This week's BBC Radio 4 "The Food Programme" was on coffee.

http://www.bbc.co.uk/radio4/factual/foodprogramme.shtml

about a quarter of the way through.

According to Prof. Robert Superco, head of molecular, genetic and
preventive cardiology at the Fuqua Heart Centre in Atlanta, Georgia,
more than about three or four cups per day of caffeinated coffee
raises blood pressure (no surprise there), but a similar amount of
decaf coffee raises LDL cholesterol (the bad stuff), neither of which
is good news for diabetics.

The caffeinated coffee in his study was made from Arabica beans, but
the decaf was from Robusta beans. Apparently the latter are commonly
used for making decaf because they are stronger flavored which makes
up for the loss of flavour that occurs during the decaf process.

Googling for coffee + cholesterol gives the following:

http://scholar.google.com/url?sa=U&q=http://www.ajcn.org/cgi/content/abstract/54
/3/599

Quote
Caffeinated and decaffeinated coffee effects on plasma lipoprotein
cholesterol, apolipoproteins, and lipase activity: a controlled,
randomized trial

HR Superko, W Bortz Jr, PT Williams, JJ Albers and PD Wood
Stanford Center for Research in Disease Prevention, Stanford
University School of Medicine, CA.

Coffee consumption has been associated with elevated plasma
cholesterol. One hundred eighty-one men consumed a standard
caffeinated coffee for 2 mo followed by randomization to continue
caffeinated coffee (control), change to decaffeinated coffee or no
coffee for 2 mo. Plasma low-density-lipoprotein (LDL) cholesterol and
apolipoprotein B concentrations increased significantly (0.12 +/- 0.65
mmol/L, P less than 0.025; 0.06 +/- 0.12 g/L, P less than 0.0004,
respectively) in the group that changed to decaffeinated coffee. In a
subgroup (n = 51), post-heparin lipoprotein lipase decreased
significantly more (-270 mmol free fatty acids.L-1.h-1, P less than
0.003) in the decaffeinated- coffee group. Resting heart rate and
blood pressure did not change significantly. Change from caffeinated
to decaffeinated coffee increased plasma LDL cholesterol and
apolipoprotein B whereas discontinuation of caffeinated coffee
revealed no change. This finding suggests that a coffee component
other than caffeine is responsible for the LDL cholesterol,
apolipoprotein B, and lipase activity changes reported in this
investigation.
Unquote

In the radio programme, Superco attributed the effect on cholesterol
as being due to the differences in the detailed composition of the two
types of beans, rather than the decaffeinating process (which here
used water).

I'm no expert but AIUI an '-ase' (as here in lipase) is an enzyme that
breaks down something, and in this case I assume that the decrease in
the 'post-heparin lipoprotein lipase' is brought about by something in
the Robusta beans and means that there's less of this cholesterol-
buster around in people who drink lots of decaf, hence the increase in
their LDL cholesterol.