Objective: To compare the efficacy of coblation nucleoplasty under computed tomography (CT) guidancein treating radicular pain related to lumbar disc herniation using the parasagittal interlaminar versus the posterolateral extrapedicular approach. Methods: This randomized, prospective, observational study included fifty-six patients with radicular pain who received coblation annuloplasty under CT guidance. Patients were randomized into two groups according to the approach used: the parasagittal interlaminar group (PIL group) or the posterolateral extrapedicular group (PEL group). The primary clinical outcome was pain assessed using the Visual Analog Scale (VAS). Secondary clinical outcomes were functional status according to modified MacNab criteria, and numbers of patients with significant pain relief (>= 50%) and significant reduction (>= 50%) in anesthetic intake. All outcome assessments were recorded for 12 months postoperatively. Results: The VAS had significantly decreased to 2.0 +/- 1.1 in the PIL group and 2.0 +/- 1.4 in the PEL group at postoperative 12 months. There was no significant difference between the two groups in VAS, modified MacNab criteria, or numbers of patients with significant pain relief (>= 50%) or significant reduction (>= 50%) in anesthetic intake. Five patients in the PIL group experienced cerebrospinal fluid leakage, but there was no significant difference in complication rate between groups. Conclusion: Under CTguidance, coblation nucleoplasty using either the parasagittal interlaminar or the posterolateral extrapedicular approach decreased pain intensity and improved modified MacNabcriteria in patients with radicular pain related to lumbar disc herniation. Both approaches were effective, safe, minimally invasive, and caused minimal discomfort.