Critical Care Management of Neurosurgical Patients

Flow-Metabolism Coupling

CBF is regionally variable, and the distribution parallels and reflects the degree of metabolic activity in different parts of the brain (flow-metabolism coupling) . The average CBF is 50 mL/100 g of tissue/min with areas of increased neuronal activity having a higher cerebral metabolic rate and therefore increased regional CBF (average in gray matter, 80–110 mL/100 g of tissue/min; average in white matter, 25 mL/100 g of tissue/min).

Metabolic activity is reflected by the cerebral metabolic rate of oxygen consumption (CMR O ₂ ) and the cerebral metabolic rate of glucose utilization (CMR _Glucose ). The close matching of flow metabolism means that although there is regional variation in CBF, CMR O ₂ , and CMR _Glucose , the overall oxygen extraction fraction (OEF) remains relatively constant (~40%). The regulatory changes in flow-metabolism coupling have a short latency (~1 sec) and are mediated by regional metabolic and neurogenic factors.

Metabolic factors exert their effect via a negative feedback mechanism. The initially increased metabolic activity increases the level of local metabolic factors precipitating an increase in perfusion and CBF. This leads to an increased washout of these factors with an associated reduction in flow. Examples of vasodilatory factors (which increase flow) include nitric oxide (NO), specific prostaglandins (PgE ₂ and PgI ₂ ), and adenosine. The local levels of these factors may also increase with increased activity during “stress” states (eg, hypoxia, hypotension, seizures). Vasoconstrictive factors (which decrease flow) include free calcium ion, thromboxane (TxA ₂ ), and endothelin.

Neurogenic mechanisms work via a combination of sympathetic activity (α-2 adrenergic activity is vasoconstrictive, and β-1 adrenergic activity is vasodilatory) and local neural factors (eg, acetylcholine [ACh], NO, serotonin [5-HT], dopamine [DA], substance P, neuropeptide Y).

Depression of cerebral function (eg, anesthesia, hypothermia, sedation) all suppress metabolism (CMR O ₂ , CMR _Glucose ) with a coupled reduction in CBF, whereas hyperthermia and seizure activity increase metabolism and CBF. It is estimated that for every 1°C change in temperature, the CMR O ₂ and CBF can change by as much as ~5%.

Autoregulation

Autoregulation describes the ability of the cerebral vasculature to maintain a relatively constant CBF despite fluctuations in CPP by regulation of the CVR of intracranial vessels. It works independently of other mechanisms of CBF control such as flow-metabolism coupling and PaCO ₂ ( Fig. 26.2 ).

Autoregulation works within CPP limits (usually between 50 and 150 mm Hg), although there is significant regional variation in the brain and among individuals. Outside of these limits, however, the CBF is directly dependent on CPP: above 150 mm Hg, an increase in CPP results in increased CBF with forced vasodilatation of cerebral arterioles, increased cerebral blood volume (CBV), ICP, and a resultant disruption of the blood-brain barrier and risk of cerebral edema or hemorrhage; conversely, below 50 mm Hg, a decrease in CPP results in a significant drop in CBF, leading to ischemia.

Autoregulation is mediated by myogenic reflexes within the resistance intracranial vessels (predominantly tone and contractility within smooth muscle in arteriole walls) and neurogenic factors (metabolic factors, neural factors, sympathetic activity as described earlier), and these factors are usually slower compared to those of flow-metabolism coupling. Other relevant clinical factors impacting autoregulation include (1) inhalational anesthetics (autoregulation impaired in a dose-dependent fashion), (2) preexisting hypertension (the autoregulation curve is shifted to the right with a narrower plateau and therefore blood pressure required clinically to maintain constant CBF may be higher than expected; targeting lower blood pressures expected with “normal” patients may place the patient in the linear part of the curve with a resultant fall in CBF and ischemic risk), and (3) disruption of the blood-brain barrier in intracranial pathologic processes (eg, trauma, tumors, stroke) where autoregulation is lost ( Fig. 26.3 ).

Impact of PaCO ₂ and PaO ₂

Arterial carbon dioxide tension is a clinically important regulator of CBF; it crosses the blood-brain barrier readily and changes extravascular pH. Increasing PaCO ₂ (hypercapnia) precipitates vasodilatation with a reduction in CVR and an increase in CBF (up to an upper limit where vasodilatory effects are maximal). Conversely, decreasing PaCO ₂ (hypocapnia) causes vasoconstriction with increased CVR and decreased CBF and CBV. It is estimated that for 1 mm Hg change in PaCO ₂ (1 kPa is equivalent to 7.5 mm Hg), the CBF may change by as much as 4% (2 mL/100 g of tissue/min). Clinically in patients with decreased intracranial compliance and therefore raised ICP, hyperventilation-mediated hypocapnia can reduce the CBF, CBV, and therefore ICP.

The manipulation of PaCO ₂ to control ICP should, however, only be used as a short-term measure until a definitive method of ICP control is instituted. The reasons are twofold: (1) persistent hypocapnia causes vasoconstriction to a maximal limit beyond which tissue hypoxia and cerebral ischemia ensue together with an associated reflex vasodilation, and (2) the effect of decreased PaCO ₂ is mediated by an increase in the perivascular pH, which is reversed by a corresponding decrease in extracellular fluid bicarbonate to normalize the pH (usually within ~6 hours).

Arterial oxygen tension impacts on CBF but not as significantly as carbon dioxide; CBF is usually unchanged with decreasing PaO ₂ until it drops below a threshold (~8 kPa) after which a vasodilatory response is seen. Hyperoxemia may produce some cerebrovascular vasoconstriction, which is postulated to be an evolutionary neuroprotective measure against oxygen free radicals ( Fig. 26.4 ).

Impact of Anesthetic Pharmacologic Agents

Benzodiazepines and the majority of intravenous (IV) anesthetic agents (eg, etomidate, propofol, thiopental) reduce CMR O ₂ and CBF and therefore ICP. This reduction is often in a dose-dependent fashion. Autoregulation, flow-metabolism coupling, and responsiveness to CO ₂ remain intact. However, in critically ill neurosurgical patients (eg, traumatic brain injury), flow-metabolism coupling may be impaired and therefore the reduction in CBF (flow) may exceed that of CMR O ₂ (metabolism), leading to a mismatch and therefore widened cerebral arteriovenous oxygen gradient and impaired perfusion and ischemia. This must be borne in mind as agents such as propofol—widely used due to a short t _1/2 allowing rapid emergence if required—may induce precipitous hypotension and a decrease in CPP, especially in hypovolemic patients. Inhalational anesthetics reduce CMR O ₂ but increase CBF via a vasodilation effect. Autoregulation is impaired in a dose-dependent fashion. Ketamine increases CMR O ₂ , CBF, and ICP and therefore is often avoided in critically ill neurosurgical patients.

Pathophysiology

The pathophysiologic processes underpinning ischemia and subsequent infarction are complex and interlinked but broadly speaking involve the following mechanisms: excitotoxicity, tissue acidosis, free radical production, inflammatory cascade activation, cerebral edema, apoptosis, and cell death.

The initial impairment in cerebral perfusion, ischemia, and the resultant lack of adenosine triphosphate (ATP) production results in a failure to maintain normal ionic gradients. Uncontrolled depolarization occurs, followed by the release of excitatory neurotransmitters (glutamate) and calcium. The excitotoxicity exacerbates the depolarization and drives other processes involved in cell necrosis and death. The absence of oxygen leads to tissue acidosis as a result of anaerobic metabolism and lactate production. The mechanism by which hyperglycemia results in poorer outcomes following neurologic injury is postulated to be because it provides a further substrate for continued anaerobic metabolism.

The excitotoxicity drives the activation of neuronal and inducible nitric oxide synthase (nNOS and iNOS, respectively) with the resultant NO production generating free radicals. Free radical–mediated DNA damage and damage to cell membranes result. Damage to cell membranes can further impact worsening ionic gradients by impairing protein channels involved in ionic transport. DNA damage and ischemia result in the activation of an inflammatory cascade via the up-regulation and nuclear transcription of pro-inflammatory cytokines (eg, adhesion molecules). Inflammatory cell recruitment and activation increase (eg, microglia), all of which continues the free radical–mediated injury.

The disruption of ionic transport mechanisms as a result of cell membrane injury precipitates uptake of fluid from extracellular spaces resulting in neuronal swelling even in the presence of an intact blood-brain barrier (cytotoxic edema). The subsequent breakdown of the blood-brain barrier then also leads to vasogenic edema, as unregulated transport of plasma proteins into cells occurs, increasing the osmotic drive for fluid influx. Cerebral edema increases ICP, worsens perfusion, and exacerbates ischemia.

The combination of excitotoxicity, tissue acidosis, free radical production, inflammatory activation, and cellular edema independently and in combination produce mitochondrial activation together with the release of cytochrome C from mitochondrial membranes (this also occurs due to mitochondrial membrane injury). The effect is activation of caspases and other proteolytic enzymes, which leads to apoptosis and cell death.

Many of the pathophysiologic processes described earlier that contribute to ischemia-related injury although damaging in the acute phase are also critical in the long term as the eventual healing and repair process begins with these mechanisms. The restoration of autoregulation including CO ₂ reactivity and eventual healing with the formation of (usually nonfunctional) scar tissue may take from 4 to 6 weeks.

Global and Focal Ischemia

Global ischemia occurs with a global reduction in CBF (eg, hypoxic-ischemia injury secondary to cardiac arrest). Focal ischemia results when there is a focal reduction in blood flow in a major vessel: the infarct core is the region usually supplied by the smallest arterioles and end arteries that rapidly undergoes irreversible cell death; the surrounding penumbra is a zone where there is evidence of neuronal dysfunction and electrical silence, but it has not yet undergone depolarization, excitotoxicity, and irreversible cell death. This penumbra can be salvaged with restoration of CBF. Failure to restore perfusion will progressively incorporate the penumbra into the core infarct and increase its size. The vascular flow thresholds at which stepwise deterioration occurs vary depending on the basal metabolic activity of the region (eg, gray matter will intrinsically have higher metabolic rates and higher CBF requirements in comparison to white matter), but on average CBF < 8 to 10 mL/100 g of tissue/min results in irreversible neuronal death ( Table 26.2 ).

The size of the penumbra and core depends on the strength of the collateral supply to the region, and principles underpinning the use of perfusion imaging are intended to help characterize the relative proportions of the core infarct versus the penumbra, which helps to predict the impact of salvage and reperfusion therapy and to guide prognosis.

Governance of Oxygen Delivery and Physiologic Concepts

Total O ₂ content within the blood is based on (1) the amount of O ₂ dissolved in plasma (as per Henry's law, dependent on the solubility of O ₂ and PaO ₂ ) and (2) the O ₂ –carrying capacity of hemoglobin (dependent on SaO ₂ and the concentration of hemoglobin). Assuming a temperature of 37°C and a hemoglobin concentration ([Hb]) of 150 g/L, in arterial blood (with SaO ₂ 97% and PaO ₂ 13.3 kPa), there are approximately 200 mL of O ₂ /L; in mixed venous blood (SaO ₂ 75% and PaO ₂ 5.6 kPa) there are 150 mL of O ₂ /L. The degree of oxygen flux and usage is approximately 50 mL/L. With a cardiac output of 5 L/min, the oxygen flux and usage are 250 mL/min. Hence it is imperative to understand that oxygen delivery depends on temperature (which determines the solubility of O ₂ ), cardiac output, PaO ₂ and SaO ₂ , and hemoglobin concentration. The meticulous optimization of these parameters is vital to ensure adequate cerebral oxygenation and perfusion and therefore limit secondary injury.

The aim should be to provide adequate oxygenation and avoid hyper- or hypocapnia. Targets vary depending on the patient comorbid status, the presence of preexisting respiratory comorbidities, and clinical status, but where feasible, the aims should be oxygen saturation (SaO ₂ ) ≥ 97%, PaO ₂ ≥ 11 to 13 kPa, and PaCO ₂ 4.7 to 5.3 kPa (normocapnia). Mechanisms to increase oxygenation include increasing the inspired oxygen fraction (FiO ₂ ), increasing the inspiratory time to expiratory time cycling (IT : ET ratio from 1 : 2 to 2 : 2, respectively), and applying positive end-expiratory pressure (PEEP ≤ 10 mm Hg). Peak inspiratory pressures should not exceed 35 mm Hg to ensure minimal interference with cerebral venous drainage.

Prophylactic hypocapnia via hyperventilation (minute ventilation [MV] = tidal volume [V _T ] × ventilatory rate [V _R ]) should be avoided, as it causes cerebral vascular vasoconstriction with the risk of worsening cerebral ischemia. For the purposes of ICP control, it is a short-term measure only until more definitive measures to control ICP are instituted. Patients that are intubated and on mechanical ventilation should have adequate sedation (intravenous hypnotic; eg, propofol, midazolam), analgesia (eg, opioid), and, if needed, a neuromuscular blocking agent to avoid ventilator-associated stress, gag, and the patient fighting against the ventilator—all of which can precipitate systemic hypertension, distress, and, via the transmission of increased intrathoracic pressures, an increase in ICP.

Attempts to wean the patient off the ventilator as soon as clinically feasible and aim for extubation should be considered. There are multiple benefits to this strategy, including the ability to perform a clinical neurologic examination and reduce the risk of ventilator-associated complications including ventilator-associated pneumonia (VAP). There are significant differences in the majority of critically ill neurosurgical patients in comparison to patients in general critical care units : most will have normal baseline pulmonary function tests (except elderly patients with preexisting chronic obstructive pulmonary disease); most will require only synchronized intermittent mandatory ventilation (SIMV) or pressure support ventilation (PSV); and other strategies such as prone ventilation, permissive hypercapnia, and inverse ratio ventilation are usually avoided and rare. As such, ventilator dependence is less common with rapid weaning expected.

Exceptions may be patients with significant polytrauma or high cervical spinal cord injury. In situations where a prolonged period of mechanical ventilation is anticipated, consideration should be made for early tracheostomy (performed as an open surgical procedure or percutaneous). Benefits include increasing patient comfort and ability to tolerate tracheal tubes without significant gag, thereby allowing neurologic examination without sedation to occur more easily, improved length of NCCU stay, more effective airway clearance and bronchial toilet of secretions, and a decreased risk of tracheolaryngeal stenosis secondary to prolonged intubation.

Fluid and Electrolyte Management

In most neurosurgical pathologies, the aim of fluid resuscitation and management should be to achieve normovolemia, normal electrolytes, normal serum osmolalities, and normoglycemia. Daily fluid requirements are approximately 40 mL/kg/24 hr with hourly requirements calculated as (1) 4 mL/kg for every kilogram from 0 to 10 kg of body weight, (2) 2 mL/kg for every kilogram from 11 to 20 kg of body weight, and (3) 1 mL/kg for every kilogram of body weight from 20 kg onward. A urine output of 1 mL/kg/hour should be the aim to avoid oliguria. Fluid management should also take into consideration variable factors such as the impact of temperature, fever, and respiratory dysfunction on increasing insensible fluid losses; the impact of anesthesia and sedation-induced cardiac depression and vasodilation; preoperative or perioperative deficits; and the redistribution of third space fluid losses. The daily sodium requirement is 1 to 2 mmol/kg (or mEq/kg), and the daily potassium requirement is 0.5 to 1 mmol/kg (or mEq/kg).