Archives of Orthopaedic and Trauma Surgery

, Volume 134, Issue 5, pp 679–683

Administration of enoxaparin 24 h after total knee arthroplasty: safer for bleeding and equally effective for deep venous thrombosis prevention


  • Fucun Liu
    • Department of OrthopedicsThe First Affiliated Hospital of Zhejiang University of Traditional Chinese Medicine
  • Xiaobing Chu
    • Department of OrthopedicsThe First Affiliated Hospital of Zhejiang University of Traditional Chinese Medicine
  • Jiefeng Huang
    • Department of OrthopedicsThe First Affiliated Hospital of Zhejiang University of Traditional Chinese Medicine
  • Kun Tian
    • Department of OrthopedicsThe First Affiliated Hospital of Zhejiang University of Traditional Chinese Medicine
  • Jiang Hua
    • Department of OrthopedicsThe First Affiliated Hospital of Zhejiang University of Traditional Chinese Medicine
    • Department of OrthopedicsThe First Affiliated Hospital of Zhejiang University of Traditional Chinese Medicine
Knee Arthroplasty

DOI: 10.1007/s00402-014-1939-2

Cite this article as:
Liu, F., Chu, X., Huang, J. et al. Arch Orthop Trauma Surg (2014) 134: 679. doi:10.1007/s00402-014-1939-2



The purpose of this study was to determine the safety and effectiveness, as assessed by risk of bleeding and incidence of deep venous thrombosis (DVT), of administering delayed low-molecular-weight heparin (LMWH) after total knee arthroplasty.


A prospective study of 210 consecutive patients undergoing primary unilateral total knee arthroplasty was undertaken. The patients were randomized into two groups: one of which was managed according to a standard LMWH program (LMWH-s group) and the other with delayed LMWH (LMWH-p). LMWH was initiated 12 h after wound closure in the LMWH-s group, and 24 h after wound closure in the LMWH-p group.


The total blood loss in the first three postoperative days was calculated and all complications were recorded. The mean total blood loss was 435 and 387 mL in the LMWH-s group and LMWH-p group, respectively (p < 0.01). No significant difference in the incidence of symptomatic DVT was observed. The mean length of hospital stay was 7.29 days in the LMWH-s group and 6.56 days in the LMWH-p group (p < 0.05).


After total knee arthroplasty, LMWH-p is safer for bleeding than LMWH-s and equally effective concerning prevention of DVT.


Deep venous thrombosisLow-molecular-weight heparinMedicationAnticoagulationTotal blood loss


The approximate risk of the common, preventable complication of deep venous thrombosis (DVT) after total knee arthroplasty (TKA) is 40–60 % [1]. Accordingly, prevention of DVT after knee replacement has become the standard of care; routinely used medications include warfarin, low-molecular-weight heparin (LMWH), and the new anticoagulant, rivaroxaban [24]. Compared with LMWH, rivaroxaban has the advantages of effectively preventing recurrence of symptomatic DVT and the convenience of oral administration [57]. However, just at the point when rivaroxaban was believed to be a safer and more effective anticoagulant in this clinical setting, a large number of patients were reported to have required repeat surgery because of wound complications induced by rivaroxaban [810]. Due to the higher onset of post-op complications with rivaroxaban and comparable DVT, most surgeons turned to LMWH again.

LMWH is quite effective at preventing lower extremity DVT after arthroplasty. However, clinicians are concerned about the associated risk of bleeding [1113]. Bleeding caused by anticoagulants characteristically occurs in the early postoperative period; its occurrence after discharge has not been reported [14]. Major bleeding events occurring within 3 days of surgery account for more than one-fourth of all bleeding events over the whole course of treatment [6, 14]. Therefore, reducing the risk of early bleeding would increase the safety of LMWH.

We designed this randomized control trial with the aim of determining whether changing the timing of LMWH administration would reduce the associated risk of bleeding. Our hypothesis was that delayed administration of LMWH after TKA might reduce the risk of bleeding without increasing the incidence of DVT.

Materials and methods


Our institutional ethical review committee approved the present study. Between January 2011 and April 2013, 210 consecutive patients who underwent primary unilateral TKA for osteoarthritis provided consent to participate in the study. The criteria for exclusion included a history of DVT, pulmonary emboli, allergy to heparin, and coagulation disorders. Patients already receiving anticoagulation therapy were also excluded. The patients were randomized by computer-generated randomization codes into two groups, one of which was managed with a standard LMWH regime (LMWH-s group) and the other with delayed LMWH (LMWH-p group) (Fig. 1).
Fig. 1

Trial profile. TKA total knee arthroplasty, LMWH low-molecular-weight heparin

Study medication and transfusion requirements

General anesthesia was employed in all patients. In the LMWH-s group, LMWH (enoxaparin) was initiated 12 h after wound closure, and in the LMWH-p group 24 h after wound closure. Both groups continued LMWH for 10–14 days. The dosage of LMWH in both groups was 40 mg once daily subcutaneously.

In both groups, patients were transfused according to clinical need alone: indications for transfusion included shortness of breath, tachycardia, low blood pressure and difficulty with mobilization. Hemoglobin measurements were not used to determine the need for postoperative transfusion.

Outcome measures

Preadmission assessment clinic included recording of each patient’s hemoglobin concentration, height and weight. The patients’ hemoglobin concentrations and hematocrits were reviewed 12 h postoperatively (post-op), just before administration of the first dose of LMWH, and 72 h post-op. We also recorded the number of patients who were transfused in each group. The total blood loss in the first three postoperative days was calculated using the formula reported by Gross [15].

Meantime, all complications were recorded and any clinical suspicion of DVT promptly investigated by Doppler ultrasound. Major and minor bleeding events were categorized according to the standards set by the Committee for Proprietary Medicinal Products [16]. The platelet count and liver function of patients who are receiving LMWH for prophylaxis against DVT be checked at least once prior to discharge.

Statistical analysis

All data were analyzed statistically using SPSS 16.0 software for windows. Relevant patient variables were compared to ensure that the randomization process had resulted in similar patient characteristics between groups. The χ2 test was used to assess group differences in categorical variables, and independent t tests to compare continuous variables. Student’s t test was used for the analysis of two independent samples. The Wilcoxon rank-sum test was used for data where the assumption of normality did not hold. Differences at a level of p < 0.05 were considered statistically significant.


Relevant patient variables, including age, sex, involved side, body mass index and operation time were similar in the two groups (Table 1). No patients were lost to follow up.
Table 1

Relevant patient variables




p value

Patients/knees (no.)




Involved side (left)

57.3 %

55.1 %


Mean age (years)a

62.7 ± 5.98

61.1 ± 6.62


Female sex (%)

64.8 %

62.6 %


Mean body mass index (kg/m2)a

28.3 ± 6.76

28.7 ± 7.20


Mean duration of surgery (min)a

88 ± 11.17

90 ± 12.60


Mean length of hospital stay (days)a

7.29 ± 1.89

6.56 ± 0.75


NS: p > 0.05

Values are mean ± standard deviation

The mean total blood loss was 435 and 387 mL in the LMWH-s and LMWH-p group, respectively (p < 0.01). Major bleeding occurred in 8 of the 103 patients in the LMWH-s group and 2 of the 107 patients in the LMWH-p group (p = 0.045). We have presented these data in detail in Table 2.
Table 2

Safety outcomes




p value

Total blood loss (ml)a

435 ± 107

387 ± 99


Major bleeding (no.)



 Fatal bleeding




 Decrease in hemoglobin of ≥ 2 g/dL within 24 h








 Transfusion of ≥2 units of packed red cells




 Bleeding warranting treatment cessation




Crystalloid first day after operation

1,475 ± 449

1,469 ± 445


Colloid first day after operation

447 ± 154

439 ± 156


Wound complications (no.)




Patients receiving blood transfusions

23.3 %

11.2 %


Non-fatal pulmonary embolism




Symptomatic deep vein thrombosis (no.)




NS: p > 0.05

Values are mean ± standard deviation

In our study, 23 % of patients in the LMWH-s group needed homologous transfusions compared to 11.2 % in the LMWH-p group; the difference is significant. There were no identifiable complications related to the blood transfusions.

There was one case of deep infection and one of delayed wound healing in the LMWH-s group, and one case of superficial infection in the LMWH-p group.

During the postoperative period, we detected 11 symptomatic DVTs by Doppler ultrasound examinations, 5 in the LMWH-s group and 6 in the LMWH-p group (p < 0.05). No pulmonary emboli occurred in either group.

All patients were discharged from hospital as soon as they had achieved 90°of knee flexion and were independently mobile and self-caring. The mean length of hospital stay was 7.29 days in the LMWH-s group and 6.56 days in LMWH-p group (p < 0.05).


In this study, administration of LMWH 24 h after surgery prevented DVTs as effectively as did commencing LMWH 12 h after surgery. Delayed administration of LMWH after TKA was effective in reducing early postoperative total blood loss. Total blood loss was reduced from an average of 435 mL per patient in the LMWH-s group to 387 mL in the LMWH-p group. Delayed administration of LMWH also reduced the incidence of major bleeding events and resulted in correspondingly fewer bleeding complications. It was also associated with shorter postoperative recovery and hospitalization times.

Without prophylactic anticoagulation, approximately 90 % of all DVTs after knee replacement occurs in the calf veins, proximal thrombosis occurring infrequently and accounting for <10 % of all thrombi [12]. Perioperative administration of anticoagulants after arthroplasty reduces the incidence of symptomatic or non-symptomatic venous thrombosis. LMWH can be given once or twice daily without laboratory monitoring; this regime results in predictable dose-dependent plasma concentrations and has a longer half-life and lower incidence of heparin-induced thrombocytopenia than does conventional heparin therapy. In a systematic review of economic analyses of venous thromboembolism prevention strategies in hospitalized patients, researchers found that low-molecular-weight heparins appear to be the most economically attractive strategy for the prevention of venous thromboembolism in most medical and surgical patients [17].

However, the use of LMWH is associated with a higher bleeding rate and occurrence of major bleeding complications in a greater percentage of patients compared to no anticoagulant regime [18]. Whereas the rate of bleeding events when LMWH has not been administered is 0.5 %, after administration of LMWH it can be as high as 5 % [1921]. The consequences of secondary complications such as infections and wound and functional defects are worse than those of venous thromboembolism [2224]. Therefore, safety should be the top priority when deciding whether to administer LMWH.

It must be noted that bleeding events associated with administration of anticoagulants often occur in the early postoperative period; bleeding events caused by anticoagulants occurring after discharge have not been reported. Major bleeding events within the first three postoperative days account for more than 1/4 of all adverse events over the whole course of treatment [14]. Therefore, we changed the timing of administration of LMWH and succeeded in increasing its safety.

Rivaroxaban has recently been recommended for routine use in patients undergoing lower-limb arthroplasty [25, 26]. Compared to enoxaparin, rivaroxaban may reduce the incidence of venous thromboembolism and mortality after orthopedic surgery by 30 %, with the same bleeding rate [27]. This is why rivaroxaban has been widely accepted by orthopedic surgeons in recent years. In addition, its oral route of administration, ease of use, lack of need for routine monitoring, minimal food and drug interactions, and acceptable safety profile make rivaroxaban an attractive option [25]. However, because new drugs are often not evaluated prospectively with efficacy studies, and because manufacturers sponsor many such studies, there is a possibility of bias [17]. In addition, they are more expensive, raising questions about their cost-effectiveness [28]. Another concern is the lack of effective antidotes for quick and consistent reversal of their anticoagulant effects. Even more serious is the incidence of wound complications resulting from rivaroxaban. Jensen et al. [8] designed trials to assess the safety and efficacy of rivaroxaban. They found that 3.94 % of patients who received rivaroxaban had to return to theatre with wound complications, compared with 1.8 % of patients who received LMWH. Thus, further randomized controlled clinical trials should be conducted to assess the safety and efficacy of rivaroxaban in clinical practice.

Though the lack of blinding in this study is a limitation, a direct comparison clarified that there is a significant difference between different LMWH regimes. Delayed administration of LMWH after surgery improves its safety and is effective in reducing the incidence of DVT. LMWH is still the gold standard for anticoagulation after total knee arthroplasty.


The authors received no grants or outside funding in support of their research or in the preparation of this manuscript. They did not receive payments or other benefits from, or have a commitment or agreement with, any commercial entity.

Conflict of interest

Each author certifies that he has no commercial associations that might pose a conflict of interest in connection with the submitted article.

Copyright information

© Springer-Verlag Berlin Heidelberg 2014