Multiple description (MD) coding is a promising alternative for robust transmission of information over non-prioritized and unpredictable networks. In this paper, according to the geometrical structure and the spatial relationship of lattice vector quantizers, a technique called shifted lattice vector quantization (SLVQ) is employed in MD image coding to generate descriptions for unbalanced or balanced channels. Experimental results validate the effectiveness of the proposed technique with better performance than the popular MD lattice vector quantization proposed by Servetto, Vaishampayan and Sloane (known as SVS technique). Especially the presented technique can effectively alleviate the gap between the central and side distortions in SVS technique.