Abstract
Every year millions of babies born too early or too small face a disadvantaged start to life when compared to their normal-weight term-born peers. Despite substantial clinical advancements, preterm birth and intrauterine growth restriction (IUGR) remain the leading cause of neonatal mortality; with the burden falling disproportionately on resource-poor settings, where access to obstetric care and neonatal intensive care infrastructure is limited. This persistent vulnerability underscores the urgent need for novel interventions that are not only effective but also feasible to implement across a diverse range of healthcare contexts. While the pathophysiology underlying the adverse outcomes in these populations is undeniably multifactorial, hypoxia has emerged as a common mechanistic thread linking various forms of perinatal compromise. Consequently, interventions that safeguard cellular energetics during hypoxic episodes may be of therapeutic benefit.
Creatine is a naturally occurring amino acid derivative that acts as a reservoir for high energy phosphate. In periods of high metabolic demand, this reservoir is used to rapidly replenish cellular ATP without a reliance on oxygen. Maternal creatine supplementation—whereby a mother consumes a dietary creatine supplement to increase fetal stores—has shown promise as a prophylactic treatment in preclinical models of acute asphyxia at term. Building on this success, there have been suggestions that creatine supplementation should become standard practice for all pregnant women, so that ifany hypoxic complications occur, a buffer is already in place to safeguard offspring cellular energy supply. However, to date, there has been little data on the safety or efficacy of creatine supplementation in chronic perinatal hypoxic scenarios such as preterm birth and IUGR. Furthermore, additional safety data is needed from “healthy” term pregnancies before translation into clinical populations can be considered.
To address these gaps, this thesis employed clinically relevant guinea pig models of perinatal compromise. Guinea pigs, whilst retaining many of the practical advantages of small research animals, have several physiological features that enhance translational relevance, including a relatively long gestation (69-71 days), precocial neonatal development, and endocrine regulation of pregnancy and parturition that closely parallels that of humans. These characteristics give rise to a unique opportunity to model conditions that are inherently linked to the in uteroenvironment and timing of birth.
This thesis brings together two complementary lines of investigation: first, the refinement and validation of guinea pig models that recapitulate key features of perinatal compromise; and second, the evaluation of maternal creatine supplementation in both uncomplicated and compromised pregnancy conditions.
Initially, a dataset of >3000 term-born guinea pigs was analysed to establish normative birthweight ranges in our outbred laboratory colony and to define pathological thresholds for IUGR (Chapter 5). Systematic comparison of published IUGR definitions demonstrated that a 10 thcentile birthweight cutoff most reliably identified pups with an increased brain-to-liver weight ratio, consistent with asymmetric growth restriction and aligning with clinical practice. These term-born references were extended to preterm pups delivered at gestational day 62, enabling gestational age-specific population norms and revealing important developmental differences between near-term and term animals that are often overlooked.
Having established a relevant definition for IUGR, attempts were then made to replicate a published maternal nutrient restriction model to produce IUGR offspring (Chapter 6). However, across two independent cohorts, the restriction paradigm proved excessively severe, preventing the ethical maintenance of pregnancy through to full term. Given that the findings were at odds with existing literature, a comprehensive characterisation of maternal responses to nutrient restriction was undertaken, including analyses of body composition, bodyweight trajectories, and food intake patterns. While the findings precluded the evaluation of creatine supplementation in an IUGR context, they contribute methodological insights for optimisation of maternal nutrient restriction protocols, thereby informing future experimental design for investigators seeking to establish ethically viable and reproducible models of nutritionally induced IUGR.
In the second half of this thesis, maternal creatine supplementation was evaluated in uncomplicated and preterm-compromised pregnancies. In uncomplicated pregnancies (Chapter 7), maternal oral creatine monohydrate supplementation effectively increased fetal creatine concentrations without adverse effects on maternal biochemistry, gestational weight gain, fetal growth, or pregnancy outcomes. Postnatal assessments revealed no detrimental metabolic effects in juvenile offspring, with a modest increase in locomotor activity observed in the absence of anxiety-like behaviours. Collectively, these findings provide a comprehensive functional assessment and indicate no overt safety concerns associated with maternal creatine supplementation in uncomplicated guinea pig pregnancies.
Building on the foundation of creatine safety in an uncomplicated pregnancy setting, the final set of experiments sought to investigate the efficacy of maternal creatine supplementation in a model of pharmacologically induced preterm labour (Chapter 8). Contrary to the original hypothesis, supplementation was associated with poorer delivery progression and adverse neonatal outcomes. Subsequent experiments demonstrated that these effects were attributable to impaired labour progression in supplemented dams rather than direct adverse effects on preterm pup physiology. These unexpected findings shifted the focus of the work toward identifying a potential safety concern and highlight the need for further mechanistic investigation before creatine supplementation can be reconsidered in a preterm context.
Collectively, these findings support that maternal creatine supplementation is safe in uncomplicated pregnancy contexts, with further investigation warranted in preterm contexts. They also underscore the importance of model fidelity and mechanistic understanding in translational research. While the results of this thesis ultimately challenge the straight-forward translation of creatine supplementation from an acute birth asphyxic context into other instances of perinatal compromise, such as preterm birth, they do not represent an insurmountable safety concern that precludes the future potential of creatine to improve neonatal outcomes. By identifying previously unrecognised safety considerations, this work prevents premature clinical trials and refines priorities for future studies: characterising creatine transport and synthesis in reproductive tissues, optimising dosing strategies, and integrating human pharmacokinetic data. These insights exemplify the value of rigorous preclinical investigation in the translational pipeline for advancing safe and effective perinatal therapeutic interventions.