Since its discovery by a medical student in 1916, heparin has enjoyed a great deal of success as an anticoagulant (1) allowing many fields of medicine to advance by leaps and bounds. The ability to “heparinize” someone for cardiopulmonary bypass has permitted heart surgeons to push the edge in the repair of congenital cardiac anomalies and cardiac revascularization procedures. Heparin has made possible peripheral bypass procedures, and many patients have enjoyed an improvement as well as an extension of life. Despite these major contributions, most medical personnel associate heparin with the prevention and treatment of deep venous thrombosis (DVT). Management of this potentially fatal disorder revolves around achieving an anticoagulated state, for which heparin has played a leading role for many years. Only within the last decade have options other than unfractionated heparin become available. An enormous amount of research has been dedicated to the understanding of heparin and more recently to the development of “the new heparins.”
Mechanism of Action
Heparin is a naturally occurring glycosaminoglycan with a negative ionic charge. Its molecular weight ranges from 5000–30,000 Daltons with an average weight of 15,000 Daltons (2). In its natural unfractionated state, heparin exists as a heterogeneous mixture of oligosaccharides composed of alternating chains of D-glucosamine and uronic acid (3). The predominant anticoagulant property of heparin is based on its interaction with antithrombin (AT) (formerly known as antithrombin III). This interaction is dependent on a unique pentasaccharide sequence located within the heparin molecule (4). A strategically positioned sulfate residue within the pentasaccharide sequence has a high affinity for a specific lysine site located on the AT molecule. Binding of heparin promotes a conformational change in AT which exposes an arginine situated in the AT reactive center. The exposed arginine binds covalently to specific serine centers found on certain clotting factors (2, 3). This conformational change accelerates the antithrombin inhibition of procoagulant serine proteases a thousand-fold over its baseline rate. Serine proteases prone to inhibition are the clotting factors XIIa, XIa, Xa, IXa and IIa (2, 3). Factor IIa, or thrombin, is the most susceptible factor (2).
Unfractionated Heparin and Low Molecular Weight Heparin
Unfractionated Heparin
When speaking of heparin, one is typically referring to unfractionated heparin (UH) as opposed to fractionated heparin or low molecular weight heparin (LMWH). The latter is an enzymatic degradation product of UH (5). The anticoagulant activity of UH varies as a consequence of its heterogeneous structure, dual mechanism of clearance, and propensity to bind plasma proteins. The heterogeneous structure of unfractionated heparin is demonstrated by its variations in oligosaccharide chain length and number of heparin molecules that contain the required pentasaccharide sequence. Only molecules of at least 18 oligosaccharides in length have the ability to inhibit thrombin (2). This length is required to bind antithrombin and thrombin simultaneously (Figure 1). In contrast, inactivation of clotting factor Xa does not require as large a heparin molecule. Chain lengths shorter than 18 oligosaccharides, which contain the high-affinity pentasaccharide sequence, retain the ability to bind AT and inactivate Xa but lack the ability to inactivate thrombin. In commercially prepared UH the ratio of anti-IIa activity to anti-Xa activity is approximately 1:1 (5). This issue of chain length becomes relevant only if the heparin molecule possesses the required high affinity sequence, and only one-third of heparin administered to patients contains this sequence (6).
Antithrombin's inactivation of factor IIa is accelerated in the presence of unfractionated heparin molecules. Large oligosaccharide chains (>18) are required to bind AT and factor IIa simultaneously. The high-affinity pentasaccharide sequence is represented by the hexagonal figures
The pharmacokinetic profile of UH is another factor that contributes to the differing levels of anticoagulation achieved among different individuals. The primary method of clearance is through binding of receptors found on the membranes of endothelial cells and macrophages (2). Binding of UH to cell surface receptors prompts internalization followed by depolymerization. Saturation of this pathway is rapid due to the limited number of surface receptors. Once receptors are saturated, the remaining UH molecules are eliminated through a slower method of renal clearance. Therefore, as the dose of UH is increased, the primary method of clearance becomes saturated and elimination becomes dependent on the slower renal pathway. A longer half-life thus occurs with increasing doses. This translates into a nonlinear relationship between dose and anticoagulant activity (2). The half-life of UH is approximately 60 minutes for therapeutic doses (100 U/kg).
The last and probably most significant factor in determining the unpredictable dose-response is the nonspecific binding of UH to plasma proteins (heparin-binding proteins). Subsequent to protein binding, the amount of UH available for activation of AT is reduced. Acute phase reactants fall under this nonspecific umbrella of heparin-binding proteins. Critically ill patients typically exhibit elevations in acute phase reactants thus increasing the dose of heparin required to achieve therapeutic anticoagulation. This laboratory phenomenon is commonly referred to as heparin resistance (2).
Secondary to the unpredictable dose-response of unfractionated heparin, clinical monitoring of the coagulation cascade is required. The effect of heparin on the clotting pathways is analyzed by measuring an activated partial thromboplastin time (aPTT). Variations in laboratory reagents prohibit a standard range for therapeutic anticoagulation; however, with rigid quality control, local laboratory reference ranges can be established (7). This is usually 1.5 to 2.5 times the normal aPTT. In addition to the inconsistencies in the level of anticoagulation achieved, another downside of using UH for therapeutic treatment is its requirement for intravenous administration. Early failures in the development of an oral formulation lead to the emergence of a subcutaneous method of administration. Initially this was accomplished using UH but more recently subcutaneous fractionated heparin (LMWH) has become popular.
Low Molecular Weight Heparin
As the name implies, LMWH has a molecular size smaller than UH. Its average size is roughly 4500 Daltons (2, 5). The smaller size bestows some unique characteristics. One of these unique traits relates to its mechanism of action. As alluded to earlier, LMWH exerts its anticoagulant effect primarily through the acceleration of antithrombin-dependent inactivation of factor Xa. Like UH, the ability of LMWH to interact with antithrombin is dependent on the presence of the high-affinity pentasaccharide sequence (Figure 2). However, in contrast to UH, less than half of LMWH contains molecules with oligosaccharide chains of sufficient length (≥18 oligosaccharides) to inhibit thrombin (5). This is obvious when comparing its anti-Xa with anti-IIa activity. Depending on the molecular size distribution, this ratio will typically vary from 4:1 to 2:1 (5).
Antithrombin's inactivation of factor Xa is accelerated in the presence of both unfractionated and fractionated heparin molecules. This interaction is not dependent on the presence of large (oligsaccharide chain lengths >18) heparin molecules. The high-affinity pentasaccharide sequence is represented by the hexagonal figures
A second distinguishing property accredited to the smaller size is a prolonged half-life. The receptor-dependent clearance mechanism have a higher affinity for larger molecules, leaving clearance of smaller molecules to the slower renal pathway (2, 5). As the average molecular size indicates, LMWH primarily consists of smaller molecules necessitating clearance by the kidney. The prolonged half-life permits once-to-twice a day dosing for therapeutic effect. As would be expected, dose adjustments are needed for renal failure patients.
The last advantageous trait attributable to smaller size is a more predictable anticoagulation profile. The smaller molecular size of fractionated heparin leads to less nonspecific plasma protein binding and a more consistent bioavailability pattern. A more predictable dose-response profile negates the need to monitor its effect on the coagulation cascade; hence, less blood drawing from patients, which inevitability leads to improved patient satisfaction. If one must measure levels of anticoagulation when using LMWH, antifactor Xa assays are available in most hospitals as send-out tests (turnaround time in our facility is 2–3 business days).
UH vs. LMWH
The gold standard for treatment of venous thromboembolic disease (VTE) has been intravenous UH followed by oral warfarin for up to 3 to 6 months (8). The natural question, “Can subcutaneous LMWH substitute for intravenous UH?” has been answered by multiple clinical trials (9–11). Prandoni and colleagues evaluated the frequency of symptomatic recurrent venous thromboembolism in patients who were initially treated with either twice a day weight-adjusted subcutaneous LMWH followed with oral warfarin or intravenous UH followed with oral warfarin (9). Patients had a baseline venogram and lung perfusion scan during their initial hospitalization. Individuals with clinically suspected extension or recurrent venous thrombosis had a repeat venogram performed for comparison. Results showed a 14% rate of recurrent venous thromboembolism in patients treated initially with intravenous UH compared with a 7% recurrence in patients treated with subcutaneous LMWH. Hull and colleagues demonstrated a similar finding (6.9% vs. 2.8%) when they compared intravenous UH with a higher once a day dose of subcutaneous LMWH (10). Neither group of investigators demonstrated a significant difference in protection from recurrent thromboembolism or rate of major bleeding between the two regimens. A more recent meta-analysis of randomized controlled trials agreed with the finding that subcutaneous LMWH was as safe and effective as intravenous UH in treating venous thromboembolic disease (12).
Outpatient Management
Once LMWH had been established as an equally safe and effective initial therapy for deep venous thrombosis, investigators sought the idea of outpatient treatment. Two early studies evaluated the effectiveness of twice a day dosing of subcutaneous LMWH as initial outpatient therapy in selected patients (13, 14). Of patients randomized to receive LMWH, 50% in one study and 36% in the other were never admitted to the hospital. Endpoints were symptomatic recurrent venous thromboembolism or major bleeding. In both studies, equally effective protection was provided against recurrent thromboembolism (5.3% vs. 6.7% and 6.9% vs. 8.6%) as well as no difference in major bleeding when compared with intravenous UH. An important finding demonstrated was a reduction in hospital admissions. Levine et al found a mean difference of 5.4 days in the length of stay between the two cohorts (14). A cost-effective analysis using Medicare reimbursement rates from 1995 was applied to these two studies and found that if 30% of patients were treated entirely as outpatients and 25% were discharged early (within 3 days) cost savings totaled $790 per patient treated (15). Savings in favor of LMWH become evident when as little as 8% of patients are treated entirely as outpatients or 13% qualify for early discharge. These encouraging statistics have enticed some clinics to set up outpatient protocols for treatment of VTE after careful selection of patients (16).
Warfarin
Currently, the only clinically available oral method of anticoagulation is warfarin, which has been the mainstay for outpatient anticoagulation for years (8). Reaching therapeutic levels of anticoagulation with warfarin frequently requires a minimum 4-day overlap with some form of heparin. This overlap is responsible for the prolonged hospital stay traditionally seen when using intravenous UH. Similar to UH, predicting levels of anticoagulation with warfarin is fraught with inconsistency. Levels of anticoagulation vary so widely that frequent monitoring is mandatory.
Other Orally Administered Anticoagulant Agents
In an effort to improve orally administered anticoagulants, ongoing research for the development of an oral heparin agent has continued over the last 60 years. Problems encountered with oral heparin are believed to center around its large size and negative ionic charge (17–19). Recently scientists have developed carrier agents to improve the gastrointestinal absorption of heparin without interfering with its anticoagulant properties. Two such carrier molecules, sodium N-[8(2-hydroxybenzoyl)amino] caprylate (SNAC) and sodium N-[10-(2-hydroxybenzoyl)amino] decanoate (SNAD) (Emisphere Technologies, Tarrytown, NY), have been designed to enhance the absorption of unfractionated and fractionated heparin, respectively. These agents facilitate absorption by neutralizing the negative ionic charge, which makes heparin more lipophilic (20). Once in the bloodstream the carrier agents dissociate from heparin permitting it to express its anticoagulant activity.
Heparin:SNAC
The initial study evaluated the efficacy of an oral heparin:SNAC complex in preventing DVT in the rat model (17). Results demonstrated a significant reduction in the amount and weight of thrombus formed when the oral heparin:SNAC compound was compared with controls. There was not a significant difference between the oral complex and intravenous heparin. A second phase of the original experiment demonstrated that reliable monitoring was possible as with the intravenous form of heparin.
Once the prevention of DVT with the use of an oral heparin:SNAC complex was established, the next logical step was to test its effectiveness in treating DVT (18). The same rat model was used; however, treatment was initiated and continued for 7 days after thrombus formation. Results were significant and convincing that the use of an oral heparin:SNAC complex was as effective in the treatment of acute DVT as subcutaneous LMWH. This study also found reductions in thrombus weight and elevations in clotting times with the administration of the oral heparin:SNAC compound.
Heparin:SNAD
Similar experiments evaluating the prevention of DVT in the rat model were conducted with an oral LMWH:SNAD combination (19). Likewise, statistically significant reductions in the amount and weight of thrombus formed were demonstrated with the use of the oral formulation. As expected, antifactor Xa levels were significantly elevated.
The next experiment tested the effectiveness of an oral LMWH:SNAD combination in treating an established DVT in a larger animal model (21). In this model, porcine inferior vena cava thrombus was induced then treated over a 7-day period with the oral combination. At the end of 7 days, there was significantly less residual thrombus in the LMWH:SNAD group versus the control. Again, levels of antifactor Xa were significantly elevated over baseline. The importance of this experiment is two-fold. First, it reiterates the earlier findings of efficient intestinal absorption, and second, it was effective in treating an established DVT in a larger mammalian model. The larger porcine model more closely resembles the human venous system in its size and flow characteristics.
Human Trials
Initial investigations in human subjects were reported in 1998 (22). In a randomized, double-blind format, researchers evaluated the effect of increasing doses of oral heparin in combination with fixed doses of SNAC. By monitoring changes in coagulation parameters, researchers demonstrated effective absorption of the heparin:SNAC compound, as well as significant elevations in coagulation parameters with increasing doses of heparin. An additional arm of the study evaluated the tolerance of the compound when mixed with a sugar-based syrup. The preparation was well tolerated by all subjects, and no significant adverse events were reported. Nausea and emesis were noted occasionally; however, no relationship could be established. Currently, a Phase II trial specifically evaluating the effectiveness of the oral heparin:SNAC combination in preventing DVT in patients undergoing elective hip arthroplasty has been completed (23). The results of this trial have not yet been made public. Phase III trials are underway and encouraging results are anticipated.
New Heparins
Focusing on the high-affinity pentasaccharide sequence required to bind AT, researchers have developed a synthetic pentasaccharide molecule with anticoagulant properties. Theoretically the new heparin molecule, Org31540/SR90107A (PS) (Figure 3), should be as equally effective in the prevention and treatment of VTE as are the various forms of heparin (unfractionated and fractionated) but lack the disadvantages associated with the heterogeneousity and large molecular size.
Org31540/SR90107A [a synthetic pentasaccharide (PS)]
On a molecular scale, the new heparin contains a pair of neighboring sulfate residues that bind strongly and exclusively to antithrombin (24). Similar to UH and LMWH, binding of the synthetic pentasaccharide to AT induces a conformational change in the AT molecule. This enhances the presentation of AT's reactive inhibitory center allowing the accelerated inactivation of factor Xa (Figure 4). This in turn inhibits the coagulation cascade at the Xa level. The uniform size of the new pentasaccharide allows it to be very specific with respect to which clotting factor is inactivated. Unlike the previous forms of heparin discussed, PS exhibits no anti-IIa activity.
Activation of antithrombin by the new synthetic pentasaccharide molecule is represented by the hexagonal figures. Notice the absence of oligosaccharides when compared with unfractionated and fractioned heparin
Dr. Conners is a Fellow in Ochsner's Vascular Surgery Section
Dr. Money is the Head of Ochsner's Vascular Surgery Section
The high specificity for antithrombin and the small molecular size prevent cross-reactivity with other plasma proteins. This, in conjunction with linear pharmacokinetics, is responsible for its anticoagulation profile being much more predictable (25). Monitoring clinical activity, as with LMWH, is not needed since anticoagulation should not differ among individuals. If activity must be assessed, the lack of anti-IIa activity prevents monitoring the aPTT so an anti-factor Xa assay would be the test of choice.
A recently published randomized double-blind phase II study evaluated the dose-response in patients undergoing elective hip arthroplasty (24). The trial design consisted of five study groups receiving different doses of subcutaneous PS (0.75mg/1.5mg/3.0mg/6.0mg/8.0mg) once a day versus the control (SQ enoxaparin 30mg bid). Two of the study arms (6.0mg and 8.0mg) were closed before the trial concluded secondary to an excessive risk of bleeding with those doses. Investigators were successful in clearly demonstrating a statistically significant dose-effect as well as identifying a 29% (1.5mg) and 82% (3.0mg) risk reduction in the formation of venous thromboembolism when compared with controls. From a bleeding standpoint, the 0.75mg and 1.5mg doses were associated with significantly less major bleeding; however, the 3.0mg dose experienced no difference in major bleeding when compared with enoxaparin. In summary, an increase in both protection from venous thromboembolism and rate of major bleeding was observed with increasing doses of PS. Success with this trial allowed identification of a dose (1.5 to 3.0mg) “that has the potential to improve significantly the risk-benefit ratio for the prevention of venous thromboembolism”(24).
Initial Phase III trials that evaluated prophylaxis from VTE in patients undergoing elective hip replacement, knee replacement, or hip fracture repair have also been completed (26–29). It appears that the new synthetic pentasaccharide will offer better protection from VTE without an increased risk of bleeding. Eriksson and coworkers evaluated prophylaxis in patients undergoing hip-fracture repair (26). Patients were randomized to receive once-daily subcutaneous injections of either 2.5mg of synthetic pentasaccharide (fondaparinux) initiated 12 to 24 hours postoperatively or 40mg of enoxaparin initiated 12 hours preoperatively. In the fondaparinux group 11% of patients received their initial dose preoperatively (secondary to delays in surgery) and 74% of patients in the enoxaparin group received their initial dose postoperatively (a consequence of very early surgery after admission or planned regional anesthesia). Treatment was scheduled to continue through postoperative days 5 to 9. At this point participants underwent bilateral ascending venograms. Results revealed a 10.8% absolute reduction or a 56.4% relative reduction in venous thromboembolism in favor of fondaparinux (p<0.001). When analyzed further, both proximal and distal deep venous thrombosis was significantly reduced with fondaparinux compared with enoxaparin. Follow-up of patients at 7 weeks demonstrated no difference in pulmonary embolism rates. Investigators point out that early detection (by venogram on days 5 to 9) and treatment of deep venous thrombosis may have prevented a difference in pulmonary embolism rates. At no point was there a significant difference in symptomatic venous thromboembolism or major bleeding. A second European study evaluated the prevention of venous thromboembolism after elective hip replacement (27). Similar results regarding relative risk reduction (56%) were demonstrated in this study as well.
Bauer and colleagues studied patients receiving either once-daily subcutaneous injections of 2.5mg of fondaparinux or 30mg of enoxaparin given subcutaneously twice daily after major knee surgery (28). Likewise treatment was continued for 5 to 9 days then bilateral ascending venograms were obtained. Patients given fondaparinux achieved an absolute reduction in venous thromboembolism of 15.3% (55.2% relative risk reduction) compared with those receiving enoxaparin (p<0.001). Unlike patients undergoing hip-fracture repair, no significant difference was seen in proximal deep venous thrombosis. Major bleeding was a problem in 11 patients receiving fondaparinux and only 1 patient in the enoxaparin group (p=0.006). Again no difference was seen in the rate of symptomatic venous thromboembolism or pulmonary embolism. Turpie et al utilized the same dosing pattern but evaluated patients undergoing elective hip replacement (29). A 25% relative risk reduction was seen in this patient cohort when compared with controls.
All four of these trials are part of a global phase III program evaluating the clinical benefit of PS and enoxaparin in prophylaxis against VTE in major orthopedic procedures. Preliminary results indicate that the new synthetic pentasaccharide may offer an overall relative risk reduction of 50% when compared with enoxaparin in this patient cohort (26–29).
Conclusion
Anticoagulation with heparin has assisted in the advancement of medicine over the last half a century. Until recently, however, anticoagulation with heparin was limited to an inpatient setting. LMWH has permitted outpatient use and eliminated the need for coagulation profile monitoring. Despite these improvements, frequent home health visits and painful injections both discourage use and raise costs. Oral warfarin therapy circumvents these issues; however, predicting levels of anticoagulation are near impossible, and frequent monitoring is required. The development of an effective carrier agent for fractionated heparin has the potential to combine the anticoagulant advantages of LMWH and the preferred oral route of warfarin. Promising results in early animal models seem to indicate that an oral formulation of heparin is effective in preventing and treating deep venous thrombosis. If similar results are obtained in human clinical studies, an oral heparin formulation will facilitate a change in the clinical practice of venous thromboembolism management.
The development of a new synthetic heparin that is more specific for antithrombin, appears to improve protection from venous thromboembolism without an increase in major bleeding. If further studies show that the synthetic heparin is as equally effective as LMWH in treating venous thromboembolic disease, its once a day dosing for therapeutic effect stands to improve patient compliance and satisfaction. The next logical step would be to develop an oral formulation of the synthetic pentasaccharide. Outpatient management of deep venous thrombosis with an oral once or twice a day regimen that does not require monitoring would revolutionize the art of managing venous thromboembolic disease.
- Ochsner Clinic and Alton Ochsner Medical Foundation
References
In this issue
Jump to section
Cited By...
- No citing articles found.