Pathophysiology of traumatic brain injury

Get Complete Project Material File(s) Now! »

Effect of intensity of care on mortality and withdrawal of life-sustaining therapies in severe traumatic brain injury patients: A post-hoc analysis of a multicenter cohort study

Peter Gerges, Lynne Moore, François Lauzier, Ryan Zarychanski, Caroline Léger, Michèle Shemilt, Xavier Neveu, Alexis F. Turgeon

Introduction

Traumatic brain injury is a major global health problem. An average of 1.4 million cases of traumatic brain injury occur in the United States every year with 50,000 deaths (1). The overall care provided to patients, including traumatic brain injury population, can be defined by its quality as well as its quantity or intensity. Despite the improvement in the quality of care, mortality following severe traumatic brain injury is still high. The case fatalities of severe traumatic brain injuries is averaging at 39% (2) without a substantial improvement in the recent decades (3-10) that matches with the concurrent improvement of quality of care. This raises the question about the impact of other determinants of care, such as the intensity of the care, on outcomes.
Intensity of care is shown to be associated with mortality in the general ICU population and, potentially, in the overall non-surgical and neurocritically ill population (11, 12). No study to date has quantified the intensity of care in severe traumatic brain injury. However, a systematic review showed that trauma centers providing aggressive treatment and monitoring of severe traumatic brain injury patients showed improved neurological outcome and reduced mortality (13). On the other hand, mortality in severe traumatic brain injury was found to vary among different trauma centers, as well as decisions to withdraw life-sustaining therapies (14). These findings may highlight a potential association of intensity of care with mortality in relation with decisions to withdraw life sustaining therapies in the specific severe traumatic brain injury population.
Our study aimed to evaluate the effect of the intensity of care on the withdrawal of life-sustaining therapies and mortality in critically ill patients with severe traumatic brain injury.

Methods

Our study is a post-hoc analysis of a large multicenter retrospective cohort study of critically ill patients with severe traumatic brain injury (14, 15). This study was conducted in six level I trauma centers from three Canadian provinces (Québec, Ontario, Alberta) and covered a 24-month period from January 2005 to December 2006 (n = 720) (14, 15). Patients were identified at all centres using the International Statistical Classification of Diseases and Related Health Problems, 10th revision (codes for traumatic brain injury: S06.0–S06.9). Seven hundred and twenty patients were included. We included patients who were 1) ≥ 16 years old 2) mechanically ventilated for ≥ 48hrs, 3) admitted to an ICU following a blunt severe traumatic brain injury with a Glasgow Coma Scale (GCS) score ≤8. We excluded patients with penetrating brain injuries and those with no record of intensive care unit stay. Trained research assistants reviewed the charts and retrieved data at each centre using a standardized case report form. Research Ethics Board approval from all participating institutions was obtained.

Intensity of care

We defined the intensity of care using interventions performed during ICU stay. We divided these interventions into two main categories: 1) traumatic brain injury related interventions, 2) interventions non-specific to the traumatic brain injury, as well as into three mutually exclusive categories of type of interventions: 1) medical, 2) surgical, and 3) diagnostic interventions (Table 1). Interventions were measured over the first 14 days of stay in the intensive care unit.

Outcome measures

Our outcome measures were hospital mortality and the incidence of withdrawal of life-sustaining therapies.

Sample size

We retrospectively collected data from six Canadian trauma centers and included 720 severe traumatic brain injury patients in the main cohort (14) following the random selection of 120 patients from each center. We retrospectively identified patients from trauma registries and/or discharge databases at all centers using the International Statistical Classification of Diseases and Related Health Problems, 10th revision (codes for traumatic brain injury: S06.0–S06.9). Four patients were excluded due to missing discharge time or censoring values. Therefore, we included 716 patients in our analyses.

Data collection

We extracted the following baseline patient characteristics: age, sex, GCS motor score, pupillary reactivity, cause of trauma, injury severity score and other associated traumatic injuries, as well as the types of interventions performed, withdrawal of life-sustaining therapies and death. (Table 1).

Statistical Analysis

We used descriptive statistics for continuous variables normally distributed using the student t test and the Wilcoxon test for those non-normally distributed. We used the Chi-square and Fisher’s exact tests for categorical variables. Continuous variables were expressed with mean ± standard deviation (SD) or median and 25-75% interquartile range (IQ25-75) for ordinal variables.
The effect of intensity of care, on mortality and the withdrawal of life-sustaining therapies, was evaluated with adjusted Cox proportional-hazards regression analyses of time-to-event data. The intensity of care was operationalized by creating a composite measure aggregating the contributing interventions (Table 1). The use of each of the contributing interventions was measured on daily basis and was treated as time-varying covariates using counting process formulation (16) to account for the timing of interventions relative to the event. Cox proportional hazard models were adjusted for sex and three basic confounding factors, measured on admission, and known to be associated with prognosis of severe traumatic brain injury: age (dichotomized at 55 years) (17-21), absence of pupillary reactivity (7, 22-25) and motor score on the Glasgow Coma Scale (GCS) (7, 23-27). Motor GCS score (28, 29) was treated in three categories; category 1 (GCS score 1), category 2 (score 2-3) and category 3 as the reference category (score 4-6) (15, 30). We used hierarchical modeling (31) to control for the possible clustering effect from trauma centers.
We conducted sensitivity analyses to assess whether there were variations across intensive care unit stay (3 day, 7 day, discharge) as well as for the type of interventions. We reported data with hazard ratios and corresponding 95% confidence intervals (95% CI). The analyses were performed using SAS version 9.4 software (SAS Institute, Cary, North Carolina, USA).

Results

A total of 716 severe traumatic brain injury patients were included in our analyses. Baseline characteristics of the cohort are presented in Table 2. Most patients were male (77.1%) with a mean age of 42.4 (SD 20.5) years and a median GCS was 3 (IQR25- 75%: 3–6). Hospital mortality was 31.8 % (n= 228) with most deaths (70.2%) in association with a decision to withdraw life-sustaining therapies.

Mortality

Adjusted Cox proportional hazards analyses showed suggested that increased intensity of care was associated with a decreased mortality (HR 0.69, 95% CI 0.63–0.74, p value < 0.0001) (Table 3). This association with a decreased mortality was observed with interventions related to the traumatic brain injury (HR 0.70, 95% CI 0.58–0.83, p < 0.0001) as well as with those not specifically related to traumatic brain injury (HR 0.68, 95% CI 0.61–0.76, p < 0.0001). When analyzed by type interventions, medical interventions (HR 0.689, 95% CI 0.629- 0.755, P < 0.0001) and diagnostic procedures (HR 0.573, 95% CI 0.395–0.832, P= 0.0034) were significantly associated with reduction of mortality hazards, whereas the association was not significant with surgical interventions (HR 0.760, 95% CI 0.571–1.012, p value 0.0601). Similar associations were observed regardless of the timing of outcome assessment (Table 4).

Withdrawal of life-sustaining therapies

Adjusted Cox proportional hazards analyses showed a significant decrease in the hazards of withdrawal of life sustaining therapies in the overall cohort with the increase of the overall intensity of care (HR 0.73, 95% CI 0.67–0.79, p < 0.0001) (Table 5). This reduction of hazards was observed for both the intensity of interventions directly related to traumatic brain injury (HR 0.74, 95% CI 0.60–0.90, p 0.0033) and for interventions not specifically related to traumatic brain injury (HR 0.72, 95% CI 0.63– 0.82, p < 0.0001). When analyzed by type interventions, intensity of medical interventions (HR 0.74, 95% CI 0.67- 0.82, p < 0. 0001) and diagnostic procedures (HR 0.58, 95% CI 0.37–0.90, p 0.0148) were significantly associated with reduction of the hazards of withdrawal of life-sustaining therapies, whereas the association was not significant with surgical interventions (HR 0.73, 95% CI 0.50-1.06, p 0.0982). When looking at withdrawal of life-sustaining therapies in non-survivors only, similar findings were observed, as well as for the sensitivity and subgroup analyses (Table 6 and 7 respectively).

Discussion

In our study, we observed that the intensity of care provided to severe traumatic brain injury patients is associated with a lower mortality and incidence of withdrawal of life-sustaining therapies. This association was observed regardless if the intensity of care was directly in relation with the traumatic brain injury or not, as well as on the type of interventions or procedures except for surgical interventions. This association of intensity of care with outcome was maintained throughout the intensive care unit stay, regardless of the timing of assessment.
No previous study specifically addressed the impact of intensity of care in the traumatic brain injury population between and across centers at a patient level. One previously published study aimed at understanding the effect intensity of care on mortality following severe traumatic brain injury. The results of this study suggested that a decreased intensity of care in the elderly patients was a possible cause for greater mortality (32). In our study, we controlled and adjusted for age and observed that this inverse relationship between intensity of care and mortality was not restricted to the elderly but present in the whole population. A systematic review with meta-analyses of cohort studies examined the effect of aggressive monitoring (mainly intracranial pressure monitoring) on mortality in patients with severe traumatic brain injury (13). Centers considered using aggressive monitoring had a lower mortality. Such observations were observed consistently in the literature. In our study, all centers were using such type of aggressive monitoring as part of the standard of care and thus allowed to address the question at a patient level by adjusting for a potential center effect using hierarchical modeling (31). Our study was also designed to quantify the intensity of care as a continuous time varying data. Interestingly, we did not observe an effect of surgical interventions, including intracranial pressure monitoring, on mortality and the incidence of life-sustaining therapies. This finding suggests that the effect observed in previous literature may not have been related to the aggressive monitoring, but to the overall greater aggressive care and secondary lower incidence of withdrawal of life-sustaining therapies in these centers.
One important limitation of our study is the absence of long-term functional outcome measures that are considered standard of care to evaluate patients with traumatic brain injury. Nonetheless, the evaluation of the short-term effects of intensity of care on mortality and incidence of withdrawal of life-sustaining therapies is clinically relevant considering that these events occur early in the care of severe traumatic brain injury patients. Second, we used data from a study not originally designed to evaluate the impact of intensity of care on outcome measures. We were then limited to the first 14 days of care for measuring intensity of care. We could however argue that the most active phase of care is in the first week of injury and that most deaths and decisions to withdraw life-sustaining therapies occur before this two-week period. Third, we quantified the intensity of care by assigning an equal weight to the different interventions; our study did not intend to evaluate the relative effects of these interventions, but the overall cumulative effect of the intensity of care defined by these interventions. Some interventions may play a greater role in the observed associations than we may think as opposed to others. Fourth, the most important limitation of our study may be the presence of residual confounding that may explain the associations with decreased mortality and incidence of withdrawal of life-sustaining therapies observed, or part of it. Despite having adjusted for the most important prognostic indicators, we cannot exclude that subjective decisions for performing these interventions, such as being more aggressive in caring for patients believed to have a more favorable prognosis, may explain these results.
On the other hand, our study used data collected from six trauma centers across Canada offering a broad representation of potential current practice variation. More so, our analyses adjusted using hierarchical modeling to account for centers, allowing analyzing data at a patient level. Our study has other methodological strengths including taking into consideration time-to-event as well as the timing of exposure. In our analyses, we used survival analyses and Cox proportional hazards with time-varying covariate model rather than using commonly used simple cross-sectional analyses. The consideration of incidence of withdrawal of life-sustaining therapies as an outcome is another strength of our study considering that more death following severe traumatic brain injury are associated with these decisions.

Conclusion

In this post-hoc analysis of a large retrospective multicenter cohort study, we observed a significant association between the overall intensity of care, defined by the different interventions and procedures commonly used, on mortality and on the incidence of withdrawal of life-sustaining therapies in critically ill patients with severe traumatic brain injury. This association was present whether interventions were specific or not specific to the traumatic brain injury, as well as whether they were medical or diagnostic procedures. The intensity of surgical interventions was not observed to be associated with these outcomes. However, we cannot exclude important potential confounding, such as a bias by indication, considering that patients with more unfavorable prognosis may not benefit from the same aggressive care and interventions during the acute phase of care. Future research should aim at understanding how decisions for high intensity of care are made in critically ill patients with severe traumatic brain injury to better control these important residual confounding, clarify the impact on clinically significant outcomes and better allocate resources.

Table of contents :

1 Chapter one: Introduction
1.1 Definition of traumatic brain injury
1.2 Pathophysiology of traumatic brain injury
1.2.1 Mechanism of traumatic brain injury
1.2.2 Primary traumatic brain injury
1.2.3 Secondary traumatic brain injuries
1.3 Classification of traumatic brain injury severity
1.3.1 Loss of consciousness
1.3.2 Post-traumatic amnesia
1.3.2 Glasgow Coma Scale
1.4 Epidemiology of traumatic brain injury
1.4.1 Traumatic brain injury incidence and prevalence
1.4.2 Traumatic brain injury contributory factors
1.4.3 Causes of traumatic brain injuries
1.4.4 Consequences of traumatic brain injuries
1.4.5 Morbidity and disability of traumatic brain injuries
1.4.6 Mortality in traumatic brain injury
1.5 Withdrawal of life-sustaining therapies in traumatic brain-injured patients
1.5.1 Variability in withdrawal of life-sustaining therapies
1.5.2 Factors associated with the withdrawal of life-sustaining therapies in traumatic brain-injured patient
1.6 Intensity of care
1.6.1 Definition of intensity of care
1.6.2 Intensity of care for traumatic brain-injured patients
1.6.3 Components of care for severe traumatic-brain injured patients
Medical component of care
Surgical component of care
Diagnostic component of care
1.7 Study rationale and objectives
Chapter two: Manuscript
Effect of intensity of care on mortality and withdrawal of life-sustaining therapies in severe traumatic brain injury patients: A post-hoc analysis of a multicenter cohort study
Résumé
Abstract
Introduction
Methods
Intensity of care
Outcome measures
Sample size
Data collection
Statistical Analysis
Results
Mortality
Withdrawal of life-sustaining therapies
Discussion
Conclusion
References
Chapter three: Conclusion
Appendices
References

GET THE COMPLETE PROJECT