In this paper, we propose to study the influence of using head-mounted displays (HMDs) for visual quality rating of omnidirectional images and their impact on the final quality scores. Because of the used display technology, these devices introduce a significant impairment called screen door effect that may alter the quality of experience. Furthermore, the extended use of such a technology may produce cyber-sickness. In this study, a subjective experiment is designed and carried out using various HMDs with various types of content. The statistically analysed results revealed a significant difference between HMDs for quality rating tasks on the overall ratings as well as per individual distortions. These findings will contribute to the development of a reliable protocol for omnidirectional subjective quality assessment, and the constructed database will be used as a ground-truth for quality metrics development.