Tải bản đầy đủ - 0trang
2 What Is PLF? Definition, Incidence, and Risk Factors
Postsurgical Liver Failure
Table 9.2 Risk factors for PLF
Parenchymal disease: cirrhosis, nonalcoholic fatty liver disease,
chemotherapy-induced liver injury (steatohepatitis and sinusoidal injury),
Age >65 years
Excessive blood loss
Extent of resection (>4 segments)
Use of vascular occlusive techniques
Ex vivo hepatic resection
Excessive blood loss and transfusion
Vascular or biliary reconstruction
Hepatic parenchymal congestion
management in the intensive care unit (ICU). However, also an improved patient
selection and management of risk factors seem to have a significant influence on the
occurrence of PLF [13, 27]. The identification of the surgical risk is imperative in
the care of any patient, especially as patients develop an increasing number of
chronic comorbid medical conditions. In resective liver surgery, three groups of risk
factors can be differentiated: patient, surgery, and miscellaneous (Table 9.2).
Patient-Related Risk Factors
Patients with a preexisting liver disease are at particularly high risk for an increased
morbidity and mortality in the postoperative period of liver resections due to both
the “generic” stress normally associated to any major surgery and specific conditions referable to the peculiarities of liver surgery itself. In a study comparing 135
patients with cirrhosis to 86 patients without cirrhosis, all undergoing non-hepatic
general surgery, 1 month mortality rates were 16.3 % in the cirrhotic and 3.5 % in
the control group . Even steatosis of the liver seems to be associated with a
higher perioperative rate of complications and increased incidence of PLF ,
whereas preoperative cholestasis was not shown to be associated with an increased
risk for PLF [13, 27]. Besides “primary” liver diseases for various reasons, preoperative chemotherapy resulting in chemotherapy-associated steatohepatitis or sinusoidal obstruction is another important defined risk factor for PLF as well.
Chemotherapy-induced liver injury is increasingly prevalent as more patients
receive chemotherapy for colorectal liver metastases before liver resection. The
liver injury varies according to the chemotherapeutic agents, duration of treatment,
and presence of preexisting parenchymal disease. The two major patterns of liver
injury are sinusoidal injury and chemotherapy-associated steatosis and steatohepatitis (CASH) [13, 27].
In summary, there is no doubt that patients suffering from any form of liver disease have an increased risk for PLF that depends on the functional reserve of the
liver preoperatively. What is further evident in the literature is that decompensated
liver disease increases the risk of postoperative complications (e.g., infections
including sepsis, bleeding, poor wound healing, and renal dysfunction).
Surgical-Related Risk Factors
With regard to the surgical-related risk factors, the extent of resection correlates closely
with the rate of PLF as failure to regenerate occurs when the remnant liver volume is
below a certain threshold. It has been reported that the incidence of PLF increases with
the number of segments resected and that death from PLF can be as high as 80 % after
resection of more than 50 % of the native parenchyma. On the other side, PLF is less
than 1 % in patients with no underlying parenchymal disease when one or two segments are resected, around 10 % when four segments are resected, and 30 % when five
or more segments are resected . It has been also estimated that a minimal functional
liver remnant volume (FLRV) of 20–25 % will be needed for an adequate liver function
after surgery in patients with a normal liver parenchyma, whereas patients with abnormal parenchyma (steatosis, fibrosis, or cirrhosis) will need an FLRV up to the 40 % of
the native liver [28, 30]. Therefore, assessing how much functioning liver will be left
after surgery is a cornerstone phase during the preoperative workout of candidates to
liver resection. To this end, preoperative radiological assessment and volumetry using
computed tomography (CT) or magnetic resonance imaging are used to enable prediction of FLRV and identification of underlying parenchymal disease, whereas CT-guided
three-dimensional reconstructions allow visualization of the hepatic venous outflow
and improve tumor localization, thus facilitating operation planning . The sensitivity
of volumetric assessment can be further enhanced by combining it with a body surface
area or bodyweight calculation.
Other surgery-specific risk factors for PLF are related to the use of techniques
temporarily occluding the liver vascular pedicle (Pringle maneuver) in the aim to
limit bleeding due to the parenchyma resection. In fact, intraoperative vascular
occlusive techniques can exacerbate the severity of postoperative hepatic dysfunction by inducing ischemia in the remnant liver. When vascular exclusion is total
(inflow + outflow occlusion), the effect is the greatest, but liver cell injury can also
occur after prolonged intermittent inflow occlusion [13, 27]. Another significant and
equally severe risk factor results from the amount of intraoperative bleeding, and
blood losses >1 L with the consequent need for a considerable amount of transfusions can significantly increase the risk of PLF [13, 14, 27]. Finally, a prolonged
operation time can also play a role in increasing the risk for PLF .
Age (the regenerative capacity of liver tissue decreases with it), malnutrition (which
is associated with an altered immune response and a reduction in hepatocyte regenerative capacity possibly due to disordered mitochondrial function), diabetes
Postsurgical Liver Failure
mellitus (possibly due to immune dysfunction or because insulin absence or resistance reduces regenerative capacity), male sex (testosterone may have immuneinhibitory effects, predisposing to septic complications), and cholestasis can play a
role in influencing the severity of PLF. In particular, cholestasis, such as from
malignant hilar obstruction, reduces hepatic metabolic and regenerative capacity.
Although preoperative biliary drainage (PBD) improves the remnant function, its
routine use in jaundiced patients is debated as it does not confer a survival benefit
and increases morbidity [15, 17]. Therefore PBD may be limited to those requiring
major resection with a predicted FLRV of less than 40 %, who require volume
manipulation or have cholangitis .
What Can We Do to Prevent PLF?
Because the therapeutic options for PLF are poor and scarce, great efforts should be
made to prevent its occurrence. However, some of the risk factors cannot be influenced (age, gender, existence of cirrhosis or fibrosis, and diagnosis of the patient).
Furthermore, patients in poor general condition have “per se” an overall higher risk
for perioperative complications. Therefore, meticulous preoperative selection and
optimization strategies, optimal intraoperative surgical and anesthetic techniques,
and cautious postoperative care should be used to prevent PLF occurrence.
Candidates to liver surgery should fulfill the criteria for general operational capability. Comorbid conditions should be optimized before surgery as much as possible.
To decrease the risk of general complications, diabetes mellitus should be screened
for and treated before surgery. The optimization of nutritional status especially in
patients with cirrhosis may be helpful, but no relationship between malnutrition and
PLF has been demonstrated . In particular, there is no evidence to support delaying liver resection for a period of nutritional preoptimization, unless the patient is
severely malnourished .
Determining the functional reserve of the liver and predicting the volume of the
remnant liver is a cornerstone in the patient preparation before liver surgery. Routine
preoperative biochemical measurements (albumin, PT, bilirubin, aminotransferases,
γ-glutamyl transferase, and alkaline phosphatase) can provide indicators of hepatic
dysfunction and may reflect ongoing parenchymal damage or cholestasis but do not
independently predict PLF . Some scores have proved to be useful tools for
assessing the functional capacity of the liver in view of a surgical intervention. The
Child–Pugh classification is one of the more used to this end (Table 9.3). There is
general consensus that liver surgery should only be conducted in patients with “stable” cirrhosis (classified as Child A) and in some very well selected Child B patients.
Table 9.3 Child–Pugh classification
Total bilirubin, μmol/l
Serum albumin, g/dl
Grade I–II (or suppressed
Moderate to severe
Grade III–IV (or
The MELD (Model for End-Stage Liver Disease) score also reflects hepatocellular
function. It is based on the formula that combines bilirubinemia, creatinine, and
MELD score = (0.957 • ln ( serum creatinine ) 0, 378 • ln ( serum bilirubin )
1,120 • ln ( INR + 0, 643) • 10
This score, initially developed to predict death within 3 months of surgery in patients
who had undergone a transjugular intrahepatic portosystemic shunt procedure, was
subsequently found to be useful in determining prognosis and prioritizing for receipt
of a liver transplant and can be used for patients with hepatic malignancy undergoing liver resection to assess their perioperative morbidity and mortality .
A MELD score >10, when compared with a score of <9, was associated with a
significant increased risk of PLF after hepatectomy for hepatocellular carcinoma in
cirrhotic patient .
Beyond deductive methods, hepatic reserve can be assessed through more objective tests. The most commonly used method is based on the use of indocyanine
green. After a bolus injection of indocyanine green, the dye binds to plasma proteins
and is removed exclusively by the liver through a carrier-mediated mechanism; the
dye is ultimately excreted unchanged into the bile. It is not metabolized and does
not undergo enterohepatic circulation. The disappearance curve of ICG has two
components, a distribution and an elimination phase, and the turning point of these
two phases is 20–30 min. ICG has a relatively high intrinsic clearance; therefore,
ICG retention at 15 min (ICG R15) represents hepatic perfusion. When hepatic
function is impaired, ICGR15 increases. If ICGR15 is less than 14 % in patients
with cirrhosis, major hepatectomy is well tolerated; when ICGR15 exceeds 20 %,
major hepatectomy should be avoided. Patients with a rate between 14 and 20 %
benefit from volume manipulation [6, 7]. Finally, there is recent evidence that intraoperative ICG clearance measurements might allow real-time monitoring of intraoperative liver function during surgery. In fact, trial clamping of arterial and
portovenous inflow predicted, in 20 patients undergoing anatomic liver resection,
immediate post-resection liver function, thus possibly helping to avoid PLF .
Another means for preventing or reducing the severity of PLF is improving size
and function of the FLRV. Strategies available for volume manipulation include
portal vein occlusion (PVO) and two-stage resection. PVO is usually performed