This chapter is most relevant to Section F5(i) from the 2017 CICM Primary Syllabus, which expects the exam candidates to be able to "describe the anatomical and physiological features of the pulmonary circulation". This represents a minor duplication of syllabus material, as the anatomy of the pulmonary and bronchial circulations has already been covered under section F1(iv). Ergo, this chapter will be concerned purely with the physiological features of the pulmonary blood vessels. All effort will be made to ensure that the pulmonary veins are not left behind as the less interesting half of the circulation; but of course the pulmonary arteries are a much sexier topic for exam purposes, and the time-poor candidate will be better off focusing on these.
- The pulmonary circulation:
- A low pressure, highly elastic system, with vessel walls which are much thinner and less muscular than the systemic circuit
- At any given time, contains ~10% of the circulating blood volume, or about 500ml.
- Pulmonary blood volume varies by around 50ml over the course of a single cardiac cycle
- Pulmonary blood flow:
- Pulmonary arterial flow is equal to the cardiac output and consists of mixed venous blood
- Pulmonary venous flow consists of oxygenated pulonary capillary blood as well as physiological shunt (blood from thebesian veins and bronchial veins)
- Red blood cells spend about 4-5 seconds in the pulmonary circulation,
- Pulmonary arterial pressure:
- Normal PA systolic pressure = 18-25 mmHg
- Normal PA diastolic pressure = 8-15 mmHg
- Normal mean pulmonary arterial pressure = 9-16 mmHg
- The pulmonary arterial waveform resembles the systemic arterial waveform, and has similar features
- Pulmonary capillary pressure:
- Pulmonary capillary pressure is usually about 8-10mmHg
- It is usually a little higher in the arterial (pre-alveolar) capillaries
- It can be estimated from the Gaar equation: PPC = LAP + 0.4 × (mPAP - LAP)
- Pulmonary venous pressure:
- The mean pressure in the pulmonary venous circulation is usually 6-12 mm Hg
- The pulmonary venous pressure waveform resembles the CVP waveform
- This waveform can be seen during pulmonary artery wedge occlusion
- Pulmonary vascular resistance:
- The resistance in the pulmonary circulation can be calculated as follows:
- PVR = 80 × (mPAP- PAOP)/CO
- The normal value for PVR is 100-200 dynes/sec/cm-5, or 255 - 285 dynes-sec/cm–5/m2 for PRVI (indexed to boy surface area)
- These values are approximately 1/10th of what one might expect from the systemic circulation.
Naeije et al (2013) and the older Waaler et al (1971) were the main resources used to construct this chapter, and are well-structured revisions of the major topics, worth reading but unfortunately paywalled. If one starts paying for things like that, one identifies oneself as a reader with infinite resources, in which case one can also splurge on Alfred Fishman's "The pulmonary circulation" (JAMA, 1978- an excellent narrative overview), or the ancient grimoire by Domingo Aviado (1965). Or one might like to read John B West's rambling discourse on the evolution of the pulmonary circulation. The only freely available peer-reviewed article found under the search term "physiology of the pulmonary circulation" is Hamilton & Augusta (1951), which is written remarkably well but which is hopelessly out of date. None of these is going to help you pass the primary exams, of course. You may as well just buy Nunn's.
Pulmonary blood volume
The lung receives all of the cardiac output, which means this 500-gram organ ends up flooded with ten times its own weight in blood, every minute; and at any given time it is home to about 10% of the circulating blood volume. This figure (500ml or so) and the evidence behind it are dealt with in greater detail elsewhere, in the chapter on the non-respiratory function of the lung where its role as a reservoir of blood is discussed. Here, it will suffice to say that:
- Pulmonary blood volume is around 400-500ml
- It varies over the course of the respiratory and cardiac cycle
- It varies in response to gravity (it is lower in the erect position, by one-third according to Nunn's)
- It increases in states of sympathetic stimulation (as systemic vasoconstriction "flushes" blood into the pulmonary circulation)
Pulmonary arterial pressure and blood flow
Because the lung receives all of the cardiac output, it is obviously very important for the pulmonary arteries to do as little as possible to hinder this flow, and so the pulmonary vascular resistance is quite low. As a result, pulmonary arterial pressure is also low, with a normal systolic maximum of about 18-25 mmHg, and a mean arterial pressure somewhere in the range of 9-16 mmHg. Any increase in the mean pressure beyond 25 mmHg is described as pulmonary arterial hypertension.
Here's a nice freshly recorded pulmonary arterial pressure waveform, albeit a little higher than one might normally expect (understandably, this being a cardiothoracic ICU patient). On monitors around the world, one expects to see this waveform is conventionally rendered in yellow, perhaps because pulmonary artery catheters are conventionally yellow, or perhaps because blue and red were already taken.
Like systemic arterial blood pressure, the pulmonary arterial pressure trace can be analysed, and its various component parts have a hidden meaning to the discerning eye:
Of these features the interesting ones are:
- The systolic pressure wave, the pressure generated by the right ventricular contraction as it pushes against the pulmonary circulation. The amplitude of this wave depends on the elastic recoil of the main pulmonary artery and the pressure generated by the RV contraction.
- The reflected pressure wave is a pressure wave sent back towards the right ventricle, reflecting off all the surfaces in the distal pulmonary arterial tree, and it depends on multiple factors. The amplitude of this wave depends on the elastic recoil of the entire vascular tree, on the pulse wave velocity, and on the distance from the transducer to the major reflecting sites in the vascular tree (i.e its various forkings and narrowings).
- The inflection point is the point in the wave where the reflected pressure starts contributing to the systolic pressure, i.e. it marks the arrival of the first reflected wave. Because under most circumstances this part of the pressure trace is smooth (i.e. there is no discernable inflection point), the inflection point is defined as the time at which the first derivative of pulmonary artery pressure reaches its first minimum.
- The augmentation index, which is the reflected pressure wave amplitude divided by the systolic pressure wave amplitude, is a measure of vascular stiffness which is usually applied to the systemic circulation (Nichols et al, 2002). It is one of the indices which can be used to discriminate between the pulmonary hypertension associated with massive PE from the "primary" variety of pulmonary hypertension (Nakayama et al, 2001
- The inflection time (Ti) is the time it takes for the reflected wave to come back to the transducer. When it arrives too early, it adds to the RV afterload, which is not ideal. The main factor which influences this variable is the location of the reflecting surfaces in the vascular tree. In pulmonary hypertension of chronic PE, the large proximal pulmonary arteries are the site of reflection, which gives rise to a shorter Ti and a higher RV afterload
The reason for the rather modest pressure in the pulmonary arteries is the fact that pulmonary blood vessels are rather thin-walled, with little muscularity (as compared to their systemic counterparts). This means the pulmonary circulation has some impressive elasticity. Just at rest, from beat to beat, the pulmonary blood volume varies by around 50ml over the course of a single cardiac cycle (Ugander et al, 2009). When cardiac output increases, say during extreme exercise, the pulmonary blood flow obviously also increases but the pressure in the arterial circulation remains the same. Trying to resist talking about resistance until the Pulmonary vascular resistance section down below is quite difficult at this stage, and so one might segue into a discussion of pulmonary capillary pressure.
Pulmonary capillary pressure
A measurement of pressure in the pulmonary capillaries is difficult to perform. In an attempt which would preempt future designs by Swan and Ganz, Hellems et al (1949) introduced a thin catheter into the pulmonary arteries of thirteen people, all of whom for some reason had primary syphilis. Presumably the investigators felt that this identified them as being of an especially agreeable character, and most likely to give consent for the experiment. The catheter was advanced until it wedged itself in a pulmonary artery of around 2-3mm diameter, and then it was transduced. The idea was that, in the absence of arterial pulsation from the pulmonary artery, such "occlusion pressure" measurements would yield the pressure of the pulmonary capillaries. The original pressure trace images created during these first recordings did not withstand the brutal attention of whatever steampunk scanner APS are using, and is unfortunately unusable even by the low standards of Deranged Physiology. Suffice to say, it demonstrates a familiar waveform which (for reasons explained later) represents pulmonary venous pressure rather than the pressure in the capillaries. This waveform (discussed in detail below) usually looks like this:
The reason as to why this pulmonary artery occlusion pressure (often confusingly referred to as pulmonary capillary wedge pressure) is not actual pulmonary capillary pressure is because there is often some considerable capillary resistance in the pulmonary circulation which ends up being abolished by the act of wedging (i.e. no forward flow is possible). Pulmonary capillary pressure is therefore usually higher than pulmonary artery occlusion pressure, but lower then pulmonary artery diastolic pressure.
The direct measurement of pressure in the pulmonary capillaries is actually quite challenging. The main barrier here is really the inaccessibility of the capillaries: it is difficult to get a measurement probe into them without destroying them and everything around them. With respiration and cardiac pulsation, these capillaries are a microscopic moving target, and the overlying visceral pleura is sufficiently tough to break your micropipette tips. Bhattacharya et al (1980) had overcome this by perfusing a dog lung with non-pulsatile flow and constant filling pressure, which allowed them to determine the pressures in vessels of different diameters, including pulmonary arterial and venous capillaries. Those micropipettes did not measure any waveforms (the flow was constant), but data from the investigators' Table 1 can be used to construct this graph of pressure changes along the pulmonary vascular tree:
Though the investigators perfused their isolated dog lungs with a blood flow which they felt to be closely resembling the physiological, it was still a continuous flow. One cannot help but ask, what would be the difference if the flow were pulsatile? The answer is, probably not much, at least in terms of mean capillary pressure. However, pulsatile blood flow in pulmonary capillaries helps recruit some of the otherwise collapsed capillaries during systole, and they remain recruited even in diastole. These lightly modified diagrams from Presson et al (2002) demonstrate, by showing alveolar capillary beds, that pulsatile flow increases the number of perfused capillaries by about 100%:
Even though pulmonary artery occlusion pressure is not the same pressure as pulmonary capillary pressure, the capillary pressure can also be estimated from the shape of the pulmonary artery occlusion pressure trace. In general some of the best images of this phenomenon and some of the most detailed discussion can be found in the freely available 2011 review by Juan Grignola, which was heavily strip-mined for its material. Without going into too much detail (that is covered in another chapter), it will suffice to say that the true pulmonary capillary pressure can be extrapolated from the fas and slow components of pressure decay, as the arterial pressure droops after a wedge occlusion.
In slightly more detail,
- As the pulmonary artery is occluded, the occluded artery discharges its blood volume:
- First into the pulmonary arterial capillaries
- Then the same blood travels into the postcapillary venules
- This produces a biphasic pressure drop:
- A fast pressure drop which occurs due to high pulmonary arterial capillary resistance (this accounts for about 2/3rds of the total pressure drop)
- A slow pressure drop which occurs due to the low pulmonary venous capillary resistance
- Thereafter, with all the excess blood discharged into the pulmonary venous circulation, the occluded artery's measured pressure equilibrates with the pressure in the pulmonary veins.
The pulmonary capillary pressure can, therefore, be determined from this graph by three main means:
- By eyeballing the curve and estimating where the inflection point is, or
- By extrapolating the slow and fast decay curves and plotting where they intersect, or
- By using the Gaar equation, described by Gaar et al (1967):
That Gaar equation is:
PPC = LAP + 0.4 × (mPAP - LAP)
- PPC = pulmonary capillary pressure
- LAP = left atrial pressure
- mPAP = mean pulmonary artery pressure
In other words, Gaar and co. determined that the capillary pressure was under most circumstances about 40% of the total fall from mean PA pressure down to left atrial pressure (which is your occlusion pressure):
Is this empirically derived shortcut accurate? Of course not. In fact, of these methods, none are particularly reliable, and so one must conclude the discussion of pulmonary capillary pressure with the statement that truly, the only thing we can safely say about it is that it is somewhere between pulmonary arterial diastolic and pulmonary venous pressure.
Regional variations in pulmonary intravascular pressure
To add an extra layer of confusion to an already inaccurate series of pressure values, one could also point out that pulmonary capillary pressure varies depending on where in the lung it is measured, mainly because of the gravity-related changes in hydrostatic pressure between different regions of the lung. Nunn's gives an excellent diagram to illustrate this concept (page 91 of the 8th edition), and though there is no supporting reference offered, we can assume it proibably came from the classic paper by West et al (1964), though the authors measured blood flow instead of pressure. This is discussed in greater detail in the chapters dealing with ventilation-perfusion matching and Wests' zones. For now, it will suffice to mention that pulmonary arterial and capillary pressures are affected by the hydrostatic pressure gradient in the erect lung:
Pulmonary venous pressure
Smiseth et al (1999) quote a mean pulmonary venous pressure of around 10.3 mmHg. Judging by the abovementioned constant flow experiments performed in isolate dog lungs, the pulmonary venous resistance must be fairly low and the pressure of the entire venous system must be pretty uniform. Flow here occurs mainly because of the pressure gradient between pulmonary venous capillaries and the larger pulmonary veins. There is a pulsatile waveform here, which resembles the central venous pressure waveform. Hellevik et al (1999) offer an excellent breakdown of this waveform, relating its features to the events of the cardiac cycle and specifically to left atrial activity. For the purpose of this (already needlessly detailed) chapter, the added complexity would add nothing to the exam performance of the readers, and so discussion of the pulmonary venous pressure waveform analysis will be limited to this diagram:
Well, perhaps some added discussion is called for. Basically, the waveforms are easily explained by the events of the cardiac cycle:
- the A-wave occurs shortly after the P-wave on the ECG, and shortly before the QRS complex. It represents the increase in pulmonary venous pressure which occurs when the left atrium contracts to eject some last-minute preload into the LV which is about to contract.
- The R-wave is the corresponding change in pulmonary venous flow during atrial systole; i.e. because of the sudden increase in atrial pressure the pulmonary venous flow stops or even reverses (that's why the flow curve dips below zero at this point on the diagram)
- The S-waves are flow peaks which occur during atrial relaxation, after the mitral valve has closed. The suddenly flaccid atrium accepts a large volume of pulmonary venous blood, and the pressure in the system drops, which corresponds to an increase in antegrade flow. Why is this flow biphasic, i.e. why are there two S-waves (S1 and S2) and what is the significance of this? Nobody can truly say. Smiseth et al (1999) thought it was the flow wave from the right ventricle, having made its way through all of the pulmonary circulation, finally making it to the left atrium just in time to contribute. Hellevik et al (1999), preferring to assume that all the pulmonary venous waveforms originate in the left heart, thought it was a delayed reflection wave of the earlier atrial contraction, which, after bouncing around the pulmonary venous circulation, returned to the left atrium.
- The D-wave is the brief rise in pulmonary venous flow (and drop in pulmonary venous pressure) which occurs during left ventricular relaxation. As the LV relaxes, the mitral valve opens and the LA decompresses into the LV, which means the pressure in the entire pulmonary venous network drops.
Pulmonary vascular resistance
Rather than digress extensively on the relative meaninglessness of this variable from a diagnostic and clinical standpoint, the objective here will be to describe its measurement and characteristics for an audience of junior trainees who will be expected to memorise facts for their exams. Thus, pulmonary vascular resistance is:
- mPAP is the mean pulmonary arterial pressure, and
- mLAP is the mean left atrial pressure
But, most often you will see this equation as:
- mPAP is the mean pulmonary arterial pressure, and
- PAOP is the pulmonary artery occlusion pressure which we assume is a reasonable surrogate for mean left atrial pressure
- 80 is a fudge factor which converts the equation's output (which ends up in mmHg⋅min⋅mL-1) into dynes-sec/cm–5.
Given that the pressure drop across the mPAP-PAOP gradient (otherwise known as the transpulmonary pressure gradient) is about 10 mmHg, one might expect the PVR to be a very low value, much lower than that of the systemic circulation where the pressure drop is much higher. A normal PVR is listed as 100-200 dynes/sec/cm-5, where the upper range is around 250 (from this Edwards Life Sciences propaganda pamphlet). When it is indexed to body surface area, it becomes PVRI, with a normal range of 255 - 285 dynes-sec/cm–5/m2.
Anyway. This resistance is spread more evenly in the pulmonary circulation than in the systemic circulation. Large pulmonary arteries capillaries and veins all contribute a roughly similar amount of resistance to the total, but the biggest drop occurs in the capillary bed, which accounts for about 40% of the total resistance. In contrastm in the systemic circulation the resistance in the arterioles is by far the dominant influence. If one were to graphically represent the distribution of resistance in the pulmonary circulation, one could do no better than this excellent diagram from Bhattacharya et al (1980), presented here in a minimally altered state: