Feasibility and Safety of a Novel Leg Exercise Apparatus for Venous Thromboembolism Prophylaxis After Total Joint Arthroplasty of the Lower Extremities

Kenta Tanaka University of Tsukuba Faculty of Medicine: Tsukuba Daigaku Igaku Iryokei Yukiyo Shimizu (  shimiyukig@md.tsukuba.ac.jp ) University of Tsukuba https://orcid.org/0000-0001-7491-4516 Hiroshi Kamada University of Tsukuba Faculty of Medicine: Tsukuba Daigaku Igaku Iryokei Shizu Aikawa Tsukuba Medical Center, Department of Cardiovascular surgery Hajime Mishima University of Tsukuba Faculty of Medicine: Tsukuba Daigaku Igaku Iryokei Akihiro Kanamori University of Tsukuba Faculty of Medicine: Tsukuba Daigaku Igaku Iryokei Tomofumi Nishino University of Tsukuba Faculty of Medicine: Tsukuba Daigaku Igaku Iryokei Masataka Sakane Tsukuba Gakuen Hospital, Depart of Orthopaedic Surgery Naoyuki Ochiai Kikkoman General Hospital, Department of Orthopaedic Surgery Masashi Yamazaki University of Tsukuba Faculty of Medicine: Tsukuba Daigaku Igaku Iryokei


Background
Venous thromboembolism (VTE) is a medical condition that includes deep vein thrombosis (DVT) and pulmonary embolism (PE) [1,2]. Orthopedic surgery of the lower limbs constitutes a high-risk factor for VTE: patients undergoing total hip and knee arthroplasty show an incidence of this complication of approximately 0.6-1.5% [3,4]. DVT of the lower limbs usually develops in the calf veins, especially those in the soleus muscle, and is a frequent source of pulmonary emboli [5][6][7].
The guidelines of the American College of Chest Physicians (ACCP) [8] and American Academy of Orthopaedic Surgeons (AAOS) [9] recommend mechanical and pharmacological thromboprophylaxis in patients undergoing orthopedic surgery. Speci c measures to prevent lower extremity VTE include mechanical graduated compression stockings (GCS), intermittent pneumatic compression (IPC), and early ambulation, in addition to the administration of antithrombotic agents. Mechanical prophylaxis aims to avoid blood ow stasis, which is considered the main risk factor for DVT in "Virchow's triad" [10,11]. IPC, which involves applying pressure on the foot and crus, has proven effective for the prevention of DVT in many clinical trials [12][13][14]. Additionally, voluntary lower extremity movements, such as those involved in early ambulation and active ankle exercising, are recommended based on their strong bene t on blood ow [5,8.15]. In this regard, using ultrasonography, Sochart and Hardinge [16] evaluated the venous blood ow during active ankle exercises and found a signi cant increase in volume and speed during automatic exercises, compared to bed rest or externally assisted exercises. Moreover, exercises involving ankle plantar exion and dorsi exion combined with varus-valgus angulation have a stronger effect on blood ow, compared to simple ankle plantar exion or dorsi exion exercises [16].
However, postoperative patients rarely perform lower extremity exercises voluntarily, even after being encouraged to do so; therefore, GCS and IPC are generally used as perioperative measures for mechanical VTE prophylaxis [12]. Many studies have con rmed the effectiveness of GCS and IPC in the prevention of thrombosis, but complications, such as lower limb pain or skin disorders, may result from prolonged usage [17].
Therefore, we developed a novel leg exercise apparatus (LEX) to facilitate active leg movement during the early postoperative period [18][19][20][21]. In a previous study on healthy adults, we found that LEX exercising is more effective than IPC alone for improving blood ow [18]. Moreover, other studies of our authorship have shown that LEX exercising of the lower extremities is signi cantly more effective in activating muscles of the lower limbs and improving blood ow than performing in-bed exercises without an assisting device [19].
The purpose of this study was to evaluate the feasibility and safety of LEX for VTE prophylaxis after total joint arthroplasty of the lower limbs.

Participants
This prospective study included 20 subjects (4 men, 16 women) who underwent total joint arthroplasty of the lower extremities for the rst time between November 2014 and January 2016 at the University of Tsukuba Hospital. The operations included 19 total hip arthroplasties (THAs) and 1 total knee arthroplasty (TKA) ( Table 1). Exclusion criteria were as follows: 1) history of DVT in the lower extremity or current anticoagulant treatment; 2) history of cerebral aneurysm or dissecting aortic aneurysm (conditions which may be aggravated by an increase in blood pressure); 3) uncontrolled hypertension; 4) serious conditions (e.g., heart, liver, and/or kidney disease); 5) history of malignancy within 5 years before the day of consent or suspicion of current malignancy; 6) con rmed or suspected pregnancy.

LEX
The LEX [18][19][20][21] is a novel device originally developed to perform in-bed lower extremity exercises in patients with impaired mobilization. The device has been designed so that exercises can be performed in a supine position without diminishing the positive effect of ankle plantar exion and dorsi exion on blood ow.
The LEX is equipped with two (left/right) pedals and a movement control mechanism. The movement control mechanism comprises the base, main shaft, arm, pedal adjuster, and hooks ( Fig. 1). The base is placed on the bed on which the patient lies, and xed to the backboard of the bed using the hooks. The main shaft is placed at the center of the base. A straight arm is placed orthogonally at the top end of the main shaft, and pedals are attached to both ends of this arm. Since the arm rotates with the main shaft as the center, if the pedal on one side is pressed, the pedal on the other side moves. The foot is xed to the pedal using a leather sole. The pedal allows ankle movements up to 30° dorsi exion, 60° plantar exion, 30°s ubtalar inversion, and 20° eversion, as the arm rotates. No extra load is added. LEX does not inhibit ankle movement, since in healthy individuals, the normal ankle movement ranges are 20° dorsi exion, 45° plantar exion, 30° varus, and 20° valgus. Moreover, since there is no frictional resistance between the leg and the bed, the device enables a smooth exercise performance that engages the knee and hip joints. Arch supports are placed inside each sole, enabling pressure to be applied to the plantar venous plexus as the pedals are pressed (Fig. 1). We intended to maintain knee exion position to allow contraction of the soleus, the most fruent site of DVT in the lower leg [5,6], to be predominant over that of the gastrocnemius; usually, with the knee in the extension position, the gastrocnemius contracts while the ankle plantar exor bends more than the soleus. Exercising with the LEX is shown in detail in Additional le 1.
Finally, based on the ndings of Sochart and Hardinge [16], the design of the LEX included pedals that enable complex exercises involving subtalar eversion or inversion in addition to dorsi exion and plantar exion.

Primary outcomes
The primary aims of the study were to determine the feasibility and safety of performing exercises using the LEX. To this purpose, we measured the degree of pain and fatigue and the effect on circulatory dynamics related to the use of the LEX. Pain and fatigue were assessed using the visual analog scale (VAS) and Borg scale [22], respectively ( Table 2). Pain was measured using a specialized scale, and fatigue was assessed using a questionnaire. Pulse rate and blood pressure were assessed before and after exercise, and blood oxygen saturation level (SpO 2 ) was monitored during exercise.

Secondary outcome
The secondary outcome was the occurrence of DVT assessed using ultrasonography of the lower extremities and blood tests including D-dimer levels, both performed within seven days after surgery.

Intervention
Exercise was performed using the LEX device placed as explained above; the body position was adjusted using a specialized leg support (Fig. 2) to attain a knee exion angle of 30° in the supine position.
The pace was set at 30 times/minute. Exercise sessions lasted for 5 minutes, and were repeated 4 times per day; in this regard, the study by Yamashita et al. [23], which investigated assisted exercise involving dorsi exion and plantar exion of the foot in postoperative patients, was used as a reference. Exercises were performed from postoperative days 1 to 7.
All patients participated in preoperative practice exercises using LEX until they were comfortable with the device, to ensure an appropriate postoperative use. A physician supervised the rst postoperative exercise session and provided instructions to conduct the exercise correctly and avoid any inappropriate movements or contraindicated positions.
Compliance with exercise was con rmed through a counter included in the device that recorded the number of times the pedals were pressed.

Perioperative management and assessment
Assessments and exercises were conducted according to the procedure shown in Table 3. Edoxaban was used in two cases according to risk analysis of VTE. (One is high BMI and the other is a TKA patient.) Thromboprophylaxis including mechanical and pharmacological methods are shown in Table 1.   Cancellation criteria If any of the following events occurred during the study period, the patient was excluded and the speci c reason and other relevant data were recorded in the case report: 1) adverse events that rendered continuation of the study di cult; 2) request from the patient to opt out of the study; 3) withdrawal of consent for participation (in this case, all data would be excluded); 4) con rmed critical or continued noncompliance to the study plan; and 5) other reasons the physician deemed su cient for cancellation.
The physician recorded all the content of the survey conducted until cancellation.

Results
Overall, 16 of 20 patients completed the exercise regimen of 7 days, while 4 dropped out. All four subjects who dropped out had undergone THA; three of them refused to initiate exercise due to leg pain related to surgery and the remaining one stopped exercising due to physical di culties after 2 days. No severe adverse events occurred during the study period.
In the rst exercise session (on the day of operation), the average VAS score was 38.5 mm (range, 0−85 mm), and the average Borg scale score was 13 (range, 7−19).
Blood pressure was assessed in 15 patients. An increase in systolic blood pressure before or after exercise was observed in seven, while an increase in diastolic blood pressure before or after exercise was observed in six. The maximum difference between systolic blood pressure before and after exercise was 11 mmHg, whereas for diastolic blood pressure, the corresponding value was 19 mmHg (Fig. 3).
Pulse rate was evaluated in 17 cases. The pulse rate increased in 14 subjects, with a maximum of 117 beats/min (bpm), and the maximum difference before and after exercise was 17 bpm (Fig. 4).
In some cases, a maximum reduction of 2% in SpO 2 was observed, but there were no complaints of breathing di culty (Fig. 5).
Regarding the secondary outcome, no DVT of the lower extremities was con rmed during postoperative week 1. The average preoperative D-dimer level was 0.85 µg/mL (range, 0.4−2.2), whereas the corresponding value in postoperative week 1 was 7.93 µg/mL (range, 2.0−13.8).

Discussion
In this study, the LEX was successfully used in patients after total joint arthroplasty of the lower extremities, and no severe adverse events occurred. Overall, 80% of subjects completed the 1-week protocol. This rate of acceptance is comparable to that of IPC, which is reported to be 81% [17,25].
With respect to the safety of a device that affects circulatory dynamics through exercise, none of the measured values for systolic or diastolic blood pressure, pulse rate, and SpO 2 exceeded the exercise cancellation standard set by the Japanese Association of Rehabilitation Medicine [26] (Table 5). This suggests that it is unlikely that exercising with the LEX will cause any hemodynamic complication. Criteria are according to the guidelines for safety management and promotion in rehabilitation medicine, proposed by the Japanese Association of Rehabilitation Medicine [26].
According to the ACCP [8] and AAOS guidelines [9] the main aim of thromboprophylaxis is the prevention of lethal or symptomatic VTE; compared to previous guidelines, anticoagulant therapy indications are more restricted due to the risk of major bleeding complications and cost-effectiveness issues. Speci cally, the AAOS guidelines clarify that pharmacological prophylaxis is only recommended for patients who are not at elevated risk of bleeding after surgery [27]. This guideline recommends mechanical prevention for asymptomatic VTE in patients with high risk of bleeding.
Sashi et al. [28] studied the VTE rates in total hip and knee arthroplasty between 2002 and 2011 and found that, despite a slight decrease in VTE incidence related to both surgeries, PE rates remained stable. In detail, the overall median DVT and PE incidences in the cited study were, respectively, 0.40 and 0.23 in primary THA, and 0.62 and 0.34 in primary TKA. Chan et al. [29] conducted a systematic review of randomized controlled trials comparing the rates of VTE and bleeding due to pharmacological prophylaxis after THA or TKA and found an overall VTE rate of 0.99%, which is similar to that of previous studies. However, the postoperative bleeding rate was 3.44%, more than three times the VTE rate. Fuji et al. (4) studied that the development, prophylaxis, and treatment of VTE and bleeding events in 36,947 patients who had undergone orthopedic surgeries of the lower extremities from 2008 to 2013 using a healthcare database. They reported that the incidences of DVT, PE, and bleeding were 1.3%, 0.2%, and 1.0% for TKA and 0.9%, 0.2%, and 1.1% for THA. Therefore, the rate of bleeding complications due to antithrombotic therapy was higher than that of PE in both studies.
In turn, Tsuda et al. [30] reported the incidence of DVT using mechanical prophylaxis solely. Their study showed that among 184 cases of hip surgery receiving only IPC and GCS (patients with trauma were excluded), 5% were diagnosed with distal thromboses according to ultrasonic tests conducted in postoperative week 3.
Considering the relatively high incidence of bleeding events, the current opinion is that individual risk analyses for VTE are needed [3,31]. In this regard, we are considering to design study protocols using LEX as mechanical prophylaxis according to VTE risk strati cation, in addition or not to pharmacological prophylaxis.
Finally, in view of the current pandemic, it is worth noting that coronavirus disease 2019 (COVID- 19) increases the risk of VTE signi cantly [32,33]. In addition to this, self-isolation periods prior to lower limb arthroplasty may increase the thrombotic risk further [31]. In a period in which close contact should be reduced to a minimum, mechanical devices which enable self-managed exercise, such as the LEX, may be valuable for both a safe DVT prophylaxis and an appropriate rehabilitation.
There were some limitations in this study. First, the sample size was too small to evaluate effectiveness in the prevention of VTE. However, considering that DVT after arthroplasty of the lower limbs is very common, it may be signi cant that no case of DVT was observed in our cohort. Second, we analyzed LEX use in patients who were independent in their ADLs before surgery and therefore could resume ambulation relatively soon after surgery. Third, patients in our study received other prophylactic methods such as IPC and anticoagulants, which can account for the absence of VTE cases. We are considering to evaluate the e cacy of the LEX in combination with established strategies to prevent VTE in bedridden patients.

Conclusions
In this study, the use of the LEX after total joint arthroplasty of the lower extremities was associated with a very good acceptance and no severe adverse events. Additionally, no cases of DVT were observed postoperatively. Our results suggest that a study protocol of lower extremity exercises using the LEX for Page 13/20 VTE prophylaxis in this group of patients is safe and feasible. We believe that LEX may be useful in patients with high VTE and bleeding risks, such as those with malignancies or undergoing spine surgery, in whom the use of pharmacological prophylaxis may be of concern. Additionally, the LEX has a potential educational value for patients and medical experts since it could be used to teach the importance of active leg exercises for thromboembolism prophylaxis.    Exercise position using the LEX with a specialized leg support Figure 3 Blood pressure before and after exercise using the LEX Page 19/20

Figure 4
Pulse rate before and after exercise using the LEX Figure 5