Normal Cardiovascular Physiology

From WickedSim

This page has two objectives:

It's an introduction to the basic physiology of the human cardiovascular system It might be used as a basis on which to construct sufficient knowledge of the cardiovascular system to pass physiology exams. (But be careful --- it's not comprehensive, nor is it eror free :-) This tutorial assumes: A basic knowledge of cardiovascular anatomy; At some time in the past you either did a basic course in physics or a basic course in physiology; If you are unfamiliar with words like proximal, distal, ventricle and arteriole this web page might not be for you; likewise if the equation "V = I * R" means little to you. In the first section we explore some fundamental concepts. Despite the heart being cast down from its role as the seat of human emotion to a mere pump, we may still sometimes attach inappropriate significance to this muscular organ. Perhaps it is because the heart is such a source of trouble, and heart problems will kill many people reading this page, that we retain a mystical belief in its power. We may still falsely believe that the heart controls the flow of blood to the tissues, while nothing could be further from the truth! Now read on ...

Contents 1 Basic Concepts 1.1 Pressure 1.1.1 Blood is thicker than water (but only just)! 1.1.2 X marks the spot 1.2 Compliance 1.3 Resistance (R = V/I) 1.3.1 The heart impedes flow! 1.4 Cardiac output (CO = HR * SV) 2 The heart is complex 2.1 Let's stop the heart 2.1.1 Implications of the mean systemic pressure 2.1.2 How does the body maintain blood pressure homeostasis? 2.1.3 Local differences 2.1.4 Two pumps don't go 'Bang'! 2.2 Intrinsic control of the heart 2.2.1 Frank-Starling curves 2.2.2 Postextrasystolic potentiation 2.2.3 Ventricular function curves 2.3 Preload, afterload 2.4 Determinants of myocardial contractility 2.5 Venous function curves 3 Autonomic and neuroendocrine control 3.1 Nervous control 3.2 Hormonal and other mechanisms 3.3 Starling forces 3.3.1 The third Starling 4 The kidney steps in --- hypertension explained 5 Myocardial oxygen demand

Basic Concepts

The heart is merely there to serve. Normally, the amount of blood put out by the heart depends on the amount of blood required by the tissues. And that is that. You might therefore conclude that "The heart sucks", but it probably can't even do this very well. There are numerous claims that the left ventricle generates a negative suction which aids filling^{[1][2][3][4][5]} but the heart appears to depend mainly on passive filling. Only when the heart fails will it start to call the shots, and then only through inadequacy rather than dominance!

In order to understand flow of blood through the circulation, it's important to grasp several basic concepts. These include:

• Pressure;
• Resistance;
• Compliance.

We'll deal with pressure first, as it's fundamental to what follows.

References

1. ↑ Am J Physiol Heart Circ Physiol 251: H47-H55, 1986
2. ↑ Basic Res Cardiol. 1998;93 Suppl 1:143-7
3. ↑ Am Heart J. 1994 Jan;127(1):143-7
4. ↑ Am J Physiol Heart Circ Physiol. 2006 Dec;291(6):H3114-21
5. ↑ Circulation. 2005 Nov 8;112(19):2921-9.

Pressure

If you're familiar with the basics of pressure, and understand the difference between a scalar and a vector, you might wish to skip to the next section. Otherwise, first read our page on Pressure, from which you can return to the current page.

Blood is thicker than water (but only just)!

Let's look at a practical example of pressure measurement. We're taking the blood pressure in somebody's brachial artery using a mercury sphygmomanometer. We determine that the pressure in the brachial artery at an arbitrary point (We use as our arbitrary reference point x the heart, more specifically the tricuspid valve). Say the systolic blood pressure measured in this way is 126 mmHg. What pressure will we measure if we elevate the brachial artery 10 cm above our reference point x, the 'level of the heart'?

Clearly the pressure will drop, but by how much? We know that the change in pressure is given by:

P = h * density * g

... but using this equation is tedious. We can convert from cm of blood to mmHg if we know the density of blood. It turns out that blood is slightly more dense than water (Blood plasma at a temperature of 37C has a density of 1025 kg.m-3 as opposed to 993.7 kg.m-3 for water, but remember that in the non-anaemic individual about 40% of blood is red blood cells with a density of nearly 1125 kg.m-3!). So to convert from cm of blood to mmHg, we divide not by 13.6 but by about 12.8. So 10 cm of blood is about 7.8 mmHg.

The bottom line is that if we elevate the arm by 10cm, we will under-estimate the blood pressure by nearly 8 mmHg; we might compensate for this by adding that number to the measured pressure, but it's smarter to simply keep the arm at the level of the heart when we measure the pressure!

X marks the spot

Why use the level of the heart as our arbitrary reference point? Apart from the convenient rather central location of the heart, there are two other reasons to use the heart as or reference point. These are:

1. It works! Arthur Guyton took dogs, moved them every which way, and measured pressures at various points, and found that a useful reference point is the right ventricular inflow (the level of the tricuspid valve).
2. As we have already mentioned, the heart is a slave to the pressure that fills it. It makes some sense to relate all pressures to that filling pressure --- the pressure at the right ventricular inflow tract.

Compliance

Consider a balloon containing a pressure transducer. We add a known volume of fluid, and measure what happens to the pressure. Let's define the compliance as:

 Compliance = (change in volume) / (change in pressure)

If the compliance is low, a small added volume will be accompanied by a large change in pressure; conversely with a high compliance you need to add a lot of volume to obtain such a large rise in pressure. You can also deduce that in a highly compliant system, adding volume will not result in a dramatic increase in pressure, and flow might suffer in consequence.

Resistance (R = V/I)

Resistance (R) is defined as pressure divided by flow, whether it's the classical definition of resistance in an electrical circuit (R=V/I), or the more familiar cardiovascular relationship of:

 vascular resistance =  blood pressure(BP) / cardiac output(CO)

There's nothing wrong with re-arranging things. For example, if we're looking at mean arterial pressure (MAP), we might say:

 CO =  MAP / SVR

But we must be careful not to assume from the above that, for example:

(WRONG ASSUMPTION: "If svr goes up, CO will go down")

If we fiddle with SVR (for example, by giving a drug which constricts the vessels) that CO will not necessarily change. More likely will be an increase in blood pressure, and an unchanged CO, as what the tissues really need is an appropriate cardiac output. The cardiovascular system is just that, a system! Simply because we define a relationship at a particular point at a particular time does not imply that we can fiddle with a particular variable like SVR and deduce what will happen next. We'll frequently encounter the fallacy of assuming that a definition can allow us to predict the behaviour of a system.

The major determinant of resistance is local tissue autoregulation. Tissues dictate the amount of blood flowing to them, and the heart provides a head of presssure, and is responsive to reasonable demands by the tissues.

Other factors apart from tissue autoregulation influence flow. The properties of the blood cannot be ignored. Blood is viscous, and the properties of the erythrocyte are particularly important --- it's been said that if erythrocytes weren't deformable, then blood would have the approximate viscosity of a brick! Depending on dimensions of tubes being perfused, several tricky and complex phenomena come into play (notably axial streaming, plasma skimming, and the Fahraeus Lindquist phenomenon). The Fahraeus Lindquist phenomenon occurs in small vessels, where viscosity seems to drop as erythrocytes tend to move near the centre of vessels.

Blood rheology is impaired in a variety of states, including sepsis (Intensive Care Med. 2003 Jul;29(7):1052-61) and poorly-controlled diabetes mellitus, but measurement of rheology still doesn't seem to have moved into the mainstream.

The heart impedes flow!

We've identified two factors which influence flow through the cardiovascular system. One was compliance, the other peripheral resistance to flow. As you might expect, the heart isn't innocent when it comes to affecting flow --- it generates the pressure differential between its input and its output needed to pump blood, but two heart properties limit flow in a significant fashion:

1. Ventricular compliance ('how much the ventricle will fill');
2. Inertia of blood flow to the ventricles.

The presence of the atria tends to mitigate the above effects --- the atria act as a buffer for storage of blood which then enters the ventricles at the next beat of the heart. In normal sinus rhythm, of course, the atria act as booster pumps, contributing perhaps 30% of the filling of the ventricle, but this effect is tiny when compared to their effects on inertia of blood flow. Without this latter atrial effect, ventricular filling would decrease to about a quarter of its normal value [Anderson]. It turns out that normally the pressure generated in the atria is 'just right' to minimise effects on inertia and maximise flow into the ventricles.

Cardiac output (CO = HR * SV)

The above equation is just a relationship between cardiac output (CO), heart rate (HR) and stroke volume (SV). In a similar fashion to the preceding section where we discussed the equation defining resistance, we must not be tempted to draw false conclusions. The equation "CO = HR * SV" does not imply that if we tinker with the heart rate or stroke volume, that the cardiac output will change. Let's say we speed up the heartbeat. Will the cardiac output increase? Almost certainly not, unless the peripheral tissues demand more output. Most likely we will find that the stroke volume decreases, and the cardiac output remains unchanged! The heart is just a pump, a slave to the periphery, providing any reasonable demands.

The heart is complex

Just because the heart is a slave of the periphery, doesn't mean it's simple. Think for a moment about the way things are set up:

1. We actually have two pumps, the right heart and the left heart.
2. In between the two pumps we have two circulations, the pulmonary circulation and the systemic circulation.
3. Each circulation has both a resistance to blood flow (defined above), and a compliance (or 'capacitance'), the amount of change in pressure you'll see if you add a certain amount of volume to that circulation.
4. Time delays are important: if you change something at some point in time in one of the two circulations, there will be a delay before the influence of this change is felt in the other circulation.
5. In addition, there are non-linear relationships between the things we change in the circulations, and the responses we see. What do we mean by `non-linear'? Let's say we make a certain change, and get a certain response. What will happen if we double the change? In a linear system, the response will double, but in a non-linear system, the response won't necessarily be twice as great as the original response. The response might not even be in the same direction as the original response!
6. A lot of neurological and endocrine 'fine tuning' takes place.

All of the above conspire to make a very complex system, which we still don't fully understand.

Let's stop the heart

Just for the sake of experiment, let's stop the heart for a few seconds (as might happen if, say the pacemaker system in the heart failed completely, or we gave a great big whack of adenosine intravenously). What will happen? In anyone, pressure will take a little while to equilibrate throughout the system. What will the final pressure be (equal throughout the arteries, heart and veins)? The answer in most normal individuals is thought to be about 15--18 centimetres of water (just above 10 mm of mercury, we already know how to make the conversion). We'll call this pressure the mean systemic pressure, in other words, the pressure averaged over the whole cardiovascular system.

Okay, that was a fine experiment, now let's start things up again. What would happen if we could somehow magically drop the mean systemic pressure before we started things up again? Before we consider the results of such a change, think about how you might do this. You have just two options, really:

1. Suck out some blood;
2. Increase the capacitance of the system, so the mean pressure drops.

The two will have similar effects --- it's the balance between filling and capacitance which counts. Let's first consider the extreme, where we drop the mean systemic pressure to zero. It should be clear that at this point, cardiac output will also drop to zero. We know this, because we already know that the heart doesn't suck.

Implications of the mean systemic pressure

Since Arthur Guyton demonstrated its existence, the concept of mean systemic pressure seems to have been held in semi-religious awe, but the reason for the existence of mean systemic pressure is simple. The cardiovascular system has compliant elements, and the 'fullness' of the system needs to be balanced against this compliance. If the system is under-filled, then the system will behave like a flaccid balloon, and not only will pushing out blood have minimal effect on pressure, but filling of the ventricle won't occur. If the system is over-filled, then pumping out small amounts of blood will result in large increases in pressure; in addition, the converse of 'under filling' will occur, and the system might conceivably be driven to a wastefully high cardiac output!

You can see that there must be some magical 'optimum' degree of fullness, and that a wise organism will maintain this fullness very carefully. Let's look at this next.

How does the body maintain blood pressure homeostasis?

The body has multiple complex mechanisms for maintaining blood pressure. In the short term, a variety of neurological reflexes and endocrine mechanisms keep blood pressure about a set point, but in the long term the set point seems to be solely determined by the kidneys, which sense and maintain mean systemic pressure. Perhaps because humans tend to study things which are easily studied (and easily altered) physiologists in the past have tended to attach more importance to hormonal and nerve regulation of the cardiovascular system than these control systems deserve.

Local differences

From the complexity of the structure of the cardiovascular system, you can deduce that tweaking the system at different points will have different effects. Let's say for example that we place a balloon around a major blood vessel, and inflate the balloon to a certain pressure, increasing the resistance of that vessel. It should be barn-door obvious that a pressure of say 15 mmHg will do little to impede flow through the aorta, but the same maneuver applied to the inferior vena cava will in all likelihood stop blood flow from the lower half of the body completely, with dramatic effects.

Even more interesting is what happens when we apply a large resistance at different points in the arterial system. If we dramatically constrict the proximal aorta close to its origin from the left ventricle, the next heartbeat won't deliver much blood. Conversely, if we apply similar constriction to the distal aorta, the compliance of the intervening aorta will still allow a lot of blood to be ejected from the ventricle.

Two pumps don't go 'Bang'!

Think about this for a moment. We have two pumps in tandem, the left and right sides of the heart. Is there some magic in the fact that blood doesn't end up pooling in one of the two 'circulations'?

Consider two mechanical pumps set up in a similar fashion. Now imagine that the two pumps don't quite pump the same amount per minute. Let's say the difference is one millilitre per minute. Clearly blood would 'dam back' in the one circulation at the expense of the other, and eventually, something would burst! Why doesn't this happen in our circulatory systems?

The answer is that the human cardiovascular system is fundamentally different from mechanical pumps with which we are familiar. As we've mentioned once or twice before, the ventricle depends largely on passive filling. The pumps don't have a fixed flow rate. So if fluid dams back in (say) the pulmonary circulation, in other words, the volume increases in that circulation, then pressure will increase (as determined by the compliance of the system). We know that cardiac output depends on filling pressure, so increased pressure at the input to the next pump in series will cause that pump to pump more, and the balance will be restored!

There are other 'buffering' features:

1. The ventricles aren't maximally filled in diastole, so increased flow into the ventricle is easily accommodated;
2. There's some "wastage" of energy over and above that needed to eject blood! If energy demands increase, flow can increase in a compensatory fashion by diverting the 'wasted' energy to moving blood!

Intrinsic control of the heart

We've already mentioned that the heart is a slave to the peripheral circulation, dependent on filling for its output. Let's explore how this might be.

Frank-Starling curves

These refer to the physiologist Otto Frank and Ernest Henry Starling. Frank pipped Starling at the post, describing the mechanism almost two decades before Starling's 1915 Linacre lecture on 'the Law of the Heart'. Simply stated, the `law' is:

The greater the heart is filled during diastole, 
the greater will be the quantity of blood 
pumped into the aorta during systole.

The mechanism by which stretched heart muscle contracts with greater vigour is still argued about. Older theories talked about overlapping of actin and myosin; newer ones refer to more complex ideas which often involve calcium flux. (For an overview of calcium flux in all its splendour, see [J Pharmacol Sci. 2006;100(5):525-37]). The bottom line is that increases in sarcomere length alter calcium ion dependent activation ('length dependent activation'). The enormous protein titin/connectin seems important in the regulation of length dependent activation. All three isoforms of nitric oxide synthase are expressed in the heart, and eNOS and nNOS contribute to the Frank Starling mechanism. For exam purposes, it's worthwhile reading a good recent review of excitation-contraction coupling. Try this link from the Journal of Physiology (J Physiol. 2006 Mar 1;571(Pt 2):253-73). The central reason why the heart 'puts out what it gets in' is the Frank-Starling mechanism. It's important to note that even where the heart works against greater loads (for example, the blood vessels constrict, increasing the work that the heart must do to pump a given amount of blood), the cardiac output will be preserved until the load on the heart becomes really excessive. The Frank-Starling mechanism is supplemented by other responses. For example, about thirty seconds after stretching the heart muscle, its metabolism increases, resulting in increased contractile strength ('homeometric autoregulation'). Stretching the right atrium increases heart rate by up to 30%, this also increasing cardiac output. Both of these mechanisms will tend to return heart muscle fibre lengths to values close to the original ones. For any given heart, there is not one but many such curves. As we increase the volume delivered to the heart, output will go up, and we can create a curve of 'stretch' versus stroke volume, but if we vary the contractile state of the heart, then we'll create a different curve. On the right is a a good drawing of both a Frank-Starling curve and a set of curves (from http://www.cvphysiology.com/Cardiac%20Function/CF003.htm --- if your browser won't display the image, then visit the link)!

Note that some draw 'Starling curves' with a descending limb on the right, where it is assumed that stretch is so great that contractility becomes impaired. There is no substantial evidence for such curves, except where the experimental design has been poor.

The terms 'homeometric' and 'heterometric' can be confusing. We've already discovered one homeometric response (increased metabolism related to stretch) --- the name refers to the fact that different forces occur at the same length (homeo + metric = same length). There are several homeometric phenomena. The response to increased pressure loading is called the Anrep effect; increased contractility with increased rate is the Bowditch effect, also known as the Treppe (staircase) phenomenon.

The Frank-Starling mechanism is termed heterometric --- changes in force are related to differing lengths, as we've described.

Postextrasystolic potentiation

It makes sense that a premature ventricular contraction is weak, as the ventricle is likely to be under-filled. The subsequent beat is stronger, but not merely because of increased filling. In fact both the premature beat and the subsequent beat are influenced by intracellular flux of calcium ions, which an important determinant of the strong second beat (post-extrasystolic potentiation).

Ventricular function curves

(This topic should be discussed in detail on a separate page --- at different blood volumes the stroke work and end-diastolic pressure are plotted; this is repeated under different conditions).

Preload, afterload

These concepts are best explained by looking at the circumstances in which they were originally used --- the laboratory. Here goes ...

Preload Consider the strip of muscle shown in the figure on the right. We fix the bottom of the muscle strip, and attach the top to a lever arm. On the other side of the fulcrum we attach a weight. This weight is the preload. (We can now stimulate the muscle to contract).

Afterload Now look at the next figure. This is the same set-up, but after the initial weight (preload) was imposed, we put a stop above the lever arm to prevent further muscle stretch. We then add an extra weight. Now, in addition to the force required to move the preload, we have to generate extra force to move the added weight, which we term the afterload.

The analogy to a functioning heart is clear. The 'preload' represents the filling pressure, or stretch of the ventricle at the end of diastole. The 'afterload' is similar to the load imposed on the ejecting ventricle by factors which limit ejection --- aortic compliance and systemic resistance.

You can see that in the normal heart, preload is determined by the return of blood from the the peripheral tissues, modulated by the mean systemic pressure and the regulatory effects of the kidneys on this pressure. Afterload depends on effective arterial compliance and resistance, the former not subject to much variation, the latter adjusted by the needs of the tissues.

You might wish to speculate what happens to preload, afterload and contractility in disease states such as heart failure, hypertension and valvular heart disease.

Determinants of myocardial contractility

We will not here discuss these in any detail. The heart is just a pump which responds to the demands of the peripheral circulation. Even in heart failure, we have begun to realise that in the long term neuroendocrine responses (fluid retention, increased sympathetic outflow) are harmful, and that fiddling with the inotropy of the heart is almost always deleterious. Beta blockers improve survival, inotropic agents impair it. Even digoxin likely has its (possible, limited) beneficial effects on heart failure despite its inotropic effect, perhaps through modulating vagal tone!

Any standard textbook of pharmacology will provide details of drugs which affect myocardial contractility; any textbook of physiology will provide boring detail of 'factors influencing contractility'.

Venous function curves

If you go back and look at Guyton's original experiments, things are simple. The peripheral tissues demand what they need, and the heart supplies the relevant cardiac output. Several modern physiology textbooks 'turn things around' probably because they haven't read Guyton's work with sufficient care. They invent some sort of mythical 'back pressure' which limits 'venous return', as if the heart dictated things (It doesn't). Ignore this folly!

For a correct overview, see Brengelmann's work, for example (J Appl Physiol 94:849--859,2003). It's not easy. Here's Fig 1 from his article:

It's all too easy to FALSELY decide based on looking at this curve that venous return/cardiac output ('dependent variable' ??) is governed by right atrial pressure. More explicit is Brengelmann's Fig 8:

Now you can see the true relationship. Right atrial pressure is the dependent variable, and variation is made in cardiac output, resulting in a new equilibrium. There's a clean, easy and mnemonic way of conceptualising what happens here.

If we have a circuit, and it has two components, each with a particular capacitance, then if we shift blood from the first to the second component, then the pressure in the second component will go up and the pressure in the first component will go down. This is obvious.

Now let's apply this to Guyton's venous function curves. If we augment cardiac output, pumping more blood into the "arterial side", then the pressure on the venous side will diminish. And that's that. The whole concept of 'venous return' is a bit suspect, as at equilibrium the 'venous return' is identical to the cardiac output (it must be)!

Autonomic and neuroendocrine control

All of the fancy autonomic and endocrine reflexes which modulate blood pressure can simply be seen as 'fine tuning' to allow the organism to stand up and run. These reflexes may be more a hindrance than an asset in the recumbent, anaesthetised patient! Several examples support the above, apparently outrageous assertion. These include:

• The patient with a high spinal cord transection. Problems arise mainly related to upright posture and/or remaining reflexes (particular post-denervation up-regulation of receptors);
• Experimental, total sympathetic and parasympathetic paralysis where the healthy cardiovascular system ticks along just fine in the recumbent position.
• Subjects with profound autonomic neuropathies, whose major cardiovascular problems are often related to posture.

Nervous control

It's extraordinarily unusual for afflictions of the nervous control of the CVS to result in long-term hypertension, telling us quite uneqivocally that the holy grail of control of mean systemic pressure doesn't reside anywhere within the reflex neurological control of the CVS. Not so the kidney.

Here's a brief summary of autonomic control of blood pressure.

1. Baroreceptor responses don't care what the current mean arterial pressure is, as long as it hasn't changed. They will accept whatever the current mean systemic pressure is, and adapt accordingly.
2. The baroreceptor reflex senses stretch in the large arteries (especially the carotids and aortic arch) when the arterial pressure is above 60 mm Hg (maxing out at about 180 mm Hg). The response to such stretch is quick (even varying with the cardiac cycle) --- sympathetic outflow to arterioles decreases resulting in less vasoconstriction, and vagal tone increases slowing the heart. In addition (and even more important) changes in sympathetic control of veins result in increased venous capacitance, a powerful control mechanism. Signals flow through the vasomotor centre in the medulla. The baroreceptor reflex is not important in long term regulation of blood pressure, as it rapidly adapts to the prevailing arterial pressure.
3. Chemoreceptors in the carotid and aortic bodies have a minor effect on the vasomotor centre.

Stretch receptors in the atria and pulmonary arteries modulate systemic arterial pressure through a variety of minor reflexes;

1. In severe disease states, diminished blood supply to the brain can set in motion the powerful `CNS ischaemic response', an emergency response system to very low CNS perfusion pressures (under 50 mm Hg).
2. Contraction of skeletal muscle pumps blood to the heart by compressing the abdomen and (during exercise) by compression of blood vessels throughout the body.
3. The 'Bezold Jarisch' response is probably mediated through receptors in the base of the heart: excessive stretch (or stimulation by e.g. ischaemia) results in inappropriate bradycardia. This reflex (otherwise known as the 'veratrinic response') may be important in the pathogenesis of syncope in many conditions.

Hormonal and other mechanisms

These act fairly quickly, and interact with the nervous control mechanisms. Mechanisms include:

1. The well-known renin-angiotensin-aldosterone system;
2. vasopressin (which may help in e.g. shocked states);
3. Atrial natriuretic peptide. It's wise to know something about this, particularly for examination purposes, but as a drug, it's rubbish.
4. Alterations in pressure at a capillary level, altering Starling forces (Starling again) across the capillary, and resulting in changes in fluid movement out of the intravascular compartment!
5. A vast array of other hormonal influences --- catecholamines, autacoids, kinins and so forth.

Starling forces

We will not discuss these mechanisms in detail here. For examination purposes it is vital to know the Starling forces which are said to balance movement of fluid across capillaries, including concepts such as reflectance.

The third Starling

By the way, there's another 'Starling' concept. The third 'Starling' is the Starling resistor. This mechanism is operative in the lungs in the upright subject --- pressure in the normal pulmonary circulation isn't quite adequate to perfuse the apices of the lungs, and the vessels start closing at a critical perfusion pressure.

The kidney steps in --- hypertension explained

It should be obvious that long-term control of mean systemic pressure (MSP) resides within the kidney. Renal dysfunction often results in hypertension, and other causes of hypertension can always be demonstrated to influence the behaviour of the kidney. Examples include Cushing's syndrome, Conn's syndrome, liquorice excess, primary hyperreninaemia, coarctation, and many experimental models where renal damage and salt or mineralocorticoid excess result in hypertension. Fiddling with other 'control systems' doesn't have a similar effect (although there have been recent arguments that baroreceptors can influence long-term blood pressure control). In order to understand hypertension, we need to understand the subtleties of renal control of MSP. Central to this role is salt and water retention.

There is widespread belief that 'the problem' with most cases of hypertension (where we have failed to establish the cause, so-called 'essential' hypertension) is a primary increase in systemic vascular resistance (SVR), because in chronic hypertension, the SVR is often raised. Looking at the cardiovascular system from a systems point of view, this confusion of association and causation becomes very silly. The peripheral tissues control cardiac output. If mean systemic pressure increases, then the kidneys will try to lower this pressure by getting rid of fluid and salt. If this renal mechanism is impaired, then it becomes very difficult to tease out the commensurate increase in pressure and resistance required to keep cardiac output pretty much the same, as demanded by the tissues. As Guyton has argued, total peripheral resistance is a dependent variable! In 1966, Guyton pointed out that the normal kidney mechanism for getting rid of excess salt and water has infinite gain --- it restores pressures to exactly the previous value. If we are to understand 'essential' hypertension, we must fully understand this feedback mechanism, and the (presumably multiple) ways that it can become impaired. A primary increase in total peripheral resistance does not result in hypertension unless the kidney's ability to regulate volume is also impaired. At present we simply do not know what the renal sensors are for volume regulation, although there are many theories.

Physiologists and hypertension experts have all focused in a monomaniac fashion on the sodium ion. This may not be wise, as the chloride ion is at least as important, and (although the few studies of its role are contradictory) it might even be argued that the culprit in hypertension is chloride. Many of the studies are flawed, as lacking a decent model of what is happening, experimenters have varied ion intake without properly taking into consideration factors like pH. If for example we administer sodium salts like sodium citrate, we are providing a net addition of sodium ions and promoting alkalosis. If we give an organic chloride salt (with e.g. glycine) we are doing the reverse. Acidosis or alkalosis has multiple effects on exchange of sodium, potassium and hydrogen ions. We must also remember that the levels of salt intake in modern 'Westernised' societies are an order of magnitude greater than salt intake in the preceding million or so years. On really low salt diets, our ancestors almost certainly did not develop hypertension.

Myocardial oxygen demand

The bottom line here is that the heart takes what it needs. For intricate detail, read a physiology book; the following is a brief precis.

In the resting individual, one hundred grams of heart muscle consumes about 9 ml of oxygen per minute, but with exercise this number can increase dramatically, related to increases in coronary blood flow (not increased extraction of oxygen). Coronary blood flow occurs mainly in diastole, at least in the left ventricle. The heart burns a variety of substrates including glucose, lactate, and fatty acids, the last mentioned accounting for about two thirds of myocardial energy production.

The heart isn't very efficient --- about 10--15% on average. You might think that stroke work is the major determinant of myocardial energy expenditure, but this isn't the case. Stroke work is determined by the area under the pressure-volume curve, but even with identical areas, energy expenditure is far greater where the heart has to pump against higher pressures. This is because tension generated accounts for a lot of energy expenditure.

Wall tension depends on the radius of a sphere, based on the well-known Laplace's law (Pressure generated is equal to twice the wall tension, divided by the radius). As radius increases, to generate the same pressure we need to generate an equally great increase in tension, with consequent increases in energy expenditure. You can see that the failing heart becomes increasingly disadvantaged as radius increases, and energy expenditure goes up.

(needs more work here)

---