Abstract Background: Adequate pain control after cardiac surgery is essential. Paravertebral block is a simple technique and avoids the potential complications of epidural catheters. The objective of this study is to compare the effect of ultrasound-guided bilateral thoracic paravertebral block with thoracic epidural block on pain control after cardiac surgery. Materials and methods: Between March 2016 and 2017, 145 patients who had cardiac surgery through median sternotomy were randomized by stratified blocked randomization into two groups. Group I (n = 70 patients) had bilateral ultrasound-guided thoracic paravertebral block and Group II (n = 75 patients) had thoracic epidural analgesia. The primary end point was the postoperative visual analogue scale (VAS). The duration of mechanical ventilation, intensive care unit (ICU), and hospital stay were the secondary end points. The study design is a randomized parallel superiority clinical trial. Results: Both groups had similar preoperative and operative characteristics. No significant difference in VAS measured immediately after endotracheal extubation then after 12, 24, and 48 hours between groups (p = 0.45). Pain score significantly declined with the repeated measures (p < 0.001) and the decline was not related to the treatment group. Postoperative pain was significantly related to diabetes mellitus (p = 0.039). Six patients in group I (8.5%) required an additional dose of morphine versus three patients (4%) in group II (p = 0.30). Patients in group I had significantly shorter ICU stay (p = 0.005) and lower incidence of urinary retention (p = 0.04) and vomiting (p = 0.018). No difference was found in operative complications between groups. Conclusion: This randomized parallel controlled trial demonstrates that ultrasound-guided paravertebral block is safe and effective method for relieving post-cardiac surgery sternotomy pain compared with thoracic epidural analgesia but not superior to it.