SCCM Resource Library-Temporal Patterns in Brain Tissue Oxygenation Associated With Mortality After Severe TBI in Children

Video Transcription

Good afternoon. My name is Jay Rucker and I'm a Pediatric Intensivist at the Children's Hospital of Pittsburgh and a T32 Scholar at the Sopper Center. Today, I'll be sharing my  research with you via this presentation. TBI is a leading cause of mortality and morbidity in  children. Treatment is aimed at optimizing intercranial pressure, brain tissue oxygenation,  as well as cerebral perfusion pressure. Brain tissue hypoxia is known to be an independent risk factor for unfavorable outcomes in TBI. Measuring and modulating partial pressure of  oxygen in the brain, or PbO2, is aimed at limiting brain tissue hypoxia. However, evidence supporting this practice is fairly limited. The current Brain Trauma Foundation  guidelines provide only a level three recommendation for monitoring brain tissue oxygenation. Additionally, the treatment thresholds in pediatric patients are also  less well defined, with a recommended threshold of more than 10 millimeters of mercury in adults  and a similar cutoff suggested in children. There are two prior studies, as shown on the right side  of the screen, that support a threshold in pediatric severe TBI of PbO2 of greater than  20 millimeters of mercury, and more specifically 28 to 30 millimeters of mercury, and this higher  threshold has been associated with improved outcomes in adult neurosurgical patients. For  clinical management of severe pediatric TBI, we utilize the approach shown in this figure. For all  patients, we optimized cerebral perfusion pressure to age-based targets for our institutional protocol  and made no changes in clinical care when PbO2 was greater than 25 millimeters of mercury and  ICP was less than 20 millimeters of mercury. When there was evidence of isolated brain tissue  hypoxia, which we defined as PbO2 less than or equal to 25 for our institutional protocol, which  is a higher threshold than what is recommended in the guidelines, we instituted mechanical ventilation  changes to alleviate this insult. In those scenarios, when both the PbO2 and ICP exceeded  treatment thresholds, we utilized a combination of mechanical ventilation changes and also provided  ICP-directed medical therapies. In addition to investigating brain tissue hypoxia in children  with severe TBI, we also explored the relationship between brain tissue oxygenation and the partial  pressure of arterial oxygen, shown here as the PbO2 to PaO2 ratio. This ratio identifies that  normal diffusion into the brain where lack of increase in PbO2 with increasing PaO2 potentially  highlights altered oxygen delivery mechanism, emphasizing the lack of PaO2 responsiveness.  This phenotype could be secondary to hypoperfusion, altered oxygen diffusion, or abnormal cellular  metabolic rate of oxygen consumption. Additionally, systemic hyperoxemia encountered in the prevention and or response to brain tissue hypoxia may also impact risk of mortality.  To that effect, we aimed to identify temporal patterns of PbO2, PaO2, and PbO2 to PaO2 ratio  in children with severe TBI and also assess their associations with in-hospital mortality.  We conducted a retrospective cohort study using electronic health record data from a coronary care pediatric intensive care unit at a level one trauma center. Our study population  was made up of severe TBI patients and restricted to those 18 years of age or younger who had ICP  and brain tissue oxygenation recorded. We followed standardized management protocol in  accordance with the published TBI guidelines. For statistical analyses, we developed time series  regressions using either the LOWs or the GAM methods depending on the sample size. We constructed  both univariate and multivariable logistic regressions with respect to mortality. And finally, we determined optimal cut points that dichotomized mortality based on maximizing  Uden's index, which is a single statistic that captures the overall performance of the model.  This table highlights the demographics of the entire study population. In total, there were 138  pediatric ICP monitored TBI patients with 71 having a PbO2 monitor contributing to around 3,400 hours  of brain tissue oxygenation data. The overall mortality rate in our cohort was 18 percent.  In the PbO2 monitored group, the patients were older, had lower GCS score, and underwent earlier  invasive intracranial instrumentation. When dichotomizing the PbO2 monitored group with respect to mortality, those that died had higher admission PLAT2 scores, which represents the  severity of organ dysfunction, and this group also had overall shorter hospital length of stays.  The temporal trends of PbO2 in monitored TBI patients over the first five days of hospitalization  are shown on this slide. The top left panel depicts the hourly PbO2 values and the bottom left panel  highlights the daily medians. There was a general trend supporting higher PbO2 values in patients  who survived hospitalization. On all five days except day two, there were significant differences  in PbO2 values in survivors versus those that did not survive their hospitalization. Evaluating in an aggregate manner, over the first five days of hospitalization, there were  significant differences in median PbO2 values between patients who survived versus those that  died. On this slide, you can see temporal trends in partial pressure of arterial oxygen or PaO2.  The top left panel depicts the hourly PaO2 values in those with a PbO2 monitor and demonstrates an  inverted U-shaped association with mortality from days one through three. The bottom left panel  highlights the daily medians. There is evidence of higher PaO2 values in patients that died versus  survived to hospital discharge on day two and a trend towards lower PaO2 values on days four  and five. Here you can see the aggregate PaO2 values over the first five days of hospitalization.  This graph shows that there were differences that were significant between patients who survived  and who died within the PbO2 monitored cohort. No similar differences exist in PaO2 values in patients without PbO2 monitor. Temporal trends in PbO2 to PaO2 ratio are shown here.  Over the first five days of hospitalization, there was a general trend towards lower values  for PbO2 to PaO2 ratios and those that did not survive to hospital discharge. Specifically,  when stratified by day of hospitalization, there were statistical significant differences on days  one, two, and five with survivors achieving higher ratios than non-survivors. The composite  median shown on the right of PbO2 to PaO2 ratios over the first five days of hospitalizations were  also significantly different when stratified by mortality as shown. Using Yudin's method, we found the optimal thresholds that best dichotomize mortality to be 30 millimeters  of mercury for brain tissue oxygenation and 0.12 for the PbO2 to PaO2 ratio. Next, we developed the multivariable logistic regression model shown here. This model included age, sex,  race, and admission GCS score and ultimately identified an association with the presence of a PbO2 monitor and in-hospital survival. A separate univariate logistic regression model  shown here evaluated the impact of different clinical characteristics on in-hospital mortality.  We found median PbO2 of less than 30, PaO2 of greater than or equal to 300 millimeters of  mercury, PaO2 less than 80 millimeters of mercury, and PbO2 to PaO2 ratio of less than 0.12 were  each associated with in-hospital mortality. In this single center study, we retrospectively analyzed temporal trends in brain tissue oxygenation as well as systemic oxygenation.  Our time series analysis showed lower PbO2 values on multiple days as well as in an aggregate manner  among non-survivors. An hourly PbO2 threshold value of 30 millimeters of mercury accurately  dichotomized mortality. This threshold is significantly higher than the generally suggested  PbO2 threshold of 10 millimeters of mercury for the most recent Brain Trauma Foundation guidelines.  This investigation also showed a higher PbO2 to PaO2 ratio in survivors on days 1, 2, and 5  and corroborates similar findings in literature published recently. We also highlight that PbO2  to PaO2 ratio of less than 0.12 and PaO2 greater than or equal to 300 millimeters of mercury  were associated with mortality, which may signify that brain tissue hypoxia refractory to systemic  hyperoxemia is a significant adverse physiological finding. There are some important limitations,  mainly our study being a retrospective single center investigation with low mortality rate and lacking an external validation. Also, we did not investigate any morbidity measures of  long-term outcomes. In conclusion, we promote the use of these clinical thresholds to optimize  outcomes for children with severe traumatic brain injury. Our data should also provide additional  support for modification of the guidelines, not only for the threshold but also in support of concomitant monitoring of PaO2 and PbO2 to PaO2 ratio in children with severe TBI.  Further studies proactively integrating these findings for clinical decision making in pediatric TBI appear warranted.

Video Summary

Pediatric traumatic brain injury (TBI) is a leading cause of mortality and morbidity in children. Treatment aims to optimize brain tissue oxygenation and cerebral perfusion pressure, but guidelines on monitoring brain tissue oxygenation are limited. This study analyzed the relationship between brain tissue oxygenation and mortality in pediatric TBI patients. The findings showed that lower brain tissue oxygenation values were associated with higher mortality rates. The study suggests that a brain tissue oxygenation threshold of 30 mmHg, higher than the current guidelines, may be more accurate in predicting mortality. The study also highlights the importance of monitoring arterial oxygen levels in conjunction with brain tissue oxygenation.

Asset Subtitle

Neuroscience, Trauma, Pediatrics, 2022

Asset Caption

INTRODUCTION: Brain tissue hypoxia is an independent risk factor for unfavorable outcomes in traumatic brain injury (TBI). Current guidelines provide only a level III recommendation and target a time-independent partial pressure of oxygen in brain tissue (PbtO2) threshold of 10 mmHg in children with severe TBI. We aimed to determine if temporal patterns of PbtO2 were associated with mortality in severe pediatric TBI and to compute the PbtO2 cut-off value that optimally dichotomized mortality. METHODS: Ten years of data from (1/2009-4/2019) were extracted from the electronic medical record of a children's hospital with a level 1 trauma center for patients ≤18 years old with severe TBI and the presence of PbtO2 and/or intracranial pressure (ICP) monitors. Data were summarized with descriptive statistics and temporal analyses performed for the first 5 days of hospitalization. Multivariable logistic regression was used to evaluate the impact of different patient and clinical characteristics on in-hospital mortality. RESULTS: 138 ICP-monitored TBI patients (7.0±5.7y; 65% male; admission Glasgow Coma Scale [GCS] score 4 [3-7]; mortality 18%), 71 with and 67 without PbtO2 monitors were included. Time-series analyses showed lower PbtO2 values (days 1, 3-5) and lower PbtO2/PaO2 ratios (days 2, 5) in patients that died vs. survived. A PbtO2 of 30 mmHg and PbtO2/PaO2 of 0.12 were identified by Youden’s method when modeled with a mortality outcome. Patients with vs. without PbtO2 monitors had higher PaO2 values and those that died had evidence of relative hyperoxemia. Multivariable logistic regression identified older age, lower admission GCS score, PbtO2 < 30 mmHg, hyperoxemia (PaO2 ≥ 300 mm Hg), hypoxemia (PaO2 < 80 mmHg), and PbtO2/PaO2 ratio < 0.12 to be independently associated with mortality. CONCLUSIONS: We identified optimal cut-off values for PbtO2 threshold and PbtO2/PaO2 ratio to discriminate in-hospital mortality in severe TBI patients in our center. Our results corroborate our prior work that suggests targeting a higher PbtO2 threshold than recommended in the Guidelines, and the need to evaluate the potential utility of the PbtO2/PaO2 ratio to manage severe pediatric TBI. Further studies using proactively targeted PbtO2 values and PbtO2/PaO2 ratios for clinical decision-making in pediatric TBI are warranted.

Meta Tag

Content Type Presentation

Knowledge Area Neuroscience

Knowledge Area Trauma

Knowledge Area Pediatrics

Knowledge Level Advanced

Membership Level Select

Tag Traumatic Brain Injury TBI

Tag Pediatrics

Tag Burns

Tag Tissue Oxygenation

Year 2022

Keywords

Pediatric traumatic brain injury

brain tissue oxygenation

mortality

cerebral perfusion pressure

arterial oxygen levels

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.