This chapter is addressed to Section O1(i) from the 2017 CICM Primary Syllabus, which expects the exam candidate to "describe the control of gastrointestinal motility, including sphincter function". This was assigned an L2 level of importance by the curriculum designers, the lowest tier of expected educational achievement which can be interpreted as "a trainee should perhaps have some vague understanding of at least the terminology used to discuss this topic". A direct reading of the syllabus document would have therefore made gastrointestinal motility a forgettable aside to be left until the last minute cramming.
However, the real importance of this topic is massive, as it has appeared in no less than five past SAQs, accounting for 25% of the total gastroenterology questions:
As you can see, rather than "gastrointestinal motility, including sphincter function", these questions mainly focused narrowly on the control of gastric emptying. How did this happen? The most likely explanation is a transcollegial migration of exam material, with anaesthesia-related material creeping across into the CICM exam from ANZCA papers. Another source of gastrocentic bias is the textbooks chosen by CICM as "official resources", which are often anaesthetic physiology textbooks, and which uniformly put the maximum emphasis on everything related to the safe instrumentation of the airway, to which an empty stomach is highly relevant. However, the intensivist often continues looking after the patient long after the excitement of intubation has passed, and has to deal with all kinds of intestinal motility problems (both too much and not enough).
In summary, specifically on the topic of gastric emptying:
- Fasting state:
- Migrating motor complexes sweep the stomach at regular intervals
- These are slow peristaltic waves which originate in the fundus
- Their role is to keep the stomach empty of secretions and food debris
- They are interrupted by eating
- Fed state:
- Receptive relaxation: shortly following swallowing, the proximal stomach relaxes by a vagally mediated reflex
- Tonic contraction: the body of the stomach contracts, propelling food slowly towards the antrum
- "Lag" phase with solid food:
- The antrum contracts forcefully against a tightly closed pylorus
- This triturate the food until it is well mixed with gastric secretions and its particles are less than 1-2mm in size
- During this phase, there is minimal passage of food into the duodenum
- Depending on the food, this takes 30-90 minutes
- Linear emptying phase with solid food:
- small food particles are pushed through the pyloric sphincter at a stable linear rate
- Time to empty half of the meal = 30 minutes for relatively nutrient-poor solids, 60-120 minutes for especially fatty solids
- Exponential emptying with liquids:
- Liquids have no lag phase and begin emptying rapidly and immediately
- Emptying rate is exponential and determined by the antral-duodenal pressure gradient, which is usually 5mmHg but which can increase to over 40 mmHg with peristalsis
- 15-20 minutes is the half-time of non-nutritive fluids, eg. water
- Factors that affect gastric emptying:
- Position (fastest when upright or right lateral)
- Consistency of the food (solids empty slower)
- The volume of liquid (larger volume empties faster)
- Gastric pH (acidic pH decreases the rate of emptying)
- Tonicity of contents (decreases gastric emptying)
- Lipid content (fatty meals slow gastric emptying)
- Caloric content (high-calorie meals slow gastric emptying)
- Hormonal influences on gastric emptying:
Hormones which: Enhance gastric empyting Slow gastric emptying
- Gastric inhibitory polypeptide
- Glucagon-like peptides 1 and 2
It is a frustrating pun, that no single resource brings all of this material together in an easily digestible way. The closest was probably Sanders et al (2012), which - though full of tough and fibrous biochemistry- at least makes an effort to lay their discussion out into a bento of easily navigated headings. For gastric emptying, Goyal et al (2019) is an excellent free alternative to paywalled works by Hellström et al (2006) or Minami & McCallum (1984), though the latter may be more comprehensive. For when you're in the right sort of mood, anything by J.N Hunt and M.T Knox is gold - these researchers, active in the sixties and seventies, are responsible for much of what we now know about gastric motility, and their studies have a hilarious fratboy dare sort of quality to them (generally being on the theme of "let's see what happens when we feed concentrated acid to these healthy volunteers").
Peristalsis has no "official" medical definition from any nomenclature-governing society, but this is fine because the vast majority of reference works will give you a decent one-liner. The term itself is ancient, originating from peri (around) and stalsis (constriction). Schemann et al (2021) attributes it to Chez Vincent (Dictionaire Raisonne’ d’Anatomie et de Physiologie, 1766):
"We give this name to the vermicular movement of the intestines, which tends to push out the excrement outside, and to facilitate the entry of the chyle into the milky vessels"
Schemann et al themselves give a more concise definition, which
"a ring of intestinal constriction traveling aborally"
"Aboral" being a zoological term meaning "in the direction away from the mouth". Though these definitions would be an excellent excuse to use the words "aboral" and "vermicular" in a written exam answer, the CICM trainees are cautioned against using anything published before 1900, even if it underwent peer review, and generally against throwing words around which the examiners may not recognise. For a short opening statement of a "outline the smooth muscle activity of the gut" kind of question, one could comfortably stoop to the use of something like Merriam-Webster, and go with:
"Successive waves of involuntary contraction passing along the walls of a hollow muscular structure (such as the esophagus or intestine) and forcing the contents onward"
Anyway: peristalsis. It was first explored and described by Carl Lüderitz (1890), who did a lot of work but didn't get anything named after him, unlike that showpony Auerbach. This wave-like constrictive movement of the intestine is present in all segments, from the oesophagus to the anus, and propagates generally in the aboral towards-the-anus direction, though this is not uniformly the case. In terms of how much detail is require for this from the CICM exam point of view, we can take the examiner comments verbatim ("an outline of peristaltic waves, the basal electrical rhythm and its modulation, the migratory motor complex (MMC) and its modulation, neural input, stretch and hormonal control"). Thus:
Peristasis consists of coordinated movements of the tunica muscularis, the muscular layer of the intestine. Specifically, this layer consists of fibres running in two directions. The inner layer is made up of circular fibres, and the outer layer is made up of longitudinal fibres. Working from the belief that everything sounds more authoritative if it is in Latin, the reader may choose to label their diagrams with stratum circulare and stratum longitudinale. Combining some images from Carl Toldt (1912) and an ancient out-of-print Soviet anatomy book:
These smooth muscle layers are the effectors of peristaltic wave activity. If you think of the intestine as a cylinder, the encircling layer of muscle decreases the diameter of the tube, and the longitudinal layer decreases its length. Behind the bolus of food, circular muscles contract and longitudinal muscles relax; whereas ahead of the bolus, the longitudinal muscles contract, and the other ones relax. This is really one of those things that is much easier for an illustrator to explain than for a writer to describe. Observe, a length of intestine trying to peristalse (peristalt?) some kind of nightmarish rounded blue food bolus:
This process clearly requires some coordination. Functionally, you'd have to have several components all acting together:
This peristaltic response to the intestinal stretch stimulus is automated and is generally referred to as the "myenteric reflex", because "Bayliss & Starlings' Law of the Intestine" sounds much too authoritarian.
This sort of stretch-induced peristaltic activity is usually guided in some way by the autonomic nervous system and hormonally, but it will quite happily carry on in the absence of these. Yes, reader, this does mean that one may have a disembodied length of bowel peristalsing quietly in a nutrient tank, as a low maintenance pet of sorts. This basically describes 90% of the early literature on the subject. "The portion to be examined was removed from the body immediately... and was placed in oxygenated Ringer’s solution kept at body temperature", wrote Walter Cannon in 1912 (the rest of that experiment can be summarised as "I poked it, and it moved"). It is coordinated by the myenteric nervous plexus, otherwise known as Auerbach's plexus. More detail can be found in Costa et al (2000), but the CICM trainee doesn't need it. Knowing the CICM examiners, they would want a highly simplified version, such as this:
Now, that's a response to stretch, i.e something distending the gut. But in fact even without any extraneous stimulus the gut will carry on with its peristaltic activity, albeit unenthusiastically. The reason for this is the basal electrical rhythm, which can be described as a pacemaker current for the gut.
"Basal electrical rhythm", "slow waves", "pacesetter potentials" and "control rhythm" are all names that have been given to the spontaneous electrical activity of intestinal smooth muscle which is rhythmic, constant, and highly variable between gut regions. Alvarez (1922), when he and Lucille Mahoney discovered and described this phenomenon, called them "action currents", but this name did not stick, clearly because it make them sound too energetic. In fact they are rather sluggish, by standards of excitable tissues. Here, in Alvarez and Mahoney's original recordings, you can see that there are only 2-3 oscillations per every 10-second interval.
Yes, that does look like a rhythm strip of VF. These are sine-wave oscillations measured from intestinal smooth muscle cells, with a rather low amplitude (lower than the amplitude of a normal neuronal or cardiac action potential). Most textbooks for some reason give 5-15 mV as the height of these waves, with no reference. Mangel et al (1982) reported that the mean amplitude in intestinal smooth muscle was about 27 mV, with a resting membrane potential of around -67 mV (although it's hard to call it "resting" with a straight face if it keeps oscillating like that). The duodenum goes a little faster (15-20 "beats" per minute), the jejunum a little slower, the ileum slowest of all (7-10). They obviously don't automatically sync with each other (otherwise the rate would be the same everywhere); for example, the slow waves from the stomach do not propagate into the small intestine because of an "electrically quiescent" region in the duodenum, which does not conduct. They are also only capable of propagating for a few centimetres before stopping or colliding with another slow wave, to paraphrase Sanders et al (2005). This makes sense, as smooth muscle cells, even when electrically coupled with gap junctions, are not very good at propagating an action potential. In any case, coordinated muscular pump activity would rapidly and efficiently move the contents of the gut in the aboral direction without giving it much of a chance to digest anything, so "rapid and efficient" is probably not what you would want from this system.
The attentive reader might at this stage point out that a smooth muscle, broadly speaking, is a feckless and indifferent tissue, estranged from ambition or initiative. Ergo, these slow waves could not possibly originate with the smooth muscle, as it usually lacks the capacity to self-depolarise. Another pacemaker tissue type is therefore required.
In fact, slow waves are thought to originate in the interstitial cells of Cajal. These are a mysterious specialised cell type, originating from ventral neural tube cells (i.e. they share their origins with neurons and glia). They form a syncytium network of electrically connected cells that interpenetrates intestinal smooth muscle tissue, which is great because intestinal smooth muscle is useless at propagating an action potential from cell to cell. Garcia-Lopez (2009) is probably the best reference for this complex histological topic, made all the better by their use of Santiago Ramón y Cajal's own art and slides in tribute to his work. Here, some of these are reproduced with an unnecessary layer of childish colouring:
The whole point of that was to illustrate how the cells are a) all-pervasive and embedded within the smooth muscle layer, and b) all interconnected in a mesh. Being neurone-like, they generate action potentials, which seem to propagate from cell to cell in a way that is completely unlike the normal propagation of action potentials in the nervous system. Even an unmyelinated neuron might have an axonal conduction velocity of 1m/sec and an action potential duration of 200 msec or so, owing to the rapid action of its voltage gated sodium channels. In contrast, the networked cells of Cajal take things nice and slow. Lammers & Stephen (2008) report a propagation rate of 10cm per second, which slows to 1cm per second in the distal bowel. The action potentials are also longer, with a slow plateau phase. Here's a recording of a typical action potential from a mouse Cajal cell by Kito et al (2005).
The process of spontaneous depolarisation and action potential propagation seems to be mediated by the activation of dihydropyridine-resistant calcium channels. Thomson et al (1998) and Sanders et al (2005) are probably the best resources to explain what these are and how they work in the cells of Cajal. Like with all excitable tissues, extracellular magnesium potassium and calcium concentrations influence this activity.
All these action potentials occurring in uninterrupted unison create "waves of electrical activity that spreads through the intestines from the stomach to the terminal ileum in a regular cycle during fasting", which is the definition given by Cheng et al for the "migratory motor complex" or MMC. These waves of electrical activity are accompanied by waves of peristaltic activity which move intestinal contents. The "during fasting" component is critical: this activity is actually interrupted by eating. Deloose et al (2012) give a more detailed account of this process and its modulation, and the easily fascinated reader is enjoined to finish their 9,030-word paper, but for the rest of us the following important points will suffice:
The point of this activity is thought to be janitorial, sweeping material from the GI tract to keep it from fermenting hideously in the various loops and pockets. "Small intestinal bacterial overgrowth" doesn't quite cover the unclean kitchen sink S-pipe imagery conjured by the very thought of it.
The whole process and its various modulating influences is incredibly complicated, and in case anybody wants to go chasing every intricate detail, Sanders et al (2012) have an excellent overview. For the purpose of functioning as an ICU doctor to understand every intricate detail would not be essential, as we really only have very crude levers to pull when it comes to pharmacological peristalsis control. The following is a non-exhaustive list of factors that can affect intestinal motility, based on the middle of this excellent review by Ravi Avvari (2019)
This brings us finally to gastric emptying, which is the main exam-focused component of this chapter. Why did it take so long to get here? Mainly, because all of the peristalsis material also describes what the stomach does.
So: finally, we are at a point where we can actually discuss "the co-ordinated emptying of chyme from the stomach into the duodenum", borrowing from the examiner comments for Question 8 from the second paper of 2018. It has taken so long to get here because it felt logical to first explain peristalsis in general and then to narrow the discussion down to the discussion of the stomach specifically. All of the general principles that govern the coordinated peristaltic activity of the gut are either present in the stomach or have some gastric analogy. Having said that, no trainee would have been harmed by skipping the peristalsis section entirely and just starting from here.
A swallowed food bolus ends up being dumped into the gastric fundus and the superior part of the body, occasionally referred to as the corpus. These anatomical regions of the stomach immediately relax to accommodate the bolus. This receptive relaxation takes place within seconds or even fractions of a second of swallowing the food bolus at the level of the pharynx, and is functionally analogous to the relaxation of the circular intestinal muscle downstream of the food bolus. Unlike the intestinal relaxation, which is managed autonomously by the enteric nervous system, this is a vagally mediated reflex first described by Cannon & Lieb in 1911, "through the kindness of Mr. A. L. Washburn, who had accustomed himself to the presence of a balloon in the cardiac end of his stomach and a tube in his esophagus". The authors did not speak a word of thanks to the countless cats whose vagus nerves they severed, but that was the key test that demonstrated its essential role in the innervation of this reflex.
At the same time, at least for a solid food bolus, the middle area of the stomach contracts, keeping the received food bolus in the body of the stomach. According to Moore et al (1986), the purpose of this "transverse midgastric band" is to separate the stored food in the reservoir parts of the stomach from the grinding and pushing parts (the antrum and pylorus). This probably regulates the rate of delivery of solid food to the antral (pyloric?) pump, which probably would become overwhelmed by and large and sudden influx of solids. This band was discovered first in animals, and was known about for close to a century before Stieve (1919) found the same structure in the food-filled stomachs of executed prisoners.
The gastric peristalsis pacemaker is somewhere in the mid-body. Following receptive relaxation, contraction of the body takes place, propelling chewed food in the direction of the pylorus (aborally, if you prefer). There, the antrum contracts forcefully against the closed pylorus, mashing and grinding the food into the stomach acid, gradually breaking down the acid-sensitive structural components and decreasing the size of the food particles. By this process, the stomach is able to "sieve" the food particles, and passes them on into the duodenum only when they are down to 1-2mm in diameter. Meyer et al (1981) referred to this process as "trituration", a term borrowed from chemistry that is usually used to describe the combined actions of mixing something while reducing its particle size. This represents the first phase of the biphasic emptying of solids in the stomach, which is usually referred to as the "lag" phase, owing to the time it usually takes. Here, a famous image is slightly altered to demonstrate the duration of this lag period on a delicious-sounding technetium-99-labelled omelette. It appears all over the place, and none of the references usually included with this figure point to the original data, which makes it abandonware; but it has little triangles on it which make it look like somebody patiently collected measurements, which means the real author of this technetium omelette is somewhere out there.
Obviously, each omelette is as unique as a snowflake, and therefore this lag phase will differ from meal to meal (especially if the consistency of the meal differs significantly). Larger meals, denser food, and larger chunks, will produce longer lag phases. For example, for Siegel et al (1988), the lag phase was about 30 minutes with scrambled egg and about 60 minutes with chicken liver. Presumably, some kind of horrendously chunky high-density food would take even longer to pulverise. The interested reader is referred to this helpful food density database from the Food and Agriculture Organization of the United Nations (disappointing spoiler alert, the densest actually edible food is sweetened jam, at 1.43g/ml).
With the food bolus smooshed into 1-2mm fragments, the emptying of the stomach can finally take place, and for solids the emptying rate is linear. It occurs by the actions of the pyloric pumps. Or is it the antral pump? Or the peristaltic pump, or the pressure pump? In fact the literature seems to refer to a whole bunch of different confusing pumps, and here is as good a place as any to discuss them.
Without enraging the reader any further with pointless digressions on the origins of medical nomenclature, it is safe to say that nobody can agree on what to call these phenomena, but the CICM trainee probably needs to be able to discuss them somehow, and the best way would probably be in terms of their function:
Both of these activities are regulated by various hormonal autonomic and local neural systems, which will be discussed below, as to make space for it in this section would interrupt the flow of this already haphazard mess of a chapter. Instead, let's focus on the pressure difference between the stomach and the duodenum. This is the main driver for the emptying of ingested liquids from the stomach.
Obviously, where there are no chunks to grind, there is nothing to stop the stomach from emptying immediately, and so there is no lag phase. Liquids, following their ingestion, distribute rapidly throughout the stomach and then start emptying as soon as they reach the pylorus. The emptying rate is directly proportional to the volume of ingested fluid, and follows first-order kinetics (i.e. it is exponential). The rate of emptying depends on the pressure gradient generated by the rest of the stomach (mainly the antrum) as it pushes the fluid volume against pyloric resistance. The fastest emptying rates are seen with nutrient-poor liquids such as water, as there is some negative hormonal feedback from the small intestine which can slow the rate of gastric emptying for particularly nutrient-rich fluids. In case anybody is wondering what that looks like when it is compared to the emptying of a solid meal, here's a nice diagram from Houghton et al (1988), who fed their volunteers some sort of barf-inducing mixture of normal saline and chicken burger (chicken liver burger, to be absolutely precise).
In case anybody wants or needs to know the precise pressures involved, impossibly granular analysis is available from Indireshkumar et al (2000), who recorded antral pyloric and duodenal pressures and plotted them over time. For the casual reader, it will suffice to know that typically resting antral pressure remains around 5mmHg, whereas duodenal pressure is essentially 0 mmHg (i.e. equal to atmospheric pressure). Food-induced waves of peristaltic activity transiently increase the pressure here, which can peak at up to 40-50 mmHg in the antrum. The corresponding tightening of the pyloric sphincter here could crack a walnut (the investigators recorded one spike of around 180 mmHg).
After all this discussion, and the introduction of unrepeatable terms like "trituration", the time-poor candidate might ask for some memorable numbers they could store in their inc sack. How fast does the stomach empty?
These numbers are half-times, i.e the time it takes to empty 50% of the ingested volume, and they come from an excellent summary by Michael Camilleri (2019). Trainees could eject these data at the examiners as a decoy (in place of real understanding). For the purpose of practising in ICU, they are of course completely meaningless, as the precise time will vary considerably with different states of critical illness, and because in any case only very few of our patients are capable of tolerating regular 600-kcal meals. The combination of continuous high-density isoosmotic liquid feeding with shock, acidosis, sepsis and a catecholamine orgy will surely make some difference to these carefully measured time intervals. Which is handly, as it bring us to:
Some of these will appear familiar because some of the same factors also affect the rate of intestinal peristalsis; others will be new, because they are unique to the stomach.
|Enhance gastric empyting||Slow gastric emptying|
Question 8 from the second paper of 2011 asked for factors that prevent gastro-oesophageal reflux. This seems to be mapped to the "...including sphincter function" part of the curriculum document, but of course the sphincter is only one of the multiple factors working to prevent you from wearing your stomach contents. Paterson (2001) is probably the most detailed overview of these factors, but unfortunately issues of Chest Surgery Clinics of North America from that era are not available in soft copy for anybody to look at. Fortunately, it was possible to cobble together the following list of factors from other resources: