Nod32 & huge *.htm files

SOED (Shorter Oxford English Dictionary v2.0, 2002) uses four big HTML files to store its dictionary:
1.htm (37.881.541 bytes)
2.htm (36.974.857 bytes)
3.htm (41.378.864 bytes)
4.htm (16.339.873 bytes)

I don't know if Nod32 has some sort of "deep scan" for *.htm files but the program to consult the dictionary becomes so slooow that it is practically impossible to use.
I tried putting the files/directory in the exclusion list but that didn't work, so I
changed all the occurrences to those files inside the dict program with
another extension and changed the extension of the files. Guess what,
it works fine.
Am I doomed to use SOED that way?

Ciao