BACKGROUND Non-invasive cardiac output monitoring (NICOM; NICOM™) may be useful in the management of extremely premature preterm infants. OBJECTIVES To evaluate a new bioreactance-based method of continuous NICOM in preterm infants following patent ductus arteriosus (PDA) ligation. METHODS Infants underwent three paired NICOM and echocardiography assessments of stroke volume (SV) and left ventricular output (LVO) in the postoperative period: at 1, 6-8, and 16-18 h postoperatively. NICOM- and echocardiography-measured SV and LVO during those periods were compared using Bland-Altman analysis and the intraclass correlation coefficient (ICC). RESULTS Twenty-five infants with a median (interquartile range) gestational age and birth weight of 25.0 weeks (24.5-25.9) and 700 g (615-775), respectively, were included. The overall systematic bias (limits of agreement) across all time points between the NICOM and echocardiography SV readings was 39% (8-69) with NICOM consistently underestimating echocardiography values. There was moderate consistency between NICOM and echocardiography SV values (ICC 0.78, p < 0.001). Compared with the 1-hour scans, the 6- to 8- and 16- to 18-hour scans had increased biases of 7.9% (95% CI 2.5-13.2) and 9.7% (95% CI 3.6-15.8), respectively. CONCLUSION Continuous LVO measurement using NICOM was feasible and demonstrated a consistent systematic bias compared with echocardiography in unstable extremely preterm infants without a PDA ligation. NICOM may be used as a trending tool for continuous monitoring in this population, but wide limits of agreement and increasing bias over time suggest it is not interchangeable with echocardiography. © 2014 S. Karger AG, Basel.