Background Posterior cervical foraminotomy and anterior cervical discectomy are routinely used operations to treat cervical brachialgia, although definitive evidence supporting superiority of either is lacking. Objective The primary objective was to investigate whether or not posterior cervical foraminotomy is superior to anterior cervical discectomy in improving clinical outcome. Design This was a Phase III, unblinded, prospective, United Kingdom multicentre, parallel-group, individually randomised controlled superiority trial comparing posterior cervical foraminotomy with anterior cervical discectomy. A rapid qualitative study was conducted during the close-down phase, involving remote semistructured interviews with trial participants and health-care professionals. Setting National Health Service trusts. Participants Patients with symptomatic unilateral cervical brachialgia for at least 6 weeks. Interventions Participants were randomised to receive posterior cervical foraminotomy or anterior cervical discectomy. Allocation was not blinded to participants, medical staff or trial staff. Health-care use from providing the initial surgical intervention to hospital discharge was measured and valued using national cost data. Main outcome measures The primary outcome measure was clinical outcome, as measured by patient-reported Neck Disability Index score 52 weeks post operation. Secondary outcome measures included complications, reoperations and restricted American Spinal Injury Association score over 6 weeks post operation, and patient-reported Eating Assessment Tool-10 items, Glasgow–Edinburgh Throat Scale, Voice Handicap Index-10 items, PainDETECT and Numerical Rating Scales for neck and upper-limb pain over 52 weeks post operation. Results The target recruitment was 252 participants. Owing to slow accrual, the trial closed after randomising 23 participants from 11 hospitals. The qualitative substudy found that there was support and enthusiasm for the posterior cervical FORaminotomy Versus Anterior cervical Discectomy in the treatment of cervical brachialgia trial and randomised clinical trials in this area. However, clinical equipoise appears to have been an issue for sites and individual surgeons. Randomisation on the day of surgery and processes for screening and approaching participants were also crucial factors in some centres. The median Neck Disability Index scores at baseline (pre surgery) and at 52 weeks was 44.0 (interquartile range 36.0–62.0 weeks) and 25.3 weeks (interquartile range 20.0–42.0 weeks), respectively, in the posterior cervical foraminotomy group (n = 14), and 35.6 weeks (interquartile range 34.0–44.0 weeks) and 45.0 weeks (interquartile range 20.0–57.0 weeks), respectively, in the anterior cervical discectomy group (n = 9). Scores appeared to reduce (i.e. improve) in the posterior cervical foraminotomy group, but not in the anterior cervical discectomy group. The median Eating Assessment Tool-10 items score for swallowing was higher (worse) after anterior cervical discectomy (13.5) than after posterior cervical foraminotomy (0) on day 1, but not at other time points, whereas the median Glasgow–Edinburgh Throat Scale score for globus was higher (worse) after anterior cervical discectomy (15, 7, 6, 6, 2, 2.5) than after posterior cervical foraminotomy (3, 0, 0, 0.5, 0, 0) at all postoperative time points. Five postoperative complications occurred within 6 weeks of surgery, all after anterior cervical discectomy. Neck pain was more severe on day 1 following posterior cervical foraminotomy (Numerical Rating Scale – Neck Pain score 8.5) than at the same time point after anterior cervical discectomy (Numerical Rating Scale – Neck Pain score 7.0). The median health-care costs of providing initial surgical intervention were £2610 for posterior cervical foraminotomy and £4411 for anterior cervical discectomy. Conclusions The data suggest that posterior cervical foraminotomy is associated with better outcomes, fewer complications and lower costs, but the trial recruited slowly and closed early. Consequently, the trial is underpowered and definitive conclusions cannot be drawn. Recruitment was impaired by lack of individual equipoise and by concern about randomising on the day of surgery. A large prospective multicentre trial comparing anterior cervical discectomy and posterior cervical foraminotomy in the treatment of cervical brachialgia is still required. Trial registration This trial is registered as ISRCTN10133661. Funding This project was funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 27, No. 21. See the NIHR Journals Library website for further project information. Plain language summary Cervical brachialgia is pain that starts in the neck and passes down into the arm. Although most people with cervical brachialgia recover quickly, in some patients pain persists, and in 15% of patients pain is so severe that they are unable to work. In the posterior cervical FORaminotomy Versus Anterior cervical Discectomy in the treatment of cervical brachialgia trial, we investigated two neck surgeries used to treat this problem: posterior cervical foraminotomy (surgery from the back of the neck) and anterior cervical discectomy (surgery from the front of the neck). This trial aimed to find out if one of them is better than the other at relieving pain and more cost-effective for the National Health Service. We assessed patients’ quality of life 1 year after their surgery and how their pain changed over the course of the year. We also measured the number of complications patients had in the first 6 weeks after their operation. Recruitment was slow and so the trial was stopped early, after only 23 patients from 11 hospitals had been randomly allocated to the two surgery groups. We had planned to recruit 252 participants to the trial; the number of participants we were able to recruit in practice was too small to enable us to determine which surgery is better at relieving pain. To find out why the trial had struggled to recruit, we asked hospital staff and participants about their experiences. We found that hospital staff sometimes struggled to organise everything needed to randomise patients on the day of surgery. Some staff also found it difficult to randomise patients as they had an opinion on which surgery they thought the patient should receive. The data collected in the trial will still be useful to help design future research. Finding out which surgery is better at relieving pain remains important, and the data we have collected will support answering this question in future. Scientific summary Background As people age, degenerative disease affects the intervertebral discs, facet joints and ligamentum flavum in the cervical spine. Consequently, the nerve root foramina may change shape and the nerve roots become compressed, causing a syndrome called cervical radiculopathy. Cervical radiculopathy is characterised by neck pain and cervical brachialgia, with the symptom of pain originating in the neck and radiating down into the upper arm. The reported incidence of cervical brachialgia is 1.79 cases per 1000 per year, with > 110,000 cases of brachialgia annually in the UK. Cervical brachialgia typically affects people aged 40–60 years, with up to 15% of patients unable to work because of the pain. In most patients, brachialgia is self-resolving with conservative management strategies. However, if symptoms persist 6 weeks to 3 months after onset, two of the available surgical techniques used (in the UK) to treat cervical brachialgia are posterior cervical foraminotomy (PCF) and anterior cervical discectomy (ACD). The most common and current standard operation is ACD, approaching from the front of the neck. The procedure is effective, but there is a high incidence of significant, potentially permanent, complications, including dysphagia and hoarse voice, which can be devastating to a patient’s quality of life. Alternatively, PCF can be undertaken from the back of the neck. Unlike ACD, PCF avoids risk to the structures in front of the spine including the carotid artery, sympathetic trunk, recurrent laryngeal nerve, larynx and pharynx, but may result in higher levels of postoperative neck pain and a greater need for revision surgery. Controversy therefore exists over which procedure is superior for the treatment of cervical brachialgia, and the decision on which procedure to use is frequently left to surgeon preference. The posterior cervical FORaminotomy Versus Anterior cervical Discectomy in the treatment of cervical brachialgia (FORVAD) trial aimed to robustly compare the two procedures, with the ambition of providing definitive evidence that would guide surgical decision-making for patients with cervical brachialgia symptoms requiring surgical intervention. Objectives The primary objective of the trial was to determine whether PCF is superior to ACD in terms of improving clinical outcome, as measured by the Neck Disability Index (NDI) at 52 weeks post operation, among patients with persistent cervical brachialgia for whom conservative management has failed. The secondary objectives were comparison of the two surgeries in terms of NDI score, patient-reported neck and upper-limb pain, dysphagia and globus, hoarse voice symptoms, incidence of revision surgery, cost effectiveness over 52 weeks post operation, the extent and severity of a patient’s spinal cord functional impairment, and incidence of surgical complications up to 6 weeks post operation. The exploratory objectives were to explore the impact of variations in the optional surgical components of PCF (open or minimal-access surgery) and ACD (surgery with or without a plate) on NDI and EuroQol-5 Dimensions, three-level version (EQ-5D-3L) scores. Design The FORVAD trial was a UK multicentre, Phase III, parallel-group, superiority, individually randomised controlled trial comparing the clinical effectiveness and cost effectiveness of PCF with those of ACD among patients experiencing symptomatic unilateral cervical brachialgia for at least 6 weeks, with confirmed nerve root compression on magnetic resonance imaging or computerised tomography myelography. Neither participants nor medical or clinical trial staff in the FORVAD trial were blinded to the treatment allocated. The trial incorporated an internal pilot phase to assess the feasibility of trial delivery against prespecified recruitment criteria. Participants were randomised (1 : 1) to receive PCF or ACD via minimisation incorporating a random element, with minimisation factors of centre, duration of upper-limb symptoms and smoking status. Participants were followed up in clinic at day 1 and 6 weeks post operation, and by post at 12, 26, 39 and 52 weeks post operation. The target sample size was 252 participants (126 per trial arm). This number was required to have 90% power to detect the minimum clinically important difference of 10% (5 points) in the change in NDI score at 52 weeks post operation, assuming a between-patient standard deviation of 23 units, two-sided 5% significance level and 10% loss to follow-up. For the analysis of the primary outcome measure, it was intended to use a multilevel linear regression model incorporating random effects with respect to centre, and adjusting for baseline (day 0) NDI score and minimisation factors (duration of upper limb symptoms and smoking status). The statistical analysis plan was amended prior to final analysis of the data to account for the restricted sample size owing to the early closure of the trial and was limited to descriptive summaries. Setting The trial aimed to recruit from 15 NHS hospitals throughout the UK. Participating surgeons were expected to perform both trial procedures and were required to have performed a minimum of 10 of each surgical procedure and to have completed a bespoke training package on the e-brain platform [www.ebrain.net (accessed 25 April 2022)]. Participants Patients were eligible to participate if they had been diagnosed with unilateral cervical brachialgia that had persisted for at least 6 weeks, resultant from single-level nerve entrapment, and conservative management had previously failed. Patients with cervical myelopathy, spinal cord compression or who had previously had cervical spine surgery were excluded. Eligibility waivers were not granted in this trial. Interventions Pre-operative investigations and preparation were as per individual site protocol. Participants received either PCF or ACD. Minimal- and open-access techniques were permitted for PCF. For ACD, the choice of fusion material and the decision to use a plate were left to surgeon discretion. Postoperative care was also as per individual site protocol. Main outcome measures The primary outcome measure was the patient-reported percentage NDI score at 52 weeks post operation. Secondary patient-reported outcomes (collected at baseline; at days 1 and 6; and at 12, 26, 39 and 52 weeks post operation) were collected using the Numerical Rating Scale–Neck Pain (NRS-NP), the Numerical Rating Scale–Arm Pain (NRS-AP), the validated PainDETECT, the Eating Assessment Tool-10 items (EAT-10), the Glasgow–Edinburgh Throat Scale (GETS) and the Voice Handicap Index-10 items (VHI-10) tools. Secondary clinical outcomes included a restricted version of the American Spinal Injury Association (ASIA) assessment scale, intraoperative and postoperative complications, incidence of reoperations within 52 weeks of operation and assessment of hoarse voice from additional voice recordings collected from randomly selected participants using the grade, roughness, breathiness, asthenia and strain scoring system. Exploratory outcomes included whether participants receiving PCF received minimal- or open-access surgery, and whether participants receiving ACD received surgery with or without a plate. Patient-reported outcomes were recorded using the EQ-5D-3L, which, alongside collected health resource use data, informed cost-effectiveness analysis. Results Owing to slower than expected accrual, the trial closed to recruitment after randomising 23 participants, 14 to PCF and 9 to ACD, from 11 sites. Therefore, results from the trial should be interpreted with caution because of the small sample size. The median NDI scores at baseline and 52 weeks were 44.0 [interquartile range (IQR) 36.0–62.0] and 25.3 (IQR 20.0–42.0) in the PCF group, respectively, indicating a reduction (improvement) from baseline. In the ACD group, the median scores at baseline and 52 weeks were 35.6 (IQR 34.0–44.0) and 45.0 (IQR 20.0–57.0), respectively. Unlike the PCF group, NDI scores did not appear to improve from baseline in the ACD group. For the NRS-NP and NRS-AP, there was an initial increase from baseline in neck pain score on day 1 post operation in both groups [median 5.5 (IQR 4.0–8.0) at baseline to 8.5 (IQR 6.0–10.0) at 1 day post operation in the PCF group; median 5.0 (IQR 4.0–7.0) at baseline to 7.0 (IQR 4.0–9.0) at 1 day post operation in the ACD group]. Neck pain decreased after day 1 in both treatment groups; at 52 weeks, the median scores were 4.0 (IQR 2.0–5.0) in the PCF group and 5.0 (IQR 3.0–7.0) in the ACD group. In both groups, arm pain improved on day 1 [median 3.0 (IQR 2.0–8.0) in the PCF group; median 4.0 (IQR 0.5–5.0) in the ACD group] and reached its lowest level 12 weeks post operation [median 3.0 (IQR 2.0–8.0) in the PCF group; median 2.5 (IQR 0.0–5.0) in the ACD group]. Thereafter, arm pain deteriorated (increased), but remained below baseline [median 5.0 (IQR 2.0–7.0) in the PCF group and median 5.0 (IQR 3.0–6.0) in the ACD group at 52 weeks]. PainDETECT category scores fluctuated over the postoperative period in the PCF group, but reduced over time in the ACD group, suggesting that in the ACD group a higher proportion of the participants had developed nociceptive, rather than neuropathic, pain. The ASIA score remained unchanged in both treatment groups over the 6-week assessment period. For the EAT-10 and the VHI-10 outcomes, the ACD group had worse outcomes at day 1 [median EAT-10 scores of 0.0 (IQR 0.0–4.0) in the PCF group and 13.5 (IQR 3.5–16.0) in the ACD group, and median VHI-10 scores of 0.0 (IQR 0.0–2.0) in the PCF group and 2.0 (IQR 0.5–8.5) in the ACD group], after which the two groups are comparable. However, the median GETS score was worse in the ACD group postoperatively, although the IQRs in each treatment group overlap at all time points except day 1. Five postoperative complications were reported in five participants throughout the trial, all occurring in the ACD group. Reported complications included two instances of dysphagia and three ‘other’ complications: ‘wound infection’, ‘urinary retention’ and ‘wound redness stitches overnight’. No serious or unexpected serious complications, deaths or reoperations were reported. The health economics and qualitative study The revised aims of the cost-effectiveness analysis were to describe the costs of the surgical interventions, health-care service use and participant out-of-pocket expenditures, and the productivity costs of losses associated with the surgery and its consequences. Responses to the EQ-5D-3L questionnaire of generic health-related quality of life and the rate of data completion are summarised. All costs are based on 2019/20 prices and are presented without discounting, as the time horizon of the analysis is 52 weeks. The costs of the intervention were heavily influenced by outliers in the PCF group, resulting in mean health-care costs of initial surgery, including devices, operation and hospital stay, of £2745 [95% confidence interval (CI) £2344 to £3147] for the PCF group (n = 12) and £4295 (95% CI £3436 to £5154) for the ACD group (n = 8). The corresponding median costs were £2622 (IQR £2402–2824) for participants undergoing PCF and £4423 (IQR £3849–4821) for participants undergoing ACD. These costs were driven by the time in theatre, which was a mean of 30 minutes (median 45 minutes) shorter for PCF (mean 61 minutes, median 52 minutes, IQR 49–60 minutes) than for ACD (mean 91 minutes, median 100 minutes, IQR 75–109 minutes). Results aggregated up to week 6 are presented owing to high attrition thereafter. The mean costs from the NHS and Personal Social Services perspective were £2716 (95% CI £2345 to £3087) for the PCF group and £4133 (95% CI £3099 to £5167) for the ACD group; the median costs were £2634 (IQR £2444–2741) and £4214 (IQR £3602–4994) for the PCF and ACD groups, respectively. The mean per-patient costs to society were £4608 (95% CI £2514 to £6703) for the PCF group and £5015 (95% CI £2286 to £7743) for the ACD group; the median costs were £4097 (IQR £2448–6591) and £4143 (IQR £4126–4284) for the PCF and ACD groups, respectively. At baseline, severe problems with anxiety and depression were more common, and pain and discomfort according to the EQ-5D-3L classification system was more severe in the PCF group than in the ACD group, resulting in mean utility scores of 0.291 (95% CI 0.07 to 0.51) and 0.595 (95% CI 0.40 to 0.78) for PCF and ACD, respectively. The corresponding median scores were 0.210 (IQR –0.01 to 0.60) and 0.689 (IQR 0.66–0.69). This is likely to have played a role in the larger observed gains from baseline in postoperative EuroQol-5 Dimensions (EQ-5D) scores for participants in the PCF group than in the ACD group [median change at 6 weeks of 0.10 (IQR 0.00–0.13) and 0.02 (IQR –0.03 to 0.06), respectively]. A rapid qualitative study was conducted during trial close-down to understand the experiences of health-care professionals and participants who participated in the FORVAD trial and why recruitment had been challenging, with the aim of informing future research in this area. Semistructured interviews were conducted with 18 health-care professionals (research nurses and surgeons) and two participants who had participated in the FORVAD trial. Interviews explored participants’ experiences of the FORVAD trial and their reasons for taking part, and staff experiences of recruiting to the FORVAD trial and neurosurgery trials in general. Interviews were audio-recorded and transcribed verbatim. Transcripts were analysed using rapid qualitative analysis. There was no single key factor that limited recruitment, and several themes were identified as common to all sites. Surgeons at participating sites supported the trial, and recognised collective clinical equipoise; however, many had preferences for one or the other procedure, linked to their usual practice. Organisation of the trial recruitment pathway varied, with some sites choosing to direct potentially eligible patients to dedicated clinics and other sites taking a more ad hoc approach. In the FORVAD trial, the dedicated clinic approach appeared to contribute to more eligible patients being identified and recruited, although staff at other sites explained that dedicated clinics did not fit easily with their clinical pathways. Randomisation on the day of surgery presented ethical and organisational challenges, and, where possible, should be avoided in future surgical trials. Organisation and implementation of a surgical trial in neurosurgery are complex and present many challenges. Future trials in neurosurgery should identify aspects of the protocol where it is possible to offer flexibility and ensure early multidisciplinary involvement at sites to maximise the effective integration of trial and clinical pathways. Conclusions The data suggest that PCF may be associated with better outcomes, fewer complications and lower cost, but the trial recruited slowly and was closed early. As a consequence, the trial is underpowered and definitive conclusions cannot be drawn. Trial recruitment was impaired by the lack of individual equipoise and concern about randomising on the day of surgery. A large prospective multicentre trial comparing ACD and PCF in the treatment of cervical brachialgia is still required. Trial registration This trial is registered as ISRCTN10133661. Funding This project was funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 27, No. 21. See the NIHR Journals Library website for further project information.