Centaur upper stages have a Propellant Utilization (PU) system. While this document is old, the fundamentals apply.
To realize optimum performance in a liquid-fueled bipropellant
space vehicle, it is necessary to control both propellants so
as to deplete them simultaneously. Such a simultaneous
depletion both minimizes vehicle burnout weight (by not
allowing any unusable amounts of one propellant or the other
to remain in the tanks) and maximizes the mission total
impulse (by using all available propellant mass in engine
reaction).
Two major factors influence simultaneous propellant
depletion. The first is accurate calibration of engine mixture
ratios, flow rates, and total thrust under flight conditions.
The second is the inability to predict the relative propellant
masses to be loaded at lift-off. Even if such a prediction
were possible, uncertainties in determining what has actually
been loaded onboard provide the second large error source. As
an example, for the Centaur two- burn vehicle, these errors
would result in a maximum error in mass ratio of
approximately 350 lb at burnout, resulting in a loss of 350
lb of payload capability from a mission requiring propellant
depletion. Clearly then, one way to improve total payload capability is to provide some sort of system for propellant
management.
For Centaur the first function of such a system for proper propellant utilization (PU) is to measure accurately
the ratio of propellants in the vehicle tanks during the
entire powered flight portion of the mission.....
The second basic function of the PU system is to control
the flow of propellant through the engines to adapt the
ratio of hydrogen and oxygen to the amounts remaining in the
tank.
The calls you quote are most likely reports of a slightly* anomalous condition in the PU system. The flights where it wasn't mentioned, likely didn't have this anomaly.
*I say "slightly" because the launch was successful.
