Thank you everyone for going to the trouble of replying and I apologise for the delay in getting back to you.

The pdf converts to word perfectly except for the parts that don't This prevents me from going to the OCR path.

The parts that don't are Greek text - they come up as squares. The font given for those squares when selected is "SymbolOOEnc". I can retrieve the Greek text by individually converting each squared word to "Symbol", but i can't use find and replace to convert all "SymbolOOEnc" to "Symbol" because SymbolOOEnc isn't one of the options in the 'find' section (cutting and pasting it in doesn't work either).

There are hundreds of such words scattered over 100s of pages so being able to use find and replace would be very helpful. Although if I have to, individually converting each word would be possible, it would be a great time saver to find a way around it, and in case I should run into the same problem again. Any suggestions?