Sepsis is a syndrome of multiple organ dysfunction caused by infection and is accompanied by dysregulation of immune, endocrine and metabolic responses. Patients with sepsis usually exhibit inflammatory factor activation, impaired endothelial function and coagulation dysfunction [8]. As a result, these patients are at high risk for thrombosis, ranging from widespread microvascular involvement, such as disseminated intravascular coagulation (DIC), to venous thromboembolism. The incidence of hospital-acquired VTE in patients with sepsis is high. Kaplan et al.[4] reported that 37% of ICU patients with severe sepsis and septic shock were diagnosed with VTE. Sepsis patients are at high risk of thrombosis, and thrombosis prevention is recommended for these patients. However, due to coagulation dysfunction, sepsis patients are also at high risk of bleeding. If patients have a high risk of bleeding or have contraindications for pharmacologic thromboprophylaxis, mechanical thromboprophylaxis should be administered. These devices include static devices, graduated compression stockings (GCS) and dynamic devices, such as intermittent pneumatic compression (IPC) devices and arteriovenous foot pumps. However, thromboprophylaxis in sepsis patients has a certain failure rate[9]. In our study, thromboprophylaxis was performed on all 223 patients who did not experience VTE at admission; VTE still occurred in 37 patients, and the incidence of VTE after thrombosis prevention was 16.6%. The incidence of VTE after pharmacologic and mechanical prophylaxis was 13.9%. In studies of critically ill patients, even with adequate pharmacologic prophylaxis within 24 hours of admission, deep vein thrombosis occurred in 5%-20% of critically ill patients, and compared with pharmacologic prophylaxis alone, dual pharmacologic and mechanical prophylaxis had no additional benefits[10–12]. Our results are consistent with this conclusion. Several procoagulant mechanisms (i.e., overactivation of platelets, increased clotting factors, and endothelial cell damage) combined with traditional VTE risk factors (i.e., central venous catheters, prolonged immobilization, and/or high-dose steroid therapy) may explain the high risk of thrombosis in these patients despite thromboprophylaxis[13]. Jorda et al. [14]showed that therapeutic doses of heparin can reduce the risk of thromboembolic events compared with prophylactic doses of heparin but can increase the risk of major bleeding. Therefore, it is important to identify patients who may still develop thrombotic events although receiving thromboprophylaxis therapy.
The occurrence of VTE affects the prognosis of critically ill patients. A study showed that the occurrence of VTE was related to the 90-day death of critically ill patients and was a risk factor for all-cause death [6, 15]. There is little research on whether VTE is a risk factor for poor prognosis in sepsis patients. Therefore, the aim of our study was to investigate the relationship between VTE and the 90-day prognosis of sepsis patients, as well as the occurrence and risk factors for hospital-acquired VTE; this could help doctors identify patients at high risk of VTE, strengthen the monitoring of VTE and adjust the strategy of preventive anticoagulant therapy.
Our study is one of the few in which the risk factors for hospital-acquired VTE and its association with outcomes have been investigated in a cohort of sepsis patients. In our study, the 90-day prognosis of sepsis patients with VTE was worse than that of sepsis patients without VTE. After adjusting for sex and age, VTE was still an independent risk factor for 90-day all-cause death in sepsis patients. Our study revealed that patients with VTE had higher SOFA scores and lactic acid levels, suggesting that VTE is related to the severity of sepsis. Kaplan et al. [4] showed that VTE patients had greater mortality within 28 days. However, there was no significant difference compared with patients without VTE, possibly because the study included fewer patients and had a shorter follow-up period. Other studies have shown that among critically ill patients, the mortality rate of patients with VTE was greater than that of patients without VTE, and VTE events were associated with all-cause in-hospital mortality (OR 1.34; 95% CI 1.18–1.52) [6, 15]. Whether VTE affects the prognosis of sepsis patients remains to be verified by large sample studies.
Several models have been used to predict VTE risk in patients, including the Caprini score, Padua score [16]and Khorana score[17]. These models have been widely used in clinical practice to identify patients at high risk for VTE. However, these models are designed for different populations and cannot be targeted to patients with a single disease. Patients with sepsis have a high risk of VTE and high mortality[18]. Risk factors in patients with sepsis were analysed separately in our study. Length of hospital stay (OR 1.059, 95% CI 1.030–1.089) and mechanical ventilation (OR 3.845, 95% CI 1.297–11.585) were found to be independent risk factors for hospital-acquired VTE in sepsis patients. Kaplan et al.[4] reported that the presence of a CVC and mechanical ventilation were risk factors for VTE, and the occurrence of VTE was related to a prolonged hospital stay. In our study, univariate analysis showed that the presence of a CVC was associated with the occurrence of hospital-acquired VTE, but multivariate analysis showed that only length of hospital stay and mechanical ventilation were independent risk factors for the occurrence of hospital-acquired VTE. Wang et al. [19] also showed that mechanical ventilation and length of hospital stay were correlated with hospital-acquired VTE. In a study of risk factors for VTE in critically ill patients, length of hospitalization was also a risk factor for VTE[20, 21]. We concluded that VTE was correlated with the severity of sepsis; however, the SOFA score and lactic acid level were not found to be independent risk factors for VTE. Patients complicated with trauma also have a greater risk of VTE[22]. A previous study showed that active cancer was associated with VTE in sepsis patients; however, it was not an independent risk factor[4]. According to our univariate analysis, trauma and cancer were associated with the occurrence of VTE in patients with sepsis, but they were not found to be independent risk factors in the multivariate analysis. Reducing the length of hospitalization and mechanical ventilation in patients with sepsis may reduce the risk of VTE, and administering higher doses of anticoagulants to patients with longer hospital stays and mechanical ventilation may reduce the risk of VTE. Age, sex, comorbidities and laboratory indicators were not risk factors for hospital-acquired VTE. More active monitoring and prevention of sepsis in patients with risk factors can reduce the occurrence of VTE and improve the prognosis of sepsis patients.
There are several limitations in our study. First, this was a single-centre retrospective study, which inevitably leads to selection bias. Second, although VTE was monitored at intervals of 7–10 days during the hospital stay and patients with VTE symptoms were examined in time, ultrasound examination was not performed for all patients upon discharge or death, and patients with hospital-acquired VTE may have been missed. Third, our study did not include an upper limb colour ultrasound examination for all patients, which may have led to missed diagnoses of patients with upper limb venous thrombosis, and included patients with more severe conditions. CTPA or pulmonary ventilation perfusion scans are not easy to perform, and the occurrence of PE could not be determined.