Abstract
Objective
Many guidelines indicate that continuous use of anticoagulant drugs reduces the incidence of venous thrombus (VT), but no studies show the effect on the incidence of symptomatic venous thrombus (SVT) in total knee arthroplasty (TKA) patients after discharge. This study aimed to investigate whether it is necessary to apply anticoagulants to TKA patients after discharge.
Methods
Patients who met the exclusion criteria requirement, underwent TKA by the same surgical team and received anticoagulant therapy after the operation were eligible for the study. Finally, a total of 567 TKA patients were recruited as participants. The patients were divided into two groups. The patients in group A were taken low molecular heparin for 5–10 days after surgery, which included but was not limited to low molecular weight heparin calcium injection (0.4 mL, ih, Qd), calcium dioxin injection (0.6 mL, ih, Qd), or enoxaparin sodium injection (0.4 mL, ih, Qd), and the patients needed to continue oral anticoagulant drug (10 mg, po, Qd) for 7–21 days after discharge. The patients in group B only took low molecular heparin 5–10 days after surgery and no treatment after discharge. The baseline characteristics of patients, total complications of SVT include lower limb vascular pain (LLVP), lower limb vascular pain no fester (LLVPNF), lower limbs swelling (LLS), lower limb fester (LLF), and death by thrombosis (DT), bleeding and mortality following discharged were compared between two groups.
Results
The study showed that the incidence of SVT patients had no significant difference between the two groups (p = 0.489). Moreover, the incidence of LLVP (p = 0.265), LLS (p = 0.84), LLVPNF (p = 0.213), LLF (p = 0.907), DT (p = 0.907), death from other causes, and bleeding (p = 0.323) had no significant differences between the two groups. However, the incidence of SVT in patients with smoking (p = 0.0001 or 0.0011) or drinking (p = 0.0002 or 0.0001) was significantly increased.
Conclusion
There is not enough evidence showing that the TKA patients given anticoagulants after discharge had benefits in decreasing the risk of SVT. Furthermore, smoking and drinking would significantly increase the risk of SVT in TKA patients.
Keywords: Anticoagulant drugs, Anticoagulants, Asymptomatic venous thrombosis, Symptomatic venous thrombosis, Total knee replacement, Venous thromboembolic disease
Anticoagulants were given after knee surgery to prevent deep vein thrombosis until discharge. No further treatment with anticoagulants after discharge from the hospital.
Introduction
Venous thromboembolic disease (VTE) includes deep venous thrombosis (DVT) and pulmonary embolism (PE), which is a severe and potentially life‐threatening complication after major orthopedic surgery, especially total knee arthroplasty (TKA). 1 Without prevention, 50% to 70% of TKA patients would suffer from VTE, significantly decreasing to 2% to 30% after prevention. Therefore, adopting effective prevention methods for patients after TKA would dramatically reduce VTE risk, relieve pain, and decrease medical costs. 2
Guidelines recommend that the patients after TKA apply at least one of the following antithrombotic drugs to prevent VTE, such as low‐molecular‐weight heparin, fondaparinux, dabigatran, apixaban, rivaroxaban, and so on. 3 Many studies on preventing VTE after TKA surgery showed that anticoagulant treatment should be used for at least 10 to 14 days in the inpatient period and up to 35 days in the outpatient period. 4 These studies suggested TKA patients should continue using the anticoagulant after discharge. These studies did not strictly distinguish between SVT and ASVT during the study, and it is necessary to make a strict distinction between different thrombosis types as they have a more significant impact on the prognosis of patients and the use of anticoagulant drugs. The symptomatic venous thrombosis (SVT) complications include lower limb vascular pain (LLVP), lower limb vascular pain no fester (LLVPNF), lower limb swelling (LLS), lower limbs fester (LLF), death by thrombosis (DT), and asymptomatic venous thrombosis (ASVT) had no other discomfort symptoms. In this study, we excluded the ASVT patients and only employed the patients who were SVT, and the incidence of SVT after discharge among TKA patients in our clinical observation was not the same as in the report above. Therefore, we designed the following experiment according to strict nadir criteria to further verify whether anticoagulants should be continued after discharge of patients after TKA.
The purposes of the study were: (i) to study whether there is any benefit in continuing anticoagulants after discharge of patients after knee arthroplasty; (ii) to classify symptomatic and asymptomatic thrombosis; and (iii) to study the effect of some factors such as smoking, alcohol consumption, age underlying medical history on asymptomatic thrombosis in patients after knee arthroplasty.
Methods
Patients
The data were collected from the Lanzhou University Second Hospital in China. The inclusion criteria were: (i) the patients who first underwent TKA; (ii) the surgery was performed by the same surgical team; and (iii) the patients received anticoagulant therapy after the operation.
Exclusion criteria included: (i) no toleration for oral anticoagulants; (ii) the positive of two lower limbs with deep vein color ultrasound; (iii) severe injury to liver and kidney; (iv) oral antithrombotic or anticoagulant drugs before TKA;(v) history of blood clots; (vi) taking medicine that affects the coagulation mechanism; (vii) any bleeding events; history of acute intracranial disease or hemorrhagic stroke; (viii) major surgery, trauma, uncontrolled hypertension or myocardial infarction within the past 3 months; (ix) gastrointestinal or urogenital bleeding or ulcer disease within the past 6 months; and (x) concomitant long‐acting non‐steroidal anti‐inflammatory drug therapy (also contraindicated during study treatment). There is no restriction on age or gender in this study. Finally, a total of 407 TKA patients (case history from January 2013 to December 2016) were involved in this study (Figure 1).
Fig. 1.
Flowchart of the study population.
We propose using data to investigate whether it is necessary to continue to apply anticoagulants after discharge on TKA patients. The collected data was only used for this retrospective research study.
Treatment Regimens
The patients were divided into two groups (groups A and B). The patients in group A were taken low molecular heparin for 5–10 days after surgery, which included but was not limited to low molecular weight heparin calcium injection (0.4 mL, ih, Qd), Calcium dioxin injection (0.6 mL, ih, Qd), or Enoxaparin sodium injection (0.4 mL, ih, Qd), and the patients needed continued oral anticoagulant drug (10 mg, po, Qd) for 7–21 days after discharge. The patients in group B only took low molecular heparin 5–10 days after surgery and no treatment after discharge.
After surgery, all indwelling anesthetic catheters were removed. Twelve hours later, patients were given low molecular heparin and encouraged to leave the bed. On the 3rd day after the operation, functional exercises were performed and insisted until the patient left the hospital. All patients were not allowed physical anti‐coagulation therapy, including elastic socks, mechanical pumps, and other drugs that may influence hemodynamics. However, concomitant treatment with low‐dose aspirin (<160 mg) and selective cyclooxygenase‐2 inhibitors was allowed during the treatment period. Elastic compression stockings were permitted, but pneumatic compression devices were prohibited.
Efficacy and Safety Evaluation
The efficacy and safety of using anticoagulant drugs by TKA patients after discharge were evaluated through the following indexes: the incidence of bleeding and death during the in‐hospital and out‐hospital; the incidence of LLVP, LLS, LLVPNF, LLF, DT, etc. All data were determined based on cases, including medical records and telephone return visit services. The information included age, gender, height, smoking, drinking, underlying disease (including hypertension, diabetes, hyperlipidemia), anticoagulants, support activity, and so forth. The above information was completed by two partners independently. The average follow‐up period was 3.4 ± 0.87 years.
Other Covariates
Based on current research on SVT, other covariates and the following potential demographic confounders were included in this study: age (<65, or ≥65 years); sex (male, female); body mass index (BMI) is divided into four groups (≤18.5; ≥18.5, ≤23.9; ≥24, ≤27.9; ≥28 kg/m2); with underlying disease (including hypertension, diabetes, and hyperlipidemia) or without underlying disease; smoking or no smoking; drinking or no drinking; course of treatment (≤5 days, 5–10 days, ≥10 days).
Statistical Analysis
Baseline data were analyzed using descriptive statistics in which the parameters were assessed, including age, gender, body mass index (BMI), underlying disease (including hypertension, diabetes, and hyperlipidemia), physical therapy, smoking, drinking, and SVT. The efficacy analyses included all patients with SVT and total complications of SVT during the follow‐up period. The incidences of LLVP, LLS, LLVPNF, LLF, or DT during the follow‐up were calculated using the observed dates in each follow‐up group. Group comparisons were tested using a t‐test and the Cochran–Mantel–Haenszel chi‐square test stratified by unilateral or bilateral surgery. Incidences of SVT and mortality during the follow‐up period have also calculated the frequency of bleeding events, and individual bleeding complications were estimated between the two groups. Differences between groups were analyzed with the Cochran–Mantel–Haenszel chi‐square test stratified by unilateral or bilateral surgery. The data were analyzed by SPSS 18.0. The significant difference was set at a level of 0.05 among groups.
Results
Study Populations
Genders, underlying disease, physical therapy, smoking, and drinking were included in this study. Demographic characteristics were similar in each group except for a slight excess of women (Table 1). The number of women is much higher than males in each group because women are more likely to develop knee arthritis. Because the thromboembolic was affected by the activity time, the mean time to the first dose of medication on the 1st day was 21.5 h for all treated patients. There was a significant difference in the duration of anticoagulant treatment between the two groups (p < 0.001), with patients treated with oral anticoagulants after discharge being treated for a significantly extended period than those in the group treated with a single LWMH.
TABLE 1.
Baseline characteristics of the 407 patients in the safety population
| Characteristic | A (low molecular heparin + rivaroxaban) (N = 180)* | B (low molecular heparin) (N = 227) † | Test statistic | p‐value |
|---|---|---|---|---|
| Female sex, N (%) ‡ | 149 (60.56) | 187 (82.4) | χ 2 = 0.011 | 0.916 |
| Age – years | ||||
| Age – years (mean ± SD) | 59.4 ± 11.3 | 62.47 ± 10.83 | T‐test | 0.872 |
| ≥65 | 72 (40) | 106 (46.6) | χ 2 = 2.98022 | 0.176 |
| <65 | 108 (60) | 121 (53.4) | ||
| Body‐mass index § | ||||
| Body‐mass index § (mean ± SD) | 27.35 ± 2.59 | 32.48 ± 3.54 | T‐test | 0.872 |
| ≥28 | 49 (27.22) | 65 (28.63) | χ 2 = 16.86834 | 0.752 |
| 24–27.9 | 57 (31.67) | 78 (34.36) | χ 2 = 0.329 | 0.566 |
| 18.5–23.9 | 74 (41.11) | 84 (37) | χ 2 = 0.713 | 0.398 |
| Underlying disease (include hypertension, diabetes and hyperlipidemia) | 66 (36.67) | 72 (31.72) | χ 2 = 1.097 | 0.295 |
| Hypertension | 27 (15) | 24 (10.57) | χ 2 = 0.581 | 0.446 |
| Diabetes | 21 (11.67) | 33 (14.54) | χ 2 = 0.718 | 0.396 |
| Hyperlipidemia | 18 (10) | 15 (6.61) | χ 2 = 1.55 | 0.213 |
| Physical therapy | 6 (3.33) | 5 (2.2) | χ 2 = 0.488 | 0.485 |
| Smoking | 7 (3.89) | 17 (7.49) | χ 2 = 2.345 | 0.126 |
| Drinking | 8 (4.44) | 14 (6.17) | χ 2 = 0.583 | 0.445 |
| Course of treatment† | ||||
| Course of treatment** (mean ± SD) | 12.4 ± 2.47 | 6.53 ± 1.21 | T‐test | 0.0002 |
| ≤5 days | 0 (0) | 152 (66.96) | FET | 0 |
| 5–10 days | 38 (21.11) | 75 (33.04) | χ 2 = 30.13699 | 0.007 |
| ≥10 days | 142 (78.89) | 0 (0) | FET | 0 |
The patients were taken to hypodermic injection of low molecular heparin after surgery, then continued oral rivaroxaban after discharge.
The patients were only taken to hypodermic injection of low molecular heparin.
The number of females is much higher than the number of males.
The body‐mass index is the weight in kilograms divided by the square of the height in meters; FET Fisher's exact test.
The course of treatment was significantly different between the A and B groups.
Efficacy
The total complications of SVT (LLVP, LLS, LLVPNF, LLF, or DT) occurred in about 2.78% (five of 180) of patients in group A and 1.76% (four of 227) in group B. There was non‐inferior between the two groups (χ 2 = 0.479; p = 0.489) (Table 2). The superior efficacy analyses were also consistently employed in the studies of subgroups defined according to age, gender, underlying disease (including hypertension, diabetes, and so on), physical therapy, body mass index, smoking, and drinking (Table 2). All factors except smoking and drinking did not influence the incidence of SVT. In all groups, the patients with smoking or drinking were more likely to develop SVT than those without smoking or drinking. The incidence of SVT in group A is 28.57% for smoking and 25.00% for drinking, but only 1.73% for no smoking and 1.57% for no drinking. Same as group A, the incidence of SVT in group B is 11.76% for smoking and 14.29% for drinking, but only 0.96% for no smoking and 0.94% for no drinking (Table 2). There is a significant difference between the smoking and no smoking groups and the drinking and no drinking groups (p < 0.001).
TABLE 2.
Efficacy outcomes during the treatment period
| Parameter | A (low molecular heparin + rivaroxaban) (N = 180)* | B (low molecular heparin) (N = 227) † | Test statistic | p‐value |
|---|---|---|---|---|
| Total complications of symptomatic venous thromboembolism, N (%) | 5 (2.78) | 4 (1.76) | χ 2 = 0.479 | 0.489 |
| Sex N (%) | 5 (2.78) | 4 (1.76) | χ 2 = 0.479 | 0.489 |
|
Female |
5 (3.36) | 3 (1.6) | χ 2 = 1.094 | 0.295 |
| N (%) (the total number of females) | ||||
|
Male |
0 (0) | 1 (2.5) | FET | 0.888 |
| N (%) (the total number of males) | ||||
| Female vs. male ‡ | FET | 0.664 | ||
| FET | 0.786 | |||
| Age‐years | 5 (2.78) | 4 (1.76) | χ 2 = 0.479 | 0.489 |
| ≥65, N (%) | 3 (4.17) | 2 (1.89) | χ 2 = 0.816 | 0.366 |
| <65, N (%) | 2 (1.22) | 2 (0.83) | χ 2 = 0.013 | 0.909 |
| ≥65 vs. <65 ‡ | χ 2 (A) = 0.898 | 0.343 | ||
| χ 2 (B) = 0.018 | 0.894 | |||
| Body‐mass index | 5 (2.78) | 4 (1.76) | χ 2 = 0.479 | 0.489 |
| ≥28, N (%) | 0 (0) | 0 (0) | NA | NA |
| 24–27.9, N (%) | 2 (3.51) | 2 (2.56) | χ 2 = 0.102 | 0.749 |
| 18.5–23.9, N (%) | 3 (4.05) | 2 (2.38) | χ 2 = 0.359 | 0.549 |
| 24–27.9 vs. ≥28 ‡ | FET | 0.543 | ||
| FET | 0.54 | |||
| 18.5–23.9 vs. ≥28 ‡ | FET (A) | 0.424 | ||
| FET (B) | 0.593 | |||
| 24–27.9 vs. 18.5–23.9 ‡ | χ 2 (A) = 0.026 | 0.871 | ||
| χ 2 (B) = 0.005 | 0.94 | |||
| Underlying disease, N (%) (the total number of underlying disease) ‡ | 4 (6.06) | 4 (5.56) | χ 2 = 0.016 | 0.899 |
| Hypertension, N (%) (the total number of hypertension y) | 1 (3.7) | 0 (0) | FET | 0.529 |
| Diabetes, N (%) (the total number of diabetes) | 1 (4.76) | 2 (6.06) | χ 2 = 0.165 | 0.684 |
| Hyperlipidemia, N (%) (the total number of hyperlipidemia) | 2 (11.11) | 2 (13.33) | FET | 0.374 |
| ‐Physical therapy, N (%) (the total number of physical therapy) | 0 (0) | 0 (0) | NA | NA |
| No physical therapy, N (%) (the total number of no physical therapy) | 5 (2.87) | 4 (1.8) | χ 2 = 0.504 | 0.478 |
| Physical therapy vs. no physical therapy ‡ | FET (A) | 0.4 | ||
| FET (B) | 0.763 | |||
| ‐Smoking, N (%) (the total number of smoking) | 2 (28.57) | 2 (11.76) | χ 2 = 1.008 | 0.315 |
| ‐No smoking, N (%) (the total number of no smoking) | 3 (1.73) | 2 (0.96) | χ 2 = 0.45 | 0.502 |
| Smoking vs. no smoking ‡ | χ 2 (A) = 17.94 | 0.0001 | ||
| χ 2 (B) =10.62 | 0.0011 | |||
| ‐Drinking, N (%) (the total number of drinking) | 2 (25) | 2 (14.29) | χ 2 = 0.393 | 0.5308 |
| ‐No drinking, N (%) (the total number of no drinking) | 3 (1.57) | 2 (0.94) | χ 2 = 0.481 | 0.488 |
| Drinking vs. no drinking ‡ | χ 2 (A) = 15.3 | 0.0001 | ||
| χ 2 (B) =13.52 | 0.0002 | |||
| Course of treatment | 5 (2.78) | 4 (1.76) | χ 2 = 0.479 | 0.489 |
| ≤5 days, N (%) | 3 (1.97) | χ 2 (B) = 0.119 | 0.73 | |
| 5–10 days, N (%) | 1 (1.33) | |||
| 5–10 days, N (%) | 1 (2.63) | χ 2 (A) = 0.0038 | 0.951 | |
| ≥10 days, N (%) | 4 (2.82) | |||
Abbreviation: FET, Fisher's exact test.
The patients were taken to hypodermic injection of low molecular heparin after surgery, then continued oral rivaroxaban after discharge.
The patients were only taken to hypodermic injection of low molecular heparin.
The number of females is much higher than the number of males.
In this study, nine patients had developed SVT, including one patient with DT in group B, three patients only had LLS (1 in group A, 2 in group B), and four patients with LLVPNF (3 in group A, 1 in group B) and one patient with LLF in group A. The incidence of LLVP was 2.22% (4/180) in group A and 0.88% (2/227) in group B (Table 3). The testing procedure did not show an apparent difference between groups in LS, LLVPNF, LLF, LLVP, and DT (p > 0.05, Table 3).
TABLE 3.
Efficacy and clinical end‐points
| Characteristic | A (low molecular heparin + rivaroxaban) (N = 180)* | B (low molecular heparin) (N = 227) † | Test statistic | p‐value |
|---|---|---|---|---|
| LLVP, N (%) | 4 (2.22) | 2 (0.88) | χ 2 = 1.243 | 0.265 |
| LLS, N (%) | 1 (0.56) | 2 (0.88) | χ 2 = 0.041 | 0.84 |
| LLVPNF, N (%) | 3 (1.67) | 1 (0.44) | χ 2 = 1.551 | 0.213 |
| LLF, N (%) | 1 (0.56) | 0 (0) | FET | 0.907 |
| DT, N (%) | 0 (0) | 1 (0.44) | FET | 0.907 |
| Death from other causes, N (%) | 0 (0) | 0 (0) | NA | NA |
Abbreviations: DT, death by thrombosis; FET, Fisher's exact test; LLF, lower limbs fester; LLS, lower limbs swelling; LLVP, lower limbs vascular pain; LLVPNF, lower limbs vascular pain no fester.
The patients were taken to hypodermic injection of low molecular heparin after surgery, then continued oral rivaroxaban after discharge.
The patients were only taken to hypodermic injection of low molecular heparin.
Safety
In this study, 16 patients experienced bleeding events, with nine (5%) in group A and seven (3.08%) in group B. There was no significant difference in bleeding events between groups A and B (p = 0.323). Further analysis also found three patients developed significant bleeding (two in group A, one in group B), and six were minor bleeding (four in group A, two in group B). All groups had no deaths from bleeding during the follow‐up period (Table 4). We analyzed each group's impact on platelets, liver, kidney, and total bilirubin. Our findings revealed that none of the drugs exhibited a noteworthy decrease in platelet levels or a significant increase in liver, kidney, or total bilirubin levels. The safety was similar in each group.
TABLE 4.
Bleeding events during the treatment period
| Characteristic | A (low molecular heparin + rivaroxaban) (N = 180)* | B (low molecular heparin) (N = 227) † | Test statistic | p‐value |
|---|---|---|---|---|
| Any bleeding, N (%) | 9 (5) | 7 (3.08) | χ 2 = 0.976 | 0.323 |
| Major bleeding, N (%) | 2 (1.11) | 1 (0.44) | χ 2 = 0.041 | 0.839 |
| Minor bleeding, N (%) | 4 (2.22) | 2 (0.88) | χ 2 = 0.491 | 0.483 |
| Death from bleeding, N (%) | 0 (0) | 0 (0) | NA | NA |
| Platelet counts, N (%) (less than 125 × 109/L) | 2 (1.11) | 1 (0.44) | χ 2 = 0.040 | 0.839 |
| Urea elevation (more than 97 μmol/l) | 0 (0) | 1 (0.44) | FET | 0.372 |
| ALT elevation (more than 50 U/L) | 3 (1.67) | 4 (1.76) | χ 2 = 0.096 | 0.756 |
| AST elevation (more than 40 U/L) | 0 (0) | 0 (0) | NA | NA |
| TBIL elevation (more than 2 6 μmol/L) | 1 (0.55) | 2 (0.88) | χ 2 = 0.041 | 0.839 |
Abbreviation: FET, Fisher's exact test.
The patients were taken to hypodermic injection of low molecular heparin after surgery, then continued oral rivaroxaban after discharge.
The patients were only taken to hypodermic injection of low molecular heparin.
Discussion
Our study showed that there is not enough evidence showing that the TKA patients given anticoagulants after discharge had benefits in decreasing the risk of SVT. Furthermore, smoking and drinking would significantly increase the risk of SVT in TKA patients.
Patients Discharged from Hospital after Knee Arthroplasty Will Not Benefit from Anticoagulants in Terms of SVT
Previous research showed that the incidence of DVT in patients with TKA was significantly decreased after receiving anticoagulants for thromboprophylaxis. 5 , 6 But it did not classify the SVT and AVST in the previous studies. However, although SVT and ASVT are blood clots that form in the blood vessels, they differ regarding harm and presentation. The SVT will affect the everyday life of patients, but the AVST will not, so there is a great need to analyze them differently during the study. This study showed that the risk of SVT was 2.78% (five of 180) of patients in group A and 1.76% (four of 227) in group B. It has no significant difference in the incidence of SVT among the two groups (p = 0.489, Table 2). However, the guideline was published by the American Academy of Orthopedic Surgeons, which suggests using pharmacologic agents and/or mechanical compressive devices to prevent VTE in patients undergoing TKA as a moderate recommendation. 7 , 8 The American College of Chest Physicians recommends using one of the following for 10 to 35 days. 9 , 10 The incidence of VTE in patients undergoing TKA appears to decline with the above method.
Analysis of Reasons for Variances
The difference between the guidelines and our study might be due to the following reasons. First, SVT and ASVT were all found among the TKA patients. And the samples of SVT and ASVT were all included in the above research. However, the plexus venous leg muscle thrombosis, microthrombotic, and deep venous thrombosis of patients with no symptoms of discomfort did not identify as SVT. In this study, we only include SVT samples to study.
Second, the incidence of SVT was not associated with anticoagulants after discharge in this study, possibly due to race and nationalities. 11 A study on 741 patients who underwent 870 primary THAs with application of an Intermittent Pneumatic Compression device (IPCD) found that three patients (0.3%) developed DVT (detected by sonography), one patient (0.1%) developed symptomatic DVT, and one patient (0.1%) developed symptomatic PE. 12 There were no reported fatal PEs. The incidence of symptomatic DVT was significantly lower compared to that of the historical control group. 13 Many studies showed that genetic traits, such as protein C, factor V, etc., are critical in evaluating the anti‐coagulation effect. In the above‐motioned guidelines and mentioned previous studies, the patients were all from Europe and North America. 14 However, the samples in our research were all Asians from Gansu province in China who share different genetic traits.
Third, the activity time after the operation was significantly associated with the thrombosis rate. The slower the blood flow speed, the more accessible thrombosis forms. Thus, the activity time after TKA can dramatically affect the speed of blood flow. The patients in this study were encouraged to get out of bed at 12 hours after TKA, which might benefit them with a lower incidence of SVT.
Last but not least, the high altitudes of Gansu Province, China, may impact the incidence of SVT. Many studies show that high altitudes can affect the coagulation system, increasing blood PGI2 levels (exogenous prostacyclin). 12 Gansu Province was a high‐altitude area with an elevation of 1500–3500 m, which would extend the blood vessels and inhibit platelet aggregation, reducing the incidence of VTE. Other studies have come to different conclusions about thrombosis for all these reasons.
Sociological Factors Influencing SVT
Some studies 15 , 16 have shown an increased risk of VTE associated with age, but the assertion is disputable. In this study, the incidence of SVT was not influenced by age (p(A) = 0.343, p(B) = 0.894, Table 2). Meanwhile, we also compared other indicators such as gender (p(A) = 0.664, p(B) = 0.786, Table 2), underlying disease (p = 0.899, Table 2) (including hypertension (p = 0.529, Table 2), diabetes (p = 0.684, Table 2), and hyperlipidemia (p = 0.374, Table 2)), body‐mass index (24–27.9 vs. ≥28: p(A) = 0.543, p(B) = 0.54; 18.5–23.9 vs. ≥28: P(A) = 0.424, p(B) = 0.593; 24–27.9 vs. 18.5–23.9: p(A) = 0.871, p(B) = 0.94, Table 2), physical therapy (p(A) = 0.4, p(B) = 0.763, Table 2), and course of treatment (p(A) = 0.73, p(B) = 951, Table 2). All these indicators have not influenced the incidence of SVT in our study.
Interestingly, this study found that smoking and drinking significantly increased the risk of SVT (p < 0.001, Table 2). Smoking raises the risk of SVT in several aspects, such as smoking can harm blood vessels, and nicotine makes blood vessels atrophy, increasing blood viscosity to more clots that can increase the risk of SVT. 17 Meanwhile, long‐term drinking can reduce the level of Protein‐C, decreasing the anti‐blood coagulation system's function and increasing the risk of SVT. 18
Discontinuing Anticoagulants after Discharge Significantly Reduces Patients' Risk of Major Bleeding
Furthermore, the safety of anti‐coagulation, especially in TKA patients, was the orthopedic surgeons are most concerned about. If bleeding occurs in the replaced joint, that can have catastrophic consequences. This study showed that the rates of major bleeding with all treatments were lower than in previous research; 19 only three patients had significant bleeding (two in group A, one in group B). The reduced bleeding risk observed in our study may be attributed to the use of LMWH and new oral anticoagulants in our patients. When administered at the same dose, LMWH generally carries a lower risk of bleeding than regular heparin. 20 This can be attributed to its more targeted action on coagulation factor Xa rather than thrombin. 21 Furthermore, the newer oral anticoagulants demonstrate a relatively lower risk of bleeding than traditional oral anticoagulants while effectively preventing thrombosis. 22
After careful consideration, we have decided to administer oral rivaroxaban for postoperative anticoagulation therapy following the patient's discharge from the hospital. Relevant studies have indicated that rivaroxaban demonstrates comparable efficacy to apixaban and dabigatran in knee replacement surgery, 23 with no significant difference observed in bleeding risk. 24 These findings indirectly support the notion that alternative anticoagulants would not yield significantly biased results, allowing us to extend this conclusion to other new anticoagulants. Although specific studies for these novel anticoagulants have not been conducted, it is reasonable to speculate that they may exhibit similar effects in postoperative anticoagulation therapy, given the comparable efficacy and safety profile shared with rivaroxaban and traditional anticoagulants.
Limitations
This study has limitations: first, the sample size is required to be more significant. However, we performed an effect analysis of the sample size by G. Power software with β = 1, proving that the sample size satisfies the statistical requirements. Second, all patients only covered one center study, not a multicenter study in this study. All patients did not make two lower limb vein color ultrasounds, but all samples were observed with SVT so that the result accuracy would not be impacted. Third, the data collected may be incomplete or biased, and the study design does not allow for control of confounding variables.
Additionally, the results may need to be more generalizable to other populations or settings, as the study was conducted at a single tertiary referral center involving only patients from high‐altitude areas. Despite these limitations and experimental evidence, it offers good clinical evidence for using anticoagulants after TKA discharge. However, there is no sufficient evidence showing that the patients who were given anticoagulants after discharge had the benefit of decreasing the risk of SVT.
Conclusion
This study is a significant finding for an early applied anticoagulants‐operative period. Not enough evidence showed that the TKA patients given anticoagulants after discharge had benefits in decreasing the risk of SVT. Furthermore, smoking and drinking would increase the risk of SVT in TKA patients. Therefore, TKA patients might not benefit from additional anticoagulants after discharge. Again, TKA patients who discontinue the habit of smoking and drinking can reduce the risk of SVT.
Author Contributions
Xiangji Dang and Yan Liu were significant contributors to manuscript writing. Xiangji Dang, Zhilong Liu, and Xiaoyun Sheng collected the data. Xiangji Dang analyzed the data. Yan Liu revised the manuscript critically. All authors read and approved the final manuscript.
Funding Information
This study was supported by the Natural Science Foundation for Gansu Province (20CX9ZA113, 20JR10RA596, and 22JR5RA982) and the Talent Innovation and Entrepreneurship Project of Lanzhou City (2020‐RC‐43).
Ethics Statement
The study was approved by Lanzhou University Second Hospital Human Research Ethics Committee (2022A‐421) and conducted following the Declaration of Helsinki. This study was retrospective. The patient data was de‐identified, so patients were not asked for informed consent.
References
- 1. Thompson SA. Complications in orthopaedics: adult reconstruction‐E‐book. Edinburgh: Elsevier Health Sciences; 2022. [Google Scholar]
- 2. Li JW, Ma YS, Xiao LK. Postoperative pain Management in total knee arthroplasty. Orthop Surg. 2019;11(5):755–761. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Johnson SA, Jones AE, Young E, Jennings C, Simon K, Fleming RP, et al. A risk‐stratified approach to venous thromboembolism prophylaxis with aspirin or warfarin following total hip and knee arthroplasty: a cohort study. Thromb Res. 2021. Oct;206:120–127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Lieberman JR, Bell JA. Venous thromboembolic prophylaxis after total hip and knee arthroplasty. J Bone Joint Surg Am. 2021;103(16):1556–1564. [DOI] [PubMed] [Google Scholar]
- 5. Matharu GS, Kunutsor SK, Judge A, Blom AW, Whitehouse MR. Clinical effectiveness and safety of aspirin for venous thromboembolism prophylaxis after Total hip and knee replacement: a systematic review and meta‐analysis of randomized clinical trials. JAMA Intern Med. 2020;180(3):376–384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Jia J. Combined use of intravenous and topical tranexamic acid in patients aged over 70 years old undergoing total hip arthroplasty. J Orthop Surg Res. 2019. Nov 7;14(1):345. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Gurunathan U, Barras M, McDougall C, Nandurkar H, Eley V. Obesity and the risk of venous thromboembolism after major lower limb Orthopaedic surgery: a literature review. Thromb Haemost. 2022;122(12):1969–1979. [DOI] [PubMed] [Google Scholar]
- 8. Sahebally SM, Healy D, Walsh SR. Aspirin in the primary prophylaxis of venous thromboembolism in surgical patients. Surgeon. 2015;13(6):348–358. [DOI] [PubMed] [Google Scholar]
- 9. Douketis JD, Spyropoulos AC, Murad MH, Arcelus JI, Dager WE, Dunn AS, et al. Perioperative management of antithrombotic therapy: an American College of Chest Physicians clinical practice guideline. Chest. 2022;162(5):e207–e243. [DOI] [PubMed] [Google Scholar]
- 10. Kristiansen A, Brandt L, Agoritsas T, Akl EA, Berge E, Flem Jacobsen A, et al. Applying new strategies for the national adaptation, updating, and disseminating trustworthy guidelines: results from the Norwegian adaptation of the antithrombotic therapy and the prevention of thrombosis: American College of Chest Physicians Evidence‐Based Clinical Practice Guidelines[J]. Chest. 2014;146(3):735–761. [DOI] [PubMed] [Google Scholar]
- 11. Rashidi F, Barco S, Kamangar F, Heresi GA, Emadi A, Kaymaz C, et al. Incidence of symptomatic venous thromboembolism following hospitalization for coronavirus disease 2019: prospective results from a multi‐center study. Thromb Res. 2021;198:135–138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Jo WL, Lee YK, Ha YC, Lee KM, Kang BJ, Koo KH. Preventing venous thromboembolism with use of intermittent pneumatic compression after Total hip arthroplasty in Korean patients. J Korean Med Sci. 2016. Aug;31(8):1319–1323. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Amarase C, Tanavalee A, Larbpaiboonpong V, et al. Asia‐Pacific venous thromboembolism consensus in knee and hip arthroplasty and hip fracture surgery: part 2. Mechanical venous thromboembolism prophylaxis. Knee Surg Relat Res. 2021;33:1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Schooling CM, Zhong Y. Plasma levels of the anti‐coagulation protein C and the risk of ischaemic heart disease. Thromb Haemost. 2017;117(2):262–268. [DOI] [PubMed] [Google Scholar]
- 15. Citla Sridhar D, Abou‐Ismail MY, Ahuja SP. Central venous catheter‐related thrombosis in children and adults. Thromb Res. 2020. Mar;187:103–112. [DOI] [PubMed] [Google Scholar]
- 16. Chopard R, Albertsen IE, Piazza G. Diagnosis and treatment of lower extremity venous thromboembolism: a review. Jama. 2020;324(17):1765–1776. [DOI] [PubMed] [Google Scholar]
- 17. Pauwels CGGM, Hintzen KFH, Talhout RC, Pennings HWJM, Smolinska JLA, AgnieszkaOpperhuizen AVS, et al. Smoking regular and low‐nicotine cigarettes results in comparable levels of volatile organic compounds in blood and exhaled breath. J Breath Res. 2021;15(1):278–297. [DOI] [PubMed] [Google Scholar]
- 18. Lewicki S, Leśniak M, Bertrandt J, et al. The long‐term effect of a protein‐deficient diet enriched with vitamin B6 on the blood parameters in unexercised and exercised rats. Food Agric Immunol. 2018;29(2):1–13. [Google Scholar]
- 19. Jia X, Yin Z, Zhang W, du S, Kang J. Efficacy and safety of novel oral anticoagulants in patients with atrial nonvalvular atrial fibrillation and diabetes mellitus: a systematic review and meta‐analysis. J Transl Med. 2022;20(1):441–453. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Belmouaz M, Goussard G, Joly F, Grand F, Sibille A, Ecotiere L, et al. Comparison of the injection of low‐molecular‐weight heparin in the arterial vs. venous blood line for preventing extracorporeal circuit clotting during hemodialysis. Hemodial Int. 2022;26(3):314–322. [DOI] [PubMed] [Google Scholar]
- 21. Thomas O, Lybeck E, Strandberg K, Tynngård N, Schött U. Monitoring low molecular weight heparins at therapeutic levels: dose‐responses of, and correlations and differences between aPTT, anti‐factor Xa and thrombin generation assays. PloS One. 2015;10(1):e0116835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Cheung KS, Leung WK. Gastrointestinal bleeding in patients on novel oral anticoagulants: risk, prevention, and management. World J Gastroenterol. 2017;23(11):1954–1963. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Harenberg J, Marx S, Zolfaghari S, Weiss C, Wehling M, Schulze A, et al. Cluster analysis followed by network meta‐analysis to compare efficacy and safety of dabigatran, rivaroxaban, and apixaban after elective Total knee and hip replacement surgery. Blood. 2012;120(21):3409. 10.1182/blood.V120.21.3409.3409 [DOI] [Google Scholar]
- 24. Marx S, Weiss C, Wehling M, Scharf HP, Dahl O. C0227 indirect comparison of the efficacy and safety of dabigatran, rivaroxaban, and apixaban in elective total knee and hip replacement surgery. Thromb Res. 2012;130:S101. 10.1016/j.thromres.2012.08.006 [DOI] [Google Scholar]