INTRODUCTION The Neuropathic Pain Symptom Inventory (NPSI) was recently validated for the assessment of neuropathic pain symptoms (1). The present prospective study aimed to confirm the psychometric propertes of the NPSI in a large cohort of neuropathic patients and compare the frequency of neuropathic descriptors. METHODS A total of 482 patients with peripheral or central neuropathic pain were recruited for the study and filled out the NPSI. A factor analysis using the principal component analysis as the method of extraction was performed to confirm the factorial structure of the NPSI. Multiple Correspondance Analysis (MCA) was performed to evidence associations between neuropathic symptoms and the topography of pain, anatomical lesion (mononeuropathy, polyneuropathy, spinal cord …) or aetiology (i.e, diabetes, postherpetic neuralgia, trauma, stroke...). RESULTS The factor analysis clearly identified 5 distinct neuropathic dimensions (superficial ongoing pain, deep pain, paroxysmal pain, evoked pain and paresthesia/dysesthesia). Minor differences were observed regarding the prevalence of descriptors and dimensions between some aetiologies of pain. Particularly the dimensions "deep pain" and « paresthesia/dysesthesia » were significantly less common in patients with PHN compared to other aetiologies. However, the MCA did not evidence any clear association between the symptoms or dimensions and the topography, lesion or aetiology of the lesion. CONCLUSION This study further validates in a large cohort of patients the multidimensional structure of the NPS and demonstrates the lack of clear association between the neuropathic symptoms and the anatomical lesion or aetiology of neuropathic pain. (1) Bouhassira D., Attal N., Fermanian J et al. Development and validation of the Neuropathic Pain Symptom Inventory. Pain. 2004;108 :248-57.