Background: Critically ill children require accurate haemodynamic assessment to evaluate the severity of illness or response to therapy. Clinical estimation of cardiac output is inaccurate. Hence, advanced haemodynamic monitoring devices may help guide physicians towards the most appropriate treatment strategy. However, none of the currently available monitors for children fulfil all the criteria of an ideal device. Methods: We evaluated two novel minimally invasive haemodynamic monitoring devices in 100 critically ill children. The transpulmonary ultrasound dilution (TPUD) method is a validated indicator dilution based technique for measuring cardiac output in children. Pressure recording analytical method (PRAM) is an arterial pulse contour based method and is not yet validated in children. We compared PRAM with TPUD both in terms of agreement with absolute values of CO and also quantified the ability of PRAM to track changes in CO in response to therapy. We also evaluated the ability of TPUD to identify, and quantify, small anatomic shunts. Finally, a range of variables measured by TPUD and PRAM were assessed for their ability to predict response to fluid bolus administration. The contribution of baseline myocardial contractility towards that response was also evaluated. Results: PRAM showed unacceptable level of error for estimation of absolute values of CO and was unable to accurately track changes in CO. TPUD could identify small anatomic shunts. All of the volumetric variables were unable to predict accurately for fluid responsiveness. Myocardial contractility was found to be an important determinant of the response of stroke volume to fluid bolus administration. Conclusion: A revision of the current algorithm of PRAM is recommended for measurement of CO in children. The predictive ability of the studied variables was poor to moderate for determining response of stroke volume to fluid bolus administration.