Agreement between USCOM and Vicorder haemodynamic measurements in pregnancy

Objective. To evaluate the agreement between maternal hae­modynamic measurements assessed with USCOM® and Vi­corder® machines.
Materials and Methods. Overall, 344 assessments from 172 participants between 30+0 and 39+6 weeks gestational age (156 healthy controls, 11 participants with small for gestation­al age and 5 with fetal growth restriction) at a tertiary level maternity unit were considered. The agreement between US­COM® and Vicorder® results was evaluated for heart rate (HR), cardiac output (CO), stroke volume (SV) and total peripheral resistance (TPR). The assessment was based on Bland-Altman plots with 95% limits of agreement (LOA) for the mean dif­ference (MD) between Vicorder® and USCOM®, reporting also the intraclass correlation coefficient (ICC).
Results. ICCs were generally low (ranging from 0.21 to 0.29), except for HR (ICC = 0.76, p < 0.001). USCOM® pro­vided, on average, higher values for HR (MD = -0.30 bpm [-18.65; 18.05]), CO (MD = -0.47 L/min [-3.77; 2.82]) and SV (MD = -5.13 mL [-48.80; 38.54]) and lower for TPR (MD = 72.7 dynes × sec/cm5 [-610.54; 755.07), compared to Vi­corder®.
Conclusions. The agreement between USCOM® and Vicord­er® measurements is low, with wide LOA for all measure­ments, emphasising a high risk of bias. Therefore, standard­isation of reference ranges in relation to the device under investigation should be used, and adjustment according to gestational age where there is a change in that parameter over time.