Although there is a growing number of learning analytics dashboards (LADs), they often fail to improve learning and learners' awareness due to their lack of adaptation capabilities. This paper presents a systematic review that follows the PRISMA statement and analyzes the adaptation features of LADs and their potential effects on learning. 24 of the 462 articles retrieved were scrutinized using an analysis framework centered on adaptation. The main finding is that there is more evidence of adaptable LADs than adaptive LADs, suggesting that adaptivity is worth exploring. The results mainly highlight 3 common LADs adaptable capabilities -most of which offer data exploration featuresand 2 adaptive ones that change or refresh indicators on dashboards. Only a few articles investigate the adaptation of indicator visualizations or organization on dashboards. Currently, there is no work on the use of advanced computing techniques such as machine learning for LADs adaptation. Additionally, only 5 articles provide some evidence of dashboards adaptation features evaluation. As a result, a preliminary research agenda on LADs adaptation is suggested for enhancing LADs adoption and utility.