Development and validation of a Taiwan version of the DN4-T questionnaire

Background: Neuropathic pain (NeP) is often under-recognized, resulting in poor pain management. Therefore, a Taiwan version of the 10-item Douleur Neuropathique 4 (DN4-T) questionnaire was developed to identify patients with NeP from a mixed population of patients with pain. Methods: A prospective, nonrandomized, multicenter study was conducted in the Neurology Departments of four Taiwanese medical centers, to develop and validate the DN4-T questionnaire as a diagnostic tool for identifying patients with NeP. Patients who experienced pain for >30 days were classified as having neuropathic, nociceptive, or mixed pain. Patients and physicians also completed the DN4-T questionnaire. The DN4-T scores were assessed with the optimal cut-off score calculated using a receiver operating characteristics (ROC) curve, and sensitivity and specificity assessed and reliability determined statistically using the Cronbach alpha coefficient. Results: Of the 318 patients who completed the DN4-T questionnaire, 189 patients were diagnosed with NeP, seven patients with mixed pain, and 122 patients with nociceptive pain. For statistical analysis, patients were categorized as having NeP (those with neuropathic pain and mixed pain) or non-neuropathic (nociceptive) pain (non-NeP). Using an optimum DN4-T cut-off score of ≥3 (ranging from 0 to 10, determined by a maximum c index value of 1.54), DN4-T scores provided a sensitivity of 0.77 and specificity of 0.78, for predicting NeP. The predictive power of DN4-T in diagnosing NeP was 0.83 (as determined by area under the curve of the ROC curve), and was significantly predictive of pain type (p < 0.0001) with a concordance of 0.785, a discordance of 0.129, and a Cronbach alpha coefficient of 0.7, suggesting that the DN4-T questionnaire is a useful predictive tool for diagnosing NeP. Conclusion: The DN4-T questionnaire has been reliably translated into Mandarin Chinese and can be used as a diagnostic tool for NeP in conjunction with clinical evaluation.