If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author at: Department of Clinical Research, NHO, National Epilepsy Center, Shizuoka Institute of Epilepsy and Neurological Disorders, 886 Urushiyama, Shizuoka 420-8688, Japan.
Department of Clinical Research, NHO, National Epilepsy Center, Shizuoka Institute of Epilepsy and Neurological Disorders, 886 Urushiyama, Shizuoka 420-8688, JapanDepartment of Clinical Pharmaceutics, Graduate School of Pharmaceutical Sciences, University of Shizuoka, 52-1 Yada, Shizuoka 422-8526, Japan
Department of Clinical Research, NHO, National Epilepsy Center, Shizuoka Institute of Epilepsy and Neurological Disorders, 886 Urushiyama, Shizuoka 420-8688, Japan
Department of Clinical Research, NHO, National Epilepsy Center, Shizuoka Institute of Epilepsy and Neurological Disorders, 886 Urushiyama, Shizuoka 420-8688, Japan
Department of Clinical Research, NHO, National Epilepsy Center, Shizuoka Institute of Epilepsy and Neurological Disorders, 886 Urushiyama, Shizuoka 420-8688, Japan
Department of Clinical Pharmaceutics, Graduate School of Pharmaceutical Sciences, University of Shizuoka, 52-1 Yada, Shizuoka 422-8526, JapanLaboratory of Clinical Pharmacokinetics and Drug Safety, Shizuoka General Hospital, 4-27-1 Kita Ando, Shizuoka 420-8527, Japan
Department of Clinical Research, NHO, National Epilepsy Center, Shizuoka Institute of Epilepsy and Neurological Disorders, 886 Urushiyama, Shizuoka 420-8688, JapanDepartment of Clinical Pharmaceutics, Graduate School of Pharmaceutical Sciences, University of Shizuoka, 52-1 Yada, Shizuoka 422-8526, Japan
To identify pediatric patients who require therapeutic drug monitoring (TDM) of levetiracetam (LEV).
Methods
We retrospectively investigated 2413 routine therapeutic drug monitoring data on serum LEV concentration from 1398 pediatric patients (age, 0–15 years). Samples were grouped by age (infants, < 1 year; preschool children, 1–5 years; primary school children, 6–11 years; and adolescents, 12–15 years), and the LEV concentration-to-dose (CD) ratio was calculated.
Results
The mean CD ratio was highest in adolescents (analysis of variance, p < 0.001); 22.5 % and 15.7 % higher in adolescents than in preschool children and school children, respectively (Scheffé test, p < 0.001); and higher in infants than in preschool children. Preschool children had the lowest ratio and tended to show an increase in the ratio from age 2 to 5 years. Use of enzyme-inducing antiseizure medication reduced the CD ratio by 6.1 % in infants, 12.2 % in preschool children, 5.9 % in primary school children, and 9.4 % in adolescents. The mean CD ratio was 2.7 %, 26.9 %, and 39.3 % higher in preschool children, primary school children, and adolescents with defined chronic kidney disease (CKD) than in the respective age group of patients without CKD. The therapeutic concentration range for a long-term LEV therapy was 11 to 32 μg/mL.
Conclusions
LEV pharmacokinetics are significantly different between infant and preschool children, so TDM of LEV is clinically useful in these patients. In pediatric patients at higher risk for CKD, glomerular filtration rate and LEV levels should be carefully monitored.
The second-generation antiseizure medication (ASM) levetiracetam (LEV) is widely used in monotherapy and as an adjunctive treatment in both pediatric and adult epilepsy patients. It is rapidly and completely absorbed after oral administration and has a high oral bioavailability (greater than95 %). Protein binding of LEV is clinically insignificant (10 %), and LEV exhibits linear and dose proportional pharmacokinetics [
]. Besides having predictable pharmacokinetics, LEV has a wide therapeutic range and good tolerability, so routine therapeutic drug monitoring (TDM) for LEV is not required [
LEV is mainly excreted unchanged in urine (more than 60 %), so adults with a creatinine clearance of less than 60 mL/min require a 50 % dose reduction [
]. Hepatic cytochrome P450 enzymes are not involved in the metabolism of LEV, but the drug is partly metabolized by type B esterase enzyme in whole blood. Previous studies showed that enzyme-inducing ASMs (e.g., phenytoin, carbamazepine, and phenobarbital) can reduce the LEV concentration by 10 % to 30 % [
Prospective assessment of levetiracetam pharmacokinetics during dose escalation in 4- to 12-year-old children with partial-onset seizures on concomitant carbamazepine or valproate.
Pharmacokinetic variability of four newer antiepileptic drugs, lamotrigine, levetiracetam, oxcarbazepine, and topiramate: a comparison of the impact of age and comedication.
Generally, drug metabolism differs between children and adults. Renal excretion is about fivefold lower in neonates than in adults, but glomerular filtration rates markedly increase after birth and reach adult levels by about age two years. Consequently, pediatric patients may benefit from TDM when using LEV [
]. Several previous studies demonstrated that pediatric patients have a high LEV clearance and require a 30 % to 40 % higher dose than adults to achieve similar concentrations [
Prospective assessment of levetiracetam pharmacokinetics during dose escalation in 4- to 12-year-old children with partial-onset seizures on concomitant carbamazepine or valproate.
Pharmacokinetic variability of four newer antiepileptic drugs, lamotrigine, levetiracetam, oxcarbazepine, and topiramate: a comparison of the impact of age and comedication.
]. However, most of these studies evaluated children older than 4 years, so data on younger patients are limited. In addition, the pharmacokinetics of LEV in pediatric patients with renal impairment is incompletely understood. Previous studies reported the effects of inducers on LEV concentration, but it remains unclear whether inducers and their combinations have different effects on LEV pharmacokinetics. Therefore, the present study evaluated routine TDM data on LEV with the aims to explore factors that influence the serum concentration, such as age, sex, and renal function, and to compare interactions between LEV and concomitant ASMs in pediatric patients stratified by age. Furthermore, we identified the therapeutic concentration range for LEV in pediatric patients receiving long-term therapy.
2. Materials and Methods
2.1 Participants and assays
The study was approved by ethics review board of the National Epilepsy Center (Shizuoka, Japan, approval number 2016–28). We retrospectively reviewed routine TDM results from January 2011 to December 2020 and obtained 8013 measurements of serum LEV concentration from 1619 pediatric patients. In addition, we evaluated the following clinical information: age, sex, body weight, ASM dose and concentration, and laboratory data. None of the patients had severe liver dysfunction (serum alanine transaminase or aspartate transaminase level greater than 200 U/L). In this study, we evaluated the serum samples collected 2 to 3 h after administration of LEV (a peak sampling). Also, we enrolled pediatric patients who received two doses of LEV evenly distributed between morning and evening.
We defined chronic kidney disease (CKD) by using the serum creatinine level specified in the Japan Committee of Measures for Pediatric CKD of the Japanese Society of Pediatric Nephrology [
Pediatric CKD Study Group; Japan committee of measures for pediatric CKD of the Japanese Society of Pediatric Nephrology. Pre-dialysis chronic kidney disease in children: results of a nationwide survey in Japan.
]. For example, in patients aged 3 years, CKD was defined as a serum creatinine level greater than or equal to 0.37 mg/dL (see supplemental file, Table S1).
The serum concentration was assayed by liquid chromatography with tandem mass spectrometry with a reverse-phase column, as described elsewhere [
]. Because the detection limit of LEV was 1 μg/mL, serum samples with levels below 1 μg/mL were excluded. In multiple dose-ranging studies, steady state concentrations of LEV were generally reached after 2 days [
]. In this study, we defined the steady state concentration as the concentration reached after a patient had received the same ASM and LEV regimens for at least 14 days. If multiple samples were obtained from one patient at different ages during the study period, we assessed two or more samples at each age. In all patients with two or more serum samples at a particular age, we used the concentration at the highest LEV dose (mg/kg/day). In total, we investigated 2413 measurements of serum LEV concentration in 1398 pediatric patients (focal epilepsy, n = 905; generalized epilepsy, n = 265; both focal and generalized epilepsy or unknown, n = 229). In this report, phenytoin, carbamazepine, and phenobarbital were studied as enzyme-inducing ASMs (inducers); the weak inducers oxcarbazepine and eslicarbazepine were not included because these drugs are not approved in Japan.
2.2 Therapeutic concentration range
To evaluate the therapeutic concentration range for LEV, we selected patients according to the following criteria [
]: (1) the clinical response was partially or completely achieved (>50 % seizure frequency reduction (lasting more than 3 months)) and (2) the patients had good tolerance and remained on LEV therapy for more than three years. When multiple samples were available from a patient had good clinical response, the sample with the highest LEV concentration was used.
2.3 Statistical methods
We examined LEV pharmacokinetics by calculating the concentration-to-dose (CD) ratio, i.e., the serum concentration (μg/mL) divided by the dose per body weight (mg/kg). To evaluate the influence of age, we divided patients into four age groups: infants, < 1 year old; preschool children, 1–5 years; primary school children, 6–11 years; and adolescents, 12–15 years.
The Kolmogorov-Smirnov test was used to determine whether each group followed a normal distribution. The significance of differences between the two groups was determined by the unpaired t test (Gaussian distribution) or Mann-Whitney U test (non-Gaussian distribution or sample size of<30). Correlations between continuous variables were tested by Spearman’s correlation coefficient analysis. For multi-group comparisons, analysis of variance (ANOVA) was performed with a post-hoc Scheffé multiple comparison test. Adjusted CD ratio means were evaluated by analysis of covariance (ANCOVA), and the significance of intergroup differences was compared by a Bonferroni test. Data were expressed as means and 95 % confidence intervals. The level of significance was set at a p value of<0.05. Statistical analyses were performed with SPSS Statistics Version 25 (IBM Corp, Armonk, NY, US).
3. Results
3.1 Patient characteristics
Table 1 shows the patient characteristics. Other than the three ASMs mentioned above, patients were being treated with the following ASMs: ethosuximide (n = 93), topiramate (n = 212), rufinamide (n = 27), stiripentol (n = 21), perampanel (n = 58), and lacosamide (n = 37). Inducers were concomitantly used in a higher percentage of patients in the infant group than in the other age groups. Among the four age groups, the adolescent group had the highest CD ratio of LEV (ANOVA, p < 0.001); the mean CD ratio in adolescents was 22.5 % and 15.7 % higher than in preschool children and school children, respectively (Scheffé test, p < 0.001).
Table 1Patient characteristics.
Age group
Infants (aged < 1 year)
Preschool children (aged 1–5 years)
School children (aged 6–11 years)
Adolescents (aged 12–15 years)
Total, n
53
769
963
628
Age, mean (95 % CI), mo or y
7.9 mo (7.1–8.7 mo)
3.2 y (3.1–3.3 y)
8.5 y (8.3–8.6 y)
13.5 y (13.4–13.6 y)
Sex, male/female
23/30
405/364
542/421
339/289
Body weight, mean (95 % CI), kg
8.0 (7.6–8.5)
13.9 (13.6–14.1)
25.8 (25.3–26.4)
45.6 (44.7–46.6)
Levetiracetam
Dose, mean (95 % CI), mg/day/kg
36.0 (31.2–40.8)
36.4 (35.1–37.7)
29.1 (27.9–30.2)
25.2 (24.0–26.4)
Concentration, mean (95 % CI), μg/mL
24.1 (20.3–27.9)
22.4 (21.5–23.3)
19.2 (18.4–20.0)
19.0 (18.0–20.0)
CD ratio, mean (95 % CI), ([μg/mL]/[mg/kg])
0.68 (0.62–0.74)
0.63 (0.61–0.64)
0.66 (0.65–0.68)
0.77 (0.74–0.79)
Concomitant ASMs
Number of ASMs, mean (95 % CI)
1.5 (1.3–1.7)
1.6 (1.5–1.7)
1.5 (1.4–1.5)
1.4 (1.3–1.5)
Inducers, n (%)
23 (43.4 %)
190 (24.7 %)
229 (23.8 %)
181 (28.8 %)
Phenytoin, n
0
19
50
47
Carbamazepine, n
8
116
151
124
Phenobarbital, n
16
63
37
23
Valproate, n
24
410
484
275
Zonisamide, n
14
111
76
48
Clobazam, n
7
109
102
79
Lamotrigine, n
2
138
140
98
ASM, antiseizure medications; CI, confidence interval; CD ratio, concentration-to-dose ratio.
Significance was determined by analysis of variance or the χ2 test.
Among the 0- to 5-year-olds, infants (aged < 1 year) had the highest mean CD ratio, and the ratio in infants was significantly higher than in the 2-year-old children (0.69 vs 0.58, Bonferroni test, p < 0.05; see supplemental file, Table S2). The mean CD ratio tended to increase from age 2 to 5 years. Adjusting the mean ratio for sex and use of inducers did not affect the results.
Fig. 1 shows the correlation between the body weight-adjusted dose and the serum concentration of LEV in the four age groups. In all age groups, the serum level of LEV showed a strong correlation with dose. Among the four age groups, the slope in adolescents was the highest, followed in order by infants, school children, and preschool children.
Fig. 1Correlation between levetiracetam dose and serum concentration in pediatric patients with epilepsy r, correlation coefficient.
Fig. 2 shows the effects of the three inducers on the CD ratios of LEV in the four age groups. Regardless of age, concomitant use of inducers tended to reduce LEV concentrations, i.e., inducers reduced the LEV CD ratio by 6.1 % in infants, 12.2 % in preschool children, 5.9 % in primary school children, and 9.4 % in adolescents. Changes were significant in all groups except the infant group, where no significant result was obtained because of the wide confidence intervals.
Fig. 2Comparison of the mean concentration-to-dose ratio of levetiracetam in pediatric patients stratified by age. The columns show the means, and the bars indicate the 95 % confidence intervals. Unpaired t test, *p < 0.001; Mann-Whitney U test, **p < 0.001, CD ratio, concentration-to-dose ratio.
Table 2 compares the CD ratio of LEV in patients receiving different inducer regimens. After the CD ratio was adjusted by age, sex, and body weight, the adjusted mean CD ratio was significantly higher in patients not taking a concomitant inducer. Concomitant use of the inducers phenytoin, carbamazepine, and phenobarbital reduced the serum LEV concentration by 11.3 %, 8.3 %, and 9.1 %, respectively, but the differences from the phenobarbital group were not statistically significant (Bonferroni test, p = 0.057). Also, the group with multiple inducers had the widest confidence interval because of the small sample size and was not significantly different from the group without inducers. Phenytoin, carbamazepine, phenobarbital, and their combination all had a similar effect on LEV pharmacokinetics, and this effect was smaller than that of age.
Table 2Effect of enzyme inducers on the concentration-to-dose ratio of levetiracetam in pediatric patients aged 0–15 years.
In this study, 63 samples were obtained from 58 pediatric patients with CKD; none of the infant patients had CKD, and CKD was stage 2 in all patients with the disease. Fig. 3 shows the relationship between kidney function and the LEV CD ratio in the three age groups with CKD. The mean CD ratios in pediatric patients with CKD were as follows: preschool children, 0.64 (95 % CI, 0.57–0.72); primary school children, 0.84 (95 % CI, 0.75–0.92); and adolescents, 1.07 (95 % CI, 0.84–1.29). The mean CD ratio was 2.7 %, 26.9 %, and 39.3 % higher in preschool children, primary school children, and adolescents with CKD than in the respective age group of patients without CKD. The differences between patients with and without CKD were significant only in school children and adolescents, indicating that kidney function affected the LEV CD ratio only in pediatric patients aged 6 years and older.
Fig. 3Comparison of the concentration-to-dose ratio of levetiracetam in pediatric patients with and without chronic kidney disease. The median concentration-to-dose ratios (interquartile range) in pediatric patients with or without CKD were as follows: preschool children, 0.64 (0.50–0.78) vs 0.60 (0.50–0.72); primary school children, 0.87 (0.74–0.96) vs 0.65 (0.52–0.78); and adolescents, 0.96 (0.93–1.12) vs 0.75 (0.60–0.92). Significance was determined by the Mann-Whitney U test. *p < 0.001, CD ratio, concentration-to-dose ratio; CKD, chronic kidney disease.
In this study, we confirmed 248 out of 1398 (17.7 %) patients who achieved partial or complete clinical response (median observation time, 1318 days). The median therapeutic concentration (interquartile range) and dose was 19.0 μg/mL (10.8–31.6) and 25.2 mg/kg (14.5–39.4), respectively. However, we could not evaluate the clinical response of 467 patients (33.4 %) because the observation period was<3 years.
4. Discussion
Although LEV is widely used as monotherapy or adjunctive therapy for pediatric epilepsy, the clinical usefulness of TDM for this drug has not been established. In this study, we clearly showed that some pediatric patients require TDM for LEV.
According to a review of recently published studies by Sourbron et al., age and concomitant use of inducers appear to be major factors that influence the pharmacokinetic profile of LEV [
]. Generally, growth and development have marked effects on drug clearance, and such effects are not apparent in adults. Our results in four age groups of pediatric patients clearly demonstrate the influence of age on LEV pharmacokinetics.
In this study, we showed that 2-year-old patients have the lowest LEV CD ratio. In contrast, infants have a relatively higher CD ratio. Because LEV has minimal plasma protein binding, it is excreted primarily by glomerular filtration. The glomerular filtration rate is quite low at birth but subsequently increases and reaches adult levels by age 2 years [
]. In our study, the mean body weight of the 2-year-old patients was 44 % higher than that of the infants; however, the kidney mas as a percentage of their body weight reaches the highest level at 2-year-old [
]. Thus, in our patients, the lower clearance in the infant group resulted in a higher CD ratio, but the rapid increase in the glomerular filtration rate in the first two years of life led to a significant decrease in CD ratios. Our findings support an earlier study in a small number of participants (n = 12), which found that children aged 6 months and older had the fastest clearance (1.57 mL/min/kg) and infants under 6 months of age had the slowest clearance (1.23 mL/min/kg) [
In addition, the LEV CD ratio tended to increase from age 2 to 5 years. By age 1 to 2 years, children have a higher clearance per bodyweight, but clearance decreases from age 2 to 5 years [
]. Our results show that TDM for LEV is clinically useful in the first 5 years of life.
We identified the therapeutic LEV concentration range for the long-term therapy in pediatric patients. Our study suggested that the reference concentration range for a LEV therapy was 11 to 34 μg/mL. Generally, LEV is initiated at a dose of 20 mg/kg. To achieve the same therapeutic levels in adolescent group, the doses in infants, preschool children, and school children need to be increased by 10 %, 23 %, and 15 %, respectively. In particular, 2-year-old patients have the lowest LEV CD ratio and need about a 31 % higher dose.
In this study, we showed that use of phenytoin, carbamazepine, and phenobarbital reduces serum LEV concentration; this finding is consistent with the results of previous studies [
Prospective assessment of levetiracetam pharmacokinetics during dose escalation in 4- to 12-year-old children with partial-onset seizures on concomitant carbamazepine or valproate.
Pharmacokinetic variability of four newer antiepileptic drugs, lamotrigine, levetiracetam, oxcarbazepine, and topiramate: a comparison of the impact of age and comedication.
]. These inducers can reduce the serum LEV concentration by increasing the activity of esterase enzymes. However, the extent of such interactions differed among studies because the studies evaluated different age groups and sample sizes. Our study demonstrated that pediatric patients are less susceptible to interactions with enzyme-inducing ASMs. Although the infant group had a higher CD ratio than the other age groups, all age groups showed similar susceptibility to inducers. The effects of enzyme induction were not different between phenytoin, carbamazepine, and phenobarbital. Also, no additive or synergetic effects were observed in patients who received multiple inducer regimens. Therefore, TDM for LEV is less clinically useful in pediatric patients when their inducer regimen is changed (i.e., after discontinuation, switching, or addition).
As mentioned in the Introduction, because LEV is primarily cleared by the kidneys, a dose reduction of 50 % is recommended in patients with moderate or severe renal impairment. Hirsch et al. reported that LEV concentrations increased linearly with decreasing creatinine clearance in both younger and older adults [
]. Generally, the incidence of renal failure in children and adolescents is lower than in adults, so no information is available on dosing adjustments for LEV in pediatric patients with CKD. Our study suggests that no dose adjustment is required in preschool children with mild CKD. However, a 25 % to 40 % dose reduction is recommended in primary school children and adolescents. In particular, the impact of impaired renal function on LEV pharmacokinetics markedly increases as children grow, so careful TDM is needed.
A population-based study in Italy found that a prevalence of CKD (creatinine clearance < 75 mL/min) of 74.7 per million in children and adolescents younger than 20 years [
]. The rate in our study was much higher, i.e., 66 of 1541 pediatric patients (4.3 %) had stage 2 CKD. Most of our patients had refractory epilepsy, and common epilepsy etiologies were prenatal or perinatal brain injury, perinatal asphyxia, short gestation, and low birth weight. According to a literature review on studies in pediatric CKD, low birth weight and small for gestational age infants have an increased risk of developing CKD [
]. Thus, assessing glomerular filtration rate and performing TDM for LEV appear to be important approaches in pediatric epilepsy patients at risk of CKD.
Our study has a number of limitations. Although we identified factors that influence the serum LEV concentration, we could not obtain trough serum concentrations. LEV has a short half-life of 5 to 6 h, but we analyzed the serum samples collected within 3 h after administration of LEV (a peak sampling). Iwasaki et al. studied LEV peak levels obtained from patients with focal epilepsy and reported that LEV concentrations showed positive correlations with efficacy [
]. However, most of the previous studies evaluated trough concentrations, which limits the comparability of our results. Because LEV concentration is related to sampling time, further investigations are needed to enable trough levels to be back-calculated. Also, the maximum dose of LEV in pediatric patients is set at 60 mg/kg/day. This was a retrospective study, and we found that 114 patients received LEV dosages greater than 60 mg/kg/day. Last, because our hospital did not routinely measure patient height, we identified the CKD stage by assessing height-for-age and serum creatinine level. Generally, patients with developmental delay show a low growth curve, and their actual height may be lower.
5. Conclusion
We identified the effects of age, concomitant use of inducers, and renal function on LEV pharmacokinetics in pediatric patients with epilepsy. LEV pharmacokinetics were significantly different between infant and preschool children, indicating that TDM for LEV is clinically useful in these age groups. In pediatric patients, concomitant use of enzyme-inducing ASMs had an insignificant effect on LEV pharmacokinetics. In pediatric patients with an increased risk of developing CKD, the glomerular filtration rate should be carefully monitored, and TDM for LEV should be performed.
Funding
This study was partly funded by a grant-in-aid for young scientists (No. 26860123) from the Japanese Ministry of Education, Science, Sports and Culture (MEXT) and by grants from the Japan Research Foundation for Clinical Pharmacology.
Author contributions
AO reviewed the clinical records. YY performed TDM for LEV under the supervision of KI and YT. Also, YY performed the data analysis and drafted the manuscript under the supervision of KI, NU, YK, and YT. All authors read and approved the final version to be published.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A. Supplementary material
The following are the Supplementary data to this article:
Prospective assessment of levetiracetam pharmacokinetics during dose escalation in 4- to 12-year-old children with partial-onset seizures on concomitant carbamazepine or valproate.
Pharmacokinetic variability of four newer antiepileptic drugs, lamotrigine, levetiracetam, oxcarbazepine, and topiramate: a comparison of the impact of age and comedication.
Pediatric CKD Study Group; Japan committee of measures for pediatric CKD of the Japanese Society of Pediatric Nephrology. Pre-dialysis chronic kidney disease in children: results of a nationwide survey in Japan.
00020-7&id=gr1.jpg){kind=link}
00020-7&id=gr2.jpg){kind=link}
00020-7&id=gr3.jpg){kind=link}