Extended-Duration Thromboprophylaxis Among Acute Medically Ill Patients

Abstract

Acute medical illnesses are associated with a prolonged elevation in inflammatory markers that predisposes patients to thrombosis beyond the duration of their hospital stay. In parallel, both observational and randomized data have demonstrated a rate of postdischarge venous thromboembolic events that often exceeds that observed in the hospital setting. Despite this significant residual risk of venous thromboembolic events following discharge among acute medically ill patients, no therapeutic strategies have been recommended to address this unmet need. Available randomized trials have demonstrated the efficacy of extending the duration of thromboprophylaxis with available anticoagulants; however, the efficacy is offset, at least in part, by an increase in bleeding events. Identification of the optimal therapeutic strategies, treatment duration, and risk assessment tools that reconcile both efficacy and safety of extended-duration thromboprophylaxis among acute medically ill patients is an area of ongoing investigation.

Venous thromboembolism (VTE) is a common cause of morbidity and mortality among patients with acute medical illnesses. Thromboprophylaxis is currently indicated only during hospitalization among medically ill patients at high risk of thrombosis, and the choice of thromboprophylaxis (mechanical vs pharmacologic) depends on the risk of bleeding. ¹ Nonetheless, a significant residual risk of VTE persists, and the majority of VTE events occurs following discharge. ^2,3 The optimal duration of thromboprophylaxis among medically ill patients currently remains poorly defined. Despite a significant reduction in VTE events among surgical patients with extended thromboprophylaxis, postdischarge VTE prophylaxis that is both safe and effective among acute medically ill patients remains an unmet medical need. Studies have consistently demonstrated that extended-duration thromboprophylaxis after hospitalization is associated with a reduction in postdischarge VTE events when compared to standard in-hospital thromboprophylaxis. However, the efficacy of extended-duration thromboprophylaxis has been offset, at least in part, by an increase in bleeding events. The optimal therapeutic strategies, treatment duration, and risk assessment tools to optimize the efficacy–safety ratio of extended-duration thromboprophylaxis among patients with acute medical illnesses following discharge are areas of ongoing investigation.

Scope of the Problem

Venous thromboembolism is a common cause of death worldwide. Global estimates of VTE range between 0.1 and 2.7 cases per 1000 and up to 3 cases per 100 hospitalizations worldwide every year. ^4,5 In the United States alone, estimates suggest approximately 900 000 VTE cases yearly, 10% to 30% of which are fatal. ^6,7 The incidence of VTE among medically ill hospitalized patients varies substantially depending upon patient risk and ranges from 0.3% to 11% through 3 months. ⁸ In-hospital thromboprophylaxis with either unfractionated heparin, low-molecular-weight heparin, or fondaparinux is indicated in approximately 90% of moderate- to high-risk medically ill patients ^{9 –11} and is associated with up to a 50% reduction in in-hospital VTE risk. ¹¹

In-hospital thromboprophylaxis, however, has not been associated with a reduction in the incidence of postdischarge VTE, which accounts for up to 74% of total VTE events. ^3,12 Despite the fact that the majority of VTE events occur following discharge, the current American College of Chest Physicians (ACCP) guidelines based upon available data from randomized trials recommend against the extension of thromboprophylaxis beyond the hospitalization period. ¹ The majority of hospitalized medically ill patients remains in the hospital for 5 days and receive a relatively short course of anticoagulant thromboprophylaxis, which is often insufficient for the prevention of VTE events that occur several days to weeks following hospital discharge.

Biology

The bidirectional association between inflammation and thrombosis sets in motion a positive feedback loop. It is hypothesized that proinflammatory cytokines, such as tumor necrosis factor-α ¹³ and interleukin-1β, ¹⁴ initially induce the expression of tissue factor by mononuclear cells and smooth muscle cells of the endothelial media and intima. ¹⁵ In turn, tissue factor binds to coagulation factor VII and activates the downstream coagulation cascade. There is accumulating evidence that proteases generated by the activated coagulation pathway induce conformational changes within protease-activated receptors, which in turn self-activate and perpetuate inflammatory responses. ^{16 –18}

Although inflammation-induced thrombosis is a physiologic phenomenon, the exaggerated prothrombotic response to inflammatory signals partly contributes to the increased risk of thrombotic events. Acute medical illnesses that involve malignancies, ^19,20 acute coronary syndromes, ^{21 –25} heart failure, ^{26 –29} chronic kidney disease, ^{30 –32} and respiratory diseases ^{33 –35} are associated with a prolonged increase in inflammatory markers, which may therefore predispose patients to thrombosis following discharge and supports the need for extended thromboprophylaxis in this population.

Observational Studies

Compared with randomized clinical trials, observational studies of large populations provide real-world insights into the magnitude of the unmet need for the extended-duration thromboprophylaxis among acute medically ill patients. Through 3 months following hospital discharge, the rate of confirmed symptomatic VTE events can range from 0.6% to 4%. ^{36 –38}

Timing of VTE

The timing of VTE has been assessed in several observational studies. The median time to a VTE event among patients with acute medical hospitalizations (8 to 23 days) is beyond the median lengths of hospitalization (5 to 7 days) in observational studies. The IMPROVE observational study (International Medical Prevention Registry on Venous Thromboembolism) demonstrated that half of the VTE events occurred following discharge (n = 15 156). ^12,36 Greater granularity of the timing of VTE was provided by the Worcester DVT study, which evaluated the medical records of patients who sustained VTE events within 3 months of medical hospitalization. ² Outpatient VTE events accounted for approximately 74% of all cases and were at least twice as likely to occur in the first 29 days following discharge than in the following months (66.9% for 0-29 days vs 33.1% for 30-90 days; n = 1897), suggesting that the early postdischarge phase may confer the highest VTE risk among medically ill patients and may provide a potential time frame for optimizing the duration of extended thromboprophylaxis. ² Another study of high-risk elderly medical patients (n = 989) demonstrated that approximately 80% of all VTE events occur within 57 days postdischarge. ³⁷

Risk Assessment of Postdischarge VTE

The rate of VTE events through 6 months following hospital admission varies among medical conditions and ranges from 2.1% in severe lung disease/chronic obstructive pulmonary disease, 3.1% in congestive heart failure, 3.4% in infectious diseases, to a high of 5.7% among patients with malignancy, more than 50% of which occur following discharge. ³⁹ An increased number of risk factors has been associated with a higher incidence of postdischarge VTE events (2.9% ≤2 risk factors vs 6.1% for ≥3 risk factors vs 8.7% for ≥4 risk factors; P = .01). ³⁷

Current Rate of Postdischarge Prophylaxis

Although it is estimated that pharmacological thromboprophylaxis is required in up to 90% of moderate- to high-risk medical patients, only 30.7% to 64.1% of patients actually receive thromboprophylaxis during their hospitalization and as few as 1.8% to 12% continue extended thromboprophylaxis following discharge (Table 1). ^{9,12,37,39 –41} The rate of extended thromboprophylaxis varies according to the primary medical diagnosis and ranges from 1.1% among patients hospitalized for infectious diseases to 18.5% among patients hospitalized for congestive heart failure. ^39,41

Table 1.

Rate of Extended-Duration Thromboprophylaxis Among Medically Ill Patients Following Discharge.

Observational Study	Sample Size (n)	Type of Thromboprophylaxis	Overall Rate of Extended-Duration Thromboprophylaxis Following Discharge (%)
Amin et al 41	9675	Pharmacologic	1.8
Hull et al 37	989	Pharmacologic	2.0
Spyropoulos et al 40	158 325	Pharmacologic	4.6
Amin et al 39	11 139	Pharmacologic	8.8
Tapson et al 12	15 156	Pharmacologic and mechanical	12

Data From Randomized Clinical Trials

MEDENOX (Prophylaxis in Medical Patients with Enoxaparin) was a double-blind randomized clinical trial that evaluated the need for in-hospital thromboprophylaxis (6-14 days) with enoxaparin among medically ill patients (n = 1102). ¹⁰ Although in-hospital administration of enoxaparin (40 mg) was associated with a 63% reduction in VTE events, a significant residual risk of VTE persisted following discharge. Compared with the in-hospital treatment period, the incidence of fatal pulmonary embolisms (PEs) was higher in the follow-up period across all study arms, and despite not being powered to evaluate mortality, there was no significant difference in mortality between treatment and placebo groups at the end of follow-up (day 110). ¹⁰ The erosion of the early benefit during the postdischarge period raises the question of whether postdischarge thromboprophylaxis would preserve the benefit of in-hospital therapy.

Findings similar to those of MEDENOX were observed in ARTEMIS (Fondaparinux for the Prevention of VTE in Acute Medically Ill Patients), a randomized double-blind clinical trial that evaluated in-hospital fondaparinux prophylaxis (6-14 days) among medical patients aged 60 years and older. Although fondaparinux prophylaxis was associated with a significant 46.7% reduction in the composite of DVT by bilateral venography at day 6 to 15 along with symptomatic VTE up to day 15, this early benefit did not persist. There was no significant difference in the rates of fatal PEs following the period of active treatment through 1 month. Four PE events (3 of which were fatal) were reported in the fondaparinux arm compared with 6 PE events (2 of which were fatal) in the placebo group following hospital discharge. ⁴² These findings highlight the limited impact of in-hospital thromboprophylaxis on postdischarge VTE risk, particularly fatal PE.

Extended Clinical Prophylaxis in Acutely Ill Medical Patients (EXCLAIM) was the first trial to evaluate the safety and efficacy of extended-duration thromboprophylaxis, investigating enoxaparin in the acute medically ill patient population (n = 6085). ⁴³ After a period of open-label enoxaparin administration for 10 ± 4 days, patients were then randomized to either enoxaparin 40 mg once daily or placebo for an additional 28 ± 4 days. Extended-duration enoxaparin was associated with a significant reduction in VTE incidence, defined as the composite of symptomatic or asymptomatic proximal DVT, symptomatic PE, or fatal PE through 28 days (2.5% vs 4.0%, absolute risk difference = −1.53%, 95% confidence interval [CI]: −2.54% to −0.52%). ⁴³ However, extended-duration enoxaparin was also associated with a significant increase in major bleeding (0.3% vs 0.8%, absolute risk difference = 0.51%, 95% CI: 0.12%-0.89%). A total of 4 intracranial bleeds were reported with the extended-enoxaparin group vs 0 in the placebo group. ⁴³ The risk of bleeding associated with postdischarge thromboprophylaxis in EXCLAIM was similar to that associated with in-hospital prophylaxis. ^10,43–44 Overall, the benefit of extended-duration enoxaparin in EXCLAIM was greater in several subgroups, namely women, patients older than 75 years, and patients with significant level 1 immobility (total bed rest or sedentary without bathroom privileges for at least 3 days before and 3 days after enrollment).

The MAGELLAN trial (Multicenter, Randomized, Parallel Group Efficacy and Safety Study for the Prevention of Venous Thromboembolism in Hospitalized Acutely Ill Medical Patients Comparing Rivaroxaban with Enoxaparin) evaluated the safety and efficacy of in-hospital and extended prophylaxis with the oral agent rivaroxaban among medically ill patients. ⁴⁵ Patients were randomized to either 40 mg of once-daily subcutaneous enoxaparin for 10 ± 4 days or 10 mg of oral once-daily rivaroxaban for an extended period of 35 ± 4 days. Rivaroxaban was noninferior to enoxaparin in the reduction of VTE events during the initial period of hospitalization (2.7% vs 2.7%, P = .003) and was superior to enoxaparin followed by placebo at 35 days (4.4% vs 5.7%, P = .02) at the expense of significantly higher major bleeding events (4.1% vs 1.7%, P < .001). ⁴⁵ Rivaroxaban administration was associated with a 35% reduction in VTE events between 11 and 35 days. ^45,46 It is unknown if even greater benefits would have been observed with even more prolonged thromboprophylaxis.

Following EXCLAIM and MAGELLAN, the Apixaban Dosing to Optimize Protection from Thrombosis (ADOPT) trial investigated extended-duration thromboprophylaxis with apixaban among acute medically ill patients. The ADOPT trial randomized subjects to either oral apixaban 2.5 mg twice daily for 30 days or subcutaneous enoxaparin 40 mg once daily for 6 to 14 days. The primary efficacy end point was a composite of fatal and nonfatal VTE during the 30-day treatment period. Although the primary efficacy end point was not met, analysis of the postparenteral treatment period demonstrated a numerically lower rate of fatal and nonfatal VTE in the extended-duration apixaban arm compared with in-hospital enoxaparin (18 vs 31, relative risk [RR] = 0.59, 95% CI: 0.33-1.05). The difference was more pronounced when only symptomatic VTE was analyzed (8 vs 18, RR = 0.44; 95% CI: 0.19-1.00). These findings support prior investigations and suggest a conceivable benefit for the strategy of extended prophylaxis among medically ill patients. ⁴⁶

Economic Costs of VTE

The economic costs of VTE include not only inpatient hospital costs but also costs related to diagnostic imaging, drug acquisition, monitoring such as international normalized ratio testing, and those of VTE-related complications such as bleeding or recurrent VTE. Among patients who develop VTE, 19% to 35% have a recurrent VTE episode and 15% to 23% develop postthrombotic syndrome within 1 year. ^37,47,48 Recurrent DVT events are associated with a prolonged hospital stay and a significant 21% increase in the total inpatient and outpatient costs compared with the initial DVT event (P = .006). ⁴⁷

Although no studies to date have been able to directly evaluate the cost effectiveness of extended-duration thromboprophylaxis among medically ill patients, in-hospital thromboprophylaxis among medically ill patients and extended-duration thromboprophylaxis among surgical patients have been demonstrated to be cost effective. ^{49 –55} The cost of VTE treatment is as much as 4 times higher than the cost of thromboprophylaxis among hospitalized medical patients. ⁵⁵ Hospitalization and human resource costs account for the largest proportion of the financial burden of VTE treatment and greatly exceed the costs of thromboprophylaxis drug acquisition. ⁵⁵

Despite bleeding risks, extended-duration thromboprophylaxis among medically ill patients may prevent 3 to 7 fewer symptomatic proximal DVT per 1000 patients treated. ⁴³ It is estimated that the total annualized health care costs of a patient who presents with DVT may be as high as US$7594 to US$10 804 and for pulmonary embolism may be as high as US$13 018 to US$16 644. ⁴⁷ Given the clinical benefits of extended-duration thromboprophylaxis among medically ill patients and the high cost associated with the management of VTE, extended-duration thromboprophylaxis may be cost effective and warrants further investigation.

Conclusion

Despite appropriate in-hospital thromboprophylaxis, a large number of patients hospitalized for an acute medical illness are not discharged on thromboprophylaxis and remain at risk for recurrent VTE events following discharge. The majority of VTE events among medically ill patients occur within a 1-month period following hospital discharge, a time during which the majority of patients are not administered thromboprophylaxis at present. Given the considerable mortality, morbidity, and associated financial burden of VTE, improved postdischarge thromboprophylaxis strategies that optimize efficacy and minimize bleeding are needed. Elderly patients and those with reduced mobility, cancer, a positive D-dimer, or prothrombotic conditions are at increased risk of events and may therefore derive greater net clinical benefit with postdischarge thromboprophylaxis. Identification of the optimal patients who have the greatest net clinical benefit as well as identification of the optimal dose and duration of thromboprophylaxis are exciting areas of ongoing investigation.