Purpose: The objectives of this survey study were to provide an estimate of the prevalence of neuropathic pain (NP) and to explore the cross-sectional and longitudinal group differences postoperatively. Method: A cohort of consecutive patients who had undergone total shoulder arthroplasty (TSA), reverse shoulder arthroplasty (RSA), or humeral head replacement (HHR) were surveyed within an average of 3.8 years after surgery. Questionnaires completed at the time of the survey were the Self-Administered Leeds Assessment of Neuropathic Symptoms and Signs (S-LANSS) pain scale, the visual analogue scale (VAS) for pain, the Western Ontario Osteoarthritis of the Shoulder (WOOS) index, the Patient Health Questionnaire–9 (PHQ–9), and a satisfaction questionnaire. Results: Of the 141 candidates who were invited to participate in the study, 115 patients participated (85 TSA, 21 HHR, and 9 RSA), for an 82% response rate. Five patients (4%) met the criteria for NP, of whom one had a loosening of the prosthesis and required further surgery. Having NP was associated with greater pain (VAS; p=0.001), greater depression (PHQ–9; p=0.001), more disability (WOOS; p=0.030), and less satisfaction with the surgery (p=0.014). There was no relationship between the presence of NP and patients' age, sex, preoperative pain, range of motion results, or WOOS scores (p>0.05). Conclusions: Persistent pain of neuropathic origin is not common after shoulder arthroplasty, but it is a significant contributor to poor mental and physical well-being and thus warrants further research.