Background Limited pharmacokinetic (PK) data of metronidazole in premature infants has led to various dosing recommendations. Surrogate efficacy targets for metronidazole are ill-defined and therefore aimed to exceed minimum inhibitory concentration of organisms responsible for intra-abdominal infections. Methods We evaluated the PK of metronidazole using plasma and dried blood spot (DBS) samples from infants ≤32 weeks gestational age in an open-label, PK, multicenter (N=3) study using population PK modeling (NONMEM). Monte Carlo simulations (N=1000 virtual subjects) were used to evaluate the surrogate efficacy target. Metabolic ratios of parent and metabolite were calculated. Results Twenty-four premature infants (111 plasma and 51 DBS samples) were enrolled: median (range) gestational age at birth 25 (23–31) weeks, postnatal age 27 (1–82) days, postmenstrual age (PMA) 31 (24–39) weeks, and weight 740 (431–1466) g. Population clearance (CL, L/h/kg) was 0.038 × (PMA/30)2.45 and volume of distribution (L/kg) of 0.93. PK parameter estimates and precision were similar between plasma and DBS samples. Metabolic ratios correlated with CL. Conclusion Simulations suggested the majority of infants in the neonatal intensive care unit (>80%) would meet the surrogate efficacy target using PMA-based dosing.

Infants are therapeutic orphans. Many drugs used in infants are used “off-label”, increasing the risk of drug toxicity and suboptimal efficacy in this vulnerable population. This knowledge gap in clinical pharmacology is partly attributed to challenges associated with conducting clinical trials in infants. Consequently, there is a need for novel and efficient study designs more suited to this population. Available literature describing the use of minimal-risk approaches to characterize the pharmacokinetics (PK) of drugs in infants was critically reviewed. Population PK approach with sparse sampling was found to be well established. Other, more recent alternatives, like dried blood spots sampling, opportunistic design, and the use of non-blood matrices are promising strategies with an increasing number of applications in infants. Physiologically based pharmacokinetic modeling provides valuable insight in drug disposition but still needs more prospective validation. Increasing experience with these methods provides understanding of their strengths and limitations and will help improve the design of future PK studies in infants.

Despite metronidazole's widespread clinical use since the 1960s, the specific enzymes involved in its biotransformation have not been previously identified. Hence, in vitro studies were conducted to identify and characterize the cytochrome P450 enzymes involved in the formation of the major metabolite, 2-hydroxymetronidazole. Formation of 2-hydroxymetronidazole in human liver microsomes was consistent with biphasic, Michaelis-Menten kinetics. Although several cDNA-expressed P450 enzymes catalyzed 2-hydroxymetronidazole formation at a supratherapeutic concentration of metronidazole (2000 mM), at a "therapeutic concentration" of 100 mM only CYPs 2A6, 3A4, 3A5, and 3A7 catalyzed metronidazole 2-hydroxylation at rates substantially greater than control vector, and CYP2A6 catalyzed 2-hydroxymetronidazole formation at rates 6-fold higher than the next most active enzyme. Kinetic studies with these recombinant enzymes revealed that CYP2A6 has a K m = 289 mM which is comparable to the K m for the high-affinity (low-K m ) enzyme in human liver microsomes, whereas the K m values for the CYP3A enzymes corresponded with the low-affinity (high-K m ) component. The sample-to-sample variation in 2-hydroxymetronidazole formation correlated significantly with CYP2A6 activity (r ‡ 0.970, P < 0.001) at substrate concentrations of 100 and 300 mM. Selective chemical inhibitors of CYP2A6 inhibited metronidazole 2-hydroxylation in a concentration-dependent manner and inhibitory antibodies against CYP2A6 virtually eliminated metronidazole 2-hydroxylation (>99%). Chemical and antibody inhibitors of other P450 enzymes had little or no effect on metronidazole 2-hydroxylation. These results suggest that CYP2A6 is the primary catalyst responsible for the 2-hydroxylation of metronidazole, a reaction that may function as a marker of CYP2A6 activity both in vitro and in vivo.