3.4 Study Comparison
There were no obvious differences in the PK of OXC due to ethnicity.Fig. 2 shows concentration-time profiles of all studies that are simulated with virtual infant (a), children (b) and adult (c) patients. The model established by Peng et al. [27] which presented lower CL, displayed a much higher trough concentration (Ctrough) than others (CL: 0.016 vs. 0.035–0.073 L/h/kg). On the contrary, the PPK model established by Yu et al. [33] which presented a higher CL, showed a considerably lower Ctrough than others (CL: 0.084 vs. 0.035–0.073 L/h/kg).
The estimated median CL was 0.078 L/h/kg for infants (10 kg, 1 year) [25], 0.058 (range: 0.016–0.084) L/h/kg for children (30 kg, 10 years) [24, 26-28, 30-33], and 0.033 (range: 0.029–0.0599) L/h/kg for adults (70 kg, 40 years) [26, 29, 33-35]. The median CL per kg in infants and children was higher than that in adults. Furthermore, children had a larger variability in CL than adults (0.016–0.084 L/h/kg vs. 0.029–0.06 L/h/kg).
Moreover, the median Vd per kg in infants and children was higher than that in adults. The determined median Vd was 1.38 L/kg for infants, 0.78 (range: 0.47–2.16) L/kg for children, and 0.45 (range: 0.2–1.48) L/kg for adults.
The final PK parameters included in our review are summarised inTable 3 . The between-subject variability (BSV) was determined from exponential models in all included studies. The median (range) BSV was as follows: CL, 16.84% (11.09%–34.21%); Vd, 35.35% (6.88%–104.4%); and Ka, 24.05% (8.27%–39.82%). The residual unexplained variability (RUV) was described using the proportional model: 12.05% (4%–32%) or additive model: 3.396 mg/L (0.93–0.512 mg/L).
All PPK analysis intended to explain the BSV of MHD PK by investigating possible covariates. The covariates investigated and identified in each study are visually shown in Fig. 3 . The commonly investigated covariates included weight, sex, age, co-administered medications, body surface area, and estimated glomerular filtration rate (eGFR). The identified covariates for CL included weight, co-administered medications, eGFR, dose. All the investigated and identified covariates in the PPK model are summarised in Table S1 .
It was found that eGFR significantly affected MHD CL, according to the observations from Lin et al. [29]. In patients with impaired renal function (eGFR: 20–80 mL/min), the CL was 34.8% (9.4%–64.2%) lower than those with normal renal function.
There was one study identified dose could affect CL, which may indicate a nonlinear elimination of MHD [27]. However, these effects may on account of the dose adjustment in clinical practice according to the TDM process, which can be misinterpreted as nonlinearity in the system [37]. Therefore, further in-depth study on the possible PK of MHD is necessary.
The influence of all included covariates on CL is shown in Fig. 4 . The ranges of continuous covariates that were scaled are as follows: the weight of adults was set at 40–100 kg and that of children was set at 16–40 kg. The age was set as 18–75 years. The range of eGFR and alanine aminotransferase (ALT) were set as 20–120 mL/min and 5–150 U/L, respectively. The dose of OXC was set as 75–1350 mg/kg/d.
Ten of all studies that investigated impact of the weight indicated that it was associated with the CL of MHD [24, 26-32, 34, 35]. Weight had a significant influence on CL. Our finding showed that the CL of patients with different weight could range from 0.54 to 1.8 times compared to the typical patient (children: 30kg, adults: 70kg). Co-administration with EIAEDs also significantly affected CL in children and adults, which approximately increased 1.17 to 1.63 times [24-26, 28, 32]; in four studies, the clinical significance of the impact of EIAED co-administration on CL was more than 20% [24-26, 32]. eGFR was identified to significantly affect the CL of adults with a range of 0.33 to 1.24 times [29] compared to a typical patient with an eGFR of 90 mL/min. ALT was also found to have a significant impact on CL in children, with a range of 0.73 to 1.35 times compared to a typical patient with an ALT level of 30 U/L [33].
Four studies conducted a model-based simulation to show the influence of covariates on the PK profile and optimize the dosage regimen to achieve the target MHD concentrations of 3 and 35 mg/L recommended by the current guideline [29-32]. For adults and paediatric patients, all the four studies recommended adjusting the dosage [29-32]. Chen et al. [31] and Rodrigues et al. [32] had a similar recommended dose range for paediatric patients based on their weight. However, Lin et al. [30] recommended a lower dose than others, which may be explained by the lower CL in their study than others (0.04 L/h/kg vs. 0.059/0.057 L/h/kg). Meanwhile, Rodrigues et al. [32] also recommended increasing the maintenance dose by 50% above the recommended maintenance dose for children treated with EIAEDs.