In the example below, you'll notice that when a \pagebreak is used right after the theorem, an unwanted vertical blank space is added after the theorem environment while it should not.
\documentclass[11pt,fleqn]{book}
\usepackage[showframe,top=3.4cm,bottom=3.4cm,left=3cm,right=3cm]{geometry}
\usepackage[utf8]{inputenc}
\usepackage{amsthm}
\usepackage[framemethod=tikz]{mdframed}
\usepackage{blindtext}
%----------------------------------------------
\theoremstyle{plain}
\newmdtheoremenv{theoreme}{Theorem}
\begin{document}%
\flushbottom
%----------------------------------------------
\chapter{Title}
%----------------------------------------------
\blindtext[3]
\begin{theoreme}text text text text text text text text text text text text text text text text text text text text text text text text text text text text text text text text text text text text text text\end{theoreme}
\pagebreak % TO BE COMMENTED FOR TESTING
text
\end{document}
%----------------------------------------------
with pagebreak

without pagebreak


theoremenvironment has vertical space built in above and below, so when you insert a\pagebreakafter the environment is ended, the space remains. space is ignored at the beginning of a page, but not at the end. – barbara beeton Sep 29 '14 at 00:52\end{theorem}, but when you add an explicit\pagebreak, that space is already applied, and\pagebreakdoesn't try to adjust, on the assumption that you knew what you were doing; sometimes end-of-page spaces are actually wanted, so it's not safe to assume otherwise. – barbara beeton Sep 29 '14 at 12:30