Paediatric traumatic Brain Injury: The relationship between Intracranial Pressure and Brain Oxygenation

Master Thesis

2009

Permanent link to this Item
Authors
Supervisors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher

University of Cape Town

License
Series
Abstract
Introduction: Intracranial pressure (ICP) monitoring is a cornerstone of care for patients with severe traumatic brain injury (TBI). The primary goal of ICP treatment is to preserve brain oxygenation, and since brain oxygenation is usually not measured, the control of ICP is used as a surrogate marker. However studies indicating that cerebral hypoxia/ischemia may occur in the face of adequate ICP and cerebral perfusion pressure (CPP) suggest that the interaction between ICP and brain oxygenation is poorly understood and warrants further investigation. This is of particular importance in the context of children in whom the interpretation of relationships between intracranial factors is even more complex due to changing physiological norms with age. To date little scientific data exists in children and treatment threshold values are often extrapolated from adult guidelines. This study aims to better understand the relationship between ICP and brain oxygenation measured as brain tissue oxygen tension (PbtO2) in a large paediatric cohort suffering from severe TBI. Specifically analysis 1) investigated ICP and PbtO2 profiles over time following TBI, 2) examined the relationship between ICP and PbtO2 from time-linked paired observations, 3) explored various critical thresholds for ICP and PbtO2, and 4) interrogated digital data trends depicting the relationship between ICP and PbtO2. The level of agreement between hourly recorded and high frequency electronic data for ICP and PbtO2 was also evaluated. Method: Paired ICP and PbtO2 data from 75 children with severe TBI were tested with correlation and regression. Additional analyses controlled for mean arterial pressure (MAP), arterial partial pressure of oxygen (PaO2), CPP, arterial partial pressure of carbon dioxide (PaCO2) and haemoglobin (Hb) using multivariate logistic regression analysis and general estimating equations. Various thresholds for ICP were examined; these included age-related thresholds to account for the potential influence of age. Receiver-operating curves (ROCs) were used to graphically demonstrate the relationships between various thresholds of ICP and various definitions of low PbtO2. These were constructed for pooled and individual patient data. Interrogation of electronically recorded data allowed for case illustrations examining the relationship between ICP and PbtO2 at selected time points. Hourly and electronic data were compared using Bland and Altman plots and by contrasting the frequency of ICP and PbtO2 perturbations recorded with each system. 5 Result: Analyses using over 8300 hours of paired observations revealed a weak relationship between ICP and PbtO2, with an initially positive but weak slope (r = 0.05) that trended downwards only at higher values of ICP. Controlling for inter-individual differences, as well as MAP, CPP, PaO2, PaCO2 and Hb did not strengthen this association. This poor relationship was further reflected in the examination of threshold ICP values with ROCs, no singular critical ICP threshold for compromised brain oxygenation was discernible. Using age-based thresholds did not improve this relationship and individual patient ROCs demonstrated inter-individual heterogeneity in the relationship between ICP and PbtO2. However, it was clear that in individual patients ICP did exhibit a strong negative relationship with PbtO2 at particular time points, but various different relationships between the 2 variables were also demonstrated. A high level of agreement was found between hourly and electronic data. Conclusion: These results suggest that the relationship between ICP and PbtO2 is highly complex. Although the relationship in individual children at specific time points may be strong, pooled data for the entire cohort of patients, and even for individual patients, suggest only a weak relationship. This is likely because several other factors affect PbtO2 outside of ICP, and some factors affect both independently of each other. These results suggest that more study should be directed at optimising ICP thresholds for treatment in children. The use of complimentary monitoring modalities may assist in this task. Depending on the adequacy of measures of brain perfusion, metabolism or oxygenation, it is possible that targeting a range of ICP values in individual patients may be appropriate; however this would require detailed investigation.
Description
Keywords

Reference:

Collections