(Chest. 1999;115:106S-112S.)
© 1999
American College of Chest Physicians
Intraoperative Fluid Management—What and How Much?*
Myer H. Rosenthal, MD, FCCP
* From the Department of Anesthesia, Stanford University School of Medicine, Stanford, CA.
Correspondence to: Myer H. Rosenthal, MD, FCCP, Department of Anesthesia—Room H3580, Stanford University School of Medicine, Stanford, CA 94305; e-mail: mhr{at}leland.stanford.edu
 |
Abstract
|
|---|
An approach to intraoperative fluid management based on a monitored
physiologic application of the Starling principles of cardiac function
is recommended to individualize therapy to optimize hemodynamic
function and tissue perfusion. The complexity of intraoperative fluid
administration, beginning with preoperative cardiovascular function
followed by innumerable intraoperative considerations, including
anesthetic pharmacology, positive pressure ventilation, operative site,
and surgical technique may lead to serious intraoperative and
postoperative complications. Emphasis must be given to intraoperative
fluid shifts resulting in hidden fluid loss and intravascular
hypovolemia that must be replaced. Explanations for this fluid
redistribution have included tissue trauma, endotoxemia, and
proinflammatory cytokines with resultant increased capillary
permeability.
|
Introduction
|
|---|
In
addition to quantitative considerations, a number of qualitative
considerations also are relevant to intraoperative fluid therapy.
Oxygen-carrying capacity and coagulation often require blood component
administration. Choice of crystalloid has potential impact on
electrolyte and acid-base equilibrium. The value of colloid
administration remains controversial with little evidence for its
historically postulated benefit of minimizing lung water accumulation.
The use of dextrose solutions, particularly involving cerebral
pathologic conditions and surgery, is not recommended owing to the
possibility of increasing cerebral acidosis.
Maintenance of renal function, avoidance of lung water accumulation,
minimizing splanchnic and hepatic circulatory insufficiency, and
ensuring GI integrity are among the principal goals of perioperative
hemodynamic stability. The maintenance of proper intravascular volume
(IVV) and ventricular preload is the foundation for cardiovascular
function. The principles set forth by Ernest Starling at the beginning
of the 20th century form the basis for our understanding of hemodynamic
physiology, the consequences of pathophysiologic mechanisms, and the
implications for treatment.1
2
Figure 1
,
using Starling’s concepts, summarizes the pathophysiologic
consequences of alterations in preload and contractility and the
expected responses to varying therapy. Differing states of myocardial
contractility, including hypodynamic sequelae of cardiac failure and
hyperdynamic sequelae as a consequence of early septic shock are
reflected in the series of Starling curves.
View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. A series of Starling ventricular function curves
(bottom: hypodynamic; middle: normal;
top: hyperdynamic) identifying pathophysiologic processes
and expected response to indicated therapy.
|
|
The maintenance of preload as the ventricular end-diastolic volume
(VEDV) is the initial approach in rectifying hemodynamic instability. A
number of factors must be addressed as general considerations in a
physiologic approach to therapy adjusting preload. Recognition of the
critical nature of VEDV as the true definition of preload and,
therefore, the potential misrepresentation of the pressure correlates
of VEDV as monitored estimates of preload must be understood. Figure 2
demonstrates the effects of normal differences in compliance of the two
ventricles on the relationships of VEDV to ventricular end-diastolic
pressure (VEDP). Figure 3
further shows the differences in the Starling relationship when using
central pressure correlates of VEDP to represent preload. Currently
available monitoring produces further limitations when using central
venous pressure to represent right VEDP and pulmonary artery occlusive
pressure to represent left VEDP.3
4
Understanding these
limitations and, equally important, relating these monitored indexes to
some identifiable indication of ventricular output allows the clinician
to structure a monitored physiologic approach to therapy as the best
means for maintaining proper hemodynamic function.5
Historical and physical indexes of adequacy of IVV and perfusion may
suffice in most surgical patients undergoing routine procedures in the
absence of complex perioperative pathophysiologic processes. However,
in more complex procedures, in patients with preexisting cardiovascular
compromise, increasingly sophisticated monitoring with central venous
and pulmonary artery catheterization and/or transesophageal
echocardiography may be necessary to more accurately assess
cardiovascular parameters of preload, afterload, and contractility to
optimize tissue perfusion.
View larger version (13K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. The differences in Starling ventricular function
curves when using pressure correlates of VEDV of the right and left
ventricles. PAWP (pulmonary artery wedge pressure) for left VEDP and
CVP (central venous pressure) for right VEDP.
|
|
Proper perioperative fluid management is the most important aspect
governing hemodynamic function in the surgical patient. Discussion of
this topic is best divided into quantitative and qualitative
considerations. Although it is of major importance to maintain
electrolyte and acid-base equilibrium and to provide sufficient
oxygen-carrying capacity and coagulation factors, the maintenance of
sufficient IVV is of paramount importance.
|
Quantitative Considerations
|
|---|
Table 1
lists the principal factors governing intraoperative fluid volume
administration. The implications of preoperative IVV status should not
be ignored in initiating intraoperative therapy. The existence of
congestive heart failure and pulmonary edema is a major cause of
perioperative morbidity and mortality.6
7
Hypovolemia is
often associated with chronic hypertension due to persistent increase
in systemic vascular resistance (SVR).
Bowel preparation with hyperosmolar solutions, vomiting, diarrhea,
polyuria, diaphoresis, hemorrhage, burns, and inadequate intake are all
common causes of preoperative hypovolemia. Often unrecognized due to
redistribution of IVV without evidence of external loss, and confusion
as a consequence of stable or even increased body weight, hypovolemia
can be present due to bowel abnormalities, sepsis, ARDS, ascites,
pleural effusions, and circulating humoral factors. In common, these
processes are often accompanied by increased capillary permeability
resulting in loss of IVV to interstitial and other compartments.
Correction of preoperative fluid deficit when time allows is critical
to minimizing severe hypotension and hypoperfusion during anesthetic
induction.
Regardless of extreme caution, the induction of anesthesia results in
decreased venous return. Commonly employed IV induction agents,
including thiopental sodium and propofol, produce a decrease in SVR and
may also reduce myocardial contractility. Alternative agents often
chosen to maintain hemodynamic stability, including etomidate,
ketamine, and high-dose opioids, may also result in hypotension by
depressing the patient’s endogenous sympathoadrenal mechanism. Muscle
relaxants used to facilitate intubation and provide improved surgical
exposure may release histamine (curare, atracurium) and decrease SVR or
produce venous pooling due to loss of muscle tone. All volatile
inhalation anesthetic agents reduce SVR and decrease contractility with
the most commonly used, isoflurane (Forane®), producing profound
vasodilation. Positive pressure ventilation instituted immediately
following anesthetic induction is particularly deleterious in the
hypovolemic patient due to preload reduction. Regional anesthetic
techniques, epidural and subarachnoid (spinal) blocks, may be chosen to
minimize the anesthetic intervention or to augment a general anesthetic
and provide postoperative pain control. The accompanying sympathetic
nervous blockade extending two to four levels higher than the sensory
block will produce considerable vasodilation with severe hypotension in
the hypovolemic patient.
Beyond the effects of anesthesia is the surgery itself. Of obvious
consideration is hemorrhage, loss of ascites or pleural fluid, or the
administration of large quantities of fluid into sites where absorption
is excessive, such as with prostate resection, influencing IVV. Patient
positioning, surgical packing and retraction, and altered temperature
have varying impact on venous return and vascular tone. Less obvious,
but of potentially greater import, are fluid shifts with redistribution
or loss from operative sites, both abdominal and thoracic.
The past 40 years has witnessed considerable controversy and eventual
alteration in approach to intraoperative fluid management as it
involves abdominal and thoracic surgery. Prior to current understanding
and documentation of redistribution of IVV, it was the belief than salt
and water retention during surgery dictated fluid restriction to avoid
fluid overload.8
Support for this view was based on
increased levels of aldosterone and antidiuretic hormone accompanying
surgery and the implications for fluid and salt retention. That
aldosterone as a consequence of the surgical stress response is
elevated is well documented. Furthermore, atrial naturetic peptide
would be expected to decrease in response to positive pressure
ventilation, further resulting in oliguria and decreased renal fluid
loss. Evidence regarding loss of IVV to the interstitium and
intracellular spaces aggravated by evaporative loss to the environment
was demonstrated by Shires et al9
and subsequently Roberts
et al.10
After some controversy over the method used to
document a loss in extracellular volume (ECV) and IVV,11
agreement was reached that replacement of this "third-spaced" fluid
was and is necessary to maintain perfusion. A number of studies have
subsequently appeared indicating the necessity to provide sufficient
fluid administration if organ hypoperfusion, particularly renal
insufficiency, is to be avoided following major intra-abdominal
surgery.12
13
For many years, particularly preceding the
availability of invasive monitoring of preload and cardiac output,
clinicians relied on guidelines for fluid management predicated on the
surgical site and duration of exposure.14
15
These
guidelines indicated a need for 10 to 15 mL/kg/h of crystalloid
administration in excess of maintenance and replacement for blood loss
while the peritoneal cavity was exposed. For thoracic procedures, the
amount was somewhat less at 5 to 7.5 mL/kg/h. Although these numbers
are no longer strictly adhered to, they serve to provide awareness of
the potential magnitude of the ECV deficit. With the introduction of
improved hemodynamic monitoring and the recognition of the implication
of surgical technique, individualized approach now guides the clinician
in determining the proper amount of fluid based on specific patient
physiology, surgical procedure, and surgeon’s technique as well as
relationship to anesthetic pharmacology.
As indicated above, in addition to maintenance, fluid administration
replacement for blood loss must also be considered. Blood loss is
accompanied by redistribution and subsequent loss of additional ECV and
intracellular volume thus often requiring excess replacement more than
that provided by a milliliter per milliliter replacement with blood. In
replacing blood loss with crystalloid, a 3:1 ratio of crystalloid
administration to blood loss is often required to maintain IVV.
The fate of added fluid administration to maintain IVV and ECV
has avoided precise identification. Studies previously discussed have
postulated evaporative loss or loss into the open abdominal and
thoracic cavities. This explanation is difficult to reconcile against
clinical experience demonstrating a need for fluid restriction and
diuresis 18 to 48 h following major abdominal surgery to avoid
pulmonary edema and congestive heart failure. Such experience supports
the impression that intraoperatively administered fluid is
redistributed to interstitial and intracellular spaces with subsequent
mobilization in the postoperative period. This redistribution is likely
a consequence of alterations in vascular permeability. The etiology of
increased permeability may be theorized from data demonstrating
increases in proinflammatory cytokines, including interleukins 6 and 8
and tumor necrosis factor-
as a result of the stress response to
major surgery.16
17
Although there is insufficient
documentation owing to lack of suitable analytical tools, endotoxemia
as a consequence of absorption from ischemic and/or traumatized
intestinal mucosa may also be implicated.
Regardless of the mechanisms involved, the last 25 years has
provided sufficient evidence to support the necessity for the
intraoperative administration of fluids in whatever quantity necessary
to provide adequate levels of preload to maintain ventricular output.
In cases of myocardial hypocontractility, fluids should be infused to
provide that minimum level of VEDP (pulmonary artery occlusive
pressure = 12 to 15 mmHg) that allows for the safe administration
of inotropic support. The need for postoperative fluid restriction and
diuresis should not be used as sole evidence of excessive
intraoperative fluid administration, but rather as an indication of
altered time- and disease-related pathophysiology.
|
Qualitative Considerations
|
|---|
Acknowledging the primary importance of assessing "how much"
volume to administer, further consideration is to evaluate "what"
fluid is preferable. Unfortunately, preferences chosen by those
providing intraoperative care are often guided by emotion and
convenience neglecting both documented evidence of superiority or the
physiologic responses expected from such choices. Table 2
identifies the qualitative considerations most often encountered in
determining the fluid to administer.
|
Oxygen-Carrying Capacity
|
|---|
The need to provide sufficient oxygen delivery
(DO2) is expressed by the following formula:
DO2 = content of arterial oxygen
(CaO2) x cardiac output (CO).
The CaO2 is a function of the hemoglobin
level, arterial oxygen saturation (SaO2), and
to a minimal extent, dissolved oxygen. Thus, maintaining RBC volume is
of necessity to maintain CaO2 and thus,
DO2. Alternatives to RBCs, including hemoglobin
solutions, liposome-encapsulated hemoglobin and perfluorocarbons
continue to be evaluated; however, none have achieved efficacy and
safety sufficient for widespread use.18
Although donor
screening and donor blood testing have markedly reduced the risks of
hepatitis and HIV infection, transfusion reactions, donor-recipient
mismatch, contaminated products, and limited supply require careful
substantiation of need before administration of blood. In determining
the need for transfusion of RBCs, the clinician must consider the
patient’s cardiac status (ie, ability to raise CO to
maintain DO2), pulmonary status (ie,
ability to maintain SaO2), and expected oxygen
demand (oxygen uptake [
O2])
(ie, ability of DO2 to match
O2) in order to assess the minimum
satisfactory level for hemoglobin. Further value should be given to the
body’s ability to unload additional oxygen at the cellular level, thus
reducing mixed venous oxygen saturation as an acceptable compensatory
mechanism. Well-tolerated increased CO, decreased mixed venous oxygen
saturation, and maintenance of high levels of
SaO2 should be viewed as reasonable, if not
preferred, alternative means of maintaining tissue oxygenation to RBC
transfusion. The relevance of O2,
however, should not be ignored as increased
O2 should be expected in the
postoperative period as a consequence of a postsurgical hypermetabolic
state.
|
Coagulation Factors
|
|---|
Depletion of coagulation factors inducing bleeding may necessitate
the administration of blood products, including fresh-frozen plasma,
platelets, or cryoprecipitate. The causes for depletion include
hemodilution, intravascular consumption, bone marrow depression,
hypersplenism, and synthetic dysfunction. Additionally, platelet
dysfunction due to endogenous (uremia) or exogenous (salicylates,
nonsteroidal anti-inflammatory drugs) factors may be present.
Regardless of cause, identification and documentation of the specific
abnormality should precede factor or platelet
replacement.19
The most common intraoperative coagulopathy
is dilutional thrombocytopenia, commonly occurring with either
large-volume (1.5 to 2 times blood volume) RBC transfusion or
large-volume crystalloid/colloid administration.20
Factor
deficiency in the absence of hepatic dysfunction is rare since stored
blood retains 20 to 30% activity of the highest risk factors VII and
VIII (labile factors), sufficient for coagulation. Documented
thrombocytopenia (< 50,000 to 75,000) should lead to platelet
transfusion in the surgical patient. Prolonged international
norma-lized ratio (> 1.2 to 1.4) warrants consideration for
fresh-frozen plasma, and fibrinogen level < 100 mg/dL in the presence
of bleeding may indicate the need for cryoprecipitate.
|
Colloid
|
|---|
The greatest controversy over the past 30 years continuing to the
present is that of colloid vs crystalloid solutions for optimal fluid
replacement in surgical patients. Ernest Starling laid the foundation
for understanding the role of oncotic forces in determining the
movement of fluid across capillary membranes.21
Figure 4
depicts the Starling equilibrium identifying the colloid oncotic
pressure (COP), of which 90% is dependent on serum albumin, as the
only pressure favoring fluid retention in the capillary. In-depth
discussion and debate over the merits of colloid solutions began with
the studies of Gaar et al22
and Guyton and
Lindsey23
demonstrating increased accumulation of lung
water accompanying reduction in COP. Opposing views were presented
indicating that increased capillary permeability permitted the free
diffusion of colloid particles minimizing their impact on
COP.24
Some studies indicated worse outcomes with colloid
as a result of decreased lymphatic capability to remove large molecular
weight particles from the pulmonary interstitium.25
A
number of conflicting studies have been published leaving the clinician
able to cite support for either crystalloid or colloid solutions.
Velanovich26
reported a meta-analysis of eight randomized
clinical trials comparing colloid vs crystalloid fluid resuscitation.
His findings showed a 12.3% difference in mortality favoring
crystalloid therapy in trauma patients and 7.8% favoring colloid in
nontrauma patients. He concluded that in patients with increased
capillary permeability, colloid administration may be deleterious, yet
in those with intact capillary permeability, colloid may be
efficacious. As a number of animal models and human studies have shown
lack of correlation of COP and no relationship between type of
fluid infused to lung water accumulation,27
28
I
have concluded that there is little support for either choice to
minimize postoperative pulmonary complications. There is consensus that
colloid solutions have a longer intravascular half-life, thus requiring
less total volume than crystalloid. This has been postulated to result
in less peripheral and intestinal edema following colloid
administration. Renal function has been investigated as to the effects
of colloid vs crystalloid fluid resuscitation showing a relationship of
increased COP to renal dysfunction in cardiac surgical
patients29
and better renal function in trauma patients
and shock primate studies following crystalloid
administration.30
31
An objective review of available
studies provides little guidance as to the best choice between
crystalloid and colloid. Available colloids currently include dextran,
albumin (5% or 25%), and 6% hydroxy-ethyl starch (hetastarch).
Dextran as a low-molecular-weight mixture is often reserved for
improving peripheral perfusion in patients with vascular insufficiency.
Hetastarch should not exceed 20 mL/kg due to possibilities of platelet
and reticuloendothelial dysfunction.
|
Electrolyte and Acid-Base Balance
|
|---|
Although assessment and correction of abnormalities of calcium
(Ca), magnesium, and phosphate should be part of complete evaluation
and care, sodium (Na), potassium (K), and chloride (Cl) are the
principal electrolytes affected by the choice of crystalloid
solution.32
Insufficient intravascular half-life and
hyponatremia usually mitigate against the use of saline solutions
< 0.9% (normal saline solution [NSS]) for intraoperative
fluid resuscitation and maintenance. The commonly used solutions are
NSS and the balanced salt solutions of which lactated Ringer’s (LR) is
the prime example. Most balanced salt solutions include K and thus must
be used with caution in hyperkalemic patients, particularly those with
renal dysfunction. Also, Ca present in LR mitigates against its use in
proximity to citrated blood transfusions.
The major sequela in choosing NSS vs LR or other balanced salt solution
is the effect on extracellular sodium to chloride ratio and acid-base
balance. As indicated earlier in this discussion, aldosterone is
increased during and immediately following surgery, thus increasing the
distal renal tubular absorption of sodium. This increased tubular
avidity for sodium requires either the accompanying absorption of a
negative ion (Cl) or the secretion of a hydrogen or K ion to maintain
renal tubular elecrical neutrality. Should the amount of CI related to
Na increase, as would occur with large-volume NSS administration,
hydrogen and K secretion would be minimized with a resultant
hyperchloremic-induced nongap metabolic acidosis. LR administration,
however, has a more physiologic (balanced) Na to Cl relationship and
will not result in acidosis. Large-volume LR administration may result
in a metabolic alkalosis postoperatively due to increased bicarbonate
arising from metabolism of lactate. It is my practice to alternate NSS
and LR provided K and Ca administration is acceptable.
|
Nutrition, Glucose, Metabolism, and Cerebral Abnormalities
|
|---|
IV nutrition utilizing dextrose solutions should be continued as
disruption of the infusion could result in hypoglycemia. If
discontinued for any reason, 10% dextrose infusion should be
considered with frequent blood glucose evaluation. Avoidance of
hyperglycemia and hypoglycemia is of increased concern in patients with
diabetes mellitus and end-stage liver disease, respectively. In the
absence of diseases that influence glucose metabolism,
dextrose-containing solutions are generally omitted.
Hyperglycemic-induced hyperosmolality, osmotic diuresis, and cerebral
acidosis are sequelae of dextrose administration to be avoided.
Cerebral abnormalities and the potential of cerebral surgical ischemia
create an environment where metabolism of glucose in the absence of
oxygen (minimally stored in the brain) leads to increasing cerebral
acidosis. As cerebral acidosis continues, death of brain cells occur
with ir-reversible cerebral damage. Thus, avoidance of
dextrose-containing solutions, unless necessary to treat hypoglycemia,
is recommended.
|
Conclusions
|
|---|
The intraoperative management of fluid therapy has great potential
for influencing intraoperative and postoperative morbidity and
mortality. Awareness of preoperative hemodynamic status, particularly
as it influences the preload/ventricular output relationship, is
critical in avoiding serious cardiovascular complications early in the
course of anesthetic induction and maintenance. The implications of
anesthetic pharmacology, positioning, thermoregulation, ventilatory
support, surgical manipulation, operative site, duration, tissue
trauma, and blood loss must be appreciated in determining how much
fluid to be administered. Providing sufficient IVV and preload is
essential for adequate vital organ perfusion. Although quantitative
considerations are of primary concern in fluid management, qualitative
considerations involving oxygen-carrying capacity, coagulation,
electrolyte and acid-base balance, and glucose metabolism are also of
critical importance. Controversy over the choice of colloid vs
crystalloid solutions continues to the present. A definitive answer as
to the best solution for resuscitation and maintenance does not exist.
Personal preference, cost, and most importantly, individualized
physiologic evaluation and approaches will guide clinical practice.
|
References
|
|---|
- Patterson, SW, Starling, EH (1914) On the mechanical factors which determine the output of the ventricles. J Physiol 48,357-359
- Starling, EH (1965) The Linacre lecture on the law of the heart. Chapman, CB Mitchell, JH eds. Starling on the heart ,119-147 Dawsons of Pall Mall London, UK.
- Mihm, FG, Rosenthal, MH (1992) Central venous catheterization. Benumof, JL eds. Clinical procedures in anesthesia and intensive care ,339-374 JB Lippincott New York, NY.
- Mihm, FG, Rosenthal, MH (1992) Pulmonary artery catheterization. Benumof, JL eds. Clinical procedure in anesthesia and intensive care ,405-441 JB Lippincott New York, NY.
- Rodriguez, RM, Rosenthal, MH (1997) Etiology and pathophysiology of shock. Murray, MJ Coursin, DB Pearl, RGet al eds. Critical care medicine—perioperative management ,185-198 Lippincott Raven New York, NY.
- Goldman, L, Caldera, DL, Nussbaum, SR, et al (1977) Multifactorial index of cardiac risk in non cardiac surgical procedures. N Engl J Med 297,845-850[Abstract]
- Lowell, JA, Schifferdecker, C, Driscoll, DF, et al (1990) Postoperative fluid overload: not a benign problem. Crit Care Med 18,728-733[ISI][Medline]
- Moore, FD (1959) Metabolic care of the surgical patient. WB Saunders Philadelphia, PA.
- Shires, T, Williams, J, Brown, F (1961) Acute changes in extracellular fluids associated with major surgical procedures. Ann Surg 154,803-810[ISI][Medline]
- Roberts, JP, Roberts, JD, Skinner, C, et al (1985) Extracellular fluid deficit following operation and its correction with Ringers lactate. Ann Surg 202,1-8[CrossRef][ISI][Medline]
- Roth, E, Lax, LC, Maloney, JV (1969) Ringers lactate solution and extracellular fluid volume in the surgical patient: a critical analysis. Ann Surg 169,149-164[ISI][Medline]
- Bush, HL, Logerfo, FW, Weisel, RD, et al (1977) Assessment of myocardial performance and optimal fluid loading during elective abdominal aortic aneurysm resection. Arch Surg 112,1301-1306[Abstract]
- Bush, HL, Huse, JB, Johnson, WC, et al (1981) Prevention of renal insufficiency after abdominal aortic aneurysm resection by optimal volume loading. Arch Surg 116,1517-1524[Abstract]
- Jenkins, MT, Gieseke, AH, Johnson, ER (1975) The postoperative patient and his fluid and electrolyte requirements. Br J Anaesth 47,143-150[Free Full Text]
- Campbell, IT, Baxter, JN, Tweedie, IE, et al (1990) IV fluids during surgery. Br J Anaesth 65,726-729[Abstract/Free Full Text]
- Crozier, TA, Muller, JE, Quittkat, D, et al (1994) Effects of anaesthesia on the cytokine responses to abdominal surgery. Br J Anaesth 72,280-285[Abstract/Free Full Text]
- McBride, WT, Armstrong, MA, McBride, SJ (1996) Immunomodulation: an important concept in modern anaesthesia. Anaesthesia 51,465-473[ISI][Medline]
- Dietz, NM, Joyner, MJ, Warner, MA (1996) Blood substitutes: fluids, drugs, or miracle solutions? Anesth Analg 82,390-405[Abstract]
- Fresh-frozen plasma, cryoprecipitate, and platelets administration: practice guidelines development task force of the College of American Pathologists: practice parameters for the use of fresh-frozen plasma, cryoprecipitate, and platelets. JAMA 1994; 271:777–781
- Miller, RD (1973) Complications of massive blood transfusions. Anesthesiology 39,82-93[CrossRef]
- Starling, EH (1896) On the absorption of fluids from the connective tissue spaces. J Physiol 19,312-326
- Gaar, KA, Taylor, AE, Owens, LJ, et al (1967) Effect of capillary pressure and plasma protein on development of pulmonary edema. Am J Physiol 213,79-82[Free Full Text]
- Guyton, AC, Lindsey, AW (1959) Effect of elevated left atrial pressure and decreased plasma protein concentration on the development of pulmonary edema. Circ Res 7,649-657[Abstract/Free Full Text]
- Robin, ED, Carey, LC, Grenvik, A, et al (1972) Capillary leak syndrome with pulmonary edema. Arch Intern Med 287,690-698
- Holcroff, JW, Trunkey, DD (1975) Pulmonary extravasation of albumin during and after hemorrhagic shock in baboons. J Surg Res 18,91-97[CrossRef][ISI][Medline]
- Velanovich, V (1989) Crystalloid versus colloid fluid resuscitation: a meta-analysis of mortality. Surgery 105,65-71[ISI][Medline]
- Feeley, TW, Mihm, FG, Halperin, BD, et al (1985) Failure of colloid oncotic-pulmonary artery wedge pressure gradient to predict changes in extravascular lung water. Crit Care Med 13,1025-1028[ISI][Medline]
- Pearl, RG, Halperin, BD, Mihm, FG, et al (1988) Pulmonary effects of crystalloid and colloid resuscitation from hemorrhagic shock in the presence of oleic acid-induced pulmonary capillary injury in the dog. Anesthesiology 68,12-20[CrossRef][ISI][Medline]
- Moran, M, Kapsner, C (1987) Acute renal failure associated with elevated plasma oncotic pressure. N Engl J Med 317,150-153[ISI][Medline]
- Siegal, DC, Cochin, A, Geocaris, T, et al (1973) Effect of saline and colloid resuscitation on renal function. Ann Surg 177,51-57[ISI][Medline]
- Carey LC, Lowery BD, Cloutier CT. Hemorrhagic shock. Curr Probl Surg 1971; 3–48
- Rosenthal, MH, Arieff, AT (1995) Fluid and electrolyte therapy in perioperative surgical and critically ill patients. Arieff, AI DeFronzo, RA eds. Fluid, electrolyte and acid-base disorders 2nd ed. ,597-632 Churchill Livingstone New York, NY.
This article has been cited by other articles:
|
|
|
|
 
G. P. Joshi
Intraoperative Fluid Restriction Improves Outcome After Major Elective Gastrointestinal Surgery
Anesth. Analg.,
August 1, 2005;
101(2):
601 - 605.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
Abstract
Introduction
