Loss of pupillary light reactivity (PLR) three days after a cardiorespiratory arrest is a prognostic factor. Its predictive value upon hospital admission remains unclear. Our objective was to determine the prognostic value of the absence of PLR upon hospital admission in patients with out-of-hospital cardiac arrest. We prospectively included all out-of-hospital cardiac arrests occurring between July 2011 and July 2017 treated by a mobile medical team (MMT) based on data from a French cardiac arrest registry database. PLR was evaluated upon hospital admission and the outcome on day 30. The prognosis was classified as good for Cerebral Performance Category (CPC) 1 or 2, and poor for CPC 3-5 or in case of death. Data from 10151 patients was analysed. The sensitivity and specificity of the absence of PLR for a poor outcome were 72.2% (71.2-73.2) and 68.8% (66.7-70.1), respectively. We identified several variables modifying the sensitivity values and the false positive fraction of a factor, ranging from 0.49 (0.35-0.69) for the Glasgow Coma Scale to 2.17 (1.09-2.48) for pupillary asymmetry. Among those living with CPC 1 or 2 on day 30 (n = 1990; 19.6%), 621 (31.2% (29.2-33.3)) had no PLR upon hospital admission. In the multivariate analysis, loss of PLR was associated with a poor outcome (OR = 3.1 (2.7-3.5)). Loss of pupillary light reactivity upon hospital admission is predictive of a poor outcome after out-of-hospital cardiac arrest. However, it does not have sufficient accuracy to determine prognosis and decision making.