Neuropathic pain affects up to 10 % of the population in developed countries. When conventional treatments fail to relieve patients suffering from such pain, one of the alternatives recognised by international guidelines is spinale medullar stimulation (SSM). SSM is a reversible and uninvasive neuromodulation technique that can be used for the treatment of various diseases, including complex regional pain syndrome (CRRS) and post-surgical chronic lumbosciatalgia (LCPC). The SSM uses electrical impulses to generate potential action on nerve cells. Electrical pulses are caused by one or more electrodes surgically implanted in the epidural area. These electrodes are connected to a subcutaneous implanted neurostimulator. The SSM’s mechanisms of action are still poorly known and are mainly based on the portillon theory, according to which the transmission of the nociceptive message at the backbone of the spinal cord can be blocked by electric stimulation on the surface of the parent. This stimulation in tonic (or conventional) mode induces paresthesies (picking sensations, scrubbing) in the affected member that affect the perception of pain and relieve the patient. The main objective of this thesis was to identify changes in sensory perception caused by spinal medull stimulation in patients with chronic pain using a set of quantitative sensory tests (TSQ). According to the gate theory, SSM should change the perception of nociceptive signals. Moreover, the presence of paresthesies should also contribute to altering sensory perceptions. A systematic literature review was carried out following the Cochrane guidelines. The search for relevant articles and summaries was carried out in all languages from the Cochrane, CINAHL, Embase, MEDLINE and Web of Knowledge databases. A total of 15 articles were included in the analyses, but due to the wide heterogeneity between the procedures used, the individuals examined and the measures assessed, we were unable to conclusively identify the impact of SSM tonics on the sensory perception of patients with chronic pain. However, available data suggest that the conventional SSM does not interfere with sensory perception. This review allowed us to identify gaps in scientific literature and inspired us to design a two-stage cross-study that was carried out in 48 SDRC and LCPC patients, featuring neuropathic pain at the legs, recruited in three establishments. Two visits were carried out (neurostimulator activated vs disabled) in a randomised order. The TQPs included 9 tests to assess detection thresholds (cold, heat, touch, vibration), painful thresholds (cold, heat, pressure), as well as the assessment of mechanical allodynia and time summation. The TQPs were repeated at each of the visits using a standardised protocol. Three sites were tested: the most painful member covered by SSM induced paresthetics (target), the contralateral member and the ipsilateral member to the target member. Our data show that SSM tonic, although effective in relieving chronic neuropathic pain, does not seem to interfere with the perception of thermal and mechanical stimuli, as well as with the perception of acute pain. These findings suggest that the portillon theory is not sufficient to explain the SSM’s mechanism of action which would rather modulate a neural track different from that of acute pain. In addition to the quantitative data collected in this study, 630 qualitative observations of perceived positive and negative sensory phenomena were collected anecdotically (i.e. no standardised procedure was used to collect these data). We classified these observations using definitions recognised by the International Association for the Study of Pain and the Encyclopedia of Pain. We also gave examples of issues that could be used for standardisation of qualitative data collection during the QTPs. These results are presented in Chapter 4 in the form of a letter to the publisher as a compact document to the TSQ study presented in Chapter 3. In conclusion, the mechanisms of action of SSM tonics do not seem to influence sensory perceptions contrary to the portillon theory, despite the presence of paresthesies. The development of a mixed research approach, including quantitative and qualitative aspects, could facilitate the monitoring of sensory abnormalities caused by chronic pain according to the different SSM parameters, such as neurostimulator condition (activated vs deactivated) and mode of stimulation (tonic, burst, high frequency, high density).