Influencing factors analysis and clinical prediction model construction of high voltage long-term pulsed radiofrequency in the treatment of herpes zoster associated pain

Abstract Background Herpes zoster-associated pain (ZAP) imposes a substantial burden with frequently suboptimal treatment outcomes. High-voltage long-duration pulsed radiofrequency (HL-PRF)) has emerged as a promising treatment. This study explored HL-PRF’s application in ZAP, identified efficacy-influencing factors, and established a clinical prediction model.​ Methods Retrospective analysis of 128 ZAP patients treated with HL-PRF (Department of Anesthesiology and Pain Medicine, First Affiliated Hospital of Wannan Medical College, Mar 2023–Dec 2024) was done. Independent predictors of treatment inefficacy were identified via univariate and multivariate logistic regression. A predictive model was developed utilizing a nomogram. Model performance was evaluated using area under the curve (AUC), calibration curves, Hosmer-Lemeshow test, and decision curve analysis (DCA). Results HL-PRF demonstrated 60% efficacy at 3-month follow-up. Univariate analysis identified significant associations ( P < 0.05) between HL-PRF efficacy and multiple factors, including age, disease duration, prolonged hormone use, early non-utilization of antiviral medications, use of antiepileptic drugs, tramadol use, coexisting connective tissue disorders, concurrent malignancy, coexisting cerebral infarction, acute-phase VAS score, and the CD4+/CD8 + T-cell ratio. Multivariate regression confirmed 7 risk factors for reduced efficacy (age, disease duration, connective tissue disease, malignant tumors, long-term hormone use, acute-phase VAS score, CD4+/CD8 + T-cell ratio; P < 0.05). The predictive model exhibited excellent discriminatory power(AUC = 0.988, 95%CI:0.976-1.000), good fit (Hosmer-Lemeshow P = 0.969), and high accuracy (93.0% overall accuracy, 90.7% sensitivity, 96.2% specificity).​ Conclusion HL-PRF is safe and effective for ZAP but affected by multiple factors. Older patients, prolonged disease duration, comorbidities (connective tissue disease/malignant tumors), long-term hormone use, acute-phase VAS > 7, and low CD4+/CD8 + ratio may have suboptimal outcomes. Future research should focus on individualized HL-PRF protocols.​