Human patient simulation in physiology teaching: designing a high fidelity cardiovascular demonstration for first year undergraduates

Undergraduate cardiovascular teaching is traditionally delivered via lectures, tutorials and practical classes. The latter enable students to record their ECG and blood pressure but are necessarily limited to using non-invasive techniques in young, healthy individuals. We are enhancing our existing teaching with a Human Patient Simulator (HPS 337, METI), a life-sized manikin equipped with mechanical and software modelling of many physiological variables. The HPS was designed for clinical training but is also well suited for demonstrating physiological principles. It has previously been used as a teaching tool to explore cardiovascular physiology [1] and we have extended the approach taken in that study by comparing the cardiovascular responses predicted by the HPS base model with published in vivo data. This has enabled us to refine the model so as to provide data that approximate more closely to in vivo measurements. We have developed a teaching session to illustrate physiological principles that include the baroreceptor reflex and Starling’s law of the heart, by using the HPS to model the response to blood loss in man of ca. 10%, 30% and 50% of total blood volume. The HPS provides real-time waveform displays of arterial blood pressure (ABP), central venous pressure (CVP) and ECG. A simulated thermodilution method can be used to determine cardiac output (CO). The waveforms are generated by a software model that is pre-configured by the manufacturer, although some parameters within the model can be adjusted by the user. In order to construct a cardiac function curve and to ‘dissect’ the baroreceptor reflex, we also used data provided by the model to derive real-time values for stroke volume (SV) and total peripheral resistance (TPR). The simulated data were compared with data from pigs [2] and dogs [3] subjected to a corresponding blood loss. A good correlation was identified between the in vivo data and the METI base model for changes in CVP, SV and CO in response to all three levels of blood loss. However, in vivo reflex control of ABP in response to haemorrhage was found to be inadequately simulated by the HPS base model. Calculation showed that the base model incorporates only a modest increase in TPR following haemorrhage. We therefore adjusted the model by increasing both the baseline systemic vascular resistance (by a factor of 1.5) and the ‘gain’ of the peripheral component of the baroreceptor reflex (by a factor of 2.0). These adjustments provided simulated data that closely approximated the porcine and canine data. We conclude that, although the METI base model is suitable for illustrating qualitative changes in systemic arterial pressure and pulse rate in response to haemorrhage, it is necessary to refine the model in order to provide more robust simulation of experimentally-derived data. Euliano TY (2001). Adv Physiol Educ 25, 36-43.