The study enrolled patients with acute/subacute ZAP who underwent R-PRF and PRF + PVB at the Pain Department of the affiliated Hospital of Jiaxing University from March 2019 to December 2021. The protocol of this retrospective cohort study was approved by the Medical Ethics Committee of the affiliated Hospital of Jiaxing University (Jiaxing, Zhejiang Province, China, approval number 2021-KY-375). This study was performed in accordance with the Helsinki Declaration of 1964 and its later amendments. Patients and their families were informed and signed the informed consent forms. This study conforms to the STROBE reporting criteria.

Inclusion Criteria: 1. The diagnosis of acute/subacute ZAP is met and is within 3 months after the eruption of herpes; 2. The numerical pain score (NRSs) is ≥ four after 1 week of standardized oral medication (patients with a rash lasting less than 72 h are started on a 7 to 10-day course of an oral antiviral, valaciclovir, 1000 mg three times a day. In addition, the patient is prescribed either pregabalin at a dose of 150 mg per day or gabapentin at a dose of 300 mg per day; 3. The eruption occurs in the unilateral cervical, thoracic, and lumbosacral segments; 4. The patient was treated with pulsed radiofrequency treatment (R-PRF) or pulsed radiofrequency combined paravertebral block treatment (PRF + PVB). Exclusion Criteria: 1. Patients who could not cooperate in completing the follow-up; 2. Patients who were not followed up or were unreachable for a follow-up; 3. Other issues that investigators identified interfered significantly with outcomes, i.e. patients with other chronic pain in the same site with ZAP.

R-PRF: In the CT procedure room, the patient was placed prone on the treatment bed and monitored for vital signs. The upper one-third of the intervertebral foramen responsible for the pain was located, and a 22-G puncture needle was used for puncture after sterilization (Fig. 1). Impedance monitoring was performed near the treatment target to determine the tissue’s resistance around the electrode tip between 250 and 550 Ω. Sensory and motor tests were performed to refine the adjustment. The sensory test was carried out at a voltage of 0.5 V and a frequency of 100 Hz. The sensory test was positive if the corresponding segment evoked the abnormal sensation. The motor test was performed at a low frequency of 2 Hz and a voltage of 0.5 V. The motor test was considered positive if fibrillation and twitching of the trunk muscles innervated by the corresponding segment were induced during the test. We performed pulsed radiofrequency treatment after repeatedly confirming the target location with the following parameters: temperature of 42 °C, period of 300 s, pulse width of 20 ms, frequency of 2 Hz, and voltage of 45 V. The intervention above was repeated within 2 weeks of hospitalization. NRSs of ≤ 3 immediate postoperatively was the success criterion of the procedure.

The needle tip was placed in the left intervertebral foramen

PRF + PVB: The PRF treatment was performed as above, except that the PVB was given to the same target following the CT guidance within 2 weeks of hospitalization, using 10 ml of mixed drug solution, comprising 4 mg of betamethasone sodium phosphate, 2 ml of 0.75% ropivacaine, 2 ml of 2% lidocaine, and 1 ml of mecobalamin in saline. The distribution of the drug was determined by CT analysis.

All patients were given pregabalin 75 mg bid po routinely and continually for 2 weeks after the intervention treatment. The dose was reduced or discontinued if severe adverse effects emerged.

Outcome information was collected by investigators pre- and post-treatment on months 1, 3, 6, 9, and 12 (with a follow-up on each patient for at least 12 months) through outpatient assessments and telephone follow-ups). We recruited suitable patients and scheduled follow-up appointments within 1 month of their receiving treatment. A portion of the data was obtained through medical records, while the rest was collected during follow-up clinic visits and phone calls. Patients were informed about the consequent follow-up of their state both during hospitalization and at their outpatient or telephone follow-up appointments.

Baseline: Having been approved by the Ethics Committee of the affiliated Hospital of Jiaxing University, researchers collected demographic and disease-related information of patients, including age, gender, underlying disease, height, weight, course of the disease, affected side, pain site, lesion space, whether early antiviral, the administered analgesic treatment, lymphocyte count, immunosuppression condition, pre-treatment Numeric Rating Scales score(NRSs) and Pittsburg Sleep Quality Index(PSQI) through the electronic medical record system.

Primary outcome: The incidence of clinically meaningful ZAP in the first month post-treatment. The incidence of clinically meaningful ZAP is indicated by an NRSs score ≥ four after herpes onset. PHN of clinical significance was defined as moderate to severe pain lasting more than 3 months [23,24,25].

Secondary outcome: NRSs:Scaling the pain level from 0 (no pain) to 10 (the worst pain you can imagine). A score of three and below was considered mild pain, four to six as moderate pain, and seven or above as severe pain. Patients were assessed for the most excruciating pain within 24 h [23].

PSQI: Assessing the quality of the patient’s sleep [26].

Short-form McGill Pain Questionnaire-2 score (SF-MPQ-2s): A measure for neuropathic and non-neuropathic pain assessment. This scale was assessed by researchers on month 12 post-treatment [27].

Follow-up interventions: Subsequent treatments such as PVB, PRF, or radiofrequency thermocoagulation (RFTC), the type and number of interventions were recorded.

The statistical analysis was conducted using R studio 4.2.2. Normality analysis was performed based on the Shapiro–Wilk test, and continuous quantitative data conforming to a normal distribution were shown as mean ± standard deviation (\(\bar\) ± SD). The non-conforming values were shown as median (interquartile range), and comparisons were made using the t test or nonparametric rank-sum test. Qualitative data were reported as frequency and proportion and compared by chi-square test, Fisher’s precision probability test, or nonparametric rank-sum test. We analyzed the repeated measures/longitudinal data using generalized estimating equations (GEE) and selected equivariant working correlation matrices. Variables that were significant in the univariable analysis at the p < 0.05 level, as well as those identified through literature review, were included in the multivariable GEE analysis. The analysis aimed to compare the effects between the two groups at each time point (separate effects in group × time). A p < 0.05 was considered statistically significant.