Prevalence Of PF4/H Antibodies, Its Effecting Factors And Association With Thrombosis In Patients Undergoing Haemodialysis

Background: Heparin-induced thrombocytopenia (HIT) is one of mostcommon agent-induced diseases, which is seriously associated with the prognosisof patients. Patients undergoing maintenance haemodialysis (MHD) wererepeatedly exposed to heparin and under higher risk for HIT. The incidence ofchronic kidney disease (CKD) is approximately10%, and about1-2millionpatients with end stage renal disease (ESRD) will depend on haemodialysis tosurvive. Based on the data of nation-wide HD patients’ registered informationsystem, about two hundred and thirty thousand patients has been receiving HDrecently, but unfortunately, there hasn’t been any clinical research on HIT inChina. Ten HD centers from three Chinese large cities of Beijing, Shenyang andDalian combined and813MHD patients had taken part in our study. It was thefirst multicentre, large-sample research on HIT in China. Injected heparin and PF4in the blood interact with HIT antibodies and produce immunological complex,leading to the activation of platelets mediated by FcγIIa receptor. Then theactivated platelets will release microparticles with procoagulant activity into theblood, resulting in the formation of thrombosis. The formation of PF4/Hantibodies is central in progression of formation of thrombosis. Therefore,prevalence of PF4/H antibodies is the focus of studies on HIT; however, differentstudies achieved significant different prevalence of PF4/H antibodies. Someforeign researchers paid their attention on the association between PF4/Hantibodies and incidence of thrombosis, while different studies achievedconflicted results. Some researchers suggested that the ELISA PF4/H antibodiescouldn’t predict thrombosis. As most applied anti-platelet agent, aspirin, isprescribed for preventing thrombosis, and it has been proved by previous in vivo and in vitro experiments that aspirin could protect patients exposed to heparinagainst HIT. However, effect of aspirin in HIT needs proved by large sampleclinical studies. In conclusion, the purpose of our research is described as follows:①the prevalence of positive PF4/H antibodies in Chinese MHD patients;②factors effecting positivity of PF4/H antibodies;③association between PF4/Hantibody positivity and incidence of thrombosis;④effect of aspirin on incidenceof thrombocytopenia and thrombosis in MHD patients;⑤discuss on the methodto predict the incidence of thrombosis.Methods: Our study included two parts: a multi-center, cross-sectional studyand a single-center, longitudinal study. We achieved the positivity of PF4/Hantibodies and factors effecting antibodies in cross-sectional study and achievedassociation between PF4/H antibodies and thrombosis and established model ofthrombosis in longitudinal study. Finally,813MHD patients were enrolled in ourstudy and all serum samples of them were collected before haemodialysis. Allsamples were collected completely and transported to key national laboratory ofnephrology of General Hospital of PLA; all clinical information had beencollected completely by physicians of each HD centers and sent together to thekey national laboratory of nephrology of General Hospital of PLA and then adatabase had been established. All samples were detected for PF4/H antibodies.Besides, we had also detected serum PF4/H antibody titers of326healthyvolunteers and achieved physiological distribution of PF4/H antibody titers inhealthy volunteers. The cutoff of positivity of PF4/H antibodies was defined byphysiological upper-limitation of healthy volunteers (mean+3SD of healthyvolunteers). Therefore,326healthy volunteers who were never exposed to heparinwere enrolled into our study from Beijing, Shenyang and Dalian. Longitudinalcohort was consisted with136patients from HD center of General Hospital ofPLA (all patients participated in cross-sectional cohort). All patients had beendivided into four groups according to PF4/H antibody positivity and aspirinprescription and platelet counts, D-dimer titers and all incidence of thrombosis of all patients during12-month follow-up had been completely recorded.Results:①PF4/H antibodies of the healthy volunteers were16.8±4.6ng/mL,and PF4/H antibodies of the MHD patients were28.1±15.9ng/mL, there weresignificantly different between two groups (p=0.001)②The cut-off of MHDpatients was based on mean+3SD of antibody titers detected in healthy volunteers(99th percentile), and255cases (31.4%)manifested antibody positivity.③Age,frequency of dialysis and heparin types (UFH or LMWH) rather than duration ofdialysis or gender were found to be significantly associated with PF4/H antibodypositivity.④Ten HD centers had achieved significant difference on positivity ofPF4/H antibodies, between14.9%and58.0%. By compared with informationamong HD centers with different PF4/H antibody titers, it was found that HDcenters with higher antibody titers were always prone to prescribe UFH ratherthan LMWH, and the trend with statistic significance.⑤Based on analyzingpatients with antibody positivity, we found aspirin could decrease the incidence ofthrombocytopenia and D-dimer titers at the pointing time.⑥In patients withpositive PF4/H antibody titers, those who were prescribed with aspirin manifestedsignificantly increased incidence of thrombocytopenia and thrombosis.⑦Thesame results were not found in patients without antibody positivity.⑧By theCOX hazard proportional model, positivity of PF4/H antibodies rather thanaspirin prescription was found a significant independent factor for thrombosis.⑨⑩The model for predicting thrombosis was established by Logistic regressionmodel and area of ROC were applied to compare the accuracy between predictedmodel and mere serum ELISA detection.Conclusion: answers to five questions of Background were described asfollows:①The prevalence of MHD patients was31.4%;②Elder age,frequently received dialysis and prescribed UFH rather than LMWH were all riskfactors for high level of PF4/H antibodies;③Increased PF4/H antibody titerswere significantly associated with higher incidence of thrombosis;④Aspirincould protect patients with antibody positivity against thrombocytopenia and thrombosis, but it’s effect on patients without antibody positivity were not foundin this study;⑤It was more accurate to combine clinical information and ELISAdetection together to predict thrombosis compared with the method merelydepending on ELISA assays.