Objective: To identify and separate the ventral root from dorsal root, which is the key for success of the artificial somatic-autonomic reflex pathway procedure for neurogenic bladder after spinal cord injury (SCI). Here we report the results of intra-operating room monitoring with 10 paralyzed patients.Methods: Ten male volunteers with complete suprasacral SCI underwent the artificial somatic-autonomic procedure under general anesthesia. Vastus medialis, tibialis anticus and gastrocnemius medialis of the left lower limb were monitored for electromyogram (EMG) activities resulted from L4, L5, and S1 stimulation respectively to differentiate the ventral root from dorsal root. A Laborie Urodynamics system was connected with a three channel urodynamic catheter inserted into the bladder. The L2 and L3 roots were stimulated separately while the intravesical pressure was monitored to evaluate the function of each root.Results: The thresholds of stimulation on ventral root were 0.02 ms duration, 0.2-0.4 mA, (mean 0.3 mA±0.07 mA), compared with 0.2-0.4 ms duration, 1.5-3 mA (mean 2.3 mA±0.5 mA)for dorsal root (P<0.01) to cause revoked potentials and EMG. Electrical stimulation on L4 roots resulted in the EMG being recorded mainly on vastus medialis, while stimulation on L5 or S1 roots caused electrical activities of tibialis anticus or gastrocnemius medialis respectively. The continuous stimulation for about 3-5 seconds on S2 or S3 ventral root (0.02 ms, 20 Hz, and 0.4 mA) could resulted in bladder detrusor contraction, but the strongest bladder contraction over 50 cm H2O was usually caused by stimulation on S3 ventral root in 7 of the 10 patients.Conclusions: Intra-operating room electrophysiological monitoring is of great help to identify and separate ventral root from dorsal root, and to select the appropriate sacral ventral root for best bladder reinnervation. Different parameters and thresholds on different roots are the most important factors to keep in mind to avoid damaging the roots and to assure the best results.