Non-Invasive Haemodynamic Monitoring in Pregnancy: A Comparative Study Using Ultrasound and Bioreactance.

Background: Due to the importance of cardiovascular dysfunction and advances in the development and use of non-invasive cardiac output (CO) monitoring devices, there is a growing interest in their use in the obstetric population. The aim of this study was to compare two commercially available, non-invasive CO monitors in the assessment of heart rate (HR), stroke volume (SV), CO, cardiac index (CI) and total peripheral resistance (TPR) in an obstetric population.

Methods: This was a prospective, comparative study including two groups. A normotensive group, which consisted of uncomplicated healthy pregnancies, and a hypertensive group, which consisted of those complicated by hypertensive disorders of pregnancy. Non-invasive cardiovascular assessments were performed using two methodologies. USCOM® is an operator-dependent Doppler ultrasound device which obtains velocity time integrals of transaortic blood flow at the left ventricular outflow tract, while NICOM® provides an operator-independent haemodynamic profile by assessing thoracic voltage changes (bioreactance) which reflect pulsatile blood flow in the thorax. Correlation coefficients were derived, and Bland-Altman analysis was performed to derive the mean percentage difference (MPD) between the devices.

Results: 598 women were recruited for this study. In the normotensive group, 524 paired results were analysed, while 74 paired results were analysed in the hypertensive group. In the normotensive group, we found excellent correlation between USCOM® and NICOM® for HR (r = 0.885, p < 0.05), and moderate correlations for SV (r = 0.445, p < 0.05), CO (r = 0.529, p < 0.05), CI (r = 0.385, p < 0.05) and TPR (r = 0.524, p < 0.05). In the hypertensive group, we obtained similar correlations (HR: r = 0.877, p < 0.05; SV: r = 0.575, p < 0.05; CO: r = 0.601, p < 0.05; TPR: r = 0.589, p < 0.05). Bland-Altman analysis found that the agreement between both methodologies improved as gestation advances, with an MPD of 34% for CO estimation in the third trimester of uncomplicated pregnancies, and 39% in pregnancies complicated by hypertensive disorders.

Conclusion: Our findings suggest that the two methodologies perform similarly in both uncomplicated pregnancies and in pregnancies complicated by hypertensive disorders. The study findings do not preclude the use of USCOM® and NICOM® devices in pregnancy, but indicate that these platforms cannot be used interchangeably. Our findings demonstrate an improvement in MPD as gestation advances and, therefore, questions the validity of previous longitudinal studies investigating maternal haemodynamics using these methodologies. Our work also highlights the need to construct reference ranges for each device and for validation of each methodology against a reference method before their use in research or clinical practice.