Screening tools for neuropathic pain: can they be of use in individuals with spinal cord injury?

Pain of both neuropathic and nociceptive aetiology is common after spinal cord injury (SCI), and classifying pain is sometimes a challenge. The objective of this study was to test the usefulness of the Swedish version of the screening tools Douleur Neuropathique 4 questions (DN4), the Leeds Assessment of Neuropathic Symptoms and Signs (LANSS), the Neuropathic Pain Questionnaire (NPQ), and the painDETECT Questionnaire (PD-Q) in individuals with SCI and pain. A further objective was to define pain descriptors able to discriminate neuropathic pain from nonneuropathic pain. Forty individuals with SCI ≥1 year and pain ≥6 months were examined by a specialised physician and assessed twice using the 4 screening tools. The analysis included tests of reliability (test-retest) and validity (calculation of sensitivity, specificity, and overall agreement), an explorative analysis of the cutoff scores and regression analysis for identifying predictors of diagnostic accuracy. Our results indicate that reliability was good to very good for 3 of the screening tools, DN4, LANSS, and NPQ with a Cohen's kappa coefficient between 0.70 and 1.00. DN4 showed the highest sensitivity (93%), followed by PD-Q (68%), NPQ (50%), and LANSS (36%). LANSS and NPQ demonstrated the highest specificity (100%), followed by PD-Q (83%) and DN4 (75%). Diagnostic accuracy for the tools was for DN4 88%, PD-Q 78%, NPQ 65%, and LANSS 55%. A final model showed that 3 items, hypoesthesia to touch, burning pain, and numbness, could discriminate pain in this cohort of individuals with SCI with a high goodness of fit.