Background: Bariatric surgery (BS) is effective for achieving significant weight loss. However, weight regain (WR) is an emerging problem. Objective: To assess the prognostic value of morning serum cortisol, a 1 mg dexamethasone suppression test (DST), 24 h urinary free cortisol (UFC) and late-night salivary cortisol (LNSC) in a cohort of patients with severe obesity (pwSO) undergoing BS in terms of weight loss and WR. Methods: Patients scheduled for BS underwent the following procedures at baseline, 12 months and 24 months after BS: medical history, anthropometric data, blood analysis and cortisol tests. We evaluated total weight loss (TWL) ≥ 30% at 1 year and WR after 2 years as an increase of ≥10% of the maximum weight lost. Results: In total, 142 subjects were included; 101 (71.1%) were females and the mean age was 45.9 ± 9.2 years. Up to 76.8% of subjects achieved ≥30% TWL, without statistically significant differences in DST results or morning serum cortisol, UFC or LNSC levels. However, a higher pre-surgery morning serum cortisol level was a significant predictor of a WR ≥ 10% (cortisol 17.8 [IQR 13.1–18.5] vs. 12.0 [IQR 8.8–15.8] μg/dL; p < 0.01); OR of 1.216 (95% CI 1.069–1.384); AUC [0.761, CI: (0.616–0.906); p < 0.01]. A cut-off value of cortisol > 13.0 μg/dL was predictive of a WR ≥ 10% (sensitivity 0.71; specificity 0.63). Conclusions: No cortisol test was useful in predicting weight loss; however, the pre-surgery morning serum cortisol level was able to predict a WR ≥ 10% in a cohort of pwSO 2 years after BS. A cut-off value of cortisol > 13 μg/dL might be an easy tool to identify patients at higher risk of WR, enabling healthcare providers to implement tailored, long-term strategies to minimize this outcome.