Virtual power plants (VPPs) have emerged as an efficient tool to manage the increasing number of distributed energy resources (DERs) in active distribution networks (ADNs). Meanwhile, this increasing proportion of DERs in the ADNs also provides an ideal opportunity for VPP based integrated energy systems (VES) to provide active as well as reactive power support. This paper proposes a bi-level energy management model to study the strategic behavior of a profit-driven VES in the distribution energy market (DEM). In the upper-level, the VES submits its offers and bids for both active and reactive powers in terms of price and quantity to the DEM. In the lower level, the overall operation cost of an ADN is minimized to clear the DEM. The network constraints are represented by an optimal power flow problem, while the uncertainty in the renewable energy is modeled by the chance constraints. For an efficient solution, the proposed bi-level energy management model is converted into a single-level mathematical program with equilibrium constraints (MPEC) that is framed as a mixed-integer linear program (MILP) using linearization methods. The simulation results from a case study on a VES in an ADN of New South Wales (NSW), Australia, demonstrate the effectiveness of the proposed model.