Nucleoplasty for cervical radicular pain due to disc herniation

Cervical radicular pain due to disc herniation presents with pain in the neck and one arm, with muscle weakness with or without numbness or tingling in the fingers or hands. Conservative treatment includes rest, analgesics, non-steroidal anti-inflammatory drugs, exercises and cervical collar. When conservative treatment fails, surgery is considered. Surgery can carry risks, and freedom from pain is not guaranteed. Recently, nucleoplasty, a new treatment for contained disc herniations, was developed. Nucleoplasty is a minimally invasive outpatient procedure that relieves nerve pressure by removing small portions of the disc's gel-like nucleus, with no reported neurological complications.

To assess the effect of nucleoplasty on pain, function, quality of life, recovery, adverse events and withdrawals due to adverse events compared to placebo, no treatment, conservative treatment, minimally invasive interventions or surgery for people with cervical radicular pain due to disc herniation.

We used CENTRAL, MEDLINE, seven other databases and two trial registers, together with reference checking, citation searching and contact with study authors and experts in the field to identify the studies that are included in the review. The latest search date was 24 February 2025.

We included randomised controlled trials (RCTs) that investigated nucleoplasty compared to placebo/sham treatment, no treatment, conservative treatment, minimally-invasive interventions or surgery for people with cervical radicular pain due to disc herniation. Major outcomes were pain in the arm and neck, neck-related function, recovery, quality of life, adverse events and withdrawals due to adverse events. The primary time point was short-term follow-up (up to three months).

Two review authors independently screened the references. We used the original Cochrane risk of bias tool for RCTs to assess the risk of bias of the included studies. We used GRADE to assess the certainty of the evidence.

We included four RCTs (259 participants). We judged all four studies to have an overall high risk of bias, due either to a high risk of detection or attrition bias. Three of four studies were at high risk of detection bias for unblinded outcome assessors. Nucleoplasty versus no treatment or placebo We did not find any RCT for this comparison. Nucleoplasty versus conservative treatment (1 study, 120 participants) Low-certainty evidence, downgraded for risk of bias and imprecision, showed that nucleoplasty may result in a large reduction in pain (0 to 100 scale, 0 = no pain) at short-term follow-up. The mean change from baseline in pain was 30.45 points lower with conservative treatment and 53.16 points lower with nucleoplasty (mean difference (MD) 22.71 points lower (95% confidence interval (CI) 30.10 lower to 15.32 lower)). Low-certainty evidence, downgraded for risk of bias and imprecision, showed that nucleoplasty may result in no difference in neck-related function at short-term follow-up (Neck Disability Index (NDI) 0 to 50, lower scores indicate less disability). The mean change from baseline function was 9.27 points lower in the conservative treatment group and 11.75 points lower in the nucleoplasty group (MD 2.48 lower, 95% CI 5.11 lower to 0.15 higher). Low-certainty evidence, downgraded for risk of bias and imprecision, showed that, compared to conservative treatment, nucleoplasty may result in little to no difference in quality of life (36-item Short-Form Health Survey, mental component summary (SF-36 MCS), 0 to 100, 100 = best score) in the short term. The mean change in quality of life from baseline was 8.04 points with conservative treatment and 6.31 points with nucleoplasty (MD 1.73 lower, 95% CI 5.32 lower to 1.86 higher). Compared to conservative treatment, it is uncertain if nucleoplasty increases the risk of adverse effects. This study did not report on recovery and there were no withdrawals. Nucleoplasty versus pulsed radiofrequency of the dorsal root ganglion (1 study, 34 participants) We are uncertain if, compared to pulsed radiofrequency, nucleoplasty has any effect on pain (0 to 100 scale, 0 = no pain; MD 7.9 lower, 95% CI 29.45 lower to 13.65 higher), neck-related function (0 to 50, 0 = best score; MD 0.30 higher, 95% CI 6.97 lower to 7.57 higher), recovery (MD 5.10 lower, 95% CI 29.92 lower to 19.72 higher) or adverse events (risk ratio (RR) 1.0, 95% CI 0.17 to 5.83) at short-term follow-up, due to very low-certainty evidence (downgraded for risk of bias, imprecision and indirectness). Nucleoplasty versus discectomy (2 studies, 105 participants) Low-certainty evidence, downgraded for risk of bias and imprecision, showed that nucleoplasty may result in little to no difference in neck pain at short-term follow-up (MD 0.33 points higher, 95% CI 0.36 lower to 1.03 higher). We are uncertain if nucleoplasty has any effect on arm pain (MD 0.74 points lower, 95% CI 1.23 lower to 0.25 lower), neck-related function (MD 0.69 points lower, 95% CI 12.63 lower to 11.25 higher), recovery (RR 0.81, 95% CI 0.51 to 1.29; 1 RCT, 48 participants), quality of life (MD 0.83 points higher, 95% CI 8.47 lower to 10.13 higher; 1 RCT, 48 participants) or adverse events (RR 0.14, 95% CI 0.01 to 2.62) compared to discectomy at short-term follow-up, due to very low-certainty evidence downgraded for risk of bias, imprecision and indirectness. No withdrawals due to adverse events were reported.

Compared to conservative treatment, low-certainty evidence showed that nucleoplasty may result in a large reduction in pain and no difference in neck-related function at short-term follow-up. For the other comparisons (pulsed radiofrequency of the dorsal root ganglion, discectomy), there was low to very low-certainty evidence for little to no effect of nucleoplasty on pain, neck-related function, recovery or adverse events. No serious complications occurred in the nucleoplasty or comparison groups. There is insufficient evidence to support the use of nucleoplasty for people with radicular pain due to disc herniation. There is a need for sufficiently powered, well-designed RCTs.