Background and Aims: Deep vein thrombosis (DVT) is a common complication of inpatients with complex etiology. How to prevent the occurrence of DVT is a major problem in clinical practice. This study was conducted to investigate the relationship between ABO blood groups and the risk of DVT through a case-control study and Meta-analysis, so as to provide evidence for the individual prevention and precise treatment of DVT. 
Methods: Five hundred hospitalized patients with DVT from October 2012 to December 2018 in the Department of Vascular Surgery of North Jiangsu People's Hospital (only the first-time admission was counted in multiple-hospitalization patients), and 500 cause (no evident precipitating event, pregnancy, injury, surgery, cancer, and long-term bedridden) and sex 1:1 matched hospitalized patients without DVT during the same period were enrolled for a case-control study. The relevant case-control studies were collected by searching several national and international databases, and the retrieval time was limited from inception to October 1, 2019. After literature screening according to the inclusion and exclusion criteria, quality assessment and data extraction, Meta-analysis was performed by Review Manager 5.3 software.
Results: In case-control study, the results of overall analysis showed that the risk of DVT in individuals with O-type blood was significant higher than those with non-O-type blood (RR=2.859, 95% CI=2.142–3.817, P<0.01); the results of subgroup analysis stratified by causes showed that non-O-type blood was a possible risk factor for DVT in those with no obvious precipitating event (RR=2.845, 95% CI=1.819–4.282, P<0.05), having injury (RR=3.11, 95% CI=1.477–6.552, P<0.05), undergoing surgery (RR=2.613, 95% CI=1.271–5.373, P<0.05) and with cancer (RR=2.962, 95% CI=1.436–6.106, P<0.05), but had no significant significance in the those with pregnancy (RR=1, 95% CI=0.034–29.807, P>0.05) and the long-term bedridden (RR=5.714, 95% CI=0.528–30.325, P>0.05). Results of Meta-analysis showed that the risk of DVT in individuals with O-type blood was lower than those with non-O-type blood (RR=0.62, 95% CI=0.56–0.69, P<0.05).  
Conclusion: For patients with no apparent cause, having injury, undergoing surgery or with cancer, incorporating the ABO blood groups into the comprehensive analysis of risk factors for DVT may have certain clinical value for individual intervention and precise treatment of thrombosis. For patients with pregnancy and long-term bedridden, ABO blood groups may not be of sufficient clinical value, which still need to be further verified.