Introduction
In 1960, Nils Lundberg, a Swedish neurosurgeon, measured the intraventricular pressure (ICP) in 64 patients (1). He inserted a plastic catheter into the frontal horn of the lateral ventricle and connected it to a strain gauge pressure transducer. Recording was carried out with an ink-writing potentiometer. Although archaic by modern standards, he used this equipment to identify three types of intracranial pressure waves. The large plateau type “A” waves suggested a very high ICP of >50 mm Hg and imminent deterioration. This phase strongly correlated with symptoms of raised intracranial pressure (ICP), including headache, vomiting, irregular respirations, and hypertension. The rhythmically oscillating “B” waves at a frequency of 1–2/min suggested a lower ICP between 30–50 mm Hg, and the higher frequency “C” waves denoted an ICP of <30 mm Hg. Initially, neurosurgeons of that era were understandably reluctant to resort to invasive monitoring. By the 1970s and 80s, Lundberg’s technique gained more acceptance and was being increasingly employed in subarachnoid hemorrhage and traumatic brain injury. However, a 1978 survey of neurosurgical training programs in the US revealed that ICP monitoring was rarely employed by neurosurgeons (2). With increasing experience, more widespread use followed; between 1995–2005, ICP monitoring in traumatic brain injury had more than doubled in the US.
Sahjpaul et al. conducted a survey of Canadian neurosurgeons regarding the utilization of ICP monitoring in traumatic brain injury more than two decades ago (3). There appeared to be universal acceptance of ICP monitoring; however, only 20.4% of respondents expressed a high level of confidence in the efficacy of ICP monitoring to improve clinical outcomes. The majority of respondents believed that a prospective trial to evaluate the efficacy of ICP monitoring is warranted, considering the uncertainty regarding its clinical utility. A year later, Stocchetti et al. performed a questionnaire-based study on severe traumatic brain injury and ICP monitoring practices in Europe (4). There was a wide variability in ICP and cerebral perfusion pressure monitoring among centers. Monitoring of ICP was a reliable tool in identifying intracranial hypertension when clinical signs alone were inconclusive. The incidence of complications related to ICP monitoring was low.
In the days before ICP monitoring became commonplace, surgical intervention for mass lesions was the only specific treatment modality available for traumatic brain injury. Identification of mass lesions was based on the limited diagnostic tools of the pre-CT era. With the advent of ICP monitoring, other therapeutic measures, including mechanical ventilation, sedation, and fluid management could be instituted and titrated to the desired effect. In the meanwhile, CT-imaging became widely available in the 1970s, allowing regular, rigorous evaluation of the injured brain, thus facilitating aggressive therapeutic interventions. The advent of CT technology led to a dramatic improvement in clinical outcomes following traumatic brain injury from the mid-1970s to the 1990s. However, from the 1990s onwards, clinical outcomes largely remained static. A 1995 survey involving trauma centers in the US revealed that only 28% of centers routinely performed ICP monitoring in traumatic brain injury (5).
Background to the trial
The evolution of ICP-based management has been intertwined with advances in other facets of care in the management of patients with severe traumatic brain injury. These advancements involved pre-hospital and emergency department care, besides groundbreaking improvements in neurointensive care and neurorehabilitation. It remained unclear if the improved clinical outcomes could be attributed to ICP measurement and monitoring or confounded by advancements in overall care. Observational studies revealed worse outcomes with an ICP-guided approach to managing severe traumatic brain injury, with a longer duration of mechanical ventilation, worse functional status, a higher risk of pneumonia, renal dysfunction, and mortality (6,7). Clinical guidelines also cite inadequate evidence to support the efficacy of ICP monitoring (8). Although a randomized controlled trial (RCT) was called for, ethical considerations arose from having a control group without ICP monitoring, especially in settings where it was routine practice in severe traumatic brain injury.
In Latin America, intensivists routinely managed severe traumatic brain injury without recourse to ICP monitoring. There was a sufficient level of equipoise among them that allowed the conduct of an RCT to evaluate the efficacy of ICP monitoring without facing an ethical dilemma. The BEST-TRIP trial was conceptualized to include patients with severe traumatic brain injury from two Latin American countries – Bolivia and Ecuador, where ICP monitoring was not routinely carried out.
Study hypothesis
The BEST-TRIP investigators compared a management protocol based on ICP monitoring with management guided by clinical examination and serial CT imaging. They hypothesized that ICP-guided management would lead to lower mortality, improved neuropsychological, and functional outcomes at 6 months.
Population and design
The BEST-TRIP trial was a multicenter RCT conducted across six sites – four in Bolivia and two in Ecuador – among patients with severe traumatic brain injury. The study was conducted between September 2008 and May 2012. Patients were randomized to undergo management based on ICP monitoring (the ICP group) or imaging and clinical examination (the I-CE group). Randomization was stratified by age, severity of injury, and study site. The study sites were staffed by intensivists and had round the clock neurosurgical cover and access to CT scan.
Inclusion criteria
The study patients were ≥13 years with traumatic brain injury and a score on the Glasgow Coma Scale (GCS) between 3–8 on presentation, or within 48 hours of injury. Among patients who were intubated and mechanically ventilated, the motor component of the GCS had to be between 1–5 for study inclusion.
Excluded
Patients with a GCS score of 3 and bilateral fixed, dilated pupils were excluded. Patients who were considered unlikely to survive were also excluded.
The ICP group
Patients allocated to the ICP group underwent insertion of an intraparenchymal catheter as soon as feasible. Treatment measures were aimed to maintain the ICP below 20 mm of Hg according to standard guidelines. Ventriculostomy was performed if cerebrospinal fluid drainage was necessary.
The I-CE (control) group
In this group, management was based on clinical examination and CT-imaging as appropriate. Therapeutic modalities included a protocol-based administration of hyperosmolar therapy, mild hyperventilation to PaCO2 levels of 30–35 mm Hg, and ventricular drainage if appropriate. If cerebral edema persisted, high-dose barbiturates and decompressive craniectomy were considered.
Common management in both groups
CT scan of the brain was performed at baseline, 48 hours, and 5–7 days after admission, or otherwise as considered necessary. Analgesia, sedation, and muscle relaxants were administered as required to both groups. Surgical evacuation of mass lesions and decompressive craniectomy were performed as appropriate. A protocolized approach was followed if neurological worsening occurred.
Sample size
The authors calculated a sample size of 324 patients to provide 80% power to identify an increase by 10 percentage points in the number of patients with a good outcome or moderate disability based on the Glasgow Outcome Score-Extended (GOS-E).
Results
The study enrolled 324 patients as planned – 157 in the ICP group and 167 in the I-CE group. The median motor component of the GCS was 5 (3–5) and 4 (3–5) respectively. In the ICP group, the catheter was maintained for a median duration of 3.6 days. There were signs of raised ICP on the initial CT scan in 90% of patients in the ICP group compared with 89% in the control group. Midline shift by more than 5 mm was evident in 34% and 39% respectively. A mass lesion was evacuated in 31% of patients in the ICP group and 35% in the control group. Osmotherapy and hyperventilation were more often resorted to in the I-CE compared to the ICP group.
The primary outcome
The primary outcome was a composite of 21 parameters. The individual components of the composite outcome included survival time, impairment of the conscious level, the functional status at 3 and 6 months, and the neuropsychological status at 6 months. The outcome was expressed as a percentile between 0–100, with a higher percentile indicating better outcomes. Evaluation was carried out at 3 and 6 months by trained assessors, who were blinded to group assignment. There was no significant difference in the composite primary outcome between the ICP and the I-CE groups. The median percentile in the ICP group was 56 (22–27) compared with 53 (21–26) in the control group (OR: 0.49; 95% CI; 0.74–1.58).
The composite primary outcome was similar between the two groups when analysis was restricted to survivors alone. Besides, on pre-defined subgroup analysis, there was no difference in the primary outcome based on age, gender, study site, and CT findings.
Secondary outcomes
Among the secondary outcomes, the 14-day mortality was similar in both groups. The cumulative mortality at 6 months was also similar – 39% in the ICP and 41% in the I-CE groups (OR: 1.10; 95% CI, 0.77–1.5). The GOS-E, ranging from 1 (death) to 8 (most favorable recovery) was evaluated at six months. A score of 2–4 was considered an unfavorable, and 5–8 a favorable outcome. The number of patients who died or experienced favorable and unfavorable outcomes were similar in both groups.
Other outcomes
The hospital length of stay was marginally lower in the I-CE group. No significant differences were observed in the ICU length of stay, the duration of ventilator support, and the incidence of non-neurological complications. A greater number of patients were treated with high-dose barbiturates, hypertonic saline, and hyperventilation in the I-CE group. The duration of treatment with mannitol or hypertonic saline was also longer in the I-CE group. No significant difference was observed in the number of patients who underwent decompressive craniectomy between the two groups.
Adverse events
There were ten adverse events related to the ICP catheter; these included catheter malfunction (3%), accidental removal (3%), and hemorrhage (1%). Serious adverse events occurred in 45% of patients in the ICP group and 46% of patients in the control group.
Study conclusions
Among patients with severe traumatic brain injury, management based on maintenance of ICP ≤20 mm Hg was not superior to imaging and clinical examination-based management.
Strengths
The BEST-TRIP trial was the first large RCT that evaluated clinical outcomes using an ICP-guided compared with a clinical examination and CT-guided approach to managing patients with severe traumatic brain injury. The multi-centric study had robust internal validity. The investigators meticulously evaluated a broad array of relevant endpoints based on in-hospital morbidity, mortality, and quality of life. The assessors were blinded to the assignment arm, thus reducing the risk of bias. Although conducted in two low-income countries, the care provided was of a high standard and complied with the trial design.
Limitations
The composite primary outcome, including 21 parameters, may not be ideal. Although the 14- and 30-day mortality were similar in both groups, the study was not powered for these endpoints.
Although the treatment in both groups was protocolized, there were some key differences in the therapeutic strategy. In the ICP group, CSF drainage was one of the modalities employed to reduce ICP, while it was not part of the treatment strategy in the IC-E group. In the IC-E group, more patients underwent osmotherapy and hyperventilation while the ICP group received barbiturates more often. The difference in treatment intensity between groups is difficult to explain. The study did not provide information regarding the number of CT scans performed in each group, which contribute to the cost of care and exposure to radiation.
Clinical outcomes following traumatic injury may depend on the specific area of compression, especially the brain stem, thalamus, and the reticular activating system. Injury to these areas occur due to local mechanical displacement of brain tissue. Therapeutic interventions aimed to reduce ICP – the average pressure within the cranial cavity – may not always alleviate injury mediated by mechanical displacement.
The commonly followed ICP threshold of 20 mm Hg based on arbitrary guidelines may be considered too low. Indeed, the window of opportunity for interventions based on intracranial hypertension may be narrow. Patients at either end of the spectrum may be unaffected by ICP-based therapeutic interventions. Those with a relatively low ICP may have favorable outcomes without specific interventions, while those with catastrophic intracranial hypertension have poor outcomes regardless of the monitoring modality employed.
The trial was conducted in countries where clinicians were well-versed with management based on clinical signs and CT findings. A deviation from the usual approach, although based on a standard protocol, may result in unfamiliarity with a novel strategy that may have impacted outcomes. On the contrary, the results of the trial may not be applicable in healthcare settings that routinely measure ICP as an important facet of care in patients with severe traumatic brain injury. Furthermore, the pre-hospital care and transfer of patients may vary considerably in other parts of the world, where the results of the trial may not be valid. Post-discharge care, including rehabilitation, hold the key to outcomes in traumatic brain injury. The details of this vital aspect of care were unclear.
The authors emphasized that the study does not challenge the use of ICP monitoring in the management of severe traumatic brain injury – they only sought to test the monitoring-based treatment algorithm. It is also plausible that a one-size fit all approach with the use of a universal ICP threshold may not be the most efficacious strategy. The ICP trend over time may be more relevant than individual, instantaneous values. Furthermore, a rigid management strategy based on a fixed level of ICP may not be the optimal approach in different subtypes of injury.
Summary
Measurement of ICP was introduced in the 1960s as a monitoring tool to evaluate the well-being of the brain by the bedside. Over the next several decades, it evolved into a standard monitoring tool and was engrained into protocols for the management of traumatic brain injury. In the meantime, questions arose regarding the utility of ICP monitoring – particularly with observational trials suggesting adverse outcomes. However, observational studies always carry a risk of bias; ICP monitoring may often be employed in the sickest of patients who are prone to poor outcomes. Clinical guidelines cited inadequate evidence to support the efficacy of ICP monitoring in traumatic brain injury. An adequately powered RCT was called for, considering the uncertainty regarding the efficacy of ICP monitoring. The BEST-TRIP trial took the up the challenge of questioning what was considered to be “standard of care”. The trial findings did not support the superiority of ICP monitoring-based management compared with a clinical examination and CT-guided approach. However, as the authors themselves aver, the study may not apply to healthcare settings that use ICP monitoring as a standard monitoring tool. Continuous monitoring may be particularly necessary in patients with severe intracranial hypertension; the trial did not specifically evaluate this subgroup of patients. The study provided evidence that a clinical examination and CT scan-based management may be appropriate in resource-poor settings where ICP monitoring is not routinely performed.
References
1. Lundberg N. Continuous recording and control of ventricular fluid pressure in neurosurgical practice. Acta Psychiatr Scand Suppl. 1960;36(149):1–193.
2. Saunders RL. Intracranial pressure monitoring survey. Neurosurgery. 1978;3(1):130.
3. Sahjpaul R, Girotti M. Intracranial pressure monitoring in severe traumatic brain injury–results of a Canadian survey. Can J Neurol Sci J Can Sci Neurol. 2000 May;27(2):143–7.
4. Stocchetti N, Longhi L, Magnoni S, Roncati Zanier E, Canavesi K. Head injury, subarachnoid hemorrhage and intracranial pressure monitoring in Italy. Acta Neurochir (Wien). 2003 Sep;145(9):761–5; discussion 765.
5. Ghajar J, Hariri RJ, Narayan RK, Iacono LA, Firlik K, Patterson RH. Survey of critical care management of comatose, head-injured patients in the United States. Crit Care Med. 1995 Mar;23(3):560–7.
6. Griesdale DEG, McEwen J, Kurth T, Chittock DR. External ventricular drains and mortality in patients with severe traumatic brain injury. Can J Neurol Sci J Can Sci Neurol. 2010 Jan;37(1):43–8.
7. Shafi S, Diaz-Arrastia R, Madden C, Gentilello L. Intracranial pressure monitoring in brain-injured patients is associated with worsening of survival. J Trauma. 2008 Feb;64(2):335–40.
8. Brain Trauma Foundation, American Association of Neurological Surgeons, Congress of Neurological Surgeons, Joint Section on Neurotrauma and Critical Care, AANS/CNS, Carney NA, Ghajar J. Guidelines for the management of severe traumatic brain injury. Introduction. J Neurotrauma. 2007;24 Suppl 1:S1-2.
Best critical care blogs: https://blog.feedspot.com/critical_care_blogs/
