Of the 120 eligible individuals screened, six had to be excluded because of missing informed consent (n = 4) or missing or faulty information in the case report form (n = 2). 114 patients (pts) (55 male (48.2%), 54 female (47.4%), undisclosed gender information in 5 individuals (4.4%) with a median age of 59 years (22 to 81 years of age), a mean height of 172.7 ± 9.81 cm, mean body weight of 81.9 ± 15.90 kg with a resulting BMI of 27.4 ± 4.80 kg/m2, could finally be included in the study. Of 114 individuals included, 37 (32.5%) received FJI and 41 (36%) received PRI exclusively, 16 patients (14.0%) received a combination of FJI and PRI, and in 20 individuals (17.5%) the treatment target was not clearly specified, although within inclusion criteria. Participation in the surveys declined over the course of the study. From 100.0% (114 pts) before therapy, participation was 99.2% (113 pts) after the first treatment, 61.4% (70 pts) before the last treatment and 46.5% (53 pts) three months after therapy. Data on the number of necessary treatments are available from 82 patients. These patients received a median of three interventions (minimum 1 and maximum 8). Demographic data and patient characteristics are summarized in Table 1.

Only 10 patients (8.8%) reported that they had not received any prior treatment from healthcare professionals. The patients received a median of 2 pretreatments from different medical professions (minimum 1 and maximum 9). Of the 104 patients with previous treatment, 71.2% (74 pts) consulted an orthopedist, 57.7% (60 pts) a general practitioner and 26.9% (28 pts) a radiologist. The heterogeneous group of professions providing previous treatment is shown in Table 2.

A total of 9 patients (7.9%) stated that they had not received any prior pain treatment. Four patients (3.5%) did not answer this question. The patients received a median of 2 different prior treatments (minimum 1 and maximum 10). Of the 101 patients with previous pain treatment, 83.2% (84 pts) received pain therapy by means of per oral analgesics, 52.5% (53 pts) physiotherapy, and 50.5% (51 pts) injections/nerve blocks. Patients did neither disclose frequency nor the exact date of the last pain pretreatment before study inclusion.The distribution of all types of previous pain treatments is shown in Table 3.

82.9% of patients (87 pts) stated that they had not yet had spinal surgery, while 12.4% (13 pts) had already had spinal surgery. This question was not answered by 4.8% of participants (5 pts). Of the 13 patients operated on, 10 specified the number of previous operations, the median number is 2 (minimum 1 and maximum 5). The specific localization of the operation was reported by 12 patients. Two patients stated that they had undergone surgery on the cervical and thoracic spine. One patient underwent surgery exclusively on the cervical spine, while nine patients had procedures solely on the lumbar spine.

The patients were also queried regarding the duration of their pain and whether a specific event could be identified as its trigger. Among the respondents, 21.2% (24 pts) reported a pain duration of 1 month to 6 months. Pain persisting for 6 months to 1 year was noted by 18.4% of patients (21 pts), while 17.5% (20 pts) indicated that the onset of their pain occurred 2 to 5 years prior.

Data regarding the exact onset date of pain was available for 25.4% of patients (29 pts), while 19.3% (22 pts) reported the occurrence of a specific triggering event associated with the onset of pain. A pension application was submitted by 7.9% (9 pts). The median pain duration over the past 4 weeks was 28 days, with a range from 1 to 30 days. When asked for a desirable treatment goal, patients regarded a pain level of 2 (0 to 9) on the 11-point NRS as tolerable. The main results are summarized in Table 4.

All characteristics related to pain frequency and duration are presented separately for each survey time point in Table 5.

Prior to treatment, the majority of patients experienced pain multiple times per day (104 pts, 91.2%), with pain episodes lasting for several hours to more than three days (84 pts, 73.7%). Following the initial treatment, the proportion of patients experiencing pain multiple times daily and for durations of several hours to more than three days decreased to 52.0% (51 out of 98 pts) and 30.6% (30 out of 98 pts), respectively. Prior to the final treatment, these proportions were further reduced to 30.0% (21 out of 70 pts) and 24.3% (17 out of 70 pts), respectively. Three months after the final treatment, 47.2% (25 out of 53 pts) of patients reported experiencing pain multiple times daily, and 50.1% (26 out of 53 pts) reported pain lasting several hours to more than three days. The percentage of patients with pain less than once per month increased ten-fold from 1.8% (2 out of 114 pts) to 18.9% (10 out of 53 pts). Prior to treatment, 43.9% (50 out of 114 pts) of patients reported being actively employed, while 23.7% (27 out of 114 pts) were on sick leave at the time of the survey. Three months after the last treatment, the rate of employment was 54.7% (29 out of 53 pts), with only 1.9% (1 out of 53 pts) on sick leave.

The severity of the pain was assessed at each time point of the survey. The evaluation of the individual parameters at the individual survey times is summarized in Table 6.

In addition, the difference from the baseline value before the start of therapy was taken into account, as visualized in Fig. 3.

Change of pain levels reported by the patients using an 11-point numerical rating scale (0 = no pain, 1 = lowest and 10 = worst pain imaginable) after the first treatment, before the last treatment and 3 months afterwards, compared to baseline values before treatment. (A) change of pain levels at the respective survey time points. (B) change of mean pain levels over the last four weeks before the respective survey time points. (C) change of the maximum pain levels within the four weeks before the respective survey time points

Before treatment, the median intensities reported for the pain at the present moment, as well as the average pain in the four weeks prior was 7.0/10. Both reduced to 5/10 after the first treatment and to 3/10 before the last treatment. Three months following the last treatment, the median present pain intensity was reported as 4.0/10, while the average recent pain intensity was 5.0/10. Maximum pain intensity was also positively affected, with a median of 9/10 before treatment, 4.5/10 before the last treatment, and 8.0/10 three months after therapy. When compared to baseline, maximum therapy effect was found at the survey time point before the last treatment, with a reduction of 4.0 points for present as well as average recent pain intensity (95% CI −5.0 to −2.0), and 3.0 points for maximum pain intensity (95% CI −5.0 to −2.0). Three months post-therapy, median pain intensities remained reduced by a mean of 2.0 points for both present and average recent pain (95% CI −4.0 to −1.0), and by 1.0 point for maximum pain intensity (95% CI −2.0 to −1.0) (Fig. 3). As zero is not included in the 95% confidence intervals at every time point and for all parameters, it can be concluded that there is a significant reduction in pain intensity in all parameters and at every time point compared to baseline.

The impairments in everyday life, leisure activities and the ability to work were considered. The statistical parameters are shown in Table 7.

In addition, the difference from the baseline value before the start of therapy was considered as visualized in Fig. 4.

Change of impairment levels reported by the patients using an 11-point numerical rating scale (0 = no impairment, 1 = lowest and 10 = total impairment) after the first treatment, before the last treatment and 3 months afterwards, compared to baseline values before treatment. (A) change of impairment levels regarding everyday life activities at the respective survey time points. (B) change of impairment regarding leisure activities. (C) change of impairment regarding the ability to work

Throughout the course of therapy, pain-related impairment showed the greatest improvement at the time point preceding the final treatment, with a median reduction of 4.0 points in all three categories (95% CI −5.0 to −3.0). At three months after therapy, improvement persisted by 1.0 point for everyday life (95% CI −2.0 to 0.0), and 2.0 points for both leisure activities and fitness for work (95% CI −3.0 to −1.0). Except for the impairment in everyday life at the last observation time point 3 months after the end of therapy zero is not included in the 95% confidence interval at all other time points and for all other parameters. It can be concluded that there is a significant reduction in pain-related impairment for these parameters and the respective time points compared to baseline.

The evaluation of all questions about therapy are summarized in the following Table 8.

Throughout the course of therapy, about 17% of patients (17 out of 98 pts) subjectively reported side effects, this percentage seemingly increased to about 21% three months after therapy (11 out of 53 pts). It is important to note that, in the questionnaires, patients were instructed to indicate the presence of suspected side effects using checkboxes (yes or no) and to provide a narrative description of any adverse events in their own words. A total of 30 different patients reported side effects at different time points, not all of them consistently, leading to 45 different observations. Three patients answered the question only at time point 4, three months after end of therapy, and retrospectively added descriptions of events encountered at a previous survey time point. For further interpretation we first excluded observations that were obviously part of the intended treatment, i.e. anesthesia in the dermatome of the treated nerve root (n = 3). Most side effects reported were typical, expectable mild adverse events known to frequently occur with PRI and FJI, especially when corticoids are added, and patients had been informed about pre therapy, such as cortisone flush (reported 14 times) and short periods (< 24 h) of palsy corresponding to the nerve root treated. Interestingly, some patients reported side effects closely related to the therapeutic procedure exclusively in retrospect at the last survey 3 months after therapy, explaining an ostensible increase at the last time point. Two individuals consistently reported a side effect resembling a cortisone flush at each survey time point, including time point 4, where no treatment was administered. One of these individuals described a ‘reddening of the face’ as an allergic reaction to corticosteroids at time point 3, but reclassified it as a cortisone flush at time point 4. Additionally, one individual reported an ‘allergic reaction’ (burning sensation in face) at time point 2, but did not report any further adverse events in subsequent surveys. This is interpreted as a description of a typical cortisone flush. Two individuals reported worsening of pain. One of them additionally described heart burn, paresthesia of the left foot, and bilateral gluteal muscle ache, with worsening of the symptoms and additional “renal problems” at 3 months after therapy. The other one reported weak and trembling legs and described paresthesia in both legs and bilateral palsy for about 3 months, that returned to nearly normal at time point 4, three months after the last therapy. The remaining adverse events reported were rash at one arm (1), nose bleed (1), drowsiness/dizziness (3), nausea (4), fatigue and sweating (1), high blood pressure (1), weight gain before last therapy (2), weight loss after first therapy (1), headache (3), xerostomia (1), deterioration of oral microflora (1), and vasovagal reaction (2). Not all of these narratives can be attributed to the procedures themselves. Nevertheless, all these reported side effects were classified as mild adverse events, Class 1 according to the SIR AE severity score [16], and no additional treatment or unplanned hospital admissions were needed.

Prior to their final treatment, 88.6% (62 out of 70 pts) of patients considered the treatment successful, and 92.9% (65 out of 70 pts) indicated they would recommend the treatment to others. Three months post-treatment, 62.3% (33 out of 53 pts) of patients still considered the treatment successful, and 75.5% (40 out of 53 pts) would recommend the procedure. Following the first treatment, only one-third (33 out of 114 pts) of patients were able to initiate physiotherapy. This proportion increased to over 50% before the final therapy session and remained approximately 45% (24 out of 53 pts) three months after the final procedure.

Data on concomitant pain medication were collected through open-ended questions in the questionnaires at each survey time point. Since this information was not coded, no statistical analyses can be carried out. There was inconsistency in the reporting of the specific drugs, often inadequate information on the dosage, and only a minority of patients gave information on pain medication at every time point. Consequently, we are limited to providing a narrative summary of certain observations. Of 74 patients that reported their medication before the first treatment, 64 were given NSAIDs, and 19 were on opioids. In 38 patients, data on pain medication from the pre-interventional visit and at least one visit after the first treatment was available. Among these, 19 patients experienced a reduction in the frequency and/or dosage of oral pain medication. Of the four who were using opioids, two were able to discontinue them entirely, while the remaining two managed to reduce their opioid doses. Seventeen patients reported no change in their pain medication, with four of them remaining on opioids. Two patients, who stated they were on opioids but not NSAIDs before interventional treatment, reported increased opioid doses. One only gave information at survey time point 2, and information on pain medication after the end of therapy is missing. The remaining patient reported initiating NSAIDs three months after the final intervention.

As participation in the study declined over time, the sizes of the two subgroups were relatively small, as presented in Table 9.

For the vast majority of parameters compared, there were no substantial differences between the two groups for any parameter. The most important findings are summarized as follows. Median age was 60 years (39 to 80 years of age) in the FJI group and 57 years (22 to 81 years of age) in the PRI group (Mann Whitney U test: p = 0.34). Pain duration before study inclusion was reported as less than one month in 10.8% (4 pts) in the FJI group versus 22.0% (9 pts) in the PRI group, and as more than five years in 18.9% (7 pts) in the FJI group versus 4.9% (2 pts) in the PRI group (Chi square test p = 0.248). Only two parameters revealed statistically significant differences, however, only at one time-point in the treatment course each. Namely, maximum pain intensity after the first treatment was reported higher in patients receiving FJI with a median of 7 (minimum 2 and maximum 9) and a mean of 6.8 ± 1.99 compared to PRI with a median of 6 (minimum 1 and maximum 10; Mann-Whitney-U test, p < 0.05) and a mean of 5.7 ± 2.08. Three months after the last treatment, maximum pain intensities were reported as a median of 8.0 (minimum 1 and maximum 10) and a mean of 6.9 ± 2.11 in the FJI group versus a median of 6 (minimum 0 and maximum 10) and a mean of 6.6 ± 2.81 in the PRI group (p = 0.98). Patients receiving FJI reported significantly more side effects (5 pts, 27.8%) compared to FJI (1 pt, 3.8%) before the last treatment (Fisher’s exact test, p < 0.05). Three months later side effects were noted by 22.2% (4 pts) in the PRI group and 18.2% (4 pts) in the FJI group (p = 1.0).