If anyone wants to know why... see below:

Some drop in the uniquely mapped reads in the 2nd pass is to be expected, but it's typically ~1-2%. It's caused by the reads with short splice ovehangs, which can be mapped to multiple junctions.

As in STAR manual, it states:

It does not increase the number of detected novel junctions, but allows to detect more splices reads mapping to novel junctions. The basic idea is to run 1st pass of STAR mapping with the usual parameters, then collect the junctions detected in the first pass, and use them as ”annotated” junctions for the 2nd pass mapping.