Genetic and nongenetic factors affecting vancomycin population pharmacokinetics and pharmacodynamics: A systematic review

Abstract

Despite frequent vancomycin prescribing, achieving optimal therapeutic outcomes is challenging due to its complex pharmacokinetic (PK) and pharmacodynamic (PD) properties. Those characteristics are influenced by various parameters, including both genetic and non-genetic variables, rendering accurate dosing and monitoring challenging for clinicians. This systematic review was conducted in accordance with the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, ensuring a standardized and comprehensive approach to the literature search, data extraction, and reporting. We searched the publications for all appropriate evidence on genetic and non-genetic factors affecting vancomycin population PKs and PDs from 2014 to January 2024. This review’s initial electronic database search found 1,543 relevant studies. Following the removal of automation-ineligible records and in-depth title relevance screening, the titles and abstracts of 67 publications were examined. After duplicate removal, 50 articles remained, and 23 were ultimately included based on the predefined inclusion and exclusion criteria. Vancomycin exhibits substantial PK and PD variability, and standard dosing often fails to achieve therapeutic targets. Area Under the Curve -guided therapeutic drug monitoring remains the most reliable strategy, while genetic factors appear to play a limited, toxicity-focused role.

Introduction

Vancomycin, a glycopeptide antibiotic, is crucial for treating Gram-positive bacterial infections that are unresponsive to alternative antibiotics.[1,2,3,4,5,6] Nevertheless, patients sometimes exhibit substantial disparities in their response to vancomycin treatment, which can result in therapeutic ineffectiveness or the incidence of severe adverse reactions.[7] Despite its widespread use, optimizing vancomycin therapy remains challenging due to its complex pharmacokinetic (PK) and pharmacodynamic (PD) properties, which are influenced by various genetic and nongenetic factors.[8] Recent studies have highlighted the significant role of genetic variability and other covariates in affecting vancomycin’s PK/PD profiles, offering insights into personalized medicine approaches. Vancomycin has a narrow therapeutic index, requiring monitoring of blood drug levels.[9,10] The medicine has undesirable effects such as hypersensitivity, nephrotoxicity, ototoxicity, and agranulocytosis.[11] In addition to that, there is also the issue of vancomycin resistance.[2] Despite comparable dosage regimes, PK investigations have revealed significant diversity in vancomycin concentrations among patients. This may be partially ascribed to hereditary variables influencing medication metabolism and elimination. Genetic variations in drug-metabolizing enzymes, particularly cytochrome P450 enzymes, can profoundly affect the rate and extent of vancomycin metabolism, hence changing the drug’s exposure and efficacy. Vancomycin, a commonly utilized glycopeptide antibiotic, is fundamental in managing severe Gram-positive bacterial infections. Although effective, the PKs and PDs of vancomycin can be markedly affected by genetic variations among individuals. Genetic variations can influence medication metabolism, elimination, and the likelihood of side effects, including nephrotoxicity and drug response with eosinophilia and systemic symptoms (DRESS). The PKs of vancomycin, encompassing absorption, distribution, metabolism, and excretion, are affected by multiple factors, such as renal function and genetic polymorphisms. Population PK modeling indicates that renal function is a crucial factor influencing vancomycin clearance (CL), as people with chronic kidney illness demonstrate modified PK profiles relative to individuals with normal renal function.[12]

Moreover, genetic differences in genes associated with renal transport and metabolism can further influence the PKs of vancomycin. In addition, genetic variants can influence the PDs of vancomycin and the drug’s biochemical and physiological effects on the body. Specific HLA alleles are linked to a heightened risk of vancomycin-induced DRESS, a severe hypersensitivity reaction marked by eosinophilia and systemic symptoms.[13] Identifying these genetic indicators can aid in forecasting patient responses to vancomycin and in customizing treatment approaches to reduce the likelihood of adverse events. This comprehensive systematic review aims to critically evaluate and synthesize the current body of research examining how genetic factors and various other covariates influence vancomycin’s PK and PD properties. By systematically analyzing the available evidence, this review seeks to provide a deeper understanding of the factors that affect vancomycin’s behavior and efficacy in the body. This review will elucidate the intricate relationship between genetic, non-genetic variables and covariates and the therapeutic effects of vancomycin by synthesizing findings from genome-wide association studies (GWAS), population PK modeling, and clinical observations. The primary objective is to augment our comprehension of customized medicine strategies for vancomycin therapy, enhancing treatment outcomes and mitigating the risk of side effects.

Methods

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 standards were utilized for this systematic review.

Eligibility criteria

This systematic review encompassed the subsequent inclusion criteria, which included research involving adult and pediatric patients that examined genetic variants influencing the metabolic or pharmacological response to vancomycin, along with studies evaluating clinically verified nephrotoxicity occurrences. This study excluded irrelevant publications, such as narrative reviews, case reports, and articles that did not focus on the relationship between genetic variation and vancomycin toxicity. The data extracted from the included articles comprised the author, sample size, age, genetic and non-genetic variables and covariates, as well as pharmacokinetic–pharmacodynamic (PK–PD) parameters. The results were evaluated subjectively and quantitatively to discern prevalent trends and emerging distinctions in the association between genetic variation and the risk of nephrotoxicity in individuals administered vancomycin.

Information source

A comprehensive literature review was performed using major online research platforms such as PubMed. The search aimed to identify relevant studies exploring the impact of genetic factors and additional covariates on PKs and PDs.

Search strategy

Articles published in English from 2014 to 2024, observational study, free full text available, study in humans, all genders, all age classifications were chosen to guarantee extensive data coverage. The search terms are listed in Table 1. The article selection process occurred in two phases: (1) screening of titles and abstracts and (2) comprehensive examination of complete texts of research that satisfied the inclusion requirements.

Table 1.

Search term in PUBMED

Search	Search term
#1	Vancomycin [MesH Term]
#2	(Vancomycin [MesH Term]) AND (Pharmacokinetics) [MesH Terms])
#3	(Vancomycin [MesH Term]) AND (pharmacodynamics) [MesH Terms])
#4	(Vancomycin [MesH Term]) AND (genetic variation) [MesH Terms])
#5	(Vancomycin [MesH Term]) AND (patient associated factor) [MesH Terms]) OR (covariates) [MesH Terms])
#6	#1 AND #2 AND #3 AND #4 AND #5

Selection process

Three independent reviewers (DAP, AW, and EM) evaluated each study using predetermined inclusion and exclusion criteria. The Joanna Briggs Institute Critical Appraisal Tools Checklist was employed to assess their quality. The authors collaboratively finalized the scores, resolving any disagreements through discussion. Microsoft Excel 365 and Mendeley were used as automation tools throughout this assessment.

Synthesis methods

Our synthesis approach was adapted from established guidelines for conducting syntheses without meta-analysis.[14] The process involved six key steps, beginning with study classification into relevant groups. Next, data synthesis was conducted by summarizing and synthesizing the findings from the grouped studies, followed by evidence quality assessment through evaluating the certainty of the collected evidence. Data visualization was then performed using suitable techniques to present the synthesized data. Subsequently, results reporting documented the outcomes of the synthesis, and finally, synthesis constraints were acknowledged to highlight the limitations of the approach. This structured method allowed for a comprehensive and systematic review of the literature without conducting a formal meta-analysis. The evaluation of potential reporting bias was conducted independently by two members of the review team. This dual assessment approach helps ensure a more objective and thorough examination of potential biases in the reporting of study results. To evaluate the reliability of the evidence, several factors must be considered, including the volume of research and the number of participants, the consistency of effects across multiple studies, the potential risk of bias within the included studies, and the degree to which the selected studies directly address the research question. A comprehensive assessment of these elements is crucial in determining the overall certainty and strength of the evidence presented in the review.

Results and Discussion

Study selections

The initial electronic database search for this review yielded 1,543 potentially relevant studies. After removing duplicates and automatically excluding ineligible entries, 67 publications were selected for title and abstract screening. Following the removal of 17 additional duplicates, 50 articles remained. These 50 articles were then subjected to predefined inclusion and exclusion criteria, resulting in a final selection of 23 studies for in-depth analysis and evaluation, as illustrated in Figure 1.

Figure 1.

PRISMA flow diagram

Study characteristics

The study involved 1,117 people, with sample sizes varying from 8 to 489 participants in each trial. The principal outcome of 22 studies was PK parameters, while three studies examined the effects of genetic variation, and two studies concentrated on PDs. The mean age of participants, with standard deviation, was 64.5 (0.7) years. This review encompassed a preponderance of papers from Asia (35%), followed by Europe (22%) and the Americas (43%).

Genetic impact on vancomycin pharmacokinetics- pharmacodynamics

This systematic review examined 23 studies in total [Table 2] but found that only a small subset – specifically two studies – addressed the impact of genetic factors on vancomycin’s PK and PD properties. This highlights a significant gap in the current research landscape regarding the role of genetics in vancomycin therapy.[7,8] Based on Van Driest et al.’s findings, a genetic variant located on chromosome 6q22.31 was significantly associated with increased serum creatinine concentration following vancomycin treatment. The most statistically significant single nucleotide polymorphism (SNP) identified was rs2789047, with the A allele conferring risk (β = −0.06, P = 1.1 × 10⁻⁷). This association was further supported by similar effect directions observed in validation cohorts, resulting in a combined meta-analysis P = 1.1 × 10⁻⁷. Variations in this area of chromosome 6, encompassing the genes TBC1D32/C6orf170 and GJA1 (which encodes connexin43), may influence the risk of vancomycin-induced nephrotoxicity.[7] GJA1 encodes connexin43, a gap junction protein found in renal proximal tubules, previously linked to renal damage.[35,36,37] To date, no research has explicitly connected the elevation of serum creatinine resulting from polymorphisms in the GJA-1 gene to the PKs and PDs of vancomycin. The research by Wang et al. indicated that vancomycin-induced DRESS in the Han Chinese demographic correlates with particular HLA alleles (HLA-A32:01, HLA-B07:05, HLA-B40:06, and HLA-B67:01).[13]

Table 2.

Impact of genetic variation and other covariates on pharmacokinetics and pharmacodynamics of vancomycin

Author-year	Sample size	Age	Genetic	Nongenetic	Clinical covariates	PK	PD
Stevenson et al., 2015[15]	20 patients	64 (54–74)	Not assessed	Age, BMI, diabetes status, peritoneal dialysis modality, anuria, causative organisms, and vancomycin serum concentrations	Vancomycin levels, age, BMI, diabetes mellitus, PD modality, anuria, and organism type	Clinical response was positively correlated with more early vancomycin exposure, and pharmacodynamic interactions were reflected in organism-specific effects	Not assessed
Ghosh et al., 2014[16]	320 patients	56.7±16.4	Not assessed	Disease severity scores, source of infection, renal function, vancomycin trough concentrations, and vancomycin AUC0–24/MICBMD	the source of bacteremia, immunosuppression status, and vancomycin AUC0–24/MICBMD below the proposed threshold	Enhanced clinical outcomes were noted across diverse infection types when the ratio of the area under the concentration-time curve to the lowest inhibitory concentration (AUC/MIC) was equal to or above 650	Improved clinical outcomes were typically linked to attaining an AUC/MIC ratio of no <650 for vancomycin treatment across various infection sources
Hahn et al., 2015[17]	230 pediatric patients	0–18 years (median age 2.6 years)	Not assessed	Age, weight, sex, serum creatinine, underlying medical conditions, and vancomycin serum concentrations	Age, weight, and serum creatinine	A PKs model of vancomycin in children, utilizing Bayesian estimation techniques to predict the area under the concentration-time curve (AUC), is an effective method for estimating vancomycin exposure in pediatric patients
Tafelski et al., 2015[18]	63 patients (32 in the I group, 31 in the II group)	Mean age was 63.6±15.5 years	Not assessed	Age, gender, severity scores, comorbidities, infection types, and serum vancomycin levels	Type of infusion administration (continuous versus intermittent)	The study evaluated continuous versus intermittent infusion regimens, revealing that continuous infusion attained therapeutic target blood concentrations more swiftly and with a reduced incidence of sub-therapeutic levels	Not assessed
Van Driest et al., 2015[7]	489	≥18 years	GWAS demonstrated that post-vancomycin increases in serum creatinine are influenced by both drug exposure and genetic determinants of renal function	Dose, duration of therapy, and cumulative exposure, as well as clinical conditions that increase renal stress, including severe infection, hemodynamic instability, and concomitant use of nephrotoxic medications	Age, sex, baseline renal function, body weight, and comorbidities	Not assessed	Research findings indicate a correlation between the chromosome 6q22.31 locus and elevated serum creatinine levels in patients undergoing vancomycin treatment. The genetic variant rs2789047 emerged as the most statistically significant, with the A allele identified as the risk allele. This variant demonstrated a beta effect −0.06 and a highly significant P value of 1.1×10−8
Park et al., 2015[19]	ECMO group: 20 patients Control group: 60 patients	Not assessed	Not assessed	EMCO	Age, body weight, renal function parameters (serum creatinine and creatinine clearance), ECMO modality and duration, and vancomycin dosing regimen	The existing vancomycin dosing protocol employed in the intensive care unit, which involves administering 15–20 mg/kg per dose at 8–12-h intervals, proved inadequate for achieving the desired trough concentration during the initial treatment phase for the majority of patients undergoing extracorporeal membrane oxygenation therapy	Not assessed
Romanelli et al., 2016[20]	98 patients with 169 serum vancomycin measurements	Mean gestational age was 33.52 weeks (SD=4.41)	Not assessed	Renal immaturity	Gestational age, postnatal age, birth weight, body surface area, renal function indicators, and dosing regimen	Researchers conclude that vancomycin displays erratic serum concentrations, and traditional empirical dosage methods inadequately attain goal levels. Their suggestions advocate for commencing therapy with elevated dosages and emphasize the vital importance of therapeutic medication monitoring in neonatal patients. This strategy seeks to enhance vancomycin therapy and elevate treatment results in this at-risk demographic	Not assessed
Moore et al., 2016[21]	14 patients	47 (19–72) years	Not assessed	Variability in patients receiving ECMO, Critical illness–related factors, including fluid shifts, organ dysfunction, and disease severity	Body weight, renal function parameters (such as serum creatinine or creatinine clearance), ECMO-related characteristics (mode and duration of ECMO), and vancomycin dosing regimen	A two-compartment PK model was established that precisely characterized the temporal dynamics of vancomycin. The model’s key parameters comprised a clearance rate of 2.83 L/h and a core volume of distribution of 24.2 L. The findings indicated that conventional vancomycin dose regimens employed for non-ECMO patients are adequate to attain target trough levels in patients undergoing ECMO support. This indicates that significant alterations to vancomycin dosing protocols may be unnecessary for patients on ECMO, contesting prior beliefs regarding the necessity for dose modifications in this demographic	Not assessed
Song and Han, 2017[22]	1027 adult patients with ISA infections	67 years (IQR: 54–74 years)	Not assessed	Vancomycin MIC values, infection severity, site of infection	Older age, MRSA infection, bacteremia, recent immunosuppressant use, severe underlying comorbidities, respiratory tract infection, severe sepsis or septic shock, and the presence of metastatic infection		This finding indicates that MIC alone is inadequate as a PD predictor of treatment outcomes in invasive S. aureus infections, underscoring the necessity for combined PK–PD methodologies
Obara et al., 2016[23]	83 patients were included in the final analysis	53–64.5 years	Not assessed	Augmented renal clearance, increased volume of distribution, and altered physiology contributed to nearly 40% of patients failing to achieve recommended therapeutic trough levels	Increased estimated body weight and the occurrence of acute renal damage were substantially correlated with increasing vancomycin trough concentrations	Standard dosage regimens often result in subtherapeutic or excessive drug exposure because of the very varied PKs of vancomycin in intensive care unit patients	The PD evaluation was restricted to trough target achievement, without an assessment of formal PK/PD indicators or exposure–response correlations
Bakke et al., 2017[24]	83 patients receiving vancomycin therapy	>18 years old	Not assessed	Physiological changes linked to critical illness, especially increased renal clearance and variations in hydration status, significantly affected vancomycin PKs despite conventional dosing methods	The existence of increased renal clearance, administration of loading doses, and discrepancies in dosing protocols among institutions	Because of the wide range of vancomycin PKs in critically ill patients, therapeutic trough concentrations were frequently not reached. The primary PK mechanisms responsible for subtherapeutic exposure were increased drug clearance and changed volume of distribution	Not assessed
Wahl et al., 2017[25]	87 patients (94 implantations)	67.2 years (SD 16.7)	Not assessed	The local tissue milieu, wound features, and elution properties of the carrier material were pivotal in influencing medication concentrations	Relevant clinical factors encompassed the nature and severity of bone or soft-tissue infection, the status of surgical debridement, and the quantity of vancomycin incorporated into the calcium sulfate carrier	The local administration of vancomycin by calcium sulphate resulted in elevated drug concentrations at the wound site while maintaining limited systemic exposure, suggesting a beneficial PK profile	Consistent local vancomycin concentrations surpassing standard pathogen MICs indicate effective antibacterial efficacy, notwithstanding the lack of thorough PK/PD studies
Alves et al., 2019[26]	Ten pediatric patients	Median age 5.5 years	Not assessed	Age-related physiological variations, especially developmental alterations in renal function and body composition, markedly affected vancomycin exposure	Age, body weight, dosing regimen, and renal function parameters	The PKs of vancomycin in children exhibit significant variability, and conventional dosage regimens often do not attain therapeutic serum values, especially in younger populations	Not assessed
Westra et al., 2019[27]	Development cohort: 17 HDF patients (159 vancomycin samples) Evaluation cohort: 21 LF-HD patients (132 vancomycin samples)	64.8 years (range 46–85)	Not assessed	The primary non-genetic factors influencing vancomycin exposure were associated with extracorporeal therapy, notably the effect of intermittent online hemodiafiltration on drug clearance, serum concentrations, extraction efficacy, and dialysis scheduling	Patient demographics (age, weight), residual renal function, and dialysis features (kind of dialysis – HDF vs. low-flux HD, extracorporeal blood flow, and session time) were important clinical factors	A two-compartment population PK model incorporating renal, nonrenal, and extracorporeal clearance accurately predicted vancomycin concentrations in HDF and LF-HD patients and supported dosing strategies to achieve target AUC(24 h) ≥400 mg·h/L	Not assessed
Kabbara et al., 2018[28]	93 patients	2 months to 95 years	Not assessed	Actual body weight and renal function	Therapy indication (empiric vs. cultureguided), dosing appropriateness, and monitoring practices	The research findings revealed a significant prevalence of improper vancomycin administration. This misuse was primarily due to inaccurate dosage calculations and insufficient monitoring of trough levels in patients	Not assessed
Mahmoud et al., 2020[29]	Eight adult patients	Median age of 58.5 years	Not assessed	Total body weight and renal function	Infection sites and dosing regimen; however, population modeling highlighted weight and creatinine clearance	Weight and renal function altered vancomycin PK by affecting volume of distribution and clearance, leading to variability in predicted exposure (AUC) and influencing target attainment	Not assessed
Downes et al., 2020[30]	20 critically ill children	Aged 2–18 years receiving intravenous vancomycin	Not assessed	Body weight (allometrically scaled) and renal function	Estimated glomerular filtration rate and patient sex	In critically ill pediatric patients, equations incorporating cystatin C measurements provide more accurate vancomycin clearance estimates than the Schwartz bedside equation. Including cystatin C assessments can enhance the precision of vancomycin dosing regimens for this specific patient group	Not assessed
Lin et al., 2021[31]	466 vancomycin- treated ICU patients	62 years (range 18–93 years)	Not assessed	Total body weight and age	Serum creatinine, CRRT, severe burn status, and dopamine co-administration	Nongenetic factors (body weight, age) and clinical covariates (serum creatinine, CRRT, burn status, dopamine use) significantly influenced vancomycin PKs in ICU patients. Integration of these variables into a population PK model improved therapeutic concentration attainment, supporting optimized therapeutic response	Not assessed
Kim et al., 2021[32]	A total of 280 non-critically ill adult patients, 134 in the PKCS group, 146 in the non-PKCS group	63 years (IQR 49–72 years)	Not assessed	Age, sex, body weight, and baseline renal function (serum creatinine and estimated GFR)	Concomitant nephrotoxic agents, site of infection, target trough range set by clinician, and duration of therapy	Nongenetic variables such as age, sex, body weight, and renal function impacted the distribution and clearance of vancomycin, thereby influencing trough concentrations. Clinical variables, especially dosing guided by PK consultation services, markedly enhanced target trough attainment versus to conventional care, signifying optimum PK exposure	Not assessed
Hirai et al., 2021[33]	208 adult patients (>15 years old) who received intravenous vancomycin	(>15 years old)	Not assessed	Age, body weight, baseline renal function	Renal hypoperfusion medications independently increased trough levels	The administration of renal hypoperfusion medicines correlated with elevated trough concentrations, indicating diminished vancomycin clearance attributable to decreased renal perfusion rather than solely intrinsic renal impairment. The results emphasize renal perfusion as a significant PK variable affecting vancomycin exposure	Not assessed
Wang et al., 2022[13]	26 patients	56.9±19.3 years	Vancomycin-induced DRESS in the Han Chinese population correlates with particular HLA alleles (HLA-A32:01, HLA-B07:05, HLA-B40:06, and HLA-B67:01)	Vancomycin exposure itself, with a mean daily dose comparable to tolerant controls, and concurrent use of other medications in a subset of patients were nongenomic factors; however, in vitro lymphocyte activation testing confirmed vancomycin as the primary causative agent rather than co-administered drugs	Hepatic injury (ALT >3 ×ULN in 50% of patients) and acute renal failure (38.5%), as well as hematologic abnormalities such as eosinophilia≥500/µL (65.4%) and atypical lymphocytosis (26.9%)	The vancomycin dosage was similar in DRESS patients and tolerant controls; DRESS development transpired following extended exposure (14–62 days), with renal impairment observed in 38.5%, perhaps influencing drug clearance	Not assessed
Ishigo et al., 2024[34]	146 patients	67 years (IQR: 56–78 years)	Not assessed	Vancomycin exposure	Critical illness status, ICU admission, and baseline renal function	Vancomycin AUC24–48 h was categorized into <500, 500–600, and ≥600 µg·h/mL. Higher AUC was associated with a higher incidence of AKI (6.5% → 28.0% → 42.9%). Bayesian AUC estimation was used for accurate TDM	The elevated risk of AKI associated with an increasing AUC indicates exposure-related nephrotoxicity. AUC limits of 500–600 µg·h/mL present a risk of nephrotoxicity, indicating that a balance between safety and efficacy is essential in critically ill patients
Ahmed et al., 2024[8]	99 adults	65 years (IQR: 50–75 years)	Race may affect drug PKs by means of factors such as genetic variances	Vancomycin exposure was influenced by dosing regimen and AUC-guided therapeutic drug monitoring. Model-based dose optimization using Monte Carlo simulations improved attainment of the target AUC24–48 (400–600 mg·h/L), highlighting the impact of dosing strategy on PK/PD target attainment	CLcr was the principal clinical covariate influencing vancomycin PKs, with a decrease in CLcr considerably diminishing vancomycin clearance. Additional variables, such as age, sex, serum albumin, and blood urea nitrogen, did not exhibit a significant effect on clearance or volume of distribution	Non-genetic variables, including the dosing regimen and AUC-guided therapeutic drug monitoring, directly affected vancomycin PKs by affecting systemic drug exposure (AUC24–48)	Model-based dose optimization modified vancomycin exposure without affecting intrinsic clearance processes, thus enhancing PD target achievement (AUC/MIC 400–600) and balancing efficacy with the risk of nephrotoxicity

ECMO=Extracorporeal membrane oxygenation, MIC=Minimum inhibitory concentration, IQR=Interquartile range, MRSA=Methicillin-resistant Staphylococcus aureus, HDF=Hemodiafiltration, LF-HD=Low-flux hemodialysis, PK/PD=Pharmacokinetic/pharmacodynamic, AUC=Area under the curve, CRRT=Continuous renal replacement therapy, ICU=Intensive care unit, eGFR=Estimated glomerular filtration rate, ALT=Alanine aminotransferase, ULN=Upper limit of normal, AKI=Acute kidney injury, CLcr=Creatinine clearance, TDM=Therapeutic drug monitoring, PKCS=Pharmacokinetic computer software, GWAS=Genome-wide association studies, BMI=Body mass index, AUC/MIC=Area under the concentration-time curve to minimum inhibitory concentration, DRESS=Drug response with eosinophilia and systemic symptoms

On the other hand, Ahmed et al. did not explicitly report the impact of specific genes on the PKs and PDs of vancomycin. Their study exclusively addressed the population PKs of a particular racial group, specifically in Sudan. Ahmed et al. indicated that population PK modeling tailored to the Sudanese ethnic group with renal impairment can be utilized to ascertain the initial vancomycin dose that optimizes the probability of a positive treatment outcome.[8]

The influence of covariates on the pharmacokinetics and pharmacogenomics of vancomycin

This article has reviewed 22 references that describe additional factors that affect the PKs–PDs of vancomycin [Table 2].

Dosage regimen

Six papers highlight the influence of dosage regimen on vancomycin PKs and PDs: Park et al.,[19] Romanelli et al.,[20] Obara et al.,[23] Westra et al.,[27] Kabara et al.,[28] and Kim et al.[32] Park et al. indicated that the current dosing regimen utilized in our ICU (15–20 mg/kg/dose every 8–12 h) was insufficient for achieving the target trough during the initial phase for the majority of patients undergoing ECMO, despite the elimination rate constant for vancomycin conforming to the population parameters of non-ECMO patients. Vigilant observation of the therapeutic trough concentration during the initial phase is crucial for critically ill patients on ECMO.[19] Romanelli et al. reported that vancomycin exhibits inconsistent serum levels, rendering empirical dosage inadequate for accurately predicting target concentrations. They advocate higher initial dosages and underscore the importance of therapeutic drug monitoring in newborns.[20] Obara et al. concluded that a significant portion (over 40%) of patients failed to achieve therapeutic initial vancomycin trough concentrations. While monitoring trough levels helped reduce subtherapeutic levels, it also increased supratherapeutic levels over time. This suggests that more frequent monitoring may be necessary to maintain optimal drug concentrations. Further research is needed to determine the most effective dosing strategies for patients in intensive care units. For critically ill patients, the administration of vancomycin without monitoring trough levels is not recommended and should be avoided.[23] Another study was reported by Westra et al. They state that the population PK model for vancomycin, initially designed for patients undergoing hemodiafiltration (HDF), has proven to be equally effective in predicting drug levels for those receiving low-flux hemodialysis (LF-HD). A preliminary dosing protocol has been established, considering the patient’s weight and the rate of extracorporeal blood flow. However, additional studies are required to validate this dosing regimen prospectively and to examine other potential influencing factors.[27] Kabbara revealed widespread misuse of vancomycin, primarily due to inaccurate dosage administration and insufficient monitoring of trough concentrations. To address these issues, it is advised that measures be implemented to enhance the quality of vancomycin prescription practices and improve the regularity and accuracy of drug-level monitoring.[28] Kim et al. conclude that for patients who are not critically ill, utilizing PK computer software (PKCS) to guide vancomycin dosing improved the likelihood of achieving target trough levels. However, this approach did not result in fewer treatment failures, shorter treatment periods, or reduced kidney-related side effects. Given the resources required to implement such a system, developing more specific criteria for determining which patients would benefit most from personalized PK-guided dosing may be necessary.[39]

Age, weight, gender, serum creatinine, and kidney function

According to the findings of Hahn et al., vancomycin PKs were predominantly affected by two patient variables: Body weight and renal function. The drug’s CL from the body was most precisely measured by combining the patient’s body weight and serum creatinine levels. The volume of distribution for vancomycin was primarily associated with the patient’s body weight alone.[17] According to the findings of Moore et al., the factors that significantly affected vancomycin PKs included age, body weight, gender, serum creatinine, creatinine CL (CLcr), and the application of renal replacement therapy.[21] Based on Alves et al.’s findings, vancomycin PKs in pediatric patients are substantially influenced by two key factors: Obesity and kidney function. These patient characteristics significantly determine how vancomycin is processed and eliminated in children’s bodies.[26] In addition, according to Ahmed et al.’s findings, CLcr emerged as a consistently significant covariate influencing vancomycin CL across multiple population PK models in previous studies. This highlights the crucial role of renal function in determining vancomycin elimination rates.[8] A recently developed comprehensive covariate model for vancomycin, incorporating the CKD-EPI method for estimating renal function, may provide appropriate dosing adjustments for patients receiving high-dose or prolonged vancomycin treatment, particularly in cases where kidney function is fluctuating.[8,38,39]

Others covariates

The source of infection has been identified as an additional factor affecting vancomycin’s PK and PD profile. Generally, achieving an area under the curve to minimum inhibitory concentration (AUC/MIC) ratio of at least 650 was correlated with enhanced clinical results across various infection sources.[16] Critical sickness is a covariate that affects the PKs and PDs of vancomycin. Fewer than 40% of critically sick patients attained therapeutic trough serum concentrations within the initial 3 days of therapy.[24] Several additional factors can impact vancomycin PKs and dosing. The use of loop or thiazide diuretics may result in renal ischemia and reduced blood volume, potentially leading to elevated initial vancomycin concentrations. The estimated glomerular filtration rate is crucial in determining vancomycin CL throughout treatment. In the later stages of therapy, both advanced age and low albumin levels can significantly affect vancomycin’s PK profile. Given these influences, healthcare providers should monitor vancomycin trough levels, particularly in patients receiving concurrent loop or thiazide diuretics.[33,40]

Clinical implications

Standard vancomycin dosing regimens often fail to reach therapeutic levels, especially in critically ill patients, neonates, children, and those undergoing extracorporeal or renal replacement therapies.[15,18,19,20,21,22,23,24,25,26,27,29,30,31] Population PK analyses and therapeutic drug monitoring consistently show significant interindividual variability, mainly caused by renal function, physiological changes from critical illness, and extracorporeal drug CL.[8,18,21,27,29,30,31,32,33] Therefore, fixed weight-based dosing alone is often insufficient in everyday clinical practice.[20,23,24,26,28] PK/PD indices, especially the AUC₀–₂₄/MIC ratio, tend to be better predictors of effectiveness and safety than trough levels.[16,22,32,34] Subtherapeutic levels have been linked to treatment failure, while excessive levels raise the risk of nephrotoxicity,[16,24,34] emphasizing the importance of AUC-guided dosing and model-informed dose individualization.[8,13,21,31,34] Evidence on how genetic factors influence vancomycin-related adverse outcomes, such as nephrotoxicity and immune-mediated reactions, is limited.[7,13] Only two studies have directly examined pharmacogenetic associations, which greatly limits the applicability of current findings.[7,13] Overall, vancomycin should be tailored to each patient using clinical factors, PK/PD principles, and therapeutic drug monitoring, while more research incorporating pharmacogenetic data into population PK models is needed to determine their potential clinical value.[7,8,13,31,33]

Risk of bias in studies

There is no significant risk of bias in this study.

The overall certainty of the evidence in this review is low to moderate. This assessment is primarily because the majority of the studies included are cohort studies, which typically provide lower-level evidence than randomized controlled trials or systematic reviews. This review has several limitations. First, the literature search was confined to the PubMed database, potentially excluding pertinent pharmacogenomic studies indexed in alternative databases such as Embase or the Cochrane Library. Second, only 2 of the 23 included articles address genetic aspects; therefore, further studies are strongly needed, particularly to strengthen the discussion of the pharmacogenetic aspects of vancomycin.

Conclusion

Vancomycin exhibits substantial interindividual PK and PD variability, particularly in patients with complex clinical conditions. Standard dosing regimens frequently fail to achieve therapeutic targets, supporting the superiority of AUC₀–₂₄/MIC-guided therapeutic drug monitoring over trough-based approaches. Although pharmacogenomic evidence remains limited, available data suggest that genetic variability predominantly influences PD toxicity rather than vancomycin PKs, and its clinical utility is not yet sufficient for routine implementation.

Publicly available

All articles contained in this manuscript are open access and therefore freely available to the public.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

The publication of this work was funded by the 2024 Regular Fundamental Research Grant from the Ministry of Education, Culture, Research, and Technology, Republic of Indonesia.