Sparse Deconvolution and Causality Analysis of Inflammatory Markers During Cardiac Surgery

Vidya Raju,Ben Gibbison,Behnam Hajihossainlou,Elizabeth B Klerman,Rose T Faghih
DOI: https://doi.org/10.1109/EMBC40787.2023.10339979
Abstract:Major bodily trauma such as cardiac surgery elicits (in response to tissue injury and other exogenous surgical factors) a whole-body inflammation response during which specialized signaling proteins called cytokines are synthesized and invoke multiple defense mechanisms. Many proinflammatory and anti-inflammatory cytokines such as interleukins (IL) and tumor necrosis factor (TNF) are produced to initiate bodily repair. Due to the adverse health consequences, including mortality, of a maladaptive cytokine response, understanding their complex dynamics using system-theoretic modeling and analysis may pave the way for controlling the inflammatory response which may eventually improve medical outcomes for patients. To this end, we use clinical data from ten patients undergoing coronary arterial bypass graft surgery to study the response of four cytokines (IL6, IL8, IL10, TNFα) and the neuroendocrine hormone cortisol. We perform deconvolution to obtain the secretory pulses underlying their pulsatile production and analyze causal interactions, mathematically uncovering some interactive relationships found in previous experimental studies.Clinical relevance- This work is a first step towards a mechanistic inference of the inflammatory response to surgery that could eventually help control the inflammatory response and could inform medical interventions to improve patient outcomes.
What problem does this paper attempt to address?