Assessing and predicting neuropathic pain after spinal cord injury: a TRACK-SCI study

Abstract

Neuropathic pain is one of the most common secondary complications occurring after spinal cord injury (SCI), and often surpasses motor and sensory deficits in the patient population preferences of the most important aspects to be treated. Despite the better understanding of the molecular and physiological mechanisms of neuropathic pain, reliable treatments are still lacking and exhibit wide variations in efficiency. Previous reports have suggested that the most effective pain management is early treatment. To this end, we utilized the TRACK-SCI prospective clinical research database to assess the neuropathic pain status of all enrolled patients and identify acute care variables that can predict the development of neuropathic pain 6- and 12-months post SCI. 36 out of 61 patients of our study cohort reported neuropathic pain at the chronic stages post SCI. Using multidimensional analytics and logistic regression we discovered that (1) the number of total injuries the patient sustained, (2) the injury severity score (ISS), (3) the lower limb total motor score, and (4) the sensory pin prick total score together predict the development of chronic neuropathic pain after SCI. The balanced accuracy of the corresponding logistic regression model is 74.3%, and repeated 5-fold cross validation showed an AUC of 0.708. Our study suggests a crucial role of polytrauma in chronic pain development after SCI and offers a predictive model using variables routinely collected at every hospital setting.