Adham Mohamed, PharmD and Timothy Berry, PharmD, BCPS
Saint Luke’s Hospital of Kansas City
Piperacillin-tazobactam (PTZ) is a combination of an expanded spectrum anti-pseudomonal penicillin and a β-lactamase inhibitor which protects piperacillin from degradation by β-lactamases produced by methicillin-sensitive Staphylococcus aureus, Haemophilus influenzae, and Morganella catarrhalis.1,2
PTZ is approved by the Food and Drug Administration (FDA) for the treatment of intra-abdominal infections (IAI), uncomplicated and complicated skin and soft tissue infections (SSTI), female pelvic infections, community-acquired pneumonia, and nosocomial pneumonia.3
PTZ provides coverage against a wide variety of gram-positive and gram-negative organisms (table 1).3 PTZ also exerts activity against some anaerobic bacteria (e.g. Bacteroides fragilis).3
The 2016 Infectious Diseases Society of America (IDSA) guidelines for the treatment of hospital-acquired pneumonia (HAP)and ventilator-associated pneumonia (VAP) suggest the use of PTZ in combination with non-β-lactam antipseudomonal agents for the treatment of VAP and suggested the use of PTZ for the treatment of HAP.4 The 2014 IDSA guidelines for the treatment of SSTI recommended PTZ in combination with anti-methicillin-resistant S. aureus (MRSA) antimicrobials for the empiric treatment of severe infections such as; necrotizing fasciitis, clostridial gas gangrene or myonecrosis, SSTIs during an episode of fever and neutropenia and cellulitis in severely compromised patients.5 The 2010 IDSA guidelines for the treatment of IAI also suggested the use of PTZ for the treatment of high-risk community acquired IAI or healthcare-associated IAI.6 The aforementioned HAP/VAP and SSTI guidelines suggested PTZ among a variety of other agents, however, PTZ always appears on the top of the list of suggested empiric antimicrobials.4,5
Based on the factors mentioned earlier, PTZ became a very common empiric first-line antimicrobial agent for critically ill patients. A 2011 survey of 183 US hospitals looking at the prevalence of antimicrobial medication use in acute care hospitals found that PTZ was the second most common antibiotic used to treat infections in critically ill patients, making up 25.6% of all antimicrobial medications given. This survey also showed that PTZ is one of the top antimicrobial used to treat infection in non-critically ill patients.7 The survey also showed that vancomycin is the most commonly used antimicrobial in critically ill patients, which reflects a widespread practice of prescribing a combination of PTZ and vancomycin empirically to treat infections in critically ill patients.7
Critical care clinicians often prefer using PTZ as a first line empiric antibiotic as it provides broad coverage against gram-positive and gram-negative organisms, especially in patients with risk factors for multi-drug resistant organisms.
Recently, there has been a growing concern with the empiric use of PTZ as a first-line antimicrobial for the treatment of infections in critically ill patients. These concerns include the increased incidence of extended-spectrum β -lactamase (ESBL)-producing bacteria and PTZ’s inferiority in treating infections caused by these bacteria and the risk of acute kidney injury (AKI) with PTZ and vancomycin combination therapy.
PTZ and ESBL-Producing Bacteria
Antibiotic-resistant infections are a growing challenge; causing two million infections per year which results in 23,000 deaths per year.8 The Center for Disease Control (CDC) categorizes ESBL-producing Enterobacteriaceae as a serious threat.8 ESBL-producing Enterobacteriaceae cause 26,000 infections per year, which results in 1700 deaths.8 Infections with ESBL-producing Enterobacteriaceae have a 57% higher mortality rate and increased healthcare cost by 40,000 USD.8 The prevalence of ESBL-producing bacteria is rising worldwide.9 CTX-M, TEM, and SHV are the major ESBL enzymes, and CTX-M is the most prevalent type globaly.9,10
As we see an increase in the incidence of infections caused by ESBL-producing bacteria in critically ill patients, PTZ may not be the best option to treat ESBL-producing bacteria, namely bacteremia. Carbapenems remain the drug of choice to treat ESBL-producing bacteria.
A retrospective cohort study in 37 hospitals from 12 countries compared β-lactam/β-lactamase inhibitor combinations (BLBLIs) to carbapenems in patients with clinically significant bloodstream infection due to ESBL or carbapenemase-producing Enterobacteriaceae.11 The study had two cohorts; empiric therapy cohort (ETC) and targeted therapy cohort (TTC). The ETC had 170 patients in BLBLIs group, and 195 patients in the carbapenem group with 65% in the BLBLIs group receiving PTZ. In the TTC, 92 patients received BLBLIs, while 509 patients received carbapenem; 83% of the patients in the BLBLI group received PTZ.11
There was no difference in the primary outcome of 30 days mortality in both cohorts In the ETC, the BLBLI group had 17.6% mortality compared to 20% in the carbapenem group (P = 0.6) and in the TTC, mortality in BLBLI group was 13.9% vs. 9.8% in the carbapenem group (P = 0.28).11
Ng et al. conducted a single-center, retrospective, cohort study that included 151 patients with E. coli and K. pneumoniae ESBL bacteremia. There were 94 patients in the PTZ group and 57 patients in the carbapenem group. The authors did not find a difference in the primary endpoint of 30 days mortality between the two groups (PTZ 30.9% vs. carbapenem 29.8%, P = 0.89). The authors observed a statistically significant higher incidence of multi-drug resistant and fungal infections in the carbapenem arm (PTZ 7.4% vs. carbapenem 24.6%, P<0.01) and relapsed bacteremia (PTZ 3.2 % vs. carbapenem 15.8%, P = 0.05).12
A retrospective study by Tamma et al. included 213 patients with ESBL bacteremia who received empiric PTZ or a carbapenem and received carbapenem as definitive therapy after blood cultures. In this study, the empiric PTZ group had a higher 14 day mortality after the first positive blood culture with empiric PTZ (17% vs. 8%). After adjusting for covariates, empiric PTZ had a 1.92 times increased risk of death (95% CI, 1.07–3.45; P = 0.03).13
Recently the MERINO trial was published, which was the first randomized control trial to compare PTZ vs. meropenem as a definitive therapy of bloodstream infection caused by ESBL-producing E. coli or Klebsiella spp resistant to ceftriaxone but susceptible to PTZ. The MERINO trial was a multicenter, open-label, parallel group, randomized clinical trial. It was conducted in 26 hospitals in 9 countries. This was a non-inferiority trial, and the primary endpoint was all-cause mortality at 30 days after randomization.14
The MERINO trial was terminated early after enrolling 391 patients (PTZ, n= 196 and meropenem, n=195). The 30 days all-cause mortality in the PTZ group was significantly higher compared to the meropenem group (12.3% vs 3.7%) [risk difference, 8.6% (1-sided 97.5% CI, −∞ to 14.5%); P = 0.90 for non-inferiority]14. The authors concluded the PTZ was inferior to meropenem in the treatment of ESBL producing E. coli or Klebsiella spp. resistant to ceftriaxone but susceptible to PTZ.14
PTZ and Vancomycin-Induced AKI
Acute kidney injury (AKI) incidence in critically ill patients ranges from 22% to 57%.15,16 AKI presents several challenges as well as higher mortality in critically ill patients.15,16 As mentioned earlier, vancomycin and PTZ are the first and second most commonly used antibiotics in critically ill patients.7 If combining PTZ and vancomycin creates an additional risk for developing AKI in critically ill patients, then we may need to re-evaluate our initial empiric antibiotic strategies in critically ill patients.
A meta-analysis was conducted by Luther et al. of 15 studies and 17 abstracts that evaluated the risk of AKI with PTZ plus vancomycin compared to other antibiotic combinations (vancomycin monotherapy, vancomycin plus cefepime or carbapenem, or PTZ monotherapy).17 This meta-analysis included 24,799 adult patients. The PTZ plus vancomycin group had a 22.2% incidence of AKI, which was significantly higher than the incidence in all the other combinations (12.9%) (P<0.0001). It is important to note that only 968 patients of 24,799 patients (3.9%) were critically ill.17
Buckley et al. conducted a retrospective cohort study that included 333 critically ill patients. The authors compared the incidence of AKI with PTZ plus vancomycin compared to cefepime plus vancomycin.18 The incidence of AKI was not different between the two groups; 19.5% in the PTZ plus vancomycin group compared to 17.3% in cefepime plus vancomycin group (P=0.61). However, there was a significant increase in need of renal replacement therapy in the PTZ plus vancomycin group (10%) compared to cefepime plus vancomycin group (3.8%) (P=0.04).18
A single-center, retrospective cohort study compared the incidence of AKI in PTZ plus vancomycin to cefepime plus vancomycin and meropenem plus vancomycin.19 This study included 2,492 critically ill patients. The incidence of AKI was 39.3% for PTZ plus vancomycin, 24.2% for cefepime plus vancomycin, and 23.5% for meropenem plus vancomycin (P<0.0001 for all 3 comparisons).19 The authors conducted a multivariate analysis which showed the PTZ plus vancomycin combination was an independent predictor of AKI (OR 2.161; 95% CI, 1.62-2.833).19
On the other hand, a single-center retrospective cohort study compared the incidence of AKI in PTZ plus vancomycin to cefepime plus vancomycin and meropenem plus vancomycin in critically ill patients receiving short courses of the combination antibiotics (24-72 hours).20 This study included 3,299 critically ill patients. The authors did not observe a difference in the incidence of AKI between the groups.20 Adjusted OR for PTZ plus vancomycin compared to cefepime plus vancomycin was 1.11 (95% CI, 0.85 - 1.45) and adjusted OR for PTZ plus vancomycin compared meropenem plus vancomycin was 1.04 (95% CI, 0.71 - 1.42).20 This study showed a lower risk of AKI if PTZ was used for less than 3 days. However, the current recommendation is to treat bacteremia for 7-14 days, which can put patients at risk if clinicians were reluctant to de-escalate early.
Where we go from here?
It is time to rethink the practice of blanket coverage of every critically ill patient with PTZ. We should individualize empiric antimicrobial therapy based on patient history, site of infection, and MDR pathogens risk factors.
The use of PTZ should be reserved for situations where anaerobic coverage is needed, such as diabetic foot infections or complicated intra-abdominal infection. Historically, anaerobic coverage was considered a cornerstone in the treatment of aspiration pneumonia. However, studies have shown that the pathogenesis of aspiration pneumonia has shifted from anaerobic to aerobic pathogens.21 Anaerobic coverage may be still needed in patients with poor dental health21. These findings raise the question, do we still need to use PTZ in pneumonia?
Based on the current evidence, it seems that there is an association between PTZ plus vancomycin and the risk of AKI. The data in critically ill patients is still conflicting — a prospective study in critically ill patients is needed to determine the causality.
PTZ is an excellent antibiotic and still offer excellent coverage in situations when broad gram-negative and anaerobic coverage is needed, but other alternatives can be used when it comes to pneumonia, SSTIs, or blood-stream infections.
Alternatives to PTZ include ceftriaxone, cefotaxime, or ampicillin/sulbactam for severe community acquired pneumonia and ceftazidime, or cefepime for HAP/VAP.4
As the incidence and prevalence of multi-drug resistance are increasing worldwide, we as critical care clinicians need to take steps to ensure the appropriate use of all antimicrobials, not just PTZ. The following are suggested measures to improve appropriate antimicrobial use:
Submit for CE