Visualization of the Intracranial Pressure and Time Burden in Childhood Brain Trauma: What We Have Learned One Decade on With KidsBrainIT.

To validate the intracranial pressure (ICP) dose-response visualization plot for the first time in a novel prospectively collected pediatric traumatic brain injury (pTBI) data set from the multi-center, multi-national KidsBrainIT consortium. Prospectively collected minute-by-minute ICP and mean arterial blood pressure time series of 104 pTBI patients were categorized in ICP intensity-duration episodes. These episodes were correlated with the 6-month Glasgow Outcome Score (GOS) and displayed in a color-coded ICP dose-response plot. The influence of cerebrovascular reactivity and cerebral perfusion pressure (CPP) were investigated. The generated ICP dose-response plot on the novel data set was similar to the previously published pediatric plot. This study confirmed that higher ICP episodes were tolerated for a shorter duration of time, with an approximately exponential decay curve delineating the positive and negative association zones. ICP above 20 mm Hg for any duration in time was associated with poor outcome in our patients. Cerebrovascular reactivity state did not influence their respective transition curves above 10 mm Hg ICP. CPP below 50 mm Hg was not tolerated, regardless of ICP and duration, and was associated with worse outcome. The ICP dose-response plot was reproduced in a novel and independent pTBI data set. ICP above 20 mm Hg and CPP below 50 mm Hg for any duration in time were associated with worse outcome. This highlighted a pressing need to reduce pediatric ICP therapeutic thresholds used at the bedside.

Resolution patterns and duration of retinal haemorrhages measured by two-dimensional retinal area pixel counts from sequential retinal imaging in childhood encephalopathies: a morphometric study.

OBJECTIVE: To determine the time to resolution of different-layered retinal haemorrhages (RHs), and to describe the main patterns of their resolution in a group of children with encephalopathies. METHODS AND ANALYSIS: From a prospective study of 114 children with traumatic and non-traumatic encephalopathies, 429 selected individual RHs (iRHs) from 18 children were serially imaged from admission using a RetCam. Photoshop and Scion Imaging software allowed calculation of RH area in pixels. RESULTS: Two patterns of the resolution were recognised on the basis of area measurements: a progressive decrease (pattern A, 60% of iRHs), and a form of asymmetrical decay in which iRHs first increased in size before then progressively decreasing (pattern B, 35% of iRHs). Within the pattern A group, the Kaplan-Meier median survival time (MST) (95% CI) was 10 (9.3 to 10.7) days for intra-RHs (IRHs) and 38 (10.8 to 65.2) days for pre-RHs (PRHs), log rank (Mantel-Cox) p=0.001. The mean percentage reduction in area per day was 12.5% for all iRHs, 14.5% for IRHs and 6.3% for PRHs. CONCLUSION: Serial area measurements of iRHs revealed that 35% haemorrhages became temporarily larger before decreasing to resolution. Serially imaged selected RHs showed a longer MST for PRHs than for IRHs.

Cerebral Perfusion Pressure Variability Between Patients and Between Centres.

INTRODUCTION: The aim of this analysis was to investigate to what extent median cerebral perfusion pressure (CPP) differs between severe traumatic brain injury (TBI) patients and between centres, and whether the 2007 change in CPP threshold in the Brain Trauma Foundation guidelines is reflected in patient data collected at several centres over different time periods. METHODS: Data were collected from the Brain-IT database, a multi-centre project between 2003 and 2005, and from a recent project in four centres between 2009 and 2013. For patients nursed with their head up at 30° and with the blood pressure transducer at atrium level, CPP was corrected by 10 mmHg. Median CPP, interquartile ranges and total CPP ranges over the monitoring time were calculated per patient and per centre. RESULTS: Per-centre medians pre-2007 were situated between 50 and 70 mmHg in 6 out of 16 centres, while 10 centres had medians above 70 mmHg and 4 above 80 mmHg. Post-2007, three out of four centres had medians between 60 and 70 mmHg and one above 80 mmHg. One out of two centres with data pre- and post-2007 shifted from a median CPP of 76 mmHg to 60 mmHg, while the other remained at 68-67 mmHg. CONCLUSIONS: CPP data are characterised by a high inter-individual variability, but the data also suggest differences in CPP policies between centres. The 2007 guideline change may have affected policies towards lower CPP in some centres. Deviations from the guidelines occur in the direction of CPP > 70 mmHg.

Visualizing Cerebrovascular Autoregulation Insults and Their Association with Outcome in Adult and Paediatric Traumatic Brain Injury.

OBJECTIVE: The aim of this study is to assess visually the impact of duration and intensity of cerebrovascular autoregulation insults on 6-month neurological outcome in severe traumatic brain injury. MATERIAL AND METHODS: Retrospective analysis of prospectively collected minute-by-minute intracranial pressure (ICP) and mean arterial blood pressure data of 259 adult and 99 paediatric traumatic brain injury (TBI) patients from multiple European centres. The relationship of the 6-month Glasgow Outcome Scale with cerebrovascular autoregulation insults (defined as the low-frequency autoregulation index above a certain threshold during a certain time) was visualized in a colour-coded plot. The analysis was performed separately for autoregulation insults occurring with cerebral perfusion pressure (CPP) below 50 mmHg, with ICP above 25 mmHg and for the subset of adult patients that did not undergo decompressive craniectomy. RESULTS: The colour-coded plots showed a time-intensity-dependent association with outcome for cerebrovascular autoregulation insults in adult and paediatric TBI patients. Insults with a low-frequency autoregulation index above 0.2 were associated with worse outcomes and below -0.6 with better outcomes, with and approximately exponentially decreasing transition curve between the two intensity thresholds. All insults were associated with worse outcomes when CPP was below 50 mmHg or ICP was above 25 mmHg. CONCLUSIONS: The colour-coded plots indicate that cerebrovascular autoregulation is disturbed in a dynamic manner, such that duration and intensity play a role in the determination of a zone associated with better neurological outcome.

Raised intracranial pressure and retinal haemorrhages in childhood encephalopathies.

AIM: To explore the relationship between raised intracranial pressure (RICP) and retinal haemorrhages in traumatic and non-traumatic childhood encephalopathies. METHOD: A prospective study of 112 children (35 females and 77 males, age range 0.01mo-17y 8.3mo; mean 5y 8.6mo, median 4y 5.6mo) included 57 accidental traumatic brain injuries (ATBIs), 21 inflicted traumatic brain injuries (ITBIs), and 34 non-traumatic encephalopathy cases. Measurements included intracranial pressure (ICP), cerebral perfusion pressure, pressure-time index of ICP, and number, zone, and layer of retinal haemorrhages on retinal imaging. RESULTS: Group I had measured elevated ICP (n=42), Group II had clinical and/or radiological signs of RICP (n=21), and Group III had normal ICP (n=49). In the combined Groups I and II, 38% had retinal haemorrhages. Multiple logistic regression confirmed that the presence of retinal haemorrhages was significantly related to the presence of RICP independent of age and aetiology; however, the occurrence and overall numbers were not significantly related to the specific ICP level. The numbers of intraretinal (nerve-fibre layer and dot blot) retinal haemorrhages were significantly greater in those with RICP. The ITBI population was significantly different from the other combined aetiological categories. INTERPRETATION: The study results indicate a complex RICP/retinal haemorrhage relationship. There was no evidence of existing retinal haemorrhages being exacerbated or new retinal haemorrhages developing during periods of confirmed RICP.

Continuous Optimal CPP Based on Minute-by-Minute Monitoring Data: A Study of a Pediatric Population.

This paper describes the use of minute-by-minute monitoring data to determine continuous optimal cerebral perfusion pressure (CPP) recommendations based on the autoregulatory status of pediatric patients with traumatic brain injury. Data from 79 children were retrospectively studied. Optimal CPP recommendations were obtained for the majority of the first 72 h of monitoring time. Actual CPP close to recommended CPP values was significantly associated with better outcome and was a significant independent predictor of better outcome when considering IMPACT model covariates in multivariate logistic regression.

Visualizing the pressure and time burden of intracranial hypertension in adult and paediatric traumatic brain injury.

PURPOSE: To assess the impact of the duration and intensity of episodes of increased intracranial pressure on 6-month neurological outcome in adult and paediatric traumatic brain injury. METHODS: Analysis of prospectively collected minute-by-minute intracranial pressure and mean arterial blood pressure data of 261 adult and 99 paediatric traumatic brain injury patients from multiple European centres. The relationship of episodes of elevated intracranial pressure (defined as a pressure above a certain threshold during a certain time) with 6-month Glasgow Outcome Scale was visualized in a colour-coded plot. RESULTS: The colour-coded plot illustrates the intuitive concept that episodes of higher intracranial pressure can only be tolerated for shorter durations: the curve that delineates the duration and intensity of those intracranial pressure episodes associated with worse outcome is an approximately exponential decay curve. In children, the curve resembles that of adults, but the delineation between episodes associated with worse outcome occurs at lower intracranial pressure thresholds. Intracranial pressures above 20 mmHg lasting longer than 37 min in adults, and longer than 8 min in children, are associated with worse outcomes. In a multivariate model, together with known baseline risk factors for outcome in severe traumatic brain injury, the cumulative intracranial pressure-time burden is independently associated with mortality. When cerebrovascular autoregulation, assessed with the low-frequency autoregulation index, is impaired, the ability to tolerate elevated intracranial pressures is reduced. When the cerebral perfusion pressure is below 50 mmHg, all intracranial pressure insults, regardless of duration, are associated with worse outcome. CONCLUSIONS: The intracranial pressure-time burden associated with worse outcome is visualised in a colour-coded plot. In children, secondary injury occurs at lower intracranial pressure thresholds as compared to adults. Impaired cerebrovascular autoregulation reduces the ability to tolerate intracranial pressure insults. Thus, 50 mmHg might be the lower acceptable threshold for cerebral perfusion pressure.

Prediction of inflicted brain injury in infants and children using retinal imaging.

BACKGROUND: Retinal hemorrhages (RHs) occur in inflicted traumatic brain injury (ITBI), accidental traumatic brain injury (ATBI), and some medical conditions, although the reported number, distribution, type, and frequency vary greatly between these different etiologies. We hypothesize that these RH characteristics reliably help to distinguish ITBI from ATBI and nontraumatic etiologies. METHODS: A 6-year prospective observational study using wide-field retinal imaging (RetCam) was conducted within 24 hours of admission to PICU, on serially recruited children with traumatic and nontraumatic encephalopathies. "Definite" and "probable" ITBI cases were confirmed by multiagency child protection case conferences. Image analysis used digital color and grayscale images for retinal "zoning" and "layering" of hemorrhages. RESULTS: Significant differences were found between the mean numbers of hemorrhages in ATBI/ITBI, and ITBI/nontraumatic etiologies for the 3 retinal zones (range, P = .003-.009) and for the dot-blot hemorrhages (range P = .001-.002). The mean numbers of RHs per ITBI patient in the peripapillary, macula, and peripheral zones were 14, 28, and 31 respectively. RHs in ATBI were near the optic disc and more superficial than in ITBI, where hemorrhages involved deeper layers (range, P = .003-.039) and were more peripheral (P = .03). The positive predictive value for ITBI in children <3 years with >25 dot-blot (intraretinal) hemorrhages was 93%. CONCLUSIONS: This prospective study, which included all potential causes of RHs, with objective retinal methodology, has confirmed that a young age and a high dot-blot count are strong predictors of ITBI. This high predictive value may support medicolegal deliberations.

Combining coma score and serum biomarker levels to predict unfavorable outcome following childhood brain trauma.

This study aims to determine if pairing the Glasgow Coma Scale (GCS) with serum biomarker levels may achieve higher outcome predictive values than using either the GCS or biomarker levels alone in childhood brain trauma. Twenty-eight critically ill children with isolated accidental brain trauma were studied in a prospective observational study. The GCS was recorded at various time points post injury. Enzyme-linked immunosorbent assay (ELISA) was used to quantify eight different serum biomarker levels (S100b, NSE, IL-6, IL-8, IL-10, L-selectin, SICAM, and endothelin) on day 1 post injury. The Glasgow Outcome Score (GOS) was used to assess global outcome at 6 months post injury. Outcome predictive values of the GCS, individual biomarker levels, and paired combinations of the GCS and biomarkers were compared using receiver operator characteristic (ROC) curve analysis and its multivariate extension, multivariate ROC curve (MultiROC). When using either the GCS or individual biomarker levels alone to predict unfavorable outcome, only the PICU discharge summated GCS achieved an area under the ROC curve (AUC) of more than 0.95. This high degree of outcome predictability was also achieved by pairing the GCS with a single biomarker level. The most pronounced improvement in outcome prediction was observed by pairing the post-resuscitation summated GCS with the day-1 serum IL-8 level, which increased the AUC from 0.78 to 0.98 and the sensitivity and specificity from 75% to 100% and 96% respectively. Paired combinations of the GCS and serum biomarker levels greatly enhanced the accuracy of post-traumatic unfavorable outcome prediction than may be achieved using either the GCS or individual biomarker levels alone.

Evaluation of pituitary function after traumatic brain injury in childhood.

OBJECTIVES: Post-traumatic hypopituitarism is well described amongst adult traumatic brain injury (TBI) survivors. We aimed to determine the prevalence and clinical significance of pituitary dysfunction after head injury in childhood. DESIGN: Retrospective exploratory study. PATIENTS: 33 survivors of accidental head injury (27 boys). Mean (range) age at study was 13·4 years (5·4-21·7 years) and median (range) interval since injury 4·3 years (1·4-7·8 years). Functional outcome at study: 15 good recovery, 16 moderate disability, two severe disability. MEASUREMENTS: Early morning urine osmolality and basal hormone evaluation were followed by the gonadotrophin releasing hormone (GnRH) and insulin tolerance (n = 25) or glucagon tests (if previous seizures, n = 8). Subjects were not primed. Head injury details were extracted from patient records. RESULTS: No subject had short stature (mean height SD score +0·50, range -1·57 to +3·00). Suboptimal GH responses (<5 µg/l) occurred in six peri-pubertal boys (one with slow growth on follow-up) and one postpubertal adolescent (peak GH 3·2 µg/l). Median peak cortisol responses to insulin tolerance or glucagon tests were 538 and 562 nm. Nine of twenty-five and two of eight subjects had suboptimal responses, respectively, two with high basal cortisol levels. None required routine glucocorticoid replacement. In three, steroid cover was recommended for moderate/severe illness or injury. One boy was prolactin deficient. Other basal endocrine results and GnRH-stimulated LH and FSH were appropriate for age, sex and pubertal stage. Abnormal endocrine findings were unrelated to the severity or other characteristics of TBI or functional outcome. CONCLUSIONS: No clinically significant endocrinopathy was identified amongst survivors of accidental childhood TBI, although minor pituitary hormone abnormalities were observed.

An interrater reliability study of a new 'zonal' classification for reporting the location of retinal haemorrhages in childhood for clinical, legal and research purposes.

BACKGROUND/AIMS: To develop and assess a zonal classification of the retina to facilitate description of the location of retinal haemorrhages in children. METHODS: A novel zonal classification of the retina was devised based on the anatomical landmarks of the optic disc and vascular arcades, by reviewing a large number of wide field digital retinal images drawn from our database of children with accidental and non-accidental head injury and other encepthalopathies. Four expert examiners then independently 'located' 142 retinal haemorrhages by zone, from 31 high quality photographs. RESULTS: Cohen's unweighted kappa scores for all possible pairs of the four raters (ie, six pairs) ranged from 0.86 to 0.92, that is 'almost perfect' agreement. Fleiss' kappa for agreement between multiple raters (four) and for multiple categories (three) was 0.8841, that is 'almost perfect' agreement. Cohen's unweighted kappa statistic for intrarater reliability gave an overall concordance that ranged from 'substantial' to 'perfect' agreement. CONCLUSION: This new retinal zone classification and the use of photographs and templates is a very reliable tool for reporting the location of retinal haemorrhages from multiple aetiologies in children, and may be useful for research and medico-legal reports.

Dynamic interactions between L-type voltage-sensitive calcium channel Cav1.2 subunits and ahnak in osteoblastic cells.

Voltage-sensitive Ca(2+) channels (VSCCs) mediate Ca(2+) permeability in osteoblasts. Association between VSCC alpha(1)- and beta-subunits targets channel complexes to the plasma membrane and modulates function. In mechanosensitive tissues, a 700-kDa ahnak protein anchors VSCCs to the actin cytoskeleton via the beta(2)-subunit of the L-type Ca(v)1.2 (alpha(1C)) VSCC complex. Ca(v)1.2 is the major alpha(1)-subunit in osteoblasts, but the cytoskeletal complex and subunit composition are unknown. Among the four beta-subtypes, the beta(2)-subunit and, to a lesser extent, the beta(3)-subunit coimmunoprecipitated with the Ca(v)1.2 subunit in MC3T3-E1 preosteoblasts. Fluorescence resonance energy transfer revealed a complex between Ca(v)1.2 and beta(2)-subunits and demonstrated their association in the plasma membrane and secretory pathway. Western blot and immunohistochemistry showed ahnak association with the channel complex in the plasma membrane via the beta(2)-subunit. Cytochalasin D exposure disrupted the actin cytoskeleton but did not disassemble or disrupt the function of the complex of L-type VSCC Ca(v)1.2 and beta(2)-subunits and ahnak. Similarly, small interfering RNA knockdown of ahnak did not disrupt the actin cytoskeleton but significantly impaired Ca(2+) influx. Collectively, we showed that Ca(v)1.2 and beta(2)-subunits and ahnak form a stable complex in osteoblastic cells that permits Ca(2+) signaling independently of association with the actin cytoskeleton.

Pediatric brain trauma outcome prediction using paired serum levels of inflammatory mediators and brain-specific proteins.

Many potential brain trauma biomarkers have been reported, but no previous study has described outcome prediction using combinations of biomarker levels. We aimed to investigate the outcome predictive values of multiple biomarkers from different mediator families and to determine whether combinations of two serum biomarkers may achieve higher outcome predictive values than individual biomarker levels. A prospective observational study was conducted involving 28 children requiring intensive care management following brain trauma. Day 1 post-injury serum concentrations of eight different biomarkers--S100b protein (S100b), neuron-specific enolase (NSE), interleukin-6 (IL-6), interleukin-8 (IL-8), interleukin-10 (IL-10), soluble intracellular adhesion molecule (SICAM), L-selectin, and endothelin--were quantified using enzyme-linked immunosorbent assay (ELISA). Global outcome was assessed at 6 months post-injury using the Glasgow Outcome Score (GOS). Receiver operator characteristic curve (ROC) analysis and its multivariate extension, Multivariate ROC (MultiROC), were used to assess the outcome predictive values of the individual and the paired biomarkers. None of the eight biomarkers assessed individually achieved an area under the ROC curve (AUC) of more than 0.95 for predicting unfavorable outcome, but five of the 20 biomarker pairs assessed had this high degree of outcome predictability. Two combinations using S100b as the "screening marker" and either L-selectin or IL-6 as the "varying marker" achieved an AUC of 0.98, and their specificity and sensitivity for unfavorable outcome prediction were 96% and 100%, respectively. Prognostic pairs combining serum levels of two biomarkers (inflammatory mediators and brain-specific proteins) offer better outcome predictive values for unfavorable outcome after childhood brain trauma than may be achieved using individual marker levels.

The role of high frequency oscillatory ventilation in the management of children with severe traumatic brain injury and concomitant lung pathology.

OBJECTIVE: To report the use of high frequency oscillatory ventilation (HFOV) in two children with severe traumatic brain injury and concurrent lung pathology where conventional mechanical ventilation was ineffective. DESIGN: : Case report. SETTING: Regional intensive care unit in a pediatric teaching hospital. PATIENTS: Two severely head-injured children (both with postresuscitation Glasgow Coma Scores of 3), one of whom was age 11 yrs and developed an invasive fungal (rhizomucor) pneumonia, while the other age 5 yrs had bilateral lung contusions. Both were treated according to local head injury guidelines, which included conventional ventilation. Despite increasing conventional ventilatory support, CO2 removal became problematic in both cases, making the intracranial pressure control and consequent maintenance of adequate cerebral perfusion pressure difficult. In both patients, a dramatic reduction in intracranial pressure and improvement in cerebral perfusion pressure was observed soon after the use of HFOV. Additionally, inotropic support was weaned by 50% in both children after commencing HFOV. A significant increase in the mean arterial blood pressure occurred in one child with HFOV. INTERVENTION: Use of HFOV as an alternative to conventional mechanical ventilation. CONCLUSION: HFOV may have utility in the management of selected cases of severe brain trauma with concurrent lung pathology where conventional ventilation is ineffective.

Incidence and demography of non-accidental head injury in southeast Scotland from a national database.

BACKGROUND: This study utilized an existing national database of cases of non-accidental head injury (NAHI; also called inflicted traumatic brain injury [inflicted TBI] and shaken baby syndrome [SBS]) in Scotland to report the incidence, confidence intervals, and demography of such cases in Southeast Scotland. METHODS: This prospective population-based study was conducted from January 1998 to September 2006. Data from the Lothian region of Scotland, where there is known full ascertainment of infant head injuries, including NAHI, have been used to calculate the incidence rate for this region of Scotland, with government statistics providing the normal annual infant population as the denominator. A new Scottish Index of Multiple Deprivation (SIMD), which assesses a very focused area (data zone population size=750) and provides novel information about social demography for education, housing, employment, health, crime, income, and geographic accessibility to services, was applied to the identified cases of NAHI during this study period. RESULTS: The mean incidence of NAHI in southeast Scotland for 8.75 years was 33.8/100,000 infants per year. The cases of NAHI were mostly located in the lowest 1 (or 2) quintiles for all SIMD domains (education, housing, employment, health, crime, income), although they had good accessibility to medical and other community services. CONCLUSIONS: The incidence rates from this prospective study for NAHI are considerably higher than other published UK surveys and are not considered to reflect a cluster effect. The perpetrators in this study fit a strongly skewed profile aggregating to the lowest socioeconomic groups in the community.

Listener perception of respiratory-induced voice tremor.

PURPOSE: The purpose of this study was to determine the relation of respiratory oscillation to the perception of voice tremor. METHOD: Forced oscillation of the respiratory system was used to simulate variations in alveolar pressure such as are characteristic of voice tremor of respiratory origin. Five healthy men served as speakers, and 6 clinically experienced women served as listeners. Speakers produced utterances while forced sinusoidal pressure changes were applied to the surface of the respiratory system. Utterances included vowels and sentences produced using usual loudness, pitch, quality, and rate, and vowels produced using different loudness, pitch, and quality. Perceptual tasks included detection threshold for voice tremor and pair comparison judgments in which listeners identified the sample with the greater magnitude of voice tremor. RESULTS: The mean detection threshold for voice tremor was 1.37 cmH(2)O (SD = 0.47) for vowel utterances and 2.16 cmH(2)O (SD = 1.52) for sentence utterances. Tremor magnitude was judged to be different for vowel and sentence utterances, but not for different vowels. Results revealed differential effects for loudness, pitch, and quality. CONCLUSIONS: These findings offer implications for the evaluation and management of voice tremor of respiratory causation.

Pressure reactivity as a guide in the treatment of cerebral perfusion pressure in patients with brain trauma.

OBJECT: The aim of this study was to compare the effects of two different treatment protocols on physiological characteristics and outcome in patients with brain trauma. One protocol was primarily oriented toward reducing intracranial pressure (ICP), and the other primarily on maintaining cerebral perfusion pressure (CPP). METHODS: A series of 67 patients in Uppsala were treated according to a protocol aimed at keeping ICP less than 20 mm Hg and, as a secondary target, CPP at approximately 60 mm Hg. Another series of 64 patients in Edinburgh were treated according to a protocol aimed primarily at maintaining CPP greater than 70 mm Hg and, secondarily, ICP less than 25 mm Hg for the first 24 hours and 30 mm Hg subsequently. The ICP and CPP insults were assessed as the percentage of monitoring time that ICP was greater than or equal to 20 mm Hg and CPP less than 60 mm Hg, respectively. Pressure reactivity in each patient was assessed based on the slope of the regression line relating mean arterial blood pressure (MABP) to ICP. Outcome was analyzed at 6 months according to the Glasgow Outcome Scale (GOS). The prognostic value of secondary insults and pressure reactivity was determined using linear methods and a neural network. In patients treated according to the CPP-oriented protocol, even short durations of CPP insults were strong predictors of death. In patients treated according to the ICP-oriented protocol, even long durations of CPP insult-mostly in the range of 50 to 60 mm Hg--were significant predictors of favorable outcome (GOS Score 4 or 5). Among those who had undergone ICP-oriented treatment, pressure-passive patients (MABP/ICP slope > or = 0.13) had a better outcome. Among those who had undergone CPP-oriented treatment, the more pressure-active (MABP/ICP slope < 0.13) patients had a better outcome. CONCLUSION: Based on data from this study, the authors concluded that ICP-oriented therapy should be used in patients whose slope of the MABP/ICP regression line is at least 0.13, that is, in pressure-passive patients. If the slope is less than 0.13, then hypertensive CPP therapy is likely to produce a better outcome.

Predicting recovery in patients suffering from traumatic brain injury by using admission variables and physiological data: a comparison between decision tree analysis and logistic regression.

OBJECT: Decision tree analysis highlights patient subgroups and critical values in variables assessed. Importantly, the results are visually informative and often present clear clinical interpretation about risk factors faced by patients in these subgroups. The aim of this prospective study was to compare results of logistic regression with those of decision tree analysis of an observational, head-injury data set, including a wide range of secondary insults and 12-month outcomes. METHODS: One hundred twenty-four adult head-injured patients were studied during their stay in an intensive care unit by using a computerized data collection system. Verified values falling outside threshold limits were analyzed according to insult grade and duration with the aid of logistic regression. A decision tree was automatically produced from root node to target classes (Glasgow Outcome Scale [GOS] score). Among 69 patients, in whom eight insult categories could be assessed, outcome at 12 months was analyzed using logistic regression to determine the relative influence of patient age, admission Glasgow Coma Scale score, Injury Severity Score (ISS), pupillary response on admission, and insult duration. The most significant predictors of mortality in this patient set were duration of hypotensive, pyrexic, and hypoxemic insults. When good and poor outcomes were compared, hypotensive insults and pupillary response on admission were significant. Using decision tree analysis, the authors found that hypotension and low cerebral perfusion pressure (CPP) are the best predictors of death, with a 9.2% improvement in predictive accuracy (PA) over that obtained by simply predicting the largest outcome category as the outcome for each patient. Hypotension was a significant predictor of poor outcome (GOS Score 1-3). Low CPP, patient age, hypocarbia, and pupillary response were also good predictors of outcome (good/poor), with a 5.1% improvement in PA. In certain subgroups of patients pyrexia was a predictor of good outcome. CONCLUSIONS: Decision tree analysis confirmed some of the results of logistic regression and challenged others. This investigation shows that there is knowledge to be gained from analyzing observational data with the aid of decision tree analysis.