Incidence and risk factors of deep vein thrombosis in critically Ill patients: a systematic review and meta-analysis

BACKGROUND: Deep vein thrombosis (DVT) is a frequent and often silent complication in intensive care unit (ICU) patients, markedly increasing the risk of pulmonary embolism and death. However, comprehensive quantitative evidence on ICU-DVT incidence and associated risk factors remains scarce. OBJECTIVE: To synthesize global data on the epidemiology of ICU-DVT and identify independent risk factors, providing a basis for evidence-based, risk-stratified prevention. METHODS: We systematically searched PubMed, Embase, Cochrane Library, and Web of Science to July 2024 for observational studies reporting ICU-DVT incidence and risk factors. Pooled odds ratios (ORs) were calculated using random-effects models (PROSPERO ID: CRD42024583844). RESULTS: A total of 186 studies were included, involving 203,880 ICU patients. The overall incidence of DVT was 10.4% (95% CI 8.9%-12.0%). The incidence was significantly higher in COVID-19 patients than in non-COVID-19 patients (12.4% vs. 8.4%, p < 0.01). Medical ICUs had the highest incidence (16.3%), followed by trauma ICUs (11.6%), general ICUs (10.1%), neuro ICUs (8.6%), and surgical ICUs (7.9%). The incidence varied by region (Asia 12.5%, Europe 11.8%; America 9.3%, Oceania 9.1%). Systematic screening significantly improved DVT detection rates compared with clinically suspected diagnosis (14.6% vs. 7.9%). Meta-analysis revealed that central venous catheterization, a history of DVT, use of vasoactive agents, mechanical ventilation, prolonged ICU stay, and malignancy were associated with increased DVT odds. CONCLUSIONS: ICU-DVT affects approximately 1 in 10 patients and is associated with distinct modifiable and non-modifiable risk factors. The influence of diagnostic strategies on DVT incidence underscores the necessity for standardized screening in the ICU. Future prospective, large-scale studies are warranted to validate causal relationships and enable the development of risk prediction tools for precision prevention.