Derived values from cardiac output measurement devices

This chapter is directly relevant to Section G7(v) of the 2017 CICM Primary Syllabus, which expects the exam candidate to "Explain the derived values from common methods of measurement of cardiac output including transpulmonary thermodilution and pulse contour analysis devices (e.g. variables such as EVLW, GEDV, SVV in addition to CI, SVI, SVRI, LVSWI etc.)​​​​​". That in itself would be a ten thousand word treatise,  but the "etc" stretches it out into something truly monstrous. Consider this print out spat from a mid-range  PA catheter output manager system, integrated with the monitor software and ventilator:

Confronted with this superabundance of abbreviations, most ICU trainees would immediately stop reading, having concluded that there is no merit in learning anything about them, as no exam anywhere could possibly ask for any detail at this level of granularity. And they would be correct in this, with the exception of SVR, which is the one variable that has been asked about repeatedly. Multiple questions from the Part 2 exam have required, directly or indirectly, for the trainee to calculate the SVR:

• Question 28 from the second paper of 2015
• Question 1 from the second paper of 2013
• Question 21 from the first paper of 2001

Out of respect for people who are looking for a quick answer, the SVR equation is:

SVR = ([MAP-CVP] × 79.9) / CO

And for the rest, equations and short descriptions are listed below, in case anybody ever needs to know this stuff. Helpfully, this content can also be found in many other places, most notably this LiDCO propaganda flier, or this paper by Wigfull & Cohen (2005).

Measured haemodynamic variables

The pulmonary catheter (even an extra-fancy one) only measures a few variables directly:

• CVP
• PA pressure (systolic and diastolic)
• PA wedge pressure
• Cardiac output (by thermodilution)
• Core body temperature
• Mixed venous oxygen saturation

Additionally, other measured variables from various sources are usually fed into the monitoring equiment interfacing with the PA catheter, which can then be used to generate even more derived numbers. These added variables are usually:

• Arterial saturation
• Haemoglobin content
• Arterial MAP, systolic and diastolic pressure
• Heart rate

From these, a thousand derived factors can be pointlessly generated.

On the indexing of haemodynamic variables

It is intuitively understood by all critical care staff that a cardiac output of 5L/min is normal for a 70kg standard human, but vastly excessive for a little mousey teenager and is totally inadequate for a morbidly obese ex-footballer. Indexing of such variables to body size is therefore required, so that they can be compared between patients, and against expected normal range values.  This idea came about in the 1930s, and its origin is generally attributed to Arthur Grollman, who described it in his 1932 book, The cardiac output of man in health and disease. Reviewers lauded the stability of this variable across many different individuals at basal conditions, and it clearly has some longevity in medicine, as we keep using it even to this day, over ninety years later.

Indexing works by taking a measured variable and dividing it by some parameter related to the patient's body size. One could potentially use body mass as that parameter, but this would probably lead to various errors. The human body, as it gains weight, does not gain it proportionally - most of the gained mass is gained as fat, and fat is not a particularly hungry tissue from the standpoint of metabolic need.  The blood supply required by adipose tissue is therefore only about 2-3 ml/100g/min (which is discussed in greater detail in the chapter on the cardiovascular consequences of obesity).  In short, the normal expected blood supply requirements of a person with a vast overabundance of fatty tissue would not be a direct mathematical scaling of a lean person's metabolic demand, if weight was the multiplier. For this reason, we tend to use body surface area (BSA) when we "normalise" haemodynamic parameters by indexing them.

The BSA is an index of body size calculated using total body weight and height, usually by using Mosteller's formula:

BSA (m²) = square root of (height ×weight / 3600)

The normal average adult BSA is 1.9 m² for males and 1.6 m² for females. Thus, a normal adult male with a cardiac output of 5.0 L/min would score a cardiac index (CI) of 2.6 L/min/m². 

The use of BSA is still an imperfect method of indexing haemodynamic variables, and may break down in various extremes. The morbidly obese, for example, can have an elevated BSA which could potentially decrease their cardiac index to the point where it ends up below the normal expected range, whereas their cardiac output might actually be quite sufficient for their body (Adler et al, 2012). Similarly, children have a higher metabolic rate and a low BSA which means their cardiac index measurement can appear normal but be in fact inadequate to meet their needs. An alternative is to use non-indexed values, or to use the BMI as another method of indexing for body size. Stelfox et al (2006) reported  there should be an extra 1L/min of cardiac output per every 12.5kg/m² of BMI above the normal range. 

And now, without further ado,

A list of common haemodynamic variables, 
derived from cardiac output measurement devices

This list is by no means exhaustive, and the amount of detail included in it is minimal. In creating it, the author had assumed that nobody would ever be interested in this material, for any reason. However, just in case, links are provided, either to local resources or peer-reviewed papers. The normal values come from this counterintelligence by Edwards Lifesciences. 

Haemodynamic Parameters which are
Derived from Cardiac Output Measurement Devices

	Formula	Description
CI	CO / BSA	Cardiac Index: cardiac output divided by body surface area. This is required for a comparison to be made between the cardiac output of patients of different size. Normal value: 2.5 - 4.0 L/min/m2 [Components and determinants of cardiac output]
SV	CO / HR × 1000	Stroke volume: "the volume of blood pumped out of the left ventricle of the heart during each systolic cardiac contraction".  This is derived from the cardiac output and heart rate, which makes it the average stroke volume over 1 minute. Normal value: 60 - 100 mL/beat [Components and determinants of cardiac output]
SVI	CI / HR  × 1000	Stroke volume index: this is the same thing as the stroke volume, except indexed for body size. It is probably better to look at this routinely than to look at the raw stroke volume itself. Normal value: 33 - 47 mL/m2 /beat [Components and determinants of cardiac output]
SVR	80 × (MAP - CVP) / CO	Systemic vascular resistance: this is a measure of the total vascular resistance to blood flow, generated by the systemic circulation. Normal value: 800 - 1200 dynes.sec.cm–5 [Factors which affect peripheral vascular resistance]
SVRI	80 × (MAP - CVP) / CI	Systemic vascular resistance index: this is just SVRI indexed to body size by using the CI instead of the CO. You will notice this is a trend here. Normal value: 1970 - 2390 dynes.sec.cm–5.m2 [Factors which affect peripheral vascular resistance]
PVR	80 × (MPAP - PAWP) / CO	Pulmonary vascular resistance: same principle as the SVR, except instead of MAP, you use MPAP (mean pulmonary artery pressure), and instead of the CVP, you use PAWP which is vaguely representative of left atrial pressure. Normal value: <250 dynes - sec/cm–5 [Factors which affect pulmonary vascular resistance]
PVRI	80 × (MPAP - PAWP) / CI	Pulmonary vascular resistance index: like SVRI, this is just PVR indexed to BSA by substituting CI instead of CO.
LVSW	SV × MAP × 0.0144
LVSWI	SVI × (MAP - PAWP) × 0.0136	Left Ventricular Stroke Work Index: this variable describes the amount of work being done by the left ventricle (work being the product of force and displacement). This is simplified by using a correction factor (0.0144) which converts a pressure difference (MAP - PAWP) into force. Again, it is indexed to BSA by using SVI instead of SV.  Normal value: 50 - 62 g/m2 /beat (Clark et al, 2020)
RVSWI	SVI × (MPAP - CVP) × 0.0136	Right Ventricular Stroke Work Index: just like the LVSWI, except it uses pressures from the right side of the circulation. Normal values: 51 - 61 g/m/m2 (Ibe et al, 2018)
LVEDV		Right ventricular end-diastolic volume: this is volume of blood left behind after diastole in the right ventricle.  This is the closest you are going to get to the slippery concept of preload.
RVEF	Holt technique	Right ventricular ejection fraction is determined by the use of the Holt method, which operates on the assumption that the ventricle empties itself of infused contrast in a "fractional" manner and the dye concentration-time curve decreases in a steplike fashion with each ventricular ejection.  Normal value: 40 - 60% (Holt, 1956)
RVEDV	SV / RVEDF	Right ventricular end-diastolic volume: this is volume of blood left behind after diastole in the right ventricle.  This is the closest you are going to get to the slippery concept of preload.  Normal value: 100 - 160 ml (Urban et al, 1987)
DO2	CaO2 × CO	Delivery of oxygen per minute: This is the rate of oxygen delivery in arterial blood. It is calculated by multiplying the oxygen content of blood by the cardiac output. All the data for calculating the CaO2 is available from ABG analysis. Normal value: 950 - 1150 ml/min [Oxygen carrying capacity of whole blood]
VO2	C(a - v)O2 × CO	Consumption of oxygen per minute: This is the rate of oxygen consumption by the tissues. This requires the monitoring of mixed venous oxygen saturation, which is available from the PA catheter.  Normal value: 200 - 250 ml/min [Oxygen extraction ratio]
O2ER	100 × VO2 / DO2	Oxygen extraction ratio: This is the ratio of O2 uptake to O2 delivery; i.e. what fraction of the delivered oxygen is being consumed in the tissues. Normal value: 20-30% [Oxygen extraction ratio]