Special Feature: ICU Management of Patients With Traumatic Brain Injury
Special Feature
ICU Management of Patients With Traumatic Brain Injury
By Grant E. O’Keefe, MD
Over the past decade, the treatment of traumatic brain injury (TBI) has evolved substantially with the development, application, and evaluation of treatment strategies and newly available technologies. These new approaches bring the potential for a reduction of morbidity and mortality attributable to TBI. However, there are few data to support the efficacy of many of the treatment strategies for TBI. Through a joint initiative, Guidelines for the Management of Traumatic Brain Injury was published in 1995 and can be viewed at the Brain Trauma Foundation website (http://www.braintrauma.org/guidelines.nsf). The objective of this work was to "clearly state the scientific basis" for the clinical management of TBI. While it is not possible to review all aspects of the care of these complex and severely injured patients in this review, I will summarize some present areas of interest and controversy. In addition, the Injuries Group of the Cochrane Collaboration has reviewed a number of components of the critical care of patients with TBI. Both of these sources provide literature-based and expert opinion regarding many aspects of the care of patients with TBI and can be referred to in addition to the references cited in this review.
Secondary Brain Injury and the Role of General Supportive Care
The prognosis following TBI appears to have improved over the past few decades, with a greater proportion of severe TBI patients surviving than in previous years, importantly without an increase in those with severe disabilities.1,2 However, most of these improvements came prior to the last decade, suggesting that the more recent technological advances and increased understanding of the pathophysiology of primary and secondary brain injury have not had a measurable effect upon survival.1 Improvements in outcome have likely been the consequence of more rapid diagnosis of intracranial space-occupying lesions with cranial computerized tomography and recognition of the importance of diagnosing and treating elevated intracranial pressure (ICP).2,3,4 It is also clear that other secondary insults (hypotension, hypoxia, seizures), in addition to intracranial hypertension, negatively impact upon outcome, and much of the supportive data originated with reports based on the large Traumatic Coma Databank.
Systemic hypotension, hypoxemia, and other potential subsequent insults occurring in the post-injury period have been shown to be major determinants of survival and neurological outcome.5 Most notably, systemic hypotension, whether occurring prior to hospital arrival, in the operating room, or in the ICU, results in secondary brain injury and markedly increases mortality.6,7,8 Therefore, the strict avoidance of hypotension must be considered important in the overall support of TBI patients. Hypoxemia has been found to contribute independently to mortality, but not to the same degree as systemic hypotension.7 The effect of a more aggressive approach to airway management is uncertain. It has been suggested that endotracheal intubation in the field improves survival in patients with severe TBI.9 However, the absolute differences in survival were not great for patients intubated in the field, and appeared limited to those who were less severely injured (i.e., Glascow Coma Score 4-8). Moreover, there was no direct evidence implicating hypoxia or hypercarbia as contributing to the poorer outcome in patients who were not intubated in the field.
In the intensive care unit (ICU), this issue is of relevance at a later time point when deciding when patients with TBI can be safely and successfully extubated. Whether it is reasonable to delay extubation because of a concern for hypoxic secondary brain injury or for concern of inadequate airway reflexes is unknown. There are little data regarding the timing of extubation following TBI. In a recently published study, it was suggested that the extubation of patients with TBI is often delayed and that this 48-72 hour delay contributes to the development of pneumonia and increases hospital costs.10 Interestingly, reintubation was required for airway obstruction or respiratory distress in 10% (13/136) of the cohort, suggesting the possibility of hypoxic secondary injury. However, there are no data regarding the clinical importance of secondary insults beyond the initial 2-3 post-injury days, once ICP and other indicators have stabilized.
Specific Interventions: Altering the Pathophysiology of TBI
Presently, there is little or no evidence for the efficacy of specific pharmacological therapies aimed at specific processes occurring after TBI. At least in part, this is due to the bewildering array of cellular and subcellular events occurring after TBI.1 Release of excitatory amino acids, cytokine-induced injury, free-radical production, and activation of the arachadonic acid cascade have all been implicated in the progression of primary and secondary brain injury. However, it has been difficult or impossible to determine which process or processes are of primary importance clinically. For instance, free-radical release and lipid peroxidation are considered to be one important process contributing to secondary brain injury, for which a number of pharmacological agents have been effective in experimental models. However, when one of these agents (Tirilizad mesylate) was subjected to a large, randomized, clinical trial, no clear benefit was found in comparison to placebo.11 Therefore, we are left with measures aimed primarily at preventing the development and limiting the consequences of secondary brain injury. I will discuss some of these treatment strategies.
Specific Support in the Intensive Care Unit
At the present time, we recognize that avoiding obvious systemic insults such as hypotension and marked hypoxemia are critical to optimizing outcomes after TBI. On the other hand, there are no specific treatments that target the pathophysiologic processes and are effective in improving the outcomes of TBI. Between these extremes, there is an array of treatments that can reasonably be expected to directly affect the secondary processes of ischemia, vasoparalysis, cerebral edema, and intracranial hypertension, and, therefore, to improve outcome. While few of these interventions have been evaluated rigorously, there are some nonexperimental data to support their application. Nonetheless, there is a clear need for well-designed and conducted clinical trials to determine which interventions effect outcome.
Intracranial hypertension adversely effects outcome after TBI.3 Data from the Traumatic Coma Databank suggest that above an ICP threshold of 20 mmHg, outcome from TBI is poor. The number and duration of hourly ICP measurements of 20 mm Hg or more were strongly related to a vegetative state or death in the 428 patients evaluated.5 It also seems that early, aggressive ICP monitoring, with treatment to maintain ICP less than 15 mm Hg, reduces mortality after severe traumatic brain injury.4 Unfortunately, the results of this provocative study that used historical controls have not been confirmed by a randomized clinical trial. Therefore, we are left with unanswered questions regarding the best ICP treatment threshold (e.g. 20 mm Hg or 15 mm Hg) although the guidelines suggest treatment at 20 mm Hg or more. It is also difficult to determine which treatments are most effective in controlling intracranial hypertension.
Most clinical protocols include cerebrospinal fluid drainage via ventriculostomy as a primary treatment modality that directly decreases intracranial volume. However, ventriculostomy monitoring and drainage is not without complications. The rate of CSF infection increases with the duration of intracranial pressure monitoring and can be expected to be 5-10% with ventriculostomy drainage.12,13 Therefore, the ability to treat intracranial hypertension with CSF drainage via ventriculostomy must be balanced against the approximate three-fold increased infection risk in comparison to intraparenchymal pressure monitors.13 Most neurosurgeons accept the defined risk of complications given the direct and effective CSF drainage afforded by ventriculostomy and the Guidelines for the Management of Severe Head Injury indicate this as the most desirable ICP monitoring technique.
Other treatments for intracranial hypertension have more obvious systemic consequences and few have been evaluated rigorously. Hyperventilation reduces ICP through a reduction in cerebral blood flow (CBF) and, therefore, cerebral blood volume. Brief hyperventilation can be beneficial in the emergency situation for the management of uncontrolled intracranial hypertension, while preparing for definitive treatment (craniotomy), or when there is impending brainstem compression or herniation. However, hyperventilation beyond brief periods may induce regional or global cerebral ischemia by excessively decreasing CBF. There is some evidence that moderate hyperventilation (PaCO2 = 30 mmHg) does not impair global cerebral metabolism, despite decreasing CBF and ICP.14 Nonetheless, other evidence suggests that moderate hyperventilation may contribute to poorer neurological outcomes after moderate to severe TBI.15 Therefore, the Guidelines recommend the avoidance of routine, early hyperventilation to a PaCO2 less than 35 mm Hg.
There is less data available regarding other systemic therapies such as the use of mannitol, which had traditionally been considered effective because of its osmotic and diuretic properties. It may be that some of mannitol’s beneficial effects are related to antioxidant and rheological properties, but the clinical importance of these effects is uncertain. However, it is clear that mannitol infusion increases CBF and reduces ICP within 10-20 minutes of infusion and probably should be used in the treatment of intracranial hypertension.16 There are a few randomized clinical trials that evaluate the effectiveness of mannitol in TBI. Unfortunately, individual studies are small and it is not possible to combine studies due to different study designs. These studies are reviewed by the Injury Group of the Cochrane Collaboration and provide no definitive evidence for an improvement in outcome. The authors of the Guidelines for the Management of Severe Head Injury, therefore, do not recommend a standard of care, but conclude that mannitol is effective in treating intracranial hypertension after TBI and recommend treatment with intermittent boluses (0.25-1.0 gm/kg body weight) rather than continuous infusion.
Cerebral perfusion pressure (CPP), defined as the difference between mean systemic arterial pressure and intracranial pressure (CPP = MAP - ICP), has become a focus of treatment strategies for reducing secondary injury. It had been observed that by stabilizing CPP at relatively high levels, ICP control was improved and cerebral ischemia was minimized. Therefore, iatrogenically increasing CPP with induced hypertension was investigated as a strategy to improve outcomes.17 The results of a small cohort study suggested a markedly reduced mortality in comparison to historical results.18 Based upon this report of 34 patients with severe TBI, whose management included ICP monitoring with cerebrospinal fluid (CSF) drainage, aggressive volume resuscitation (crystalloid and colloid), vasopressor infusion, mannitol infusion, and normocapneic ventilation (PaCO2 = 35 mm Hg), it was suggested that maintaining CPP of 70 mm Hg or more was the major determinant of the observed overall excellent outcomes (79% survival, with only 8% of patients dying from uncontrolled intracranial hypertension). These observations have been extended to a larger cohort with similar conclusions.17
However, there are numerous potential limitations to this aggressive manipulation of hemodynamics. In one reported series, 40% of patients required vasopressor support, most had pulmonary artery catheters placed for monitoring, and all were treated with neuromuscular blockade. The entire treatment protocol was resource intensive and included a number of unproven interventions, such as dopamine for renal protection in all cases where vasopressors were used.17 The benefits of aggressively pushing CPP to ³ 70 mm Hg or even ³ 80 mm Hg are controversial and there is some evidence that intracranial hypertension (³ 20 mm Hg) is a much more important determinant of poor neurological outcome than CPP is. In a review of a large prospective series, investigators determined that there is no advantage to raising the CPP beyond 60 mm Hg.19 Clearly, this is an area of controversy that could be addressed scientifically. Critical care physicians must be careful when applying such a protocol in the absence of good supportive evidence.
Summary
There are a number of other treatment strategies and monitoring techniques that have been used and studied in patients with TBI. In addition to those discussed here, others, such as controlled hypothermia, sedation strategies, neuromuscular blockade, and jugular venous oxygen saturation monitoring, each may be of potential benefit. Moderate hypothermia as a potential treatment for TBI was evaluated in a well-conducted clinical trial and found not to improve outcome.20 However, the main message from this negative trial is not necessarily the result, but rather that such a study could be done. Other present treatment strategies, most notably those which involve aggressive hemodynamic management to keep CPP greater than 70 mm Hg, should not be accepted until critically studied in a randomized trial.
Presently, most treatments for TBI are based upon increasing knowledge of the pathophysiology of primary and secondary brain injury, cohort studies, and a few small clinical trials. The early diagnosis of progression of mass lesions, prevention of hypotension and hypoxia, and minimizing elevations in ICP are likely the most important aspects of the present management of TBI. Altering cerebral hemodynamics and perfusion through aggressive strategies are a promising but unproven strategy, for which results of effectiveness are anxiously awaited. Until such a time when specific, beneficial treatments are identified, the role of the critical care physician remains to provide systemic support and avoid secondary brain injury.
References
1. Marshall LF. Neurosurgery 2000;47(3):546-561.
2. Luerssen TG, et al. Improved outcome as a result of recognition of absent and compressed cisterns on initial CT scans, in Hoff JT, Betz AL (eds): Intracranial Pressure VII. New York, Springer-Verlag, 1989, pp 598-602.
3. Miller JD, et al. J Neurosurg 1977;47:503-516.
4. Saul TG, Ducker TB. J Neurosurg 1982; 56:498-503.
5. Maramou A, et al. J Neurosurg 1991;75: S59-S66.
6. Pietropaoli JA, et al. J Trauma. (September) 1992; 33(3):403-407.
7. Chesnut RM, et al. J Trauma 1993;34(2): 216-222.
8. Winchell RJ, et al. Arch Surg 1996;131:533-539.
9. Winchell RJ, et al. Arch Surg 1997;132:592-597.
10. Coplin WM, et al. Am J Respir Crit Care Med 2000; 161:1530-1536.
11. Marshall LF, et al. J Neurosurg 1998; 89:519-525.
12. Holloway KL, et al. J Neurosurg 1996;85:419-424.
13. Rebuck JA, et al. J Neurol Neurosurg Psychiatry. 2000;69:381-384.
14. Diringer MN, et al. J Neurosurg 2000;92:7-13.
15. Muizelaar JP, et al. J Neurosurg 1991;75:731-739.
16. Mendelow AD, et al. J Neurosurg 1985;63(1):43-48.
17. Rosner MJ, et al. J Neurosurg 1995;83:949-962.
18. Rosner MJ, et al. J Trauma 1990;30(8):933-941.
19. Juul N, et al. J Neurosurg 2000;92:1-6.
20. Marion DW, et al. N Engl J Med 1997; 336(8):540-546.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.