[BRX]

The csv is nothing more than the means to provide MSP with the data to save in its db initially. If you've imported the pdf and the metadata you want you don't need the csv any more and probably the index in the book neither since it's all there if you've done it right,

You don't have to split your books manually. It's sufficient to import the pdf as one file and let MSP do the referencing. I usually have my fakebooks sorted under albums, too (but other groups would do as well) so I can browse them as a whole book (well, alphabetically) or use the entries as alphabetical index. 

IMO you'll see that you don't need any more than that. (But you have to do the work with the csv import initially).