Determinants of contractility

This chapter is relevant to Section G2(ii)  of the 2017 CICM Primary Syllabus, which asks the exam candidate to "define the components and determinants of cardiac output". Specifically, this chapter focuses on cardiac contractility, the forgotten and ignored element. Unlike afterload and preload, contractility has only ever come up once in the exams, and that was in Question 4 from the second paper of 2012. "Briefly describe dP/dT, the end systolic pressure volume (ESPV) relationship and the ejection fraction (EF)", they asked. And "define myocardial contractility". 

The pass rate was 13.6%.

Without descending to the lowest level of discourse, after a short break and some deep breaths, the author was able to calmly acknowledge that, though there is no fair way to test the trainee's knowledge of a definition that does not exist, it is still probably sensible to determine whether they understand the surrounding concepts well enough to recombine them during a short answer question. That acts as an IQ test of sorts, equivalent to an exercise in mental rotation. If you can synthesise a passable definition over that short a timeframe, you must have some substantial background physiology knowledge, and the sort of confident quick-thinking chutspah that would be valuable on the ICU floor. 

In summary:

 

There are few good resources to help the reader through this problematic topic. Muir & Hamlin (2020) present a superb bird's eye view of the main problems facing anybody who is trying to define and quantify cardiac contractility. For the author, this had value by the same mechanism as visiting a support group, in that it normalised the feelings of frustration and confusion as a natural reaction to the subject matter. We are all here for the same reason, the other authors seemed to say. 

Definition of contractility

Undeterred by the lack of scientific consensus on the matter, the examiners appear to give their own definition for contractility in the surprisingly comprehensive notes for Question 4 from the second paper of 2012:

"Contractility represents the performance of the heart at a given preload and afterload. It is 
the change in peak isometric force (isovolumic pressure) at a given initial fibre length (end 
diastolic volume)."

This definition has its origin in Berne & Levy (p. 250 of the 4th edition), in the sense that it was plagiarised from there verbatim:

"Contractility represents the performance of the heart at a given preload and afterload and at constant heart rate. Contractility may be determined experimentally as the change in peak isometric force (isovolumic pressure) at a given initial fibre length (end-diastolic volume)"

And you know this is an official definition because in the original textbook it appears in shouty all-caps. Though Berne & Levy is not on the Official Reading List for CICM Part One, the contractility entry in Pappano & Weir (p.78 of the 10th edition) is identical, a direct copy and paste. So, either this definition is particularly good, or the editors are particularly lazy. In either case, it appears the trainees need to memorise this specific definition for the primary.

As with everything else in the baffling hell dimension of cardiac output physiology, contractility has several other definitions, none of which are clearly superior to one another. Vincent & Hall (2012) give us this:

"Cardiac contractility can be defined as the tension developed and velocity of shortening (i.e., the “strength” of contraction) of myocardial fibers at a given preload and afterload. It represents a unique and intrinsic ability of cardiac muscle to generate a force that is independent of any load or stretch applied."

Stunned by the lack of progress in this area in the last two hundred years, Muir & Hamlin (2020) retreated into etymology:

"Literally defined the term contract infers that something has become smaller, shrunk or shortened. The addition of the suffix “ility” implies the quality of this process."

Cardiovascular Hemodynamics by Anvaruddin et al (2013), in an excellent sidestep of the question, instead decided to define contractility in terms of what it's not: 

"Contractility describes the factors other than heart rate, preload, and afterload that are responsible for changes in myocardial performance."

This definition also appears in Part One, which places it on the next uppermost pedestal just below the CICM examiners' definition.  Anyway, one could clearly continue with this sarcastic autopsy of textbooks for some more paragraphs, but the growing pile of mangled definitions would bring no further satisfaction to the author, and certainly no additional understanding to the reader. Beyond committing the CICM examiners' definition to memory, no useful recommendations can be made.

Determinants of contractility

After trawling the literature, it became clear that Penefsky (1994) is the single most useful resource on this subject, as all the parameters which affect afterload are laid out in a logical pattern by the author. A clear effort is made to produce some sort of conceptual union between the macroscopic factors which influence cardiovascular system performance as a whole and the microscopic factors which influence the performance of cell preparations.  

Properties of the cardiovascular system which affect contractility are:

• Preload
• Afterload
• Heart rate

Biochemical and cellular factors which affect contractility are:

• Calcium concentration 
  • Catecholamines and the autonomic nervous system
  • ATP availability (eg. ischaemia)
  • Extracellular calcium
• Temperature

Effects of preload on contractility

Preload is a major determinant of contraction. The degree of sarcomere stretch at the very end of diastole is an important factor in determining the force of contraction, as we might recall from the Frank-Starling relationship. The more volume, the greater the force of contraction, until beyond a certain point the sarcomeres stretch becomes too 

But that's the force of contraction. What about contractility, the "quality of this process" of contraction?  That changes too, in a predictable pattern. Volume loading (a fluid bolus of around 250-600ml of Hartmann's) increased the contractility of dog ventricles in a study by Mahler et al (1975) by about 11% (it was measured by dP/dT, which is discussed later). 

However, that is not the most interesting or exam-point-scoring element of this. Changes in contractility change the relationship between ventricular pressure and ventricular volume.  And at this point we are compelled to discuss LV pressure-volume loops.

The pressure-volume loop as a narrative device

For the explanation of the relationship between contractility and preload, the use of pressure-volume loops is made inevitable by some of the statements made by college examiners. They started with something rather noncommittal like "a diagram of a pressure volume loop is very helpful when describing the ESPV",  but finished with an aggressive warning that "absence of a diagram (correctly labelled and scaled) was a weakness in many answers". In short, you clearly need this diagram for your answer to score highly.  When it is correctly labelled and scaled, the LV pressure-volume loop looks a bit like this:

Without preempting the contents of the entire PV loop chapter, the discussion of PV loops here will mainly focus on their use for describing contractility, and in particular its changes with preload and afterload. 

End-systolic pressure-volume relationship (ESPVR)

The specific use of the PV loop in the discussion of cardiac contractility is for the purpose of describing the change in end-systolic pressure with increasing end-diastolic volume. This relationship, abbreviated as ESPVR, describes the maximal end-systolic pressure which can be achieved with that volume. 

How does this factor into contractility? Well: 

• As you increase the preload (here represented by the end-diastolic volume), the blood pressure increases.
• Both the systolic blood pressure and the diastolic pressure increase.
• Thus, the aortic valve closes at a higher pressure
• This higher pressure at the end of systole means the end-systolic volume is also higher
• Thus, the end-systolic pressure and volume point (and the rest of the  loop) is shifted to the right

Thus, if you were to plot the loop several times at different end-diastolic volume conditions, the end-systolic pressure-volume point would migrate northeast:

The relationship of these end-systolic pressure-volume points can be plotted as a line, which is the end-systolic pressure-volume relationship (ESPVR):

So... good story, but again, how does this integrate into a discussion of contractility?

Like so:

The more "contractile" the ventricle, the greater the change in pressure from a given level of preload. Ergo, the slope of the ESPVR line describes contractility, or at least how contractility affects the response to changes in LV volume.   

A reader well acquainted with Deranged Physiology traditions will at this stage be wondering when the author will try to support this theory by dredging up the experimental results of some abominable vivisection. So, here is a recording of pressure-volume loops at different ventricular volumes from Kass et al (1986), who captured these data from dog ventricles. One set demonstrates the effects of autonomic blockade (with hexamethonium chloride), and the other demonstrates the effects of dobutamine.

So, ESPVR seems like a good surrogate measure of contractility. However, it is not perfect:

• ESPVR slope decreases progressively as ventricular size increases without this change necessarily indicating a change in contractility (Nakano et al, 1990)
• Direct in-vivo measurement of this is frustrated by the fact that, during the preload challenge, in response to increased cardiac output baroreceptors and stretch receptors would decrease heart rate and therefore contractility, obscuring the true relationship

Of course, the ESPVR is not the only way of representing contractility. A multitude of other methods is made possible by the lack of agreed-upon definition. This segues nicely into...

Measures of contractility

Yes, there are several. The most common ones are:

• ESPVR, as discussed above; the relationship of contractility to the effect of preload on LV end-systolic pressure. 
• Ejection fraction,  the relationship of stroke volume to the end-diastolic volume expressed as a percentage. It is basically SV/EDV ×100.
• Myocardial strain (Abraham & Nishimura, 2001)
• Mean velocity of fibre shortening (Vcfc; Karliner et al, 1971)
• dP/dT, the maximum rate of change in LV pressure, which is the topic of the next heading: 

dP/dT as a measure of contractility

The dP/dT (or ΔP/ΔT) is a change in pressure per unit time. Specifically, in this setting, it is the maximum rate of change in left ventricular pressure during isovolumetric contraction:

This is not bad, as far as measures of contractility go. A more "contractile" ventricle should contract better (harder, faster, stronger) and this parameter will reflect that in a shorter isovolumetric contraction, or a higher pressure achieved over the same timeframe. Ditto, the feeble useless ventricle will take longer to achieve a lower pressure, so it goes:

 

Obviously, contractility is not that simple, and this parameter has its drawbacks. Borrowing from Mason (1969):

• dP/dT is affected by alterations of arterial diastolic pressure, i.e. a raised diastolic pressure results in a rise of peak dP/dT.
• dP/dT is dependent on the heart rate, which means it is impossible to assess the effects of an inotrope if it also has chronotropic effects.

So, dP/dT is influenced by some major haemodynamic parameters, which are difficult to control for. It is far from perfect, and probably the kindest thing that can be said about it is that it "changes in max. dp/dt can and frequently do reflect changes in myocardial contractility" (Wallace et al, 1963).

In their answer to Question 4 from the second paper of 2012, the examiners mentioned that this parameter is preload dependant and afterload independent. Where does this assertion come from? Well, it appears to be a logical outcome of using isovolumetric contraction as the dT period. Consider: most definitions of afterload involve aortic pressure to some degree or another (or,  they assert that afterload is aortic pressure). However, during the period of isovolumetric contraction, the aortic valve remains closed. So, they argue, how can dP/dT be affected by afterload, if it is observed before afterload has its effect on the LV?

This line of reasoning is somewhat suspicious. First of all, aortic diastolic pressure is definitely a factor which affects dP/dT, and it is certainly related to afterload. Also one must take into account that dP/dT_max (i.e. the maximum slope of the curve, the steepest tangent) might be observed at some stage after the aortic valve opens. 

So, what is the experimental evidence? To test these ideas, Quiñones et al (1976), because it was 1976, were able to convince elective outpatients to have huge boluses of angiotensin. Peak wall stress was increased by 44%, but the dP/dT barely budged (the change was 2.5%). Similarly, Kass et al (1987) found that dP/dT did not vary much over a range of high afterload values, becoming afterload-dependent only where afterload was extremely low (i.e. where the aortic diastolic pressure was so low that the maximum dP/dT value was observed long after the aortic valve opened). In summary, it is fair to say that within a normal range of afterload values, dP/dT should be relatively afterload-independent. Which is going to be something of a problem for its quality as a measure of contractility, as contractility is clearly affected by afterload.

Effects of afterload on contractility (Anrep effect)

Afterload affects contractility. It is a known thing. Gleb von Anrep detected this in 1912 after he clamped a dog aorta, even though he had no idea what he was looking at. The heart, with an abrupt increase of afterload, markedly and immediately increased its force of contraction - and then gradually, even more so, over the subsequent minutes. Here's a recording of what that looks like, made by Cingolani et al (2013) from a rat papillary muscle they were torturing:

The mechanism behind the abrupt phase of the increase is pure Frank-Starling:

• The increased afterload causes an increased end-systolic volume
• This increases the sarcomere stretch 
• That leads to an increase in the force of contraction

Thereafter, a gradual creeping increase in the intracellular calcium occurs, driven mainly by neurohormonal influences. Cingolani et al (2013) go through it in much more detail than even a patient reader would tolerate. In a nutshell there is an increased activity of the  Na⁺/Ca²⁺ exchanger because of an aldosterone-related uptake of intracellular sodium, and this is supported by the fact that this increase in contractility was totally blocked by eplerenone. 

Effects of heart rate on contractility (Bowditch effect)

Eminent authors have also called this the Treppe phenomenon, the staircase phenomenon (treppe being the German word for staircase), and frequency-dependent activation. As with the Anrep effect, it all boils down to having more calcium in the myocytes, which is the final common pathway for all increases in contractility. At a fundamental level, the mechanism is as follows:

• Myocyte contraction is the consequence of significant calcium influx into the myocytes
• Relaxation is mainly due to this calcium being ejected back out of the cell, or re-sequestered into the sarcolemma
• This expulsion of calcium is a chemical process with a finite reaction time
• Ergo, with increased heart rate, the time left for the removal of calcium is decreased
• Ergo, the residual calcium will increase the contractility of the myocytes wherever a high heart rate is sustained.

To some extent, the mechanisms which enhance relaxation with increased heart rate also help the calcium removal, but these are fighting against the fact that intracellular calcium modulates itself (eg. release of calcium from the sarcolemma is triggered by intracellular calcium). 

So, how fast do you have to go, to produce a sizeable Bowditch effect in your myocytes? To produce nice-looking publishable effect sizes, investigators usually have to crank up the heart rate. Here, Haizlip et al paced the rabbit ventricle fibres at a rate of 240 to produce a satisfying increase in generated force:

At this stage, the reader may point out that any increase in contractility which depends on a preposterous heart rate must surely be offset by the complete failure of diastolic filling produced by such a rate. Recall the cruel studies on syncope-prone volunteers which produced stroke volumes of 20ml and systolic pressures of 50 mmHg at a rate of 200. In short, though this effect is a known phenomenon and needs to be discussed in an exam setting, most reasonable people will acknowledge that it has minimal utility at the bedside.

Woodworth effect

This probably deserves a mention as well, as it is another version of the staircase phenomenon - or rather, it is whatever the opposite of a staircase is. Essentially, this effect describes the positive inotropic effect of a prolonged period between contractions - "the recuperative effect of a long pause", to borrow words from Woodworth himself (1902). Here's an illustrative diagram from the original Woodworth paper, labelled with the relevant effect.

Yes, that's all there is: a higher than normal systolic peak following a period of tachycardia.  Again, this is calcium related. By washing the muscle fibres in a calcium-free solution, the effect was completely abolished by Hajdu (1969).

Some authors also seem to attribute the name "Woodworth effect" to the observation that bradycardia increases the apparent force of contraction, but this may in fact be merely the effect of better preload. There are very few mentions of this phenomenon in modern literature.

The influence of calcium on contractility

Its central role in excitation-contraction coupling makes intracellular calcium the final common pathway for the activity of most inotrope drugs and physiological factors which influence contractility. It is basically the lever you pull when you want to modify contractility in one way or another. The basis for its central role in this process is discussed elsewhere; for an instant overview one may be directed to Eisener et al (2017). In brief:

• Calcium entry into cardiac myocytes triggers a calcium-dependent release of calcium from the sarcolemma
• Calcium binds to troponin resulting in sliding of the thick and thin filaments
• Force of contraction depends on the amount of calcium bound to troponin
• Therefore, the major factor that regulates the force of contraction is the level of intracellular calcium

From this, it follows that intracellular calcium concentration is a determinant of contractility. That's a fairly indirect thing to discuss, as we don't normally measure it, or titrate our interventions to it, or really think about it in any meaningful sense. And yet, it is there. Any discussion of cardiac contractility would need to include the contribution of calcium, and the factors which modify it. Which are:

Catecholamines. The inotropic effects of systemic catecholamines and of the sympathetic nervous system is mediated by the β-1 receptors, which are G_s-protein coupled receptors. The increase in cyclic AMP which results from their activation increases the activity of protein kinase A, which in turn phosphorylates calcium channels. Calcium influx ensues. Sperelakis (1990) and Rüegg (1998) put together give more detail than most people would be able to handle, when it comes to this aspect.

Ischaemia. Though the depletion of ATP which is expected to occur in the absence of oxygen is a convenient mechanism to blame for the ischaemia-associated decrease in contractility, in actual fact the amount of ATP in acute ischaemic cells is not reduced for a while, whereas the contractility suffers immediately. This impairment of contractile function is thought to be due to a decrease in the ability of intracellular calcium to trigger the release of more calcium from the sarcolemma (Gomez et al, 2001).

Extracellular calcium. That calcium - when it pours into the cell during the action potential - has to come from somewhere. Bathing the cells in a fluid devoid of calcium is a sure way to abolish all contraction. Lang et al (1988) dialysed seven chronic renal failure patients to achieve different serum calcium levels and were able to demonstrate that the Vcfc (their chosen measure of contractility) declined significantly with hypocalcemia. In fact, the relationship between the calcium levels and contractility appeared to be linear, over the ethically permissible range of calcium concentrations.

The influence of temperature on contractility

Moderate hypothermia (32-38º C) impairs contractility, and there is a well known temperature-proportional decrease in cardiac output. One might think of this as something to do with the catecholamine receptors losing their affinity (and to be fair, they do) but there are also other factors at play. Specifically, hypothermia causes decreased cardiac myofilament sensitivity to calcium (Han et al, 2010) and the activity of cardiac actin-activated myosin ATPase decreases (de Tombe et al, 1990).