PATIENTS AND METHODS

All neonates admitted within the first 24 hours of life to the neonatal intensive care unit and with an arterial line in place were considered for inclusion if propacetamol was administered. The decision to prescribe propacetamol or any other analgesic was made by the attending neonatologist. Propacetamol was administered when minor, painful procedures were carried out, such as insertion of a peripheral arterial or venous line, insertion of a central venous line, or placement of a chest tube, or as additional treatment in infants receiving opioids. Exclusion criteria were major congenital malformations and severe birth asphyxia (Apgar score < 4 at five minutes) in line with other studies performed on neonates. The initial dose (20 mg (10 mg paracetamol)/kg) was based on literature data, with the intention to change this dose if interim analysis of the paracetamol levels in the first 15 infants was inadequate (plasma levels < 5 mg/l within 8–10 hours of administration).^10–13 Maternal use of analgesics (besides paracetamol) was not an exclusion criterion in this single-dose pharmacokinetic study.

As part of standard nursing care in the neonatal intensive care unit, a multidimensional pain scale was used to document pain/comfort. With this pain scale (Leuven neonatal pain scale), three different levels can be discriminated: level 1, < 4/14 (no pain); level 2, 4–6/14 (mild discomfort); level 3, > 6/14 (pain).¹⁶ An algorithm is used within the unit to administer and adapt analgesics based on this pain scale.¹⁷

The number and dose of other analgesics or sedatives prescribed in the first day of life were recorded. Birth weight was documented on admission to the unit. GA was estimated by routine ultrasound examination before 20 weeks of gestation if available or was based on the last menstrual period of the mother and postnatal physical characteristics.

Propacetamol was administered as a 15 minute infusion to avoid local discomfort.¹⁸ Blood samples (0.2 ml) were taken from an arterial line 30, 60, 90, 120, 180, 240, and 600 minutes after the start of intravenous administration. The maximum total amount of blood allowed to be collected in a single neonate was 1 ml/kg. After centrifugation, samples were stored at −20°C until analysis. Plasma paracetamol concentrations were determined using fluorescence polarisation immunoassay (Adx system; Abbott Laboratories, North Chicago, IL, USA). The determination limit was 1 mg/l, and the precision was 7%.

Pharmacokinetics were calculated assuming a linear one compartment model with instantaneous input and first order output. For every patient, a logarithmic trend line (y = a ln(x) + b) was calculated based on at least three plasma samples. The relative distribution volume (litres/kg) (V_d) and concentration at t = 0 (Cmax₀) were calculated. The slope of the curve (slope = (logCt₂ − logCt₁)(t₂ − t₁)) was used to calculate the time constant K (slope × 2.303), elimination half life (0.693/K) (t_½), and total clearance (K × V_d) (CL_t). Results are expressed as mean (SD) and range.

Student’s t test (normal distribution) or the Wilcoxon test was used to compare clinical and pharmacokinetic findings in preterm (< 37 weeks GA) and term (⩾ 37 weeks GA) infants. Linear regression analysis of the effect of GA and birth weight on t_½ were calculated (MedCalc; Medi Software, Mariukerke, Belgium). The protocol was approved by the local ethics committee (Gasthuisberg, Leuven, Belgium), and infants were only included after written informed consent had been obtained from the parents.

RESULTS

Thirty neonates of variable GA were included in this single-dose study. Fifteen received the 20 mg (10 mg paracetamol)/kg dose, and the remaining 15 received a 40 mg (20 mg paracetamol)/kg dose. Table 1 summarises the clinical characteristics. The overall mean (SD) birth weight was 2111 (1094) g, and GA at inclusion was 33.8 (3.9) weeks. Postnatal age at inclusion was 12.7 (6.4) hours. Ten infants had a GA of ⩾ 37 weeks. Twenty infants were preterm (< 37 weeks GA), 10 of whom were younger then 32 weeks GA. Twenty six infants received respiratory support, 16 of whom (53%) were ventilated. Fifteen (50%) received other analgesics in the first 24 hours of life.

In total, 213 blood samples were collected and analysed. Figure 1 gives the results for all the plasma samples; great variability can be observed.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Plasma levels of paracetamol in all infants (n = 213) after a single intravenous administration of 20 mg/kg (A) or 40 mg/kg (B) of propacetamol.

Table 2 summarises the pharmacokinetic characteristics. No significant difference in relative V_d (litres/kg) between preterm and term infants was found. t_½ and CL_t were significantly (both p < 0.05) different between preterm and term infants. Mean t_½ was 277 minutes in preterm infants and 172 minutes in term infants. Mean CL_t was significantly lower in preterm than term infants (0.116 v 0.170 litre/kg/h) In infants of < 32 weeks GA, mean t_½ was 290 minutes, whereas in more mature infants (32–36 weeks GA) it was 265 minutes. Correlation of GA with t_½ (r = −0.46) was stronger than birth weight with t_½ (r = −0.39). Figure 2 shows linear regression analysis of the effect of GA on t_½ with 95% confidence intervals. We found no difference in t_½ or other clinical characteristics (birth weight, GA) between preterms (< 35 weeks GA) who received (n = 13) and preterms who did not receive (n = 6) prenatal steroids (betamethasone) for lung maturation. Neither did we find any sex related differences.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Linear regression analysis (with 95% confidence intervals) of the effect of gestational age on serum half life (r = −0.46).

As this is a single-dose study, other analgesics were allowed. Half of the infants received at least one other analgesic (fentanyl (11), tramadol (6), ibuprofen-lysine (1)) during the first 24 hours based on the standardised evaluation by pain score (Leuven neonatal pain scale). Level 1 pain (pain scale < 4) was documented in 26/30 infants in the hours before paracetamol administration, in 30/30 infants during the period when a therapeutic level (> 5 mg/l) of paracetamol had been reached, and in 24/30 infants afterwards. If we consider only infants (n = 15) who did not receive any analgesic besides paracetamol in the first 24 hours, level 1 analgesia was documented in 14/15 infants before administration, in 15/15 infants in the period when a therapeutic level had been reached, and in 12/15 infants after this period.

DISCUSSION

The serum half life was 277 (143) minutes in preterm infants and 172 (59) minutes in term infants (p < 0.05). Clearance was 0.116 (0.08) litre/kg/h in preterm infants and 0.170 (0.06) litre/kg/h in term infants (p < 0.05). The pharmacokinetics and pharmacodynamics of propacetamol are well documented in adults and children, but there is only one study on its pharmacokinetics in infants younger than 1 year (n = 12, of which five were < 10 postnatal days), and there are no data on propacetamol in preterm neonates. The pharmacokinetics of propacetamol in this study were compared with the pharmacokinetics of paracetamol and propacetamol in other cohorts described in the literature.

Term neonates

Our findings in term infants are in line with the single study on intravenous propacetamol.¹⁰ Autret et al documented the pharmacokinetics in 12 infants, five of whom were less than 10 days old. The serum half life in these five neonates was 210 (30) minutes, CL_t was 0.149 (0.067) litre/kg/h, and V_d was 0.7 (0.2) (table 3). Pharmacokinetics after rectal administration of paracetamol in term neonates were studied by Van Lingen et al¹¹ and Hopkins et al.¹⁹ Van Lingen et al found a t_½ of 162 (84) minutes (n = 10), and Hopkins et al found a t_½ of 228 minutes (n = 9). There are no studies on the pharmacokinetics of paracetamol after nasogastric administration in the first day of life. Studies in neonates by Hopkins et al¹⁹ (n = 3) and Anderson et al⁹ (n = 16) after nasogastric administration found a serum t_½ of 168 minutes and 576 minutes respectively. Co-administration of opioids and its effect on gastric motility may, at least partially, explain these differences. There is a recent report on unintentional intramuscular administration of propacetamol in one term neonate; in that single case, the calculated serum half life was 210 minutes.²⁰

Preterm infants

Mean t_½ (< 37 weeks GA; n = 20) and CL_t after a single dose were 277 minutes and 0.116 litre/kg/h in preterm infants, and the relative V_d was 0.61 litre/kg. In infants of < 32 weeks GA, mean t_½ was 290 minutes, and in more mature infants (32–36 weeks GA), it was 265 minutes. Data on the pharmacokinetics of paracetamol in preterm neonates are only available after rectal administration. Van Lingen et al¹¹ studied pharmacokinetics after rectal administration of paracetamol in 28 preterm neonates in the first day of life (28–36 weeks GA). t_½ was 660 (342) minutes in the 28–32 week GA group and 450 (240) minutes in the 32–36 week GA group. Mean maximal concentration was 12.5 and 7.5 mg/l, and mean time to reach maximal concentration was 234 and 306 minutes (28–32 and 32–36 weeks). Lin et al¹² found a mean (SD) maximum concentration of 8.38 (3.9) mg/l and a mean (SD) time to reach maximum concentration of 78 (40) minutes after rectal administration of 20 mg/kg in five preterm neonates.¹² These findings are formulation specific, but may be relevant in clinical care as therapeutic drug concentration after intravenous administration will be reached sooner.

Combining pharmacokinetic data in term and preterm neonates in our population with the findings of Autret et al¹⁰ in neonates and infants, a maturational trend during the first year of life is observed (table 3). This is in line with the developmental pharmacokinetics described after oral or rectal administration of paracetamol.^7,9,10

Although we observed a maturational trend in the pharmacokinetics of paracetamol after intravenous administration, overall correlation (r = −0.46) between GA and t_½ is still weak. In contrast with rectal and oral administration, differences in bioavailability (venous rectal drainage, gastrointestinal motility) cannot explain this variability. Further study of other variables potentially responsible for this variability is needed. Prenatal administration of betamethasone for lung maturation had no maturational effect on t_½ in this study. We did not find any sex related differences, in contrast with the findings reported after rectal administration.¹¹

Pharmacodynamic data suggest an analgesic effect of intravenous paracetamol in this population. The design of this study (not blinded, other analgesics allowed) does not allow us to draw conclusions other than that multiple dose administration of intravenous paracetamol should be adjusted for GA. Based on the longer t_½ in preterm infants, either the interval should be longer or the dose should be lower, in line with reported regimens for rectal and oral administration.^7,9,11

Because of the major interindividual variability of the pharmokinetics in preterm infants, we believe it is too early to make any multiple dose recommendations. In term infants, a loading dose of 30 mg/kg propacetamol—that is, 15 mg paracetamol—followed by 20 mg/kg every six hours could be considered. As accumulation may still occur in the individual neonate, it is safest and feasible to determine plasma concentrations until additional data are available.