Surgical bleeding may adversely affect the clinical outcome following major orthopaedic surgery and is a serious concern for orthopaedic surgeons [19
]. It is a particular problem in elderly patients with reduced renal function in whom standard doses of antithrombotic drugs can increase bleeding [4
]. A recent international survey emphasised physicians’ concerns about bleeding risk in patients undergoing orthopaedic surgery, with the majority of respondents favouring the development of new anticoagulants that offered a reduced bleeding risk whilst maintaining similar VTE prevention compared with existing agents [19
]. In this study, we found that a lower than standard dose (150 mg qd instead of 220 mg qd) of postoperatively administered dabigatran was as effective as preoperatively injected 40 mg qd enoxaparin for VTE prevention (major VTE or VTE-related death: 4.3% vs 6.4%, respectively) and was associated with numerically lower rates of bleeding (MBE 1.3% vs 3.3%; MBE/CRBE: 8.0% vs 9.3%, respectively). These bleeding rates appear to compare favourably with those observed in placebo-treated patients following hip replacement [21
]. However, it is difficult to directly compare bleeding rates between trials because of the different definitions used [16
]. The reduction in bleeding found in our study in the elderly and/or moderately renally impaired patients was not statistically significant but showed a trend towards lower bleeding with the lower dose. Given the limited number of patients over 75 years or with moderate renal impairment in the database, we did not have power to show statistically significant differences. However, based on a power of 80%, approximately 2 × 975
1,950 patients would need to have been evaluated to enable conclusions that reached statistical significance based on our MBE findings (Fig. ).
Although the subgroup of patients with moderate renal impairment was smaller and the statistical significance should be interpreted with caution, there were fewer major bleeding events when these patients were treated with 150 mg qd dabigatran compared with enoxaparin (p
Even with the relatively few patients we traced in the database, results of this study support the view of the EMA of using a lower dose of dabigatran for patients at higher risk of bleeding. In the same type of orthopaedic population, reduced plasma clearance, and hence increased risk of bleeding with the standard dose of fondaparinux (2.5 mg) in certain fragile patient groups, resulted in the EMA recommending a lower fondaparinux dose (1.5 mg) in patients with a CrCl 20–50 ml/min [22
]. Furthermore, a recent report suggests that 1.5 mg fondaparinux administered to patients with moderate renal impairment offers a similar predicted exposure as does 2.5 mg in patients with normal renal function [23
]. Although the low-molecular-weight heparin dalteparin does not seem to increase bleeding in patients with severe renal impairment [23
], the principal impression from available data suggests an increased risk of bleeding even in these patient groups [22
]. The medicinal authorities have applied a similar approach to dabigatran by approving two doses for VTE prophylaxis in patients undergoing major joint replacement: a higher dose (220 mg qd) for healthy adults and a lower dose (150 mg qd) for elderly patients and those with reduced renal function (CrCl ≥30 to <50 ml/min). This strategy is supported by the findings in this study that daily administration of 150 mg dabigatran is at least as safe as preoperative injections of the standard European dose of enoxaparin these patients. Furthermore, a recent international clinical guideline committee noted that renal function should be considered and that lower than standard doses of anticoagulant therapy should be prescribed [7
]. This view is supported by a recent comment by van Thiel et al. that additional studies with lower drug doses be undertaken if there is concern over bleeding with new antithrombotic agents [28
No difference was seen in surgical-site bleeding during surgery or in drainage volume after surgery. As dabigatran was not administered until one to four hours after surgery and did not trigger any additional perioperative bleeding compared with enoxaparin, this indicates that dabigatran can be administered safely without any increase in the risk of wound bleeding. There was also no difference between treatments in the number of patients receiving transfusions or in units transfused per person, and there were few postoperative clinical bleeding events at the surgical site or elsewhere (Table ).
The rate of other wound complications, such as infections, abscesses and healing disturbances, were few and equally distributed, indicating that the studied doses of dabigatran and enoxaparin do not interfere with the inflammatory healing process. This is in line with earlier enoxaparin studies in which there were no higher rates of wound complications compared with placebo-treated patients [29
]. The low wound infection rate was similar to a study of enoxaparin and fondaparinux (0.7%) [31
], and no correlation between wound infection and postoperative anticoagulation therapy following joint replacement was reported in two other studies [32
]. Of particular concern to orthopaedic surgeons is the impact of anticoagulant treatment on wound healing. In this study, there was no significant difference in the rate of impaired wound healing. Similar results (no difference) were found in a pilot study of wound healing comparing fondaparinux with enoxaparin in patients undergoing knee replacement surgery [34
In summary, for patients at higher risk of bleeding (i.e. over 75 years or with moderate renal impairment), 150 mg qd dabigatran is as effective as enoxaparin for preventing VTE following elective total hip or knee replacement surgery. In addition, 150 mg qd dabigatran is associated with numerically less bleeding events (including wound bleeding) and a comparable safety profile to enoxaparin.