Interventional procedure overview of neurostimulation of lumbar muscles for refractory non-specific chronic low back pain

Efficacy summary

Improvement in back pain

A randomised, blinded sham-controlled trial (RCT) of 204 patients with chronic mechanical low back pain (CMLBP) and impaired multifidus control who had an implanted neuromuscular restoration stimulator reported that the proportion of 'responders' in the intention to treat analysis (that is, more than 30% relief on the low back pain visual-analog-scale [VAS] without analgesics increase) at 120‑day follow up were not significantly different in the therapeutic stimulation group than in the low-level sham stimulation control group (57% compared with 47%; difference of 10%; 95% confidence interval [CI], -3.3% to 24.1%, p=0.138). (Gilligan 2021). After 2-year follow up, the proportion of patients experiencing ≥30% improvement in VAS was 82.6% (128/155) (95% CI 76.6% to 88.6%), the proportion of patients experiencing ≥50% improvement in VAS was 71.6% (111/155) (95% CI 64.5% to 78.7%), and the proportion of patients experiencing ≥70% improvement in VAS was 61.9% (95% CI 54.3% to 69.6%). The proportion of patients experiencing LBP resolution (VAS ≤ 2.5 cm) was 66.5% (103/155) after 2 years (95% CI 59.0% to 73.9%) (Gilligan 2021).

In the RCT, the average LBP-VAS improved from 7.3 ± 0.7cm at baseline to 4.0 ± 2.7cm in the therapeutic group and 4.8±2.9cm in the sham control group. The mean group difference was significantly in favour of the therapeutic stimulation treatment group (-3.3 compared with -2.4; difference of -0.9 cm; 95% CI -1.6 to -0.1 cm; p=0.032). The cumulative-proportion-of-responders analysis showed that therapeutic stimulation was superior to sham-control (p=0.0499). The difference in proportion of LBP-resolution (VAS less than 2.5 cm) was not statistically significant between the 2 groups (34% compared with 28%; difference of 6%; 95% CI -6.5 to 19.0%; p=0.335). At 1‑year follow up in the overall cohort (n=176), mean average LBP had improved by -4.3±2.6 cm (95% CI -4.7 to -3.8; p<0.0001) or -58.9±35.0% (95% CI -53.6 to -64.1%; p<0.0001) and 74% (130/176) of patients had a 30% or greater improvement; 64% of patients had a 50% or greater improvement; and 52% reported LBP-resolution (VAS≤2.5cm). A reduction in mean LBP VAS from 7.3±0.7 at baseline to 2.4±0.2 at 2-year follow up for completed cases (difference of -4.8± 0.2, 95% CI -4.6 to -3.8, or % difference of -66.7±2.6, 95% CI -71.7 to -61.6, p<0.0001) was reported. (Gilligan 2021).

A prospective case series of 53 patients with chronic mechanical low back pain (CMLBP), implanted with a neurostimulator for contraction of the lumbar multifidus, reported satisfactory improvement in low back pain (LBP). Back pain (7 day average, evaluated on a 10-point numerical rating scale [NRS], with 0 indicating no pain and 10 worse pain) reduced from 6.8±0.8 at baseline to -2.5 ± 0.3, (p < 0.0001) at 90‑day follow up. The responder rate (defined as patients with 2-point reduction in the mean NRS pain score from baseline to 90 days post-stimulation without a clinically meaningful increase in LBP medications) was 58% (30/52). The percentage of patients with an improvement of at least the minimal clinically important difference (MCID) of 2 points in LBP in the single day NRS (without a clinically meaningful increase in LBP medications at 90 days) was 63% (33/52), 61% (31/51), and 57% (27/47) at 90 days, 6 months and 1 year, respectively (Deckers 2018). After 4-year follow up, 73% of patients experienced a clinically meaningful improvement of at least the MCIC on NRS (defined as a change in at least 2 points). There was a reduction in mean NRS from 6.8±0.8 at baseline to 3.2±0.4 at 4 years (difference of 3.6, p<0.001) (Mitchell 2021).

A prospective case series of 42 patients with chronic mechanical low back pain (CMLBP), implanted with a neurostimulator for contraction of the lumbar multifidus, reported that the mean NRS improved from 7.0±0.2 at baseline to 3.5 ± 0.3 (p<0.0001) after 2-year follow up for complete cases. 68% (25/37) of patients experienced ≥30% improvement in NRS and 57% (21/37) experienced at least a ≥50% improvement in NRS after 2 years. 65% (24/37) reported mild to negligible pain (NRS≤3) after 2 years (Thomson 2021).

A case series of 28 patients with continuing chronic low back pain (CLBP) despite physical therapy and medical treatment and no prior surgery, implanted with pulse generators and leads, reported that average LBP (measured on a VAS 100 mm scale) improved significantly at 3 and 5 months follow up (therapy withdrawal phase) (decreased from 67.3 ± 11.1 mm at baseline to 40.8 ± 23.8 mm at 3 months, change of 26.4 ± 22.3 mm [p <0.0001] and 39.7 ± 33.4 at 5 months, change of 27.6±27.3 [p=0.0005]). A minimally important change of either ≥15 mm or ≥30% in VAS was reported in 74% (14/19) of patients and 67% (12/18) of patients at 3 and 5‑month follow up (Deckers 2015).

Oswestry Disability Index (ODI)

In the RCT of 204 patients comparing therapeutic stimulation (n=102) with low-level sham-stimulation (n=102), disability measured using the 100-point Oswestry Disability Index (ODI) showed that ODI scores were statistically significantly better in the therapeutic stimulation treatment group at 120‑day follow up compared with baseline. The ODI scores improved from 39.1±10.3 at baseline to 22.3±14.5 in the therapeutic group and 25.7±15.0 in the sham control group (mean difference -17.5±15.1 and -12.2±14.6; difference of -5.4 points between groups; 95% CI - 9.5 to -1.2 points; p=0.011). At 1‑year follow up, in the overall combined cohort (n=176), ODI scores improved by -19.9±15.8 points from baseline (95% CI -2.3 to -17.6; p<0.0001) or 50.5±38.7% (95% CI -44.8 to -56.3; p<0.0001). At 2-year follow up, a reduction in mean ODI from 39.1±10.3 at baseline to 17.6±1.2 for completed cases was reported (difference of -21.5, 95% CI -24.0 to -18.7 or % difference of -54.3%±3.2 (-95% CI -60.6% to -48.0%, p<0.0001). 61.3% (95/155) of patients experienced a ≥20 point improvement in ODI (95% CI 53.6% to 69.0%) (Gilligan 2021).

In the prospective case series of 53 patients, disability measured using the 100-point ODI (with scores of 21 to 40% indicating moderate disability and scores of 41 to 60% indicating severe disability) showed that the percentage of patients with MCID improvement of more than 10 points in ODI was 52% (27/52), 57% (29/51), and 60% (28/47) at 90 days, 6 months, and 1 year, respectively (Deckers 2018). After 4-year follow up, 76% of patients experienced a clinically meaningful improvement of at least the MCIC on ODI (defined as a change in at least 10 points). There was a reduction in mean ODI in completed cases from 44.9±10.1 at baseline to 23.0±0.4 at 4 years (difference of 21.9, p<0.001) (Mitchell 2021).

A prospective case series of 42 patients with chronic mechanical low back pain (CMLBP), implanted with a neurostimulator for contraction of the lumbar multifidus, reported a reduction in mean ODI for complete cases from 46.2±2.2 at baseline to 29.2±3.1 (difference of 17.0, p<0.0001) after 2 years. 51.4% of patients (19/37) experienced a ≥15 point improvement in ODI and 43.2% (16/37) of patients experienced a ≥20 point improvement in ODI after 2 years (Thomson 2021).

The case series of 28 patients reported that disability scores (measured using the 100-point ODI scale) significantly improved at 3 and 5‑month follow up (therapy withdrawal phase) (decreased from 38.5 ± 14.6 at baseline to 27.6 ± 15.6 at 3 months, change of 10.9 ± 9.6 [p = 0.0001] and 29.6 ± 29.3 at 5 months, change of 12.1±14.4 [p=0.0017]). A minimally important change of either ≥10 points or ≥30% on the ODI score was reported in 63% (12/19) of patients and 53% (10/19) of patients at 3 and 5‑month follow up. 45% (5/11) of patients on disability leave returned to work by 3 months (Deckers 2015).

Quality of life (QoL; EQ-5D)

In the RCT of 204 patients comparing therapeutic stimulation (n=102) with low-level sham-stimulation (n=102), quality of life measured using the European quality of life score on 5 dimensions (EQ-5D-5L) questionnaire showed that the EQ-5D-5L index scores were statistically significantly better in the therapeutic stimulation treatment group at 120‑day follow up compared with baseline. The scores improved from 0.585 ± 0.174 at baseline to 0.758 ± 0.160 in the therapeutic stimulation group and 0.713 ± 0.160 in the sham control group (mean difference 0.186 compared with 0.115; difference of 0.071 between groups; 95% CI 0.018 to 0.123; p=0.009). At 1‑year follow up, in the overall combined cohort (n=176), the EQ-5D-5L index improved by 0.198±0.207 (95% CI 0.167 to 0.229; p<0.0001). At 2 year follow up, an increase in EQ-5D from 0.585±0.174 at baseline to 0.798±0.013 in complete cases was reported (difference of 0.213±0.017, 95% CI 0.184 to 0.253, p<0.0001) (Gilligan 2021).

In the prospective case series of 53 patients, quality of life measured using the European quality of life score on 5 dimensions (EQ-5D) questionnaire showed that the percentage of patients with MCID improvement of at least 0.03 points in EQ-5D was 88% (46/52), 82% (42/51), and 81% (38/47) at 90 days, 6 months, and 1 year, respectively (Deckers 2018). After 4-year follow up, the mean EQ-5D increased from 0.434±0.185 at baseline to 0.721±0.035 (difference of 0.287, p<0.001) (Mitchell 2021).

A prospective case series of 42 patients with chronic mechanical low back pain (CMLBP), implanted with a neurostimulator for contraction of the lumbar multifidus, reported an increase in EQ-5D from 0.425±0.035 at baseline to 0.680±0.030 at 2 years for complete cases (difference of 0.254, p<0.0001) (Thomson 2021).

The case series of 28 patients reported that quality of life (measured using EQ-5D questionnaire) significantly improved at 3 and 5‑month follow up (at therapy withdrawal phase) (EQ-5D score increased from 0.43 ± 0.34 at baseline to 0.70 ± 0.21 at 3 months, improvement of 0.27 ± 0.24 points [p = 0.0002] and 0.20±0.43 at 5 months [p=0.06]). At 3 months, 84% (16/19) of patients reported an increase in EQ-5D scores and none reported a decrease (Deckers 2015).

Patient satisfaction

In the prospective case series of 53 patients, 89%, 84%, and 81% of patients reported being either 'satisfied' or 'very satisfied' with their treatment at 90 days, 6 months, and 1 year (Deckers 2018). At 4 years 97% (32/33) of participants reported being 'very satisfied' with the treatment (Mitchell 2021).

In the open label follow up of the RCT of 204 patients with chronic mechanical low back pain (CMLBP), 80% (124/155) of patients answered 'definitely satisfied' on the treatment satisfaction questionnaire (95% CI 73.7% to 86.3%) (Gilligan 2021).

In the RCT of 204 patients comparing therapeutic stimulation (n=102) with low-level sham-stimulation (n=102), the difference in proportion of patients 'definitely satisfied' (measured using a treatment satisfaction questionnaire) was statistically significantly better in the therapeutic stimulation treatment group at 120‑day follow up (61.0 compared with 40.0; difference of 21%; 95% CI 7.9 to 34.9%; p=0.002). At 1‑year follow up in the combined cohort (n=176), 78% of patients answered 'definitely satisfied' on the treatment satisfaction questionnaire (Gilligan 2021).