Background: Psychosocial stress is a significant public health problem inducing consequences for quality of life. Results about the use of dehydroepiandrosterone (DHEA) as a biomarker of acute stress are conflicting. We conducted a systematic review and meta-analysis to demonstrate that DHEA levels could be a biomarker of stress. Methods: PubMed, Cochrane Library, Embase, and ScienceDirect databases were searched on March 19, 2021 using the keywords “acute stress” AND “DHEA” OR “Dehydroepiandrosterone.” Articles needed to describe our primary outcome, i.e., induction of acute stress and at least two measures of DHEA. Results: We included 14 studies, with a total of 631 participants, in our meta-analysis. The DHEA levels increased overtime after acute stress [standardized mean difference (SMD) = 1.56, 95%CI = 1.13–1.99]. Stratification by time showed a main peak at the end of stress (SMD = 2.43, 95%CI = 1.59–3.27), followed by a progressive decrease (coefficient = −0.11, 95%CI = −0.19 to −0.17, p = 0.020). There was no significant change 1 h after the end of acute stress. Metaregressions showed an impact of mental stress (SMD = 2.04, 95%CI = 1.43–2.65), sex (SMD = 0.02, 95%CI = 0.00–0.04), age (SMD = −0.12, 95%CI = −0.2 to −0.05), and obesity (SMD = 0.31, 95%CI = −0.00 to 0.63). There was no difference whatever the type of fluid (blood or saliva) and the measurement technique used. Conclusions: DHEA is a biomarker of acute stress, with a short-term increase (1 h). DHEA increases following acute mental stress, whatever the type and duration of mental stress. Women, young people, and obese individuals had a higher response. Blood and saliva measures were comparable.