Accepted Manuscript Population pharmacokinetic-pharmacodynamic target attainment analysis of sulbactam in patients with impaired renal function: dosing considerations for Acinetobacter baumannii infections Yuta Yokoyama, Kazuaki Matsumoto, Kazuro Ikawa, Erika Watanabe, Norifumi Morikawa, Yasuo Takeda PII:
S1341-321X(14)00420-6
DOI:
10.1016/j.jiac.2014.12.005
Reference:
JIC 230
To appear in:
Journal of Infection and Chemotherapy
Received Date: 18 September 2014 Revised Date:
16 December 2014
Accepted Date: 18 December 2014
Please cite this article as: Yokoyama Y, Matsumoto K, Ikawa K, Watanabe E, Morikawa N, Takeda Y, Population pharmacokinetic-pharmacodynamic target attainment analysis of sulbactam in patients with impaired renal function: dosing considerations for Acinetobacter baumannii infections, Journal of Infection and Chemotherapy (2015), doi: 10.1016/j.jiac.2014.12.005. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
1
Journal of Infection and Chemotherapy
2
Original article
RI PT
3
Population pharmacokinetic-pharmacodynamic target attainment analysis of
5
sulbactam in patients with impaired renal function: dosing considerations for
6
Acinetobacter baumannii infections
SC
4
M AN U
7 8
Yuta Yokoyamaa,b , Kazuaki Matsumotoa , Kazuro Ikawab , Erika Watanabea ,
9
Norifumi Morikawab , Yasuo Takedaa, *
10 11
a
12
Medical and Dental Sciences, Kagoshima University
13
b
TE D
EP
Department of Clinical Pharmacotherapy, Hiroshima University
AC C
14
Department of Clinical Pharmacy and Pharmacology, Graduate School of
15
*Corresponding author
16
Address: 8-35-1 Sakuragaoka, Kagoshima 890-8520, Japan
17
Tel.: +81-99-275-5543; Fax: +81-99-265-5293
18
E-mail: [email protected]
1
ACCEPTED MANUSCRIPT 19 20
Abstract This study aimed to perform a pharmacokinetic (PK)-pharmacodynamic (PD) target attainment analysis of sulbactam against
22
Acinetobacter baumannii in patients with impaired renal function. The PK
23
data (188 plasma samples and 27 urine samples) were modeled simultaneously.
24
The mean population parameters were CL r (l/h) = 0.0792 × CLcr (ml/min),
25
CLnr (l/h) = 2.35, Vc (l) = 12.2, Q (l/h) = 4.68 and Vp (l) = 4.44, where CL r and
26
CLnr are the renal and non-renal clearances, Vc and Vp are the distribution
27
volumes of the central and peripheral compartments and Q is
28
intercompartmental clearance. The creatinine clearance (CL cr) was the most
29
significant covariate. The determined MIC of sulbactam against A. baumannii
30
clinical isolates (n = 27) was 0.75–6.0 µg/ml with MIC50 and MIC90 of 1 and 4
31
µg/ml, respectively. For sulbactam regimens, a Monte Carlo simulation
32
estimated the probabilities of attaining the bactericidal target (60% of the time
33
above the MIC) and determined the PK-PD breakpoints (the highest MICs at
34
which the probabilities were 90% or more). In a patient with a CLcr of 15
35
ml/min, a regimen of 1 g twice daily achieved a 90% or more probability
36
against the A. baumannii isolate population; however, 2 g four times daily was
37
needed for a 90% or more probability in a patient with a CL cr of 90 ml/min.
38
The results of the PK-PD target attainment analysis are useful when choosing
39
the sulbactam regimen based on the CLcr of the patient and the susceptibility
40
of A. baumannii. Registration number: UMIN000007356.
AC C
EP
TE D
M AN U
SC
RI PT
21
2
ACCEPTED MANUSCRIPT Keywords
42
Sulbactam, Acinetobacter baumannii, Population pharmacokinetics, Monte
43
Carlo simulation, Dosing regimen
AC C
EP
TE D
M AN U
SC
RI PT
41
3
ACCEPTED MANUSCRIPT 44
45
1. Introduction Acinetobacter baumannii causes outbreaks of infections and health care-associated infections, including pneumonia, bacteremia, wound
47
infections, urinary tract infections and meningitis [1]. The increasing
48
prevalence of A. baumannii strains which are resistant to aminoglycosides,
49
carbapenems and quinolones has resulted in the need to modify the therapeutic
50
options, taking into account the susceptibility of strains in particular medical
51
centers [2]. Sulbactam, a β-lactamase inhibitor that is commercially available
52
in a combined formulation with β-lactam antibiotics, demonstrates intrinsic
53
activity against A. baumannii [3]. However, the appropriate sulbactam dosing
54
for A. baumannii infections is currently unknown.
SC
M AN U
The efficacy of antibacterial chemotherapy is determined by the
TE D
55
RI PT
46
interrelationship between the pharmacokinetics (PK) and pharmacodynamics
57
(PD). As reported by Drusano et al. [4], the integration of PK and PD targets
58
derived from both PK data and exposure-response data can be utilized to
59
optimize the dosing regimen. The PK of sulbactam have been well
60
characterized in individual healthy volunteers and adult patients [5,6]. The
61
renal function was integrated into the population PK models as a significant
62
covariate. However, urine concentration data were not used as part of the
63
basic PK model [5]. Sulbactam is a time-dependent bactericidal agent; thus,
64
its antibacterial effects correlate with the exposure time that the free drug
65
concentration remains above the minimum inhibitory concentration (MIC) for
AC C
EP
56
4
ACCEPTED MANUSCRIPT the bacterium (fT > MIC). The fT > MIC targets required for near-maximal
67
bactericidal effects are considered to be 60% of the dosing interval [7].
68
However, the MIC values for sulbactam against clinical isolates have not been
69
reported. In the present study, the MIC distribution for sulbactam was
70
determined for A. baumannii blood culture isolates.
71
RI PT
66
It is important to thoroughly characterize the population PK of
sulbactam in patients, particularly focusing on the drug urinary excretion
73
process and assessing the PK-PD target attainment at 60% fT > MIC based on
74
the susceptibility patterns of A. baumannii. Population PK models reflecting
75
the renal function will make it possible to optimize appropriate sulbactam
76
dosing regimens for patients with impaired renal function.
M AN U
This study therefore conducted a population PK analysis in patients
TE D
77
SC
72
using plasma and urine concentration data and assessed the probability of
79
target attainment (PTA) for sulbactam by the fraction of the population of A.
80
baumannii in each sulbactam MIC category.
AC C
EP
78
5
ACCEPTED MANUSCRIPT 81
2. Methods
82
2.1. Sulbactam administration and sample collection Patients with varying renal function were intravenously administered
84
single and multiple doses of ampicillin-sulbactam (1.5 or 3.0 g) for 0.00138–
85
1.25 h infusions. Blood samples were immediately collected and centrifuged at
86
3,000 × g for 10 min, and plasma samples were stored at -70°C until measured
87
as below at Kagoshima University Hospital.
M AN U
88
90
2.2. Measurement of sulbactam concentrations in plasma
The concentrations of sulbactam in plasma were measured by
TE D
89
SC
RI PT
83
high-performance liquid chromatography, with minor modifications of the
92
methods described by Martin et al. [8]. The analytical column was a Mightysil
93
RP-18, 5 µm, 250 × 4.6 mm (Kanto Chemical, Tokyo, Japan). For sulbactam,
94
the ultraviolet wavelength was 322 nm, and the mobile phase consisted of 1%
95
acetonitrile and 99% potassium phosphate buffer (0.1 mol/l, pH 6.4). The flow
96
rate was 1.0 ml/min. The lowest detectable concentration of sulbactam was 0.1
97
µg/ml. For sulbactam, the intra- and inter-day accuracy (as absolute values for
98
the relative errors of the means) and precision (as coefficients of variation)
99
were within 10%.
AC C
EP
91
6
ACCEPTED MANUSCRIPT
100 101
2.3. PK data The above-mentioned drug concentration data (153 plasma samples from 51 patients) were obtained in Department of Surgery, Kagoshima
103
University Hospital. This PK study was approved by the Ethics Review Board
104
of Kagoshima University Hospital. In addition to the observational PK data, the
105
PK data (35 plasma concentrations and 27 cumulative urinary excretions from
106
15 patients) were obtained by extracting data from five reports on separate
107
studies [9–13], which disclosed detailed raw data for individual patients. The
108
plasma and urinary concentrations in the literature were determined by
109
microbiological assay and gas chromatography-mass spectrometry methods,
110
which were compatible with high-performance liquid chromatography method
111
[14, 15].
114
SC
M AN U
TE D
EP
113
2.4. Population PK modeling
AC C
112
RI PT
102
The plasma and urine data were combined with demographic variables
115
to produce nonlinear mixed-effects modeling program datasets that were used
116
in the population analyses. The sulbactam PK data were modeled using the
117
NONMEM program, version 7.2.0 (ICON Development Solutions, Ellicott, MD,
118
USA). In the modeling, the NONMEM subroutine ADVAN6 with the first-order 7
ACCEPTED MANUSCRIPT
conditional estimation method were used. A two-compartment model was fitted
120
to the plasma and urine data, and was selected as the basic PK model based on
121
a preliminary analysis (Fig. 1). Therefore, a multi-compartment model [16] was
122
chosen to describe both the plasma and urine data as follows:
RI PT
119
dXc/dt = R – (CL r/Vc + CLnr/Vc + Q/Vc ) × Xc + Q/Vp × Xp
124
dXp /dt = Q/Vc × Xc – Q/Vp × Xp
125
dXu /dt = CL r/Vc × Xc
M AN U
SC
123
where Xc , Xp , and Xu are the amounts of sulbactam in the central, peripheral
127
and urine compartments (mg); R is the drug infusion rate (mg/h); CL r and CLnr
128
are the renal and nonrenal clearances (l/h); Vc and Vp are the volumes of
129
distribution of the central and peripheral compartments (l) and Q is the
130
intercompartmental (central–peripheral) clearance (l/h). The inter-individual variability was modeled using an exponential
EP
131
TE D
126
error model: θi = θ × exp(ηi ), where θ i is the fixed-effects parameter for the
133
i-th subject, θ is the mean value of the fixed-effects parameter in the
134
population and η is a random inter-individual variable which is normally
135
distributed with mean of 0, and variance of ω 2.
136 137
AC C
132
The residual variability, including drug measurement error from differences in the measurement methods, was modeled using a combined 8
ACCEPTED MANUSCRIPT
proportional and additive error model: Cobs,ij = Cpred,ij × (1 + ε1 ) + ε2 , where
139
Cobs,ij and Cpred,ij denote the jth observed and predicted concentrations for the
140
ith subject, and ε is a random intraindividual error which is normally
141
distributed with mean of 0, and variance of σ2.
142
RI PT
138
The individual PK parameters obtained from the basic model were
plotted against continuous covariates (age, body weight, blood urea nitrogen,
144
serum creatinine and creatinine clearance) and categorical covariate (sex)
145
separately. Then, covariates that showed a correlation with the individual PK
146
parameters obtained from the basic model were introduced into the model one
147
at a time. A decrease in the objective function value of >3.84 from the basic
148
model was regarded as statistically significant during the covariate screening
149
process. The full model was built by incorporating the significant covariates,
150
and the final model was developed by a backward deletion method. The
151
coefficients in the full model were excluded from the model one at a time, and
152
an increase in the objective function value of >6.63 from the full model was
153
regarded as statistically significant.
154
AC C
EP
TE D
M AN U
SC
143
The validity of the final model was evaluated by the bootstrap method
155
(1,000 replications) using the Perl-speaks-NONMEM program, version 3.7.6
156
(Karlsson M, Hooker A, Nordgren R, Harling K; Uppsala University, Uppsala, 9
ACCEPTED MANUSCRIPT
157
Sweden).
158
160
2.5. Microbiological data
RI PT
159
The MIC distributions for sulbactam were determined by the Etest
(Sysmex bioMérieux Co. Ltd., Tokyo, Japan) for 27 A. baumannii blood culture
162
isolates obtained between 2003 and 2013 from Kagoshima University Hospital.
163
The MIC for sulbactam was determined using the standardized agar dilution
164
method according to the Clinical and Laboratory Standards Institute guidelines
165
[17]. A suspension of bacteria equivalent to the 0.5 McFarland turbidity
166
standards was inoculated onto Mueller-Hinton agar plates (Becton, Dickinson
167
and Company; Tokyo, Japan). An Etest for sulbactam was plated onto the agar.
168
The MIC value was read following 16–20 h of incubation at 37°C.
TE D
M AN U
SC
161
EP
169
2.6. Probability of the target attainment analysis using a Monte Carlo
171
simulation
172
AC C
170
A simulation using the Monte Carlo sampling method [16] was
173
conducted to assess the PD profile of the sulbactam regimens. For each tested
174
regimen, the following process was repeated 10,000 times using the Crystal
175
Ball software, version 2000 (Oracle, Redwood Shores, CA, USA). A set of 10
ACCEPTED MANUSCRIPT
fixed-effects parameters (CL r, CLnr, Vc , Q, and Vp ) was randomly generated
177
according to each θ and ω value for the final PK model. The plasma versus
178
time curve of the unbound drug concentration was simulated using the
179
fixed-effects parameters, where a value of 38% protein binding [18] was used.
180
The time point at which the free drug concentration coincided with a specific
181
MIC value was determined, and fT > MIC was calculated as the cumulative
182
percentage of a 24-h period. The PTA (%) was determined as the fraction that
183
achieved at least 60% fT > MIC (the bactericidal target against A. baumannii)
184
[7] of 10,000 estimates. For each sulbactam regimen, the PK-PD breakpoint
185
was determined at the highest MIC at which the probability of 60% fT > MIC
186
attainment in plasma was 90% or more. The probability at a specific MIC was
187
then multiplied by the fraction of the population of the bacterium in each MIC
188
category, and the sum of the individual products was defined as the expected
189
population PTA (%).
AC C
EP
TE D
M AN U
SC
RI PT
176
11
ACCEPTED MANUSCRIPT
190
3. Results
191
3.1. Population PK modeling A total of 188 plasma samples and 27 urine samples from 66 adult
193
patients were used for the population PK modeling. The demographic and
194
physiopathological parameters of the study subjects are shown in Table 1. The final population PK parameters are listed in Table 2. The
SC
195
RI PT
192
inter-individual variability for Q was finally fixed as zero, because it was
197
negligibly small (η < 0.0025). Incorporation of CL cr into CL r caused the
198
largest change in the objective function value, although age and body weight
199
also had a significant effect on the CL r. Because age and body weight showed
200
a high correlation with the CLcr, it was not additionally incorporated into the
201
CL r to avoid a collinearity effect. None of the examined covariates had a
202
significant effect on the CLnr, Vc , Q or Vp . Thus, the mean fixed-effects
203
parameters were CL r (l/h) = 0.0792 × CL cr (ml/min), CL nr (l/h) = 2.35, Vc (l) =
204
12.2, Q (l/h) = 4.68 and Vp (l) = 4.44. All relative standard error (standard
205
error/estimate) values via the covariance step in the NONMEM modeling were
206
MIC (required for the bactericidal activity of β-lactam), with a MIC
290
of 8 µg/ml observed when sulbactam was administered by a 4-h infusion of 1 g
291
three times daily. This was determined using PK data from healthy volunteers
292
without consideration of the renal function. A 4-h infusion of sulbactam
293
resulted in a greater PD index of T > MIC compared with a 1-h infusion. The
294
pharmacodynamic assessment demonstrated that the approved regimens of 1-12
295
g/day should be effective [6].
EP
AC C
296
TE D
289
Oliveira et al. demonstrated that sulbactam (1 g every 8 h)-containing
297
regimens appeared to be comparable to regimens of other agents when the
298
infecting organisms were susceptible to sulbactam [20]. For serious A.
299
baumannii infections, Fishbain et al. recommended dosing of at least 6 g daily 17
ACCEPTED MANUSCRIPT
in divided doses for patient with normal renal function [21]. According to the
301
EUCAST database, we estimated the population probabilities of attaining 60%
302
fT > MIC in plasma against the EUCAST-MIC distribution of A. baumannii for
303
sulbactam regimens in patient populations with various degrees of CLcr (data
304
not shown). The expected population probability fT > MIC attainment was
305
similarly potent against the present A. baumannii isolate population as against
306
the EUCAST-MIC distribution of A. baumannii.
M AN U
SC
RI PT
300
This study has some limitations. The sulbactam concentration data were
308
obtained in the presence of ampicillin, because sulbactam is clinically available
309
only as a combination preparation at this time. We believe that the PK are
310
independent between the two drugs, although the possibility that ampicillin
311
affected sulbactam cannot be excluded. Regarding the PK-PD, the target value
312
employed (60% fT > MIC) comes from experiments on animals in the absence
313
of ampicillin [22]. We believe that 60% fT > MIC is the best target value
314
currently available, because in vitro activity of ampicillin-sulbactam against A.
315
baumannii results from sulbactam alone, and there is no synergy effect between
316
the two drugs [3]. Therefore, the results of our PK-PD target attainment
317
analysis are considered to provide useful information on the dosing strategy for
318
sulbactam alone and also in the presence of ampicillin. However, the
AC C
EP
TE D
307
18
ACCEPTED MANUSCRIPT
319
implications of our findings and proposals should be verified in the clinical
320
setting. In conclusion, the population PK modeling fitted plasma and urine data
RI PT
321
to precisely describe the population PK of sulbactam. A Monte Carlo
323
simulation based on this model demonstrated that the approved regimens of 2–8
324
g/day in divided doses (1-h infusions) had sufficient bactericidal effects against
325
A. baumannii in most typical patients. The simulation also indicated that a
326
shorter dosing interval was useful for maintaining the plasma sulbactam
327
concentrations above the MIC in patients with various degrees of CLcr. The
328
data are also useful for choosing a sulbactam regimen based on the CLcr of the
329
patient and the susceptibility of the A. baumannii.
AC C
EP
TE D
M AN U
SC
322
19
ACCEPTED MANUSCRIPT
Conflict of interest
331
None declared.
AC C
EP
TE D
M AN U
SC
RI PT
330
20
ACCEPTED MANUSCRIPT
332
References
333
[1]
Maragakis LL, Perl TM. Acinetobacter baumannii: epidemiology, antimicrobial resistance, and treatment options. Clin Infect Dis. 2008; 46:
335
1254–63.
336
[2]
RI PT
334
Sieniawski K, Kaczka K, Rucińska M, Gagis L, Pomorski L.
Acinetobacter baumannii nosocomial infections. Pol Przegl Chir. 2013;
338
85: 483–90. [3]
M AN U
339
SC
337
Corbella X, Ariza J, Ardanuy C, Vuelta M, Tubau F, Sora M, et al. Efficacy of sulbactam alone and in combination with ampicillin in
341
nosocomial infections caused by multiresistant Acinetobacter baumannii.
342
J Antimicrob Chemother. 1998; 42: 793–802. [4]
antimicrobialagents. Clin Infect Dis. 2003; 36(Suppl. 1): S42–50.
344 [5]
pharmacokinetics of ampicillin and sulbactam in patients with
346
community-acquired pneumonia: evaluation of impact of renal
347
impairment. Br J Clin Pharmacol. 2014; 77: 509–21.
348 349 350
Soto E, Shoji S, Muto C, Tomono Y, Marshall S. Population
AC C
345
Drusano GL. Prevention of resistance: a goal for dose selection for
EP
343
TE D
340
[6]
Jaruratanasirikul S, Wongpoowarak W, Aeinlang N, Jullangkoon M. Pharmacodynamics modeling to optimize dosage regimens of sulbactam. 21
ACCEPTED MANUSCRIPT
Antimicrob Agents Chemother. 2013; 57: 3441–4.
351 352
[7]
Yokoyama Y, Matsumoto K, Ikawa K, Watanabe E, Shigemi A, Umezaki Y, et al. Pharmacokinetic and pharmacodynamic evaluation of sulbactam
354
against Acinetobacter baumannii in in vitro and murine thigh and lung
355
infection models. Int J Antimicrob Agents. 2014; 43: 547–552. [8]
Martin C, Cotin A, Giraud A, Beccani-Argeme M, Alliot P, Mallet MN, et
SC
356
RI PT
353
al. Comparison of concentrations of sulbactam-ampicillin administered by
358
bolus injections or bolus plus continuous infusion in tissues of patients
359
undergoing colorectal surgery. Antimicrob Agents Chemother. 1998; 42:
360
1093–7.
363 364 365
TE D
Sekine O, Usuda Y, Aoki N, Yuasa Y, Kohda Y. Clinical study of sulbactam•ampicillin. Jpn J Chemother. 1988; 36: 130–6.
[10] Yura J, Shinagawa N, Mizuno A, Mashita K, Taniguchi M, Hori K, et al.
EP
362
[9]
Basic and clinical studies on sulbactam•ampicillin in the surgical field.
AC C
361
M AN U
357
Jpn J Chemother. 1988; 36: 324–33.
366
[11] Ohta N, Morinaga O, Hirai M, Yamaguchi Y, Ushiyama T, Tama A, et al.
367
Clinical and pharmacokinetic studies on sulbactam•ampicillin. Jpn J
368
Chemother. 1988; 36: 382–94.
369
[12] Cho N, Fukunaga K, Kunii Katsuaki, Kimura T, Saito Y, Suzuki A, et al. 22
ACCEPTED MANUSCRIPT
370
Basic and clinical studies on sulbactam•ampicillin in obstetrics and
371
gynecology. Jpn J Chemother. 1988; 36: 466–89. [13] Yokoyama Y, Matsumoto K, Yamamoto H, Iguro Y, Imoto Y, Ikawa K, et
373
al. Pharmacokinetics of ampicillin-sulbactam and the renal function-based
374
optimization of dosing regimens for prophylaxis in patients undergoing
375
cardiovascular surgery. J Infect Chemother. 2012; 18: 878–82.
SC
[14] Shimooka K, Ito M, Matsumoto K, Niimi H, Matsunaga T, Kawasaki K.
M AN U
376
RI PT
372
Sensitive high-performance liquid chromatographic-amperometric and
378
microbiological assays for determining sulbactam in serum. Jpn J
379
Chemother. 1988; 36: 81–9.
380
TE D
377
[15] Kano H, Sekiguchi K, Takematsu K, Shimooka K, Oki T. Assay methods
of sulbactam and cefoperazone in body fluids and tissues. Jpn J
382
Chemother. 1988; 32: 131–41.
384 385
[16] Ikawa K, Morikawa N, Uehara S, Monden K, Yamada Y, Honda N, et al.
AC C
383
EP
381
Pharmacokinetic-pharmacodynamic target attainment analysis of doripenem in infected patients. Int J Antimicrob Agents. 2009; 33: 276–9.
386
[17] Wayne PA. Methods for dilution antimicrobial susceptibility tests for
387
bacteria that grow aerobically: approved standard 8th ed. Clinical and
388
Laboratory Standards Institute. 2009; Document M7-A8. 23
ACCEPTED MANUSCRIPT
[18] Trotman RL, Williamson JC, Shoemaker DM, Salzer WL. Antibiotic
390
dosing in critically ill adult patients receiving continuous renal
391
replacement therapy. Clin Infect Dis. 2005; 41: 1159–66.
392
RI PT
389
[19] De la Pena A, Derendorf H. Pharmacokinetic properties of beta-lactamase
inhibitors. Int J Clin Pharmacol Ther. 1999; 37: 63–75.
394
[20] Oliveira MS, Prado GV, Costa SF, Grinbaum RS, Levin AS.
SC
393
Ampicillin/sulbactam compared with polymyxins for the treatment of
396
infections caused by carbapenem-resistant Acinetobacter spp. J
397
Antimicrob Chemother. 2008; 61: 1369–75.
401
TE D
400
Dis. 2010; 51: 79–84.
[22] Craig WA. Pharmacokinetic/pharmacodynamic parameters: rationale for
antibacterial dosing of mice and men. Clin Infect Dis. 1998; 26: 1–10.
EP
399
[21] Fishbain J, Peleg AY. Treatment of Acinetobacter infections. Clin Infect
AC C
398
M AN U
395
24
ACCEPTED MANUSCRIPT
RI PT
Figure legends
Fig. 1. The multicompartment pharmacokinetic model for sulbactam. Xc, Xp and Xu
SC
are the amounts of sulbactam (mg) in the central, peripheral and urine
M AN U
compartments; R, the drug infusion rate (mg/h); CLr and CLnr, the renal and non-renal clearances (l/h); Vc and Vp, the volumes of distribution of the central and
TE D
peripheral compartments (l) and Q, the intercompartmental clearance (l/h).
Fig. 2. Diagnostic scatter plots of the final population pharmacokinetic model for
EP
sulbactam (◇, 188 plasma samples; ×, 27 urine samples). The observed versus
AC C
population-predicted plasma concentration (A), observed versus population-predicted cumulative urinary excretion (B), conditional weighted residual for plasma versus time (C) and conditional weighted residual for urine versus time (D).
Fig. 3. The MIC distribution for sulbactam against Acinetobacter baumannii (27 25
ACCEPTED MANUSCRIPT
RI PT
clinical isolates).
Fig. 4. The probabilities of attaining 60% fT > MIC in the plasma at specific MICs for sulbactam regimens (1-h infusions) in patients with a creatinine clearance (CLcr)
SC
of 15 ml/min (A), 30 ml/min (B), 60 ml/min (C) or 90 ml/min (D). The dotted lines
AC C
EP
TE D
M AN U
represent the 90% probability.
26
ACCEPTED MANUSCRIPT
Number or mean ± SD (range)
Sex (male:female)
51:15
Age (years)
67.3 ± 15.2 (21–90)
Body weight (kg)
57.7. ± 10.3 (35.7–78.6)
Blood urea nitrogen (mg/dl)
17.2 ± 10.8 (7–71)
Serum creatinine (mg/dl)
1.1 ± 1.8 (0.4–13.9)
Creatinine clearance (ml/min)
61.7 ± 28.6 (5.1–128.9)
AC C
EP
TE D
M AN U
SC
Characteristics
RI PT
Table 1 Demographic and physiopathological parameters of the study subjects (n = 66)
ACCEPTED MANUSCRIPT
Table 2
Relative standard error (covariance step)
0.107
EP
AC C
0.0619–0.0965
0.190
1.31–3.38
0.0434
10.7–13.7
0.0935
3.34–6.01
TE D
Fixed-effects parameter CLr (l/h) = θ1 × CLcr (ml/min) θ1 0.0792 CLnr (l/h) = θ2 θ2 2.35 Vc (l) = θ3 θ3 12.2 Q (l/h) = θ4 θ4 4.68 Vp (l) = θ5 θ5 4.44 Interindividual variability (exponential model) ηCLr 0.242 (ωCLr = 0.523) ηCLnr 0.257 (ωCLnr = 0.541) ηVc 0.0523 (ωVc = 0.232) ηQ 0 ηVp 0.557 (ωVp = 0.863) Residual variability (combined proportion and additive model) εprop 0.00247 (σprop = 4.97%) εadd 0.265 (σadd = 0.515 µg/ml)
95% confidence interval (bootstrap method)
SC
Estimate
M AN U
Parameter
RI PT
The final population parameters of sulbactam in the multicompartment pharmacokinetic model (Fig. 1)
0.133
3.00–5.87
0.337 0.416 0.432 None 0.487
0.104–0.379 0.034–0.481 0.0142–0.0903 None 0.007–1.179
0.453 0.496
0.00105–0.00388 0.079–0.564
θ, population mean value; η, random variable which is normally distributed with a mean of zero and variance, ω2; ε, random error which is normally distributed with a mean of zero and variance, σ2
ACCEPTED MANUSCRIPT
Table 3
a
CLcr = 60 ml/min MIC attainment (%) CLcr = CLcr = CLcr = CLcr = 15 ml/min 30 ml/min 60 ml/min 90 ml/min 93.9 83.6 57.9 37.5 98.7 95.8 84.1 67.8 99.8 98.7 94.4 84.6 98.4 94.1 74.9 55.4 99.8 98.8 94.3 83.3 100 99.8 98.6 94.5
AC C
Sulbactam regimen (1-h infusion)
ACCEPTED MANUSCRIPT
Vp
TE D
Q R
AC C
EP
Central Xc
Vc
SC
M AN U
Peripheral Xp
RI PT
Fig. 1
CLr CLnr
Urine Xu
ACCEPTED MANUSCRIPT
(A)
60 50 40
70 60 50 40 30 20
SC
30
80
RI PT
70
20 10 0
(C)
TE D
5
10 0
0 10 20 30 40 50 60 70 80 90 100 Population-predicted cumulative urinary excretion (% of dose)
10 20 30 40 50 60 70 80 Population-predicted plasma concentration (µg/ml)
4 3
EP
2 1
AC C
0 -1 -2 -3 -4 -5
Conditional weighted residual for urine
0
Conditional weighted residual for plasma
(B)
100 90
Observed cumulative urinary excretion (% of dose)
80
M AN U
Observed plasma concentration (µg/ml)
Fig. 2
(D)
5 4 3 2 1 0
-1 -2 -3 -4 -5
0
12
24 36 Time (h)
48
60
0
4
8
12 Time (h)
16
20
24
ACCEPTED MANUSCRIPT
RI PT
Fig. 3
0.9
0.6
0.8
SC
■
0.9
0.7
M AN U
0.7
0.5
0.6 0.5 0.4
TE D
0.4 0.3
0.3 0.2
EP
0.2 0.1 0 0.75
AC C
Fraction (
)
0.8
1
1
0.1 0
1.5
2
MIC (µg/ml)
3
4
6
Cumulative fraction (—)
1
ACCEPTED MANUSCRIPT
(A)
(B)
80 70 50
CLcr = 15 ml/min ○ 1 g twice daily △ 1 g three times daily □ 1 g four times daily ● 2 g twice daily ▲ 2 g three times daily ■ 2 g four times daily
40 30 20 10 0
70 60 50 40 30
CLcr = 30 ml/min ○ 1 g twice daily △ 1 g three times daily □ 1 g four times daily ● 2 g twice daily ▲ 2 g three times daily ■ 2 g four times daily
20 10 0
0.5
1
2
4 8 MIC (µg/ml)
16
32
100 90
EP
80 60 50 CLcr = 60 ml/min ○ 1 g twice daily △ 1 g three times daily □ 1 g four times daily ● 2 g twice daily ▲ 2 g three times daily ■ 2 g four times daily
40 30 20 10
AC C
70
0.5
64
TE D
(C) Probability of 60% fT > MIC attainment (%)
80
M AN U
60
90
RI PT
90
100
SC
Probability of 60% fT > MIC attainment (%)
100
1
2
4 8 MIC (µg/ml)
16
32
64
(D)
100 Probability of 60% fT > MIC attainment (%)
Probability of 60% fT > MIC attainment (%)
Fig. 4
90 CLcr = 90 ml/min ○ 1 g twice daily △ 1 g three times daily □ 1 g four times daily ● 2 g twice daily ▲ 2 g three times daily ■ 2 g four times daily
80 70 60 50 40 30 20 10 0
0 0.5
1
2
4 8 MIC (µg/ml)
16
32
64
0.5
1
2
4
8
MIC (µg/ml)
16
32
64
S1341-321X(14)00420-6
DOI:
10.1016/j.jiac.2014.12.005
Reference:
JIC 230
To appear in:
Journal of Infection and Chemotherapy
Received Date: 18 September 2014 Revised Date:
16 December 2014
Accepted Date: 18 December 2014
Please cite this article as: Yokoyama Y, Matsumoto K, Ikawa K, Watanabe E, Morikawa N, Takeda Y, Population pharmacokinetic-pharmacodynamic target attainment analysis of sulbactam in patients with impaired renal function: dosing considerations for Acinetobacter baumannii infections, Journal of Infection and Chemotherapy (2015), doi: 10.1016/j.jiac.2014.12.005. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
1
Journal of Infection and Chemotherapy
2
Original article
RI PT
3
Population pharmacokinetic-pharmacodynamic target attainment analysis of
5
sulbactam in patients with impaired renal function: dosing considerations for
6
Acinetobacter baumannii infections
SC
4
M AN U
7 8
Yuta Yokoyamaa,b , Kazuaki Matsumotoa , Kazuro Ikawab , Erika Watanabea ,
9
Norifumi Morikawab , Yasuo Takedaa, *
10 11
a
12
Medical and Dental Sciences, Kagoshima University
13
b
TE D
EP
Department of Clinical Pharmacotherapy, Hiroshima University
AC C
14
Department of Clinical Pharmacy and Pharmacology, Graduate School of
15
*Corresponding author
16
Address: 8-35-1 Sakuragaoka, Kagoshima 890-8520, Japan
17
Tel.: +81-99-275-5543; Fax: +81-99-265-5293
18
E-mail: [email protected]
1
ACCEPTED MANUSCRIPT 19 20
Abstract This study aimed to perform a pharmacokinetic (PK)-pharmacodynamic (PD) target attainment analysis of sulbactam against
22
Acinetobacter baumannii in patients with impaired renal function. The PK
23
data (188 plasma samples and 27 urine samples) were modeled simultaneously.
24
The mean population parameters were CL r (l/h) = 0.0792 × CLcr (ml/min),
25
CLnr (l/h) = 2.35, Vc (l) = 12.2, Q (l/h) = 4.68 and Vp (l) = 4.44, where CL r and
26
CLnr are the renal and non-renal clearances, Vc and Vp are the distribution
27
volumes of the central and peripheral compartments and Q is
28
intercompartmental clearance. The creatinine clearance (CL cr) was the most
29
significant covariate. The determined MIC of sulbactam against A. baumannii
30
clinical isolates (n = 27) was 0.75–6.0 µg/ml with MIC50 and MIC90 of 1 and 4
31
µg/ml, respectively. For sulbactam regimens, a Monte Carlo simulation
32
estimated the probabilities of attaining the bactericidal target (60% of the time
33
above the MIC) and determined the PK-PD breakpoints (the highest MICs at
34
which the probabilities were 90% or more). In a patient with a CLcr of 15
35
ml/min, a regimen of 1 g twice daily achieved a 90% or more probability
36
against the A. baumannii isolate population; however, 2 g four times daily was
37
needed for a 90% or more probability in a patient with a CL cr of 90 ml/min.
38
The results of the PK-PD target attainment analysis are useful when choosing
39
the sulbactam regimen based on the CLcr of the patient and the susceptibility
40
of A. baumannii. Registration number: UMIN000007356.
AC C
EP
TE D
M AN U
SC
RI PT
21
2
ACCEPTED MANUSCRIPT Keywords
42
Sulbactam, Acinetobacter baumannii, Population pharmacokinetics, Monte
43
Carlo simulation, Dosing regimen
AC C
EP
TE D
M AN U
SC
RI PT
41
3
ACCEPTED MANUSCRIPT 44
45
1. Introduction Acinetobacter baumannii causes outbreaks of infections and health care-associated infections, including pneumonia, bacteremia, wound
47
infections, urinary tract infections and meningitis [1]. The increasing
48
prevalence of A. baumannii strains which are resistant to aminoglycosides,
49
carbapenems and quinolones has resulted in the need to modify the therapeutic
50
options, taking into account the susceptibility of strains in particular medical
51
centers [2]. Sulbactam, a β-lactamase inhibitor that is commercially available
52
in a combined formulation with β-lactam antibiotics, demonstrates intrinsic
53
activity against A. baumannii [3]. However, the appropriate sulbactam dosing
54
for A. baumannii infections is currently unknown.
SC
M AN U
The efficacy of antibacterial chemotherapy is determined by the
TE D
55
RI PT
46
interrelationship between the pharmacokinetics (PK) and pharmacodynamics
57
(PD). As reported by Drusano et al. [4], the integration of PK and PD targets
58
derived from both PK data and exposure-response data can be utilized to
59
optimize the dosing regimen. The PK of sulbactam have been well
60
characterized in individual healthy volunteers and adult patients [5,6]. The
61
renal function was integrated into the population PK models as a significant
62
covariate. However, urine concentration data were not used as part of the
63
basic PK model [5]. Sulbactam is a time-dependent bactericidal agent; thus,
64
its antibacterial effects correlate with the exposure time that the free drug
65
concentration remains above the minimum inhibitory concentration (MIC) for
AC C
EP
56
4
ACCEPTED MANUSCRIPT the bacterium (fT > MIC). The fT > MIC targets required for near-maximal
67
bactericidal effects are considered to be 60% of the dosing interval [7].
68
However, the MIC values for sulbactam against clinical isolates have not been
69
reported. In the present study, the MIC distribution for sulbactam was
70
determined for A. baumannii blood culture isolates.
71
RI PT
66
It is important to thoroughly characterize the population PK of
sulbactam in patients, particularly focusing on the drug urinary excretion
73
process and assessing the PK-PD target attainment at 60% fT > MIC based on
74
the susceptibility patterns of A. baumannii. Population PK models reflecting
75
the renal function will make it possible to optimize appropriate sulbactam
76
dosing regimens for patients with impaired renal function.
M AN U
This study therefore conducted a population PK analysis in patients
TE D
77
SC
72
using plasma and urine concentration data and assessed the probability of
79
target attainment (PTA) for sulbactam by the fraction of the population of A.
80
baumannii in each sulbactam MIC category.
AC C
EP
78
5
ACCEPTED MANUSCRIPT 81
2. Methods
82
2.1. Sulbactam administration and sample collection Patients with varying renal function were intravenously administered
84
single and multiple doses of ampicillin-sulbactam (1.5 or 3.0 g) for 0.00138–
85
1.25 h infusions. Blood samples were immediately collected and centrifuged at
86
3,000 × g for 10 min, and plasma samples were stored at -70°C until measured
87
as below at Kagoshima University Hospital.
M AN U
88
90
2.2. Measurement of sulbactam concentrations in plasma
The concentrations of sulbactam in plasma were measured by
TE D
89
SC
RI PT
83
high-performance liquid chromatography, with minor modifications of the
92
methods described by Martin et al. [8]. The analytical column was a Mightysil
93
RP-18, 5 µm, 250 × 4.6 mm (Kanto Chemical, Tokyo, Japan). For sulbactam,
94
the ultraviolet wavelength was 322 nm, and the mobile phase consisted of 1%
95
acetonitrile and 99% potassium phosphate buffer (0.1 mol/l, pH 6.4). The flow
96
rate was 1.0 ml/min. The lowest detectable concentration of sulbactam was 0.1
97
µg/ml. For sulbactam, the intra- and inter-day accuracy (as absolute values for
98
the relative errors of the means) and precision (as coefficients of variation)
99
were within 10%.
AC C
EP
91
6
ACCEPTED MANUSCRIPT
100 101
2.3. PK data The above-mentioned drug concentration data (153 plasma samples from 51 patients) were obtained in Department of Surgery, Kagoshima
103
University Hospital. This PK study was approved by the Ethics Review Board
104
of Kagoshima University Hospital. In addition to the observational PK data, the
105
PK data (35 plasma concentrations and 27 cumulative urinary excretions from
106
15 patients) were obtained by extracting data from five reports on separate
107
studies [9–13], which disclosed detailed raw data for individual patients. The
108
plasma and urinary concentrations in the literature were determined by
109
microbiological assay and gas chromatography-mass spectrometry methods,
110
which were compatible with high-performance liquid chromatography method
111
[14, 15].
114
SC
M AN U
TE D
EP
113
2.4. Population PK modeling
AC C
112
RI PT
102
The plasma and urine data were combined with demographic variables
115
to produce nonlinear mixed-effects modeling program datasets that were used
116
in the population analyses. The sulbactam PK data were modeled using the
117
NONMEM program, version 7.2.0 (ICON Development Solutions, Ellicott, MD,
118
USA). In the modeling, the NONMEM subroutine ADVAN6 with the first-order 7
ACCEPTED MANUSCRIPT
conditional estimation method were used. A two-compartment model was fitted
120
to the plasma and urine data, and was selected as the basic PK model based on
121
a preliminary analysis (Fig. 1). Therefore, a multi-compartment model [16] was
122
chosen to describe both the plasma and urine data as follows:
RI PT
119
dXc/dt = R – (CL r/Vc + CLnr/Vc + Q/Vc ) × Xc + Q/Vp × Xp
124
dXp /dt = Q/Vc × Xc – Q/Vp × Xp
125
dXu /dt = CL r/Vc × Xc
M AN U
SC
123
where Xc , Xp , and Xu are the amounts of sulbactam in the central, peripheral
127
and urine compartments (mg); R is the drug infusion rate (mg/h); CL r and CLnr
128
are the renal and nonrenal clearances (l/h); Vc and Vp are the volumes of
129
distribution of the central and peripheral compartments (l) and Q is the
130
intercompartmental (central–peripheral) clearance (l/h). The inter-individual variability was modeled using an exponential
EP
131
TE D
126
error model: θi = θ × exp(ηi ), where θ i is the fixed-effects parameter for the
133
i-th subject, θ is the mean value of the fixed-effects parameter in the
134
population and η is a random inter-individual variable which is normally
135
distributed with mean of 0, and variance of ω 2.
136 137
AC C
132
The residual variability, including drug measurement error from differences in the measurement methods, was modeled using a combined 8
ACCEPTED MANUSCRIPT
proportional and additive error model: Cobs,ij = Cpred,ij × (1 + ε1 ) + ε2 , where
139
Cobs,ij and Cpred,ij denote the jth observed and predicted concentrations for the
140
ith subject, and ε is a random intraindividual error which is normally
141
distributed with mean of 0, and variance of σ2.
142
RI PT
138
The individual PK parameters obtained from the basic model were
plotted against continuous covariates (age, body weight, blood urea nitrogen,
144
serum creatinine and creatinine clearance) and categorical covariate (sex)
145
separately. Then, covariates that showed a correlation with the individual PK
146
parameters obtained from the basic model were introduced into the model one
147
at a time. A decrease in the objective function value of >3.84 from the basic
148
model was regarded as statistically significant during the covariate screening
149
process. The full model was built by incorporating the significant covariates,
150
and the final model was developed by a backward deletion method. The
151
coefficients in the full model were excluded from the model one at a time, and
152
an increase in the objective function value of >6.63 from the full model was
153
regarded as statistically significant.
154
AC C
EP
TE D
M AN U
SC
143
The validity of the final model was evaluated by the bootstrap method
155
(1,000 replications) using the Perl-speaks-NONMEM program, version 3.7.6
156
(Karlsson M, Hooker A, Nordgren R, Harling K; Uppsala University, Uppsala, 9
ACCEPTED MANUSCRIPT
157
Sweden).
158
160
2.5. Microbiological data
RI PT
159
The MIC distributions for sulbactam were determined by the Etest
(Sysmex bioMérieux Co. Ltd., Tokyo, Japan) for 27 A. baumannii blood culture
162
isolates obtained between 2003 and 2013 from Kagoshima University Hospital.
163
The MIC for sulbactam was determined using the standardized agar dilution
164
method according to the Clinical and Laboratory Standards Institute guidelines
165
[17]. A suspension of bacteria equivalent to the 0.5 McFarland turbidity
166
standards was inoculated onto Mueller-Hinton agar plates (Becton, Dickinson
167
and Company; Tokyo, Japan). An Etest for sulbactam was plated onto the agar.
168
The MIC value was read following 16–20 h of incubation at 37°C.
TE D
M AN U
SC
161
EP
169
2.6. Probability of the target attainment analysis using a Monte Carlo
171
simulation
172
AC C
170
A simulation using the Monte Carlo sampling method [16] was
173
conducted to assess the PD profile of the sulbactam regimens. For each tested
174
regimen, the following process was repeated 10,000 times using the Crystal
175
Ball software, version 2000 (Oracle, Redwood Shores, CA, USA). A set of 10
ACCEPTED MANUSCRIPT
fixed-effects parameters (CL r, CLnr, Vc , Q, and Vp ) was randomly generated
177
according to each θ and ω value for the final PK model. The plasma versus
178
time curve of the unbound drug concentration was simulated using the
179
fixed-effects parameters, where a value of 38% protein binding [18] was used.
180
The time point at which the free drug concentration coincided with a specific
181
MIC value was determined, and fT > MIC was calculated as the cumulative
182
percentage of a 24-h period. The PTA (%) was determined as the fraction that
183
achieved at least 60% fT > MIC (the bactericidal target against A. baumannii)
184
[7] of 10,000 estimates. For each sulbactam regimen, the PK-PD breakpoint
185
was determined at the highest MIC at which the probability of 60% fT > MIC
186
attainment in plasma was 90% or more. The probability at a specific MIC was
187
then multiplied by the fraction of the population of the bacterium in each MIC
188
category, and the sum of the individual products was defined as the expected
189
population PTA (%).
AC C
EP
TE D
M AN U
SC
RI PT
176
11
ACCEPTED MANUSCRIPT
190
3. Results
191
3.1. Population PK modeling A total of 188 plasma samples and 27 urine samples from 66 adult
193
patients were used for the population PK modeling. The demographic and
194
physiopathological parameters of the study subjects are shown in Table 1. The final population PK parameters are listed in Table 2. The
SC
195
RI PT
192
inter-individual variability for Q was finally fixed as zero, because it was
197
negligibly small (η < 0.0025). Incorporation of CL cr into CL r caused the
198
largest change in the objective function value, although age and body weight
199
also had a significant effect on the CL r. Because age and body weight showed
200
a high correlation with the CLcr, it was not additionally incorporated into the
201
CL r to avoid a collinearity effect. None of the examined covariates had a
202
significant effect on the CLnr, Vc , Q or Vp . Thus, the mean fixed-effects
203
parameters were CL r (l/h) = 0.0792 × CL cr (ml/min), CL nr (l/h) = 2.35, Vc (l) =
204
12.2, Q (l/h) = 4.68 and Vp (l) = 4.44. All relative standard error (standard
205
error/estimate) values via the covariance step in the NONMEM modeling were
206
MIC (required for the bactericidal activity of β-lactam), with a MIC
290
of 8 µg/ml observed when sulbactam was administered by a 4-h infusion of 1 g
291
three times daily. This was determined using PK data from healthy volunteers
292
without consideration of the renal function. A 4-h infusion of sulbactam
293
resulted in a greater PD index of T > MIC compared with a 1-h infusion. The
294
pharmacodynamic assessment demonstrated that the approved regimens of 1-12
295
g/day should be effective [6].
EP
AC C
296
TE D
289
Oliveira et al. demonstrated that sulbactam (1 g every 8 h)-containing
297
regimens appeared to be comparable to regimens of other agents when the
298
infecting organisms were susceptible to sulbactam [20]. For serious A.
299
baumannii infections, Fishbain et al. recommended dosing of at least 6 g daily 17
ACCEPTED MANUSCRIPT
in divided doses for patient with normal renal function [21]. According to the
301
EUCAST database, we estimated the population probabilities of attaining 60%
302
fT > MIC in plasma against the EUCAST-MIC distribution of A. baumannii for
303
sulbactam regimens in patient populations with various degrees of CLcr (data
304
not shown). The expected population probability fT > MIC attainment was
305
similarly potent against the present A. baumannii isolate population as against
306
the EUCAST-MIC distribution of A. baumannii.
M AN U
SC
RI PT
300
This study has some limitations. The sulbactam concentration data were
308
obtained in the presence of ampicillin, because sulbactam is clinically available
309
only as a combination preparation at this time. We believe that the PK are
310
independent between the two drugs, although the possibility that ampicillin
311
affected sulbactam cannot be excluded. Regarding the PK-PD, the target value
312
employed (60% fT > MIC) comes from experiments on animals in the absence
313
of ampicillin [22]. We believe that 60% fT > MIC is the best target value
314
currently available, because in vitro activity of ampicillin-sulbactam against A.
315
baumannii results from sulbactam alone, and there is no synergy effect between
316
the two drugs [3]. Therefore, the results of our PK-PD target attainment
317
analysis are considered to provide useful information on the dosing strategy for
318
sulbactam alone and also in the presence of ampicillin. However, the
AC C
EP
TE D
307
18
ACCEPTED MANUSCRIPT
319
implications of our findings and proposals should be verified in the clinical
320
setting. In conclusion, the population PK modeling fitted plasma and urine data
RI PT
321
to precisely describe the population PK of sulbactam. A Monte Carlo
323
simulation based on this model demonstrated that the approved regimens of 2–8
324
g/day in divided doses (1-h infusions) had sufficient bactericidal effects against
325
A. baumannii in most typical patients. The simulation also indicated that a
326
shorter dosing interval was useful for maintaining the plasma sulbactam
327
concentrations above the MIC in patients with various degrees of CLcr. The
328
data are also useful for choosing a sulbactam regimen based on the CLcr of the
329
patient and the susceptibility of the A. baumannii.
AC C
EP
TE D
M AN U
SC
322
19
ACCEPTED MANUSCRIPT
Conflict of interest
331
None declared.
AC C
EP
TE D
M AN U
SC
RI PT
330
20
ACCEPTED MANUSCRIPT
332
References
333
[1]
Maragakis LL, Perl TM. Acinetobacter baumannii: epidemiology, antimicrobial resistance, and treatment options. Clin Infect Dis. 2008; 46:
335
1254–63.
336
[2]
RI PT
334
Sieniawski K, Kaczka K, Rucińska M, Gagis L, Pomorski L.
Acinetobacter baumannii nosocomial infections. Pol Przegl Chir. 2013;
338
85: 483–90. [3]
M AN U
339
SC
337
Corbella X, Ariza J, Ardanuy C, Vuelta M, Tubau F, Sora M, et al. Efficacy of sulbactam alone and in combination with ampicillin in
341
nosocomial infections caused by multiresistant Acinetobacter baumannii.
342
J Antimicrob Chemother. 1998; 42: 793–802. [4]
antimicrobialagents. Clin Infect Dis. 2003; 36(Suppl. 1): S42–50.
344 [5]
pharmacokinetics of ampicillin and sulbactam in patients with
346
community-acquired pneumonia: evaluation of impact of renal
347
impairment. Br J Clin Pharmacol. 2014; 77: 509–21.
348 349 350
Soto E, Shoji S, Muto C, Tomono Y, Marshall S. Population
AC C
345
Drusano GL. Prevention of resistance: a goal for dose selection for
EP
343
TE D
340
[6]
Jaruratanasirikul S, Wongpoowarak W, Aeinlang N, Jullangkoon M. Pharmacodynamics modeling to optimize dosage regimens of sulbactam. 21
ACCEPTED MANUSCRIPT
Antimicrob Agents Chemother. 2013; 57: 3441–4.
351 352
[7]
Yokoyama Y, Matsumoto K, Ikawa K, Watanabe E, Shigemi A, Umezaki Y, et al. Pharmacokinetic and pharmacodynamic evaluation of sulbactam
354
against Acinetobacter baumannii in in vitro and murine thigh and lung
355
infection models. Int J Antimicrob Agents. 2014; 43: 547–552. [8]
Martin C, Cotin A, Giraud A, Beccani-Argeme M, Alliot P, Mallet MN, et
SC
356
RI PT
353
al. Comparison of concentrations of sulbactam-ampicillin administered by
358
bolus injections or bolus plus continuous infusion in tissues of patients
359
undergoing colorectal surgery. Antimicrob Agents Chemother. 1998; 42:
360
1093–7.
363 364 365
TE D
Sekine O, Usuda Y, Aoki N, Yuasa Y, Kohda Y. Clinical study of sulbactam•ampicillin. Jpn J Chemother. 1988; 36: 130–6.
[10] Yura J, Shinagawa N, Mizuno A, Mashita K, Taniguchi M, Hori K, et al.
EP
362
[9]
Basic and clinical studies on sulbactam•ampicillin in the surgical field.
AC C
361
M AN U
357
Jpn J Chemother. 1988; 36: 324–33.
366
[11] Ohta N, Morinaga O, Hirai M, Yamaguchi Y, Ushiyama T, Tama A, et al.
367
Clinical and pharmacokinetic studies on sulbactam•ampicillin. Jpn J
368
Chemother. 1988; 36: 382–94.
369
[12] Cho N, Fukunaga K, Kunii Katsuaki, Kimura T, Saito Y, Suzuki A, et al. 22
ACCEPTED MANUSCRIPT
370
Basic and clinical studies on sulbactam•ampicillin in obstetrics and
371
gynecology. Jpn J Chemother. 1988; 36: 466–89. [13] Yokoyama Y, Matsumoto K, Yamamoto H, Iguro Y, Imoto Y, Ikawa K, et
373
al. Pharmacokinetics of ampicillin-sulbactam and the renal function-based
374
optimization of dosing regimens for prophylaxis in patients undergoing
375
cardiovascular surgery. J Infect Chemother. 2012; 18: 878–82.
SC
[14] Shimooka K, Ito M, Matsumoto K, Niimi H, Matsunaga T, Kawasaki K.
M AN U
376
RI PT
372
Sensitive high-performance liquid chromatographic-amperometric and
378
microbiological assays for determining sulbactam in serum. Jpn J
379
Chemother. 1988; 36: 81–9.
380
TE D
377
[15] Kano H, Sekiguchi K, Takematsu K, Shimooka K, Oki T. Assay methods
of sulbactam and cefoperazone in body fluids and tissues. Jpn J
382
Chemother. 1988; 32: 131–41.
384 385
[16] Ikawa K, Morikawa N, Uehara S, Monden K, Yamada Y, Honda N, et al.
AC C
383
EP
381
Pharmacokinetic-pharmacodynamic target attainment analysis of doripenem in infected patients. Int J Antimicrob Agents. 2009; 33: 276–9.
386
[17] Wayne PA. Methods for dilution antimicrobial susceptibility tests for
387
bacteria that grow aerobically: approved standard 8th ed. Clinical and
388
Laboratory Standards Institute. 2009; Document M7-A8. 23
ACCEPTED MANUSCRIPT
[18] Trotman RL, Williamson JC, Shoemaker DM, Salzer WL. Antibiotic
390
dosing in critically ill adult patients receiving continuous renal
391
replacement therapy. Clin Infect Dis. 2005; 41: 1159–66.
392
RI PT
389
[19] De la Pena A, Derendorf H. Pharmacokinetic properties of beta-lactamase
inhibitors. Int J Clin Pharmacol Ther. 1999; 37: 63–75.
394
[20] Oliveira MS, Prado GV, Costa SF, Grinbaum RS, Levin AS.
SC
393
Ampicillin/sulbactam compared with polymyxins for the treatment of
396
infections caused by carbapenem-resistant Acinetobacter spp. J
397
Antimicrob Chemother. 2008; 61: 1369–75.
401
TE D
400
Dis. 2010; 51: 79–84.
[22] Craig WA. Pharmacokinetic/pharmacodynamic parameters: rationale for
antibacterial dosing of mice and men. Clin Infect Dis. 1998; 26: 1–10.
EP
399
[21] Fishbain J, Peleg AY. Treatment of Acinetobacter infections. Clin Infect
AC C
398
M AN U
395
24
ACCEPTED MANUSCRIPT
RI PT
Figure legends
Fig. 1. The multicompartment pharmacokinetic model for sulbactam. Xc, Xp and Xu
SC
are the amounts of sulbactam (mg) in the central, peripheral and urine
M AN U
compartments; R, the drug infusion rate (mg/h); CLr and CLnr, the renal and non-renal clearances (l/h); Vc and Vp, the volumes of distribution of the central and
TE D
peripheral compartments (l) and Q, the intercompartmental clearance (l/h).
Fig. 2. Diagnostic scatter plots of the final population pharmacokinetic model for
EP
sulbactam (◇, 188 plasma samples; ×, 27 urine samples). The observed versus
AC C
population-predicted plasma concentration (A), observed versus population-predicted cumulative urinary excretion (B), conditional weighted residual for plasma versus time (C) and conditional weighted residual for urine versus time (D).
Fig. 3. The MIC distribution for sulbactam against Acinetobacter baumannii (27 25
ACCEPTED MANUSCRIPT
RI PT
clinical isolates).
Fig. 4. The probabilities of attaining 60% fT > MIC in the plasma at specific MICs for sulbactam regimens (1-h infusions) in patients with a creatinine clearance (CLcr)
SC
of 15 ml/min (A), 30 ml/min (B), 60 ml/min (C) or 90 ml/min (D). The dotted lines
AC C
EP
TE D
M AN U
represent the 90% probability.
26
ACCEPTED MANUSCRIPT
Number or mean ± SD (range)
Sex (male:female)
51:15
Age (years)
67.3 ± 15.2 (21–90)
Body weight (kg)
57.7. ± 10.3 (35.7–78.6)
Blood urea nitrogen (mg/dl)
17.2 ± 10.8 (7–71)
Serum creatinine (mg/dl)
1.1 ± 1.8 (0.4–13.9)
Creatinine clearance (ml/min)
61.7 ± 28.6 (5.1–128.9)
AC C
EP
TE D
M AN U
SC
Characteristics
RI PT
Table 1 Demographic and physiopathological parameters of the study subjects (n = 66)
ACCEPTED MANUSCRIPT
Table 2
Relative standard error (covariance step)
0.107
EP
AC C
0.0619–0.0965
0.190
1.31–3.38
0.0434
10.7–13.7
0.0935
3.34–6.01
TE D
Fixed-effects parameter CLr (l/h) = θ1 × CLcr (ml/min) θ1 0.0792 CLnr (l/h) = θ2 θ2 2.35 Vc (l) = θ3 θ3 12.2 Q (l/h) = θ4 θ4 4.68 Vp (l) = θ5 θ5 4.44 Interindividual variability (exponential model) ηCLr 0.242 (ωCLr = 0.523) ηCLnr 0.257 (ωCLnr = 0.541) ηVc 0.0523 (ωVc = 0.232) ηQ 0 ηVp 0.557 (ωVp = 0.863) Residual variability (combined proportion and additive model) εprop 0.00247 (σprop = 4.97%) εadd 0.265 (σadd = 0.515 µg/ml)
95% confidence interval (bootstrap method)
SC
Estimate
M AN U
Parameter
RI PT
The final population parameters of sulbactam in the multicompartment pharmacokinetic model (Fig. 1)
0.133
3.00–5.87
0.337 0.416 0.432 None 0.487
0.104–0.379 0.034–0.481 0.0142–0.0903 None 0.007–1.179
0.453 0.496
0.00105–0.00388 0.079–0.564
θ, population mean value; η, random variable which is normally distributed with a mean of zero and variance, ω2; ε, random error which is normally distributed with a mean of zero and variance, σ2
ACCEPTED MANUSCRIPT
Table 3
a
CLcr = 60 ml/min MIC attainment (%) CLcr = CLcr = CLcr = CLcr = 15 ml/min 30 ml/min 60 ml/min 90 ml/min 93.9 83.6 57.9 37.5 98.7 95.8 84.1 67.8 99.8 98.7 94.4 84.6 98.4 94.1 74.9 55.4 99.8 98.8 94.3 83.3 100 99.8 98.6 94.5
AC C
Sulbactam regimen (1-h infusion)
ACCEPTED MANUSCRIPT
Vp
TE D
Q R
AC C
EP
Central Xc
Vc
SC
M AN U
Peripheral Xp
RI PT
Fig. 1
CLr CLnr
Urine Xu
ACCEPTED MANUSCRIPT
(A)
60 50 40
70 60 50 40 30 20
SC
30
80
RI PT
70
20 10 0
(C)
TE D
5
10 0
0 10 20 30 40 50 60 70 80 90 100 Population-predicted cumulative urinary excretion (% of dose)
10 20 30 40 50 60 70 80 Population-predicted plasma concentration (µg/ml)
4 3
EP
2 1
AC C
0 -1 -2 -3 -4 -5
Conditional weighted residual for urine
0
Conditional weighted residual for plasma
(B)
100 90
Observed cumulative urinary excretion (% of dose)
80
M AN U
Observed plasma concentration (µg/ml)
Fig. 2
(D)
5 4 3 2 1 0
-1 -2 -3 -4 -5
0
12
24 36 Time (h)
48
60
0
4
8
12 Time (h)
16
20
24
ACCEPTED MANUSCRIPT
RI PT
Fig. 3
0.9
0.6
0.8
SC
■
0.9
0.7
M AN U
0.7
0.5
0.6 0.5 0.4
TE D
0.4 0.3
0.3 0.2
EP
0.2 0.1 0 0.75
AC C
Fraction (
)
0.8
1
1
0.1 0
1.5
2
MIC (µg/ml)
3
4
6
Cumulative fraction (—)
1
ACCEPTED MANUSCRIPT
(A)
(B)
80 70 50
CLcr = 15 ml/min ○ 1 g twice daily △ 1 g three times daily □ 1 g four times daily ● 2 g twice daily ▲ 2 g three times daily ■ 2 g four times daily
40 30 20 10 0
70 60 50 40 30
CLcr = 30 ml/min ○ 1 g twice daily △ 1 g three times daily □ 1 g four times daily ● 2 g twice daily ▲ 2 g three times daily ■ 2 g four times daily
20 10 0
0.5
1
2
4 8 MIC (µg/ml)
16
32
100 90
EP
80 60 50 CLcr = 60 ml/min ○ 1 g twice daily △ 1 g three times daily □ 1 g four times daily ● 2 g twice daily ▲ 2 g three times daily ■ 2 g four times daily
40 30 20 10
AC C
70
0.5
64
TE D
(C) Probability of 60% fT > MIC attainment (%)
80
M AN U
60
90
RI PT
90
100
SC
Probability of 60% fT > MIC attainment (%)
100
1
2
4 8 MIC (µg/ml)
16
32
64
(D)
100 Probability of 60% fT > MIC attainment (%)
Probability of 60% fT > MIC attainment (%)
Fig. 4
90 CLcr = 90 ml/min ○ 1 g twice daily △ 1 g three times daily □ 1 g four times daily ● 2 g twice daily ▲ 2 g three times daily ■ 2 g four times daily
80 70 60 50 40 30 20 10 0
0 0.5
1
2
4 8 MIC (µg/ml)
16
32
64
0.5
1
2
4
8
MIC (µg/ml)
16
32
64
