Infectious Diseases

ISSN: 2374-4235 (Print) 2374-4243 (Online) Journal homepage: https://www.tandfonline.com/loi/infd20

Does nephrotoxicity develop less frequently when

vancomycin is combined with imipenem-cilastatin
than with meropenem? A comparative study

Hakeam A. Hakeam, Lina AlAnazi, Reem Mansour, Shahad AlFudail & Filwah
AlMarzouq

To cite this article: Hakeam A. Hakeam, Lina AlAnazi, Reem Mansour, Shahad AlFudail & Filwah
AlMarzouq (2019) Does nephrotoxicity develop less frequently when vancomycin is combined
with imipenem-cilastatin than with meropenem? A comparative study, Infectious Diseases, 51:8,
578-584, DOI: 10.1080/23744235.2019.1619934

To link to this article: https://doi.org/10.1080/23744235.2019.1619934

Published online: 24 May 2019.

Submit your article to this journal

Article views: 136

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=infd20
INFECTIOUS DISEASES, https://doi.org/10.1080/23744235.2019.1619934
2019; VOL. 51,
NO. 8, 578–584

ORIGINAL ARTICLE

Does nephrotoxicity develop less frequently when

vancomycin is combined with imipenem-cilastatin
than with meropenem? A comparative study

Hakeam A. Hakeama,b, Lina AlAnazic, Reem Mansourc, Shahad AlFudailc and Filwah AlMarzouqc
a
Pharmaceutical Care Division, King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia; bCollege of Medicine,
Alfaisal University, Riyadh, Saudi Arabia; cCollege of Pharmacy, Princess Nora Bint Abdulrhman University, Riyadh, Saudi Arabia

ABSTRACT
Introduction: Nephrotoxicity is a frequent complication of vancomycin therapy. Experimental studies in different animal
species have demonstrated the attenuation of vancomycin-associated nephrotoxicity with cilastatin administration. This
study aimed to evaluate if imipenem-cilastatin attenuates vancomycin-associated nephrotoxicity, in patients treated with
combinations of vancomycin and carbapenems.

Methods: This retrospective, propensity-score matched study was conducted at King Faisal Specialist Hospital and
Research Centre, Riyadh and Jeddah. Nephrotoxicity was compared in patients who received imipenem-cilasta-
tin þ vancomycin or meropenem þ vancomycin. Patients with no history of renal disease who received imipenem-cilasta-
tin þ vancomycin or meropenem þ vancomycin for a minimum of 72 h, from 1 January 2017 to 31 December 2017, were
included. Nephrotoxicity was defined according to the RIFLE criteria (Risk, Injury, Failure, Loss, End-stage renal disease) if
sustained for least 72 h.

Results: A total of 227 patients were included in the analysis, consisting of 121 patients in the imipenem-cilasta-
tin þ vancomycin group, and 106 patients in the meropenem þ vancomycin group. In the unmatched data set the rate of
nephrotoxicity was 8.2% in imipenem-cilastatin þ vancomycin group and 20.7% in the meropenem þ vancomycin group
(p ¼ .007). Logistic regression analysis showed that imipenem-cilastatin þ vancomycin therapy was associated with a 56%
lower rate of nephrotoxicity compared to meropenem þ vancomycin therapy. Propensity-score matching resulted in rates
of nephrotoxicity of 6.2% and 17.1% in the imipenem-cilastatin þ vancomycin group and the meropenem þ vancomycin
groups, respectively (p ¼ .034).

Conclusion: Vancomycin-associated nephrotoxicity developed less frequently when vancomycin was combined with imi-
penem-cilastatin than when combined with meropenem.

KEYWORDS ARTICLE HISTORY CONTACT

Vancomycin Received 4 March 2019 Hakeam A. Hakeam
nephrotoxicity Revised 9 May 2019 hakeam@kfshrc.edu.sa
imipenem Accepted 10 May 2019 Pharmaceutical Care Division, King Faisal
cilastatin Specialist Hospital & Research Centre, Adjunct
Assistant Clinical Professor, College of Medicine,
Alfaisal University, P.O Box 3354 Riyadh 11211
MBC# 11, Riyadh, Saudi Arabia

ß 2019 Society for Scandinavian Journal of Infectious Diseases


Introduction
Nephrotoxicity has been associated with systemic use of
vancomycin (VAN) since its introduction into clinical
practice. With modern VAN preparations, the reported
incidence of nephrotoxicity ranges from 5% to greater
than 40% [1]. This wide range could be due to various
risk factors predisposing to this renal complication [2].
Although VAN-associated nephrotoxicity is deemed
reversible, current reports have shown prolonged hospi-
talization and increased mortality following acute kidney
injury (AKI) of any cause [3]. During the last two-deca-
des, VAN use has increased more than 100-fold due to
its efficacy in the treatment of methicillin-resistant
Staphylococcus aureus (MRSA) infections [4]. The expan-
sion of VAN use has increased the interest in preventive
strategies to attenuate the associated nephrotoxicity.
Carbapenems are b-lactam antibiotics, which include
imipenem-cilastatin (IMIC) and meropenem (MERO).
These antibiotics are commonly used to treat severe
bacterial infections and are overall safe to use [5]. VAN
is frequently added to IMIC or MERO in empirical treat-
ment of severe infections when MRSA must be cov-
ered [6].
Imipenem is rapidly catalyzed to toxic metabolites by
the dehydropeptidase-I (DHP-I) enzyme at the luminal
face of the proximal tubular cells of the kidneys [7],
which results in nephrotoxicity. Cilastatin is a specific
inhibitor of DHP-I, thus functioning as a renoprotective
agent. Therefore, imipenem is formulated at a 1 mg to
1 mg ratio with cilastatin to prevent rapid hydrolysis to
toxic metabolites with accumulation in the tubular cells
[8]. Experimental studies found that cilastatin exerts a
renoprotective activity in different animal species admin-
istered VAN [9–12]. In these studies, cilastatin was used
either solely or as the clinically used formula in combin-
ation with imipenem.
To our knowledge, no clinical study has investigated
the rate of nephrotoxicity when VAN is combined with
IMIC to treat various infections. This study aimed to eval-
ute if IMIC attenuates VAN-associated nephrotoxicity in
patients treated with combinations of VAN and
carbapenems.
Materials and methods
Study design and setting
This was a retrospective, propensity-score (PS) matched
study, conducted at King Faisal Specialist Hospital and
Research Centre (KFSH & RC), the main tertiary care
INFECTIOUS DISEASES 579
organization in Saudi Arabia, consisting of two hospitals
in Riyadh and Jeddah. The institutional review board at
KFSH & RC approved the study prior to data collection.
Study population
The study population consisted of patients aged
18 years admitted to KFSH & RC at Riyadh or Jeddah
between 1 January 2017 and 31 December 2017.
Patients were included in the study if they had received
VAN in combination with either IMIC (IMIC þ VAN) or
meropenem (MERO þ VAN) for a minimum of 72 h. For
patients who received the antibiotic combinations mul-
tiple times during hospitalization, only the initial regi-
men was included. At KFSH & RC, VAN therapy is
optimized by clinical pharmacists to maintain steady-
state VAN trough levels at a target of either 10–15 mg/L
or 15–20 mg/L according to infection severity. Exclusion
criteria were: chronic kidney disease (CKD) of any degree
(glomerular filtration rate (GFR)<60 mL/min/per 1.73 m2),
AKI within 30 days before VAN therapy, do not resusci-
tate (DNR) status prior to VAN therapy, and receipt of
any of the following medications by intravenously con-
comitantly with VAN for > 24 h: aminoglycosides, colis-
tin, acyclovir, and amphotericin B.
Data collection
Clinical characteristics and laboratory data were
extracted from electronic medical records. Data collected
included patient demographics, baseline serum creatin-
ine levels, admission to intensive care unit (ICU), indica-
tion for antibiotic therapy, receipt of potential
nephrotoxins (cyclosporin, non-steroidal anti-inflamma-
tory drugs (NSAIDs), furosemide, angiotensin-converting
enzyme inhibitors (ACE-I), or angiotensin II receptor
blockers (ARBs)) concomitantly with VAN, receipt of a
platinum chemotherapy or intravenous contrast during
or within 3-days prior to the VAN therapy, VAN therapy
characteristics (receipt of a loading dose, total doses,
daily dose >4 g, and duration of therapy), and VAN
trough levels (first and maximum). The VAN loading
dose was defined as an initial dose higher than subse-
quent maintenance doses. VAN-carbapenems combin-
ation duration and carbapenem doses were recorded as
well. The lowest mean arterial pressure (MAP) was calcu-
lated and recorded for all patients. Serum creatinine lev-
els were screened starting from the first day of VAN
therapy until the fourth day after cessation. GFR was cal-
culated based on serum creatinine levels after 4 days of
580 H. A. HAKEAM ET AL.

Table 1. Clinical characteristics and laboratory values of included patients who received imipenem-
cilastatin þ vancomycin (IMIC þ VAN) or meropenem þ VAN (MERO þ VAN).
IMIC þ VAN, n ¼ 121 MERO þ VAN, n ¼ 106 p
Male sex, n (%) 62 (51.24) 63 (59.43) .21
Age (years) 50.7 ± 17.4 45.7 ± 17.3 .029
Weight (kg) 64.4 ± 18.4 67.0 ± 20.8 .33
BMI (kg/m2) 24.6 ± 7.2 25.4 ± 7.4 .45
BMI > 30 kg/m2, n (%) 57 (47.1) 50 (47.1) .99
Diabetes mellitus, n (%) 46 (38.0) 45 (42.4) .49
Type of infection, n (%) .033
Community acquired pneumonia 4 (3.3) 4(3.7) 1
Skin and soft tissue infection 8 (6.6) 9 (8.4) .59
Febrile neutropenia 19 (15.7) 27 (25.4) .067
HAP/VAP 24 (19.8) 31(29.2) .098
Intra-abdominal infection 25 (20.6) 12 (11.3) .057
Central catheter infection 2 (1.6) 2 (1.8) 1
Septic shock 25 (20.6) 16 (15.0) .27
Urinary tract infection 5 (4.1) 1 (0.9) .21
Others 9 (7.4) 4 (3.7) .12
Laboratory values
Basal serum creatinine,(mmol/L) 54.8 ± 16 55.2 ± 18.9 .89
Baseline glomerular filtration rate (mL/min/1.73 m2) 139.0 ± 55 144.1 ± 61.1 .69
BMI: body mass index; HAP/VAP: hospital-acquired pneumonia/ventilator-associated pneumonia.
Calculated based on the Cockcroft and Gault equation.

stopping VAN. For patients admitted to the ICU, the
Sequential Organ Failure Assessment (SOFA) score was
calculated and recorded, as were vasopressors use and
daily urine output. The following characteristics of
nephrotoxicity were obtained: onset, duration, highest
serum creatinine level, lowest GFR calculated based on
the Cockcroft and Gault equation, and the lowest urine
output. Nephrotoxicity outcomes were recorded, includ-
ing time to resolution, need for dialysis, and death.
Study end points
The primary endpoint was the difference in nephrotox-
icity rates in patients treated with IMIC þ VAN versus
MERO þ VAN. Nephrotoxicity was defined according to
the RIFLE (Risk, Injury, Failure, Loss, End-stage renal dis-
ease) criteria and the vancomycin therapeutic monitor-
ing consensus definition [13,14]. Presence of any RIFLE
stage was considered as nephrotoxicity if it lasted for
72 h. Secondary endpoints included the differences
between the two groups in nephrotoxicity characteris-
tics, including onset, duration, and recovery.
Associations of different risk factors for nephrotoxicity in
VAN-carbapenem combinations were studied as second-
ary endpoints.
combined with two different b-lactam antibiotics. The
reported absolute difference in the risk of VAN-associ-
ated nephrotoxicity was 18% [15]. Therefore, a minimum
of 88 patients per group was sufficient to achieve 80%
power using a two-sided test at a ¼ 0.05.
Demographics, clinical characteristics, and secondary
outcomes were compared between the two groups
using Student’s t-tests or Wilcoxon rank-sums for con-
tinuous variables and Chi-square or Fisher’s exact tests
for categorical variables. Logistic regression analysis was
used to determine the associations with VAN-associated
nephrotoxicity, and to adjust for potential confounders.
Variables with p < .2 following univariate analyses were
considered for multivariate logistic regression analysis.
To minimize the impact of risk factors predisposing
patients to VAN-associated nephrotoxicity, patients in
the study groups were entered in a propensity score
(PS) matching, at a caliper between PS 0.001 and PS
þ0.001. The PS was based on the following risk factors:
VAN duration, VAN trough level higher than 15 mg/L,
and body mass index (BMI) [2,13,16,17]. Statistical signifi-
cance was defined as p < .05. SPSS software (SPSS for
Windows, Version 22, IBM SPSS Inc., Armonk, NY, USA)
was used for all statistical analysis.

Statistical analyses Results

We included all patients who met the inclusion criteria
for the study during the year of 2017. To ensure statis-
tical power, the minimum number of subjects to be
enrolled was determined according to a study that com-
pared nephrotoxicity in patients who received VAN
A total of 1574 patients were reviewed: 485 patients
were excluded due to renal dysfunction; 470 patients
received <72 h of VAN-carbapenem combination ther-
apy, 380 patients received intravenous aminoglycoside,
colistin, acyclovir, or amphotericin B, and 12 patients
 INFECTIOUS DISEASES 581

Table 2. Characteristics of vancomycin (VAN) therapy in patients who received imipenem-cilastatin þ VAN (IMIC þ VAN) or
meropenem þ VAN (MERO þ VAN).
IMIC þ VAN, n ¼ 121 MERO þ VAN, n ¼ 106 p
Durationa, days 7.7 ± 6.2 8.9 ± 6.9 .16
Number of doses 16.8 ± 14.1 20.0 ± 18.1 .13
Total dose (g) 17.1 ± 16.7 20.5 ± 18.6 .15
Dose > 4 g, n (%) 3 (2.4) 3 (2.8) .45
Therapeutic drug monitoring
First trough level (mg/L) 9.5 ± 4.7 9.3 ± 5.7 .74
Maximum trough level (mg/L) 15.9 ± 8.6 16.1 ± 7.8 .88
Trough level > 15 mg/L, n(%) 55 (45.4) 50 (47.1) .80
Serum creatinine
Day 1 after VAN stop (mmol/L) 52.1 ± 24.3 55.1 ± 29.2 .40
Day 4 after VAN stop (mmol/L) 56.8 ± 24.5 66.1 ± 46.0 .11
Glomerular filtration rate 4 days after stopping VAN (mL/min/1.73 m2) 137 ± 60.7 127 ± 70.7 .69
Vasopressors, n (%) 26 (21.4) 21 (19.8) .75
Nephrotoxins
Intravenous contrast, n (%) 37 (30.5) 24 (22.6) .17
Furosemide, n (%) 40 (33.0) 44 (41.5) .18
ACE-I/ARB, n (%) 6 (5) 11 (10.3) .12
NSAIDs, n (%) 15 (12.4) 2 (1.8) .0027
Cyclosporin, n (%) 2 (1.6) 8 (7.5) .048
ACE-I: angiotensin-converting enzyme inhibitor; ARB: angiotensin II receptor blockers; NSAIDs: non-steroidal anti-inflammatory drugs;
VAN: vancomycin.
Calculated based on Cockcroft and Gault equation.
a
Total days of vancomycin therapy.

were excluded due to DNR status. A total of 227
patients were included in the analysis, consisting of 121
patients in the IMIC þ VAN group and 106 patients in
the MERO þ VAN group.
Baseline patient demographics and laboratory values
were comparable between the two patient groups,
except for age as patients in the IMIC þ VAN group were
likely to be older (Table 1). The overall antibiotic indica-
tions were variable between the two groups; however,
the IMIC þ VAN and MERO þ VAN groups contained simi-
lar percentages of each indication for anti-
biotic treatment.
The two groups were similar in terms of dosing and
therapeutic drug monitoring of VAN (Table 2). VAN was
combined with IMIC for 6.9 ± 4.9 days, and with MERO
for 8.2 ± 5.8 days (p ¼ .059). IMIC was dosed 0.5 g every
6 h in 109 patients (90%), and 1 g every 8 h in 12
patients (10%). None of the patients who received the
higher dose of IMIC developed nephrotoxicity. MERO
was administered every 8 h at a dose of either 1 g in 98
(92.4%) patients, and 0.5 g in 8 (7.5%) patients. Of the
patients who developed nephrotoxicity in the
MERO þ VAN group, only one (4.5%) patient received
0.5 g of MERO.
Use of vasopressors or intravenous contrast media
was similar in the two patient groups. Potential nephro-
toxic medications were also comparable between the
two groups. However, there was a higher number of
recipients of NSAIDs in the IMIC þ VAC group, while
cyclosporin was used more frequently in patients in the
MERO þ VAN group (Table 2). Among patients who
developed nephrotoxicity, one patient in the IMIC þ VAN
group received an NSAID, and two patients in the
MERO þ VAN group received cyclosporin.
A comparable number of patients in both groups
were admitted to the ICU during VAN therapy (38% and
33.9% in the IMIC þ VAN and MERO þ VAN groups,
respectively; p ¼ .52). The mean SOFA score was 3.1 ± 1.9
in the IMIC þ VAN group and 3.3 ± 2.2 in the
MERO þ VAN group (p ¼ .22).
The mean of lowest MAP reading was 67.8 ± 12.1 mmHg
in the IMIC þ VAN group and 66.9 ± 11.7 mmHg in the
MERO þ VAN group (p ¼ .98). The percentages of patients
with MAP readings below 65 mmHg were similar in the
IMIC þ VAN and MERO þ VAN groups (53.5% and 46.4%,
respectively, p ¼ .87).
In the unmatched data, the rate of nephrotoxicity
was lower in the IMIC þ VAN group (8.2%) compared to
the MERO þ VAN group (20.7%) (p ¼ .007) (Table 3).
There was no difference in the need for renal replace-
ment therapy between the two groups. Among the
patients who developed nephrotoxicity, two patients
(20%) in the IMIC þ VAN group died within 18 days
(interquartile range (IQR), 8 to 29 days), and 8 (36%) in
the MERO þ VAN group died within 7 days (IQR, 3 to 19
days) (p ¼ .136). The overall RIFLE scores were not statis-
tically different between the IMIC þ VAN and the
MERO þ VAN groups (p ¼ .26). One patient in the
IMIC þ VAN group progressed to the loss stratification of
the RIFLE classification of nephrotoxicity. All other
patients with nephrotoxicity returned to their baseline
serum creatinine level before discharge from hospital.
582 H. A. HAKEAM ET AL.

Table 3. Characteristics of patients who received imipenem-cilastatin þ vancomycin (IMIC þ VAN) or

meropenem þ VAN (MERO þ VAN) and developed nephrotoxicity.
IMIC þ VAN, n ¼ 10 MERO þ VAN, n ¼ 22 p
Nephrotoxicity, n (%) 10 (8.2) 22 (20.7) .007
Onset of nephrotoxicity (days) 8.6 ± 6.9 11.4 ± 9.4 .35
Need for renal replacement therapy, n (%) 1 (10) 2 (9) 1
Maximum serum creatinine (mmol/L) 149 ± 122.4 125.2 ± 43 .42
Duration of nephrotoxicity (days) 11.2 ± 9 11.0 ± 8.7 .99
In-hospital mortality, n (%) 2 (20) 8(36.3) .428
Lowest urine output (mL)a 1253.1±788 1039 ± 812 .23
Lowest urine output (mL/kg/h)a 0.53 ± 0.6 0.66 ± 0.7 .44
RIFLE criteria .26
R 3 6 .31
I 1 8 .21
F 5 8 .69
L 1 0 .31
RIFLE: Risk, Injury, Failure, Loss, End-stage renal disease.
a
Urine output during the nephrotoxicity event.

Logistic regression analysis was performed on the
study subjects’ characteristics to determine predictors of
nephrotoxicity. Patients who received IMIC þ VAN were
at lower risk of nephrotoxicity than those who received
MERO þ VAN (odds ratio (OR) 0.34, 95% confidence inter-
val (CI) 0.15–0.76; p ¼ .0089) with a relative risk of 0.54
(95% CI 0.34–0.9; p ¼ .007). Nephrotoxicity was associ-
ated with the receipt of vasopressors (OR 2.74, 95% CI
1.22–6.12; p ¼ .0114), and furosemide (OR 2.16, 95% CI
1.01–4.60; p ¼ .0448), independent of treatment group.
Although not statistically significant, VAN trough levels
higher than 15 mg/L (p ¼ .069) were entered into multi-
variate logistic regression analysis together with vaso-
pressor and furosemide use. However, none of these
factors were associated with nephrotoxicity (VAN trough
level higher than 15 mg/L p ¼ .17, vasopressor use
p ¼ .18, and furosemide use p ¼ .15).
Overall, 74 pairs of patients were matched by PS for
duration of VAN therapy, VAN trough higher than
15 mg/L and BMI. After matching, the rates of nephro-
toxicity were 6.2%, and 17.1% in the IMIC þ VAN group
and the MERO þ VAN group, respectively (p ¼ .034).
Discussion
In this retrospective study of nephrotoxicity in patients
who received VAN combined with either IMIC or MERO,
we found that nephrotoxicity developed less frequently
in the IMIC þ VAN group than in the MERO þ VAN group.
To our knowledge, this is the first study to examine
VAN-associated nephrotoxicity in patients receiving
these two combinations.
There is little in the literature regarding the rate
of VAN-associated nephrotoxicity when combined
with IMIC. This scarcity may be a result of more
attention directed towards b-lactams, such as
piperacillin-tazobactam, that potentiate nephrotoxicity
when combined with VAN [15,18]. Hundeshagen et al.
examined pediatric and adult burn patients who
received VAN solely or in combination with either IMIC
or piperacillin-tazobactam [19]. After 7 days of initial
IMIC þ VAN therapy, the rate of nephrotoxicity in adults
was 0%. Our study showed a higher rate of nephrotox-
icity in the IMIC þ VAN group (8.2%). This could be
explained by lower target of VAN trough levels of
10–15 mg/L, and a study population limited to burn
patients with altered VAN pharmacokinetics, resulting in
low serum levels in the study by Hundeshagen. In our
study, we assessed a heterogeneous patient population
with various indications for antibiotic treatment, includ-
ing severe infections requiring higher target VAN trough
levels of 15–20 mg/L. However, both studies indicated
lower rates of nephrotoxicity when VAN was combined
with IMIC than with comparators.
The high proportion of patients with diabetes mellitus
in our study might contribute to the high rates of VAN-
associated nephrotoxicity observed in the IMIC þ VAN
and the MERO þ VAN groups [20]. Studies assessing
VAN-associated nephrotoxicity with VAN þ MERO have
reported a rate of 5.3% to 15.4% [18,21], lower than in
our study (22.7%).
In both the IMIC þ VAN and the MERO þ VAN groups,
VAN-associated nephrotoxicity developed within 4–17
days. This finding is consistent with the results of a sys-
tematic review of renal outcomes in patients receiving
VAN [22].
Cilastatin has shown efficacy in attenuating VAN-asso-
ciated nephrotoxicity in different animal species.
Cilastatin administrated concomitantly with VAN in equal
doses of 100 mg/kg to rats, resulted in a significant
increase in total VAN clearance [9]. Another experiment
utilized a rabbit model in which rabbits were given

intravenous VAN (300 mg/kg) followed by saline or vari-
able doses of cilastatin (75, 150, or 300 mg/kg). Lower
serum creatinine levels and renal cortex concentrations
of VAN were observed in rabbits given VAN combined
with cilastatin compared to VAN alone [11]. The exact
mechanism of renoprotection by cilastatin is not fully
understood. However, many hypotheses have been
postulated. Cilastatin may reduce VAN uptake and accu-
mulation in renal tubular cells [10,11]. Alternatively, cilas-
tatin competes with VAN for binding to megalin, a
transport glycoprotein expressed on the apical mem-
brane of the proximal tubular epithelial cells [23].
Although it is difficult to precisely assess VAN-
associated nephrotoxicity independently from confound-
ers, we excluded patients with any degree of renal
impairment prior to the initiation of VAN therapy.
Furthermore, we excluded patients who received antimi-
crobials known to induce high rates of nephrotoxicity
(>50%) such as amphotericin B, colistin, and aminogly-
cosides [24–26].
It is unlikely that the more frequent use of cyclo-
sporin in the MERO þ VAN group explains the higher
rate of nephrotoxicity observed in this group compared
to the IMIC þ VAN group. Less than 25% of patients who
developed nephrotoxicity in the MERO þ VAN group
were exposed to cyclosporin. On the other hand,
patients in the IMIC þ VAN group were older and were
treated more frequently with NSAIDs, known risk factors
for nephrotoxicity.
Univariate analysis indicated that furosemide and
vasopressor use were associated with nephrotoxicity
when VAN was combined with carbapenems in our
study. These results are consistent with the knowledge
that both factors predispose to VAN-associated nephro-
toxicity [2]. The lack of significant association in the
multivariate analysis could be attributed to insufficient
statistical power to distinguish effects adequately.
Combinations of VAN with MERO or IMIC are used as
empirical therapy in many healthcare-associated infec-
tions. The Infectious Diseases Society of America (IDSA)
and the American Thoracic Society (ATS), recommended
these combinations to treat hospital-acquired (HAP) and
ventilator-associated pneumonia (VAP) [6]. Patients with
VAP are common in ICUs and are at high-risk for VAN-
associated nephrotoxicity. Therefore, the findings of this
study may decrease the risk of VAN-induced nephrotox-
icity in these vulnerable, high-risk patients by indicating
the benefit of using IMIC instead of MERO in combin-
ation with VAN.
INFECTIOUS DISEASES 583
When combined with VAN, cilastatin does not inhibit
the antibacterial activity of VAN but exerts a renoprotec-
tive activity [10,23]. Future investigations could assess
the efficacy and safety of combining larger doses of
both VAN and cilastatin in the treatment of severe
Gram-positive infections, with or without imipenem [27].
One limitation of this study is the retrospective
design. However, the PS matching offers a balance of
measured covariates between the IMIC þ VAN and the
MERO þ VAN groups. Many nephrotoxins and patients
with any renal impairment were excluded, thus the
study was not representative for the complete spectrum
of hospitalized patients. Because this is the first clinical
study to assess cilastatin attenuation of VAN-associated
nephrotoxicity, tight control of certain confounders may
help to more explicit endorse this outcome. It might
have been better if the sample size had been calculated
according to the nephrotoxicity rate associated with
IMIC þ VAN therapy. However, the only prior study on
nephrotoxicity with the IMIC þ VAN combination had 0%
nephrotoxicity, which would underestimate the neces-
sary sample size. Future studies with larger patient num-
ber might support our findings.
In conclusion, VAN-associated nephrotoxicity devel-
oped less frequently when VAN was combined with
IMIC compared to MERO. Cilastatin likely played a role in
the attenuation of VAN-associated nephrotoxicity. IMIC
may be a better choice when empirical therapy with
VAN and carbapenem is indicated and nephrotoxicity is
a concern. Further prospective investigations are needed
to confirm the results of this study.
Disclosure statement
No potential conflict of interest was reported by the authors.
References
[1] Mergenhagen KA, Borton AR. Vancomycin nephrotoxicity: a
review. J Pharm Pract. 2014;27:545–553.
[2] Filippone EJ, Kraft WK, Farber JL. The nephrotoxicity of
vancomycin. Clin Pharmacol Ther. 2017;102:459–469.
[3] Prendecki M, Blacker E, Sadeghi-Alavijeh O, et al. Improving
outcomes in patients with acute kidney injury: the impact
of hospital based automated AKI alerts. Postgrad Med J.
2016;92:9–13.
[4] Nolin TD. Vancomycin and the risk of AKI: now clearer than
Mississippi mud. Clin J Am Soc Nephrol. 2016;11:
2101–2103.
[5] Slama TG. Clinical review: balancing the therapeutic, safety,
and economic issues underlying effective antipseudomonal
carbapenem use. Crit Care. 2008;12:233.
584 H. A. HAKEAM ET AL.
[6] Kalil AC, Metersky ML, Klompas M, et al. Management of
adults with hospital-acquired and ventilator-associated
pneumonia: 2016 clinical practice guidelines by the
Infectious Diseases Society of America and the American
Thoracic Society. Clin Infect Dis. 2016;63:e61–e111.
[7] Kahan FM, Kropp H, Sundelof JG, et al. Thienamycin: devel-
opment of imipenen-cilastatin. J Antimicrob Chemother.
1983;12:1–35.
[8] Norrby SR. Imipenem/cilastatin: rationale for a fixed com-
bination. Rev Infect Dis. 1985;7(3):S447–S451.
[9] Kusama M, Yamamoto K, Yamada H, et al. Effect of cilasta-
tin on renal handling of vancomycin in rats. J Pharm Sci.
1998;87:1173–1176.
[10] Nakamura T, Hashimoto Y, Kokuryo T, et al. Effects of fosfo-
mycin and imipenem/cilastatin on nephrotoxicity and renal
excretion of vancomycin in rats. Pharm Res. 1998;15:
734–738.
[11] Toyoguchi T, Takahashi S, Hosoya J, et al. Nephrotoxicity of
vancomycin and drug interaction study with cilastatin in
rabbits. Antimicrob Agents Chemother. 1997;41:1985–1990.
[12] Nakamura T, Kokuryo T, Hashimoto Y, et al. Effects of fosfo-
mycin and imipenem-cilastatin on the nephrotoxicity of
vancomycin and cisplatin in rats. J Pharm Pharmacol. 1999;
51:227–232.
[13] Rybak M, Lomaestro B, Rotschafer JC, et al. Therapeutic
monitoring of vancomycin in adult patients: a consensus
review of the American Society of Health-System
Pharmacists, the Infectious Diseases Society of America,
and the Society of Infectious Diseases Pharmacists. Am J
Health Syst Pharm. 2009;66:82–98.
[14] Bellomo R, Ronco C, Kellum JA, et al. Acute renal failure -
Definition, outcome measures, animal models, fluid therapy
and information technology needs: the Second
International Consensus Conference of the Acute Dialysis
Quality Initiative (ADQI) Group. Crit Care. 2004;8:
R204–R212.
[15] Navalkele B, Pogue JM, Karino S, et al. Risk of acute kidney
injury in patients on concomitant vancomycin and pipera-
cillin-tazobactam compared to those on vancomycin and
cefepime. Clin Infect Dis. 2017;64:116–123.
[16] van Hal SJ, Lodise TP, Paterson DL. The clinical significance
of vancomycin minimum inhibitory concentration in
Staphylococcus aureus infections: a systematic review and
meta-analysis. Clin Infect Dis. 2012;54:755–771.
[17] Lodise TP, Lomaestro B, Graves J, et al. Larger vancomycin
doses (at least four grams per day) are associated with an
increased incidence of nephrotoxicity. Antimicrob Agents
Chemother. 2008;52:1330–1336.
[18] Rutter WC, Burgess DS. Incidence of acute kidney injury
among patients treated with piperacillin-tazobactam or
meropenem in combination with vancomycin. Antimicrob
Agents Chemother. 2018;62:pii: e00264-18.
[19] Hundeshagen G, Herndon DN, Capek KD, et al. Co-adminis-
tration of vancomycin and piperacillin-tazobactam is associ-
ated with increased renal dysfunction in adult and
pediatric burn patients. Crit Care. 2017;21:318.
[20] Girman CJ, Kou TD, Brodovicz K, et al. Risk of acute renal
failure in patients with Type 2 diabetes mellitus. Diabet
Med. 2012;29:614–621.
[21] Al Yami MS. Comparison of the incidence of acute kidney
injury during treatment with vancomycin in combination
with piperacillin-tazobactam or with meropenem. J Infect
Public Health. 2017;10:770–773.
[22] van Hal SJ, Paterson DL, Lodise TP. Systematic review and
meta-analysis of vancomycin-induced nephrotoxicity associ-
ated with dosing schedules that maintain troughs between
15 and 20 milligrams per liter. Antimicrob Agents
Chemother. 2013;57:734–744.
[23] Hori Y, Aoki N, Kuwahara S, et al. Megalin blockade with
cilastatin suppresses drug-induced nephrotoxicity. J Am
Soc Nephrol. 2017;28:1783–1791.
[24] Kwon JA, Lee JE, Huh W, et al. Predictors of acute kidney
injury associated with intravenous colistin treatment. Int J
Antimicrob Agents. 2010;35:473–477.
[25] Luber AD, Maa L, Lam M, et al. Risk factors for amphoteri-
cin B-induced nephrotoxicity. J Antimicrob Chemother.
1999;43:267–271.
[26] Kumin GD. Clinical nephrotoxicity of tobramycin and gen-
tamicin. A prospective study. JAMA. 1980;244:1808–1810.
[27] Soriano A, Marco F, Martinez JA, et al. Influence of vanco-
mycin minimum inhibitory concentration on the treatment
of methicillin-resistant Staphylococcus aureus bacteremia.
Clin Infect Dis. 2008;46:193–200.