Pharmacokinetics of teicoplanin in paediatric patients-A systematic review of current literature.

1
INTRODUCTION
Infections caused by gram‐positive bacteria pose a significant threat to critically ill children.
1
,
2

Teicoplanin (Targocid®), a glycopeptide antibiotic, is widely used to treat such infections, including those caused by Staphylococcus aureus and streptococcal species.
2
Although not registered for prophylactic use, it is frequently used off‐label for surgical or oncology‐related prophylaxis.
3
,
4
Because of its favourable spectrum of activity, efficacy and minimal reported toxicity, its use in children is increasing.
1
,
2

Chemically related to vancomycin, teicoplanin is produced by Actinoplanes teichomycetius and acts by inhibiting bacterial cell wall synthesis.
5
,
6
,
7
Compared with vancomycin, teicoplanin causes fewer adverse events and is better tolerated.
8
It is administered intravenously due to poor oral absorption and has a long distribution half‐life.
9
,
10
In adults, the volume of distribution (Vd) is 0.9–1.6 L/kg at steady state.
9
Teicoplanin binds strongly to serum albumin, with a free fraction of 6%–12%.
8
,
9
Teicoplanin is renally eliminated, making dose adjustment essential in patients with renal impairment to prevent toxic serum trough levels.
11
,
12
,
13
,
14

Understanding teicoplanin pharmacokinetics (PK) and pharmacodynamics (PD) in children is critical for effective treatment.
8
However, paediatric PK/PD data are limited, and most dosing regimens are extrapolated from adult studies (Supporting Information S1).
4
,
8
,
15
,
16
,
17
The recent publication by Mouton et al.
18
provided a comprehensive summary of all reported data on PK of teicoplanin. However, an in‐depth analysis of specific paediatric subpopulations is lacking, whereas PK vary significantly across paediatric subgroups. First, teicoplanin clearance (CL) in children is higher compared with adults, neonates and infants.
19
Second, renal impairment can influence the PK of teicoplanin, as it is exclusively renally cleared.
1
Third, in critically ill children, the PK of teicoplanin can vary due to changes in organ function, which can alter distribution, elimination and interaction with receptors.
20
Finally, paediatric oncology patients have altered drug absorption, distribution, metabolism and excretion, for example, due to chemotherapy or malnutrition.
21
They represent a specific subset of children who are highly susceptible to infections.
22
Adequate dosing of empirical antibacterial therapy is therefore vital.
22
Clinical guidelines recommend target trough concentrations (C
min) of ≥15 mg/L for most gram‐positive infections and ≥20 mg/L for bone, joint or MRSA infections
23
(Supporting Information S1). These parameters are based on adult target levels.
23
Although recent evidence supports AUC/MIC as the preferred PK/PD target, C
min remains a common surrogate.
24
Studies show that target levels are frequently not achieved in paediatric practice, and therapeutic drug monitoring (TDM) is not routinely implemented despite recommendations.
25
,
26

Given the increasing use of teicoplanin in various groups of children,
1
a better understanding of its PK across different subgroups is essential. The aim of this systematic review is therefore to (1) summarize and evaluate the reported PK data and target attainment of teicoplanin in children, (2) identify variations in PK and variables explaining this variation and (3) provide—when reported—dosing recommendations for different subgroups of children.

2
METHODS
2.1
Search strategy
The systematic literature review adhered to the PRISMA guidelines of 2020.
27
For the PRISMA flow diagram, see Figure 1. A systematic literature search of studies published up until 31 December 2024 was conducted using the PubMed and Embase databases. The search terms included a combination of ‘teicoplanin’, ‘child’ and ‘pharmacokinetics’. An information specialist was consulted to ensure the thoroughness and accuracy of the search. The detailed search strategy for both databases is available in Supporting Information S2. The systematic review study protocol can be found in Supporting Information S4.

2.2
Inclusion criteria
Eligible study designs were randomized controlled trials (RCTs), non‐randomized controlled trials or prospective/retrospective cohort studies. Studies were included if they assessed at least one PK parameter or teicoplanin exposure in children. Both intramuscular and intravenous dosage formulations were included. Only articles written in English were included. For studies that reported both paediatric and adult data, inclusion was dependent on the ability to extract stratified paediatric data.

2.3
Exclusion criteria
Studies were excluded if the study design was a review, a conference abstract, a letter to the editor, an animal study or a case report. As systematic reviews aim to provide evidence that is applicable to broader groups of patients, case reports were not considered relevant for this systematic review.

2.4
Study selection
The literature search was conducted with an information specialist and duplicate articles were removed. Subsequently, titles and abstracts were independently screened by two authors (N.W. and P.M.). Discrepancies in article selection were discussed. In the case of persistent disagreement, a third researcher (D.T.) was consulted. Full‐texts were screened to ascertain their eligibility for inclusion in the review (N.W. and P.M.).

2.5
Quality assessment
The methodological quality of each included article was assessed using the ClinPK checklist for PK studies.
28
Two authors (N.W. and P.M.) independently assessed the quality of each article. Discrepancies were solved via discussion. An overall summary of the scores is presented in Supporting Information S3.

2.6
Data extraction
Data extraction was performed by N.W. and checked by P.M. for all included studies. In the case of disagreement, a third author (D.T.) was consulted. A Microsoft Word table was used which included the following information: study design, PK analysis method, sample size, dose and formulation, clinical condition, age, weight, covariates within the population PK model, PK parameters (e.g., C
max, C
min, t
max, area under the curve [AUC], t
1/2, CL, Vd and Q), PTA and dose advice. For studies reporting unbound PK parameters, we reported these values separately from total concentrations. Data on toxicity and clinical efficacy were collected when described. In Tables 1, 2, 3, 4 and 5, the results are presented in their original units, and the PK parameters in the figures were converted to a consistent unit. For example, CL values originally reported in L/h were converted to L/kg/h by dividing by the mean or median body weight reported in each study.

2.7
Definitions
Different subgroups were defined as follows: Neonates are aged up until and with 28 days
29
; infants are >28 days and <1 year of age; children are ≥1 year of age. Renal impairment refers to children with mild (eGFR <90 mL/min/1.73 m2), moderate (eGFR <60 mL/min/1.73 m2) or severe (eGFR <30 mL/min/1.73 m2) reductions in kidney function.
30
Augmented kidney function is defined as an eGFR >130 mL/min/1.73m2.
31
Critically ill children are those necessitating high or intensive care.
17
Finally, paediatric oncology patients are defined as children with a diagnosis of cancer who are undergoing treatment.

3
RESULTS
3.1
Characteristics of included studies and study population
The systematic literature search in PubMed and Embase yielded 312 articles (Figure 1). After screening, 26 studies were considered eligible. Data extraction was performed on these 26 full‐text articles. The patient characteristics and PK parameters of the studies are presented in Tables 1, 2, 3, 4 and 5. The number of participants in the different studies ranged from 5 to 280. The mean age ranged from 28 weeks postmenstrual age (PMA) to 9.9 years. The mean body weight ranged from 1130 g to 24.2 kg. Indications for teicoplanin included prophylactic use before surgery and treatment of infections such as MRSA.

3.2
Quality assessment
The quality of the 26 studies included in the final analysis was assessed using the ClinPK checklist and is summarized in Supporting Information S3. The average quality score was 86.3%, with a range from 11.5% to 100%. Overall, the majority of studies adequately reported objectives, dosing regimens, sampling strategies and pharmacokinetic modelling methods. The least frequently addressed items were Item 16 (study withdrawals or subjects lost to follow‐up, 11.5%) and Item 17 (quantification of missing or excluded data, 11.5%). Only 57.7% of studies disclosed funding sources (Item 24).

3.3
Neonates and infants
PK differ across paediatric age groups due to developmental changes. In neonates and infants, absorption is affected by a higher gastric pH and immature digestive enzyme activity.
31
Distribution is altered by higher total body water and lower fat stores, resulting in a higher Vd for hydrophilic drugs and a lower Vd for lipophilic drugs.
31
Metabolism of many drugs is slower due to immature enzyme activity.
31
Renal elimination, the major determinant of teicoplanin clearance, is immature at birth (especially in preterm neonates) and increases rapidly over the first few weeks to months of life, typically reaching adult levels by 1 year of age.
31
This might lead to high interindividual variability in the clearance, particularly in the neonatal period.
Six studies examined teicoplanin PK in neonates and infants (n = 173) (see Table 1; NB: the number of neonates/infants in Strenger et al. is unclear; this study is presented in Table 2).
32
,
33
,
34
,
35
,
36
,
37
Ages ranged from a gestational age (GA) of 28.4 to 50 weeks and a postnatal age (PNA) of 5 days to 69 days. Birth weight ranged from 1130 to 4200 g. Most studies investigated treatment of suspected or proven infections
32
,
34
,
35
,
36
,
37
; one study evaluated prophylaxis in very low birthweight infants.
33
Loading doses varied (4–24 mg/kg), as did maintenance doses (2–15 mg/kg/day; Table 1).
32
,
34
,
35
,
36
,
37

PK findings varied among studies. C
max and C
min were reported in three studies and differed according to dosing regimen. Vd ranged from 0.18 L/kg (Kontou et al.) to 0.56 L/kg (Ramos‐Martín et al. 2016), despite similar populations and dosing.
32
,
34
Möller et al. reported a Vd of 0.25–0.31 L/kg, depending on regimen (Figure 3).
33
This variability could be due to differences in the methodologies used to calculate Vd (e.g., noncompartmental analysis vs. population PK models with one, two or three compartments).
CL also differed (0.015 vs. 0.067 L/kg/h in Kontou et al. and Ramos‐Martín et al., respectively; Figure 2).
32
,
34
Although Kontou et al. identified eGFR or serum creatinine as explanatory covariates,
32
Ramos‐Martín et al. found no such correlation, potentially due to the small sample size.
34

PTA was inconsistent, and target trough levels were not always achieved.
32
,
36
One‐third of the patients in Degrauewe et al. failed to reach the target C
min of 10 mg/L, even though treatment rates for gram‐positive septicaemia were 90%.
36
In Kontou et al., attainment was lower in neonates <1 kg, leading the authors to recommend higher maintenance doses of 10–11 mg/kg for smaller neonates.
32
Only 38.8% of patients in Ramos‐Martín et al. achieved the predefined target of C
min > 15 mg/L
34
; in Strenger et al., 13% and 59% of patients did not reach <10 and <20 mg/L, respectively.
37
Yamada et al. showed improved target attainment (>15 mg/L) with maintenance dosing of 6–8 mg/kg/day compared with <6 mg/kg/day (83.3% vs. 15.4%).
35
Finally, in their study on prophylaxis, Möller et al. reported adequate troughs above the minimum inhibitory concentration at steady state (MIC; 8 μg/L for Staphylococcus epidermis in their institution) by Day 4 with a twice‐daily regimen of 3 mg/kg.
33

Overall, these studies demonstrate considerable variability in teicoplanin PK among neonates and infants, mostly related to body weight, renal function and dosing regimen.
32
,
34
Current dosing strategies often fail to achieve target concentrations, varying from a C
min of 10–20 mg/L. This limits the ability to derive dosing recommendations based on these results.
32
,
34
,
35
,
36

3.4
Children aged ≥1 year
Ten studies reported teicoplanin PK in children aged ≥1 year (n = 997). Ages ranged from 0.2 to 17 years and weights from 4.9 to 68.2 kg. Two studies also included patients aged <1 year; however, these data could not be stratified.
38
,
39
Indications included MRSA treatment,
40
,
41
,
42
prophylaxis prior to elective surgery
10
,
43
and management of suspected or proven infections.
2
,
25
,
37
,
38
,
39
Loading doses differed across studies, but a common regimen consisted of 10 mg/kg every 12 h for three loading doses. Maintenance doses ranged from 6 to 16 mg/kg/day (Table 2).
Vd was consistent across several studies, ranging from 0.24 to 0.68 L/kg.
2
,
39
,
42
,
44
However, Vd was notably higher in Terragna et al. (1.25 L/kg) and lower in Zhang et al. (0.08 L/kg).
40
,
43
CL values were also comparable across different studies, ranging from 0.014 to 0.042 L/h/kg (Figures 2 and 3).
2
,
39
,
40
,
43
,
44

Six studies evaluated target attainment, which was highly variable. Ito et al. found that troughs ≥20 mg/L were rarely achieved in Japanese children treated for MRSA,
41
despite similar regimens to Loane et al., who reported 81% of patients treated for suspected sepsis achieved the therapeutic range of 10–30 mg/L, without subtherapeutic levels.
38
The therapeutic range in Loane et al. (starting from 10 mg/L) was much lower compared with that in Ito et al. (≥20 mg/L).
38
,
41
Ramos‐Martín et al. investigated the same dosing regimens in patients with a median age of 4 (4.3) years and found that 54% of the patients had serum levels of <10 mg/L at steady state.
2
Their 2017 study showed that only 30% of the patients achieved 15–60 mg/L with three 25 mg/kg loading doses followed by maintenance doses of 10 mg/kg/day, whereas 13% of the patients reached potentially toxic levels (>60 mg/L). Strenger et al. investigated teicoplanin PK among four different age groups (neonates/infants [<1.0 year], see Section 3.3, toddlers [1.0–5.9 year], children [6.0–11.9 year], adolescents [12–18 year]). Over 70% of all patients did not achieve the target of >20 mg/L during the maintenance dosing regimen. Among toddlers, particularly, target attainment was often not achieved (<10 mg/L in 24.6%, <20 mg/L in 82.6%). These findings were similar to those of adolescents (<10 mg/L in 13.7%, <20 mg/L in 79.5%).
37
Finally, in Yamada et al., 15.4% of the patients (median age 7.3 years) had trough levels of <10 mg/L and 46.2% had trough levels of <15 mg/L at steady state, whereas the target trough level was 15–30 mg/L. Hence, many patients had serum trough levels below the target.
25

Providing dose recommendations is challenging, as it is strongly dependent on the predefined target level. For a target level of >10 mg/L, three 10 mg/kg loading doses every 12 h followed by 10 mg/kg/day appeared adequate.
38
For target levels of >15 mg/L, none of these studies provided adequate dose recommendations to consistently reach this target level.
In conclusion, the PK of teicoplanin in children aged ≥1 year exhibit relatively low variability in Vd and CL. The CL of teicoplanin is significantly higher in children than in adults.
19
Targets varied widely among studies and were often not achieved.

3.5
Children with renal impairment
Changes in kidney function not only impact drug clearance but also absorption, distribution and metabolism. For example, hypoalbuminemia and fluid retention can influence drug distribution. Also, accumulation of uremic toxins can suppress hepatic enzyme activity, leading to altered metabolism.
45
Two of the included studies primarily investigated the PK of teicoplanin in children with differences in kidney function (n = 167).
30
,
46
Loading doses were 3 or 10 mg/kg every 12 h,
46
and maintenance doses were 10 mg/kg/day.
46
In Xu et al., doses of 3, 6 and 10 mg/kg were investigated.
In Gao et al., CL was higher in children with augmented (eGFR >130 mL/min/1.73m2: CL = 0.015 L/h/kg) and normal (eGFR ≥90 mL/min/1.73m2: CL = 0.013 L/h/kg) kidney function, as compared to children with impaired kidney function (eGFR ≤30 mL/min/1.73m2: CL = 0.008 L/h/kg), confirming a positive correlation with eGFR.
46
In the simulation presented by Xu et al., the area under the curve (AUC) values were 1.25‐fold, 1.95‐fold and 2.82‐fold higher in paediatric patients with mild, moderate and severe renal impairment, respectively, than the normal population. Otherwise, there were only minimal variations in C
max (Figures 2 and 5).
Regarding PTA, Xu et al. concluded that standard dosing often failed to achieve therapeutic exposure. Adjusted regimens of 12 mg/kg (normal kidney function), 9.5 mg/kg (mild impairment), 6 mg/kg (moderate impairment) and 4 mg/kg (severe impairment) every 12 h achieved PTAs of >90%.
30
In contrast, Gao et al. did not report on PTA. However, they did provide optimal dosing recommendations to achieve the therapeutic target of 10–30 mg/mL across different categories of kidney function (Table 3).
46

Overall, both studies recommend reducing maintenance doses in renal impairment given the positive relationship between CL and eGFR. For patients with normal kidney function, three loading doses of 12 mg/kg every 12 h, followed by a maintenance dose of 10 mg/kg/day is recommended. For patients with mildly impaired kidney function (eGFR <90 mL/min/1.73 m2), three loading doses of either 9.5 or 12 mg/kg every 12 h, followed by maintenance dose of either 10 or 8 mg/kg/day, respectively, is recommended. For patients with moderately impaired kidney function (eGFR <60 mL/min/1.73 m2), three loading doses of 6 mg/kg every 12 h, followed by a maintenance dose of 5 mg/kg/day is recommended. For patients with severely impaired kidney function (eGFR <30 mL/min/1.73 m2), a loading dose of 4 mg/kg is recommended.
30
,
46

In summary, impaired kidney function correlates with decreased CL and consequently higher serum trough values. Thus, eGFR should be taken into account when tailoring both loading and maintenance dosing regimens for optimal therapeutic outcomes.

3.6
Critically ill children
Inflammation, a key feature of critical illness, impacts drug PK in critically ill children.
48
It can either increase or decrease drug absorption, and it is associated with reduced plasma protein levels, particularly hypoalbuminemia, which reduces protein binding and increases free drug concentrations for highly protein‐bound drugs, such as teicoplanin.
48
Also, inflammation impairs drug metabolism such that metabolism decreases as inflammation increases.
48
Regarding active renal excretion, inflammation may cause reduced clearance. In addition, critical illness is often associated with hemodynamic instability, use of vasoactive agents and fluid resuscitation, which can all impact renal perfusion and glomerular filtration rate. This may result in either decreased or increased renal clearance. In all, it has been suggested that inflammatory biomarkers should be considered in PK‐related clinical trials, not only in children.
48

Six studies examined teicoplanin PK in critically ill children (n = 220). Age ranged from 7 days to 15.6 years and weight ranged from 3.2 to 56 kg. Notably, these studies only included five neonates: Sanchez et al. included a 7‐day‐old neonate (newborn), and Tarral et al. included four neonates.
20
,
47
These data could not be stratified and were therefore not analysed in the ‘Neonates and Infants’ subgroup. PK was investigated in children admitted to the PICU
1
,
44
for various critical conditions such as gram‐positive infections,
19
nosocomial infections,
20
severe burns
49
and severe urinary tract infections.
47
The standard loading dose was three doses of 10 mg/kg every 12 h except in the studies by Steer et al. and Tarral et al., both of whom did not apply a loading dose. Maintenance doses were 6,
47
10,
1
,
19
,
20
12
49
or 15 mg/kg/day
19
(Table 4).
PK parameters varied widely. Lukas et al. reported higher C
max in infants (70.75 mg/L) than in older children (54.9 mg/L). Similarly, infant C
min levels were higher (12.85 mg/L) compared with that of older children (8.45 mg/L), possibly reflecting differences in clearance or protein binding—an aspect that has been underexplored.
19
Future research should include measurement of unbound teicoplanin concentrations, as suggested by the recent publication of Mouton et al.
18
Sanchez et al. also reported on C
max (26.2 mg/L) and C
min (5.8 mg/L) but found no differences between age groups.
20
Dose regimens in Lukas et al. and Sanchez et al. were similar. Additionally, Tarral et al. concluded that the mean Vd was higher in neonates (0.61 L/kg) than in older children (0.54 L/kg). Redistribution from the peripheral to central compartments proceeded more quickly in neonates than in children (Figures 3 and 4).
47

CL values ranged from 0.018 to 0.21 L/kg/h.
1
,
20
,
49
Sanchez et al. and Lukas et al. found no age‐related differences in CL,
19
,
20
whereas Tarral et al. reported a lower CL in neonates (0.016 L/kg/h) than in children (0.028 L/kg/h) (Figure 2).
47
Regarding kidney function, Aulin et al. did not find a significant correlation between eGFR and CL. However, this could be explained by the variation in renal biomarkers having been too narrow in this study and the fact that there were no patients with renal failure included.
1

Target attainment was generally poor. Aulin et al. demonstrated that only 29.4% of the patients achieved the common clinical target C
min > 20 mg/L.
1
Lukas et al. reported subtherapeutic levels (<10 mg/L) in 35% of older children as compared with 8% of infants, due to higher CL in the older group. They concluded that younger infants (<12 months) in the PICU required nearly half the dose of older children (>12 months–10 years) to achieve similar serum concentrations.
19
Sanchez et al. also found that with a daily dosage of 10 mg/kg, most patients (89%) did not achieve the target level of >10 mg/kg at steady state.
20
Steer et al. showed all patients maintained a serum trough level of 4 mg/L for 24 h after a single intravenous dose of 12 mg/kg, though no formal target was defined.
49
Similarly, in Tarral et al., a target level was not defined.
47

Comparisons with adults suggested children with severe burns had higher CL (0.018 vs. 0.010 L/kg/h, respectively), whereas Vd in children was lower (0.69 vs. 1.0 L/kg, respectively) (Figure 2).
49
Aulin et al. investigated the plasma protein binding of teicoplanin in critically ill children. In contrast to adult studies, they did not observe a relationship between albumin levels and teicoplanin levels.
1

Providing dose recommendations based on these studies is challenging because of the significant variability in illness, age and predefined target levels. Nevertheless, the studies indicate that critically ill children generally show a somewhat higher CL and larger Vd, implying a need for higher dosing, although younger children may require dose reductions due to lower clearance. However, this is mainly based on the study by Aulin et al. We note that the methods used to calculate PK parameters varied significantly between studies (e.g., noncompartmental analysis vs. population PK models with one, two or three compartments). Aulin et al. recommended a dosing regimen optimized for critically ill children aiming to achieve concentrations of >20 mg/L—levels often necessary to treat severe infections. The regimen included three loading doses of 10 mg/kg, followed by a daily maintenance dose of 10 mg/kg. However, with this dose regimen, only 29.4% of the patients achieved the target of >20 mg/L. Hence, the loading and maintenance doses appear inadequate, illustrated by the AUC calculation for Days 1, 3 and 5.
1

These studies highlight the complexity associated with administering teicoplanin in critically ill paediatric patients, as therapeutic targets are very often not achieved. Therefore, these studies recommend routine TDM to optimize treatment, after an indication‐dependent target is defined.
1

3.7
Paediatric oncology patients
Altered PK is well established in adult oncology patients, but paediatric data remain limited.
22
Paediatric oncology patients have altered drug absorption due to gastrointestinal changes caused by chemotherapy and malnutrition, for example.
21
Changes in body composition, particularly in patients undergoing long‐term chemotherapy treatment, can influence drug distribution. Protein binding can also be affected. Additionally, metabolism is often altered due to interactions with chemotherapeutic agents. Liver dysfunction due to the disease or chemotherapy‐induced hepatotoxicity can also affect drug metabolism.
50
Finally, renal function may be impacted both acutely and chronically in these patients. Nephrotoxic chemotherapeutic agents (e.g., cisplatin and methotrexate), tumour lysis syndrome and dehydration can all lead to impaired glomerular filtration and tubular damage, thereby reducing renal clearance of teicoplanin. In some cases, supportive care measures or hyperhydration protocols may also lead to increased renal clearance, further complicating PK predictability. Paediatric oncology patients represent a high‐risk population due to their increased susceptibility to infections and frequent antibiotic use.
22
Given the substantial variability in PK in this group, adequate dosing of antibiotics is vital.
Three studies examined teicoplanin PK in paediatric oncology patients with suspected or proven infections (n = 141).
22
,
51
,
52
Age varied widely, with a median age of 7 years (0.5–16.9 years). There were no neonates included in these studies. The loading doses were three doses of 10 mg/kg every 12 h,
22
,
51
or two loading doses of 6, 8 or 10 mg/kg.
52
The maintenance doses were 10
22
,
52
or 20 mg/kg/day
51
(Table 5).
Dufort et al. and Lemerle et al. both reported higher C
max and C
min with increased maintenance dosages, suggesting dose‐dependent distribution and elimination.
51
,
52
In Dufort et al., CL was 0.029 L/kg/h, whereas in Zhao et al., CL was 0.018 L/kg/h (Figures 2 and 4).
22
,
51

Regarding PTA, Dufort et al. showed that higher daily maintenance dosages (20 mg/kg) ensured serum levels above the therapeutic threshold of >10 mg/L,
51
whereas Zhao et al. concluded that 41% of patients receiving the standard maintenance dose of 10 mg/kg/day had C
ss,min < 10 mg/L (Figure 5). This highlights the variability in metabolism and excretion among these patients. Zhao et al. provide dosing recommendations to reach the target AUC of 750 mg/L/h (Table 5).
22
Lemerle et al. suggested a maintenance dose of 10 mg/kg/day as adequate, based on a 100% successful treatment rate of four patients treated with a higher dose compared with a 33% successful treatment rate of those treated with lower doses.
52
However, the sample size in this study (n = 4) is too small to draw conclusions and a target level was not defined.
Providing dose recommendations based on these three studies is challenging, as sample sizes are small.
51
,
52
However, the results of these studies imply that increased dosing regimens are required for paediatric oncology patients, as these patients frequently have higher CL rates and thereby often fail to achieve target attainment levels. It is particularly critical to attain such levels in this population, given their vulnerability to infections.
22
,
51
,
52

4
DISCUSSION
This systematic review aimed to comprehensively evaluate the PK of teicoplanin in various groups of children and to provide—when reported—evidence‐based dosing recommendations. Although considerable PK studies have been conducted, definitive conclusions remain difficult as PK parameters vary widely across paediatric subgroups and therapeutic targets are often unmet. This highlights the necessity for age‐ and condition‐specific dosing strategies to achieve optimal therapeutic outcomes.
Various covariates explain part of this variability. CL was higher in paediatric oncology patients (based on Zhao et al.), likely due to fluid overload during oncology treatments. Critically ill children showed somewhat larger Vd, a trend also observed with other hydrophilic medications. CL correlated positively with eGFR, indicating decreased CL and higher serum trough values in renal impairment. CL was variable in neonates and infants, as the GFR undergoes maturation up to 1 year of age.
1
Although these findings indicate that the PK are influenced by covariates such as kidney function, illness and age, substantial variability remains unexplained.
Therapeutic targets ranged across studies (C
min ≥ 4 to ≥20 mg/L) and were frequently not achieved. Notably, studies reporting on target attainment focused on measurements at steady state. It remains somewhat unclear whether the inadequacies extend to both maintenance and loading doses, as together they define the (fast) achievement of steady state concentrations. The loading phase, which occurs over 36 h, plays a key role, with clearance becoming increasingly important with each dose.
Clinical outcome data were limited, although a few studies described a correlation between achievement of the target level and clinical recovery.
10
,
38
,
52
In contrast, Degrauwe et al. reported that a third of the patients failed to reach the target C
min of 10 mg/L during the maintenance dosing regimen, whereas the microbiologically and clinically successful treatment rates were 90% in gram‐positive septicaemia. Establishing evidence‐based targets is essential when providing dosing recommendations. However, it remains unclear whether the current recommended trough concentrations correlate with clinical outcome.
26
Current guidelines suggest target levels of ≥15 mg/L for most gram‐positive infections and ≥20 mg/L for bone and joint infections, MRSA infections and for intensive care and burn patients.
25
,
26
A recent study by Hanai et al. suggested that an optimal target trough level of 10–20 mg/L ensures clinical efficacy and reduced toxicity in paediatric patients.
53
In an updated guideline for adult patients, target trough concentrations of 15–30 mg/L are recommended for treatment of noncomplicated MRSA infections and 20–40 mg/L for patients with serious and/or complicated MRSA infections.
24

Most studies used C
min/MIC as targets, whereas recent evidence suggests that C
min/MIC is not as adequate of a correlate for AUC/MIC as previously thought. AUC/MIC is considered the key PK/PD parameter for teicoplanin. However, trough concentrations can be used as surrogate markers in clinical settings where AUC estimation software is unavailable, as highlighted by Hanai et al.
24
Also, unlike vancomycin, teicoplanin has a much longer half‐life, which reduces the variability in the relationship between AUC and C
min. This makes C
min a more reliable predictor of AUC for teicoplanin compared with vancomycin, including in paediatric patients where half‐life remains prolonged.
All target concentrations reported in the included studies refer to total plasma teicoplanin levels; none of the targets were age‐specific. Given teicoplanin's high protein binding, total concentrations may not consistently reflect the pharmacologically active fraction, particularly in paediatric patients with hypoalbuminemia or other conditions altering protein binding. The recently published review by Mouton et al.
24
highlighted that monitoring unbound teicoplanin concentrations could improve assessment of target attainment and dose optimization, as unbound exposure (e.g., fAUC/MIC) is more directly linked to antibacterial activity. However, current paediatric data on unbound teicoplanin are scarce, and only one included study (Aulin et al.
1
) reported unbound PK parameters. Future research should prioritize evaluating the role of unbound teicoplanin monitoring, next to defining age‐specific targets.
The variability in teicoplanin PK across paediatric populations presents a significant challenge for establishing universal dosing recommendations. Targets vary widely across studies, using C
min/MIC as well as AUC/MIC. Therefore, we recommend conducting a population PK meta‐analysis, requiring access to raw data from all included studies, which was beyond the scope of the current work. This meta‐analysis should apply indication‐dependent targets, defined through multidisciplinary consensus. This approach would help to establish evidence‐based, optimized dosing recommendations for paediatric patients.
When such indication‐dependent targets are set, TDM should be considered a part of treatment, as it can improve clinical efficacy and prevent toxic serum trough levels (>60 mg/L), though robust clinical toxicity thresholds remain scarce.
14
,
23
,
24
According to teicoplanin product guidelines, TDM is advised at the start of treatment, after steady state attainment (Days 4–5
2
) and weekly during treatment.
26
The notion that teicoplanin has fewer dosing difficulties than vancomycin cannot be supported, as PK of teicoplanin is interindividual highly variable and TDM for teicoplanin is not as universally implemented as for vancomycin. As a consequence, subtherapeutic levels will be left unidentified.
Prophylactic use of teicoplanin was investigated in three studies.
10
,
33
,
43
Of these, only Möller reported on target attainment (MIC: 8 μg/L for Staphylococcus epidermis). Aarons et al. concluded that none of the children who received teicoplanin prophylactically after surgical removal of the tonsils developed an infection, although they did not report on targets.
10
Evidence in paediatric oncology patients (Boztug et al.) suggests potential benefit of prophylactic teicoplanin in this vulnerable group.
3
The currently ongoing Pro‐Teico study, an open‐label, randomized clinical trial, is investigating the prophylactic efficacy of teicoplanin in paediatric AML patients (EudraCT Number: 2020‐000508‐13). This study is also investigating PK among different age groups, aiming to establish optimal prophylactic dosing regimens.
This systematic review has limitations. First, we primarily focused on PK parameters and PTA, rather than clinical efficacy or toxicity, limiting direct clinical interpretation. Nonetheless, we have collected these data from the PK studies that assessed clinical efficacy and safety in Table 1. Future studies should investigate the correlation between recommended trough levels and clinical outcomes.
26
Second, study designs varied, including noncompartmental analysis and population PK modelling. A future meta‐analysis could harmonize these data. This would allow for the model‐based development of both loading and maintenance doses, which should then be evaluated prospectively for clinical outcomes.
In conclusion, teicoplanin PK in children is highly variable, with targets often not achieved, making the development of universal dosing recommendations challenging. Future studies should define indication‐dependent targets. Consequently, TDM should be incorporated to improve efficacy and prevent toxic serum trough levels. Prophylactic use in oncology patients warrants further study. Defining unbound exposure and its clinical correlates should be a priority for future research.
4.1
Nomenclature of targets and ligands
Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org and are permanently archived in the Concise Guide to PHARMACOLOGY 2021/22.
54

AUTHOR CONTRIBUTIONS

Noa E. Wijnen: Conceptualization; investigation; methodology; writing. Daan J. Touw: Investigation; writing—review and editing. Kim Klein: Writing—review and editing. Gertjan J. L. Kaspers: Writing—review and editing. Paola Mian: Conceptualization; investigation; methodology; writing—review and editing.

CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.

Supporting information

Data S1 Dosing recommendations for treatment of infections with teicoplanin in children.

Data S2 Overview of the final search strategy in PubMed and Embase.

Data S3: Quality assessment.

Data S4: Protocol template: Systematic review.