Introduction Residual pain is a major factor in patient dissatisfaction following total hip arthroplasty or total knee arthroplasty (THA/TKA). The proportion of patients with unfavourable long-term residual pain is high, ranging from 7% to 34%. There are studies indicating that a preoperative degree of central sensitisation (CS) is associated with poorer postoperative outcomes and residual pain. It is thus hypothesised that preoperative treatment of CS could enhance postoperative outcomes. Duloxetine has been shown to be effective for several chronic pain syndromes, including knee osteoarthritis (OA), in which CS is most likely one of the underlying pain mechanisms. This study aims to evaluate the postoperative effects of preoperative screening and targeted duloxetine treatment of CS on residual pain compared with care-as-usual. Methods and analysis This multicentre, pragmatic, prospective, open-label, randomised controlled trial includes patients with idiopathic hip/knee OA who are on a waiting list for primary THA/TKA. Patients at risk for CS will be randomly allocated to the preoperative duloxetine treatment programme group or the care-as-usual control group. The primary end point is the degree of postoperative pain 6months after THA/TKA. Secondary end points at multiple time points up to 12months postoperatively are: pain, neuropathic pain-like symptoms, (pain) sensitisation, pain catastrophising, joint-associated problems, physical activity, health-related quality of life, depressive and anxiety symptoms, and perceived improvement. Data will be analysed on an intention-to-treat basis. Ethics and dissemination The study is approved by the local Medical Ethics Committee (METc 2014/087) and will be conducted according to the principles of the Declaration of Helsinki (64th, 2013) and the Good Clinical Practice standard (GCP), and in compliance with the Medical Research Involving Human Subjects Act (WMO). Trial registration number 2013-004313-41; Pre-results.