Information Hep C Information Station Assessment of Prognosis of Cirrhosis
| Assessment of Prognosis of Cirrhosis |
 |
 |
 |
| Written by Linda | |
| Thursday, 08 May 2008 09:03 | |
François Durand, M.D.; Dominique Valla, M.D.AbstractOnce patients with cirrhosis experience decompensation, early mortality risk increases sharply. Liver transplantation has transformed the prognosis of decompensated cirrhosis. Child-Pugh score has been the reference for many years for assessing the prognosis of cirrhosis. However, Child-Pugh score has important limitations among which is subjective interpretation of some of its variables, making it difficult to categorize patients according to their own disease severity. The model for end-stage liver disease (MELD) score, which was originally designed for assessing the prognosis of cirrhotic patients undergoing transjugular intrahepatic portosystemic shunt (TIPS), is a continuous score relying on three objective variables. Along with TIPS, MELD score proved to be a robust marker of early mortality across a wide spectrum of causes of cirrhosis, even though 10 to 20% of patients are still misclassified. MELD is especially useful for prioritizing candidates for transplantation according to a “sickest first” policy. However, MELD is not a universal prognostic marker of cirrhosis and several MELD exceptions require more specific approaches. General Principles and Natural History of CirrhosisAny patient with cirrhosis carries a risk of specific life-threatening complications such as variceal bleeding, sepsis, or hepatorenal syndrome. There is also a significant risk of nonspecific life-threatening complications due to the frequent association of comorbidities. The general course of the disease is characterized by a longstanding phase of compensated cirrhosis, followed by the occurrence of specific complications. It has been shown that 10 years after diagnosis, the probability of developing decompensated cirrhosis is ?60%, ascites being the most frequent complication (?50%).[1] Once patients have developed the first episode of decompensation, complications tend to accumulate and life expectancy is markedly reduced. A schematic representation of the course of cirrhosis according to four consecutive clinical stages is shown in Table 1 .[2] The course of cirrhosis is extremely variable from patient to patient due to several factors, including hepatic synthetic function (or “hepatic reserve”), the cause of cirrhosis, the possibility of stopping or slowing the underlying damaging process to the liver, and the occurrence of liver malignancy. Therefore, establishing a prognosis in a given patient with cirrhosis remains a challenging issue. In addition to the simple estimation of life expectancy, more complex issues must be taken into account, such as the capacity of a cirrhotic patient to withstand a given therapeutic intervention, or whether a given therapeutic option offers an acceptable chance of survival. Over the last couple of decades, additional complex issues have emerged with the generalization of liver transplantation, namely, the optimal timing for transplantation and, on a collectivity basis, the optimization of allocation policy in a context of organ shortage. Even though the course of cirrhosis varies according to several factors, the need for prognostic models and scoring systems is obvious in order to manage individuals faced with different therapeutic options. Scoring systems are even more crucial for managing populations of patients in the setting of transplantation, for instance. Major efforts have been made in recent years to achieve these goals and develop prognostic tools which are detailed below. Hepatocellular carcinoma (HCC) will not be discussed in this article. Child's ScoreDefinitionsChild's score, initially termed Child-Turcotte score, was proposed more than 30 years ago.[3] It was originally designed for predicting the outcome after surgery for portal hypertension (portocaval shunting and trans-section of the esophagus) in patients with cirrhosis. Child-Turcotte score included two continuous variables (bilirubin and albumin) and three discrete variables (ascites, encephalopathy, and nutritional status) which were empirically selected because they were felt to have their own influence on the prognosis in this context.[4] A modified version termed Child-Pugh score was proposed ?20 years ago ( Table 2 ).[5] The only change in this modified version was that nutritional status was replaced by prothrombin time. Initially, prothrombin time was expressed in seconds. However, a limitation comes from the fact that prothrombin can also be expressed as either a percentage of normal (prothrombin index) or as international normalized ratio (INR), this latter being now the reference in many countries. The original cut-off values of 4 and 6 seconds for prothrombin time prolongation correspond approximately to a prothrombin index of 50% and 40%, respectively. These same values roughly correspond to an INR of 1.7 and 2, respectively. Child-Pugh score corresponds to the total of points for each item. According to the sum of these points, patients can be categorized into Child-Pugh grades A (5 to 6 points), B (7 to 9 points), or C (10 to 15 points). The variables included in Child-Pugh score are not specific markers of the synthesis (albumin and prothrombin) and elimination (bilirubin) functions of the liver. Indeed, changes in serum albumin may be also related to increased vascular permeability,[6] especially in cases of sepsis, and large-volume ascites.[7] Similarly, bilirubin can be increased as a consequence of impaired renal function, hemolysis, or sepsis.[8] Prolonged prothrombin time can be a consequence of an intravascular activation of coagulation during sepsis.[9] Overall, the individual components of the Child-Pugh score encompass a broader spectrum of conditions than the single impairment of “liver function.” Child-Pugh score as a whole is also a marker of the multiorgan changes resulting from cirrhosis. ApplicationsSeveral studies have shown that Child-Pugh score is an independent prognostic marker in the settings of ascites,[10] ruptured esophageal varices,[11] alcoholic cirrhosis,[12] hepatitis C virus- (HCV-) related cirrhosis,[13] primary biliary cirrhosis (PBC),[14] primary sclerosing cholangitis (PSC),[15] and Budd-Chiari syndrome.[16] Child-Pugh score, which can be easily calculated at the bedside, has been widely used for selecting candidates for resection of HCC[17] and nonhepatic surgery.[18] Attempts have been made to improve the accuracy of Child-Pugh score by incorporating other markers of liver function (such as the clearance of galactose, indocyanine green, aminopyrine, and lidocaine).[19-22] However, these additional markers had a limited value. The incorporation of conventional markers of nutritional status to Child-Pughscore also had a limited value.[23] LimitationsSince the five variables of Child-Pugh score were selected empirically, it can be anticipated that not all are independent predictors of prognosis. For example, albumin and prothrombin time are somewhat redundant. Including both variables in a single score may result in overweighting their own influence. The cut-off value for each variable has been empirically selected. There is no evidence that moving from one class to the next one translates into a proportional change in mortality risk. As an example, patients with serum bilirubin above 100 ?mol/L may be under-scored with Child-Pugh, because the limit for the upper class of bilirubin is only 51 mmol/L. This ceiling effect of discrete classes does not exist with continuous variables. In addition, the limits for qualitative variables (ascites and encephalopathy) are still vague. They may be influenced by subjective interpretation. The five variables of Child-Pugh score are empirically given the same weight, which is also questionable. Multivariate analysis has shown that the proper weight of predictive factors is quite variable. For example, the weight of INR is three times as high as that of bilirubin in MELD score.[24] Child-Pugh score does not take into account specific variables, serum creatinine in particular, which have been shown to have a determinant impact on the prognosis of cirrhosis.[10,24] Similarly, it has been shown that the addition of markers of portal hypertension, such as esophageal varices, portal blood velocity, and hepatic venous pressure gradient (HVPG), would improve the accuracy of Child-Pugh score.[20,25,26] Finally, Child-Pugh score does not take into account the cause of cirrhosis and the possibility of stopping (or slowing) the damaging process to the liver. This limitation is especially relevant in patients with alcoholic cirrhosis or with hepatitis B virus- (HBV-) related cirrhosis with viral replication. Meld ScoreDefinitionsWhile Child score was originally designed for assessing the prognosis of cirrhotic patients undergoing surgical treatment of portal hypertension, MELD score was designed for assessing the prognosis of cirrhotic patients undergoing transjugular portosystemic intrahepatic shunt (TIPS).[24] Multivariate analysis using Cox regression analysis showed that among a list of pre-determined variables, four variables had an independent impact on survival, namely bilirubin, creatinine, INR, and the cause of cirrhosis (alcoholic and cholestatic versus other causes). To lessen the influence of extreme values, the natural logarithm of bilirubin INR and creatinine were entered into the model. Based on regression analysis, a coefficient was attached to each variable, according to the weight of each variable on mortality risk. In the original series, the resulting score was slightly more accurate than Child-Pugh score for predicting survival after TIPS. With the expansion of liver transplantation, attention moved from the management of portal hypertension to waiting list mortality and organ allocation policy in patients listed for transplantation. Indeed, waiting time alone proved to be an inaccurate marker of waiting list mortality[27] and more relevant markers were needed. A slightly modified risk score, termed the MELD score, was tested in populations of cirrhotic patients for assessing early (3 months) mortality risk after placement on the waiting list.[28] This modified score proved to be a robust marker of early mortality which could be generalized to patients with various causes of cirrhosis and various degrees of severity. Excluding the variable “cause of cirrhosis” had a minimal impact on the model accuracy. As a result, a simplified version of MELD score including only three objective variables (bilirubin, creatinine, and INR) was eventually proposed for easier use ( Table 3 ).[29] According to this modified score, patients with bilirubin and creatinine values below 1 mg/dL (17 and 90 ?mol/L, respectively) are rounded off to 1 mg/dL to avoid negative logarithmic values. Similarly, patients with INR below 1 are rounded off to 1. Whatever the individual values, the score is empirically capped at 40. Consequently, MELD score represents a continuous variable ranging from 6 to 40. ApplicationsMELD score has been adopted since 2002 for organ allocation to patients listed for liver transplantation in the United States.[30] According to the MELD-based policy, patients with the highest score have a priority for organ allocation. More recently, MELD score has also been adopted in several European countries as well as in South America. The impact of MELD score on liver transplantation is discussed below. In addition to organ allocation, several studies have confirmed that MELD score is a reliable tool for predicting outcome after TIPS.[31,32] The “c” statistic represents a global estimate of the ability of a score to predict an event. This statistic, which is derived from the area under the receiver operating characteristic curve, ranges from 0 to 1. A “c” statistic of 0.5 means that the score is of no value for predicting a given event. A “c” statistic of 1 means that the score is perfect (a goal which is never achieved in clinical practice). The “c” statistic of the MELD score for predicting 1-year survival after TIPS is ?0.7, which means that it is clinically useful. MELD score also proved to be a reliable marker of 1-year and 5-year survival across a broad spectrum of liver diseases including alcoholic cirrhosis and alcoholic hepatitis.[33] In addition, MELD score has been shown to be a good prognostic marker in cases of variceal bleeding,[34] spontaneous bacterial peritonitis,[35] and hepatorenal syndrome.[36] Independent of the cause of cirrhosis, high MELD score was shown to be associated with a decrease in residual liver function as measured by monoethylglycinexylidine test.[37] The particular issues of alcoholic cirrhosis, HCC, nontransplant surgery, and transplantation are detailed below. LimitationsUsing statistical analysis rather than empirical selection of variables for establishing a prognostic score can be anticipated as more rigorous, on a methodological basis, and eventually, more efficacious. However, it must be kept in mind that even if the three components of the MELD score have been selected on the basis of a statistical analysis, all the variables entered in the model have been empirically selected, because they were felt to have an influence on the outcome or, more simply, because they were available. It cannot be excluded that other important variables may not have been taken into account. The three variables entered in the MELD score (bilirubin, creatinine, and INR) are supposed to be objective (in contrast to ascites and encephalopathy, the grading of which is subjective). However, even these “objective” markers may be subjected to significant variations resulting from either changes over time or from different laboratory methodologies. There are substantial interlaboratory variations in INR depending on the methods used for determination. It has been shown that interlaboratory variation in INR is ?25%. Among the three variables of MELD score, INR has the highest multiplicative value. Therefore variations in INR may translate to up to 20% differences in MELD score.[38] In single individuals with cirrhosis, substantial changes in serum creatinine may occur, especially in those undergoing large-volume paracentesis and/or receiving diuretics. Laboratory methods may also interfere with the value of serum creatinine. There is a poor agreement among different creatinine assays, especially as serum bilirubin rises.[39] Overall, MELD score is not as objective as it was expected to be. A major limitation of MELD score is the need for computation, which makes it less friendly to use than Child-Pugh score at the bedside. Logarithmic transformation has been chosen to optimize the statistical model. Originally, this statistical model and the derived score were not designed to be routinely used in clinical practice. With the expansion of MELD score in many fields of hepatology, the need for computation represents a source of difficulties. There is no evidence that a simplified score based on the original values of bilirubin, INR, and creatinine (without logarithmic transformation) would be less accurate. In addition, there are no clear-cut values of MELD score for easily categorizing individual patients according to their own mortality risk. Meld Score DerivativesMELD-NaWith the implementation of the MELD score, refractory ascites was removed from the list of variables used for assessing the prognosis. However, even though its interpretation can be subjective, ascites was shown to be associated with the poorest prognosis.[40] It was felt that patients with refractory ascites, normal creatinine, and preserved hepatic function could be under-scored with MELD. In particular, it was shown that persistent ascites and low serum sodium identified a subset of patients with relatively low MELD score (below 21) and a high risk of early death.[41] Serum sodium is a simple, readily available, and objective marker of disease severity. During cirrhosis, hyponatremia results from solute-free water retention. Systemic arterial vasodilation leads to the release of antidiuretic hormone which, in turn, induces dilution hyponatremia. The activation of these mechanisms correlates with the degree of portal hypertension.[42] In this view, hyponatremia can be considered an indirect marker of portal hypertension during cirrhosis. Several studies have shown that hyponatremia is a strong predictor of early mortality, independent of MELD score.[41,43-45] Changes in survival are especially pronounced for sodium concentrations ranging from 120 to 135 mEq/L. Within this range, a decrease in serum sodium of 1 mEq/L corresponds to a 12% decrease in 3-month probability of survival.[45] A modified score including serum sodium, termed MELD-Na, has been proposed as an alternative to MELD score ( Table 3 ).[44] The accuracy of MELD-Na was shown to be slightly superior to that of MELD in candidates for transplantation.[43-45] The effect of hyponatremia is higher in patients with low MELD score compared with those with high MELD score. A limitation to the incorporation of serum sodium into MELD is that during cirrhosis, marked changes in serum sodium concentration can result from several factors, including the administration of diuretics and intravenous hypotonic fluids. For example, the administration of diuretics leads to a 4 mEq/L decrease in serum sodium, on average.[46] In some patients the decrease may reach 10 mEq/L. In contrast, the use of V2-receptor antagonists for treating refractory ascites is encouraging. These agents induce a significant increase in serum sodium. Again, serum sodium is not as objective as it was thought to be. Further validation is needed and practical guidelines regarding the incorporation of sodium should be proposed to avoid misclassification. MELD-XIINR is the variable which has the highest weight in MELD score ( Table 3 ). Unfortunately, INR is hardly interpretable in patients receiving anticoagulation therapy. It is not exceptional that patients with cirrhosis receive anticoagulation due to portal vein thrombosis, an underlying prothrombotic state, or any other concomitant condition.[47] Most patients with Budd-Chiari syndrome also receive anticoagulation with anti-vitamin K. In this population, INR is artificially increased. Using MELD score in this context would result in overestimating disease severity. With the aim of overcoming this difficulty, a modified MELD score termed MELD-XI (for MELD excluding INR) has been proposed.[48] This modified score relies only on bilirubin and creatinine. Thus, neither INR nor any other marker of coagulation is taken into account ( Table 3 ). The coefficients ascribed to creatinine and bilirubin have been changed to obtain the optimal linear correlation between MELD and MELD-XI. In other words, the adjusted coefficients mean that patients with a given MELD-XI score have a mortality risk comparable to that of patients with interpretable INR and a similar MELD score. The validation of MELD-XI score shows that its accuracy for assessing 3-month mortality risk is comparable to that of MELD (with a c statistic of ?0.83).[48] Omission of INR and the use of adjusted coefficients did not much change the predictive accuracy of the score, which is somewhat surprising since INR was a strong, independent prognostic marker in the original series from which MELD score is derived. Moreover, it must be noted that both creation and validation of MELD-XI score have been made in populations of patients who did not receive anticoagulation. Patients receiving anticoagulation are expected to have comorbidities and/or a different natural history. These patients may have specific risk profiles with a higher risk compared with patients without thrombotic complications. Therefore, further validation in this particular population is needed. Delta MELDIntuitively, it can be anticipated that taking into account changes in MELD score over time may add prognostic information.[49] Patients with a rapid increase in MELD over time might be expected to have a worse outcome than those with stable or even decreasing MELD score. Delta MELD is defined as the difference between current MELD and the lowest MELD measured within 30 days prior to current MELD ( Table 3 ). Delta MELD was shown to be predictive of early mortality in patients with cirrhosis on univariate analysis. However, delta MELD was no longer predictive of mortality when entered into a multivariate model with current MELD score.[50] These results suggest that current MELD score is the only predictor of mortality regardless of how that score was reached. Prognosis According to Specific Causes of CirrhosisAlcoholic Cirrhosis and Alcoholic HepatitisA particularity of alcoholic cirrhosis is that the majority of patients with high disease severity indexes do have superimposed alcoholic hepatitis. Alcoholic hepatitis is a potentially reversible condition, which means that some of these patients are likely to improve within the first months following discontinuation of alcohol. Such patients may return to a state of compensated cirrhosis. Accumulated evidence has shown that corticosteroids improve short-term survival in patients with severe alcoholic hepatitis.[51] Therefore, it would be difficult to assess the prognosis of patients with alcoholic cirrhosis without taking into account the existence of alcoholic hepatitis and, in those with severe alcoholic hepatitis, response to steroids. In most surveys, severe alcoholic hepatitis has been defined by a “discriminant function” above 32 ( Table 4 ).[52] In addition to this discriminant function, generally termed as “Maddrey score” (or Maddrey discriminant function), several specific scores have been created to predict early mortality in patients with severe alcoholic hepatitis.[53,54] The more general MELD score has also been assessed in this setting. MELD score proved to be as efficacious as or even superior to the original Maddrey discriminant function.[55,56] This finding is not surprising since MELD score includes the two variables included in the Maddrey discriminant function (bilirubin and prothrombin). However, none of these scores takes into account the progression over time and the response to steroids which, again, may be determinant. Recently, an original model termed Lille model was created on the basis of a large series of patients with alcoholic hepatitis treated with steroids.[57] The corresponding score incorporates six objective variables as shown in Table 4 . This score, which includes a dynamic variable corresponding to early response to steroids (change in bilirubin between day 0 and day 7), was more accurate than MELD score, Child-Pugh score, and Maddrey score for predicting 6-month survival in this population. Patients with a score above 0.45 had a 6-month mortality rate of 75%, while those with a score below 0.45 had a mortality rate of only 15%. It must be noted that these scores have been designed for predicting early mortality. Along with response to steroids, long-term outcome may be related to the residual liver function but also to return to alcohol abuse. As a result, long-term outcome is much more difficult to assess. In addition, long-term survival should necessarily be assessed in view of comorbidities such as cancer, diabetes, or respiratory diseases, which are frequently associated. HBV- and HCV-Related CirrhosisIn recent years, major advances have been achieved in the treatment of chronic HBV infection with the advent of antinucleot(s)ide analogues. Patients with decompen-sated HBV-related cirrhosis receiving antiviral therapy have a biphasic survival pattern. Mortality rate within the first 6 months after initiation of antiviral therapy is ?15%.[58] After the first 6 months, mortality rate is 10 times lower. In the subgroup of patients who survive more than 6 months, 3-year survival exceeds 85%. Compared with historical controls, patients with HBV-related decompensated cirrhosis receiving antinucleot(s)ide analogues have better survival rates. Based on a large cohort of patients with decompensated HBV-related cirrhosis receiving lamivudine, a specific prognostic score has been proposed ( Table 4 ). This score incorporates three variables: bilirubin, creatinine, and the presence of HBVDNA before treatment. It is somewhat paradoxical that patients with undetectable HBV-DNA before initiation of antiviral therapy had a better survival. Indeed, it could have been expected that only those with evidence of viral replication (those who are positive for serum HBVDNA) would benefit from antiviral therapy. Nonetheless, other data suggest that 20% of patients initially considered for transplantation can eventually be removed from the waiting list after receiving adefovir-dipivoxil as a result of clinical improvement.[59] Advances in the treatment of HCV infection have been more limited, although significant. In patients with HVC-related cirrhosis, sustained virological response to interferon was shown to improve long-term outcome by reducing the incidence of liver-related complications.[60] However, the combination of interferon and ribavirin is generally contraindicated in patients with decompen-sated cirrhosis. There is no specific model for predicting the outcome according to viral load, genotype, and response to therapy. PBCPBC is one of the causes of cirrhosis for which specific prognostic scores were first proposed.[61,62] The aim of scoring was to determine the optimal timing for transplantation. The Mayo risk score for PBC includes four objective variables and one subjective variable (i.e., edema) ( Table 4 ). It has been shown that the probability of survival without transplantation for a risk score of 7.8 is 63% and 39% at 1 and 2 years, respectively. It has also been shown that the risk of post-transplant mortality increases significantly when the risk score exceeds 7.8. Therefore, it is recommended that patients be referred to transplantation centers before reaching this value. Another prognostic model has been created based upon a large European series.[63] This model includes variables which are relatively close to those included in the Mayo model, which are bilirubin, ascites, albumin, age, and gastrointestinal bleeding. According to this model, the survival benefit from transplantation increases when the probability of survival without transplantation falls below 0.85. In nontransplanted patients, this occurs on average 8 months before death which, in most cases, gives sufficient time to bridge patients to transplantation. Even though specific scores exist for PBC, there is no evidence that patients with PBC are misclassified with MELD score. Nor there is evidence that specific scores are superior to MELD. However, no discriminant value of MELD score has been established to specifically identify PBC patients who may benefit from transplantation. PSCThe course of PSC is much more variable than that of PBC. Therefore, it is more difficult to create reliable prognostic scores, especially for assessing long-term outcome. It has been shown that the specific risk score (termed Mayo risk score for PSC) shown in Table 4 is more accurate than Child-Pugh for predicting survival, especially in patients with less-advanced disease.[64] This score allows the identification of three groups at low (score < 0), intermediate (0 ? score > 2), or high (score ? 2) risk. Five-year survival is above 90% in patients at low risk while it is less than 40% in patients at high risk. MELD score has not been specifically assessed for PSC. However, most patients with advanced PSC have high bilirubin level. In these patients, it is unlikely that disease severity is underestimated by MELD score compared with other chronic liver diseases. However, some patients with relatively low markers of severity have repeated episodes of cholangitis and a rapid deterioration. In this group, bilirubin may be fluctuating. It can be assumed that, independent of the severity of the underlying parenchymal disease, repeated episodes of cholangitis have a deleterious impact on the outcome. Unfortunately, the own influence of the repitition of episodes of cholangitis, response to antibiotics, and the possible selection of resistant strains on prognosis have not been clearly assessed. Prognosis in the Particular Setting of Nontransplant SurgeryIndependent of liver resection for HCC, the probability is relatively high that patients with cirrhosis will require surgery (whether intra- or extra-abdominal) at some time. Unfortunately, patients with cirrhosis also represent a population at especially high risk of surgical morbidity and mortality. The most recent reports indicate that in this population, in-hospital mortality may be as high as 10 to 20%, even though it can be assumed that most patients were carefully selected.[65,66] Mortality is the consequence of a high rate of postoperative decompensation of cirrhosis (especially in cases of intra-abdominal surgery) and an increased risk of bacterial infection. The issue of surgery and cirrhosis depends on whether there is an alternative to surgery or not. When nonsurgical alternatives exist, prognostic markers should help determine whether the risk of surgery is justified. Child-Pugh score has been used for more than two decades for addressing these issues. More recently, MELD score has also been assessed for predicting non-transplant surgical mortality.[65] The results proved relatively good. In general, there is approximately a 1% increase in mortality risk per MELD point below a score of 20. There is a 2% increase in mortality risk per MELD point over 20.[65] Mortality is higher for intra-abdominal surgery (up to 25%) compared with other types of surgery. The c statistic of the MELD score for predicting 30-day mortality was found to be 0.72 in the whole population of patients undergoing surgery and 0.8 in the subgroup with intra-abdominal surgery. However, there are no simple limits with MELD score such as Child-Pugh grades A, B, and C for estimating patients' risk. In addition, the proper risk of surgery is not balanced against the expected benefit and the existence of nonsurgical alternatives.[67] Simple algorithms based on MELD score for different types of surgery would be helpful to replace Child-Pugh score. The issue of “rescue” transplantation in cirrhotic patients who have severe decompensation and profound liver insufficiency after liver resection is also important. Indeed, cirrhotic patients have limited liver regeneration capacity. It has been shown that the persistence of a decrease in prothrombin index below 50% of normal (INR of ?1.7) and an increase in serum bilirubin above 50 ?mol/L on postoperative day 5 is associated with a 60% risk of early mortality.[68] These criteria allow early identification of patients who may need emergency transplantation, provided there is no general contraindication. Prognosis in the Particular Setting of the Intensive Care UnitIn general, the prognosis of cirrhotic patients admitted to the intensive care unit (ICU) due to multiorgan failure is especially poor. Mortality rates in patients with failure of two or three organ systems are estimated to be ?75% and 95%, respectively.[69] Mortality is much higher than that of noncirrhotic patients with multiorgan failure. As an example, mortality rate among noncirrhotic patients with failure of two or three organ systems in a context of severe sepsis is ?26% and 34%, respectively.[70] Among other factors, the poor outcome of cirrhotic patients with multiorgan failure results from a rapid alteration of liver function, a limited capacity for liver regeneration, and the absence of efficient artificial liver support systems. Predicting the outcome in this context may help optimize resource utilization, as aggressive management is not justified in all patients. However, it must be kept in mind that on the other hand, not all patients admitted to the ICU have a fatal outcome. In particular, some patients may be efficiently bridged to “rescue” transplantation with aggressive management. In the particular setting of ICU, it can be reasonably assumed that Child-Pugh and MELD score have significant limitations for predicting very short-term survival. The more general Acute Physiology and Chronic Health Evaluation (APACHE) II[71] and sequential organ failure assessment (SOFA) scores[72] have been extensively validated in ICU patients. SOFA score is a relatively complex score based on respiration, coagulation, liver function, cardiovascular status, neurological status, and renal function. In cirrhotic patients admitted to the ICU, the accuracy of SOFA score was shown to be superior to that of APACHE II and Child-Pugh score with a c statistic of 0.94.[69] Although the assessment of MELD score has been limited in this context, it seems that its prognostic value is lower than that of general ICU scores.[73] Overall, general ICU scores seem to be superior to “liver-oriented” prognostic scores in this context. However, some variables which are not included in general ICU scores may have a significant impact for predicting survival and justifying resource utilization. In particular, whether multiorgan failure is the consequence of terminal illness alone or the consequence of one or more iatrogenic factors may weight heavily on the probability of recovery. Prognosis in the Particular Setting of TransplantationAs it has been clearly shown that waiting time is not an accurate marker of waiting list mortality,[27] the “sickest first” policy has been widely adopted for organ allocation, with the aim of reducing waiting list mortality. Until now, liver transplantation has been the main application for MELD score. The main reasons the MELD score has been widely adopted in this context were discussed above. Briefly, MELD score is a robust marker of early mortality in cirrhotic patients across a wide spectrum of causes. It is a continuous score based on three readily available and relatively objective variables. Patients with particular conditions such as HCC can receive extra points corresponding to a given mortality risk. MELD can be updated in each patient according to the progression of the disease. The implementation of MELD in the United States has been associated with a reduction in waiting list mortality.[30] In parallel, this “sickest first” policy has been associated with an increasing number of patients with advanced cirrhosis undergoing transplantation. Importantly, this shift in the indications for transplantation did not affect post-transplantation survival. In other words, transplanting patients with high MELD score does not necessarily translate into a significant increased post-transplant mortality, except for extreme values (over 30 to 35).[74,75] As HCC patients receive extra points, the implementation of MELD score also led to a significant decrease in the waiting list dropouts related to excessive tumor growth, without affecting post-transplant survival.[76] As shown above, MELD score proved highly efficient for prioritizing patients who are at high risk of dying without transplantation. However, an original approach consisting of comparing liver transplant recipients' survival to that of comparable candidates without transplantation offered the possibility of assessing the transplant survival benefit. This comparison showed that transplant survival benefit steadily increased with increasing MELD score.[77] A very important finding is that only patients with a MELD score exceeding 15 to 17 derive a significant benefit from transplantation. Patients with a lower MELD would have a higher risk of dying from transplantation than they have of dying from the complications of cirrhosis. Transplantation would be futile in this subgroup. However, a subset of patients with low MELD score and uncommon complications such as hepatopulmonary syndrome or mild portopulmonary hypertension are at high risk of dying in the absence of transplantation ( Table 5 ). More studies in these patients are needed for refining the assessment of the prognosis, with and without transplantation. Which Prognostic Tool to Use for Assessing the Prognosis of Cirrhosis?Child-Pugh score has been the reference for assessing the prognosis of cirrhosis for about three decades. The longevity of the Child-Pugh score can be explained by its empirical simplicity, its intuitiveness, and, overall, its good accuracy across a broad spectrum of causes and specific situations. In recent years, MELD score emerged as a “modern” alternative to Child-Pugh score. There is no clear evidence that MELD is superior to Child-Pugh score in terms of accuracy. Studies comparing these scores have shown that the accuracy of Child-Pugh score for predicting 3-month to 3-year survival is not always inferior (and is sometimes even slightly superior) to that of MELD score.[4] In addition, for many physicians, Child-Pugh score remains more convenient to use at the bedside and more explicit than MELD score. However, MELD score has several strengths compared with Child-Pugh. The variables incorporated into the MELD score are simple and more objective. The weight of each variable has been determined by statistical analysis. MELD is a continuous score, which makes it more convenient for scoring individuals within large populations. In addition to organ allocation, MELD score has been validated across a large spectrum of causes of liver diseases. All these reasons make the MELD score likely to be the core tool for assessing the prognosis of cirrhosis in the future. By using MELD score, it can be reasonably assumed that physicians will get landmarks as simple as those they had with Child-Pugh score. However, as indicated above, the outcome of cirrhosis is quite variable from patient to patient according to different causes, different stages, and different therapeutic options. In parallel, with the expansion of MELD score, several “MELD exceptions” emerged. Therefore, the quest for a universal, simple, and objective scoring system for cirrhosis is likely to remain unsuccessful. As many as 11 different scores mentioned above are available for addressing general or more specific issues regarding the prognosis of cirrhosis ( Table 2 , Table 3 , and Table 4 ). Other scores which are not mentioned or detailed in this article have been proposed. This assortment is confusing. Again, MELD score could represent a core system. However, it is not a universal prognostic marker of liver diseases. Adjustments of MELD score and/or the adjunction of additional variables are needed to address some specific issues. More studies are needed to create adapted MELD score derivates. While patients with decompensated cirrhosis are exposed to severe complications, short-term and middle-term mortality in patients with compensated cirrhosis is very low. Predicting decompensation is more relevant than predicting mortality in this group.[2] Prognostic tools for predicting decompensation would be useful. Perspectives: Beyond MELD ScoreSeveral aspects or conditions associated with chronic liver disease may be of prognostic value in addition to the MELD score. Portal hypertension, nutritional status, and glucose tolerance have long been considered in this regard and deserve further attention. Portal HypertensionIn models elaborated for predicting death, variables related with portal hypertension (size of esophageal varices and history of gastrointestinal bleeding) generally bring additional information to Child-Pugh score or its components.[2,78] Moreover, HVPG was shown to have an independent predictive value in the majority of the available multivariate analyses, including Child-Pugh score or its components.[2,79] However, various HVPG thresholds have been used across studies for their positive as well as negative predictive value. Furthermore, it is still not clear when the most predictive HVPG value should be collected, at presentation or after a few months of follow-up.[80] It was recently suggested that HVPG would be more informative in patients with compensated than with decompensated cirrhosis.[2] Indeed, HVPG, MELD, and albumin were recently identified as independent predictors of decompensation in patients with compensated cirrhosis: an HVPG < 10 mm Hg was associated with a 90% probability of not developing clinical decompensation in a median follow-up of 4 years.[81] Considerable interest recently arose from data on the prognostic value for mortality of a reduction in HVPG obtained with pharmacological agents (mainly nadolol or propranolol, with or without mononitrates). A multivariate analysis identified age, HVPG reduction by ?20% or to ?12 mm Hg, and serum albumin, not Child-Pugh score, as independent predictors for 8-year mortality.[82] In a recent systematic review, such an HVPG reduction was associated with an odds ratio of 0.39 (95%; CI 0.19 to 0.81) for mortality,[83] although it is still unclear what the added value of HVPG reduction to Child-Pugh or MELD scores would be. Further limitation in interpreting the data is the uncertainty regarding the optimal timing of the second measurement following the introduction of beta blockers. Further developments are expected as HVPG reduction in a single acute pharmacological study might prove of prognostic value for survival. Nutritional StatusVarious indices for nutritional status have been tested for their predictive value independently from Child-Pugh class.[23,84-86] Indeed, as mentioned above, the transition from Child-Turcotte classification to Child-Pugh score consisted in part in the substitution of nutritional status by prothrombin time. The results are not completely clear. As a rule, indices of poor nutrition are strongly associated with worsening Child-Pugh class. The prognostic information added to Child-Pugh class may be more marked in Child-Pugh class A and B patients.[23,84] What remains to be clarified is the optimal index for nutrition in terms of reproducibility, clinical availability, and prognostic performance. Glucose ToleranceThe relationship between diabetes and cirrhosis is difficult to interpret as diabetes can be a causal factor for, as well as a consequence of, cirrhosis. There are a limited number of studies addressing the prognostic impact of diabetes. Multivariate analyses have usually shown an independent negative effect of glucose intolerance or frank diabetes, after Child-Pugh score or its components were taken into account.[87-91] Interestingly, cause of death in diabetic patients was more frequently related to liver failure than to complications of diabetes.[87,91] Tight glycemic control was related to a better outcome in HCV-related cirrhosis, although not in HBV-related cirrhosis.[90] Statistical ModelsAs shown above, general prognostic scores based upon readily available and objective variables help predict the outcome of patients with cirrhosis at different stages. More specific prognostic scores can be even more accurate for particular causes of cirrhosis. Prognostic scores make it possible to assess the prognosis of cirrhotic patients undergoing a given therapeutic intervention. Scores also help identify patients who are the most likely to benefit from transplantation in a context of organ shortage. However, the usefulness of prognostic scores remains limited for identifying the optimal first-line option among different strategies. These scores are also limited for defining when to use different therapeutic options when the options are not exclusive one to the other and can be used in a stepwise approach. Finally, existing scores have obvious limitations for identifying which sequence of therapeutic options provides the optimal benefit in terms of survival and cost-effectiveness. Comparing different stepwise strategies with controlled studies would be especially difficult due to the large number of patients which might be involved in any comparative studies. Statistical models represent an attractive way to address these complex issues. In recent years, several studies based upon Markov models or multistate models have been published in this area.[92,93] Most studies focused on the optimal timing for transplantation or cost-effectiveness issues.[94] Several unresolved issues concerning the management of patients with cirrhosis and the overall prognosis could be investigated with multistate models. However, the strengths and limitations of statistical models require further assessment. http://www.medscape.com/viewarticle/572659?src=mp&spon=17&uac=117179DZ
Set as favorite
Bookmark
Email This
Hits: 568 Comments (0)
 Write comment
|
Subscribe to this comment's feed