In the recent few years, there has been an explosive rise in the use of whole-body positron emission tomography-computed tomography (PET-CT) due to increasing clinical acceptance and the establishment of medical insurance reimbursement. However, PET-CT study is associated with a higher radiation dose burden to the patient when compared with an individual CT or PET study. Thus, the question: patient dose in PET-CT: does it matter? The answer to this query is viewed from two contrasting hypotheses, namely, the linear no-threshold (LNT) hypothesis and radiation hormesis. The former hypothesis is mainly based on mathematical linear extrapolation from effects observed by high doses. This notion advocates the linear assumption of radiation risk and that there is no lower limit to the amount of radiation that causes short- or long-term damage. The latter is the countervailing hypothesis-radiation hormesis. This is supported by scientific data from extensive epidemiological observations and experimental studies, denotes that there is indeed a threshold for radiation harm in human and that the radiation doses associated with PET-CT may lie below this threshold and may even be beneficial that may reduced mortality from all causes including decreased cancer mortality, thereby extending longevity. In this paper, previous experimental observations and human epidemiological studies on radiation hormesis as well as its mechanism will be reviewed.