In recent years, with the popularity of the concept of national sports, an increasing number of people participate in sports, followed by an increase in the number of injuries associated with sports, especially in the process of non-contact sports. Cruciate ligament injury have become a more common knee joint sports injury, among which the anterior cruciate ligament (ACL) injuries are more common (1). A study of ACL injury rates in college sports showed that 59% of ACL injuries in men were contact injuries and 60% of ACL injuries in women were non-contact injuries (2). Posterior cruciate ligament (PCL) injuries are usually associated with other ligament, menisci, and cartilage injuries (3). Inadequate treatment of cruciate ligament injuries can have a serious impact on the knee function and quality of life. It has been shown (4, 5) that the short-term outcome of ACL rupture is an inability to participate in activities and a long functional rehabilitation time, and that in terms of long-term outcomes, ACL reconstruction with or without ACL reconstruction increases the patient’s risk of developing osteoarthritis and is often accompanied by meniscus damage.

Surgery is an effective treatment option for patients with knee injuries. It has been suggested that the first-line treatment for patients with ACL rupture is ACL reconstruction followed by rehabilitation (6). Santiago et al. (3) stated that surgical treatment of symptomatic simple and combined PCL injuries is recommended to restore joint stability and improve knee function. Surgical treatment of simple PCL also reduces the incidence of secondary osteoarthritis (7).

Lower limb venous thrombosis (LLVT) is the formation of new blood clots in the veins of the lower limbs, which can be severe enough to lead to the development of pulmonary embolism (PE) and death. When the key components of Virchow’s triad, namely slow blood flow, endothelial damage, and hypercoagulability, are met simultaneously, the chances of thrombosis are significantly elevated. More and more people are opting for arthroscopic surgery to treat cruciate ligament injuries for early recovery of knee function, improvement of symptoms, and quality of life; however, the potential for deep vein thrombosis and its sequelae are also present. A Meta-analysis noted that the incidence of DVT after knee arthroscopy was 3.1-17.9% (8). It has been shown (9) that post-thrombotic syndrome without intervention gradually increases from 23% at 2 years to more than 49% at 5-10 years, and may lead to amputation in the worst cases. However, controversy remains regarding whether prophylactic anticoagulation should be administered. Yeo et al. (10) suggested that there is a significant increase in the incidence of DVT in the perioperative period of knee arthroscopy and that prophylactic anticoagulation is necessary in some patients. It has been suggested that patients with no risk factors for thrombosis, not at high risk for thrombosis, less than 45 years of age, and early functional exercise do not need anticoagulant therapy for thromboprophylaxis (11–13).

It is now clinically observed that venous thrombosis occurs after a knee injury and before surgical treatment and that the probability of LLVT increases considerably after surgical treatment. Song et al. (14) noted that patients with a preoperative diagnosis of LLVT had a 66.7% probability of developing thrombus at the same site postoperatively. Thus, the risk of venous thrombosis is unpredictable until symptoms develop, and both patients and clinicians may overlook this danger. This study aimed to investigate the risk factors for preoperative LLVT in patients with knee ligament injuries and to provide clinicians and patients with a portable predictive model to assist in the diagnosis of LLVT in order to adopt a personalized treatment plan.

Wakabayashi et al. (15) noted that the incidence of preoperative asymptomatic deep vein thrombosis was as high as 17.4% in patients admitted for knee osteoarthritis (KOA). In this study, the incidence of preoperative LLVT in patients admitted for knee ligament injuries was 8.7%, which is lower than the incidence in patients with KOA. However, it has been shown that in total hip arthroplasty (THA) patients with a preoperative diagnosis of LLVT, the probability of recurrent thrombosis at the same site after THA is as high as 66.7% (14). Smith et al. (16) noted that IVT may dislodge intraoperatively and during other therapeutic procedures, progressing to pulmonary embolism (PE) or even death (17). Patients with LLVT have poor clinical prognostic outcomes, and post venous thrombosis syndrome can greatly distress patients and may eventually lead to amputation (9, 13). All these factors prolong the hospitalization time and increase the financial burden of patients (18). Therefore, it is important to identify patients at risk of preoperative LLVT. Preoperative LLVT in knee ligament injuries is currently understudied, and this study will help to explore this direction, identify risk factors in high-risk individuals, and develop a diagnostic tool.

This study focused on the demographic information, relevant medical history, and routine examination of patients. This study showed that the prevalence of preoperative LLVT in patients with knee ligament injuries was approximately 8.7%. Univariate logistic regression analysis showed that age, gender, comorbid cerebral infarction, damaged ligament site, injury-examination time, LDL, G, D-dimer, and FDP levels were risk factors for developing LLVT (Table 2). Among patients with preoperative LLVT, the prevalence was higher in those with advanced age, female (70.4%), combined cerebral infarction (18.5%), injury ACL/PCL+MCL/LCL (66.7%), injury-examination time ≤ 21 days (85.2%), and high levels of LDL, G, D-dimer, and FDP (Table 1). In the multivariate logistic regression analysis, gender, damaged ligament site, and D-dimer level were independent risk factors for the development of preoperative LLVT in patients with knee ligament injuries (Table 3).

Previous studies have pointed to age as a risk factor for DVT, which needs to be brought to the attention of physicians in patients of advanced age (10, 12, 13). In one study, the risk of venous thrombosis was 14.2 per 10,000 women in their 40s, increasing to 34 per 10,000 women over 45 (19). In this study, there was a significant difference between the mean age of the LLVT and None-LLVT groups, and age was shown to significantly affect the occurrence of LLVT in univariate logistic regression analysis, but not in multivariate logistic regression analysis. Considering the results of previous studies, age was included in the model construction, as increasing age causes an increase in coagulation factor levels as well as a decrease in vascular quality. Female, who have a high prevalence of knee ligament injuries in non-contact sports, also have a high prevalence of VTE (2, 20, 21). Similar results to those of previous studies were observed in this study. Some studies have shown that estrogen can promote blood clotting by increasing the levels of clotting factors II, VII, VIII, and X (22). Hormone-induced changes are a delicate balance between hemostasis and thrombosis but may increase the overall risk of thrombosis (23). In a rat study (24), injection of estradiol dipropionate (EDP) was found to increase the incidence of thrombin-induced pulmonary embolism, and the higher the estrogen level, the higher the likelihood of thrombosis.

This study showed that damaged ligament site is an independent risk factor for preoperative LLVT. Preoperative LLVT in patients with PCL injury alone was 14.3%, which was much higher than that in patients with ACL injury alone (3.7%), and patients with ACL/PCL combined with MCL/LCL injury had the highest prevalence of preoperative LLVT (15.9%). This may be related to the structure of the PCL, which is the main structure for knee extension and flexion activities and rotational activities and is twice as strong against external forces such as the ACL, which are often caused by great violence, and the injuries will lead to knee instability, pain, and weakness. This may have a greater impact on the limitation of knee motion than the ACL, and reduced lower extremity activity, among other factors, increases the risk of LLVT. D-dimer level is a significant influencing factor in thrombosis and is highly sensitive. D-dimer levels were higher in patients with proximal thrombosis than in those with distal thrombosis (25, 26).

Pharmacological anticoagulation is an important tool in the prevention of DVT, but the choice of prophylaxis is currently controversial. Related studies have suggested that early mobility exercise may be an option without routine anticoagulation prophylaxis after knee arthroscopy, but the use of low molecular heparin is recommended in patients with potential risk factors (11–13, 27). On the contrary view, aggressive preventive measures should be taken in patients with potential risk factors (10, 21, 27, 28). Wirth et al. (28) suggested that prophylaxis be given aggressively to high-risk individuals; the incidence of thrombosis receiving pharmacologic prophylaxis was 0.85%, which was much lower than the control group’s 4.1%, and the authors recommended 10 days of prophylaxis, which is effective and safe. Aggressive imaging should also be performed in high-risk patients for preoperative DVT (15). In response to this controversy, further research is needed to determine the clinical benefits of prophylactic anticoagulant use. In this study, we constructed a personalized diagnostic prediction model based on the identified risk factors, which can assist clinicians in screening patients at high risk of developing LLVT and adopting more aggressive preventive strategies for this group of patients to reduce the incidence of LLVT and alleviate the burden on patients. Our internal validation showed that the AUC of the model was 0.888 with good diagnostic efficacy, and the decision curve showed good clinical benefits.

Our study had some limitations. Our study was a retrospective study, and there may have been some unavoidable biases. In addition, we did not find a suitable public dataset for external validation, and further validation is needed to generalize the prediction model using more data and multi-center studies. In contrast to the currently published studies, we found no significant difference in the incidence of DVT in patients who did not receive thromboprophylaxis. Therefore, prospective randomized clinical trials are needed to evaluate the clinical benefit of prophylactic anticoagulation in the perioperative period for knee ligament injuries.

In summary, the independent risk factors for the preoperative occurrence of LLVT in patients with knee ligament injuries were gender, damaged ligament site, and high D-dimer level. This study is the first to establish a nomogram diagnostic prediction model for the preoperative occurrence of LLVT in knee ligament injuries, which will help clinicians identify the high-risk group for preoperative LLVT and actively perform imaging screening and preventive therapeutic measures for this group.