Pulsed radiofrequency or surgery for anterior cutaneous nerve entrapment syndrome: Long‐term results of a randomized controlled trial

Patients with anterior cutaneous nerve entrapment syndrome (ACNES) often require a step‐up treatment strategy including abdominal wall injections, pulsed radiofrequency (PRF) or a neurectomy. Long‐term success rates of PRF and surgery are largely unknown. The aim of the current study was to report on the long‐term efficacy of PRF and neurectomy in ACNES patients who earlier participated in the randomized controlled PULSE trial.

muscle and sensory innervation of overlying abdominal wall skin. 3,4CNES is characterized by an area of continuous abdominal pain that can be localized with a fingertip.Findings during physical examination are altered skin sensibility at the painful spot such as dysesthesia, changed cold perception, a disproportionate pain while pinching the skin (hyperalgesia), and a positive Carnett's sign at the tender point (maximum pain point, which can be localized with a fingertip). 5][7][8] After confirming the diagnosis ACNES, a step-up treatment strategy is advised.The initial step is an injection regimen using local anesthetic subfascial tender point injections (TPIs) over the course of several weeks. 8,9f pain persists or the patient is dissatisfied, minimal invasive pulsed radiofrequency (PRF) at the tender point is a second step. 10,11If these treatments fail to reach satisfactory pain relief, removal of terminal nerve branches may be considered at the level of the anterior muscle sheath (anterior neurectomy).][14] The potential role of PRF for ACNES was initially documented in a retrospective cohort study encompassing 26 patients.After a median 15 months follow-up, approximately one in four reported a successful outcome. 10 subsequent randomized controlled trial (RCT) comparing the effect of PRF with a neurectomy (PULSE trial) found that 38% of the PRF patients reported success after 8 weeks, and 13% after 6 months. 11Another study on the efficacy of the anterior neurectomy with a 32 months follow-up reported a considerably higher success rate up to 61%. 12 However, efficacy rates beyond these 15 and 32 months are unknown.The aim of the present study was to determine long-term success rates of PRF and neurectomy in ACNES patients previously participating in the PULSE trial, allowing for an optimization of the current treatment protocol.

M ET HOD S
For the present study, only patients who earlier participated in the PULSE trial, a RCT comparing PRF versus neurectomy as a second step in the treatment for ACNES, were considered eligible. 11Inclusion criteria of this RCT were age ≥18 years, unilateral ACNES, pain duration ≥3 months before diagnosis, and temporary pain relief following TPI.Between 2015 and 2018, a total of 60 patients were randomized to undergo either PRF or neurectomy.Patients received one of the two treatments with a crossover option to surgery if they were not satisfied 8 weeks after PRF.Final follow-up was performed 6 months after the last received treatment.
Both procedures were performed at the tender point, which was determined after a physical examination prior to the procedure.PRF was performed under ultrasound guidance to identify the anterior fascia of the rectus abdominis muscle.Once imaged, the RF cannula is positioned in the rectus abdominis muscle at the tender point, after which a very low electrical current was used to stimulate the affected nerve.If a cannula is positioned correctly, patients report paresthesia, numbness, or a tingling sensation in the area of the ACNES pain.When the patient confirmed this sensation, PRF was performed.The neurectomy was performed under general anesthesia.A transverse skin incision was made over the tender point, after which the subcutaneous area was dissected to expose the anterior sheath of the rectus abdominis muscle.Any perforating neurovascular bundles were ligated or coagulated within the tender point's 5 cm radius.Both procedures are described in detail in the PULSE trial. 11or the present study, all patients were contacted by phone.They provided verbal informed consent, considering that written informed consent was previously obtained for inclusion in the PULSE trial.Deceased patients were potentially included using their collected electronic medical data.Patients were excluded if no informed consent was given, or when telephone contact failed.If phone numbers were out of use, patients were contacted by email.The Daily Board of the Medical Ethics Committee Máxima MC confirmed that no further ethical approval was required due to the observational character of the study (local reference number N22.003).

Outcome measures
The 6-month follow-up questionnaires of the PULSE trial were used to compare baseline data of included and excluded patients to analyze potential selection bias.If a patient experienced some type of pain, they were asked if the pain was abdominal, and whether that pain was present at the location of the previous intervention for ACNES.Patients were asked to score their abdominal pain in the previous weeks with a mean Numeric Rating Scale (NRS) (range 0-10) as recommended by the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT). 15f patients had received additional treatment, type(s) of treatment, date(s), and effect of the treatment(s) were registered.The onset of ACNES at another location was classified as a separate disease.To determine treatment outcomes in chronic pain patients, we followed IMMPACT stating that treatment success was defined as >50% reduction or ≥4 points reduction using NRS.A reduction in NRS between 30% and 50% or ≥2 points was defined as moderate success.NRS reduction <30%, no change, or an increase in NRS was defined as failure. 16Patients were stratified and analyzed by their initial randomized treatment in the PULSE trial.
Primary outcomes were treatment outcomes comparing baseline NRS scores from the PULSE trial with current NRS scores, and the need for additional treatment(s) since the initial PRF or neurectomy.
Secondary outcomes were the time until re-intervention and number of re-interventions.Another outcome was treatment success at the end of follow-up based on treatment satisfaction measured by the Patient Global Impression of Change (PGIC). 15To determine if the treatment outcome based on NRS was in concordance with patient's perception of treatment outcome, treatment outcomes based on NRS and PGIC were compared.

Statistics
Due to the limited number of patients, data were expected to be skewed.Therefore, continuous and categorical data were presented as median and interquartile range (IQR; Q1-Q3) or absolute numbers and percentages, as appropriate.Mann-Whitney U-test was used to test categorical and continuous data.Fisher's exact test was used for binary data.Time until re-intervention was presented using Kaplan Meier curves and the Log Rank test was used to test equality of distribution for both curves.Correlation between treatment success defined by NRS and patient's perception of treatment success by PGIC was tested with Spearman's rank correlation.
All analyses were performed using SPSS software Version 22.0 for Windows (IBM, Armonk, NY, USA).

R E SU LT S
All 60 patients were approached by phone between May 2022 and July 2022.A total of 44 patients were reached.Verbal informed consent was obtained in all but one patient in the neurectomy group who refused participation.One deceased patient remained included.Therefore, a 73.3% response rate was attained (44/60).Median followup was 71.5 months (IQR 69.5-74.0).

PRF group
Figure 1 shows the results of the original 32 patients who were previously randomized to PRF in the PULSE trial and their subsequent treatments during follow-up.Of the 23 PRF patients included in the current study, 13 crossed over to a neurectomy at the 8 weeks' follow-up.Of the 10 remaining PRF patients, one decided not to crossover and refused additional treatment, although the outcome of the PRF treatment was unsuccessful.Eight of the remaining nine PRF patients had recurrent or ongoing pain and received one or more additional treatments including repeated PRF, TPI, or neurectomy.Those treatment(s) were successful in all but one.The one remaining patient (4.3%) was still free of pain without additional treatment 74 months after the received PRF treatment during the PULSE trial.Moreover, five of the PRF patients (21.7%) had a successful outcome without surgical intervention.

Neurectomy group
Figure 2 shows the results of all patients who were randomized to an anterior neurectomy in the PULSE trial (n = 28).Of the 21 neurectomy patients included in the current study, seven patients received additional treatment (PRF, posterior neurectomy) for ongoing or recurrent pain.At the end of the current follow-up, 15 patients (71.4%) reported a successful outcome.Two of these 15 patients had received additional treatment.Therefore, 13 patients (61.9%) had a successful outcome after the initial anterior neurectomy.

Re-interventions for ongoing or recurrent ACNES
Median time until the first re-intervention was significantly longer (p < 0.001) in the neurectomy group compared to the PRF group (69 months (IQR 15-73) vs. 3 months (IQR 2-10), Figure 3), suggesting recurrence of ACNES pain, rather than residual pain after a neurectomy.PRF patients required more additional re-interventions compared to the neurectomy group (median, two (IQR 1-3) vs. one (IQR 1-1.5), p = 0.043).All initial re-interventions occurred in the first 24 months after the initial treatment.

Patients' perception of treatment outcome
PGIC scores are presented in Figure 4. PGIC scores were obtained from 43 patients, as one patient was deceased in the neurectomy group.At the 71.5 months median follow-up, 36 of these 43 patients (83.7%) reported 'very much' or 'much improvement' of their pain, whereas four patients (9.3%) experienced a minimal improvement or no change at all.Three remaining patients (7.0%) experienced an increase in pain.Patient perception of treatment success based on PGIC was similar in both treatment groups (p = 0.865) and corresponded with the defined treatment success based on NRS scores (Spearman's rank correlation, r s (41) = 0.69, p < 0.001).Of note, the groups are based on the initially randomized treatment in the PULSE trial, PGIC scores were obtained after the additional treatments (as illustrated in Figures 1 and 2).

DI SC US SION
In 2019, the randomized controlled PULSE trial demonstrated that PRF treatment was an alternative treatment option for ACNES patients with persistent pain after an injection regimen.Although it reported that PRF success rates were inferior compared to a neurectomy, its minimal invasive nature makes PRF a viable option before embarking upon surgery. 11Since this RCT, PRF treatment has become a separate step in the treatment protocol in most patients with ACNES in our department.However, its long-term efficacy was unknown.The present study is the first to describe long-term outcomes (median of almost 6 years) of PRF treatment and neurectomy in patients with ACNES.
A very limited long-term success rate after PRF treatment was expected.Indeed, just one patient (4.3%) was still free of pain 6 years after a single PRF treatment.Previous studies of PRF for peripheral neuropathic pain conditions such as ACNES and pudendal neuralgia have generally shown a fading effect in most patients after 3 to 4 months. 10,17This finding is underscored by the present study demonstrating a median 3 month interval until re-intervention.Interestingly, five patients (22%) reported success after repeated PRF or TPI, without surgical intervention.It is concluded that one or a series of PRF treatments are long-term effective in approximately one in five ACNES patients.
The present study found that almost two-thirds (13 patients, 62%) of the population who received a neurectomy still reported success after 6 years.This result is similar to the 61% success rate at 32 months' follow-up of an earlier study encompassing 181 anterior neurectomies. 12This ongoing success rate is confirmed by the present study indicating that all primary re-interventions occurred within the first 2 years after the initial treatment (PRF or neurectomy).It may be concluded that recurrent pain after a successful treatment beyond this 2 year time period is rare.These findings are important in the counseling process in selected patients who consider to undergo a neurectomy.
Data on the overall long-term success rates of various therapies for ACNES are scarce.At the end of the current median 6 year follow-up period, success rates of both the initially randomized PRF and neurectomy group were similar, regardless of the needed re-intervention(s).However, type and number of re-interventions differed between both groups.By then, the overall success rate of the included patients was 80% (35/44).However, this success rate is only applicable to PRF and neurectomy treatment, not to complete treatment success rates.After all, patients who became free of pain after an injection regimen with TPI's were not included in the PULSE trial. 11Previous studies showed a significant, long-term pain relief in approximately 33% of the patients after injection regimen. 6,18The current 80% overall success rate is based on a subset of patients (67% of the total), who still experienced pain after the injection regimen.The combination of these success rates (33% and 80% of the remaining 67%), results in an overall success rate of 87% for the entire treatment strategy.This success rate is not far from an earlier published 90% success rate at the end of the complete treatment strategy. 14utcome after a treatment can be assessed using a variety of tools.Treatment outcome in the current study was based on NRS scores as suggested by IMMPACT recommendations. 16Additionally, patients were asked their perception of change in pain by the use of the PGIC questionnaire. 15Interestingly, the patients' perception was strongly related to treatment outcome as based on NRS (r s = 0.69).It may be concluded that the level of pain as scored using NRS by an ACNES patient is a good indication of the patient's own perception.
A limitation of this study is the restricted number of patients with a risk of bias, although the response rate was considerable (73%).This percentage is well above the expected 60% that is reported in most published questionnaire surveys. 19,20No significant differences were found between responders and non-responders (Table 2), reducing the risks of selection (non-response) bias.During the six follow-up years, different treatment strategies were explored resulting in heterogeneous treatment protocols (Figures 1 and 2).Any provided treatment sequence was not controlled but based on shared decision making, which is a daily practice at our outpatient clinic.Nonetheless, regardless of the different additional treatments, treatment outcomes at the end of the current Future research should explore other less invasive pain treatments as possible new treatment options for ACNES.An improved understanding of the pathophysiology is crucial in optimizing treatment protocols for ACNES.We conclude that PRF treatment is a long-term valuable minimal invasive treatment option for ACNES as a surgical intervention can be prevented in one in five patients.

T A B L E 1
Major and minor criteria for the diagnosis of ACNES, adapted from Scheltinga and Roumen. 5Abdominal pain in a circumscript area of the abdominal region Pain increases due to muscle tensing, movement, or exercise, or is related to posture Physical examination Somatosensory disturbances in skin area covering the painful area A positive pinch test A localized fingertip large maximal pain point situated between midline and lateral edge of the rectus abdominis muscle Deep palpation of this small area repeatedly evokes a predictable intense pain A positive Carnett's test Female Age 20-50 years Right-sided abdominal pain location Recent trauma including abdominal surgery or pregnancy Autonomic dysfunctions resulting in visceral manifestations such as nausea, bloating, swelling of abdomen, loss of appetite and body weight, or altered defecation patterns Painful areas along ipsilateral costal margin, in flank or paravertebral area (erector trunci)

F I G U R E 1
Flowchart of treatment outcome in patients randomized to pulsed radiofrequency.Numbers represent the absolute number of patients.AN, anterior neurectomy; Cryo, cryotherapy; PN, posterior neurectomy; PRF, pulsed radiofrequency; TENS, transcutaneous electrical neurostimulation; TPI, tender point injection.follow-upwere similar in both initial randomized treatment groups.

F I G U R E 2
Flowchart of treatment outcome in patients randomized to neurectomy.Numbers represent the absolute number of patients.PN, posterior neurectomy; PRF, pulsed radiofrequency.# One patient did not give informed consent.

F I G U R E 3
Rates of freedom from re-intervention after neurectomy or pulsed radiofrequency (PRF) as dictated by the original randomization.Number at risk are absolute numbers.PRF, pulsed radiofrequency.
Characteristics of included and excluded study patients who previously completed the PULSE trial (n = 60).