LMWH for the treatment of acute VTE
How should LMWH be initiated, including baseline laboratory tests and dosing?
Enoxaparin (Lovenox) and dalteparin (Fragmin) are the LMWHs available in the US. Enoxaparin is approved for inpatient treatment of acute DVT with or without PE and for outpatient treatment of acute DVT without PE using 1 mg/kg SQ q12 h or 1.5 mg/kg q24 h . Dalteparin is approved for VTE in patients with cancer at a dose of 200 units/kg SQ q24 h . Dosing is generally based on total body weight and renal function, evaluated using the Cockcroft-Gault method, further influences dosing requirement .
Use of enoxaparin 1.5 mg/kg once daily for the treatment of VTE is controversial. This dosing option is based on a single randomized, clinical trial comparing unfractionated heparin (UFH) to enoxaparin 1.5 mg/kg daily or 1 mg/kg twice daily in 900 patients with VTE . While there was no difference in recurrent VTE or major bleeding between the groups as a whole, only 32 % of the patients enrolled had PE at the time of randomization. Patients with symptomatic PE, obesity and malignancy all had higher rates of recurrent VTE when treated with 1.5 mg/kg daily versus 1 mg/kg twice daily. Current guidelines suggest that when enoxaparin is used for the treatment of VTE, it should be dosed at 1 mg/kg twice daily and that the reduced dose delivered by 1.5 mg/kg once daily be avoided [11, 60].
Limiting dalteparin to the treatment of cancer-associated VTE is not necessary. Dalteparin is highly effective for the treatment of VTE in patients without malignancy using 200 units/kg once daily or 100 units/kg twice daily as determined in a number of clinical trials [11, 61, 62].
Although the risk of HIT with LMWH is lower than with UFH, a baseline platelet count is recommended as a basis from which to consider the development of HIT. Re-exposure to LMWH should be avoided in a patient with a known history of HIT .
Therapy should be initiated as soon as possible, as long as it is determined that fibrinolytics are not going to be administered for acute VTE. A pre-treatment hemoglobin and/or hematocrit are used as a baseline from which to assess subsequent changes that may reflect bleeding. Finally, an elevated pre-treatment PT or aPTT may detect the presence of an underlying coagulation defect.
We suggest that total body weight, baseline serum creatinine, CBC, PT and aPTT be obtained prior to initiating LMWH. We suggest that when enoxaparin is used for the treatment of VTE, only the twice daily dosing strategy be used, except in patients with severe renal insufficiency (see below). Further, we suggest that once daily dalteparin can be used for the treatment of both cancer- and non-cancer-associated VTE.
What weight should be used to calculate dosing, and should obese and low body weight patients be treated differently?
In clinical trials evaluating the effectiveness and safety of LMWH for the treatment of VTE, total body weight has been used to calculate dosing. While data evaluating the safety and efficacy of LMWH in VTE patients with extremes of weight is limited, total body weight is recommended for LMWH dosing . Due to concerns that dosing based on total body weight may lead to over-anticoagulation in obesity, dose capping has been suggested and is recommended in the product information for dalteparin. However, several studies show little to no accumulation in patients given uncapped doses with body weights up to 190 kg with dalteparin and 159 kg with enoxaparin [59, 64]. In addition, limiting the dose of LMWH by using dose capping may result in inadequate anticoagulation and an increased risk of recurrent VTE .
Pharmacodynamic studies involving LMWHs have included patients weighing up to 190 kg, and the maximum weight of patients enrolled in clinical trials is 196 kg [64, 66]. In a retrospective cohort of 300 patients receiving enoxaparin for VTE treatment, the incidence of bleeding events was similar between patients with a BMI ≥ 40 kg/m2; (maximum 66.4 kg/m2) and non-obese patients (29 % vs. 23.1 %, p = 0.43) . A multivariate analysis concluded that obesity was not associated with an increased risk of bleeding. The average dose of enoxaparin was clinically similar between the groups (0.98 mg/kg vs. 1.04 mg/kg) with the majority of patients receiving twice daily dosing (97 % and 91.5 % respectively). The incidence of new thromboembolic events was statistically similar (3.5 % vs. 2 %, p = 0.72).
As noted previously, a subgroup analysis from a retrospective study suggested that VTE may recur more often in overweight and obese patients (BMI > 27 kg/m2) treated with enoxaparin once daily compared to twice daily (7.3 % vs. 3.4 %; OR 4.0 [CI 1.08–15]) . This difference may reflect the benefits of a higher total daily dose with the twice daily regimen. A retrospective study of 193 patients weighing >90 kg treated with dalteparin 200 International units/kg total body weight for VTE revealed only 2 major bleeding events which were deemed unlikely to be caused by dalteparin .
There are limited data on dosing LMWH in patients with low body weight. The lowest body weight reported in an enoxaparin VTE clinical trial was 44 kg and patients <40 kg were excluded from the major dalteparin VTE clinical trial [43, 69]. A registry of 7962 patients receiving LMWH for acute VTE analyzed clinical outcomes based on weight ranges: (less than 50 kg vs. 50–100 kg, vs. greater than 100 kg] . The majority of patients weighed between 50–100 kg; only 242 patients weighed >100 kg and only 161 patients weighed <50 kg. Compared to patients weighing 50–100 kg, patients <50 kg had a significant increase in the incidence of major bleeding (3 % vs. 1.3 %) and minor bleeding (5.3 % vs. 2.5 % [OR 2.2; 95 %CI 1.2–4]). Mean daily doses were significantly higher in the <50 kg group with 54 % receiving >200 international units/kg daily. The incidence of recurrent VTE was similar between the <50 kg and 50–100 kg groups. Patients >100 kg experienced similar bleeding and thromboembolic complications compared with the 50–100 kg group.
For patients weighting >190 kg, peak anti-Xa monitoring has been suggested . However, an open label prospective trial in 233 patients showed that mean peak anti-Xa levels were similar between obese and healthy weight individuals receiving enoxaparin 1.5 mg/kg once daily or 1 mg/kg twice daily  and peak anti-Xa levels have not been correlated with effectiveness (see below).
We suggest that in all patients, including underweight and obese, LMWH dosing should be based on total body weight. For patients <40 kg, UFH may be more appropriate. For enoxaparin dosing in obese patients, 1 mg/kg BID is preferred over 1.5 mg/kg daily. Dose capping should be avoided. Routine monitoring of peak anti-Xa levels is not suggested in patients on LMWH, whether obese or non-obese.
How should patients with renal impairment be treated?
LMWHs are cleared renally. There is an inverse relationship between CrCl and anti-Xa levels, with accumulation of anti-Xa activity at the end of the dosing interval as renal function declines [12, 71, 72].
Enoxaparin appears to be more dependent on renal function for elimination than is dalteparin . Product information for enoxaparin includes a dose reduction to 1 mg/kg daily for patients with CrCl < 30 mL/min . A pharmacokinetic study showed that using this reduced dose of enoxaparin resulted in 74 % of peak anti-Xa levels being within an expected range of values . After repeated dosing, higher peak anti-Xa levels were reported in patients receiving enoxaparin 1 mg/kg BID compared to 1.5 mg/kg daily in both moderate (CrCl 30–50 mL/min) and severe (CrCl < 30 mL/min) renal dysfunction, indicating greater accumulation with the BID regimen in renal insufficiency .
In comparison, dalteparin product information includes no dose adjustment for patients with severe renal impairment, and recommends to use with caution and “monitor anti-Xa levels” in patients with CrCl < 30 mL/min) . One prospective study evaluating dalteparin 100 International units/kg every 12 h found no difference in peak anti-Xa levels in patients with CrCl < 40 mL/min compared to patients with normal renal function (0.47 U/mL vs. 0.55 U/mL, p > 0.5.) .
Compared to patients with normal renal function, the risk of major bleeding increases in patients with renal insufficiency exposed to LMWH. In a prospective registry of 1037 patients on LMWH, patients with CrCl < 30 mL/min had an increased incidence of major bleeding (7.3 % vs. 2.3 %; p < 0.001) . A systematic review and meta-analysis of 18 LMWH studies (4971 patients) showed that patients with CrCl ≤ 30 mL/min had a significant increase in major bleeding compared to patients with CrCl > 30 mL/min (5 % vs. 2.4 %; odds ratio 2.25 [95 % CI, 1.19–4.27]; p = 0.013) . Fifteen of the 18 studies evaluated enoxaparin and seven of those involved the use of therapeutic dosing rather than dosing for VTE prophylaxis. When data were analyzed based on LMWH preparation, major bleeding was increased with standard dose enoxaparin (8.3 % vs. 2.4 %; odds ratio, 3.88 [CI 1.78–8.45]) but not when dose was adjusted for CrCl (0.9 % vs. 1.9 %; odds ratio, 0.58 [CI, 0.09–3.78] p = 0.23). There were no data on bleeding associated with the use of dalteparin.
An increased risk of bleeding has also been observed in patients with moderate renal impairment. A retrospective study compared major bleeding in patients receiving treatment dose enoxaparin with normal renal function (CrCl > 80 mL/min) and moderate renal impairment (CrCl 30–50 mL/min) . The incidence of major bleeding was 5.7 % with normal renal function compared to 22 % with moderate renal impairment, unadjusted odds ratio of 4.7 (95 % CI, 1.7–13; p = 0.002). A dose reduction for enoxaparin use in patients with mild or moderate renal impairment (CrCl 30–80 mL/min) has not been established.
Extended use (>10 days) of enoxaparin in patients with renal insufficiency may require trough anti-Xa measurement and dose adjustments if accumulation is noted. More data on dose adjustment in renal impairment are needed. One clinical approach to dosing enoxaparin in renal insufficiency is to utilize manufacture dose recommendations for CrCl < 30 mL/min but avoid LMWH if the CrCl < 20 mL/min .
LMWH is routinely avoided in patients on renal replacement therapy (RRT) because of the numerous variables that can affect clearance (filter type, interruption, regimen change). LMWH dose adjustments for RRT are not well defined .
When LMWH is used for acute treatment of VTE in patients with renal impairment, we suggest that vigilant attention to potential bleeding risk and monitoring for signs and symptoms of bleeding be employed. Renal function should be estimated using the Cockcroft-Gault method for calculating CrCl. In patients with a CrCl < 30 mL/min the use of UFH may be preferred over LMWH and if enoxaparin is used, it should be dosed at 1 mg/kg daily. If LMWH is used for an extended period beyond the usual 5–7 days of treatment, trough anti-Xa measurement may be considered in patients with severe renal dysfunction. LMWH should be avoided in patients with CrCl < 20 mL/min and those receiving renal replacement therapy.
How should routine treatment be monitored?
LMWHs have predictable pharmacodynamic profiles and wide therapeutic windows that do not require routine coagulation monitoring in clinically stable and uncomplicated patients. There are currently no commercial assays available for LMWH. PT and aPTT are insensitive measures of LMWH activity. Anti-Xa activity is a surrogate marker that measures the anticoagulant effect of LMWH and is assumed to correlate with hemorrhagic and thromboembolic events. While LMWH anti-Xa concentrations may be helpful in evaluating dosing in special patient populations, routine LMWH anti-Xa monitoring is unnecessary and potentially harmful if misinterpreted [12, 59, 78].
Although the risk of HIT is < 1 % in patients on LMWH, the consequences of HIT can be devastating. Therefore, in patients with acute VTE, we suggest that a baseline platelet count be obtained prior to initiation of LMWH, and occasionally during the first 2 weeks of LMWH use. Circulating HIT antibodies may remain present for a median of 50–85 days depending on assay performed and re-exposure can lead to a large decrease in platelet count within 24 h. Therefore, in patients recently treated with heparin/LMWH, a baseline platelet count should be obtained prior to initiating LMWH and repeated 24 h later .
LMWHs are excreted by the kidney and accumulation may occur in renal impairment. Occasional monitoring of renal function using serum creatinine, and calculation of CrCl using the Cockcroft-Gault method may be useful to assess changes in renal function that may indicate the need for a dosing adjustment.
We suggest all patients receiving LMWH be monitored for signs and symptoms of bleeding and be observed for changes in renal function that may require a dose adjustment. We suggest against the routine use of LMWH anti-Xa monitoring. CBC, platelet count and Scr should be assessed periodically during LMWH treatment.
Is there a role for peak anti-Xa monitoring and for trough anti-Xa monitoring?
The clinical trials evaluating LMWH did not use anti-Xa levels to guide dosing and anti-Xa levels have not been evaluated in large studies. Although anti-Xa levels have been used as a marker of LMWH activity they are not routinely evaluated in clinically stable or uncomplicated patients. The interpretation of anti-Xa levels depends on the dose and time of last LMWH administration. Trough anti-Xa levels may be used to evaluate accumulation of anticoagulant effect at the end of dosing interval. The value of peak anti-Xa levels is less clear. Peak levels occur 3–5 h after a LMWH dose and if obtained, should be measured at steady state [12, 79]. In a retrospective review, the majority of anti-Xa levels were drawn inappropriately, limiting their utility for interpretation .
Data supporting a relationship between elevated LMWH anti-Xa levels and bleeding are quite limited and include a study in which dalteparin was administered by continuous infusion and bleeding was increased in those with mean levels above 0.8 u/mL [80, 81]. In the uncommon situations in which anti-Xa activity is monitored, it should be determined using a chromogenic method and a calibration curve based on the LMWH used. Target anti-Xa levels are not clinically validated, and there is no standardized method for adjusting doses based on anti-Xa level . Peak anti-Xa levels observed in patients treated with enoxaparin range from 0.6–1 IU/mL for twice daily dosing and >1 IU/mL for once daily dosing. For dalteparin, observed peaks may be somewhat higher simply because the total dose is given as a single injection (200 units/kg SQ once daily) rather than divided into two doses as in the case of enoxaparin (1 mg/kg SQ q12 h) [12, 79, 82]. Importantly, there are no data to suggest that making dosing adjustments based on peak levels is correlated with improved safety or efficacy.
While there is no consensus on an acceptable trough anti-Xa level for treatment dose LMWH, at the end of the 12 or 24 h dosing interval, these values should not be ‘high’ . In an acute coronary syndrome study, trough anti-Xa levels >0.5 IU/mL were considered to be elevated . Elevated troughs reflect lack of LMWH clearance and may suggest both an increased risk of bleeding and the need for a prolonged dosing interval.
Trough anti-Xa concentrations may be helpful to evaluate the safety of LMWH dosing in special patient populations including patients with severe renal impairment (although usefulness undetermined in patients on RRT) and extremely low body weight [12, 59]. The role of peak anti-Xa concentrations for evaluating efficacy in special populations including pregnancy and extremes of body weight is not defined.
We suggest that in limited populations, including patients with severe renal failure, trough anti-Xa levels may have a role in evaluating LMWH accumulation and the need to prolong the dosing interval. We suggest that peak anti-Xa levels not be utilized to evaluate dosing regimens in clinical practice.
What is the appropriate duration of therapy when transitioning to oral anticoagulant therapy?
Traditional anticoagulation involves concurrent initiation of LMWH and warfarin on the same day, with continuation of LMWH for a minimum of 5 days and until the INR is above 2.0 (see above and Table 3)  When dabigatran or edoxaban are used for VTE treatment, LMWH must be started first and continued for a minimum of 5 days prior to initiation of these oral anticoagulants [47, 49, 50].
LMWH alone is an option for patients in whom INR is difficult to control or in whom oral anticoagulation is not an option, and is more effective than VKA therapy in patients with cancer . Several randomized, controlled trials have evaluated the safety and efficacy of LMWH for the full course of VTE treatment compared to traditional short term LMWH followed by oral vitamin K antagonist (VKA) therapy. Among trials with the highest methodological quality, a recent meta-analysis showed a non-significant reduction in the odds of recurrent VTE (OR 0.80, 95 % CI 0.54–1.18) and in the odds of bleeding (OR 0.62, 95 % CI 0.36–1.07) favoring LMWH . Nevertheless, oral anticoagulation remains a more common approach due to the expense of LMWH and the need for drug delivery by injection.
Parenteral anticoagulation with LMWH should be overlapped with warfarin for at least 5 days and until a single INR is 2.0 or greater. Treatment of VTE with rivaroxaban and apixaban does not require initial parenteral anticoagulation while dabigatran and edoxaban require a minimum of 5 days of parenteral anticoagulation prior to initiation. See Table 4for additional details. The timing of the first dose of a TSOAC is based on when the next scheduled dose of LMWH would be due.
Which patients are acceptable candidates for outpatient treatment of VTE with LMWH?
A number of randomized trials have compared outpatient treatment of DVT with LMWH versus inpatient treatment of DVT with either UFH or LMWH. A Cochrane review of 6 randomized controlled trials including 1708 patients with DVT showed that outpatient therapy was associated with a lower rate of recurrent VTE, reduced mortality and no difference in minor bleeding .
While PE has historically been treated on an inpatient basis, a systematic review and meta-analysis of 11 studies, including 1258 patients, showed that low risk patients with PE can safety be treated as outpatients . Studies included in the meta-analysis utilized either a risk stratification method or clinical judgment to determine low risk patients. The incidence of VTE recurrence and major bleeding was low in the studies and the event rates between the studies that used a risk stratification model versus clinical judgment were similar. Approximately one-third to one-half of acute PE patients may be classified as low-risk .
Clinical prediction rules, including the Pulmonary Embolism Severity Index (PESI) and Simplified PESI are simple tools to identify low-risk PE patients (mortality <1 %) with excellent negative predictive performance . In a randomized controlled trial, acute PE patients with low risk PESI scores who were treated as outpatients with enoxaparin 1 mg/kg twice daily had non-inferior outcomes compared to those treated as inpatients .
Characteristics of patients who may be less suitable for outpatient VTE management have been identified and include body weight <70 kg, active malignancy, recent immobility, chronic heart failure, renal insufficiency, and bilateral DVT. In the REITE registry, these factors were independently associated with an increased risk of symptomatic PE, recurrent DVT, major bleeding or death . In addition, DVT patients who may require hospitalization include those with venous gangrene or extensive iliofemoral involvement, severe acute obstruction (phlegmasia cerula dolens), poor social circumstances, active bleeding or a high risk of bleeding, severe pain, renal impairment, significant communication deficits or mobility problems [11, 92]. Home circumstances for adequate outpatient treatment include well-maintained living conditions, strong support from family or friends, phone access and ability to quickly return to hospital if there is clinical deterioration .
We suggest patients with VTE be evaluated to determine treatment setting. Patients with DVT and/or PE who are identified as having a low risk of complications should be treated in the outpatient setting as long as they have adequate home support.
How should LMWH-induced over-anticoagulation, thrombocytopenia and bleeding be managed?
The overall risk of major bleeding associated with LMWH ranges from 1–4 % . When significant bleeding or over-anticoagulation occurs, LMWH should be discontinued immediately. Observation without intervention is appropriate if bleeding is not present as demonstrated in a case series of intentional LMWH overdose . Protamine sulfate may be used as a reversal agent for LMWH, but only reverses 60–80 % of the anticoagulant activity of LMWH. While it fully reverses the anti-IIa fraction of LMWH, it only partially reverses the anti-Xa component of LMWH due to the reduced sulfate charge in the ultra-low molecular weight heparin fragments present. Enoxaparin appears to be less susceptible to protamine sulfate reversal than dalteparin because its structure has less sulfonation .
There are limited clinical data on the use of protamine sulfate to reverse LMWH [93, 96]. A retrospective, single center study that evaluated the use of protamine sulfate to emergently reverse LMWH found that 4 of 14 patients with active bleeding continued to bleed or rebled after protamine administration . No correlation was evident between anti-Xa levels and bleeding cessation.
The need for and dose of protamine sulfate is based on the timing of the last dose of LMWH, the severity of bleeding and the estimated clearance of LWMH based on the patient’s renal function. In patients with impaired renal function the anticoagulant effect of LMWH may persist and the treatment window for protamine sulfate may be extended. Current guidelines suggest that if LMWH was given within the previous 8 h, protamine sulfate may be administered in a dose of 1 mg per 100 anti-Xa units of LMWH up to a maximum single dose of 50 mg (1 mg of enoxaparin equals approximately 100 anti-Xa units) . A second dose of 0.5 mg protamine sulfate per 100 anti-Xa units should be administered if bleeding continues or if the aPTT is prolonged 2–4 h after the initial protamine dose. A lower initial dose of protamine sulfate (0.5 mg per 100 anti-Xa units) may be administered if the time since LMWH administration is greater than 8 h. If greater than 12 h has elapsed since administration of LMWH, protamine sulfate may not be effective and supportive measures to control bleeding should be used.
The risk of HIT is lower for LMWH than for UFH . Nevertheless, if thrombocytopenia or thrombosis develops during LMWH treatment, the patient should be evaluated for HIT as outlined previously for UFH, and treated according to current guidelines .
We suggest that protamine sulfate be used to reverse LWMH if major bleeding occurs. The timing of the last dose of LMWH should be assessed when determining if protamine sulfate should be administered and the appropriate dose to be administered. A repeat dose of protamine may be administered if bleeding continues or if the aPTT is prolonged 2–4 h after the initial dose. Patients who develop HIT in response to LMWH should be treated according to current guidelines for HIT.
2. Hull, RD, Raskob, GE, Rosenbloom, D, et al (1992) Optimal therapeutic level of heparin therapy in patients with venous thrombosis. Arch Intern Med 152,1589-1595
3. Elliot, GC, Hiltunen, SJ, Suchyta, M, et al (1994) Physician-guided treatment compared with a heparin protocol for deep vein thrombosis. Arch Intern Med 154,999-1004.
4. Flaker, GC, Bartolozzi, J, Davis, V, et al (1994) Use of a standardized heparin nomogram to achieve therapeutic anticoagulation after thrombolytic therapy in myocardial infarction. Arch Intern Med 154,1492-1496