A New Algorithm for Arterial Inflection Point Identification in Critically Ill Patients: A Preliminary Study

The morphological characteristics of arterial blood pressure (ABP) waveform represent an important source of information about the cardiovascular system. However, the identification of ABP features, such as the inflection point (IP), may be challenge in cases of high-frequency noise or important morphological changes of ABP morphology as it happens, for example, in critically ill patients. In this work we propose a new method for IP identification based on a bandpass filtering operation, and we tested it on ABP signals collected during a swine experimental protocol of polymicrobial septic shock and resuscitation, to evaluate its performance on non-physiological ABP waveforms. Cardiac echography images of left ventricular outflow tract will be used to extract independent measures of IP occurrence time in order to have a reference measure to validate the proposed method in comparison with traditional methods. The results show that the bandpass filtering method proposed may represent a good approach for IP identification in cases of very smooth waveform induced by the septic shock condition, where traditional methods tend to underestimate and to anticipate the IP instead. This work represents a preliminary exploration of alternative methods for IP identification to be used in critical cardiovascular conditions such as septic shock. Further studies are necessary to refine the IP detection algorithm so to translate into clinical practice markers and indices related to pulse wave dynamics which are currently neglected.