In my opinion, the different entries are obtained in different conditions of cristallization and other factors, by different laboratories..... . I did not observe the particular entries that you mentioned, but in general, sometimes the protein is completed, and sometimes is just a fragment. Try to select one entry with the complete protein, obtained under mild conditions, preferentially without added factors (as inhibitors, heavy metal ions and so on) and recent entries should be better than oldones.
For a protein as common as lysozyme, you can find it in many different organisms. Furthermore, many proteins can have homologues and isoforms in a given organism. Sometimes researchers do mutational studies on an enzyme, generating even more entries. For a protein with a given amino acid sequence, researchers can study ligand binding, and effects chemical environment (such as pH) can have on the structure, complicating things further. Given that lysozyme is one of the easiest protein to crystallise, there will be more entries compared to your average protein. (Since PDB is mainly a depository for crystal structures, it's skewed towards proteins that are prone to be crystallised and give good diffractions.)
The most straight forward way to tell them apart is to read the description and primary citation on the pdb entry.