63 Intracranial Hypertension and Monitoring ROBERT C TASKER AND ALIREZA AKHONDI-ASL PEARLS • As intracranial hypertension progresses, changes may occur in the vital signs, with an elevation of blood pressure, decrease or increase in pulse, and irregularity in the respiratory rhythm These signs, sometimes associated with episodes of decerebrate rigidity, indicate the occurrence of transtentorial herniation—or “coning”—that will lead to death if the process cannot be reversed • The continuous measurement of intracranial pressure is an essential modality in most brain monitoring systems After a In most organs in the human body, the environmental pressure for blood perfusion is either low or coupled to atmospheric pressure The environmental pressure for the brain differs in this respect because the brain is surrounded and protected by a stiff skull Thus, a rise in environmental pressure—intracranial pressure (ICP)—may impede cerebral blood flow (CBF) and cause is chemia In pediatric critical care, ICP may be of acute significance in a number of instances (e.g., traumatic brain injury [TBI], bacte rial meningitis, hydrocephalus, and the Fontan circulation) In this chapter, we discuss how information from ICP monitoring helps our understanding and treatment of brain disorders Clinical Background In critical illness, the early recognition and treatment of intracra nial hypertension is important because it is a major cause of mortality and morbidity Therefore, an attempt should be made to collate the clinical evidence for and against its presence How ever, the early symptoms and signs of this complication, which are invariably subtle and nonspecific (Table 63.1), make this form of assessment somewhat limited As will be discussed later, as intra cranial hypertension progresses, changes may occur in the vital signs, with an elevation of blood pressure, decrease or increase in pulse, and irregularity in respiratory rhythm These signs, some times associated with episodes of decerebrate rigidity, indicate the occurrence of transtentorial herniation (or “coning”) and imply the possibility of impending death if the process cannot be re versed Unfortunately, recognition at this stage is often too late to prevent death Brain tissue shifts may produce various syndromes (see Chapter 62) First, transtentorial or cerebellar herniation may result in midbrain or medullary compression Many of the clinical 768 decade of enthusiastic attempts to introduce newer modalities for brain monitoring (e.g., tissue oxygenation, microdialysis, cortical blood flow, transcranial Doppler ultrasonography, and jugular bulb oxygen saturation), the measurement of intracranial pressure remains a robust and only moderately invasive modality that can be realistically conducted in most pediatric critical care units signs observed in association with herniation result from direct compression of structures by the impacted tissue or are due to angulation of nerves or arteries against normal structures in the area These herniations can cause increasing coma, with distortion of the brainstem leading to midbrain and pontine hemorrhages Cerebellar herniation is likely to occur when the increase in ICP is maximal in the posterior fossa Such herniation occurs more commonly downward, squeezing one or more of the cerebellar tonsils through the foramen magnum; compressing the medulla; and leading to neck stiffness, head tilt, lower cranial nerve palsies, respiratory irregularities, or sudden cardiorespiratory arrest Cer ebellar herniation may occur upward through the tentorial notch, causing midbrain compression and leading to paralysis of upward gaze, dilated and fixed pupils, and respiratory abnormalities, al though this type of cerebellar herniation is uncommon When intracranial hypertension is more marked in the supra tentorial compartment—for instance, with acute intracerebral hematoma—the temporal lobe on the affected side may be dis placed into the tentorial notch and result in unilateral transtento rial herniation The herniation may be more marked anteriorly (uncal) or posteriorly (hippocampal) and is usually accompanied by displacement of the ipsilateral cingulate gyrus under the falx (cingulate herniation) Clinical manifestations of this condition may include ipsilateral third cranial nerve palsy, contralateral hemiparesis, respiratory irregularities, deepening coma with de cerebrate posturing, and, ultimately, cardiorespiratory arrest Finally, when bilateral or a general increase in ICP in the su pratentorial compartment occurs, as in diffuse cerebral edema, central transtentorial herniation may occur This condition leads to impairment of upward gaze, pupillary constriction, hyperto nus, and decerebrate posturing Temperature irregularities and diabetes insipidus may develop; eventually, cardiorespiratory CHAPTER 63 Intracranial Hypertension and Monitoring arrest may occur A summary of the clinical features of “central syndrome” is provided in Table 63.2 Physiology of the Intracranial Vault A physiologic process underlies the clinical picture of intracranial hypertension In this section of the chapter, our intention is to famil iarize the pediatric intensivist with new approaches to understanding the hydrodynamic function of the intracranial vault In other words, what happens before brain tissue shifts occur? In common with any aspect of physiology, the tasks have been straightforward: Can it be measured? Can it be modeled? How the models help in the un derstanding of the underlying homeostasis and the mechanism of derangement and perturbation? We highlight these topics and hope that the reader will see the obvious application to pediatric critical care—that is, use of this form of cerebral monitoring and potential approaches to ICP assessment In many specific conditions, how ever, knowledge is still lacking with regard to which parts of the adult-generated theory may be fully applicable to children TABLE Early, Subtle Symptoms and Signs of Raised 63.1 Intracranial Pressure Infant Child Poor feeding Vomiting Irritability to coma Seizures Anorexia and nausea Vomiting Lethargy to coma Seizures Head/eyes Full fontanelle Scalp vein distention False localizing signs False localizing signs Other Altered vital signs Hypertension Pulmonary edema Altered vital signs Hypertension Pulmonary edema General state 769 Intracranial Pressure The brain is an expansile structure that expands and contracts with each beat of the heart Because there are no valves within the venous drainage from the brain, any changes in intrathoracic pres sure are transmitted to ICP Such phenomena can be seen and palpated simply through the examination of an infant’s fontanelle and quantified with cerebrospinal fluid (CSF) pressure recording Once the cranial sutures have fused, any change in cerebral blood volume (CBV) on the arterial-to-arteriolar side of the cerebral circulation must be compensated by either reduction in cerebral venous volume or by phasic movement of CSF out of the intra cranial vault through the foramen magnum Such expansion of the cerebral mantle with compression of the lateral ventricles dur ing systole and movement of CSF through the aqueduct of Syl vius and to-and-fro through the foramen magnum was first visu alized with pneumoencephalography These changes now can be seen more easily with dynamic magnetic resonance imaging (MRI) Inevitably, a lag phase exists between the systolic increase in CBV and the effect of the compensatory mechanisms so that CSF pressure increases to reflect, in part, the systolic waveform The CSF pressure waveform is not an exact replica of the arterial waveform because it has been “filtered” by the combined effects of arterial wall compliance of the cerebral arteries, cerebrovascular resistance, and intracranial compliance (Fig 63.1).1 Utility of Hydrodynamic and Electrical Analog Models of Intracranial Pressure ICP is a function of the circulation of CBF and CSF in which ICP is related to a vascular component (ICPvascular) and a CSF component (ICPCSF) There is considerable interest in modeling these relationships as an aid to understanding some of the com plex phenomena seen in critically ill patients The vascular com ponent is difficult to express quantitatively However, multiple variables—such as arterial blood pressure (ABP), autoregulation, TABLE Clinical Features of Central Syndrome or Rostrocaudal Deterioration 63.2 Stage Level of Consciousness Respiration Pupil Size and Reactivity Oculocephalic and Oculovestibular Responses Posture and Tone Diencephalic (early to late) Agitation Drowsiness Stupor Deep sighs or yawns Occasional pauses Cheyne-Stokes or periodic breathing Small (1–3 mm) with brisk reaction to light Conjugate at rest and responds quickly Normal or slightly increased Generalized muscular hypertonus Midbrain to upper pontine Coma Central hyperventilation Midposition (3–5 mm) with sluggish reaction to light Dysconjugate Decorticate posturing and increased tone Lower pontine to upper medullary Deep coma Midposition and fixed Absent Flaccid: (1) retained bilateral extensor plantars and (2) occasional flexor responses in the lower limbs Medullary (terminal) Deep coma May be unequal Absent Flaccid Irregular breathing interrupted by deep sighs, gasps, and then terminal apnea Modified from Plum F, Posner JB Diagnosis of Stupor and Coma Philadelphia: FA Davis; 1966 770 S E C T I O N V I Pediatric Critical Care: Neurologic and cerebral venous outflow—all contribute to it Regarding the other component, circulation of CSF, up to 80% of CSF is the product of active secretion by the choroid plexus, and movement of interstitial fluid into the ventricles and subarachnoid space contributes the remainder in the uninjured brain Drainage is largely passive via arachnoid villi and granulations into the supe rior sagittal sinus and spinal root sleeve venous drainage Some drainage, which is currently unquantifiable, occurs through the olfactory bulb and mucosa into the deep cervical lymphatics The equation by Davson and colleagues2 shows the immediate relationships controlling CSF pressure: ICP (mm Hg) 70.0 52.5 35.0 17.5 ABP (mm Hg) 150 120 90 60 ICPCSF (Resistance to CSF outflow) (CSF formation) (Pressure in sagittal sinus) 30 0 10 10 ABP (mm Hg) ICP (mm Hg) Time (sec) 37.5 25.0 12.5 150 120 90 60 30 0 Time (sec) • Fig 63.1 Examples of waveforms of intracranial pressure (ICP) and arterial blood pressure (ABP) recorded at a high level of ICP (upper plots) and a lower level of ICP (lower plots) Note that the pattern of ABP waveform is relatively invariant, but ICP changes its shape considerably With these two components—vascular and CSF—taken together, Fig 63.2 shows the electrical analog circuit equivalent of CBF and CSF circulation.3 Such models can be used to interrogate ICP and CSF dynamics clinically These models can also be used to estimate ICP noninvasively For example, Kashif et al.3 described a model-based noninvasive estimation of ICP from middle cere bral artery CBF velocity using transcranial Doppler (TCD) and arterial pressure from 35 hours of data in 37 adult patients with severe TBI The authors generated ICP estimates and achieved a mean error (bias) of 1.6 mm Hg More recently, Fanelli et al.4 evaluated a similar model in 12 pediatric and young adult patients (aged 2–25 years) and described a method for real-time estima tion of noninvasive ICP with values falling within 67 mm Hg of the reference ICP measurement (Fig 63.3) Overall performance gave a mean error (bias) of 1.0 mm Hg, standard deviation of the error (precision) of 5.1 mm Hg, and root-mean-square error of 5.2 mm Hg The associated mean and median absolute errors were 4.2 mm Hg and 3.3 mm Hg, respectively The assessment of ICP is needed to direct medical interven tions in infants, children, and adolescents with severe TBI.5,6 However, there are instances in which physicians not carry out invasive ICP monitoring but being able to follow it in real-time Skull Cerebrospinal fluid Skull pa(t) Cerebrospinal fluid R q(t) q1(t) Cerebral blood flow Brain tissue Arteries C Veins Brain A Cerebral arterial network B Arterial blood Collapsed pressure Intracranial pressure venous segment ICP C • Fig 63.2 Progressive abstraction of a hydrodynamic to electrical analog of cerebrovascular physiology (A) Relevant cerebrovascular anatomy: brain tissue (BT), cerebrospinal fluid (CSF), and cerebral arterial network (AN) (B) Schematic hydrodynamic representation of the main cerebrovascular compartments and associated physiologic variables: cerebral blood flow (CBF), arterial blood pressure (ABP), and intracranial pressure (ICP); the collapsed venous segment is also shown (C) Lumped circuit-model representation of cerebrovascular physiology: CBF q(t), cerebral arteriovenous flow q1(t), and ABP pa(t) ICP denotes both extraluminal pressure and the effective downstream pressure for cerebral perfusion C, Compliance; pa(t), arterial blood pressure; q1(t), cerebral arteriovenous flow; q(t), cerebral blood flow; R, resistance ICP + – CHAPTER 63 Intracranial Hypertension and Monitoring ABP(t) 771 R CBF(t) q1(t) C ICP + – ICP A ICP (mm Hg) ICP 20 15 10 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Time (s) CBFV (cm/s) CBFV 150 100 50 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Time (s) ABP (mm Hg) ABP 140 120 100 80 60 40 20 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Time (s) B • Fig 63.3 (A) Electrical analog model of cerebral blood flow (CBF), arterial blood pressure (ABP), and intracranial pressure (ICP) similar to Fig 63.2 (B) Simultaneous data collection of ICP, cerebral blood flow velocity (CBFV) from transcranial Doppler of the middle cerebral artery and ABP used in the validation of noninvasive ICP estimation C, compliance; q1(t), cerebral arteriovenous flow; t, time; R, resistance would be useful, for example, in moderate-severity TBI, certain cases of craniosynostosis, metabolic disorders, central nervous system (CNS) infection, fulminant liver failure, and certain other forms of acute encephalopathy.5,6 It has been suggested that non invasive ultrasound measurement of the optic nerve sheath diam eter (ONSD) may be used to identify and monitor elevated ICP Unfortunately, such measurements are neither suitable for continuous real-time monitoring nor are they able to provide absolute ICP.7 However, one approach used in adults in the emer gency department is to measure ONSD in order to give some indication of whether ICP is elevated or not.7 Patient-specific ICP determination would be a key advancement for the care of chil dren, particularly in the range of values when symptoms are not obvious: 15 to 25 mm Hg Of note, the technique described by Fanelli et al.4 performs robustly in the ICP range 15 to 25 mm Hg, which—if supported by advancements in noninvasive ABP technology and miniaturization of TCD technology8—could be transformative to practice Just consider surveillance and inter ventions that would be possible beyond the care of pediatric pa tients with severe TBI to a number of coma-inducing conditions complicated by intracranial hypertension worldwide, such as CNS infection 772 S E C T I O N V I Pediatric Critical Care: Neurologic Cerebral Vasodilation and Cerebral Spinal Fluid Pressure Three major factors regulate CBF: cerebral perfusion pressure (CPP), partial pressure of arterial carbon dioxide (Paco2), and partial pressure of arterial oxygen (Pao2) Hypercapnia causes cerebral vasodilation, increases CBF, and increases ICP Hypoxia also causes cerebral vasodilation and a rise in ICP CPP is taken as the difference between mean ABP and ICP and represents the pressure gradient acting across the cerebrovascular bed Therefore, it is an important factor in regulation of CBF.9 Under normal circumstances, over a wide range of CPP, CBF is autoregulated (i.e., it remains constant when CPP varies) Thus, regarding the effect of this phenomenon on ICP, active cerebral arteriolar con striction occurs and ICP consequently falls to maintain CBF when ABP is increased (i.e., hypertension) At the other extreme, systemic hypotension (within the autoregulatory range) provokes cerebral vasodilation and an increase in ICP When autoregulation is defective, ICP increases and decreases with AB—that is, it is pressure passive In practice, measurement of ICP can be used to estimate the CPP when it is the most significant downstream pressure acting on vascular perfusion However, in some instances, venous pressure may be of more significance, such as critical Fontan circulation Cerebral Perfusion Pressure and Cerebral Autoregulation In adults, the lower limit for CPP is taken as a threshold of 60 to 70 mm Hg In children, however, it is evident from measurement of normal ABP that the lower limit for CPP must be lower than this adult level for much of childhood (Fig 63.4).10 To date, two general approaches are used to study indirectly whole brain circulation pathophysiology.11 First, there is an empiric approach using observational and epidemiologic evi dence to establish statistical relationships and correlations be tween inputs and outputs This approach is best exemplified by studies in severe TBI using PRx (i.e., pressure reactivity index in 150 95% CI for mean BP CPP = BP – ICP ICP “normal” Mean BP (mm Hg) 125 100 75 50 25 0 12 15 18 21 Age (years) • Fig 63.4 Estimated lower limit of cerebral perfusion pressure (CPP) by age in the pediatric range on the basis of normal blood pressure (BP) and intracranial pressure (ICP) data CI, confidence interval which 5–7 minutes of time-averaged ICP and ABP data are used to calculate mean Fisher-transformed Pearson correlation) to derive, by statistical analyses, the CPP optimal for an indi vidual (i.e., CPPOpt) Briefly, the assumption in this approach is that when vascular reactivity is absent, ABP and ICP are posi tively correlated By comparison, strong cerebrovascular vaso motor constriction in response to increased ABP results in de creased CBV, which leads to decreased ICP and negative PRx (Fig 63.5).12–15 The CPPOpt is defined as the CPP at which PRx is at its minimum (Fig 63.6) In severe TBI, the difference between contemporaneous time-series median CPP (CPPMed) and CPPOpt (i.e., DCPP) to mm Hg is associated with favor able outcome, whereas DCPP –10 to –15 mm Hg occurs only in those with unfavorable outcome.13 Such empiric associations and observations not provide better understanding of patho physiology or even the determinants of CPPOpt.16 PRx-derived CPPOpt, with calculation of DCPP and DCategory, are basically statistical constructs that have proven value in patient riskstratification and in epidemiologic association with poor out come in adult cases of severe TBI.12–15,17–19 The main reason here is that CPP is not a physiologically controlled parameter in homeostasis; there is no sensor, no error signal, no set point or optimum within the operating range between the upper and lower limits of cerebral autoregulation Hence, CPPOpt-guided management may be a testable pragmatic therapeutic approach for future clinical trials20 but, intrinsically, empiricism has limitations to understanding mechanistic functions In the second approach to studying whole brain circulation pathophysiology, investigators have a mechanistic approach to indirect assessment of global cerebrovascular behavior using electrical analog models of autoregulatory control pathways, feedback loops, and arterial network structure.11 Here, com puter modeling evidence is rationalized according to the fluid dynamics or electronics of inputs and outputs In 1988, Ursino21,22 proposed the first lumped compartmental model of cerebral circulation physiology that included autoregulation Since then, the modelling-mechanistic approach has been used to better understand patient data from, for example, TCD studies of middle cerebral artery flow velocity.11 Recently, one such model,16 shown in Fig 63.7, has been used to provide mechanistic insight into the questions “What does PRx-derived CPPOpt mean?” and “What is the effect of manipulating ABP, ICP, and Paco2?” For example, Akhondi-Asl et al.16 found that computer modeling targeting of CPPOpt, a shift in ABPMean or mean ICP, with intact cerebral autoregulation, impacts CPPOpt (Fig 63.8) Second, a positive PRx at CPPOpt, which should signify impaired autoregulation, in the lumped model also oc curs with intact autoregulation Last, in relation to DCPP, the compartmental model confirms that greater deviation from zero is associated with more severe 6-hour profiles in ICP Thus, in answer to the questions posed about meaning and effect of manipulations, the simulations provide mechanistic insight The empirically derived variables (e.g., PRx, CPPOpt, and DCategory) are able to provide a numerical summary of complex states in volving well-described clinical progression and deterioration with raised ICP.23 This information has epidemiologic impor tance but is not proven useful in directing management in pedi atric patients—it may alert clinicians that ICP is high or low and that ABP is high or low, but this is already known from the direct measurements! That said, model-based mechanistic approaches in critical care medicine should be considered as complementary and potentially synergistic to other bedside approaches.24 ... Intracranial Vault A physiologic process underlies the clinical picture of intracranial hypertension In this section of the chapter, our intention is to famil iarize the pediatric intensivist with new... hope that the reader will see the obvious application to pediatric critical care—that is, use of this form of cerebral monitoring and potential approaches to ICP assessment In many specific conditions,... structure that expands and contracts with each beat of the heart Because there are no valves within the venous drainage from the brain, any changes in intrathoracic pres sure are transmitted