The use of Monte Carlo simulation to predict vancomycin dosage for methicillin-resistant  in Thai patients of various ages and with varying degrees of renal function

Vancomycin plays an important role in treating methicillin-resistant Staphylococcus aureus (MRSA) infection in Thai patients. Although some evidence suggests appropriate vancomycin dosing for patients infected with MRSA, vancomycin dosing may be affected by various conditions, such as race [1, 2], age [3], body weight [4], and renal function [5]. Therefore, vancomycin dosing calculated for patients in other countries may not be appropriate to use for Thai patients. Purwonugroho et al. conducted a study to provide specific vancomycin population pharmacokinetics in Thai patients [6]. The study illustrated that creatinine clearance (CL_cr) calculated by the Cockcroft–Gault equation and age were covariates of vancomycin clearance (CL_v) and volume of the central compartment (V_c), respectively. Thus, suitable vancomycin dosage could vary depending on CL_cr and age. In addition, MRSA susceptibility data to vancomycin is a crucial factor for evaluating proper vancomycin dosing. Canut et al. determined suitable vancomycin dosages for European patients with MRSA infection [7]. The study revealed that Belgian patients can use vancomycin 1 g every 12 h as adequate dosing, but Spanish patients have to use vancomycin at least 1 g every 8 h because they are infected with MRSA with lower susceptibility to vancomycin.

Monte Carlo simulation is useful to predict a suitable vancomycin dosing regimen because it can provide the outcome without using data from actual patients. Although previous studies have already used simulations to determine appropriate vancomycin dosing [8, 9, 10], to our knowledge, no published study has investigated appropriate vancomycin dosing for Thai patients of various ages and with varying degrees of renal function. Therefore, the aim of the present study was to predict appropriate vancomycin dosing regimens for Thai patients of various ages and with varying degrees of renal function.

Materials and methods

Susceptibility data for MRSA isolates

After the approval from institutional review boards of Silpakorn University (approval No. 1/2558) and Chaoprayayomraj Hospital (approval No. YM 014/2558), 105 unduplicated MRSA isolates from Chaoprayayomraj Hospital, Thailand, were accumulated during 2014. The MRSA vancomycin susceptibility data were determined using an Epsilometer test or E-test (M.I.C.E; Thermo Fisher Scientific) in accordance with CLSI 2014.

Pharmacokinetic parameters of vancomycin

A simulation was conducted using previously published population pharmacokinetic models derived from 212 Thai patients with 319 vancomycin serum concentration data to develop pharmacokinetic parameters. The 2-compartment model was the best structural model to fit the population data. Pharmacokinetic parameters were vancomycin clearance (CL_v) (L/h) = 0.044 × CL_cr (mL/min), the central volume of distribution (V_c) (L) = 0.542 × age (years), intercompartment clearance (Q) = 6.95 L/h, and volume of peripheral compartment (V_p) = 44.2 L [6].

Pharmacodynamic index

The ratio of the 24-h area under the concentration–time curve (AUC₂₄) and minimum inhibitory concentration (MIC) ≥400 was the marker of a good outcome of MRSA treatment with vancomycin [11]. Thus, AUC₂₄/MIC ≥400 was used as a target value of pharmacodynamic index.

Monte Carlo simulation

Monte Carlo simulation is a stochastic simulation method that randomly selects a parameter value from a known distribution. The process is repeated many times (usually 10,000) to generate pharmacokinetic parameters [12]. The generated pharmacokinetic parameters of each simulated patient are incorporated with the structural pharmacokinetics model, MIC distribution, and target value of the pharmacodynamic index to predict the efficacy of antibiotic dosing [13].

A total of 10,000 simulated patients were generated by Monte Carlo simulation using Oracle Crystal Ball software (Oracle Crystal Ball Faculty Classroom Edition) for each vancomycin dosing including 0.75, 1, and 1.5 g every 24 h; 1, 1.25, 1.75, and 2 g every 12 h; and 1 g every 8 h. The age and CL_cr of simulated patients were specified ranging from 20 to 80 years and 20 to 100 mL/min, respectively. The mean (standard deviation) values of age and CL_cr were assumed to be 50 (30) years and 60 (40) mL/min, respectively. V_c and CL_v were assumed to be lognormal distributions. AUC₂₄ at steady state was computed using the trapezoidal rule. The 10,000 simulated patients who were each administrated vancomycin dosage were grouped by age and renal function. The age group of patients was defined as <40, 40–60, and >60 years. In addition, the CL_cr group of patients was defined as <40, 40–60, >60–80, and >80 mL/min. After the simulated patients were grouped, the probability of target attainment (PTA) of each MIC was computed, and then, the cumulative fraction of response (CFR) for each vancomycin dosing administered in each patient group was calculated as follows:

$\begin{matrix} CFR = Σ_{i = 1}^{n} {PTA}_{i} \times F_{i} \end{matrix}$ $$\begin{array}{} \displaystyle \text{CFR}\,\,=\,\,\Sigma^n_{i=1} {\text{PTA}_i}\times F_i \end{array}{} $$

where the subscript i indicates the MIC group classified from lowest to highest MIC value of MRSA isolates, PTA_i stands for the PTA for AUC₂₄/MIC ≥400 of each MIC, and F is the ratio of the MRSA isolates at each MIC.

To evaluate the risk of renal toxicity, this study assessed the probability of attaining a trough concentration (C_trough) of vancomycin of ≤20 mg/L, >20–35 mg/L, and >35 mg/L in each group of patients who received vancomycin.

Results

The distribution of simulated patients in various age groups, including <40, 40–60, and >60 years, was approximately 45%, 35%, and 20%, respectively, whereas that in various renal function groups, including <40, 40–60, >60–80, and >80 mL/min, was approximately 36%, 32%, 20%, and 12%, respectively (data not shown). Sources of specimens of 105 non-duplicated MRSA isolates and MIC distribution are given in Table 1. The PTA of each MIC according to patient age and renal function is listed in Tables 2 and 3, respectively. Figure 1 illustrates that vancomycin dosing at least 2.5 g per day can attain CFR ≥90% in every age group. In addition, for simulated patients with CL_cr <40, 40–60, >60–80, and >80 mL/min, vancomycin dosages that can achieve CFR ≥90% are 1.5, 2.5, 3, and 3.5 g per day, respectively, as shown in Figure 2. More than a half of simulated patients using these regimens attained a C_trough of >20 mg/L.

The cumulative fraction of response (CFR) achieving an area under the curve (AUC)/minimal inhibitory concentration (MIC) >400 and the percentage of trough concentration (C_trough) of vancomycin at steady state for each dosing regimen in simulated patients of various ages. (A) <40 years, (B) 40–60 years, and (C) >60 years. q = quaque or every; l = liter

Table 1

The numbers of methicillin-resistant Staphylococcus aureus (MRSA) clinical isolates according to source of specimens and minimal inhibitory concentration (MIC) distribution (n = 105)

MIC (mg/L)	Source of specimens		Total (%)

	Sputum	Blood
0.5	5	0	5 (4.8)
1	56	1	57 (54.3)
2	41	1	42 (40)
4	1	0	1 (1)

Table 2

The probability of target attainment (PTA) of an area under the curve (AUC)/minimal inhibitory concentration (MIC) > 400 at each MIC according to age of patients

Age(years)	Vancomycin dosing	PTA of each MIC (%)

		0.5	1	2	4
<40	0.75 g every 24 h	92.3	42.0	1.9	0.0
	1 g every 24 h	100	64.6	13.8	0.0
	1.5 g every 24 h	100	92.4	39.9	1.5
	1 g every 12 h	100	100	62.3	14.1
	1.25 g every 12 h	100	100	80.2	27.5
	1 g every 8 h	100	100	92.3	41.2
	1.75 g every 12 h	100	100	99.6	53.3
	2 g every 12 h	100	100	100	63.9
40–60	0.75 g every 24 h	91.9	41.8	1.8	0.0
	1 g every 24 h	100	60.8	13.3	0.0
	1.5 g every 24 h	100	92.7	41.5	1.6
	1 g every 12 h	100	100	63.5	13.9
	1.25 g every 12 h	100	100	80.2	26.8
	1 g every 8 h	100	100	92.6	41.1
	1.75 g every 12 h	100	100	99.7	54.1
	2 g every 12 h	100	100	100	65.5
>60	0.75 g every 24 h	91.3	39.9	1.2	0.0
	1 g every 24 h	100	62.8	13.3	0.0
	1.5 g every 24 h	100	93.2	41.6	1.3
	1 g every 12 h	100	100	62.3	14.2
	1.25 g every 12h	100	100	79.6	28.1
	1 g every 8 h	100	100	92.3	39.8
	1.75 g every 12 h	100	100	99.7	52.0
	2 g every 12 h	100	100	100	64.1

Table 3

The probability of target attainment (PTA) of an area under the curve (AUC)/minimal inhibitory concentration (MIC) >400 at each MIC according to renal function of patients

CL_cr(mL/min)	Vancomycin dosing 0.75 g every 24 h	PTA of each MIC (%)

		0.5	1	2	4
<40	0.75 g every 24 h	100	100	4.7	0.0
	1 g every 24 h	100	100	38.7	0.0
	1.5 g every 24 h	100	100	100	4.1
	1 g every 12 h	100	100	100	40.0
	1.25 g every 12 h	100	100	100	77.3
	1 g every 8 h	100	100	100	100
	1.75 g every 12 h	100	100	100	100
	2 g every 12 h	100	100	100	100
40–60	0.75 g every 24 h	100	14.7	0.0	0.0
	1 g every 24 h	100	86.3	0.0	0.0
	1.5 g every 24 h	100	100	14.4	0.0
	1 g every 12 h	100	100	86.3	0.0
	1.25 g every 12 h	100	100	100	0.0
	1 g every 8 h	100	100	100	14.6
	1.75 g every 12 h	100	100	100	53.1
	2 g every 12 h	100	100	100	87.0
>60–80	0.75 g every 24 h	100	0.0	0.0	0.0
	1 g every 24 h	100	0.0	0.0	0.0
	1.5 g every 24 h	100	100	0.0	0.0
	1 g every 12 h	100	100	0.0	0.0
	1.25 g every 12 h	100	100	59.6	0.0
	1 g every 8 h	100	100	100	0.0
	1.75 g every 12 h	100	100	100	0.0
	2 g every 12 h	100	100	100	0.0
>80	0.75 g every 24 h	31.9	0.0	0.0	0.0
	1 g every 24 h	100.0	0.0	0.0	0.0
	1.5 g every 24 h	100.0	33.1	0.0	0.0
	1 g every 12 h	100.0	100	0.0	0.0
	1.25 g every 12 h	100.0	100	0.0	0.0
	1 g every 8 h	100.0	100	32.5	0.0
	1.75 g every 12 h	100.0	100	97.2	0.0
	2 g every 12 h	100.0	100	100	0.0

CL_cr, creatinine clearance calculated using the Cockcroft–Gault equation

Discussion

Both PTA and CFR are outcomes of simulation applied to evaluate effectiveness of each vancomycin dosing regimen. The higher the trough concentration of vancomycin, the higher the risk of nephrotoxicity [14, 15]. Therefore, to evaluate the safety of vancomycin administration, percentage of achieving trough concentration for each vancomycin dosage should be taken into consideration.

Simulated patients in the present study presented with vancomycin dosages based on CL_cr and age. It is not surprising that CL_cr is a significant factor for CL_v [6], but that an increasing V_c is related to an increasing age.

We found that a vancomycin dose of1.5 g per day can attain CFR ≥90% in all age groups. Variation in age may rarely influence the change in AUC₂₄. However, the result is inconsistent with findings by Revilla et al. [3] who compared the estimated CFR between critically ill patients aged >65 years and younger patients with no severe renal impairment (CL_cr >60 mL/min).

Both groups were given vancomycin at 1 g every 12 h. They found that CFR was inversely proportional to age. Their estimated CFR of critically ill patients aged >65 years was 65% but was 30% in younger patients. The present study found that a vancomycin dosage of 2 g daily, a usual dose for a patient with normal renal function (estimated from the recommended dose of 15–20 mg/kg every 12 h) [16], reached the CFR in only 64.3% of patients with CL_cr >80 mL/min. However, this was not surprising because the usual dose suggested by clinical guidelines is recommended for patients infected with MRSA with an MIC of ≤1 mg/L. By contrast, the frequency of MRSA with an MIC of 2 mg/L in the present study was high (40%). Accordingly, Thai patients need a higher vancomycin dose for empirical treatment of MRSA infection. Vancomycin is excreted primarily by kidney [17]. Patients with good kidney function need higher vancomycin dosing than patients with poor kidney function to successfully treat MRSA infection. The present study showed that vancomycin dosing of at least 1.5, 2.5, 3, and 3.5 g per day could achieve CFR of ≥90% in patients with CL_cr of <40, 40–60, >60–80, >80 mL/min, respectively. Thus, besides the data for vancomycin susceptibility of MRSA isolates, patient renal function has a strong effect on CFR for each vancomycin dosing.

Trough concentration of vancomycin of >20 mg/L is related to undesirable nephrotoxicity. A higher trough level is associated with a higher incidence of renal toxicity [18]. Therefore, trough concentration of vancomycin will need to be considered. As shown in Figures 1 and 2, as a result of high MIC, high vancomycin doses were necessary to attain a CFR of ≥90%. Therefore, the patients using these regimens will have a high possibility of having a serum trough concentration of >20 mg/L and vancomycin-induced kidney injury may occur in them. Consequently, patients using these regimens should be monitored closely for renal function, especially patients receiving concomitant nephrotoxic agents [19]. Nevertheless, we note that vancomycin can be used effectively and safely in patients infected with MRSA with a vancomycin MIC of ≤1 mg/L. Vancomycin doses of 0.75, 1.5, 1.5, and 2 g daily for patients with a CL_cr of <40, 40–60, >60–80, and >80 mL/min, respectively, should be given to patients infected with MRSA with a vancomycin MIC ≤1 mg/L because these regimens can attain a PTA of 100% and the majority of patients can attain a C_trough of ≤20 mg/L. The results are similar to the recommendation from clinical guidelines suggesting that the use of vancomycin should be limited to patients with MRSA infection with a vancomycin MIC of ≤1 mg/L [20].

The present study has some limitations. First, vancomycin susceptibility data against MRSA were accumulated from a regional hospital in Thailand. Thus, the sensitivity profile of MRSA to vancomycin in the present study may not be applicable for the prediction of appropriate vancomycin doses for all Thai patients. Further studies collecting MRSA sensitivity data from several types of hospitals in various regions are required to predict the vancomycin dosage more widely. Second, despite that we could find therapeutic vancomycin doses for patients with a CL_cr of <40 mL/min, vancomycin concentration in these patients may take some time to reach a steady state. The initial low concentration of vancomycin may increase the rate of mortality, and MRSA resistance to vancomycin may develop [21, 22]. Thus, a suitable loading dose may be necessary to reach a therapeutic concentration promptly.

Conclusion

Although vancomycin doses attained a CFR of ≥90% for effectively treating MRSA infection, these regimens may cause kidney injury. We recommend vancomycin doses of 0.75, 1.5, 1.5, and 2 g daily for patients with a CL_cr of <40, 40–60, >60–80, and >80 mL/min, respectively, should be given to patients infected with MRSA with a vancomycin MIC of ≤1 mg/L because these regimens will attain a PTA of 100% and the majority of patients will attain a C_trough of ≤20 mg/L.

eISSN:: 1875-855X
Sprache:: Englisch

Zeitrahmen der Veröffentlichung:: 6 Hefte pro Jahr
Fachgebiete der Zeitschrift:: Medizin, Gesundheitsfachberufe, Vorklinische Medizin, Grundlagenmedizin, andere, Klinische Medizin

Zeitschrift RSS Feed

The use of Monte Carlo simulation to predict vancomycin dosage for methicillin-resistant Staphylococcus aureus in Thai patients of various ages and with varying degrees of renal function

Article Category: Brief communication

Online veröffentlicht: 21. März 2018

Seitenbereich: 379 - 386

DOI: https://doi.org/10.1515/abm-2018-0011

SchlüsselwörterMonte Carlo simulation, MRSA, Thai patients, vancomycin dosing

© 2017 Krairerk Pitaksontayothin, Wichai Santimaleeworagun, Manat Pongchaidecha, Jantana Houngsaitong, Panuwit Srisena

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Figure 1

Figure 2

Schlüsselwörter
Monte Carlo simulation, MRSA, Thai patients, vancomycin dosing