This study describes the development and validation of the Neuropathic Pain Symptom Inventory (NPSI), a new self-questionnaire specifically designed to evaluate the different symptoms of neuropathic pain. Following a development phase and a pilot study, we generated a list of descriptors reflecting spontaneous ongoing or paroxysmal pain, evoked pain (i.e. mechanical and thermal allodynia/hyperalgesia) and dysesthesia/paresthesia. Each of these items was quantified on a (0-10) numerical scale. The validation procedure was performed in 176 consecutive patients with neuropathic pain of peripheral (n = 120) or central (n = 56) origin, recruited in five pain centers in France and Belgium. It included: (i) assessment of the test-retest reliability of each item, (ii) determination of the factorial structure of the questionnaire and analysis of convergent and divergent validities (i.e. construct validity), and (iii) evaluation of the ability of the NPSI to detect the effects of treatment (i.e. sensitivity to change). The final version of the NPSI includes 10 descriptors (plus two temporal items) that allow discrimination and quantification of five distinct clinically relevant dimensions of neuropathic pain syndromes and that are sensitive to treatment. The psychometric properties of the NPSI suggest that it might be used to characterize subgroups of neuropathic pain patients and verify whether they respond differentially to various pharmacological agents or other therapeutic interventions.

Several recent studies have suggested that cognitive complaints among chronic pain patients could result from an interference between ongoing pain and mental tasks, as they share common and limited attentional resources. This study was intended to further explore that presumed relationship between chronic pain and attentional disorders. For this purpose, a more sensitive version of the conventional colour-word Stroop task, with four subtasks of increasing difficulty, was used in a group of 33 consecutive patients with chronic non-malignant pain and a group of 20 healthy subjects as controls. This task assesses specifically selective attention. Since the Stroop task is sensitive to dysthymic states, levels of anxiety and depression were also assessed in chronic pain patients.As expected, the increase of response times was positively related to the difficulty of the subtasks. However, only the patients with chronic pain of high intensity presented a significant increase in the response times to each subtask as compared to controls. Response accuracy was not affected. Patients with high pain had higher scores on trait-anxiety than those with low-intensity pain. However, ANCOVA showed that trait-anxiety scores x pain groups interactions were not significant and thus that trait-anxiety was not useful for predicting task performance.These results point to a disturbance of selective attention as a function of pain intensity in chronic pain patients. Copyright 1999 European Federation of Chapters of the International Association for the Study of Pain.

Memory deficits in chronic pain patients are frequently observed. The objective of this study was to explore memory performances of chronic pain patients by using the Process Dissociation Procedure developed by Jacoby (J. Mem. Lang. 30 (1991) 513). This procedure permits to separate the contribution of controlled processes from automatic processes operating within a memory task. The results show a significant decrease of controlled processes in chronic pain patients. Furthermore for both groups, automatic processes contribute in a similar extent to the memory performance. The estimates of controlled processes in the chronic pain patients are significantly related to the fear of pain and catastrophic beliefs. This is interpreted as a sign of interference between the attention consumed by pain experience (namely fear related to pain) and the attention to be allocated to the memory task.

Evidence exists that chronic pain partially consumes the limited attentional resources, with the consequence that controlled processes sustaining cognitive tasks are affected and that automatic processes are preserved. Fibromyalgia syndrome is consistently rated as more severe than other chronic painful conditions. It is assumed here that fibromyalgia is more attention-demanding, leading to a more pronounced decrease of the controlled processes in comparison with other chronic painful conditions. In this perspective, Study 1 compares fibromyalgia patients, patients with localized pain and healthy subjects in a procedure separately estimating the within-task contributions of controlled and automatic processes in a cued recall task. As predicted, controlled processes are more strongly affected in fibromyalgia patients related to the group with localized pain. Unexpectedly, contribution of automatic processes is increased in fibromyalgia. Study 2 replicates these results and reveals that memory functioning in fibromyalgia patients is related to their painful condition as a whole rather than to any particular patient's characteristics.