RESUMO
INTRODUCTION AND OBJECTIVE: To obtain a new cut-off point (CP) for a lumbar flexion-relaxation (RF) test established with tetrapolar (e.) electrodes, from values already defined with bipolar devices. MATERIALS AND METHODS: The study sample consists of 47 patients in a situation of temporary disability due to low back pain (DL). They were evaluated by means of an isometric dynamometry test, a kinematic test and an assessment of the FR phenomenon. Two experiments with ROC curves are proposed. The first, with 47 patients who consecutively performed the RF test with both types of electrodes, using the cut-off point (CP) known for the e. bipolar (2.49µV). In the second, with the EMG data recorded with e. tetrapolar in 17 patients, a DeLong test was performed that compares the 2 ROC curves that were constructed on the one hand, by classifying the sample from dynamometry and kinematic tests, and on the other, by classifying them with the bipolar EMG values. RESULTS: A total of 34 patients adequately completed the evaluations of the first experiment and 17 patients the second. The first study shows a cut-off point of 1.2µV, with an AUC of 87.7%; Sensitivity 84.2% and Specificity 80%. The second shows a PC for e. bipolars of 1.21µV (AUC 87.5%) and for e. tetrapolar values of 1.43 (AUC 82.5%) with a DeLong test without significant differences between both curves (p>0.4065). CONCLUSIONS: The validation methodology with ROC curves has made it possible to obtain a new PC for the RF test in a practical way, simply by simultaneously performing both tests on the same group of patients until a significant sample is obtained.