Unfractionated and low molecular weight heparin

This chapter is relevant to Section Q2(i) of the 2023 CICM Primary Syllabus, which expects the exam candidates to "understanding of the pharmacology of anti-coagulants, anti-platelet drugs, thrombolytic drugs and anti-fibrinolytic drugs". Heparin is a staple of ICU anticoagulation, and one would do well to become very familiar with its properties, for various pragmatic reasons but also because it appears in several historical exam papers:

• Question 20 from the first paper of 2024 (heparin vs. enoxaparin)
• Question 20  from the second paper of 2018 (heparin vs. enoxaparin)
• Question 5 from the second paper of 2017 (heparin vs. enoxaparin)
• Question 4 from the second paper of 2011 (heparin vs. hirudin)
• Question 8(p.2) from the second paper of 2009 (heparin vs. enoxaparin)
• Question 2 from the first paper of 2008 (heparin vs. enoxaparin)

As you can see, it's all heparin vs. enoxaparin, except for that one question involving hirudin (hirudin!) which barely anybody passed. As such, it seemed logical to present the summary as a big table comparing these two drugs to one another. 

Name	Heparin	Enoxaparin
Class	Parenteral anticoagulant	Parenteral anticoagulant
Chemistry	Glycosaminoglycan	Glycosaminoglycan
Routes of administration	IV ands subcut	IV ands subcut
Absorption	Minimal oral bioavailability (~ 1%)	Minimal oral bioavailability (~ 1%)
Solubility	pKa -2.0 to -4.0, excellent solubility in water	pKa -2.8, excellent solubility in water
Distribution	Highly protein-bound, mainy to lipoproteins (LDL)	VOD=0.05L/kg, basically confined to the bloodstream. Somewhat protein-bound, but less than unfractionated heparin (only the chains which are more than 6000 Da are protein-bound)
Target receptor	Antithrombin III	Antithrombin III
Metabolism	Sequestered into reticuloendothelial cells and degraded gradually into inactive and renally cleared metabolites.	Mainly metabolized by the liver via desulfation and depolymerization to lower molecular weight fragments, which end up being either less potent or totally inactive
Elimination	Biphasic (saturable) metabolism: with low doses, a rapid saturable clearance (by reticuloendothelial tissues), which becomes slower with high doses when this system is saturated. Monitored by APTT, which incorporates an assessment of thrombin activity. Reversal is using protamine, 1mg binds 100 units of heparin	About 40% of active and inactive fragments combined are excreted renally, which is why low molecular weight heparin is not especially well suited to renal failure patients. Monitoring is by measurement of anti-Xa activity. Reversal
Time course of action	Half-life of 25 units per Kg = 30 minutes Half-life of 100 units per Kg = 60 minutes Half-life of 400 units per Kg = 150 minutes	Half life is about 4-7 hours.
Mechanism of action	By binding to antithrombin III and causing the active site to undergo a conformational change, heparin increases its availability to its normal ligands, including factor Xa and thrombin. The result is an increase in the activity of antithrombin, which manifests in the form of the anticoagulant effect	By binding to antithrombin III and causing the active site to undergo a conformational change, low molecular weight heparin increases its affinity for factor Xa (but not thrombin). The result is an increase in the activity of antithrombin on Factor Xa, which manifests in the form of the anticoagulant effect.
Clinical effects	Anticoagulation, bleeding, the possibility of HITS. Also osteopenia, mineralocorticoid deficiency alopecia and LFT derangement	Anticoagulation is the only clinically apparent effect; no significant side effects apart from the possibility of HITS (which is much smaller than with UFH)
Single best reference for further information	TGA PI document	TGA PI document

Merli & Groce (2010) seem like a perfect reference to recommend because their article is exactly a comparison of unfractionated and low molecular weight heparin.

Chemical properties and chemical relatives

Heparin is a heterogeneous mixture of mucopolysaccharides, termed glycosaminoglycans. It is essentially a polymerised disaccharide, a starch. Each repeated disaccharide is variably sulfated. Here's an image ripped off directly from Wikipedia:

Because the disaccharide polymers are of varying lengths, heparin has an average molecular weight of 3 to 30 kDa. Unless it is fractionated, in which case you can control to only have the low molecular weight version. Its polymer length is really very very random, mainly because heparin is an animal product (the first heparin was in fact derived from canine liver cells in 1916, by a second-year medical student) and animals are largely disinterested in quality control of industrial chemistry. The heparin in your hospital is derived from bovine lung or porcine gut, and we are occasionally offered a helpful reminder of this each time there is a swine flu outbreak among the pigs of China. 

The chemical relatives of heparin would have to include all the fractionated heparins (eg. enoxaparin), as well as a some of the alternatives, into which one must go if one is (for example) having to systemically anticoagulate a patient who is heparin-intolerant. These will be mentioned here in some minimal detal. 

Naturally occurring heparin-like molecules

Naturally occurring mucopolysaccharides (MPS) of mammals include the chondroitin sulphates, hyaluronic acid, heparitin sulphate, keratosulphate, chondroitin, and heparin. As mentioned above, heparin is already getting scraped out of pig intestines and purified for human use. The other heparin-like substances are also available on the market as a blended mixture known as danaparoid. Danaparoid has roughly the same molecular weight as dalteparin (about 6000 Da) and is therefore considered as one of the low molecular weight heparins.

Low molecular weight heparins

Enoxaparin, dalterparin and fondaparinux are grouped as "low molecular weight heparins" because their oligosaccharide chains are shorter than heparin. They are produced from heparin by "fractionation", which actually means a different thing depending on which drug you are looking at, and mainly means "reduction in size" rather than implying some sort of fractionation column. Still, this is why the normal version of heparin is referred to as "unfractionated", as it has not undergone this process. Dalteparin is about 6,000 Da and enoxaparin is about 4500 Da on average (Merli & Groce, 2010) Fondaparinux is an even smaller molecule (1500 Da), completely synthetic (i.e. not reliant on the fickle Chinese pork industry) and does not produce a risk of HITS as it has no affinity for PF4. 

Polypeptide direct thrombin inhbitors

Hirudin was the one required by the college in Question 4 from the second paper of 2011. That's the main ingredient of leech saliva, a 65 amino acid polypeptide secreted by Hirudo medicinalis, and leeches remain the only source of this substance (though it does not appear to be available on the Australian market). Because the process of pureeing hundreds of leeches was laborious and distasteful, a recombinant option (lepirudin) was ultimately developed in 1997, which differs from the natural product by its lack of one out of three sulfate bridges. As the result, its affinity for thrombin is lower (Greinacher et a, 1991). Bivalirudin, a much shorter peptide analog (20 amino acids), is also available.

What the hell is a "unit" of heparin?

One unit of heparin is the quantity required to keep 1ml of cats blood liquid for 24 hrs at 0 degrees Celsius. This unusually animal-unfriendly definition comes from one WH Howell, who left some cats blood overnight in the refrigerator. It didn't clot - it half-clotted- but it remained liquid all the same. These days the International Heparin Standard uses sheep plasma, presumably because sheep are easier to capture and exsanguinate.

Administration and absorption

Heparin is typically given intravenously, or subcutaneously (wherefrom it gradually dissociates). The subcutaneous route of administration takes 1-2 hrs to reach peak effect. Thereafter, its volume of distribution is 40-70ml/kg, essentially confined to the intravascular volume. Being a huge molecule, unfractionated heparin  does not penetrate the placenta, which makes it especially useful in pregnancy.

Oral heparin has very poor bioavailability. It is too large a molecule, and its charge is too negative to be absorbed easily. Brave men have tried cooking with heparin to make it more orally bioavailable. Additionally, 45 volunteers drank 20,000 units of heparin and then allowed their APTT to be tested - turns out it increases by 2.3 seconds on average, which is not much, but which demonstrates that there is some absorption. The authors did not mention anything about the taste. Similarly, low molecular weight heparins have minimal oral bioavailability.

Name	Bioavailability	pKa and water solubility
Heparin	Minimal oral bioavailability (~ 1%)	pKa -2.0 to -4.0, excellent solubility in water
Enoxaparin	Minimal oral bioavailability (~ 1%)	pKa -2.8, excellent solubility in water
Dalteparin	Minimal oral bioavailability (~ 1%)	pKa -2.6, excellent solubility in water
Fondaparinux	Oral bioavailability up to 20%, but only when administered as lipid nanocapsules	pKa -3.0, excellent solubility in water
Danaparoid	Oral bioavailability is minimal, less than 1%.	pKa 3.0; good water solubility
Hirudin	Oral bioavailability 10%	pKa 7.1-9.2; good water solubility
Lepirudin	Oral bioavailability 10%	pKa 4.4; good water solubility
Bivalirudin	Minimal oral bioavailability; poorly absorbed	pKa 2.78, reasonable water solubility
Argatroban	Oral bioavailability is minimal, less than 1%.	pKa 13.4 (thus, minimal oral absorption);

The chemical relatives of heparin are all generally parenteral agents. For example, direct thrombin inhibitor polypeptides like hirudin and lepirudin have modest oral bioavailability, probably because they are massive peptides and get destroyed by intestinal peptidases on their way through the gut (Cen et al, 2006). 

Distribution

In the circulation, the heparin drugs are highly protein bound, mainly to lipoprotein (unfractionated heparin seems to especially enjoy LDL). The same is the case for enoxaparin and dalteparin, though apparently to a much lesser extent, as the bulk of their molecules are less than 6000 Daltons. Because of this predilection for plasma protein binding, as well as their relatively large size,  these substances have a very small volume of distribution, i.e they are basically confined to the circulating blood volume.

Heparin	VOD= 0.1L/kg, highly protein-bound, mainy to lipoproteins (LDL)
Enoxaparin	VOD=0.05L/kg, basically confined to the bloodstream. Somewhat protein-bound, but less than unfractionated heparin (only the chains which are more than 6000 Da are protein-bound)
Dalteparin	VOD=0.05L/kg, basically confined to the bloodstream. Somewhat protein-bound, but less than unfractionated heparin (only the chains which are more than 6000 Da are protein-bound)
Fondaparinux	VOD=0.2L/kg; does not bind to any plasma proteins
Danaparoid	VOD=0.1L/kg; minimal plasma protein binding
Hirudin	VOD=0.3L/kg; minimally protein bound (only to thrombin)
Lepirudin	VOD=0.3L/kg; minimally protein bound (only to thrombin)
Bivalirudin	VOD=0.1L/kg; minimally protein bound (only to thrombin)
Argatroban	VOD=0.2L/kg; 54% plasma protein bound

The other heparin substitutes have similar properties, in the sense that most of them are confined to the circulating blood volume. 

Metabolism and clearance

Unfractionated heparin enjoys a rapid saturable clearance at low doses, and a slow first-order clearance at higher doses. Thus, the half-life depends on the dose.​​​​​

At low doses, due to binding to heparin-binding-proteins, macrophages and endothelial cells, heparin becomes sequestered and biologically useless, i.e. cleared for all intents and purposes. It is eventually degraded by depolymerization. To the observer, this looks like zero-order linear elimination.

At higher doses, some additional mechanisms take effect. After huge doses, a small amount is excreted in the urine, which probably represents the lower end of the molecular weight spectrum. Most likely the reticuloendothelial system plays a role in sequestering and destroying heparin. After Dawes and Pepper (1979) injected radiolabelled heparin into animals, the radioactivity is concentrated in the liver, spleen, bone marrow and lungs. Then something happens, and desulfated forms appear in the bloodstream, with basically unchanged molecular weight but absolutely none of the original antithrombin-binding activity. These zombie heparin molecules degrade gradually over hours, eventually turning into fragments small enough to be excreted renally. 

Name	Metabolism	Elimination and monitoring	Half-life
Heparin	Sequestered into reticuloendothelial cells and degraded gradually into inactive and renally cleared metabolites.	Biphasic (saturable) metabolism: with low doses, a rapid saturable clearance (by reticuloendothelial tissues), which becomes slower with high doses when this system is saturated. Monitored by APTT, which incorporates an assessment of thrombin activity	Half-life of 25 units per Kg = 30 minutes Half-life of 100 units per Kg = 60 minutes Half-life of 400 units per Kg = 150 minutes
Enoxaparin	Mainly metabolized by the liver via desulfation and depolymerization to lower molecular weight fragments, which end up being either less potent or totally inactive	About 40% of active and inactive fragments combined are excreted renally, which is why low molecular weight heparin is not especially well suited to renal failure patients. Monitoring is by measurement of anti-Xa activity	Half life is about 4-7 hours.
Dalteparin	Small amount of dalteparin is metabolised into lower molecular weright framents, mainly in the liver	70% of dalteparin is eliminated via the kidneys. Monitoring is by measurement of anti-Xa activity	Half life is about 3-5 hours.
Fondaparinux	Undergoes minimal metabolism.	The majority of the administered dose is eliminated unchanged in urine in individuals with normal kidney function. Usually, no monitoring is required, but anti-Xa activity can be used (though it may overestimate the dose)	Half-life is about 17-20 hours
Danaparoid	Minimally metabolised	50% of danaparoid is eliminated via the kidneys. Monitoring is by measurement of anti-Xa activity	Half-life is about 25 hours
Hirudin	Minimally metabolised; some hydrolysis in the liver occurs, whcih liberates [eptide fragments and amino acids	90% of the drug is cleared renally. Monitored by APTT or ECT(ecarin time)	Half life is about 0.8-1.7 hrs
Lepirudin	Minimally metabolised; some hydrolysis in the liver occurs, whcih liberates [eptide fragments and amino acids	48% of the drug is cleared renally. 35% as unchanged drug and the rest as fragments Monitored by APTT or ECT(ecarin time)	Half-life is about 1.3 hours
Bivalirudin	Mainly metabolised by proteolysis in the liver (80%)	20% of the drug is cleared renally as unchanged drug; the rest is metabolised Monitored by APTT or ECT(ecarin time)	Half life is about 20 minutes
Argatroban	Metabolised in the liver by hydroxylation and aromatisation of the 3- methyltetrahydroquinoline ring	Inactive metabolites are renally cleared. Monitored by APTT	Half life is about 45 minutes

Mechanism of action

Heparin is present in the body in the secretory granules of mast cells. It is also found in numerous animals, including various invertebrates which don't have anything even remotely resembling the human coagulation cascade. Which is weird. So nobody really knows exactly what its purpose is. But, in humans, heparin enhances the activity of antithrombin-III by a factor 1000. It does this by binding to antithrombin III and causing the active site to undergo a conformational change.

In this childish diagram, the flicking away of the antithrombin-III molecular tail represents the increased availability of the active site. Thus activated, antithrombin III inactivates several factors – but most notably, Xa and IIa (Thrombin). 

The inactivation of thrombin depends on heparin molecule length. Specifically, 18 disaccharide units is the key number (about 5kDa). On the other hand, inactivation of Xa is independent of length:  so long as any sort of heparin is bound to it, antithrombin-III will inactivate Xa. This underlies the difference in pharmacodynamic of low molecular weight heparin and unfractionated heparin.

Thus, in summary, unfractionated heparin affects thrombin, whereas low molecular weight heparin only affects Xa. This also explains why measuring APTT is not going to tell you whether the low molecular weight heparin dose is therapeutic.

The unfractionated heparin also affects the activity of Factor 9, but not the activity of Factor 7. Thus, the intrinsic and common pathways are affected, which increases the APTT. The extrinsic pathway is unaffected, and the PT does not rise very much. Because thrombin is unaffected by low molecular weight heparin, the APTT remains essentially unchanged.

Unlike heparin and the heparin-like drugs which make their effect felt by binding to antithrombin, most of the available alternatives (hirudin, lepirudin, bivalirudin, argatroban) are direct thrombin inhibitors. 

Side effects 

Apart from bleeding, which is an obvious answer, the only additional side effect which needs to be mentioned is HITS (Heparin-Induced Thrombocytopenia Syndrome), which is discussed at greater length elsewhere. In short:

• It is an immune-mediated thrombocytopenia well discussed by Franchini in 2005
  • More frequently associated with unfractionated heparin
  • More frequent in the elderly; unheard of in children
  • Cardiac and orthopedic surgery patients are at greater risk
  • Typically occurs 5-10 days after start of heparin
• Comes in 2 flavours: type 1 and type 2.
  • HITS Type 1:
    • Mild transient thrombocytopenia, platelet count above 100
    • Totally reversed by heparin cessation
    • Occurs in up to 10% of patents
    • NOT associated with an increased risk of thrombosis
    • Probably not even immune in origin
  • HITS Type 2:
    • Nasty severe thrombocytopenia, platelet count might drop to nil
    • Occurs in something like 1% of patients
    • Associated with thrombosis in 30% of cases
    • Due to the formation of antibodies to the complex made up of platelet factor 4 (PF4) and heparin; this complex forms on the surface of platelets.
      • When the HIT antibody binds to this complex, it causes platelet activation and aggregation, and so there is a tendency towards clotting (because all the platelets are activates) as well as a simultaneous tendency towards bleeding (as there is a destruction of antibody-coated platelets in the reticuloendothelial system)

If one has overdone one's heparinisation, th APTT will rise dramatically, and one may have some sort of bleeding complications. One may be caught thinking, "I wish I could put the coagulation cascade back together". This can be accomplished with protamine:

• 1mg reverses 100 units
• No more than 50mg at any one time

Protamine sulfate is far from benign. It is a foreign, unusual substance- a strongly alkaline polypeptide which binds to strongly acidic heparin irreversibly, and thereby decreases its anticoagulant effect on antithrombin-3. However, in ridiculous doses, protamine itself will act as an anticoagulant.

Among its many adverse effects are the following:

• Catastrophic hypotension due to vasodilation, which is thankfully brief (only about 3-4 min) - this seems to be the result of systemic histamine release, triggered in some sort of directly-complement-activating way by the circulation heparin-protamine complexes
• Pulmonary hypertension due to the localised vasoconstrictor activity of thromboxane, activated by an anaphylactoid reaction to protamine
• Anaphylaxis (it is after all a fish product)

The full pharmacology of protamine has never been a strongly emphasised element of the exam process, but perhaps somewhere nestled among all the useful material we might find the space for asides like this. So:

Protamine, in broad overview

When reading in any detail about the chemistry and origins of this product, most normal people are taken aback by the sheer weirdness of infusing into other humans.  Protamine is a small protein (approximately 5-13 kDa), originally derived from salmon sperm (and in fact called salmine for a time).   Its effect on heparin were discovered by Chargaff & Olson (1937) when they were trying to increase the duration of the anticoagulant effect of heparin by following in the footsteps of Hagedorn et al (1936), who successfully produced the first long-acting insulin by combining it with salmon protamine. If insulin, they reasoned, why not heparin? But "in the case of heparin the effect was unexpected: the anticoagulant action of heparin in vivo was entirely stopped by protamine", the surprised authors wrote. 

Pharmacokinetically, protamine was something of an unknown actor until Butterworth et al (2002) performed an extensive exploration of its properties, as they were baffled by how we could keep using this drug without any of this information. Its only bioavailable as an intravenous agent, which makes perfect sense as for your digestive system it is a nutrient indistinguishable from steak. It has a high pKa but is water soluble only with the help of a small amount of sulfuric acid, buffering the solution down to a pH of around 6. The immediate fate of protamine is something of a mystery; it appears to be confined the the circulating volume initially but then is found distributed to the lung and liver, where some of it is eliminated. Some sources also claim that some of it is eliminated renally (20-30%, according to DeLucia et al, 1993), but this does not seem entirely plausible considering its charge and size. Either way, the disappearance is so abrupt (half life of 5-7 minutes) that it often fails to clear all of the available heparin, resulting in "rebound" re-anticoagulation. 

How does it work? In terms of charge polarity, protamine is the anti-albumin, the Ahriman to albumin's Ahura Mazda. It is perhaps the most positively charged substance you will ever infuse into a person. Arginine and lysine, the amino acids which protamine is mostly composed of, leave abundant positive charge on the surface of the molecule, such that at physiological pH there are 7-8 positive charges per kDa of protamine. That is the basis of its role in its usual domain (the sperm) - it binds to  the negatively charged phosphate backbone of DNA and protects it from degradation during spermatogenesis. Infused into the heparinised human, it rapidly binds to the anionic heparin, and forms an inactive complex that has no anticoagulant activity. That complex is then eliminated by mechanisms which remain obscure, though one study in rats found it was eliminated by liver macrophages. 

One might comment that something with this much positive charge on its surface will not lay about indolently, but is likely to have all kinds of unexpected reactions; and this would be entirely accurate. The excellent review by Sokolowska et al (2016) is to be recommended for the reader interested in the finest detail. In short, apart from heparin, there are many other anionic drugs of interest (beta-lactam antibiotics, for example) and protamine will readily remove these from the circulation. Moreover it has a tendency to have an anticoagulant effect all of its own, by an antiplatelet effect as well as by inhibition of the extrinsic pathway (Factor V activation by thrombin is reduced by protamine). It can also cause anaphylaxis, because it is a foreign substance, as well as myocardial depression, pulmonary hypertension, systemic vasodilation, pulmonary oedema and haemolysis.   The literature on protamine toxicology is mostly a series of comments on the sad fact that protamine is the only option for heparin reversal for almost a hundred years, and that we tolerate it only because we have nothing better.

But does protamine reverse low molecular weight heparins? Yes, but incompletely. Crowther et al (2002) reports on the mechanisms of why and why not. In summary, the thrombin-inhibitory part of LMWH activity is completely neutralised, but not the anti-FXa activity, and the main factor seems to be reduced sulphate charge density (the specific anionic groups which attract protamine). If there were more sulfate charges to bind to, the protamine would be more effective, and the more sulfated products (eg. tinzaparin and dalteparin) are more efficiently inactivated, whereas only about 33-66% of the activity of relatively less sulfated enoxaparin is reversed by protamine. In fact, Crowther et al sulfated the bejesus out of some commercially available LMWH (to a density of three SO₃^- groups per disaccharide) and found that protamine comfortably reversed all of their activity, demonstrating the concept nicely. 

Resistance to heparin therapy

There are situations in which vast quantities of IV heparin fail to increase the APTT in spite of your every effort. One might call this "heparin resistance", or "heparin insensitivity".

There are several reasons one might be resistant to heparin:

• Increased heparin-binding protein levels (all of them are acute phase reactants)
• Low antithrombin-III levels (i.e. nothing for heparin to bind)
• Increased heparin clearance (eg. in liver disease)
• High Factor VIII levels

UpToDate offers a good article about Antithrombin III deficiency. Either you hereditarily fail to synthesise enough of it, or your liver is so damaged that it cannot produce enough. Or, it has been used up somehow, eg. in the context of DIC, MAHA, or in a bypass circuit. Lastly, it is possible that you are losing it along with other proteins via your leaky nephrotic kidneys. The management of AT-III deficiency is, predictably, supplementation with AT-III. If the expensive purified factor is not available, FFP will suffice. 

Effective coagulation of the heparin-resistant patient

There are several strategies one can employ. The specific choice relies on what exactly is causing the heparin resistance.  There are some good articles on this. Most of them do not touch upon the routine anticoagulation of some random patient who happens to have escalating doses of heparin; I suppose it is generally assumed that one will continue to escalate the dose until such time as therapetic goals are met. However, there are situations when anticoagulation is critically important, and one such scenario is the cardiopulmonary bypass circuit.

• Change to low molecular heparin, instead of unfractionated heparin
• Give  cryoprecipitate  and/or  fresh  frozen  plasma  (if  there  is  confirmed ATIII deficiency )
• Give antithrombin III concentrate

Or, you could consider using something else, such as a direct thrombin inhibitor (hirudin or argobatran)