Theoretically, the simplest method is to add the indicator substance at a known concentration and constant rate into the flow and to measure its concentration downstream. Providing that perfect mixing has occurred, the dilution factor gives the ratio of the two flows in the steady state. Stewart (1897), who was the first to use this principle for measuring cardiac output, infused sodium chloride over 10-15 s and measured the change in electrical resistance of the blood. When this had fallen to a steady level during the constant infusion he withdrew a blood sample and derived the dilution factor.
A more practical method is to add the indicator
as a bolus and to measure the concentration time profile of the passage
of this bolus at a point downstream. Typically the concentration rises to
a rounded peak and falls away more slowly as the remaining indicator is
washed out. Such a curve may be considered as a large number of elements
of sufficiently short time duration that for each element the flow and concentration
are effectively constant. The concentration of the marker multiplied by
the flow and the time duration of each element gives the quantity of marker
passed by that element. The sum of these quantities should therefore equal
the total quantity of indicator added as the bolus. If the flow is constant
for the duration of the curve, i.e. is the same for each of the elements,
it may be calculated simply as the quantity of indicator injected divided
by the total area under the dilution curve. The shape of the curve is unimportant
providing that the total area can be found.
The bolus injection method of measuring cardiac output was first described by Henriques (1913), who injected sodium thiocyanate and then withdrew arterial blood samples at 1 s intervals for in vitro colorimetric analysis using ferrous chloride. He identified the problem of the indicator starting to recirculate before the first pass was complete. See also Henriques - further details. White (1947) derived cardiac output from continuously recorded curves and the method was developed by Hamilton et al. (1932) and others for use in man.
Many different indicators have been tried but the most successful has been indocyanine green whose concentration curve can be measured in blood by light absorption. Like the Fick method, it is accurate but difficult to carry out. The densitometer, which measures indocyanine green concentration, has to be calibrated with samples of the individual patient's blood containing known concentrations of indocyanine green. This is time consuming. Other disadvantages are that the dye is very expensive, and if used repeatedly makes the patient green. A new variant of this method is that of pulse dye densitometry. Based on the principle of pulse oximetry, a photodetector is used to record the indocyanine green curve non-invasively. Imai et al (1997) found a considerable degree of error in some patients and that the agreement with thermodilution was less good at low cardiac outputs. Thermodilution, in which cold fluid (or sometimes heat) is injected, is now the most commonly used method.
Fegler first described the use of thermodilution for measuring cardiac output (Fegler, 1954). This method was then adapted for use in man by Branthwaite and Bradley (1968) at St. Thomas' Hospital and the method developed further by Ganz et al (1971) and others. There is now a huge literature on this subject. There are three slightly different thermodilution methods in clinical use, all of which use a thermal indicator (cold or heat) injected into the right side of the heart and detected either in the pulmonary arterial or arterial blood. By far the commonest clinical method of measuring cardiac output is by thermodilution using a Swan-Ganz catheter.
A Swan-Ganz catheter is inserted via a central
vein (usually the internal jugular or subclavian) through the right atrium
and ventricle so that its tip lies in a branch of the pulmonary artery.
Cold dextrose is injected rapidly via one port of the catheter which ends
at a side hole in the right atrium. The cold dextrose mixes with the blood
in the right atrium and ventricle before passing into the pulmonary artery
where the fall in temperature is sensed by a thermistor on the side of the
catheter. Cardiac output is then calculated from the temperature-time curve.
The usual clinical practice is to perform three or more measurements in
rapid succession and to take the average of those which agree closely. For
an excellent review of pulmonary artery catheterization see Gomez and Palazzo 1998.
This is a variant of thermodilution in which heat is introduced into the blood in the right ventricle, by passing a current through a heating coil wrapped around this part of the catheter (Yelderman et al.,1992,1992; Haller et al., 1995; Jakobsen et al., 1995; Lefrant et al., 1995). The rise in temperature which can safely be produced is very small and therefore a random series of pulses of heat is given and signal analysis used to derive the mean thermodilution curve (Yelderman, 1990). The advantage of this method over cold thermodilution is that since no fluid is injected the heat pulses can be continued indefinitely and repeated estimates of cardiac output obtained at intervals of a few minutes. Infusion of cold fluids will of course interfere with this method, but the main drawback is the need for an adapted Swan-Ganz catheter, which is bigger and stiffer than the normal Swan-Ganz catheter, and its attendant risks.
It is possible to inject the cold indicator on the right side of the heart and measure the fall in temperature this produces in the blood in the aorta and so derive cardiac output from this 'transpulmonary thermodilution' curve. First described by Fegler, this method is sometimes used in children too small for insertion of a Swan-Ganz catheter. A fine thermistor is inserted into the aorta via a femoral artery. The agreement with Fick and indocyanine green is generally good although experience with this device is at present limited. This method is also used in the Pulsion PiCCO system to provide calibration for the pulse contour analysis method.