Christopher Hawthorne

Monitoring of intracranial pressure in patients with traumatic brain injury.

Since Monro published his observations on the nature of the contents of the intracranial space in 1783, there has been investigation of the unique relationship between the contents of the skull and the intracranial pressure (ICP). This is particularly true following traumatic brain injury (TBI), where it is clear that elevated ICP due to the underlying pathological processes is associated with a poorer clinical outcome. Consequently, there is considerable interest in monitoring and manipulating ICP in patients with TBI. The two techniques most commonly used in clinical practice to monitor ICP are via an intraventricular or intraparenchymal catheter with a microtransducer system. Both of these techniques are invasive and are thus associated with complications such as hemorrhage and infection. For this reason, significant research effort has been directed toward development of a non-invasive method to measure ICP. The principle aims of ICP monitoring in TBI are to allow early detection of secondary hemorrhage and to guide therapies that limit intracranial hypertension (ICH) and optimize cerebral perfusion. However, information from the ICP value and the ICP waveform can also be used to assess the intracranial volume–pressure relationship, estimate cerebrovascular pressure reactivity, and attempt to forecast future episodes of ICH.

Sharing of Big Data in Healthcare: Public Opinion, Trust, and Privacy Considerations for Health Informatics Researchers

Advances in technology has transformed clinical medicine; electronic patient records routinely store clinical notes, internet-enabled mobile apps support self-management of chronic diseases, point-of-care testing enables laboratory tests to be performed outside of hospital environments, patient treatment can be delivered over wide geographic areas and wireless sensor networks are able to collect and send physiological data. Increasingly, this technology leads to the development of large databases of sensitive electronic patient information. There is public interest into the secondary use of this data; many concerns are voiced about the involvement of private companies and the security and privacy of this data, but at the same time, these databases present a valuable source of clinical information which can drive health informatics and clinical research, leading to improved patient treatment. In this position paper, we argue that for health informatics projects to be successful, public concerns over the secondary use of patient data need to be addressed in the design and implementation of the technology and conduct of the research project.

Transcranial Bioimpedance Measurement as a Non-invasive Estimate of Intracranial Pressure

OBJECTIVES We have previously demonstrated a relationship between transcranial bioimpedance (TCB) measurements and intracranial pressure (ICP) in an animal model of raised ICP. The primary objective of this study was to explore the relationship between non-invasive bioelectrical impedance measurements of the brain and skull and ICP in traumatic brain injury (TBI) patients. MATERIALS AND METHODS Included patients were adults admitted to the Neurological Intensive Care Unit with TBI and undergoing invasive ICP monitoring as part of their routine clinical care. Multi-frequency TCB measurements were performed hourly through bi-temporal electrodes. The bioimpedance parameters of Z c (impedance at the characteristic frequency) and R 0 (resistance to a direct current) were then modelled against ICP using unadjusted and adjusted linear models. RESULTS One hundred and sixty-eight TCB measurements were available from ten study participants. Using an unadjusted linear modelling approach, there was no significant relationship between measured ICP and Zc or R0. The most significant relationship between ICP and TCB parameters was found by adjusting for multiple patient specific variables and using Zc and R0 normalised per patient (p < 0.0001, r 2 = 0.32). CONCLUSIONS These pilot results confirm some degree of relationship between TCB parameters and invasively measured ICP. The magnitude of this relationship is small and, on the basis of the current study, TCB is unlikely to provide a clinically useful estimate of ICP in patients admitted with TBI.

Investigation of the Relationship Between the Burden of Raised ICP and the Length of Stay in a Neuro-Intensive Care Unit

OBJECTIVES Raised intracranial pressure (ICP) is well known to be indicative of a poor outcome in traumatic brain injury (TBI). This phenomenon was quantified using a pressure time index (PTI) model of raised ICP burden in a paediatric population. Using the PTI methodology, this pilot study is aimed at investigating the relationship between raised ICP and length of stay (LOS) in adults admitted to a neurological intensive care unit (neuro-ICU). MATERIALS AND METHODS In 10 patients admitted to the neuro-ICU following TBI, ICP was measured and data from the first 24 h were analysed. The PTI is a bounded area under the curve, where the bound is the threshold limit of interest for the signal. The upper bound of 20 mmHg for ICP is commonly used in clinical practice. To fully investigate the relationship between ICP and LOS, further bounds from 1 to 40 mmHg were used during the PTI calculations. A backwards step Poisson regression model with a log link function was used to find the important thresholds for the prediction of full LOS, measured in hours, in the neuro-ICU. RESULTS The fit was assessed using a Chi-squared deviance goodness of fit method, which showed a non-significant p value of 0.97, indicating a correctly specified model. The backwards step strategy, minimising the models Akaike information criteria (AIC) at each change, found that levels 13-16, 18 and 20-21 combined were the most predictive. From this model it can be shown that for every 1 mmHg/h increase in burden, as measured by the PTI, the LOS has a base exponential increase of approximately 2 h, with the largest increases in the LOS given at the 20-mmHg threshold level. CONCLUSIONS This model demonstrates that increased duration of raised ICP in the early monitoring period is associated with a prolonged LOS in the neuro-ICU. Further validation of the PTI model in a larger cohort is currently underway as part of the CHART-ADAPT project. Second, further adjustment with known predictors of outcome, such as severity of injury, would help to improve the fit and validate the current combination of predictors.

1204: AUTOMATED VERSUS MANUAL APACHE II CALCULATION IN A SCOTTISH TERTIARY NEUROINTENSIVE CARE UNIT (NICU)

www.ccmjournal.org Critical Care Medicine • Volume 46 • Number 1 (Supplement) Learning Objectives: Automated data extraction for calculation of Acute Physiology and Chronic Health Evaluation II (APACHE II), within a tertiary NICU, has not been validated over gold standard manual scoring. This study aims to evaluate automated scoring efficacy, through assessing: 1) agreement between automated and manual scores for the same patients, 2) manual scoring inter-rater reliability and 3) manual scoring time. Methods: The automated system, Wardwatcher, extracts data from the clinical information system IntelliSpace Critical Care and Anaesthesia (ICCA) for APACHE II scoring. From 230 patient admissions to NICU in 2016, with valid APACHE II scores, 17 were randomly sampled and each patient retrospectively manually scored by two clinicians. Seventeen clinicians manually scored two patients each. Clinicians examined ICCA data from the first 24 hours in NICU and extracted most deviated from normal variable values, for the Acute Physiology Score (APS) component of APACHE II. For manual score calculation, clinician APS values were combined with age, chronic health and surgery data. Automated and manual score agreement was assessed using a Friedman test, linear regression and Bland-Altman analysis. Inter-rater reliability was assessed using Krippendorff ’s alpha reliability coefficient. Results: For the 17 patients, median automated APACHE II was 15 (IQR 14–19). Median manual APACHE II was 14 (IQR 12–18.8). 79.4% of manual scores differed from automated; 29.4% higher and 50% lower. Overall, Friedman analysis showed no statistically significant difference between manual and automated scores (p = 0.134). However, 94% of manual score pairs differed and Krippendorff ’s alpha was -0.0057, indicating poor inter-rater reliability. Total manual scoring took 384.7 minutes, median 10 minutes 59 seconds per patient. Conclusions: There is no statistically significant difference between APACHE II scores generated manually or automatically. However, poor inter-rater reliability highlights fundamental interclinician variability in manual scoring. Automated scoring is therefore a reasonable and timesaving alternative.

Multi-resolution Convolution Methodology for ICP Waveform Morphology Analysis

Intracranial pressure (ICP) monitoring is a key clinical tool in the assessment and treatment of patients in neurointensive care. ICP morphology analysis can be useful in the classification of waveform features.A methodology for the decomposition of an ICP signal into clinically relevant dimensions has been devised that allows the identification of important ICP waveform types. It has three main components. First, multi-resolution convolution analysis is used for the main signal decomposition. Then, an impulse function is created, with multiple parameters, that can represent any form in the signal under analysis. Finally, a simple, localised optimisation technique is used to find morphologies of interest in the decomposed data.A pilot application of this methodology using a simple signal has been performed. This has shown that the technique works with performance receiver operator characteristic area under the curve values for each of the waveform types: plateau wave, B wave and high and low compliance states of 0.936, 0.694, 0.676 and 0.698, respectively.This is a novel technique that showed some promise during the pilot analysis. However, it requires further optimisation to become a usable clinical tool for the automated analysis of ICP signals.

Identification of Clinically Relevant Groups of Patients Through the Application of Cluster Analysis to a Complex Traumatic Brain Injury Data Set

In neurological intensive care units (NICUs) we are collecting an ever increasing quantity of data. These range from patient demographics and physiological monitoring to treatment strategies and outcomes. The BrainIT database is an example of this type of rich data source. It contains validated data on 264 patients who suffered traumatic brain injury (TBI) admitted to 22 NICUs in 11 European countries between March 2003 and July 2005 [1, 6].

Artefact in Physiological Data Collected from Patients with Brain Injury: Quantifying the Problem and Providing a Solution Using a Factorial Switching Linear Dynamical Systems Approach.

INTRODUCTION High-resolution, artefact-free and accurately annotated physiological data are desirable in patients with brain injury both to inform clinical decision-making and for intelligent analysis of the data in applications such as predictive modelling. We have quantified the quality of annotation surrounding artefactual events and propose a factorial switching linear dynamical systems (FSLDS) approach to automatically detect artefact in physiological data collected in the neurological intensive care unit (NICU). METHODS Retrospective analysis of the BrainIT data set to discover potential hypotensive events corrupted by artefact and identify the annotation of associated clinical interventions. Training of an FSLDS model on clinician-annotated artefactual events in five patients with severe traumatic brain injury. RESULTS In a subset of 187 patients in the BrainIT database, 26.5 % of potential hypotensive events were abandoned because of artefactual data. Only 30 % of these episodes could be attributed to an annotated clinical intervention. As assessed by the area under the receiver operating characteristic curve metric, FSLDS model performance in automatically identifying the events of blood sampling, arterial line damping and patient handling was 0.978, 0.987 and 0.765, respectively. DISCUSSION The influence of artefact on physiological data collected in the NICU is a significant problem. This pilot study using an FSLDS approach shows real promise and is under further development.