Perhaps my information is old, but it states that up to 80% of adults
have evidence of previous infection with CMV, but only a small
percentage of them have been shown to be excreting virus, and
therefore infectious.  Presumably they are infectious only when
excreting virus, no?

Most CMV infection in postnatal children and adults is asymptomatic,
as it usually is in the pregnant woman. The problem is the infection
which occurs in the first trimester.  I am not sure what the ID people
are talking about when they say they are concerned about the "viral
load".  Is there any evidence at all of continued exposure to the
virus being of any danger to the baby who is already been affected by
the virus (including premature babies)?

I did write to the original questioner that the baby was infectious,
likely, and could be shedding virus for months or years and thus was a
danger to pregnant women in particular.

Jack Newman, MD, FRCPC