My recollection is that the indexing technology used by Catalog is proprietary and works only with the Search plug-in distributed with Acrobat and Reader.
What you can do, however, is perform PDF->text conversion and use the results to create your own full-text index. One existing application that can be used to do just that is ht::/dig (www.htdig.org).