Acute kidney injury (AKI) is a common complication, occurring in approximately 20% of hospitalized cirrhotic patients and has a significant negative impact on patients' outcomes according to either the initial stage (at the time of the first fulfillment of AKI criteria), or the peak stage (at the peak value of serum creatinine concentration during hospitalization). Among all the precipitating factors to cirrhotic AKI, acute gastrointestinal hemorrhage is a common cause that leads to a decrease in effective arterial blood volume in the hyperdynamic circulatory status of cirrhosis. However, there is still lack of optimal predictors to developing AKI in cirrhotic patients suffering from acute GI bleeding. A number of biomarkers associated with AKI were recently described. Some studies have shown that these novel biomarkers increase with the severity of liver injury and are predictive of clinical outcomes. However, the effective prediction, definitive diagnosis and differentiation of AKI by these biomarkers are still controversial. Furthermore, there is no clinical studies focus on the applicability and potential alteration in the setting of acute gastrointestinal hemorrhage in patients with cirrhosis. Aim and significance: In this study, we aim to investigate the capability of novel renal biomarkers in predicting development of acute kidney injury, differentiating causes (between pre-renal AKI, acute tubular necrosis, and hepatorenal syndrome), and predicting the response to renal treatment as well as the hepatic and overall outcomes in patients with cirrhosis suffering from acute gastrointestinal hemorrhage.

Specific aims of this study :

1. To examine the influences of acute gastrointestinal hemorrhage on the serological or urinary level of novel renal biomarkers in patients with cirrhosis.
2. To investigate the ability of novel biomarkers to predict the development of acute kidney injury and the response to renal salvage treatment in patients with cirrhosis suffering from acute gastrointestinal hemorrhage.
3. To investigate the ability of novel biomarkers to differentiate the precipitating factors (such as sepsis-induced AKI, hypervolemia-induced AKI, etc.) and the causes of AKI between pre-renal AKI, acute tubular necrosis, and hepatorenal syndrome.
4. To investigate the ability of these novel biomarkers to predict the hepatic and overall outcomes in patients with cirrhosis suffering from acute gastrointestinal hemorrhage.

Estimation of sample size :

According to previous studies, the estimated incidence of AKI in cirrhotic patients suffering from gastrointestinal hemorrhage is about 10-15%. The type 1 error is set as 0.05; and the type 2 error is set as 0.2. Then, the calculated sample size is about 80. Considering the possibility of loss of follow-up is about 10%, the estimated sample size will be about 90 patients.

Sample collection and laboratory experiments:

Patients will be prospectively followed from admission until discharge.

1. Sample collection: A fresh 30-mL of urine sample will be collected either by way of clean catch or Foley catheter tubing will be collected at the time in admission to our hospital, at the peak stage of AKI, and after the recovery (if recovery occurs) 10cc of blood will be collected from the peripheral vessel once at the time in admission to our hospital, at the peak stage of AKI, and after the recovery (if recovery occurs).
2. Abdominal sonography (including evaluation of bilateral kidneys) will be arranged if no image study available within 3 months upon admission.

Urine examination:

urinary neutrophil gelatinase-associated lipocalin (NGAL), IL-18, spot urine protein, albumin, creatinine, urea nitrogen, sodium, kidney injury molecule 1 (KIM-1), liver-type fatty acid binding protein (L-FABP), insulin-like growth factor, and tissue inhibitor metalloproteinase will be measured.

Serum examination:

serum NGAL, cystatin C, blood urea nitrogen, creatinine, uric acid, IL-1, tumor necrosis factor alfa will be measured.

Record clinical information and regular follow-up:

Record any precipitating factors, including: presence of active infection or sepsis, the dosage and types of diuretics, presence of acute hemorrhage, the frequency and the drainage volume of each therapeutic paracentesis, supplement of albumin or other colloid fluid, presence of heart failure or active pulmonary problems, prescription of non-selective beta blockers, non-steroidal anti-inflammatory drugs (NSAIDs), antibiotics, herbs and other nephrotoxic drugs, as well as others. Record any chronic underlying diseases, including: diabetes mellitus, hypertension, congestive heart failure (the baseline left ventricular ejection rate if available), chronic kidney disease (baseline glomerular filtration rate), autoimmune disease, anemia of chronic disorder, the severity of liver diseases (such as status of ascites, serum albumin level, presence of varices, etc.), and so on.

Regular follow-up of renal function and fluid status at least twice per week as the routine management in clinical practice. Regular measurements of patients' clinical data, including vital signs will be done in daily practice.

Detailed laboratory methods:

Urine samples will be immediately refrigerated and then centrifuged at 5,000g for 10 minutes at -4°C. Aliquots of 1 mL of supernatant will be subsequently stored within 6 hours of collection in cryovials at -80°C for NGAL, IL-18, KIM-1, L-FABP, albumin, sodium, and creatinine measurements. No additives or protease inhibitors will be used. All biomarkers will be measured from frozen aliquots that will not undergo any additional freeze-thaw cycles. Laboratory measurements will be performed by personnel blinded to patient information.

Inclusion Criteria:

Consecutive adult cirrhotic patients admitted to Taipei Veterans General Hospital for suffering from acute gastrointestinal hemorrhage without intervention within 3 days

Exclusion Criteria:

1. Patients less than 20 years old.
2. Patients with antecedent or ongoing active malignancy.
3. Patients with human immunodeficiency virus coinfection or severe comorbidities, such as end-stage renal disease or uremia without regular dialysis, recent acute coronary syndrome or stroke, severe heart failure, severe arrhythmia, and major trauma.
4. Patients with severe hyperbilirubinemia with total bilirubin > 10mg/dL.
5. Patients who underwent trans-jugular intrahepatic portosystemic shunt.
6. Patients who are pregnant or breastfeeding.
7. Patients who refuse to join the study.

Consecutive adult cirrhotic patients admitted to Taipei Veterans General Hospital for suffering from acute gastrointestinal hemorrhage without intervention within 3 days