Thrombocytopenia In HBV Related Acute On Chronic Liver Failure Patients

BackgroundHepatitis B virus (HBV) infection remains a worldwide public health problem, with more than 2 billion people infected, including 350 million suffering from chronic hepatitis. It is estimated that HBV related liver failure, cirrhosis and hepatocellμlar carcinoma leads to about 1 million deaths annually. According to a national epidemiological survey in 2006, there were about 93 million HBsAg-positive subjects in China, of whom about 20 million had chronic hepatitis B infection. Therefore, China is still considered to be a highly endemic region, with about 1% CHB patients will develop into liver failure each year. Liver failure is a life-threatening condition with poor prognosis and high mortality, mainly caused by infection, alcohol and hepatotoxic drugs, which leads to rapid deterioration of liver function. The main symptoms include jaundice, coag μ lopathy, ascites, encephalopathy with or without hepatorenal syndrome and mμltiple organ failure. In China, acute-on-chronic liver failure (ACLF) is the most common type of liver failure, in which HBV related acute-on-chronic liver failure accounted for more than 80%.Thrombocytopenia (platelet counts<150,000/μL) is common in patients with chronic liver disease (CLD) which has been observed in up to 76% of patients with liver cirrhosis. Among patients with liver failure, more severe thrombocytopenia was observed, which leads to increased risk of hemorrhage, prolonged hospitalization and higher costs. Mμltiple factors contribute to the development of thrombocytopenia in CLD patients, including decreased activity of the hematopoietic growth factor thrombopoietin (TPO),splenic platelet sequestration, bone marrow suppression, interferon-based antiviral therapy, and increased platelet destruction. In this study, we described clinical characteristics of platelets in HBV-ACLF patients, explored the relationship between platelet count and prognosis, and elucidated the major factors which may contribute to thrombocytopenia through mμlti-factor analysis.Objectivesa) To describe the clinical characteristics of platelet counts in patients with HBV-ACLF.b) To explore the relationship between platelet count and short-term mortality in patients with chronic liver disease.c) To study different platelets dynamic patterns and their potential correlation with prognosis in patients with HBV-ACLF.d) To compare the platelet related parameters among different chronic liver diseases.e) To identify risk factors associated with thrombocytopenia in patients with ACLF through mμlti-factor analysis and to establish a mathematic model for the factors.Methodsa) A total of 191 patients were enrolled in this study, including chronic hepatitis B hepatitis(CHB) 68 cases, liver cirrhosis (LC) 48 cases, acute-on-chronic liver failure (ACLF) 75 cases, all of whom were hospitalized in Hepatology Unit andDepartment of Infectious Diseases, Nan fang hospital between December 2011 and August 2014. Inclusion criteria:CHB and ACLF were diagnosed according to "The guideline of prevention and treatment for chronic hepatitis B (2010 version)" and "Guideline for diagnosis and treatment of liver failure (2012) " respectively, while CIR was diagnosed based on either liver biopsy or the combination of physical examination, laboratory tests, imaging examination, and the symptom of portal hypertension (splenomegaly, ascites, and esophageal varices).Exclusion criteria:1) Previously or newly diagnosed primary hepatocellμlar carcinoma (HCC) or patients who had received liver transplantation before.2)Patients suffered from hematological disorders, such as hematological malignancies and myelosuppression. 3) Patients with history of bleeding or blood transfusion before administration.4) Patients complicated with other liver diseases.5) Patients received anti-platelet agents or toxins which may affect platelet count in recent 6 months.6) Post-splenectomy patients.b) Demographic characteristics and laboratory results, as well as spleen width determined by Abdominal ultrasound or computed tomography (CT) scan were collected. Laboratory examinations included platelet count (PLT), biochemical tests (ALT, AST, albumin, TBIL), blood coagulation parameters (plasma fibrinogen, PT-INR, plasma D-dimer). Besides, reticulated platelet ratio, serum thrombopoietin level, platelet activation ratio (PAC-1, CD62P, and CD63), glycocalicin, serum soluble CD 163, and apoptosis ratio of platelet were measured. Univariate and multivariate linear regression analyses were performed to identify the major fators responsible for thrombocytopenia in ACLF patients. Platelet counts on the 1st,7th day of hospitalization and the last follow-up were collected, while patients were followed for 28-day survival by telephone survey.c) We retrospectively analyzed a total of 114 patients diagnosed with HBV related ACLF according to " Guideline for diagnosis and treatment of liver failure (2012) " who were hospitalized in Hepatology Unit and Department of Infectious Diseases, Nan fang hospital between February 2011 and January 2015. Platelet counts on the 1st,7th,15th day,1st and 2nd month of hospitalization and the last follow-up were collected, while patients were followed for survival by telephone survey. We further divided patients into groups according to platelet dynamic patterns and performed Kaplan-Meier analyses to compare survival rates among groups using Log-rank tests.d) Platelets preparations. Venous blood samples were collected into one-tenth volume of sodium citrate-based anticoagμlant tubes. Platelet-rich plasma (PRP) was prepared by centrifugation at 125g for 20 minutes, and platelets were separated from PRP by centrifugation for 5 minutes at 850g. The platelet pellet was then resuspended with platelet buffer. Plasma and serum were collected and stored at-20℃ in the meanwhile.e) Flow Cytometric Analysis of Reticμlated Platelets Percetages of RP were measured by flow cytometer with fluorescent RNA dye (Thiazole orange). In brief, Some 5μl of whole blood was added with 10μl of PerCP-labelled CD61 antibody for platelet identification and 1ml of 0.1% Thiazole orange. A control tube was used for each sample with 10μl of CD61-PerCP and lml of phosphate buffer saline (PBS). After incubation for 30 min in the dark at room temperature, the tubes were centrifuged at 1,200 g for 2.5 minutes to form cellμlar pellet which was then suspended in 1mL of PBS. A dot plot cytogram (CD61-PerCP versus Thiazole orange fluorescence) was generated, and the RP rate was expressed as a percentage of both a Thiazole orange and CD61-PerCP-positive popμlation among 10,000 identified platelets.f) Flow cytometric detection of activated platelets 5 tubes of whole blood (5μl) was added with 10μl of PerCP-labelled CD61 antibody for platelet identification.5μl of PE-labelled CD62P,5μl of PE-labelled CD63,20μl of FITC-labelled PAC-1 were added to three different tubes respectively. In the meanwhile, two control tubes were incubated with 20μl of FITC-labelled PAC-1 or 5μl of PE-labelled Mouse IgG combined with 10μl of RGDS. After incubation for 30 min in the dark at room temperature, cells were washed with PBS buffer twice, resuspended with 200μl of 2% paraformaldehyde and stored in 2～8℃ for 2 hours before detection.g) Flow cytometric detection of phosphatidylserine (PS) exposure. Platelets (1x108) were incubated with 250μl of Annexin-V binding Buffer and 5μl of FITC-Annexin V for 30 minutes in the dark before detection with Flow Cytometry.Resμtsa) Clinical characteristics and short-term kinetics of platelet counts in patients with HBV related chronic liver disease. A total of 191 patients were enrolled, including 68 cases of CHB,48 cases of LC,75 cases of ACLF. Platelet counts (G/L) in patients with ACLF[98 (3-253) 10^9/L] were lower than CHB patients [172 (24-327) 10^9/L] (P<0.0001), but comparable to LC patients [88 (22-244) 10^9/L] (P=0.913). However, ACLF patients always had rapid drop in the platelet count during the first week of administration, which declined from [97 (21-267) 10^9/L]to [71 (4-208) 10^9/L] (P<0.0001), while in patients with CHB and LC, platelet count remained stable relatively.b) Relationship between platelet count at admission and short-term mortality in patients with chronic liver disease. Receiver operating characteristic (ROC) cureve analysis revealed that platelet count at admission can predict 28-day survival (AUROC=0.72, P<0.0001).c) Comparison of platelet related parameters am28-day and long-term survival in ACLF patients with different platelets dynamic patterns. According to platelets dynamic patterns, ACLF patients were divided into sustained declining group, decline-recovery group and steady low-level group. The 28-day and long-term survival rates in decline-recovery group was 100% and 95.8%, remarkably higher than that in sustained declining group (47.4% and 13.2) and in steady low-level group (64.9% and 40.4%).d) ong different chronic liver diseases. There were no significant differences in the ratio of reticμlated platelet, TPO level, platelet activation markers CD62P and CD63, glycocalicin and platelet apoptosis ratio between patients with CHB and ACLF. However, compared to CHB, ACLF patients had higher sCD163 level (53.3 vs.18.1ug/mL, P<0.0001)、spleen width(40.4 vs.35.2mm, p=0.002)、D-dimer(443.5 vs.90 ug/L, p=0.002)and PT-INR(2.14 vs.1.26, P<0.0001),but lower PAC-1(+) ratio (3.34 vs 10.65,P=0.002)、fibrinogen level (1.24 vs.2.25 g/L,P<0.0001)e) Risk factors associated with thrombocytopenia in patients with ACLF. Platelet count was positively correlated with fibrinogen level (Rs=0.478, P<0.0001), albumin level(Rs=0.335, P<0.0001), alanine aminotransferase(Rs=0.199,P=0.007) and glycocalicin level (Rs=0.484, P<0.0001), while negatively correlated with spleen width (Rs=-0.488, P<0.0001), PT-INR (Rs=-0.270, P<0.0001), serum total bilirubin (Rs=-0.171, P=0.025) and platelet apoptosis ratio (Rs=-0.251, P=0.027). Multiple linear regression analysis revealed that spleen width, fibrinogen level and glycocalicin level were independent risk factors for platelet count. Platelet count=137.937+21.658 fibrinogen level (g/L)-2.249 spleen width (mm)+0.020 glycocalicin level (In mg/ml) (R2=0.597, P=0.011).Conclusiona) Platelet counts (G/L) in ACLF and LC patients were significantly lower than that in CHB patients. During the first week, platelet counts dropped rapidly in ALCF patients but remained stable in LC or CHB patients.b) The short-term survival of patients with HBV related chronic liver disease had a close relationship with platelet count at admission. The platelet count at admission can predict short-term survival of these patients.c) The short-term and long-term survival rates in the decline-recovery group were remarkably higher than other groups, indicating that the dynamic patterns of platelet counts have a close relationship with prognosis.d) The ratio of retic μ lated platelet, TPO level, platelet activation marker CD62P and CD63, glycocalicin and ratio of platelet apoptosis in CHB patients were all comparable with those in ACLF cases, indicating that thrombopoiesis、apoptosis rate、activation rate didn’t affect much. Compared to patients with CHB, ACLF patients had higher spleen width、D-dimer、PT-INR、and sCD 163 level, but lower PAC-1(+) ratio and fibrinogen level, suggesting that macrophage activation、splenic platelet sequestration、clotting consumption contribute to thrombocytopenia in ACLF patients.e) Platelet count was positively correlated with fibrinogen, albumin level, alanine aminotransferase and glycocalicin level, while negatively correlated with spleen width, PT-INR, serum total bilirubin and platelet apoptosis ratio. Multiple linear regression analysis revealed that spleen width, fibrinogen levels and glycocalicin level were independent factors associated with platelet counts in patients with HBV related chronic liver disease.