This chapter is relevant to Section N1(i) from the 2023 CICM Primary Syllabuss, which expects the trainee to "describe the functions of the liver." The college have not specifically asked about storage functions as a stand-alone SAQ, but some mention of "storage" does tend to appear in the official answers, suggesting this is something of importance.
Questions asking about the functions of the liver have included:
The liver is a storage organ that packages mainly nutrient molecules (macro and micro) in order to buffer periods of nutritional scarcity. That does not necessarily mean prolonged nutritional deprivation - though it does play a role in the physiological adaptation to starvation. Rather, the main purpose of storing things in your liver is to produce a steady stream of metabolic fuel between regular meals, thereby smoothing the peaks and troughs of nutrient availability and maintaining a stable plasma nutrient content.
Storage functions of the liver include:
Metabolic fuel storage:
- Glycogen storage
- The liver buffers the flux of blood glucose
- Immediately after a meal, about 20% of ingested dietary carbohydrate is stored in the liver, and then released slowly between meals
- Hepatic carbohydrate is stored as glycogen, a branching glucose polymer
- Normal hepatocyte glycogen storage is less than 75g (400 kcal), which is about 20% of the total body glycogen reserve (the rest is in skeletal muscle)
- Lipid storage
- Fat is stored in the liver as triglycerides in hepatocyte vacuoles
- Normal hepatic fat content is less than 5% (75g, or 675 kcal)
- Triglyceride is created from excess glucose when carbohydrate intake exceeds the normal glycogen storage capacity (about 100g)
- Under conditions of impaired lipid handling by adipocytes (eg. in obesity) the liver also becomes a destination for free fatty acids, contributing to their storage
- The lipid fraction of TPN can also cause hepatic steatosis because of its high free fatty acid content
- Excess hepatic fat storage leads to hepatic steatosis, non-alcoholic fatty liver disease, insulin resistance, cirrhosis, and hepatocellular carcinoma
- Vitamins: fat soluble vitamins A, D, E and K as well as water-soluble B12
- Vitamin A: 50-70% of the total; stored in stellate cells
- Vitamin D: minor site (most of it is in adipose tissue); stored in hepatocytes
- Vitamin E: a third of the total content is in the liver; in hepatocyte mitochondria
- Vitamin K: mostly stored in the liver; small reserve, rapidly depleted (over days);
- Vitamin B12: 50% of the total body stores is in the liver
- Iron: stored in hepatocytes as ferritin and (in excess) as haemosiderin
- Copper: one third of the total body reserve is stored in the liver as complexes with "metallochaperone" proteins in hepatocytes
- Trace elements (zink, selenium, tin, manganese, molybdenum and cobalt)
Blood reservoir function:
- Low vascular resistance circulation
- Acts as a venous reservoir (along with skin, spleen and lungs)
- A large volume of blood can be mobilised into, or out of, the hepatic circulation, in order to buffer fluctuations in cardiac preload
- Some estimates state this reservoir can contribute 27% of the total blood volume
"I propose to describe as storage that amount ... laid down in excess of the actual physiological level needed for normal functioning of the organ", is how E. Kodicek explained what is meant by "storage", as opposed to "contents". For example, the average liver may consist of approximately 20% amino acids by weight, but these are not "stored", as each is incorporated into essential structural and functional proteins.
Glycogen is a branching polysaccharide created out of glucose molecules; each glycogen deposit can contain around 55,000 glucose residues, which makes it noticeable on electron microscopy (Adeva-Andany et al, 2016, describe hepatic glycogen deposits measuring up to 290 nm). These deposits are stored in a hydrated potassium-enriched form, with each gram of glycogen requiring 3g of water and 0.45mmol of potassium (Kreitzman et al, 1992). The liver is not unique in being able to produce and degrade this substance - it is ubiquitous in the body; of the total body glycogen content (about 15g/kg, or 600-700g), only about 75g is in the liver (Björntorp & Sjöström, 1978). The majority of it is stored in skeletal muscle, as well as brain, myocardium, kidney tissue and various other depots. These deposits are somewhat less useful, as most of these tissues lack glucose-6-phosphatase, and are therefore incapable of releasing large amounts of glucose into the bloodstream (i.e. their stored glycogen is destined for local use only). The liver is therefore the main source of blood glucose during the time between meals.
Glycogen acts as a reservoir of glucose which can be deposited into, and withdrawn from, depending on the wax and wane of nutrient availability. There is a constant balance of glycogen synthesis (by glycogen synthase) and glycogen degradation (glycogenolysis shall we call it), which is under hormonal regulatory control:
This method of storing glucose in the form of glycogen is ancient, and probably predates multicellular life and plasmid endosymbiosis (i.e it's older than our alliance with our mitochondria). The only organisms that seem to lack this system are those that lead a parasitic lifestyle - in which case it appears that they have lost it to disuse, because they can mooch off the host's glycogen. To criminally oversimplify the detailed analysis by Wang et al (2019), one could summarise that glycogen storage is one of the fundamental elements that make up life on Earth, along with DNA, RNA, and the lipid bilayer.
Hepatic glycogen stores appear to max out at around 100g, or about 4-5% of the total liver mass, which equates to approximately 400 kcal - beyond that, according to Acheson et al (1988), the liver converts the excess carbohydrates into fat. The weight-conscious reader will be horrified to learn that these data were collected from volunteers who agreed to undergo "massive carbohydrate overfeeding", consisting of 5000 kcal/day of pure carb (supplied mainly as delicious-sounding "sugared fruit juices"). After four days of this ridiculous juice orgy, the investigators found that no further glycogen was being deposited, and the excess dietary carbohydrate was being converted into fat, with up to 150g of fat being deposited every day. "We demonstrated that this pathway can readily dispose of nearly 500 g of glucose per day", they belched happily.
This fat accumulates harmlessly at first, filling small lipid vacuoles in hepatocytes, but as the process continues it begins to interfere with liver function, earning itself a pathological descriptor (hepatic steatosis). The hepatology community recognises several grades, determined by the percentage of the liver mass accounted for by fat. Anything less than 5% is considered safe, 5-33% is mild, 33-66% is moderate, and anything over 66% is severe.
A jarring image is added here as a speedbump, to keep the conversation from moving lightly forward over this horrifying percentage. Yes, reader, it does mean that, in some people, over two-thirds of the liver's mass is replaced by pure fat. This foie gras level of hepatic steatosis is remarkably common (5% of the obese population in some studies), and comes with a number of unpleasant consequences - for example insulin resistance, cirrhosis, and primary hepatocellular malignancy (Nassir et al, 2015). For the record, and to complete this nightmarish comparison, the average percentage of fat in foie gras is only 55.8%.
Interestingly, where glycogen and fat vacuoles are used to store excess dietary carbohydrate in the liver between meals, the same cannot be said for dietary fat itself. Periprandial buffering of fatty acids seems to be the role of the adipose tissue, i.e. fatty acids entering the bloodstream seem to be incorporated into adipocytes, and then released from them as needed during periods of fasting (Frayn, 2002). However the liver does seem capable of absorbing free fatty acids and converting them into triglycerides for storage. This can happen in a number of circumstances - for example, when the adipocytes are unable to perform their role (eg. obesity and diabetes), or where there is an absurd excess of fatty acids in the bloodstream (eg. where the patient is being infused with TPN). TPN usually contains large supraphysiological amounts of free fatty acids and Cavicci et al (2007) found evidence of harmful hepatic triglyceride deposition with certain formulae, for example 20% Intralipid.
Returning to the point of this section, it being the storage of dietary lipid, one can summarise by saying that the liver acts as an ectopic site for fat storage, which usually does not exceed 5% of its total mass (75g), or approximately 675 kcal in total metabolic energy. Under normal circumstances, outside of cruel goose-stuffing and the unrestrained carbohydrate excesses of the North American diet, this fat reserve plays a minimal role in the total flux of fatty acids in the body.
The liver stores fat-soluble vitamins A,D,E and K, as well as vitamin B12. Without digressing extensively on the topic of their absorption or function, the role of the liver in their storage can be summarised as follows:
Vitamin A is stored in hepatic stellate cells as retinyl palmitate. Textbooks often quote a range of 50%-80% to describe the amount of total vitamin A stored in the body, which implies that there is also some large extrahepatic reservoir of this vitamin, but this may actually be untrue. As Nagy et al (1997) have rightly pointed out, these numbers probably come from animal studies where the animals were oversupplemented with vitamin A. Still, there are pericytes everywhere, and even in rats on a normal diet there are small vitamin A deposits in lung and intestine. According to an autopsy study of dead New Zealanders from 1963, the average vitamin A content of a human liver is approximately 1000 IU per gram, which equates to approximately 500 days worth (assuming a recommended daily intake of 3,000 IU and a normal-sized 1500g liver).
Vitamin D is really a large number of different molecules, of which 25(OH)D3 is the one most people would be interested in, as it is the biologically active form of vitamin D, synthesised by the liver from cholecalciferol (itself formed by the action of ultraviolet light on 7-dehydrocholesterol). All of these various daughter and parent molecules are basically sterols, and therefore highly fat-soluble. From this, it unsurprisingly follows that they distribute into all fat everywhere, i.e.not exclusively the liver. Cadaveric tissue assays demonstrated that only about 2% of injected radiolabeled D ended up in the viscera, with the vast majority divided between adipose tissue, bone marrow, skin and skeletal muscle (Mawer et al, 1972).
Interestingly, though vitamin D can (and does) accumulate massively in adipose tissue, there is no apparent beneficial reason for this to happen, nor is there any serious harm from even the most preposterous level of its accruement. The normal recommended daily intake is only 400-800 IU; but in order to develop honest toxicity from this substance, one really needs to be consuming more than 10 000 IU per day, as it distributes harmlessly into fatty tissues. This can go on seemingly forever, and with no apparent downsides- when seventy-six subjects were administered a weekly dose of 20,000 IU for five years, Martinaityte et al (2017) found no major adverse effects (also the levels remained high even a year after the period of hypersupplementation had ended). Even a normal store of vitamin D is massive, with Mawer et al (1972) reporting a minimum total body content of something like 127,000 IU (enough to last for 150-300 days of deprivation). This raises even more puzzling questions, such as why one should have such a huge reserve of a substance that one's own skin can easily synthesise at a rate of 1000 IU per every 6 minutes.
Vitamin E, like vitamin D, is actually a group of molecules, except unlike vitamin D these are not parent and daughter metabolites, but a clone of eight near-identical stereoisomers (tocopherols), of which α-tocopherol usually gets the most mentions because it has the highest concentration (humans seem to preferentially absorb and store it to the exclusion of other molecular variants). The liver is not the only organ that accumulates it - Wiss et al (1962) reported a high fraction of radiolabeled α-tocopherol uptake in the heart, kidneys, adrenal glands, adipose tissue and the intestine. Even though it normally contains only about one-third of the total body vitamin E content, the liver is usually described as the primary store, perhaps because it has a vast capacity to accommodate an unexpected excess - when the dietary intake is increased, hepatic store expand dramatically, whereas other stores do not (Drevon, 2009). It is mainly stored in hepatocytes (interestingly, in the mitochondria). Under normal circumstances, the total body store is estimated to be 25,900 μmol, or approximately 740 days worth (going by the recommended daily intake of 35 μmol).
Vitamin K is another term that refers to a group of different molecules, mainly referring to dietary phylloquinone (vitamin K1) and menaquinones (vitamin K2). Unlike the other fat-soluble vitamins, which the human body seems to greedily hoard, vitamin K has a fairly small pool, and we are highly dependent on regular intake to top up our stores. When Novotny et al (2010) experimented with radioactive kale, the total body content of their subjects ended up being about 41-46 μg (for phylloquinone), of which the majority was stored in the liver (though there is also some in the heart and pancreas). This is actually less than one day's worth of recommended intake (70 μg/day). This probably explains the wide variety seen in the vitamin K content of donated post-mortem livers - the range is from 1.1 to 21.4 ng/g of tissue, i.e it varies by a factor of twenty, depending on the last time you ate your greens.
Vitamin B12 (cyanocobalamin) is stored in the liver, unlike the vast majority of water-soluble vitamins. According to biopsies analysed by Joske in 1963, the human liver tends to contain about 0.5-0.6 μg/g of B12, or 750-900 μg in total (which is about 50% of the total B12 in the body). Given a recommended daily intake of 2.4 μg/day, this stored reserve represents about 315-370 days worth. It does tend to hang around: Glass et al (1958) determined that the half-life of stored B12 is approximately twelve months.
Iron is mainly stored in the liver. Hepatocytes accumulate it via receptor-mediated endocytosis, accepting it mainly in the form of circulating transferrin or ferritin; additionally there is a constant stream of ferritin from Kupffer cells, as they ingest effete red cells. Because it is a toxic horror chemical, iron needs to be stored in a specific protective form, and ferritin is usually that form. Normal liver iron content is said to be something less than 1.8 mg/g dry weight. In total, a normal male can be expected to have 600-1000mg of elemental iron on their person at any given time, whereas a normal adult female usually only has 200-300mg. This reserve is usually enough to last for a very long time. Anderson & Shah (2014) describe in detail how the elemental iron in the body is reclaimed and recycled, so much so that the recommended daily intake is only about 1-1.5mg.
When ferritin storage is fully saturated, some of it becomes degraded into haemosiderin, an insoluble complex that consists of degenerate fibrin remnants and all kinds of other random garbage. Haemosiderin tends to be aggregated into siderosomes, organelles that sequester the toxic iron atoms even further, making it safer to store them, but also making it more difficult to reclaim it (Iancu, 1992). This form of overflow storage accommodates the iron which is excess to need, for example in iron overload states. There does not seem to be an upper limit on how high hepatic iron content can go in haemosiderosis, except perhaps cirrhosis and death. As an example, Olynic et al (1998) found some haemochromatosis patients with hepatic iron content values up to 40 mg/g, i.e. twenty times the norm, and these people still only had the same phlebotomy requirements as the rest of the haemochromatosis crowd, suggesting that there might be a population of sicker patients with even more iron in their livers.
Copper is also a highly reactive oxidant species that would be dangerous on its own, which means its storage in the body is mainly in the form of complexes with "metallochaperone" proteins (Roberts & Sarkar, 2008). In the bloodstream, it travels as the relatively harmless ceruloplasmin, or in a loose association with albumin. Throughout the livers of the animal kingdom hepatocytes preferentially absorb and store this mineral, apparently because they are helplessly fascinated by it, and powerless against its charm - Beck (1956) suggested that "the high liver copper level characteristic of some species is due, not to a higher intake of copper or to a greater absorption, but to a lesser ability to restrict the storage of copper in the liver". The normal range in humans is around 15-55 μg/g dry liver weight, or up to 250 μg/g in Wilson's disease, and the total body store of copper is approximately 50-120mg, of which the majority (at least two thirds) is stored in the bone and muscle. The recommended daily intake is about 900 μg, meaning that it could take up to 130 days to become depleted (see Milne, 1994, for details of what clinical copper deficiency might look like).
Trace elements essential for survival (zink, selenium, tin, manganese, molybdenum and cobalt) as well as unfriendly heavy metals (which is most of the rest of them) are often stored in the liver. This seems to also be a common trait among vertebrates. Some have clear importance, in that their deficiency leads to problems, and others have theoretical importance which is only guessed at. The boundaries of what is known about trace elements are so narrow that one can only assume CICM examiners will never ask about them, and the trainee should only know that the liver is one of the potential sites of their storage. This table from Linder (1984) is left here so that the reader's eyes can slip frictionlessly across it on their way to the next section.
Phosphate should probably be mentioned here, even though its presence in the liver somewhat violates the agreement about the definition of what "storage" is, established at the beginning of this chapter. All phosphate in the liver is doing something useful, eg. busy being ATP, or participating in glucose metabolism, or putting the "phospho" into phospholipids. According to a landmark paper by Woodard & White (1986), the liver is second only to the brain in phosphate content (elemental phosphorus comprises 0.3% of its mass). The reason this is brought up here is that in states where the liver is regenerating (eg. following liver surgery), serum phosphate levels are seen to drop- a phenomenon that has usually been attributed to increased uptake of phosphate into dividing hepatocytes.
This role of the liver seems to be brought out in CICM model answers and various official textbooks as if it were a genuine "job" of the organ. This might sound puzzling, as all blood in the body is generally expected by the reader to be doing something circulatory, i.e. to remain in constant motion is essential to its task, and so to have a weird blood-filled gourd in the abdomen seems counterintuitive. But still, this is exactly what the liver is, and how it acts. Specifically, the low resistance vessels in the liver act as a buffer system to absorb fluctuations in volume: when volume is rapidly added to the circulation, most of it is rapidly sequestered in the liver, reducing the fluctuations in central venous pressure. Conversely lower central venous pressure and higher sympathetic tone vasoconstricts the hepatic venous circuit, permitting more blood to reflux from the hepatic venous beds. This certainly seems to be the case in other mammals (eg. cats, in Lautt & Greenway from 1976), and in fact Greenway (1983) claimed that up to 27% of the total blood volume could be reclaimed from the splanchnic vascular bed in this way (in general it seems most of the work in this area was done by Clive V. Greenway and Wayne Lautt, as their articles are quoted wherever a textbook mentions the blood reservoir functions of the liver and has the decency to include proper references). In summary, together with the skin lungs and spleen, the liver is a venous blood reservoir, from which blood can be mobilised in times of haemorrhage, and into which fluid boluses can be buffered.