over 24 hours), extended (4 g/0.5 g every 8 hours), and intermittent bolus (4 g/0.5 g every 8 hours) infusion of PIP/TAZ of respectively 0.55, 0.43, and 0.26. Several retrospective clinical studies have demonstrated that larger drug exposures are required, with β-lactam concentrations up to 4 times the MIC for the entire dosing (TN4×MIC) [31,32]. In critically ill patients with pathophysiology changes, this high PK/PD target can be obtained by using continuous infusion. Again, the literature shows that this purpose can be obtained with more frequent dosing or by extended or continuous infusions [10,33]. Duszynska et al [32,34] elegantly proved the relationship between the percentages of TN4×MIC and improved clinical outcome. In 16 patients with ventilator-associated pneumonia (VAP), these investigators showed that continuous infusion of 10.0/1.25 g PIP/TAZ produced adequate therapeutic drug concentrations (defined as TN4×MIC) on the first day of treatment for 71% of the isolated pathogens. Clinical cure was achieved in 90% of the patients with adequate drug concentrations vs 50% in patients with insufficient levels [32,34]. Few studies have investigated tissue penetration of PIP/TAZ. The results of these studies suggest that bolus infusion can achieve tissue concentration of PIP/TAZ above the MIC breakpoint according to EUCAST or Clinical and Laboratory Standards Institute, but the concentration might be suboptimal. Joukhadar et al [35] showed that mean piperacillin concentrations in subcutaneous adipose tissue never exceeded 11 mg/L in septic shock patients who were given 4 g/0.5 g PIP/ TAZ. This tissue concentration is just higher than one-step dilution of MIC 16 mg/L. Another group investigated concurrent plasma and subcutaneous tissue concentrations in critically ill septic patients after receiving continuous vs intermittent bolus of PIP/TAZ. They concluded that continuous infusion more successfully achieved tissue PD targets and enabled to maintain higher trough concentrations compared with standard bolus dosing [36]. In critically ill patients with severe bacterial pneumonia treated with 4 g/0.5 g PIP/TAZ every 8 hours, Boselli et al [37] found PIP/TAZ epithelial lining fluid concentration (SD) of 13.6 (9.4) mg/L/2.1 (1.1) mg/L. They concluded that the given PIP/TAZ regimen might provide insufficient concentrations into lung tissue to exceed the MIC of many causative pathogens [37]. 3. Systematic review on comparing clinical outcome of prolonged vs intermittent bolus infusion of PIP/TAZ 3.1. Literature search and data extraction Together with medical librarians, we searched Medline, Science Citation Index through Web of Science, Embase, and Cochrane up to April 2014. Search terms were “piperacillin” or “piperacillin/tazobactam” and “intensive care unit” or “critically ill” or “critical illness” or “critical care” or “intensive care unit” and “pharmacokinetics” or “pharmacodynamics” or “extended infusion” or “continuous infusion.” Only English language articles were reviewed. We excluded data from critically ill children and from patients undergoing renal replacement therapy. Complete search strategies are shown in Appendix I. Two reviewers read the title and the abstracts of all retrieved references for obvious exclusions. They subsequently read the full text of remaining references. References of included studies were screened
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
E. Yusuf et al. / Journal of Critical Care xxx (2014) xxx–xxx
to find further relevant articles. Studies were included if they contained data comparing clinical outcomes of prolonged vs intermittent bolus of PIP/TAZ. Studies on more than 1 β-lactam antibiotic were included only when the data regarding PIP/TAZ could be extracted and analyzed separately. The literature flow to find relevant studies is shown in Fig. 1.
3
PIP/TAZ every 6 to 8 hours (19% vs 38%, P = .01). Lodise et al [6] identified the APACHE II score as a possible confounder. The difference in 14-day mortality was only found among the subgroup of patients with APACHE II scores less than or equal to 17 (12.2% for patients treated with extended infusion [3.375 g for 4 hours every 8 hours] vs 31.6%, for those with intermittent bolus [3.375 g every 4 or 6 hours], P = .04) [6]. These authors also noticed a shorter hospital length of stay in patients with an APACHE II score greater than or equal to 17 who received extended infusions. Differences in mortality and hospital length of stay were not observed in patients with APACHE II scores less than or equal to 17 [6]. Lorente et al [40] did not find a difference in mortality between patients who received 16 g/2 g PIP/TAZ continuously over 24 hours and those who received the same dose 6 hourly. However, clinical cure rate was higher in patients with continuous compared with those with intermittent bolus (89.2% vs 56.5%, P = .001) [40]. Finally, Waxier and Hu [22] showed shorter duration of mechanical ventilation in patients who received extended than those who received intermittent bolus (5.6 vs 9.7 days, P = .05). Two studies failed to show any difference in clinical outcome between the 2 infusion regimens [20,21]. Fahimi et al [20] compared 2 PIP/TAZ infusion methods (3 g/0.375 g every 8 hours administered as a 4-hour infusion vs 3 g/0.375 g every 6 hours) in patients with VAP. They showed no difference in mortality, duration of hospital stay, and duration of mechanical ventilation. Furthermore, no difference in the duration of mechanical ventilation was noted. Goncalves-Pereira et al [21] did not find a difference in 28-day mortality between continuous and intermittent bolus infusion of 16 g/2 g PIP/TAZ in patients with pneumonia and intraabdominal infection. They also did not observe a difference in 28-day mortality in more severely ill patients (ie, patients with a Simplified Acute Physiology II Score, N42).
3.2. Characteristics of the included studies Demographic characteristics, drug regimen comparisons, and treatment results with either prolonged or intermittent bolus of PIP/TAZ were extracted from all studies. Table 1 summarizes the studies that compare prolonged (continuous or extended) with intermittent bolus infusion of PIP/TAZ in critically ill patients. From one paper [22], only an abstract and no full paper could be obtained. Two randomized controlled trials (RCTs) specifically focused on PIP/TAZ [27,38]. In 40 critically ill septic patients, Rafati et al [27]did not find a difference in mortality between continuous (8 g/2 g over 24 hours) and intermittent bolus (3 g/0.375 g every 6 hours). However, a significantly greater decline in mean Acute Physiology and Chronic Health Evaluation (APACHE) II score after 4 days of treatment was noted in the continuous treatment group (5.2 vs 2.8, P = .04) [27]. Lau et al [38] observed no difference in clinical and microbiological cure rates between patients who received continuous (12 g/1.5 g) and intermittent bolus (3 g/0.375 g every 6 hours). The clinical cure rate in continuous infusion was 86.4% vs 88.4% in intermittent bolus, P = .817. The microbiological cure rate was 83.9% in continuous infusion vs 87.9% in intermittent bolus, P = .597. In this study, only 5% of the patients had an APACHE II score above 20, and the microorganisms involved had low MICs (as low as 0.016 mg/L). Six retrospective cohort studies were identified [6,20-22,39,40]. All used mortality and duration of ICU or hospital stay as outcome variable. A benefit of prolonged above extended infusion was reported in 4 studies [6,22,39,40]. Lee et al [39] showed significantly lower 30-day mortality in patients initiated on an extended infusion of 3 g/0.375 g for 4 hours every 8 hours than in patients treated with intermittent bolus of 2 to 4 g/0.25 to 0.5 g
3.3. Adverse events Information on adverse events was only available in the 2 included RCT. The occurrence of adverse events was not registered in the observational studies. Therefore, pooling data on adverse events is therefore impossible. Lau et al [38] showed treatment-related adverse
Identified references, titles reviewed 1041 Obvious exclusions after reading titles 920 Possibly relevant references, abstract reviewed 121
No original data or not relevant 106
Possibly relevant, full text articles read 15
Relevant article retrieved from included reference 1
Exclusion due to: - No relevant data: 2 - No data on piperacillin/ tazobactam only: 6
Total included studies 8
Randomized Control Trials 2
Retrospective studies 6
Fig. 1. Literature flow.
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
4
Author, publication year
Study population
Setting, country
No. of patients
Type of isolated bacteria
Dose of PIP/TAZ in prolonged infusion
Dose of PIP/TAZ in intermittent infusion
Clinical outcomes prolonged vs intermittent infusion
Gram negative Continuous: (n = 10 in 8 continuous and n = 6 in 4 intermittent). MIC 16 mg/L in 6 and 32 mg/L in 10. 2 g/0.25 g over 30 min loading dose followed by 8 g/1 g over 24 h.
3 g/0.375 g every 6 h.
Mortality (no information on follow-up): RR = 0.8, P = .72. Change in APACHE II score from baseline till end of the fourth day: 5.2 vs 2.8 (P = .04).
3 g/0.375 g every 6 h
81 continuous vs 86 intermittent mean APACHE II score 8 in both groups
Gram negative (n = 153) and Viridans group streptococci (n = 35). MICs between 0.016 and 16 mg/L.
Treatment success at end of therapy (ie, 10-21 d after the last dose): RR = 0.98, P = .817. Microbiological success: RR = 0.95, P = .597.
Pneumonia, bloodstream infections, and other severe infections. 68 extended vs 80 intermittent Median SOFA score 9 in both groups.
Gram negative, comparable between groups. No information on MIC.
2-4 g/0.25-0.5 g every 6-8 h.
Mortality (30 d):
Disease severity or organ failure score RCT Rafati et al [27], 2006 single ICU, Iran.
Sepsis from any source. 20 continuous vs 20 intermittent. Mean APACHE II score 15 in both groups.
Lau et al [38], 2006
All patients in both groups also received amikacin. Intraabdominal infection
33 centers, the United States
Retrospective cohort study Lee et al [39] 2012
2 ICUs, the United States.
Goncalves-Pereira et al [21] 2012 7 ICUs, Portugal.
Significant higher fluoroquinolone or aminoglycoside use in the intermittent (60%) than in extended infusion group (40%). Mainly pneumonia and Gram negative (n = 283) intraabdominal infection. and Gram positive (n = 63). 173 patients, continuous vs intermittent infusion. Mean SAPS II score 48 in both No information on MIC. propensity-matched groups. One-third of the patients received a second antibiotic; no information whether the second antibiotic use is comparable between groups.
Continuous:
2 g/0.25 g over 30 min loading dose followed by 12 g/1.5 g over 24 h.
Extended:
3 g/.375 g for 4 h every 8 h.
Continuous:
16 g/2 g (in 81% of the study population).
RR = 0.5, P = .01. Duration of hospital stay (median): no information on duration itself, P = .41.
Comparable proportion of Mortality (28 d): patients with 16 g/2 g as continuous group. RR = 1, P =1.00. Duration of hospital stay (median): 30 vs 31 d, P = .475.
E. Yusuf et al. / Journal of Critical Care xxx (2014) xxx–xxx
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
Table 1 Clinical studies comparing prolonged with intermittent PIP/TAZ administration
Table 1 (continued) Author, publication year
Study population
Setting, country
No. of patients
Dose of PIP/TAZ in prolonged infusion
Dose of PIP/TAZ in intermittent infusion
Clinical outcomes prolonged vs intermittent infusion
Gram negative. 15/25 multidrug resistant.
Extended: 3 g/0.375 g every 8 h administered as a 4-h infusion.
3 g/0.375 g every 6 h.
Mortality (no information on follow-up): ND (no. of patients not reported).
Disease severity or organ failure score Fahimi et al [20], 2012 Single ICU, India. VAP. 31 continuous vs 30 intermittent. Mean APACHE II score 20 in both groups. No difference in no. of antibiotics used between continuous and intermittent group.
Waxier and Hu [22], 2012.
200 continuous vs 200 intermittent.
Duration of hospital stay:
No information on isolated microorganisms.
Single ICU, the United States.
Extended: No information on dose, administered as a 4-h infusion.
Lorente et al [40], 2009.
VAP.
Gram negative.
Continuous:
Single ICU, Spain.
37 continuous vs 46 intermittent.
No information on MIC.
4 g/0.5 g loading dose over 30 min followed by 16 g/2 g over 24 h.
Lodise et al [6], 2007. Single center, (critically ill patients only), the United States.
No information on dose
Mean APACHE II score 16 in both groups. All patients in both groups also received tobramycin. Pulmonary, skin and soft tissue, intraabdominal, and urinary infections 102 extended vs 92 intermittent. Mean APACHE II score 16 in both groups.
4 g/0.5 g every 6 h.
ND (no. of patients not reported). Duration of mechanical ventilation (ND). Mortality (in-hospital, no information on follow-up): RR = 0.8, P = .3. Duration of hospital stay: 9 vs 10.7 d, P = .39 Duration of mechanical ventilation: 5.6 vs 9.7 d, P = .05. Mortality (no information on follow-up): RR = 0.7, P = .5. Clinical cure (no time point reported):
RR = 1.6, P = .001. Duration of hospital stay (mean): 22 vs 26 d, P = .62 P aeruginosa.
Extended:
No information on MIC.
3 g/0.375 g every 8 h administered as a 4-h infusion.
Comparable fluoroquinolone (6% vs 11%) and aminoglycoside (23% vs 26%) use in the extended and intermittent groups, respectively.
3 g/0.375 g every 4 or 6 h
Mortality (14 d): Overall: RR = 0.6, P = .7
APACHE II score ≥17:
RR = 0.4, P = .04 Duration of hospital stay (median): 18 vs 22.5 d, P = .09. APACHE II score ≥17: 21 vs 38 d, P = .02.
E. Yusuf et al. / Journal of Critical Care xxx (2014) xxx–xxx
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
Type of isolated bacteria
RR indicates relative risk; SAPS, Simplified Acute Physiology Score; SOFA, Sequential Organ Failure Assessment; ND, no difference.
5
6
E. Yusuf et al. / Journal of Critical Care xxx (2014) xxx–xxx
events in 16.9% of the patients receiving continuous infusion and in 13.6% of the patients treated with intermittent bolus. Type and severity of adverse events were comparable between treatments groups; the most common adverse event was gastrointestinal related. Another included RCT only indicated that the adverse events did not vary significantly between the 2 infusion groups [27]. However, no further data on adverse events can be derived from the study. 3.4. Study quality assessment Study quality was evaluated by using the Dutch Cochrane Collaboration questionnaire (http://dcc.cochrane.org/beoordelingsformulieren-enandere-downloads), and a quality assessment of all included studies is displayed in Appendix II. The 2 RCTs were considerably biased. Randomization was not clearly described. The patients, the treating physicians, and the clinical outcome assessment team were not blinded [27,38]. Evaluation of clinical outcome was deemed inappropriate in 1 trial [25], which used the APACHE II score, representing severity of disease within the first 24 hours after admission, to assess evolution of illness over time. In general, the cohort studies did not adjust for possible confounders such as disease severity and use of other antibiotics. 3.5. Rating the level of evidence Patient population, antibiotic dose, and clinical outcome parameters largely varied among studies. Together with the earlier described methodological pitfalls, this rendered pooling of the available evidence for statistical analysis, including meta-analysis, unreliable [41]. A potential alternative for summarizing evidence in heterogenic studies is the best-evidence synthesis. This method originates from the guideline on systematic review of the Cochrane Collaboration Back Review Group [42]. The hypothesis of the systematic review was that the continuous infusion gives better clinical outcome than intermittent bolus infusion. Best-evidence synthesis has 5 predetermined levels of evidence that add more weight on RCTs than on retrospective studies: (1) strong, when general consistent findings that continuous infusion led to better clinical outcome than intermittent bolus that were reported in multiple high-quality RCTs; (2) moderate, when 1 highquality RCT and at least 2 high-quality retrospective studies show general consistent findings or when at least 3 high-quality retrospective studies report general consistent findings; (3) limited, when the finding was found in a single RCT, or general consistent findings were found in maximum 2 retrospective studies; (4) conflicting, when no consistent findings were reported; and (5) no evidence, when no study could be found. Clinical outcome variables that were considered for calculation of the best-evidence synthesis were mortality, clinical cure, duration of hospital stay (or duration of ICU stay when data on hospital stay were not available), or any related clinical outcome compatible with the disease under study (eg, duration of medical ventilation in patients with VAP). If a study showed statistically significant better outcome with prolonged than with intermittent bolus infusion in any of these clinical outcome variables, the study was counted as one “positive study.” Using this method, the level of evidence is that the clinical outcome in favor of prolonged PIP/TAZ is moderate. This level of evidence is based on 1 RCT [27] and 4 cohort studies [6,22,39,40]. However, conflicting level of evidence is found when mortality alone was considered as primary clinical outcome variable. 4. Discussion The current review differs from previously published systematic reviews [15-18] because it specifically focuses on PIP/TAZ use in ICU patients. We know from published reviews that continuous infusion of β-lactam antibiotics in hospitalized patients does not provide better clinical outcome than “traditional” intermittent bolus. In contrast, the
present review underscores that prolonged infusion of PIP/TAZ in critically ill patients may be more beneficial than intermittent dosing. Because the included studies show considerable heterogeneity in terms of PIP/TAZ dose, severity of illness, MIC of pathogens, duration of follow-up, and outcome definition, we are unable to quantify the outcome gain obtained with continuous infusion compared with intermittent bolus. Study results appear to be particularly determined by severity of illness and MIC levels. The RCT that failed to show benefit of prolonged infusion mainly enrolled less severely ill patients (N 70% had APACHE II scores below 10) [38]. This is in accordance with results from a retrospective study showing that prolonged infusion was only beneficial in patients with a high (cut-off at 17) APACHE II score [6]. Our narrative review provides solid evidence that prolonged infusion procures a higher TNMIC, which theoretically should lead to better clinical outcome. However, when MIC is low, intermittent bolus can produce sufficient TNMIC and is as effective as continuous infusion. Therefore, no outcome difference between the 2 treatment regimens can be detected. Minimum inhibitory concentrations of pathogens infecting ICU patients are usually higher than in non-ICU patients, and consequently, PIP/TAZ might be withheld in a large part of this population. During our search, we encountered several studies that compare continuous and intermittent bolus of β-lactam antibiotics, where no separate data on PIP/TAZ can be extracted [28,43-45]. In general, these studies pointed toward a more favorable outcome of prolonged vs intermittent bolus infusion. An RCT that included 18 patients on PIP/TAZ of 30 patients on continuous β-lactam antibiotics and 17 patients on PIP/TAZ of 30 patients on intermittent bolus β-lactam antibiotic showed that more patients in the continuous group were clinically cured (76.7% vs 50.0%) [28]. These results were corroborated by 2 cohort studies [43,44], which observed lower mortality in prolonged compared with intermittent bolus infusion of β-lactam antibiotics among which patients on PIP/TAZ. Only 1 study on βlactam antibiotics that compared prolonged with intermittent bolus for the treatment of Gram-negative infections, where patients with PIP/TAZ were included showed no difference in clinical success rate [45]. Well-conducted and sufficiently powered clinical trials and cohort studies are definitely needed to prove whether prolonged PIP/TAZ infusion is superior to intermittent bolus for treatment of infections in critically ill patients. Patient groups should be stratified based on the diagnosis (eg, VAP and intraabdominal infection) and disease severity. Subanalysis should be done on the patients because the benefit of prolonged infusion is potentially most useful in microorganisms with high MIC values; subgroup analysis should also be performed in patients infected with pathogens expressing MICs around the breakpoint. Equivalent PIP/TAZ doses are required for prolonged and intermittent bolus. Data analysis should be adjusted for possible confounders such as concomitant use of other antibiotics than PIP/TAZ and disease severity. Primary clinical outcome parameters should be mortality or cure at a predefined time point, for example, 28 days after start of treatment. Secondary clinical outcome variables should compose evaluation of various morbidity parameters, microbiological eradication, and adverse events. In this context, the multicenter Betalactam Infusion Group II RCT should be mentioned. This trial was started in July 2012 and intends to compare continuous with intermittent bolus of β-lactams, independent of dose, in 420 critically ill patients [46]. The primary outcome of the Beta-lactam Infusion Group II study is ICU-free days at day 28. Secondary outcomes include 90-day survival, clinical cure at 14 days after cessation of the study antibiotic, organ failure–free days at day 14, and duration of bacteremia. In conclusion, PK/PD studies provide a robust rationale to prefer prolonged above intermittent bolus infusion of PIP/TAZ. There is a moderate level of evidence that clinical outcome in critically ill patients on prolonged infusion is better than in patients on
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
E. Yusuf et al. / Journal of Critical Care xxx (2014) xxx–xxx
intermittent infusion. Prolonged infusion of PIP/TAZ should be preferred in treating critically ill patients. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jcrc.2014.07.033. Acknowledgments We thank the medical librarians Jan Schoones (Leiden University Medical Center, Leiden, The Netherlands) and Katrien Alewaeters (Universitair Ziekenhuis Brussel, Brussels, Belgium) for their assistance in the literature research. 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(5):500–3 [Epub 2010/03/17]. [2] Sorgel F, Kinzig M. The chemistry, pharmacokinetics and tissue distribution of piperacillin/tazobactam. J Antimicrob Chemother 1993;31(Suppl A):39–60 [Epub 1993/01/01]. [3] Acar JF, Goldstein FW, Kitzis MD. Susceptibility survey of piperacillin alone and in the presence of tazobactam. J Antimicrob Chemother 1993;31(Suppl A): 23–8 [Epub 1993/01/01]. [4] Andes D, Anon J, Jacobs MR, Craig WA. Application of pharmacokinetics and pharmacodynamics to antimicrobial therapy of respiratory tract infections. Clin Lab Med 2004;24(2):477–502 [Epub 2004/06/05]. [5] Barbosa TM, Levy SB. The impact of antibiotic use on resistance development and persistence. Drug Resist Updat 2000;3(5):303–11 [Epub 2001/08/11]. [6] Lodise Jr TP, Lomaestro B, Drusano GL. Piperacillin-tazobactam for Pseudomonas aeruginosa infection: clinical implications of an extended-infusion dosing strategy. Clin Infect Dis 2007;44(3):357–63 [Epub 2007/01/06]. [7] Lodise Jr TP, Lomaestro B, Rodvold KA, Danziger LH, Drusano GL. Pharmacodynamic profiling of piperacillin in the presence of tazobactam in patients through the use of population pharmacokinetic models and Monte Carlo simulation. Antimicrob Agents Chemother 2004;48(12):4718–24 [Epub 2004/11/25]. [8] Bulitta JB, Duffull SB, Kinzig-Schippers M, Holzgrabe U, Stephan U, Drusano GL, et al. Systematic comparison of the population pharmacokinetics and pharmacodynamics of piperacillin in cystic fibrosis patients and healthy volunteers. Antimicrob Agents Chemother 2007;51(7):2497–507 [Epub 2007/05/09]. [9] Craig WA. Basic pharmacodynamics of antibacterials with clinical applications to the use of beta-lactams, glycopeptides, and linezolid. Infect Dis Clin N Am 2003;17 (3):479–501 [Epub 2004/01/09]. [10] Roberts JA, Lipman J. Pharmacokinetic issues for antibiotics in the critically ill patient. Crit Care Med 2009;37(3):840–51 [quiz 59; Epub 2009/02/25]. [11] Langgartner J, Lehn N, Gluck T, Herzig H, Kees F. Comparison of the pharmacokinetics of piperacillin and sulbactam during intermittent and continuous intravenous infusion. Chemotherapy 2007;53(5):370–7 [Epub 2007/09/06]. [12] Patel GW, Patel N, Lat A, Trombley K, Enbawe S, Manor K, et al. Outcomes of extended infusion piperacillin/tazobactam for documented Gram-negative infections. Diagn Microbiol Infect Dis 2009;64(2):236–40 [Epub 2009/06/09]. [13] Abdul-Aziz MH, Dulhunty JM, Bellomo R, Lipman J, Roberts JA. Continuous beta-lactam infusion in critically ill patients: the clinical evidence. Ann Intensive Care 2012;2(1):37 [Epub 2012/08/18]. [14] McKenzie C. Antibiotic dosing in critical illness. J Antimicrob Chemother 2011;66 (Suppl 2):ii25–31. [15] Falagas ME, Tansarli GS, Ikawa K, Vardakas KZ. Clinical outcomes with extended or continuous versus short-term intravenous infusion of carbapenems and piperacillin/tazobactam: a systematic review and meta-analysis. Clin Infect Dis 2013;56(2):272–82 [Epub 2012/10/18]. [16] Hayashi Y, Roberts JA, Paterson DL, Lipman J. Pharmacokinetic evaluation of piperacillintazobactam. Expert Opin Drug Metab Toxicol 2010;6(8):1017–31 [Epub 2010/07/20]. [17] Kasiakou SK, Sermaides GJ, Michalopoulos A, Soteriades ES, Falagas ME. Continuous versus intermittent intravenous administration of antibiotics: a meta-analysis of randomised controlled trials. Lancet Infect Dis 2005;5(9):581–9 [Epub 2005/08/27]. [18] Mah GT, Mabasa VH, Chow I, Ensom MH. Evaluating outcomes associated with alternative dosing strategies for piperacillin/tazobactam: a qualitative systematic review. Ann Pharmacother 2012;46(2):265–75 [Epub 2012/01/26]. [19] Roberts JA, Lipman J, Blot S, Rello J. Better outcomes through continuous infusion of time-dependent antibiotics to critically ill patients? Curr Opin Crit Care 2008;14 (4):390–6 [Epub 2008/07/11]. [20] Fahimi F, Ghafari S, Jamaati H, Baniasadi S, Tabarsi P, Najafi A, et al. Continuous versus intermittent administration of piperacillin-tazobactam in intensive care unit patients with ventilator-associated pneumonia. Indian J Crit Care Med 2012; 16(3):141–7 [Epub 2012/11/29]. [21] Goncalves-Pereira J, Oliveira BS, Janeiro S, Estilita J, Monteiro C, Salgueiro A, et al. Continuous infusion of piperacillin/tazobactam in septic critically ill patients—a multicenter propensity matched analysis. PLoS One 2012;7(11):e49845. [22] Waxier C, Hu D. Clinical outcomes of prolonged-infusion piperacillin/tazobactam in patients admitted to the intensive care unit. Crit Care Med 2012;40(12):U101. [23] Aird WC. The role of the endothelium in severe sepsis and multiple organ dysfunction syndrome. Blood 2003;101(10):3765–77 [Epub 2003/01/25].
7
[24] Smith BS, Yogaratnam D, Levasseur-Franklin KE, Forni A, Fong J. Introduction to drug pharmacokinetics in the critically ill patient. Chest 2012;141(5):1327–36 [Epub 2012/05/04]. [25] Udy AA, Roberts JA, Boots RJ, Paterson DL, Lipman J. Augmented renal clearance: implications for antibacterial dosing in the critically ill. Clin Pharmacokinet 2010; 49(1):1–16 [Epub 2009/12/17]. [26] Udy AA, Putt MT, Shanmugathasan S, Roberts JA, Lipman J. Augmented renal clearance in the intensive care unit: an illustrative case series. Int J Antimicrob Agents 2010;35(6):606–8 [Epub 2010/03/24]. [27] Rafati MR, Rouini MR, Mojtahedzadeh M, Najafi A, Tavakoli H, Gholami K, et al. Clinical efficacy of continuous infusion of piperacillin compared with intermittent dosing in septic critically ill patients. Int J Antimicrob Agents 2006;28(2):122–7 [Epub 2006/07/04]. [28] Dulhunty JM, Roberts JA, Davis JS, Webb SA, Bellomo R, Gomersall C, et al. Continuous infusion of beta-lactam antibiotics in severe sepsis: a multicenter double-blind, randomized controlled trial. Clin Infect Dis 2013;56(2):236–44 [Epub 2012/10/18]. [29] Zelenitsky SA, Ariano RE, Zhanel GG. Pharmacodynamics of empirical antibiotic monotherapies for an intensive care unit (ICU) population based on Canadian surveillance data. J Antimicrob Chemother 2011;66(2):343–9 [Epub 2010/10/12]. [30] Roberts JA, Kirkpatrick CM, Roberts MS, Dalley AJ, Lipman J. First-dose and steadystate population pharmacokinetics and pharmacodynamics of piperacillin by continuous or intermittent dosing in critically ill patients with sepsis. Int J Antimicrob Agents 2010;35(2):156–63 [Epub 2009/12/19]. [31] Angus BJ, Smith MD, Suputtamongkol Y, Mattie H, Walsh AL, Wuthiekanun V, et al. Pharmacokinetic-pharmacodynamic evaluation of ceftazidime continuous infusion vs intermittent bolus injection in septicaemic melioidosis. Br J Clin Pharmacol 2000;50(2):184–91 [Epub 2000/08/10]. [32] Tam VH, McKinnon PS, Akins RL, Rybak MJ, Drusano GL. Pharmacodynamics of cefepime in patients with Gram-negative infections. J Antimicrob Chemother 2002;50(3):425–8 [Epub 2002/09/03]. [33] Mouton JW, den Hollander JG. Killing of Pseudomonas aeruginosa during continuous and intermittent infusion of ceftazidime in an in vitro pharmacokinetic model. Antimicrob Agents Chemother 1994;38(5):931–6 [Epub 1994/05/01]. [34] Duszynska W, Taccone FS, Switala M, Hurkacz M, Kowalska-Krochmal B, Kübler A. Continuous infusion of piperacillin/tazobactam in ventilator-associated pneumonia: a pilot study on efficacy and costs. Int J Antimicrob Agents 2012;39(2):153–8. [35] Joukhadar C, Frossard M, Mayer BX, Brunner M, Klein N, Siostrzonek P, et al. Impaired target site penetration of beta-lactams may account for therapeutic failure in patients with septic shock. Crit Care Med 2001;29(2):385–91 [Epub 2001/03/14]. [36] Roberts JA, Roberts MS, Robertson TA, Dalley AJ, Lipman J. Piperacillin penetration into tissue of critically ill patients with sepsis—bolus versus continuous administration? Crit Care Med 2009;37(3):926–33 [Epub 2009/02/25]. [37] Boselli E, Breilh D, Cannesson M, Xuereb F, Rimmele T, Chassard D, et al. Steady-state plasma and intrapulmonary concentrations of piperacillin/tazobactam 4 g/0.5 g administered to critically ill patients with severe nosocomial pneumonia. Intensive Care Med 2004;30(5):976–9. [38] Lau WK, Mercer D, Itani KM, Nicolau DP, Kuti JL, Mansfield D, et al. Randomized, open-label, comparative study of piperacillin-tazobactam administered by continuous infusion versus intermittent infusion for treatment of hospitalized patients with complicated intra-abdominal infection. Antimicrob Agents Chemother 2006;50(11):3556–61 [Epub 2006/08/31]. [39] Lee GC, Liou H, Yee R, Quan CF, Neldner K. Outcomes of extended-infusion piperacillin-tazobactam: a retrospective analysis of critically ill patients. Clin Ther 2012;34(12):2297–300 [Epub 2012/12/01]. [40] Lorente L, Jimenez A, Martin MM, Iribarren JL, Jimenez JJ, Mora ML. Clinical cure of ventilator-associated pneumonia treated with piperacillin/tazobactam administered by continuous or intermittent infusion. Int J Antimicrob Agents 2009;33(5): 464–8 [Epub 2009/01/20]. [41] Falagas ME, Lourida P, Poulikakos P, Rafailidis PI, Tansarli GS. Antibiotic treatment of infections due to carbapenem-resistant Enterobacteriaceae: systematic evaluation of the available evidence. Antimicrob Agents Chemother 2014;58(2): 654–63 [Epub 2013/10/02]. [42] van Tulder M, Furlan A, Bombardier C, Bouter L, Editorial Board of the Cochrane Collaboration Back Review G. Updated method guidelines for systematic reviews in the cochrane collaboration back review group. Spine 2003;28(12):1290–9 [Epub 2003/06/18]. [43] Dow RJ, Rose WE, Fox BC, Thorpe JM, Fish JT. Retrospective study of prolonged versus intermittent infusion piperacillin-tazobactam and meropenem in intensive care unit patients at an academic medical center. Infect Dis Clin Pract 2011;19(6):413–7. [44] Yost RJ, Cappelletty DM. The Retrospective Cohort of Extended-Infusion Piperacillin-Tazobactam (RECEIPT) study: a multicenter study. Pharmacotherapy 2011;31(8):767–75. [45] Arnold HM, Hollands JM, Skrupky LP, Smith JR, Juang PH, Hampton NB, et al. Prolonged infusion antibiotics for suspected gram-negative infections in the ICU: a before-after study. Ann Pharmacother 2013;47(2):170–80 [Epub 2013/01/24]. [46] Dulhunty JM, Roberts JA, Davis JS, Webb SA, Bellomo R, Gomersall C, et al. A protocol for a multicentre randomised controlled trial of continuous beta-lactam infusion compared with intermittent beta-lactam dosing in critically ill patients with severe sepsis: the BLING II study. Crit Care Resusc 2013;15(3):179–85 [Epub 2013/08/16].
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review☆ Erlangga Yusuf, MD, PhD a,⁎, Herbert Spapen, MD, PhD b, Denis Piérard, MD, PhD a a b
Department of Medical Microbiology and Infection Control, Universitair Ziekenhuis Brussel, Brussels, Belgium Intensive Care Unit, Universitair Ziekenhuis Brussel, Brussels, Belgium
a r t i c l e
i n f o
Keywords: Piperacillin/tazobactam Infection Intensive care Prolonged infusion Pharmacokinetic Pharmacodynamic
a b s t r a c t Purpose: The purpose of this study is to review the rationale of prolonged (ie, extended or continuous) infusion of piperacillin/tazobactam (PIP/TAZ) in critically ill patients and to perform a systematic review that compare the effectiveness of prolonged infusion with intermittent bolus of PIP/TAZ. Materials and methods: A search of Medline, Web of Science, Embase, and Cochrane databases was conducted up to April 2014. For systematic review, studies comparing the effectiveness of prolonged and bolus administration of PIP/TAZ were included. The level of evidence is determined using best-evidence synthesis, which consisted of 5 possible levels of evidence: strong, moderate, limited, conflicting, or no evidence. Results: The pharmacokinetic/pharmacodynamic studies that account for an eventual benefit of prolonged PIP/TAZ infusion were reviewed. In the systematic review, 1 randomized controlled trial was identified that showed higher “cure” in the prolonged than in the intermittent infusion group, yet the chosen clinical outcome in this study, decline in mean Acute Physiology and Chronic Health Evaluation II score is controversial. Of 6 retrospective cohort studies, 4 showed either less mortality, a higher clinical cure rate, or shorter length of hospital stay with prolonged PIP/TAZ treatment. The level of evidence supporting a better clinical outcome with prolonged infusion of PIP/TAZ is moderate. Conclusion: Pharmacokinetic/pharmacodynamic studies provide a robust rationale to prefer prolonged above intermittent infusion of PIP/TAZ. However, although some studies suggest a better outcome in critically ill patients receiving prolonged infusion, the level of evidence is moderate. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Piperacillin/tazobactam (PIP/TAZ) is a broad spectrum combination antibiotic commonly used to treat severe infections in the intensive care unit (ICU) [1]. It consists of 2 active components: piperacillin with high-antimicrobial activity and tazobactam, a βlactamase inhibitor with limited antimicrobial activity [2]. Tazobactam inactivates β-lactamase enzymes produced by bacteria [2], thereby restoring their susceptibility [3]. Despite this interesting microbiological profile, optimization of pharmacokinetic/pharmacodynamic (PK/PD) behavior of PIP/TAZ remains needed to improve treatment outcome and to prevent selection and spread of resistant strains [4-7]. Pharmacokinetics provides information about the movement of a drug from its administration site to the site of action and its elimination from the body. Pharmacodynamics for a given antibiotic refers to its ability to kill or inhibit the growth of
☆ Conflict of interest: D. Piérard received sponsoring for a clinical study from Pfizer, Bayer, and Astra-Zeneca and was member of an advisory board organized by Astra-Zeneca. ⁎ Corresponding author at: Department of Medical Microbiology and Infection Control, Universitair Ziekenhuis Brussel, Laarbeeklaan 101, 1090 Jette, Belgium. Tel.: + 32 2 477 50 00. E-mail address: [email protected] (E. Yusuf).
microorganisms. A key PD feature of an antimicrobial is the minimum inhibitory concentration (MIC). Minimum inhibitory concentration is the term that is used to express the lowest concentration of an antibiotic that inhibits bacterial growth. Pharmacokinetic/pharmacodynamic study combines pharmacokinetic and pharmacodynamics features to predict the probability of successful antibiotic treatment [8]. From a PK/PD perspective, various antibiotic classes show different bacterial kill characteristics. Preclinical studies have defined PIP/TAZ as time-dependent antibiotic; the time during which the free (unbound) antibiotic concentration is maintained above the MIC (fTNMIC), and not the magnitude of its concentration, is the determining factor for bacterial killing [9,10]. Piperacillin/tazobactam has a short half-life (between 0.8 and 1.1 hour) [11] and is given intravenously because of its poor oral absorption [2], generally in bolus dose (ie, infused over 20-60 minutes every 6 or 8 hours) [6,12]. Arguably, giving PIP/TAZ PD features to produce sustained/PD features to produce sustained fTNMIC [13]. For this reason, continuous administration has been proposed as a valuable alternative. However, ICU physicians often argue against this strategy because it requires a dedicated venous access and may cause unwarranted incompatibility with other intravenous therapy. Another proposed method is extended administration, which can be defined as an infusion time beyond 1 hour [14]. Several reviews have
http://dx.doi.org/10.1016/j.jcrc.2014.07.033 0883-9441/© 2014 Elsevier Inc. All rights reserved.
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
2
E. Yusuf et al. / Journal of Critical Care xxx (2014) xxx–xxx
compared continuous with intermittent administration of β-lactam antibiotics [15-18], but none has specifically focused on PIP/TAZ in an ICU setting. Critically ill patients in particular might benefit from prolonged PIP/TAZ infusion [19], for the reasons that will be explained below. We also noticed that the published reviews often did not discuss other types of prolonged infusion than continuous infusion, and they discussed only PK/PD characteristics [16] or only clinical outcome [15]. Furthermore, new studies have been published since the last review [20-22]. Therefore, a systematic review with updated information on PK/PD and clinical outcome of PIP/TAZ use in ICU setting is needed. The present article has 2 aims. First, to assess the rationale of prolonged (ie, extended or continuous) infusion of PIP/TAZ in critically ill patients. Second, to perform a qualitative systematic review comparing the effectiveness of prolonged vs intermittent bolus of PIP/TAZ. 2. Rationale of prolonged infusion of PIP/TAZ in critically ill patients Intensive care unit patients differ from other hospitalized patients in terms of pathophysiology and disease severity. Both factors will affect drug metabolism and PK/PD behavior. For example, the microvascular endothelium becomes highly permeable during sepsis [23]. This will augment the distribution volume (a theoretical volume that relates the plasma concentration of a drug to the administered dose) of hydrophilic drugs such as PIP/TAZ [19]. Volume resuscitation during the early stage of severe sepsis also highly increases cardiac output, thereby enhancing renal and hepatic blood flow [24]. This will significantly affect PIP/TAZ metabolism and excretion rate because piperacillin is mainly (50%-60%) excreted by the kidney and partly in bile [2]. Increased distribution volume and clearance will finally result in lower plasma piperacillin concentrations. This is confirmed by studies showing maximum plasma piperacillin concentrations of 380 mg/L in healthy volunteers [10] and 231 mg/L [11] in ICU patients after intermittent infusion of 4 g piperacillin [10,11]. Augmented renal clearance (ie, elevated renal elimination resulting in subtherapeutic plasma concentrations) of antibiotics is increasingly reported in critically ill patients [25,26]. This phenomenon is probably related to the innate immune response to infection and inflammation (with its associated systemic and hemodynamic consequences) but also to fluid loading and use of vasoactive medications. As a result, cardiac output and renal blood flow increase, which subsequently lead to enhanced glomerular filtration. Increased filtration induces substantial drug elimination and causes subtherapeutic antibiotic plasma levels [25]. Therefore, alternative antibiotic dosing regimens (ie, other than “traditional” intermittent bolus infusion) must be considered when augmented renal clearance is present to offer better treatment options and to reduce the risk of resistance development. As mentioned above, the fTNMIC is an important determinant of bacterial killing when β-lactam antibiotics are used. Animal studies showed that TNMIC between 40% and 70% of the dosing interval is required [11]. Rafati et al [27] studied TNMIC in 40 septic critically ill patients. The TNMIC (many studies mentioned TNMIC instead of fTNMIC which does not specifically mention the measurement of free antibiotic concentration) of continuous infusion of PIP/TAZ (2 g/0.25 g loading dose over 30 minutes followed by 8 g/1 g daily) was higher than the TNMIC after intermittent bolus of PIP/TAZ (3 g/0.375 over 30 minutes every 6 hours): 100% vs 62% and 65% vs 39% for an MIC of 16 mg/L and 32 mg/L, respectively. Noteworthy, these MICs values are high. The European Committee on Antimicrobial Susceptibility Testing (EUCAST) considers Pseudomonas aeruginosa with an MIC higher as 16 mg/L as resistant (EUCAST breakpoint table, version 3.1). Dulhunty et al [28] showed that patients under continuous infusion of PIP/TAZ more often reached plasma antibiotic concentrations above the MIC than patients receiving intermittent bolus (9 [75%] of 12 vs 4 [36%] of 11, respectively). In this study, the 24-hour dose was chosen at the clinician's discretion. Two studies using a Monte Carlo simulation to calculate probability of target attainment (ie, the
probability that a specific value of a PD index [eg, 50% fTNMIC] is achieved at a certain MIC) also confirmed the benefit of prolonged infusion [29,30]. Using MIC distributions from Canadian ICU surveillance data, Zelenitsky et al [29] showed that the cumulative target attainment, determined by integrating each probability of target attainment with the corresponding pathogen and MIC distributions from Canadian ICU surveillance data of 50% fTNMIC in extended infusion (3-hour infusion of 3 g/0.375 g PIP/TAZ every 6 hours) was higher than when an intermittent bolus (30-minute infusion of an equivalent PIP/TAZ dose at similar frequency) was provided (0.84 vs 0.79) . This difference became even more significant in favor of extended infusion at the 100% fTNMIC (0.63 vs 0.36). For an MIC of 1 mg/L, Roberts et al [30] demonstrated a 50% fTNMIC for continuous (8 g/1 g over 24 hours), extended (4 g/0.5 g every 8 hours), and intermittent bolus (4 g/0.5 g every 8 hours) infusion of PIP/TAZ of respectively 0.55, 0.43, and 0.26. Several retrospective clinical studies have demonstrated that larger drug exposures are required, with β-lactam concentrations up to 4 times the MIC for the entire dosing (TN4×MIC) [31,32]. In critically ill patients with pathophysiology changes, this high PK/PD target can be obtained by using continuous infusion. Again, the literature shows that this purpose can be obtained with more frequent dosing or by extended or continuous infusions [10,33]. Duszynska et al [32,34] elegantly proved the relationship between the percentages of TN4×MIC and improved clinical outcome. In 16 patients with ventilator-associated pneumonia (VAP), these investigators showed that continuous infusion of 10.0/1.25 g PIP/TAZ produced adequate therapeutic drug concentrations (defined as TN4×MIC) on the first day of treatment for 71% of the isolated pathogens. Clinical cure was achieved in 90% of the patients with adequate drug concentrations vs 50% in patients with insufficient levels [32,34]. Few studies have investigated tissue penetration of PIP/TAZ. The results of these studies suggest that bolus infusion can achieve tissue concentration of PIP/TAZ above the MIC breakpoint according to EUCAST or Clinical and Laboratory Standards Institute, but the concentration might be suboptimal. Joukhadar et al [35] showed that mean piperacillin concentrations in subcutaneous adipose tissue never exceeded 11 mg/L in septic shock patients who were given 4 g/0.5 g PIP/ TAZ. This tissue concentration is just higher than one-step dilution of MIC 16 mg/L. Another group investigated concurrent plasma and subcutaneous tissue concentrations in critically ill septic patients after receiving continuous vs intermittent bolus of PIP/TAZ. They concluded that continuous infusion more successfully achieved tissue PD targets and enabled to maintain higher trough concentrations compared with standard bolus dosing [36]. In critically ill patients with severe bacterial pneumonia treated with 4 g/0.5 g PIP/TAZ every 8 hours, Boselli et al [37] found PIP/TAZ epithelial lining fluid concentration (SD) of 13.6 (9.4) mg/L/2.1 (1.1) mg/L. They concluded that the given PIP/TAZ regimen might provide insufficient concentrations into lung tissue to exceed the MIC of many causative pathogens [37]. 3. Systematic review on comparing clinical outcome of prolonged vs intermittent bolus infusion of PIP/TAZ 3.1. Literature search and data extraction Together with medical librarians, we searched Medline, Science Citation Index through Web of Science, Embase, and Cochrane up to April 2014. Search terms were “piperacillin” or “piperacillin/tazobactam” and “intensive care unit” or “critically ill” or “critical illness” or “critical care” or “intensive care unit” and “pharmacokinetics” or “pharmacodynamics” or “extended infusion” or “continuous infusion.” Only English language articles were reviewed. We excluded data from critically ill children and from patients undergoing renal replacement therapy. Complete search strategies are shown in Appendix I. Two reviewers read the title and the abstracts of all retrieved references for obvious exclusions. They subsequently read the full text of remaining references. References of included studies were screened
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
E. Yusuf et al. / Journal of Critical Care xxx (2014) xxx–xxx
to find further relevant articles. Studies were included if they contained data comparing clinical outcomes of prolonged vs intermittent bolus of PIP/TAZ. Studies on more than 1 β-lactam antibiotic were included only when the data regarding PIP/TAZ could be extracted and analyzed separately. The literature flow to find relevant studies is shown in Fig. 1.
3
PIP/TAZ every 6 to 8 hours (19% vs 38%, P = .01). Lodise et al [6] identified the APACHE II score as a possible confounder. The difference in 14-day mortality was only found among the subgroup of patients with APACHE II scores less than or equal to 17 (12.2% for patients treated with extended infusion [3.375 g for 4 hours every 8 hours] vs 31.6%, for those with intermittent bolus [3.375 g every 4 or 6 hours], P = .04) [6]. These authors also noticed a shorter hospital length of stay in patients with an APACHE II score greater than or equal to 17 who received extended infusions. Differences in mortality and hospital length of stay were not observed in patients with APACHE II scores less than or equal to 17 [6]. Lorente et al [40] did not find a difference in mortality between patients who received 16 g/2 g PIP/TAZ continuously over 24 hours and those who received the same dose 6 hourly. However, clinical cure rate was higher in patients with continuous compared with those with intermittent bolus (89.2% vs 56.5%, P = .001) [40]. Finally, Waxier and Hu [22] showed shorter duration of mechanical ventilation in patients who received extended than those who received intermittent bolus (5.6 vs 9.7 days, P = .05). Two studies failed to show any difference in clinical outcome between the 2 infusion regimens [20,21]. Fahimi et al [20] compared 2 PIP/TAZ infusion methods (3 g/0.375 g every 8 hours administered as a 4-hour infusion vs 3 g/0.375 g every 6 hours) in patients with VAP. They showed no difference in mortality, duration of hospital stay, and duration of mechanical ventilation. Furthermore, no difference in the duration of mechanical ventilation was noted. Goncalves-Pereira et al [21] did not find a difference in 28-day mortality between continuous and intermittent bolus infusion of 16 g/2 g PIP/TAZ in patients with pneumonia and intraabdominal infection. They also did not observe a difference in 28-day mortality in more severely ill patients (ie, patients with a Simplified Acute Physiology II Score, N42).
3.2. Characteristics of the included studies Demographic characteristics, drug regimen comparisons, and treatment results with either prolonged or intermittent bolus of PIP/TAZ were extracted from all studies. Table 1 summarizes the studies that compare prolonged (continuous or extended) with intermittent bolus infusion of PIP/TAZ in critically ill patients. From one paper [22], only an abstract and no full paper could be obtained. Two randomized controlled trials (RCTs) specifically focused on PIP/TAZ [27,38]. In 40 critically ill septic patients, Rafati et al [27]did not find a difference in mortality between continuous (8 g/2 g over 24 hours) and intermittent bolus (3 g/0.375 g every 6 hours). However, a significantly greater decline in mean Acute Physiology and Chronic Health Evaluation (APACHE) II score after 4 days of treatment was noted in the continuous treatment group (5.2 vs 2.8, P = .04) [27]. Lau et al [38] observed no difference in clinical and microbiological cure rates between patients who received continuous (12 g/1.5 g) and intermittent bolus (3 g/0.375 g every 6 hours). The clinical cure rate in continuous infusion was 86.4% vs 88.4% in intermittent bolus, P = .817. The microbiological cure rate was 83.9% in continuous infusion vs 87.9% in intermittent bolus, P = .597. In this study, only 5% of the patients had an APACHE II score above 20, and the microorganisms involved had low MICs (as low as 0.016 mg/L). Six retrospective cohort studies were identified [6,20-22,39,40]. All used mortality and duration of ICU or hospital stay as outcome variable. A benefit of prolonged above extended infusion was reported in 4 studies [6,22,39,40]. Lee et al [39] showed significantly lower 30-day mortality in patients initiated on an extended infusion of 3 g/0.375 g for 4 hours every 8 hours than in patients treated with intermittent bolus of 2 to 4 g/0.25 to 0.5 g
3.3. Adverse events Information on adverse events was only available in the 2 included RCT. The occurrence of adverse events was not registered in the observational studies. Therefore, pooling data on adverse events is therefore impossible. Lau et al [38] showed treatment-related adverse
Identified references, titles reviewed 1041 Obvious exclusions after reading titles 920 Possibly relevant references, abstract reviewed 121
No original data or not relevant 106
Possibly relevant, full text articles read 15
Relevant article retrieved from included reference 1
Exclusion due to: - No relevant data: 2 - No data on piperacillin/ tazobactam only: 6
Total included studies 8
Randomized Control Trials 2
Retrospective studies 6
Fig. 1. Literature flow.
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
4
Author, publication year
Study population
Setting, country
No. of patients
Type of isolated bacteria
Dose of PIP/TAZ in prolonged infusion
Dose of PIP/TAZ in intermittent infusion
Clinical outcomes prolonged vs intermittent infusion
Gram negative Continuous: (n = 10 in 8 continuous and n = 6 in 4 intermittent). MIC 16 mg/L in 6 and 32 mg/L in 10. 2 g/0.25 g over 30 min loading dose followed by 8 g/1 g over 24 h.
3 g/0.375 g every 6 h.
Mortality (no information on follow-up): RR = 0.8, P = .72. Change in APACHE II score from baseline till end of the fourth day: 5.2 vs 2.8 (P = .04).
3 g/0.375 g every 6 h
81 continuous vs 86 intermittent mean APACHE II score 8 in both groups
Gram negative (n = 153) and Viridans group streptococci (n = 35). MICs between 0.016 and 16 mg/L.
Treatment success at end of therapy (ie, 10-21 d after the last dose): RR = 0.98, P = .817. Microbiological success: RR = 0.95, P = .597.
Pneumonia, bloodstream infections, and other severe infections. 68 extended vs 80 intermittent Median SOFA score 9 in both groups.
Gram negative, comparable between groups. No information on MIC.
2-4 g/0.25-0.5 g every 6-8 h.
Mortality (30 d):
Disease severity or organ failure score RCT Rafati et al [27], 2006 single ICU, Iran.
Sepsis from any source. 20 continuous vs 20 intermittent. Mean APACHE II score 15 in both groups.
Lau et al [38], 2006
All patients in both groups also received amikacin. Intraabdominal infection
33 centers, the United States
Retrospective cohort study Lee et al [39] 2012
2 ICUs, the United States.
Goncalves-Pereira et al [21] 2012 7 ICUs, Portugal.
Significant higher fluoroquinolone or aminoglycoside use in the intermittent (60%) than in extended infusion group (40%). Mainly pneumonia and Gram negative (n = 283) intraabdominal infection. and Gram positive (n = 63). 173 patients, continuous vs intermittent infusion. Mean SAPS II score 48 in both No information on MIC. propensity-matched groups. One-third of the patients received a second antibiotic; no information whether the second antibiotic use is comparable between groups.
Continuous:
2 g/0.25 g over 30 min loading dose followed by 12 g/1.5 g over 24 h.
Extended:
3 g/.375 g for 4 h every 8 h.
Continuous:
16 g/2 g (in 81% of the study population).
RR = 0.5, P = .01. Duration of hospital stay (median): no information on duration itself, P = .41.
Comparable proportion of Mortality (28 d): patients with 16 g/2 g as continuous group. RR = 1, P =1.00. Duration of hospital stay (median): 30 vs 31 d, P = .475.
E. Yusuf et al. / Journal of Critical Care xxx (2014) xxx–xxx
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
Table 1 Clinical studies comparing prolonged with intermittent PIP/TAZ administration
Table 1 (continued) Author, publication year
Study population
Setting, country
No. of patients
Dose of PIP/TAZ in prolonged infusion
Dose of PIP/TAZ in intermittent infusion
Clinical outcomes prolonged vs intermittent infusion
Gram negative. 15/25 multidrug resistant.
Extended: 3 g/0.375 g every 8 h administered as a 4-h infusion.
3 g/0.375 g every 6 h.
Mortality (no information on follow-up): ND (no. of patients not reported).
Disease severity or organ failure score Fahimi et al [20], 2012 Single ICU, India. VAP. 31 continuous vs 30 intermittent. Mean APACHE II score 20 in both groups. No difference in no. of antibiotics used between continuous and intermittent group.
Waxier and Hu [22], 2012.
200 continuous vs 200 intermittent.
Duration of hospital stay:
No information on isolated microorganisms.
Single ICU, the United States.
Extended: No information on dose, administered as a 4-h infusion.
Lorente et al [40], 2009.
VAP.
Gram negative.
Continuous:
Single ICU, Spain.
37 continuous vs 46 intermittent.
No information on MIC.
4 g/0.5 g loading dose over 30 min followed by 16 g/2 g over 24 h.
Lodise et al [6], 2007. Single center, (critically ill patients only), the United States.
No information on dose
Mean APACHE II score 16 in both groups. All patients in both groups also received tobramycin. Pulmonary, skin and soft tissue, intraabdominal, and urinary infections 102 extended vs 92 intermittent. Mean APACHE II score 16 in both groups.
4 g/0.5 g every 6 h.
ND (no. of patients not reported). Duration of mechanical ventilation (ND). Mortality (in-hospital, no information on follow-up): RR = 0.8, P = .3. Duration of hospital stay: 9 vs 10.7 d, P = .39 Duration of mechanical ventilation: 5.6 vs 9.7 d, P = .05. Mortality (no information on follow-up): RR = 0.7, P = .5. Clinical cure (no time point reported):
RR = 1.6, P = .001. Duration of hospital stay (mean): 22 vs 26 d, P = .62 P aeruginosa.
Extended:
No information on MIC.
3 g/0.375 g every 8 h administered as a 4-h infusion.
Comparable fluoroquinolone (6% vs 11%) and aminoglycoside (23% vs 26%) use in the extended and intermittent groups, respectively.
3 g/0.375 g every 4 or 6 h
Mortality (14 d): Overall: RR = 0.6, P = .7
APACHE II score ≥17:
RR = 0.4, P = .04 Duration of hospital stay (median): 18 vs 22.5 d, P = .09. APACHE II score ≥17: 21 vs 38 d, P = .02.
E. Yusuf et al. / Journal of Critical Care xxx (2014) xxx–xxx
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
Type of isolated bacteria
RR indicates relative risk; SAPS, Simplified Acute Physiology Score; SOFA, Sequential Organ Failure Assessment; ND, no difference.
5
6
E. Yusuf et al. / Journal of Critical Care xxx (2014) xxx–xxx
events in 16.9% of the patients receiving continuous infusion and in 13.6% of the patients treated with intermittent bolus. Type and severity of adverse events were comparable between treatments groups; the most common adverse event was gastrointestinal related. Another included RCT only indicated that the adverse events did not vary significantly between the 2 infusion groups [27]. However, no further data on adverse events can be derived from the study. 3.4. Study quality assessment Study quality was evaluated by using the Dutch Cochrane Collaboration questionnaire (http://dcc.cochrane.org/beoordelingsformulieren-enandere-downloads), and a quality assessment of all included studies is displayed in Appendix II. The 2 RCTs were considerably biased. Randomization was not clearly described. The patients, the treating physicians, and the clinical outcome assessment team were not blinded [27,38]. Evaluation of clinical outcome was deemed inappropriate in 1 trial [25], which used the APACHE II score, representing severity of disease within the first 24 hours after admission, to assess evolution of illness over time. In general, the cohort studies did not adjust for possible confounders such as disease severity and use of other antibiotics. 3.5. Rating the level of evidence Patient population, antibiotic dose, and clinical outcome parameters largely varied among studies. Together with the earlier described methodological pitfalls, this rendered pooling of the available evidence for statistical analysis, including meta-analysis, unreliable [41]. A potential alternative for summarizing evidence in heterogenic studies is the best-evidence synthesis. This method originates from the guideline on systematic review of the Cochrane Collaboration Back Review Group [42]. The hypothesis of the systematic review was that the continuous infusion gives better clinical outcome than intermittent bolus infusion. Best-evidence synthesis has 5 predetermined levels of evidence that add more weight on RCTs than on retrospective studies: (1) strong, when general consistent findings that continuous infusion led to better clinical outcome than intermittent bolus that were reported in multiple high-quality RCTs; (2) moderate, when 1 highquality RCT and at least 2 high-quality retrospective studies show general consistent findings or when at least 3 high-quality retrospective studies report general consistent findings; (3) limited, when the finding was found in a single RCT, or general consistent findings were found in maximum 2 retrospective studies; (4) conflicting, when no consistent findings were reported; and (5) no evidence, when no study could be found. Clinical outcome variables that were considered for calculation of the best-evidence synthesis were mortality, clinical cure, duration of hospital stay (or duration of ICU stay when data on hospital stay were not available), or any related clinical outcome compatible with the disease under study (eg, duration of medical ventilation in patients with VAP). If a study showed statistically significant better outcome with prolonged than with intermittent bolus infusion in any of these clinical outcome variables, the study was counted as one “positive study.” Using this method, the level of evidence is that the clinical outcome in favor of prolonged PIP/TAZ is moderate. This level of evidence is based on 1 RCT [27] and 4 cohort studies [6,22,39,40]. However, conflicting level of evidence is found when mortality alone was considered as primary clinical outcome variable. 4. Discussion The current review differs from previously published systematic reviews [15-18] because it specifically focuses on PIP/TAZ use in ICU patients. We know from published reviews that continuous infusion of β-lactam antibiotics in hospitalized patients does not provide better clinical outcome than “traditional” intermittent bolus. In contrast, the
present review underscores that prolonged infusion of PIP/TAZ in critically ill patients may be more beneficial than intermittent dosing. Because the included studies show considerable heterogeneity in terms of PIP/TAZ dose, severity of illness, MIC of pathogens, duration of follow-up, and outcome definition, we are unable to quantify the outcome gain obtained with continuous infusion compared with intermittent bolus. Study results appear to be particularly determined by severity of illness and MIC levels. The RCT that failed to show benefit of prolonged infusion mainly enrolled less severely ill patients (N 70% had APACHE II scores below 10) [38]. This is in accordance with results from a retrospective study showing that prolonged infusion was only beneficial in patients with a high (cut-off at 17) APACHE II score [6]. Our narrative review provides solid evidence that prolonged infusion procures a higher TNMIC, which theoretically should lead to better clinical outcome. However, when MIC is low, intermittent bolus can produce sufficient TNMIC and is as effective as continuous infusion. Therefore, no outcome difference between the 2 treatment regimens can be detected. Minimum inhibitory concentrations of pathogens infecting ICU patients are usually higher than in non-ICU patients, and consequently, PIP/TAZ might be withheld in a large part of this population. During our search, we encountered several studies that compare continuous and intermittent bolus of β-lactam antibiotics, where no separate data on PIP/TAZ can be extracted [28,43-45]. In general, these studies pointed toward a more favorable outcome of prolonged vs intermittent bolus infusion. An RCT that included 18 patients on PIP/TAZ of 30 patients on continuous β-lactam antibiotics and 17 patients on PIP/TAZ of 30 patients on intermittent bolus β-lactam antibiotic showed that more patients in the continuous group were clinically cured (76.7% vs 50.0%) [28]. These results were corroborated by 2 cohort studies [43,44], which observed lower mortality in prolonged compared with intermittent bolus infusion of β-lactam antibiotics among which patients on PIP/TAZ. Only 1 study on βlactam antibiotics that compared prolonged with intermittent bolus for the treatment of Gram-negative infections, where patients with PIP/TAZ were included showed no difference in clinical success rate [45]. Well-conducted and sufficiently powered clinical trials and cohort studies are definitely needed to prove whether prolonged PIP/TAZ infusion is superior to intermittent bolus for treatment of infections in critically ill patients. Patient groups should be stratified based on the diagnosis (eg, VAP and intraabdominal infection) and disease severity. Subanalysis should be done on the patients because the benefit of prolonged infusion is potentially most useful in microorganisms with high MIC values; subgroup analysis should also be performed in patients infected with pathogens expressing MICs around the breakpoint. Equivalent PIP/TAZ doses are required for prolonged and intermittent bolus. Data analysis should be adjusted for possible confounders such as concomitant use of other antibiotics than PIP/TAZ and disease severity. Primary clinical outcome parameters should be mortality or cure at a predefined time point, for example, 28 days after start of treatment. Secondary clinical outcome variables should compose evaluation of various morbidity parameters, microbiological eradication, and adverse events. In this context, the multicenter Betalactam Infusion Group II RCT should be mentioned. This trial was started in July 2012 and intends to compare continuous with intermittent bolus of β-lactams, independent of dose, in 420 critically ill patients [46]. The primary outcome of the Beta-lactam Infusion Group II study is ICU-free days at day 28. Secondary outcomes include 90-day survival, clinical cure at 14 days after cessation of the study antibiotic, organ failure–free days at day 14, and duration of bacteremia. In conclusion, PK/PD studies provide a robust rationale to prefer prolonged above intermittent bolus infusion of PIP/TAZ. There is a moderate level of evidence that clinical outcome in critically ill patients on prolonged infusion is better than in patients on
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
E. Yusuf et al. / Journal of Critical Care xxx (2014) xxx–xxx
intermittent infusion. Prolonged infusion of PIP/TAZ should be preferred in treating critically ill patients. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jcrc.2014.07.033. Acknowledgments We thank the medical librarians Jan Schoones (Leiden University Medical Center, Leiden, The Netherlands) and Katrien Alewaeters (Universitair Ziekenhuis Brussel, Brussels, Belgium) for their assistance in the literature research. 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(5):500–3 [Epub 2010/03/17]. [2] Sorgel F, Kinzig M. The chemistry, pharmacokinetics and tissue distribution of piperacillin/tazobactam. J Antimicrob Chemother 1993;31(Suppl A):39–60 [Epub 1993/01/01]. [3] Acar JF, Goldstein FW, Kitzis MD. Susceptibility survey of piperacillin alone and in the presence of tazobactam. J Antimicrob Chemother 1993;31(Suppl A): 23–8 [Epub 1993/01/01]. [4] Andes D, Anon J, Jacobs MR, Craig WA. Application of pharmacokinetics and pharmacodynamics to antimicrobial therapy of respiratory tract infections. Clin Lab Med 2004;24(2):477–502 [Epub 2004/06/05]. [5] Barbosa TM, Levy SB. The impact of antibiotic use on resistance development and persistence. Drug Resist Updat 2000;3(5):303–11 [Epub 2001/08/11]. [6] Lodise Jr TP, Lomaestro B, Drusano GL. Piperacillin-tazobactam for Pseudomonas aeruginosa infection: clinical implications of an extended-infusion dosing strategy. Clin Infect Dis 2007;44(3):357–63 [Epub 2007/01/06]. [7] Lodise Jr TP, Lomaestro B, Rodvold KA, Danziger LH, Drusano GL. Pharmacodynamic profiling of piperacillin in the presence of tazobactam in patients through the use of population pharmacokinetic models and Monte Carlo simulation. Antimicrob Agents Chemother 2004;48(12):4718–24 [Epub 2004/11/25]. [8] Bulitta JB, Duffull SB, Kinzig-Schippers M, Holzgrabe U, Stephan U, Drusano GL, et al. Systematic comparison of the population pharmacokinetics and pharmacodynamics of piperacillin in cystic fibrosis patients and healthy volunteers. Antimicrob Agents Chemother 2007;51(7):2497–507 [Epub 2007/05/09]. [9] Craig WA. Basic pharmacodynamics of antibacterials with clinical applications to the use of beta-lactams, glycopeptides, and linezolid. Infect Dis Clin N Am 2003;17 (3):479–501 [Epub 2004/01/09]. [10] Roberts JA, Lipman J. Pharmacokinetic issues for antibiotics in the critically ill patient. Crit Care Med 2009;37(3):840–51 [quiz 59; Epub 2009/02/25]. [11] Langgartner J, Lehn N, Gluck T, Herzig H, Kees F. Comparison of the pharmacokinetics of piperacillin and sulbactam during intermittent and continuous intravenous infusion. Chemotherapy 2007;53(5):370–7 [Epub 2007/09/06]. [12] Patel GW, Patel N, Lat A, Trombley K, Enbawe S, Manor K, et al. Outcomes of extended infusion piperacillin/tazobactam for documented Gram-negative infections. Diagn Microbiol Infect Dis 2009;64(2):236–40 [Epub 2009/06/09]. [13] Abdul-Aziz MH, Dulhunty JM, Bellomo R, Lipman J, Roberts JA. Continuous beta-lactam infusion in critically ill patients: the clinical evidence. Ann Intensive Care 2012;2(1):37 [Epub 2012/08/18]. [14] McKenzie C. Antibiotic dosing in critical illness. J Antimicrob Chemother 2011;66 (Suppl 2):ii25–31. [15] Falagas ME, Tansarli GS, Ikawa K, Vardakas KZ. Clinical outcomes with extended or continuous versus short-term intravenous infusion of carbapenems and piperacillin/tazobactam: a systematic review and meta-analysis. Clin Infect Dis 2013;56(2):272–82 [Epub 2012/10/18]. [16] Hayashi Y, Roberts JA, Paterson DL, Lipman J. Pharmacokinetic evaluation of piperacillintazobactam. Expert Opin Drug Metab Toxicol 2010;6(8):1017–31 [Epub 2010/07/20]. [17] Kasiakou SK, Sermaides GJ, Michalopoulos A, Soteriades ES, Falagas ME. Continuous versus intermittent intravenous administration of antibiotics: a meta-analysis of randomised controlled trials. Lancet Infect Dis 2005;5(9):581–9 [Epub 2005/08/27]. [18] Mah GT, Mabasa VH, Chow I, Ensom MH. Evaluating outcomes associated with alternative dosing strategies for piperacillin/tazobactam: a qualitative systematic review. Ann Pharmacother 2012;46(2):265–75 [Epub 2012/01/26]. [19] Roberts JA, Lipman J, Blot S, Rello J. Better outcomes through continuous infusion of time-dependent antibiotics to critically ill patients? Curr Opin Crit Care 2008;14 (4):390–6 [Epub 2008/07/11]. [20] Fahimi F, Ghafari S, Jamaati H, Baniasadi S, Tabarsi P, Najafi A, et al. Continuous versus intermittent administration of piperacillin-tazobactam in intensive care unit patients with ventilator-associated pneumonia. Indian J Crit Care Med 2012; 16(3):141–7 [Epub 2012/11/29]. [21] Goncalves-Pereira J, Oliveira BS, Janeiro S, Estilita J, Monteiro C, Salgueiro A, et al. Continuous infusion of piperacillin/tazobactam in septic critically ill patients—a multicenter propensity matched analysis. PLoS One 2012;7(11):e49845. [22] Waxier C, Hu D. Clinical outcomes of prolonged-infusion piperacillin/tazobactam in patients admitted to the intensive care unit. Crit Care Med 2012;40(12):U101. [23] Aird WC. The role of the endothelium in severe sepsis and multiple organ dysfunction syndrome. Blood 2003;101(10):3765–77 [Epub 2003/01/25].
7
[24] Smith BS, Yogaratnam D, Levasseur-Franklin KE, Forni A, Fong J. Introduction to drug pharmacokinetics in the critically ill patient. Chest 2012;141(5):1327–36 [Epub 2012/05/04]. [25] Udy AA, Roberts JA, Boots RJ, Paterson DL, Lipman J. Augmented renal clearance: implications for antibacterial dosing in the critically ill. Clin Pharmacokinet 2010; 49(1):1–16 [Epub 2009/12/17]. [26] Udy AA, Putt MT, Shanmugathasan S, Roberts JA, Lipman J. Augmented renal clearance in the intensive care unit: an illustrative case series. Int J Antimicrob Agents 2010;35(6):606–8 [Epub 2010/03/24]. [27] Rafati MR, Rouini MR, Mojtahedzadeh M, Najafi A, Tavakoli H, Gholami K, et al. Clinical efficacy of continuous infusion of piperacillin compared with intermittent dosing in septic critically ill patients. Int J Antimicrob Agents 2006;28(2):122–7 [Epub 2006/07/04]. [28] Dulhunty JM, Roberts JA, Davis JS, Webb SA, Bellomo R, Gomersall C, et al. Continuous infusion of beta-lactam antibiotics in severe sepsis: a multicenter double-blind, randomized controlled trial. Clin Infect Dis 2013;56(2):236–44 [Epub 2012/10/18]. [29] Zelenitsky SA, Ariano RE, Zhanel GG. Pharmacodynamics of empirical antibiotic monotherapies for an intensive care unit (ICU) population based on Canadian surveillance data. J Antimicrob Chemother 2011;66(2):343–9 [Epub 2010/10/12]. [30] Roberts JA, Kirkpatrick CM, Roberts MS, Dalley AJ, Lipman J. First-dose and steadystate population pharmacokinetics and pharmacodynamics of piperacillin by continuous or intermittent dosing in critically ill patients with sepsis. Int J Antimicrob Agents 2010;35(2):156–63 [Epub 2009/12/19]. [31] Angus BJ, Smith MD, Suputtamongkol Y, Mattie H, Walsh AL, Wuthiekanun V, et al. Pharmacokinetic-pharmacodynamic evaluation of ceftazidime continuous infusion vs intermittent bolus injection in septicaemic melioidosis. Br J Clin Pharmacol 2000;50(2):184–91 [Epub 2000/08/10]. [32] Tam VH, McKinnon PS, Akins RL, Rybak MJ, Drusano GL. Pharmacodynamics of cefepime in patients with Gram-negative infections. J Antimicrob Chemother 2002;50(3):425–8 [Epub 2002/09/03]. [33] Mouton JW, den Hollander JG. Killing of Pseudomonas aeruginosa during continuous and intermittent infusion of ceftazidime in an in vitro pharmacokinetic model. Antimicrob Agents Chemother 1994;38(5):931–6 [Epub 1994/05/01]. [34] Duszynska W, Taccone FS, Switala M, Hurkacz M, Kowalska-Krochmal B, Kübler A. Continuous infusion of piperacillin/tazobactam in ventilator-associated pneumonia: a pilot study on efficacy and costs. Int J Antimicrob Agents 2012;39(2):153–8. [35] Joukhadar C, Frossard M, Mayer BX, Brunner M, Klein N, Siostrzonek P, et al. Impaired target site penetration of beta-lactams may account for therapeutic failure in patients with septic shock. Crit Care Med 2001;29(2):385–91 [Epub 2001/03/14]. [36] Roberts JA, Roberts MS, Robertson TA, Dalley AJ, Lipman J. Piperacillin penetration into tissue of critically ill patients with sepsis—bolus versus continuous administration? Crit Care Med 2009;37(3):926–33 [Epub 2009/02/25]. [37] Boselli E, Breilh D, Cannesson M, Xuereb F, Rimmele T, Chassard D, et al. Steady-state plasma and intrapulmonary concentrations of piperacillin/tazobactam 4 g/0.5 g administered to critically ill patients with severe nosocomial pneumonia. Intensive Care Med 2004;30(5):976–9. [38] Lau WK, Mercer D, Itani KM, Nicolau DP, Kuti JL, Mansfield D, et al. Randomized, open-label, comparative study of piperacillin-tazobactam administered by continuous infusion versus intermittent infusion for treatment of hospitalized patients with complicated intra-abdominal infection. Antimicrob Agents Chemother 2006;50(11):3556–61 [Epub 2006/08/31]. [39] Lee GC, Liou H, Yee R, Quan CF, Neldner K. Outcomes of extended-infusion piperacillin-tazobactam: a retrospective analysis of critically ill patients. Clin Ther 2012;34(12):2297–300 [Epub 2012/12/01]. [40] Lorente L, Jimenez A, Martin MM, Iribarren JL, Jimenez JJ, Mora ML. Clinical cure of ventilator-associated pneumonia treated with piperacillin/tazobactam administered by continuous or intermittent infusion. Int J Antimicrob Agents 2009;33(5): 464–8 [Epub 2009/01/20]. [41] Falagas ME, Lourida P, Poulikakos P, Rafailidis PI, Tansarli GS. Antibiotic treatment of infections due to carbapenem-resistant Enterobacteriaceae: systematic evaluation of the available evidence. Antimicrob Agents Chemother 2014;58(2): 654–63 [Epub 2013/10/02]. [42] van Tulder M, Furlan A, Bombardier C, Bouter L, Editorial Board of the Cochrane Collaboration Back Review G. Updated method guidelines for systematic reviews in the cochrane collaboration back review group. Spine 2003;28(12):1290–9 [Epub 2003/06/18]. [43] Dow RJ, Rose WE, Fox BC, Thorpe JM, Fish JT. Retrospective study of prolonged versus intermittent infusion piperacillin-tazobactam and meropenem in intensive care unit patients at an academic medical center. Infect Dis Clin Pract 2011;19(6):413–7. [44] Yost RJ, Cappelletty DM. The Retrospective Cohort of Extended-Infusion Piperacillin-Tazobactam (RECEIPT) study: a multicenter study. Pharmacotherapy 2011;31(8):767–75. [45] Arnold HM, Hollands JM, Skrupky LP, Smith JR, Juang PH, Hampton NB, et al. Prolonged infusion antibiotics for suspected gram-negative infections in the ICU: a before-after study. Ann Pharmacother 2013;47(2):170–80 [Epub 2013/01/24]. [46] Dulhunty JM, Roberts JA, Davis JS, Webb SA, Bellomo R, Gomersall C, et al. A protocol for a multicentre randomised controlled trial of continuous beta-lactam infusion compared with intermittent beta-lactam dosing in critically ill patients with severe sepsis: the BLING II study. Crit Care Resusc 2013;15(3):179–85 [Epub 2013/08/16].
Please cite this article as: Yusuf E, et al, Prolonged vs intermittent infusion of piperacillin/tazobactam in critically ill patients: A narrative and systematic review, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.07.033
