Could someone explain to me why I get ~100 ms latency when I query with a -E pr=500 option while it usually takes 5 ms if I query the same without the option ?

Question is raised because I use a solution which unfortunately query with the option even though there is no real need to do so, and can make up to 100K requests in a row,
So eventually with a 100 ms latency x 100 k request, the global wait time can get very ugly.

Thanks in advance,