Infection In Decompensated Cirrhosis Research Articles

Infections in decompensated cirrhosis: Pathophysiology, management, and research agenda.

Bacterial infections in patients with cirrhosis lead to a 4-fold increase in mortality. Immune dysfunction in cirrhosis further increases the risk of bacterial infections, in addition to alterations in the gut microbiome, which increase the risk of pathogenic bacteria. High rates of empiric antibiotic use contribute to increased incidence of multidrug-resistant organisms and further increases in mortality. Despite continous advances in the field, major unknowns regarding interactions between the immune system and the gut microbiome and strategies to reduce infection risk and improve mortality deserve further investigation. Here, we highlight the unknowns in these major research areas and make a proposal for a research agenda to move toward improving disease progression and outcomes in patients with cirrhosis and infections.

Incidence, risk factors and clinical outcomes of multidrug-resistant microorganism infections among patients admitted for decompensated cirrhosis: A prospective study.

Bacterial infections remain one of the main complications in cirrhosis and worsen patients' prognosis and quality of life. An increase in multidrug resistant microorganism (MDRM) infections among patients with cirrhosis, together with infection-related mortality rates, have been reported in recent years. Therefore, adaptation of the initial empiric antibiotic approach to different factors, particularly the local epidemiology of MDRM infections, has been recommended. We aim to describe the main features, outcomes and risk factors of MDRM infections in patients with cirrhosis. Prospective registry of all episodes of in-hospital infections occurring among cirrhotic patients admitted within a 2-year period at a single center. Clinical and microbiological data were collected at the time of infection diagnosis, and the in-hospital mortality rate of the infectious episode was registered. A total of 139 infectious episodes were included. The disease-causing microorganism was identified in 90 episodes (65%), of which 31 (22%) were caused by MDRM. The only two factors independently associated with MDRM infections were rectal colonization by MDRM and a nosocomial or healthcare-associated source. The infection-related mortality rate was 18.7%. MDRM infection and a past history of hepatic encephalopathy were independently associated with in-hospital mortality. Almost one fourth of bacterial infections occurring in admitted cirrhotic patients were due to MDRM. Rectal colonization was the most important risk factor for MDRM infections in decompensated cirrhosis. Screening for MDRM rectal colonization in patients admitted for decompensated cirrhosis should be assessed as a tool to improve local empiric antibiotic strategies.

O03 Rapid Infection Diagnostics in Cirrhosis (RIDiC): a novel molecular-based approach to address the ever increasing challenge of antimicrobial resistant infections in decompensated cirrhosis and acute-on-chronic liver failure

Abstract Background Global prevalence of MDR infections in cirrhotic patients has risen to 34% and associates with worsening clinical trajectory and mortality (Fernandez J et al. 2019). A major unmet need in addressing the challenge of antimicrobial resistance (AMR) in cirrhosis is rapid pathogen diagnosis and antimicrobial resistance (AMR) (Patel VC et al. 2019). These can facilitate selection of more effective antibiotic therapy, reducing selection for MDR organisms. The Curetis Unyvero™ platform (CUp) employs multiplex PCR technology for pathogen and AMR gene detection from multiple bodily tissues and fluids requiring minimal pre- processing. The investigator-initiated pilot study aimed to assess the utility of this platform for hospitalized cirrhotic patients. Methods Ascitic fluid, bronchoalveolar lavage (BAL) and blood culture samples were obtained from hospitalized cirrhotic patients with acute decompensation (AD) or acute-on-chronic liver failure (ACLF), in whom there was clinical suspicion of infection. Samples were tested using the IAI (intra-abdominal infection) panel, HPN (hospital pneumonia) panel and BCU (blood culture) panel based on sample type available. Results were not shared with clinicians for decision-making. Data were compared with standard microbiology diagnostics. Results 31 samples from 21 patients were tested: ascites (25), BAL (4), blood culture (2). Time to positive culture result of all samples was reduced with CUp compared with standard microbiology (4.59 versus 73.5 h). CUp showed 100% specificity (12/12) and detected the same organism in 100% of cases (10/12). In 6 samples, CUp identified pathogens which were consistent with clinical evidence of spontaneous bacterial peritonitis and previous positive isolates which standard microbiological diagnostics failed to identify. AMR genes identified by the CUp correlated with AMR patterns reported by standard diagnostics in 100% (5/5) of cases. In 2 patients, broad-spectrum antibiotic therapy continued although the CUp supported infection resolution. Four patients in whom CUp screening was negative continued to receive antibiotics whilst awaiting negative standard cultures that arrived 72 h later. CUp did not identify an Aspergillus species in one BAL sample as this is not part of the BAL panel. Conclusions These preliminary data suggest CUp results are compatible with standard laboratory diagnostics. The CUp may be utilized for earlier infection diagnosis and improve the targeting and de-escalation of antimicrobial therapy in AD and ACLF. These are crucial components in enhancing antibiotic stewardship strategies and to reduce MDR infections (Figure 1). Further work is however required to evaluate cost implications, validation and impact on clinical outcomes and mortality in cirrhosis.

S2776 A Rare Case of Pseudomonas Aeruginosa Colitis and Subsequent Bacteremia in the Setting of Alcohol-Induced Cirrhosis

Introduction: Pseudomonas aeruginosa is an opportunistic bacterium that commonly causes pneumonia and urinary tract infections. While cases of P. aeruginosa colitis have been infrequently reported, we identify a unique case of P. aeruginosa bacteremia occurring after antibiotic-treated P. aeruginosa colitis in a patient with alcohol-induced cirrhosis. Case Description/Methods: A 38-year-old male with class III obesity and alcohol use disorder presented with cramping abdominal pain, vomiting and dark stool. There was no recent hospitalization and no antibiotic or intravenous drug use. Last alcoholic drink was reported to be several days prior to presentation. Physical exam identified scleral icterus, abdominal distension without tenderness, and hepatomegaly. Laboratory test results are outlined in Table 1. Serum acetaminophen and alcohol levels were not detected. Urinalysis and chest x-ray were unremarkable. Ultrasound demonstrated cirrhotic liver morphology and ascites, but no cholelithiasis. Ceftriaxone was started due to concerns for an upper gastrointestinal bleed. On day 2, he developed watery diarrhea. Stool cultures were positive for P. aeruginosa, prompting antibiotic transition to cefepime. Clostridioides difficile testing was negative. He completed a five-day course of cefepime with resolution of diarrhea and leukocytosis. Subsequently, he was started on methylprednisolone for treatment of alcoholic hepatitis, and later discharged on a tapering dose. Eleven days later, he returned with vomiting and generalized weakness, and was admitted to the intensive care unit with pansusceptible P. aeruginosa bacteremia and septic shock. Unfortunately, clinical condition continued to deteriorate despite maximal medical therapy, and he ultimately expired. Discussion: P. aeruginosa has rarely been described in association with colitis. To our knowledge, this is the first reported case of P. aeruginosa colitis in a patient with alcohol-induced cirrhosis, and the first case of P. aeruginosa bacteremia after previously treated colitis. While the optimal duration of treatment for P. aeruginosa colitis is not well defined, this case supports extended antibiotic courses for the treatment of P. aeruginosa infections in decompensated cirrhosis and cautious post-treatment steroid utilization.Table 1.: This table depicts initial admission laboratory testing results for the patient’s two separate admissions.

Biomarkers in Liver Disease: From Diagnosis to Prognosis

A study published in Afro-Egypt J Infect Endem Dis under the title of Serum MiRNA-122 as a Diagnostic Marker in HCV Related Liver Cirrhosis aimed to evaluate serum miRNA-122 expression as a potential biomarker for diagnosis and monitoring different stages of disease in chronic hepatitis C patients. The study revealed that miRNA-122 expression levels were significantly increased among liver disease patients than healthy controls. Furthermore, miRNA-122 expression levels were significantly higher in the compensated patients compared to the decompensated patients. Also, this study elucidated that serum miRNA-122 levels were decreased with the progression of liver disease (from Child-Pugh class A to C) but without reaching to a significant difference. According to an interesting study published in the current issue of the Afro-Egyptian Journal of Infectious and Endemic Disease Presepsin and Resistin as Diagnostic Markers for Bacterial Infection in Patients with Decompensated CirrhosisPresepsin and resistin were significantly higher among patients with infection and positively correlated with Model for End-stage Liver Disease score (MELD), Child-pough score (CPS), CRP and PCT. At 1205 pg/ml as cutoff, Presepsin could predict infection at sensitivity 83.8%, specificity 93% and accuracy 88.7%. While using 21ng/ml as cutoff, Resistin could predict infection at sensitivity 64.6%, specificity 68.4% and accuracy 66.7%. Adding CRP to PCT or presepsin increased sensitivity to 99%, specificity 73.7%, and accuracy 85.4%. Adding presepsin to PCT or resistin increased sensitivity to 94.9%. Yet combined presepsin and PCT had higher specificity than combined presepsin and resistin. Conclusion: Presepsin has comparable diagnostic performances to CRP and PCT for bacterial infection in decompensated cirrhosis while resistin has poor sensitivity and specificity. Adding presepsin to CRP yields the same diagnostic performance as combined CRP and PCT. So, combining any of them to CRP helps to early diagnose bacterial infection in those patients

Short-term mortality in patients with cirrhosis of the liver and acute kidney injury: A prospective observational study.

Renal failure is a common and severe complication of cirrhosis and confers poor prognosis. Serum creatinine is the most practical biomarker of renal function. Serum creatinine estimation in cirrhosis of the liver is affected by decreased formation, increased tubular secretion, increased volume of distribution, and interference by elevated bilirubin. Studies on the prognosis of cirrhotic patients using creatinine kinetics as a definition of acute kidney injury (AKI) proposed by the International Ascites Club are limited. In this single-center prospective observational study, decompensated cirrhotics with AKI defined by the International Ascites Club as the rise of serum creatinine ≥ 0.3mg/dL within 48h of admission or increase of serum creatinine ≥ 50% from stable baseline creatinine over the previous 3months were followed and assessed for the development of complications during hospital course and in-hospital and 30-day mortality. AKI developed in 142 out of 499 (28.45%) patients with cirrhosis. Twenty patients were excluded. The most common etiology of cirrhosis was alcohol (n = 64, 52%), and ascites was present in 115 (94%) patients. Eighty-two (67.21%) patients presented with AKI at the time of admission. Thirty-day mortality was 46.72% (57/122 patients). Hepatorenal syndrome had the highest mortality followed by AKI related to infection. Presence of jaundice and hepatic encephalopathy (HE) was associated with poor survival with adjusted hazard ratio of 3.54 and 2.17, respectively. On bivariate logistic regression analysis, jaundice, HE, type of AKI, AKI stage at maximum creatinine, bilirubin, serum glutamic oxaloacetic transaminase (SGOT), international normalized ratio (INR), and Child-Turcotte-Pugh (CTP) and model for end-stage liver disease (MELD) scores were predictors of mortality (p < 0.05). Sensitivity, specificity, and accuracy of MELD > 29 and CTP score > 11 were 75.44%, 82%, and 78.70% and 66.67%, 81.54%, and 74.60%, respectively for predicting 30-day mortality. Development of AKI as defined by the International Ascites Club in cirrhosis confers high short-term mortality. Jaundice, HE, AKI stage, creatinine at enrollment, bilirubin, CTP, and MELD score were the predictors of mortality. Bullet points of the study highlights What is already known? • Renal failure is a common and severe complication of cirrhosis. • Serum creatinine is the most practical biomarker of renal function but it has many limitations in cirrhotic patients. • Creatinine kinetics-based definition of acute kidney injury (AKI) was proposed by the International Ascites Club. What is new in this study? • Short-term mortality (30days) in decompensated cirrhotic patients with AKI as defined by the International Ascites Club using creatinine kinetics was high. • AKI due to hepatorenal syndrome (HRS) has the highest short-term (30days) mortality followed by AKI due to infection in decompensated cirrhosis. • Detection of AKI using creatinine kinetics-based definition mayprompt anearly appropriate intervention. What are the future clinical and research implications of study findings? • Creatinine kinetics-based definition of AKI diagnose renal injury at anearlier stage; anappropriate intervention should be initiated at the earliestin these patients to improve patient survival.

Presepsin and Resistin as Diagnostic Markers for Bacterial Infection in Patients with Decompensated Cirrhosis

Background and study aim: Bacterial infections mess up prognosis of cirrhotic patients. Presepsin and resistin are favorable infection markers that can help in diagnosis of such condition. This study aimed to assess performance of presepsin and resistin in diagnosis of infection compared with C reactive protein (CRP) and procalcitonin (PCT) among patients with decompensated cirrhosis. Patients and Methods: Two hundred and thirteen patients with decompensated cirrhosis admitted to Internal Medicine hospital, Zagazig University, were included in this study. All patients underwent history taking, thorough clinical examination and laboratory investigations including measuring CRP, PCT, presepsin and resistin. Results: About 47% of patients have infections. Presepsin and resistin were significantly higher among patients with infection and positively correlated with Model for End-stage Liver Disease score (MELD), Child-pough score (CPS), CRP and PCT. Presepsin cutoff≥1205 pg/ml could predict infection at sensitivity 83.8%, specificity 93% and accuracy 88.7%. Resistin cutoff≥21 ng/ml could predict infection at sensitivity 64.6%, specificity 68.4% and accuracy 66.7%. Adding CRP to PCT or presepsin increased sensitivity to 99%, specificity 73.7%, and accuracy 85.4%. Adding presepsin to PCT or resistin increased sensitivity to 94.9%. Yet combined presepsin and PCT had higher specificity than combined presepsin and resistin. Conclusion: Presepsin has comparable diagnostic performances to CRP and PCT for bacterial infection in decompensated cirrhosis while resistin has poor sensitivity and specificity. Adding presepsin to CRP yields the same diagnostic performance as combined CRP and PCT. So, combining any of them to CRP helps to early diagnose bacterial infection in those patients

Trends in Hospitalizations for Clostridioides difficile Infection in End-Stage Liver Disease, 2005–2014

Data on the current estimates of the disease burden of Clostridioides difficile (C. difficile) infection in the setting of end-stage liver disease (ESLD) are emerging. We examined the recent trends and predictors of hospitalizations and in-hospital mortality from C. difficile infection among hospitalizations with ESLD in the USA. We performed a retrospective analysis using the National Inpatient Sample, 2005-2014. We defined ESLD and C. difficile infection using the International Classification of Diseases, Ninth Revision, Clinical Modification. Multivariable logistic regression was used to determine the risk factors that impacted hospitalization and mortality. The prevalence of coding for C. difficile infection in decompensated cirrhosis increased from 1.3% in 2005 to 2.7% in 2014, with an annual rate of 7.8%. In hospitalizations with hepatocellular carcinoma, C. difficile infection increased steadily from 1.0 to 1.7% with an annual incremental rate of 6.4%. Among hospitalizations with ESLD, each passing 2-year period, increasing age, female, higher Charlson index, accompanying infection, hepatorenal syndrome, and ascites were associated with C. difficile infection. Although C. difficile infection was an independent predictor of in-hospital mortality during hospitalization with decompensated cirrhosis (odds ratio 1.53, 95% confidence interval 1.44-1.63), the proportion of in-hospital mortality during hospitalization with C. difficile infection and decompensated cirrhosis decreased from 15.4% in 2005 to 11.1% in 2014, with an annual rate of - 3.1% (95% CI - 5.7% to - 0.3%). While the prevalence of C. difficile infection in hospitalized patients with ESLD increased approximately twofold, the in-hospital mortality decreased significantly during the past decade.

Are presepsin and resistin better markers for bacterial infection in patients with decompensated liver cirrhosis?

BackgroundBacterial infections impair prognosis in patients with cirrhosis. Presepsin and, more recently, resistin are promising markers of infection and sepsis in patients without cirrhosis. AimsThe aim of our study was to assess the performance of presepsin and resistin as early markers of infection compared with C reactive protein (CRP) and procalcitonin (PCT), and their prognostic relevance in patients with decompensated cirrhosis. MethodsOne hundred and fourteen consecutive patients with decompensated cirrhosis were enrolled and followed-up for 28 days. Diagnostic performances of CRP, PCT, presepsin and resistin were assessed. ResultsFifty-three (46.5%) patients had bacterial infections of which 30 (56%) had sepsis. Presepsin and resistin had similar performance as CRP and PCT for the diagnosis of infection (best cut-off of 1444 pg/ml and 20 ng/ml, respectively) and sepsis. Presepsin (HR = 5.5; 95%CI: 2.36–13.21, p < 0.0001) and the ≥500 pg/ml increase of presepsin at 48 h (HR = 9.24; 95%CI: 3.66–23.27, p < 0.008) were independently associated with 28-day mortality. ConclusionsPresepsin and resistin have similar diagnostic performances to CRP and PCT for bacterial infection in decompensated cirrhosis. Presepsin and Δ presepsin ≥500 pg/ml have also a prognostic relevance for 28-day mortality.

Clarifying the role of C-reactive protein as a bacterial infection predictor in decompensated cirrhosis.

Bacterial infections are frequent in cirrhosis and may induce other deleterious complications. Ultrasensitive C-reactive protein (US-CRP), like other acute-phase proteins, is often considered useful in predicting bacterial infection in decompensated cirrhosis. However, US-CRP's reliability remains inconclusive, as inflammation in cirrhosis causes US-CRP synthesis independently of infection. The aim of this study was to clarify US-CRP's role as an infection predictor in decompensated cirrhosis. This was a prospective single-center study with systematic inclusion of cirrhotic patients admitted because of decompensation. A total of 118 patients were enrolled, of whom 47 (39.8%) had an overt infection, defined by clinical and laboratory/imaging criteria. Within those, 17 had infection confirmed by culture bacterial identification. Escherichia coli was the most frequent isolated bacteria. Seventeen patients had spontaneous bacterial peritonitis, but only four (23.5%) had positive ascitic fluid cultures. US-CRP levels were significantly higher in cases of overt infection and positive culture groups than the no infection group (median: 4.14 and 6.40 vs. 1.11 mg/dl, P<0.0001 for both). When considering both overt infection and positive culture groups, the US-CRP values of area under the curve as an infection predictor were, respectively, 0.824 and 0.870, P<0.0001 for both, with associated cutoff values of 2.40 and 3.92 mg/dl, and sensitivity and specificity of 78.7/74.6 and 82.4/79.2%, respectively. The ideal US-CRP infection confirmatory cutoff is probably situated between 2.40 and 3.92 mg/dl. However, as infection is somewhat concealed and hazardous in cirrhotic patients, if not considered with lower US-CRP levels according to specific clinic scenarios, it should be carefully considered, at least, if US-CRP is greater than 2.40 mg/dl (0.5 mg/dl normal upper cutoff).

Neutrophil-lymphocyte ratio predicts hospital-acquired bacterial infections in decompensated cirrhosis

BackgroundBacterial infection is a frequent complication and severe burden in cirrhotic patients. We determined the utility of neutrophil-to-lymphocyte ratio (NLR) to predict the hospital-acquired (HA) bacterial infections episode in patients with decompensated cirrhosis. MethodsWe retrospectively included 2066 consecutive decompensated cirrhotic patients from two separate tertiary hospitals, divided into training (n=1377) and validation (n=689) set. All data were collected on admission and all overt bacterial infections occurring after >48h of hospital stay were registered. ResultsThe incidence of HA bacterial infections in training and validation cohort was 35.87% and 31.05% respectively. Multivariate analysis showed that total bilirubin (TBil), albumin, white blood cell count (WBC) and NLR were independent predictors of HA bacterial infections. We established a Model_NTWA using these four variables and a Model_TWA which did not include NLR. Areas under the curves (AUC) of Model_NTWA (0.859) and NLR (0.824) were higher than which of Model_TWA (0.713), WBC (0.675), TBil (0.593) and Albumin (0.583). Consistent with training cohort, validation cohort showed similar results. Patients with NLR of at least 4.33 had a significantly lower survival (P<0.001). ConclusionsNLR can be used as a novel noninvasive marker to predict the occurrence of HA bacterial infections in decompensated cirrhotic patients.

Treat chronic hepatitis C virus infection in decompensated cirrhosis - pre- or post-liver transplantation? the ironic conundrum in the era of effective and well-tolerated therapy.

The management of hepatitis C virus (HCV) infection in patients with decompensated cirrhosis has evolved dramatically over the past few years mainly due to the availability of all-oral antiviral regimens. The currently approved all-oral direct-acting antivirals (DAA) containing sofosbuvir, ledipasvir, daclatasvir and ribavirin, in various combinations, have shown to be safe and effective in patients with decompensated cirrhosis with sustained virological response (SVR) rates nearly comparable to those with well-compensated liver disease. Unique issues yet remain such as the challenges with renal insufficiency, tolerability of ribavirin and risk of further hepatic decompensation with a protease inhibitor-based regimen. While most patients who achieve SVR have demonstrated improvement in hepatic synthetic function over the short course of follow, the long-term beneficial effects are unknown. Further, the baseline predictors of improvement in hepatic function have not been well delineated and thus have left us in a quandary as to what we might expect with successful therapy and thus we are at a loss to well educate our patients. The major concern, in potential liver transplant candidates, is of unintended 'harm' by achieving SVR but without improvement in hepatic function to an extent where the patients might function well. As HCV therapies are as effective in liver transplant recipients, there is a growing sentiment in some of the transplant quarters that those with decompensated liver disease and awaiting liver transplant be treated for HCV after liver transplant. This strategy would thus eliminate any concern of leaving a patient in 'no person's' land by treating HCV successfully pretransplant but not to the point of functional normalcy, while also would maintain the risk of HCC. Yet a contrarian view would be that not all patients have access to liver transplantation (LT), cannot bear the cost, have comorbidities or contraindications to LT. While the debate continues, it is essential that we develop robust predictors of improvement in liver function so that we can carefully select our patients for therapy in the context of liver transplantation.

Role of Portal Hypertension in Prediction of Bacterial Infection in Decompensated Cirrhosis

Background: Bacterial infection in cirrhotic patients is a fatal complication. The high incidence of bacterial infections in those patients may be related to several alterations in the defensive mechanisms against infections and increased intestinal permeability with bacterial translocation. Aim: To evaluate the role of portal hypertension (PH) in predicting the occurrence of bacterial infections in decompensated cirrhosis. Patients and Methods: In this retrospective cohort study, 99 patients—56 males and 43 females, with decompensated liver cirrhosis were included. Diagnosis of liver cirrhosis was based on clinical, laboratory and ultrasonographic examinations. Patients were classified according to the presence of bacterial infection into patients with infection—Group 1, and those without infection—Group 2. Laboratory, abdominal US and upper endoscopic data for all patients were collected. Logistic regression analysis was done to detect the independent factors for prediction of bacterial infection. Results: The mean age of patients was 50.5 ± 14.2 years. Bacterial infection was found in 41 patients (41.4%) and no infection in 58 patients (58.6%). Infected patients showed statistically significant higher values in the level of bilirubin, PT and Child-Pugh score (P value = 0.000) and lower values in the level of albumin, total serum protein and PC than those without infection (P value = 0.006, 0.000 and 0.000 respectively). Portal vein diameter (PVD) and splenic diameter (SD) showed statistically significant higher values in infected patients than in those without infection (P value = 0.028 and 0.000 respectively), also infection was more significantly prevalent in patients with varices than those without varices (P value = 0.000). The independent predictors for bacterial infection were: the age, total serum bilirubin, serum albumin, PT, PC, child score, PVD, SD and the presence of varices. Conclusion: Presence of varices (as a complication of PH) is an independent risk factor for the development of bacterial infection in decompensated cirrhotic patients and reduction of PH by any way could decrease this fatal complication.

Diagnosis and management of bacterial infections in decompensated cirrhosis

Bacterial infections are one of the most frequent complications in cirrhosis and result in high mortality rates. Patients with cirrhosis have altered and impaired immunity, which favours bacterial translocation. Episodes of infections are more frequent in patients with decompensated cirrhosis than those with compensated liver disease. The most common and life-threatening infection in cirrhosis is spontaneous bacterial peritonitis followed by urinary tract infections, pneumonia, endocarditis and skin and soft-tissue infections. Patients with decompensated cirrhosis have increased risk of developing sepsis, multiple organ failure and death. Risk factors associated with the development of infections are severe liver failure, variceal bleeding, low ascitic protein level and prior episodes of spontaneous bacterial peritonitis (SBP). The prognosis of these patients is closely related to a prompt and accurate diagnosis. An appropriate treatment decreases the mortality rates. Preventive strategies are the mainstay of the management of these patients. Empirical antibiotics should be started immediately following the diagnosis of SBP and the first-line antibiotic treatment is third-generation cephalosporins. However, the efficacy of currently recommended empirical antibiotic therapy is very low in nosocomial infections including SBP, compared to community-acquired episodes. This may be associated with the emergence of infections caused by Enterococcus faecium and extended-spectrum β-lactamase-producing Enterobacteriaceae, which are resistant to the first line antimicrobial agents used for treatment. The emergence of resistant bacteria, underlines the need to restrict the use of prophylactic antibiotics to patients with the greatest risk of infections. Nosocomial infections should be treated with wide spectrum antibiotics. Further studies of early diagnosis, prevention and treatment are needed to improve the outcomes in patients with decompensated cirrhosis.

Commentary: PPIs and risk of serious infection in decompensated cirrhosis – authors' reply

*Division of Gastroenterology, Hepatology and Nutrition, Virginia Commonwealth University and McGuire Veterans Affairs Medical Center, Richmond, VA, USA. Department of Epidemiology and Community Health, Virginia Commonwealth University and McGuire Veterans Affairs Medical Center, Richmond, VA, USA. Department of Quantitative Health Sciences, University of Massachusetts Medical School, Worcester, MA, USA. E-mail: jsbajaj@vcu.edu

Commentary: PPIs and risk of serious infection in decompensated cirrhosis

Commentary: <scp>PPI</scp>s and risk of serious infection in decompensated cirrhosis

PPI use linked with risk for infection in decompensated cirrhosis

PPI use linked with risk for infection in decompensated cirrhosis

Bedside Sublingual Video Imaging of Microcirculation in Assessing Bacterial Infection in Cirrhosis

Bacterial infections are common in cirrhosis and can lead to life-threatening complications. Sidestream dark-field (SDF) imaging has recently emerged as a noninvasive tool for capturing real-time video images of sublingual microcirculation in critically ill patients with sepsis. The objective of this study was to assess the utility of SDF in determining underlying infection in patients with cirrhosis. Sublingual microcirculation was compared among patients with compensated cirrhosis (Group A, n = 13), cirrhosis without sepsis (Group B, n = 18), cirrhosis with sepsis (Group C, n = 14), and sepsis only (Group D, n = 10). The blood flow was semi-quantitatively evaluated in four equal quadrants in small (10–25 mm); medium (26–50 mm); and large (51–100 mm) sublingual capillaries. The blood flow was described as no flow (0), intermittent flow (1), sluggish flow (2), and continuous flow (3). The overall flow score or microvascular flow index (MFI) was measured for quantitative assessment of microcirculation and predicting power for concurrent infection in cirrhosis. Marked impairment was observed at all levels of microvasculature in Groups B and C when compared with Group A. This effect was restricted to small vessels only when Group B was compared with Group C. MFI < 1.5 was found to have highest sensitivity (100%) and specificity (100%) for infection in decompensated cirrhosis. SDF imaging of sublingual microcirculation can be a useful bedside diagnostic tool to assess bacterial infection in cirrhosis.

Interleukin-6: an early marker of bacterial infection in decompensated cirrhosis

Fifty-seven patients with decompensated cirrhosis were studied prospectively to assess the sensitivity and specificity of early clinical or biological signs of bacterial infection. Among them, 19 had proven infection on admission (7 spontaneous bacterial peritonitis, 5 bacteraemia, 3 urinary tract infections, 2 pneumonia, 1 dental abscess and 1 cholangitis). Fever, polymorphonuclear cell count, fibrinogen and C-reactive protein levels were found to be of little or no help in diagnosing bacterial infection on admission. Interleukin-6 plasma levels were, however, significantly different between infected (median: 1386 pg/ml, range: 237-20,000) and non-infected patients (median: 34 pg/ml, range: 0-4500, p < 0.00001). Levels above 200 pg/ml were always found in infected patients, giving a sensitivity of 100% and a specificity of 74%. C-reactive protein correlated weakly with interleukin-6 levels, indicating a defective acute-phase response in cirrhosis. Tumor necrosis factor alpha plasma levels were less sensitive (95%) and specific (68%) for the diagnosis of bacterial infection at a threshold of 50 pg/ml, but were more closely related to a poor patient outcome. In decompensated cirrhosis, interleukin-6 plasma levels on admission provided the most sensitive and specific tool for the diagnosis of bacterial infection.