As base stations (BS) are responsible for the large amount of energy consumed in cellular networks, energy efficient BS sleep mode techniques have the potential to save a significant amount of energy. However, assuming that BSs are able to alternate between sleeping and active states as frequently as possible may have a negative impact on network reliability, shortening BS lifetime. In this paper we propose a multiobjective optimization framework aimed at minimizing the power consumption and number of BS sleep mode switchings in heterogeneous cellular networks (HetNet), by jointly considering Quality of Service (QoS) requirements. We focus on the HetNet scenario in which macro and micro cells coexist. The Mixed Integer Quadratic Programming (MIQP) optimization technique is used to minimize the power consumption together with the number of BS sleep mode operations of both macro and micro cells. The trade-off between power consumption, sleep mode switchings and performance of the network is shown for different energy saving solutions. Results show that the proposed optimization can guarantee QoS target throughput for users and significant reduction of 50% for macro and 73% for micro BS respectively daily number of switchings, while still achieving 8% savings in terms of daily energy consumption.