Assessment of Sepsis-3 criteria and quick SOFA in patients with cirrhosis and bacterial infections

INTRODUCTION

Patients with cirrhosis have a high risk of sepsis, which confers a poor prognosis. The systemic inflammatory response syndrome (SIRS) criteria have several limitations in cirrhosis. Recently, new criteria for sepsis (Sepsis-3) have been suggested in the general population (increase of Sequential Organ Failure Assessment (SOFA) ≥2 points from baseline). Outside the intensive care unit (ICU), the quick SOFA (qSOFA (at least two among alteration in mental status, systolic blood pressure ≤100 mm Hg or respiratory rate ≥22/min)) was suggested to screen for sepsis. These criteria have never been evaluated in patients with cirrhosis. The aim of the study was to assess the ability of Sepsis-3 criteria in predicting in-hospital mortality in patients with cirrhosis and bacterial/fungal infections.

METHODS

259 consecutive patients with cirrhosis and bacterial/fungal infections were prospectively included. Demographic, laboratory and microbiological data were collected at diagnosis of infection. Baseline SOFA was assessed using preadmission data. Patients were followed up until death, liver transplantation or discharge. Findings were externally validated (197 patients).

RESULTS

Sepsis-3 and qSOFA had significantly greater discrimination for in-hospital mortality (area under the receiver operating characteristic (AUROC)=0.784 and 0.732, respectively) than SIRS (AUROC=0.606) (p<0.01 for both). Similar results were observed in the validation cohort. Sepsis-3 (subdistribution HR (sHR)=5.47; p=0.006), qSOFA (sHR=1.99; p=0.020), Chronic Liver Failure Consortium Acute Decompensation score (sHR=1.05; p=0.001) and C reactive protein (sHR=1.01;p=0.034) were found to be independent predictors of in-hospital mortality. Patients with Sepsis-3 had higher incidence of acute-on-chronic liver failure, septic shock and transfer to ICU than those without Sepsis-3.

CONCLUSIONS

Sepsis-3 criteria are more accurate than SIRS criteria in predicting the severity of infections in patients with cirrhosis. qSOFA is a useful bedside tool to assess risk for worse outcomes in these patients. Patients with Sepsis-3 and positive qSOFA deserve more intensive management and strict surveillance.

Incidence, predictors and outcomes of acute-on-chronic liver failure in outpatients with cirrhosis

BACKGROUND & AIMS

Acute-on-chronic liver failure (ACLF) is the most life-threatening complication of cirrhosis. Prevalence and outcomes of ACLF have recently been described in hospitalized patients with cirrhosis. However, no data is currently available on the prevalence and the risk factors of ACLF in outpatients with cirrhosis. The aim of this study was to evaluate incidence, predictors and outcomes of ACLF in a large cohort of outpatients with cirrhosis.

METHODS

A total of 466 patients with cirrhosis consecutively evaluated in the outpatient clinic of a tertiary hospital were included and followed up until death and/or liver transplantation for a mean of 45±44months. Data on development of hepatic and extrahepatic organ failures were collected during this period. ACLF was defined and graded according to the EASL-CLIF Consortium definition.

RESULTS

During the follow-up, 118 patients (25%) developed ACLF: 57 grade-1, 33 grade-2 and 28 grade-3. The probability of developing ACLF was 14%, 29%, and 41% at 1year, 5years, and 10years, respectively. In the multivariate analysis, baseline mean arterial pressure (hazard ratio [HR] 0.96; p=0.012), ascites (HR 2.53; p=0.019), model of end-stage liver disease score (HR 1.26; p<0.001) and baseline hemoglobin (HR 0.07; p=0.012) were found to be independent predictors of the development of ACLF at one year. As expected, ACLF was associated with a poor prognosis, with a 3-month probability of transplant-free survival of 56%.

CONCLUSIONS

Outpatients with cirrhosis have a high risk of developing ACLF. The degree of liver failure and circulatory dysfunction are associated with the development of ACLF, as well as low values of hemoglobin. These simple variables may help to identify patients at a high risk of developing ACLF and to plan a program of close surveillance and prevention in these patients.

LAY SUMMARY

There is a need to identify predictors of acute-on-chronic liver failure (ACLF) in patients with cirrhosis in order to identify patients at high risk of developing ACLF and to plan strategies of prevention. In this study, we identified four simple predictors of ACLF: model of end-stage liver disease (MELD) score, ascites, mean arterial pressure and hemoglobin. These variables may help to identify patients with cirrhosis, at a high risk of developing ACLF, that are candidates for new strategies of surveillance and prevention. Anemia is a potential new target for treating these patients.

Validation of a Staging System for Acute Kidney Injury in Patients With Cirrhosis and Association With Acute-on-Chronic Liver Failure

BACKGROUND & AIMS

In patients with cirrhosis of the liver, acute kidney injury (AKI) is classified into 3 stages. Recent studies indicate that there are 2 subgroups of stage 1 disease, associated with different outcomes and serum levels of creatinine (SCr): stage 1A (SCr <1.5 mg/dL) and stage 1B (SCr ≥1.5 mg/dL). We performed a prospective study to validate, in a large series of patients with cirrhosis, the association between this new description and patient outcomes, and assess the relationship between AKI stage and the presence of acute-on-chronic liver failure.

METHODS

We collected data from 547 consecutive patients admitted for cirrhosis with acute decompensation to 2 tertiary hospitals (Italy and Spain), from February 2011 through June 2015. A total of 290 patients had AKI (53%; 197 had stage 1 disease); AKI stages were determined based on levels of SCr at diagnosis. Patients were followed up until death, liver transplantation, or for 90 days. The primary outcome was 90-day survival; secondary outcomes were progression and resolution of AKI and association with acute-on-chronic liver failure.

RESULTS

Based on level of sCr at diagnosis, 58 patients had stage 1A disease and 139 had stage 1B disease. Of patients with stage 1A disease, 82% survived for 90 days; of patients with stage 1B disease, 55% survived for 90 days (P = .001). Hepatorenal syndrome and acute tubular necrosis were the most common causes of stage 1B AKI, and hypovolemia was the most common cause of stage 1A AKI. AKI progressed in a higher proportion of patients with 1B than 1A AKI (31% vs 15%; P = .017) and resolved in a higher proportion of patients with 1A disease (90% vs 52% of patients with stage 1B; P < .001). Stage 1B disease, but not 1A, was an independent predictor of AKI progression and mortality. ACLF developed in a significantly greater proportion of patients with stage 1B disease (76%) than stage 1A disease (22%; P < .001), which could account for the poor outcomes of patients with stage 1B disease.

CONCLUSIONS

In a large group of patients with decompensated cirrhosis, we validated the association between AKI stages IA and IB (based on level of sCR) with survival times and AKI progression. We also associated these subgroups of AKI with development of acute-on-chronic liver failure. These findings are important for management of patients with decompensated cirrhosis.

Terlipressin given by continuous intravenous infusion versus intravenous boluses in the treatment of hepatorenal syndrome: A randomized controlled study

In patients with cirrhosis and hepatorenal syndrome (HRS), terlipressin has been used either as continuous intravenous infusion or as intravenous boluses. To date, these two approaches have never been compared. The goal of this study was to compare the administration of terlipressin as continuous intravenous infusion versus intravenous boluses in the treatment of type 1 HRS. Seventy-eight patients were randomly assigned to receive either continuous intravenous infusion (TERLI-INF group) at the initial dose of 2 mg/day or intravenous boluses of terlipressin (TERLI-BOL group) at the initial dose of 0.5 mg every 4 hours. In case of no response, the dose was progressively increased to a final dose of 12 mg/day in both groups. Albumin was given at the same dose in both groups (1 g/kg of body weight at the first day followed by 20-40 g/day). Complete response was defined by decrease of serum creatinine (sCr) from baseline to a final value ≤133 μmol/L, partial response by a decrease ≥50% of sCr from baseline to a final value >133 μmol/L. The rate of adverse events was lower in the TERLI-INF group (35.29%) than in the TERLI-BOL group (62.16%, P < 0.025). The rate of response to treatment, including both complete and partial response, was not significantly different between the two groups (76.47% versus 64.85%; P value not significant). The mean daily effective dose of terlipressin was lower in the TERLI-INF group than in the TERLI-BOL group (2.23 ± 0.65 versus 3.51 ± 1.77 mg/day; P < 0.05).

CONCLUSION

Terlipressin given by continuous intravenous infusion is better tolerated than intravenous boluses in the treatment of type 1 HRS. Moreover, it is effective at doses lower than those required for intravenous bolus administration.

Adherence to a moderate sodium restriction diet in outpatients with cirrhosis and ascites: a real-life cross-sectional study

BACKGROUND & AIMS

A moderate sodium restriction diet should be indicated in patients with cirrhosis and ascites. Nevertheless, there is a lack of specific investigation on its correct application. To evaluate the adherence of patients with cirrhosis and ascites to a moderately low-salt diet and the impact on intake of total calories and serum sodium concentration.

METHODS

A total of 120 outpatients with cirrhosis and ascites were interviewed with a pre-established questionnaire. A quantitative assessment of nutrient and salt intake was performed.

RESULT

A moderately low-salt diet was followed by 37 patients (Group A). Of the 83 patients who did not follow the diet (Group B), 54 thought that they were following it. The mean daily sodium intake was 79.5 ± 5.5 mmol/day (Group A) and 205.9 ± 14.1 mmol/day (Group B), P < 0.0001. The adherence to diet was related to the severity of cirrhosis, and was higher among candidates for liver transplantation and in patients followed through the Care Management Program. Patients of Group A had reduced the mean daily calorie intake by 20% compared with Group B patients (P < 0.0005), while there was no difference on the occurrence of hyponatraemia.

CONCLUSIONS

This study shows a poor adherence of patients with cirrhosis and ascites to a moderate dietary sodium restriction. Adherence to a diet seems to increase with the worsening of liver disease, probably because of the reduction of alternative therapeutic options. In addition, a deficiency in the educational process can lead the patient to follow a sodium-reduced diet by means of dangerous tools, such as reducing the overall daily food intake.