A bolus of injectate, usually cold fluid, given into a large central vessel (eg: Superior vena cava) can be detected in a large artery (eg: femoral or axillary), as a temperature change, giving rise to a pulse contour. This is similar in theory to the thermo dilution principle used in the PA catheter where cold fluid of a known volume and temperature is injected proximally into the catheter, and a temperature change detected more distally by a temperature sensor as the colder fluid mixes with blood. The magnitude of the temperature change detected at the thermistor can be used to estimate the volume of the blood into which the cold fluid was diluted, and hence the cardiac output. When this is done across the entire pulmonary circulation however, and is detected in an artery the "pulse contour" generated by the temperature change is flatter and longer, but nevertheless still gives reliable results when extrapolated to predict cardiac output.
Estimation of intra-thoracic blood volume and extra-vascular lung water
The derivation of these parameters using this technique is not simple, nor intuitive, and requires extensive extrapolation of data.
The leap of faith then, is the belief that from wave form analysis produced by a transpulmonary thermo dilution technique, one can calculate the cardiac output of a patient, and derive a number of fluid volumes which allow estimation of both preload (ITBV – intra thoracic blood volumes), and the amount of extravasated fluid into the lungs (EVLW – extra vascular lung water).
Once you accept this principle you can use these numbers to guide both further fluid therapy and administration of inotropes or vasopressors.