Fig. 2.1
Schematic diagram of the stages of cirrhosis. The compensated patient has no ascites, variceal hemorrhage, encephalopathy or jaundice. The main stratifying factors are the presence or absence gastroesophageal varices, although the presence or absence of subclinical ascites may also be a useful stratifying factor. In patients without varices, an hepatic venous pressure gradient (HVPG) > or < 10 mmHg is the main stratifying factor. The principal mechanism in the development of decompensation is increasing portal pressure. Decompensation is defined by the presence of clinically evident events, specifically variceal hemorrhage, ascites and hepatic encephalopathy. The lowest mortality is associated with variceal hemorrhage as the initial event, followed by an isolated nonbleeding event (mostly ascites) and highest mortality with a second decompensating event. A stage of “further decompensation” occurs with worsening of the hyperdynamic circulatory state (HCS) and liver dysfunction and usually follows an acute insult (infection). The highest mortality is associated with renal failure. The number of organ failures is proportional with mortality
Different prognostic sub-stages have been identified in patients with compensated cirrhosis based on the following stratifying factors:
a.
Gastroesophageal varices
The initial mechanism leading to portal hypertension is an increase in intrahepatic vascular resistance to portal flow. One of the early consequences of portal hypertension is the formation of porto-systemic collaterals, the most important being those that form via the coronary or the short gastric veins and constitute gastroesophageal varices . Although varices are a complication of cirrhosis, they are asymptomatic (unless they rupture) and are only diagnosed by endoscopy. About a third to half of patients with compensated cirrhosis have varices when first diagnosed [10, 13]. Patients with varices (without ascites, HE or jaundice) that have not bled are still in the low-mortality compensated stage, although studies have shown that rates of mortality and evolution to decompensation are higher in these patients than in those without varices [10, 15, 16]. This has led to the classification of compensated cirrhosis into two sub-stages: stage 1 are patients with compensated cirrhosis who do not have varices (the very compensated patient) while patients with varices are designated as being at stage 2 [9, 10].
One may assume that visualization of portosystemic collaterals on imaging studies may have the same prognostic significance as the presence of endoscopically-proven gastroesophageal varices, but this remains to be determined.
b.
Clinically significant portal hypertension
As mentioned above, in patients with compensated cirrhosis without varices, the main predictor of the development of varices is the HVPG. The cutoff that best predicts variceal development is 10 mmHg. While the probability of developing varices at 2 and 5 years in patients with an HVPG ≥ 10 mmHg was 18 and 45 %, respectively, these probabilities were 7 and 30 % in those with an HVPG < 10 mmHg [13]. In fact, an HVPG ≥ 10 mmHg is also the best predictor of the development clinical decompensation [17] and HCC [18]. The probability of developing decompensation at 2 and 5 years in patients with an HVPG ≥ 10 mmHg was 13 and 29 %, respectively, while in patients with an HVPG < 10 mmHg, it was 6 and 15 %, respectively [17]. An HVPG ≥ 10 mmHg has been termed “clinically significant portal hypertension” (CSPH).
Therefore, it can be proposed that patients without varices could be stratified into those with an HVPG < 10 (without CSPH), that would be the extremely compensated patients (a redefined stage 1), and those with an HVPG ≥ 10 mmHg (with CSPH). The staging system in cirrhosis is clearly in evolution and at this point, simply describing the populations at risk (e.g., patients with compensated cirrhosis with varices or patients with compensated cirrhosis without varices and an HVPG < 10 mmHg) is recommended.
c.
Ascites detectable only by ultrasound
Patients who have ascites detectable only by ultrasound have no symptoms or signs referable to ascites and therefore are still compensated. However, a recent study evaluated the prognostic significance of subclinical ascites (n = 38) in a population of patients with predominantly alcoholic cirrhosis and compared them to patients without ascites (n = 153) and to patients with clinically-evident ascites (n = 252) [20]. Patients with subclinical ascites had a survival that was intermediate between patients with overt ascites and those without ascites. This situation would be akin that of the patient with cirrhosis (without ascites or HE) and varices that have never bled who is considered compensated because the presence of varices cannot be established by physical examination. Therefore, as for nonbleeding varices, patients with subclinical ascites should be considered compensated, albeit at a higher risk of death and clinical decompensation than those without any ascites.
d.
Portosystemic encephalopathy without liver insufficiency
Perhaps one exception to the definition of decompensation is the case of HE that presents in patients with compensated cirrhosis (no variceal hemorrhage, no ascites) and essentially normal liver synthetic function, in whom HE is the result of a large spontaneous portosystemic shunt [21]. It has been shown that patients with a MELD score less than 11 (i.e., compensated) are more likely to respond to occlusion of the spontaneous shunt, without changes in MELD score in the short-term [22]. The long-term course of patients with HE due to these shunts (and the effect of their occlusion) needs to be further evaluated to determine their prognostic significance. Until then, these (rare) patients could be considered compensated.
2.
Decompensated Stage of Cirrhosis
This is the symptomatic stage of cirrhosis and is defined by the presence of ascites, variceal hemorrhage , HE or liver insufficiency (jaundice). The main pathogenic mechanisms are portal hypertension and the hyperdynamic circulatory state [23]. This hemodynamic abnormality is the result of splanchnic and systemic vasodilatation that increases as HVPG surpasses 10 mmHg and portosystemic collaterals develop. The vasodilatation (manifested clinically as arterial hypotension) leads to activation of the neurohumoral systems, sodium and water retention , increased blood volume and increased cardiac output, that is, a hyperdynamic circulatory state.
Of the decompensating events, overt ascites is clearly the most common, accounting for 60–80 % of initial clinical events, followed by gastrointestinal hemorrhage, while HE and jaundice occur as the first clinical event in only a minority of patients [10, 24].
Sub-staging of patients with decompensated cirrhosis is not as well-defined as compensated cirrhosis and requires further investigation. The following are different proposed prognostic sub-stages based on the following stratifying factors:
a.
Type and number of decompensating clinical events
Even though each of the individual complications of cirrhosis has an impact on survival in patients with cirrhosis, the magnitude of the impact is different. The Baveno IV consensus conference, based on results from a large Italian cohort, had stratified patients with decompensated cirrhosis into two sub-stages based on the type of initial decompensating event: (1) patients with ascites with or without varices (stage 3) and (2) patients with gastrointestinal bleeding with or without ascites (stage 4) [25]. However, it was shown in another cohort that decompensated patients with ascites have a significantly poorer outcome than those presenting with variceal hemorrhage as the only decompensating event [16]. This led to a re-staging of cirrhosis, based on an Italian prospective inception cohort study of 464 patients in which patient flow across stages was assessed by competing risk analysis [10]. In this re-staging, decompensated patients would be placed in three strata: (1) bleeding without other complications; (2) first nonbleeding decompensation (mainly ascites); and (3) patients with any second decompensating event [10]. Five-year mortality rates for each of these three stages was 20, 30, and 88 %, respectively. The mortality rate difference between patients who present with variceal hemorrhage (no other complication) and those that presented with one nonbleeding complication was not large, similar to findings in another cohort followed for a median of 33 months in which a poor outcome (death or LT) was 20 % in patients with variceal hemorrhage and 36 % in those with ascites [24]. It is not unexpected that patients that develop more than one complication have the highest mortality. The higher mortality in the different sub-stages was confirmed in a retrospective study of patients on a transplant list with a MELD score < 20 that combined patients with compensated and decompensated cirrhosis [26].
b.
Complications of the initial complication or “further” decompensation
Patients who die of decompensated cirrhosis often do so after development of “further decompensation”—worsening of the pathophysiological mechanisms (portal hypertension, hyperdynamic circulatory state and/or liver insufficiency) lead to a subsequent complication after the initial event. Specifically, patients with ascites would develop diuretic-refractory ascites, hyponatremia or hepatorenal syndrome (HRS) as a result of worsening vasodilatation (and decreasing mean arterial pressure) and activation of neurohumoral systems [27]; patients with variceal hemorrhage would develop recurrent variceal hemorrhage as a result of worsening portal pressure and/or worsening of the hyperdynamic circulatory state [28–30]; and patients would develop recurrent/persistent HE, coagulopathy and jaundice as a result of further impairment in liver function. The development of these added decompensating events may have a trigger that is not clinically evident (e.g., overt bacterial infection versus. bacterial translocation) . Bacterial infections occur in both compensated and decompensated cirrhosis and are a frequent precipitant for acute decompensation (see below) and therefore do not represent a separate stage.
There is evidence demonstrating that refractory ascites has a higher mortality than diuretic-responsive ascites [31], that the presence of hyponatremia is associated with a significantly poorer survival in patients on the liver transplant waiting list, independent of MELD score [32] and that HRS type 1 (acute renal failure in cirrhosis) has a higher mortality than HRS type 2 (renal failure associated mostly with refractory ascites), which in turn has a higher mortality than refractory ascites [33]. In fact, while the median survival in compensated cirrhosis is greater than 12 years (as long as the patient remains compensated), the median survival in decompensated cirrhosis , refractory ascites and in patients with untreated HRS type 1 is approximately 2 years, 7 months and 1 month, respectively. Therefore, another way to stratify patients with decompensated cirrhosis would be to divide them into those who are decompensated by virtue of the development of ascites, variceal hemorrhage or HE and those that have other complications that denote a more advanced liver disease: refractory ascites, hyponatremia, HRS, recurrent/persistent HE, and jaundice.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
