In this paper, a list sphere decoding method based on an extension of the signal noise ratio (SNR)-adaptive sphere decoder (SSD) is proposed to perform joint iterative detection and decoding in coded multiple-input multiple-output (MIMO) systems. The SSD offers almost optimal symbol error rate (SER) performance over the entire range of SNRs, while keeping its complexity roughly constant for uncoded MIMO systems. The proposed list SSD (LSSD) further improves the complexity of detection schemes in coded MIMO systems, which can greatly reduce the number of nodes visited when generating the candidate list. The simulation results show that the LSSD achieves mostly equivalent performance to the original list sphere decoder (LSD), and much lower complexity compared to the LSD.