Perioperative Fluid

Therapy
a, b
Denise Fantoni, DVM, PhD *, Andre C. Shih, DVM

KEYWORDS
 Fluid responsiveness  Cardiac output  Anesthesia  Perioperative time

KEY POINTS
 Anesthesia can lead to pathophysiologic changes that dramatically alter the fluid balance
of the body compartments and the intravascular space.
 Fluid administration can be monitored and evaluated using static and dynamic indexes.
 Guidelines for fluid rates during anesthesia begin with 3 mL/kg/h in cats and 5 mL/kg/h in
dogs.
 If at all possible, patients should be stabilized and electrolyte disturbances should be cor-
rected before general anesthesia.

INTRODUCTION

Fluid administration during anesthesia is necessary to control vascular tone, maintain

circulating volume, and improve cardiac output (CO). Overhydration and excessive
fluid administration can be detrimental to patients just as dehydration and hypovole-
mia can lead to adverse consequences.1 Potential adverse effects of overly aggres-
sive fluid therapy include volume overload, pulmonary edema, detrimental fluid
shifts (eg, edema of the brain, kidneys, and intestinal tract), electrolyte and acid
base derangements, exacerbation of hemorrhage, and hemodilution coagulopathy.2
The decision to cease or decrease fluid administration can be as important as the
decision to start or increase fluid administration. The ideal rate of fluid administration
during anesthesia must be tailored to a patient’s volume status as well as serum elec-
trolyte and acid base status. Although fluid therapy is routinely performed in most
anesthetized patients, little progress has been made on estimation of circulating blood
volume status and fluid responsiveness. The volume status of patients typically is

The authors have nothing to disclose.

a
Surgery Department, Faculdade de Medicina Veterinária e Zootecnia da Universidade de São
Paulo (FMVZ USP), Av. Orlando Marques de Paiva 80, São Paulo 55800, Brazil; b Large Animal
Clinical Science, University of Florida College of Veterinary Medicine, 2015 Southwest 16th
Avenue, Gainesville, FL 32609, USA
* Corresponding author.
E-mail address: dfantoni@usp.br

Vet Clin Small Anim 47 (2017) 423–434

http://dx.doi.org/10.1016/j.cvsm.2016.11.004 vetsmall.theclinics.com
0195-5616/17/ª 2016 Elsevier Inc. All rights reserved.
424 Fantoni & Shih

evaluated indirectly by means of parameters that reflect perfusion, but the limitations
of such methods do not assure adequate volume maintenance in many situations.
Blood pressure, urine output, and heart rate are parameters frequently used to es-
timate the adequacy of blood volume and the response to fluid administration.3,4 Other
parameters, such as peripheral pulses, mucous membrane color, capillary refill time,
and tissue turgor, also are used to evaluate circulating blood volume. Sole use of these
parameters, however, can lead to clinically relevant errors in blood volume estimation.
Furthermore, sick patients under anesthesia can develop pathophysiologic changes
that dramatically alter fluid balance within the body’s fluid compartments. Hypoprotei-
nemia, electrolyte disturbances, renal disease, hepatic insufficiency, and cardiovas-
cular dysfunction are common conditions encountered in surgical patients that can
markedly affect fluid balance.
For all of these reasons, a comprehensive understanding of the factors that influ-
ence hemodynamic stability and new concepts regarding the evaluation of fluid
administration and circulating blood volume status is crucial to successful administra-
tion of fluids in the perioperative period.

HEMODYNAMIC STABILITY

Hemodynamic stability is the primary therapeutic goal in every patient and the main
reason fluids are administered to surgical high-risk patients. If a patient is hemody-
namically stable and blood oxygen content is adequate for the clinical situation, oxy-
gen demand is met. In situations in which oxygen content decreases (eg, acute
hemorrhage), increasing venous return by fluid loading assures oxygen demand.3,5,6
Conversely, if the hemodynamic state is compromised, the oxygen demand may
not be adequately met. To understand this basic equilibrium, it is important to define
the constituents of hemodynamic stability. If there is awareness of the clinical rele-
vance of each constituent, the importance of adequate assessment of circulating
blood volume and the role of fluid therapy in maintaining it is readily apparent.
Hemodynamic stability depends on heart rate and stroke volume, which together
generate CO (Fig. 1). Stroke volume in turn is the result of preload, afterload, and
myocardial contractility, and all of these variables are closely related. Preload is the
result of venous return and venous tone, afterload depends on arterial vascular resis-
tance and aortic impedance, and contractility is an intrinsic property of the myocar-
dium. Blood pressure ultimately reflects CO because it is the product of peripheral
vascular resistance and CO. For this reason, blood pressure can be normal, despite
low CO, if peripheral vascular resistance is high. The elements that determine CO
can be altered during anesthesia by different conditions and also in sick patients.
For example, vasodilatation caused by administration of inhaled anesthetic agents de-
creases afterload and if heart rate does not increase accordingly, CO decreases. This
is common during inhalation anesthesia because the ability to increase heart rate to

Fig. 1. Hemodynamic stability.

 Perioperative Fluid Therapy 425

counteract vasodilatation is impaired by a decrease in baroreceptor sensitivity.3,5,6 In

some circumstances, fluids can be administrated to overcome hypotension caused by
vasodilatation, but this approach is not always effective because a fluid deficit is not
present and the heart is not preload responsive. Fluid administration can even be
harmful in clinical situations associated with a tendency for fluid to shift to the third
space (eg, hypoproteinemia or sepsis) or in cardiac patients. It has been demon-
strated that fluid therapy does not effectively counteract severe hypotension pro-
moted by deep anesthesia.7
Conversely, during acute hemorrhage, fluid administration can help meet oxygen
demand because the increase in preload augments stroke volume and increased CO
counteracts the decrease in oxygen content. Anesthesia does not blunt this response
if anesthetics are administered at 1 MAC (minimum alveolar concentration),8,9 but in a
deeper plane, the intense vasodilatation and decrease in contractility may negate the
benefit of the increase in preload. Even in an awake patient or during light anesthesia,
there is a point at which the increase in CO is not sufficient to overcome the decrease in
oxygen content. This point is difficult to determine in anesthetized patients, especially if
the adequacy of blood volume is underestimated or the response to fluid administration
carefully evaluated.6
In recent decades, monitoring fluid administration using more reliable instruments
has received more attention and today is considered one of the most important ap-
proaches in high-risk and unstable surgical patients.

MONITORING FLUID THERAPY

How can the adequacy of fluid administration be effectively evaluated? Basically,

the adequacy of fluid administration is evaluated using well-known static and dy-
namic indices. The static indices are measurement of arterial blood pressure and
central venous pressure (CVP). These indices do not reflect fluid responsiveness
with sufficient accuracy because they only indirectly reflect cardiac preload and
function. In a meta-analysis of previous studies conducted in humans and animals,
CVP was determined not to be an accurate method to evaluate fluid responsiveness
and it was recommended to no longer use CVP for this purpose.10,11 Studies have
indicated that the most commonly monitored indices (heart rate, body temperature,
CVP, and urine output) were the worst predictors of survival.3 In dogs with sepsis
associated with pyometra, the increase in CVP in response to a fluid load was
not accompanied by a stepwise increase in blood pressure or CO measured by a
pulmonary artery catheter.12 In an experimental study conducted in pigs, the au-
thors demonstrated that dynamic parameters, such as pulse pressure variation
(PPV), are superior to CVP and even pulmonary artery occlusion pressure in predict-
ing changes in preload.10,11,13 As discussed previously, vascular resistance can
greatly influence blood pressure. For this reason, in states of severe vasodilatation,
including septic patients and those under general anesthesia, fluid administration
alone is not able to correct hypotension appropriately. If only blood pressure is
used to evaluate the response to the fluid load, an excessive volume of fluids can
be administered before or the fluids may be stopped before an effective increase
in preload can be accomplished. These facts emphasize the need for better tools
to monitor fluid therapy.
One reason to adopt better tools for the evaluation of fluid responsiveness is based
on the Frank-Starling curve. Observing the curve (Fig. 2), it is easy to see that, to a
point, an increase in preload results in a stepwise increase in CO. On the other
hand, when this limit is reached, increasing preload does not result in a further
426 Fantoni & Shih

Fig. 2. Frank-Starling curve showing the relationship between preload and systolic volume.

increase in CO.3 This crucial transition point is not easily identifiable using static pa-
rameters for assessment.
Among the dynamic indices, PPV, systolic pressure variation, stroke volume varia-
tion, and plethysmographic waveform variation are the most studied and used indices
in humans and experimental animals.10,11,13 Also, the inferior vena cava distensibility
rate and superior vena cava collapsing rate are being studied extensively. The major
disadvantage of these indices is that they rely on patients mechanically ventilated with
a known tidal volume (8–10 mL/kg), without arrhythmias, and with no inspiratory effort.
Taking into consideration that mechanical ventilation is becoming more common in
veterinary anesthesia and for veterinary patients in ICU, these indices should find
more use in veterinary medicine.
Of these indices, PPV has been extensively studied in humans and experimental an-
imals and has proved an excellent method to evaluate fluid responsiveness.14–17 As
with other indices, it is based on the respiratory variation in arterial pulse pressure
and requires an arterial line and software incorporated in a multiparametric monitor
to automatically calculate it. Advantages are that its measurement does not depend
on operator experience, and it is continuously measured. Its use in humans is associ-
ated with better outcomes both in high-risk ICU patients and in major surgical pa-
tients,15,16 and it has been incorporated into algorithms to minimize uninformed
infusion of fluids. Of great importance is the definition of the appropriate cutoff value
for each species that differentiates responders from nonresponders.17 In humans, the
cutoff value is 13%; in pigs, the authors found a value of 16%,15 whereas in dogs, this
value is 15% (Fantoni et al, unpublished data, 2016). For example, if the PPV is more
than 15% and blood pressure is low, fluid infusion results in a clinically relevant in-
crease in CO of at least 15% (a percentage increase in CO that is considered ideal
for fluid responsiveness). Corroborating results of other studies, the area under the
receiver operator characteristic curve for PPV in dogs was significantly higher than
that for CVP or MAP, meaning that PPV is a better guide for fluid responsiveness
than is CVP or MAP (Fantoni et al, unpublished data, 2016). Pulse oximetry plethysmo-
graphic waveform amplitude is another accurate method to predict fluid responsive-
ness.18 This method is being studied by many veterinary investigators to establish if
respiratory variation above 15% as used in humans can also be used in dogs. The
great advantage of pulse oximetry plethysmographic waveform variation is that it
does not require an arterial line.
The static volumetric index also may play a role in monitoring patient fluid require-
ments during anesthesia. Ultrasound velocity dilution (UDCO) (COstatus, Transonic,
 Perioperative Fluid Therapy 427

Ithaca, New York) is a novel technique for determining CO that may resolve some of
the disadvantages of previous CO determination methods. UDCO is minimally inva-
sive, does not involve blood loss, and uses a physiologic noncumulative signal (saline
solution).19,20 This technique has been used in anesthetized dogs, pigs, and juvenile
horses with good success. UDCO also allows clinicians to monitor other preload volu-
metric variables, such as total end diastolic volume, that can have predictive value for
fluid responsiveness in patients.19,20

FLUID RATES DURING ANESTHESIA

Recently published guidelines for animals help veterinarians choose types of fluids
and adequate rates of administration in different clinical scenarios (2013 American An-
imal Hospital Association/American Association of Feline Practitioners Fluid Therapy
Guidelines for Dogs and Cats).21 In critically ill and high-risk surgical patients, howev-
er, a more refined method to evaluate and monitor fluid therapy should be used.
Choosing a fixed rate of fluid administration in the perioperative period or for main-
tenance is based on assumptions that take into consideration the amount of blood and
other organic fluids that can be lost during surgery or in a specific clinical situation. Ac-
cording to the guidelines published in 2013, fluid rates during anesthesia should start
at 3 mL/kg/h in cats and 5 mL/kg/h in dogs. In the ensuing hours, these rates can be
decreased according to the status of the patient, amount of blood loss, and type of
surgery.21 The previous recommendation to administer 10 mL/kg/h during procedures
with increased blood loss was abandoned because it was considered excessive.
Blood loss should be assessed intraoperatively to ensure that infused fluid restores
circulating blood volume adequately. Evaluation of blood loss during surgery by
means of weighing saturated gauze sponges, cotton balls, and swabs as well as blood
from suction bottles should be used more frequently in veterinary patients (Table 1).

Recommendation anesthesia fluid rates

Species Rate of Crystalloid Administration (mL/kg/h)

Dog 5
Cat 3

Table 1
Signs of shock according to the amount of blood loss

Classification of Shock
Grade I Grade II Grade III Grade IV
Heart rate Tachycardia (1) Tachycardia (11) Tachycardia (111) Tachycardia (1111)
Arterial Normal YY YYY YYYY
pressure
Pulse Normal Accelerate Weak Imperceptive
Mucous Pink Pale White Cyanotic
membrane
color
Blood loss (%) 10 20 30 40
Urinary output Normal Diminished Oliguria Anuria
Blood loss Hemorrhagic shock
428 Fantoni & Shih

Fluid replacement should be optimized individually, and aspects highlighted by the

International Fluid Administration Group,22 such as type of fluid administered, timing of
fluid administration and total amount of fluid administered should always be
considered.21
Another aspect that was reviewed by the task force and must be considered is the
rate of fluid administration in patients with sepsis. In this situation, to overcome the
physiologic response to hypotension (ie, fluid shifting to the heart and brain with renal
and splanchnic hypoperfusion), a large amount of fluid must be administered in a short
time period (20–30 mL/kg intravenous [IV] bolus followed by 5–10 mL/kg/h). A smaller
fluid bolus, such as 10 mL/kg to 15 mL/kg over 10 to 15 minutes, is another good op-
tion for initial fluid resuscitation. This volume should be prescribed according to indi-
vidual patient needs and fluid responsiveness.23 If the clinical or hemodynamic goal is
achieved in this period of time, fluid infusion is stopped, whereas if it is not effective, an
alternative therapeutic maneuver can be instituted to avoid administration of exces-
sive amounts of fluid.23

USE OF COLLOIDS VERSUS CRYSTALLOIDS DURING ANESTHESIA

Crystalloids are fluids that contain small solutes with molecular weights less than
500 Da. Crystalloids easily cross the intravascular barrier and equilibrate rapidly
with the extracellular space. Isotonic crystalloids are the used most commonly during
the perioperative period. A bolus of crystalloid increases intravascular volume, but its
effect lasts only 30 minutes. On the other hand, colloids contain large molecules
(>10,000 Da).24 Colloids do not readily cross the vascular barrier and tend to remain
in intravascular space for a longer period of time. Colloids increase colloid osmotic
pressure and may be a good alternative for animals with low oncotic pressure (eg, pa-
tients with hypoproteinemia). During anesthesia, colloids are more effective at
increasing patient blood pressure as compared to crystalloids.
Synthetic colloids, however, can have clinically relevant adverse effects. Synthetic
starches interfere with coagulation and platelet function, decrease von Willebrand fac-
tor and factor VIII, and can prolong bleeding time. Clinical bleeding is not observed in
healthy patients, except when high doses are used (>20 mL/kg). Colloids, however,
are not recommended in patients with coagulopathy.24 Synthetic starches also have
been associated with an increased risk of acute kidney injury in human patients.
This correlation has not been established in veterinary patients. One possible reason
for this disparity could be that dogs are more effective at metabolizing colloids than
are humans. The authors continue to recommend caution when using these products
in patients with underlying kidney disease.24

ELECTROLYTE DISTURBANCES

As a rule of thumb, it is always preferable to stabilize patients and correct electrolyte

disturbance before general anesthesia. Occasionally, however, the emergency nature
of a procedure does not allow sufficient time for stabilization. Electrolyte abnormalities
can potentiate the adverse effects of anesthetic drugs and lead to increased risk of ar-
rhythmias, hypotension, and decreased CO. Furthermore, general anesthesia blunts
sympathetic tone, which can unmask the adverse effects of electrolyte disturbances.
Electrolyte abnormalities are common in critically ill patients undergoing general anes-
thesia. The most common electrolyte disturbances faced in the perioperative period are
sodium disorders, potassium disorders, and calcium disorders.25 For example, hyper-
kalemia is commonly found in patients with uroabdomen (eg, those undergoing surgical
correction of ruptured bladder). Hypocalcemia can be encountered in patients
 Perioperative Fluid Therapy 429

undergoing cesarean section, and hypernatremia may be found in patients with a free
water deficit.

SODIUM ABNORMALITIES
Hypernatremia
Sodium (Na1) is crucial for electrical conduction in neurons, for muscle contraction, and
in acid-base balance. It is also vital for the maintenance of extracellular fluid volume.
Hypernatremia is defined as a sodium concentration greater than 160 mEq/L. Most an-
imals under anesthesia with hypernatremia have excessive free water loss. To decrease
the risk of vomiting and aspiration pneumonia, food and water are withheld for a variable
period of time before anesthesia. This enforced decrease in water consumption puts an
animal at risk for hypernatremia, especially if the animal has an ongoing increase in free
water loss. Examples include patients with high losses of sodium-poor fluid (eg, diarrhea,
vomiting, polyuria, or excessive panting). Also, patients with free water loss, such as
those with diabetes insipidus, or those with disturbances in the renin-angiotensin-
aldosterone system or vasopressin (antidiuretic hormone [ADH]) axis.25–28 Patients
with diabetes insipidus have lost their ability to reabsorb water in distal tubule and collect-
ing duct and are sensitive to water restriction. Such patients should have free access to
drinking water until close to the time of premedication for anesthesia. Unfortunately, a
clinically important cause of sodium imbalance is iatrogenic. Hypernatremia can occur
can if large quantities of sodium are administered, as can occur with excessive adminis-
tration of hypertonic saline or sodium bicarbonate.25–28
Hypernatremia can cause free water to move out of the brain and other intracellular
spaces and into the extracellular space. Given an appropriate amount of time, the
brain can protect itself and adapt physiologically to reverse this osmotic gradient.
These changes can initially increase sodium movement from the cerebrospinal fluid
into cerebral tissue and decrease hydrostatic pressure in the cerebrospinal fluid rela-
tive to the cerebral interstitium.
Within 24 hours, brain cells produce and accumulate organic solutes (so-called idio-
genic osmoles), such as inositol, glutamine, and glutamate. These idiogenic osmoles
attract water back into the cells and reverse the osmotic shift. Patients that develop
hypernatremia gradually usually are asymptomatic. On the other hand, severe acute
hypernatremia can produce marked clinical signs. Hypernatremia can lead to central
nervous system (CNS) signs, such as lethargy, weakness, head pressing, obtundation,
seizures, and coma. Such signs are not clinically apparent in patients already under
general anesthesia. Most of these neurologic signs (including seizures) are not
apparent to a clinician until a patient has recovered from general anesthesia, and by
this time substantial brain damage may have occurred.25
Overly aggressive treatment of hypernatremia also can cause CNS signs. Rapid
correction of hypernatremia can lead to excessive movement of free water back
into the intracellular space and cause substantial brain edema and neurologic signs.
In patients with chronic hypernatremia, serum sodium concentration should be
decreased gradually and in a consistent manner at a rate of 0.5 mEq/L/h to 1 mEq/
L/h. This gradual decrease in serum sodium concentration is important to prevent
development of brain edema. If cerebral edema occurs, it can be treated with mannitol
(1 g/kg IV over 15–30 minutes).25
As discussed previously, patients under anesthesia cannot show CNS signs. Further-
more, anesthetists often need to administer fluids at a rapid rate to ensure intravascular
volume expansion and correct hypotension. One solution to this problem is to administer
fluids that have a sodium concentration that matches that of the patient (within 6 mEq/L).
430 Fantoni & Shih

For example, if a patient has severe hypernatremia (eg, serum sodium concentration of
180 mEq/L) and must be anesthetized, a good perioperative fluid is one with a sodium
concentration of 175 to 180 mEq/L. The simplest way to create such a fluid is to add
NaCl to a balanced electrolyte solution. A 23.4% NaCl solution contains 4 mEq NaCl
per milliliter of solution, and hypertonic saline (7.4%) increases serum sodium concen-
tration by 1.3 mEq/L for every 1 mL administered. So, in this example, if 12 mL of the
23.4% hypertonic NaCl solution were added to a 1-L bag of lactated Ringer solution
(lactated Ringer solution has a sodium concentration of 130 mEq/L), the resultant solu-
tion would have a sodium concentration of 178 mEq/L.25–27 This solution allows anes-
thetists to be liberal with fluid therapy without fear of a rapid decrease in a patient’s
serum sodium concentration. Once a patient has recovered from anesthesia, a gradual
decrease in the patient’s serum sodium concentration can be reinstituted.

Hyponatremia
Hyponatremia is defined as a serum sodium concentration less than 135 mEq/L. Hypo-
natremia is not a common finding in anesthetized patients. It occurs when there has been
an increase in sodium loss together with an increase in the retention of free water. This
clinical scenario can occur in patients treated with loop diuretics, uncontrolled hypoa-
drenocorticism, or chronically decreased effective circulating volume. The decrease
in circulating volume stimulates endogenous ADH release, which increases water reab-
sorption from the distal renal tubule and collecting ducts. Patients with gastrointestinal
parasites, duodenal perforation, or pregnancy also can be affected by hyponatremia
mediated by nonosmotic stimulation of ADH release and what has been called pseudo-
hypodrenocorticism. Hyponatremia also can be due to iatrogenic administration of
sodium-poor fluids (eg, 0.45% NaCl, 5% dextrose in water). Pseudohyponatremia
can occur when effective osmoles are present in the blood at high concentration. The
most common causes of pseudohyponatremia are severe hyperglycemia and severe
hyperlipidemia. Pseudohyponatremia does not require specific treatment and serum so-
dium concentration normalizes once the effective osmoles are removed.25–28
Mild to moderate hyponatremia causes few clinical signs, but acute or severe hypo-
natremia (<120 mEq/L) can cause clinically relevant CNS signs. Acute hyponatremia
can be corrected by administering a sodium-rich fluid, such as 0.9% saline or hyper-
tonic saline (7.4%). Another proactive approach to achieve free water excretion is to
administer mannitol along with furosemide. The goal is a slow rate of increase in serum
sodium concentration (0.5–1 mEq/L/h). This approach is especially important in pa-
tients with chronic hyponatremia. A rapid increase in serum sodium concentration in
patients with chronic hyponatremia can lead to myelinolysis and CNS signs. Myelinol-
ysis is due to rapid shrinking of neuronal cells causing myelin disruption. Clinical signs
may only occur days later.29 For this reason, unless hyponatremia is acute, it usually is
preferable not to correct a patient’s serum sodium concentration during the short
intraoperative period. If a patient has hyponatremia and must undergo anesthesia,
the ideal fluid is one with a low sodium concentration.25 Most of the time, lactated
Ringer solution (sodium concentration, 130 mEq/L) is adequate. Clinicians should
avoid use of hypertonic saline or sodium bicarbonate and excessive use of heparin-
ized saline for catheter irrigation (Table 2).

POTASSIUM ABNORMALITIES
Hyperkalemia
Hyperkalemia is defined as a serum potassium concentration greater than 5.5 mEq/L.
The most common cause of intraoperative hyperkalemia is decreased urinary
 Perioperative Fluid Therapy 431

Table 2
Ideal fluids for anesthetized patients

Example of Perioperative
Sodium Disturbance Fluid Sodium
Hypernatremia, acute (h) Lactated Ringer solution or 130–140 mEq/L
Plasma-Lyte R
Hypernatremia, chronic (d) Lactated Ringer solution Match patient’s serum
with appropriate volume sodium concentration
of hypertonic saline
(23.4%)a
Hyponatremia, acute (h) Normal saline solution 154 mEq/L
(0.9% NaCl)
Hyponatremia, chronic (d) Lactated Ringer solution 130 mEq/L
a
Hypertonic saline (23.4%) increases Na concentration by 4 mEq/L for every 1 mL administered.
Hypertonic saline (7.4%) increases Na concentration by 1.3 mEq/L for every 1 mL administered.

excretion due to prerenal, renal, or postrenal disease. Patients with uroabdomen and
complete urinary obstruction can present with life-threatening hyperkalemia. Other
causes of hyperkalemia include acute tumor lysis syndrome, severe intraoperative
cell damage, diabetes mellitus with ketoacidosis, and some gastrointestinal diseases
(eg, trichuriasis and perforated duodenal ulcer). Iatrogenic cause of hyperkalemia can
occur with administration of potassium-containing fluids, excessive amounts of
angiotensin-converting enzyme inhibitors, or use of expired banked blood for transfu-
sion. Hyperkalemia can lead to CNS signs, muscle weakness, hypotension,
decreased CO, and severe bradyarrhythmias. ECG abnormalities include decreased
P waves, atrial standstill, and prolonged QRS complexes and PR intervals. If not
treated, severe hyperkalemia can result in ventricular tachycardia, ventricular fibrilla-
tion, and asystole.29–31
Treatment of moderate to severe hyperkalemia includes administration of drugs that
shift potassium back into the intracellular space (eg, insulin, sodium bicarbonate, or
terbutaline) or that minimize cardiovascular signs (eg, calcium gluconate). Sodium bi-
carbonate has not been shown effective in cohort studies of humans. Furthermore, the
use of sodium bicarbonate during anesthesia can have clinically important adverse ef-
fects, such as respiratory acidosis, a shift in the oxygen-hemoglobin dissociation
curve to the left, and paradoxical intracellular acidosis.
It is usually recommended that the fluid of choice be a potassium-free fluid, such as
normal saline (0.9% NaCl). Normal saline, however, has an excessive amount of chlo-
ride (154 mEq/L) that can lead to dilutional acidosis (hyperchloremic acidosis), which
may further worsen existing metabolic acidosis. Use of a more balanced solution,
such as lactated Ringer solution, may be as effective in decreasing the serum potas-
sium concentration and more effective in correcting metabolic acidosis. If the surgery
is an abdominal exploration, abdominal lavage can be one of the most effective way to
decrease serum potassium concentration (Table 3).29–31
Hypokalemia
Hypokalemia by defined as a serum potassium concentration less than 3.5 mEq/L. Low
serum potassium concentration can lead to CNS abnormalities, muscle weakness, and
arrhythmias. ECG abnormalities in dogs include ST segment depression and prolonga-
tion of the QT interval. An increase in P-wave amplitude and prolongation of PR interval
also may be observed. Common causes of hypokalemia during the perioperative period
432 Fantoni & Shih

Table 3
Treatment options for hyperkalemia

Drug Recommended Dosage

Calcium gluconate, 10% 0.5–1 mL/kg IV over 10 min
Sodium bicarbonate 1–2 mEq/kg IV over 15–30 min
25% Dextrose with regular insulin 0.1–0.3 U/kg regular insulin with dextrose 1–2 g/kg IV
Terbutaline 0.01 mg/kg IV over 5–10 min

include metabolic acidosis, diabetic ketoacidosis, hyperaldosteronism, chronic liver

disease, and severe diarrhea. Hypokalemia can occur iatrogenically by overdose of in-
sulin, overdose of a b-agonist drug, or use of a potassium-wasting diuretic (eg, loop
diuretics).29
For clinically relevant hypokalemia, treatment consists of correcting the underlying
disease process and potassium replacement. Mild hypokalemia should not be cor-
rected during the intraoperative time period. Moderate to severe hypokalemia war-
rants treatment. The rate of potassium infusion should not exceed 0.5 mEq/kg/h. It
is important to closely monitor the ECG during potassium infusion. It is also important
to clearly label the potassium-containing fluid with its potassium concentration and
the maximally permissible fluid infusion rate. During anesthesia, the fluid administra-
tion rate usually is higher than that used in awake patients and sometimes a bolus
of fluids is administered. If an extension set is used during anesthesia, flushing the
line or administering bolus of another drug can inadvertently result in a bolus of fluid
with a high concentration of potassium, which could result in bradyarrhythmias, a
decrease in CO, and even cardiac arrest (Table 4).30

CALCIUM DISTURBANCES
Hypocalcemia
Most animals show few signs of hypocalcemia until it becomes severe (ie, a total
serum calcium concentration <6.5 mg/dL or a serum ionized calcium
concentration <1 mmol/L). Patients with low serum total calcium concentration but
normal serum ionized calcium concentration do not require treatment.32 Clinical signs
of severe hypocalcemia include muscle weakness, muscle tremors, CNS abnormal-
ities, seizures, hypotension, and decreased CO. Hypocalcemia can be a result of
hypoalbuminemia, renal disease, hypoparathyroidism, pancreatitis, eclampsia, or
acute tumor lysis syndrome. Iatrogenic causes of hypocalcemia include excessive
administration of citrate-containing anticoagulant (eg, after several blood product
transfusions).33

Table 4
Levels for potassium replacement

Serum Potassium Potassium Chloride Maximal Infusion

Concentration (mEq/L) Added to 1 Liter of Fluids (mEq) Rate (mL/kg/h)
<2.0 80–60 5
2.1–2.5 60–40 8
2.5–30 20 10
3.1–4.0 No treatment intraoperatively NA
 Perioperative Fluid Therapy 433

Intraoperative treatment of mild to moderate hypocalcemia usually is not necessary.

Severe hypocalcemia can be treated with calcium gluconate supplementation (ie,
10% calcium gluconate administered at a dosage of 0.5–1 mL/kg IV over 15–30 mi-
nutes).32 The ECG should be closely monitored during calcium administration. Exces-
sively rapid infusion can lead to severe bradycardia. Calcium should be administered
separately from other drugs and never in the same line as a blood transfusion. Calcium
administration can precipitate with several drugs if used during a constant rate infu-
sion. It also binds to anticoagulants and increase the risk of clotting in blood
products.32,33

SUMMARY

Fluid administration during the perioperative time period is different from fluid admin-
istration when a patient is awake. A more acute procedure requires more emphasis on
volumetric monitoring and careful assessment of serum electrolytes. Furthermore, to
ensure patient safety and well-being, the perioperative time period should not be used
for correction of dehydration and electrolyte abnormalities unless doing so is unavoid-
able as a consequence of a patient’s clinical condition.

REFERENCES

1. Kelm DJ, Perrin JT, Cartin R, et al. Fluid overload in patients with severe sepsis
and septic shock treatd with early goal directed therapy is associated with acute
need for fluid-related medical intervention and hospital death. Shock 2015;43(1):
68–73.
2. Aldrich J. Shock fluid and fluid challenge. In: Silverstein D, Hopper K, editors.
Small animal critical care medicine. 2nd edition. Philadelphia: Saunders Elsevier;
2015. p. 276–80.
3. Mellema M. Cardiac output monitoring. In: Silverstein D, Hopper K, editors. Small
animal critical care medicine. Philadelphia: Saunders Elsevier; 2009. p. 894–8.
4. Tantalean JA, Leon RJ, Santos AA, et al. Multiple organ dysfunction syndrome in
children. Pediatr Crit Care Med 2003;4:181–5.
5. Hatherill M, Waggie Z, Purves L, et al. Mortality and the nature of metabolic
acidosis in children with shock. Intensive Care Med 2003;29:286–91.
6. Perel A. Bench-to-bedside review: the initial hemodynamic resuscitation of the
septic patient according to surviving sepsis campaign guidelines–does one
size fit all? Crit Care 2008;12:223.
7. Valverde A. Balanced anesthesia and constant rate infusion. Vet Clin North Am
Equine Pract 2013;29:89–122.
8. Fantoni DT, Otsuki DA, Ambrosio AM, et al. A comparative evaluation of inhaled
halothane isoflurane and sevoflurane during acute normovolemic hemodilution in
dogs. Anesth Analg 2005;100:1014–9.
9. Geeraedts LM Jr, Kaasjager HA, van Vugt AB, et al. Exsanguination in trauma: a
review of diagnostics and treatment options. Injury 2009;40:11–20.
10. Marik PE, Baram M, Vahid B. Does central venous pressure predict fluid respon-
siveness? a systematic review of the literature and the tale of seven mares. Chest
2008;134:172–8.
11. Kumar A, Anel R, Bunnell E, et al. Pulmonary artery occlusion pressure and central
venous pressure fail to predict ventricular filling volume, cardiac performance, or
the response to volume infusion in normal subjects. Crit Care Med 2004;32:691–9.
12. Fantoni DT, Auler J Jr, Futema F, et al. Intravenous administration of hypertonic
sodium chloride with dextran solution or isotonic sodium chloride solution for
434 Fantoni & Shih

treatment of septic shock secondary to pyometra in dogs. J Am Vet Med Assoc

1999;215(9):1283–7.
13. Durairaj L, Schmidt GA. Fluid therapy in resuscitated sepsis: less is more. Chest
2008;133:252–63.
14. Kim JJ, Dreyer WJ, Chang AC, et al. Arterial pulse wave analysis: an accurate means
of determining cardiac output in children. Pediatr Crit Care Med 2006;7:532–5.
15. Tibby SM, Murdoch IA. Monitoring cardiac function in intensive care. Arch Dis
Child 2003;88:46–52.
16. Sant ‘Ana AJ, Otsuki D, Fantoni D, et al. Use of pulse pressure variation to esti-
mate changes in preload during experimental acute normovolemic hemodilution.
Minerva Anestesiol 2012;78:426–33.
17. Auler J, Galas F, Hajjar L, et al. Arterial pulse pressure variation predicting fluid
responsiveness in critically ill patients. Shock 2008;30:18–22.
18. Cannesson M, Besnard C, Jacques D, et al. Relation between respiratory varia-
tion in pulse oximetry plethysmographic waveform amplitude an arterial pulse
pressure in ventilated patients. Crit Care 2005;9:R562–8.
19. Shih A, Giguere S, Sanchez LC, et al. Determination of cardiac output in neonatal
foals by ultrasound velocity dilution and its comparison to the lithium dilution
method. J Vet Emerg Crit Care (San Antonio) 2009;19:438–43.
20. Vigani A, Shih A, Queiroz P, et al. Quantitative response of volumetric variables
measured by a new ultrasound dilution method in a juvenile model of hemorrhag-
ic shock and resuscitation. Resuscitation 2012;83:1031–7.
21. Davis H, Jensen T, Shafford H, et al. 2013 AAHA/AAFP fluid therapy guidelines for
dogs and cats. J Am Anim Hosp Assoc 2013;49:149–59.
22. Navarro LHCN, Bloomstone JA, Auler JOC Jr, et al. Perioperative fluid therapy: a
statement from the International Fluid Optimization Group. Preoperative Med
2015;4(3):2–20.
23. Glassford N, Eastwood G, Bellomo R. Physiological changes after fluid bolus
therapy in sepsis. Crit Care 2014;27:696.
24. Adamik K, Yozova I, Regenscheit N. Controversies in the use of hydroxyethyl
starch solution in small animal emergency and critical care. J Vet Emerg Crit
Care (San Antonio) 2015;25(1):20–47.
25. Burkitt J. Sodium disorder. In: Silverstein D, Hopper K, editors. Small animal crit-
ical care medicine. 2nd edition. Philadelphia: Saunders Elsevier; 2015. p. 224–9.
26. DiBartola SP. Disorders of sodium: hypernatraemia and hyponatraemia. J Feline
Med Surg 2001;3(4):185–7.
27. de Morais HA, DiBartola SP. Hypernatremia: a quick reference. Vet Clin North Am
Small Anim Pract 2008;38(3):485–9.
28. Ellison DH. Disorders of sodium and water. Am J Kidney Dis 2005;46(2):356–61.
29. Schaer M. Therapeutic approach to electrolyte emergencies. Vet Clin North Am
Small Anim Pract 2008;38(3):513–33.
30. Riordan L, Schaer M. Potassium disorder. In: Silverstein D, Hopper K, editors.
Small animal critical care medicine. 2nd edition. Philadelphia: Saunders Elsevier;
2015. p. 229–35.
31. DiBartola S, DeMorais H. Perioperative fluid therapy. In: DiBartola SP, editor. Fluid
therapy in small animal practice. 4th edition. Philadelphia: Saunders; 2006. p. 92–6.
32. Schenck P, Chew D. Anesthesia fluid therapy. In: DiBartola SP, editor. Fluid ther-
apy in small animal practice. 4th edition. Philadelphia: Saunders; 2006. p. 120–5.
33. Green T, Chew D. Calcium disorder. In: Silverstein D, Hopper K, editors. Small an-
imal critical care medicine. 2nd edition. Philadelphia: Saunders Elsevier; 2015.
p. 233–40.