Multi-modality Neurologic Monitoring

Technical Aspects

There are two main methods of monitoring ICP—an intraventricular catheter or parenchymal microtransducer device. Other techniques, such as subarachnoid or epidural devices, are less accurate and now rarely used.

Ventricular catheters measure global ICP (CSF pressure in the lateral ventricles) in one of two ways: using a standard catheter connected via a fluid-filled system to an external pressure transducer, or a catheter incorporating microstrain gauge or fiberoptic technology. In vivo calibration is possible with both methods. ICP monitoring via a ventricular catheter is the technique of choice in patients with established or incipient hydrocephalus as it allows therapeutic drainage of CSF. Ventricular catheter insertion can be difficult, and there is a risk of placement-related hemorrhage and catheter-associated ventriculitis. The risk of ventriculitis increases with the length of time since catheter insertion, and can be reduced but not abolished by the use of antibiotic-impregnated or silver-coated catheters.

Parenchymal microtransducer ICP monitoring systems are of two types. Solid-state piezoelectric strain gauge devices incorporate pressure-sensitive resistors which translate pressure generated changes in resistance to an ICP value. Fiberoptic devices transmit light via a fiberoptic cable towards a displaceable mirror at the catheter tip. Changes in ICP distort the mirror and the difference in intensity of reflected light is converted into an ICP value. Both systems are easy to insert and are usually placed about 2 cm into brain parenchyma through a cranial access device or at craniotomy when they can also be sited in the subdural space. While intraventricular catheters have historically been considered the “gold standard” for ICP monitoring, intraparenchymal devices provide equivalent pressure measurement and are safer. In particular, the risk of hematoma and infection is low. Microtransducer systems are considered reliable, but zero drift can result in measurement error over several days and in vivo calibration is not possible.

Several noninvasive ICP monitoring techniques, which would be applicable in broader patient groups than invasive monitoring, have been described. Transcranial Doppler (TCD) ultrasonography-derived pulsatility index is an imprecise assessment of ICP compared with invasively measured pressure, and there is large intra- and interoperator variability. Optic nerve sheath diameter (ONSD) can be measured using ultrasound or computed tomography (CT), and is able to predict intracranial hypertension. Although lacking the risks of invasive methods, noninvasive ICP monitoring techniques currently fail to measure ICP sufficiently accurately for routine clinical use. They are also unable to monitor intracranial dynamics continuously.

Indications

Despite the absence of high-quality evidence demonstrating outcome benefits of ICP-guided management, ICP monitoring has become a standard of care after traumatic brain injury (TBI). It can also provide valuable information in other brain injury types, as well as in patients with hydrocephalus (it can also be useful for chronic monitoring in normal pressure hydrocephalus), and those undergoing craniotomy for lesions producing mass effect.

The Brain Trauma Foundation (BTF) recommends ICP monitoring to guide ICP- and CPP-directed therapy after severe TBI in all salvageable patients with an abnormal cranial CT scan, as well as in those with a normal scan and two or more of: age > 40 years, unilateral or bilateral motor posturing, and/or systolic blood pressure < 90 mmHg. A recent expert statement provides more detailed and updated guidance. It recommends ICP monitoring in comatose TBI patients with cerebral contusions when interruption of sedation to check neurologic status is dangerous, or when the clinical examination is unreliable. ICP monitoring is also recommended following a secondary decompressive craniectomy and should be considered after evacuation of an acute supratentorial intracranial hematoma in salvageable patients at increased risk of intracranial hypertension, including those with a GCS motor score ≤ 5, pupillary abnormalities, prolonged/severe hypoxia and/or hypotension, cranial CT findings suggestive of raised ICP, or intraoperative brain swelling. The expert group recommended that ICP monitoring should also be considered in TBI patients with extracranial injuries requiring multiple surgical procedures and/or prolonged analgesia and sedation.

ICP monitoring is increasingly being incorporated into protocols for the critical care management of subarachnoid (SAH) and intracerebral hemorrhage (ICH), although these indications are not as well-defined or well-studied compared with TBI.

Thresholds for Treatment and Evidence

Normal ICP varies with age and body position; in healthy, resting supine adults, normal mean ICP is 5–10 mmHg. ICP greater than 20–25 mmHg is widely considered to require treatment after TBI, although higher and lower thresholds are described. The presence of intracranial hypertension is detrimental and the time spent above a defined ICP threshold, as well as absolute ICP, is an important determinant of poor outcome. Changes in the ICP waveform occur as ICP increases, and waveform analysis has been used to predict the development of intracranial hypertension. While indication of critical reductions in intracranial compliance, before intracranial compensatory mechanisms have become exhausted, would allow more timely clinical intervention, clinical translation of this technique remains elusive.

A meta-analysis of 14 studies and 24,792 patients with severe TBI found that ICP monitoring-guided management of intracranial hypertension had no significant mortality benefit overall compared with treatment in the absence of ICP monitoring, although mortality was lower in patients who underwent ICP monitoring in studies published after 2012. The only randomized controlled trial of ICP-guided management after TBI found no difference in 3- or 6-month outcomes in 324 patients in whom treatment was guided by ICP monitoring compared with imaging and clinical examination in the absence of ICP monitoring. The non-ICP monitoring group in this study received protocol-specified but empirical treatment on a fixed schedule basis, and the wider applicability of such an approach is questionable given evidence that one of the interventions (mannitol) has a more beneficial effect when directed by monitored rises in ICP. In contrast to previous studies, those undergoing ICP monitoring received significantly fewer days of ICP-directed treatment (hyperventilation, hypertonic saline/mannitol and barbiturates) than those in the ICP monitored group, although the length of intensive care unit (ICU) stay was similar. It remains to be seen whether the findings of this study, conducted in Bolivia and Ecuador, are applicable to populations with access to superior pre-hospital care and rehabilitation services. When interpreting this study it is also important to note that evaluation and diagnosis of intracranial hypertension, either by monitoring ICP or assessment of clinical and imaging variables, was fundamental to the management of all patients. It therefore reinforces the widely held view that assessment of ICP is an integral part of the monitoring and management of severe TBI.

Multimodal monitoring incorporating ICP, brain tissue partial pressure of oxygen (PbtO ₂ ) and cerebral microdialysis (CMD) is significantly more accurate in predicting cerebral hypoperfusion than ICP monitoring alone, and ICP monitoring is, therefore, best regarded as one part of a multimodal monitoring strategy rather than as a monitoring modality in isolation.

Technical Aspects

CPP is calculated as the difference between mean arterial pressure (MAP) and ICP, and modifiable through this relationship. For the calculation of CPP to be accurate the arterial pressure transducer must be referenced at the level of the foramen of Monro (tragus of the ear), and the implications of failing to do this are profound. When the head of the bed is elevated, measuring arterial blood pressure (ABP) at the level of the heart results in a calculated CPP that is erroneously high; for example, a measured CPP reading of 60 mmHg may actually represent a “true” CPP of < 45 mmHg. Such measurement discrepancies are exacerbated in tall patients, with varying elevations of the head of the bed, and different sites of arterial cannulation.

Indications

The main indications for CPP monitoring are similar to those for ICP. Although CPP-guided management is primarily applied in patients with TBI there is emerging evidence for a role in other brain injury types. Postoperative ICP monitoring is indicated if there is a risk of intracranial hypertension, such as after surgery for large brain tumors with mass effect.

Thresholds for Treatment and Evidence

Recommendations for a CPP threshold have changed over time. Current BTF guidelines recommend that CPP be maintained between 50 and 70 mmHg after TBI, with evidence of adverse outcomes if CPP is lower or higher. While lower CPP risks cerebral ischemia, higher CPP does not necessarily achieve a favorable outcome, and the administration of large fluid volumes and inotropes/vasopressors to maintain CPP carries a substantial risk of acute lung injury. Rather than a single threshold for CPP, multimodal monitoring can be used to identify an “optimal” value for an individual, which aims to reduce the risks of an excessive CPP while minimizing the risks of cerebral hypoperfusion and worsening secondary brain injury. Optimal CPP (CPPopt) can be determined using autoregulatory indices, which are discussed in the next section.

Technical Aspects

The ICP response to changes in ABP depends on the pressure reactivity of cerebral vessels, and disturbed reactivity implies disturbed pressure autoregulation. Continuous monitoring and analysis of slow waves in ABP allow derivation of a pressure reactivity index (PRx) as a surrogate and continuous marker of global CA. Under normal circumstances, an increase in ABP leads to cerebral vasoconstriction within 5–15 seconds and a secondary reduction in cerebral blood volume (CBV) and ICP. When cerebrovascular reactivity is impaired, CBV and ICP increase passively with ABP with opposite effects occurring during reduced ABP. PRx is calculated as the moving correlation coefficient of consecutive time averaged data points of ICP and ABP recorded over a 4-minute period. A negative value for PRx, when ABP is inversely correlated with ICP, indicates a normal cerebrovascular reactivity, and a positive value a nonreactive cerebrovascular circulation. In the injured brain, cerebral vasoreactivity varies with CPP and optimizes within a narrow CPP range that is specific to an individual patient—referred to as CPP _opt . The continuous monitoring of PRx allows management of CPP levels according to an individual’s pathophysiological requirements, rather than at a generic predetermined target.

An oxygen reactivity index (ORx) has similarly been described as the moving correlation between PbtO ₂ and CPP. Cerebrovascular reactivity can also be assessed noninvasively using the correlation between ABP and TCD-derived mean and systolic blood flow velocity, and several near-infrared spectroscopy (NIRS)-derived variables.

Indications

Measurement of cerebrovascular reactivity using PRx and ORx has gained popularity in the management of TBI and, more recently, after SAH and ICH. NIRS-derived measures of cerebrovascular reactivity have been used to guide brain protection protocols during cardiac surgery.

Thresholds for Treatment and Evidence

Abnormal autoregulatory responses, as indicated by positive PRx and ORx values, are associated with poor outcome, and PRx-guided optimization of CPP is associated with improved outcome after TBI. Unlike PRx, which is a global measure of autoregulatory status, ORx represents regional autoregulation because of the focal nature of PbtO2. Thus, deranged ORx but normal PRx after ICH strongly suggests the presence of focal but not global autoregulatory failure. The duration and magnitude of blood pressure below the lower limit of CA, as determined by NIRS-derived autoregulatory indices, have been shown to be independently associated with major morbidity and mortality after cardiac surgery.

The incorporation of cerebrovascular reactivity into a multimodal monitoring strategy was the subject of a systematic review in 2014. Monitoring cerebrovascular reactivity was found not only to be important for the optimization of CPP, but also to inform interpretation of, and interventions targeted to, other monitored variables, specifically the assessment of the relationships between CBF, oxygen delivery and demand, and cellular metabolism.