Cyberattacks on Internet of Things (IoT) networks have been increasing rapidly over the last few years to reduce the operating efficiency of connected objects. For this reason, deep learning (DL) and machine learning (ML) are presenting techniques to track abnormal behavior by intrusion detection systems (IDS). This paper performs a systematic literature review (SLR) on using deep learning to detect intrusion in IoT environments. In this work, we used three databases Scopus, Science Direct, and Web of Sciences to collect 355 papers with a textual request. Through the PRISMA approach, we found 194 papers to be analyzed with Nvivo and Vosviewer tools. After analyzing the corpus, we establish that the most important topics are anomaly-based IDS and intrusion detection in IoT networks, therefore we found many results regarding to the publishers, the type of papers, the database that was published more than the other, as well as the number of publications per year.