Correlations of vancomycin trough concentration and its efficacy and toxicity in patients in the intensive care unit

Abstract

BACKGROUND

Plasma concentration monitoring is crucial for optimizing vancomycin use, particularly in patients in the intensive care unit (ICU). However, the reference interval for vancomycin plasma concentration remains undetermined.

AIM

To evaluate the correlations of area under the curve (AUC_0-24) and trough concentration (C_min) with efficacy and nephrotoxicity in patients in the ICU.

METHODS

A total of 103 patients treated with vancomycin for methicillin-resistant Staphylococcus aureus infections were analyzed in this study. The associations of clinicodemographic characteristics (including sex, age, weight, infection sites, main etiologies of ICU cases, comorbidities, acute physiological chronic health evaluation II score, and mechanical ventilation) and pharmacokinetics (daily dose, C_min, AUC_0-24, and AUC_0-24/minimum inhibitory concentration) with efficacy and nephrotoxicity of vancomycin were evaluated with univariate and multivariate logistic regression analyses. AUC_0-24 was calculated using VCM-TDM software based on vancomycin population pharmacokinetics and Bayesian feedback method.

RESULTS

C_min over 9.4 μg/mL and AUC_0-24 exceeding 359.6 μg × hour/mL indicated good efficacy against infection. C_min below 14.0 μg/mL predicted no significant nephrotoxicity.

CONCLUSION

In this study, the effective and safe concentration interval for vancomycin in patients in the ICU was C_min 9.4-14.0 μg/mL. Close attention should be paid to adverse effects and renal function during vancomycin treatment.

Key Words: Vancomycin; Therapeutic drug monitoring; Intensive care unit; Trough concentration; Methicillin-resistant Staphylococcus aureus infection; Area under the curve

Core Tip: Plasma concentration monitoring is essential for optimizing vancomycin therapy. This study investigated the relationship between clinical characteristics, pharmacokinetic parameters of vancomycin, specifically the trough concentration and area under the curve, and its efficacy and nephrotoxicity in Chinese patients in the intensive care unit to inform personalized vancomycin administration. The results showed that the effective and safe concentration interval for vancomycin for patients in the intensive care unit was a trough concentration of 9.4-14.0 μg/mL.

INTRODUCTION

Vancomycin has always been the first-line drug treatment for methicillin-resistant Staphylococcus aureus (MRSA), especially in patients in the intensive care unit (ICU)[1]. However, the optimal vancomycin plasma concentration range remains a challenge to define. The Infectious Diseases Society of America recommends an area under the curve (AUC_0-24)/minimum inhibitory concentration (MIC) ratio above 400 for MRSA infection with MIC not exceeding 1 mg/L, while the monitoring of trough concentration (C_min) with a target of 15-20 μg/mL is no longer recommended[2]. Nevertheless, different AUC_0-24/MIC breakpoints reported in different studies have cast doubt on the target of 400[3,4].

In China’s vancomycin consensus guideline, a C_min of 10-15 μg/mL in common infections and a C_min of 10-20 μg/mL in serious MRSA infections is recommended[5]. However, Shen et al[6] reported no significant correlation between pharmacokinetic indices and clinical or microbiological efficacy of vancomycin. Liang et al[7] found that vancomycin C_min was not statistically correlated with clinical efficacy but was an indicator of nephrotoxicity. The cutoff point of C_min on nephrotoxicity was 13 μg/mL. Van Hal et al[8] found that C_min above 15-20 μg/mL increased renal function impairment. Therefore, the threshold for vancomycin-induced nephrotoxicity remains undefined[9]. Clinical factors may contribute to these different study findings.

This study aimed to collect data from 103 patients in the ICU who were treated with vancomycin in a Chinese hospital, monitor the entire treatment process, and analyze the relationship between vancomycin dose, C_min, AUC_0-24, and AUC_0-24/MIC with efficacy and adverse reactions. We also examined the influence of clinical physiological and pathological parameters on treatment outcomes to support individualized treatment and rational use of vancomycin.

MATERIALS AND METHODS

Study objects and vancomycin treatment

The study enrolled patients admitted to the ICU at the Fourth Hospital of Hebei Medical University between October 2021 and June 2024. This study was approved by the Ethics Committee of the Fourth Hospital of Hebei Medical University, No. 2021KS031. All study participants, or their legal guardian, provided informed written consent prior to study enrollment. Patients were included if they had MRSA infections. Infection sites included bloodstream, lung, chest cavity, and abdominal cavity. Many patients had comorbidities, including hypertension, diabetes, tumors, coronary disease, anemia, and hypoproteinemia. Bacterial culture and drug-sensitivity experiments were performed. The bacterial culture experiment was conducted once every 2-3 days. Each experiment lasted 3-7 days to obtain the result. The MIC experiment was conducted at or before vancomycin application. The MIC was determined using the broth microdilution method[10]. If no MRSA bacteria were found or vancomycin insensitivity was indicated, vancomycin treatment was discontinued, and the case was excluded in the study.

Vancomycin dosage regimens varied among patients. The dose included 0.5 g, 0.75 g, 1.0 g, or 1.25 g per dose. The dosing interval included 8 hours, 12 hours, or 24 hours. Each intravenous vancomycin infusion lasted 1 h. The study for each patient began with vancomycin administration and ended with its cessation. The application of vancomycin in every enrolled case started during the ICU treatment. Even if the patient was transferred from the ICU to the general ward after vancomycin application, the study for the case would continue until vancomycin cessation.

Inclusion and exclusion criteria

Inclusion criteria included: (1) MRSA infection; (2) Intravenous vancomycin injection for over 3 days; (3) Monitoring vancomycin C_min; and (4) For all patients undergoing surgical treatment source control was addressed. Exclusion criteria included: (1) Not MRSA infection; (2) No C_min tested; (3) Source uncontrolled; and (4) Pediatric, pregnant, or lactating patients.

Detection method

Ultra-performance liquid chromatography with tandem mass spectrometry method was established for quantifying vancomycin in human plasma[11]. The results of vancomycin detection by this method passed the consistency assessment of the Beijing Clinical Laboratory Center. Plasma samples were collected 0.5 h before the fourth or subsequent vancomycin dose, with detection concentration serving as C_min. In this study, 214 C_min from 103 patients were examined. Among them, 51 cases were tested for C_min once, 21 cases were tested 2 times, 15 cases were tested 3 times, 10 cases were tested 4 times, 4 cases were tested 5 times, 1 case was tested 6 times, and 1 case was tested 10 times.

Efficacy and toxicity evaluation

Bacterial culture results were used for efficacy evaluation. Effective was defined as the disappearance or full recovery of all symptoms and signs, recovery of all non-microbial indicators such as imaging and laboratory tests, and no MRSA pathogenic bacteria cultured from the infection site. Ineffective was defined as persistent or worsening symptoms or signs after treatment, new symptoms or signs of the disease, or other antimicrobial treatments for this disease with the original infected pathogenic bacteria tested in the specimens from the infection site. Vancomycin-related adverse effects were judged by the clinical symptoms and laboratory test results. Nephrotoxicity was defined as an increase in serum creatinine (Scr) ≥ 26.5 μmol/L within 48 h or Scr increase ≥ 1.5 times the baseline within 7 days[12].

AUC calculation

Real-time vancomycin concentration time curves were simulated using VCM-TDM software based on the vancomycin population pharmacokinetic model, Bayesian feedback method, and C_min detection values. Patient information, dosing regimen, and C_min detection values were input to the process to obtain the concentration values and concentration time curves for AUC calculation.

Statistical analysis

Statistical analyses were performed by GraphPad Prism 9.0. Clinicodemographic characteristics were analyzed as descriptive parameters. The Mann-Whitney U test was applied to compare continuous data. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic accuracy and specificity of pharmacokinetic/pharmacodynamic parameters. The value corresponding to the maximum of “sensitivity + specificity-1” was the optimal diagnostic cutoff point. Univariate and multivariate logistic regression analyses were performed to identify independently influencing factors. P < 0.05 or lower was considered statistically significant. The statistical methods of this study were reviewed by a statistician in the Fourth Hospital of Hebei Medical University.

RESULTS

A total of 103 patients were included in this study. Among these patients, 71 were male and 32 were female. The age range was 18-98 years, with a median age of 62.3 years. The average weight was 65.6 kg. The main etiologies of ICU cases included postoperative infection (80.6%), respiratory failure (12.6%), shock (3.9%), and severe injury (2.9%). A total of 91 patients (88.3%) had comorbidities. The infection sites included abdominal cavity (21.4%), lung (34.0%), chest cavity (9.7%), and bloodstream (35.0%). Disease status was recorded with the acute physiological chronic health evaluation II (APACHE II) score. The scoring range was 0-71. Of these, 54 cases (52.4%) were ≤ 35, and 49 cases (47.6%) were > 35. Seventy-four cases (71.8%) received mechanical ventilation, and 29 patients (28.2%) did not.

According to the effective evaluation criteria, among the 103 cases, 77 were effective, and 26 were ineffective, yielding an effective rate of 74.8%. The mean daily dose was not significantly different between the two groups (P = 0.4689) (Table 1). However, C_min, AUC_0-24, and AUC_0-24/MIC of the effective group were significantly higher than those of the ineffective group (P < 0.001). Adverse reactions were also observed during the vancomycin treatment, including 2 cases of skin rashes, 3 cases of liver function impairment, and 27 cases of nephrotoxicity. Thus, nephrotoxicity was the primary adverse effect of vancomycin treatment, with a nephrotoxic rate of 26.2%. No significant differences were shown in the mean daily dose, C_min, AUC_0-24 and AUC_0-24/MIC between the nephrotoxic and non-nephrotoxic groups (P = 0.1371, 0.5211, 0.2931, and 0.9458, respectively) (Table 2).

Table 1 Effect of pharmacokinetic indices on efficacy.

Characteristics	Ineffective (n = 26)	Effective (n = 77)	P value
Mean daily dose (g)	1.7 ± 0.5	1.8± 0.5	0.4689
Cmin (μg/mL)	6.9 ± 3.3	14.5 ± 6.2	< 0.0001
AUC0-24	342.8 ± 119.1	546.8 ± 171.8	< 0.0001
AUC0-24/MIC	467.0 ± 248.1	696.2 ± 327.1	0.0015

AUC: Area under the curve; C_min: Trough concentration; MIC: Minimum inhibitory concentration.

Table 2 Effect of pharmacokinetic indices on nephrotoxicity.

Characteristics	Non-nephrotoxic (n = 76)	Nephrotoxic (n = 27)	P value
Mean daily dose (g)	1.7 ± 0.5	1.9 ± 0.4	0.1371
Cmin (μg/mL)	12.4 ± 6.3	13.3 ± 6.9	0.5211
AUC0-24	484.9 ± 154.4	528.1 ± 245.6	0.2931
AUC0-24/MIC	640.6 ± 322.0	635.7 ± 332.5	0.9458

AUC: Area under the curve; C_min: Trough concentration; MIC: Minimum inhibitory concentration.

ROC curves were used to evaluate the accuracy and specificity of pharmacokinetics/efficacy (Figure 1). The AUC_ROC was 0.5147 with a 95% confidence interval of 0.3831 to 0.6464, 0.8854 (0.8145, 0.9562), 0.8761 (0.7945, 0.9578), and 0.6958 (0.5742, 0.8174) for mean daily dose, C_min, AUC_0-24, and AUC_0-24/MIC, respectively, with P = 0.8228, P < 0.0001, P < 0.0001, and P = 0.0029. The cutoff point of daily dose was 1.125 μg/mL with 88.31% sensitivity and 26.92% specificity. The cutoff point of C_min was 9.4 μg/mL with 85.71% sensitivity and 84.62% specificity. The cutoff point of AUC_0-24 was 359.6 μg × hour/mL with 93.51% sensitivity and 53.85% specificity. The cutoff point of AUC_0-24/MIC was 359.6 μg/mL with 92.21% sensitivity and 38.46% specificity.

Figure 1 Receiver operating characteristic curves for daily dose, trough concentration, area under the curve, area under the curve/minimum inhibitory concentration, and efficacy. AUC: Area under the curve; C_min: Trough concentration; MIC: Minimum inhibitory concentration.

ROC curves were also used to evaluate the diagnostic accuracy and specificity of pharmacokinetic/nephrotoxicity (Figure 2). The AUC_ROC was 0.5765 with a 95% confidence interval of 0.4594 to 0.6936, 0.5436 (0.4143, 0.6729), 0.5402 (0.4060, 0.6744), and 0.5041 (0.3783, 0.6300) for daily dose, C_min, AUC_0-24, and AUC_0-24/MIC, respectively, with P = 0.2391, P = 0.5021, P = 0.5362, and P = 0.9492. The cutoff point of daily dose was 1.65 μg/mL with 85.19% sensitivity and 38.16% specificity. The cutoff point of C_min was 14.00 μg/mL with 48.15% sensitivity and 67.11% specificity. The cutoff point of AUC_0-24 was 500.5 μg × hour/mL with 55.56% sensitivity and 60.53% specificity. The cutoff point of AUC_0-24/MIC was 522.1 μg/mL with 62.96% sensitivity and 44.74% specificity.

Figure 2 Receiver operating characteristic curves for daily dose, trough concentration, area under the curve, area under the curve/minimum inhibitory concentration, and nephrotoxicity. AUC: Area under the curve; C_min: Trough concentration; MIC: Minimum inhibitory concentration.

The associations of clinical characteristics and pharmacokinetic indices on efficacy were analyzed by univariate logistic regression analysis. No significant differences were found in the baseline clinical data between the effective and ineffective cases, including sex, age, weight, main etiology, comorbidity, and infectious site, APACHE II score, and mechanical ventilation (Table 3). The results suggested that C_min, AUC_0-24, and AUC_0-24/MIC were all significantly associated with clinical efficacy, with odds ratios (OR) of 33.0, 19.64, and 7.396, respectively (P < 0.001). In addition, there was a certain trend of correlation between daily dose and clinical efficacy, with an OR of 2.784 (P = 0.0709).

Table 3 Univariate logistic analyses between pharmacokinetics and clinical characteristics and efficacy, n (%).

Variable	Ineffective (n = 26)	Effective (n = 77)	OR	95%CI	P value
Sex			0.9815	0.360-2.517	0.9696
Female	8 (30.8)	24 (31.2)	-	-	-
Male	18 (69.2)	53 (68.8)	-	-	-
Age (years)			0.8780	0.358-2.150	0.7745
≤ 65	13 (50.0)	41 (53.2)	-	-	-
> 65	13 (50.0)	36 (46.8)	-	-	-
Weight (kg)			0.6029	0.231-1.492	0.2830
≤ 60	9 (34.6)	36 (46.8)	-	-	-
> 60	17 (65.4)	41 (53.2)	-	-	-
Main etiologies of ICU cases			0.5000	0.183-1.421	0.1805
Postoperative infection	18 (69.2)	63 (81.8)	-	-	-
Other	8 (30.7)	14 (18.2)	-	-	-
Comorbidities			0.5583	0.082-2.315	0.4720
No	2 (7.7)	10 (13.0)	-	-	-
Yes	24 (92.3)	67 (87.0)	-	-	-
Infectious site			0.7960	0.504-1.257	0.3235
Abdominal cavity	5 (19.2)	17 (22.1)	-	-	-
Lung	9 (34.6)	26 (33.8)	-	-	-
Blood	7 (26.9)	29 (37.7)	-	-	-
Chest cavity	5 (19.2)	5 (6.5)	-	-	-
Daily dose (g)			2.7840	0.892-8.484	0.0709
≤ 1.125	7 (26.9)	9 (77.0)	-	-	-
> 1.125	19 (73.1)	68 (88.3)	-	-	-
APACHE II score			0.7143	0.289-1.743	0.4597
≤ 35	12 (46.2)	42 (54.5)	-	-	-
> 35	14 (53.8)	35 (45.5)	-	-	-
Mechanical ventilation			1.1850	0.431-3.076	0.7320
No	8 (30.8)	21 (27.3)	-	-	-
Yes	18 (69.2)	56 (72.7)	-	-	-
Cmin			33.0000	10.440-130.700	< 0.0001
≤ 9.4	22 (84.6)	11 (14.3)	-	-	-
> 9.4	4 (15.4)	66 (85.7)	-	-	-
AUC0-24			19.6400	6.321-71.150	< 0.0001
≤ 359.6	15 (57.7)	5 (6.5)	-	-	-
> 359.6	11 (42.3)	72 (93.5)	-	-	-
AUC0-24/MIC			7.3960	2.405-24.680	0.0006
≤ 359.6	10 (38.5)	6 (7.8)	-	-	-
> 359.6	16 (61.5)	71 (92.2)	-	-	-

APACHE: Acute physiological chronic health evaluation; AUC: Area under the curve; CI: Confidence interval; C_min: Trough concentration; ICU: Intensive care unit; MIC: Minimum inhibitory concentration; OR: Odds ratio.

The associations of clinical characteristics and pharmacokinetic indices on nephrotoxicity were also analyzed by univariate logistic regression analysis. The results demonstrated that no significant differences were found in the baseline clinical data between the non-nephrotoxic and nephrotoxic cases, including sex, age, weight, main etiology, comorbidity, infectious site, APACHE II score, and mechanical ventilation (Table 4). The results showed that daily dose, C_min, AUC_0-24, and AUC_0-24/MIC were not significantly associated with clinical nephrotoxicity, with ORs of 1.465, 1.894, 1.917, and 1.376 and P = 0.4289, P = 0.1612, P = 0.1508, and P = 0.4880, respectively.

Table 4 Univariate logistic analyses between pharmacokinetics and clinical characteristics and nephrotoxicity, n (%).

Variable	Non-nephrotoxic (n = 76)	Nephrotoxic (n = 27)	OR	95%CI	P value
Sex			1.0960	0.430-2.977	0.8509
Female	24 (31.6)	8 (29.6)	-	-	-
Male	52 (68.4)	19 (70.4)	-	-	-
Age (years)			1.2620	0.522-3.068	0.6046
≤ 65	41 (53.9)	13 (46.4)	-	-	-
> 65	35 (46.1)	14 (53.6)	-	-	-
Weight (kg)			1.1770	0.486-2.927	0.7193
≤ 60	34 (44.7)	11 (40.7)	-	-	-
> 60	42 (55.3)	16 (59.3)	-	-	-
Main etiologies of ICU cases			1.0710	0.347-2.994	0.8987
Postoperative infection	60 (78.9)	21 (77.8)	-	-	-
Other	16 (21.1)	6 (22.2)	-	-	-
Comorbidities			1.0750	0.292-5.141	0.9190
No	9 (11.8)	3 (11.1)	-	-	-
Yes	67 (88.2)	24 (88.9)	-	-	-
Infectious site			0.7187	0.432-1.147	0.1808
Abdominal cavity	18 (23.7)	4 (14.8)	-	-	-
Lung	23 (30.2)	12 (44.4)	-	-	-
Blood	27 (35.5)	9 (33.3)	-	-	-
Chest cavity	7 (9.2)	3 (11.1)	-	-	-
Daily dose (g)			1.4650	0.582-3.947	0.4289
≤ 1.65	29 (38.2)	8 (29.6)	-	-	-
> 1.65	47 (61.8)	19 (70.4)	-	-	-
APACHE II score			0.5580	0.220-1.356	0.2047
≤ 35	37 (48.7)	17 (63.0)	-	-	-
> 35	39 (51.3)	10 (37.0)	-	-	-
Mechanical ventilation			0.4510	0.177-1.160	0.0945
No	18 (23.7)	11 (40.7)	-	-	-
Yes	58 (76.3)	16 (38.2)	-	-	-
Cmin (μg/mL)			1.8940	0.771-4.663	0.1612
≤ 14.0	51 (67.1)	14 (51.9)	-	-	-
> 14.0	25 (32.9)	13 (48.1)	-	-	-
AUC0-24			1.9170	0.792-4.731	0.1508
≤ 500.5	46 (60.5)	12 (44.4)	-	-	-
> 500.5	30 (39.5)	15 (55.5)	-	-	-
AUC0-24/MIC			1.3760	0.565-3.485	0.4880
≤ 522.1	34 (44.7)	10 (37.0)	-	-	-
> 522.1	42 (55.3)	17 (63.0)	-	-	-

Data are presented as n (%). APACHE: Acute physiological chronic health evaluation; AUC: Area under the curve; CI: Confidence interval; C_min: Trough concentration; ICU: Intensive care unit; MIC: Minimum inhibitory concentration; OR: Odds ratio.

The association of clinical characteristics and pharmacokinetic indices with clinical efficacy and nephrotoxicity were performed by multivariate logistic regression analysis. Compared with AUC_0-24/MIC, daily dose, C_min, and AUC_0-24 were significantly correlated with efficacy, with P values of 0.0178, 0.0003, and 0.0353, respectively (Table 5). Relative to other clinical characteristics and pharmacokinetic indices, C_min emerged as an independent influencing factor affecting vancomycin nephrotoxicity (P = 0.0429) (Table 6). Collectively, C_min and AUC_0-24 were significantly associated with efficacy and nephrotoxicity. C_min over 9.4 μg/mL and AUC_0-24 over 359.6 μg × hour/mL indicated good efficacy of vancomycin against infection. C_min below 14.0 μg/mL predicted no significant nephrotoxicity. Therefore, in this study, C_min over 9.4 μg/mL and 14.0 μg/mL predicted good efficacy without significant nephrotoxicity.

Table 5 Multivariate logistic analyses between pharmacokinetics and efficacy.

Variables	OR	95%CI	Wald χ2	P value
Intercept	0.3617	0.011-8.888	0.6165	0.5375
Sex	0.5565	0.079-3.229	0.6345	0.5257
Age	0.3449	0.030-2.656	0.9746	0.3298
Weight	0.5051	0.089-2.624	0.8085	0.4188
Main etiologies of ICU cases	0.5206	0.076-3.206	0.7011	0.4833
Comorbidities	0.1924	0.016-1.535	1.4520	0.1464
Infectious site	0.9723	0.405-2.311	0.0651	0.9481
Daily dose	11.0400	1.752-109.200	2.3700	0.0178
Cmin	75.0000	9.551-1201.000	3.6430	0.0003
AUC0-24	28.7700	1.710-2026.000	1.9320	0.0353
AUC0-24/MIC	0.1049	0.002-1.789	1.3610	0.1736
APACHE II score	0.5606	0.100-2.929	0.6910	0.4895
Mechanical ventilation	2.5520	0.363-22.680	0.9161	0.3596

APACHE: Acute physiological chronic health evaluation; AUC: Area under the curve; CI: Confidence interval; C_min: Trough concentration; ICU: Intensive care unit; MIC: Minimum inhibitory concentration; OR: Odds ratio.

Table 6 Multivariate logistic analyses between pharmacokinetics and nephrotoxicity.

Variables	OR	95%CI	Wald χ2	P value
Intercept	0.2826	0.026-2.565	1.0970	0.2727
Sex	0.8383	0.255-2.805	0.2919	0.7704
Age	1.1870	0.449-3.148	0.3484	0.7275
Weight	1.1540	0.403-3.393	0.2666	0.7898
Main etiologies of ICU cases	1.1600	0.338-3.700	0.2468	0.8051
Comorbidities	1.0740	0.247-5.781	0.0913	0.9272
Infectious site	0.6946	0.395-1.164	1.3400	0.1803
Daily dose	2.5480	0.502-17.650	1.0620	0.2882
Cmin	2.9900	1.003-10.470	1.8600	0.0429
AUC0-24	1.0810	0.151-5.831	0.0872	0.9305
AUC0-24/MIC	0.8763	0.258-2.906	0.2160	0.8290
APACHE II score	0.6485	0.212-1.960	0.7714	0.4404
Mechanical ventilation	0.5025	0.156-1.592	1.1730	0.2407

APACHE: Acute physiological chronic health evaluation; AUC: Area under the curve; CI: Confidence interval; C_min: Trough concentration; ICU: Intensive care unit; MIC: Minimum inhibitory concentration; OR: Odds ratio.

DISCUSSION

There are two primary approaches for calculating vancomycin AUC_0-24 with test plasma concentration. The first is the Bayesian method used in this study, which employs C_min to generate accurate and reliable estimates of the AUC_0-24 using a software program embedded with a population pharmacokinetic model and Bayesian feedback method[13]. Such software is readily available for bedside use to identify optimal vancomycin dosage[14]. In China, two Bayesian dose-optimizing software programs are commonly used: Smart dose application on Wechat; and the VCM-TDM on the web (http://www.pharmado.net/). Other software based on vancomycin population pharmacokinetics and Bayesian theory can achieve an equal results[15]. The second method is the equation method, which determines the AUC_0-24 value with C_max, C_min, and a simple first-order pharmacokinetic formula[14,15]. The approach also has reasonable precision with low bias and has been recently validated for AUC_0-24 calculation[13-15]. Both methods are based on vancomycin pharmacokinetic formulas.

Vancomycin, one of the glycopeptide antibiotics, primarily inhibits peptidoglycan synthesis by combining D-alanyl-D-alanine dipeptide and the precursor lipid II ends of peptidoglycan synthesis[16]. Vancomycin can inhibit cell wall synthesis of gram positive bacteria but is ineffective against gram negative bacteria. Smirnova et al[17] conducted a pharmacodynamic study of vancomycin and found a correlation between the minimum bacterial survival number small control growth curve and time bactericidal curve between the area and AUC_0-24/MIC. A pharmacodynamic study of vancomycin treatment in 108 patients with a respiratory tract infection showed that AUC/MIC greater than 400 could achieve good clinical efficacy, and pathogen eradication was poor when AUC_0-24/MIC was less than 400[18]. In the latest study, the starting dose of vancomycin should be calculated based on actual body weight, including patients with obesity, with dose adjustments to reach the target therapeutic concentration based on the actual serum concentration[2]. Continuous infusion administration is not considered more effective than brief administration but has a better safety profile[19].

In this study, the efficacy evaluation of vancomycin on MRSA infection, through U test, ROC evaluation, univariate logistic analysis, and multivariate logistic analysis showed that C_min and AUC_0-24 were significantly associated with efficacy. C_min exceeding 9.4 μg/mL and AUC_0-24 over 359.6 suggested better clinical efficacy. In addition, the daily dose and AUC_0-24/MIC were also correlated with efficacy but was weaker than C_min and AUC_0-24. These results suggest that good clinical benefit may be obtained by adjusting C_min above 9.4 μg/mL or AUC_0-24 above 356.9. The results were generally consistent with the above reports, but the C_min and AUC_0-24 were slightly lower than the reference values. The subtle differences possibly arise from ethnic differences. AUC_0-24/MIC has a slightly lower correlation with efficacy, and one possible reason is detection method or detection error.

The exact cause of acute kidney injury from vancomycin is not yet definite, but the biopsies show two possible reasons[20]. On one hand, vancomycin induces proximal renal tubular epithelial cells to produce oxidative stress, causing cell damage and apoptosis. One the other hand, vancomycin causes obstructive thrombosis, damaging renal tubular epithelial cells and then develops into acute kidney injury. It has been proven that vancomycin can increase the production of superoxide, destroy the mitochondrial structure, and then cause apoptosis. This damage positively correlated with vancomycin dose and action time[21].

In this study, the nephrotoxicity evaluation of vancomycin on MRSA infection through U test, ROC evaluation, and univariate logistic analysis showed that daily dose, C_min, and AUC_0-24 were not significantly associated with nephrotoxicity significance. However, in the multivariate logistic analysis, C_min was a factor affecting nephrotoxicity, with C_min exceeding 14 μg/mL, suggesting potential nephrotoxicity. These results suggest that higher C_min of vancomycin may lead to increased nephrotoxicity in patients in the ICU. However, C_min did not show a significant correlation with nephrotoxicity in the univariate logic analysis, suggesting that the effect of vancomycin C_min on nephrotoxicity may be influenced by other factors.

The thresholds of the C_min and AUC_0-24 causing nephrotoxicity vary in different studies, with some concluding that C_min and AUC_0-24 are not associated with nephrotoxicity. This difference may come from two causes. The difference may come from the different patient characteristics, including the varying bases of renal function in different patients and the physiological and pathological factor status of the patients. The cases in this study came from the ICU and had complex clinical conditions and reduced compensatory capacities of renal functions, which increases the vulnerability to renal dysfunction. In addition, the extent of renal impairment was defined differently by different studies. Some reports have shown that an AUC_0-24 greater than 1300 increases acute renal injury by 2.5 times[22,23]. We selected an Scr increase of 26.5 μmol/L within 48 h or a 1.5-fold increase as the definition of acute renal injury.

There were some limitations in this study. First, this was a single-center observational study with a small sample size and a long study duration, primarily due to the relatively small number of patients with MRSA infection. This study also excluded patients with combined effective antibiotics against gram positive bacteria, which may disturb the efficacy of vancomycin. These factors cause difficulties in patient collection and lengthen the experiment. Second, there was a lack of comprehensive baseline treatment data as a control. Vancomycin is currently the first-line treatment for MRSA infections, making it difficult to find patients with MRSA infection who have not received vancomycin treatment. Additionally, vancomycin plasma concentration monitoring is commonly used in the clinic, and clinicians will monitor vancomycin plasma concentration during treatment to avoid risks. Therefore, it is also difficult to find patients who have not received concentration monitoring as controls. The lack of baseline data may bias the findings. These limitations may limit the accuracy and applicability of the study findings.

CONCLUSION

It is beneficial to monitor C_min and AUC_0-24 of vancomycin during the entire treatment process. C_min exceeding 9.4 μg/mL and AUC_0-24 over 359.6 suggested better clinical efficacy. Additionally, C_min exceeding 14 μg/mL suggests potential nephrotoxicity. A C_min range of 9.4-14.0 μg/mL is suggested as an effective and safe concentration interval. Moreover, during vancomycin treatment of MRSA, close attention should be paid to adverse effects and renal function, as different patients may respond differently to vancomycin.

ACKNOWLEDGEMENTS

VCM-TDM software program used in this study was developed by the Preparation Team of “China Vancomycin Treatment Drug Monitoring Guidelines” and Development Team of “PharmVAN” of the Individualized Drug Delivery Platform of Vancomycin in the Third Hospital of Beijing University. We are grateful for the free use of the software.