TABLE OF CONTENTS
Page
ORGANIZATION ---------------------------------------------------------------------------
1
SECTION I – PATHOPHYSIOLOGY ----------------------------------------------------------HEPATIC BLOOD SUPPLY -------------------------------------------------------------RISK ASSESSMENT ----------------------------------------------------------------------HEPATIC DISEASE AND RISK --------------------------------------------------------HEPATIC ASSESSMENT ----------------------------------------------------------------REFERENCES ------------------------------------------------------------------------------THE DIFFERENTIAL DIAGNOSIS OF POSTOPERATIVE JAUNDICE -------POSTOPERATIVE RESPIRATORY FUNCTION -----------------------------------REFERENCES ------------------------------------------------------------------------------
3
3
4
4
8
10
12
15
18
SECTION II – AN APPROACH TO BLOOD UTILIZATION ----------------------------BLOOD COMPONENT THERAPY ----------------------------------------------------COAGULATION FACTORS ------------------------------------------------------------DELIVERY SYSTEMS -------------------------------------------------------------------COMPLICATIONS -------------------------------------------------------------------------
20
20
26
29
33
SECTION III – ANESTHESIA FOR LIVER TRANSPLANATION ---------------------PREOPERATIVE EVALUATION ------------------------------------------------------ANESTHESIA OVERVIEW -------------------------------------------------------------FULMINANT HEPATIC FAILURE ----------------------------------------------------CIRCULATORY PATHOPHYSIOLOGY AND OPTIONS IN HEMODYNAMIC
MANAGEMENT DURING LIVER TRANSPLANTATION -----------------------PULMONARY HYPERTENSION ------------------------------------------------------OR SET UP ---------------------------------------------------------------------------------THERAPEUTIC PROTOCOLS
NEW ANESTHETIC RECORD ---------------------------------------------------------BLOOD COMPONENT THERAPY IN LIVER TRANSPLANT
RECIPIENTS --------------------------------------------------------------------------------
34
34
34
38
M.S. Mandell, M.D.
Director of Anesthesia for Liver Transplantation
UCHSC
Revised: 10/03
41
54
62
66
67
68
ORGANIZATION
University Hospital has a multidisciplinary team of health care providers who interact to form the Liver Transplant
Team. Subspecialty participants include: Anesthesiology, Blood Bank, Hepatology, Psychiatry, Transplant Surgery
and Social Work. Members of the Liver Transplant Team work together to identify appropriate candidates for liver
transplantation, optimize care for patients prior to and after transplantation and perform transplants. The Department
of Anesthesiology has faculty dedicated to the care of liver transplant patients. These faculty are members of the
University Hospital Liver Transplant Team. Anesthesiology faculty responsibilities include the evaluation of patients
for the Liver Transplant Selection Committee, the preoperative explanation of the risks and benefits of anesthesia care
and intraoperative care.
It is anticipated that all residents in the Department of Anesthesiology will participate in the intraoperative care of liver
transplant patients. Assignment of resident staff to a liver transplant will be at the discretion of the Charge
Anesthesiologist after consultation with the attending anesthesiologist for each transplant. In the best judgment of the
attending anesthesiologist, it is possible that call residents other than the C1such as CV call or C2 residents may be
asked to help care for a liver transplant recipient after regular work hours.
The anesthesia faculty recommend that residents at all levels of training familiarize themselves with the contents of the
Hepatolbiliary Survival Manual in preparation for the care of liver transplant recipients.
Assignments and Responsibilities
Preoperative consults for inpatients will be performed by the resident and attending together when possible. This is
intended to be an educational experience and the resident is NOT responsible for submission of evaluation to the Liver
Transplant Committee. A standard preoperative evaluation, however, must be performed on all patients admitted for
liver transplant.
Postoperative visits and notes are the responsibility of the attending anesthesiologist. However, residents should
perform postoperative visits documented by a note in the patient chart.
The Liver Transplant Selection Committee convenes Thursday at 0700h. This is a multi disciplinary committee and
attendance by the transplant resident or at least part of the meeting is encouraged.
We will try to conform to the ABA suggested content for educational purposes. A list of topics is given below to act
as a checklist during the rotation.
LIVER PHYSIOLOGY AND PATHOLOGY
1.
2.
3.
4.
5.
Hepatic blood flow and autoregulation
Chronic end stage liver disease
Portal hypertension
Fulminant hepatic failure
Hepatorenal syndrome
RENAL PHYSIOLOGY AND PATHOLOGY
1.
2.
3.
4.
Renal blood flow and autoregulation
Chronic end stage renal disease
Acute renal failure
Hemodialysis
-1-
1
RELATED TOPICS
1.
2.
3.
4.
5.
Interpretation of Liver Function Tests
Coagulation abnormalities
Blood transfusion
Pharmacokinetic and dynamic changes of end stage liver and renal disease
Differential diagnosis of postoperative hepatic dysfunction
SPECIALTY TOPICS
1.
2.
3.
Rapid infusion systems
Introduction to Veno venous bypass circuits
Physiology of vena caval crossclamping
-2-
2
SECTION I - PATHOPHYSIOLOGY
HEPATIC BLOOD SUPPLY
The liver is the largest discrete organ in the human body, occupying 2% of the total body weight (1). As an
intra-abdominal organ, the liver is the last link in the much larger splanchnic system. Basic hepatic functions include
vascular storage and filtration, secretion of bile, metabolism and synthetic activity (2).
Splanchnic Circulation
The splanchnic organs include the large and small intestine, pancreas and spleen. These are an integrated
intra-abdominal system that receives blood from the celiac, superior and inferior mesenteric arteries. This system
receives an average of 25% of the total cardiac output (2). The capacitance of the system is large and usually holds
30% of the estimated blood volume under normal conditions (1). The liver by itself may contain 10% of the estimated
blood volume. Under the influence of the sympathetic nervous system, the liver can extrude 5OOcc of blood acutely
into the circulation (1).
The hepatic bed has a dual blood supply consisting of the portal vein and hepatic artery (3). The portal vein is a
product of the confluence of venous drainage from the splanchnic organs. These include the splenic, superior and
inferior mesenteric veins. The hepatic artery is a direct branch of the celiac artery.
Although the portal vein supplies up to 75% of the total hepatic blood flow, only 45-55% of the oxygen requirements
are provided by this part of the circulation (4). Instead, hepatic artery, which only supplies 25% of total hepatic blood
flow delivers up to 45-55% of the oxygen requirements (4,5). Together, the dual vascular supply provides the total
hepatic oxygen requirement and a small reserve. Certain zones, which are distant to the main blood vessels, have
minimal oxygen reserve and are predisposed to ischemic injury (3).
Hepatic Reserve
It is well known that the liver has impressive functional reserve. This is easily appreciated after hemihepatectomy in
patients with mass lesions as there is usually no impairment in physiological function postoperatively. Furthermore,
regenerative properties of the liver are well documented (4). A hepatocyte is capable of recovery up to the point of
mitochondrial damage, while an acinus can regenerate if a single layer of cells adjacent to the hepatic arteriole
survives (6).
Hepatic function is protected by autoregulation of splanchnic blood flow. Both intrinsic and extrinsic mechanisms
control blood and oxygen supply. Control of total hepatic blood flow within the liver has three different aspects.
Regulation occurs through: 1.) hepatic arterial, 2.) portal venous flow and 3.) the interrelationship between these
circuits (7). These responses in the liver are less efficient than in other organ systems. A pressure-flow relationship is
the basis for hepatic arterial regulation (8). The liver differs from other systems, however, in that autoregulation
primarily occurs in the metabolically active (postprandial) state (8).
In contrast, venous blood delivery to the portal bed is passive and is controlled by the extrahepatic splanchnic
circulation (8,9). Both metabolic and neurogenic mechanisms have been identified as factors controlling mesenteric
and splenic blood flow (7). Portal pressure is the regulated variable rather than venous blood flow. This is opposite to
most other autoregulated systems (7). In other words, portal venous pressure and not flow is the factor that is kept
constant. Extrinsic regulation of the splanchnic bed is mediated through autonomic and catecholamine stimulation of
all splanchnic organs in response to metabolic and hemodynamic demands.
The interaction between the hepatic arterial and portal venous vascular beds has been termed "reciprocity" (10). This
describes the reciprocal relationship between flow and resistance in arterial and venous supply. Reduction of flow in
the portal vein decreases hepatic arterial resistance and consequently increases arterial blood flow. Reduction of
arterial flow lowers portal resistance, however, does not directly alter flow. Instead, the latter is regulated by the
prehepatic splanchnic vasculature. Studies on hepatic reciprocity have shown that it is an intrinsic response and not
dependent on liver innervation (8). Complete occlusion of one circuit reduces vascular resistance in the other by up to
-3-
3
20% (11). In spite of these compensatory changes in the hepatic blood flow, single circuit occlusion in most normal
individuals results in insufficient oxygen delivery and complete hepatic failure usually ensues (11).
RISK ASSESSMENT
HEPATIC DISEASE AND RISK
Due to the large reserve of the liver, significant impairment of physiological function must occur before clinical signs
and symptoms of hepatic failure become evident. However actual functional impairment may exist at earlier stages in
the natural history of the disease that consequently place the liver at higher risk of subsequent injury. There is
increased risk of toxic and ischemic damage due to the decreased hepatic reserve. With these attendant problems,
alterations in splanchnic blood flow or changes in pressure-resistance relationships within the liver may generate
recurrent hepatocellular injury.
Both anesthesia and surgery may play a role in hepatocellular injury (12,13). Regional and general anesthesia can
cause circulatory changes that can reduce vital blood flow to the liver (12,13). Furthermore, some intravenous and
inhalation anesthetics agents may produce widespread toxic damage within a compromised liver (14). Simple surgical
manipulation can challenge the relatively weak autoregulatory capabilities of the hepatic bed.
The concept of increased risk in underlying hepatic disease as presented above is an important consideration since it
forms the basis for preventative management. Issues that follow from this relate to identification, quantification and
assessment of the patient at risk. These topics will be dealt with briefly in the following discussion.
Identification of the Patient at Risk
Patients that have hepatic impairment and are at increased risk of further injury may be difficult to identify during
preoperative assessment. Risk factors and symptoms of liver impairment are not as well defined as in other organ
systems. Also, signs of end organ damage are often not apparent until late in the course of the disease. This leaves the
anesthesiologist with a poor means of screening by history and physical examination. At the same time we know that
1/700 asymptomatic patients admitted for elective surgery have unexplained abnormalities in liver function tests
during routine pre-operative evaluation (1 5). One third of these patients experience an episode of jaundice over the
next two weeks. Other studies have confirmed at least a 0.135% incidence of significant liver disease in completely
asymptomatic patients presenting for elective surgery (16). Patients with cirrhosis have twice the incidence of
cholelithiasis (17). Identification of these patients may result in a change of anesthetic or surgical plan depending on
the nature of the nature of the underlying illness. There is, however, no easy means of reliable identification or
screening.
Degree of Risk
There is little doubt that intra-abdominal surgery in the presence of hepatic disease is associated with increased
morbidity and mortality. It is well known that hepatic failure is second only to coronary artery disease as a cause of
death following cholecystectomy (17). In a study from Britain of patients that had unsuspected liver disease at the time
of laparotomy, the morbidity and mortality within one month of operation was 61 and 3l % respectively (18). All
patients with acute viral or alcoholic hepatitis died.
To date, risk assessment has been primarily performed in patients with known liver disease and the quantification of
risk in asymptomatic patients can only be extrapolated from these studies. Table 1 illustrates the averaged risk in
patients with varying degrees of hepatic impairment secondary to cirrhosis undergoing cholecystectomy (l9). This is in
comparison to a control mortality rate of less than 0.5%.
-4-
4
Source/
Year
Schwartz
1981
Aranha, et al.
1982
Castaing, et al.
1983
Manfredi, et al.
1983
No. Of
Patients
21
Garrison, et al.
1984
Present Study
Table 1. REVIEW OF LITERATURE
Mortality
Morbidity
%
%
9.5
4.8
55
25.5
23.6
14
7.5
7.1
19
21
25
39
20.5
49
10.2
12.2
Requiting
Transfusion (%)
61.9
42.6
44.9
Further analysis of this problem is shown in Table 2. In this review specific risk factors have been identified that
appear in cases associated with increased morbidity and mortality (20). Low albumin and prolonged PT, which
indicates markedly impaired synthetic ability, were associated with increased risk. In addition, abnormalities in
serum bilirubin and bromsulphalein excretion also correlate with poor outcome. Emergency surgery was the most
impressive predictor of poor outcome in this study.
-5-
5
Table 2. STUDIES OF NONPORTOSYSTEMIC SHUNT SURGICAL RISK
IN PATIENTS WITH CIRRHOSIS
INVESTIGATOR
No.
INDICATION
CAUSE
MORTALITY MORBIDITY
RISK
Of
FOR
OF
RATE
RATE
FACTORS
PTS
SURGERY
CIRRHOSIS
Guyer
35 Mostly
NS
19%
NS
Low albumin,
(1955)
abdominal
anemia,
prolonged PT,
ascites
Lindenmoth
104 Mixed
NS
7%
25%
BSP>10%,
(1963)
albumin <3 gm
per dl
Wirthlin
83 Nonvariceal
NS
57%
NS
Emergency
(1974)
gastroduodenal
(emergency)
surgery,
bleeding
8%
bilirubin >2
(elective)
mg per dl,
albumin <3gm
per dl, PT > 16
sec. elevated
ammonia
Schwartz (1981)
33 Biliary
Mixed
15%
39%
Bile duct
obstruction
Aranha
55 Biliary
Alcoholic
26%
Prolonged PT
9% PT ≤ 2.5
(1983)
17%
sec. prolonged; 83%
PT >2.5 sec
Doberneck
102 Mixed
NS
prolonged
47°/a
Bilirubin > 3.5
(1983)
20%
mg per dl,
alkaline
phosphates >
70 IV. PT >2
see prolonged
Gastrointest.
surg.ascites,
emer. surg.,
Operative
blood loss
> 1000 ml
postop complication
The Child scoring system, shown below (Table 3), documents the degree of hepatic impairment in individual patients.
The system was originally designed to risk stratify patients undergoing porto-systemic shunting procedure. Using this
method mortality rates of 10%, 31% and 76% were identified in Child class A, B, and C patients respectively (21).
Subsequently, it has gained wide use as a basis for risk assessment and long-term prognosis. This scoring system has
been tested on patient populations and found to have reasonable predictive value for operative outcome and estimated
blood loss.
-6-
6
Table 3. CLINICAL AND LABORATORY CLASSIFICATION OF PATIENTS WITH
CIRRHOSIS IN TERMS OF HEPATIC FUNCTIONAL RESERVE
(Data from Child 1964)
GROUP A
GROUP B
GROUP C
Serum bilirubin
< 40
40-50
> 50
(μmol litre-1)
Serum albumin
> 35
30-35
< 40
(g litre-1)
Ascites
None
Easily controlled
Poorly controlled
Neurological Disorder
None
Minimal
Advanced coma
Nutrition
Excellent
Good
Poor-wasting
Risk of Operation
Good
Moderate
Poor
The Pugh scoring system (Table 4) is a modification of the original Child classification where prothrombin time
has been substituted for nutritional status (21). The latter substitution has improved predictive value.
Numerical scores can be assigned which gives quantitative index of risk, similar to the cardiac Goldman
classification. Overall, only Child Class A (Pugh 5-6) are considered a reasonable risk for intra abdominal
surgery.
Table 4. GRADING OF SEVERITY OF LIVER DISEASE
(Data from Pugh and others, 1973)
Clinical and Biochemical
Points Scored for Increasing Abnormalities
Measurement
1
2
3
Encephalopathy (grade)
None
1and 2
3 and 4
Bilirubin (μmol litre-1)
.< 25
25-40
> 40
Albumin (g litre-1)
35
28-35
< 28
Prothrombin time (seconds prolonged
I-4
4-6
>6
-7-
7
HEPATIC ASSESSMENT
As mentioned previously, clinical evidence of hepatic disease is often not apparent until late in the course of the
illness. Furthermore, the clinical problems associated with liver insufficiency are relatively nonspecific on history and
physical exam. Despite these limitations there are clinical scenarios that should raise suspicion of underlying liver
disease. Information regarding family history, inflammatory bowel disease, environmental exposure and alcohol
consumption should be obtained as well any contact with known infected individuals, prior transfusion, tattoos and
ingestion of potential hepatotoxins. This knowledge together with a description of clinical symptoms can help in
identifying hepatic impairment.
Liver function tests can measure different aspects of hepatic function. As a group of tests, they lack specificity and are
often affected by non hepatic function (6). Quantification of hepatic impairment is not obtained from a single set of
tests. Consequently, results need to be interpreted in the total clinical context. These observations are not indicative of
good screening tests.
Common biochemical markers used in the evaluation of hepatobiliary function consist of serum bilirubin, enzymes,
proteins and coagulation profile. Since serum enzymes are frequently used as diagnostic indicators, their role and
limitations will be briefly discussed.
The cytosolic aminotransferases consists of AST (aspartate aminotransferase) and ALT (alanine aminotransferase).
Elevation of AST is caused by damage to hepatocytes, myocardium or skeletal muscle. ALT is more hepto specific,
reflecting cell membrane damage and necrosis. In the absence of myocardial or muscle injury, high levels of serum
transaminases suggest hepatocellular injury. Moderate elevations (2-20X normal) are often seen in anicteric or
subclinical viral hepatitis. High values occur in severe acute viral or toxic hepatitis. Low to normal levels of serum
enzymes may be observed in chronic hepatocellular destruction, complete liver failure and cholestatic disease.
Alkaline phosphatase (ALP), a membrane bound enzyme, is produced by many tissues and is excreted in the bile. In
hepatobiliary disease, enzyme synthesis increases and there is a large release of ALP into the bloodstream. In the
absence of bone disease and pregnancy, an elevated ALP reflects impaired biliary function. Slight to moderate
increases in ALP occur in all types of liver disease. The most striking elevations (IOX normal) occur in extra or
intrahepatic cholestasis.
A sensitive indicator of hepatocellular and renal damage is gamma-glutamyl transpeptidase (GGT). GGT is elevated in
the serum of virtually all patients with cellular liver damage. Significantly raised levels are usually observed in acute
hepatic injury before other liver function tests show abnormalities. Similar to ALP, 5'nudeotidase is a hepatobiliary
enzyme. It is relatively specific but insensitive indicator of cholestasis. It is very useful in equivocal elevations of the
ALP. Changes in serum albumin and coagulation parameters are considered late signs of liver disease and are
primarily used to mark the severity of hepatic impairment.
The diagnosis of liver disease requires a high degree of suspicion and a careful probing of the clinical history.
Biochemical tests add to the accuracy of diagnosis. In contrast, anesthesiologists are often confronted with abnormal
hepatic function tests in asymptomatic patients. Some of these abnormalities may be secondary to the surgical illness
or coexisting liver disease. Patients with underlying hepatic impairment are important to identify since they are at
increased risk for further injury due to alterations in splanchnic blood flow, which occurs during anesthesia and
intra-abdominal surgery. Medical management of these patients may be changed if liver disease is identified.
The question of whether further investigation should be pursued in all cases of subtle changes in liver function tests is
widely debated (15,19,20). Most authors agree that the clinical history must be examined closely and that any findings
associated with liver disease be investigated. Misinterpretation of liver function tests is, however, a problem that may
result in a failure to recognize primary hepatic impairment (17). Inability to distinguish hepatocellular injury from
cholestatic patterns has been identified as a recurring mistake. This alone may lead to an incorrect diagnosis and
inappropriate treatment.
Further controversy exists over the use of liver function tests as a screening tool in patients undergoing
intra-abdominal surgery (6,16,20). Some studies have supported the use of screening in such patient populations,
showing the fortuitous diagnosis of liver disease, which either led to an appropriate cancellation of the procedure or an
alteration in anesthetic technique.
-8-
8
The disposition of a patient presenting for surgery with a previous history of jaundice also poses a special dilemma for
the anesthesiologist. Clinical history can be very helpful in establishing a differential diagnosis. But again, whether
these patients should undergo further investigations is controversial depending on the clinical history. Current
literature favors the review of liver function tests in these patients (20,21).
In summary, the liver, which is the largest organ in the body, is an integral part of the splanchnic system. Although
there is a dual blood supply, oxygen delivery is highly dependent on hepatic arterial blood flow. Reciprocity of flow
occurs between the hepatic arterial and portal venous circuits but there is incomplete compensation with hepatic
necrosis often occurring after the interruption of arterial flow. A pressure-flow relationship has been identified for
hepatic arterial supply in the postprandial liver but this response is not as well defined as in other organ systems. Portal
flow is controlled primarily by extrahepatic factors to maintain a constant pressure gradient. This results in a weak
autoregulatory response to alterations in cellular demands and places the liver at increased risk of ischemic injury.
Both anesthesia and surgery may impair liver function through alterations in splanchnic flow or toxic injury.
Hepatocellular failure may occur post-operatively in patients that present for intra-abdominal surgery with decreased
hepatic reserve. Furthermore, there is a direct correlation between the severity of liver disease and the resulting
morbidity and mortality. A good predictor of outcome is the Child or Child-Pugh classification. Patients who have
liver disease, however, are not always easily identified due to the nonspecific nature of the signs and symptoms.
Liver function tests have been found to be useful in the diagnosis of hepatic disease when interpreted in the clinical
context. The use of these tests for screening is widely debated. Some investigators suggest the routine use of liver
function tests on all patients undergoing intra-abdominal surgery since this population is at increased risk. The lack of
specificity of these tests results in a problem of interpretation in the asymptomatic patient. The decision to further
investigate the latter group of patients is, therefore difficult. Incorrect interpretation of liver function tests has resulted
in failure to recognize hepatocellular compromise that will affect management and outcome. Continuing education on
the topic of pre-operative liver function test may prevent this problem.
The preoperative management of the asymptomatic patient with a prior history of jaundice has also been debated.
Most authors agree that these patients warrant further probing into the clinical history and evaluation by screening
with liver function tests. Any risk factors or diseases states associated with abnormal hepatic function should be
identified.
Patients with liver disease are not always easily identified by brief clinical examination. There is at least a 0.135%
incidence of significant hepatic impairment in asymptomatic patients that present for elective surgery. A thorough
understanding of assessment and identification of the patient at risk will lend guidance to the use and interpretation
liver function tests as part of a rational clinical approach.
-9-
9
REFERENCES
1
Lautt WW and Greenway C: Hepatic venous compliance and role of liver as a blood reservoir Am. J. Physiol.
1976;231:292-295
2.
Guyton AC: The Liver as an Organ. Textbook of Medical Physiology (8th ed). WB Saunders Co.,
Philadelphia, 1991;771-788
3.
Maze M: Hepatic Physiology (3rd ed) in Anesthesia ed by Miller. Churchill Livingstone, New York,
1990;585600
4.
Gelman S: Anesthesia and the Liver in Management of Anesthesia ed by Barash. JB Lippincott Co.,
Philadelphia, 1989;1133-1162
5.
Cooperman LH: Effects of anesthetics on the splanchnic circulation. Br. J. Anaesth 1972;44:967-970
6.
Gornall AG, Goldberg DM: Hepatobiliary disorders. Applied Biochemistry of Clinical Disorders. Harper and
Row Publishers, Philadelphia, 1980;164-192
7.
Richardson PD, Withrington PG: Liver Blood Flow I. Intrinsic and nervous control of liver blood flow.
Gastroenterology 1981;81:159-173
8.
Hanson KM, Johnson PC: Local control of hepatic arterial and portal venous flow in the dog. Am. J. Physiol.
1966;211:712-720
9.
Richardson PD, Withrington PG: Pressure-flow relationships and effects of noradrenalin and isoprenaline on
the hepatic arterial and portal venous vascular beds of the dog. J. Physiol. 1978;282:451-470
10.
Hirsch LJ, Ayabe T, Glick G: direct effects of various catecholamines on liver circulation in dogs. Am. J.
Physiol. 1976;230:1394-1399
11.
Kim DK, Kinne DW and Fortner JG: Occlusion of the hepatic artery in man. Surg. Gynecol. Obstet
1973;136:966-968
12.
Gelman S. Fowler KC, Smith LR: Liver circulation and function during isoflurane and halothane anesthesia,
Anesthesiology 1984;61:726-730
13.
Kennedy WF, Everett GB, Cobb LA, Allen GD: Anesth. Analg. 1970;49:1016-1022
14.
Kenna JG and Van Pelt FN The metabolism and toxicity of inhaled anesthetic agents. Anesth Pharmacol Rev.
1994;2:29-42
15.
Schemel WH: Unexpected hepatic dysfunction found by multiple laboratory screening Anesth. Analg.
1976;55:1810-1812
16.
Wataneeyawech M, Kelly KA: Hepatic diseases unsuspected before surgery N.Y. State J. Med.
1975;1278-1281
17.
Bloch RS, Allaben RD, Walt AJ: Cholecystectomy in patients with cirrhosis Arch. Surg.
1985;120:669-672
18.
Powell-Jackson P, Greenway B, Williams R: Adverse effects of exploratory laparotomy in patients with
unsuspected liver disease Br. J. Surg. 1982;69:449-451
- 10 -
10
19.
Garrison RN, Cryer HM, Howard DA, Polk HC: Clarification of risk factors for abdominal operations in
patients with cirrhosis Ann. Surg. 1984;199:648-655
20.
Friedman LS, Maddrey WC: Surgery in the patient with liver disease Med. Clin N.A.
1987;71:453-476
21.
Strunin L: Preoperative assessment of the patient with liver dysfunction Br. J. Anaesth. 1978;50:25-31
- 11 -
11
THE DIFFERENTIAL DIAGNOSIS OF POSTOPERATIVE JAUNDICE
The development of jaundice in the postoperative period indicates significant physiological dysfunction. Mechanisms
underlying abnormal bilirubin metabolism are complex and usually multifactorial in this setting. Anesthesiologists are
often asked to help in the assessment of these patients so that an etiology can be identified. Therefore the following
approach to the differential diagnosis of jaundice in the postoperative patient is briefly outlined.
Unconjugated Bilirubin
The first step is to differentiate conjugated from unconjugated bilirubin. Unconjugated hyperbilirubinemia is defined
as an elevation of the serum bilirubin of which the conjugated fraction does not exceed 15%. This is most often related
to abnormalities in the turnover of red blood cells. A list of considerations in the differential diagnosis is listed below.
I.
Hemolysis
A.
B.
C.
Extravascular hemolysis of red blood cells (positive indirect immunoglogulin)
Arterio/veno bypass circuits
Congenital or acquired defects listed below
1.
2.
3.
4.
G6PD deficiency
autoimmune hemolytic anemia
drug induced hemolytic anemia
Paroxysmal cold/nocturnal hematuria (as rare as chicken's teeth)
II.
Hematoma resorption
III.
Gilbert's/Crigler Najjar Syndrome
Gilbert's syndrome is a hereditary nonhemolytic intermittent jaundice associated with increases in unconjugated
bilirubin. There is impaired hepatic clearance but otherwise normal hepatocellular function. It occurs in 3-7% of the
population and is 2-7 times more common in males. A deficiency of hepatic UDP glucuronyl transferase has been
identified as the defect. Decreased caloric intake, intercurrent illness and vomiting can cause jaundice.
Crigler Najjar, types I and II are the extreme expressions of Gilbert's syndrome and associated with minimal levels to
complete absence of UDP glucuronyl transferase. These syndromes are usually detected early and are often associated
with severe neurological damage due to kernicterus.
Diagnostic Tests
The following tests should be sent to aid diagnosis:
1.
2.
3.
CBC and smear
Indirect immunoglobin (indirect Coomb's)
ANA, and RF
Conjugated Bilirubin
Any elevation in the conjugated bilirubin fraction always signifies hepatobiliary dysfunction. This is found in both
biliary and hepatocellular disease and therefore does not help to distinguish between the two disorders. More
commonly, a mixed picture of conjugated and unconjugated hyperbilirubinemia exists with most hepatobiliary
disorders.
Elevation of the alanine transaminase with or without aspartate transaminase usually indicates hepatocellular
inflammation or necrosis. In contrast, increases in alkaline phosphatase, gamma glutamyl transferase or 5'nucleotidase
incriminate cholestasis or obstructive jaundice.
- 12 -
12
Inn and extrahepatic biliary obstruction (obstructive jaundice) present with similar clinical and laboratory findings and
are usually identified on the basis of ultrasound imaging. A differential diagnosis is listed below.
I.
Obstructive Jaundice (Conjugated hyperbilirubinemia)
A.
Extrahepatic Obstruction
1.
2.
3.
4.
5.
6.
B.
Tumor: biliary, pancreas and duodenal
Calculous and acalculous cholecystitis
Pancreatitis
Biliary stricture
Sclerosing cholangitis
Ascending cholangitis
Intrahepatic Obstruction
This is almost a diagnosis of exclusion or one that is established by liver biopsy. A list is
given below.
1.
2.
3.
4.
5.
II.
Primary biliary cirrhosis
Drugs
TPN
Steroids
Liver transplant rejection
Hepatocellular Disease
Hepatocellular disease is the other category that may give rise to unconjugated or mixed
hyperbilirubinemia. The differential diagnosis is extensive and includes the following problems.
A. Infectious hepatitis (Hepatitis A, B, C, D, and E, CMV, Epstein Barr virus and
others)
B. Drugs/-Toxins; There are over 600 drugs listed that cause hepatic injury. Generally two
categories exist; idiosyncratic reactions and dose-dependent reactions.
C. Sepsis
D. TPN (abnormal LFTs in 68-93% of individuals given > 2weeks therapy)
E. Metabolic (low PaC02/PaO2)
F. Ischemia. This is a complex topic but generally a fall in perfusion pressure due to
increased venous or decreased arterial pressure will result in suboptimal oxygen delivery.
Contributing factors are given in the following list.
1.
Increased Venous Pressure
a.
b.
c.
d.
e.
f.
Increased infra-abdominal pressure
IPPV/PEEP
Congestive heart failure
Pulmonary hypertension
Pulmonary embolus
Surgical manipulation
- 13 -
13
2.
Decreased Arterial Pressure
a.
b.
c.
d.
Prolonged hypovolemia. Systemic shock is not necessary to cause hepatic
ischemia.
Vasopressors with alpha activity
Aortic crossclamp
Surgical manipulation
Halothane Hepatitis
There are direct toxic effects of drugs that are dose related such as acetaminophen. In addition there are idiosyncratic
or unpredictable reactions that are often immune based. Inhalational agents, most notably halothane has been
incriminated in mild to severe cases of hepatitis. It appears that halothane is capable of causing both toxic and immune
mediated reactions.
Toxic reactions are often identified by a transient elevation of the LFTs immediately following an anesthetic. This
occurs in up to 30% of individuals exposed to halothane, but does not result in permanent hepatocellular impairment.
In contrast immune mediated hepatitis, observed in 1/30,000 exposures, causes significant morbility and mortality.
Halothane is metabolized by the liver and produces reactive acyl chloride, which acts Immunologically as a hapten.
This results in trifluoroacetylation of hepatocyte membranes. The membrane-hapten complex induces an immune
response resulting in hepatocyte necrosis.
Risk factors associated with halothane hepatitis include obesity, female gender, familial factors and prior exposure.
The response is delayed and usually observed 7 to 28 days following exposure. Signs and symptoms of autoimmune
dysfunction are observed and include:
1.
2.
3.
4.
5.
6.
pyrexia
arthralgia
rash
eosinophilia
autoantibodies
circulating immune complexes
Diagnostic tests must include a full panel LFT, CBC and differential and autoimmune panel. If halothane hepatitis is
strongly suspected, there are several centers in the USA and Europe that test for immune-modified hepatocyte
complexes.
Enflurane has also been associated with immune based hepatitis and has an incidence of 1/800,000. There also have
been a few case reports of isoflurane hepatitis. In general, the potential of an inhalational agent to induce immune
complexes may be partially related to the extent of metabolism as shown.
Halothane > Sevoflurane > Enflurane > Isoflurane > Desflurane
- 14 -
14
POSTOPERATIVE RESPIRATORY FUNCTION
The introduction of general anesthetic agents facilitated intra-abdominal exploration in surgical practice. As these
procedures became more common, however, the incidence of postoperative pulmonary complications also increased
(1). Early observations suggested that generalized reduction in lung volumes may cause pulmonary dysfunction
observed following surgery (2). Subsequently, in 1932 functional residual capacity (FRC) was identified as one of the
most important lung volumes to predict outcome (3). Since that time, there has been a growing interest in the effects of
intra-abdominal surgery on lung function.
One of the primary interests in pulmonary function has been the prediction of outcome following surgery. In this
respect, pre-operative pulmonary function tests have been tested and used extensively. Post-operative pulmonary
function tests are, also studies of outcome and may be used as predictors of potential complications. In this way, they
may aid in therapeutic intervention.
Studies of post-operative lung function have been mainly descriptive and concentrated on changes in pulmonary
mechanics. In spite of these detailed studies, very little is known about control of respiration and changes in
ventilatory patterns. The following will provide an outline of current information regarding descriptive pulmonary
mechanics.
Pulmonary Mechanics
Studies of pulmonary mechanics have used clinical outcome to measure of predictive potential. The data is derived
from both prospective clinical studies and laboratory investigations. Experiments have shown that while there is a
decrease in total lung capacity post-operatively, the major volume loss occurs in FRC and vital capacity (VC) (4,5,6).
The importance of these latter changes lies in their ability to identify possible respiratory morbidity.
The changes in VC following intra-abdominal surgery are larger than those in FRC, but do not correlate well with
pulmonary complications (4,5,6). Instead, the best correlation is between FRC, hypoxemia and pulmonary
complications (6,7). Following intra-abdominal surgery, VC is decreased by 60 percent of the pre-operative value and
remains depressed for at least 5 to 7 days (4,5). In contrast, a 30 percent fall in FRC does not occur until 16 hours
following surgery and returns to within normal limits by day 7 (4,5). The degree of hypoxemia is closely associated
with losses in FRC. Both demonstrate a similar time delay in onset and larger reductions in FRC are also associated
with greater falls in arterial oxygenation (4). This is not as obvious with changes in VC. Finally, several studies have
shown a predictable relationship between degree of hypoxemia and incidence of pulmonary complications (4-6).
Therefore, it appears that measurements of FRC may provide the best insight into respiratory outcome, although the
precise specificity and sensitivity are unknown. The role of VC in determining outcome is not as clear as FRC. There
is, however, a weaker but apparent relationship between the magnitude of change in VC and ventilatory impairment in
some studies (8).
Site of Surgery
Not all types of surgical procedures produce comparable changes in lung volumes post-operatively (4). In terms of
volume changes, the site of surgery appears to influence the degree of ventilatory change. Upper abdominal surgery
demonstrates the largest impact on post-operative lung volumes and requires the longest recovery times (3,4). This is
followed by lower abdominal and then posterior surgery (4). Superficial and peripheral surgery under general
anesthesia is not associated with significant or ongoing respiratory problems (4). The use of regional anesthesia for
superficial surgery does not result in any changes of lung volumes from pre-operative values (4).
Within the category of upper abdominal surgery, there are differences in pulmonary outcome based upon type of
incision and surgical approach. Following open cholecystectomy, greater changes in lung capacities and arterial
oxygenation are identified after midline rather than subcostal incisions (8). For other types of upper abdominal
procedures, both midline and paramedian surgical approaches have the greatest respiratory impact (9). Esophagectomy
patients have greater changes following thoracolaparotomy than a two incision laparotomy and thoracotomy (10).
Laparoscopic upper abdominal surgery is associated with significantly smaller changes in pulmonary mechanics than
open approaches (11).
- 15 -
15
From the preceding information it is apparent that both site of surgery and type of incision can be important
determinants of respiratory outcome. The underlying reasons for these differences are unclear; however, alterations in
central respiratory control have been suggested (12). Ventilatory rate and pattern, which are representative functions of
central control, are altered postoperatively. The type and degree of abnormality is most apparent following midline
incision for upper abdominal surgery (8). These changes may represent one of the final links to understanding
respiratory dysfunction.
Mechanism of Hypoxemia
Prolonged post-operative hypoxemia is an early clinical manifestation of the respiratory changes, which result in
pulmonary complications. The principal causes of late and prolonged desaturation have been examined in several
studies (13-15). In particular, the relative contribution of altered ventilation/perfusion ratios and shunt fraction have
been investigated. Results indicate that venous admixture secondary to increased shunt fraction is the main cause of
hypoxemia following upper abdominal surgery (14-1 5).
The cause of increased shunt fraction is due to changes in the relationship between closing volume (CV) and FRC
(6,15). Closing volume, the lung volume at which small airway begin to obstruct, contains the majority of shunt
fraction. Venous admixture results from gas trapping and atelectasis secondary to collapse of small airways (15).
Although, CV remains relatively constant following surgery, FRC has been shown to drop significantly
post-operatively (6,16). Closing volume, then occupies a larger proportion of the total lung capacity and tidal
ventilation occurs within or just above CV. The result is little or no reserve remaining during respiration.
Alterations in FRC
To date, the reasons for the fall in lung volumes and particularly FRC following upper abdominal surgery are not fully
understood. A number of potential contributing factors have been studied and their roles clarified. The following is a
review of these considerations:
General Anesthesia
Initially many investigators felt that continued the pulmonary effects of general anesthesia into the post-operative
period may have accounted for some of the prolonged respiratory defects identified. Indeed, similar changes in FRC
were identified following induction of general anesthesia (17). These mechanical alterations also correlated with a fall
in arterial oxygenation (18). However, ventilation/perfusion mismatch plays a larger role in producing hypoxemia
during anesthesia than post-operatively. Furthermore, recovery of FRC and hypoxemia have been observed in the
early post-operative period before the subsequent fall that is associated with prolonged desaturation and pulmonary
complications following upper abdominal surgery (11, 18). This alone strongly suggests a different cause.
Pain
Post-operative pain has also been cited as a factor that may alter pulmonary mechanics. Clinical studies have
compared various forms of analgesia to determine the impact upon the respiratory system.
A series of investigators used epidural block in cross over studies to provide analgesia to the thoracic dermatome four
level with minimal motor impairment (19-20). It had been demonstrated that epidural by itself at that level did not
affect pulmonary function. Following surgery, epidural local anesthetics provided good analgesia but failed to correct
FRC. There was a significant improvement in VC. Overall there appeared to be less pulmonary complications in the
epidural as compared to narcotic group (20). This has been attributed to the improved ability to cough and clear
secretions, based upon greater VC. However, even with good control of pain, post-operative respiratory, complications
still occurred. The effect of improved VC on outcome demonstrated the complexity of this situation.
Unilateral intercostal blockade demonstrated small but significant improvements in VC (21). This was also
accompanied by slight improvements in patient outcome. Bilateral intercostal blockade, however, had an adverse
effect on both pulmonary function and patient outcome. Transcutaneous nerve stimulation (TENS) has been reported
to provide sparing effects on both FRC and VC (22). Further studies are needed to confirm these beneficial effects.
- 16 -
16
Position and Age
It has been shown that pulmonary gas exchange is affected by posture in specific age groups (16). Post-operatively,
patients are usually placed supine. This position has been shown to increase the severity of hypoxemia by a reduction
in lung volume. This results from pressure on the diaphragm by the intra-abdominal contents. There is a decrease in all
lung volumes except residual volume (18). The expiratory reserve volume is significantly decreased. Closing volume
does not change significantly with position, but FRC decreased. This increased the chance of breathing within CV,
resulting in underventilated and over perfused regions.
Age by itself is an independent variable determining the relationship between FRC and CV (17). As age increases,
FRC remains constant, but CV increases. It has been found that once breathing occurs within CV, that the supine
position will worsen gas exchange due to the mechanisms outlined above. In general, by age 44, breathing occurs
within CV in the supine position. This was observed in the seated position by age 65.
Other Factors
Other patient attributes have also been shown to affect gas exchange. These can be divided into two distinct groups:
those that decrease FRC and those that increase CV. Within the first group, obesity and pregnancy are significant
factors (12). Smoking and pulmonary edema contribute to the second group. Pulmonary disease represents a
combination of the above features, depending on the nature of the disease. Pneumoperitoneum has also been suggested
to affect respiratory outcome by causing a fall in FRC (4,6).
- 17 -
17
REFERENCES
1.
Pasteur W. Active lobar collapse of the lung after abdominal operations: a contribution to the
study of postoperative lung complications. Lancet 1910, 2, 1080-1083
2.
Haldane J.S., Meakins J.L., Priestley J.G. The effect of shallow breathing. J. Physiol., 1919, 52,
433-453.
3.
Beecher H.K. Effect of laparotomy on lung volume; demonstration of a new type of collapse. J. Clin Invest.,
1932, 12,651-658.
4.
Ali J. Hechman H. Weisel R . D ., Layug A . B., Kripke B . J., B . Consequences of post-operative alterations
in respiratory mechanics. Amer. J. Surg., 1974, 128, 376-382.
5.
Meyers JR., Lembeck L., O'Kane A.E. Changes in functional residual capacity of the lung after operation.
Arch. Surg., 1975, 110, 576-582.
6.
Alexander J.I., Spemce A.A., Pankh R.K., Stuart B. The role of airway closure in postoperative hypoxemia.
Br. J. Anaesth., 1973, 5, 34-40
7.
Hechtman H.B., Weisel R.D., Vito L., Ali J., Berger R.L. Independence of pulmonary shunting in pulmonary
edema. Surgery, 1973, 300.
8.
Ali J., Khan T . A . The comparative effects of muscle transection and median upper abdominal incisions on
postoperative pulmonary function. Surg. Gynecol., Obstet., 1979, 148, 836-866.
9.
Knudsen J., Duration of hypoxemia after uncomplicated upper abdominal and thoraco-abdominal operations.
Anesthesia, 1970, 25, 372-377.
10.
Black J., Kalloor G.J., Colli- J.L. The effect of the surgical approach on respiratory function after
esophagectomy. Br. J. Surg., 1977, 64, 624-627.
11.
Putensen-Himmer G., Putensen C., Lammer H., Lingnauw., Algner F. Benzer H. Comparison of
postoperative respiratory, function after laparoscopy or open laparoscopy for cholecystectomy.
Anesthesiology, 1992, 77. 675-680.
1.
Craig D.B. Postoperative recovery of pulmonary function. Anesthesia and Analgesia, 1981, 60.
46-52.
13.
Vaughan R. W., Wise L. Postoperative arterial blood gas measurement in obese patients: effect of position on
gas exchange. Ann. Surg.,1975, 183, 705-709.
14.
Colgan F.J., Mahoney P. The effects of major surgery on cardiac output and shunting. Anesthesiology, 1969,
31, 213-221.
15.
Siler J.N., Rosenberg H., Mull T.D., Kaplan J.A. Hypoxemia after upper abdominal surgery: comparison of
venous admixture and ventilation /perfusion inequality components, using a digital computer. Ann . Surg .,
1974, 179, 149-155.
16.
Craig D.B., Wahba W.M., Don H. F., Couture J.G., Becklake M.R. "Closing volume" and its relationship to
gas exchange is seated and supine positions. J. Appl. Physiol., 1971, 31, 717-721.
17.
Rehder K. Anesthesia and the respiratory system. Can. Anaes. Soc. J., 1979, 26, 451-462.
I8.
Marshall B . E ., Wyche M .Q . Hypoxemia during and after anesthesia. Anesthesiology, 1972, 37, 178-209.
- 18 -
18
19.
Wahba W . M., Don H . F., Craig D . B . Post-operative epidural analgesia effects on lung volumes. Can.
Anaes. Soc. J., 1975, 22, 519-527.
20.
Spence A . A., Smith G. Postoperative analgesia and lung function: a comparison of morphine with
extradural block. Br. J. Anaes., 1971, 43, 144-148.
21.
Engberg G. Relief of postoperative pain with intercostal blockade compared with the use of narcotic drugs.
Acta Anaesthesia Scand. (Suppl.)., 1978, 70, 36-38.
22.
Ali J., Yaffe C., Serrette C., The effect of transcutaneous electrical nerve stimulation on postoperative pain
and pulmonary function. Surgery, 1981.
- 19 -
19
SECTION II – AN APPROACH TO BLOOD UTILIZATION
BLOOD COMPONENT THERAPY
A recent subcommittee survey of the American Society of Anesthesiologist found that anesthesiologists administer
greater than half of all blood products given to patients. It is therefore important to develop expertise in the
management of blood component therapy.
Blood Components
A major advance in the field of blood banking is the fractionation of whole blood into individual components that are
subsequently used to treat specific deficiencies. Figure I demonstrates the method for blood fractionation and the
commonly available therapeutic components.
Fig I
Typing, Crossmatch and Screen
Red blood cells ("packed cells") are the most common component administered. There are several standard blood bank
orders depending on the urgency of a situation or most cost effective route in an elective case. Rational decision
making regarding packed red cells requests requires an understanding of how red cells are prepared for transfusion.
The following is a synopsis of these procedures.
Type
Typing refers to the determination of a patient's ABO and Rh status. This is done by testing the red cells for A and B
antigens while the serum is tested for A and B antibodies. Additional Rh antigen testing completes red blood cell
typing. A blood type compatibility and recipient chart is shown in Fig. 2
- 20 -
20
Blood Group
A
B
AB
O
ABO COMPATIBILITY TESTING
Red Cells Tested With
Anti-A
Anti-B
+
+
++
-
Serum Tested With
A Cells
B Cells
+
+
+
+
Fig. 2
Crossmatch
The crossmatch is performed when DONOR red blood cells are mixed with RECIPIENT serum. This is an in vitro
trial transfusion that is carried out in three distinct stages. A complete crossmatch requires 45-60 min regardless of
circumstances. The following diagram shows the phases of crossmatching and the antigens detected in each.
- 21 -
21
Screen
This refers to the detection of ANTIBODIES in the serum of donor or patient blood. The test is conducted in three
phases similar to crossmatch studies. All donated blood is screened shortly after collection for unusual antibodies by
combining serum with commercially obtained red blood cells that contain antigens known to cause hemolytic
reactions.
Emergency Protocols
In an emergent situation blood may have to be administered without a full crossmatch. There are set orders that the
blood bank can respond to under these circumstances. The following are the most common emergency orders.
Type Specific Partial Crossmatch
Typed RECIPIENT serum is combined with DONOR red blood cells (from previously screened blood) in saline at
room temperature and centrifuged for 5 minutes. This is called an immediate spin crossmatch. Macroscopic
agglutination is checked. This requires 5-10 minutes.
Type Specific Uncrossmatched Blood
All blood that is collected from donors undergoes serum screening to identify potentially dangerous antibodies. In an
emergency situation where blood is required immediately, screened type specific packed red cells are available. This
is considered relatively safe since in a patient without prior transfusion or pregnancy, an unexpected antibody is found
in only 1/1000 crossmatches.
Type O Rh-Negative/Rh-Positive Uncrossmatched Blood
Generally type 0 red blood cells lack the A and B antigens and cannot be hemolyzed by antibodies in the recipient’s
blood (Fig 5). However, some 0 donors contain high titers of hemolytic anti A and B ANTIBODIES. Emergency units
of type 0 blood are available when there is insufficient time for patient typing. There are several safety features to
protect patients. First, only type 0 blood with low serum antibody titers are chosen as emergency units. Second, this is
only available as packed red cells so that exposure to donor serum can be minimized. This is most important when
transfusing females of childbearing age. Finally, Rh negative blood can be chosen to prevent alloimmunization. In the
operating room at University Hospital 10 units of low titre type 0 Rh negative/positive blood is kept in Central
Laboratory for emergency use.
- 22 -
22
DONOR BLOOD GROUPS WHOSE BLOOD PATIENT CAN RECEIVE
DONOR
RECIPIENT
0
0, A, B, AB
A
A, AB
B
B, AB
AB
AB
Fig. 5
Indications for Blood Administration
Blood is mainly given to increase the oxygen carrying capacity. The hematocrit at which blood should be administered
is not simply answered. A young healthy patient with normal cardiorespiratory function may easily compensate for
anemia while an elderly patient with vital organ impairment may be unable to tolerate an identical hematocrit.
General guidelines are, chronic anemia is tolerated much better than acute blood loss. Decreased physiological reserve
as found in disease of the cardiorespiratory, renal or hepatic systems may influence the decision when to transfuse.
The anticipation of ongoing losses may also influence transfusion decisions. A standard "safe" hematocrit has not been
established under these conditions.
The FDA proposed that adequate oxygen carrying capacity can be met by a hemoglobin of 7g/dl or less if perfusion is
maintained by adequate intravascular volume. Certain medical conditions may justify, giving blood at a higher
hematoglobin. At University Hospital there is a Blood Utilization Committee that reviews blood component ordering
and administration for quality assurance.
The figure below gives a means of estimating red cell replacement based on physiological findings in emergent
conditions. In elective cases, an equation is given that will help determine the timing of transfusion based on known
patient variables. Always indicate the reason for transfusion in some manner on the anesthesia chart for example by a
recent hematocrit, exceeding the calculated allowable blood loss or hemodynamic variable.
- 23 -
23
Emergency
ESTIMATED FLUID AND BLOOD REQUIREMENTS 1N A 70KG MALE
INITIAL PRESENTATIONS
CLASS I
CLASS II
CLASS III
Blood loss (ml)
< 750
750-1500
1500-2000
Blood loss (% BV)
<15%
15-30%
30-40%
Pulse rate
< 100
> 100
> 120
Blood pressure
Normal
Normal
Decreased
Pulse pressure (mm Hg)
Normal or Increased
Decreased
Decreased
Capillary blanch test
Normal
Positive
Positive
Respiratory rate
14-20
20-30
30-40
Urine output (mL/h)
30 or more
20-30
5-15
CNS-mental status
Slightly anxious
Mildly anxious
Anxious and
confused
Fluid Replacement
Crystalloid
Crystalloid
Crystalloid +
(3:1 rule)
blood
Elective
CLASS IV
z 2000
z 40%
z 140
Decreased
Decreased
Positive
> 35
Negligible
Confused and
lethargic
Crystalloid +
blood
Allowable Loss = EBV x (Hbinitial – Hbtarget)/Hbinitial
Fig. 6
Storage
Belle Bonfils Blood Center supplies blood units for transfusion to University Hospital. Citrate, phosphate and dextrose
(CPD) are added to donated blood for storage at 4C. Citrate acts as an anticoagulant while phosphate buffers changes
in pH. Dextrose provides a source of energy for red blood cell metabolism. To further prolong shelf life, adenine saline
solution, another energy substrate is added to the CPD solution. This extends the shelf life to 42 days.
Short Dated Blood
A unit of blood marked as short dated means that its shelf life will expire within 14 days. This is important information
as changes take place in the biochemical composition of blood solutions with time. These changes are illustrated
below.
- 24 -
24
PROPERTIES OF WHOLE BLOOD AND PACKED CELL CONCENTRATES
STORED IN CPDA-l
DAYS OF STORAGE
Parameter
0
35 (Whole Blood)
35 (Packed Cells)
pH
7.55
6.98
6.71
Plasma hemoglobin (mg/dl)
8.2
46.1
246.0
Plasma potassium (mEq/L)
4.2
27.3
76.0
Plasma sodium (mEq/L)
169
155
122
Blood dextrose (mg/dl)
440
229
84
2,3-Diphosphoglycerate (μ M/ml)
13.2
<1
<1
Percent survival
79
71
a
Percent recovery of OR-tagged red blood cells at 24 hours.
Abbreviation: CPDA-1, citrate phosphate dextrose adenine-1.
Fig. 7
Autologous Blood
There are three methods of collecting autologous blood units: preoperative donation and storage, acute preoperative
phlebotomy and hemodilution and Perioperative blood salvage from the surgical site. Despite the relative safety of
autologous transfusion, it is currently not recommended to transfuse these units unless there is a specific indication
since the most common cause of lethal hemolytic transfusion reaction is clerical error, which may occur in this setting.
Frozen Blood
Packed red blood cells can be frozen which allows storage for up to 10 years. This is commonly done for rare blood
types and occasionally with multiple collections of autologous units. It is an expensive process and therefore not often
used.
Dilutants
A compatibility list is given in figure 8. Some dilutants can cause hemolysis due to a reduced tonicity of the solution.
Other solutions such as Lactated Ringers may promote clotting due to the presence of calcium. The latter occurs as
micro debris.
- 25 -
25
COMPATIBILITY OF BLOOD WITH INTRAVENOUS SOLUTIONS
Blood to Intravenous Solution
Hemolysis at 30 Min
1 : 1 Ratio
Room Temperature
37°C
5% dextrose in water
Plasmanate
5% dextrose in 0.2% saline
5% dextrose in 0.4% saline
5% dextrose in 0.9% saline
0.9% saline
Normosol-R, pH, 4
Lactated Ringer's Solution
1+
1+
0
0
0
0
0
0 (clotted)
4+
3+
3+
0
0
0
0
0 (clotted)
Fig.8
COAGULATION FACTORS
Coagulation components may be required peri or intraoperatively for treatment of patients with bleeding disorders. It
is important to have guidelines for effective treatment. Figure 9 shows the major stages involved in hemostasis. It is
important to remember that coagulation is a multistage process that is influenced through injury and treatment at many
points.
- 26 -
26
Coagulation Tests
Bleeding disorders can vary from subtle defects that increase the risk of blood loss to major deficits that are
immediately life threatening. Diagnosis is often based on the clinical and/or laboratory observations and are
complimentary for diagnosis. Figure 10 shows the routine tests of coagulation and their indications. An isolated
laboratory number should never be treated without consideration of the clinical scenario.
PRIMARY (SCREENING) TESTS OF HEMOSTASIS USED TO CONFIRM
THE PRESENCE OF A DISORDER HEMOSTASIS
Activated coagulation time (ACT)
Can be done in OR: observe for clot retraction and lysis
Fibrinogen level
Depressed in DIC
Prothrombin time (PT)
Prolonged in liver disease, vitamin K deficiency, coumarin anticoagulation, DIC
Partial thromboplastin time (aPTT)
Prolonged in Factors V, VIII deficiency (massive transfusion), the hemophilias, or
The presence of heparin
Platelet count
Decreased in thrombocytopenia and DIC
Fig 10
Platelets
Platelets are involved in all phases of the coagulation cascade. They form the initial hemostatic plug and activate the
coagulation cascade. Clinical coagulopathy during massive blood transfusion is often initially due to dilutional
thrombocytopenia. However, the spleen and bone marrow provide a reserve supply. Therefore, the platelet count is
never as low as predicted on a dilutional basis alone. Figure 11 illustrates this point.
- 27 -
27
In the past platelets transfusions were only available from pooled or multiple donors. This increases the risk of
infectivity and autoimmunization. Currently pheresis units are available which are derived from one donor. At
University Hospital, PLP are platelets derived from a single donor are equivalent to 3-4 pooled units. PLA is single
donor platelets equivalent to approximately 6-8 pooled units.
Pooled units are only used in times of shortage. Once fractionated, platelets are stored at room temperature and should
not be refrigerated. They have a shelf life of 5 days and must be agitated. Assuming normal platelet function, the
correlation between platelet count and bleeding is shown in figure 12.
CORRELATION BETWEEN PLATELET COUNT AND
INCIDENCE OF BLEEDING
PLATELET COUNT
TOTAL NUMBER
NUMBER OF PATIENTS
(cells/mm3)
OF PATIENTS
WITH BLEEDING
More than 100,000
21
0
75,000-100,000
14
3
50,000-75,000
11
7
Less than 50,000
5
5
Fig. 12
Fresh Frozen Plasma
These units are obtained from centrifuged whole blood from which the platelets have been removed. Each unit is
derived from one donor and can be held frozen up to a one year period. Currently 40 minutes are required for thawing
and the units cannot be refrozen. Plasma units are only usable for 24 hours following thawing. Fresh frozen plasma
contains soluble coagulation factors required for hemostasis. It should never be used for volume expansion in a patient
with normal coagulation.
Cryoprecipitate
This is a concentrated source of fibrinogen, Von Willebrand's factor, fibronectin and factor VIII and is derived from
platelet poor plasma. Once thawed, the shelf life is only 4 hours. Each unit is derived from a single donor and is
subsequently pooled. Although cryoprecipitate is usually administered as ABO compatible, this is not critical since the
concentration of antibodies is extremely low. Prevention of alloimmunization due to Rh is more important as
cryoprecipitate can contain red blood cell fragments. In addition to the above mentioned products, there are factor
concentrates available for treatment of congenital hemophilia. There is now a recombinant DNA factor VIII
concentrate available for Hemophilia A. Figure 13 shows the minimal factor levels needed for hemostasis and the
appropriate therapeutic agent.
- 28 -
28
HEMOSTATIC
FACTOR
I
II
V
VII
VIII
von
Willebrand's
IX
X
XI
XII
XIII
Platelets
BLOOD PRODUCTS FOR HEMOSTATIC DISORDERS
MINIMUM LEVEL NEEDED FOR
IN VIVO
THERAPEUTIC AGENT
SURGICAL HEMOSTATIS
HALF LIFE
(% Normal)
50-100
3-6 days
Cryoprecipitate
20-40
3-4 days
Plasma
5-20
12 hours
Fresh Plasma
Fresh Frozen Plasma
10-20
4-6 hours
Plasma
30
10-18 hours
Cryoprecipitate
Antihemophilic factor
30
Desmopressin
Plasma
20-25
18-24 hours
Plasma
Prothrombin complex concentrate
10-20
2-4 days
Plasma
20-30
2-3 days
Plasma
0
Plasma
1-3
5+ days
Plasma .
50,000-100,000
variable
Platelet concentrates
Fig. 13
DELIVERY SYSTEMS
Rapid blood component delivery may be critical in emergency situations. There are a number of physical factors,
which govern flow velocity. Poiseuille’s law is shown below (Fig 14) and illustrates the variables that determine the
rate of flow.
- 29 -
29
Radius or Size
As can be seen from this formula, intravenous catheter and tubing size are the most important determinants of flow
velocity in a delivery system. Due to this it is important to select sizes that do not limit the rate of delivery. The effect
of sizing on flow rate is illustrated in Figure 15. It is also important to remember that length is negatively related to
flow. It becomes apparent that short and wide radius delivery systems promote rapid flow.
FLOW RESISTANCE OF SELECTED CATHETERS
ED*
RL,+
Rr.+
(mm)
(mmHg.s/ml)
(mmHg.s/ml)
10 G,7.6cm
3.4
0.2
0.2
12G,7.6cm
2.8
0.3
0.3
14 G, 5.1 cm
2.1
2.0
0.9
14G,5.7cm
2.1
4.9
1.9
16 G, 5.1 cm
1.7
1 1.2
8.7
18 G, 5.1 cm
1.2
28.0
25.0
20 G, 3.2 cm
0.9
57.5
57.4
22 G, 2.5 cm
0.7
76.8
162.1
*Stubb’s needle gauge, CRC Handbook of Chemistry and Physics. +Data from Philip and Philip.ED=approximate external
diameter, RL = resistance to laminar flow.Rr = resistance to turbulent flow.
DESCRIPTION
Viscosity
Blood viscosity is related to hematocrit (Fig. 16). As observed from Poiseuille’s law, viscosity affects rate of delivery.
Since the hematocrit of packed red blood cells is approximately 70%, dilution may be required in certain systems to
ensure rapid delivery.
- 30 -
30
Pressure Gradient
Finally it is noted that the pressure gradient also determines flow velocity. A gradient is established in a delivery
system by pressurizing the delivery system at the source. Pneumatic bags are provided in the operating room for this
purpose. In addition the Level 1 warmers incorporate pneumatic devices into their system. In contrast, the Rapid
Infusion System (RIS) found in OR 2 uses a roller device to push fluid at a constant pressure along its tubing. The
combined effects of catheter size, viscosity and pressure gradient are shown in Figure 17.
Warming Devices
There are a wide variety of products on the market available for warming intravenous solutions. Each system functions
within an optimal range for flow rates and this should be checked. If flow rates are too fast, there is insufficient time
for equilibration between the heating device and fluid. If the flow rate is too slow or there is copious tubing beyond the
heating box then the fluid will cool before it reaches the patient.
At University Hospital, we commonly use the Hotline and Level 1 warmers. The Hotline is unique in that it heats
tubing outside the warmer box by a water heated jacket. It is not advantageous to add multiple extensions to the
system since this will increase resistance to flow and promote cooling. The Level I and RIS are single channel counter
current exchangers and capable of heating solutions during high flow. The heating capacity is shown in Figure 18
while cost is illustrated in Figure 19.
- 31 -
31
APPARENT THERMAL CLEARANCE OF WARMING DEVICES
Vrc(ml/min)
TYPE
Q CALCULATED
(ml/min)
Saline*
PRBC*
Single-coil immersion (Hemokinetitherm)
184
15
8
Single-channel dry wall (Fenwal)
256
275
148
Multiple-channel countercurrent (Infuser 37TM)
831
341
231
Single-channel countercurrent (System 500TM)
1356
658
446
*Data from Flancbaum et al. VTC= apparent thermal clearance. PRBC=packed red blood cells. Q calculated=the
calculated flow rate of PRBC infused at 5°C producing an outlet temperature of 32°C.
Fig. 18
COST OF WARMING BLOOD
TYPE OF WARMER
UNIT
DISPOSABLE SET
American Pharmaseal DWI000A (Single-channel dry wall)
$500
$10
System 2501TM
$2,500
$25
System 500 TM
$5,000
$50
Infuser 37 TM
$4,000
$330
Rapid Infusion System TM
$51,700
$590
Approximate prices of commercially available warmers and their disposable sets (December 1990)
Fig 19
- 32 -
32
COMPLICATIONS
Adverse outcome related to the administration of blood products can be divided into two major groups: immune-based
and nonimmune. Immunological reactions result in hemolytic or nonhemolytic (allergic) reactions. These
complications are shown below (Fig 20).
I.
II.
Immunologic transfusion reactions
A.
Hemolytic
1.
Acute hemolytic reactions
2.
Delayed hemolytic reactions
3.
Interdonor hemolytic reactions
4.
Lysis of recipient cells by antibodies in donor serum
B.
Nonhemolytic (allergic) reactions
1.
Febrile reactions
2.
Anaphylactic reactions
3.
Urticaria
4.
Noncardiogenic pulmonary edema
5.
Posttransfusion purpura
Nonimmunologic transfusion reactions
Fig. 20
Acute Intravascular Hemolysis
The most serious reactions are due to acute intravascular hemolysis. This is usually the result of ABO incompatibility
from packed red cells or IgA induced hemolysis from giving fresh frozen plasma to an IgA deficient but sensitized
individual. An intravascular reaction may be difficult to immediately recognize during aggressive resuscitation and
under anesthesia. Therefore a high degree of suspicion is needed. If a reaction is suspected, the following protocol
should be followed. (Fig 21).
Some of the most serious consequences of intravascular hemolysis are due to subsequent disseminated intravascular
coagulation, hypotension with low perfusion state, pulmonary edema and renal failure. Mortality from hemolytic
transfusion reaction varies from 17-53%. Most cases are preventable by careful clerical checking of donor blood units
supplied for transfusion.
Delayed (Extravascular) Hemolytic Reaction
This usually occurs in patients previously sensitized to foreign antigens by such events as pregnancy or prior
transfusion. Sensitized donor red blood cells are cleared by the reticuloendothelial system with cell survival times of
only 2 to 21 days. Although this is usually less serious a reaction than intravascular hemolysis, jaundice and renal
dysfunction can develop. Rh and Kidd are the antibodies most commonly involved.
Nonimmunological
These problems are related to catheter insertion, delivery systems, fluid balance and infectivity of blood products.
Infectivity continues to be a severe problem despite improved detection methods. A rational approach to blood
utilization should help reduce complications by preventing unnecessary transfusion.
- 33 -
33
SECTION III - ANESTHESIA FOR LIVER TRANSPLANTATION
The following information is intended as an abbreviated GUIDE to anesthesia for liver transplantation. It does not
contain background information necessary for critical decision making during the surgical and perioperative periods.
This information is available from the standard liver manual in the library and should be reviewed at the beginning of
your liver transplant rotation.
PREOPERATIVE EVALUATION
All patients listed and approved for liver transplantation have been seen during their initial assessment by
Anesthesiology. Typed consultations are available on file. These should be obtained and reviewed. Since many
patients are placed on a waiting list, the consult may not be current and the health status of the patient may have
changed. It is therefore vital that each patient undergo early pre-operative assessment by the resident on call. All
patients must be regarded as full stomachs. Please use the Pre-operative Liver Transplantation Anesthesiology
Assessment form included with the Anesthesiology Record. This form will help you categorize disease complications
important for formulating a plan for patient care. The MELD (Model for End Stage Liver Disease) is a scoring
system for prioritizing patients waiting for liver transplantation similar in principle to Apache scoring. MELD is
calculated from the bilirubin, INR and creatinine. You can access the MELD calculator at www.UNOS.org for further
information.
Review of laboratory results: Dobutamine echocardiography, complete blood counts, electrolytes, coagulation indices,
and PFT's are usually available. Always obtain a pre-operative set of vital signs, as hypotension, hypoxia and
respiratory alkalosis are part of the disease process. It is common to request both a family member and patient
signature on the Consent form since many patients have cognitive impairment due to encephalopathy. Do not give any
preoperative medication without consultation with the attending anesthesiologist.
To protect confidentiality of the patient and comply with HIPPA rules, ALL paper copies of the consult MUST be
returned to the Anesthesiology Office.
It is the attending anesthesiologists responsibility to advise the Blood bank how many blood products to prepare. There
are currently three standard requests. These requests are outlined on pages 59-61. All products are brought to the
Operating Room in a thermally insulating container and checked to confirm correct identity prior to surgery. Two
individuals, RN or MD, are needed to confirm identity. Products must not be left at room temperature and should only
be removed from the container with the intent of immediate transfusion. Sign all transfusion forms.
The operating room should be prepared in advance for anesthetic administration. The setup can be completed with the
exception of the Rapid Infusion System (RIS). All pressurized lines must be free of heparin. Following final clearance
by the attending surgeon, the RIS may be assembled.
The patient will then be ready for transport to the operating room. At this point, a usual nursing pre-operative check
will be carried out.
INTRAOERATIVE ANESTHESIA OVERVIEW
The surgical procedure is divided into three distinct stages, each with its own special set of anesthetic considerations
and accompanying problems. The protocol outlined below is based upon this classification and therefore each stage
will be briefly described.
Stage I PRE-ANHEPATIC STAGE
The pre-anhepatic stage I begins with the surgical incision and includes native liver dissection and mobilization.
During this stage, the surgeons will identify the hepatic artery, portal vein and the inferior vena cava above and below
the liver. If required, veno-venous bypass is implemented during this stage. Stage I concludes with vascular isolation
of the liver.
Stage II ANHEPATIC STAGE
- 34 -
34
Stage II commences with the occlusion of the hepatic artery and portal vein. Following removal of the native liver,
vascular anastomoses of the donor graft are usually then performed in the following order: suprahepatic vena cava,
intrahepatic vena cava, portal vein and hepatic artery. Hepatic arterial anastomoses may be performed in Stage III. The
inferior vena cava may remain partially patent during a "piggyback procedures" as described on page 44. The
anhepatic stage concludes with removal of the vascular clamps resulting in reperfusion of the donor liver graft.
Stage III POST ANHEPATIC STAGE
Stage III starts during reperfusion and extends to the conclusion of the operation. It mainly encompasses the process of
biliary reconstruction.
Protocol
Pre-operative Preparation
All patients listed and approved for liver transplantation have been seen and discussed by the multidisciplinary Liver
Transplant Team at the Selection Committee. Consultations performed by the attending anesthesiologist for the
Selection Committee are available on Clinical Work Station or on paper in the Anesthesiology Administrative Offices.
Stage I
Patients should be transported to a heated operating room. The temperature within the room must be raised to at
least 80° F in advance to prevent early cooling. Noninvasive monitors are applied after transfer to the table. Pre
oxygenation begins immediately since hypoxia is common in liver disease. Ativan 1-2 mg can be given iv for sedation
and prolonged amnesia if there is no history of encephalopathy. Careful observation of the patient is required
following the administration of sedatives since end stage liver disease increases the sensitivity to sedatives. The
induction is carried out as a rapid sequence.
The arterial and central venous access are commonly placed post induction but the timing of placement is at the
discretion of the anesthesiologist. All lines must be placed under strict aseptic conditions. Many of these patients have
profound coagulopathies and treatment with clotting factors may be elected prior to the insertion of invasive monitors
or vascular access. A baseline thrombelastogram (TEG) can be obtained before the administration of coagulation
factors to help pinpoint specific clotting defects.
A 9.0 Fr introducer is placed into the internal jugular or subclavian artery. We do not routinely place pulmonary artery
catheters in liver transplant patients at University Hospital since there is still considerable debate in the nation
regarding the risks and benefits of routine catheterization in these patients patients. Rather, the decision to insert a
pulmonary artery catheter is based upon the identification of an additional underlying condition that would warrant
such an intervention in any surgical candidate.
The RIS is connected to a 9.0 or 8.5 Fr introducer. Common sites for 8.5 Fr Rapid Infusor Catheters (RIC) are the
brachial or cephalic vein. The RIS entry site should be pressure tested and inspected for interstitial extravasation. The
RIS reservoir will have a Hct of 27% if the RIS is primed with 25Occ crystalloid, 1 unit FFP and 1 unit red blood
cells. If other blood product proportions are used, a hematocrit (Hct) can be sent from the reservoir of the RIS to the
blood bank in a heparinized syringe.
Attention to pressure points is important. The patient is placed supine on a gel pad with a heating pad under the gel
pad. The gel should be checked to assure that the heating pad does not contact the patient directly. Heels are placed in
foam pads and soft support is placed behind the knees. Each arm is positioned with slight flexion of the elbow and
functional position of the hand. The left arm is placed on an extension with an angle of less than 90 degrees to the
body., while the right is commonly tucked at the patient’s side. The neck should be in neutral position with support.
- 35 -
35
The right shoulder should not contact the retractor pole.. Once upper and lower extremities Bair Huggers are placed,
the room temperature may be reduced for the comfort of the operating room staff.
Antibiotics are given prior to the start of surgery. The following infusions are started after induction unless there are
specific reasons not to: renal dose dopamine (2ug/kg/min), 20% mannitol at 20-50 cc/hr (4-10 g/hr), magnesium 5g
over a5hr (1g/hr). After consultation with the attending surgeon, the anesthesia team may elect to use octreotide
(50ug/hr) to reduce splanchnic flow and portal hypertension during the pre and anhepatic stage. The attending
anesthesiologist may also elect to use aprotinin. A test dose of aprotinin (1cc) must be given prior to infusion and
documented on the record.
Stage II
Significant changes in blood chemistries occur during the anhepatic stage warranting careful attention to monitoring
arterial blood gases at this time. During stage II the patient is placed on 100% oxygen. The anesthesia goals during
Stage II are to maintain normal electrolytes, pH and an Hct of approximately 30-35%. Electrolytes are corrected to
within normal limits since compensatory responses by the patient cannot occur in the absence of liver function.
Hyperventilation during this stage to approximately 30mmHg will prepare the patient for a large increase of pC02
during the unclamping phase. Similarly correction of the base deficit prepares the patient for a large acid load
following removal of the portal venous crossclamp. The base buffer THAM (triethanolamine) is used in preference to
bicarbonate since it is sodium free. Large fluxes in the serum sodium level, especially in patients with pre-existing
hyponatremia can cause central pontine myelinosis, a debilitation chronic neurological disease. If sodium bicarbonate
must be used to treat a base deficit, one formula for determining dosing is: HC03meq = [Wt (Kg) X 0.2] X BE. Where
0.2 is the volume of distribution of bicarbonate and BE is the base excess (or deficit). Calcium can fall significantly
with the administration of blood due to the chelation of calcium by citrate in the blood preservative. This is
particularly severe in the anhepatic stage since there is no liver to metabolize the citrate. Potassium levls greater than
3.5 meq/dl are associated with reperfusion syndrome, necessitating a purposeful reduction in serum potassium to less
than 3.5meq in preparation for reperfusion of the liver graft.
Veno-venous bypass has not been used at the University of Colorado for hemodynamic support of blood pressure
during the anhepatic stage since 1995. If special conditions arise necessitating the use of veno-venous bypass, the
room temperature must be elevated to prevent hypothermia as the bypass pump and lines are unheated. Flows from
the femoral and/or portal veins to the bypass pump must be kept greater than 800 ml/min. Should flows fall below this
critical level, the risk of thrombus formation will necessitate discontinuing bypass
During the infrahepatic caval anastomoses, the donor liver is flushed with crystalloid to reduce the amount of
preservative fluid and air in the donor graft. The preservative solution contains 150 meq/dl of potassium which can
caused diastolic cardiac arrest in the recipient during reperfusion. Once this procedure is initiated, the conclusion of
stage II is near. If veno-veno bypass has been used, the portal cannula is removed just prior to portal vein anastomoses.
Venous return and consequently blood pressure can fall precipitously.
Solu-medrol is commonly given just prior to release of the crossclamp. Because steroids are contraindicated in some
patients and the dose may vary, check with the surgeon before drug administration.
One ampule of calcium chloride (1g) is usually administered just prior to release of the cross-clamp to counteract the
high potassium levels that exit from the donor graft during reperfusion. Rapid blood pressure changes and arrhythmias
may last 1-5 min following reperfusion.
Stage III
Correction of any underlying acid-base or electrolyte abnormalities should be based on the understanding that the new
liver may be able to compensate. The status of the new liver must be checked with the surgeon. Assessment of
function is usually determined by the production of bile and color of the organ. There are no reliable clinical markers
of organ function. However, the ability of the patient to increase their body temperature, hemodynamic stability,
continued urine output, increased carbon dioxide production and a stable serum biocarbonate level suggest graft
- 36 -
36
function. Post reperfusion bleeding that cannot be surgically localized suggests graft dysfunction. It is often difficult
to differentiate between bleeding due to coagulation defects and surgical loss. Laboratory tests and TEG may be
helpful.
Correction of the coagulation status may be required at this time. Therapy should be guided by analysis of the TEG
and clinical impression. Each unit of FFP contains 5% of the coagulation requirement and under stable conditions
only 30% of factors are required for normal clotting function. Under ideal circumstances one platelet concentrate will
usually produce an increase of 7-10,000 platelets/m3 1 hour after transfusion in a 70kg adult without hypersplenism.
When coagulopathies are treated with platelet concentrates, the need for FFP may be reduced. A plasmapheresis unit
of platelet contains approximates the plasma coagulation factor content of one unit of FFP.
A cholangiogram (in expiration) may be performed before the abdomen is closed. Because closure of the surgical
incision increases central filling volumes and pressures, the CVP must be monitored closely and optimally reduced to
a pressure less than 12 mm Hg to prevent venous back pressure on the liver graft.
The room is reheated towards the end of stage III and at the conclusion of the operative procedure. Patients are either
prepared for immediate postoperative extubation or transport to the ICU with continuous hemodynamic monitoring.
Patients who are extubated in the Operating Room can be transferred to the Post Operative Care Unit or the SICU.
The postoperative site of care is determined by consensus between the attending anesthesiologist and surgeon.
Common Clinical Problems
Pre-operative
One of the most common problems identified in the chronic liver patient is the presence of antibodies from previous
transfusions. If compatible blood products are limited, the attending anesthesiologist and blood bank director must a
identify a plan in the case of massive blood loss. In severe circumstances, plasmapheresis may have to be performed
during the surgical procedure.
A careful review of medications is essential, since diuretics are commonly used drugs. It may be difficult to control
potassium levels during surgery since spironolactone blocks potassium secretion sites in the distal renal tubule. This
can be further complicated by the presence of hepatorenal syndrome.
If dialysis is required during the procedure, the Fellow on call for dialysis must be informed at an early stage and a one
hour start up time provided for the machines. The type of dialysis bath is discussed with the Fellow in advance.
As outlined earlier, care must be taken in giving sedatives to liver failure patients due to their increased sensitivity.
Medications including metoclopramide may lead to an exaggerated response including oversedation and extra
pyramidal side effects.
Vascular Access
Difficulty may be encountered in obtaining vascular access. The following is a list of access sites given in preferred
order. Site-Rite and Sonosite equipment is available in the Operating Room.
A.
Arterial lines
1.
L radial
2.
R radial
3.
R femoral
4.
R axillary
B.
RIS line venous access
1.
L antecubital
2.
L internal jugular
- 37 -
37
3.
4.
L eternal jugular
L subclavian
Elevated serum potassium can be a life threatening complication leading to cardiac arrest and reperfusion syndrome
during surgery. Conditions that predispose to elevated serum potassium levels include prerenal fluid deficits caused
by diuretics, potassium sparing diuretics and varying degrees of hepatorenal syndrome. Rehydration will dilute
potassium while Lasix can promote excretion in some patients. Alkalosis, which shifts potassium intracellularly, can
be induced by hyperventilation and base buffer administration. Glucose and insulin can also decrease serum levels by
intracellular shifts. However, if insulin is administered careful follow up of the blood sugar is necessary because
insulin is poorly metabolized in liver disease and hepatic glycogen supplies are often depleted. A D10/W drip is often
required to maintain normoglycemia after insulin administration.
Blood loss during this stage can be significant. This should be anticipated in any patient who had previous intra
abdominal surgery especially in the right upper quadrant. Underlying coagulopathies may exacerbate the above
problems. Mechanical reasons for excessive bleeding in liver failure patients include increased portal pressures with
excessive splanchnic venous filling, high central venous pressures and elevated cardiac output. Preliminary evidence
at the University of Colorado suggests that a reduction in splanchnic filling by octreotide and cardiac output by beta
blockers can reduce blood loss without any adverse effects upon patient or graft outcome.
Stage II
Ligation of the inferior vena cava often results in hemodynamic instability in the absence of veno-veno bypass. This is
more pronounced in older patients and those with severe portal hypertension pressures and is due to a fall in venous
return. Consequently the CVP may read very low during stage II and central filling pressures do not reflect the total
fluid status. The surgeon often performs a trial of cross clamping. During this 2-3 minute period vasopressors and
fluids are used to maintain blood pressure. Vasopressors are preferred since large amounts of fluids can result in high
venous pressure below the crossclamp site resulting in renal impairment and bowel engorgement with ischemia. The
new liver will also be compromised by the resulting high venous pressures following reperfusion. Phenylepherine is
the preferred vasopressor since it acts at venous and arterial sites, improving venous return and compliance.
The initial effects of caval ligation are attenuated by veno-veno bypass. Blood pressure and central filling pressures are
partially maintained by pump return. Normal flow rates are approximately 1.5-2.5L/min. Hypotension during bypass
may result from kinking or obstruction in the pump system. The two inflow cannulae usually arise from the portal and
femoral veins. A single cannula from the femoral vein may be used. The cannulae must be checked carefully in the
presence of low flow. Flow rates less than 0.8L/min are unacceptable since this leads to thrombus formation within the
system. This may occur during the portal vein anastomoses once the portal cannula is removed and bypass should be
discontinued. Venous air embolism is a rare but significant complication. Bypass is contraindicated in portal vein
thromboses and Budd-Chiari syndrome.
Common electrolyte abnormalities during the anhepatic stage include metabolic acidosis, hyperkalemia and
hypocalcemia. These abnormalities must be completely corrected since the anhepatic patient has no means of
physiologic compensation. Hypernatremia caused by the treatment of metabolic acidosis with bicarbonate can be
treated with free water (D5/W). Ventilation parameters must be carefully adjusted when administering bicarbonate to
compensate for the rapidly changing pC02. As discussed above, serum potassium must be aggressively lowered to less
than 3.5meq/L to prevent asystole and reperfusion syndrome. Hypocalcemia should be treated early to maintain cardic
performance. The severity of these electrolyte abnormalities is related to the length of stage II, severity of ptient
illness, hypotension and the number of blood products administered.
Stage III
Stage III begins with reperfusion of the graft liver. The sequence of unclamping usually is: suprahepatic vena cava,
infrahepatic vena cava, portal vein and hepatic artery. Blood pressure can precipitously increase following release of
the infrahepatic vena cava due to the large amount of blood returning from the splanchnic to the systemic circulation.
Release of the portal vein clamp directs blood through the liver graft to the heart. Products of cell death and residual
- 38 -
38
preservation fluid can cause severe hypotension, bradycardia, supraventricular and ventricular arrhythmias,
electromechanical dissociation and occasionally cardiac arrest. Many of these complications can be attenuated by
meticulous management during stage II.
Approximately 30% of patients will exhibit profound cardiovascular collapse on reperfusion irrespective of attentive
management during stage II. This lasts for 1-5 minutes following unclamping. "Reperfusion Syndrome" is
characterized by bradycardia, myocardial depression and systemic vasodilation. Generally treatment with calcium
and/or epinephrine improves cardiovascular function. The use of fluids should be judicious since transfusion can
aggravate the already increased filling pressures and this may result in impairment of hepatic perfusion. Although the
hemodynamic changes generally subside within 10-15 minutes, pulmonary hypertension, elevated CVP and
hypotension may persist. The etiology of this syndrome is thought to be caused by the release of vasoactive substances
from the grafted liver and is therefore not predictable.
Hypotension may also be due to surgical bleeding as the anastomatic sites are exposed to venous pressure. Mild
coagulation defects are common during stage III as the new liver requires time to resume its synthetic functions.
Treatment should be guided by the TEG. If the surgeon anastomoses the hepatic artery to the aorta, continuous
bleeding can occur during retroperitoneal dissection.
FULMINANT HEPATIC FAILURE
Most patients that present for transplantation have chronic liver disease. The clinical course and medical complications
of end stage liver disease are better understood than those associated with fulminant hepatic failure. Because
fulminant hepatic failure is an indication for urgent liver transplantation a brief description of observation on the
clinical course of these patients is described below.
Fulminant hepatic failure is defined as the presence of encephalopathy within eight weeks of the onset of jaundice.
Mortality rates of 80% to 90% occur in these patients, in spite of aggressive medical intervention. In contrast, survival
rates of 60% and 67% have been achieved in Pittsburgh and UCLA respectively with emergency liver transplantation.
Common causes of fulminant failure include: viral hepatitis (A,B), idiosyncratic drug reactions, toxic drug reactions,
Wilson's Disease and indeterminant. Patients commonly present at the time of transplantation with symptoms of
cerebral edema, coagulopathy, hypoglycemia, respiratory failure and renal impairment. Death most commonly occurs
in these patients due to elevated ICP with consequent brainstem herniation or sepsis.
The etiology of cerebral edema in these patients is poorly understood, but may be due to both vasogenic and cytotoxic
mechanisms. Fluid shifts and drugs given during surgery may affect outcome. The important issues in the care of these
patients include the prevention of brainstem herniation prior to surgery, selection of patients for transplantation and
maintenance of cerebral perfusion pressure intra-operatively.
Pre-operatively, patients are closely monitored in an ICU setting. ICP monitoring is becoming increasingly common
and pressure increases are treated aggressively. Decerebrate posturing is not considered a contra-indication to
transplantation and full neurological recovery has occurred in these patients. A different mechanism may underlie
posturing in liver failure patients as opposed to those who are head injured. Clinical findings indicative of poor
neurological outcome are: the absence of brainstem reflexes, a flat EEG over 24h or persistent elevation of the ICP
above 40mmHg. There is a poor correlation between findings on CT scan and ICP measurement.
Despite the different etiology and presentation of cerebral edema in these patients, intra-operative management utilizes
standard techniques of ICP reduction. Most transplant candidates will have a subarachnoid bolt in place linked to a
Camino monitor. Both a waveform and a digital readout are required for pressure interpretation. By enlarge, ICP is
usually lowest in the anhepatic stage (II) in the absence of veno-veno bypass. However, vasopressors and fluid boluses
may influence pressure values. Veno-veno bypass seems to be associated with higher ICP values. This may reflect the
higher central venous filling pressures. Reperfusion (Stage III) is associated with higher ICP values and precipitous
increases may occur. In addition, hypotension during these stages may result in diminished cerebral perfusion
pressure. Associated coagulopathy increases the risk of intracerebral hemorrhage.
- 39 -
39
Intracranial hypertension is treated with hyperventilation, mannitol, loop diuretics and barbiturates. Since cerebral
blood flow may be reduced in the encephalopathy of fulminant hepatic failure, the safe limits of hypocarbia are
unknown. The administration of mannitol may also have to be considered carefully in the presence of renal failure.
Hemodialysis or CAVH (continuous arterio-venous hemodialysis) can be used to control filling pressures. However,
hemodialysis has been associated with paradoxical change in the ICP. Finally, both sodium thiopental and
pentobarbital have been found useful in decreasing intracranial hypertension. Pentobarbital is associated with less
hypotension than pentothal. The usual dose is 2-5mg/kg/h. Pentothal can be used for precipitous increases but
hypotension must be avoided due to reduced perfusion pressures.
Other co-existent conditions may contribute to poor neurological outcome. Hypoglycemia, which occurs commonly in
these patients must be avoided. On the other hand, the relationship between elevated blood glucose and neural injury is
not as well defined as in head injury. Elevated serum glucose levels are common in stage III after the administration of
steroids. Airway and central filling pressures must also be controlled.
Respiratory, failure commonly occurs as neurological status deteriorates. Initially oxygenation and ventilation are
adequate. However pneumonitis often develops over time. The etiology of this problem may arise from both infective
and toxic mechanisms. Patients with high airway pressures may require an ICU ventilator in the OR.
Coagulopathy is initially due to loss of the synthetic function of the liver. This is usually correctable with blood
components. Early in the course of the disease, platelet number and function are adequate. As multiple organ system
failure progresses, secondary fibrinolysis becomes a problem with consumption of both factors and platelets. There
may also be acute bone marrow depression with a resulting decrease in all cellular elements. Often, coagulopathy
improves once the diseased liver is removed.
Renal failure may have a variety of causes including: hepatorenal syndrome, drug toxicity, sepsis and DIC. If oliguria
is present or potassium levels are not well controlled, intra-operative dialysis will be necessary. Alternatively, if only
fluid control is required, CAVH can be used.
Finally, it is important to remember that many patients with fulminant failure of unknown etiology may actually have
viral causes. Universal precautions should be exercised in all cases irrespective of the diagnosis.
- 40 -
40
CIRCULATORY PATHOPHYSIOLOGY AND OPTIONS IN HEMODYNAMIC MANAGEMENT DURING
LIVER TRANSPLANTATION
Liver transplantation is an accepted mode of therapy, which improves life expectancy in patients with advanced and
progressive hepatic failure. Despite scientific advances in this area, hepatic transplantation is still associated with
increased risk of morbidity and mortality. Vascular isolation and reperfusion of the liver can cause significant
cardiovascular stress leading to perioperative complications. The hemodynamic response to vascular isolation is
complex, modified by the presence of a hyperdynamic circulation, identified in most patients with liver disease.
The development of increased cardiac output and low peripheral vascular resistance, or hyperdynamic circulation
differs among patients with liver disease. The response to surgical stimuli is therefore variable, modified by individual
differences in patient cardiovascular physiology. Consequently, a single approach to intra operative management of
the anhepatic stage may not be optimal for all patients. Rather, a selection of techniques designed to modify
hemodynamic parameters may help improve patient response to surgical intervention.
Innovative surgical techniques designed to reduce intraoperative hemodynamic stress associated with vascular
isolation of the liver include venous bypass (VBP) and preservation of the inferior vena cava or "piggyback
technique". Management of caval crossclamping, by optimization of intravascular volume and use of vasopressors has
been used as an alternative to surgical intervention for hemodynamic support. Despite an increased understanding of
cardiovascular responses during liver transplantation, the success and indications for use of these techniques has
remained controversial.
Cardiovascular Response to Hepatic Vascular Isolation and Donor Graft Reperfusion
The dissection stage, which can present considerable surgical challenge, is often uneventful hemodynamically when
compared to the anhepatic stage2,3. Hypotension, however, can occur during a difficult dissection requiring urgent or
early vascular isolation4,5. In anticipation of a difficult dissection, stage II may be initiated early to maintain vascular
control and prevent excessive bleeding.4 This will prolong the anhepatic stage along with its associated hemodynamic
abnormalities.
Hemodynamic changes during the anhepatic stage mainly result from occlusion of the vena cava 6(Fig 1). Upon
interruption of caval flow there is a fall in central fall in central filling pressures combined with increased resistance in
both systemic and pulmonary vascular beds. As a result, cardiac output decreases by an average of 48% 6. A similar
decrease in pulmonary wedge pressures illustrates the effects of acute left ventricular preload reduction2,7. As stoke
volume falls, a compensatory 15% increase in heart rate and 66% elevation of systemic vascular resistance help
maintain arterial blood pressure4,6,7. Venous hypertension can develop inferior to the crossclamp leading to splanchnic
and renal congestion4 and impaired tissue perfusion. Isolated occlusion of the portal vein in patients with portal
hypertension does not usually cause significant hypotension 8.
Increased venous return following reperfusion leads to a rise in cardiac output with lower extremity and splanchnic
venous decompression4,6. Systemic vascular resistance falls and contributes to development or exacerbation of
hyperdynamic circulatory state7,9,10-12.
A subset of individuals experience significant hypotension most pronounced
after portal venous flow is re-established10,13,14. A fall in mean arterial blood pressure greater than 30% of baseline for
at least one of the first five minutes after unclamping15 bradycardia and elevated central filling pressures has been
defined as postreperfusion syndrome6. The incidence and severity of reperfusion syndrome may be influenced by
management during the anhepatic phase.
Pathophysiology of Circulatory Changes Associated with the Anhepatic Stage
There is no direct correlation between baseline cardiac output and intraoperative hemodynamic stability during the
anhepatic stage16. This suggests that additional factors modify hemodynamic changes. An acute fall in venous return
following caval occlusion is one of the mechanisms leading to hypotension. Individual variation in venous return and
therefore cardiac output is determined by several factors, including redistribution of blood flow, and patterns of
collateral flow.
- 41 -
41
Similar to aortic crossclamping17, caval occlusion creates a pressure gradient which drives redistribution of the
intravascular volume4. There is relative hyperemia inferior to the cross clamp and venous return depends on the type
and extent of collateral flow18. The development of vascular shunts, which varies among patients with hepatic failure,
constitutes part of the collateral flow19. Shunts may arise from neovascularization or pressure dependent enlargement
of pre existent potential portasystemic connections. Shunts inferior to the crossclamp may act as capacitance vessels,
sequestering intravascular volume and regulating rates of venous return.
Angiogenic activity influences the density and distribution of vascular shunt formation 20. Hepatocyte growth factor,
elevated in chronic liver disease21, has been shown to act as a potent paracrine mediator for new blood vessel
formation in vivo22, 23. Serum hepatocyte growth factor generally correlates with the severity of liver disease and
decreases following transplantation21,24. However, variations in cell expression may determine differences in
individual neovascularization and consequently collateral blood flow. Expression of new vessel formation in addition
to enlargement of portasystemic connections will influence central venous return and therefore cardiac output
following caval occlusion.
During the anhepatic stage, reflex changes in systemic vascular resistance, which varies among patients probably
modifies arterial blood pressure and organ perfusion 7. Changes in peripheral resistance are in part due to
vasoconstrictive responses to acute preload reduction following redistribution of the intravascular blood volume.
Abnormalities in vascular resting tone and vasogenic activity, correlate with severity of liver disease as an expression
of the hyperdynamic circulatory state1. These changes will influence cardiovascular compensatory reflexes to acute
preload reduction.
It has been shown that an endothelial vascular defect contributes to impaired vascular contractility that leads to
vasodilation25. Various endogenous factors, including nitric oxide25, tumor necrosis factor26, 27 and possibly other
cytokines or endotoxins have been implicated as vasodilatory agents acting indirectly through endothelial cell
receptors28-29. The study of impaired vascular response in cirrhotic rats has shown good correlation with increased
levels of the constitutive form of nitric oxide synthetase in major conductance vessels 25. Although similar studies have
not been carried out on capacitance vessels, the use of nitric oxide synthesis inhibitors normalizes hyperdynamic
cardiovascular indices in experimental animals30.
Despite the presence of elevated cardiac output in hyperdynamic circulatory state, impaired myocardial function has
been reported in portal hypertension associated with and without liver disease in humans and experimental
animals31,32. Abnormalities in both systolic and diastolic function, described in a subset of patient 31, may inadequately
pair cardiac output to the degree of vasodilation. Similar observations on patients with septic shock 33-35 have been
attributed to abnormalities in right ventricular compliance 33-35. Once contractility has been maximized, increases in
heart rate are required to promote forward flow in an afterload reduced circulatory system. In patients with sepsis,
compensatory increases in heart rate has been correlated with negative outcom 34. Under resting condition in patients
with hepatic failure, right and left systolic ventricular function appears normal. However, similar to mural
regurgitation, ventricular function may seem normal when there is minimal impedance to ejection. Alterations in
preload and afterload conditions described during clamping of the inferior vena cava 7 may enhance myocardial
dysfunction in some patients, leading to suboptimal stroke volume ejection. Myocardial dysfunction may occur due to
increased loading of the ventricle.
Although there are no reliable predictors of hemodynamic instability during venous occlusion, marked hyperdynamic
circulatory state, severe portal hypertension, and advanced age have been associated with adverse hemodynamic
events36-39. The underlying etiology of this association is unclear. However, abnormalities in vascular response,
collateral blood flow and myocardial compliance, which commonly occur in patients with hepatic disease, may be
accentuated in these coexistent conditions.
Immediately following reperfusion, up to 30% of patients exhibit transient hemodynamic instability 15. The etiology of
abrupt decreases in systemic vascular resistances observed in recipients at this time include rapid increases in serum
potassium, decreases in temperature and acute acidosis7, 10, 14, 40, 41. Vasoactive mediators from the donor liver
including prostaglandins, kallikrein, platelet activating factor and leukotrienes have also been implicated 40,41.
Increased age, hypothermia and larger donor organs are considered risk factors 42.
- 42 -
42
Normal cardiac function has been reported in most patients following reperfusion 43. Right ventricular failure, however,
observed in some patients has been implicated as a possible factor in reperfusion hemodynamic instability 44,45. Both
transesophageal echocardiography44 and right ventricular ejection catheters have suggested transient right ventricular
pressure overload most pronounced after portal venous flow is re-established10, 45. Bulging of the interventricular
septum into the left ventricular chamber and the development of tricuspid regurgitation have been reported 45,46.
Vasoactive substances reperfusion may play a released from the donor liver during role in right ventricular
dysfunction'°. Thrombo-emboli and venous air emboli, identified during reperfusion may also contribute to elevated
pulmonary vascular resistance44, 46. Release of the enzyme L-arginase during reperfusion of the graft has been
observed and can cause a transient decrease in arginine, the precursor of nitric oxide. This may consequently cause a
temporary imbalance in pulmonary vasogenic mediators and increase resistance to right ventricular output47.
Management of the Anhepatic Stage and Hemodynamic Effects: Venous bypass
Extracorporeal circulatory assist devices designed to improve venous return during the anhepatic stage 48 were
introduced into clinical practice in 1983 49-50. The use of venous bypass prevented most hemodynamic alterations
observed during caval crossclamping 51,52. Initial studies suggested better hemodynamic control reduced
intraoperative bleeding, postoperative renal dysfunction and improved thirty day mortality. This led investigators to
encourage the use of VBP in all liver transplant recipients51, 52.
Numerous studies during the anhepatie stage in patients supported with VBP have shown that cardiac output and
stroke volume are maintained near baseline values36, 51-53. Systemic and pulmonary resistances increased, but
significantly less than following simple venous occlusion. Heart rate and mean arterial pressure were stabilized
without volume loading or vasopressor support in most cases. Decompression inferior to the crossclamp by the bypass
circuit reduces intravascular redistribution, improving venous return, central filling pressures and hemodynamic
stability (Fig 2).
A correlation between VBP flow rates, cardiac output and pulmonary artery diastolic pressures has been identified for
individual patients54-55. In contrast, poor correlations between these variables observed across a population of
transplant recipients indicate diversity in cardiovascular reserve.
Renal failure, associated with increased post operative mortality56-58, became an important issue regarding the use of
VBP. Renal perfusion pressure, calculated as the mean arterial blood pressure minus inferior vena cava pressure has
been used as an index of organ perfusion and a guide to management during surgery4. Improved values were observed
with the use of VBP59. Intuitively it would seem that preservation of renal perfusion pressure should decrease the
incidence of postoperative renal failure. However, no direct correlation between values for renal perfusion pressure
and renal outcome has been found39.
In contrast, the important determinants of perioperative renal function include preoperative renal damage 58,
immunosuppressive toxicity58-60 and donor liver function61 rather than the use of VBP. Many investigators currently
view postoperative renal failure as an expression of poor outcome associated with multisystem organ failure and not a
primary event that causes increased mortality.
The benefits and indications for VBP remain controversial due to the lack of randomized clinical trials39. Outcome
studies suggest VBP is not required in all transplant recipients8, 37-39, 61, 62. It is recognized, however, that certain
patients exhibit excessive hemodynamic instability following venous occlusion. Although patient factors have been
associated with increased risk of hemodynamic deterioration during stage II, accurate prediction is not always
possible.
It is accepted, however, that bypass provides stable central filling pressures throughout surgery44,46. Continual venous
decompression with VBP also controls high volume return following release of the inferior vena caval crossclamp.
This may be advantageous in patients with limited cardiac reserve due to coronary artery disease, cardiomyopathy,
valvular heart disease or pulmonary hypertension.
Considerations regarding the use of VBP must be tempered with the recognition of reported complications. Both fatal
pulmonary thromboembolism63 and venous air embolus64, 65 have resulted from the use of bypass circuits.
- 43 -
43
Hypothermia from use of unheated circuits66 and acute superior vena cava syndrome in the presence of superior vena
caval thrombosis or shunt67 have been described.
Medical Management
Approaches to the management of venous occlusion during stage II without VBP include intravascular fluid loading
and in several studies, vasopressor support. During hemodynamic studies, fluid has been given prior to crossclamping
until a predesignated pulmonary diastolic or wedge pressure was obtained 36,67-69. Optimal intravascular volume may
vary significantly for individual recipients depending on the severity of liver disease and underlying cardiovascular
reserve. Furthermore, acute changes in intravascular volume distribution following vena caval occlusion may
unpredictably alter central filling pressures. Unfortunately no study has evaluated this relationship in a population of
liver transplant candidates. Therefore use of a single central filling pressure as a criterion of optimal intravascular
volume for all patients prior to crossclamping may have inherent error. The length of the anhepatic stage, a possible
confounding variable modifying outcome, has also not been independently evaluated in these studies.
Some investigators add vasopressors if hemodynamic instability occurs following vascular occlusion4,8,61. The use of
drugs with combined alpha and beta activity has been described 4, 13. The chronotropic and inotropic stimulation of beta
receptors may help to increase cardiac output and peripheral perfusion.
Venoconstriction of capacitance vessels following alpha receptor stimulation may aid venous return partially by
decompression of the splanchnic circulation. The redistributive effects of venoconstriction may be as important as
chronotropic or inotropic stimulation in improving cardiac output70. In addition, alpha stimulation may increase
abnormally low vascular tone in conductance vessels of patients with liver disease 25.
A decreased incidence of hemodynamic instability during reperfusion has been reported in patients not supported with
VBP4. Fluid and often vasopressors were used to support blood pressure during the anhepatic stage 4. Results of this
study suggest that the incidence of reperfusion syndrome may be influenced by vasoactive support during the
anhepatic and early reperfusion stage. Increased venous return following splanchnic decompression may also improve
cardiac compliance curves and consequently cardiac output.
Right ventricular enlargement with pressure overload reported in some patients following reperfusion44, 46 has not
been identified in most patients supported during the anhepatic stage with VBP 43,44. Aggressive volume
administration during the anhepatic stage can cause transient volume overload due to excessive venous return during
reperfusion in patients not supported with VBP. Further studies will be required to carefully examine the relationship
between bypass and changes in right ventricular function and pressure.
Trial of Crossclamp
Some investigators have performed a trial of vena caval occlusion, evaluating the patient for hemodynamic stability
prior to crossclamping. Patients that fail to meet predesignated hemodynamic variables have undergone hepatectomy
with VBP support. Various criteria for hemodynamic stability have been cited in the literature and are usually based
on relationships between cardiac output, central filling pressure and mean arterial blood pressure 37,38,71. These criteria
have not been evaluated for independent outcome.
Studies have shown that patients who tolerate a trial period of hepatic vascular exclusion may not require the use of
VBP for hemodynamic support8,39,61,73. Whether the use of VBP in these patients would have resulted in better
outcome has not been addressed39. In addition, the benefit of VBP in patients who do not achieve pre specified
hemodynamic indices during a trial of vascular exclusion is also unknown. A trial of crossclamp, however, identifies
at least two groups of transplant candidates who exhibit different hemodynamic responses to a surgical event. This
observation alone underscores the need for options in intraoperative management.
The Piggyback Procedure
A technique of recipient hepatectomy with direct anastomosis of the donor inferior vena cava with recipient hepatic
veins74, "the piggyback approach" has been described 75,76. Used with VBP support, the piggyback technique was
introduced to reduce bleeding associated with retrocaval and retroperitoneal dissection 75. A further modification
- 44 -
44
which placed a tangential clamp across the vena cava isolated the hepatic veins for donor anastomosis, yet preserved
caval flow throughout the anhepatic stage77-79. (Fig 3)
Venous bypass was not used since the maintenance of venous return during stage II provided hemodynamic stability7981
.
Under optimal conditions, hemodynamic measurements during the anhepatic stage are comparable to those recorded
during the dissection stage80, and the technique compares favorably with VBP. This generally results in a smooth
anhepatic stage where patients are not exposed to the potential complications associated with the use of bypass82.
Similar to VBP, stable central filling pressures with minimal vascular redistribution may underlie improved
hemodynamic response to hepatectomy. Studies comparing renal function, intraoperative blood use, operative time
and mortality have show similar outcomes between piggyback and VBP 83. Variability in hemodynamic stability
during stage II is largely related to placement of the tangential caval crossclamp which can impair venous return.
Surgical dissection of the native liver from the vena cava can be technically challenging in patients with extensive
portal hypertension or hepatobiliary scarring. In view of these problems, successful completion of the piggyback
technique has varied from all79 to seventy percent of recipients80,83. In difficult situations, loss of vascular control may
require precipitous application of a complete vena cava crossclamp. Depending on underlying patient physiology,
further hemodynamic deterioration may occur with few options for intervention. This has prompted some investigators
to use the piggyback technique only in patients that have demonstrated hemodynamic stability following a trial of
crossclamp78. Should hemodynamic instability occur during the trial period the piggyback technique with VBP support
has been used.
Within the last two decades there has been an impressive increase in the understanding of the cardiovascular
alterations of liver disease and the hemodynamic response to liver transplantation. This has led to the development of
intraoperative management strategies designed to modify patient response to surgical events. Introduction of each
method seems to have resulted in an initial clinical enthusiasm that is eventually tempered by recognition of the varied
response of patients to each technique. This pendulum of clinical opinion divided transplant centers into camps for and
against specific management styles. To date, there has been few randomized studies designed to compare different
management choices and evaluate their specific outcomes. This has made it difficult to asses the isolated effects of a
single management strategy. Despite these limitations, the combined understanding of individual patient problems and
response to surgical events allow clinicians to choose from well described options in intraoperative management based
upon a logical approach.
- 45 -
45
- 46 -
46
Fig 2
During venous bypass, blood from the femoral and portal veins are drawn
by a centripetal pump and delivered to the heart through the axillary or
subclavian vein. Venous return is maintained despite complete occlusion
of the inferior vena cava.
- 47 -
47
- 48 -
48
REFERENCES
1.
Grozmann RJ. Hyperdynamic circulation of liver disease 40 years later: Pathophysiology and clinical
consequences. Hepatology. 1994: 20:1359-1363.
2.
Rettke SR, Chantigian RC, Jonossoy TA, Chantigian RC, Burritt MF, Van Dyke RA, et al. Anesthesia
approach to hepatic transplantation. Mayo Clin Proc 1989; 64: 232-240.
3.
Carton EG, Plevak W, Kranner PW, Rettke SR, Geiger HJ, Coursin DB. Perioperative care of the liver
transplant patient: Part 2. Anesth Analges 1994; 78:382-399.
4.
Estrin JA, Belani KG, Ascher NL, Lura D, Payne W, Najarian JS. Hemodynamic changes on clamping and
unclamping of major vessels. Transplant Proc 1989; 21:3500-3505.
5.
Merritt WT, Beattie C, Peck R, Burdick JF, Klein AS, Dickstein R. Vena caval pressure gradients during
liver transplantation. Transplantation 1990; 50:336-338.
6.
Kang YG, Freeman S, Aggarwall S, De Wolf AM. Hemodynamic instability during liver transplantation.
1989; 21:3489-3492.
7.
Carmichael FJ, Lindop MJ, Farman JV. Anesthesia for hepatic transplantation: Cardiovascular and metabolic
alterations and their management. Anesth Analg 1985; 64:108-116.
8.
Stegall MD, Mandell S, Karrer F, Kam I. Liver transplantation without venovenous bypass. Transplant Proc
1995; 27:1254-1255.
9.
Plevak DJ. Hyperdynamic circulatory state after liver transplantation. Transplant Proc 1993; 25:1839.
10.
Zetlaoui P, Catoire P, Gillon MC, Descorps-Declere A. Changes in cardiac output during liver reperfusion in
hepatic transplantation [Abstract]. Anesthesiology 1989; 71:A182
11.
Hadengue A, Moreau R, Sogni P, Durand F, Bernuau J, Belghiti J, et al. High cardiac output after liver
transplantation (OLT) is due to persistent portosystemic collateral blood flow and elevated hepatic blood flow
[Abstract]. Hepatology 1991; 14:57A.
12.
Henderson MJ, McKay GJ, Hooks M, Chezmar JL, Galloway JR, Dodson TF, et al. High cardiac output of
advanced liver disease persists after orthotopic liver transplantation. Hepatology 1992; 15:258-262.
13.
Khoury GF, Kaufman RD, Musich JA, Mogard M. Neurotensin and vasoactive intestinal peptide levels
during orthotopic liver transplantation [Abstract]. Anesth Analg 1986; 65:S I
14.
Paulsen WA, Whitten CW, Ramsey MA, Klintmalm GB. Considerations for anesthetic management during
veno-venous bypass in adult hepatic transplantation. Anesth Analg 1989; 68:489-496.
15.
Aggarwall S, Kand Y, Freeman J, Fortunato FL, Pinsky MR. Postreperfusion syndrome: Cardiovascular
collapse following hepatic reperfusion during liver transplantation. Transplant Proc. 1987; 19 [Supplement
3]:54-55.
16.
Siefert RD, Kang YG, Begliomini B, Miller SR. Baseline cardiac output does not predict hemodynamic
instability during orthotopic liver transplantation. Transplant Proc 1989; 21:3523-3524.
17.
Gelman S, Khazaeli MB, Orr R, Henderson T. Blood volume redistribution during cross-clamping of the
descending aorta. Anesth Analg 1994;78:219-224.
- 49 -
49
18.
Picaache RS, Kapur BM, Starzel TE. The effect of liver disease on the need for venous decompression during
the anhepatic phase of canine orthotopic liver transplantation. Surgery 1970; 67: 319.
19.
Hawker F. Liver failure in the patient with chronic liver disease. In Critical Care Management: The Liver (ed
Park GR). WB Saunders, London, Philadelphia, Toronto, 1993;136-195
20.
Folkman J. Clinical applications of research on angiogenesis. New Engl J Med 1995; 333:1757-1763.
21.
Shiota G, Kawasake H, Kawamoto T, Nakamura T. Serum hepatocyte growth factor levels in liver diseases:
Clinical implications. Hepatology 1995;21:106-112.
22.
Grant DS, Kleiman HK, Goldberg I, Bhargava MM, Nickoloff BJ, Kinsella JL, et al. Scatter factor produces
blood vessel formation in vivo. Proc Natl Acad Sci 1993;90:1937-1941.
23.
Bussolino F, DiRenzo MF, Bocchietto E, Olivero M, Naldini L, Gaudino G, et al. Hepatocyte growth factor is
a potent angiogenic factos which stimulates endothelial cell motility and growth. J Cell Biol
1992;119:629-641.
24.
Hrusovsky S, Tsubouchi H, Remy-Martin A, Jouvenot M, Adessi GL, Rouget C, et al. Human hepatocyte
growth factor serum levels in early stages after liver transplantation. Transpl Proc 1994:26:3618-3619.
25.
Weigert AL, Martin PY, Niederberger M, Higa EM, McMurtry IF, Gines P, et al. Endothelium-dependent
vascular hyporesponsiveness without detection of nitric oxide synthetase induction in aortas of cirrhotic rats.
Hepatology 1995;22:1856-1862.
26.
Khourts A, Stahnke L McClain C. Logan G, Allen JL. Circulating tumor necrosis factor, interleukin-1 and
interleukin-6 concentration in chronic alcoholic patients. Hepatology 1991; 13:267-276.
27.
Mehta R Gottstein J, Zeller PW, Lichtenberg R, Blei AT. Endotoxin and the hyperdynamic circulation of
portal vein-ligated rats. Hepatology 1990; 12:1152-1156.
28.
Sieber CC, Grozmann RJ. Nitric oxide mediates in vitro hyporeactivity to vasopressors in mesenteric vessels
of portal hypertensive rats. Gastroenterology 1992; 103:235-239.
29.
Sieber CC, Lopez-Talavera JC, Grozmann RJ. Role of nitric oxide in the in vitro splanchnic vascular
hyporeactivity in ascitic cirrhotic rats. Gastroenterology 1993; 104:1750-1754.
30.
Niederberger M, Martin PY, Gines P, Morris K, Tsai P, Xu DL, et al. Normalization of nitric oxide
production corrects arterial vasodilation and hyperdynamic circulation in cirrhotic rats. Gastroenterology
1995;109:1624-1630.
31.
Battarbee HD, Zavecz JH, Roering SC, O'Donnell JM. Portal vein stenosis and cardiac impairment
[Abstract]. Gastroenterology 1993; 104:A875.
32.
Ingles AC, Hernandez I, Garcia-Estan J, Quesada T, Carbonell LF. Limited cardiac preload reserve in
conscious cirrhotic rats. Am J Physiol 1991; 260:H912-H917.
33.
Mitsuo T, Shimazaki S, Matsuda H. Right ventricular dysfunction in septic patients. Crit Care Med
1992;20:630-634.
34.
Jardin F, Valtier B, Beauchet A, Dubourg 0, Bourdarias JP. Invasive monitoring combined with
two-dimensional echocardiographic study in septic shock. Intensiv Care Med 1994.
35.
Redl G, Germann P, Hammerle A. Right ventricular function in early septic shock states. Intensiv Care Med
1993; I 9:3-7.
- 50 -
50
36.
Shaw B. Some further notes on venous bypass for orthotopic transplantation of the liver. Transplant Proc
1987; 19:13-16.
37.
Taura P, Beltran J, Garcia-Valdecasas, Zavala E, Lacy AM, Balust J, et al. The need for venovenous bypass
in orthotopic liver transplantation. Transplantation 1991; 52:730-733.
38.
Beltran J, Taura P, Grande L, Garcia- Valdecasas 1C, Rimola A, Cugat E. Venovenous bypass and liver
transplantation [letter]. Transplantation 1993; 77:642.
39.
Kelley SD. Venovenous bypass during liver transplantation. Anesth Analg 1992; 75:481-483.
40.
Paulsen WA, Valek TR, Blessing WS, Johnson DD, Parks RI, Pyron JT, et al. Hemodynamics during liver
transplantation with veno-venous bypass. Transplant Proc 1987; 19:2417-2419.
41.
Aggarwall S, Kang Y, Freeman J, DeWolf AM, Begliomini B. Is there a post-reperfusion syndrome.
Transplant Proc 1989 21:34973499.
42.
Gabrielson G, Nam SH, Reich DL. Metabolic and ECG changes associated with reperfusion during liver
transplantation [Abstract]. Anesthesiology 1991; 75:A989.
43.
De Wolf AM, Begliomini B, Gasior TA, Kang YG, Pinsky MR. Right ventricular function during orthotopic
liver transplantation. Anesth Analg 1993; 76:562-568.
44.
Ellis JE, Lichtor JL, Feinstein SB, Chung MR, Polk SL, Broelsch C, et al. Right heart dysfunction,
pulmonary embolism and paradoxical embolization during liver transplantation: A transesophageal two
dimensional echocardiographic study. Anesth Analg 1989; 68:777-782.
45.
Gabriel A, Muller C, Tuchy G, Hackl W, Zimpfer M. Reperfusion during orthotopic liver transplantation:
Analysis of right ventricular dynamics. Transplant Proc 1993; 25:1811-1812.
46.
Steltzer H, Blazek G, Gabriel A, Leitner K, Muller C, Tuchy GL. Zimpfer M. Two dimensional
transesophageal echocardiography in early diagnosis and treatment of hemodynamic disturbances during liver
transplantation. Transplant Proc 1991; 23: l 957-1958.
47.
a
Roth E, Steinninger R, Winkler S, Langle R, Grunberger T, Fugger R, Muhlbacher F. L-Arginine deficiency
after liver transplantation as an effect of arginase efflux from the graft. Transplantation 1994; 57:665-669.
48.
Calne RY, Smith DP, McMasterP,Craddock GN, Rolles K, Farman JV, et al. Use of partial cardiopulmonary
bypass during the anhepatic phase of orthotopic liver grafting. Lancet 1979; 2:612-614.
49.
Denmark SW, Shaw BW, Starzl TE, Griffith BP. Veno-venous bypass without systemic anticoagulation in
canine and human liver transplantation. Surg Forum 1983; 34:380-302.
50.
Griffiths BP, Shaw BW, Hardesty RI, Shunzaburo I, Bahnson HT, Starzl TE. Veno-venous bypass without
systemic anticoagulation for transplantation of the human liver. Surg Gynecol Obstet 1985; 160:271-272.
51.
Shaw BW, Martin M, Marquez JM, Kang YG, Bugbee AC, Shunzaburo I, et al. Venous bypass in clinical
liver transplantation Ann Surq. 1984; 200:524-534.
52.
Shaw BW, Martin DJ, Marquez JM, Kang YG, Bugbee AC, Shunzaburo I, et al. Advantages of venous
bypass during orthotopic transplantation of the liver. Sem Liver Dis 1985; 5:344-348.
53.
Ringe B, Bomscheuer A, Blumhardt G, Bechstein WO, Wongeit K, Pichlmayr R. Experience with
veno-venous bypass in human liver transplantation. Transplant Proc 1987; 19:2416.
- 51 -
51
54.
Avolio AW, Agnes S, Pelosi G, Sollazzi L, Magalini SC, Citterio F, et al. Determinants of hemodynamic
compensation during veno-venous bypass in liver transplantation. Transplant Proc. 1990; 22:I535-1536.
55.
Kang Y. Hemodynamic changes during intra-abdominal organ transplantation. Transplant Proc 1993;
25:2583-2587.
56.
Kirby RM, McMasterP, Clements D, Hubscher SG, Angrisani L, Sealey M, et al. Orthotopic liver
transplantation: Postoperative complications and their management. Br J Surg 1987; 74:3-11.
57.
McCauley J, Van Thiel DH, Starzl TE, Puschett JB. Acute and chronic renal failure in liver transplantation.
Nephron 1990; 55:121-128.
58.
Danovitch GM, Wilkinson AH, Colonna JO, Busuttil RW. Determinants of renal failure in patients receiving
orthotopic liver transplants. Kidney Int 1987; 31:195.
59.
Gunning TC, Brown MC, Swygert TH, Goldstein R, Husberg BS, Klintmalm GB, et al. Perioperative renal
function in patients undergoing orthotopic liver transplantation. Transplantation 1991; 51:422-427.
60.
Swygert TH, Roberts LC, Valek TR, Brajtbord D, Brown MR, Gunning TC, et al. Effect of intraoperative
low dose dopamine on renal function in liver transplant recipients. Anesthesiology 1991; 75:571-576.
61.
Wall WJ, Grant DR, Duff JH, Kutt JL, Ghent CN. Blood transfusion requirements and renal function in
patients undergoing liver transplantation without venous bypass. Transplant Proc 1987; 19:17-20.
62.
Zaballos JM, Riano D, Davila PA, Perez-Cereda F, Martinez PA. Venovenous bypass vs. no bypass in
orthotopic liver transplantation: Metabolic values during reperfusion. Transplant Proc 1993; 25:1865-1866.
63.
Navalgund AA, Kang Y, Sarner JB, Jahr JS, Gieraerts R. Massive pulmonary thromboembolism during liver
transplantation. Anesth Analg 1988; 67:400-402.
64.
Khoury GF, Mann ME, Porot MJ, Abdul-Rasool IH, Busuttil RW. Air embolism associated with
veno-venous bypass during orthotopic liver transplantation. Anesthesiology 1987; 67:848-851.
65.
Bohrer HR, Luz M. Bypass-associated air embolism during liver transplantation. Anaesth Intens Care 1990;
18:265-266.
66.
Khoury GF, Kaufman RD, Musich JA. Hypothermia related to the use of veno-venous bypass during liver
transplantation in man. Eur J Anaesth 1990; 7:501-503.
67.
Kang Y, Videira R, DeWolf AM, Casavilla A, Freeman JA, Aggarwal S, DeRiso B. The superior vena caval
syndrome during orthotopic liver transplantation: A complication of LeVeen shunt and venovenous bypass.
Transplant Proc 1991; 23; 1998-1999.
68.
Veroli P, El Hage C, Ecoffey C. Does adult liver transplantation without venovenous bypass result in renal
failure? Anesth Analg 1992; 75:489-494.
69.
Stock PG, Payne WD, Ascher NL, Roberts JP, Belani K, Estrin J, Najarian JS.Rapid infusion technique as a
safe alternative of venovenous bypass. Transplant Proc 1989; 21:2322-2325.
70.
Lawson NL. Clinical Use of Pressors and Inotropes. In Annual Refresher Course Lectures, American Society
of Anesthesiologists, 1991, San Francisco. 114:1-4.
71.
Jugan E, Albanladejo P, Jayais P, Ecoffey C. The failure of venovenous bypass to prevent graft liver
postreperfusion syndrome. Transplantation 1992; 54:81-84.
- 52 -
52
72.
Ramanathan S, Grant GJ. Vasopressor therapy for hypotension due to epidural anesthesia for caesarian
section. Acta Anesthesiol Scand 1988;32:1-7.
73.
Wall W1, Grant DR, Duff JH, Kutt JI, Ghent CN, Bloch MS. Liver transplantation without venous bypass.
Transplantation 1987; 43:56-61.
74.
Belghiti J, Panis Y, Sauvanet A, Gayet B, Fekete F. A new technique of side to side anastomosis during
orthotopic hepatic transplantation without inferior vena cava occlusion. Surgery 1992; 175:271-272.
75.
Stieber AC, Marsh JW, Starzl TE. Preservation of the retrohepatic vena cava during recipient hepatectomy
for orthotopic transplantation of the liver. Surg Gynec Obstet 1989; 168:543-544.
76.
Tzakis A, Todo S, Starzl TE. Orthotopic liver transplantation with preservation of the inferior vena cava. Ann
Surg 1989; 210:649-652.
77.
Lerut J, Gertsch P, Blumgart LH. "Piggyback" adult orthotopic liver transplantation. Helv chir Acta
1989;56:527-530.
78.
Jones R, Hardy KJ, Fletcher DR, Michell I, McNicol PL, Angus PW. Preservation of the inferior vena cava in
orthotopic liver transplantation with selective use of veno-venous bypass: The piggyback operation.
Transplant Proc 1992; 24:189-191.
79.
Fleitas MG, Casanova D, Martino E, Maestre JM, Herrera L, Hemanz F, et al. Could the piggyback operation
in liver transplantation be routinely used? Arch Surg 1994; 129:842-815.
80.
Figueras J, Sabate A, Fabergat J, Drudis TR, Rafecas A, Dalmau A, et al. Hemodynamics during the
anhepatic phase in orthotopic liver transplantation with vena cava preservation: A comparative study.
Transplant Proc 1993;25:2588-2589.
81.
Solares G, Maestre J, Pulgar S,Casanueva J, Casanova D, Martino E, Gomez-Fleitas M. Hemodynamic
changes during adult liver transplantation with partial vena cava clamping. Transplant Proc 1993; 25:1850.
82.
Sallizzoni M, Andomo E, Bossuto E, Cerutti E, Liviqni S, Lupo F, et al. Piggyback techniques versus
classical technique in orthotopic liver transplantation: A review of 75 cases. Transplant Proc 1994;
26:3552-3553.
83.
Mandell MS, Katz J, Kam I. Preservation of the inferior vena eava during liver transplantation. Joint
Congress on Liver Transplantation; London UK, 1995:September 27-30.
- 53 -
53
PULMONARY HYPERTENSION IN LIVER DISEASE
Within the last ten years physicians have begun to focus their attention on the cardiopulmonary system of patients
with liver disease. This fascination stems from the influence that pulmonary vascular pathology can have on the
outcome of liver transplantation. Morbidity and mortality have been reported in multiple transplant candidates
and recipients due to pulmonary vasculopathy caused by liver disease. To appreciate how liver disease can
disrupt the pulmonary circulation, we will first review key physiological concepts that govern the behavior of the
pulmonary circulation and then examine the effects of liver dysfunction on pulmonary vascular function. The rest
of this review will critically appraise current knowledge and its application to the clinical management of patients
with portopulmonary hypertension.
Pulmonary Vascular Physiology
Early physiological studies portrayed the pulmonary circulation as a passive and low pressure circuit that behaved
as a conduit between the right and left ventricle. However, it is now clear that the pulmonary vasculature is a
dynamic and complex organ system that influences physiological homeostasis. Even though regulatory function
of the pulmonary vasculature is complex, there are certain physical laws that govern blood flow through all
vessels. In fact all vascular flow is entirely determined by two factors; 1) the pressure difference between two
ends of the vessel and 2) the impediment to flow which is called the vascular resistance. (1,2) This is shown in
the equation Q = P / R where Q is flow, P is the pressure differential and R is resistance. Rearranged this
equation is expressed as R = P / Q or vascular resistance equals the pressure differential divided by the cardiac
output. Thus resistance is directly proportional to the pressure gradient and inverse of flow.
There is a clear downstream pressure gradient between the arterial and venous vessels in the systemic circulation
that helps propel blood forward. Because resistance of the pulmonary arterial and venous circulation are
approximately the same, the driving force to forward flow primarily depends on the pressure differential between
the right ventricle and left atrium.(1,3) Although the pulmonary vessels relax in response to high flow, increased
blood volumes are also conducted to the left atrium through the recruitment of blood vessels in the physiological
dead space.(4) This unique feature allows the normal pulmonary circulation to conduct a greater blood volume
during physiological states of high flow such as exercise. Laminar flow in the pulmonary arterioles also aids rapid
conduction of an increased volume associated with higher cardiac outputs.(2) Because vascular relaxation,
recruitment and laminar flow act to decrease resistance, the pulmonary artery pressures usually do not increase in
high flow states.(1)
Almost all patients with liver disease develop systemic circulatory changes that result in a hyperkinetic blood flow
in most vascular beds.(5) Consequently, the increase in cardiac output associated with this hyperdynamic
circulation must be accommodated by the pulmonary vasculature. A low pulmonary vascular resistance (PVR)
observed in most patients with liver disease at rest has been interpreted as an adaptation of the pulmonary
circulation to the increased systemic blood flow.(6) However, the pulmonary vascular response to exercise is
distinctly abnormal in patients with liver disease that do not have any recognized pulmonary pathology and
suggests that either the normal compensatory mechanisms have been exhausted or pulmonary vascular mechanics
have been altered.(7) During exercise patients with a hyperkinetic circulation due to liver disease exhibit an
increase in pulmonary capillary wedge pressures (PCWP) that parallels the increase in cardiac output. Thus the
relationship between the measured pressures and flow become almost linear compared to the exponential
relationship that occurs in normal subjects. The linear relationship implies that flow becomes dependent on the
pressure differential and that resistance is no longer a modifying factor. In fact, a normal pressure differential
measured from the right ventricle to the left atrium during exercise in patients with liver disease confirms that the
elevations observed in the PCWP are flow and therefore volume related. That PVR fails to fall in patients with a
hyperdynamic circulation in response to exercise suggests a change in pulmonary vascular regulation or that the
pulmonary vessels are already maximally dilated and recruited at rest.(8)
Classification of Pulmonary Hypertension
Pulmonary hypertension is a not a disease but a pathological condition where the mean pulmonary artery pressure
(mPAP) is greater than 25 mm Hg at rest. Pathological lesions that cause pulmonary hypertension can arise in the
- 54 -
54
pulmonary vessels before or after the pulmonary capillary bed. Postcapillary defects have been identified in the
pulmonary veins (venocclusive disease), left atrium (myxoma), mitral valve (stenosis), or left ventricle
(myocardial failure). These lesions share the common effect of increasing pulmonary venous pressure and
therefore PCWP. Because there is a minimal difference between the pulmonary arterial and venous resistances,
the increased pulmonary venous pressure is readily transmitted to the capillary vascular bed and ultimately to the
right ventricle causing symptoms of congestive heart failure.
An abnormal pressure gradient can also develop between
Pulmonary hypertension
the pulmonary capillary bed and right ventricle due to
(mPAP > 25 mm Hg at rest)
pathological lesions that arise in the pulmonary arterioles.
/
\
Pulmonary hypertension caused by precapillary lesions are
postcapillary precapillary
classified as Primary by the National Institute of Health
(venous PH) (arteriolar PH)
Registry if pulmonary arterial hypertension is not associated
/ \
Primary PH Secondary PH
with any identifiable secondary causes.(9) A recent
|
consensus statement from the American College of Chest
Portopulmonary hypertension
Physicians included congenital right to left shunts, chronic
hypoxemia, chronic thromboembolism, the vasculitities and
Fig 1. Classification of pulmonary hypertension
portal hypertension as causes of secondary pulmonary
arterial hypertension.(10) At this time the classification of
portal hypertension as a secondary cause of pulmonary
arterial hypertension seems arbitrary since the etiology of
both portal and primary pulmonary hypertension are
unknown.(Fig 1) However, there is no question that the complex circulatory changes in patients with liver disease
makes the simple comparison of primary and portal hypertension impossible.
Criteria for the Diagnosis of Portal Pulmonary Hypertension
Because well established epidemiological and hemodynamic criteria are used to identify patients with PPH, it
seemed reasonable to construct similar criteria to identify those patients with PPHTN.(10) Most investigators
would agree that portosystemic shunting is necessary to make a diagnosis of PPHTN.(11,12) Once portal
hypertension is present it is obvious that shunting is present. The clinical symptoms caused by portal
hypertension are therefore acceptable criteria as the direct measurement of portal pressure is not always practical.
However shunting can occur without an increase in portal pressures (13) and may explain the presence of
pulmonary hypertension in patients with hepatobiliary disease who do not exhibit obvious signs of portal
hypertension. (14-16) Since there is no practical method to detect portosystemic shunting without a
corresponding elevation in pressure, the diagnosis of PPHTN in some cases may remain questionable.
The selection of hemodynamic criteria used to diagnose PPHTN presents a more complicated picture due to the
effects of the hyperdynamic circulation on the pulmonary vasculature. A mPAP greater than 25 mm Hg at rest
with a PCWP less than 15 mm Hg has become a standard definition to identify pulmonary arteriopathy of various
etiologies.(9) Errors can occur if we indiscriminately apply this definition to PPHTN patients. For example,
pulmonary artery hypertension can be incorrectly diagnosed when high capillary pressure caused by the
hyperdynamic flow is transmitted backwards to the pulmonary artery. In this case the elevated PCWP is helpful
to exclude an incorrect diagnosis. However, a diagnosis of pulmonary artery hypertension may be prematurely
dismissed as volume overload when the PCWP is elevated by excessive flow in patients who have exceeded the
compensatory pulmonary vascular mechanisms for accommodating increased volume. In this case an elevated
PCWP could preclude the correct diagnosis. Or, the diagnosis of PPHTN can also be missed when aggressive
diuresis or -blockade decrease flow and therefore pulmonary artery pressures. Similar problems arise when PVR
is used as a criterion for diagnosis. Pulmonary artery resistance cannot be simply measured. A quick review of
the equation R = P / Q shows that as flow increases, the measured resistance decreases. Therefore high
pulmonary artery pressures may be accompanied by a low PVR. This problem is realized by the difficult
categorization of patients who have a mPAP greater than 25 mm Hg, normal PCWP but a PVR less than 120
dynes.s.cm-5 . With these limitation in mind the following guidelines have been published to help identify
patients with PPHTN.(16) It is likely that these criteria will change in the future as PPHTN is better understood.
- 55 -
55
1.
2.
3.
4.
Presence of portosystemic shunting
Mean pulmonary artery pressures greater than 25 mm Hg
Pulmonary vascular resistance greater than 120 dynes.s.cm-5
Pulmonary capillary wedge pressure less than 15 mm Hg
Mechanism of Disease
Although more than 20 percent of patients with portal hypertension have a mPAP greater than 25 mm Hg only 2 to
4 percent of patients meet the current guidelines for the diagnosis of PPHTN. (18). Investigators hypothesize that
arteriopathic changes leading to pulmonary hypertension are initiated by factors derived from the splanchnic blood
that would normally be cleared by hepatic metabolism but reach the pulmonary circulation due to portosystemic
shunting and impaired hepatic reticuloendothelial function.(11) Because not all patients with portal hypertension
develop PPHTN, it is possible that compounds derived from the splanchnic circulation unmask or potentiate a
preexisting genetic susceptibility to pulmonary hypertension.
Histological studies have shown that endothelial cell injury is a common initiating event in patients with primary
and secondary pulmonary hypertension. (19,20) The injury induces changes in cellular behavior that promotes cell
growth vasoconstriction and increased endothelial reactivity to coagulation proteins.(20,21) Mitogenic activity is
directed at the formation of new blood vessels and microvascular thrombosis promotes the remodeling of older
vessels (22) These changes creates a vascular network with increased resistance to blood flow. Three subtypes of
pulmonary hypertension are recognized based upon the histological appearance of pulmonary vascular lesions.
These subtypes include: isolated medial hypertrophy, plexogenic pulmonary arteriopathy and thrombotic
arteriopathy.(23) Cellular growth causes medial hypertrophy and plexogenic lesions, but thombotic arteriopathy
results from microvascular clot formation.(23) Vasoconstriction probably enhances both types of lesion. Patients
usually exhibit a mosaic of lesions, indicating that all three pathological mechanisms are probably active in most
patients. (23) Although microvascular thrombosis is observed to some degree in 50% of PPH patients, it is the
predominant lesion in 58% of patients with PPHTN.( 23,24) Excessive clot formation can originate from defects
in three major pathways: 1) an excess of thrombin; 2) a deficiency of fibrinolytic activity; or 3) intrinsic platelet
abnormalities that favor clot formation. The first and second pathways are likely mechanisms leading to
microvascular thrombosis in PPHTN (Fig 2) because in chronic liver disease 1) there is often a low grade
consumptive coagulopathy which leads to increased amounts of prothrombin reaching the pulmonary circulation
through portosystemic shunts (25) 2) there are elevated systemic concentrations of
Fig. 2 Coagulation defects
cytokines including endotoxin, tumor necrosis factor
leading to Pulmonary Hypertension
and interleukin-1 that increase thrombin activity and down
regulate inhibitors of thrombin and fibrinolysis (26, 27) 3)
Antithrombin-III, which inactivates circulating thrombin is
often low (25) and 4) the hyperdynamic circulatory state
induces high shear stress in the pulmonary circulation that can
aggravate endothelial injury and increase the synthesis of
factors that activates thrombin (28) and recruit platelets .(29)
Because these findings suggest a large role of microvascular
thrombosis in PPHTN compared to other types of pulmonary
hypertension, we have to determine if these lesions result from
specific pathological mechanisms that favor clot formation or
are associated with more advanced disease due to a delayed diagnosis.
- 56 -
56
Diagnosis
An algorithm formulated by the National Institutes of Health to diagnose and evaluate patients with PPH has been
applied to patient with pulmonary
hypertension of other etiologies.(Fig 3)
Suspicion
|
When the history and physical
Echo
examination suggest pulmonary
/
\
hypertension, an echocardiogram is
No PH
PH
performed to exclude common
/
\
Shunts, valves
No secondary causes
secondary causes and to assess right
Cardiomyopathy
|
ventricular function. Ventilation and
V/Q
perfusion scanning is used to exclude
/
\
thromboembolic disease while
Normal/Low prob
>1 segmental defect
|
|
conditions that cause chronic hypoxic
PFT, ABG
Angiography
lung disease are identified by
Autoimmune serology
pulmonary function tests. Patients
\
with suspected PPH undergo cardiac
No significant disease
|
catheterization to measure intracardiac
Cardiac catheterization
pressures and evaluate the efficacy of
pulmonary vasodilators.(10)
Figure 3 Algorithm for diagnosis of PPH taken from Rubin1993.
V/Q = ventilation/perfusion scan: PFT = pulmonary function tests:
ABG = arterial blood gas
There is no question that it is difficult
to detect pulmonary hypertension in
patients who have liver disease and it is
only when these patients develop
decompensated right heart failure does
the diagnosis become evident. Because subtle findings in the history and physical examination are easily
confused with signs and symptoms of liver disease, the physician must have a high degree of suspicion. Patients
with pulmonary hypertension complain of dyspnea on exertion, fatigue and peripheral edema.(9) These are also
common complaints of patients who have liver disease but no coexistent pulmonary problems. A prospective
study found that 60% of patients who had PPHTN were asymptomatic and forty percent who had dyspnea on
exertion were overlooked during the clinical examination.(30) In contrast, reports of chest pain and syncope are
late and ominous symptoms that usually indicate severe cardiac failure. Physical signs suggestive of pulmonary
hypertension such as increased splitting of the second heart sound or right-sided third and fourth heart sounds can
also be caused by hyperdynamic flow. It can be difficult to distinguish the diagnostic murmur of tricuspid
regurgitation from commonly found high flow murmurs. The pulmonary hyperemia usually seen in patients with
liver disease masks the classical diagnostic findings of pulmonary hypertension on the chest radiograph.(31)
Pulmonary function tests, including arterial blood gases are not useful for the detection or characterization of
PPHTN, but do help exclude other possible causes of pulmonary hypertension.(32) Although nonspecific defects
in both the ventilation and perfusion phases can make interpretation of lung scans difficult in patients with liver
disease, the presence of one or more segmental or larger defects suggest thromboembolic disease.(33) Fortunately,
signs observed on the standard 12-lead electrocardiogram have helped identify up to 80% of patients diagnosed
with PPHTN.(34) Conduction defects including right bundle branch block and premature atrial beats usually
precede the development of advanced findings associated with right ventricular pressure overload such as right
axis deviation and right ventricular hypertrophy with or without strain.
The echocardiogram is pivotal in the evaluation of patients with suspected pulmonary hypertension. A single
examination can identify common causes of postcapillary pulmonary hypertension, assess cardiac function and
estimate the pulmonary artery pressure. The right heart is not well visualized by transthoracic echocardiography.
It is therefore important to direct the examiner’s attention to findings that could suggest pulmonary
hypertension.(35) Premature closure of the pulmonic valve accompanied by regurgitant flow is an early finding of
pressure overload in the pulmonary circulation. Right ventricular hypertrophy, tricuspid regurgitation or
interventricular wall thickening are adaptations to the chronic pressure overload found in more advanced disease.
Right atrial enlargement or bulging of the interventricular septum into the left ventricle are signs of cardiac
decompensation. Echocardiographic changes that suggest pulmonary hypertension however must be interpreted
- 57 -
57
with care because changes in intravascular volume, ascites and pleural and pericardial effusions can also cause
pulmonary and tricuspid regurgitation. Impaired right ventricular contractility caused by sepsis or systemic
inflammatory states can also lead to chamber dilation and tricuspid regurgitation. The resulting elevation in
pulmonary artery pressure is due to causes other than precapillary pulmonary arteriopathy. Direct measurement of
pulmonary artery pressures and PCWP from the cardiac catheterization should help to distinguish between pre and
postcapillary pulmonary hypertension.
Treatment
Similar to PPH, there is no definitive treatment for patients with PPHTN. The care of PPHTN patients is
further complicated by the presence of progressive liver disease that may by itself cause death of the patient.
Because of the association between portosystemic shunting and pulmonary hypertension, it seemed likely that
liver transplantation would act as a single therapy for the treatment of both pulmonary and hepatic disease.
However, multiple reports of perioperative mortalities forced clinicians to reevaluate the benefits of
transplantation in PPHTN patients and examine the role of vasodilators in controlling pulmonary artery
pressures.(36, 37)
Vasodilators have reduced pulmonary artery pressures and prolonged survival in some PPH patients.
However, PPH patients are only placed on long-term therapy if they respond favorably to an acute vasodilator
challenge during cardiac catheterization.(38) A favorable response has been defined as 1) an increase in
cardiac output as a single finding 2) a fall in PVR of 20% or greater 3) a fall in mPAP of 20% or greater and
4) a greater fall in PVR compared to systemic vascular resistance.(39) Unlike those with PPH, few PPHTN
patients who were subjected to similar testing demonstrated a favorable response. The inability to modify
vasoreactivity in the pulmonary vasculature led investigators to consider most PPHTN patients as poor
candidates for transplantation. However, recent studies have shown that chronic epoprostenol infusions can
reduce pulmonary artery pressures in PPHTN patients who do not acutely respond to vasodilators.(40)
Epoprostenol given in doses ranging from 10 to 28ng/kg/min caused a progressive fall in pulmonary artery
pressures over a 6 to 14 month period that was sustained following transplantation. Antimitogenic and
antithrombotic properties of epoprostenol leading to favorable vascular remodeling may explain the slow but
progressive fall in pulmonary circulatory resistance.(20) This observation has opened a new window of
opportunity for PPHTN patients who were previously not considered operative candidates. The effects of
nitric oxide on the pulmonary vasculature in PPHTN patients have not been as clear. Nitric oxide inhaled in
doses up to 80 ppm has acutely reduced pulmonary artery pressures is a small number of PPHTN
patients.(41,42) The reason that nitric oxide is not effective in all PPHTN patients is unknown but those who
respond may have a greater component of pulmonary vasoconstriction.(43) Whether pulmonary
vasoconstriction represents an earlier stage of disease or is an expression of disease heterogeneity is
unknown. It would be interesting to know if inhaled nitric oxide used over several months would have
favorable effects similar to those seen with epoprostenol. Positive effects of prolonged use of nitric oxide
would suggest that vascular remodeling and not vasoconstriction is a prominent mechanism causing PPHTN.
Investigators have evaluated other vasodilators but these do not appear to be as effective as epoprostenol and
nitric oxide. Although combined transplantation of heart, lung and liver has been performed in patients with
coexistent thoracic disease due to sarcoidosis and cystic fibrosis,(44) the current experience with PPHTN
patients is too small to draw any conclusions.(45)
The Effects of Liver Transplantation on the Pulmonary Circulation
In contrast to most transplant recipients, PPHTN patients have an increased risk of right ventricular
decompensation and death during transplantation.(46) The pulmonary vessels of PPHTN patients are unable
to accommodate or conduct the increased blood flow that commonly occurs during transplantation due to
rapid transfusion or exacerbation of the hyperdynamic state. The resulting acute dilation of the right ventricle
compresses the right coronary vessels and left ventricle, leading to myocardial ischemia with right ventricular
failure and a fall in left ventricular stroke volume due to partial chamber obliteration. Electrolyte and
acid/base imbalances that occur during surgery can increase PVR and reduce the threshold for the
development of a lethal arrhythmia.(47) Vasoactive substances released from the new liver during portal
reperfusion may further raise PVR, impair myocardial contraction and lower systemic vascular resistance,
- 58 -
58
causing severe hemodynamic instability. Any additional patient or surgical factors that would lead to
worsening of these events will increase the burden of risk that already exists.
Outcomes
A reduction in portal pressures and improved hepatocellular function following liver transplantation should
prevent substances that cause endothelial injury from reaching the pulmonary vasculature. There are reports
of improvement, persistence and de novo development of pulmonary hypertension following liver
transplantation.(48-50) Investigators have speculated whether persistent and progressive pulmonary
hypertension following transplantation is the result of continuous pulmonary vascular injury or if pathological
changes in the pulmonary endothelium become self-generating. The presence of low pressure portosystemic
collateral flow following transplantation would argue for continuous endothelial stimulation as a causative
factor. However, the recent success using epoprostenol suggests that remodeling of pulmonary vessels may
initiate a chain of self-perpetuating events that require metabolic correction.
Selection Criteria
Criteria based upon mPAP, right ventricular function and the response to pulmonary vasodilators have been
important in the selection of PPHTN patients for transplantation. There is a general consensus that there is a
marked risk of intra and postoperative mortality when the mPAP is greater than 45 mm Hg and the PVR exceeds
120 dynes.s.cm-5. (18, 37, 46) In contrast a mPAP greater than 25 mm Hg with a PVR less than 120 dynes.s.cm-5
did not affect intra or postoperative outcome.(18) That elevated pulmonary artery pressures result from a high
cardiac output and not arteriopathy may explain the better outcome in PPHTN patients with a normal PVR. The
exact perioperative risk associated with mPAPs between 25 and 45 mm Hg when the PVR is elevated is unknown.
The current small collection of case reports is insufficient information to reach a conclusion on this issue.
Many clinician feel that patients with right ventricular dysfunction are not operative candidates.(50) However,
impaired right ventricular function in pulmonary hypertensive patients is caused by pressure overload and is not a
primary pathological process. The right ventricle of PPH patients who undergo lung transplantation recovers
rapidly. Thus, right ventricular dysfunction in PPHTN patients is a probably a measure of the severity of pressure
overload. A reduction in pulmonary artery pressures following long-term therapy with epoprostenol could lead to
a significant improvement in right ventricular function. Patients should therefore, not be immediately dismissed
as transplant candidates if right ventricular function is abnormal during the initial evaluation.
A favorable response to vasodilators is also used as a criterion for transplantation by some investigators. (51)
Because failure to control pulmonary artery pressures has led to perioperative death, physicians feel that they must
be able to treat precipitous increases in pulmonary artery pressures that result from intraoperative events including
hepatic reperfusion. Although this seems logical, no study has shown that patients who respond to an acute
challenge with vasodilators have an improved outcome. It is also important to consider that patients who do not
respond to an acute challenge with vasodilators during an initial evaluation may exhibit a positive response
following long-term treatment with epoprostenol. This is suggested by the recent report of a patient who briskly
and reversibly responded to nitric oxide while receiving chronic epoprostenol.(52) The latter report supports
current hypotheses that constant remodeling of the pulmonary vessels occurs in health and disease and it is
unlikely that pulmonary hypertension is ever “fixed”.
Questions and Future Directions
Although a better understanding of hyperdynamic flow has helped explain some of the pulmonary circulatory
responses in patients with liver disease, the complexity of these changes have hindered the identification and
characterization of PPHTN patients. A simple and inclusive definition of PPHTN is still lacking while the
classification of patients with elevated pulmonary artery pressures but low PVR remains uncertain. It is thus
difficult to identify disease characteristics that are associated with a good outcome and decide which patients will
benefit from transplantation and at what risk. If we are unable to reliably identify characteristics associated with
increased risk, is it reasonable to use living donors for these potential recipients? Should patients with moderate
- 59 -
59
pulmonary hypertension but mild liver disease receive a liver transplant to treat the pulmonary arteriopathy? Or
should the UNOS status of PPHTN patients be upgraded because the natural history is unknown? The systematic
collection of information through the Multicenter Database for Hepatopulmonary Syndrome and Pulmonary
Hypertension should be able to provide answers to most of these questions. Only once a distinct study cohort is
identified will clinical trials become meaningful and the mechanism of disease apparent.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Mecham R. Conference summary: biology and pathobiology of the lung circulation. Chest 1998, 114:106s111s.
Comroe J. The main functions of the pulmonary circulation. Circulation 1996, 33:146-58
Fahri L. Pulmonary circulation and systemic circulation: similar problems, different
solutions. Adv Exp Med Biol. 1990, 277:579-86.
Johnson R, Hsia C. Functional recruitment of pulmonary capillaries. J Appl Physiol 1994, 76:1405-7.
Groszmann R. Hyperdynamic circulation of liver disease 40 years later: pathophysiology and clinical
consequences. Hepatology 1994, 20:1359-63.
Daoud F, Reeves J, Schaefer J. Failure of hypoxic pulmonary vasoconstriction in patients with liver
cirrhosis. J Clin Invest 1972, 143:1076-80.
Agusti A, Roca J, Rodriguez-Roisin R, Wagner P, Bosch J. Pulmonary hemodynamics and gas exchange
during exercise in liver cirrhosis. Am Rev Resp Dis 1989,139:485-91.
Naeije R, Hallermans R, Mols P, Melot C. Hypoxic pulmonary vasoconstriction in liver cirrhosis. Chest
1981, 80:570-4.
Rich S, Dantzker D, Ayres S, Bergofsky E, Brundage B, et al., Primary pulmonary hypertension. A national
prospective study. Ann Intern Med 1987, 107:216-23.
Rubin L. Primary pulmonary hypertension. ACCP consensus statement. Chest 1993,104:236-50.
Lockhart A. Pulmonary arterial hypertension in portal hypertension. Clin Gastroenterol 1985, 14:123-37.
Lebrec D, Capron J, Dhumeaux, Benhamou J. Pulmonary hypertension complicating portal hypertension.
Am Rev Respir Dis 1979, 120:849-56.
Chezmar J, Redvanly R, Nelson R, Henderson M. Persistence of portosystemic collaterals and splenomegaly
on CT after orthotopic liver transplantation. Am J Radiol 1992, 159:317-20.
Senior R, Britton R, Turino G, Wood J, Langer G, Fishman A. Pulmonary hypertension associated with
cirrhosis of the liver and with portocaval shunts. Circulation 1968, 28:88-96.
Shah H, Piris J, Finlayson N. Primary pulmonary hypertension developing 11 years after a splenorenal shunt
for portal hypertension in hepatic cirrhosis. Eur J Gastroenter Hepatol 1995, 7:283-6.
Mandell M, Groves B, Duke J. Progressive pulmonary hypertension following liver transplantation.
Transplantation 1994, 59:1488-90.
Mandell M, Krowka M. Formation of a national database on pulmonary hypertension and hepatobiliary
disease in chronic liver disease. Anesthesiology 1997, 87:450-1.
Castro M, Krowka M, Schroeder R, Beck K, Plevak D, et al., Frequency and clinical implication of increased
pulmonary artery pressures in liver transplant patients. Mayo Clin Pro 1996, 71:543-51.
Tuder R, Groves B, Badesch D, Voelkel N. Exuberant endothelial cell growth and elements of inflammation
are present in plexiform lesions of pulmonary hypertension. Am J Pathol 1994, 144:275-85.
Higgenbottom T. Pathophysiology of pulmonary hypertension. A role for endothelial dysfunction. Chest
1994, 105:7s-12s.
Vane J, Anggard E, Botting R. Regulatory functions of the vascular endothelium. N Engl J Med 1990,
323:27-36.
Chaouat A. Weitzenblum E, Higgenbottom T. The role of thrombosis in severe pulmonary hypertension.
Eur Respir J 1996, 9:356-63.
Pietra G. The histopathology of primary pulmonary hypertension. Chest 1994, 105:2s-6s.
Edwards B, Weir K, Edwards W. Coexistent pulmonary and portal hypertension. Morphological and clinical
features. J Am Coll Cardiol 1987, 10:1233-38.
Violi F, Ferrro D, Basili S, Quintarelli C, Musca A, et al., Hyperfibrinolysis resulting from clotting
activation in patients with different degrees of cirrhosis. Hepatology 1993, 17:78-83.
- 60 -
60
25 Terry C, Callahan K. Protein C-kinase regulates cytokine-induced tissue factor transcription and
procoagulant activity in human endothelial cells. J Lab Clin Med 1996, 127:81-93.
26 Conway E, Rosenberg R. Tumor necrosis factor suppresses the transcription of the thrombomodulin gene in
endothelial cells. Mol Cell Biol 1988, 8:5588-92.
27 Diquelou, Lemozy S, Dupouy D, Boneu B, Sakariassen K, et al ., Effects of blood flow on thrombin
generation is dependent on the nature of the thromobgenic surface. Blood 1994, 84:2206-13.
28 Harker L, Hanson S. Platelet factors predisposing to arterial thrombosis. Balliere’s Clin Haem 1994, 7:499522.
29 Hadengue A, Benhayoun K, Lebrec D, Benhamou J. Pulmonary hypertension complicating portal
hypertension: Prevalence and relation to splanchnic hemodynamics. Gastroenterology 1991, 1000:520-8.
30 Chan T, Palevsky H, Miller W, Pulmonary hypertension complicating portal hypertension. Findings on chest
radiographs. Am J Radiol 1988, 151:909-14.
31 Robalino B, Moodie D. Association between primary pulmonary hypertension and portal hypertension:
Analysis of its pathophysiology and clinical, laboratory and hemodynamic manifestations. Am J Cardiol
1991, 17:492-8.
32 Moser K, Ashburn G, Fedullo P. Perfusion lung scans provide a guide to which patients with apparent
primary pulmonary hypertension merit angiography. West J Med 1988, 148:167-70.
33 Kanemoto N. Electrocardiogram in primary pulmonary hypertension. Eur J Cardiol 1980, 12:181-93.
34 Reeves J, Turkevich D, Morrison D. Noninvasive detection of pulmonary hypertension. Semin Respir Med
1985, 7:147-159.
35 Dewolf A, Scott V, Gasior T, Kang Y. Pulmonary hypertension and liver disease. Anesthesiology 1993,
78:213.
36 Ramsay M, Simpson B, Nguyen A, Ramsay K, East C, et al., Severe pulmonary hypertension in liver
transplant candidates. Liver Transpl Surg 1997, 3:494-500.
37 Rich S, Martinez J, Lam W, Levy P, Rosen K. Reassessment of the effects of vasodilator drugs in primary
pulmonary hypertension: Guidelines for determining a pulmonary vasodilator response. Am Heart J 1983,
105:119-127.
38 Weir E, Rubin L, Ayres S, Bergofsky E, Brundage B, et al., The acute administration of vasodilators in
primary pulmonary hypertension. Experience from the National Institutes of Health Registry on primary
pulmonary hypertension. Am Rev Respir Dis 1989, 140:1623-30.
39 Kuo P, Johnson L, Plotkin J, Howell C, Bartlett S, et al., Continuous intravenous infusion of epoprostenol for
the treatment of portopulmonary hypertension. Transplantation 1997, 63:604-16.
40 Mandell M, Duke J. Nitric oxide reduces pulmonary hypertension during hepatic transplantation. 1994,
81:1538-41.
41 Tarquino M, Geggel R, Stauss R, Rhodes J, Wunderlich B, et al., Treatment of pulmonary hypertension with
inhaled nitric oxide during hepatic transplantation in an adolescent: Reversibility of pulmonary hypertension
after transplantation. Clin Pediat 1998,37:505-10.
42 Ramsay M, Schmidt A, Hein T, Nguyen A, Lynch K, et al., Nitric oxide does not reverse pulmonary
hypertension associated with end-stage liver disease: A preliminary report. Hepatology 1997, 25:524-27.
43 Couetil J, Houssin D, Soubrane O, Chevalier P, Dousset B. et al., Combined lung and liver transplantation in
patients with cystic fibrosis. J Thorac Cardiovasc Surg 1995,110:1422-3.
44 Wall W, Grant D, Scudmore C, Shackleton C, Fradet G. Single-lung versus liver transplantation for the
treatment of portopulmonary hypertension-A comparison of two patients. Transplantation 1992, 55:686-90.
45 Mandell M, Forum on Critical Care Issues in Liver Transplantation: Cardiopulmonary concern: Pulmonary
hypertension. Liver Transpl Surg 1996, 2:321-26.
46 Mandell M, Katz J, Wachs M, Gill E, Kam I. Circulatory pathophysiology and options in hemodynamic
management during adult liver transplantation. Liver Transpl Surg 1997, 3:379-87.
47 Cheng E, Woehlick H. Pulmonary artery hypertension complicating anesthesia for liver transplantation.
Anesthesiology 1992, 77:389-92.
48 Koneru B, Ahmed S, Weisse A, Grant G, McKim K. Resolution of pulmonary hypertension after liver
transplantation. Transplantation 1994, 58:1133-35.
49 DeWolf A, Gasior T, Kang Y. Pulmonary hypertension in a patient undergoing liver transplantation. Transpl
Proc 1991, 23:1000-1.
50 Taura P, Garcia-Valdecasas J, Beltran J, Izquierdo E, Navasa M, et al., Moderate primary pulmonary
hypertension in patients undergoing liver transplantation. Anesth Analg 1996, 83:675-80.
- 61 -
61
51 Ramsay M, Spikes C, East C, Lynch K, Hein T, et al., The perioperative management of portopulmonary
hypertension with nitric oxide and epoprostenol. Anesthesiology 1999, 90:299-301.
OR SET-UP
Liver transplants are usually performed in Room 2. This room must be set up prior to each case. The following list
consists of materials required for all transplants. Other equipment and drugs may be needed in more complicated
cases.
I.
Drugs
A.
Anesthetic
1.
2.
3.
4.
5.
Propofol
Succinylcholine
Lorazepam
Fentanyl
CIS-Atricurium
B.
Resuscitation
1.
C.
Dopamine
Vasopressin
Mannitol
Magnesium
10µg/ml
100µg/ml
1g
1 amp
5mg/ml
100µg/ml
0.4mg/ml
150mg
50 ug/ml
1600µg/ml
0.04U/min
20%
5g/250 cc
500cc
Albumin
5%
Transplant Box Contents
1.
2.
3.
4.
5.
6.
7.
8.
II.
Epinephrine
Drugs to have in Room
1.
E.
20mg vials x 2
2.
Calcium Chloride
3.
Sodium bicarbonate
4.
Ephedrine
5.
Phenylephrine
6.
Atropine
7.
Amioderone
8.
Nitroglycerine
Infusion
1.
2.
3.
4.
D.
20cc (10mg/ml)
10cc (20mg/ml)
1-2mg
Mannitol
Furosemide
Solumedrol
Aminocaproic acid
Magnesium
0.9% NaCl
Magnesium Sulfate
Propranolol
20% (500cc x 2)
10mg/ml (l0cc x 3)
1g
250mg/ml (20cc x 2)
l mg ampule x 2
100cc bag x 2
1 g ampule x 2
I mg ampule x 5
Vascular Supplies and Set-up
- 62 -
62
A.
Arterial
1.
2.
Hand board and Arrow introducers x 2
Cook catheter 4.0 F x 12cm
- 63 -
63
B.
Venous
1.
2.
C.
Pressure Transducers
1.
2.
D.
III.
Triple or double transducer kit on L for Art/PA/CVP*
Single transducer on L for Art line if necessary
Intravenous Solutions
1.
1000cc Plasmacyte on R pole to Level 1 warmer
2.
1000cc Plasmacyte for peripheral IV to be started in PACU
3.
RIS: prime solution* or albumin/hextend
Monitors
A.
B.
C.
D.
E.
F.
G.
H.
I.
J.
IV.
Arrow Percutaneous Sheath Introducer (9.0 F) x 2
Arrow Rapid Infusion Catheter (8.5 F x 6.35 cm) x 3
Blood pressure cuff
Arterial line
Pulmonary Artery Catheter*
Cardiac Output Setup*
EKG: 5 lead system: Monitor II/V5 and ST
Temperature: esophageal
Esophageal stethoscope
End tidal C02 monitor
Pulse Oximeter
Foley and Urometer
Warming Devices
A.
B.
C.
Warming Pad set at 40°C
Level 1 fluid warmer
Bair Huggers x 2
*To be discussed with attending anesthesiologist
- 64 -
64
- 65 -
65
THERAPEUTIC PROTOCOLS
Vasopressin-Octreotide:
Markers for the use of intraoperative Vasopressin or Octreotide;
Evidence of portal hypertension. Refractory ascites a history of SBP, large volume paracentesis, current albumin
infusion, Aldactone dose of 400 mg/OD with or without the addition of other loop diuretics, diagnosis of portal
hypertensive gastropathy, > 2 variceal bleeds, occluded TIPS
Starting dose of Octreotide 200 ug in 200 mls at 50 ug/hr and maximum dose of 75ug/ hr is two or more of the above
criteria are present
Octreotide is discontinued at the initiation of the anhepatic stage so this should be enough for a procedure.
Starting dose of Vasopressin is 2.4 U/hr (0.04U/min)
rVIIa
No prophylactic dosing
Intervention for uncontrolled bleeding only after discussion with blood bank attending which we need to document on
our chart
Must meet current protocol criteria
AT-III
No prophylactic dosing
Intervention for uncontrolled bleeding only after discussion with blood bank attending which we need to document on
our chart
Antiemetics (serotonin antagonists)
We do not have any significant problems with PONV and can therefore forgo treatment with these agents except in the
rare case of rescue therapy in the PACU
Benzodiazepines
Single dose Lorazepam is the drug of choice due to its metabolic pathway. Patients who have greater than Grade 1
encephalopathy or who are on lactulose do not need prophylactic amnesic treatment
Albumin
5% albumin as needed in patients who come with a serum albumin > 3.0 and an infusion of 25% albumin when the
serum albumin < 3.0.
HBIG
This is the a surgical decision but the administration must be documented. The incidence (and cost) of antibodymediated complications can be reduced by starting the HBIG immediately after induction and running it over 5 hr.
Using this regime, anaphylaxix/anaphalactoid reactions are rare
THAM
THAM is the buffer of choice in hyponatremic patients
Blood Orders
In most circumstances we can order a low risk profile during normal working hours since the blood bank staff can be
mobilized very quickly, otherwise we will follow the criteria set out in the liver manual guiding blood orders
Early Extubation
Always adhere to our published criteria for early extubation and transfer to the surgical ward
- 66 -
66
NEW ANESTHETIC RECORD
A new anesthesia record has been designed specifically for Liver Transplantation. There are five original forms
consisting of
1.
2.
3.
4.
5.
Preanesthesia Evaluation
Intraoperative Events
Anesthesia Record
Blood Gas and Laboratory Data
Transfusion Data
These forms along with billing and a Quality Assurance questionnaire have been assembled and placed in a package
labeled Liver Transplantation Anesthesia Record. The package is available at the operating room front desk.
A set of instructions have been included in each package for reference.
Immediately following a Liver Transplant, place the yellow copies of all forms used for recording in the envelope and
put the package with other forms in the PACU. The record will be reviewed and any incomplete forms will be returned
to the resident for completion.
Please remember the importance of completing a Quality Assurance questionnaire. Most of our cases have a
recordable event and these must be documented. Failure to tolerate vena caval crossclamping, requiring VVBP is
considerable a recordable event as is sustained hypotension or requirement for vasoactive support.
The following description outlines a format for recording anesthetic information during the procedure.
Preoperative Assessment
An anesthesia consult, dictated by a liver transplant attending, is available on each patient. These are in the filing
cabinet located in the OR reading room. These consults are to be used for reference information and do not constitute a
preoperative assessment.
The new preoperative evaluation form contains two parts: part one outlines risk factors used in the assessment of all
patients presenting for liver transplant. The following topics are explained and an approach to recording outlined:
ASA
All patients are either ASA 4E or SE. Most ICU patients are 5E.
SHUNT
Surgical shunt for decompression of portal hypertension such as mesocaval, portocaval etc.
GI BLEED
Usually upper gastrointestinal hemorrhage associated with portal hypertension. Record
most recent event.
ASCITES
Recorded as controlled on diuretics, intractable and/or requiring paracentesis.
RENAL
Record presence of acute or chronic renal insufficiency.
TRANSFUSION
History of prior blood transfusion.
NUTRITION
Discuss this with your attending and record status as good, poor or cachectic.
ALLEN'S TEST
Despite the limitations of this test please document results as most patients have
bilateral, radial, arterial lines. Record the number of seconds required for refill.
- 67 -
67
The second section deals with patients in the intensive care unit. Most of these patients are extremely ill and awaiting
transplantation on an urgent basis. Please note that current CXR interpretations must be documented on all patients in
the ICU. Dialysis catheter site and type of exchange should be recorded (i.e. CAVHD vs. standard hemodialysis).
The forms are self explanatory, however if there is any difficulty in completing the information requested, please
consult with your attending.
Anesthesia Record/Intraoperative Events
This form was designed to simplify the recording of standard events such as line insertion, monitoring and positioning.
It will be assumed that all intravascular access is established under strict aseptic conditions and all central venous
access performed using the Seldinger technique unless otherwise stated. Any alteration from this protocol can be
recorded in the comment section along with the number of attempts or complications. Positioning and padding have
been standardized and can simply be checked off on the list provided. Only deviations from the documentation
protocol require description.
We are currently using Baxter pumps for infusion, Level 1 heat exchanger and the Rapid Infusion System. The
anesthetic setup has been outlined in the UCHSC Liver Manual. Any change from the standard setup must be outlined
in the Comment section. Accessory equipment (i.e., Camino or ICU ventilator) should be added to the Comment
Section.
Anesthesia Record
There is more room available for recording of medications and designated lines for hemodynamics. Separate lines
have been provided for calcium and bicarbonate administration. Record the latter two drugs as the number of ampules
given. Immunosuppressant medications have been listed in the right hand column and must be documented.
Anticoagulants refer to Antithrombin III, Amicar or Cofactor IX administration.
For dialysis and bypass lines, mark: as present or absent, start and finish time. A more detailed account is provided by
the involved service. The term "oozing" refers to profound stage III coagulopathy and is often associated with initial
poor function of the liver. Please consult with your attending before completing this section.
The following is a format for recording anesthetic information during surgery. These measurements represent the
minimum data points for the record.
Hemodynamic Measurements
Lines have been provided for recording of PAP (s/d) if used, CVP, and CO. Time intervals for these measurements are
as follows:
PAP and CVP recorded every 30mln
PCWP and
CO recorded
Stage I
1)
60min post induction
2)
90min post induction or just
prior to crossclamping the inferior vena cava
Stage II
3)
30min post crossclamp
4)
60min post crossclamp or
just prior to release of the crossclamp
Stage III
5)
6)
10min after unclamping
90min after unclamping
Laboratory Measurements
- 68 -
68
Arterial blood gases should be drawn at least every hour or as the situation dicatates. These may have to be drawn
more frequently in Stage II.
Laboratory Data Sheet
This is self explanatory. Please remember to fill out the upper section, most notably F102 and PEEP on all blood gas
results. The TEG data can be recorded in the bottom section. Fibrinolysis can be marked as + (present) or - (absent).
Fluid and Transfusion Sheet
It is important to mark times that fluids are given. At a minimum, document the stage of procedure during which
fluids/transfusions are given. All fluids are recorded on this sheet. Please write in unit numbers.
BLOOD COMPONENT THERAPY IN LIVER TRANSPLANT RECIPIENTS
Protocol for Supplying CMV "Reduced Risk" Components
I.
Recipient CMV Negative (Donor Liver CMV Negative)
Recipient CMV Negative (Donor Liver CMV Positive)
Give CMV antibody negative cellular components (WB/RBC and platelets).
OR
Give leukocyte-reduced cellular components (WB/RBC and platelets).
CMV positive donor, discuss treatment with antivirals
The type of component will be in part based on availability at Blood Bank.
II.
Recipient CMV positive, donor liver CMV negative or positive
No special components necessary.
III.
Other Issues
1.
Packed red cells, no older than 10 days.
2.
Short dated units are not used.
3.
Transfusion service must be provided with the CMV status of both the patient and
the donor.
4.
Blood Bank will perform an initial recipient evaluation to include ABO/Rh type,
antibody screen and CMV antibody testing.
5.
Should blood component demands exceed the ability of Belle Bonfils to
provide CMV antibody negative/leukocyte-reduced products, the Blood Bank
physician and the anesthesiologist will be notified.
"PLEASE NOTIFY BlOOD BANK OF AN IMPENDING LIVER TRANSPLANT AS SOON AS POSSIBLE**
Blood Utilization Protocols for Liver Transplantation
- 69 -
69
In 1988 the liver transplant program was introduced at the University of Colorado. A protocol was established by
members of the transplant team and blood bank to guarantee an adequate supply of blood components during surgery.
There are three standard orders available, high, medium and low risk.:
In an attempt to reduce the waste of plasma products, a new protocol has been designed and is outlined below.
High Risk
20 units RBC, FFP and PLA X 2.
A maximum of 10 units of FFP can be thawed and available at the start of the surgical case. The remaining FFP is to
be frozen but consigned. As units are used they should be replaced with a total of 20 units always available but only 10
units thawed.
Medium Risk
10 units RBC, FFP and PLA.
The same approach will be taken as in high risk cases but only 5 units of FFP will be thawed with the remaining 5
units consigned. Units will be replaced as they are used to maintain a total of 10 units of FFP with half frozen but
consigned.
Low Risk
5 units RBC and FFP. Platelets will be ordered on an individual basis. This is an uncommon category but has been
used in the past for certain groups of patients with relatively preserved hepatic function undergoing liver
transplantation for reasons other than cirrhosis.
Special Notes
The anesthesiologist may change the risk status of the patient at any time pre or intraoperatively after discussion with
the surgical team.
Blood components administered to the patient pre operatively should not be counted as part of the intra operative
order.
Under special circumstances the anesthesiologist may request the complete intra operative blood order be thawed and
immediately available. The reasons for this will always be documented on the anesthesia record and therefore
available for review.
During massive transfusion (greater than 20 units of packed cells), the physician on call for the blood bank should be
notified.
Patients with unusual antibodies will be treated as high risk candidates.
- 70 -
70
All unused blood components will be returned to the blood bank as soon as possible for reconsignment.
Until FFP and cryoprecipitate are more rapidly available the transplant team will have to order components in advance
to ensure patient safety. This protocol should reduce plasma component waste without influencing patient care.
Criteria for Blood Utilization and Risk Stratification
The criteria used for risk stratification are the following:
High Risk
Prior upper intra-abdominal surgery a single criteria (excluding cholecystectomy)
Two or more of the following:
Prior SBP
Renal failure requiring dialysis
Retransplant greater than 6 months following first transplant
Severe portal hypertension identified by at least two of the following:
1. ascites requiring paracentesis for control
2. poorly controlled upper GI bleed
3. marked splenomegaly with thrombocytopenia
Previous open cholecystectomy
Difficult blood crossmatch
Extended criteria donor
Low Risk
Age < 50
Liver synthetic function PT/albumin within 20% of normal
No signs of advanced portal hypertension (see above)
No prior upper intra-abdominal surgery
No obesity
Surgical confidence in graft at time of procurement
Criteria for Immediate Postoperative Extubation of Liver Transplant Patients
Patients are candidates for immediate postoperative extubation when they meet the standard criteria for extubation.
These criteria include: awake, positive gag, normothermia, VT > 5 mL/kg, ETCO2 within 20% of 35 mmHg, no
controlled bleeding and hemodynamically stable
Patients that meet the preoperative criteria should be discussed with the surgical staff regarding possible early
extubation. If the patient is found to be a candidate, the following changes should be made in anesthetic management.
1.
2.
3.
4.
5.
6.
7.
A maximum of 1 mg of Ativan may be given on induction
Fentanyl dose should be controlled since there is still increased brain sensitivity postreperfusion (i.e.,
maximum total dose of 10 ug/kg)
If it becomes necessary to discontinue the inhalational agent during times of hemodynamic
instability, then 1-2mg of midazolam can be administered
Desflurane run at flows less then 2L/min once steady state has been established
Neuromuscular reversal of 50-70ug/kg of neostigmine and 0.07-0.1 mg/kg of glycopyrrolate in ALL
candidates
The use of flumazenil and naloxone are up to the discretion of the anesthesiologist, but both are
potentially useful drugs in view of the known preexistent neuroreceptor alterations in these patients
Patients must meet the usual standards for extubation in the operating room
Please note that any patient who met preoperative criteria may be removed from this protocol if there are any
significant adverse intraoperative events.
- 71 -
71
INFECTIOUS DISEASE AND THE ANESTHESIOLOGIST
VIRAL HEPATITIS
CHARACTERISTIC
HEPATITIS A
HEPATITIS B
HEPATITIS C
Incubation (dy)
15-45
40-180
15-150
Epidemiology
faecal-oral
parenteral
Venereal
Perinatal
parenteral
faecal-oral
perinatal
Transfusion
Related
rare
5-10%
90-95%
Chronic
Hepatitis
no
5-10%
5-50%
Fulminant
Hepatitis
rare
1%
<5%
Hepatocellular
Carcinoma
no
yes
yes
HEPATITIS A
 Small risk of contracting virus during patient care
 Potentially infectious for 2 weeks after onset of jaundice
 Higher risk in pediatric setting
 Requires enteric precautions
 Prophylaxis – immune globulin
Given pre-exposure; 80-90% effective; efficacy decreases over 14dys
Only recommended after known exposure or enoemic area
 Vaccine in clinical trial
HEPATITIS B
ACUTE HEPATITIS B INFECTION
Complete recovery
90%
Fulminant
Carriage
1%
5-10%
Full recovery
50%
Chronic Active
50%
Increased Risk of Chronic Active
1)
immunodeficiency state-natural or secondary to other infection
- 72 -
72
Postexposure Prophylaxis
 risk of HBV infection in individual without immunity after needle stick exposure in HBsAg positive patient is
30%
 CDC recommendations for use under revision
PROTOCOL FOR PARENTERAL OR MUCOUS MEMBRANE EXPOSURE
1)
sample of patient’s blood to lab for HBsAg testing
2)
sample of HCW blood to lab for testing
3)
if patient HBsAg negative, do nothing
4)
if patient HBsAg positive and HCW anti-HBsAg positive then booster suggested if titre >10 RIA
5)
if patient HBsAG positive and HCW anti-HBsAg negative then HBIG given and repeat in 1 month,
also start vaccine
6)
if sample not available give both HBIG and vaccine
HEPATITIS C
Epidemiology
 most cases sporadic and without known etiology
 15% occur after blood transfusion
 90% of post-transfusion hepatitis secondary to HCV prior to RIA screen of blood products
 now most frequent cause of hepatitis in patients and HCW of dialysis patients
 frequency of occurrence in HCW unknown
Clinical







incubation 15-150 days
acute HCV clinically less severe than HBV
only 25% develop jaundice
fulminant course rare
at least 50% carrier rate
prevalence of carriers in USA may be as high as 3-7%
NOW SEROLOGICAL TESTING FOR HCV BY RIA-SENSITIVITY AND SPECIFICITY NOT
YET ESTABLISHED
Transmission
 blood products
 other body fluid risks uncertain
 epidemic form in Asia transmitted by possible faecal-oral route; has not been described in USA
 chance of transmission after needle stick is less than with HBV but conversion rate not yet
established
Prevention
 NO VACCINE
 EFFECTIVENESS OF IMMUNE GLOBULIN NOT DOCUMENTED
 HBV IMMUNE GLOBULIN NOT EFFECTIVE
- 73 -
73
PREVENTION PROTOCOLS
1) identification of high risk individuals
a)
homosexual or bisexual men
b)
intravenous drug abusers
c)
recipients of frequent blood products
d)
people of endemic areas
e)
dialysis patients
f)
institutionalized patients
g)
prostitutes
h)
ALL PATIENTS
2) body fluid precautions for ALL patients including protective eyewear
3) care with all sharp instruments, remember all sharps disposal boxes MUST be replaced when ¾ full!
4) use 3 way stop cocks and no recapping used needles if at all possible
5) regular hand washing
IF EXPOSURE OCCURS REPORT TO EMPLOYEE HEALTH FOR TESTING AND POSSIBLE TREATMENT
IMMEDIATELY
- 74 -
74