Clinical and MRI-associated outcomes after warm-needle acupuncture for cervical radiculopathy: A retrospective comparative cohort study

Objective This study mainly explored the clinical and magnetic resonance imaging (MRI)-associated outcomes after conventional acupuncture (CA) combined with warm-needle acupuncture (WA) for cervical radiculopathy. Methods In this single-center retrospective comparative cohort, 93 patients with cervical radiculopathy were divided into CA group (n = 45) and CA + WA group (n = 48). Scoring metrics for traditional Chinese medicine (TCM) symptom manifestations, mobility performance of the cervical vertebrae system, alongside MRI measurement indicators [C2∼7 Cobb angle, C1∼7 sagittal vertical axis (SVA), C2∼7 SVA, T1 slope (T1S), T1 slope minus cervical lordosis mismatch (T1S-CL), foraminal width (FW), segmental lordosis (SL)] received quantitative measurement at three time nodes: before treatment (T0), one day after the last treatment (T1), and 12 months after treatment (T2). Pain was assessed using the pain rating index (PRI), visual analogue scale (VAS), and present pain intensity (PPI) components of the McGill pain questionnaire. Results The CA + WA group showed greater improvements in VAS, PRI, and PPI than the CA group at T1 [−1.63 (95% CI −2.10 ∼ −1.16), −1.78 (−2.45 ∼ −1.11), −0.71 (−0.92 ∼ −0.50)] and at T2 [−1.82 (−2.30 ∼ −1.34), −1.83 (−2.50 ∼ −1.16), −0.73 (−0.94 ∼ −0.52)]; at T1 and T2, the CA + WA group showed a more pronounced decrease in the comprehensive TCM syndrome score than the CA group, and more pronounced increases in cervical mobility and cervical curvature ( P < 0.05). At T1 and T2, the total treatment response rate of the CA + WA group was higher than that of the CA group. Between-group differences were present in MRI-derived parameters: the reductions in C2∼7 Cobb angle and FW in the CA + WA group were greater than those in the CA group, and the increases in C1∼7 SVA, C2∼7 SVA, T1S, T1S-CL, and SL were greater than those in the CA group ( P < 0.05). There was no statistically significant between-group difference in the 12-month recurrence rate (2.08% vs 8.89%, P = 0.194). Conclusion Compared with CA alone, the combining WA with CA showed greater improvement in the treatment of cervical radiculopathy at the end of treatment and at 12 months, especially in TCM syndromes, cervical function, and pain intensity. In exploratory analyses, between-group differences were also observed in MRI-derived anatomical and sagittal parameters. However, these findings support prospective, adequately powered randomized evaluation rather than definitive claims of long-term efficacy.