This chapter addresses the problem of mean-square stability of systems described by impulsive stochastic linear differential equations. We consider a general class of stochastic impulsive differential equations that encompasses impulsive stochastic linear differential equations of Itô type as well as impulsive linear differential equations subject to Markovian jumping as particular cases. In Sect. 3.2, we first introduce a Lyapunov-type jump differential equations that belong to the category of GLJLDEs introduced in Chap.  2 . Such an equation is associated to the considered class of impulsive stochastic linear differential equations, and it plays a key role in characterizing mean-square stability properties for such systems. In Sect. 3.3, several mean-square stability concepts are defined. The different connections/relations between these stability notions are then resumed in Propositions 3.2 and 3.3 and Theorem 3.1, respectively. A particular emphasis is then put on the so-called exponential stability in the mean-square sense (ESMS). This emphasis is motivated by the fact that ESMS benefits from a Lyapunov-type characterization. Hence, in Sect. 3.4, Lyapunov criteria for ESMS are given in Theorem 3.2 for the general time-varying case, as well as for the time-periodic case, Theorem 3.3 and the time-invariant case, Theorem 3.4, respectively. The main theoretical support used in this section is the theory of jump linear differential equations with positive evolution developed in Chap.  2 . Finally, in Sect. 3.5, we shall study some useful properties of the solutions of a class of non-homogeneous jump stochastic linear differential equations often named affine jump stochastic linear differential equations.