The RFFIT test can be used in two ways: to determine a neutralization titer (e.g., 1/5) or to determine a value for rabies antibody concentration (e.g., 0.5 IU). The IU stands for international unit and is measured against a standard reference. So, you have a test serum (e.g., your blood) and you run it through the RFFIT. The serum sample is mixed with a certain amount of rabies virus and incubated. Whatever antibody is there will neutralize the virus. Then you add some fresh cells and incubate it all together. Whatever rabies virus is left (i.e., that which has not been neutralized by the antibody in your blood), will infect the fresh cells and this can be seen under the microscope. So, let's say you have a lot of antibody in your blood. The virus will all be neutralized and not infect the cells. Ok, how much do you really have? Let's dilute your blood and then see if it still neutrralizes the virus. Ok, we did a twofold dilution and it's still going strong--must have a lot of antibody. Let's do a fivefold dilution--still going strong. So, at a fivefold dilution (1/5) your blood still contains enough neutralizing antibody to be protective (by the definition of the standard 1/5). If your blood petered out at say, a 1/2 dilution, you wouldn't be protected. The bigger the denominator, the more antibody initially in your blood (e.g., a titer of 1/75 has more antibody than a titer of 1/5 because it still neutralized at a much higher dilution). It is a dimensionless number because it just refers to the amount of dilution of the sample that will give a measured (i.e., standard) response. It doesn't say how many antibodies are actually in your blood (i.e., it doesn't give a measurement of it). The titer is a proxy for that value. (The whole thing can be further complicated by serial dilutions in the amount of virus used too.) But the RFFIT can be used to get the actual concentration of antibody (by working against the reference standards). This is measured as IU/ml. So, the 1/5 dilution is approximately equivalent to an actual antibody level of 0.5 IU/ml.
Thanks to Danny Brass for simplifying a complicated topic.
