EFFECTIVENESS OF RESUSCITATION IN TRAUMATIC
BRAIN INJURY WITH SYSTEMIC ACIDOSIS
Laurie Greer Munro
B.S., University of Colorado at Denver, 1992
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
This thesis for the Master of Arts
Laurie Greer Munro
has been approved
Munro, Laurie (M.A., Biology)
Effectiveness of Resuscitation in Traumatic Brain Injury
with Systemic Acidosis
Thesis directed by Professor Teresa Audesirk
Severe traumatic brain injury is one of the most serious problems
facing modern medicine. Brain injury has the highest rate of mortality in
the traumatic population, and trauma is the most common cause of
mortality in people under 45 years of age. Currently, much remains
unknown concerning the mechanisms involved in brain injury and the
most appropriate ways to treat a brain injured patient. Clinical research
suggests that following brain injury, systemic respiratory and metabolic
acidosis occurs. This secondary injury of acidosis may irreversibly
damage viable brain cells, and has been correlated with poor clinical
outcome. Because bicarbonate appears to be effective in managing
intracranial pressure (ICP) based on clinical observations, this study
was proposed to support or dispute this capability under controlled
experimental conditions and to determine whether the effect is due to
buffering mechanisms or osmotic mechanisms. The results
demonstrate that the osmotic agent properties of both the bicarbonate
and the 6 % NaCI effectively managed ICP. Both groups succeeded in
maintaining pretreatment ICP levels for almost four hours post
treatment. The control group, however, maintained pretreatment ICP
levels for only 45 minutes, and then climbed to a mean peak of 44.75
3.50 mmHg after four hours of treatment. Normal ICP is under 15
While there was a significant difference between the control and
bicarbonate (p < 0.001) and control and 6% NaCI group (p = 0.029),
there was no significant difference between the bicarbonate and 6%
NaCI groups (p = 0.686). These results strongly suggest the influence
of the osmotic agent properties of both the bicarbonate and the 6%
NaCI on controlling ICP.
Therefore, based on the I CP level durations, there does not
appear to be an obvious benefit from the bicarbonate buffering system
on the regulation of elevations of ICP. The effect of controlling ICP
appears to be solely a result of the hyperosmotic property of
This abstract accurately represents the content of the candidate's
thesis. I recommend its publication.
1.1 Background Physiology..........................................2
1.1.1 Regulation of Acid-Base Balance.............................2
184.108.40.206 Chemical Acid-Base Buffer Systems..........................4
220.127.116.11 Respiratory Removal of C02.................................5
18.104.22.168 Renal Control of Acid-Base Balance.........................6
22.214.171.124 , Metabolic Acidosis.......................................7
126.96.36.199 Respiratory Acidosis.......................................9
1.1.3 Cerebral Autoregulation....................................10
1.1.4 Blood-Brain Barriers.......................................11
1.2 Problem Statement.........................................12
1.3 Previous Research.............................................13
1.4 Basis of the Experimental Protocol...........................19
1.4.1 Brain Injury with Systemic Acidosis Model.....................19
188.8.131.52 Brain Injury................................................19
184.108.40.206 Brain Edema.................................................20
220.127.116.11 Experimental Groups.........................................24
18.104.22.168.1 Bicarbonate Group.........................................24
22.214.171.124.2 6% Saline Group...........................................24
126.96.36.199.3 Control Group.............................................25
188.8.131.52 Intracranial Pressure.......................................27
184.108.40.206 Arterial pH and Blood Gases.................................29
220.127.116.11 Vital Signs.................................................29
2.1 Methods Overview...............................................30
2.2 Experimental Set-Up Surgical Procedures........................31
2.3 Experimental Groups............................................34
2.4 Measurement of Brain Injury....................................36
2.5 Exclusion Criteria.............................................36
2.6 Data Analysis.................................................37
3.1 Intracranial Pressure..........................................39
3.4 Base Excess...................................................47
3.5 Mean Arterial Blood Pressure..................................49
3.6 Carbon Dioxide................................................51
3.7 Lesion Diameter...............................................53
4.1 Management of I CP.............................................55
4.2 Volume Control................................................56
4.3 Osmotic Agents................................................57
4.4 Buffer Capabilities...........................................60
4.5 Metabolic Acidosis............................................65
4.6 Future Research...............................................66
4.6.1 Laboratory Studies..........................................66
18.104.22.168 Corrections to the Current Study...........................66
22.214.171.124 Future Laboratory Studies..................................69
4.6.2 Clinical Studies
126.96.36.199 Retrospective Studies.....................................71
188.8.131.52 Prospective Studies.......................................72
Severe traumatic brain injury (TBI) is one of the most serious
problems facing modern medicine. Brain injury has the highest rate of
mortality in the traumatic population, and trauma is the most common
cause of mortality in people younger than 45 years of age (Chesnut and
Marshall, 1993). As a result of an organized and comprehensive
treatment approach which begins at the scene of the accident and is
maintained throughout, the mortality of brain injury has been halved, to
approximately 35% since the 1970's. (Chesnut and Marshall, 1993).
Yet currently, much remains unknown concerning the mechanisms
involved in brain injury and the most appropriate ways to treat a brain
Severe traumatic brain injury involves a complex series of
damaging events, which is further compounded by the deleterious
effects of secondary injury, occurring after the initial insult (Marmarou,
1992). Current clinical research has shown that following brain injury,
systemic respiratory and metabolic acidosis occurs. This secondary
injury of acidosis may result in changes that irreversibly damage viable
brain cells, and has been correlated with poor clinical outcome (Rabow
et al, 1986). Therefore, it is critical to address the issue of correcting
1.1 Background Physiology
1.1.1 Regulation of Systemic Acid-Base Balance
To achieve homeostasis, the intake or production of hydrogen
ions must be balanced with the net removal of hydrogen ions from the
body. The activity of almost all enzyme systems in the body are
influenced by hydrogen ion concentration, and any change in the
hydrogen concentration can alter virtually all cell and body functions
(Guyton and Hall, 1996).
Strong acids, such as lactic acid, rapidly dissociate and release
large amounts of hydrogen ion in solution, while weak acids, such as
carbonic acid (^COJ, have less tendency to dissociate. Strong bases,
such as the hydroxyl ion (OH ), rapidly remove hydrogen ions from
solution, while weak bases, such as the bicarbonate ion, (HC03), bind
hydrogen ions more weakly (Guyton and Hall, 1996).
Hydrogen ion concentration is expressed in equivalents per liter
using a logarithmic scale of pH units, which is inversely proportional to
the hydrogen ion concentration.
Equation 1.1 pH = log _1_________= log [H*]
When the hydrogen ion concentration increases above normal
values, the pH is lowered and the body is considered to be in an
acidotic state. Three primary systems regulate the hydrogen ion
concentration in body fluids to prevent acidosis. The first line of
defense is the chemical acid-base buffer systems of the body fluids.
This system neutralizes any excess acid or base to prevent hydrogen
ion concentration changes. The second system, the respiratory center,
regulates the removal of C02 and H2C03 from the extracellular fluid.
The third system, the kidneys, excrete acid urine to readjust the
extracellular fluid concentration toward normal (Guyton and Hall, 1996).
184.108.40.206 Chemical Acid-Base Buffer Systems
The bicarbonate buffer system consists of a water solution
containing carbonic acid (l-^COg), which acts as a weak acid, and
sodium bicarbonate (NaHCOg), which acts as a weak base. The
enzyme carbonic anhydrase, which catalyzes the reaction between
carbonic acid and carbon dioxide (COJ with water (H20), is abundant in
the walls of the lung alveoli where C02 is released and in the epithelial
cells of the renal tubules, where the reaction of C02 with water to form
carbonic acid occurs. Sodium bicarbonate is present in the extracellular
fluid. Equation 1.2 describes this buffer system.
Equation 1.2 C02 + H20 H2C03 H+ + HC03*
When the concentration of a strong acid, such as lactic acid,
increases in the body fluids, the increased hydrogen ions released from
the lactic acid are buffered by the bicarbonate ion as shown in the
t H+ + HC03- H2C03 -* C02 + H20
This results in increased carbon dioxide and water production
due to greater formation of carbonic acid. The excess carbon dioxide
causes increased respiration to expire the excess C02, which results in
a normal concentration of carbon dioxide in the extracellular fluid
(Guyton and Hall, 1996).
220.127.116.11 Respiratory Removal of C02
The lungs act as the second line of defense against acid-base
imbalances by controlling the concentration of carbon dioxide in the
extracellular fluid. Increased ventilation eliminates C02 from the
extracellular fluid, resulting in an increase in pH. Under normal
conditions, the pulmonary expiration of C02 balances the metabolic
formation of C02. Carbon dioxide, formed as a by-product of cellular
respiration, diffuses from cells into the interstitial fluids and then into
blood where it is transported to the lungs. In the lungs, it diffuses into
the alveoli and is removed to the atmosphere by pulmonary ventilation.
The respiratory system has between one and two times the
buffering power of the combined chemical buffer mechanisms in the
extracellular fluid (Guyton and Hall, 1996). The fast acting respiratory
system keeps the hydrogen ion concentration from changing too much
until the slower responding kidneys begin to eliminate the imbalance.
18.104.22.168 Renal Control of Acid-Base Balance
The kidneys excrete acid urine to control acidosis. The kidneys
use three basic mechanisms to control the extracellular concentration of
hydrogen ions. First, acid is removed from the blood by the secretion
of large numbers of hydrogen ions into the tubular lumen by epithelial
cells. There is a net loss of acid from the extracellular fluids when
more hydrogen ions are secreted than bicarbonate ions are filtered.
Second, the kidneys prevent loss of bicarbonate in the urine by
reabsorption of filtered bicarbonate in the tubule. Third, the kidneys
produce new bicarbonate which is added to the extracellular fluid
(Guyton and Hall, 1996).
22.214.171.124 Metabolic Acidosis
In metabolic acidosis, the ratio of the bicarbonate ion
concentration to C02 concentration falls resulting in depressed pH.
This relationship is described by the Henderson-Hasselbalch equation
(Equation 1.4), which gives the pH resulting from the solution of C02 in
the blood and the consequent dissociation of carbonic acid.
Equation 1.4 pH = pKa + log fHCCyi
The bicarbonate ion concentration may be lowered by the
accumulation of acids in the blood, for example, by the lactic acid
released during tissue hypoxia. Respiratory compensation involves
lowering the resulting pC02 by an increase in ventilation which raises
the depressed [HC03] / [C02 ] ratio. As a result, there is a negative
base excess (West, 1990). Base excess is an empirical expression for
the approximate quantity of acid or base required to titrate one liter of
blood to obtain a normal blood pH level of 7.4 (Ciba Corning
Diagnostics Corp., 1982). Equation 1.5 gives the formula used by the
170 Ciba Corning Blood Gas Analyzer in calculating the base excess.
Hemoglobin (Hb) is the oxygen carrying pigment of red blood cells.
B.E. = (1- 0.014 Hb) ([HC03-] 24) + (9.5 + 1.63 Hb)(pH 7.4)
oBase Excess as a Measure of Renal Compensation and an
Indicator of Acid-Base Imbalances
Acid-Base Condition Base Excess
Metabolic Acidosis < 0
Respiratory Acidosis > 0
Metabolic Alkalosis < 0
Respiratory Alkalosis > 0
Base excess is a measure of renal compensation. As shown in
Table 1.1 a negative base excess (BE < 0) indicates possible metabolic
acidosis or respiratory alkalosis. A positive base excess (BE > 0)
indicates respiratory acidosis or metabolic alkalosis. If base excess
equals zero, a neutral state exists (West, 1990).
The expected values for metabolic acidosis would be a low pH, a
low plasma bicarbonate concentration, and a reduction in pC02 after
partial respiratory compensation (Guyton and Hall, 1996).
126.96.36.199 Respiratory Acidosis
Respiratory acidosis is caused by an increase in pC02 which
reduces the HC03/pC02 ratio, resulting in lower pH. When the pC02
rises, the bicarbonate must also increase because of the dissociation of
the produced carbonic acid. Respiratory acidosis may result from
hypoventilation (West, 1990).
If this condition persists, the kidney is prompted to conserve
bicarbonate triggered by the increased pC02 in the renal tubule cells.
To compensate for the increased hydrogen ions, the kidneys secrete a
more acid urine composed of H2P04- and NH4+. Bicarbonate ions are
reabsorbed into the tubule. As a result, plasma bicarbonate levels
increase, thus increasing the ratio of [HC03]/pC02 towards normal
Normal pH levels of 7.4 are rarely restored completely as a result
of this compensation mechanism of the renal system. The base excess
determines the extent of the renal compensation. The expected
values for respiratory acidosis would be a reduced pH, increased pC02,
and increased plasma bicarbonate concentration after partial renal
compensation (Guyton and Hall, 1996).
1.1.3 Cerebral Autoregulation
Cerebral vessels can respond to altered physiological conditions
in a unique fashion and have the ability to alter vessel diameter
(Kandel, Schwartz and Jessell, 1991). There are two types of
autoregulation. In the first type, the brain arterioles constrict when the
systemic blood pressure is increased and dilate when the blood
pressure is decreased. These adjustments help maintain optimal blood
flow resulting in constant cerebral blood flow (CBF) for mean arterial
pressures between 60 and 150 mmHg under normal conditions. When
mean arterial pressure is below 60 or above 150 mmHg, the CBF
changes linearly with pressure. In the second type, when arterial
pC02 is increased (hypercarbia), the brain arterioles dilate and CBF
increases. When pC02 is decreased (hypocarbia), brain arterioles
vasoconstrict and CBF decreases (Kandel et al, 1991).
1.1.4 Blood-Brain Barriers
The blood-cerebrospinal fluid barrier exists between the
blood and the CSF and the blood-brain barrier exists between the blood
and the brain fluid both at the choroid plexus and at the tissue capillary
membranes in almost all areas of brain parenchyma. The barriers are
not present in some areas of the hypothalamus, pineal gland, and area
postrema, because substances diffuse easily into the tissue spaces in
these areas. These areas of the brain have sensory receptors that
respond to changes in the body fluids, such as osmolality, thus
providing the signals for feedback regulation (Guyton and Hall, 1996).
The barriers are highly permeable to water, carbon dioxide,
oxygen, and most lipid-soluble substances. They are slightly permeable
to sodium, potassium and chloride. They are impermeable to plasma
proteins and most non-lipid-soluble large organic molecules. The low
permeability of the barriers is due to the tight junctions formed between
the membranes of adjacent endothelial cells (Guyton and Hall, 1996).
1.2 Problem Statement
The purpose of this study was to investigate the therapeutic
benefits of bicarbonate on the treatment of head injury. Since
bicarbonate is both a buffering agent and an osmotic agent, this study
proposed to investigate the efficacy of using bicarbonate to manage
traumatic head injury. Specifically, the ability of bicarbonate as a
therapeutic agent to maintain low intracranial pressure, either through
the elimination of the systemic acidosis normally associated with head
injury or through elevation of the serum osmolality to reduce intracranial
edema, was evaluated.
1.3 Previous Research
The control of intracranial hypertension is essential to the
management of the brain injured patient (Cooper, 1993). Continuous
monitoring of intracranial pressure (ICP), considered to be accurate and
relatively safe, is currently the most useful real-time diagnostic tool
currently used to evaluate brain injury. The effects of elevated
intracranial pressure are the most frequent cause of death in head-
injured patients (Becker et al, 1977; Marshall et al, 1979). Intracranial
hypertension develops in approximately 40% of traumatic brain injured
patients (Miller et al, 1977) and plays a significant role in the mortality
of 50% of brain injured patient deaths (Marshall et al, 1979). The
level of ICP has been demonstrated to be a strong predictor of
outcome, based on analysis of the data from the Trauma Coma Data
Bank (TCDB) (Marmarou, 1990).
Another strong predictor of outcome is central nervous system
(CNS) lactic acidosis (DeSalles et al, 1986; Rabow et al, 1986;
Rosner et al, 1984) which results from severe injury to the brain.
Rosner et al (1986) demonstrated that sustained high concentrations of
ventricular lactic acidosis occurring in patients with poor outcomes
indicated that a persistent dysfunction in brain metabolism must be
present following severe brain injury. These acid-imbalanced patients
were clinically difficult to manage and developed severe intracranial
hypertension. The cause of increased lactate levels in the CSF
following brain injury are still under debate (Rabow et al, 1986;
DeSalles et al, 1986; Siesjo et al, 1993). One possible mechanism,
however, is that the initial high lactate levels result from the mechanical
injury, which result in lack of oxygen availability due to edema
compressing blood vessels or trauma damaging blood vessels. As a
result of the traumatized brain tissues and capillaries, capillary fluid
leaks into the traumatized tissues. Once brain edema is initiated, two
vicious cycles begin because of positive feedback mechanisms. In one
feedback mechanism, edema compresses the vessels, which decreases
blood flow and causes brain ischemia. The ischemia causes dilation of
the arteries, resulting in further increases in capillary pressure, which
results in yet more edematous fluid. This cycle results in a progression
of brain edema. In the second positive feedback mechanism, oxygen
delivery to the brain is decreased as a result of decreased cerebral
blood flow, which increases capillary permeability, thus allowing more
fluid leakage. The decreased blood flow also turns off the sodium
pumps of the brain tissue cells and as a result the cells swell (Guyton
and Hall, 1996).
Systemic metabolic acidosis is also known to accompany severe
brain injuries, especially in multiple trauma patients with associated
systemic trauma. Arterial lactic acidosis in the multi-trauma patient with
severe brain injury could possibly result from at least three potential
sources (Levy et al, unpublished). First, hypoperfusion of tissues will
result in elevated lactate levels and metabolic acidosis if hypotension is
present, either from hypovolemic shock or as a direct result of
neurologic injury. Second, many patients with severe brain injury and
multiple trauma experience hypoxia prior to airway management, which
results in accumulation of lactate due to decreased tissue oxygenation.
Third, a catecholamine surge following primary brain injury occurs as a
result of activation of the sympathetic nervous system and adrenal
glands which leads to increased blood lactate levels.
The time course of lactic acid production and clearance in brain
tissue, CSF, and serum following experimental brain injury was studied
by Inao, et al. (1988). Arterial and CSF lactate levels were found to
peak within 15 minutes post injury, decreased toward control values
during the next two hours and then subsequently increased gradually
from four to eight hours following injury. The brain tissue lactate,
however, peaked two hours post injury, and then gradually decreased.
Inao et al. concluded that the catecholamine surge induced from brain
injury is responsible for the immediate arterial and CSF lactate level
increase, while the secondary rise of lactate is due to the slow seepage
of lactate produced in brain tissue at the injury site. These results
suggest that systemic lactic acid does enter the CNS and that systemic
metabolic acidosis may have direct effects on CNS acid-base balance
following brain injury at the injury site (Inao, 1988; Marmarou, 1993).
Previous research also indicates that systemic acidosis increases the
ability of lactic acid to enter the brain (Rosner and Becker, 1984).
Current studies evaluating the relationship between acidosis and
brain injury have focused primarily on the influence of CNS acidosis.
As a result, the effect of systemic metabolic acidosis on I CP following
brain injury and the importance of correcting a systemic lactic acidosis
in the management of intracranial hypertension remains under-
represented in the literature. Preliminary clinical observations (UCHSC
Neurosurgery Division) suggest that multitrauma patients with severe
head injuries and accompanying systemic lactic acidosis develop
intracranial hypertension which is very difficult to control. Intravenously
administered NaHC03 to correct the metabolic acidosis, in conjunction
with hyperventilation to prevent ongoing lactic acid production, appears
to allow better control of elevations in ICP. A pilot study performed in
the UCHSC laboratory (presented at the 1994 American Association of
Neurological Surgeons) supported the correction of systemic lactic
acidosis by the administration of NaHCOa. Bicarbonate was found to
significantly improve the efficacy and duration of the action of mannitol
and hyperventilation in the treatment of cold lesion induced intracranial
hypertension (Levy et al, 1994).
Osmotic diuretics act to reverse the blood-brain osmotic gradient,
thereby reducing extracellular fluid volume in normal and damaged
brain. They also lower blood viscosity, resulting in reflex
vasoconstriction and reduced ICP, which is why osmotics are given as
a bolus, a concentrated mass, rather than constant infusion (Ropper,
1993; Cooper, 1993). Most clinical protocols for the treatment of
elevated intracranial hypertension recommend the use of mannitol as
an osmotic diuretic.
Elevation in ICP and systemic acidosis are damaging conditions
that often result from head injury. Bicarbonate should have both an
osmotic effect, by reducing fluid in the brain, and a buffering effect, by
restoring systemic acid-base balance. The current study is based on
current literature and clinical experience. While bicarbonate has
( clinically been demonstrated to be effective in managing ICP, this
study was proposed to support or dispute this capability under
controlled experimental conditions and to determine whether its effect is
due to buffering mechanisms in systemic circulation or the ability to act
as an osmotic agent in the brain.
1.4 Basis of the Experimental Protocol
1.4.1. Brain Injury with Systemic Acidosis Model
188.8.131.52 Brain Injury
This study was designed to simulate systemic metabolic acidosis
in a model of brain injury. The brain injury model is the standard cold
lesion model first developed by Clasen (1953). It involves freezing a
small section of brain tissue. This injury most closely models vasogenic
brain edema (Tominaga and Ohnishi, 1995). The model was chosen
in part because the injury is consistent and reproducible from subject to
subject. Mongrel dogs were chosen based on the existence of a large
data bank of previous research which used the cold lesion dog model.
In addition, the dog brain is structurally comparable to human brain,
and is of sufficient size to accommodate the monitoring equipment.
184.108.40.206 Brain Edema
As a result of the traumatized brain tissues and capillaries,
capillary fluid leaks into the traumatized tissues. Once brain edema is
initiated, two vicious cycles begin because of positive feedback
mechanisms. In one feedback mechanism, edema compresses the
vessels which decreases blood flow and causes brain ischemia. The
ischemia causes dilation of the arteries, resulting in further increases in
capillary pressure, which results in yet more edematous fluid. This
cycle results in a progression of brain edema. In the second positive
feedback mechanism, oxygen delivery to the brain is decreased as a
result of decreased cerebral blood flow which increases capillary
permeability thus allowing more fluid leakage. The decreased blood
flow also turns off the sodium pumps of the brain tissue cells and as a
result the cells swell (Guyton and Hall, 1996).
The sodium pumps play an essential role in brain metabolism.
When a neuron conducts an action potential, the ions move through the
membranes, thus increasing the need for membrane transport to
restore baseline ionic concentrations. Most neural activity depends on
the immediate delivery of oxygen and glucose from the blood, since
little glycogen or oxygen is stored in brain tissues. Therefore, a sudden
cessation of blood flow to the brain or lack of oxygen in the blood
results in reduced production of ATP due to lack of available glycogen
and oxygen. (Guyton and Hall, 1996).
To create and maintain an acidotic state, lactic acid was
continuously infused at a rate which lowered the arterial pH to 7.25 +
.03 (initially established in each dog prior to treatment). This pH level
was chosen based on pilot studies that defined a pH level sufficiently
acidotic to simulate acidosis in a patient while not severe enough to
increase the risk of mortality prior to completion of the experiment.
Lactic acid was chosen as the acid because it is a natural by-product of
anaerobic glycolysis and is believed to be the major contributor in
Table 1.2 is an acid-base imbalance table based on information
in Guyton and Hall (1996) of arterial blood pH, arterial plasma HC03-
and pC02 values for normal and acid-base imbalances. The
relationship between pH and the bicarbonate-carbonic acid buffer
system in plasma is expressed by the Henderson-Hasselbach equation,
pH= 6.1 + log rHCO?-1
The normal ratio of salt (bicarbonate) to acid (carbonic acid) is
20:1, producing an arterial blood pH of 7.4. As the bicarbonate
decreases in relation to carbonic acid the pH decreases, resulting in
acidosis. If the bicarbonate increases in relation to carbonic acid, the
pH increases, resulting in alkalosis. Metabolic acidosis refers to the
bicarbonate concentration in the Henderson-Hasselbach equation and
respiratory acidosis refers to the carbonic acid concentration (ALBA,
Arterial Bicarbonate Carbon Dioxide and pH Levels in Conditions
of Acid-Base Imbalance
Acid-Base Imbalance Arterial Blood pH Arterial Plasma [HC03-] (mEq/L) pC02 (mmHg)
Metabolic Acidosis < 7.4 < 24 < 40 (after partial renal compensation)
Respiratory Acidosis < 7.4 > 24 (after partial renal compensation) > 40
Metabolic Alkalosis > 7.4 > 24 > 40
Respiratory Alkalosis >7.4 < 24 < 40
Normal 7.4 24 40
220.127.116.11 Experimental Groups
18.104.22.168.1 Bicarbonate Group
This study uses hyperventilation as a short-term therapy in
conjunction with bicarbonate therapy. Bicarbonate is an osmotic agent
and also a buffering agent. As an osmotic agent, bicarbonate should
draw water out of the brain and into the blood, thus resulting in reduced
intracranial pressure. As a buffering agent, bicarbonate should help to
neutralize systemic blood pH.
22.214.171.124.2 6% Saline Group
The 6% NaCI is the study control for the osmolality. 6% NaCI is
hyperosmotic to the ECF. Therefore, it should act as an osmotic agent
and draw water out of the brain and into the blood, resulting in reduced
intracranial pressure. Previous animals studies have demonstrated that
hypertonic saline is effective in reducing I CP while maintaining
perfusion pressure (Ropper, 1993). The 6% saline concentration was
chosen for this experiment because it has the same osmolality as the
bicarbonate. Since osmotic agents are effective in reducing ICP, then
both the bicarbonate group and the 6% NaCI group should have lower
ICP values. If, however, systemic acidosis exacerbates intracranial
hypertension, the bicarbonate group should maintain a lower ICP than
the 6% NaCI group.
126.96.36.199.3 Control Group
The control group is only treated with hyperventilation. It is the
control for the osmotic agents and bicarbonate. Hyperventilation is
currently used clinically for the treatment of intracranial hypertension
(Chesnut and Marshall, 1993). While hyperventilation induces
hypocapnia and reduces CBF and ICP levels through chemical
autoregulation effects, it also has the potential to induce detrimental
changes in both general and cerebral circulation metabolic activities. A
major reduction in the arterial pC02 produces an abrupt increase in the
pH of brain extracellular fluid, because C02 quickly equilibrates across
the blood-brain barrier. This results in reduced concentrations of
carbonic acid in the brain extracellular fluid (ECF). Bicarbonate does
not cross the blood brain barrier in intact tissues and therefore the
concentration of bicarbonate does not immediately change. As a result,
the pH of brain ECF increases with acute hypocapnia regardless of
what happens to the arterial bicarbonate concentration. The
bicarbonate concentration is then reduced within hours, and the pH
resumes its normal value by 30 hours (Chesnut and Marshall, 1993).
Vasoconstriction and reduction of cerebral blood volume secondary to
increased pH are thought to be the mechanisms involved in the
reduction of ICP with acute hyperventilation, and therefore, the
beneficial effect is short-lived.
The use of hyperventilation as a treatment for intracranial
hypertension has several complications. First, vasoconstriction, which
tends to reduce ICP, may eventually jeopardize adequate perfusion.
Second, the longer the period of hyperventilation (over 24 hours), the
greater the reduction in bicarbonate, and if pC02 is increased, it may
produce a decrease in brain pH and an increase in ICP. It is strongly
recommended that hyperventilation be used as a short-term therapy in
conjunction with other measures used to lower ICP (Chesnut and
188.8.131.52 Intracranial Pressure
Intracranial pressure (ICP) was measured in the left cerebral
hemisphere using a fiber optic transducer (the standard clinical method
to measure ICP). Since the cranium and spinal canal are a closed
system, an increase in either the brain tissue, blood, cerebral spinal
fluid (CSF) or brain fluids must be accompanied by a decrease in
volume in another component of the calvarium, according to the Monro-
Kellie doctrine. If not, the intracranial pressure will increase because
the calvarium rigidly limits the total cranial volume (Kandel, Schwartz
and Jessel, 1991). Increased absorption or decreased formation of
CSF compensates for chronic changes in intracranial pressure. But
when these compensatory mechanism fail, the intracranial pressure
increases while the cerebral blood flow decreases.
Osmolality was measured by the freezing-point depression
method using samples of arterial blood. Osmosis is the net diffusion of
water from a region of higher water concentration to a region of lower
water concentration. The blood brain barrier is relatively impermeable
to most solutes but highly permeable to water. As a result, when
there is a higher concentration of solute on one side of the blood-brain
barrier, water diffuses across the barrier to the region of higher solute
concentration. In this experiment, two solutions of equal osmolality are
infused (sodium bicarbonate and 6% sodium chloride) in an effort to
increase the blood osmolality. By increasing the blood osmolality, the
water in the brain should diffuse across the blood-brain barrier into the
blood, resulting in decreased intracranial pressure.
184.108.40.206 Arterial pH and Blood Gases
Serum blood levels were monitored every fifteen minutes to
maintain the pH and C02 levels within their fixed protocol criteria.
Since this was an acidotic model, the pH was maintained at about 7.25.
Also, since this model is based on clinical treatments of systemic
acidosis in a brain injured patient, and hyperventilation is a standard
clinical protocol, hyperventilation was used as a treatment for the
control, osmotic control (6% NaCI), and bicarbonate groups. Oxygen
levels (p02 and 02 saturation), bicarbonate and base excess were also
220.127.116.11 Vital Signs
Arterial mean, systolic and diastolic pressure, and heart rate
were monitored continuously and recorded every fifteen minutes.
2.1 Methods Overview
Approval from the University of Colorado Health Sciences Center
Institutional Animal Care and Use Committee was obtained for this
protocol (#63102497(05) 1B) prior to experimentation. In this study, we
proposed to create a canine model of brain injury with systemic
acidosis. Using the standard canine cold lesion model (Clasen, 1953),
lactic acid was infused (.056 g/ml) to reduce blood pH to 7.25 +/- .03.
The 7.25 +/- .03 range was determined by previous pilot studies to be a
level tolerated by canines for an extended period of time (over eight
hours). In two of the three groups that comprise this study, the
bicarbonate infusion (n=4) and 6% NaCI infusion (n=4) groups, the
serum osmolality was maintained at a constant range by infusion of
either bicarbonate or 6% NaCI. Osmolality was not controlled in the no
infusion group (n=4). In all three groups the animals were
hyperventilated. One control group was infused with 6% NaCI, which
has the same osmolality (1600 mosm/L) as bicarbonate. Treatment
was initiated when the intracranial pressure (ICP) increased naturally to
20 mmHg (corresponding to a pathologic state) and the blood pH had
decreased to 7.25 (corresponding to an acidotic state). Following four
hours of continuous monitoring, the animal was euthanized without
regaining consciousness and the brain was immediately removed for
measurement of the lesion.
2.2 Experimental Set-Up and Surgical Procedures
Mongrel dogs were obtained through the Center for Laboratory
Animal Care at the University of Colorado Health Sciences Center.
Prior to transportation to the laboratory, each dog was sedated with
xylazine (1-2 mg/kg) by intramuscular injection into the right thigh.
Sedated animals were transported to the Neurosurgery Research
Laboratory where they were anesthetized using an intravenous injection
of pentobarbital (26-31 mg/kg). Using a Miller blade laryngoscope, the
animals were intubated with a 7.0 mmID cuffed tracheal tube. They
were ventilated on 100 % oxygen (Bird 6400 ventilator) at a tidal
volume and respiratory rate sufficient to maintain a end tidal C02 of 35
+ 3 mmHg. Oxygen saturation was kept above 98% and p02 was
maintained above 100 mmHg. End tidal C02 levels, heart rate and
inspired 02 were continuously monitored via a tongue sensor using a
pulse oximeter/capmonitor (Ohmeda 4700 Oxicap monitor). A surgical
dissection was performed on the right inner thigh to expose the right
femoral artery and vein. The femoral vein was ligated with 2-0 silk
suture. A small incision into the vein using a number 11 surgical blade
was made just above the ligation, and an 18 gauge Teflon catheter was
inserted into the vein and secured with 2-0 silk suture. This procedure
was repeated for the femoral artery. The right arterial line was
inserted for the purpose of obtaining blood samples for blood gas
analysis. The right venous line was inserted for the purpose of infusing
A surgical dissection was performed on the left inner thigh to
expose the left femoral artery and vein. The femoral vein was ligated
with 2-0 silk suture. A small incision into the vein using a number 11
surgical blade was made just above the ligation, and an 18 gauge
Teflon catheter was inserted into the vein and secured with 2-0 silk
suture. The femoral artery was ligated with 2-0 silk suture. A small
incision into the artery using a number 11 surgical blade was made just
above the ligation, and a fiberoptic pressure transducer (model 110-4B,
Camino Laboratories) was inserted into the artery and secured with 2-0
silk. The left venous line was inserted for the purpose of infusing
maintenance fluid (0.9% NaCI) at a rate of 4 cc/kg/hr. The left arterial
fiberoptic pressure transducer was inserted for the purpose of obtaining
continuous readings of systolic, diastolic, and mean arterial blood
Using an electric scalpel and cautery (Valley Lab), the skin and
muscle was retracted from the skull. A 3 mm hole was drilled into the
left parietal skull, 2mm posterior to the cranial suture and 2 mm lateral
from the midline leaving the dura intact. A Camino bolt was screwed
into this hole, and a fiberoptic intracranial pressure transducer (model
110-4B, Camino Laboratories) was inserted through the bolt into the left
cerebral hemisphere. The purpose of the fiberoptic intracranial
pressure transducer was to continuously measure intracranial pressure
in the left hemisphere. Using a grinding drill bit, a 9 mm trephination
into the right parietal region of the skull, 4 mm dorsal to the cranial
suture and 5 mm lateral from the midline was made. A cold lesion was
made by freezing the 9 mm area of exposed brain (dura mater intact)
with liquid nitrogen for 90 seconds via nitrogen in a self contained
copper funnel held gently on the dura.
2.3 Experimental Groups
Each group was infused with lactic acid (.056 g/ml) (Model 77
Peristaltic pump, Harvard Apparatus, Inc.) post cold lesion injury at a
rate sufficient to create a blood pH level of 7.25 + .03. This lactic acid
infusion level was maintained throughout the experiment, regardless of
subsequent pH readings. When the intracranial pressure was equal to
or greater than 20 mmHg (a level considered to be clinically serious)
and the blood pH was 7.25 + .03 for 30 minutes, preassigned treatment
protocols were commenced. An initial treatment 50cc bolus of either
6%NaCI or NaHC03 was given to the appropriate group. These
boluses were given to simulate actual clinical resuscitation protocol.
Animals were assigned to one of three treatment protocols. The
control group received no treatment infusion of either an osmotic or
buffering agent, but did receive an initial bolus of 6% NaCI to control for
the effects of bolus administration. This group was hyperventilated
only. The 6% saline group received an initial 50 cc bolus of 6% NaCI
followed by continuous infusion of 6% NaCI (Model 77 Peristaltic pump,
Harvard Apparatus, Inc.) adjusted to achieve a blood osmolality of 325
+ 3 mosm/L (equivalent to the osmolality achieved with bicarbonate).
Infusion continued to maintain osmolality of 318 < OSM < 335. The
dog was hyperventilated to keep pC02 at 25 + 2 mmHg.
The experimental group, the bicarbonate group, received an
initial bolus of NaHC03 (50mEq) to correct the metabolic acidosis,
followed by continuous infusion of sodium bicarbonate to correct base
excess to 0 + 2. The blood osmolality range for this treatment was 318
< OSM < 335. The dog was hyperventilated to keep pC02 at 25 + 2
Arterial blood samples (1 cc) were drawn from the arterial line
every fifteen minutes for four hours for osmolality (20 ul, 3
microosmometer, Advanced Instruments, Inc.) and pH and blood gas
analysis (model 640 blood gas analyzer, Ciba Corning). All variables in
addition to the blood analysis were recorded every fifteen minutes for
2.4 Measurement of Brain Injury
After four hours of treatment, the animals were euthanized
without regaining consciousness with an overdose of pentobarbital (50
mg/kg). Using a circular bone saw (Autopsy model, Stryker) around the
circumference of the skull, the top of the skull is lifted and removed with
a periosteal elevator. The dura was opened and dissected off the
brain. The brain was removed via the elevator and the size of the
lesion in three dimensions was measured.
2.5 Exclusion Criteria
This experiment involved sustained elevations in intracranial
pressure and low levels of blood pH. Animals were excluded from this
study and were immediately euthanized with pentobarbital overdose if
the intracranial pressure dramatically increased immediately post lesion
(30 mmHg and above), signally a different and more serious injury e.g.
hematoma, or if the ICP did not increase to a minimum of 20 mmHg
within four hours post lesion, or the mean arterial pressure dropped to
critical levels (MAP < 60 mmHg).
2.6 Data Analysis
Repeated measures ANOVA and t-tests were used on hour
interval data. The repeated measures ANOVA analyzes whether there
are differences between the treatment groups and whether there are
interaction effects for the treatment groups with respect to time.
All other statistics, such as comparisons of the major means of
baselines and durations, were analyzed using non-paired t-tests.
The experimental, 6% saline, and no infusion control groups
were similar in all parameters prior to the start of treatment. The time
from baseline to the start of treatment varied between individual
subjects due to the differences in the time required for each to reach an
ICP > 20 mmHg and a pH of 7.25 + .03 for two consecutive readings
15 minutes apart. Table 3.1 lists the mean pretreatment peak ICP and
low ICP values.
Pretreatment Maximum and Minimum ICP (mmHg)
Treatment Group Mean Maximum ICP Max ICP Standard Deviation Mean Minimum ICP Min ICP Standard Deviation
Control 27.5 3.7 14.5 5.07
hco3- 29.5 7.55 8.75 0.96
6% Saline 23.25 3.77 11.00 2.71
3.1 Intracranial Pressure
The pretreatment peak I CP in the experimental group was 29.5 +
3.70 mmHg, 23.25 + 3.77 in the 6% saline group, and 27.5 + 3.70 in
the no infusion control group. There was no significant difference
between the peak pretreatment ICP between the groups as shown in
Table 3.2. Therefore, these groups were identical with respect to ICP
prior to the treatment point.
Comparison of Pretreatment Maximum ICP Between Groups
Comparison between groups Statistically Significant? Probability
control vs bicarbonate not significant p = 0.651
control vs 6% saline not significant p = 0.159
bicarbonate vs 6% saline not significant p = 0.189
The duration of ICP control within + 2 mmHg of the pretreatment
ICP peak was 45 + 30 minutes following initial hyperventilation for the
no infusion control group. The duration of ICP control was still
maintained after four hours for both the bicarbonate and the 6% NaCI
control groups. There were significant differences in ICP control
between the no infusion control and the bicarbonate treatment group (p
< 0.001) and the no infusion control and the 6% NaCI (p = 0.029).
There was no difference between the bicarbonate treatment group and
the 6% NaCI group (p = 0.686).
At the conclusion of the experiment, four hours after treatment,
the mean ICP in the bicarbonate group was 32.5 + 21.9 mmHg, in the
6% NaCI group was 23.3 + 5.5 mmHg, and in the no infusion control
was 44.75 + 3.50 mmHg. There was a significant difference between
the pre and post treatment ICPs for the no infusion control group (p <
0.001). There was no significant difference for the pre and post
treatment ICPs for either the bicarbonate treatment (p = 0.804) or 6%
NaCI control (p = 0.814) groups. Given the short term time constraint
of this protocol, it is unlikely that a significant difference would be
observed between the 6% saline and bicarbonate group with a larger
sample size. The time course of I CP for the three groups is shown in
ICP Levels During Four Hours Post Treatment
Graph 3.2 shows a plot of osmolality vs time for the three
treatment groups. There were no significant differences between the
groups in mean pretreatment osmolalities. By 60 minutes post
treatment, however, there was a significant difference between the no
infusion control group and the controlled osmolality bicarbonate and 6%
NaCI control groups (p=0.003). The osmolality of the no infusion
control group dropped below 310 mOsm/kg, but this interaction for
treatment with respect to time was not significant (p = 0.287). This
possible downward trend in osmolality may be due to a lack of
adequate hydration and electrolyte imbalance since this group was not
being infused with fluid. The bicarbonate and 6% NaCI were kept
between 318 and 335 mmHg.
Serum Osmolality Levels During Four Hours Post Treatment
In all three groups, the pH of arterial blood was reduced to 7.25
+ .03 prior to treatment. As early as 15 minutes post treatment, the
mean pH levels for each group diverged. All groups were significantly
different from each other (p = 0.002). The pH for the bicarbonate
treatment group became mildly alkalotic, remaining just under 7.55 but
above 7.5. The pH for the no infusion control group remained mildly
acidic, generally staying under 7.4 but above 7.3. The 6% NaCI
remained the most acidic, reaching a low pH of just above 7.2. The
pH data are shown in Graph 3.3.
pH Levels During Four Hours Post Treatment
c Bicarbonate j
'30 210 240
3.4 Base Excess
As shown in Graph 3.4, the base excess was controlled in the
bicarbonate group between 0 and under 5 mEq/L, indicating a neutral
state of pH. The no infusion control group hovered between -5 and -9
mEq/L, which indicated a state of metabolic acidosis. The 6% NaCi
control group had the most negative base excess, ranging from -10 to -
17.5 mEq/L, which indicated a state of metabolic acidosis.
All groups were significantly different from each other (p =
0.002). The interaction effect of treatment course with time is
significant (p = 0.003). The 6% saline group appears to have a
downward trend with time.
Base Excess During Four Hours Post Treatment
90 -60 -30 0 30 50 90 120 150 180 210 240 270
3.5 Mean Arterial Blood Pressure
There was no significant differences between the three groups in
the pretreatment mean arterial blood pressures. There was no
significant difference between the bicarbonate treatment group and the
6% NaCI control group throughout the experiment (p = 0.208). The
mean arterial blood pressure for the no infusion control was somewhat
hypertensive to the bicarbonate and 6% NaCI groups. However, this
difference was not significant (p = 0.208). There were no interaction
effects (p = 0.357). Graph 3.5 shows these trends in mean arterial
Mean Arterial Blood Pressure During Four Hours Post Treatment
3.6 Carbon Dioxide
As shown in Graph 3.6, the mean carbon dioxide levels for all
three groups were held under 30 mmHg and above 23 mmHg by
hyperventilation throughout the four hours post treatment. There were
no significant differences between the three groups (p = 0.281) and no
interaction effects (p = 0.620).
Arterial pC02 Levels During Four Hours Post Treatment
I Control ; ~i
I o Blcarbonst* i i
< a 6% Saline j -
30 50 90 '20 150 130 210 240 270
3.7 Lesion Diameter
The mean lesion diameter was 18.2 + 1.6 mm for the
bicarbonate treatment group, 17.5 + 0.7 mm for the no infusion control
group, and 18.0 4.3 mm for the 6% NaCI control group. These data
confirm that all three groups received comparable brain injuries.
While bicarbonate has clinically been demonstrated to be
effective in managing ICP, this study was proposed to support or
dispute this capability under controlled experimental conditions and to
determine whether its effect is due to the ability to neutralize systemic
acidosis or the ability to act as an osmotic agent. The results
demonstrate that the osmotic agent properties of both the bicarbonate
and the 6 % NaCI played a critical role in effectively managing ICP.
Both groups succeeded in maintaining pretreatment ICP levels for
almost four hours post treatment. The no infusion control group,
however, maintained pretreatment ICP levels for 45 minutes, and then
climbed to 45 mmHg after four hours of treatment. These results
suggest that the duration of pretreatment ICP levels were due solely to
the osmotic effects of bicarbonate.
4.1 Management of ICP
Although there is debate in the literature in the definition of
critical levels of ICP, there is some consensus that ICP is normally less
than 10 mmHg when measured at the level of the foramen of Monro
(midhead) in a patient lying face upwards. Elevations of ICP above 15
mmHg are generally considered abnormal. In one study, 160 patients
who were unable to obey commands at hospital admission were
monitored for ICP. 44% of the patients had ICPs greater than 20
mmHg, and 82% had ICPs greater than 10 mmHg. All patients whose
ICP could not be controlled below 20 mmHg died (Ropper, 1993).
Therefore, the ICP level of 44.75 + 3.5 mmHg seen in the control
group is extremely critical, and based on current case studies in the
literature, would have very poor outcome (DeSalles et al, 1986).
While there was a significant difference between the control and
bicarbonate (p < 0.001) and control and 6% NaCI group (p = 0.029),
there was no significant difference between the bicarbonate and 6%
NaCI groups (p = 0.686). These results strongly suggest the influence
of the osmotic agent properties of both the bicarbonate and the 6%
NaCI. Both of these groups were infused to maintain the same serum
osmolality range (318 < OSM < 332). The duration of ICP control was
maintained throughout the experiment for both the 6% NaCI group and
the bicarbonate group. Therefore, based on these ICP level durations,
there does not appear to be an obvious benefit from the bicarbonate
buffering system on the regulation of elevations of ICP at least for the
time period studied. Although the concentration of sodium ions was a
common component of each osmotic agent and might suggest a
possible "non-osmotic1 influence by Na+, it seems unlikely since no
other side effects occurred, such as increased blood pressure. The
effect of controlling ICP appears to be solely a result of the
hyperosmotic properties of each fluid.
4.2 Volume Control
This is not to say, however, that 6% NaCI or bicarbonate is
necessarily the fluid of choice in the management of intracranial
hypertension. The core medical treatment for reduction of ICP is fluid
restriction, osmotic agents, and hyperventilation (Loftus, 1994). An
important consideration in this model is that the volume of fluid infused
was not controlled because the rate of fluid infusion, whether
bicarbonate or 6% NaCI, was determined by maintaining a specific
range of serum osmolality level. So, infusion rates varied between
individual group subjects throughout various points in time in each
experiment (infusion rates varied between approximately 0.5 cc/min to
3.0 cc/min), resulting in a lack of fluid restriction. The infusion
volumes of bicarbonate and 6% NaCI were greater than the rates of the
core medical treatment osmotic agent mannitol. Mannitol is unable to
cross the blood-brain barrier and thus reduces I CP by drawing water
from the brain into the intravascular space, achieving maximum ICP
reduction in 20 60 minutes. Mannitol is administered as a bolus every
four hours in relatively small volumes (Loftus, 1994).
4.3 Osmotic Agents
The use of a 6% NaCI solution as an osmotic agent in the
volumes administered in this experiment may also have a questionable
long-term, and possibly short-term, effect on the brain and kidneys.
The normal concentration of extracellular sodium in a mammalian
neuron is 145 mM, and the intracellular concentration is between 5 and
15 mM (Purves et al, 1997). Normal saline, which is 0.9 % NaCI, is
isotonic to the blood. Therefore, the 6% NaCI was very hyperosmotic
to body fluid levels, while also adding a large elevation in sodium levels.
However, hypertonic saline has been suggested as a primary treatment
for intracranial hypertension, especially in situations when the effects of
mannitol have diminished (Ropper, 1993). It has been noted in the
literature that hypertonic saline causes sustained reductions in
intracranial pressure without diuresis, and improved renal function in
patients that were severely dehydrated. Hypertonic saline has also
proven useful in small amounts in preventing hypotension in patients
that previously received large doses of mannitol. In animal models,
hypertonic saline has been shown to be advantageous for combined
hemodynamic and cerebral resuscitation in shock by reducing the
intracranial hypertension while maintaining the perfusion pressure
(Ropper, 1993). Since the focus of this study was purely to
demonstrate or disprove the ability of bicarbonate to maintain I CP, and
whether this effect was due to its buffering or osmotic properties, the
role of the 6% NaCI was simply as an osmotic agent equivalent to
bicarbonate in osmolality.
Osmotic agents, such as the bicarbonate and 6% NaCI in this
current study, or the commonly used agents such as mannitol and 3%
NaCI, have several potential disadvantages. Hypotension, which is
abnormally low blood pressure, and hypokalemia, which is an
abnormally low potassium level in the blood, which may lead to
neuromuscular and renal disorders and to electrocardiographic
abnormalities may result from the use of osmotic agents (Ropper,
1993). Repeated administration of mannitol, for example, can result in
acute renal failure. Hypernatremia, an excess of sodium in the blood,
results from the renal loss of water through osmotic diuresis. The
diuresis effect may also cause volume contraction which results in an
increased risk of hypotension and ischemia. In addition, the continuous
high serum mannitol levels may force mannitol into injured brain regions
and result in intracranial hypertension. In this current study, a slight
drop in blood pressure was noted in two trials of the bicarbonate group,
but this blood pressure change was not significant (p = 0.357).
There are several advantages to the use of osmotic agents,
however, and these advantages tend to overcome the possible risks.
They have rapid onset, are titratable, and predictable (Ropper, 1993).
These advantages were demonstrated in this current study. Both the
bicarbonate and the 6% NaCI were effective in management of the ICP,
both were easily titrated to adjust for osmolality levels, and both
attained very predictable results in each trial.
4.4 Buffering Capabilities
The buffering capabilities of bicarbonate did not appear to play a
role in the management of elevated ICP in this study. The buffering
capabilities are important, however, in the management of systemic
Sustained systemic acidosis can be a fatal condition. For
example, a person can only live a few hours when the systemic pH is
6.8 (Guyton and Hall, 1996). The pH of most body fluids is between
6.5 and 8.0, and sustained acidosis can result in deleterious effects if
enzymes cease to function due to the acidic environment. The buffer
systems of the body respond within a fraction of a second to slight
changes in hydrogen ion concentration (Guyton and Hall, 1996). First,
the chemical system keeps them tied up with conversion to carbonic
acid until the balance can be reestablished. Within a few minutes the
respiratory system begins to eliminate C02 and low pH increases the
efficiency of oxygen unloading from hemoglobin (Mathew and van
Holde, 1996). Third, the kidneys eliminate excess acid from the body.
Acidosis also has negative effects on the brain. It is suggested
that it affects ischemic brain damage by accelerating the evolution of
the damage, by disrupting the blood-brain barrier, and may trigger
postischemic seizures (Katsura et al, 1994).
On the cellular level, acidosis strongly depresses neuronal
activity. A comatose state may result from a drop in pH from 7.4 to
7.0 (Guyton, 1991).
Uninjured sites in the brain are also affected as a result of
acidosis. As the edema initiated by acidosis compresses the cerebral
vessels, blood flow is decreased resulting in brain ischemia.
The effects of acidosis at the injured site predispose the brain
tissue to infarction, which is an ischemic condition that causes a
persistent focal neurologic deficit in the affected area. Acidosis is also
thought to accelerate tissue damage mechanisms.(Katsura et al, 1994).
During the first hour of treatment, hyperventilation was
successful in raising the pH for all three groups. The bicarbonate group
became mildly alkalotic, reaching a pH just under 7.55. The control
group was returned to a normal pH of 7.4, while the 6% saline group
increased in pH, yet remained under 7.4.
Following the first hour of treatment, however, hyperventilation
used alone to manage systemic acidosis began to lose its
effectiveness. The bicarbonate group maintained its alkalotic pH
throughout the remainder of the trial. Alkalosis increases neuronal
excitability. Elevations in arterial pH from 7.4 to 7.8 to 8.0 may cause
cerebral convulsions (Guyton, 1991). The alkalotic state of the
bicarbonate group may have been due to the method used to correct
the acidosis. Because we were adjusting the bicarbonate infusion level
to maintain a base excess level at 0 + 2 instead of regulating the pH,
the bicarbonate group over corrected and became slightly alkalotic
ranging between pH 7.50 and 7.55. Another possible explanation is the
volume of lactic acid infused into this group. The rate of lactic acid
infusion was based on lowering the pH to 7.25 + .03 for two arterial
blood sample pH readings fifteen minutes apart. Due to variances in
animal age and health conditions, the amount of lactic acid infused to
reach this end point varied between trials. At the end of the study,
infusion rates were analyzed. Table 4.1 lists the mean infusion rates
for each group based on weight. Retrospectively, a rate of around 0.05
ml/min kg would have been ideal. Both the control and bicarbonate
groups received too little lactic acid, at 0.032 ml/min kg and 0.034 ml/kg
respectively, while the 6% saline group received too much lactic acid, at
a rate of 0.061 ml/min kg.
Mean Rates of Lactic Acid Infusion for Experimental Groups
Experimental Group Lactic Acid Infusion Rate (ml/min) Weight of Animal (kg) Lactic Acid Rate (ml/min kg)
Control 0.73 23.25 0.036
Bicarbonate 0.64 19.00 0.034
6% Saline 1.35 22.00 0.062
The control group, whose base excess was not controlled for by
infusion, hovered in the mildly acidotic range, with a pH of 7.33 being
the minimum pH after four hours of treatment. Again, the reason that
this group was not more acidotic was probably due to the infusion rate
of the lactic acid being below the retrospectively calculated ideal rate of
0.05 ml/min kg.
The 6% NaCI group experienced the lowest pH, dropping below
7.30 approximately one hour post treatment. Although it would be
expected that this group would have low pH values because there was
no correction of the systemic acidosis other than hyperventilation, it
would not be expected to have a lower pH than the control. This result
is probably due to the lactic acid infusion rates. The 6% saline group
received a higher rate of lactic acid infusion compared to both the
control and bicarbonate groups. This group showed a very slight
increase in pH due to the effects of hyperventilation treatment in the
first hour, but this small benefit was overwhelmed as the experiment
progressed. Also, the acidic pH of the 6% NaCI may be an indirect
result of the large salt concentration assault on the kidneys. Since the
renal system is responsible for correcting acid base imbalances by
regulation of the secretion and reabsorption of salts, bicarbonate, and
hydrogen ions, the concentration of sodium in the 6% NaCI infused over
a four hour period may have overloaded the filtering capacity of the
kidneys (Guyton and Hall, 1996). This current study was not sufficient
in duration to study the long term effects of this salt concentration on
the kidneys, but in theory, the sustained infusion of 6% NaCI would lead
to renal failure.
4.5 Metabolic Acidosis
The post treatment arterial blood gases were significantly
different for pH (p = 0.002), HCOa- concentration, and base excess (p =
0.002) between groups, but there was no difference in the paC02
between groups (p = 0.281). These data suggest that the respiratory
component of the acid-base parameters was not affecting the acid-base
imbalance, or that we were controlling the degree of respiratory
acidosis, and that any differences between the groups were due to the
metabolic acid-base imbalance.
4.6 Future Research
4.6.1 Laboratory Studies
18.104.22.168 Corrections to the Current Study
This current study used an induced model of systemic metabolic
acidosis. There were several areas of the study that could be improved
upon. For instance, the design of the control groups could be
improved. This current study was based on the previous pilot study
(Levy et al, 1994) which hypothesized that systemic acidosis blunted
the efficacy of hyperventilation and contributed to cerebral hyperemia
and brain swelling, thus adversely affecting ICP. In that study, eight
dogs were given a cold lesion, infused with lactic acid, and treated
when the ICP reached 20 mmHg. Both groups were treated with 1
gm/kg of mannitol and hyperventilated to a pC02 of 20-25 mmHg. The
treatment group received a 50mEq of NaHC03 to correct the acidosis.
The control group received a 50 mEq of 6% NaCI to control for the
short-lived physiologic effects of bolus administration, which include the
rapid rise and fall of ICP and increased blood pressure. The study
found that the correction of systemic lactic acidosis by the
administration of NaHCOa significantly improved the efficacy and
duration of the action of mannitol and hyperventilation in the treatment
of cold lesion induced intracranial trauma. So, as a result of the pilot
study, the 6% NaCI group was used to control for osmolality in the
current study. The 6% NaCI had the advantage of being an equivalent
osmolality to the bicarbonate and was easy and inexpensive to prepare.
It had the disadvantage of being of greater hyperosmolality than the
clinically used saline hyperosmotic, 3% NaCI. This is not a factor in the
study outcome, however, because the 6% saline served only as a
control, and not as a recommended clinical treatment.
In addition, there should have been a control group that received
no bolus and no treatment fluid. Although bolus administration is a
standard in clinical trauma settings, the laboratory offers the opportunity
to control for many of the variables that cannot be controlled for in a
clinical setting. As a result, we should have controlled for the effect of
the bolus by having a control group that did not receive a bolus.
This current study should also have had tighter control of the
lactic acid infusion, systemic pH, pC02, and [HC03], The rate of lactic
acid infusion presented a problem. In this study, the lactic acid
infusions rate was adjusted to reduce pH to 7.25 and maintain that pH
for two pH readings 15 minutes apart. However, due to variances in
the dogs, such as age, general health and physical condition, the
infusion rate varied considerably in each trial. As a result, some
animals were less acidotic than others, and therefore had less acidosis
to adjust and compensate for. The control and bicarbonate groups
received less lactic acid than the 6 % saline group. This explains the
pH graph, Graph 3.3, which shows a near normal pH for the control
group and an overcompensated to the point of alkalosis, bicarbonate
group. After evaluating the infusion rates for all of the trials, a rate of
approximately 0.05 ml/min would probably be an ideal rate for lactic
acid infusion to achieve systemic acidosis.
However, the variables of lactic acid infusion, systemic pH, pC02
and [HCOJ are somewhat difficult to control at times since they are
interconnected variables. Also they can change rapidly or slowly under
different conditions. Tighter control of pH was attempted originally with
continuous pH monitoring via an arterial pH probe. This micro pH
probe was inserted directly into the femoral artery for continuous
monitoring, but it complicated the experiment by being prone to clotting
and breakage. Thus, due to the additional time delays in microprobe
adjustments and financial constraints due to tip breakage, this
technique was abandoned in favor of arterial blood pH and gases
analyzed every fifteen minutes.
The severity of this model was very discouraging in terms of the
number of animals that met the exclusion criteria. Animals were
excluded due to complicating factors such as animals experiencing
existing health problems which made the infused lactic acid fatal and
animals that received a more serious injury which escalated their
intracranial pressures too high too quickly.
22.214.171.124 Future Laboratory Studies
It would be interesting to perform a long term study, of over 24
hours, to evaluate the effects of bicarbonate and 6% NaCI on a model
that would gradually and naturally develop metabolic acidosis as a
result of brain injury. This proposed study would allow for greater
evaluation of the other systems, such as the renal and cardiac systems,
affected by brain injury and its subsequent treatment. This proposed
study should also record more parameters, such as serum electrolytes,
CSF pH and lactate levels, brain tissue pH and lactate levels, and a
method to assess the extent of tissue damage. Gravimetric analysis or
magnetic resonance spectroscopy (MRI) could be used to analyze
water content of the brain.
Currently, researchers are testing the effect of another buffering
agent, Tris-hydroxymethyl aminomethane (THAM), on the treatment of
head injury in patients (Marmarou, 1992). This clinical study
demonstrated THAM to be effective in reducing I CP and the amount of
therapy, such as the amount of mannitol administered. The current
study could be expanded to include a THAM group, which would
provide additional comparison to the bicarbonate group. THAM, unlike
bicarbonate, has the advantage of diffusing into the intracellular space.
It can cross the blood-brain barrier. THAM is a weak base amino-
alcohol that has a favorable pK and pH buffer range for physiologic
reactions and is currently proposed as an alternative to sodium
bicarbonate in resuscitation procedures (Sirieix et al, 1997). It is also
thought to be a possible osmotic agent. A current study using an
isolated heart model suggested that there are deleterious myocardial
effects of metabolic acidosis, and that sodium bicarbonate further
impairs myocardial performance. Administration of THAM alone failed
to completely buffer metabolic acidosis, but the combination of THAM
and sodium bicarbonate was thought to be successful due to the ability
of THAM to capture the C02 produced as a result of sodium
bicarbonate buffering. The THAM and sodium bicarbonate combination
was able to correct metabolic acidosis while improving myocardial
performance (Sirieix, 1997).
4.6.2 Clinical Studies
126.96.36.199 Retrospective Studies
There is a wealth of information contained in patient charts
that could be obtained by a medical record search using ICD9 injury
codes. Cases could be analyzed by controlling for age of patient, type
of brain injury, presenting ICP, treatment course, complications and
outcome. Intensive Care Unit (ICU) records could be analyzed by the
specifics of the treatments used, such as when, how much, and any
effects. While this would be a very time intensive task, requiring a
very large number of subjects to achieve statistical significance due to
such wide variances in type and location of injury, patient age, patient
health condition, treatment course, it would help identify specific
correlations between variables to provide information on which to base
future study designs.
188.8.131.52 Prospective Studies
Making the leap from a laboratory trial to a clinical trial can
occasionally present some ethical dilemmas. For instance, designing a
clinical trial based on this current study is possible, though problematic.
The first step should be to investigate in the laboratory the THAM arm
of the experiment. If the results are promising, a combination
bicarbonate/THAM should also be tested in the laboratory to investigate
whether the results support the literature. Based on the laboratory
experiment results and laboratory and clinical studies in the literature,
which currently show no increased morbidity or mortality due to
bicarbonate and THAM, the study would possibly be ready for clinical
When designing the trial, it is important to present to the Human
Research Subjects Committee, who oversee all clinical trials, that there
are currently no ideal, or perfect, treatments for intracranial
hypertension at this time. All current therapies come with deleterious
effects that vary during the length of the treatment course and from
individual to individual. For example, hyperventilation has been shown
to have deleterious short term effects (3-6 months) on patient disability
outcome (Marmarou, 1992). Also, the use of mannitol could result in
renal failure. Therefore, it is possible to ethically justify the design of
a study that restricts certain treatments. In example, a clinical study
investigated the use of THAM for the treatment of head injury by
randomly assigning patients into one of three groups. All groups
received mannitol when CSF drainage was ineffective and I CP
exceeded 20 mmHg. The control group received no additional
treatment. The second group received hyperventilation treatment, while
the third group received hyperventilation and THAM (Marmarou, 1992).
Was this study design ideal? No, but it did allow for all patients to be
treated for elevated intracranial pressure while evaluating the
effectiveness of THAM.
In conclusion, I propose that further laboratory and retrospective
studies be conducted to test the effectiveness of sodium bicarbonate,
THAM and the combination of THAM and sodium bicarbonate in the
treatment of intracranial hypertension and systemic acidosis. If these
results are statistically significant for either of these treatments, then a
prospective clinical study be pursued.
Barrow, D. Complications and Sequelae of Head Injury. Park Ridge,
IL: AANS Publications; 1992.
Berenyi, K; Wolk, M.; Killip, T. Cerebrospinal fluid acidosis
complicating therapy of experimental cardiopulmonary arrest.
Circulation. 52:319-324; 1975.
Chesnut, R.; Marshall, L. Management of Severe Head Injury:
Neurological and Neurosurgical Intensive Care. New York, NY: Raven
Press, Ltd.; 1993.
Ciba Corning Diagnostics Corp. 170 pH/Blood Gas Analyzer Instruction
Cooper, P. Head Injury. Baltimore, MD: Williams and Wilkins; 1993.
DeSalles, A.; Kontos, H.; Becker, D.; Yang, M.; Ward, J.; Moulton, R.;
Gruemer, H.; Lutz, H.; Maset, A.; Jenkins, L; Marmarou, A.; Muizelaar,
P. Prognostic significance of ventricular CSF lactic acidosis in severe
head injury. J. Neurosurgery. 65:615-624; 1986.
Duthie, S.; Goulin, G.; Zornow, M.; Scheller, M.; Peterson, B. Effects of
THAM and sodium bicarbonate on intracranial pressure and mean
arterial pressure in an animal model of focal cerebral injury. J
Neurosurgical Anesthesiology. 6(3):201-208; 1994.
Eleff, S.; Sugimoto, H.; Shaffner, H.; Traystman, R.; Koehler, R.
Acidemia and brain pH during prolonged cardiopulmonary resuscitation
in dogs. Stroke. 26:1028-1034; 1995.
Gordon, E. Some correlations between the clinical outcome and the
acid-base status of blood and cerebrospinal fluid in patients with
traumatic brain injury. Acta anaesth. Scandinav. 15:209-228; 1971.
Guyton, A. Basic Neuroscience. Philadelphia: W.B. Saunders
Guyton, A.; Hall, J. Textbook of Medical Physiology. Philadelphia:
W.B. Saunders Company; 1996.
Hurn, P.; Koehler, R.; Bizzard, K.; Traystman. Deferoxamine reduces
early metabolic failure associated with severe cerebral ischemic
acidosis in dogs. Stroke. 26:688-695; 1995.
Kandel, E.; Schwartz, J.; Jessell, T. Principles of Neural Science. New
York, NY: Elsevier Science Publishing Co, Inc.; 1991.
Katsura, K.; Kristian, T.; Smith, M.; Siejo, B. Acidosis induced by
hypercapnia exaggerates ischemic brain damage. Journal of Cerebral
Blood Flow and Metabolism. 14:243-250; 1994.
Levy, A.; Munro, L; Beel, J.; Nichols, J. Correction of systemic lactic
acidosis enhances the treatment of increased intracranial pressure after
experimental brain injury. American Association of Neurological
Levy, A.; Munro, L.; Beel, J.; Nichols, J. Correction of systemic lactic
acidosis enhances treatment of increased intracranial pressure,
(manuscript in progress)
Loftus, C. Neurosurgical Emergencies, Volume I. USA: AANS Press;
Marmarou, A. Intracellular Acidosis in Human and Experimental Brain
Injury. Journal of Neurotrauma. 9:S551-S562; 1992.
Mathews, C.; vanHold, K Biochemistry. Menlo Park, CA: The
Benjamin/Cummings Publishing Company, Inc; 1996.
Nickel, K. ALBA's Medical Technology. Denver, Co: Berkeley
Scientific Publications; 1991.
Ohnishi, S.; Ohnishi T. Central Nervous System Trauma Research
Techniques. Boca Raton, FL: CRC Press; 1995.
Paljarvi, L.; Soderfeldt, B.; Kalimo, H.; Olsson, Y.; Siejo, B. The brain
in extreme respiratory acidosis. Acta Neuropathologica. 58:87-94;
Purves, D.; Augustine, G.; Fitzpatrick, D.; Katz, L; LaMantia, A.;
McNamara, J. Neuroscience. Sunderland, MA: Sinauer Associates,
Rabow, L.; DeSalles, A.; Becker, D.; Yang, M.; Kontos, H.; Ward, J.;
Moulton, R.; Clifton, G.; Gruemer, H.; Muizelaar, J.; Marmarou, A. CSF
brain creatine kinase levels and lactic acidosis in severe head injury. J
Neurosurg. 65:625-629; 1986.
Ropper, A. Neurological and Neurosurgical Intensive Care. New York,
NY: Raven Press, Ltd; 1993.
Siesjo, B.; Katsura, K; Mellergard, P.; Ekholm, A.; Lundgren, J.;
Smith, M. Acidosis-related brain damage. Progress in Brain Research.
Sirieix, D.; Delayance, S.; Paris, M.; Massonnet-Castel, S.; Carpentier,
A.; Baron, J. Tris-hydroxymethyl aminomethane and sodium
bicarbonate to buffer metabolic acidosis in an isolated heart model. Am
J Respir Crit Care Med. 155:957-963; 1997.
Taylor, G.; Myers, S.; Kurth, C.; Duhaime, A.; Yu, M.; McKernan, M.;
Gallagher, P.; O'Neill, J.; Templeton, J. Hypertonic saline improves
brain resuscitation in a pediatric model of head injury and hemorrhagic
shock. J Pediatric Surg. 31:65-70; 1996.
Temppema, L.; Barts, P.; Evers, J. Effects of metabolic arterial pH
changes on medullary ecf pH, scf pH and ventilation in peripherally
chemodenervated cats with intact blood-brain barrier. Respiration
Physiology. 58:123-36; 1984.
Vomov, J.; Thomas, A.; Jo, D. Protective effects of extracellular
acidosis and blockage of sodium/hydrogen ion exchange during
recovery from metabolic inhibition in neuronal tissue culture. J
Neurochemistry. 67:2379-2389; 1996.
Wagerle, L; Kumar, S.; Belick, J.; Delivoria-Papadopoulos, M. Blood-
brain barrier to hydrogen ion during acute metabolic acidosis in piglets.
J Applied Physiology. 65:776-781; 1988.
West, J. Respiratory Physiology the Essentials. Baltimore, MD:
Williams and Wilkins; 1990.
Zorrow, M.; Prough, D. Fluid management in patients with traumatic
brain injury. New Horizons. 3:488-98; 1995.
Zupping, R. Cerebral acid-base and gas metabolism in brain injury. J.
Neurosurg. 33:498-505; 1970.
Zupping, R.; Magi, M.; Tikk, A.; Raudam, E. Cerebral metabolic
disorders during prolonged unconsciousness after severe head injury.
Europ. Neurol. 8:145-150; 1972.