Influence of further decompensation on survival across clinical stages of decompensated cirrhosis: The role of portal hypertension and HVPG changes

BACKGROUND AND AIMS

Decompensated-cirrhosis encompasses several stages with different prognosis, such as bleeding, ascites and bleeding-plus-ascites. Development of further-decompensation worsens survival, while non-selective β-blockers (NSBBs) can modify the risk. However, how this applies to each stage is uncertain. We aimed to investigate, in each stage of decompensated-cirrhosis, the influence of further-decompensation on mortality and whether changes in portal-pressure (HVPG) under NSBBs influence these outcomes.

METHODS

Patients with variceal bleeding were consecutively included differentiating those with bleeding-alone from those who also had ascites. Patients with ascites and high-risk varices referred for primary-prophylaxis were also investigated. A baseline haemodynamic study was performed and was repeated after 1-3-months under NSBBs. Outcomes were investigated by competing-risk.

RESULTS

Totally 103 patients had bleeding-alone, 186 bleeding-plus-ascites and 187 ascites-alone. Mean follow-up was 32-months (IQR, 12-60). Patients with bleeding-plus-ascites had higher HVPG and were more hyperdynamic than patients with ascites-alone and these than those with bleeding-alone. At each stage, the mortality risk was more than twice in patients developing further-decompensation vs. those without (p < .001). In each stage, HVPG-decrease under NSBBs showed better discrimination to predict further-decompensation than the baseline MELD, Child-Pugh or HVPG, by time-dependent ROC-curves (c-statistic >70%). At each stage, patients without HVPG-decreases, either ≥10% or ≥20% from the baseline, had higher risk of further-decompensation (sHR from 2.43 to 6.73, p < .01) and worse survival.

CONCLUSIONS

In each stage of decompensated cirrhosis, mortality risk significantly and very markedly increase with further-decompensation. HVPG-non-response to NSBBs may adequately stratify the risk of further decompensation and death, in each stage. This suggests potential benefit with pre-emptive therapies in HVPG-non-responders at each-stage.

Impact of the COVID-19 pandemic on the incidence and type of infections in hospitalized patients with cirrhosis: a retrospective study

Infections are a major cause of morbidity and mortality in cirrhosis, especially those caused by multi-drug resistant bacteria. During the COVID-19 pandemic, the incidence and type of infection in these patients may have been influenced by the restrictive measures implemented. We aimed to compare the infections in patients with cirrhosis hospitalized before the COVID-19 pandemic versus those hospitalized during the pandemic. We retrospectively compared infections in patients with cirrhosis hospitalized in the hepatology unit during the pre-pandemic period (3/2019-2/2020) with infections in patients hospitalized during the pandemic (3/2020-2/2021). Baseline characteristics, type of infections, type of bacteria, antimicrobial resistance and mortality were evaluated. There were 251 hospitalizations in 170 patients during the pre-pandemic period and 169 hospitalizations in 114 patients during the pandemic period. One or more infections were identified in 40.6% of hospitalizations during the pre-pandemic period and 43.8% of hospitalizations during the pandemic, P = 0.52. We found 131 infections in the pre-pandemic period and 75 infections during the pandemic. The percentage of nosocomial infections decreased in the pandemic period (25.3% vs. 37.4% in the pre-pandemic period, P = 0.06). We found a non-significant trend to a higher incidence of infections by multi-drug resistant organisms (MDRO) in the pandemic period than in the pre-pandemic period (6.5% vs. 4%). The incidence of infections was similar in both periods. However, during the pandemic, we observed a trend to a lower incidence of nosocomial infections with a higher incidence of MDRO infections.

Improving primary prophylaxis of variceal bleeding by adapting therapy to the clinical stage of cirrhosis. A competing-risk meta-analysis of individual participant data

BACKGROUND & AIMS

Non-selective β-blockers (NSBBs) and endoscopic variceal-ligation (EVL) have similar efficacy preventing first variceal bleeding. Compensated and decompensated cirrhosis are markedly different stages, which may impact treatment outcomes. We aimed to assess the efficacy of NSBBs vs EVL on survival in patients with high-risk varices without previous bleeding, stratifying risk according to compensated/decompensated stage of cirrhosis.

METHODS

By systematic review, we identified RCTs comparing NSBBs vs EVL, in monotherapy or combined, for primary bleeding prevention. We performed a competing-risk, time-to-event meta-analysis, using individual patient data (IPD) obtained from principal investigators of RCTs. Analyses were stratified according to previous decompensation of cirrhosis.

RESULTS

Of 25 RCTs eligible, 14 failed to provide IPD and 11 were included, comprising 1400 patients (656 compensated, 744 decompensated), treated with NSBBs (N = 625), EVL (N = 546) or NSBB+EVL (N = 229). Baseline characteristics were similar between groups. Overall, mortality risk was similar with EVL vs. NSBBs (subdistribution hazard-ratio (sHR) = 1.05, 95% CI = 0.75-1.49) and with EVL + NSBBs vs either monotherapy, with low heterogeneity (I2  = 28.7%). In compensated patients, mortality risk was higher with EVL vs NSBBs (sHR = 1.76, 95% CI = 1.11-2.77) and not significantly lower with NSBBs+EVL vs NSBBs, without heterogeneity (I2  = 0%). In decompensated patients, mortality risk was similar with EVL vs. NSBBs and with NSBBs+EVL vs. either monotherapy.

CONCLUSIONS

In patients with compensated cirrhosis and high-risk varices on primary prophylaxis, NSBBs significantly improved survival vs EVL, with no additional benefit noted adding EVL to NSBBs. In decompensated patients, survival was similar with both therapies. The study suggests that NSBBs are preferable when advising preventive therapy in compensated patients.

Assessment of portal hypertension severity using machine learning models in patients with compensated cirrhosis

BACKGROUND & AIMS

In individuals with compensated advanced chronic liver disease (cACLD), the severity of portal hypertension (PH) determines the risk of decompensation. Invasive measurement of the hepatic venous pressure gradient (HVPG) is the diagnostic gold standard for PH. We evaluated the utility of machine learning models (MLMs) based on standard laboratory parameters to predict the severity of PH in individuals with cACLD.

METHODS

A detailed laboratory workup of individuals with cACLD recruited from the Vienna cohort (NCT03267615) was utilised to predict clinically significant portal hypertension (CSPH, i.e., HVPG ≥10 mmHg) and severe PH (i.e., HVPG ≥16 mmHg). The MLMs were then evaluated in individual external datasets and optimised in the merged cohort.

RESULTS

Among 1,232 participants with cACLD, the prevalence of CSPH/severe PH was similar in the Vienna (n = 163, 67.4%/35.0%) and validation (n = 1,069, 70.3%/34.7%) cohorts. The MLMs were based on 3 (3P: platelet count, bilirubin, international normalised ratio) or 5 (5P: +cholinesterase, +gamma-glutamyl transferase, +activated partial thromboplastin time replacing international normalised ratio) laboratory parameters. The MLMs performed robustly in the Vienna cohort. 5P-MLM had the best AUCs for CSPH (0.813) and severe PH (0.887) and compared favourably to liver stiffness measurement (AUC: 0.808). Their performance in external validation datasets was heterogeneous (AUCs: 0.589-0.887). Training on the merged cohort optimised model performance for CSPH (AUCs for 3P and 5P: 0.775 and 0.789, respectively) and severe PH (0.737 and 0.828, respectively).

CONCLUSIONS

Internally trained MLMs reliably predicted PH severity in the Vienna cACLD cohort but exhibited heterogeneous results on external validation. The proposed 3P/5P online tool can reliably identify individuals with CSPH or severe PH, who are thus at risk of hepatic decompensation.

IMPACT AND IMPLICATIONS

We used machine learning models based on widely available laboratory parameters to develop a non-invasive model to predict the severity of portal hypertension in individuals with compensated cirrhosis, who currently require invasive measurement of hepatic venous pressure gradient. We validated our findings in a large multicentre cohort of individuals with advanced chronic liver disease (cACLD) of any cause. Finally, we provide a readily available online calculator, based on 3 (platelet count, bilirubin, international normalised ratio) or 5 (platelet count, bilirubin, activated partial thromboplastin time, gamma-glutamyltransferase, choline-esterase) widely available laboratory parameters, that clinicians can use to predict the likelihood of their patients with cACLD having clinically significant or severe portal hypertension.

Carvedilol reduces the risk of decompensation and mortality in patients with compensated cirrhosis in a competing-risk meta-analysis

BACKGROUND & AIMS

Whether non-selective β-blockers can prevent decompensation of cirrhosis warrants clarification. Carvedilol might be particularly effective since its intrinsic vasodilatory activity may ameliorate hepatic vascular resistance, a major mechanism of portal hypertension in early cirrhosis. We assessed whether carvedilol may prevent decompensation and improve survival in patients with compensated cirrhosis and clinically significant portal hypertension (CSPH).

METHODS

By systematic review we identified randomized-controlled trials (RCTs) comparing carvedilol vs. control therapy (no-active treatment or endoscopic variceal ligation [EVL]) in patients with cirrhosis and CSPH without previous bleeding. We performed a competing-risk time-to-event meta-analysis using individual patient data (IPD) obtained from principal investigators of RCTs. Only compensated patients were included. Primary outcomes were prevention of decompensation (liver transplantation and death were competing events) and death (liver transplantation was a competing event). Models were adjusted using propensity scores for baseline covariates with the inverse probability of treatment weighting (IPTW) approach.

RESULTS

Among 125 full-text studies evaluated, 4 RCTs were eligible. The 4 provided IPD and were included, comprising 352 patients with compensated cirrhosis, 181 treated with carvedilol and 171 controls (79 received EVL and 92 placebo). Baseline characteristics were similar between groups. Standardized differences were <10% by IPTW. The risk of developing decompensation of cirrhosis was lower with carvedilol than in controls (subdistribution hazard ratio [SHR] 0.506; 95% CI 0.289-0.887; p = 0.017; I2 = 0.0%, Q-statistic-p = 0.880), mainly due to a reduced risk of ascites (SHR 0.491; 95% CI 0.247-0.974; p = 0.042; I2 = 0.0%, Q-statistic-p = 0.384). The risk of death was also lower with carvedilol (SHR 0.417; 95% CI 0.194-0.896; p = 0.025; I2 = 0.0%, Q-statistic-p = 0.989).

CONCLUSIONS

Long-term carvedilol therapy reduced decompensation of cirrhosis and significantly improved survival in compensated patients with CSPH. This suggests that screening patients with compensated cirrhosis for CSPH to enable the prompt initiation of carvedilol could improve outcomes.

PROSPERO REGISTRATION NUMBER

CRD42019144786.

LAY SUMMARY

The transition from compensated cirrhosis to decompensated cirrhosis is associated with markedly reduced life expectancy. Therefore, preventing decompensation in patients with compensated cirrhosis would be associated with greatly improved patient outcomes. There has been controversy regarding the use of non-selective β-blockers (portal pressure-lowering medications) in patients with cirrhosis and elevated portal blood pressure (portal hypertension). Herein, using a competing-risk meta-analysis to optimize sample size and properly investigate cirrhosis as a multistate disease and outcomes as time-dependent events, we show that carvedilol (a non-selective β-blocker) is associated with a reduced risk of decompensating events and improved survival in patients with cirrhosis and portal hypertension.

Bacterial infections adversely influence the risk of decompensation and survival in compensated cirrhosis

BACKGROUND & AIMS

The prognosis of compensated cirrhosis is good until decompensation. In decompensated cirrhosis, bacterial infections (BIs) are common and increase the risk of death. The incidence and prognostic implications of BIs in compensated cirrhosis are less-well characterized. This study aimed to assess whether BIs influence the risk of decompensation and survival in patients with compensated cirrhosis.

METHODS

This is a cohort study nested to the PREDESCI study, a double-blind, multicenter, randomized controlled trial designed to assess whether β-blockers could prevent decompensation of cirrhosis. Patients with compensated cirrhosis and hepatic venous pressure gradient ≥10 mmHg were included. Development of BIs during follow-up was prospectively registered. Using a competing-risk time-dependent regression analysis, we investigated whether BIs affect the risk of decompensation and survival. Decompensation was defined as development of ascites, bleeding or overt encephalopathy.

RESULTS

A total of 201 patients were randomized and followed for a median of 36 months (IQR 24-47 months); 34 patients (17%) developed BIs, which occurred before decompensation in 33 cases, and 29 (14%) developed ascites. Respiratory and urinary tract infections were the most frequent BIs. Decompensation occurred in 26% patients with BIs vs. 16% without BIs. Patients with BIs were at higher risk of decompensation (subdistribution hazard ratio [SHR] 2.93; 95% CI 1.02-8.42; p = 0.047) and of developing ascites (SHR 3.55; 95% CI 1.21-10.47; p = 0.022) than those without BIs. Risk of death was also higher in patients with BIs (subdistribution HR 6.93; 95% CI 2.64-18.18; p <0.001), although decompensation occurred before death in 71% of such cases.

CONCLUSIONS

BIs have a marked impact on the natural history of compensated cirrhosis, significantly increasing the risk of decompensation, mainly that of ascites, and increasing the risk of death, which usually occurs after decompensation. Our results suggest that BIs may constitute a target to prevent decompensation.

LAY SUMMARY

It is widely known that bacterial infections are common and increase the mortality risk in patients with decompensated cirrhosis. However, the relevance of bacterial infections in compensated cirrhosis has not been well studied. This study shows that in patients with compensated cirrhosis and clinically significant portal hypertension, bacterial infections occur as frequently as the development of ascites, which is the most frequent decompensating event. Bacterial infections increase the risk of progression to decompensation, mainly by increasing the risk of ascites, and also increase the risk of death, which usually occurs after decompensation. CLINICALTRIALS.

GOV IDENTIFIER

NCT01059396.

Short-term hemodynamic effects of β-blockers influence survival of patients with decompensated cirrhosis

BACKGROUND & AIMS

Whether the effect of β-blockers on arterial pressure and/or cardiac function may offset the benefit of reducing portal pressure in advanced cirrhosis is controversial. Herein, we aimed to evaluate the systemic and splanchnic hemodynamic effects of β-blockers in decompensated vs. compensated cirrhosis and to investigate the influence of systemic hemodynamic changes on survival times in decompensated cirrhosis.

METHODS

Patients with cirrhosis and high-risk esophageal varices, without previous bleeding, were consecutively included and grouped according to the presence or absence of decompensation (ascites with or without overt encephalopathy). Systemic and hepatic hemodynamic measurements were performed before starting β-blockers and again after 1 to 3 months of treatment (short-term).

RESULTS

Four hundred and three patients were included (190 decompensated and 213 compensated). At baseline, decompensated patients had higher portal pressure than compensated patients and were more hyperdynamic, with higher cardiac output (CO) and lower arterial pressure. Under β-blockers, decompensated patients had lower portal pressure decrease (10 ± 18% vs. 15 ± 12%; p <0.05) and had greater reductions in heart rate (p <0.001) and CO (17 ± 15% vs. 10 ± 21%; p <0.01). Among patients with decompensated cirrhosis, those who died had a greater decrease in CO with β-blockers than survivors (21 ± 14% vs. 15 ± 16%; p <0.05) and CO under β-blockers independently predicted death by competing-risk regression analysis, with good diagnostic accuracy (C-index 0.74; 95% CI 0.66-0.83). Death risk was higher in decompensated patients with CO <5 L/min vs. CO ≥5 L/min (subdistribution hazard ratio 0.44; 95% CI 0.25-0.77; p = 0.004).

CONCLUSIONS

In patients with high-risk varices treated to prevent first bleeding, the systemic hemodynamic response to β-blockers is greater and the portal pressure decrease is smaller in those with decompensated cirrhosis. The short-term effect of β-blockers on CO might adversely influence survival in decompensated cirrhosis.

LAY SUMMARY

β-blockers are often used to reduce the risk of variceal bleeding in patients with cirrhosis. However, it is not known whether the effect of β-blockers on arterial pressure and/or cardiac function may offset the benefit of reducing portal pressure. Herein, we show that in patients with decompensated cirrhosis the potentially detrimental systemic effects of β-blockers are greater than in compensated patients, while the beneficial pressure lowering effects are reduced. The short-term effect of β-blockers on cardiac output may adversely influence survival in patients with decompensated cirrhosis.

Early rebleeding increases mortality of variecal bleeders on secondary prophylaxis with β-blockers and ligation

BACKGROUND/AIMS

Despite secondary-prophylaxis with β-blockers and endoscopic-variceal-ligation rebleeding is frequent, particularly within the first-6-weeks. Early-rebleeding may have greater impact on death-risk than late rebleeding, which may affect therapy. We assessed whether the influence of rebleeding on long-term survival of patients on secondary-prophylaxis is greater in patients with early-rebleeding.

METHODS

369 patients with cirrhosis were consecutively included once recovered from first variceal-bleeding. The impact of rebleeding on survival was investigated according to whether it occurred within 6-weeks (early-rebleeding) or later (late-rebleeding).

RESULTS

During 46-months of follow-up (IQR: 14-61), 45 patients (12%) had early-rebleeding, 74(20%) had late-rebleeding and 250(68%) had not rebleeding. Mortality risk was higher in early-rebleeding group vs. late-rebleeding (HR = 0.476, 95%CI = 0.318-0.712, p < 0.001) and was similar in late-rebleeding group vs. no-rebleeding (HR = 0.902, 95%CI = 0.749-1.086, p = 0.271). Adjusting for baseline risk-factors, early-rebleeding was independently associated with mortality-risk (HR = 1.58, 95%CI = 1.02-2.45; p = 0.04). Child-Pugh&MELD scores improved at 3rd-4th-week only in patients without early-rebleeding (p < 0.05). Presence of ascites or encephalopathy, MELD-score>12 and HVPG>20 mmHg identified patients at risk of early-rebleeding.

CONCLUSIONS

Patients with early-rebleeding have higher risk of death than patients without rebleeding and even than those rebleeding later. Our results suggest that patients at risk of early rebleeding might benefit from preemptive therapies such as early-TIPS.