Haemodynamics in preterm infants with Patent Ductus Arteriosus

Haemodynamics in preterm infants with patent ductus arteriosus. Dr Charalampos Kotidis The primary aims addressed by my thesis were a multimodal, integrated assessment of influences of patent ductus arteriosus (PDA) on cerebral haemodynamics in extremely preterm infants using echocardiography, cerebral near infrared spectroscopy, amplitude integrated electroencephalography (aEEG) and cerebral ultrasonography, and the evaluation of a pre-defined two-step model for the causation of brain injury by PDA shunt (Chapter 3). The dataset was rich and provided an opportunity to characterise components of the multimodal assessment including: influences on the aEEG (Chapter 4) and antenatal influences on elements of the multimodal assessment (Chapter 5). Moreover, I investigated further existing biomarkers of aortic haemodynamics [pulse transit time (PTT) and pulse wave velocity (PWV)] and related them to the PDA and cerebral biomarkers and devised a novel biomarker [phase difference (PD)] (Chapter 6). The analysis demonstrated that severe intraventricular haemorrhage (IVH) was associated with the presence of large PDA and lower cerebral oxygenation. Furthermore, there was a significant association between lower cerebral oxygenation, lower cerebral electrical activity and higher resistivity index on anterior cerebral artery with larger PDA size, but not in interaction of these variables with IVH severity. This means that separate steps of a proposed two step model were valid, but not the entire two-step model. IVH happens early during extrauterine life or even during pregnancy and during the study there were more observations reflecting the aftermath following the IVH. Hence, it is uncertain whether the observed relationships are the cause or the epiphenomenon of brain injury. The strong association between PDA size and all the cerebral biomarkers implies that PDA has important and simultaneous effects on global cerebral function. PDA in some preterm infants has a different natural course and maybe linked to cerebral injury. However, many preterm infants with large PDAs did not develop severe IVH and had normal cranial ultrasound at discharge. This supports the ongoing controversy regarding the definition of the haemodynamically significant PDA. Whether and at which critical point the PDA becomes a disease is still uncertain and it is very possible this goalpost to be shifting depending on the rest of the neonatal comorbidities. Chapter 4 demonstrated that cerebral electrical activity increases with postnatal age and baseline gestational age, but decreases with morphine administration and the presence of a larger PDA. Chapter 5 investigated the neonatal cardiovascular and cerebral function following antenatal maternal MgSO4 administration. Maternal BMI and duration of MgSO4 infusion (and their interaction) appeared to have significant effect on neonatal Mg2+ that may merit changes to the dosing regimen. No associations were demonstrated between neonatal serum Mg2+ with PDA score, cerebral oxygenation and cerebral electrical activity. Future dose finding studies should be based on multicompartmental population pharmacokinetic studies that include maternal and neonatal pharmacodynamic measures and sample size calculation was performed to inform these future studies. Chapter 6 demonstrated that PDA has significant effects on aortic haemodynamics as measured by PTT, PWV and PD. In a small population of extremely preterm infants a strong positive correlation was found between PDA and PD which renders further investigation with validation studies. A continuous predictive biomarker for PDA size can be a helpful screening tool for infants needing echocardiography in the neonatal unit in an effort to use human resources more effectively and reduce the disturbance of the vulnerable preterm infants by repeated echocardiograms. Moreover, the analysis demonstrated the clinical potentials of combining PWV and blood pressure data to predict left ventricular cardiac output in the descending aorta without the need for echocardiographic assessment, which can be useful clinical information for the management of this population