Portal hypertension: Critical care considerations

Portal hypertension is not the only driver of complications of cirrhosis

Cirrhosis is typically classified as compensated or decompensated, based on the absence or presence (or previous history) of variceal bleeding, ascites, jaundice, or encephalopathy. The significantly longer survival—usually symptomless—and better quality of life experienced by patients with compensated cirrhosis compared with those with decompensated cirrhosis have brought about the concept that compensated and decompensated cirrhosis are distinct clinical stages of the disease. ^, Further disease stages have been identified according to the presence of esophageal varices and to the presence of only one or more disease complications. Four stages were proposed at the Baveno IV Consensus Conference: stage 1, compensated cirrhosis without esophageal varices; stage 2, compensated cirrhosis with varices; stage 3, ascites with or without varices; and stage 4, bleeding with or without ascites. The four stages are characterized by a significant increase in the risk of death. We went on for decades thinking that the only driver in the stage progression of cirrhosis was the progressive increase in portal pressure. In accordance with the theory of splanchnic arterial vasodilation, the progressive impairment of the cardiocirculatory function characterized by arterial vasodilatation and inadequate cardiac output, in addition to the progressive increase in the degree of activation of the endogenous vasoconstrictive and sodium-retaining systems were in fact attributed only to the progressive increase in portal pressure ^, ( Fig. 83.1 ). However, today, we can affirm that this is only partially true on the basis of the following, more recent findings.

First, it has not been confirmed in larger cohorts of patients that there is a progressive increase of the grade of activation of the endogenous vasoconstrictive and sodium-retaining systems, at least in the more advanced phases of liver disease. Second, it is possible to reduce significantly the grade of activation of those systems without achieving any clinical result in terms of reduction of the rate of cirrhosis-related complications and/or mortality. Third, during the last 8 years we have learned that another syndrome may develop in any patient with cirrhosis, but particularly in those with decompensated cirrhosis. This syndrome has been called acute on chronic liver failure (ACLF) . ACLF is characterized by an acute decompensation (AD) of the liver disease, the development of organ failure, and a 28-day mortality rate >15%. ACLF can develop even in the presence of a normal grade of activation of the endogenous vasoconstrictive and sodium-retaining systems, but never without an increased plasma level of proinflammatory cytokines in absence of systemic inflammation (SI). Taking into account that SI precedes the development of a clinically overt ascites and organ failure and that there is a strong relationship between its grade and both the stage progression of cirrhosis and the severity of ACLF, SI should be considered as another driver of decompensation in patients with cirrhosis. There is some evidence that an SI-related oxidative stress (OS) can be another potential driver of decompensation in these patients. How can we put together these three drivers of decompensation? By proposing a new pathophysiologic theory. We have recently proposed a new theory according to which the main consequence of portal hypertension is thought to be bacterial translocation (BT). BT is the main cause of SI and OS, which, in turn, are the main drivers of the derangement of the cardiovascular function and of the consequent reduction of the effective circulating volume. All these factors together are responsible for the onset of ascites and for other major complications of cirrhosis. In addition, SI and OS can generate a vicious circle whose consequence is a sequential chain of related events, including a further increase in portal pressure and/or in the severity of liver failure and thus in the extent of BT. Thus the final ring of this chain is a further increase in the degree of SI and OS. This may be the most credible, plausible, and acceptable explanation for the possibility of the onset of organ failure in patients with cirrhosis when a precipitating event cannot be detected ( Fig. 83.2 ). In an attempt to quantify, if possible, the relative weight of individual heart failure drivers in the various stages of cirrhosis, it has recently been observed that progression of portal pressure occurs mostly in patients with compensated cirrhosis, whereas SI increases substantially only across the different stages of decompensated cirrhosis. The mechanisms of the effect of SI and OS on organ integrity and function are beyond the aim of this chapter; however they are several and not mutually exclusive. The first mechanism is quite obvious and is an increase in the nitric oxide–mediated vasodilatory effect on arterial splanchnic circulation, resulting in overactivation of the endogenous vasoconstrictor systems, which elicit intense vasoconstriction and hypoperfusion of some organs, in particular, the kidney. The second mechanism is immunopathology, which is an immune-mediated tissue damage that is extremely variable in terms of pathways and impact on the different organs. The third mechanism is related to the relative storage of essential nutrients in several organs because of the fact that SI is energetically quite an expensive process, requiring reallocation of nutrients to fuel immuneactivation.

A new vision on the dynamics and classification of decompensation in patients with cirrhosis

AD has been defined as “the acute development of large ascites, hepatic encephalopathy, gastrointestinal hemorrhage, bacterial infection, or any combination of these requiring an emergent hospitalization.” However, the majority of patients with cirrhosis may develop progressive ascites or mild to moderate hepatic encephalopathy (HE) without requiring hospitalization at the first stage of decompensation. This pathway of decompensation can be defined as nonacute decompensation (NAD), and it often precedes AD. Thus focusing on ascites, we should outline that, in keeping with the new pathophysiologic theory, SI is already present in patients with grade 1 ascites, which is defined as an ascites that can be detected only by abdominal ultrasonography and that ascites of any grade, together with a high Model for End-Stage Liver Disease (MELD) score, a low mean arterial pressure (MAP), and the presence of anemia, is a predictive factor in the development of organ failure and ACLF. Thus we can state that NAD and AD represent the two main ways of transitioning from compensated to decompensated cirrhosis in clinical practice. Patients with NAD may then develop one or more episodes of AD, and when this happens, the subsequent evolution of the liver disease can follow three main trajectories. They may develop ACLF, unstable decompensated cirrhosis (UDC), or stable decompensated cirrhosis (SDC). Patients who develop ACLF have 3-month and 1-year mortality rates of 53.7% and 67.4%, respectively. Those who develop UDC, defined by the absence of organ failure but with a need of at least one hospital readmission during the following 3 months, have a 3-month and 1-year mortality of 21.0% and 35.6%, respectively. Patients with SDC, defined by the absence of both the development of ACLF and the need for hospital readmission during the 3 months after AD, have a 1-year mortality of 9.5%. Therefore this new vision of the dynamics and classification of decompensation can be summarized as shown in Fig. 83.3 . Several aspects of decompensation should be further investigated, and thus the new vision is certainly new but also equally certainly partial. In fact, the temporal relationship of the sequence from NAD to AD and the long-term outcome of patients with ACLF, UDC, and SDC are not known. In addition, we do not know the outcome of patients recovering from decompensation. We need a consensus definition on how to define recompensation. It would probably require a symptom-free interval from the previous decompensation and the ability to maintain this stage without treatment. However, this can sometimes occur as a result of effective antiviral treatment in cirrhosis patients with hepatitis B virus (HBV) or hepatitis C virus (HCV) and of abstinence in those with alcohol-related cirrhosis. If recompensation is not taken for granted for etiologic treatments, it is not even expected as a result of disease-modifying agents such as long-term use of human albumin, beta-blockers, and/or statins when exposure to the etiologic agent is still ongoing. However, these treatments may prevent further episodes of decompensation. ^,