Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

Pediatric Neurocritical Care is an emerging multidisciplinary field of clinical and experimental medicine. While there are challenges to implementing specialized units for pediatric neurologic critical care, and limited evidence on which to base clinical practice, several models now exist for pediatric neurocritical care programs in centers that are contributing to the development of treatment guidelines and protocols; including multimodal neuromonitoring [1, 2]. Recent multi-center studies investigating pediatric brain injury in stroke, status epilepticus, and hypothermia after cardiac arrest and traumatic brain injury have served to promote the feasibility of accomplishing brain-directed research in children [3–7]. Several organizations, such as the Pediatric Emergency Care Applied Research Network (PECARN) and the Pediatric Neurocritical Care Research Group (PNCRG) are working on advancing care in highly specialized training programs across the United States. Pediatric neurocritical care exists thanks to technological advances in pediatric critical care, neurology, neurosurgery, and anesthesiology. The main goal of pediatric neurocritical care is to improve outcomes in infants and children with life-threatening neurologic injuries, and to prevent the development of secondary neurologic and non-neurologic injuries. This chapter briefly covers some of the most common neurologic conditions encountered in the pediatric intensive care unit (PICU).

Traumatic Brain Injury

Traumatic brain injury (TBI) remains a leading cause of disability and mortality in infants, children and adolescents across the United States, and constitutes a significant portion of PICU admissions [8, 9]. In the United States alone, TBI affects over half a million children ages 0–19 annually, including 630,000 emergency department visits, 60,000 hospitalizations and over 6100 deaths [10]. It is estimated that five times as many children will succumb from the devastating effects of acute brain injury (hypoxic-ischemic or traumatic) than childhood neoplasias [9]. Guidelines for the care and management of pediatric severe traumatic brain injury were first published in 2003 and updated in 2012 [11, 12].

TBI pathophysiology occurs in two known different phases of care: primary and secondary injury. Currently, there is little that can be done to reverse primary injury, which is the damage resulting at the moment of trauma. Secondary or delayed injury is most commonly caused by physiologic insults such as hypotension and hypoxia. Inflammatory, metabolic, and excitotoxic mechanisms also represent a wide spectrum of potential secondary insults, leading to cerebral edema and intracranial hypertension. Fortunately, secondary injury is a potentially preventable and treatable condition.

The Kennard principle proposes the idea that pediatric recovery after TBI would be enhanced due to a higher degree of neural plasticity in the developing brain [13]. For years, it was thought that children had a greater ability to recover from TBI. However, the Kennard principal referred to localized or focal lesions, not diffuse brain injuries; and outcome following TBI in children may not be better than in adults [14, 15]. Children suffering from TBI classified into favorable outcome groups still may exhibit long-term underappreciated cognitive and behavioral impairments [16, 17]. Moreover, in comparison with their peers, TBI leads to decreased academic achievement, lower scores on intelligence testing, reduced ability to focus, and other neurocognitive deficits [18, 19].

TBI has special connotations in the pediatric population as the impact of the injury may not be entirely evident until many years later as developmental stages are acquired. Falls, bicycle, motor pedestrian and motor vehicle collisions are the most common causes of injury in young children and adolescents. However, inflicted or abusive head trauma (AHT)- also termed non-accidental trauma (NAT)- deserves special mention as one of the leading causes of TBI in infants <2 years [20], as this is frequently associated with significantly worse outcomes than accidental mechanisms [21, 22]. Abusive TBI is typically defined by a triad of physical and radiologic findings including, hyperdense and hypodense subdural collections, retinal hemorrhages, and some degree of encephalopathy [23]. Brain injury in these patients is associated with repetitive trauma, delay in seeking medical care and high incidence of hypoxic-ischemic injury and seizures, which all may serve to worsen neurologic outcomes [24]. Seizures have been reported to occur in as many as 77 % of AHT patients who had continuous EEG monitoring, significantly higher than rates of older children and adults with accidental mechanisms. Of particular importance is the high frequency of subclinical events. In a report by Arndt et al., subclinical seizures occurred in 16 % of children with TBI; however all of the children with only subclinical seizures were < 1 year old and subclinical status epilepticus occurred in 45 % of AHT infants [25]. In this study subclinical seizures and subclinical status epilepticus were associated with worse hospital discharge outcome scores. Thus AHT and young children may benefit from additional EEG monitoring to detect subclinical seizure activity [25]. Every year, 3 million cases of child abuse and neglect are reported to child welfare systems in the USA, and of these one-third are substantiated [26]. Despite well described injury patterns, NAT is a difficult and serious diagnosis due to the widespread implications beyond the patient [27]. Currently, serum and cerebrospinal biomarkers are under investigation to determine differences between accidental and NAT [28].

As compared to adolescents, children have a higher incidence of diffuse axonal injury (DAI), SDH, and cerebral edema [29]. In turn, adolescents show a higher incidence of DAI and contusions compared to the adult population [30]. There are also reports of a higher magnitude of cerebral edema after TBI in pediatric populations, likely due to a more heterogeneous vascular and inflammatory response [31, 32]. Also, a child’s skull is more susceptible to suffer a higher degree of deformity before reaching its compliance limit [33, 34].

Animal data suggests that immature and developing cerebral tissue is at higher risk of apoptotic cell death [35]. In fact, increased levels of apoptosis-related proteins such as cytochrome c, Fas, and caspase-1 have been observed in children after TBI [36]. In addition, increased CSF levels of neuron-specific enolase which is a pro-apoptotic protein have been found after TBI [37].

Pediatric TBI Management

Cervical Spine and Airway Management

Although not as frequent as in adults, associated cervical spine (C-spine) injury can be seen after sustaining severe blunt cranial trauma. It is estimated that up to 25 % of patients suffering C-spine injury develop neurologic deficits caused by pre-hospital manipulation [38]. C-spine evaluation in children must take into consideration the anatomic development (Table 6.1). Any child at risk of having a C-spine injury must be immobilized in a neutral position until injury is ruled out. Patients less than 8 years of age are more susceptible to injury of the upper cervical spine as the maximal motion occurs at C1–C3. After 12 years of age, maximal movement occurs around C5–C6. C-spine clearance should follow the current pediatric guidelines for children with either reliable or unreliable physical examinations. In children, spinal cord injury may occur without radiographic evidence (SCIWORI) necessitating reliance on the clinical exam or magnetic resonance imaging to detect injuries that may not be evident on conventional radiographic imaging. Patients with severe TBI (i.e., TBI with coma; GCS <9), or worsening mentation should have a definitive secure airway established [39]. Initially, the airway can be opened with a jaw thrust and chin-lift maneuver while maintaining cervical immobilization (by an assistant). Endotracheal intubation is always a potentially challenging scenario in trauma patients. It is recommended to perform orotracheal intubation with in-line manual stabilization to prevent further spinal cord injury. In-line manual stabilization should be performed by an experienced provider. Nasotracheal intubation should be avoided in patients with facial trauma and/or signs of skull base fracture.

Table 6.1 Pediatric C-spine anatomic considerations
Full size table

Endotracheal intubation is best achieved using rapid sequence induction/intubation with application of gentle cricoid pressure. Etomidate is frequently used for this purpose as it decreases ICP without significant reductions in mean arterial pressure [40]. There is no definitive evidence showing that succinylcholine increases ICP in humans with brain injury [41, 42]. Therefore, the use of succinylcholine vs. rocuronium for rapid sequence induction should be based on other clinical factors and provider expertise. Prolonged hyperventilation during mechanical ventilation should be avoided as it may cause cerebral tissue ischemia resulting from oligemia [43]. Arterial PaCO2 should be monitored and normocapnea, PaCO2 35-40 mmHg, should be targeted except in the setting of reversal of clinical herniation syndrome where lower PaCO2 levels can be temporarily used. The head should be elevated, either at 30° (or equivalent reverese trendelenburg tilt) to improve cerebral venous drainage. Studies performed in adults suggest that head elevation to 30° improves CPP and reduces ICP [44]. Also, the head should be maintained in a neutral position to avoid obstruction of jugular venous outflow.

As in adults, children who are hypotensive during the first hours of hospital care have worse outcomes [45, 46]. The lower limit of systolic blood pressure should be maintained greater than the 5th percentile for age (estimated by 70 mmHg + (2 × age in years). Evidence supports that better outcomes are achieved in children who receive early fluid resuscitation [47]. Therefore, it is imperative to urgently initiate resuscitation with isotonic fluids to correct hypotension and hypovolemia.

Current guidelines for the management of pediatric TBI recommend the use of intravenous agents such as analgesics, sedatives, and neuromuscular blockers as adjuvants to prevent or minimize secondary brain injury and intracranial hypertension [12, 43]. There are few studies addressing the choice of agent, however the use of these agents should be limited to patients who are hemodynamically stable with a secure airway. Decompressive craniectomy and barbiturate therapy are also used to reduce ICP and improve CPP. While there are currently few studies that would lend support to adopting a standard of care recommendation, these therapies may be consider using effective in “control of refractory ICP” or “treating refractory ICP” control and should be considered in children with a salvageable or recoverable injury [12]. Contrary to adult guidelines, pediatric use of propofol infusion is not FDA approved due to its associated morbidity [11]. The incidence of early post-traumatic seizures in children ranges from 5 to 43 %; risk factors include young age (<2 years), AHT, skull fracture and severe head injury [25, 48, 49]. In a randomized trial of 102 children with acute TBI, Young et al., found empiric phenytoin versus placebo did not affect the incidence of post-traumatic seizures (7 % vs. 5 %) or outcome [48]. Prophylactic anticonvulsant use varies widely among centers (10–35 %), current recommendations state that prophylactic treatment with anti-seizure medications can be considered and might decrease the onset of post-traumatic seizures in children, and improve outcomes [50, 51].

It is well known that hyperglycemia is associated with worse outcomes after TBI in adults and children. This may be secondary to worsening of lactic acidosis at a brain tissue level [52]. Sharma et al. [53] in 2009 observed that predictors of hyperglycemia were children <4 years old, GCS ≤ 8, and multiple traumatic injuries including SDH. Currently, it is not completely clear what the upper cutoff for hyperglycemia in children should be [43]. The majority of centers recommend that hyperglycemia should be corrected in acute childhood TBI. Of note, steroid administration in children following TBI has not been associated with additional benefit or improved outcomes [11]. In fact, evidence suggests increased morbidity and mortality after its use [54].

Intracranial Hypertension Management (See Also Table 6.2)

Table 6.2 Pediatric ICP management
Full size table

Multiple clinical trials have shown the beneficial effect of hyperosmolar therapy (mannitol or hypertonic saline) in decreasing ICP in children [55, 56]. Potential concerns with using hypertonic saline include dehydration, natriuresis, central pontine myelinolysis, and the theoretical concern for rebound intracranial hypertension in the setting of a disrupted blood brain barrier [57]. The 2012 severe pediatric TBI guidelines recommend that either bolus or infusion therapy is effective in reducing intracranial pressure in children [11]. The use of mannitol in children has not been well studied. Several potential complications have led to a decrement of its use, i.e. volume depletion, hypotension, acute renal injury particularly in hypovolemic patients, a lower reflection coefficient than sodium chloride, and the potential reverse osmotic effect leading to an increase in ICP.

Hyperventilation therapy linearly reduces CBF and ICP due to hypocapnia via cerebral vasoconstriction; sustained hyperventilation (PaCO2 < 30 mmHg) has been associated with regional cerebral ischemia in up to 73 % of patients and poorer long-term outcomes [58, 59]. Current guidelines recommend against routine use of hyperventilation to a PaCO2 less than 30 mmHg. If used, concomitant advanced neuromonitoring should be provided [11, 43].

CSF drainage works by reducing the amount of intracranial fluid and achieves immediate reductions in intracranial pressure. An external ventricular drain can be used to both remove CSF and to monitor ICP. CSF drainage is highly effective until cerebral edema produces ventricular collapse.

Barbiturate coma is recommended in the hemodynamically stable patient when maximal medical and surgical therapy has failed to manage elevated ICP. Barbiturates decrease the cerebral metabolic rate (~50 % at the point of burst suppression), with concomitant decreases in CBF, brain bulk, and subsequently ICP [60]. Ionotropic and vasopressor support to avoid hypotension and maintain adequate mean arterial pressures are commonly required after starting barbiturate coma. Continuous EEG monitoring is used to guide therapy, both to maintain a burst suppression profile as well as to monitor for subclinical seizures (Fig. 6.1).

Fig. 6.1
figure 1

Typical burst-suppression pattern observed during EEG recording while in pentobarbital coma

Full size image

Decompressive craniectomy should be considered in pediatric patients who are experiencing signs of cerebral herniation or in those with intracranial hypertension refractory to medical treatment. The ideal timing for performance is still a debate (early vs. late/rescue). Decompressive craniectomy can clearly be life-saving with good neurologic outcome in select cases. However, the application and timing of this surgery is debated as the evidence base is limited, particularly in children. Decompressive craniectomy has been reported in small pediatric case series and retrospective studies to be effective in lowering ICP in children with refractory ICP elevation with reports of good outcomes [61, 62]. Recommendations are limited by the small sample size, single center and retrospective design and lack of adequate case controls for comparison. Taylor et al., randomized 27 children with severe TBI and refractory ICP to early bitemporal decompressive craniectomy (mean 19.2 h form injury) versus maximal medical therapy. The mean ICP was lower in the craniectomy group 48 h after randomization and outcome appeared to be improved (normal or mild disability 54 % in craniectomy group versus 14 % in the medical group) [63]. The Decompressive Craniectomy in Diffuse Traumatic Brain Injury (DECRA) trial did not show benefit to bifrontal craniectomy in comparison to medical therapy in adults (Median age ~24 years) [64]. ICP was reduced, along with length of stay in the intensive care unit, but neurologic outcomes at 6 months were worse. The rescueICP (Randomised Evaluation of Surgery with Craniectomy for Uncontrollable Elevation of Intra-Cranial Pressure) trial has finished enrollment but results have not yet been published [65]. Randomized trials are needed in children to determine the safety, efficacy, timing, and optimal patient population for decompressive craniectomy in pediatric TBI patients.

Therapeutic hypothermia can be used to lower ICP (decreases cerebral metabolic rate ~6 % per degree Celsius) and it remains a therapeutic option for controlling refractory ICP [66], however 2 prospective randomized clinical trials have failed to demonstrate a benefit of early prophylactic hypothermia on neurologic outcome [3, 4]. In these studies, children with severe TBI were randomized to hypothermia 32–34 °C within 6–8 h of injury for either 24 or 48–72 h, versus normothermia 36.5–37.5 °C, with slow rewarming. The authors found no difference in the proportion of children with unfavorable outcomes at 3 or 6 months. Hyperthermia is thought to be injurious in the setting of acute brain injury of any etiology. Therefore, ‘targeted temperature management’ is becoming standard of care in the patient with acute neurologic injury [67]. A prospective multi-center international trial is ongoing, Approaches and Decisions for Acute Pediatric TBI (ADAPT) trial that will use a comparative effectiveness strategy to test intracranial hypertension therapies, brain tissue oxygenation monitoring, hyperventilation, and nutrition on neurologic outcome in severe pediatric TBI. With a proposed enrollment of 1000 children, this will be the largest prospective dataset in pediatric TBI obtained to date.

Hypoxic Ischemic Brain Injury

Neonatal asphyxia remains a common cause of hypoxic brain insults in the pediatric population. Unlike TBI, high level evidence exists for the efficacy of therapeutic hypothermia in this setting. Standard of care in neonatal resuscitation is the induction of moderate hypothermia for the treatment of moderate-severe hypoxic ischemic encephalopathy [68]. American Heart Association Guidelines [68] recommend a target temperature of 33.5–34.5 °C (initiated within 6 h of birth) for 72 h followed by a controlled rewarming period, avoiding overshoot hyperthermia. Neonates seem particularly sensitive/vulnerable to hyperoxia, so FiO2 should be minimized with pulse oximetry guidance.

Unlike adults, the primary mechanism for cardiac arrest in children is secondary to respiratory failure, thus establishment of an airway, oxygenation and ventilation with bag mask ventilation (BMV) or endotracheal intubation should be rapidly instituted. Standard therapy for post-anoxic cerebral resuscitation should be targeted at optimizing systemic hemodynamic and physiologic variables and avoiding secondary insults. Poor prognostic factors include myoclonic status, non-reactive or burst suppression pattern on EEG, absent pupillary reflex at 72 h and abnormal somatosensory evoked potentials.

The role of induced moderate hypothermia in neuroprotection of children beyond the neonatal period is less clear. Pediatric Advanced Life Support (PALS) guidelines suggested that the immediate induction of therapeutic hypothermia (Temperature 32-34 °C) may be beneficial but that further study is needed [69]. In a multicenter study of Therapeutic Hypothermia after Pediatric Cardiac Arrest (THAPCA) trial, 295 comatose children were randomized within 6–8 h of return of spontaneous circulation after cardiac arrest to either a target temperature of 33.0 for 48 hours vs. 36.8. I am not sure if it is important but in this study normothermia was also controlled for 5 days after the injury or 36.8 °C) [5]. In this study, which first published the out-of–hospital cohort, there was no difference in 1-year neurologic function or survival in the hypothermia group vs. the controlled normothermia group. Analysis of the in-hospital arrest cohort and subgroup analysis may provide additional insight data. Hyperthermia should be avoided after cardiac arrest as it increases cellular energy metabolism and release of excitotoxicity chemicals and accelerates apoptotic pathways. In neonates after HIE, the odds of death or disability are increased 3.6–4 fold for each 1 °C increase above 38 °C) [70]. Additionally, hypotension and hyperventilation should be avoided in these patients to avoid cerebral hypoperfusion and oligemia.

Stroke

Although more common in the elderly, arterial and venous strokes also occur in neonates, infants, and young adolescents, and result in significant mortality and long-term disability [71]. Overall, the occurrence is at least as frequent as the number of pediatric tumors [72]. Reported incidence is variable, ranging from 0.9 to 13 cases per 100,000 children [73, 74]. Neonates represent the group at the highest risk, with an incidence for ischemic and hemorrhagic stroke of 1 per 3500, and 1 per 16,000 live births respectively [73, 75]. Whereas in adults ischemic stroke is more common, hemorrhagic stroke is more equally distributed in children due to the higher prevalence of vascular malformations.

Common causes of acute ischemic stroke in children include embolic sources from congenital cardiac defects, sickle cell disease, and head and neck infections (Table 6.3). Venous strokes can be due to dehydration, infection, and hyperosmolar states. Etiology varies again in young adults, with vasculopathy, cardiac defects, smoking, pregnancy, drug use, hypercoagulable states, and premature atherosclerosis among the most common [76]. Hemorrhagic strokes are most commonly due to arteriovenous malformations, cavernomas, tumors, and coagulopathy [77].

Table 6.3 Causes of acute ischemic stroke in children [76, 78, 79]
Full size table

Clinical Presentation

Diagnosis of stroke in children is frequently delayed, with an average time to establish a diagnosis of >24 h [80–82]. The delay in diagnosis may be related to the community misperception that children are not at risk for strokes and that pediatric patients are likely to manifest more subtle non-focal signs such as seizures or altered mental status [83]. Also, a low clinical suspicion for acute ischemic stroke by healthcare providers plays an important role in the diagnostic delay as stroke symptoms are often attributed to other more common diagnoses [84]. Even with more classic signs such as hemiparesis and aphasia, presentation to a tertiary pediatric facility is delayed; advanced stroke centers and evidence based protocols are lacking for children. Whereas in neonates seizures often are the presenting symptom, the most common clinical presentation of stroke in older children is hemiparesis, with the middle cerebral artery territory being most frequently affected [85].

Evaluation and Management

Neuroimaging is a cornerstone of stroke diagnosis in children. CT is no longer recognized as the gold standard initial test in children. Current United Kingdom guidelines recommend performing MRI as soon as possible after admission, as MRI is more sensitive than CT in detecting ischemic stroke [86]. Also, MRI will help differentiating clinical conditions mimicking stroke (migraine, seizure, encephalitis, other intracranial lesion) which may be seen in a significant proportion of patients.

As stroke etiology in children can encompass a broad spectrum of conditions, beyond a general laboratory and toxicology screens, patients should be evaluated for hypercoagulable states (i.e. Protein C and S deficiency, homocysteine, Lupus, Factor V Leiden), intracardiac lesions, vasculitis, and mitochondrial disorders.

Pediatric stroke management closely follows the guidelines adapted for adults (https://www.rcplondon.ac.uk/sites/default/files/documents/stroke-in-childhood-guideline.pdf) [87]. However, anticoagulation and thrombolysis therapy may differ. Currently, alteplase (rt-PA) is not FDA approved for use in children less than 18 years of age with ischemic stroke, and endovascular thrombolysis/mechanical thrombectomy are not routinely used in children <14 years of age [88]. For adolescents >15 years, thrombolytic use should be considered on an individual basis. Now in 2015, with 5 prospective randomized trials [89] demonstrating the superiority of stent-like retrievers over intravenous rt-PA in improving the outcome of adults with acute large vessel stroke, the application of mechanical thrombectomy to pediatric acute ischemic stroke is likely to expand.

Currently, there are no trials showing the efficacy of anticoagulation or antithrombotic therapy in children with acute arterial ischemic stroke. The American Heart Association considers reasonable the use of LMWH or unfractionated heparin until full work-up is completed [88]. The Royal College of Physicians recommends the use of aspirin as initial therapy (https://www.rcplondon.ac.uk/sites/default/files/documents/stroke-in-childhood-guideline.pdf) (Table 6.4).

Table 6.4 Recommendations vary for management of specific causes of ischemic stroke
Full size table

Intensive care unit management of pediatric stroke patients largely follows adult treatment goals. Intubation is instituted for airway protection due to depressed level of conciousness or for maintenance of oxygenation and ventilation. Euvolemia and adequate mean arterial systemic pressure should be maintained. Anticonvulsant medications may be considered in individual cases, and patients should have their temperature controlled to prevent fever. Depending on the stroke etiology pediatric neurosurgery, neurology and neuro-interventional radiology consults may be required. Management of intracranial hypertension may require ICP monitoring, sedatives, hyperosmolar therapy and barbiturate therapy. Decompressive craniectomy is reserved for children presenting with large stroke(s) involving the middle cerebral artery territory causing malignant (i.e., life threatening) cerebral edema with intracranial hypertension, midline shift, and decline in neurologic exam [90].

Status Epilepticus (SE)

SE is one of the most common pediatric neurologic emergencies. Traditionally, SE is defined as a seizure that lasts more than 30 min, or occurs frequently enough that the patient does not recover consciousness in between episodes [91]. However, some experts suggest that SE definition should include those patients with seizures lasting more than 5–10 min [92], as the risk for a worse outcome and the potential for seizures to be refractory to anti-seizure medications increases with longer ictal duration [93].

The highest incidence of SE is observed during the first year of life due to febrile seizures [94]. Risk factors identified with recurrent SE are symptomatic established epilepsy [91], young age at onset, and genetic syndromes (i.e. Angelman syndrome, Dravet syndrome) [95]. SE may also be the manifestation of metabolic abnormalities, CNS infections, tumors, illicit drug abuse, hypoxic-ischemic injuries, child abuse, heat stroke, TBI and fever, among others [94].

Of great importance is to diminish secondary complications associated with SE, including hypoxia, acidosis, myoglobinuria, hyperkalemia, intracranial hypertension, and hemodynamic instability [96]. SE can be fatal in some cases, accounting for mortality close to 10 % [95, 97–99]. Long-term outcomes depend on the underlying cause, the duration of event, and the child’s age [98].

Management

In patients with epilepsy, it is crucial to know the response to previous antiepileptic drugs to guide the treatment approach, and to obtain a focused history from parents or caregivers. Common causes for SE in children are intercurrent infection, recent changes in medications, missed medications, or inadequate antiepileptic medication dosing. The approach should include an assessment of respiratory and circulatory status, intravenous access, and neurologic examination to determine type and possible precipitants. Laboratory workup should include screening for infection, level of current AEDs, sodium, and glucose levels. Neuroimaging studies (CT and/or MRI) are used to exclude other pathologies (such as hematoma, tumor, or stroke) and should be used in patients who have new onset SE, focal neurologic deficits or have not responded to initial therapy by regaining consciousness. Continuous video EEG may also be appropriate for SE management for patients with persistent encephalopathy, especially if the child’s neurologic status is impaired beyond baseline making clinical correlation difficult or for assessment for non-convulsive SE.

Pharmacologic management is based on the guidelines published by the Neurocritical Care Society in 2012 [100]. Benzodiazepines (lorazepam, midazolam, or diazepam) are first-line treatment as they can quickly achieve seizure control. If seizures persist for 10 min after at least 2 doses of BDZ, fosphenytoin should be loaded at a dose of 20 mg/kg IV. If seizures persist, a third-line drug is initiated (phenobarbital, valproic acid, levetiracetam, lacosamide) and placement of a secure airway should be considered. For refractory SE (RSE) cases not responsive to standard therapies treatment options are multiple. Intravenous anesthetics are administered with continuous EEG guidance. Intravenous infusions of midazolam at high doses may be used to achieve seizure control; or pentobarbital infusion to produce burst-suppression pattern. Severe hypotension and or respiratory depression may occur with initiation of IV infusions requiring mechanical ventilation or vasopressor support, thus airway and continuous hemodynamic monitoring should be available when starting these therapies. After 24-48 h of seizure control, the infusion is slowly titrated to off (over many hours) and the patient is monitored for seizure recurrence. Ketamine infusions have been reported in small case series to be effective in RSE in children who failed to respond to barbiturate therapy [101]. Propofol is infrequently used in pediatrics, due to the risk of propofol infusion syndrome [102].

Multimodal Cerebral Monitoring

A fundamental goal of neurocritical care is to prevent the development of secondary neurologic injury after the initial (and typically irreversible) cerebral insult. In the brain-injured child multiple physiologic parameters need to be simultaneously managed and optimized in order to achieve the best posible outcome. In addition to standard physiologic monitors (pulse oximeter, EKG, blood pressure), patients with neurologic conditions frequently require specific cerebral monitoring to avoid and promptly recognize the occurrence of secondary injuries. Such monitors include ICP measurement devices (Fig. 6.2), measures of global and regional cerebral oxygenation [near infrared spectroscopy (NIRS), jugular venous oxygen saturation (SjvO2), or partial brain tissue oxygen tension (PbtO2)], measures of cerebral metabolism (cerebral microdialysis), continuous EEG monitoring (discussed above), transcranial Doppler ultrasound (Fig. 6.3), and cerebral blood flow (CBF) measurements [103]. Multimodal neuromonitoring involves understanding not only displayed numbers, but also data acquisition, informatics challenges, device interoperability-related issues, and longitudinal data analysis, among others. Observational studies suggest that multimodal neuromonitoring provides accurate and unique information when used to individualize management of severe head injured patients. Clinical outcome data are sparse.

Fig. 6.2
figure 2

Diagram demonstrating multimodal monitoring

Full size image
Fig. 6.3
figure 3

Circle of Willis defined by color-coded transcranial Doppler ultrasound

Full size image

Partial Brain Tissue Oxygen Tension (PbtO2)

PbtO2 can be measured by inserting an oxygen electrode into the brain parenchyma. Cerebral blood tissue oxygen tension is continuously measured and threshold values for treatment are generally to maintain PbtO2 > 15–20 mmHg [11, 104]. Most of the current pediatric data is focused on TBI, however, there are some other potential applications for this technology including pediatric stroke, and management of cerebral edema during diabetic ketoacidosis [105, 106]. Low levels have been associated with poor outcomes after TBI [107].

Jugular Venous Oximetry (SjvO2)

Through retrograde cannulation of the internal jugular vein, SjvO2 can be measured at the level of the jugular bulb. Normal values in adults are considered to be between 55 and 75 mmHg [108]. There is limited evidence for the utility of its use in pediatric or adult population.

Cerebral Microdialysis

Cerebral microdialysis allows the determination of the metabolic state of the brain. An intraparenchymal probe is inserted into the brain tissue to determine levels of extracellular pyruvate, lactate, glucose, glutamate, and glycerol [109]. In adults, elevation of lactate, and/or a lactate:pyruvate ratio >40, are suggestive of anaerobic metabolism which could exacerbate secondary cerebral injury [110]. Currently, there is limited evidence for its application in children. A small pilot study by Richards et al., in 2003, showed that decreased glutamine:glutamate ratio could be an outcome predictor after brain injury [111]. Overall, its current use can be considered experimental [112].

Thermal Diffusion Cerebral Blood Flow and Cerebral Oximetry

Point-of-care continuous cerebral blood flow monitoring at a regional (such as lobar) or global level is a holy grail in neurocritical care but beyond currently available technology. Quantitative CBF can be obtained with CT perfusion, MR perfusion, or positron emission tomography (PET), but not in a frequent or continuous manner that would allow titration of hemodynamic and intracranial pressure therapies. Focal cerebral blood flow can be monitored continuously with thermal diffusion technology using a parenchymal probe inserted through a cranial bolt. However, whether such data allows optimization of outcome, or simply exposes the patient to added risk, requires further study. Trancutaneous near-infrared cerebral oximetry can be used to monitor cerebral oxygen saturation in a continuous manner in the intensive care unit. While new monitors are being developed, data to support widespread use in neurocritical care is very limited [103].

Transcranial Doppler Ultrasound

Doppler ultrasound (typically 2 MHz) insonated through the temporal bone ‘windows’, foramen magnum, and orbits can be used to assess cerebral blood flow velocity and direction. Coupled with B-mode sonography, transcranial Duplex can be performed (Fig. 6.3) in order to obtain some information regarding vascular anatomy [103]. TCD is routinely used for vascular screening in sickle cell disease but has not seen widespread adoption in pediatric neurocritical care; however, it is extremely low risk and can be used to assess pediatric cerebrovascular disease including vasospasm, TBI/post-traumatic vasospasm, and for the assessment of cerebral blood flow (ie, global oligemia vs. hyperemia) [113, 114].

Conclusion

In summary, pediatric neurocritical care is an emerging field in which providers must be able to integrate multiple neurologic specialties in the care of the brain-injured child. Neurologic injury and disorders are common in the PICU, representing approximately 20 % of all admissions, and are associated with a longer length of stay and higher mortality than general ICU patients [1, 115]. Pathways into pediatric neurocritical care are less established than adult neurocritical care programs and need to be matured. Finally, an ever expanding evidence base, both for neurocritical care pharmacotherapeutics and advanced cerebral monitoring technologies, is shaping the field and will warrant the need for increased numbers of subspecialized pediatric neurointensivists.