ICP Responsivity, a Metric to Evaluate Cerebrovascular Dynamics in Pediatric Traumatic Brain Injury

Introduction: Intracranial pressure (ICP) typically remains stable during minor physiological fluctuations in cardiovascular dynamics and CO2 levels. However, under pathological conditions such as severe traumatic brain injury (TBI), systemic physiological changes can result in aberrant responses in ICP, indicating disturbed autoregulation and/or poor intracranial compliance. We introduce a quantitative metric, ICP responsivity (RICP), derived from high-frequency physiological signals, and describe its characteristics in pediatric TBI patients.


Methods: This IRB-approved study analyzed 0.2 Hz bedside monitor data from 21 prospectively enrolled children (4 mo-15 y, 62% male) with TBI (initial GCS < 8, median GCS 4 [IQR 2.5]) at a large children’s hospital. Penalized regression models were developed using continuous ICP data with heart rate (HR), mean arterial pressure (MAP), and end-tidal CO2 (ETCO2) as predictor variables. A 6-hour weighted rolling window with normalized variables was used to derive a time-varying RICP (calculated as the R² value of the model fit) and regression coefficients for MAP, HR, and ETCO2.


Results: RICP and the contribution of individual regression coefficients varied temporally and amongst patients. RICP tended to be higher in patients that died vs survived (RICP: 0.58 vs. 0.39, p=0.08). The CO2 coefficient was higher in patients that survived vs died (0.36 vs. 0.08, p=0.04) consistent with intact CO2 reactivity whereas the MAP coefficient tended to be higher in patients that died vs survived (0.32 vs. 0.05, p=0.13) consistent with disturbed autoregulation. RICP was higher when cerebral perfusion pressure (CPP) was < vs > 50mmHg (0.67 [0.46-0.84] vs 0.38 [0.19-0.60]) and was driven by changes in MAP (coefficient 0.64 [0.25-0.82] vs 0.06 [-0.15-0.30]) and ETCO2 (coefficient (0.31 [0.03-0.60] vs 0.08 [-0.10-0.32], consistent with CPP below the lower limit of autoregulation.


Conclusions: RICP reflects aberrant changes in ICP in response to physiologic fluctuations in MAP and CO2 and can be obtained using standard ICU monitoring. RICP may be a valuable tool for the real time assessment of intracranial physiology including blood pressure autoregulation and CO2 reactivity in TBI patients, warranting further development as a bedside clinical decision support tool.