The patents have long been recognized as a rich and potentially fertile source of data for the study of innovation and technological evolution, which, in turn, would eventually bring value to business. In particular, patent data includes citations to previous patents and to other scientific literature, and patent citations allow one to create an indicator of the "importance" of individual patents. However, existing methods of measuring the importance of patents simply count the number of citations, which is insufficient to capture the large variance in the technological and economic significance of individual patents. In this paper, we propose a ranking-based metric to measure the importance of patents. The metric not only takes into account of the citation counts, also considers the importance ranking score of citing patents. That is, the importance of patents propagates through the citation links. Using the defined ranking-based measure, we conduct an empirical study to show the effects of forward truncation problem in real patent citations data.