INTRODUCTION — It is estimated that over half of hospitalized medical patients are at risk for venous thromboembolism (VTE; ie, deep vein thrombosis [DVT] and/or pulmonary embolus [PE]) [1]. In addition, it is widely believed that PE is the most common preventable cause of hospital death [2-8].
This topic review discusses the prevention of VTE in patients hospitalized for acute medical illnesses [9]. Approaches to the prevention of VTE in surgical patients and patients with cancer and stroke, as well as in pregnancy are presented separately. (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients" and "Risk and prevention of venous thromboembolism in adults with cancer" and "Deep vein thrombosis and pulmonary embolism in pregnancy: Prevention" and "Prevention and treatment of venous thromboembolism in patients with acute stroke", section on 'Approach to VTE prevention'.)
EPIDEMIOLOGY — While many epidemiologic studies report venous thromboembolism (VTE) rates, in the absence of prophylaxis, that range from 10 to 80 percent, these rates are likely overestimated [2,3,5,10-26].
Accurate estimates of the incidence of clinically meaningful venous thromboembolic disease (ie, pulmonary embolus [PE] and deep vein thrombosis [DVT]) in hospitalized medical patients have been hampered by factors including the derivation of data from mixed surgical and medical populations, the reporting of both symptomatic and asymptomatic events, the increasing use of thromboprophylaxis and early ambulation during hospital admission, and the trend toward reducing length of hospital stay.
Thromboprophylaxis has been shown to reduce the risk of VTE in hospitalized medical and surgical patients. While thromboprophylaxis has been reported to reduce the risk of death in surgical patients [27,28], most studies and a meta-analysis have not been able to show a consistent beneficial effect of thromboprophylaxis on mortality in hospitalized medical patients [29-37]. The reasons for this difference between medical and surgical patients are unclear but may be related to a greater number of comorbidities in medical patients that contribute to overall deaths. In addition, the baseline risk in medical patients is lower than in surgical patients, and therefore, most studies or meta-analyses may be underpowered to demonstrate an effect of thromboprophylaxis in mortality [38]. Finally, clinicians should be aware that VTE prophylaxis does not eliminate the risk of VTE or VTE-related death in hospitalized patients.
The risk of VTE in specific medical populations is discussed in this section, while the effect of thromboprophylaxis in these groups is described below. (See 'Methods of thromboprophylaxis' below.)
General medical patients — Compared with the community, most patients admitted with an acute medical illness are at increased risk for VTE during and following hospital admission [2,3,5,26,39]. One retrospective medical record review of 911 patients with VTE reported that the age- and sex-adjusted incidence of VTE was more than 130 times greater among hospitalized patients than among community residents with 60 percent of all cases occurring in hospitalized, recently discharged, or nursing home patients [26]. Older age and male sex appeared to be risk factors for the development of in-hospital VTE. Other studies have reported that PE is responsible for death in up to one-third of hospitalized medical patients and 45 percent of VTE events occurs in the three months following discharge, although some of these figures may be an over estimate based upon interpretation of autopsy results [2,3,5,39].
High-risk populations — Compared with other patient groups, the groups of medical patients listed below are at particularly high risk for the development of VTE during hospital admission or after discharge.
Intensive care unit patients — All patients admitted to intensive care units (ICUs) are considered high risk for VTE (both upper and lower extremity venous thrombosis; approximately 10 percent), even after routine prophylactic anticoagulation [21,40-45]. As examples:
●A retrospective study of over 150,000 critically ill adults reported mortality was higher in those who received delayed thromboprophylaxis (ie, within the first 24 hours) compared with those who received prophylaxis from the time of admission (8 versus 6 percent [ICU mortality] and 11 versus 10 percent [hospital mortality]). [46]. The increase in mortality attributed to delayed thromboprophylaxis varied depending upon the population, and was reported as 4 percent (trauma patients), 8 percent (sepsis patients), 9 percent (cancer patients), and 15 percent (patients who survived a cardiac arrest). (See "Overview of inpatient management of the adult trauma patient", section on 'Thromboprophylaxis' and "Venous thromboembolism risk and prevention in the severely injured trauma patient", section on 'Thromboprophylaxis'.)
●In a post-hoc analysis of a randomized trial of 3746 critically ill patients, the incidence of VTE was 8 percent (5 percent DVT, 2 percent PE, 1 percent DVT plus PE) [45]. Predictors of thromboprophylaxis failure included a personal or family history of VTE and obesity).
●In a retrospective analysis of 678 ICU patients, 5 percent had DVT despite being on thromboprophylaxis. A history of DVT was the strongest predictor for the development of proximal DVT [47].
Cancer — VTE is a common complication in patients with active malignancy, defined as cancer in the advanced stages or for which patients are receiving antineoplastic treatment. Detailed discussion of the risk of VTE in patients with active malignancy is provided separately. (See "Risk and prevention of venous thromboembolism in adults with cancer".)
Stroke — VTE is a common complication of stroke, particularly in those with lower limb paralysis. Detailed discussion of the risk of VTE in patients with stroke is provided separately. (See "Prevention and treatment of venous thromboembolism in patients with acute stroke".)
Other — Pregnancy is a risk factor for VTE. Detailed discussion of the overall risk of VTE during pregnancy as well as risk in pregnant women admitted to hospital are provided separately. (See "Deep vein thrombosis in pregnancy: Epidemiology, pathogenesis, and diagnosis" and "Deep vein thrombosis and pulmonary embolism in pregnancy: Prevention", section on 'Inpatient thromboprophylaxis'.)
There are numerous additional risk factors for VTE that may be present in hospitalized medical patients including heart failure, myocardial infarction, and age >60 years, as well as previous VTE, prolonged immobility, renal failure, obesity, inherited or acquired hypercoagulable states, and coronavirus disease 2019 (COVID-19). These and other causes of VTE are discussed separately. (See "Overview of the causes of venous thrombosis" and "COVID-19: Hypercoagulability".)
Effect of statin or aspirin on risk — We believe that the nature of the acute illness dictates the overall risk of VTE such that the approach to patients on statins and aspirin should be no different to that in general medical or high-risk populations, and that statins or aspirin should not be used as thromboprophylaxis agents. Despite studies that suggest a lower risk of VTE in nonhospitalized patients on statins and aspirin, several trials did not show benefit to the baseline risk of VTE. Conflicting outcomes may be due to the impact of variables common in patients taking statins or aspirin that may concurrently increase or decrease the risk of VTE (eg, age, ischemic heart disease, other antiplatelet agents).
Data that report the risk of VTE associated with statin and aspirin use in the general population are described below:
●Statins – Several studies including randomized trials, national database registries, population cohorts, and meta-analyses have reported conflicting effects of statins on the prevention of VTE [48-59].
•In the JUPITER trial, 17,802 healthy adults with marginal elevations in cholesterol were randomly assigned to treatment with oral rosuvastatin (20 mg/day) or placebo [49]. Compared with placebo, rosuvastatin was associated with a reduction in the rate of symptomatic VTE (0.18 versus 0.32 events/100 person-years). Subgroup analysis showed similar decreases in the rates of deep venous thrombosis and pulmonary embolism, regardless of whether the event was provoked or unprovoked.
•One meta-analysis of 29 randomized clinical trials that included patients from JUPITER reported that statin therapy did not significantly reduce the risk of VTE events (odds ratio [OR] 0.89, 95% CI 0.78-1.01) [60]. Compared with standard dose therapy, the same lack of benefit was reported in patients taking high-dose statin regimens (OR 0.98, 95% CI 0.8-1.2).
●Aspirin – Most epidemiologic studies suggest that aspirin in healthy individuals does not reduce the risk of VTE. As examples:
•In the Women's Health Study, VTE rates were no different in healthy females that were randomly assigned to receive either low-dose aspirin (100 mg orally every other day) or placebo for 10 years (1.18 versus 1.25 events per 1000 patient-years) [61]. On subgroup analysis, no VTE risk factor (ie, age, obesity, menopausal status, reported VTE at baseline, factor V Leiden, prothrombin gene mutation) modified the relationship between aspirin and the overall risk of VTE. However, the use of aspirin was associated with an increased risk of gastrointestinal bleeding, peptic ulcer, hematuria, bruising, and epistaxis.
•In a population-based case-control study, an analysis restricted to individuals without a history of cardiovascular events indicated that the current use of aspirin did not significantly reduce the incidence of VTE [48].
DEFINITION OF VTE PROPHYLAXIS — Prevention of venous thromboembolism (VTE) can be defined as primary or secondary. Primary thromboprophylaxis is preferred because its efficacy is well established and it is more cost effective than treatment of complications once they occur [62].
Primary prophylaxis — Primary prophylaxis, the preferred method for VTE prevention, is carried out using either drugs (eg, heparin) or mechanical methods (eg, intermittent pneumatic compression boots) that are effective for preventing deep vein thrombosis (DVT). The characteristics of an ideal primary prophylactic method include ease of administration, effectiveness, safety (particularly with respect to bleeding), and cost-effectiveness or at least cost-neutrality. The prophylactic measures available for hospitalized medical patients include low-dose unfractionated heparin; low-molecular-weight (LMW) heparins; fondaparinux; intermittent pneumatic compression (IPC) and/or graduated compression stockings (GCS) [63]; and where available, oral factor Xa or direct thrombin inhibitors. (See "Direct oral anticoagulants (DOACs) and parenteral direct-acting anticoagulants: Dosing and adverse effects", section on 'Direct thrombin inhibitors' and "Direct oral anticoagulants (DOACs) and parenteral direct-acting anticoagulants: Dosing and adverse effects", section on 'Direct factor Xa inhibitors' and "Heparin and LMW heparin: Dosing and adverse effects" and "Fondaparinux: Dosing and adverse effects" and "Prevention of venous thromboembolism in adults undergoing hip fracture repair or hip or knee replacement".)
Secondary prophylaxis — Secondary prevention involves the early detection and treatment of subclinical venous thrombosis by screening medical patients with objective tests that are sensitive for the presence of DVT. However, the benefits of available screening methods (eg, venous ultrasound, magnetic resonance imaging venography) on patient important outcomes is not well-established [64,65]. Accordingly, secondary prevention is often reserved for patients in whom primary prophylaxis is either contraindicated or shown to be ineffective. Secondary prophylaxis is also used during pregnancy in women with a high clinical suspicion for DVT but a negative compressive lower extremity ultrasound, the details of which are discussed separately. (See "Deep vein thrombosis in pregnancy: Epidemiology, pathogenesis, and diagnosis", section on 'Serial compression ultrasonography'.)
OUR APPROACH — Our approach is, for the most part, consistent with guidelines issued by several societies including the American College of Chest Physicians, the Neurocritical Care Society, the Society of Critical Care Medicine, and the American Society of Hematology [66-69].
Assess risk
Thrombosis risk assessment — The risk of venous thromboembolism (VTE) depends upon the nature of the acute illness and the presence of individual risk factors. A full history and examination should be obtained to assess this risk in all acutely ill hospitalized medical patients. We consider that medical patients admitted to hospital with at least one risk factor for VTE are at risk for the development of deep venous thrombosis and/or pulmonary embolism (PE). Risk factors commonly encountered in patients admitted to hospital include heart failure, acute respiratory failure (eg, acute exacerbations of chronic pulmonary disease), sepsis, inflammatory bowel disease, known thrombophilia, prolonged immobility ≥3 days, age >60 years, previous VTE, those with multiple risk factors, and possibly patients with an elevated D-dimer [19,29,38-41,63,70-83]. In addition, patients at particularly high risk of VTE include those who are critically ill, patients with lower limb paralysis from stroke, and patients with active cancer. (See "Overview of the causes of venous thrombosis" and 'High-risk populations' above.)
Most clinicians empirically assess risk and administer thromboprophylaxis based upon this assessment (see 'Selection of method of prophylaxis' below). However, a number of empirically generated or data-derived risk models, based upon a number of the above-noted risk factors have been proposed [84]. These prediction scores require further validation from independent, prospective studies before they can be used in routine practice [85] but are described in brief below:
●The Padua Prediction Score was used to determine VTE risk in 1180 consecutive medical patients. Patients were followed for up to 90 days following admission to assess the occurrence of symptomatic VTE [86,87]. The rate of VTE was as follows (calculator 1):
•Low-risk patients (score <4): 0.3 percent
•High-risk patients (score ≥4): 2.2 (receiving adequate in-hospital thromboprophylaxis) and 11 percent (not receiving adequate in-hospital thromboprophylaxis)
●The IMPROVE risk score was used to determine VTE risk in 15,156 medical patients enrolled in the observational International Medical Prevention Registry on Venous Thromboembolism (IMPROVE) study [39]. The observed rate of VTE within 92 days of admission was 0.4 to 0.5 percent if none of four risk factors was present (prior VTE, active cancer, age >60 years, and thrombophilia), and was in the range of 8 to 11 percent in those with the highest risk scores (three to four factors present) (IMPROVE thrombosis risk model).
A modified IMPROVE risk score incorporates a D-dimer level, measured during hospitalization, in which a score of 2 or 3 plus a D-dimer greater than twice the upper limit of normal identifies patients at increased risk for symptomatic VTE [88,89].
•A score of 0 or 1 identifies patients with a risk for VTE of 0.4 to 0.6 percent at 42 days and a risk of 0.5 to 0.7 percent at 77 days.
•A score of 2 or 3 (considered high risk) identifies patients with a risk for VTE of 0.8 to 1 percent at 42 days and a risk of 1 to 1.2 percent at 77 days.
•A score of 4 or 5 identifies patients with a risk for VTE of 1.6 to 1.9 percent at 42 days and a risk of 2.2 to 2.7 percent at 77 days.
•A score of 6 or 7 has a similar risk for VTE as a score of 4 or 5.
●The GENEVA risk score was tested in a multicenter validation study that included 1478 hospitalized medical patients, 43 percent of whom did not receive thromboprophylaxis [19,90]. The risk of symptomatic VTE or VTE-related death at 90 days was the following (calculator 2) [90]:
•Low-risk patients (score <3): 0.6 percent (receiving adequate in-hospital thromboprophylaxis) and 0.8 percent (not receiving adequate in-hospital thromboprophylaxis.
•High-risk patients (score ≥3): 3.2 percent (receiving adequate in-hospital thromboprophylaxis) and 3.5 percent (not receiving adequate in-hospital thromboprophylaxis).
Bleeding risk assessment — A full history and examination should be obtained to assess the risk of bleeding in all acutely ill hospitalized medical patients in whom pharmacologic VTE prophylaxis is warranted. Examples of medical patients at high risk of bleeding in whom pharmacologic thromboprophylaxis is typically contraindicated, include those with active bleeding or intracranial hemorrhage, those in whom a surgical procedure is planned in the immediate 6 to 12 hours (eg, spinal neuroaxial anesthesia), patients who have a moderate or severe coagulopathy, and patients with a severe bleeding diathesis or thrombocytopenia (eg, platelet count <50,000/microL or <100,000/microL plus additional risk factors for bleeding). Epistaxis and menstrual bleeding are not contraindications to pharmacologic thromboprophylaxis.
Validated models for evaluating the risk of bleeding in hospitalized medical patients are lacking [91,92]. Retrospective analysis of data from the International Medical Prevention Registry on Venous Thromboembolism (IMPROVE) has been used to assess in-hospital bleeding incidence in 15,156 medical patients and to identify risk factors at the time of admission associated with in-hospital bleeding risk [91]. The cumulative incidence of major and nonmajor in-hospital bleeding within 14 days of admission was 3.2 percent. The strongest independent risk factors for bleeding at the time of admission were the following:
●Active gastroduodenal ulcer (odds ratio [OR] 4.15, 95% CI 2.21-7.77)
●Bleeding within the three months prior to admission (OR 3.64, 95% CI 2.21-5.99)
●Platelet count <50,000/microL (OR 3.37, 95% CI 1.84-6.18)
Other bleeding risk factors included increased age, hepatic and/or renal failure, intensive care unit stay, presence of a central venous catheter, rheumatic disease, cancer, and male sex (ie, factors that also increase the risk of VTE). Each of the above risk factors, with appropriate weighting, was entered into a risk model (IMPROVE Bleeding risk model) with scores ranging from 0 to 15. Overall rates of bleeding were the following in patients with:
●Risk scores of 1: 0.5 percent
●Risk scores of 4: 1.6 percent
●Risk scores of 7: 4.1 percent
●Risk scores of 15: 14 percent
Further validation of this model is warranted before it can be routinely used to assist clinicians in estimating the risks of bleeding from pharmacologic thromboprophylaxis in patients at risk for VTE.
Selection of method of prophylaxis — Selecting a method of thromboprophylaxis is dependent upon many factors including the nature of the acute medical illness, the risk of hemorrhage and thrombosis, preferences and values of the patient, institutional policy, and cost. Although thromboprophylaxis is typically individualized, our approach in hospitalized medical patients is outlined in the sections below. These suggestions apply to hospitalized medical patients regardless of whether or not they are receiving aspirin or statins on admission. (See 'Effect of statin or aspirin on risk' above.)
Low-risk patients — For most patients hospitalized with an acute medical illness and who are without obvious risk factors for VTE (eg, young patients admitted for a 12 hour observation following an episode of syncope from hypoglycemia), pharmacologic thromboprophylaxis is not warranted. Options for this low-risk group include early ambulation with or without mechanical methods of thromboprophylaxis. (See 'Mechanical methods of thromboprophylaxis' below.)
Moderate-risk patients — For most patients hospitalized with an acute medical illness, who have at least one risk factor for VTE and do not have an increased risk of bleeding, we recommend the use of pharmacologic thromboprophylaxis rather than mechanical methods or no prophylaxis. Low-molecular-weight (LMW) heparin is generally the preferred anticoagulant based upon randomized trials that suggest it is superior for preventing deep vein thrombosis (DVT). For those with renal failure (creatinine clearance <30 mL/min) or for those in whom cost is an issue, unfractionated heparin (UFH) is a reasonable alternative to LMW heparin. (See 'Pharmacologic thromboprophylaxis' below.)
High-risk patients — For most patients hospitalized with an acute medical illness who are considered to be at high risk for VTE (eg, critically ill, cancer, stroke) and at low risk of bleeding, we recommend the use of pharmacologic thromboprophylaxis rather than mechanical methods or no prophylaxis. In general, we prefer LMW heparin rather than other anticoagulants but UFH is an alternative in those with renal failure or in whom cost is an issue. (See 'Pharmacologic thromboprophylaxis' below.)
Special populations
Heparin-induced thrombocytopenia — Fondaparinux may be used as an alternative to heparin in patients with a history of heparin-induced thrombocytopenia. (See 'Pharmacologic thromboprophylaxis' below and "Management of heparin-induced thrombocytopenia".)
At risk of bleeding — For most patients hospitalized with an acute medical illness who have risk factors for VTE and who are at high risk of bleeding or in whom anticoagulation is contraindicated (eg, gastrointestinal or intracranial hemorrhage), mechanical methods of VTE prevention (eg, intermittent pneumatic compression, graduated compression stockings, venous foot pump) are suggested over no prophylaxis [66]. Among these methods, pneumatic compression devices and graduated compression stockings are the most common forms of mechanical prophylaxis used. When mechanical forms of prophylaxis are used, transition to a pharmacologic agent should occur as soon as the bleeding risk becomes acceptably low (eg, often within 24 to 48 hours) or has been reversed [66]. (See 'Mechanical methods of thromboprophylaxis' below.)
Other — Special populations of medical patients require an individualized approach to thromboprophylaxis during an acute hospitalization. Detailed discussion of those populations is provided separately:
●Patients undergoing neuraxial anaesthesia or analgesia (see "Neuraxial anesthesia/analgesia techniques in the patient receiving anticoagulant or antiplatelet medication")
●Patients with stroke (see "Prevention and treatment of venous thromboembolism in patients with acute stroke")
●Patients who are pregnant (see "Deep vein thrombosis and pulmonary embolism in pregnancy: Prevention")
●Patients with cancer (see "Risk and prevention of venous thromboembolism in adults with cancer")
●Patients traveling for extended periods (see "Prevention of venous thromboembolism in adult travelers")
●Patients with spinal cord injury (see "Respiratory complications in the adult patient with chronic spinal cord injury", section on 'Venous thromboembolism')
Duration of prophylaxis — VTE prophylaxis should ideally continue until the patient is fully ambulatory or discharged from the hospital. Although data do not support routinely extending the duration of thromboprophylaxis in acutely ill medical patients beyond admission, in our experience select populations should probably receive extended thromboprophylaxis (eg, nonambulatory patients, patients unable to ambulate independently or mechanically ventilated patients admitted to acute rehabilitation for physical therapy or ventilator weaning). Thromboprophylaxis is typically not administered in chronically immobilized patients residing at home or in a nursing home [66]. The optimal degree of ambulation that provides protection or reduced risk for DVT is unknown.
Although extended duration prophylaxis (ie, beyond the acute hospital stay), has been shown to benefit some high-risk surgical patients (eg, patients with total hip replacement (see "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients", section on 'Duration')), similar benefits have not been consistently observed in patients admitted for acute medical illness [93,94]. Other meta-analyses showed a benefit of extended duration prophylaxis [95,96].
●Best illustrating this is a meta-analysis that included the three major randomized trials (EXCLAIM [30,97], ADOPT [98], MAGELLAN [99]), which reported that compared with traditional duration thromboprophylaxis, extended duration thromboprophylaxis (usually with enoxaparin or rivaroxaban) was not associated with a reduction in individual rates of symptomatic or asymptomatic DVT, PE, or VTE-related death [93]. Bleeding events were two to three times as likely to occur with extended duration thromboprophylaxis (number needed to harm 194). A subsequent meta-analysis of five trials showed similar results [100]. Other meta-analyses have suggested a benefit of extended duration prophylaxis in reducing symptomatic VTE, but the absolute risk reduction was low (number needed to treat [NNT] to prevent 1 DVT = 313, NNT to prevent 1 nonfatal PE = 625) [95].
●Similarly, in another randomized trial of acutely ill hospitalized medical patients (≥40 years) who had an elevated D-dimer, extended prophylaxis for 35 to 42 days with oral betrixaban (80 mg daily) was compared with enoxaparin (40 mg daily) administered for 6 to 14 days; three groups were analyzed, patients with an elevated D-dimer, patients with an elevated D-dimer or age >75 years, and all patients [94,101]. Betrixaban was associated with a reduction in the rate of VTE (symptomatic and asymptomatic) in the group as a whole (5.3 versus 7 percent) and in patients with an elevated D-Dimer or patients >75 years (5.6 versus 7.1 percent) but was not significant in patients with an elevated D-Dimer alone (6.9 versus 8.5 percent). There was no excess bleeding with extended prophylaxis. However, the analysis plan specified that if the between-group difference in any of the analyses was not significant, the other analyses would be considered exploratory, thereby prohibiting conclusive interpretation. Additional analyses and research are needed to identify specific patient subgroups that may benefit from extended prophylaxis with this agent. Betrixaban is no longer available.
●In another trial (MARINER), 12,019 patients considered at high risk of VTE were randomly assigned to receive rivaroxaban (10 mg orally daily) or placebo for 45 days following hospital discharge [102]. A 7.5 mg dose was used in a subgroup of patients with impaired renal function (creatinine clearance <50 mL per minute and ≥30 mL per minute). Rates of the composite outcome of symptomatic VTE or VTE-related death were similar among the groups (0.83 for rivaroxaban versus 1.1 percent for placebo; hazard ratio [HR] 0.76, 95% CI 0.52-1.09). However, similar to the effects of rivaroxaban reported in MAGELLAN [99], rivaroxaban in this study lowered the rates of symptomatic nonfatal VTE (0.18 versus 0.42 percent; HR 0.44, 95% CI 0.22-0.89). Although rates of major bleeding were higher in the rivaroxaban group compared with placebo, the overall incidence of major bleeding was low in both groups and without a statistically significant difference (0.28 versus 0.15 percent). The lower rate of VTE events in the placebo group (1.1 percent instead of the expected 2.5 percent) in this study may have resulted in an underestimation of the prophylactic effect of rivaroxaban.
●In a post-hoc analysis of MARINER and MAGELLAN, rivaroxaban at 10mg per day had a similar efficacy and safety profile in a small subgroup of patients with renal dysfunction compared with the majority of study participants who did not have renal dysfunction [103]. Further trials are needed before rivaroxaban could be considered in this setting.
●In a subsequent meta-analysis of three randomized trials, extended thromboprophylaxis with betrixaban or rivaroxaban reduced VTE (risk difference [RD] -1.51 percent, 95% CI -2.32 to -0.69 percent) without excess major bleeding (RD 0.12 percent, 95% CI -0.05 to -0.29 percent) [104]. Patients with an elevated D-dimer (ie, greater than twice the upper limit of normal) experienced a greater reduction in VTE than those without an elevated D-Dimer (RD -2.39 versus -0.26 percent). A similar trend was seen in patients aged ≥75 years compared with those younger than 75 years (RD for VTE -2.29 versus -0.63 percent). Betrixaban is no longer available for clinical use.
METHODS OF THROMBOPROPHYLAXIS
Pharmacologic thromboprophylaxis — In randomized trials, pharmacologic prophylaxis with low-molecular-weight (LMW) heparin, unfractionated heparin (UFH), or fondaparinux have all been shown to be superior to placebo or mechanical devices in preventing VTE [38,75,76,105-113]. Among these, we prefer LMW heparin because meta-analyses suggest it is superior to UFH (twice daily or three times daily dosing regimens), particularly in high-risk populations (see 'High-risk populations' above). Most data from randomized studies suggest that there is no appreciable effect of heparin prophylaxis on mortality and limited effect on bleeding. The quality of data is limited by the heterogeneity of included study populations (low risk and high risk) and the failure to distinguish symptomatic from asymptomatic VTE. Aspirin and warfarin should not be used as primary agents for VTE prevention.
Low-molecular-weight heparin
Efficacy — Trials comparing LMW heparin with UFH and/or placebo in specific populations are discussed below:
●General medical populations – Examples of randomized studies and meta-analyses that illustrate the value of pharmacologic prophylaxis in general medical populations include the following:
•One meta-analysis of 36 randomized trials of a mixed population of hospitalized medical patients that included those at low and high risk for VTE, compared heparin with placebo for the prevention of deep vein thrombosis (DVT) and pulmonary embolism (PE) [75]. Both UFH and LMW heparin reduced the risk of DVT (risk ratio [RR] 0.33, 95% CI 0.26-0.42; RR 0.56, 95% CI 0.45-0.7, respectively) and PE (RR 0.64, 95% CI 0.50-0.82; RR 0.37, 95% CI 0.21-0.64, respectively) without any effect on mortality. When compared with UFH, use of LMW heparin was associated with a lower risk of DVT (RR 0.68, 95% CI 0.52-0.88) but no difference was reported between the two agents in the risk of bleeding or thrombocytopenia.
•Another meta-analysis of 16 randomized studies reported similar results in a population of over 34,000 medical patients that excluded those at high risk for VTE (myocardial infarction, stroke, intensive care unit patients) [112]. In that population, compared with placebo, both UFH and LMW heparin reduced the risk of DVT (odds ratio [OR] 0.38, 95% CI 0.29-0.51) as well as symptomatic and fatal PE (OR 0.65, 95% CI 0.42-1) with no effect on mortality and thrombocytopenia. However, an increase in major hemorrhage was reported with use of heparin (OR 1.81, 95% CI 1.1-2.98). Compared with UFH, LMW heparin was associated with a reduced risk of DVT (OR 0.77, 95% CI 0.62-0.96) and lower rates of major bleeding (OR 0.43, 95% CI 0.22-0.83).
•Another meta-analysis of 44 randomized trials (over 90,000 mostly medical patients) reported that all pharmacologic anticoagulants reduced symptomatic VTE compared with placebo and that none offered a mortality benefit [114]. Among the agents, those that offered superior efficacy for preventing VTE included intermediate dose LMW heparin (OR 0.66, 95% CI 0.46-0.93), direct oral anticoagulants (0.68, 95% CI 0.33-1.34), and intermediate dose unfractionated heparin (0.71, 95% CI 0.43-1.19) (very low to low quality evidence). However, bleeding rates were higher with intermediate dose unfractionated heparin (2.63, 95% CI 1.00 to 6.21) and direct oral anticoagulants (2.31, 95% CI 0.82-6.47) (low to moderate quality evidence).
●High-risk populations – Studies have also investigated the use of prophylaxis in select populations considered to be at high risk for VTE:
•Stroke – In patients with leg paralysis from ischemic stroke, data derived from randomized studies reported that, compared with UFH, LMW heparin was associated with a greater risk reduction for VTE with no significant increase in clinically relevant bleeding [80,115-117]. (See "Prevention and treatment of venous thromboembolism in patients with acute stroke", section on 'Approach to VTE prevention'.)
•Critically ill – Randomized trials and meta-analyses have examined the importance of prophylaxis in the critically ill [118-122]. As examples:
-One network meta-analysis of 13 randomized trials that enrolled a total of 9619 critically ill patients reported that LMW heparin reduced the incidence of all DVT (OR 0.59, 95% CI 0.33-0.90; high certainty) when compared with control treatment (a composite of no prophylaxis, placebo, or compression stockings only) [122]. LMW heparin appeared to be more effective than UFH (OR 0.72, 95% CI 0.46-0.98; moderate certainty), and mechanical methods of thromboprophylaxis (eg, intermittent compressive devices) were least effective (OR 0.85, 95% CI 0.50-1.50; low certainty). The effect of combination therapy was unclear.
-Another randomized trial (PROTECT) of 3764 critically ill patients directly compared UFH with the LMW heparin, dalteparin [119]. Dalteparin was associated with a rate reduction in PE (1 versus 2 percent; hazard ratio 0.5) but had no effect on DVT, bleeding, mortality, or cost [119,121].
•Older patients – A meta-analysis of 5657 patients ≥75 years of age showed that, compared with placebo, prophylaxis reduced the incidence of VTE (4 versus 11 percent) without affecting bleeding events, or all-cause mortality [111]. However, among all the VTE events seen in the placebo group (symptomatic and asymptomatic DVT, PE and fatal PE), two-thirds were asymptomatic or ultrasound-detected DVT. Although major bleeding events were rare (1.1 versus 0.7 percent; OR 1.77, 95% CI 0.52-6.08), the wide confidence limits suggest that some patients are more at risk for bleeding than others. Further randomized studies are required to assess the safety and efficacy of VTE prophylaxis in the older population. (See 'Assess risk' above.)
•Patients hospitalized with COVID-19 – Such patients represent a special population at markedly increased risk for VTE, especially if critically ill in an intensive care unit setting. The use of anticoagulants as prophylaxis against COVID-19-associated VTE is covered elsewhere. (See "COVID-19: Hypercoagulability".)
Dosing — A number of LMW heparin preparations are available (enoxaparin, dalteparin, tinzaparin, nadroparin), none of which have proven superiority over the other when administered as agents to prevent VTE. Suggested prophylactic doses for medical patients with a creatinine clearance >30 mL/minute and assuming no extremes in body weight are the following:
●Enoxaparin 40 mg subcutaneously once daily
●Dalteparin 5000 units subcutaneously once daily
Less commonly used are:
●Tinzaparin 4500 anti-Xa subcutaneously once daily
●Nadroparin 3800 anti-Xa units/day in patients ≤70 kg and 5700 units per day in patients >70 kg once daily
The platelet count may be monitored regularly (eg, day 5 and 9) in all patients receiving LMW heparin to detect the development of heparin-induced thrombocytopenia (HIT). All LMW heparin agents are contraindicated in patients with HIT. Management of heparin anticoagulants in patients with HIT is discussed separately. (See "Management of heparin-induced thrombocytopenia".)
The ideal dose for patients with obesity, typically defined as those with a body weight greater than 100 kg, is unknown. Dosing should be individualized on a case-by-case basis. We generally prefer to empirically treat these patients with UFH (see 'Low-dose unfractionated heparin' below). An alternative approach used with LMW heparin prophylaxis is to increase the dose by 50 percent (eg, enoxaparin 60 mg once daily or dalteparin 7500 units once daily).
Based upon data extrapolated from patients receiving therapeutic doses of LMW heparin, some experts prefer the avoidance of these agents in patients with severe renal insufficiency (eg, creatinine clearance <30 mL/min) should it be present upon admission or develop during hospitalization. However, we and other experts continue to use selected LMW heparins in patients with renal insufficiency (eg, tinzaparin and dalteparin) but a dose reduction of enoxaparin is needed for those with severe renal insufficiency (table 1). For those who develop severe renal insufficiency during hospitalization, it is prudent that selected LMW heparins be discontinued and replaced with UFH. (See "Heparin and LMW heparin: Dosing and adverse effects".)
Low-dose unfractionated heparin
Efficacy — Thromboprophylactic doses of UFH are effective at preventing VTE when compared with placebo or mechanical devices. However, when compared with UFH, LMW heparin appears to be marginally superior for the prevention of VTE, the details of which are discussed separately. (See 'Low-molecular-weight heparin' above.)
Dosing — Typical prophylactic dosing for UFH is 5000 units subcutaneously twice or three times daily. The frequency of UFH dosing is controversial. While one meta-analysis of 12 randomized studies, reported a nonsignificant reduction in the overall rate of VTE with three times daily compared with twice daily UFH heparin dosing (3.5 versus 5.4 events/1000 patient-days), the risk for major bleeding was significantly greater (0.96/1000 patient-days versus 0.35/1000 patient-days) [123]. In a second meta-analysis, when compared with placebo, UFH injected three times daily was more effective in preventing DVT than a twice daily dosing regimen (RR 0.27; versus RR 0.52; 95% CI 0.28-0.96) [75]. However, a third meta-analysis found no difference between twice and three times daily UFH dosing in the rates of VTE, VTE-related death, or major bleeding [124]. Our preference is for twice daily UFH dosing since the evidence to support three times daily is inconclusive. Factors including cost, institutional policy, body weight, and risk of bleeding may also be used to help the clinician make the decision to choose the frequency of dosing.
The ideal dose for patients with obesity is unknown, such that dosing should be individualized on a case-by-case basis. We generally prefer to empirically treat with UFH 5000 to 7500 units three times daily; alterations in the dosing and frequency should be tailored to individual patients.
Compared with LMW heparin, the dose of UFH does not need to be adjusted for patients with renal failure. In addition, in some countries, low-dose UFH has the advantage of being inexpensive. However, similar to LMW heparin, UFH is contraindicated in patients with heparin-induced thrombocytopenia (HIT) and the platelet count should be monitored regularly (eg, days 5 and 9) in all patients receiving low-dose UFH to detect the development of HIT [125,126]. (See "Heparin and LMW heparin: Dosing and adverse effects", section on 'Platelet count monitoring'.)
Fondaparinux
Efficacy — Fondaparinux is superior to placebo and probably as effective as LMW heparin for patients who are not critically ill, although compared with LMW heparin, very little data are available to support its routine use.
Fondaparinux (2.5 mg subcutaneously once daily) was compared with placebo in 849 medical patients ≥60 years of age, hospitalized due to acute cardiac, respiratory, infectious, or inflammatory disease, who were expected to remain in bed for at least four days, and considered to be at moderate risk for VTE (ARTEMIS) [72]. There was a significant reduction in the rate of symptomatic and asymptomatic VTE (11 versus 6 percent). Major bleeding occurred in one patient (0.2 percent) in each group. This study is discussed in greater detail separately. (See "Fondaparinux: Dosing and adverse effects".)
Dosing — Fondaparinux is typically administered as 2.5 mg subcutaneously once daily. Fondaparinux should be avoided in those with a creatinine clearance <30 mL/min. If necessary, a dose reduction to 1.5 mg subcutaneously daily can be used in those with a creatinine clearance in the range of 30 to 50 mL/min [127]. The safety and efficacy of reduced dosing was tested in a multicenter prospective cohort study in 206 acutely ill older (mean age 82 years) medical patients (mean creatinine clearance 33 mL/minute [range 20 to 50 mL/min]) at high risk for both bleeding and thrombosis [128]. Symptomatic VTE developed in three patients (1.5 percent) and major bleeding occurred in only one patient (0.5 percent).
Aspirin — We agree with others that aspirin not be used, either alone or in combination, as prophylaxis against VTE in any medical patient group [66]. Although aspirin, is highly effective in reducing major arterial thrombotic events, there is little evidence that aspirin and/or other antiplatelet agents (eg, clopidogrel) can prevent venous thromboembolic events in hospitalized medical patients. Although a 1994 meta-analysis suggested that aspirin reduced the incidence of VTE by approximately 20 percent compared with placebo or no treatment [129,130], studies since then have shown either no significant benefit or inferiority when compared with other modalities such as LMW heparin [130].
In the community setting, aspirin does not appear to reduce the overall risk of VTE prior to admission, the details of which are discussed separately. (See 'Effect of statin or aspirin on risk' above.)
In contrast, aspirin has been shown to decrease the recurrence rate in patients with a first episode of VTE, the details of which are also discussed separately. (See "Selecting adult patients with lower extremity deep venous thrombosis and pulmonary embolism for indefinite anticoagulation".)
Warfarin — Initiating warfarin in patients who are not already anticoagulated is not appropriate for immediate and short-term (ie, one month or less) prevention of VTE; the anticoagulant effect is delayed and does not occur until 36 to 72 hours after drug administration. In addition, hospitalized medical patients often have comorbidities (eg, impaired liver function) and/or are on medications that interfere with consistent prediction of the anticoagulant effect. (See "Warfarin and other VKAs: Dosing and adverse effects" and "Biology of warfarin and modulators of INR control", section on 'Mechanism of action'.)
Direct oral anticoagulants — Rivaroxaban is approved for use in hospitalized medically ill patients [131,132]. In a study of over 7000 hospitalized medically ill patients, compared with enoxaparin, the composite outcome of asymptomatic and symptomatic VTE plus VTE-related death was reduced in patients taking betrixaban (which is no longer available) compared with subcutaneous enoxaparin (4.4 versus 6 percent) without any increase in the bleeding rate [101]. Rivaroxaban has been studied in the extended duration setting [99,133]. These studies are discussed above. (See 'Duration of prophylaxis' above.)
Dosing in patients with obesity is unclear. The International Society on Thrombosis and Hemostasis suggest that for patients with a body mass index (BMI) >40 kg/m2 or weight >120 kg, that standard dosing for rivaroxaban and apixaban rather than other DOACs can be used in this population [134].
Mechanical methods of thromboprophylaxis — We agree with others that mechanical methods for the prevention of VTE are primarily indicated in patients at high risk of bleeding or in whom anticoagulation is contraindicated (eg, gastrointestinal or intracranial hemorrhage) [66]. When used in these circumstances, transition to a pharmacologic agent should be considered as soon as the bleeding risk becomes acceptably low or has been reversed [66]. Data that supports this approach are derived mostly from surgical patients and limited data performed in acutely-ill hospitalized medical patients.
Options include intermittent pneumatic compression, graduated compression stockings, and venous foot pumps. Neuromuscular electrical stimulation remains investigational [135].
Intermittent pneumatic compression — Intermittent pneumatic compression (IPC) (picture 1) prevents DVT by enhancing blood flow in the deep veins of the legs, thereby preventing venous stasis [136]. IPC also reduces plasminogen activator inhibitor-1 (PAI-1), thereby increasing endogenous fibrinolytic activity [137]. (See "Thrombotic and hemorrhagic disorders due to abnormal fibrinolysis", section on 'PAI-1'.)
IPC devices are an alternative for VTE prevention in medical patients with a high risk of bleeding or in whom anticoagulant drugs are contraindicated (eg, GI bleeding, intracranial hemorrhage) [66]. Although there are no data available on skin complications of IPC use, skin breakdown is a known complication, especially in the frail older adult population. IPC devices are also contraindicated in patients with evidence of leg ischemia due to peripheral vascular disease. Attention must be paid to optimal compliance, as well as proper fit of the IPC device.
Data on the efficacy and safety of IPCs are limited. However, one large randomized trial in patients with stroke suggested that IPCs reduce the incidence of VTE [138]. A multicenter, randomized trial of 2876 immobile patients with acute stroke (CLOTS 3) reported that, compared with no device, IPC use was associated with a lower rate of VTE at 30 days (12 versus 8.5 percent) without altering mortality (13 versus 11 percent). However, the use of LMW heparin was similar in both groups (32 versus 30 percent). Further results of this trial and clot prevention in patients with stroke are discussed separately. (See "Prevention and treatment of venous thromboembolism in patients with acute stroke", section on 'Intermittent pneumatic compression'.)
Contraindications to anticoagulant prophylaxis are discussed above. (See 'Bleeding risk assessment' above.)
Graduated compression stockings and venous foot pump — There is even less convincing evidence regarding the efficacy of graduated compression stockings (GCS) and venous foot pumps (VFP) in medical patients (picture 2).
In a meta-analysis, GCS was found to be ineffective in the prevention of VTE in patients with ischemic stroke [139]. One randomized trial of stroke patients showed that thigh-length GCS was associated with no benefit with respect to reduction of VTE, along with a four-fold increase in skin ulcers and necrosis [140]. The use of GCS to prevent VTE following stroke is discussed separately. (See "Prevention and treatment of venous thromboembolism in patients with acute stroke", section on 'Ineffective or unproven treatments'.)
Similarly, randomized studies showing efficacy of VFP devices in medical patients are lacking and most data to support their use are derived from the surgical population [141]. In theory, they prevent thrombosis by stimulating lower limb venous flow at the level of the foot [142].
Combined methods — Despite the widespread practice of combining mechanical and pharmacologic methods of thromboprophylaxis, few studies have systematically examined the efficacy of this approach on the rates of VTE or VTE-related death in acutely ill hospitalized medical patients. One large randomized trial assessing the use of GCS in medical patients, with or without low-dose LMW heparin, found no added mortality benefit with the use of LMW heparin. However, this trial may have been underpowered to detect a significant difference in mortality [34]. A second randomized trial (PREVENT) of 2003 critically ill patients, 80 percent of whom were medical intensive care unit patients, reported that the addition of intermittent pneumatic compression (IPC) for at least 18 hours a day to unfractionated or LMW heparin did not reduce the rate of ultrasound-detected deep venous thrombosis (3.9 versus 4.2 percent) or symptomatic pulmonary embolism (0.8 versus 1.0 percent) or death from any cause (26 percent each) [143]. However the incidence of VTE was lower than expected in the control group, thereby reducing the power of the trial to detect a difference between the groups. The value of combined prophylaxis in surgical patients is discussed separately. (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients", section on 'With low bleeding risk: Combined prophylaxis'.)
ANTICOAGULANTS IN DEVELOPMENT — Anticoagulants in development for thrombosis prophylaxis are discussed separately. (See "Investigational anticoagulants".)
QUALITY IMPROVEMENT INITIATIVES
●National programs – The prevention of venous thromboembolism (VTE) is a major patient safety issue in hospitals according to the Agency for Health Care Research and Quality [144]. As an example, the Center for Medicare and Medicaid Services (CMS) considers appropriate VTE prophylaxis to be a pay-for-performance quality measure for specific surgical procedures [145].
Thus, in the United States there have been a number of initiatives aimed at calling attention to the prevalence of VTE and advocating for increased use of VTE prophylaxis in hospitalized patients. These initiatives have come from The National Quality Forum [146], The Surgical Care Improvement Project [115], the Centres for Medicine and Medicinal Services, the Joint Commission on Accreditation of Health Care Organizations [147], and the Office of the Surgeon General of the United States [148]. Similar initiatives have been developed in Canada [149], the United Kingdom [70,150], and Europe.
However, despite these efforts, numerous audits and reviews have demonstrated that appropriate thromboprophylaxis is not being offered to large numbers of patients, particularly those hospitalized with medical conditions [105,151-160]. Reasons for this may include contraindications to anticoagulants or altered risk as the patient progresses throughout their admission as well as clinician reluctance to change practice.
●Hospital decision support programs – Various strategies designed to improve the appropriate use of thromboprophylaxis have shown inconsistent or unproven effects on reducing VTE incidence [161-165]. Such strategies include clinician alerts, computerized order sets with electronic alerts or pre-printed orders, risk assessment models, and quality improvement in the form of audit and feedback. Despite the existence of such programs, formal educational programs for involved health care professionals may still be needed in order to significantly improve the appropriate use of thromboprophylaxis [166,167]. Further efforts are required at improving the translation of data from clinical trials into clinical practice [86,168-177].
●Predictive models – Predictive models that identify those at greatest risk and in need of thromboprophylaxis have been reported. As an example, one database analysis identified factors independently associated with VTE as previous VTE, known thrombophilia, cancer, age >60 years, lower-limb paralysis, immobilization for seven days or longer, and admission to an ICU or coronary care unit [39]. Points were assigned to each factor identified to give a total risk score for each patient. During hospitalization, a score ≥2 was associated with higher overall and VTE-related mortality. Predictive models such as these need validation before they can be routinely used at the time of admission or during hospitalization to identify those at greatest risk in whom thromboprophylaxis should be administered.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Superficial vein thrombosis, deep vein thrombosis, and pulmonary embolism".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Beyond the Basics topics (see "Patient education: Deep vein thrombosis (DVT) (Beyond the Basics)" and "Patient education: Warfarin (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Epidemiology and thrombosis risk – While many studies report venous thromboembolism (VTE) rates in the absence of prophylaxis that range from 10 to 80 percent, these rates are likely overestimated.
•General medical patients – While thromboprophylaxis reduces the risk of VTE in hospitalized medical patients, mortality benefits are unclear in this population. Importantly, clinicians should be aware that VTE prophylaxis does not eliminate the risk of VTE or VTE-related death. (See 'Epidemiology' above.)
•High-risk patient groups – Patients at highest risk of VTE include patients who are critically ill, patients with cancer or stroke, and patients with multiple risk factors for VTE including pregnancy, heart failure, myocardial infarction, old age (>75 years), previous VTE, prolonged immobility, renal failure, obesity, and inherited or acquired hypercoagulable states. (See 'Epidemiology' above and "Overview of the causes of venous thrombosis".)
•Risk assessment – For patients admitted to the hospital with an acute medical illness, a full history and examination should be performed to assess both the risk of VTE and bleeding. However, validated models for evaluating this risk in hospitalized medical patients are lacking. (See 'Assess risk' above.)
●Low-risk patients – For hospitalized medical patients without obvious risk factors for VTE, we recommend that pharmacologic thromboprophylaxis not be employed. Options for this low-risk group include early ambulation with or without mechanical methods of thromboprophylaxis. (See 'Selection of method of prophylaxis' above and 'Mechanical methods of thromboprophylaxis' above.)
●Moderate- or high-risk patients – For most patients hospitalized with an acute medical illness who have at least one risk factor for VTE (moderate or high risk) and do not have an increased risk of bleeding, we recommend the use of pharmacologic thromboprophylaxis rather than mechanical methods or no prophylaxis (Grade 1B).
•Agent of choice – Among the available agents, we prefer low-molecular-weight heparin rather than other anticoagulants (table 1) (Grade 2C), particularly in high-risk populations (eg, critically ill, cancer, stroke).
•Alternatives – For those with renal failure (creatinine clearance <30 mL/min) or for those in whom cost is an issue, unfractionated heparin is a reasonable alternative. For patients with heparin-induced thrombocytopenia, fondaparinux may be used as an alternative to heparin. (See 'Selection of method of prophylaxis' above and 'Pharmacologic thromboprophylaxis' above.)
•Combined prophylaxis – While some experts administer more aggressive prophylaxis in the form of increased intensity of a pharmacologic agent or the addition of a mechanical device, data do not support this approach in medical patients. (See 'Combined methods' above.)
●Patients at risk of bleeding – For most patients hospitalized with an acute medical illness who have risk factors for VTE and who are at high risk of bleeding or in whom anticoagulation is contraindicated (eg, gastrointestinal or intracranial hemorrhage), mechanical methods of VTE prevention (eg, intermittent pneumatic compression, graduated compression stockings, venous foot pump (picture 1)) are suggested over no prophylaxis (Grade 2C). Transition to a pharmacologic agent should occur as soon as the bleeding risk becomes acceptably low or has been reversed. (See 'Mechanical methods of thromboprophylaxis' above and 'At risk of bleeding' above.)
●Special populations – Special populations of medical patients require an individualized approach to thromboprophylaxis during an acute hospitalization. These include patients with heparin-induced thrombocytopenia, patients undergoing neuraxial anaesthesia or analgesia, patients with stroke or cancer, patients who are traveling for extended periods, and patients who are pregnant. (See "Neuraxial anesthesia/analgesia techniques in the patient receiving anticoagulant or antiplatelet medication" and "Risk and prevention of venous thromboembolism in adults with cancer" and "Deep vein thrombosis and pulmonary embolism in pregnancy: Prevention" and "Prevention and treatment of venous thromboembolism in patients with acute stroke".)
●Duration – VTE prophylaxis should typically continue until the patient is ambulatory or discharged from the hospital. We suggest not routinely extending the duration of thromboprophylaxis beyond the period of the acute hospital stay (Grade 2B). However, select populations may benefit from extended duration prophylaxis including nonambulatory patients admitted to acute rehabilitation for weaning from mechanical ventilation. (See 'Duration of prophylaxis' above.)
