Article Text
PDF
Abstract
Traumatic brain injury (TBI) is a leading cause of morbidity and mortality in young individuals. Management of TBI is complex and aims to prevent secondary injury and optimise conditions for neurological recovery. Vasospasm is a possible complication of TBI, and its significance is unknown. Its incidence is underestimated and there are currently no preventive or therapeutic approaches with proven efficacy. The occurrence of vasospasm contributes to secondary brain injury and worsens prognosis. The diagnosis of vasospasm in TBI is challenging due to the difficulty in perceiving neurological deterioration in these patients. We present a case of a young patient admitted to the neurocritical care unit following TBI. He presented a partial neurological recovery, followed by clinical deterioration and persistent coma. The diagnosis of extensive ischaemic lesions due to severe vasospasm was established. Suspicion of vasospasm and timely screening, particularly in high-risk patients, may improve survival and outcomes in TBI.
- Brain Injuries, Traumatic
- Adult intensive care
- Neuro ITU
- Neurosurgery
Statistics from Altmetric.com
Background
Traumatic brain injury (TBI) is a leading cause of morbidity and mortality in young individuals. With an incidence of up to 800 cases/100 000 inhabitants/year, its social and economic burden is very significant.1 It can be classified as mild, moderate or severe based on the Glasgow Coma Scale (GCS).2
Management of moderate and severe TBI (GCS 3–12) is complex and aims to prevent the development of secondary brain injury and optimise conditions for neurological recovery. Secondary brain injury can occur due to multiple mechanisms, including disturbances in cerebral blood flow (CBF), neuroinflammation, excitotoxicity, oxidative stress and changes in cerebral metabolism.2 3 Inadequacy of CBF is multifactorial, resulting from the interaction of raised intracranial pressure (ICP), systemic hypotension, impaired autoregulation and vasospasm.2 3 Optimal neurocritical care aims to preserve brain homeostasis and prevent secondary insult. This is achieved through several strategies, including the control of arterial blood pressure, ICP and brain oxygenation, as well as management of factors influencing cerebral homeostasis, including ventilation, temperature and sedoanalgesia.4
Vasospasm is a frequent complication of aneurysmatic subarachnoid haemorrhage (aSAH), with devastating consequences associated with secondary ischaemia.5 It is multifactorial, with endothelial damage, formation of microthrombi, lysis of subarachnoid blood clots and blood degradation products, oxidative stress and other mechanisms contributing to its pathophysiology. The risk of vasospasm in aSAH typically peaks in days 7–10 and resolves after day 21.6 7
The diagnosis of vasospasm can be made with different techniques, including transcranial Doppler (TCD) ultrasound, CT angiography, digital subtraction angiography (DSA) and continuous electroencephalography.7 Although DSA is the gold standard for diagnosis, it is technically difficult and requires contrast administration. TCD, on the other hand, is a bedside non-invasive examination that can be repeated to evolution assessment. TCD is currently used as a screening tool for vasospasm in aSAH patients.7–9
Vasospasm can occur in TBI patients and its significance in this population is not yet determined. The incidence of post-traumatic vasospasm is largely underestimated. Angiography and TCD studies suggest that it can be present in up to 40%–60% of TBI. Vasospasm onset in TBI typically occurs earlier than in aSAH, with maximal TCD velocities in days 5–7 post injury.10 11 Younger age, lower GCS, larger cisternal/intracerebral haemorrhage and higher severity of lesions seem to be associated with increased risk.10–13 The pathophysiology of vasospasm in TBI appears to have some differences from findings in aSAH. Vasospasm can occur in the absence of SAH and there seems to be a pathological role of the primary mechanical insult.10 12 14 15 Figure 1 summarises the proposed pathophysiology for the occurrence of vasospasm in TBI.
Diagram of the pathophysiology of post-traumatic vasospasm. Although the exact mechanisms causing vasospasm in traumatic brain injury are not fully understood, there seems to be a role both for direct mechanical damage to endothelium and blood vessels and for the effects of the presence of blood products in the subarachnoid space. The release of vasoactive substances promotes inflammation, with associated oxidative stress, calcium imbalance and multifactorial endothelial dysfunction leading to vasospasm and delayed cerebral ischaemia. Diagram created by the authors. CBF, cerebral blood flow. NO, nitric oxide
The diagnosis of clinically relevant vasospasm is challenging in TBI patients, with coexisting lesions and depressed consciousness making clinical symptoms of neurologic deterioration difficult to recognise. No strategy for prevention or treatment of post-traumatic vasospasm has been proven to be effective. Treatment with calcium-channel blockers and endovascular treatment may have good results, with improvement of imagiological findings and clinical outcomes.10–12 15 16
We present a case of a moderate TBI admitted to the neurocritical care unit (ICU). A partial neurological recovery was observed in the first week after admission, but clinical worsening occurred shortly after. A diagnosis of delayed cerebral ischaemia associated with severe vasospasm was made.
Case presentation
A man in his 30s was admitted to the neurocritical ICU after TBI due to a quad bike accident. He had medical history of hypertension and anxiety and was treated with perindopril 4 mg and lorazepam 1.25 mg daily. On scene, he was described as having a GCS of 12, with otorrhagia and a scalp laceration. Significant psychomotor agitation and non-cooperation led to the need for sedation and subsequent decrease in consciousness and orotracheal intubation and the patient was transferred to a trauma centre.
On admission, he had been sedated with etomidate and midazolam and mechanically ventilated. He was hypertensive (blood pressure of 180/120 mm Hg, normalised after optimisation of sedoanalgesia) and otherwise hemodynamically stable. He was normoventilated (arterial pH 7.39, pCO2 35.5 mm Hg, pO2 155 mm Hg) and his pupils were symmetrical, miotic and reactive to light. The initial blood analysis showed no relevant alterations.
The cerebral CT scan on admission revealed a sunken fracture of the right pterional, parietal and temporal region with distortion of underlying brain parenchyma, diffuse subarachnoid haemorrhage and bilateral temporal and frontobasal contusional oedema (figure 2). The trauma study revealed a right glenohumeral fracture, as well as multiple fractures of the face and a fracture of the greater wing of the right sphenoid.
CT of the head shows depressed skull fracture (white arrows) with extensive subarachnoid haemorrhage, with diffuse cisternal distribution (black arrows).
He underwent emergent right frontotemporoparietal decompressive craniectomy. The postsurgical CT scan of the brain showed a right serohematic fronto-temporo-parietal collection and pneumocephalus, as well as a large haemorrhagic contusion on the right cerebellar hemisphere, including the right cerebellar amygdala and vermis, causing distortion of the fourth ventricle and obliteration of the cisternal spaces, along with dilation of the supratentorial ventricular system (figure 2). The patient maintained ICP<22 mm Hg and haemodynamic stability under vasopressor support (noradrenaline<1 µg/kg/min). He was under ICP and cerebral perfusion pressure (CPP) evaluation, as well as cerebral autoregulation assessment (pulse reactivity index (PRx), ICM+ software, Cambridge University)17. Subsequent re-evaluation CT scans at 24 hours and 3 days post surgery showed no signs of further worsening of the lesions.
He evolved with haemodynamic and neurological stability and no signs of complications. The cerebral magnetic resonance (MR) performed on the 7th day post admission revealed a right subdural fronto-temporo-parietal haemorrhagic contusion, subarachnoid haemorrhage in reabsorption, right frontal edematous contusion and a right cerebellar contusion with a reduced fourth ventricle and supratentorial ventricular dilation, with no signs of tension or brainstem lesions.
Weaning of sedoanalgesia was initiated on day 5. On day 7, a partial neurological recovery was observed, with a GCS of 10 (E3V1tM6). Due to persistent vomiting with orotracheal tube reactivity, the patient needed to be resedated on day 8. Several attempts to reduce sedoanalgesia were complicated by vomiting and paroxysmal sympathetic hyperactivity syndrome. Withdrawal of intravenous sedatives was completed on the 15th day and a persistent non-reactive coma was observed, with a GCS of 5 (E2VTM2).
Investigations
An EEG was performed, showing diffuse non-specific encephalopathy and excluding epileptic activity. A CT scan of the brain was performed 48 hours after interruption of sedation. It revealed extensive bilateral cortical and subcortical frontoparietal and opercular hypodense areas compatible with ischaemia, with greater extent on the right hemisphere, involving the frontobasal, superior temporal, temporopolar and insular cortex (figure 3). The remaining cerebral lesions were improved, with no signs of further complications. The CT-angio scan revealed irregularities in the filling of contrast in the M1 distal portion of the right medial cerebral artery (MCA), P1/P2 transition of the right posterior cerebral artery (PCA) and P1 and P3 segments of the left PCA, as well as focal ectasia of the V4 segment of the right vertebral artery, compatible with vasospasm (figure 4).
CT of the head shows extensive, bilateral cortical ischaemic lesions (*).
CT-angio revealed diffuse and severe cerebral vasospasm, namely in the terminal carotid arteries and M1 and M2 segments of the middle cerebral artery (arrows).
The TCD revealed globally increased flow velocities, predominantly on the right MCA (Lindegaard ratio of 7), also involving the left MCA (Lindegaard ratio of 6) and the anterior and PCAs and vertebral arteries bilaterally. These findings were suggestive of diffuse vasospasm. No periods of arterial hypotension, intracranial hypertension or inadequate CPP (evaluated through revision of pulse reactivity index (PRx) data) were observed during ICU stay. An echocardiogram excluded structural pathology and the presence of intracardiac thrombi.
Differential diagnosis
Neurological deterioration in TBI patients can be multifactorial. Seizures and non-convulsive epileptic status are common in this population and can cause diminished consciousness. An EEG was performed after withdrawal of sedatives, excluding epileptic activity. Worsening of oedema can cause deterioration of neurological status, usually occurring in the first days after trauma. In our patient, oedema was evolving favourably, and no local complications were identified in the follow-up CT scans of the brain in the early days after admission. Intracranial hypertension and hypoperfusion can cause compression and ischaemia and worsen disability in neurocritical care patients. This was excluded by continuous ICP monitoring and CPP monitoring and revision in our patient.
In our patient, other causes of neurological deterioration were excluded through EEG, neuromonitoring and CT scans of the brain, with Doppler findings diagnostic of vasospasm and subsequent delayed cerebral ischaemia.
Treatment
The clinical case was discussed in a multidisciplinary meeting including neurointensivists, neuroradiologists experienced in interventional radiology, neurologists and neurosurgeons. Due to the established ischaemic lesions, the team concluded that there was no benefit in endovascular treatment of vasospasm. He was kept under ICP, CPP and haemodynamic monitoring.
TCD was repeated on the 25th day post admission, showing improvement of the flow velocities. A new CT scan of the brain was performed, showing cortical and subcortical hypodensities in the right frontal, parietal, occipital and temporal areas and left frontal and parietal lobe, with hyperdense foci of haemorrhage in their interior. Best GCS Score was 9T (E4VTM4). Serial CT scans of the brain (days 25, 30, 38 and 45 post injury) showed no further signs of intracranial bleeding. An MRI was repeated on the 54th day post injury, with the additional finding of active hydrocephalus. He was submitted to cranioplasty and placement of a ventriculoperitoneal shunt on the 56th day post admission.
No further neurological recovery was observed, with a persistent coma with GCS 7 (E4VTM2). Figure 5 presents a timeline summary of events.
- Timeline of events from admission to the diagnosis of vasospasm. Diagram created by the authors. GCS, Glasgow Coma Scale; ICU, neurocritical care unit; SAH, subarachnoid haemorrhage; TCD, transcranial Doppler.
Outcome and follow-up
There was no neurological recovery, despite reduction of the dimensions of the ventricular system after ventriculoperitoneal shunt placement. On the 65th day post admission, he evolved with leucocytosis and high C reactive protein. A diagnosis of bacteremia caused by Serratia marcescens was made. On the 70th day post admission, a cerebral MR revealed the presence of a right fronto-temporo-parietal empyema (figure 6). Despite surgical and antibiotic treatment, there was infection progression and persistent coma, leading to the decision of withdrawal of care and subsequent demise on day 85.
MRI with empyema: signs of right fronto-temporo-parietal craniotomy, with a subjacent extradural collection with 16 mm of greater width; it contains heterogeneous content, predominantly hypodense in T1 and hyperintense in T2; the imagiological findings are suggestive of empyema.
Discussion
TBI is a frequent cause of morbidity and mortality in young patients. Vasospasm is a potential complication of TBI and may contribute to secondary ischaemic lesion. Studies in TBI patients have shown that haemodynamically significant vasospasm is associated with worse neurological outcomes.18 19 Vasospasm is not classically considered as a major complication of TBI, and routine screening is not seldom performed in these patients. Its diagnosis in TBI patients is challenging and its incidence is underestimated. No current standardised screening or diagnostic approach has been established.10–12 15
We present a paradigmatic example of secondary brain lesion in TBI caused by vasospasm and delayed cerebral ischaemia. Our patient was young, had severe injuries on admission, with significant subarachnoid, cisternal and intracerebral haemorrhage, conferring higher risk for vasospasm.10–12 Vasospasm and subsequent ischaemia led to worsening of the neurological prognosis despite optimal neurocritical care.
TBI patients’ clinical deterioration is often difficult to perceive and multiple confounders may cause delays in the diagnostic process. In our case, persistent vomiting, reactivity to the orotracheal tube and sympathetic hyperactivity led to increasing needs of sedoanalgesia, limiting neurological evaluation. Vasospasm diagnosis was made after the establishment of ischaemic areas, with no opportunity for intervention or treatment.
The impact of vasospasm in TBI may be significant, contributing to a worse neurological prognosis. It can occur as early as the first 48 hours, with maximal TCD velocities reported in the 5th–7th days post injury.10 11 Later onset and longer duration of vasospasm have also been reported.13 Our patient presented an atypical course for this complication, as he evolved with partial neurological recovery in the first days post injury and delayed deterioration (after the 8th day). The presence of diffuse SAH on admission may have contributed to the occurrence and temporal evolution of vasospasm in our patient.
Although no consensus exists on an appropriate protocol for vasospasm screening, experts seem to agree that systematic screening in high-risk patients may allow detection and treatment of this complication.10 11 20 Risk factors for post-traumatic vasospasm are not completely understood, with several studies reporting heterogeneous results.13 15 Low GCS on admission and younger age seem to be the most common findings associated with an increased risk for vasospasm. Correlations have been established between this complication and other factors, including fever on admission, higher severity of lesions and presence of both subarachnoid and epidural, subdural and parenchymal haemorrhage.11 13 15
TCD constitutes a cost-effective examination for the diagnosis and serial monitoring of vasospasm.21 22 Its use in the TBI population may lead to a potential improvement of neurological prognosis due to early diagnosis of vasospasm. In our neurocritical ICU, TCD screening is routinely used for diagnosis of vasospasm in aSAH, but no protocol existed regarding TBI patients. Due to greater awareness of this potentially modifiable complication, routine TCD screening is now performed in the first days after admission (due to the earlier onset of vasospasm in TBI when compared to aSAH) in high risk TBI patients in whom neurological examination is limited.
Learning points
Vasospasm is a possible complication of traumatic brain injury (TBI) and its incidence and significance seem to be largely underestimated.
Delayed cerebral ischaemia due to vasospasm can contribute significantly to patient prognosis in TBI.
The approach of TBI patients is complex and more studies are needed to determine the impact and management of vasospasm on the neurological outcome.
Suspicion and screening for vasospasm can contribute to better outcomes.
Transcranial Doppler is a safe, bedside examination that can be used for screening and monitoring of vasospasm and its use can allow for timely diagnosis and treatment of this complication.
Ethics statements
Patient consent for publication
References
Footnotes
Contributors All authors (IMC, AAS, EM and LA) were responsible for drafting of the text, sourcing and editing of clinical images, investigation results, drawing original diagrams and algorithms and critical revision for important intellectual content. All authors gave final approval of the manuscript. Guarantor: IMC.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.