Medical Diseases and Metabolic Encephalopathies

" "

Neurological emergencies in medical diseases (secondary brain
injury), e.g., renal coma, hepatic coma, salt and water imbalance,
disturbance of glucose metabolism, other endocrinal causes of
coma, disturbances of calcium and magnesium metabolism,
drugs and intoxication, represent a good part of the patient
population in the neurocritical care unit. Understanding the
underlying mechanisms of secondary brain injury which include
hypoxia, hypoperfusion, reperfusion injury with free radical
formations, release of excitatory amino acids and harmful
mediators from injured cells, and electrolyte and acid base
changes from systemic or regional ischemia, are very important
for proper management of such conditions. Management rules
will be specified according to each cause and pathogenesis.
Metabolic encephalopathies are a group of neurological deficits
affecting the brain causing delirium, confusion, or coma, caused
by different mechanisms involving toxin production or
interference with metabolic biochemical processes. Metabolic
encephalopathies are usually multifactorial in origin, and are
important complications of many diseases of patients treated in
critical care units. Confusion is clinically defined as the inability

to maintain a coherent stream of thought or action. Delirium is a
confusional state with superimposed hyperactivity of the
sympathetic limb of the autonomic nervous system with
consequent signs including tremor, tachycardia, diaphoresis,
and mydriasis. Acute toxic-metabolic encephalopathy (TME),
which encompasses delirium and the acute confusional state, is
an acute condition of global cerebral dysfunction in the absence
of primary structural brain disease (Chen 1996).
Examination of the Encephalopathic Patient
1. Mental state assessment
a. Level of consciousness using Glasgow Coma Scale (GCS),
b. Memory and attention by Mini-Mental State
Examination (MMSE),
c. Mood (depression, elation, mania or irritability),
d. Hallucination, which is usually visual.
2. Physical examination
a. Temperature,
b. Pulse (for tachycardia),
c. Pupillary dysfunctions and extraocular movements,
N.B. pupillary functions are ones that last and resist
changes in metabolic encephalopathies.
d. Respiratory pattern,
e. Motor response,
N.B. during examination of motor response look for
presence of asterixis, which are drops of fully extended
wrists for less than one second with re-extension again.
3. Investigations
a. CSF examination,
b. Full metabolic scanning of the blood,
c. Neurophysiologic tests, e.g., EEG, evoked potentials,
i. EEG patterns in metabolic encephalopathies are not
specific (e.g., triphasic waves in hepatic
encephalopathy). On the other hand, multifocal

spikes are specific to lithium intoxication (Kaplan
2011).
ii. Brain stem auditory evoked potentials (BAER) are
resistant to metabolic changes.
d. Neuroimaging, e.g., CT scan, MRI.
General Pathophysiology
Normal neuronal activity requires a balanced environment of
electrolytes, water, amino acids, excitatory and inhibitory
neurotransmitters, and metabolic substrates (Earnest 1993). In
addition, normal blood flow, normal temperature, normal
osmolality, and physiologic pH are required for optimal central
nervous system function. Complex systems, including those
mediating arousal and awareness and those involved in higher
cognitive functions, are more likely to malfunction when the
local body environment is deranged (Young 1998).
All forms of acute metabolic encephalopathy (ME) interfere
with the function of the ascending reticular activating system
and its projections to the cerebral cortex, leading to impairment
of arousal and awareness. Ultimately, the neurophysiologic
mechanisms of ME include interruption of polysynaptic
pathways and altered excitatory-inhibitory amino acid balance
(Lipton 1994). The pathophysiology of ME varies according to the
underlying etiology.
Hepatic Encephalopathy
Hepatic encephalopathy (HE) appears as a complication of
fulminant hepatic failure (FHF) and in chronic liver failure.
Initially it is characterized by minor mental and personality
changes with some cognitive impairment. With disease
progression there are obvious motor abnormalities and
increasing loss of consciousness until deep coma.
Etiology of fulminant hepatic failure includes viruses, drugs
(such as halothane, acetaminophen, valproate and INH), and

acute fatty liver in pregnancy, toxins (such as amatoxins and
phosphorus), Wilson disease and Rye’s syndrome.
Pathophysiology of hepatic encephalopathy includes
neurotoxins (like ammonia, short- and medium-chain fatty
acids, mercaptans, phenols, etc), and altered neurotransmission
(due to benzodiazepine-like substances, and neurotransmitters’
hypothesis including 5-HT and glutaminergic transmission). This
may end in cerebral edema (Young 1998).
All of the following conditions may precipitate hepatic
encephalopathy:
– Increased GIT protein absorption like in GIT hemorrhage or
increased dietary protein
– Drugs like benzodiazepines or INH
– Renal dysfunction
– Catabolic states like infections and surgery
– Dehydration, hypokalemia, constipation
– Chronic hepatopathies which can present with FHF, e.g.,
cirrhosis with portal hypertension, portal vein thrombosis,
Wilson disease, ornithine carbamoyltransferase deficiency
and chronic valproate hepatopathy
Complications of hepatic encephalopathy (HE)
Epilepsy may complicate hepatic encephalopathy in 10–30% of
cases, related to hypoglycemia. Cerebral edema may complicate
hepatic encephalopathy in 80% of cases. Bleeding and renal
dysfunction (i.e., renal failure in presence of normal-sized
kidney (hepatorenal syndrome), with urine sodium
concentration below 20 mmol/L, while the urine is
hyperosmolar) may complicate hepatic encephalopathy also.
This syndrome is reversible with normalization of renal
functions. Another form of renal dysfunction is pre-renal
azotemia. Hypotension and derangement of acid base balance
with acidosis is reported with cerebral edema and alkalosis, and
is found with vomiting and hypokalaemia in cases of hepatic
encephalopathy.

Table 9.1 – Differences between FHF and chronic hepatic
encephalopathy
Feature FHF Chronic HE
History
Onset
Mental state
Precipitation
History of liver disease
Acute
Mania → coma
Viral infection or
hepatoxin
No
Insidious or subacute
Blunted → coma
GIT hemorrhage,
exogenous protein,
uremia
Yes
Symptoms
Nausea & vomiting
Signs
Liver
Nutritional state
Ascitis
Common
Small, soft, tender
Normal
Absent
Unusual
Large, firm, painless
Cachectic
May be present
Lab tests
Transaminases
Coagulopathy
Very high
Present
Normal or slightly high
Often present
Table 9.2 – Clinical staging of hepatic encephalopathy
Grade Consciousness Intellect Behavior Motor Psychometric
0-1 Normal Normal Normal Normal Poor performance
1 Inverted sleep
Insomnia
Hypersomnia
Short
attention
Low
perception
Impaired
calculation
Anxiety
Apathy
Irritability
Low
monotonous
voice
Incoordination
Poor hand
writing
Tremors
Prolonged
2 Slow response
Lethargy
Disorientation
for time
Disinhibition
Disobedience
Asterixis
Ataxia
Dysarthria
Hyperreflexia
Very prolonged
3 Confusion
Delirium
Paranoia
Semistupor
Disorientation
for place
Amnesia
Perseverate
Bizarre
behavior
Hyperreflexia
Nystagmus
Rigidity
Yawing
Incontinence
Unable to
perform
4 Coma, arousal to
pain
Coma
unresponsive
None absent Decorticate /
Decerebrate
posturing

Treatment of hepatic encephalopathy
Fulminant Hepatic Failure (FHF). First of all, patients must be
nourished by intravenous infusion of glucose 20–40%, then be
given retention enema with 250 ml lactulose in 750 ml
electrolyte solution, neomycin 2 g 2-3x daily, along with fresh
frozen plasma (FFP) for coagulopathy. Plasma exchange can be
used daily, until the prothrombin time is near normal (Quick
score prolonged below 25%) or patient is awake. Hypotension
must be treated with volume expanders. GIT bleeding is the
second cause of death in FHF, administration of H2 antagonists or
proton pump inhibitors may help. Cerebral edema is treated by
20% manitol, 100 ml every 8 hours in absence of renal failure, but
steroids are contraindicated. In cases of metabolic
encephalopathy due to organ failure, transplantation may be an
option. In In case of acetaminophen-induced fulminant hepatic
failure, gastric lavage, forced diarrhea and N-acetyl cysteine
infusion of 150 mg/kg diluted with glucose solution may help.
Hepatic encephalopathy in chronic liver disease. Most
porto-systemic encephalopathies (PSE) occur as a consequence
of dietary mistakes, GIT bleeding, hemorrhage, infection,
alkalosis or hypokalemia as a result of diuretic therapy.
Hepatic encephalopathy in this case can be managed by a diet
restricted in proteins, allow 30-40 g/day. Lactulose can be given
at about 60-180 ml, in divided doses, to give 2-3 soft stools daily.
In addition, neomycin (2 g twice daily) and flumazenil (a
benzodiazepine receptor antagonist) should be given in 2 mg
dose infused in 15 minutes (Onyekwere 2011).
Renal Encephalopathies
The most disabling feature of both renal failure and dialysis is
encephalopathy. It is probably caused by the accumulation of
renal toxins. Other important causes are related to the
underlying disorders that cause renal failure, particularly
hypertension. The clinical manifestations of renal (uremic)
encephalopathy spans from mild confusional states to deep

coma, and movement disorders such as asterixis may be
associated. Cognitive impairment is considered to be the major
indication for the initiation of renal dialysis with or without
subsequent transplantation. Sleep disorders including restless
legs syndrome are also common in patients with kidney failure.
Renal dialysis is also associated with neurologic complications
including acute dialysis encephalopathy and chronic dialysis
encephalopathy, formerly known as dialysis dementia (Seifter
2011).
Five major syndromes of CNS affection are seen in renal failure.
– Acute renal failure: asterixis, myoclonus, seizures,
irritability and raised deep tendon reflexes.
– Chronic renal failure: early personality changes,
polyneuropathy.
– Dialysis disequilibrium syndrome: usually when there is
rapid hemodialysis, it presents by headache, lethargy,
nausea and vomiting.
– Chronic dialysis encephalopathy (dialysis dementia): always
late in the course of chronic renal failure. It gives dysarthria,
deterioration in memory, progressing to myoclonus,
mutism, coma and death (Seifter 2011).
– Renal transplantation: patients with renal transplantation
may experience encephalopathy (soon) after
transplantation. Immunosuppressive drugs, like
cyclosporine and acyclovir may produce confusion, lethargy
and coma at toxic levels. (Later) after transplantation,
primary CNS lymphoma becomes a major concern in
patients developing symptoms of brain dysfunction.
Management of acute and chronic renal failure is by
hemodialysis; dialysis disequilibrium syndrome is managed by
prolonging the time of dialysis; management of chronic dialysis
encephalopathy is by renal transplantation

Fluid and Electrolyte Imbalance
Osmolarity disorders
Hypernatremia: Hypernatremia indicates a deficit of body
water relative to sodium concentration. Clinically, it is similar to
hyponatremia where encephalopathy possibly develops, due to
dehydration. Usually, hypernatremic patients are hypovolemic.
Common causes of hyponatremia are
– Pure water loss (in renal diabetes insipidus and external
insensible losses via the skin and lungs).
– Combined water and sodium loss (in renal osmotic diuresis
combined with inadequate water intake, and external
excessive sweating).
– Inadequate sodium gain (in cases of excessive sodium
administration, like hypertonic solutions; adrenal
hyperfunction, like hyperaldosteronism, Cushing
Syndrome; and intake of exogenous steroids).
Hypernatremia should be corrected slowly. When volume
depletion with circulatory insufficiency is predominant,
vigorous treatment with isotonic saline solution is mandatory.
When the cause is diabetes insipidus, administer 2-5 units of
aqueous vasopressin, or 1-5 mcg of desmopressin (DDAVP)
should be given subcutaneously or intranasally. When
hypernatremia is due to excessive gain, hypotonic (0.45%) saline
is used to replace, in part, additional water deficits.
Hyponatremia: Three types of hyponatremia are described:
Hypovolemic hyponatremia: patients with low intake of sodiumcontaining
fluids and have attempted replacement with free
water may present with encephalopathy.
Hypervolemic hyponatremia: usually seen in congestive heart
failure or hypoalbuminemia. This condition can be treated with
fluid restriction, a wise use of diuretics as well as treatment of
the primary cause.
Euvolemic hyponatremia: This condition is seen in syndromes of
inappropriate secretion of ADH (SIADH) adrenal insufficiency,
hypothyroidism, severe psychogenic polydipsia, and

hypoglycemia; also in pancreatitis with hyperlipidemia and
hyperproteinemia. The degree of encephalopathy produced by
hyponatremia depends on the rate of fall of serum sodium rather
than its value.
All cases of euvolemic hyponatremia are treated with fluid
restriction (800-1000 ml/d) and removal of precipitants (Young
1998).
Central pontine myelinolysis (CPM): Due to rapid correction
of hyponatremia by more than 10 meq/d. Clinically, patients
present with quadriparesis and cranial nerve dysfunction over
several days, which may be followed by encephalopathy. The
maximal lesion is seen in the basis pontis, but supratentorial
white matter is also affected.
Syndrome of inappropriate secretion of antidiuretic
hormone (SIADH): It is a common syndrome in neurological
diseases; it leads to hyponatremia and increases salt
concentration in urine (>20 mmoI/L). Serum ADH is high. Causes
of SIADH include
– Malignant neoplasms likes oat-cell carcinoma of lung, and
Hodgkin disease
– Non-malignant pulmonary diseases, e.g., TB, emphysema,
pneumothorax
– CNS diseases like subarachnoid hemorrhage, cerebral
venous thrombosis, encephalitis, and meningitis, and PNS
diseases like Guillain-Barré syndrome.
– Use of drugs like vincristine, carbamazepine, tricyclic
antidepressants, etc.
Slow correction of hyponatremia by IV 3% sodium solution is
recommended. IV 100 cc given over one-hour interval, until
serum sodium level reach 125 mmol/l. Do not exceed correction
rate of 2 mmol/h.
Hypercalcemia: The encephalopathy of hypercalcemia is not
different from any metabolic encephalopathy except in early
anosmia.

Other findings in hypercalcemia are myopathy, polyuria,
pruritis, nausea and vomiting. Patients start to complain at
serum calcium level of 13 mg/dl, when abnormal EEG changes
start to appear. Patients suffering from hyperparathyroidism
may manifest seizures independent of serum calcium level due
to elevated serum parathormone.
Management: Hypercalcemia is corrected by saline diuresis,
augmented with furosemide, followed by a choice of
mithramycin steroids, phosphate or etidronate.
Encephalopathy in Diabetic Patients
Hypoglycemia: Clinically, patients who develop hypoglycemia
are graded:
– At 20 mg/dl, immediate loss of consciousness in adults and
children, neonates resist hypoglycemia better,
– At 45 mg/dl, confusion, irritability. Sometimes unexplained
focal lesions appear with hypoglycemia.
Management: give IV glucose at 1 g/kg body weight, plus
thiamine 1 mg/kg to prevent Wernicke’s encephalopathy (Quinn
2002).
Nonketotic hyperosmolar hyperglycemia (NHH): Usually
occurs in diabetic patients whose insulin production is adequate
to inhibit lipolysis, but insufficient to prevent hyperglycemia,
which result in a marked osmotic diuresis. Diuresis leads to
dehydration and hyperosmolarity. In such situations, serum
glucose may rise to 800-1200 mg/dl, and serum osmolarity may
exceed 350 mOsm/L, which may invite development of brain
edema.
Osmolarity= 2(Na+K) + (glucose/18) + (BUN/2.8)
Clinically, patients present with encephalopathy, focal
neurological signs, and partial seizures that do not respond to
conventional antiepileptic medication. Such encephalopathy
must be treated by rehydration.
Management: Normal saline is infused slowly to correct
hypotension and improve osmolality, in addition to insulin

infusion at the rate of 10 IU/h, with regular checking of plasma
glucose, since these patients are very sensitive to insulin.
Glucose should be added to saline when plasma glucose is
approximately 300 mg/dl (Quinn 2002).
Diabetic ketoacidosis (DKA): About 80% of DKA patients have
encephalopathy and 10% are comatose. Management: Like NHH,
but with higher amounts of insulin. If there is evidence of brain
edema mannitol is used. If there is evidence of electrolyte
imbalance, mandate correction. The use of IV sodium
bicarbonate to compensate for metabolic acidosis is debatable
(Quinn 2002).
Hypoxic Ischemic Encephalopathy (HIE)
Following cardiac or respiratory arrest, CO poisoning or
cyanide poisoning, one of four clinical syndromes might appear:
– Global encephalopathy
– Memory loss
– Postanoxic Parkinsonism
– Lance-Adams syndrome (intention myoclonus)
Findings predicting good prognosis are preserved pupillary
responses, preserved roving eye movement, decorticate posture
or better at initial examination. We predict good prognosis when
we find in clinical examination after 24 hours, motor withdrawal
from noxious stimuli or improvement of 2 grades in eye
movement. Also, finding motor withdrawal or better, and
normal spontaneous eye movements at 72 hours examination,
carries a good prognosis. Also, when a patient obeys commands
at the 1-week examination.
Management is by hyperventilation and osmotic diuresis, for
cerebral edema. Seizure control is live saving and has an impact
on prognosis, as patients suffering from GTCS have a better
outcome than those who suffer from myoclonic seizures.
108

Septic Encephalopathy
Septic encephalopathy is a frequent sequel of severe sepsis, with
no definite therapeutic strategies available that can prevent
associated neurological dysfunction and damage. It is caused by
a number of processes, such as direct bacterial invasion, toxic
effects of endotoxins, inflammatory mediators, impairment of
microcirculation, and neuroendocrine changes. The exact
cellular and molecular mechanisms remain an enigma. Several
mediators of inflammation have been assigned a key role in
etiogenesis of encephalopathy, including cytokines, chemokines
and complement cascade. With the observations that brain
dysfunction in such sepsis disorders can be alleviated by
regulation of the cytokines and complements in various species
of animals, optimism is building for a possible therapy of the
sepsis-damaged brain (Jacob 2011).
Early aggressive treatment with antibiotics is key, along with
modulators of cytokines and complements and antiinflammatory
medicines (Jacob 2011).
Drug-induced Encephalopathies
Commonly implicated drugs in encephalopathy etiology include
salicylates, tricyclic antidepressants, lithium, sedatives,
neuroleptics, methyldopa, amantadine, acyclovir, digitalis,
propranolol, hydantoins, etc (Jain 2001).
Drug-induced delirium results from disruption of the normal
integration of neurotransmitters, including dopamine,
acetylcholine, glutamate, gamma-aminobutyric acid (GABA),
and/or serotonin (Young 1998).

General Neurological Treatment Strategies

" "

The concept of neurocritical care has been developed to
coordinate the management of critically ill neurological patients
within a single specialist unit and to include clinical situations
such as swallowing disturbances, respiratory problems
management in neurocritical care, infection control in the unit,
pain relief and sedation in some patients, as well as diagnosing
brain death. Acute rehabilitation is important in securing
improved long-term neurological outcomes after many brain
insults, trauma, ischemia or hemorrhage. Intervention from
neurophysiotherapists, as part of the neurocritical care
multiprofessional team, must occur as early as possible.
Respiratory muscle impairment is the most common reason for
admission to the ICU in patients with neuromuscular disorders.
Objective measures of respiratory muscle function are necessary
because significant respiratory muscle impairment may exist
despite a paucity of symptoms.
Analgesia in the neurocritical care unit is indicated in many
situations such as postoperative pain, traumatic injury, and
subacute or chronic pains. Although it is mandatory and
beneficial in many situations, precautions must be taken before

applying many agents; e.g., equipment and personnel to intubate
and mechanically ventilate patients must be readily available
with use of narcotic agonists. Some agents may cause decreased
level of consciousness or obtundation leading to impairment of
neurological exam. This chapter will cover management of these
issues in the neurocritical care setting.
Swallowing Disturbances
Weakness, spasticity or both of the pharynx and tongue cause
dysphagia and tendency for aspiration. A feeding tube through a
percutaneous endoscopic gastrostomy (PEG), cervical
esophagostomy or jejunostomy is a reliable method of patient
feeding when prolonged deficit is expected. Nutrition support by
enteral feeding through either a nasogastric or an orogastric
tube should be maintained in all intubated patients whenever
possible. In patients with a normal baseline nutritional state,
support should be initiated within 7 days. In malnourished
patients, nutritional support should be initiated within 72 h.
Delayed gastric emptying is common in critically ill patients on
sedative medications but often responds to promotility agents
such as domperidone, and metoclopramide (Gomes 2010).
Parenteral nutrition is an alternative to enteral nutrition in
patients with severe gastrointestinal pathology. When patients
improve the gastrostomy is easy to close. It is better to try to
provide adequate and safe nutrition by mouth in an alert patient
before placing a feeding tube. Nasogastric tubes have been used
temporarily for feeding but they are uncomfortable, cause
pressure necrosis of the nares when used chronically, and allow
aspiration, so it should not be used for long. In ambulatory
patients with severe dysphagia, cervical esophagostomy may be
useful, as a patient can insert the tube during feedings and the
ostium can be covered with a dressing at other times. Patients
with severe weakness or ataxia of the upper extremities are
usually unable to feed themselves, so percutaneous endoscopic
gastrostomy is the method of choice (Wanklyn 1995). Tube

feeding needs high caloric diet (1200 to 2400 Kcal/day) for active
patients, and liquid foods are usually concentrated to deliver
1 Kcal/ml. To avoid clogging of the tube, each feeding is followed
with water. Tube feeding is best started with about one half the
total desired calories diluted in water, with gradual increase in
concentration and calories, to avoid diarrhea and malabsorption,
reaching a maximum volume of about 200 ml (150 ml food and 50
ml of water). If aspiration of saliva and nasal secretions is a
problem, a cuffed endotracheal tube is necessary and the use of
tricyclic antidepressants or anticholinergic drugs (if there is no
absolute contraindication for its use) might reduce salivation
and prevent drooling (Fjærtoft 2011). Adequate nutritional
feeding, trace elements, minerals and vitamins constitute the
most important basic brain supplements.
Respiratory Management in Neurocritical Care
Most patients who are started on ventilatory support receive
synchronized intermittent mandatory ventilation (SIMV),
because this ensures user-specified backup minute ventilation in
the event that the patient fails to initiate respiratory efforts.
Once the intubated patient has been stabilized with respect to
oxygenation, definitive therapy for the underlying process
responsible for respiratory failure is formulated and initiated.
Subsequent modifications in ventilator therapy must be
provided in parallel with changes in the patient’s clinical status.
As improvement in respiratory function is noted, the first
priorities are to reduce PEEP and supplemental O2 and once a
patient can achieve adequate arterial saturation with an FIO2
≤0.5 and 5 cm H2O PEEP, attempts should be made to reduce the
level of mechanical ventilatory support and weaning should be
planned and started. Patients previously on full ventilator
support should be switched to a ventilator mode that allows for
weaning, such as SIMV, PSV (pressure support ventilation), or
SIMV combined with PSV. Ventilator therapy can then be
gradually removed while patients whose condition continues to

deteriorate after ventilator support is initiated may require
increased O2, PEEP, and alternative modes of ventilation such as
IRV or OLV (Borel 2000). Patients who are started on mechanical
ventilation usually require some form of sedation and analgesia
to maintain an acceptable level of comfort. Often, this regimen
consists of a combination of a benzodiazepine and opiate
administered intravenously. Medications commonly used for this
purpose include lorazepam, midazolam, diazepam, morphine,
and fentanyl. Immobilized patients in the intensive care unit on
mechanical ventilatory support are at increased risk for deep
venous thrombosis; accepted practice consists of administering
prophylaxis in the form of subcutaneous heparin and/or
pneumatic compression boots. Fractionated low-molecularweight
heparin has also been used for this purpose; it appears to
be equally effective and is associated with a decreased incidence
of heparin-associated thrombocytopenia (Pelosi 2011).
Prophylaxis against diffuse gastrointestinal mucosal injury is
indicated for patients who have suffered a neurologic insult, so
histamine receptor antagonists (H2 receptor antagonists), proton
pump inhibitors, and cytoprotective agents such as Carafate
have all been used for this purpose and appear to be effective.
Recent data suggest that Carafate use is associated with a
reduction in the incidence of nosocomial pneumonias, since it
does not cause changes in stomach PH and is less likely to permit
colonization of the gastrointestinal tract by nosocomial
organisms at neutral PH.
Endotracheal intubation and positive-pressure mechanical
ventilation have direct and indirect effects on several organ
systems, including the lung and upper airways, the
cardiovascular system, and the gastrointestinal system.
Pulmonary complications include barotraumas, nosocomial
pneumonia, oxygen toxicity, tracheal stenosis, and
deconditioning of respiratory muscles (Hurford 2002).
Upper airway function must be intact for a patient to remain
extubated but is difficult to assess in the intubated patient.
Therefore, if a patient can breathe on his own, through an

endotracheal tube, but develops stridor or recurrent aspiration
once the tube is removed, upper airway dysfunction or an
abnormal swallowing mechanism should be suspected, and plans
for achieving a stable airway be developed. An intact cough
reflex during suctioning is a good indicator of a patient’s ability
to mobilize secretions. Respiratory drive and chest wall
functions are assessed by observation of respiratory rate, tidal
volume, inspiratory pressure, and vital capacity (Hardin 2006).
The weaning index, defined as the ratio of breathing frequency
to tidal volume (breaths per minute per liter), is both sensitive
and specific for predicting the likelihood of successful
extubation. When this ratio is less than 105, and the patient can
breathe without mechanical assistance through an endotracheal
tube, successful extubation is likely. An inspiratory pressure of
more than −30 cm H2O and a vital capacity of greater than 10
ml/kg are considered indicators of acceptable chest wall and
diaphragm functions. Alveolar ventilation is generally adequate
when elimination of CO2 is sufficient to maintain arterial pH in
the range of 7.35 to 7.40, and SaO2 >90% can be achieved with an
FIO2 <0.5 and a PEEP ≤5 cm H2O. Although many patients may not
meet all criteria for weaning, the likelihood that a patient will
tolerate extubation without difficulty increases as more criteria
are met (Hurford 2002).
Many approaches to wean patients from ventilator support
have been advocated. T-piece and CPAP weaning are best
tolerated by patients who have undergone mechanical
ventilation for brief periods and require little respiratory muscle
reconditioning, whereas SIMV and PSV are best for patients who
have been intubated for extended periods and require gradual
respiratory muscle reconditioning. Weaning by means of SIMV
involves gradual tapering of the mandatory backup rate, in
increments of 2 to 4 breaths per minute, while monitoring blood
gas parameters and respiratory rates (Webb 1999). Rates of
greater than 25 breaths per minute, on withdrawal of mandatory
ventilator breaths, generally indicate respiratory muscle fatigue
and the need to combine periods of exercise with periods of rest.

Exercise periods are gradually increased until a patient remains
stable on SIMV at 4 breaths per minute or less without needing
rest at higher SIMV rates. A CPAP or T-piece trial can then be
attempted before planned extubation (Bernard 1994).
Infection Control in Neurocritical Care
Sepsis (and the systemic inflammatory response to sepsis)
remains the major cause of organ failure and death in the
intensive care unit, being either directly or indirectly
responsible for 75% of all deaths (Valles 1997). Common sites of
infection include the urinary tract, respiratory tract (especially
ventilator associated pneumonia), vascular cannulae (catheter
related sepsis) and long-term use of nasogastric feeding tubes.
Vascular cannulae sepsis, particularly those associated with
internal jugular and subclavian catheters, constitute the
majority, but peripheral catheters also carry a considerable risk
of infections. Thus, placement of intravenous lines requires
careful aseptic technique and regular changing of lines. It is
important to culture specimens from the tips of catheters that
have been removed. Catheter related infections are usually
caused by Staphylococcus epidermidis or Staphylococcus aureus, and
its treatment is empiric depending on use of vancomycin and
cephalosporins.
Most nosocomial infections seen in the NICU are endogenous,
caused by colonization of the patient’s GI tract by pathogenic
bacteria which then translocate through the intestinal mucosa to
reach distant sites by hematogenous spread. With vascular
cannulation, Gram-negative organisms such as Escherichia coli,
Klebsiella, and Pseudomonas species stated to be traditionally
responsible, but Gram-positive organisms (Streptococcus and
Staphylococcus species) are increasingly suspected. Fungi may
also be implicated and considered the most serious infection
(Vincent 1995).
Patients intubated for longer than 72 hours are at high risk for
ventilator-associated pneumonia as a result of aspiration from

the upper airways through small leaks around the endotracheal
tube cuff; the most common organisms responsible for these
conditions are enteric gram-negative rods, Staphylococcus aureus,
and anaerobic bacteria. Because the endotracheal tube and
upper airways of patients on mechanical ventilation are
commonly colonized with bacteria, the diagnosis of nosocomial
pneumonia requires “protected brush” bronchoscopic sampling
of airway secretions coupled with quantitative microbiologic
techniques to differentiate colonization from infection.
Precautions and ways to combat nosocomial infections involve
the isolation of the infected patient whenever possible,
meticulous staff hygiene (hand washing before and after each
patient contact, aseptic techniques for invasive procedures, etc),
early identification and treatment of infection by the routine
sending of blood, urine, sputum, etc, for culture, use of
disposable equipments and, most importantly, joint daily ward
rounds between microbiologists and the ICU team (Fagon 1993).
Treatment of nosocomial infection, with or without septicemia,
requires the administration of appropriate antibacterial drugs in
adequate doses for an appropriate period.
Pain Relief and Sedation
In neurocritical care units different scores are used to evaluate
patient anxiety, agitation and response to surroundings.
Different scores for pain, anxiety and agitation are used as
guidelines for medication when pain is common, with a variety
of causes such as anxiety, confusion, sleep deprivation, sepsis,
drug withdrawal – especially sedatives, metabolic (hypo- or
hyperglycemia, hypernatremia, uremia, hepatic precoma or
coma), respiratory (infection, hypoxemia, hypercapnia) and
cardiac (low output state, hypotension). It may manifest as
discomfort, pulling at intravenous and bladder catheters,
tracheal and nasogastric tubes, shouting, aggressive behavior,
extreme restlessness, and confusion (Schnakers 2007). Pain is
particularly common and often unrecognized because of

confusion and the difficulties with communication in the
aphonic, aphasic or paralyzed patient. Clinical assessment may
direct attention through finding of profuse sweating, sustained
tachycardia and blood pressure fluctuations, and dilated pupils.
Most patients will require sedation but there is a natural
reluctance to sedate patients with an evolving CNS disorder
(McMillian 2011).
The first line of management is to reassure and calm the
patient, ensuring a quiet environment and normal diurnal cycle.
Next, there should be careful nursing and treatment of the
underlying causes, including positioning, splinting, bed cages,
catheterization, and physical treatments. In spite of reluctance
to sedate patients with an evolving CNS disorder, sedative
medications are mandatory and must be used, when indicated, to
reduce pain, distress, and anxiety. Also they may be used to help
patient toleration of tracheal tubes, IPPV, tracheal suction, and
physiotherapy. Patients with neurologically-induced respiratory
failure often require prolonged periods of mechanical
ventilation in the NICU, and it is not desirable to keep such
patients continuously sedated throughout their stay (Jacobi
2002). Furthermore, assessment of their condition and
subsequent weaning is impossible when sedated. However,
during periods of cardiorespiratory instability, raised ICP, and in
cases of intubation, sedation may be essential. In the ICU
environment, however, medications are often needed to calm
patients. As many as 30 different medications are used, and the
agents most frequently used are midazolam, olanzapine,
propofol, lorazepam and opioid analgesics (Jacobi 2002).
Bedside approach to the agitated patient
Assess and manage immediate threat to life (airway, breathing,
circulation and temperature).
Assess pain by querying the patient about pain, and assess for
noxious stimuli after measurement of pain score. Correct any
identified causes, and if the patient is hemodynamically
unstable, give fentanyl 25-100 μgm IV q5-15 minutes until

desired effect is achieved, or hydromorphone 0.25-0.75 mg IV
q 5-15 minutes; if patient is hemodynamically stable, give
morphine 2-5 mg IV every 5-15 minutes.
Assess anxiety. When patients have been on sedative and
analgesic drug infusions for longer than 24 hours and begin to
recover, daily interruptions of drug infusions for a time period
sufficient to allow awakening is recommended (Blanchard 2002).
Role of Rehabilitation
Rehabilitation is a reiterative active, educational, problemsolving
process, focused on the patient disability. The
operational characteristics of rehabilitation services comprise a
multidisciplinary group of people who work together towards
common goals for each patient, involve and educate the patient
and family in the process, have relevant expertise and
experience (knowledge and skills) and can, between them,
resolve most of the common problems faced by patients (Arnow
1993). The stages of rehabilitation involve general assessment to
identify the nature and extent of the patient problems and the
factors relevant to their solution, goal setting, then intervention,
which may include treatments that affect the process of change
and support (care), which maintain life and safety, and finally
evaluation to check on the effects of the intervention.
The aims of rehabilitation are to maximize the participation of
the patient in his/her social setting, minimize the pain and
distress experienced by the patient and family and/or carers
(Collen 1990). The development of models for disability has
fostered and clarified discussion about the nature of
rehabilitation. Definition of rehabilitation refers to the
operational characteristics of a rehabilitation service (structure),
how rehabilitation service works (process) and the aims of
rehabilitation services (outcome).
Three specific core skills are particularly associated with

rehabilitation:

1. An ability to assess all relevant aspects of patients’ situation
not simply their disease and its symptoms and signs, formulating
the important interactions.
2. An ability to set realistic but challenging goals in both the
short and long term, a skill that depends upon an accurate
evaluation of the likely prognosis and scope for effective
intervention.
3. An ability to participate in teamwork, working co-operatively
with a group of other experts towards agreed common goals
(Cunningham 2000).
However, recent research, mostly related to stroke, does
support various hypotheses. First, there is now evidence that
even quite small levels of intervention can have powerful and
specific effects. There is also some evidence of a dose response
relationship between therapeutic input and outcome. Second,
there is some evidence that even the simple provision of
information may be effective and acting on the patient’s
personal context is an important component of rehabilitation
(Badley 1993). Third, patients should be encouraged to seek
review of medication at regular intervals, perhaps every 6
months, especially for drugs that have cognitive side effects and
either slow or hinder the process of recovery. Use drugs to treat
specific impairments like gabapentin to improve visual acuity
when nystagmus is present, and a dopamine agonist (such as
bromocriptine) to ameliorate the reduced initiation after frontal
lobe damage. Acute-onset disability is often considered the
easiest to manage, because recovery occurs and may be complete
(Wade 2000).
Diagnostic Findings in Cerebral Death
Brain death is the irreversible cessation of cerebral functions.
Cerebral death, which is of medico-legal importance, cannot be
accurately assessed on the basis of a criterion related to a single
functional system. It is basically characterized by the absence of
three main brain functions:

(1) Cerebral activity known as “cerebral responsivity”
(2) Vital functions
(3) Cephalic reflexes
Cephalic reflexes are mediated by the cranial nerves, and are
considered as important indicators of the integrity of the
brainstem, as absence of the cephalic reflexes are essential for
considering the diagnosis of brain death, yet they differ in
importance as a criterion of death. Some cephalic reflexes are
considered to be the most sensitive and discriminative of
brainstem functions, and accepted as criteria of cerebral brain
death, e.g., pupillary reflex was absent in 98.4% of cases, corneal
reflex was absent in practically all cases, and returns early if the
patient shows any signs of survival (unless the cornea has
become insensitive due to edema or dehydration), the
oculocephalic reflex (Doll’s eyes response) is quite
discriminative, and it returns early, with evidence of recovery.
The vestibular reflex is also quite discriminative for cerebral
death, but upon recovery, it is slightly slower to return than the
pupillary and corneal reflexes. The audio-ocular reflex, which is
a blink of the eyelids in response to a sudden clap, is not as
discriminative as the other cephalic reflexes, and is somewhat
slower to recover. Snout reflex, pharyngeal (gag) reflex,
swallowing reflex and cough reflex are not particularly
discriminative.
In the collaborative study of cerebral death, absence of all
cephalic reflexes was noted in more than half of the cases,
whereas audio-ocular reflex was absent in 99% of cases, the
pupillary reflex was absent in only 76.6%, and the audio-ocular
reflex was extremely sensitive to brain injury.
Certain combinations of cranial nerve reflexes have been
specified as essential for the diagnosis of brain death. The
absence of pupillary, corneal, vestibular, audio-ocular and
oculocephalic reflexes showed significant correlation with
cerebral death (Smith 1973), so all clinical tests are needed to
declare brain death and are likely essential.

Spinal reflexes may sometimes be present but they are not
relevant in establishing brain death. Isolated clinical studies,
electroencephalographic (EEG) examination and even blood flow
determination, fall short of an absolute diagnosis of cerebral
death.
Repetitive studies of a single functional system provide greater
confidence but the time period is long and many patients die of
cardiac arrest before they meet the criteria for cerebral death
(George 1991). Within limits, the shorter the period of
observation the more individuals may be diagnosed as cerebrally
dead (Smith 1973), so after the first rapid evaluation, the clinical
data should be confirmed by the following tests:
– conventional or CT angiography which shows no
intracerebral filling at the level of the carotid bifurcation or
Circle of Willis, patency of the external carotid circulation,
and a delay in the filling of the superior longitudinal sinus
(Brodac 1974; Frampas 2009),
– electroencephalography which shows no electrical activity
for at least 30 minutes of recording in suspected brain death,
as adopted by the American Electroencephalographic
Society (Benett 1978), including 16 channel
electroencephalographic instruments,
– transcranial Doppler ultrasonography; ten per cent of
patients may not have temporal insonation windows,
therefore the initial absence of Doppler signals cannot be
interpreted as consistent with brain death; small systolic
peaks in early systole without diastolic flow or reverberating
flow, indicating very high vascular resistance associated
with greatly increased intracranial pressure (Ropper 1987),
– technetium 99m hexamethyl propylene amineoxime brain
scan: no uptake of isotope in brain (Hollow skull
phenomenon),
– somatosensory evoked potentials which showed bilateral
absence of N20-P23 response with median nerve stimulation,
and its recordings should adhere to the minimal technical

criteria for somatosensory evoked potentials recording in
suspected brain death as adopted by the American
electrophysiological society (Benett 1978).
Conclusion
A neurological intensive care unit requires a multidisciplinary
approach to the management of critically ill patients. The
intensivists’ and neurologists’ attention to communication, daily
nursing care, physical therapy, and infection control will ensure
the best outcome (Rivers 2001).

Cerebral Edema

" "

Cerebral edema is a challenging problem in the neurocritical
care setting. Different etiologies may cause increased
intracranial pressure. Secondary brain injury may ensue as a
result of cerebral edema, and may result in different herniation
syndromes.
Brain monitoring for increased intracranial pressure may by
employed in certain patient populations. Serial neuroimaging
may be useful in monitoring exacerbations of brain edema.
Osmotherapy has been recommended for management of
cerebral edema. Mannitol and hypertonic saline are the two
agents widely used for this purpose. Knowledge of possible side
effects of osmotherapeutic agents is necessary. Common
concerns of such therapies include renal insufficiency,
pulmonary edema, and exacerbation of congestive heart failure,
hypernatremia, hemolysis, and hypotension. Specific measures
as controlled ventilation, sedation and analgesia, pharmacologic
coma, hypothermia and surgical decompression may be required
in patient subpopulations. Important questions still need to be
answered regarding the timing of the decompressive surgery
and patient selection criteria.
Surgical decompression may be applicable in certain patients.
Recent studies indicate that surgical decompression may

significantly reduce mortality in young patients with malignant
cerebral infarcts.
General medical management is focused toward limiting
secondary brain damage. General measures include head and
neck position, optimization of cerebral perfusion and
oxygenation, management of fever, nutritional support and
glycemic control.
Abnormalities of intracranial pressure may result in pathology
requiring urgent evaluation and intervention to prevent lifethreatening
consequences. This pathology may represent
intracranial hyper- or hypotension, or it may manifest as an
abnormality of cerebrospinal fluid (CSF) dynamics, such as
hydrocephalus. Elevated intracerebral pressure is the final
common pathway for almost all pathology leading to brain
death, and interventions to treat ICP may preserve life and
improve neurologic function after head trauma, stroke, or other
neurologic emergencies.
Common causes of raised intracranial pressure are shown in
Table 7.1, symptoms and signs in Table 7.2.

Table 7.1 – Some common causes of increased intracranial pressure
(Czosnyka 1999)
Head injury Intracranial hematoma (extradural, subdural, and
intracerebral)
Diffuse brain swelling
Contusion
Cerebrovascular Subarachnoid hemorrhage
Intracerebral hematoma
Cerebral venous thrombosis
Major cerebral infarct
Hypertensive encephalopathy (malignant hypertension,
eclampsia)
Hydrocephalus Congenital or acquired
Obstructive or communicating
Craniocerebral
disproportion
Brain “tumour” (cysts; benign or malignant tumours)
Secondary hydrocephalus
Mass effect
Oedema
“Benign” intracranial hypertension (pseudotumor cerebri;
idiopathic intracranial hypertension)
CNS infection Meningitis
Encephalitis
Abscess
Cerebral malaria
Metabolic
encephalopathy
Hypoxic-ischemic
Reye’s syndrome, etc.
Lead encephalopathy
Hepatic coma
Renal failure
Diabetic ketoacidosis
Burns
Near drowning
Hyponatremia
Status epilepticus
Types of Cerebral Edema
Cerebral swelling or edema can complicate many intracranial
pathologic processes including neoplasms, hemorrhage, trauma,
autoimmune diseases, hyperemia, or ischemia. There are
essentially three types of cerebral edema:
1. Cytotoxic edema is associated with cell death and failure of
ion homeostasis. Cytotoxic edema results from energy
failure of a cell as a result of hypoxic or ischemic stress,

which leads to cell death. Intracellular swelling occurs and
results in the CT and MR appearance of both gray and
white matter edema, usually in the distribution of a
vascular or borderzone territory after hypoxia or stroke.
2. Vasogenic edema is associated with breakdown of the
blood-brain barrier. Vasogenic edema represents
breakdown of the blood-brain barrier, appears mostly in
the white matter, and is more likely to be associated with
neoplasms or cerebral abscesses. In reality, cerebral edema
in many situations, usually exhibit a combination of
vasogenic and cytotoxic edema.
3. Interstitial (hydrostatic or hydrocephalic) edema is
associated with hydrocephalus, in which there is increased
tension of CSF across the ependyma. Interstitial edema, or
transependymal flow, is radiographically seen with
hypodense areas surrounding the ventricular system and is
associated with increased CSF volume or pressure.
Table 7.2 – Signs and symptoms of increased intracranial pressure and
edema
Symptoms Physical Signs
Headache, worsened with Valsalva
Decreased visual acuity
Diplopia
Nausea
Vomiting
Progressive decline in level of
consciousness
Decreased upward gaze
Cranial nerve VI palsy
Papilledema
Loss of normal venous pulsations in the
fundus
Field cut or enlarged physiologic blind
spot
Alterations in vital signs
Management of Cerebral Edema
The treatment of cerebral edema depends mainly on treating the
underlying cause. In cytotoxic edema, osmotic therapy with
mannitol and hypertonic saline may not reduce edema in the

lesion itself, but may reduce the volume of normal brain
allowing for some increased margin of safety by decreasing
intracranial pressure (Raslan 2007). Steroids are of no value in
cytotoxic edema due to stroke, and may be harmful in the
settings of brain trauma. Surgical decompression of cytotoxic
edema with decompressive craniectomy may be therapeutic, and
life-saving (Hofmeijer 2009). Vasogenic edema responds to
steroids and surgical resection of the lesion, and may also
benefit from osmotic therapy with mannitol or hypertonic saline
(Oddo 2009). Hydrostatic edema is treated surgically with CSF
removal or shunting, and it is treated medically with agents to
decrease production of CSF, such as acetazolamide and
furosemide.

Neurocritical Monitoring

" "

Neurocritical care relies on monitoring cerebral functions.
Intracranial pressure monitoring may indicate high pressure in
several acute neurological conditions. Massive stroke may cause
life-threatening brain edema and occur in about 10% of patients
with supratentorial stroke. Massive brain edema usually occurs
between the second and the fifth day after stroke onset. Case
fatality rates may exceed 80%. Other neurologic conditions that
may be accompanied with increased intracranial pressure
include severe head injuries, status epilepticus, fulminant
hepatic failure, Reye’s syndrome, and metabolic
encephalopathies.
Neuro-Specific Monitoring
Accurate neurological assessment is fundamental for the
management of patients with intracranial pathology. This
consists of repeated clinical examinations (particularly GCS and
pupillary response) and the use of specific monitoring
techniques, including serial CT scans of the brain. This chapter
provides an overview of the more common monitoring
modalities found within the neurocritical care environment

In general terms, a combination of assessments is more likely
to detect change than any one specific modality. Real-time
continuous monitoring (e.g., measurement of intracranial
pressure, ICP) will provide more timely warning about adverse
events (e.g., an expanding hematoma) compared to static
assessments such as sedation holds or serial CT brain scans.
Clinical Assessment
The Glasgow Coma Scale
The Glasgow Coma Scale (GCS) provides a standardized and
internationally recognized method for evaluating a patient’s CNS
function by recording their best response to verbal and physical
stimuli. A drop of two or more GCS points (or one or more motor
points) should prompt urgent re-evaluation and a repeat CT
scan.
Eye opening is not synonymous with awareness, and can be
seen in both coma and persistent vegetative state (PVS). The
important detail is that the patients either open their eyes to a
specific command or shows ability to fix eye on a specific target
or follows a visual stimulus.
Pupillary response
Changes in pupil size and reaction may provide useful additional
information:
– Sudden unilateral fixed pupil: Compression of the third
nerve, e.g., ipsilateral uncal herniation or posterior
communicating artery aneurysm
– Unilateral miosis: Horner syndrome (consider vascular
injury)
– Bilateral miosis: Narcotics, pontine hemorrhage
– Bilateral fixed, dilated pupils: Brainstem death, massive
overdose (e.g., tricyclic antidepressants)
In the non-specialist center, neurological assessment of the
ventilated patient consists of serial CT brain scans, pupillary

response, and assessment of GCS during sedation holds. A
reduction in sedation level will usually be at the suggestion of
the Regional Neurosurgical Center (RNC) and its timing will
depend upon a number of factors. Responses such as unilateral
pupillary dilatation, extensor posturing, seizures, or severe
hypertension should prompt rapid re-sedation, repeat CT scan,
and contact with the RNC. In the patient with multiple injuries,
consideration must be given to their analgesic requirements
prior to any decrease in sedation levels.
Invasive Monitoring
Cerebral perfusion pressure (CPP) reflects the pressure gradient
that drives cerebral blood flow (CBF), and hence cerebral oxygen
delivery. Measurement of intracranial pressure (ICP) allows
estimation of CPP. CPP is mean arterial pressure minus ICP.
Sufficient CPP is needed to allow CBF to meet the metabolic
requirements of the brain. An inadequate CPP may result in the
failure of autoregulation of flow to meet metabolic needs whilst
an artificially induced high CPP may result in hyperemia and
vasogenic edema, thereby worsening ICP. The CPP needs to be
assessed for each individual and other monitoring modality (e.g.,
jugular venous oximetry, brain tissue oxygenation) may be
required to assess its adequacy.
Despite its almost universal acceptance, there are no properly
controlled trials demonstrating improved outcome from either
ICP- or CPP-targeted therapy. However, in the early 1990s
Marmarou et al. showed that patients with ICP values
consistently greater than 20 mmHg suffered worse outcomes
than matched controls, and poorer outcomes have been
described in patients, whose CPP dropped below 60 mmHg
(Marmarou 1991; Juul 2000; Young 2003). As such, ICP- and CPPtargeted
therapy have become an accepted standard of care in
head injury management.
The 2007 Brain Trauma Foundation Guidelines (Brain Trauma
Foundation 2007) recommend treating ICP values above

20 mmHg and to target CPP in the range of 50-70 mmHg. Patients
with intact pressure autoregulation will tolerate higher CPP
values. Aggressive attempts to maintain CPP >70 mmHg should
be avoided because of the risk of ARDS.
Table 6.1 – ICP Values
Normal ICP <15 mmHg
Focal ischemia occurs at ICP >20 mmHg
Global ischemia occurs at ICP >50 mmHg
Usual treatment threshold is 20 mmHg
Measuring ICP
Intraventricular devices consist of a drain inserted into the
lateral ventricle via a burr hole, and connected to a pressure
transducer, manometer, or fiber optic catheter. Although
associated with a higher incidence of infection and greater
potential for brain injury during placement, this remains the
gold standard. It has the added benefit of allowing CSF drainage.
Historically, saline could be injected to assess brain compliance.
Extraventricular systems are placed in parenchymal tissue, the
subarachnoid space, or in the epidural space via a burr hole.
These can be inserted at the bedside in the ICU. These systems
are tipped with a transducer requiring calibration, and are
subject to drift (particularly after long-term placement).
Examples of extraventricular systems are the Codman and
Camino devices. These devices have a tendency to underestimate
ICP.
In general, both types of device are left in situ for as short a
time as possible to minimize the risk of introducing infection.
Prophylactic antibiotics are not generally used.
Indications for ICP monitoring
In any case of head injury, if brain CT is positive for pathology,
and the patient fulfills the criteria for use of a ventilator,

monitoring for intracranial pressure (ICP) becomes mandatory.
More specific indications are shown in Table 6.2.
Table 6.2 – Indications for intracranial pressure (ICP) monitoring
1) Traumatic brain injury, in particular:
– Severe head injury (GCS <8)
– Focal pathology on CT brain scan
– Head injury and age >40
– Normal CT brain scan but systolic blood pressure persistently <90 mmHg
– Where other injuries and their treatment necessitate the use of sedation
or anesthesia
2) Subarachnoid hemorrhage with associated hydrocephalus
3) Hydrocephalus
4) Hypoxic brain injury, for example, after near-drowning
5) Postoperative in patients at risk of severe cerebral edema
6) Encephalopathy (e.g., in liver failure)
Coagulopathy is the primary contraindication to insertion. The
ICP device will generally be removed as soon as the patient is
awake with satisfactory neurology (GCS motor score M5 or M6)
or when physiological challenges (removal of sedation,
normalizing PaCO2) no longer produce a sustained rise in ICP.
Intracranial Pressure Waveforms and Analysis
The normal ICP waveform is a modified arterial trace and
consists of three characteristic peaks. The “percussive” P1 wave
results from arterial pressure being transmitted from the
choroid plexi, the “tidal” P2 wave varies with brain compliance,
whilst P3 represents the dicrotic notch and closure of the aortic
valve. It is important to establish the accuracy of the ICP trace
and value before initiating therapy based upon the numbers
generated. Transient sequential occlusion of the internal jugular
veins or removing the head-up tilt should produce an increase in
ICP.

The ICP waveform: The ICP waveform can be divided into
systolic and diastolic components and demonstrates cardiac and
respiratory variations.
– P1 (percussion wave): originates from pulsations in choroid
plexus, sharp peak, consistent in amplitude
– P2 (tidal wave): variable in shape, ends on dicrotic notch
– P3 (dicrotic wave): begins immediately after dicrotic notch
When ICP increases and compliance decreases, P2 and P3 elevate
causing a rounder waveform.
In addition to simple pressure measurement, if ICP is recorded
against time, a number of characteristic wave forms (Lundeberg
waves) can be seen.
– A waves: Pathological sustained plateau waves of 50-80
mmHg lasting between 5 and 10 min, possibly representing
cerebral vasodilatation and an increase in CBF response to a
low CPP.
– B waves: Small, transient waves of limited amplitude every
1-2 min representing fluctuations in cerebral blood volume.
These may be seen in normal subjects, but are indicative of
intracranial pathology when the amplitude increases above
10 mmHg.
– C waves: Small oscillations in ICP that reflect changes in
systemic arterial pressure.
With cerebral autoregulation intact, a rise in MAP produces
vasoconstriction and a fall in ICP. However, when autoregulation
fails, the circulation becomes pressure passive and changes in

MAP are reflected in changes in the ICP. Continuous analysis of
MAP and ICP allows a correlation coefficient called the pressure
reactivity index to be derived (PRx). Positive values indicate
disturbed cerebral vascular reactivity, whilst negative values
indicate that reactivity remains intact (Gupta 2002).
Despite the fact that trial results have not always been
compelling, most clinicians regard the ICP monitor as an
essential tool that allows estimation of CPP (Czosnyka 2004;
Czosnyka 1996), gives early warning of developing pathology,
allows the response to therapy to be objectively measured, and
has value as a prognostic indicator (Joseph 2005).
Methods of intracranial pressure ICP measurement:
Methods for the measurement of intracranial pressure are
ventriculostomy, subdural catheter, epidural transducer, and
fiberoptic microtransducer. Ventriculostomy remains the gold
standard for monitoring ICP as it offers an accurate and reliable
means of calibration. Disadvantages include a <2% risk for
infection, a <10% risk for hemorrhage and difficulty in placing
the catheter (Clark 1989).
One of the widely used forms of ICP monitoring is the fiberoptic
or bolt ICP monitor. This method is relatively less invasive with
lower morbidity. It lacks, however therapeutic CSF drainage.
Steps to ICP bolt placement (Crutchfeld 1990):
1. ICP bolt placement takes place in the ICU or the OR
2. A small skin incision to the skull bone is made
3. The periosteum is stripped off the bone and a drill burr
hole is made to match the size of a bolt adapter. The bolt
screw is then advanced into the skull and a dilator is used
to dilate a tract in the dura for the fiberoptic probe
4. The skin is closed around the bolt
5. The fiberoptic probe is zeroed at atmospheric pressure
6. The probe is then placed through a retaining cap into the
subarachnoid space or less commonly intraparenchymally
7. Probe placement can be verified by observing ICP
waveforms

8. The ICP monitor is then connected to the bedside
monitorFiberoptic transducers cannot be recalibrated
externally.
Table 6.3 – Comparison between fiberoptic ICP monitoring and
ventriculostomy (Andrew 2010)
Fiberoptic bolt Ventriculostomy
Accuracy Subject to shift Gold standard
Placement Relatively easy Relatively hard
Feasibility of use No recalibration Requires height adjustment
and zeroing
Clinical use Measurement only Measurement and CSF
draining
ICP measurement Focal pressure variation is a
disadvantage
Focal variation disadvantage
is less
Continuous care Lower burden Higher burden
Risk of infection Lower Increases after >5 days
ICP waveform analysis: Analysis of the relationship between
ICP waveforms and ABP waveforms, i.e., pressure reactivity
index (PRx), has been outlined (Czosnyka 1996).
1. PRx varies from low values (no correlation) to values
approaching 1.0 (strong correlation)
2. With lower BP, lower blood vessel wall tension results in an
increase in transmission of the BP waveform to the ICP
3. With elevated ICP brain compliance is reduced thus
increasing the transmission of the BP waveform
Approximate Entropy (ApEn) is a logarithmic measure of
system regularity or randomness that can be used in physiologic
systems (Pincus 1991). Reductions in ApEn imply reduced
randomness or increased order and may indicate pathology in
the cardiovascular, respiratory and endocrine systems.
Cerebral blood flow CBF measurement is feasible and may be
considered in some patients. On the other hand, cerebral tissue
metabolic demand is not currently available.
Neurocritical

Indirect measures of CBF include measurement of surrogates of
cerebral tissue physiology as jugular venous oxygen saturation,
tissue oxygen tension, and microdialysis. Other indirect markers
for CBF measurement include cerebral perfusion pressure (CPP)
and less directly continuous EEG.
Jugular Venous Oximetry (SjvO2)
SjvO2 is an indicator of global oxygen extraction of the brain.
Jugular venous desaturation suggests an increase in cerebral
oxygen extraction which indirectly implies that there has been a
decrease in cerebral oxygen delivery, and hence perfusion.
The internal jugular vein drains the majority of blood from the
brain, and in most patients the right lateral sinus is larger.
Despite the fact that flow is different on the two sides, oxygen
saturations are normally very similar. This also appears to be the
case in diffuse brain injury, whilst in focal injuries there tends to
be a greater difference in the saturation of the two veins.
Jugular venous saturation can be measured using the principle
of infrared refractometry via a specially designed catheter
(Gopinath 1994). SjvO2 values are:
– 55-75% - normal
– >75% - luxury perfusion
– <54% hypoperfusion
– <40% suggests global ischemia and is associated with
increased cerebral lactate production.
Insertion of Jugular Venous Saturation Catheter: Insertion
involves retrograde cannulation of the internal jugular vein. A
pediatric pulmonary artery catheter introducer can be used
through which the fiber optic SjvO2 catheter is advanced beyond
the outlet of the common facial vein to the level of the jugular
bulb at the base of the skull. Ultrasound is often used for
accurate identification of vein position to avoid arterial
puncture, and to minimize the risk of hematoma formation
which can in turn impede venous drainage. Correct positioning

is confirmed on a lateral neck X-ray with the catheter tip lying at
the level of the mastoid air cells.
Indications for SjvO2 Monitoring:
– Acute brain injury. An association between SjvO2
desaturation and poor neurological outcome has been
observed. Fandino showed that in traumatic head injury
SjvO2 was the only factor associated with outcome, whilst
Gopinath showed that multiple SjvO2 desaturations were
associated with an increased incidence of poor neurological
outcome compared to those who showed no desaturations
(Moppett 2004).
– Optimal CPP would appear to be at the point when further
increases in MAP do not lead to a rise in SjvO2.
– Monitoring of therapy response. If ICP and SjvO2 are both
raised, hyperemia is implied and hyperventilation is
appropriate. SjvO2 should be monitored and kept above 55%
in these circumstances, as excessive hyperventilation may
cause profound cerebral vasoconstriction and cerebral
ischemia. More recent work using PET scanning, however,
has cast some doubt on the value of SjvO2, with
hyperventilation appearing to increase ischemic brain
volume without necessarily producing a fall in jugular
venous saturation.
– To guide optimal blood pressure and PaCO2 management
during operative treatment of aneurysms following SAH.
During the operative treatment of an aneurysm,
hypertension must be avoided because of the risk of rupture
and bleeding. However, excessive reductions in blood
pressure may risk cerebral ischemia, especially in those
patients with preoperative hypertension. SjvO2 monitoring
allows the anesthetist to assess the degree to which blood
pressure can be safely lowered during the operative period.
Similarly, a low PaCO2 will cause SjvO2 desaturation.

Problems with SjvO2 Monitoring: The major criticism of SjvO2
is that it is a measure of global oxygen delivery and does not
reflect metabolic inadequacies in focal areas of injury and hence
may miss regional areas of ischemia. Inaccuracies can occur with
catheter misplacement, contamination with extra cerebral
blood, when the catheter abuts the vessel wall, or if thrombosis
occurs around the catheter tip. Contraindications and
complications are similar to those of an IJV central line.
Interpretation of Changes in SjvO2:
If cerebral oxygen delivery is impaired, oxygen extraction
increases and SjvO2 decreases. If autoregulation is intact, CBF
increases to meet metabolic demand and SjvO2 is restored.
However, in the injured brain autoregulation is often impaired
and cerebral ischemia ensues.
– Decreased SjvO2: This implies inadequate cerebral oxygen
delivery that may be due to decreased oxygen delivery
(systemic hypoxia, anemia), decreased CBF (hypotension,
raised ICP, excessive hypocapnia or vasospasm), or increased
cerebral oxygen consumption (seizures, hyperthermia, pain)
– Increased SjvO2: This is somewhat more difficult to
interpret, and may represent either hyperemia (e.g., when
the autoregulation mechanisms are lost) or reduced oxygen
consumption (e.g., hypothermia, deep sedation, or cerebral
infarction).
– Lactate Oxygen Index: During cerebral hypoperfusion the
brain can become a net producer of lactate, with the jugular
venous lactate rising above arterial values.
Brain Tissue Oximetry
Interest in measuring brain tissue oxygenation via implantable
sensors has grown in recent years. The Licox sensor is an
implantable polarographic electrode that measures tissue
oxygen tensions. It is inserted through a compatible bolt and
ideally should be placed into the penumbral area of the injury.
Oxygen diffuses from the tissue through the catheter into an

electrolyte chamber where an electrical current is generated.
Brain tissue oxygen tension is normally lower than arterial
oxygen tension (15-50 mmHg); whilst tissue CO2 is normally
higher (range 40-70 mmHg). The sensors are useful in
monitoring local changes and trends in tissue oxygenation that
might be missed by SjvO2 measurements.
At present it is primarily used in severe head injury and poorgrade
subarachnoid hemorrhage, and in conjunction with other
monitoring modalities. The technique allows a continuous
method of monitoring of regional tissue oxygenation and in
particular, monitoring areas of high ischemic risk, and is a
promising and reliable clinical tool.
Direct measures of CBF:
Measurement of injectable tracers that reach the brain after
peripheral injection using signal intensity changes.
Xenon-Enhanced CT:
In this technique xenon, a diffusible agent is used. The patient
inhales a mixture of 28% xenon and 72% oxygen for
approximately 4 minutes after baseline CT scans are obtained.
Sequential scanning of the same slices occurs during the
inhalation period. Tissue attenuation vs. time data is then
obtained. The arterial concentration is proportional to the
expired xenon concentration. Advantages are the relatively low
cost and high ease of use. The downside is a high sensitivity to
motion artifact (Andrew 2010).
SPECT (Single Photo Emission CT):
In SPECT scanning, the radioisotope technetium-99m (Tc-99m)
is combined with hexamethylpropyleneamine (HMPAO) or ethyl
cysteinate dimer.
SPECT scanning has the advantage of being easy to perform.
However, there are limitations with regards to assembly of the
compound.
MR Perfusion: “Dynamic Susceptibility Contrast Imaging” also
called “first-pass” or “bolus tracking” MR perfusion imaging is
based on rapid acquisition of MR signal intensity data from the
brain during the injection of a contrast agent. Signal intensity

Noninvasive Monitoring
Continuous measures of CBF by Transcranial Doppler
Transcranial Doppler (TCD) is a noninvasive technique that
calculates blood flow velocity in the cerebral vasculature. An
ultrasound beam is reflected back by the moving bloodstream at
a different frequency than it was transmitted (Doppler shift),
and from the Doppler equation, the velocity of blood flow (FV)
can be calculated. Changes in FV correlate well with changes in
CBF, as long as the orientation of the transducer and the vessel
diameter remain constant. It is used clinically to diagnose
vasospasm, to test cerebral autoregulation, and to detect emboli
during cardiac surgery and carotid endarterectomy (Moppett
2004).
Normal Values. From the FV waveform systolic, diastolic, and
mean velocities can be calculated. The mean FV in the middle
cerebral artery (MCA) is usually 35-90 cm/s and correlates well
with CBF. The FV can be influenced by age, being lowest at birth
(24 cm/s), highest at age 4-6 years (100 cm/s), and then declining
until the seventh decade of life (40 cm/s). FV is also 3-5% higher
in females and increases in hemodilutional states.
Technique for Insonating the Middle Cerebral Artery
(MCA): The M1 branch of the MCA is the commonest vessel to be
insonated, and is visualized through a transtemporal window
with a 2 MHz pulsed Doppler signal. The anterior and posterior
cerebral arteries can also be accessed through this window,
whilst a transorbital approach allows access to the carotid
siphon and the suboccipital route to the basilar and vertebral
arteries.

Analysis of Doppler waveform: Analysis of the Doppler
waveform gives rise to useful derived variables as well as blood
velocity information.
– Pulsatility Index (PI): FVsys-FVdias/FVmean (normal value:
0.6-1.1). This reflects distal cerebrovascular resistance and
correlates with CPP.
– Change in CBF with arterial CO2 tension (cerebral vascular
reactivity).
Uses of TCD in Intensive Care Head Injury: Three distinct
phases have been shown in severe head injury with regard to
CBF and MCA FV.
– Phase 1 occurs on the day of injury and has a normal CBF,
normal MCA FV, and normal AVDO2.
– Phase 2 occurring 1-2 days post-injury, a hyperemic state is
encountered with an increased CBF, MCA FV and decreased
AVDO2.
– The final phase seen at days 4-15 is the vasospastic phase
and is associated with a significantly decreased CBF and
increased MCA FV. The use of TCD allows interpretation of
the dynamic physiological changes seen in severe head
injury, and in combination with other modalities allows
perfusion and oxygenation to be optimized for the
individual patient.
The highest MCA FV recorded at any stage is an independent
predictor of outcome from head injury, and the loss of
autoregulation (calculated by regression of CPP on MCA FV) has
also been shown to be a predictor of poor outcome from head
injury.
Subarachnoid Hemorrhage: Vasospasm occurs in
approximately 50% of people with subarachnoid hemorrhage
between 2-17 days post-event, and is associated with significant
morbidity and mortality. TCD may be used to detect vasospasm
by the increase in MCA FV associated with vessel narrowing.
Spasm is also assumed to be occurring when blood velocity is

>120 cm/s. High MCA FV is associated with worse-grade SAH,
larger blood loads on CT (assessed by Fisher Grade) and hence
worse outcome (Steiger 1994).
Near Infrared Spectroscopy
While the criticism of jugular venous oximetry is that it is
representative of global oxygen delivery, near infrared
spectroscopy (NIRS) is a noninvasive technique that measures
regional cerebral oxygenation.
Light in the near infrared wavelength (700-1,000 nm) can pass
through bone, skin, and other tissues with minimal absorption,
but is partly scattered and partly absorbed by brain tissue. The
amount of light absorbed is proportional to the concentration of
chromophobes (iron in hemoglobin, and copper in cytochromes),
and measurement of absorption at a number of wavelengths
provides an estimate of oxygenation (Owen-Reece 1999).
The probes illuminate a volume of about 8-10 ml of tissue and
are ideally suited for use in neonates because of their thin skull,
but have been used with success in adults.
Advantages of this technique are that it is non-invasive, and
provides a regional indicator of cerebral oxygenation. Its major
limitation is its inability to distinguish between intra- and extracranial
changes in blood flow.
Electrophysiological Monitoring
An electroencephalogram (EEG) is obtained using the
standardized system of electrode placement. Practically, this is
not often readily available and requires expert interpretation.
The EEG is affected by anesthetic agents and physiological
abnormalities such as hypoxia, hypoperfusion and hypercarbia.
A number of methods have been developed to simplify and
summarize the EEG data:
– Cerebral Function Monitor (CFM): This is a modified device
from a conventional EEG. It uses a single biparietal or

bitemporal lead, and is processed to give an overall
representation of average cortical activity.
– Cerebral function analyzing monitor: Developed from the
CFM but displays information about amplitude and
frequency separately.
– Bispectral Analysis: This modification of the EEG analyzes
the phase and power between any two EEG frequencies. The
bispectral index (BIS) is a dimensionless number statistically
derived from these phased and power frequencies and
ranges from 0 to 100 (100-awake, 60-unconscious, 0-
isoelectric EEG). This technology was derived with normal
subjects and is not readily transferable to the injured brain,
but may have a use in guiding sedation and analgesia.
Spectral Edge Frequency: Compressed Spectral Array: Raw
EEG data is processed into a number of sine waves (Fourier
analysis). Power spectral analysis then investigates the
relationship between power and frequency of the sine waves
over a short time period (Epoch). The compressed spectral array
is obtained by superimposing linear plots of successive epochs to
produce a three-dimensional “hill and valley” plot. The spectral
edge frequency looks at the frequency below which a determined
power of the total power spectrum occurs. SEF90 indicates a
spectral edge frequency of 90% and is the frequency below which
90% of activity is occurring.
Continuous Electroencephalogram Monitoring: When a
comatose, critically ill patient arrives in the intensive care unit
(ICU), he is connected to a pulse oximetry monitor, ECG monitor,
respiration monitor, arterial blood pressure monitor, etc, to
provide physicians and nurses with real-time information about
cardiopulmonary physiology. Monitoring for the brain has been
unavailable to ICU staff until recently.
In comatose or sedated patients, there may be too few
examination findings that can be reliably followed to assess
worsening brain injury. Neuroimaging cannot reveal functional
changes, such as seizures and level of sedation; it provides

information about structural brain injury often after it is
irreversible. There is great need for central nervous system
monitoring for at-risk patients, as more interventions are
becoming available to manage neurologic injury and “time is
brain”.
It is now possible to monitor and record the continuous digital
electroencephalogram, with full electrode placement, of many
critically ill patients simultaneously. Continuous EEG monitoring
(cEEG) provides real-time dynamic information about brain
function, which is especially useful when the clinical
examination is limited. Nonconvulsive seizures and
nonconvulsive status epilepticus are common in comatose
critically ill patients and can have multiple negative effects on
the injured brain.
Nonconvulsive status epilepticus (NCSE) seems to be an
important issue in stroke; NCSE is a frequent finding reaching
18% in a recent multicenter study (Kitchener 2010), thus
requiring a high degree of suspicion in an acute stroke setting to
avoid further neuronal injury and morbidity. The majority of
seizure activity in these patients cannot be detected without
cEEG. So, cEEG monitoring is mandatory to detect and guide
management of nonconvulsive status, including those that occur
following convulsive status epilepticus. In addition, it is used to
guide management of pharmacological coma sometimes used for
treatment of increased intracranial pressure. There are
emerging applications for cEEG, one of which is to detect new or
worsening brain ischemia in patients at high risk, especially
those with subarachnoid hemorrhage. As qEEG software is
continuously improving, full scalp cEEG monitoring is feasible,
and can provide continuous information about changes in brain
function in real time at the bedside and to alert clinicians to any
acute brain event, including seizures, ischemia, increasing
intracranial pressure, hemorrhage, and even systemic
abnormalities affecting the brain, such as hypoxia, hypotension,
acidosis, and others. cEEG monitoring without expert review of
the raw EEG, must not be allowed as false positives and false

negatives are common. When cEEG is combined with
individualized multimodality brain monitoring, intensivists can
identify when the brain is at risk for injury or when neuronal
injury is already occurring and intervene before there is
permanent damage. We believe that cEEG has significant
potential to improve neurologic outcomes in a variety of
settings.
Application of the EEG in the ICU:
– Seizure management: Confirms the diagnosis of seizures and
identifies a focal or lateralized source of activity. It also
helps to distinguish between involuntary movements,
posturing, and eye signs that are common in intensive care
and true seizure activity.
– Nonconvulsive status epilepticus (NCSE): This represents a
state that lasts more than 30 min with clinical evidence in
alteration in mental state from normal, and seizure activity
on the EEG. Between 4 and 20% of patients with status
epilepticus have nonconvulsive episodes. NCSE is a frequent
finding in ischemic stroke reaching 18% of ischemic stroke
patients admitted to neurocritical care units.
– Metabolic suppression: Burst suppression (isoelectric EEG) is
a definable end point when pharmacological reduction of
the cerebral metabolic rate of the injured brain is required
for either neuroprotection or intractable intracranial
hypertension.
– Ensuring adequate sedation in patients who require
prolonged neuromuscular paralysis.
– Prognosis: The EEG can be of prognostic value following
brain injury, with absence of spontaneous variability being
associated with poor outcome.

Multimodal Monitoring
In any type of brain injury, the available monitoring modalities
are prone to artifact and misinterpretation. By utilizing more
than one monitoring technique, the observer is more likely to
determine whether a genuine change in cerebral physiology has
occurred and what the most appropriate intervention should be.
For instance, in traumatic brain-injured patients we routinely
monitor ICP, processed EEG, SjvO2 and brain-tissue oxygen
tension (PbtO2), allowing us to observe both local and regional
changes in cerebral hemodynamics. General rules cannot always
be applied to individual patients, and multimodal monitoring
can allow more informed decision making such as determining
CPP thresholds or the ability of the cerebral vasculature to
autoregulate (Cecil 2011; Czosnyka 1996).
Conclusions
A wide range of monitoring techniques is available, each with
different strengths and limitations. Multimodal monitoring
using a combination of techniques can overcome some of the
limitations of the individual methods discussed. The choice of
monitoring is often guided by clinical familiarity and local
policy.
Key points:
1. Repeated clinical assessment through the Glasgow Coma
Scale (GCS) is the cornerstone of neurological evaluation.
2. Ventilated head-injured patients with intracranial
pathology on CT require ICP monitoring.
3. Invasive or non-invasive neurospecific monitoring requires
careful interpretation when assisting goal-directed
therapies.
4. Multimodal monitoring using a combination of techniques
can overcome some of the limitations of individual
methods.

See also:
Brain Trauma Foundation Guidelines 2007
www.braintrauma.org
Anaesthesia UK
www.frca.co.uk

Basic Hemodynamic Monitoring of Neurocritical Patients

" "

The importance of basic hemodynamic monitoring of
neurocritical patients comes from the goal of maintaining brain
autoregulation. Brain autoregulation and other biological signals
are the variables to be monitored by using biomedical sensors.
Complications that may occur in neurocritical patients (e.g.,
sepsis, dehydration, post-cardiac arrest status) make
hemodynamic monitoring of greater importance.
The goals of hemodynamic monitoring in neurocritical care
units are to assess the magnitude of physiological derangements
in critically ill patients and to institute measures to correct the
imbalance.
The following steps should be taken to reach these goals.
Although our review of data may be helpful, the attending
physician should decide what to use and when to use it. Menu to
work with for proper patient management:
1. Pulse oximeter (SpO2) is regarded as one of the most
important advances in critical care monitoring. SpO2
provides a continuous non-invasive method to measure
arterial oxygen saturation, and should be used on every
neurocritical patient. The absorption spectra of both

oxyhemoglobin and deoxyhemoglobin and the
characteristics of pulsatile blood can thus be determined.
SpO2 is accurate to within ± 2% for saturations >70%. SpO2 is
widely used in monitoring patients who have a variety of
neurological conditions (Adams 1997), and calculations
made from the processed signals provide estimates of the
tissue or venous and arterial blood and provide an estimate
of the amount of oxygenated hemoglobin and the percent
saturation of hemoglobin by oxygen SaO2, which is not the
same as the PaO2 (partial pressure of oxygen) in the blood
(Adams 1997). The PaO2 and SaO2 measurements of
oxygenation are related through the oxyhemoglobin
dissociation curve. Importantly, SpO2 is a measure of
arterial oxygenation saturation, not arterial oxygen tension
(PaO2). Given the characteristics of the oxygen dissociation
curve, large fluctuations in PaO2 can occur despite minimal
changes in SpO2. In addition to its inability to measure PaO2,
SpO2 provides no measure of ventilation or acid-base status.
Therefore, it cannot be used to determine pH or arterial
carbon dioxide tension. Significant increases in arterial
carbon dioxide can occur with normal readings in SpO2.
Although useful for arterial oxygen saturation, SpO2 should
not be assumed to provide information about ventilation.
Studies have shown that to assure a saturation of 60 torr
(8.0 kPa), an SpO2 of 92% should be maintained in patients
with light skin, whereas 94% saturation may be needed in
patients with dark skin. Oxygenation is considered
adequate if the arterial oxygen saturation is above 95%. The
majority of these patients are placed on positive end
expiratory pressure (PEEP) at 5cm H2O (Curley 1990).
Also, for patients with manifestations consistent with
hypoxemia (e.g., tachycardia, hypotension, anxiety, and
agitation) there is a time delay in pulse oximeter to express
fluctuation in SpO2. Again in hypothermia, low CO2, and
vasoconstriction secondary to drugs or peripheral hypoxia

all increase bias, imprecision, and response time for
hypoxic episodes, so we proceed to the next step.
2. Arterial blood gas analysis is widely available in hospitals
and offers direct measurements of many critical
parameters (pH, PaO2, PaCO2). Arterial blood gas analysis is
among the most precise measurements of oxygen tension
and pressure that will reflect tissue oxygenation (García
2011).
3. Non-invasive automated blood pressure devices are
frequently used to obtain non-invasive, intermittent blood
pressure measurements. Measurements of systolic and
diastolic pressure to calculate the mean arterial pressure
(MAP) is mandatory to calculate the cerebral perfusion
pressure. These devices are less accurate in critically ill
patients as well as in those with secondary brain injury.
These less accurate readings can distract the attention of
the caregiver. The evaluation of blood pressure will be
significantly affected by the use of vasopressors. Therefore,
the numeric reading may reflect vasoconstriction in spite
of decreasing perfusion with adequate blood pressure.
4. Invasive blood pressure monitoring for continuous
monitoring and recording of the arterial pressure via an
arterial catheter is preferable to the use of an automated
blood pressure device in hemodynamically unstable
patients (García 2011). The radial artery is most commonly
cannulated because of its superficial location, relative ease
of cannulation, and in most patients, adequate collateral
flow from the ulnar artery. Other potential sites for
percutaneous arterial cannulation include the femoral,
brachial, axillary, ulnar, dorsalis pedis, and posterior tibial
arteries. Possible complications of intra-arterial
monitoring include hematoma, neurologic injury, arterial
embolization, limb ischemia, infection, and inadvertent
intra-arterial injection of drugs. Intra-arterial catheters are
not placed in extremities with potential vascular
insufficiency. However, with proper selection, the

complication rate associated with intra-arterial
cannulation is low and the benefits can be important.
5. A Foley catheter, for monitoring of urine output on an
hourly basis or for 24 hours, is a simple and important tool
to monitor volume status of the patient besides renal
perfusion and function. The hourly urine output is a cheap,
simple and indirect method of assessing adequacy of
cardiac output and tissue perfusion.
6. A temperature probe is also indicated for purposes of
monitoring core temperature.
7. Continuous monitoring of volume status and other
parameters:
If the patient is hypotensive, a fluid supplement with 1-2
liters Ringer’s lactate for 30-60 minutes is reasonable.
Subsequent fluid management should be based upon urine
output and maintaining central venous pressure between 5
and 10 mmHg. Monitor potassium, sodium, glucose and
arterial blood gases every 4 hours. Especially if there is
respiratory embarrassment at initial evaluation,
measurement of hematocrit, magnesium, blood urea
nitrogen, creatinine, calcium, liver function tests, urine
analysis, prothrombin, partial thromboplastin times and
phosphate levels are mandatory. If hematocrit is below
30%, transfuse cross-matched blood (Amin 1993).
8. Common indications for central venous cannulation:
measurement of mean central venous pressure, large bore
venous access, administration of irritating drugs and or
parenteral nutrition, hemodialysis, placement of a
pulmonary artery catheter.
Placement of a pulmonary artery catheter is indicated to
obtain direct and calculated hemodynamic data that cannot
be obtained through less invasive means (Sakr 2005).
The goal for all critically ill patients is to provide
adequate oxygen for cellular use through suitable oxygen
consumption, which is variable between tissues, and the
changes of the basal or active metabolic rate for each cell.

Oxygen delivery to tissues and organs responds to many
local systemic variables to keep cellular homeostasis.
Pulmonary artery balloon flotation catheter insertion may
be necessary for full assessment of these parameters, and of
evaluation of determinants of cardiac output and the
oxygen content of circulating blood, on which oxygen
delivery is dependent. In the presence of positive pressure
ventilation with PEEP, central venous and pulmonary
artery occlusion pressures may be falsely elevated and need
to be interpreted with caution. The fluid challenge is the
only way to interpret Central Venous Pressure (CVP) or
Pulmonary Artery Occlusion Pressure (PAOP).
Assessment of the determinants of cardiac output will
proceed as follows:
a. Heart rate and rhythm assisted by pulse oximeter and
electrocardiogram.
b. Preload assessed right and left heart; right heart
through neck vein distension, liver enlargement and
central venous pressure assessment; left heart through
dyspnea on exertion, orthopnea, arterial blood
pressure; pulmonary artery occlusion pressure and
arterial pressure through waveform analysis (Sakr
2005).
c. Afterload assisted by mean arterial blood pressure and
systemic vascular resistance; contractility can be
assessed by ejection fraction and echocardiography.
As discussed earlier, the goal of hemodynamic monitoring in
neurocritical care units is to assess the magnitude of
physiological derangements in critically ill patients and to
institute measures to correct the imbalance. Basic hemodynamic
monitoring consists of clinical examination, invasive arterial
monitoring, central venous pressure monitoring, hourly urine
output, central venous oxygen saturation and echocardiography.
Dynamic indices of fluid responsiveness such as the pulse
pressure variation and stroke volume variation can guide

decision making for fluid resuscitation. Cardiac output is
traditionally measured using the pulmonary artery catheter; less
invasive methods now available include the pulse contour
analysis and arterial pulse pressure derived methods. It is
essential to determine whether the hemodynamic therapy is
resulting in an adequate supply of oxygen to the tissues
proportionate to their demand. Mixed and central venous
oxygen saturation and lactate levels are commonly used to
determine the balance between oxygen supply and demand
(Walley 2011).