757497

research-article2018
AOPXXX10.1177/1060028018757497Annals of PharmacotherapyMullins et al

Research Report
Annals of Pharmacotherapy

Comparison of the Nephrotoxicity

1­–6
© The Author(s) 2018
Reprints and permissions:
of Vancomycin in Combination sagepub.com/journalsPermissions.nav
DOI: 10.1177/1060028018757497
https://doi.org/10.1177/1060028018757497

With Cefepime, Meropenem, or journals.sagepub.com/home/aop

Piperacillin/Tazobactam: A
Prospective, Multicenter Study

Brandon P. Mullins, PharmD1, C. Joseph Kramer, PharmD2,

Billie J. Bartel, PharmD3, Jennifer S. Catlin, PharmD4,
and Richard E. Gilder, RN-BC, MS2

Abstract
Background: Patients often receive broad-spectrum antibiotics for nosocomial infections commonly with activity against
Pseudomonas aeruginosa and methicillin-resistant Staphylococcus aureus. Previous retrospective and/or single-center studies
have suggested that the combination of vancomycin and piperacillin/tazobactam might be associated with an increased
risk of acute kidney injury. Objectives: To compare the incidence of nephrotoxicity in patients receiving intravenous
vancomycin in combination with cefepime, meropenem, or piperacillin/tazobactam. Methods: This was a prospective,
multicenter observational study of patients receiving vancomycin in combination with piperacillin/tazobactam versus
cefepime or meropenem. Adult patients 18 years of age or older who were hospitalized and received 72 or more hours of
intravenous vancomycin and 72 hours or more of cefepime, meropenem, or piperacillin/tazobactam were eligible. Patient
and medication characteristics were examined for the 242 patients included. Results: The incidence of acute kidney
injury for patients treated with vancomycin and piperacillin/tazobactam was significantly higher than for those treated
with vancomycin and cefepime or meropenem, 29.8% versus 8.8%, respectively, P < 0.001. Binary logistic regression
demonstrated that patients receiving vancomycin and piperacillin/tazobactam were 6.7 times more likely to develop acute
kidney injury compared with the other cohort. Conclusions: The combination of vancomycin with piperacillin/tazobactam
significantly increases the risk of acute kidney injury compared with other broad-spectrum antibiotic combinations.
Clinicians should be vigilant when employing this regimen.

Keywords
vancomycin, nephrotoxicity, antibiotics, beta-lactams, cephalosporins

Background trend is likely to continue and it is important to ensure that

Empiric antimicrobial coverage for health care–related
infections typically includes multiple broad-spectrum anti-
biotics, specifically those with activity against Pseudomonas
aeruginosa and methicillin-resistant Staphylococcus
aureus, as well as other likely pathogens.1,2 Choice of which
agents to use depends on several factors including local sus-
ceptibility patterns as well as patient-specific characteristics
(ie, recent history of antibiotic use and drug allergies). In
the previous decade, antibiotic use has increased 36%, with
the largest increases seen with broad-spectrum penicillins,
cephalosporins, flouroquinolones, and carbapenems.3 This
all safety concerns with these agents are fully understood.
Until recently, acute kidney injury (AKI) was not a
major concern when selecting between anti-pseudomonal
1
St Luke’s Hospital, Chesterfield, MO, USA
2
Baylor Scott and White Research Institute, Dallas, TX, USA
3
Avera McKennan Hospital and University Health Center, Sioux Falls,
SD, USA
4
CoxHealth Medical Centers, Springfield, Missouri, MO, USA
Corresponding Author:
Brandon P. Mullins, St Luke’s Hospital, 232 S Woods Mill Road,
Chesterfield, MO 63017, USA.
Email: Brandon.Mullins@stlukes-stl.com
2 Annals of Pharmacotherapy 00(0)
beta-lactams; however, recent literature has raised con-
cerns about the potential AKI occurrence with piperacil-
lin/tazobactam (PT) in combination with vancomycin.
Vancomycin nephrotoxicity is a well-documented concern
that has persisted since the drug’s introduction, but cases
are generally mild and reversible.4 Many risk factors for
vancomycin-induced nephrotoxicity exist: prolonged
treatment duration, vancomycin concentrations >15 mg/L,
use of concurrent nephrotoxic agents, and certain host fac-
tors (history of AKI, chronic kidney disease [CKD], and
admission to the intensive care unit).4
Previously, 3 retrospective studies were published exam-
ining the correlation of AKI in conjunction with PT or
cefepime. In one study of patients treated for osteomyelitis,
Moenster and colleagues5 reported the incidence of AKI for
vancomycin + PT (VPT) and vancomycin + cefepime (VC)
to be 31.2% and 19.5%, respectively, but this difference
was not statistically significant. In another retrospective
study, it was found that VPT patients had a 34.8% incidence
of AKI compared with only 12.5% of VC patient (95% con-
fidence interval [CI] 1.89-7.39, P < 0.0001).6 Last, a third
study published looked at patients receiving vancomycin
with and without PT and reported that patients in their study
who were receiving VPT were almost 2.5 times more likely
to develop AKI than those patient receiving vancomycin
alone (95% CI >1.11, P = 0.032).7
While this study was being conducted, 3 additional pub-
lications were found examining this subject. A ­single-center,
retrospective study examining VPT versus VC in critically
ill patients conducted by Hammond and colleagues8 did not
find a difference in the rates of AKI (VPT 32.7% vs 28.8%,
P = 0.761). Likewise, a meta-analysis of 10 studies of van-
comycin versus VPT and 5 studies of VPT versus vancomy-
cin combined with another beta-lactam found similar
results.9 Vancomycin alone had lower rates of AKI com-
pared with VPT (odds ratio [OR] 3.980, P < 0.001) but the
comparison of VPT versus vancomycin + beta-lactam had
similar rates of AKI (OR 3.029, P = 0.063). Last, a small,
single-center study compared VPT with VC or vancomycin
+ meropenem (VM), similar to our study, and found the
rates of AKI for VPT to be 37.3% versus 7.7% for the VC/
VM group, P = 0.005.10
Based on the conflicting results of the aforementioned
studies and the uncertainties that remain with certain broad-
spectrum antibiotic combinations, this prospective study
was designed to compare the incidence of AKI in patients
receiving intravenous (IV) vancomycin in combination
with cefepime, meropenem, or PT.
Methods
Study Design and Outcomes
This was a prospective, multicenter observational study
examining the incidence of AKI in patients receiving IV
vancomycin in combination with cefepime, meropenem, or
PT. This study was conducted at 4 medical centers around
the United States (see the appendix for full list of participat-
ing institutions). The study was approved by all institutional
review boards of the participating institutions before collec-
tion of data. The primary objective of this study was to
determine if there is a difference in the incidence of AKI in
patients receiving a combination of IV vancomycin with PT
versus meropenem or cefepime. For the purposes of this
investigation, the definition of AKI is a minimum 1.5-fold
increase in serum creatinine (SCr; baseline vs maximum
within first 7 days of antimicrobial therapy) similar to pre-
vious studies.5-7 Secondary objectives were to determine if
the presence of the following risk factors increased the inci-
dence of AKI for patients receiving the above-mentioned
antibiotic combinations: intensive care unit admission, con-
comitant use of nephrotoxic agents (see the appendix), age
≥65, certain comorbid disease states (see the appendix), or
vancomycin trough concentrations ≥15 mg/L. All partici-
pating institutions used a 4-hour extended-infusion of PT
rather than standard intermittent infusion of 30 minutes.
Participants
Adult patients 18 years of age or older who were hospital-
ized and received 72 or more hours of IV vancomycin and
72 hours or more of cefepime, meropenem, or PT were eli-
gible for inclusion into this study (antibiotics must overlap
for at least 48 hours). Patients were excluded from the study
if they had a documented history of CKD (K/DOQI stage 3
or higher), baseline SCr ≥ 1.5 mg/dL, or AKI at any time
during their hospital stay prior to receipt of study antibiotics
(defined as creatinine clearance [CrCl] < 30 mL/min via
Cockroft-Gault or a 1.5-fold increase in SCr or greater). If
at any time a random or trough vancomycin concentration
was measured <10 mg/L before discontinuation of therapy
the patient was excluded as this does not represent an ade-
quate steady-state vancomycin regimen. Additionally, if a
patient switched from one stratification of antibiotics to
another or if they received any dose of the study antibiotics
between hospital admission and enrollment into the study
they were excluded. Furthermore, if a patient suffered car-
diac arrest before initiation of antimicrobials they were
excluded.
Data Collection
The following patient information and laboratory values
were recorded from the electronic medical records of the
participating institutions: age, gender, weight, height, con-
comitant nephrotoxic agents, comorbid disease states,
Charlson Comorbidity Index, dosing regimen and duration
of specified antibiotics, SCr concentrations (initial and max-
imum), vancomycin concentrations (trough and random),
hospital unit, length of hospital stay, incidence of renal
Mullins et al 3

replacement therapies (RRT), time to resolution of AKI, pri- Table 1. Justification for Exclusiona.
mary infection site, if sepsis type of sepsis (not indicative of
n
primary site of infection), and SCr at day of hospital dis-
charge or last measured SCr. Vancomycin concentrations CKD 3 or higher 139
obtained before steady state (before fourth scheduled dose) Baseline SCr 1.5 mg/dL or higher 32
or trough concentrations drawn >2 hours before the next Baseline CrCl <30 ml/min 33
scheduled dose were regarded as random concentrations. If AKI before antibiotics 103
a steady-state trough concentration was obtained between 2 [Vancomycin] <10 mg/L 91
and 1 hours prior to the next scheduled dose the value was Antibiotic switch 59
extrapolated using population-based pharmacokinetic equa- Prior receipt of study drug 31
tions (see the appendix). Patients were followed from time Cardiac arrest prior to antibiotics 2
No vancomycin concentration 8
of inclusion until hospital discharge.
Total 498

Abbreviations: CKD, chronic kidney disease; SCr, serum creatinine;

Statistical Analysis CrCl, creatinine clearance; AKI, acute kidney injury.
a
To detect a 10% difference in the incidence of AKI between Data are from total number of patients.
VPT and VC or (VM) groups, a sample size of 540 patients
(270 in each arm) was estimated to achieve a statistical Table 2. Baseline Patient Characteristics.
power of 80% based on the estimates of a 25% risk of AKI VC/VM
in the VPT group and 15% AKI in both the VC and VM Characteristica VPT (n = 94) (n = 148) P Value
groups. Chi-squared or Fisher exact test were used for dif-
ferences in nominal data. Comparisons between continuous Age (years) 67.1 ± 1425 65.1 ± 15.18 0.883
data were performed using Student’s t test, median test, or Male gender (%) 60.6 (57/94) 71.4 (80/148) 0.314
Mann-Whitney U test, as appropriate. A multivariate analy- Height (cm) 170 ± 18.74 171.4 ± 10.73 0.692
Weight (kg) 91.3 ± 38.66 87.4 ± 27.10 0.736
sis was performed to determine if certain characteristics or
Baseline SCr (mg/dL) 0.86 ± 0.20 0.84 ± 0.23 0.305
medications were independent risk factors for development
Baseline CrCl (mL/min) 77.3 ± 29.44 80 ± 34.08 0.769
of AKI. The Type I error (α) probability is 0.05. A P value
ICU (%) 34 (32/94) 41 (61/148) 0.263
of 0.05 or less was considered statistically significant for
PMH diabetes (%) 37.2 (35/94) 33.1 (49/148) 0.511
the purposes of this study. A prespecified subgroup analysis PMH COPD (%) 30.9 (29/94) 35.1 (53/148) 0.491
was performed of patients in the VPT arm to identify any PMH CHF (%) 23.4 (22/94) 18.2 (27/148) 0.33
additional risk factors for developing AKI. A preplanned PMH CAD (%) 35.1 (33/94) 31.8 (47/148) 0.589
interim analysis at 200 patients was conducted to surveil for PMH PVD (%) 12.8 (12/94) 12.2 (18/148) 0.89
an overwhelming difference or no difference between CCI 2.9 ± 1.92 2.6 ± 1.80 0.182
cohorts based on previous reports. This was done to ensure
no further delay of potentially significant results would Abbreviations: VPT, vancomycin + piperacillin/tazobactam; VC,
vancomycin + cefepime; VM, vancomycin + meropenem; SCr, serum
ensue. creatinine; CrCl, creatinine clearance; ICU, intensive care unit; PMH,
past medical history; COPD, chronic obstructive pulmonary disease;
CHF, congestive heart failure; CAD, coronary artery disease; PVD,
Results peripheral vascular disease.
a
Data are mean ± SD values unless otherwise indicated.
Between December 2014 and August 2016, 740 patients
were screened for inclusion into the study, of which 242
patients were included. Ninety-four patients received VPT Table 3. Primary Site of Infection.
and 148 received either VM (47) or VC (101). A total of 498
patients were excluded mostly due to having baseline CKD Site of Infection VPT (n = 94) VC/VM (n = 148) P Value
stage 3 or higher or development of AKI before initiation of Blood (%) 3.2 (3/94) 6.8 (10/148) 0.231
antimicrobials (Table 1). There were no differences in Intraabdominal (%) 8.5 (8/94) 4.1 (6/148) 0.148
Charlson Comorbidity Index or any other baseline charac- Osteomyelitis (%) 1.1 (1/94) 2 (3/148) 1.000
teristics between patient groups (Table 2). During a pre- Other (%) 4.3 (4/94) 9.5 (14/148) 0.133
specified interim analysis, there was a significant difference Respiratory (%) 60.1 (57/94) 60.8 (90/148) 0.979
between the 2 groups for the primary outcome, and thus, Skin (%) 20.2 (19/94) 15.5 (23/148) 0.350
patient enrollment was stopped before the target sample UTI (%) 2.1 (2/94) 1.4 (2/148) 0.643
size was reached.
Abbreviations: VPT, vancomycin + piperacillin/tazobactam; VC,
Most patients in the study were being treated for respira- vancomycin + cefepime; VM, vancomycin + meropenem; UTI, urinary
tory infections or skin and skin structure infections (Table 3). tract infection.
4 Annals of Pharmacotherapy 00(0)

Table 4. Sepsis Type If Present. Table 6. Concomitant Nephrotoxic Agents.

Sepsis Category VPT (n = 94) VC/VM (n = 148) P Value Medication (%) VPT (n = 94) VC/VM (n = 148) P Value
None (%) 60.6 (57/94) 64.1 (94/148) 0.577 ACEI/ARB 29.8 31.8 0.777
Sepsis (%) 25.5 (24/94) 20.3 (31/148) 0.338 Acyclovir 1.1 2 1.000
Severe sepsis (%) 8.5 (8/94) 8.8 (13/148) 0.941 Aminoglycosides 1.1 1.4 1.000
Septic shock (%) 5.3 (5/94) 6.8 (10/148) 0.651 Amphotericin B 0 0.7 1.000
Calcineurin inhibitor 0 3.4 0.160
Abbreviations: VPT, vancomycin + piperacillin/tazobactam; VC,
IV contrast 44.8 37.2 0.282
vancomycin + cefepime; VM, vancomycin + meropenem.
IV immunoglobulin 0 0.7 1.000
Loop diuretic 50 50.7 1.000
Table 5. Vancomycin Characteristicsa. NSAID 8.5 11.5 0.404
Sulfonamide 2.1 3.4 0.709
VPT VC/VM Vasopressors 7.5 11.5 0.380
Characteristic (n = 94) (n = 148) P Value
Abbreviations: VPT, vancomycin + piperacillin/tazobactam; VC,
Median individual dose (mg) 1250 1250 0.598 vancomycin + cefepime; VM, vancomycin + meropenem; ACEI,
Duration of Tx (hours) 131.7 153.7 0.681 angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor
Minimum trough (mg/L)b 16.3 15.2 0.180 blocker; NSAID, nonsteroidal antiinflammatory drug.
Maximum trough (mg/L) 19.6 18.9 0.619
mg/kg/dose 14.9 15.0 0.895 those who received VC or VM (OR 6.65, 95% CI 2.79-
mg/kg/day 29.1 29.7 0.692
15.84, P < 0.001). Additionally, patients receiving loop
Abbreviations: VPT, vancomycin + piperacillin/tazobactam; VC, diuretics (OR 3.27, 95% CI 1.42-7.53, P = 0.005) or vaso-
vancomycin + cefepime; VM, vancomycin + meropenem. pressors (OR 5.04, 95% CI 1.66-15.35, P = 0.004) were
a
Data are mean values unless otherwise indicated. also at increased risk of developing AKI. Last, any patient
b
Trough concentrations at steady state.
with a maximum vancomycin trough >30 mg/L was also at
increased risk of AKI (OR 13.33, 95% CI 3.13-56.77, P <
Just over a third of patients (37.2%) had a diagnosis of sepsis 0.001). Other variables analyzed were not found to be asso-
(90/242). Of those that did, 22.3% (54/242) had sepsis and ciated with the development of AKI.
14.9% (36/242) had a diagnosis of either severe sepsis or sep- A prespecified subgroup analysis of the 94 patients
tic shock (Table 4). Vancomycin dose and duration were receiving VPT showed 3 significant risk factors for the
comparable between groups. Both the minimum and maxi- development of AKI. Patients receiving VPT plus either a
mum trough concentrations were similar between groups loop diuretic (42.5%, P = 0.012) or vasopressors (85.7%,
(Table 5). There was no difference in the incidence of con- P = 0.003) had a further increased risk of AKI occurrence.
comitant nephrotoxic agents between groups (Table 6). Additionally, VPT patients that had a maximum trough
The incidence of AKI for patients in the VPT group was value ≥30 mg/L also had the same increased risk (100%,
29.8% (28/94) versus 8.8% (13/148) in the VC/VM group, P = 0.024); however, this only encompassed 3 patients.
P < 0.001. When analyzing the combined treatment arm
VC/VM separately compared with VPT the same difference
Discussion
was seen for VPT versus VC (29.8% vs 5.9%; P < 0.001).
There was no statistical difference in the incidence of AKI This was the first prospective, multicenter study exploring
between VPT and VM (29.8% vs 14.9%, P = 0.054). differences in the development of AKI between patients
The maximum mean recorded SCr was higher in the receiving VPT and VC/VM. Like other studies, VPT was
VPT group compared with the VC/VM group (1.4 mg/dL vs associated with an independent, increased risk of AKI,
1.0 mg/dL; P < 0.001). Of the 41 patients that developed which was further compounded by the concomitant use of
AKI during treatment, 23 (56%) never returned to their loop diuretics, vasopressors, and, potentially, vancomycin
baseline SCr. Of the 18 patients (44%) that did return to concentrations greater than 30 µg/mL. These risk factors
baseline, the average time to baseline was 164 hours. Three seem to make logical sense and can be interpreted in a vari-
patients in the VPT group required RRT while no patients in ety of ways given the difficulty in assessing the onset of
the VC/VM group required RRT (P = 0.057). There was no injury and the delayed manifestation of kidney failure. It is
difference in the length of stay between the 2 groups (12.2 unclear in our study whether the loop diuretics and vaso-
days vs 13 days, P = 0.791). pressors were the cause of kidney injury, a signal of a more
A binary logistic regression of the entire cohort was con- complex patient population, or more likely, a combination
ducted to determine independent predictors of AKI. Patients of the two. While focusing on the associated toxicity of van-
receiving VPT were at a higher risk of developing AKI than comycin concentrations, it is not clear whether elevated
Mullins et al
vancomycin concentrations were causative of AKI or rather
kidney injury that led to accumulation.
The mechanism of observed nephrotoxicity with the
combination of VPT is still unclear, as is whether the com-
bination is truly nephrotoxic. While vancomycin has
clearly reported risk factors associated with the develop-
ment of AKI, PT is not as simple. In point of fact, one pub-
lished report evaluated rates of AKI with amikacin with or
without piperacillin, and suggested a lower rate of AKI
with the combination than with amikacin alone.11 It is
unclear and needs further evaluation as to whether the
combination of VPT itself is nephrotoxic or if piperacillin
and/or tazobactam enhances the already nephrotoxic poten-
tial of vancomycin. The previously reported data on this
topic unfortunately did not routinely evaluate the require-
ment of renal replacement therapy in this context.
Interestingly, 2 studies did report this and found lower rates
of renal replacement therapy in the VPT group than in the
control group.6,8 In the present study, all patients that
required renal replacement therapy (n = 3) were in the VPT
cohort. Questions have also been raised regarding the accu-
racy of SCr in diagnosing AKI in this setting. As a competi-
tive inhibitor of tubular secretion, piperacillin can increase
the accumulation of nephrotoxic drugs into the renal
tubules, as well as reduce the excretion of creatinine. This
may raise questions about using SCr as a marker of AKI in
this context; however, 56% of patients with AKI never
returned to baseline during the observation period, and
those that did return to baseline took, on average, 164 hours
to return to baseline, which may argue against tubular
secretion as a sole contributor.
Using a prospective, multicenter, observational design,
we attempted to enhance the strength and quality of this
study over previous retrospective and/or single-center
cohorts. A previously reported meta-analysis attempts to
remedy the limitations of the observational studies, but nat-
urally remains inadequate by inherent limitations of these
analyses within the larger model.9 There are also limitations
within our analysis. A possible type II error exists in the
separate comparison of vancomycin and meropenem. Due
to the study stopping early and the smaller amount of
patients in the VM arm, it is possible that VM indeed does
have a higher rate of nephrotoxicity, and we feel this should
be explored further before coming to a conclusion.
Controlling for the impact of AKI risk and patient complex-
ity is imperfect, although we did evaluate Charlson
Comorbidity Index and concomitant nephrotoxins, as in
previous publications. Additionally, our study was not able
to examine the mechanism behind the cause of nephrotoxic-
ity and is an area for future research to pursue. Last, this
study does not answer the question of whether novel bio-
markers, such as tissue inhibitor of metalloproteinase-2 and
insulin-like growth factor binding protein-7, can assist in
antibiotic management decisions through measuring renal
stress and damage before AKI develops. Future studies
5
should consider using these methods to help delineate renal
damage risk and occurrence rates.
The findings of our research, in addition to other previ-
ously published studies, should allow clinicians to clearly
recognize the risks of using VPT as initial empiric antimi-
crobial therapy. The decision of which anti-pseudomonal
agent to use is largely based on local susceptibility patterns,
site of infection, and individual patient factors. Clinicians
should incorporate these into the decision-making process
to adequately weigh risks and benefits for each patient. In
this study, we did not include any patients with a history of
CKD stage 3 or higher or with ongoing AKI prior to enroll-
ment, but it might be prudent to avoid VPT combination
altogether for this group of patients. More important, it is
the opinion of these authors that early, broad-spectrum anti-
biotic use may be helpful in the management of severe
infections, but commonly, indefinite combination therapy is
rarely warranted. Thus, we recommend that clinicians
employ appropriate culture-based and non–culture-based
organism identification techniques as soon as possible to
encourage antimicrobial deescalation and, thus, minimize
the risk of this particularly significant adverse event.
Conclusion
Our data further substantiates that of earlier data to suggest
an independent association between the combination of
VPT and AKI. We suggest that our data, along with other
published data, should be used to guide clinicians in the
management of antibiotic selection. Keeping in line with
core measures and best practice statements from national
organizations, empiric antibiotic selection should continue
to be based on local susceptibility patterns and suspected
sites of infection. Additionally, the focus should be primar-
ily on reducing overall antibiotic exposure, and secondarily,
a consideration should be made to select other combina-
tions of broad-spectrum antibiotics to avoid the toxic com-
bination of vancomycin and piperacillin/tazobactam. Last,
the authors recommend daily monitoring of SCr, appropri-
ate antibiotic deescalation, and avoidance of identified con-
comitant nephrotoxins if possible.
Appendix
Concomitant nephrotoxic agents surveilled for the following:
•• ACEIs/ARBs
•• Acyclovir
•• Aminoglycosides
•• Amphotericin B
•• Calcineurin inhibitors (cyclosporine, tacrolimus)
•• Colistin
•• IV contrast
•• Intravenous immune globulin
•• Loop diuretics
6 Annals of Pharmacotherapy 00(0)

•• NSAIDs References
•• Rifampin 1. American Thoracic Society. Guidelines for the management
•• Sirolimus of adults with hospital-acquired, ventilator-associated, and
•• Sulfonamide antibiotics healthcare-associated Pneumonia. Am J Respir Crit Care
•• Tenofovir Med. 2005;171:388-416.
•• Vasopressors, inotropes 2. Dellinger RP, Levy MM, Rhodes A, et al. Surviving sepsis cam-
paign: international guidelines for managements of severe sep-
Comorbid disease states surveilled for the following: sis and septic shock: 2012. Crit Care Med. 2013;41:580-637.
3. Van Boeckel T, Gandra S, Ashok A, et al. Global antibi-
otic consumption 2000 to 2010: an analysis of National
•• Diabetes
Pharmaceutical Sales Data. Lancet Infect Dis. 2014;14:742-
•• Chronic obstructive pulmonary disease 750.
•• Heart failure (all diagnoses) 4. Carreno JJ, Kenney RM, Lomaestro B. Vancomycin-associated
•• Coronary artery disease renal dysfunction: where are we now? Pharmacotherapy.
•• Peripheral vascular disease 2014;34:1259-1268. doi:10.1002/phar.1488.
5. Moenster RP, Linneman TW, Finnegan PM, Hand S, Thomas
Pharmacokinetic equations: Z, McDonald JR. Acute renal failure associated with vanco-
mycin and β-lactams for the treatment of osteomyelitis in dia-
•• ke = 0.00083 (CrCl) + 0.0044 betics: piperacillin-tazobactam as compared with cefepime.
•• Extrapolated trough = vancomycin concentration Clin Microbiol Infect. 2014;20:O384-O389.
* e(−ke*t) 6. Gomes DM, Smotherman C, Birch A, et al. Comparison
|| t = time difference from lab value to actual
of acute kidney injury during treatment with vancomycin
in combination with piperacillin-tazobactam or cefepime.
trough time Pharmacotherapy. 2014;34:662-669.
7. Burgess LD, Drew RH. Comparison of the incidence of
List of participating institutions vancomycin-induced nephrotoxicity in hospitalized patients
with and without concomitant piperacillin-tazobactam.
•• St. Luke’s Hospital, Chesterfield, MO Pharmacotherapy. 2014;34:670-676.
•• CoxHealth Medical Center, Springfield, MO 8. Hammond DA, Smith MN, Painter JT, Meena NK, Lusardi
•• Baylor Scott and White Health, Dallas, TX K. Comparative incidence of acute kidney injury in critically
•• Avera McKennan Hospital and University Health ill patients receiving vancomycin with concomitant piper-
Center, Sioux Falls, SD acillin-tazobactam or cefepime: a retrospective cohort study.
Pharmacotherapy. 2016;36:463-471.
9. Giuliano CA, Patel CR, Kale-Pradhan PB. Is the combina-
Declaration of Conflicting Interests
tion of piperacillin-tazobactam and vancomycin associated
The author(s) declared no potential conflicts of interest with respect with development of acute kidney injury? A meta-analysis.
to the research, authorship, and/or publication of this article. Pharmacotherapy. 2016;36:1217-1228.
10. Peyko V, Smalley S, Cohen, H. Prospective comparison of
Funding acute kidney injury during treatment with the combination of
The author(s) received no financial support for the research, piperacillin-tazobactam and vancomycin versus the combina-
tion of cefepime or meropenem and vancomycin. J Pharm
authorship, and/or publication of this article.
Pract. 2017;2:209-213.
11. Kusafuka H, Obayashi T, Iwayama N, et al. Protective effect
Supplemental Material by piperacillin against renal impairment caused by amikacin
Supplementary material is available for this article online. [in Japanese]. Nihon Jinzo Gakkai Shi. 1992;34:153-62.