Segmental cervical instability, heart rate variability, and sympathetic symptoms in cervical spondylosis: an exploratory retrospective cohort study

Sympathetic symptoms in cervical spondylosis represent a diagnostically challenging entity. Whether such symptoms arise from direct mechanical irritation of the cervical sympathetic trunk or from systemic autonomic dysregulation remains highly contested. This study aimed to investigate the segment-specific associations between cervical instability, systemic autonomic tone evaluated by heart rate variability (HRV), and sympathetic symptoms, and to explore whether systemic autonomic tone statistically accounts for any part of the association between segmental instability and sympathetic symptoms. This retrospective cohort study included 67 patients with cervical spondylosis, dichotomized into sympathetic and non-sympathetic symptom groups. Segmental instability was evaluated via dynamic cervical radiographs with inter-rater reliability assessed by Cohen’s Kappa. HRV parameters were extracted from 24-hour Holter monitoring. Multivariable associations were analyzed using Firth penalized logistic regression. A nomogram with ROC analysis, bootstrap calibration and causal mediation analysis were conducted as exploratory internal evaluations. Twenty-seven patients had sympathetic symptoms and 40 did not. Radiographic assessment demonstrated substantial inter-rater agreement (Cohen’s Kappa = 0.82). In the Firth penalized model, the SDNN index (OR 1.033, 95% CI 1.007–1.069, P = 0.011) and C3/4 instability (OR 7.855, 95% CI 2.352–31.456, P < 0.001) were the variables associated with sympathetic symptoms, whereas age and sex were not. An exploratory nomogram based on this model showed moderate discrimination (AUC 0.794) on internal assessment. In exploratory mediation analysis, the indirect effect through the SDNN index was small and non-significant (ACME 0.013, 95% CI -0.081 to 0.105, P = 0.782), the direct effect remained significant (ADE 0.410, 95% CI 0.204–0.629, P < 0.001) and the total effect was significant (0.424, 95% CI 0.203–0.647, P = 0.002), with only about 3.2% of the total effect estimated as mediated. In this single-centre cohort, C3/4 instability and a higher SDNN index were the factors most strongly associated with sympathetic symptoms, and most of the observed association between C3/4 instability and symptoms operated through a direct statistical pathway rather than through the SDNN index. Given the small sample and lack of external validation, these findings should be regarded as hypothesis‑generating and do not establish causality or a confirmed anatomical mechanism.