Abstract
Mrs. Smith, a 24-year-old woman without prior past medical history, is admitted to the intensive care unit after a motor vehicle collision with ejection from the vehicle, during which she suffered a severe traumatic brain injury (TBI). The patient was intubated at the scene and remained intubated throughout the hospital course. The patient’s Glasgow Coma Scale after resuscitation was 6T (M4, E1, V1(T)). Serial computed tomography (CT) scans of the head revealed left frontal and temporal contusions in evolution with an increasing amount of pericontusional edema, and mass effect with midline shift of the septum pellucidum and compression of the mesencephalic cisterns (Figure 12.1). The patient was treated according to Brain Trauma Foundation guidelines with intracranial pressure (ICP) monitoring for elevated ICP. Initially the patient responded well to medical management with deep sedation, osmotherapy (mannitol and 23.4% hypertonic saline alternating resulting in a serum sodium level of 165 mmol/dL), and normothermia. Sedation holidays were omitted owing to ICPs spiking to 40 mmHg off sedation. On hospital day 5, the patient’s ICP continued to intermittently spike to 30 mmHg, and she no longer responded to osmotherapy. A repeat head CT scan revealed an interval increase in the midline shift and pericontusional cerebral edema. Given her age and active lifestyle before the injury, the family wanted “everything done.” You call the neurosurgeon and in collaboration with the neurosurgical team you walk into a meeting with the family to discuss secondary decompressive craniectomy (DC) vs. additional medical management with significant potential side effects (induced hypothermia, pentobarbital coma).
Mrs. Smith, a 24-year-old woman without prior past medical history, is admitted to the intensive care unit after a motor vehicle collision with ejection from the vehicle, during which she suffered a severe traumatic brain injury (TBI). The patient was intubated at the scene and remained intubated throughout the hospital course. The patient’s Glasgow Coma Scale after resuscitation was 6T (M4, E1, V1(T)). Serial computed tomography (CT) scans of the head revealed left frontal and temporal contusions in evolution with an increasing amount of pericontusional edema, and mass effect with midline shift of the septum pellucidum and compression of the mesencephalic cisterns (Figure 12.1). The patient was treated according to Brain Trauma Foundation guidelines with intracranial pressure (ICP) monitoring for elevated ICP. Initially the patient responded well to medical management with deep sedation, osmotherapy (mannitol and 23.4% hypertonic saline alternating resulting in a serum sodium level of 165 mmol/dL), and normothermia. Sedation holidays were omitted owing to ICPs spiking to 40 mmHg off sedation. On hospital day 5, the patient’s ICP continued to intermittently spike to 30 mmHg, and she no longer responded to osmotherapy. A repeat head CT scan revealed an interval increase in the midline shift and pericontusional cerebral edema. Given her age and active lifestyle before the injury, the family wanted “everything done.” You call the neurosurgeon and in collaboration with the neurosurgical team you walk into a meeting with the family to discuss secondary decompressive craniectomy (DC) vs. additional medical management with significant potential side effects (induced hypothermia, pentobarbital coma).
Figure 12.1 Pre- and postcraniectomy head CT scan.
(A) Three slices of the precraniectomy head CT scan of the patient. The patient suffered frontal and temporal hemorrhagic contusions on the left (hyperdense lesions), with pericontusional edema (dark borders), resulting in a left-to-right midline shift of the septum pellucidum, partial compression of the left lateral ventricle, and obliteration of the basal cisterns.
(B) The same three slices of the immediate after craniectomy head CT scan of the same patient, status post left frontotemporal craniectomy and evacuation of the prior left temporal hemorrhagic contusion. The previous midline shift and the compression of the lateral ventricle are improved. The basal cisterns are open.
TBI remains a major public health issue, as 69 million individuals worldwide suffer a TBI each year1 and TBI contributes to one third of all injury-related deaths in the United States. Nearly 300,000 TBI-related hospitalizations occur annually in the United States, with more than 56,000 Americans dying annually after sustaining a TBI, of which the majority are due to severe TBI (Glasgow Coma Scale of ≤8).2 Additionally, the total number of TBI-related emergency department visits, hospitalizations, and deaths, has increased by 53% since 2006.3
Although treatment protocols and management algorithms have been developed over the last two decades for the clinical management of patients with severe TBI,4–6 the prognostication of outcomes after TBI often remains uncertain. In addition to the limited availability of validated and well-calibrated prediction models,7, 8 physicians have variable confidence in their own accuracy of outcome prediction in TBI9 and generally low confidence in using the most validated prediction models in patients.10 Reasons for difficulty with prognostication include that TBI has multiple different pathomechanisms, may co-occur with other injuries and complications, and may present implementation and compliance challenges with respect to existing guidelines.11 In this Chapter, we focus on the use of DC to relieve refractory intracranial hypertension after severe TBI.
12.1 Common Scales Used in TBI
The severity of TBI is graded using the Glasgow Coma Scale, which describes a patient’s level of consciousness based on their verbal, motor, and eye-opening responses to stimulus.12 The most widely used outcome measure has been the 5-point Glasgow Outcome Scale,13 although more recently, the 8-point extended Glasgow Outcome Scale – Extended (GOS-E) has been shown to have increased sensitivity to small changes in functional status14 and good reliability, even when administered over the phone.15 By definition, the GOS-E is an ordinal scale with eight different categories (Table 12.1). The GOS-E is frequently dichotomized into “unfavorable” (GOS-E 1–4) and “favorable” (GOS-E 5–8) outcomes for ease of interpretation in the statistical analysis.14, 16
Table 12.1. Glasgow Outcome Scale – Extended12
| 1 | Death |
| 2 | Vegetative state |
| 3 | Lower severe disability |
| 4 | Upper severe disability |
| 5 | Lower moderate disability |
| 6 | Upper moderate disability |
| 7 | Lower good recovery |
| 8 | Upper good recovery |
12.2 Decompressive Craniectomy
DC refers to the removal of a portion of skull and the opening of the underlying dura to create room for the swollen brain to expand beyond the confines of the cranial vault, thereby relieving intracranial hypertension. DC techniques can be classified in three ways – a unilateral DC (hemicraniectomy) where a large frontotemporoparietal bone flap is removed, a bilateral hemicraniectomy, and a bifrontal DC where the removed bone flap extends from the floor of the anterior cranial fossa to the coronal suture and to the middle cranial fossa floor bilaterally.17
“Primary DC” is the use of DC after the evacuation of a space-occupying lesion without replacing the bone flap owing to intraoperative brain swelling or an anticipation that brain swelling will worsen postoperatively, i.e., owing to evolving contusions. “Secondary DC” is used as a part of the tiered management for refractory intracranial hypertension. This event is critical when the patient develops a mass effect from traumatic lesions, resulting in eventual brain herniation and death. Guidelines suggest that, if there is a high index of suspicion that ICP will be elevated in patients who cannot be evaluated clinically, ICP monitoring should be instituted with the goal of maintaining an ICP of less than 23 mmHg and a cerebral perfusion pressure of 60–70 mmHg.5 Elevated ICP is associated with increased mortality and worse functional status, as well as poor neuropsychological functional outcomes.18 Medically uncontrollable intracranial hypertension can be managed using DC, mild hypothermia, or barbiturate coma.4
12.3 The Role of Secondary DC in TBI
A number of studies in the 1980s and 1990s established the groundwork on the use of DC. These studies showed that patients with a lower ICP had better outcomes,19 and that DC could be used efficaciously as second-tier therapy.20
Jiang et al. published the first multicenter randomized controlled trial in adults assessing the management of refractory intracranial hypertension in 486 patients.21 This trial suggested that the standard large frontotemporoparietal DC resulted in significantly better ICP control than the smaller diameter temporoparietal DC and was associated with fewer complications and better outcomes. This conclusion has been supported by findings from other studies,22 and large DC has consequently been recommended.23
Most recently, two large, multicenter, randomized controlled trials examined two different hypotheses about the role of DC in refractory intracranial hypertension in severe TBI: (1) the Decompressive Craniectomy in Diffuse Traumatic Brain Injury (DECRA) Trial examined the role of early/neuroprotective bifrontal DC24 and (2) the Randomized Evaluation of Surgery with Craniectomy for Uncontrollable Elevation of Intracranial Pressure (RESCUEicp) examined the role of unilateral or bifrontal DC as a last-tier therapy for severe and refractory intracranial hypertension.25
The DECRA trial randomized 155 patients to receive either bilateral DC or standard of care. All patients received first tier interventions for intracranial hypertension whenever ICP increased to more than 20 mmHg. If the ICP spontaneously increased for more than 15 minutes within 1 hour despite first-tier interventions, patients randomized to surgery underwent large bifrontotemporoparietal DC with bilateral dural opening plus standard care.
Importantly, the DECRA investigators defined “unfavorable outcome” as a GOS-E of 4 or less at 6 months after TBI. In DECRA, patients in the DC group had significantly lower mean ICP, shorter duration of mechanical ventilation, and shorter intensive care unit stays at 6 months, but had higher rates of “unfavorable outcomes” compared with the standard of care control group (70% vs 51%; p = .02). This result was likely due to the fact that more patients in the surgery group had bilateral unreactive pupils than the control group (27% vs. 12%; p = .04); after post hoc adjustment for baseline pupil reactivity, unfavorable outcome rates were no longer significantly different between the groups.
The RESCUEicp trial took a different approach to examining the role of DC in the management of increased ICP. Investigators assessed DC as a last-tier intervention, enrolling 389 patients who presented with increased ICP of more than 25 mmHg for 1–12 hours despite tier 1 and 2 measures. Patients were randomized to receive either medical management with the option of barbiturates after randomization, or a DC, with the decision about the exact type (i.e., unilateral or bifrontal) dependent on the patients’ injury and surgeon discretion. In contrast with DECRA, the predefined primary analysis of RESCUEicp was an ordinal analysis, which is in keeping with recent recommendations.26 The primary analysis showed a significant between-group difference in the GOS-E distribution and a substantial reduction in mortality with surgery. The prespecified sensitivity analysis dichotomized at upper severe disability (GOS-E ≥4) or better was significant at 12 months (i.e., 45% of the patients in the surgical group were at least independent at home, as compared with 32% of patients in the medical group; p = .01). At 6 months, the difference in the favorable outcome rates was not statistically significant (43% vs. 35%; p = .12).
A comparison of the two studies is informative. The DECRA trial, as compared with the RESCUEicp trial, enrolled patients with a lower ICP threshold (20 mmHg vs. 25 mmHg) for shorter intervals (15 minutes vs. 1–12 hours), and after lower intensities of therapy (stage 1 interventions vs. stage 1 and 2 interventions). At enrollment, the populations also differed with respect to expected outcome, as it has been shown that the requirement for stage 2 interventions increases the relative risk of death by 60%.27 This fact also accounts for why the pooled mortality of 38% in the RESCUEicp trial vs. 19% in the DECRA trial is unsurprising.
With these conclusions on and new questions raised about the clinical application of DC from these two most recent randomized trials, the International Consensus Meeting on the Role of Decompressive Craniectomy in the Management of Traumatic Brain Injury was organized. Participating delegates met to discuss key topics around DC and published the first consensus statement on the topic.28 Consensus statements concerning secondary DC are summarized in Table 12.2 .
Table 12.2. Summary of recommendations from the International Consensus Meeting on Secondary Decompressive Craniectomy28
Related posts:
Chapter 2 – How Much Does the Family Want to Be Involved in Decision-Making?
Chapter 4 – Communication Skills for Critical Care Family Meetings
Chapter 5 – The Do-Not-Resuscitate Order
Chapter 6 – The Do-Not-Intubate Order
Chapter 19 – Measuring and Evaluating Shared Decision-Making in the Intensive Care Unit
Chapter 20 – Brain Death Discussions
Full access? Get Clinical Tree
