Subcutaneous low dose unfractionated heparin (LDUH) 5000 units every 12 hours, or subcutaneous low molecular weight heparin (LMWH) is recommended for patients undergoing general abdominal, thoracic, or gynecologic surgery [5-8,29,63,77,78]. The two types of heparin are equally effective, but in meta-analyses less bleeding has been seen with low molecular weight heparin. Subcutaneous LMWH has the advantage of a once daily injection but is more expensive. Intermittent pneumatic compression until the patient is ambulatory is an alternative for patients at high risk of bleeding. Pharmacologic methods may be combined with IPC or graduated compression stockings in selected patients.
Patients undergoing an intracranial neurosurgical procedure should receive intermittent pneumatic compression due to its efficacy and safety profile in this population; any bleeding risk may be unacceptable for patients undergoing intracranial or spinal cord procedures. This approach may be used in conjunction with graduated compression stockings [63]. LDUH or LMWH are acceptable alternatives but are usually started 48 postoperatively and may be combined with physical prophylaxis modalities in high-risk patients [63].
High risk surgical patients — The following recommendations for prevention of venous thrombosis apply to high risk patients. They vary somewhat according to the clinical setting. High risk general surgery patients are typically treated with subcutaneous LMWH.
Pharmacologic therapy may be combined with intermittent pneumatic compression in patients at highest risk [63]. Adjusted dose heparin requires frequent monitoring and is seldom used.
Knee replacement — Following elective knee replacement, either LMWH (given at the usual high-risk dose), fondaparinux, or adjusted dose oral anticoagulants (target INR 2.5, range 2.0 to 3.0) can be used [21,30,31,39,63,79]. Prophylaxis should be continued for seven to ten days [63]. One trial found that LMWH started within eight hours of total knee replacement surgery significantly reduced the incidence of DVT when compared with adjusted dose warfarin [79].
Prolonging therapy for an additional three to six weeks does not appear to provide further benefit [23,30]. External pneumatic compression was found to be effective in earlier studies and is a useful alternative [63].
Hip replacement — Several approaches are effective for patients undergoing total hip replacement. Subcutaneous LMWH (given once or twice daily), fondaparinux, or adjusted dose oral anticoagulation (target INR 2.5, range 2.0 to 3.0) is effective and safe in patients undergoing hip replacement [16,17,21,30,33,63,66,80-82]. Prophylaxis should be continued for at least ten days (see below).
In North America, LMWH has usually been started between 12 and 24 hours after surgery. One trial found that LMWH given within two hours preoperatively, or four to six hours postoperatively, significantly decreased both total and proximal DVT compared to oral warfarin [82]. The timing of the first dose of LMWH following total hip replacement was further assessed in a meta-analysis of four trials [36]. Early initiation of half-dose LMWH between two and six hours postoperatively was associated with decreased clot formation when compared with warfarin or LMWH treatment beginning 12 or more hours postoperatively. The risk of bleeding was similar in both groups.
A number of randomized trials have shown that extended LMWH prophylaxis through postoperative day 27 to 35 significantly reduced the incidence of total DVT and, in some studies, proximal DVT, when compared with placebo [17,21,33,80-83]. Two separate meta-analyses demonstrated a significant decrease in the incidence of DVT or pulmonary embolism when compared with placebo, without an increase in major bleeding episodes [23,84].
Hip fracture — Thromboembolic prophylaxis is becoming a routine aspect of the care of the patient with hip fracture. However, a number of questions remain, including the choice of the optimal agent and the timing and duration of prophylaxis. A complete discussion of the therapeutic options is presented separately. (See "Medical consultation for patients with hip fracture", section on Thromboembolic prophylaxis).
We recommend either one of three approaches for prophylaxis against venous thromboembolism in patients with hip fractures: oral anticoagulation with warfarin (target INR 2.5, range 2.0 to 3.0) [63], fondaparinux [67], or fixed dose subcutaneous low molecular weight heparin started preoperatively [63]. The combined use of graduated compression stockings with either approach may provide additional benefit in certain patients. Extended prophylaxis with fondaparinux significant decreases the incidence of both venographic and clinical VTE when compared with in hospital fondaparinux alone [68].
Multiple trauma — Low molecular weight heparin is effective and superior to low dose heparin for patients who have suffered multiple trauma [85]. LMW heparin prophylaxis should be started as soon as it has been considered safe to do so [63]. Intermittent pneumatic compression has been recommended, when feasible, because it eliminates any risk for bleeding.
Inferior vena cava filters are not recommended for primary prophylaxis [63]. For patients in whom an inferior vena cava filter has been inserted, chronic anticoagulation is recommended where feasible. (See "Inferior vena caval filters", section on Anticoagulant therapy following filter placement).
Acute spinal cord injury — Low molecular weight heparin is the most effective prophylaxis for patients who have had an acute spinal cord injury associated with paralysis [63]. Adjusted dose heparin is also beneficial in this setting, while low dose heparin and intermittent pneumatic compression are less effective. Combining intermittent pneumatic compression with low molecular weight heparin or adjusted dose heparin may provide additional benefit, but this approach is not supported by convincing data [63]. (See "Respiratory complications of spinal cord injury").
Surgery in cancer patients — In a survey of clinical trials of thromboprophylaxis in surgical patients with cancer, the average incidence of DVT in untreated patients was 29 percent, placing them in the "high risk" category (see "Surgery" above) [86-88]. This high historic incidence was reduced in the randomized Enoxaparin and Cancer (ENOXACAN) I study, in which the incidence of DVT in patients recovering from abdominal surgery for cancer after receiving 10 days of prophylaxis with either enoxaparin or unfractionated heparin was similar at 15 and 18 percent, respectively [86].
In the double-blind placebo-controlled ENOXACAN II study, four weeks of prophylactic treatment with enoxaparin (40 mg/day SQ) was significantly better than one week of treatment in reducing the incidence of venous thromboembolism (VTE) in patients undergoing planned curative open surgery for abdominal or pelvic cancer [87]. At the three-month observation period, there were 23 VTEs in the 167 patients (13.8 percent) treated with enoxaparin for one week (and placebo for the remaining three weeks) and nine events in the 165 patients (5.5 percent) treated with enoxaparin for four weeks, for a relative risk reduction of 60 percent (95 percent confidence interval: 17 to 81 percent). The incidence of bleeding was similar in the two treatment arms.
A retrospective single institution study evaluated the effectiveness of a clinical prophylactic pathway on the incidence of symptomatic VTE in 3898 patients undergoing breast cancer surgery [89]. All patients received preoperative teaching about the importance of immediate and frequent postoperative ambulation as well as the use of elastic compression stockings and intermittent pneumatic compression devices. The stockings and compression devices were applied before induction of anesthesia and continued through the time of discharge. Despite a median age of 54 years and median operative times in excess of 117 minutes, which otherwise would have defined such surgery as moderate risk, the cumulative 60-day incidence of symptomatic VTE was 0.16 percent.
Medical patients — Medically ill patients are at increased risk for developing VTE while hospitalized [63,87]; many studies indicate that such patients often do not receive VTE prophylaxis [90,91]. As an example, in one study, 75 percent of patients admitted to a medical service were characterized as b**** at increased risk for VTE, yet only 43 percent received prophylaxis of any sort [92].
One single institution retrospective study analyzed patterns of VTE prophylaxis and in-hospital mortality in 272 critically ill medical patients receiving at least 24 hours of care in an intensive care unit. Results included [93]:
Some form of VTE prophylaxis was used in 75 percent of patients (pharmacologic 20 percent, mechanical 38 percent, mechanical and pharmacologic 18 percent, none 24 percent)
In-hospital mortality rates were 23 versus 36 percent for those receiving or not receiving pharmacologic VTE prophylaxis, respectively (adjusted odds ratio 0.45, 95% CI 0.22-0.93)
These results, which suggest that pharmacologic prophylaxis of VTE improves survival in patients with critical medical illness, require confirmation in a prospective randomized trial.