Clinical Effectiveness of a Pneumatic Compression Device Combined with Low-Molecular-Weight Heparin for the Prevention of Deep Vein Thrombosis in Trauma Patients: a Retrospective Cohort Study


 Purpose: To investigate the clinical effectiveness of a pneumatic compression device (PCD) combined with low-molecular-weight heparin (LMWH) for the prevention and treatment of deep vein thrombosis (DVT) in trauma patients.Materials and Methods: This study retrospectively analyzed 286 patients with mild craniocerebral injury and clavicular fractures admitted to our department from January 2016 to February 2020. Patients treated with only LMWH were used as the control group, and patients treated with a PCD combined with LMWH were used as the observation group. The incidence of DVT, the postoperative change in the visual analog scale (VAS) score and coagulation function were observed and compared between the two groups. Excluding the influence of other single factors, binary logistic regression analysis can indicate whether the use of a PCD effectively improves the patient's postoperative coagulation function.Result: The incidence of DVT in the observation group was significantly lower than that in the control group (7.1% vs. 16.7%, respectively, χ2=4.605, P<0.05). The postoperative VAS scores of the two groups were lower than those before surgery (P<0.05). The coagulation function of the LMWH+PCD group was significantly higher than that of the LMWH group, and the combined anticoagulant effect was better (P<0.05). There were no significant differences between the two groups in terms of preoperative and postoperative Glasgow coma scores, intraoperative blood loss, postoperative infection rate, or length of hospital stay (P>0.05). According to binary logistic regression analysis, the postoperative risk of DVT in patients with LMWH alone was 1.764 times that with the combination of LMWH+PCD (P <0.05). The incidence of DVT with LMWH alone was 1.37 times that with the combined application. Activated partial thromboplastin time (APTT), platelet (PLT) count, and DVT are independent risk factors that indicate the need for a PCD (P <0.05). Excluding the influence of other variables, LMWH+PCD can effectively improve the coagulation function of patients. The area under the curve data indicate that APTT and PLT are sensitive indicators of anticoagulant efficacy (Z inspection, inspection level α = 0.05). Conclusion: Compared with LMWH alone, PCD combined with LMWH can improve blood rheology and coagulation function in patients with traumatic brain injury and clavicular fracture, significantly reduce the incidence of DVT, shorten the hospital stay and improve the clinical effectiveness of treatment.


Introduction
With the rapid development of the global economy, people's living standards have continuously improved, and the incidence of work-related injuries and tra c injuries with fractures and brain trauma has also increased. In the treatment of patients with multiple injuries, we initially focus on addressing potentially fatal injuries. After the patient's vital signs are initially stabilized, fractures are addressed. The risk of postoperative deep vein thrombosis (DVT) is signi cantly higher in patients with multiple traumatic injuries than in other patients. 1 As the number of injuries increases, the incidence of DVT increases. More than 50% of all hospitalized patients have a risk of developing DVT, and patients undergoing surgery are at higher risk than medical patients. 2 DVT can be life-threatening after trauma; thus, the prevention of DVT is critical for the treatment of trauma patients. 3 This study aimed to compare the effectiveness of a pneumatic compression device (PCD) combined with low-molecular-weight heparin (LMWH) and LMWH alone in trauma patients and provide an experimental and theoretical basis for the development of a clinical anticoagulation strategy.

Study Design and Grouping
This study retrospectively analyzed 286 patients with mild craniocerebral injury and clavicular fractures admitted to our department from January 2016 to February 2020. After excluding 34 patients who did not meet the inclusion criteria or were lost to follow-up, patients treated with LMWH alone were used as a control group, and patients treated with a PCD and LMWH were used as an observation group (LMWP+PCD). The incidence of DVT, the postoperative change in visual analog scale (VAS) scores and coagulation function were observed and compared between the two groups. The research procedure was approved by the Ethics Committee of our hospital.
Inclusion criteria: The inclusion criteria were as follows: age < 75 years; Caprini risk scores indicating high risk and very high risk; 4 indications for the surgical treatment of clavicle fractures; no surgical contraindications and no history of preoperative DVT; Glasgow coma scores (GCSs) of 13-15 points; 5 and mild craniocerebral injury with disordered consciousness <30 min. 6 Exclusion criteria: The exclusion criteria were as follows: age ≥ 75 years; GCS < 13 points; fracture surgery contraindications; malignant tumor; severe organ dysfunction; severe coagulopathy; preoperative vascular ultrasound indicating DVT; allergy to the drug in this study; refusal to provide informed consent; failure to meet the experimental ethical standards; a history of mental disorders or a history of craniotomy; and pregnancy.

Ethics statement
We declare that we have no nancial or personal relationships with other people or organizations that could inappropriately in uence our work, we have no professional or other personal interests of any nature in any product, service or company that could be construed as in uencing the position presented in, or the review of, this manuscript. There are no con icts of interest. The study complied with the principles of the Declaration of Helsinki (1964), and this study was approved by the institutional review board. The requirement for informed consent was waived for this study.

Clinical treatment
All patients included in the study had no need for neurosurgical treatment for mild craniocerebral injury and were treated conservatively with nutritional support and antineurodegenerative drugs. The clavicle fractures were treated with open reduction and internal xation ( Figure 1). All patients were treated with anticoagulation based on their scores on the Caprini Risk Assessment Model. In the control group (LMWH group), 4 000 IU of LMWH calcium was administered once daily after surgery for 7 days. In the observation group (LMWH+PCD group), subcutaneous injections of LMWH calcium 4 000 IU per day after surgery were administered combined with the use of a PCD on the lower limbs ( Figure 2) twice a day after surgery for 7 days.

Observation index
The general patient characteristics, past medical history, Caprini score, GCS score, use of anticoagulant drugs, and type of clavicular fracture were recorded. The VAS was used to assess the pain experienced by patients before and after surgery. 7 Both groups of patients underwent deep venous color Doppler ultrasonography before and after surgery, and each patient underwent an ultrasound examination every two weeks. Outpatient follow-up visits continued after discharge. Platelet (PLT) counts, brinogen (FIB) levels, plasma activated partial thromboplastin times (APTTs), plasma prothrombin times (PTs), and Ddimer levels were recorded in both groups. The blood coagulation parameters of all patients before, 1 day after, and 7 days after surgery were observed and recorded. The preoperative and postoperative GCS scores, the amount of bleeding during clavicular fracture internal xation, the postoperative infection rate, and the length of hospital stay were recorded.

Statistical analysis
The measurement data of continuous variables that conformed to a normal distribution or an approximately normal distribution are expressed as the mean ± standard deviation, and the comparisons between the two samples were performed using independent sample t tests. Classi cation data, such as Caprini scores, fracture classi cations and GCS scores, were compared using nonparametric rank sum tests. The chi-square (χ 2 ) test was used to compare the count data between the groups. P<0.05 was considered signi cant. The dependent variable was whether the patient experienced a DVT after operation, while gender, age, medication use, Caprini score, length of stay, type of clavicle fracture, preoperative GCS score, intraoperative blood loss, postoperative infection rate, blood coagulation function index, etc., were independent variables. For these independent variables, a binary multivariate logistic analysis was introduced. Odds ratios and corresponding 95% con dence intervals were calculated for each characteristic. According to the ROC curve and the area under the curve (AUC) to evaluate the area under the curve of a single factor variable, it was found that APTT and PLT are sensitive indicators. (Z inspection, inspection level α = 0.05). All statistical tests were performed at a signi cance level of α=0.05. The statistical software IBM SPSS for macs, version 26, was used.

Results And Conclusions
Comparison of basic data Between the two study groups, there were no signi cant differences in sex, age, hypertension, diabetes, antiplatelet therapy, oral anticoagulant therapy, Caprini risk scores, GCS scores, or clavicle fracture types (all, P>0.05) ( Table 1).

Comparison of VAS scores
There were no signi cant differences between the two groups in the VAS scores before and 7 days after surgery (P>0.05). However, the postoperative VAS scores of the two groups were signi cantly lower than those before surgery, which indicates that the surgical effect on clavicle fracture-related pain in trauma patients is signi cant (P < 0.05) (Table 2, Figure 3).

Comparison of the incidence of deep vein thrombosis
Deep venous color Doppler ultrasonography of both lower limbs was recorded after the operation, and the incidences of DVT were compared. The incidence of DVT was signi cantly lower in the observation group than in the control group (7.1% vs. 16.7%, respectively; χ 2 =4.605, P<0.05) (Table 3, Figure 4).

Comparison of coagulation function indicators
In the comparison of the coagulation parameters of all patients, there was no signi cant difference in preoperative coagulation function between the two groups (P>0.05). Repeated measures ANOVA showed signi cant differences in the APTT, PT, PLT count, and FIB level before the operation and 1 and 7 days after the operation. The PLT count and FIB level in the LMWH group were signi cantly higher than those in the LMWH+PCD group; the APTT and PT in the LMWH group were signi cantly lower than those in the LMWH+PCD group. This nding indicates that the coagulation ability in the LMWH+PCD group was signi cantly greater than that in the LMWH group, and the combined anticoagulant effect was better. The difference was signi cant (P<0.05) (Table 4. Figure 5).

Comparison of other related indicators during hospitalization
There were no signi cant differences between the two groups in terms of preoperative and postoperative GCS scores, intraoperative blood loss, postoperative infection rate, or length of hospital stay (P>0.05) ( Table 5).

Multivariate logistic regression analysis of thrombosis
Taking the occurrence of DVT after the operation as the dependent variable, the basic data were used as the independent variable to introduce the binary multivariate logistic analysis. The results showed that the postoperative risk of DVT in patients with LMWH alone was 1.764 times that of patients who received the combination of LMWH+PCD (P <0.05). This nding indicates that the combined anticoagulation effect of LMWH+PCD is superior to that of LMWH alone (Table 6).
Multivariate logistic regression analysis of whether to add PCD Taking the addition of PCD as the dependent variable, these independent variables were introduced into a binary multivariate logistic analysis. The results indicated that the incidence of DVT with LMWH alone was 1.37 times that with the combined application. APTT, PLT, DVT are independent risk factors that indicate the need for a PCD (P <0.05). Excluding the in uence of other variables, LMWH+PCD can effectively improve the coagulation function of patients (Table 7).
Area under the curve of a single factor variable According to the ROC curve and the AUC of a single factor variable, APTT and PLT are sensitive indicators of anticoagulant e cacy (Z inspection, inspection level α = 0.05). Combined with multivariate logical retrospective analysis, the results indicated that LMWH+PCD can affect the patient's APTT and the number of PLTs (Table 8, Figure 6).

Discussion
In patients with multiple injuries, potentially fatal injuries are treated rst. Early internal xation needs to adhere to the principle of the preservation of life as the rst goal. 8 After the patient's vital signs are initially stable, other treatments can be initiated. Epiphyseal growth is rapid in patients with brain trauma. A large number of epiphyses in the subclavian region may compress the subclavian structure, stimulate the brachial plexus nerve, and cause traumatic arthritis. The internal xation of fractures should be performed early. 9 However, the incidence of venous thromboembolism (VTE) in patients with trauma is high and can be life-threatening after trauma and cause various complications. Therefore, the prevention of DVT is very important in the treatment of trauma patients. DVT and pulmonary embolism (PE) are collectively referred to as VTE events. 10 In clinical experience, chest fractures, pelvic fractures and limb fractures, craniocerebral trauma and bed rest are high risk factors for VTEs. 11 This study analyzed various factors in trauma patients and provides a theoretical basis for the formulation of clinical anticoagulation strategies.
In patients with multiple traumatic injuries, most deaths are caused by traumatic brain injury (TBI). According to the GCS, clinical TBI can be divided into mild, moderate and severe. The related permanent disability rates are 10%, 60% and 100%, respectively. 12 The inclusion criteria in this study were patients with mild injuries, with a GCS score of 13 or higher. The incidence of VTE and its related complications increases the length of hospital stay and medical costs; it is also the most common preventable cause of death in hospitals. 13 The prevalence of DVT in trauma and orthopedic patients is estimated to be 1.16%. With the continuous improvement of medical knowledge and treatment methods, DVT can be controlled and prevented. 14 In this study, some patients experienced a DVT of the lower limbs. Despite drug and mechanical anticoagulation, craniocerebral injury, chest trauma, and bed rest are still high risk factors for DVT. Hospitalization for acute trauma is an independent risk factor for incident VTE; however, VTE preventive measures increase the risk of bleeding in patients with acute trauma. 15 According to previous experience, there is concern regarding the worsening of intracranial hemorrhage in patients with brain trauma, and drugs are not usually used in the early stage after injury. 16 According to a 2011 study by Scudday with 812 patients with craniocerebral trauma, the incidence of VTE was lower in patients receiving preventive drugs than in those not receiving preventive drugs. 17 Shelan proposed in 2012 that low-risk TBI patients start enoxaparin treatment within 24 hours after injury. 18 In this study, based on the Caprini risk scores of patients admitted to the hospital and the BTF recommendations, 19 if the injury is stable and the bene t of prevention outweighs the risk of bleeding progression, then preventive drugs should be administered. All patients at high and very high risk were treated with anticoagulation drugs in this study.
Many risk factors have been identi ed for venous thrombosis that alter blood ow, activate the endothelium, and increase blood coagulation. Three important factors affecting thrombosis are blood ow, blood composition and blood vessels. Severe trauma often leads to the emergence of the triad of Virchow, with hypercoagulability, endothelial damage, and venous stasis, which increases the risk of thromboembolism. 20 Prolonged bed rest and limb fractures in the patients in this study promoted venous stasis. Increased hematocrit levels, increased FIB levels, and shortened clotting time in patients with trauma can lead to the local accumulation of coagulation activation products and cause hypercoagulability in patients. 21 In this study, patients' blood coagulation was monitored to determine whether there was thrombosis in the blood.
The contraindications caused by brain trauma often limit the choice of strategies to prevent VTE in trauma patients. This study mainly explores the choice of methods to prevent DVT in trauma patients. The available methods for thrombosis prevention in trauma patients are divided into pharmacological anticoagulation, mechanical prevention and inferior vena cava lters (IVCs). 22 LMWH was used for anticoagulation in this study. Its advantages are fast pharmacokinetics, convenient administration and few side effects; it has strong activity against Xa because of its low molecular weight and charge. 23 LMWH became the most (or only) effective method for preventing DVT in trauma patients in the late 1990s. 24 Several studies in recent years have shown that LMWH reduces the incidence of DVT in trauma patients. 25 Due to the need for safer and more effective preventive measures in patients with high-risk trauma, the experimental group in our study used drugs and mechanical anticoagulation. Conventional mechanical prevention includes graded compression socks (GCS), sequential pneumatic compression equipment (PCD), and pneumatic sole (A-V) foot pumps. 26 These devices reduce the risk of thrombosis and related bleeding by reducing the luminal diameter of the veins, which leads to an increase in venous blood ow velocity; they are often used in the treatment of trauma patients. 27 Kurtoglu prospectively randomized 120 trauma patients and compared PCDs and LMWH with regard to the prevention of VTE; the authors concluded that PCDs can be safely used to prevent DVT. 28 It is recommended to combine LMWH in with intermittent pneumatic pressure devices. Based on the results and conclusions of this study, the combination of PCD and LMWH appears to be an advantageous way to prevent DVT in patients with multiple traumatic injuries.
According to the American College of Chest Physicians (ACCP) guidelines for the prevention of VTE in 2008, 29 LMWH should be administered as soon as possible to patients with severe trauma; an acceptable alternative is to combine LMWH with the best mechanical methods. For patients with severe trauma, ACCP recommends therapy to prevent thrombosis until discharge, and patients in this study received therapy for the duration of hospitalization. The Eastern Trauma Surgery Association (EAST) recommends secondary LMWH for patients with multiple traumatic injuries as an anticoagulation therapy. The level III recommendation is that PCD alone can be used in patients with partial head injuries. 30 The 9th ACCP guidelines only suggest the "use of low dose unfractionated heparin (LDUH), LMWH, or mechanical prophylaxis over no prophylaxis" in major trauma patients and give this recommendation the lowest grade of evidence. 31 Based on the latest guidelines for the prevention of traumatic DVT, in this study, the clinical effectiveness of LMWH combined with PCD was analyzed and compared with that of LMWH alone; the results show that LMWH should be used in trauma patients. 32 Obviously, VTE is one of the main problems in trauma patients. As demonstrated in the extensive literature on preventing VTE in trauma patients, there is currently a lack of high-quality clinical studies supporting the selection of VTE preventions methods by clinicians for this group of patients. Although there is no preventive method that can completely prevent VTE, it is clear that without precautionary measures, the incidence of DVT is higher, which may increase the risk of VTE-related morbidity and mortality. Although LMWH anticoagulation was used in all patients in this study, the best VTE prevention strategy in trauma patients remains controversial. Large-scale, randomized prospective clinical studies are needed to provide evidence regarding the optimal clinical VTE prevention measures in clinical practice.
This study has some limitations, including the retrospective data collection of case records from outpatient databases, a small sample size, and a short study period. Further research with a longer study period is needed to collect more cases. Multicenter follow-up in a prospective study is needed. In addition, the changes in index values at multiple time points, the risk of thrombosis over a long period of time, and different doses of drugs need to be investigated. Embedding the VTE risk assessment scale in electronic medical records and guiding intervention based on those scores can effectively improve VTE prevention and control strategies and reduce the incidence of VTE in the hospital. Although angiography is the gold standard for diagnosing DVT, vascular Doppler ultrasound was used in this study because of its noninvasive nature and relatively low cost. In addition, our records allowed the calculation of infection rates during hospitalization. The advantage of this study is that the results provide clinical evidence for clinicians selecting anticoagulation regimens for patients with multiple traumatic injuries. Declarations up the patients. Hui-ming Zhong and Guang-feng Zhao prepared the gures and tables. Nan Li performed the statistical analyses. All the authors interpreted the data and contributed to the preparation of the manuscript. Peng-chao Guo, Nan Li and Mao Zhang contributed equally to this manuscript. The authors read and approved the nal manuscript.

Funding
There is no funding received.

Availability of data and materials
The data and materials contributing to this article may be made available upon request by sending an email to the rst author.

Ethics approval and consent to participate
This study was approved by the Medical Ethical Committee of our hospital with the participants' written informed consent (ZEYYMECT:L20195236), and all the participants had written the informed consent.

Consent for publication
Consent to publish was obtained from the patient detailed in this study.
Competing interests  Postoperative VAS scores were lower than those before surgery. The incidence of DVT in the LMWH+PCD group was signi cantly lower than that in the LMWH group.  There was a slight difference in blood loss between the two groups. OR, odds ratio; 95% CI, 95% con dence interval. In case of statistical signi cance (P < 0.05), "1"represents the occurrence of statistical events. "1 a ", female; "1 b ", past medical history of anticoagulant use; "1 c ", LMWH alone; "1 d ", clavicle mild fracture, "e", P<0.05. LMWH+PCD is more e cient for preventing DVT. OR, odds ratio; 95% CI, 95% con dence interval. In case of statistical signi cance (P < 0.05), "1"represents the occurrence of statistical events. "1 a ", female; "1 b ", past medical history of anticoagulant use; "1 c ", clavicle mild fracture; "1 d ", LMWH alone; "e\f\g", P<0.05. The incidence of DVT with LMWH alone is 1.37 times that with LMWH+PCD application. The test result variable(s) age, Caprini score, hospital stay, GCS, blood loss, age, APTT, PLT have at least one tie between the positive actual state group and the negative actual state group. Statistics may be biased. a, b The area of APTT or PLT under the curve is greater than 0.6.