Pharmacological and Clinical Differences Between Low-Molecular-Weight Heparins

Abstract

Thanks to their predictable pharmacokinetics and ease of use, low-molecular-weight heparins (LMWHs) have established uses in the prevention and treatment of thrombotic diseases and as a replacement for unfractionated heparin (UFH). Although LMWHs as a class have similar antithrombotic effects, they comprise a diverse group of agents with distinct biochemical and pharmacological profiles. In light of the ongoing pressure to contain pharmacy costs, the diversity among the LMWHs and their benefits over UFH are important considerations in clinical practice.

INTRODUCTION

Both unfractionated heparin (UFH) and low-molecular-weight heparins (LMWHs) have established roles in preventing and treating venous thromboembolism (VTE) and as adjuvant therapies for atherothrombotic syndromes.^1,2 Numerous studies have compared the efficacy and safety of these agents.

For example, in a meta-analysis of 13 randomized, controlled trials that compared LMWHs and UFH in the treatment of VTE, the relative risks of recurrent VTE with LMWHs were 0.85 (95% confidence interval [CI], 0.65–1.21) and 1.02 for pulmonary embolism (PE) (95% CI, 0.64–1.12). Similarly, the relative risk of major bleeding was 0.63 (95% CI, 0.37–1.05) and the relative risk of minor bleeding was 1.18 (95% CI, 0.87–1.61).³

LMWHs are replacing UFH for therapeutic anticoagulation owing to a number of advantages, including a more predictable pharmacokinetic profile and their ease of use.⁴ The 2008 guidelines of the American College of Chest Physicians (ACCP) recommend LMWHs, UFH, or fondaparinux (Arixtra, Glaxo-SmithKline) as VTE prophylaxis for surgical and acutely ill medical patients.¹

Even though their antithrombotic effects are similar, the LMWHs are heterogeneous compounds; they are produced by different processes and have distinct biochemical and pharmacological properties (Table 1).^5,6 This diversity among the LMWHs has important implications for clinical practice because the continual pressure on pharmacy budgets has resulted in initiatives to reduce costs by switching to the product with the lowest price. This approach relies on the assumption that different LMWHs are pharmacologically and clinically equivalent, but such equivalence has not been demonstrated.⁵ Nevertheless, the practice of substituting one LMWH for another (so-called therapeutic interchange) continues to have an appeal, despite statements by the FDA, World Health Organization, ACCP, American Heart Association (AHA), and American College of Cardiology (ACC)^4,7,8 that LMWHs are not clinically interchangeable. The increasing availability of biosimilar LMWH formulations raises further concerns in this context because it cannot be assumed that such formulations are biochemically or pharmacologically equivalent to the branded formulations.⁹

Table 1.

Pharmacological Properties of Low-Molecular-Weight Heparins Available in the U.S.

	Enoxaparin	Dalteparin	Tinzaparin
Brand name	Lovenox	Fragmin	Innohep
Manufacturer	Sanofi-Aventis	Pfizer (for Eisai)	Leo (formerly DuPont)
Manufacturing process	Benzylation followed by alkaline hydrolysis	Controlled nitrous acid depolymerization	Heparinase digestion
Mean molecular weight (daltons)	4,500	6,000	6,500
Elimination half-life (hours)	4.5	3–5	3.4
Bioavailability (%)	90–92	87	87
Anti-Xa/anti-IIa ratio	3.8	2.7	2.8
Anti-Xa activity (IU/mg)	100	156	100

From Merli GJ, Vanscoy GJ, Rihn TL, et al. J Thromb Thrombolysis 2001;11:247–259. With kind permission of Springer Science and Business Media.⁵

Another consideration in treating atherothrombotic syndromes is the medicolegal implication of interchanging one LMWH for another if the substituted LMWH does not have an FDA-approved indication.⁵ Self-insured health systems make every effort to reduce such exposure for their physician staff. The FDA-licensed indications of a drug do not restrict prescribing of that agent by a physician for off-label use.¹⁰ Standards of care, such as clinical guidelines from authoritative bodies (such as the ACCP, ACC, and AHA), can provide some additional guidance on the suitability of the various parenteral anticoagulants.^1,2,8,11

In this review, we discuss therapeutic interchange programs for currently available LMWHs in the context of their pharmacological and clinical profiles and the experience with therapeutic interchange in order to facilitate the appropriate prescribing of these agents.

CURRENT USE

Anticoagulants are widely used for preventing VTE in a broad range of surgical and medical patients. Surveys have consistently shown that the most widely used agents are LMWHs, UFH, and vitamin K antagonists such as warfarin (Coumadin, Bristol-Myers Squibb).^12–16 For example, data from a registry of 3,778 patients in 38 hospitals in the U.S. show that approximately half of all patients undergoing total hip or knee replacement received warfarin and 45% received LMWHs.¹³ Similarly, a report from the International Medical Prevention Registry on Venous Thromboembolism (IMPROVE), which included data from more than 15,000 acutely ill, hospitalized medical patients from 12 countries worldwide, found that LMWHs were the most commonly prescribed anticoagulants; LMWHs were used in 34% of patients overall, and UFH was used in 11%.¹⁴ However, UFH tended to be used more often in the U.S. (21%) than in other countries (14%), whereas LMWH usage (14%) was lower than in other countries (140%).¹⁴

Among the LMWHs, enoxaparin (Lovenox, Sanofi-Aventis) is the most widely used agent, as assessed by studies of clinicians’ practices and clinical care^15,16 and by its inclusion in hospital formularies.^12,17 In a survey of 224 acute-care hospitals in the U.S., enoxaparin was on the formulary of 81.1% of hospitals, dalteparin (Fragmin, Eisai, Pfizer) was used in 17.3% of hospitals, and tinzaparin (Innohep, Celgene/Leo) in 1.6%.¹²

CHEMICAL AND PHARMACOLOGICAL DIFFERENCES

LMWHs are a diverse group of chemically distinct compounds. Most are prepared by various chemical or physical depolymerization techniques; others, such as tinzaparin, are produced by enzymatic depolymerization. As a result of these differences in manufacturing techniques, the available products differ markedly in terms of their mean molecular weights, mix of polysaccharide chain lengths, and pharmacological properties (see Table 1).^5,6

Compared with UFH, LMWHs have higher anti-Xa/anti-IIa ratios, which are typically between 2.0 and 4.0, depending on the mix of chain lengths and the molecular weight of individual preparations.⁴ These biochemical differences influence the in vivo properties of LMWHs, for example, nonspecific binding to body protein and cell surfaces (such as platelets or endothelia). In turn, this influences their pharmacokinetics, bioavailability, and plasma half-life (Table 1).^5,6

LMWHs also seem to differ in their effects on platelet function. In a study that compared the effects of different anticoagulants on the release of von Willebrand factor (vWF) in patients with unstable angina or non–Q-wave myocardial infarction (MI), significant increases in vWF release were noted in patients who were treated with dalteparin or UFH, whereas no such increase was seen in enoxaparin-treated patients.¹⁸ Increased vWF release was associated with an increased risk of death, MI, or revascularization.¹⁸ However, in another study, the differences seen in the effects of enoxaparin or dalteparin on vWF levels did not reach statistical significance.¹⁹ Variations between LMWHs have also been observed in preclinical models of hemorrhage, depending on the route of administration.²⁰

CLINICAL PROFILES

Three LMWHs are currently available in the U.S.: enoxaparin (Lovenox); dalteparin (Fragmin); and tinzaparin (Innohep). Each of these medications has its own indications and dosage recommendations (Table 2),^21–23 which reflect the available evidence of their efficacy and safety (Table 3).^24–48

Table 2.

FDA-Approved Indications for Low-Molecular-Weight Heparins in the U.S.

Indication	Enoxaparin		Dalteparin	Tinzaparin
	Standard Dose	Severe Renal Impairment (CrCl < 30 mL/minute)
Venous thromboembolism
Prophylaxis in hip replacement surgery	SQ 30 mg b.i.d. or SQ 40 mg q.d.a	SQ 30 mg q.d.	SQ 5,000 IU q.d. in post-operative period; initial doses depend on timingb	NA
Prophylaxis in knee replacement surgery	SQ 30 mg b.i.d. starting 12–24 hours after surgery	SQ 30 mg q.d. starting 12–24 hours after surgery	NA	NA
Prophylaxis in abdominal surgery	SQ 40 mg q.d. starting 2 hours prior to surgeryc	SQ 30 mg q.d. starting 2 hours prior to surgeryc	SQ 2,500 IU q.d. or SQ 5,000 IU q.d. starting 1–2 hours prior to surgeryc	NA
Prophylaxis in acutely ill medical patients	SQ 40 mg q.d.	SQ 30 mg q.d.	SQ 5,000 IU q.d.	NA
Secondary prophylaxis/extended treatment in cancer patients	NA	NA	SQ 200 IU/kg (max. 18,000 IU) q.d. for 1 month, then SQ 150 IU/kg (max. 18,000 IU) q.d. for 5 months	NA
Inpatient treatment of DVT, without PE	SQ 1 mg/kg b.i.d. with warfarin or SQ 1.5 mg/kg q.d. with warfarin	SQ 1 mg/kg q.d. with warfarin	NA	SQ 175 IU/kg q.d. until stable with warfarin
Outpatient treatment of DVT with/without PE	SQ 1 mg/kg b.i.d. with warfarin	SQ 1 mg/kg q.d. with warfarin	NA	NA
Arterial indications
Unstable angina and non-ST elevated MI	SQ 1 mg/kg b.i.d. with aspirin	SQ 1 mg/kg q.d. with aspirin	120 IU/kg b.i.d. with aspirin (cap: 10,000 IU/dose)	NA
Acute ST-elevation MI	IV 30 mg single bolus plus SQ 1 mg/kg b.i.d., thereafter with aspirind (cap first two SQ doses: 100 mg)	IV 30 mg single bolus plus SQ 1 mg/kg q.d., thereafter with aspirind (cap first two SQ doses: 100 mg)	NA	NA

b.i.d. = twice daily; CrCl = creatinine clearance; DVT = deep vein thrombosis; IU = International Units; IV = intravenous; kg = kilograms; MI = myocardial infarction; NA = not applicable; PE = pulmonary embolism; q.d. = every day; SQ = subcutaneous.

^aSQ 30 mg b.i.d. to be initiated 12 to 24 hours after surgery; SQ 40 mg q.d. to be initiated 12 ± 3 hours before surgery.

^bPostoperative start, 2,500 IU 4 to 8 hours after surgery; preoperative start on day of surgery, 2,500 IU within 2 hours before surgery and 4 to 8 hours after surgery; preoperative start on the evening before surgery, 5,000 IU 10–14 hours before surgery and 4 to 8 hours after surgery.

^cIn practice, prophylaxis is often initiated 12 to 24 hours postoperatively.

^dDose regimen in patients 75 years of age or older, no bolus; SQ 75 mg/kg twice daily (first dose not to exceed 75 mg). Data from prescribing information for Lovenox (enoxaparin), 2007;²¹ Fragmin (dalteparin), 2007²²; and Innohep (tinzaparin), 2008.²³

Table 3.

Controlled Clinical Studies of Enoxaparin, Dalteparin, and Tinzaparin^*

Indication/Agent	Study	n	Comparator	Efficacy, OR (95% CI)	Safety, OR (95% CI)
Venous thromboembolism
Prophylaxis in hip replacement surgery				VTE	Major bleeding
Enoxaparin	Turpie et al., 198624	100	Placebo	0.19 (0.07–0.52)a	0.49 (0.04–5.58)a
	Levine et al., 199125	665	UFH	0.88 (0.59–1.31)a	0.56 (0.26–1.20)a
Dalteparin	North American Fragmin Trial26	1,472	Warfarin	0.43 (0.30–0.60)a	1.78 (1.01–2.90)a
Tinzaparin	—	—	—	—	—
Prophylaxis in knee replacement surgery				VTE	Major bleeding
Enoxaparin	Fitzgerald et al., 200127	349	Warfarin	0.41 (0.26–0.64)a	2.36 (0.71–7.81)a
Dalteparin	—	—	—	—	—
Tinzaparin	—	—	—	—	—
Prophylaxis in acutely ill medical patients				VTE	Major bleeding
Enoxaparin	MEDENOX28	1,102	Placebo	0.37 (0.22–0.63)b	1.52 (0.42–5.42)a,b
	THE-PRINCE29	665	UFH	0.79 (0.42–1.49)a	1.00 (0.06–16.10)b
Dalteparin	PREVENT30	3,706	Placebo	0.55 (0.38–0.80)	2.99 (0.81–11.04)a
Tinzaparin	—	—	—	—	—
Initial treatment and secondary prophylaxis/extended treatment in cancer patients				Recurrent VTE	Major bleeding
Enoxaparin	ONCENOX31	122	Warfarin	0.47 (0.06–3.55)a	3.16 (0.36–27.55)a
Dalteparin	CLOT32	672	Coumarin	0.48 (0.30–0.77)	1.60 (0.77–3.36)a
Tinzaparin	Hull et al., 200633	200	UFH and warfarin	0.57 (0.20–1.65)a	1.00 (0.34–2.96)a
Inpatient treatment of DVT with/without pulmonary embolism				Recurrent VTE	Major bleeding
Enoxaparin	Merli et al., 200134	900	UFH	1.06 (0.47–2.36)a,c	0.81 (0.24–2.67)a,c
				0.69 (0.29–1.66)a,d	0.61 (0.17–2.20)a,d
Dalteparin	Lindmarker et al., 199434	204	UFH	1.74 (0.40–7.46)a	—
	Fiessinger et al., 199636	268	UFH	2.21 (4.40–12.29)a	—
Tinzaparin	THESEE37	612	UFH	0.84 (0.25–2.79)a	0.76 (0.26–2.02)a
Outpatient treatment of DVT with/without pulmonary embolism				Recurrent VTE	Major bleeding
Enoxaparin	Levine et al., 199638	500	UFH	0.77 (0.37–1.62)a	1.72 (0.41–7.28)a
Dalteparin	—	—	—	—	—
Tinzaparin	—	—	—	—	—
Arterial indications
Unstable angina and non–ST-elevation myocardial infarction				30-day death/MI	Major bleeding
Enoxaparin	ESSENCE39	3,171	UFH	0.76 (0.58–1.01)	0.90 (0.63–1.27)
	TIMI 11B40	3,917	UFH	0.88 (0.70–1.11)	1.52 (0.85–2.70)
	ACUTE II41	525	UFH	0.97 (0.51–1.83)	NA
	INTERACT42	746	UFH	0.54 (0.30–0.96)	0.47 (0.24–0.95)
	A to Z43	3,987	UFH	0.94 (0.73–1.20)	NA
	SYNERGY44	9,978	UFH	0.96 (0.86–1.07)	1.17 (0.99–1.39)
Dalteparin	FRISC45	1,506	Placebo	0.75 (0.54–1.03)	1.53 (0.43–5.45)a
	FRIC46	1,482	UFH	1.18 (0.84–1.66)	NA
Tinzaparin	—	—	—	—	—
Acute ST-elevation myocardial infarction				30-day death/MI	Major bleeding
Enoxaparin	ExTRACT-TIMI 2547	20,506	UFH	0.83 (0.77–0.90)	1.53 (1.23–1.89)
Dalteparin	FAMI48	1,128	No/placebo	—	—
Tinzaparin	—	—	—	—	—

^*The selected studies were based on main trials supporting FDA-approved indications and other main studies.

CI = confidence interval; DVT = deep vein thrombosis; NA = not applicable; OR = odds ratio; UFH = unfractionated heparin; VTE = venous thromboembolism.

Trials: ACUTE II = Antithrombotic Combination Using Tirofiban and Enoxaparin; CLOT = Comparison of LMWH versus Oral Anticoagulant Therapy for Long-Term Anticoagulation in Cancer Patients with VTE; ESSENCE = Efficacy and Safety of SQ Enoxaparin in Non–Q-Wave Coronary Events (ESSENCE); ExTRACT–TIMI = Enoxaparin and Thrombolysis Reperfusion for Acute MI Treatment–Thrombolysis in MI; FAMI = Fragmin in Acute MI; FRISC = Fragmin during Instability in Coronary Artery Disease; FRIC = Fragmin in Unstable Coronary Artery Disease; MEDENOX = Prophylaxis of VTE in Medical Patients with Enoxaparin; INTERACT = INTEnsive blood pressure Reduction in Acute Cerebral haemorrhage Trial; ONCENOX = ONCE daily ENOXaparin in the secondary prevention of venous thromboembolic events in patient with active cancer; PREVENT = Prospective Evaluation of Dalteparin Efficacy for Prevention of VTE in Immobilized Patients Trial; SYNERGY = Superior Yield of the New Strategy of Enoxaparin, Revascularization and Glycoprotein IIb/IIIa Inhibitors; THE-PRINCE = The Thromboembolism Prevention in Cardiac or Respiratory Disease with Enoxaparin; THESEE = Tinzaparin ou Heparin Standard: Evaluation dans l’Embolie Pulmonaire Study; TIMI = Thrombosis in MI.

^aCalculated.

^bFor the 40-mg once-daily dose.

^cFor the 1.5-mg/kg once-daily dose.

^dFor the 1-mg/kg twice-daily dose.

Enoxaparin

Enoxaparin has the broadest range of FDA-approved indications, based on evidence for the product’s efficacy and safety in multiple patient populations. It is indicated for the prevention of VTE in patients undergoing major orthopedic or abdominal surgery or in acutely ill medical patients; the treatment of deep vein thrombosis (DVT) with or without PE; and the treatment of patients with acute coronary syndromes (ACS), including ST-segment elevation MI (STEMI) and unstable angina or non-STEMI (UA/NSTEMI).

In 2007, the STEMI indication for enoxaparin was included for patients receiving thrombolytic therapy. This indication is based on the results of the Enoxaparin and Thrombolysis Reperfusion for Acute Myocardial Infarction Treatment–Thrombolysis in Myocardial Infarction (ExTRACT–TIMI 25) study.⁴⁷ The primary endpoint of death or recurrent MI at 30 days had occurred in 12% of patients receiving UFH and in 9.9% of those receiving enoxaparin; the initial dose was a 30-mg intravenous (IV) bolus, followed by 1 mg/kg subcutaneously (P < 0.001). Major bleeding occurred in 1.4% of the UFH-treated patients and in 2.1% of those receiving enoxaparin (P < 0.001).

Dalteparin

FDA-approved indications for dalteparin include the prevention of VTE in patients undergoing hip replacement or abdominal surgery and in acutely ill medical patients and the prevention of arterial thrombosis in patients with UA/NSTEMI ACS. More recently, dalteparin has also been approved for the long-term secondary prevention of VTE in patients with VTE and cancer. This indication is based largely on the randomized CLOT study (Comparison of Low-Molecular-Weight Heparin versus Oral Anticoagulant Therapy for Long-Term Anticoagulation in Cancer Patients with VTE).³² CLOT included 676 patients who received either dalteparin 200 IU/kg once daily for five to seven days, followed by a coumarin derivative for six months, or dalteparin alone for six months (200 IU/kg once daily for one month, followed by approximately 150 IU/kg once daily for five months). The probability of recurrent VTE at six months was 9% with dalteparin and 17% with the oral anticoagulant, resulting in a hazard ratio of 0.48 (95% CI, 0.30–0.77; P = 0.002). The difference in major bleeding complications between the two treatments was not significant (6% for dalteparin and 4% with oral anticoagulants; P = 0.27).

Tinzaparin

Of the available LMWHs, tinzaparin has the fewest indications. It is given at a dose of 175 IU/kg per day for the treatment of acute symptomatic DVT with or without PE.⁴⁹ In July 2008, the company revised the labeling to restrict the agent’s use in patients 90 years of age or older. However, preliminary data from the Innohep in Renal Insufficiency Study (IRIS) suggested that the increased risk of mortality was not limited to this population. Therefore, a label warning was then issued to mention an increased risk of death in elderly patients with renal insufficiency (70 years of age or older with a creatinine clearance [CrCl] of 30 mL/minute or below, or 75 years of age with a CrCl of 60 mL/minute or less).²³

Few studies have compared LMWHs for clinical equivalence (Table 4). Although some trials have reported comparable efficacy between LMWHs in the prophylaxis and treatment of VTE,^50–52 others have reported significant differences in efficacy or pharmacological properties.^53,54 Clinical guidelines from the ACCP on VTE prevention and treatment provide recommendations on the use of LMWHs as a class^1,11 and recommend that clinicians follow the manufacturer-suggested dosing guidelines for each of the antithrombotic agents.¹ Clinical guidelines from the ACC/AHA for the treatment of STEMI and UA/NSTEMI specifically recommend enoxaparin.^2,8

Table 4.

Studies Comparing Efficacy or Safety of Various Low-Molecular-Weight Heparins (LMWHs)

Author	Study Design	Patient Group	LMWHs	N	Outcome
Chiou-Tan et al., 200350	Prospective, randomized, open-label	Prevention of VTE acute spinal cord injury	Enoxaparin 30 mg twice daily vs. dalteparin 5,000 IU daily	50 vs. 45	DVT: 6% vs. 4% (P = 0.51); bleeding: 2% vs. 4% (P = 0.72)
Wells et al., 200551	Randomized controlled trial	Outpatient treatment of DVT and PE	Tinzaparin 75 IU/kg daily vs. dalteparin 200 IU/kg daily	254 vs. 251	Composite of recurrent VTE and major bleeding: 5.9% vs. 4.4% (P = 0.44)
Slavik, DETECT, 200752	Retrospective, cohort	Prevention of VTE in acute spinal cord injury and major orthopedic trauma	Enoxaparin 30 mg twice daily vs. dalteparin 5,000 IU daily	63 vs. 72	Symptomatic proximal DVT or PE; 1.6 % vs. 9.7% (P = 0.103 for non-inferiority)
Michalis, EVET, 200353	Randomized controlled trial	NSTEMI	Enoxaparin 1 mg/kg twice daily vs. tinzaparin 175 IU/kg daily	220 vs. 218	Recurrent angina, myocardial infarction, or death: 12.3% vs. 21.1% (P = 0.15)

DVT = deep vein thrombosis; IU = International Units; NSTEMI = non–ST-segment elevation myocardial infarction; PE = pulmonary embolism; VTE = venous thromboembolism.

Trials: DETECT = Dalteparin versus Enoxaparin for venous Thromboembolism prophylaxis in acutE spinal Cord injury and major orthopedic Trauma patients; EVET = Enoxaparin VErsus Tinzaparin.

DOSING

As with all forms of anticoagulant therapy, correct dosing of LMWHs is essential for achieving an optimal balance between therapeutic efficacy and the risk of bleeding. However, as a result of the marked pharmacological differences between LMWHs, therapeutically equivalent doses have not been established; thus, substituting one product for another is difficult.

For example, because of the high risk of VTE in hip or knee replacement surgery, high-dose prophylaxis with subcutaneous (SQ) enoxaparin 30 mg twice daily is recommended for these patients. For patients undergoing hip replacement, standard prophylaxis with SQ enoxaparin 40 mg once daily is an alternative. Dalteparin is indicated at a standard dose of 5,000 IU once daily to prevent VTE in patients receiving hip replacement, but it is not indicated in knee replacement because of a lack of published prospective randomized trials in this patient population.

Similarly, the doses of enoxaparin and dalteparin recommended for preventing arterial thrombosis in patients with ACS differ. In TIMI-11B⁴⁰ and ESSENCE (Efficacy and Safety of Subcutaneous Enoxaparin in Non–Q-Wave Coronary Events),⁵⁵ the dose of enoxaparin was 1 mg/kg twice daily; however, in FRISC (Fragmin during Instability in Coronary Artery Disease)⁴⁵ and FRIC (Fragmin in Unstable Coronary Artery Disease),⁴⁶ the dose of dalteparin was 120 IU/kg twice daily.

In contrast to UFH, anticoagulant monitoring is not usually required with LMWHs, although it has been suggested that this may be appropriate in certain populations, such as obese patients or those with renal impairment.⁴ In certain indications, such as the treatment of established VTE or treatment of patients with ACS, enoxaparin or dalteparin is given in doses adjusted according to body weight. Because there is no linear relationship between intravascular volume and total body weight, it is possible that this approach could lead to overdosing in obese patients; conversely, the use of fixed doses could lead to underdosing.⁴

There is no maximum dose for enoxaparin in patients with VTE or UA/NSTEMI; however, in patients with UA/NSTEMI who are treated with dalteparin, the maximum dose is capped at 10,000 IU twice daily. Similarly, when enoxaparin is used in patients with STEMI, the first two doses are capped at 100 mg. In clinical practice, enoxaparin is often administered below the recommended weight-based dose in patients with UA/NSTEMI who weigh more than 150 kg (342 pounds). Use of the recommended dose of 1 mg/kg seems to be associated with a higher bleeding risk in these patients.⁵⁶

Because LMWHs are excreted via the kidneys, caution is necessary in patients with impaired renal function, including elderly patients. However, LMWHs differ in their dependence on renal clearance.⁴ Currently, enoxaparin is the only LMWH for which the FDA has approved a dose adjustment for patients with severe renal impairment (CrCl below 30 mL/minute). It would not be appropriate to propose similar dose adjustments for dalteparin and tinzaparin in this situation.

COST

The cost of LMWH therapy varies according to the preparation, and this consideration has been a driving force in the debate about the potential for therapeutic interchange of LMWHs. In general, the acquisition costs of LMWHs are higher than those of UFH, but this greater expenditure is offset by decreases in costs because of improved clinical outcomes, shorter hospital stays, and less need for anticoagulant monitoring.^57,58

In a 2004 study that used a decision-tree approach to model the costs of thromboprophylaxis with enoxaparin 40 mg once daily or UFH 5,000 IU twice daily in acutely ill medical inpatients, enoxaparin was found to be less expensive than UFH because of the lower costs of adverse events and treatment of VTE (Figure 1).⁵⁷ In the Prophylaxis in Medical patients with Enoxaparin (MEDENOX) study, which investigated the efficacy of thromboprophylaxis with enoxaparin in acutely ill patients, the incremental cost effectiveness of enoxaparin, when compared with no prophylaxis, ranged from $1,249 to 3,088 per VTE event prevented. Moreover, if the incidence of treated VTE in the absence of prophylaxis was at least 3% to 4%, the costs of enoxaparin could be recouped through cost savings by the reduced incidence and treatment of VTE.⁵⁸

Figure 1.

Estimated costs of venous thromboembolism prophylaxis and treatment over 30 days in a hypothetical cohort of 10,000 acutely ill medical inpatients receiving enoxaparin 40 mg once daily; unfractionated heparin 5,000 IU twice daily; or no prophylaxis. b.i.d. = twice daily; DVT = deep vein thrombosis; PE = pulmonary embolism; q.d. = every day; UFH = unfractionated heparin. (Reprinted with permission from McGarry LJ, Thompson D, Weinstein MC, et al. Am J Manag Care 2004;10:632–642. Courtesy of Elsevier Publishing.⁵⁷)

THERAPEUTIC INTERCHANGE

As a result of the pressure to reduce drug and other health care costs, and with the expanding numbers of medications in each therapeutic or related class prescribed for the same indication, drug interchange has become an increasingly common practice in hospital pharmacies.

There are two forms of interchange:

• Therapeutic interchange is the substitution of a drug that is therapeutically equivalent to, but chemically distinct from, a drug with equivalent clinical (both therapeutic and adverse) effects.

• Biosimilar exchange describes the substitution of one drug formulation with a chemically and biologically similar version of the same product.

Therapeutic interchange programs have been established in most hospitals in the U.S. In a survey of 463 hospitals, therapeutic interchange policies had been established in 88% of teaching hospitals, in 89% of non-teaching hospitals, and in 100% of investor-owned hospitals.⁵⁹ Therapeutic interchange in hospitals usually involved the histamine H₂-receptor antagonists (in 91% of hospitals), proton pump inhibitors (in 71%), and antacids (in 57%). In comparison, therapeutic interchange of UFH and LMWHs was reported in 9% to 17% of hospitals.⁵⁹

In approximately 70% of the hospitals surveyed, physicians were involved in making decisions about the process of therapeutic interchange.⁵⁹ However, therapeutic interchange was then automatically implemented so that few hospitals required consent from a prescribing physician before substitution by the pharmacist could be implemented.

Although interchange programs often aim to reduce costs, only 20% of hospitals in the survey were able to quantify the exact annual cost savings that resulted from this strategy.⁵⁹ In addition, a large number of hospitals estimated cost savings to be in the lower ranges of $10,000 to 15,000,⁵⁹ which falls short of savings often purported for these programs. Cost analyses should take into account elements such as duration and cost of hospitalization, the need for nursing care, and the cost of managing adverse outcomes, in addition to drug acquisition costs. It is essential that therapeutic interchange programs be evidence-based and not justified solely on the basis of potential cost reductions.

Some of the reasons cited by hospitals that did not utilize therapeutic interchange programs included resistance by physicians (50% to 63%), expected benefits that did not justify the cost of initiating the therapeutic interchange (12% to 15%), concerns about violating state law (10% to 15%), and fear of civil liability (4% to 7%).⁵⁹

Experience with Therapeutic Interchange

Therapeutic interchange of LMWHs has been studied in various patient populations (Table 5).^60–63 The focus has been on interchange programs for VTE prophylaxis in total knee replacement surgery. Although these studies have generally reported that therapeutic interchange is cost-neutral⁶¹ or results in cost savings,^60,62 the clinical basis of these claims has been questioned because of (1) small patient populations with inadequate statistical power to detect significant differences in clinical outcomes, (2) inappropriate control groups, and (3) a lack of applicability to the range of indications for LMWHs.^64,65

Table 5.

Therapeutic Interchange Programs Using Dalteparin Instead of Enoxaparin for Prophylaxis Of Venous Thromboembolism

Author	Patient Population	No. of Patients (Dalteparin/Enoxaparin)	Symptomatic VTE, %		Bleeding, %		Cost Savings
			Dalteparin	Enoxaparin	Dalteparin	Enoxaparin
Bollinger et al., 200060	Trauma, orthopedic surgery, general surgery, medical	90/5,578 (literature control)	2.2 (1.1 PE)	Estimated, 3.4 (0.2 PE)	3.3	1.7	Annual reduction $90,000 in acquisition costs
Allen et al., 200361	Total knee replacement	101/88	4.0	3.4	0	3.4	$172 (95% CI, −17 to 395) per patient
Krotenberg et al., 200162	Total hip or knee replacement	161/300	0.3	1.9	2.3	3.7	−$129 (95% CI, −108 to −150) per patient
Komorney et al., 200263	Total knee replacement	40/–	2.5	—	2.5	—	Not reported

CI = confidence interval; PE = pulmonary embolism; VTE = venous thromboembolism.

Two surveys have been conducted to investigate hospital practices concerning LMWH therapeutic interchange polices and guidelines governing LMWH use.^12,17 The first survey, by Schumock and colleagues in 2002,¹⁷ reported that 29% of hospitals employed guidelines concerning LMWH use, whereas only 10% had policies on the therapeutic interchange of LMWHs.¹⁷ The largest barrier to therapeutic interchange of LMWHs, identified by 87% of respondents, included the lack of therapeutic equivalence and differences in labeled indications between products.¹⁷

In a second survey in 2007, the presence of LMWH guidelines in hospitals had risen to 55%, and 25% of hospitals at that time had therapeutic interchange policies for LMWHs.¹²

Biosimilar Low-Molecular-Weight Heparins

Biosimilar products (also called follow-on biologics), which are often regarded as therapeutically interchangeable with the original products on the basis of identical chemical and pharmacokinetic properties, are an additional consideration. Ideally, biosimilar agents should demonstrate reproducibility of pharmacological activities beyond surrogate markers, despite their highly complex molecular structure or biological origin. Because they have been studied in large populations that include special patient groups, they should, preferably, also show reliability in their clinical use in various clinical settings. Such formulations then offer the potential advantage of lower acquisition costs compared with the branded product. In practice, however, biosimilar products are not supported by the same weight of clinical evidence as the branded product, and they are not required by regulatory authorities to provide such evidence.

Decisions on therapeutic interchange with biosimilar products should not be made on economic grounds alone; the primary consideration should be the most appropriate therapy for the patient. In addition, even minor differences in a drug’s production process can have profound consequences in clinical settings. For example, modifications to the production of epoetin alfa (Eprex, Janssen-Ortho), a synthetic analogue of erythropoietin, dramatically changed its immunogenic profile.⁶⁶ Eprex stimulates the production of red blood cells; accordingly, one of its indications is for use in patients with anemia. However, following treatment with a new formulation of Eprex, some patients developed antibodies that neutralized both the drug itself as well as the body’s own erythropoietin. This resulted in pure red blood cell aplasia, a potentially life-threatening condition,^66,67 and it prompted urgent safety restrictions from the manufacturers and national regulatory bodies.

Biosimilar LMWHs have been developed in India and South America, and regulatory approval of a biosimilar formulation of enoxaparin has been requested in the U.S.⁹ However, LMWHs are produced by a complicated process involving chemical depolymerization (as with enoxaparin) or enzymatic depolymerization (as with tinzaparin). If this process is not replicated exactly in the production of a biosimilar formulation, equivalence of the resulting product cannot be ensured.⁶⁸

The potential significance of this problem is highlighted by a study of the pharmacological properties of three biosimilar formulations of enoxaparin from India and Brazil and the branded formulation available in the U.S.⁹ Although the four products were similar in terms of their molecular profiles and anticoagulant potencies, as measured by activated partial thromboplastin time (aPTT), there were marked differences between the biosimilar formulations in thrombin time (TT) and in Heptest (American Diagnostica), a heparin clotting assay kit (Figure 2).⁹ Therefore, caution is necessary when one is considering therapeutic interchange with biosimilar LMWHs. Indeed, it has been proposed that new regulatory guidelines be developed before such substitutions can be recommended.⁶⁸ In November 2007, the FDA did not approve M-Enoxaparin (a generic product) because the application did not adequately address the potential for immunogenicity of the drug.⁶⁹

Figure 2.

Thrombin time (A) and clotting time (B) measured by the Heptest assay for three biosimilar formulations of enoxaparin (two batches from India and one from Brazil), and the commercial formulation available in the U.S. (Lovenox). (From Maddineni J, Walenga JM, Jeske WP, et al. Clin Appl Thromb Hemost 2006;12:267–276. Reprinted by permission of Sage Publications.⁹)

Fondaparinux

Fondaparinux (Arixtra) is a synthetic preparation of the high-affinity pentasaccharide required for binding of heparin to antithrombin. Although sometimes described as an ultra-low-weight heparin, this agent is chemically distinct from other LMWHs and has a different spectrum of indications. Fonda-parinux is indicated for the prevention of VTE in orthopedic and abdominal surgery patients and for the treatment of VTE. It is not currently approved for patients with ACS, although it is included in the ACC/AHA guideline recommendations for ACS.^2,8

In the Fifth Organization to Assess Strategies in Acute Is-chemic Syndromes (OASIS-5) trial in patients with NSTEMI, fondaparinux 2.5 mg/kg daily was associated with a similar rate of death, MI, or refractory ischemia as enoxaparin 1 mg/kg twice daily (5.8% vs. 5.7%; P = 0.007 for non-inferiority) but with a lower incidence of major bleeding (2.2% vs. 4.1%; P < 0.001). Fondaparinux cannot be used as the sole anticoagulant in patients with ACS who are undergoing percutaneous coronary intervention, because an anticoagulant possessing anti-IIa activity (such as UFH) must be added.^2,8 This requirement clinically differentiates fondaparinux from enoxaparin, which can be used as the sole anticoagulant for the medical and invasive management of ACS. Furthermore, in clinical trials, fondaparinux and LMWHs had different efficacy and safety profiles.⁷⁰

In a meta-analysis of trials in major orthopedic surgery, fondaparinux at a dose of 2.5 mg once daily was associated with a lower incidence of symptomatic or asymptomatic VTE, compared with enoxaparin (6.8% vs. 13.7%, respectively; P < 0.001), but it was also related to a higher incidence of major bleeding (2.7% vs. 1.7%; P = 0.008). For these reasons, fondaparinux and LMWHs cannot be considered interchangeable.

CONCLUSION

LMWHs are chemically and pharmacologically distinct agents, and each has its own unique efficacy and safety profile. Enoxaparin has the most extensive clinical evidence of efficacy and safety in various situations and hence has the broadest range of indications. It is the LMWH licensed for the largest number of patients in the management of thromboembolism in both arterial and venous indications. As such, it is the most commonly used LMWH in hospital pharmacies.

The diversity of pharmacological and clinical properties, as well as the poor-quality, inconsistent evidence for equivalence between agents, makes the LMWHs unsuitable candidates for therapeutic interchange programs. Similarly, although attention is being devoted to the potential for therapeutic interchange with biosimilar LMWHs, this approach cannot be recommended because no currently available biosimilar product has been shown to be identical to the original product. Moreover, the potential risks to patients, such as immunogenicity, need to be assessed.