Pain after total knee arthroplasty (TKA) is a well-known clinical problem potentially delaying ambulation and recovery. Perioperative glucocorticoids reduce pain and facilitate early recovery, but the optimal timing and dose are still unknown. High pain catastrophizers have an increased risk of poorly controlled postoperative pain, and moderate to severe pain at 24 h is associated with a risk of pain relapse at 48 h. To evaluate the effect of a repeat moderate dose of glucocorticoids after TKA in high pain catastrophizers presenting with moderate to severe pain 24 h postoperatively, having received preoperative high-dose glucocorticoids. High pain catastrophizers (Pain Catastrophizing Scale > 20) undergoing TKA are screened 24 h postoperatively and are included if they experience moderate to severe pain (VAS > 30) during a 5 m walk test. The included patients will receive either oral 24 mg dexamethasone (n = 55) or placebo (n = 55) on the evening of Day 1 (~30–37 h) after surgery. In addition, patients receive a standard multimodal analgesic regimen, including paracetamol, celecoxib, local infiltration analgesia, and preoperative dexamethasone (1 mg/kg). Patients will fill out a pain diary for 7 days after surgery. The primary outcome is moderate to severe pain (VAS > 30) during a 5 m walk test on the morning of Day 2 after surgery. The secondary outcomes include cumulated pain at rest and during ambulation, cumulated use of rescue analgesics, quality of sleep, lethargy, dizziness, nausea, satisfaction with the analgesic regimen, length of stay, morbidity, mortality, and reasons for readmissions. Follow-up is at 8 and 30 days. The data from this study will provide evidence for the effect of a repeated dose of dexamethasone as an analgesic adjuvant in patients undergoing TKA with a high risk of postoperative pain.