Table 3Benefits and Comparative Benefits of Nonpharmacological Therapies

Key question/intervention Recommendation Strength of recommendation Quality of evidence

PICO 1. For patients with lumbar spinal stenosis, should multimodal rehabilitation interventions versus another treatment be used to decrease pain, and improve function, quality of life, and return to function?

Multimodal therapy For patients with LSS and neurogenic claudication with or without LBP, we suggest offering a combination of education and advice, manual therapy and home-based exercise for improvement in walking capacity and symptoms/physical function in the short and long term. Conditional/Weak Moderate
(⊕⊕⊕O)

Definition: Multimodal rehabilitation interventions may include sedentary and nutrition lifestyle changes, ⁷¹^,¹²⁵ behavioral change techniques in conjunction with manual therapy, exercise and/or rehabilitation, and ancillary non-pharmacological treatment.

Included studies: We identified 3 RCTs ³^,⁸⁶^,¹⁰⁸ in which a comprehensive program, including various combination of self-management strategy, with or without cognitive behavioral approach, patient education and advice to stay active, supervised and home exercises (strengthening, stretching, and conditioning exercises, and stationary cycling), and manual therapy (thrust and non-thrust manipulation, manual spine stretching) was compared to home exercises or to medical care plus exercise (Appendix 6, Table 1).

Primary outcomes: Functional disability (ODI), leg pain (NRS), physical performance scale of the Zurich Claudication Questionnaire (ZCQ) or Swiss Spinal Stenosis (SSS) questionnaire, physical function (SF-36), and walking distance (Self-Paced Walk Test (SPWT))/gait disturbance (Japanese Orthopaedic Association Back Pain Evaluation Questionnaire).

Key results: In one RCT (Ammendolia 2018) ³, the adjusted mean difference (MD) in walking distance in the comprehensive group vs the self-directed group was 304.1 m (95% CI, 77.9 to 530.3) at 3 mo and 421.0 m (95% CI, 181.4 to 660.6) at 6 mo. At 6 mo, 82% of participants in the comprehensive group and 63% in the self-directed group achieved the MCID, (adjusted RR 1.3; 95% CI, 1.0 to 1.7). Both primary treatment effects persisted at 12-mo favoring the comprehensive program. At 6-mo, the comprehensive program showed significantly greater improvements in the ODI walk scale (-0.8; 95% CI, -1.3 to -0.4) and at 12-mo in the ZCQ, SF-36 physical function and bodily pain scores.

In one other RCT (Schneider 2019), ¹⁰⁸ manual therapy/individualized exercise had greater, but non-clinically important improvement of symptoms/physical function (Swiss Spinal Stenosis (SSS) questionnaire) at 2 mo, compared to medical care (adjusted mean difference -2.0; 95% CI: -3.6 to -0.4) or group exercise (-2.4; 95% CI: -4.1 to -0.8). Using the >30% responder criterion (secondary responder analyses), manual therapy/exercise had a greater proportion of responders in symptoms/physical function (20%; omnibus P = .002) and walking capacity (Self-Paced Walking Test) (65.3%; omnibus P = .04) at 2-mo compared to medical care (7.6% and 48.7%) or group exercise (3% and 46.2%). Group exercise also had greater improvement in average daily physical activity (armband accelerometer) at 2 -mo compared to medical care (28.7; 95% CI: 2.7 to 54.7). At 6-mo, there were no between-group differences in mean outcome scores or responder rates.

In the third RCT (Minetama (2019),⁸⁶ the supervised physical therapy group showed significant greater improvement at 6 wk vs home exercise in ZCQ symptom severity and physical function (mean difference (MD) −0.4; 95% CI: −0.6 to −0.2), walking distance on the SPWT (MD 455.9 m; 95% CI: 308.5 to -603.2), leg pain (MD −1.4; 95%CI: −2.5 to −0.3), gait disturbance (MD 16.0; 95%CI: 5.4 to -26.7), and physical functioning (MD 9.2; 95%CI: 2.1 to -16.3).

Comment: The panel determined a moderate certainty in the evidence, with minor and transitory undesirable effects and no major adverse events reported.

Remarks: Multimodal rehabilitation intervention was delivered twice weekly over 6 wk. It included individualized instruction on exercise and self-management strategies using a cognitive behavioral approach. At the end of the program, daily home exercise (30 min cycling plus 30 min of structured exercises) and self-care strategies should be maintained.³^,¹⁰⁸

PICO 2. For patients with lumbar spinal stenosis, should acupuncture versus another treatment be used to decrease pain, and improve function, quality of life, and return to function?

Acupuncture For patients with LSS and neurogenic claudication with or without LBP, we suggest considering traditional acupuncture on a trial basis to improve pain and physical function in the short-term. Conditional/Weak Very low
(⊕000)

Definition: Needle acupuncture (eg, Hwato Acupuncture, Suzhou, China; 0.30×40 mm/0.30×75 mm) at various sites (eg, Acupoints of Shenshu (BL23), Dachangshu (BL25), Weizhong (BL40), Chengshan (BL57), and Taixi (KI3)) ¹⁰³ or outward from the spinous process bilaterally at L2, L4, S2, and S4, middle of the popliteal fossa, inferior recess in the fibular head, lower end of the groove of the inner and outer head of the gastrocnemius).⁸⁷

Included studies: We identified 2 RCTs ⁸⁷^,¹⁰³ investigating the effect of acupuncture in patients with NC caused by LSS (Appendix 6 Table 2).
Primary outcomes: Physical function (RMDQ) and physical performance (ZCQ)

Key results: A RCT by Qin et al (2020) ¹⁰³ compared acupuncture to noninsertive sham acupuncture for 24 treatments over 8-wk in patients LSS with NC. The acupuncture group showed significant greater improvement in disability at 8 wk (adjusted mean difference (MD) -2.6 [95% CI, -3.7 to -1.4]) and at 3 mo (MD -2.3 [95% CI, -3.9 to -0.7], but not at 6-mo. The acupuncture group also showed greater improvement in leg and buttock pain intensity (NRS) at 8 wk (MD -2.9 [95% CI, (-3.8 to -2.0)], 3 mo MD -2.4 [95% CI, -3.3 to -1.4)] and 6 mo (MD -2.1 [95% CI, -3.0 to -1.2]), and back pain (NRS) at 8 wk (MD -2.3 [95% CI, -3.0 to -1.5]) and 3 mo [95% CI -1.7 (-2.6 to -0.8]).

A low-quality comparative study by Oka et al (2018) ⁸⁷ assigned 119 Japanese patients with LSS and L5 radiculopathy (mixed population) to receive either acupuncture (5 sessions in one month), back flexion exercises and an educational manual or pain medication (acetaminophen). Significant reduction in symptom severity was observed in all 3 groups, while improved physical function was found in the acupuncture group only (MD − 2.1, 95% CI − 0.40 to − 0.01). The acupuncture group also demonstrated better physical function compared to exercise group at 1 month (between-group difference in ZCQ least-square mean = 2.17, P = .02).

Comment: The panel determined a moderate certainty in the evidence, with minor and transitory undesirable effects and no major adverse events reported. The most frequently reported transient minor adverse events were worsening of symptoms, general discomfort, pain at the treated areas, and body ache. ⁵⁷ The resources required for an acupuncture intervention are relatively small (cost of care and equipment needed), with the exception of training and certification to provide the technique.

Remarks: There is very low quality evidence from 2 small trials that acupuncture provides marginal short-term improvement in pain and functional recovery for degenerative LSS. Current evidence provides borderline clinically important short-term improvement and is insufficient to suggest long-term benefit.

PICO 3. In patients who underwent spinal fusion with or without decompression, should supervised training after surgery versus another treatment be used to decrease pain, and improve function, quality of life, and return to function?

Supervised training after surgery For patients with LSS and neurogenic claudication, we suggest offering post-operative rehabilitation with CBT to reduce pain and improve function at 1 month and 12 mo postsurgery. Conditional/Weak Low
(⊕⊕00)

Definition: Post-operative rehabilitation was defined as a supervised program of exercises and/or educational materials encouraging activity 12 wk after surgery. Supervised exercise may include active spinal mobilization, strengthening of spinal deep muscles, stretching of lower limb and low back, functional exercise, walking, and ergonomic advice.

Included studies: A RCT by Monticone et al (2014) ⁸⁷ compared individual 60-min sessions twice/wk of cognitive-behavioral therapy (CBT) for 4 wk combined with exercise (90-min session 5 times/wk for 4 wk) to exercise therapy alone in of patients with post-operatively following lumbar fusion due to LSS with NC (Appendix 6, Table 3).

Outcomes: Functional disability (ODI), back and leg pain intensity (NRS)

Key results: At 1 month, CBT + exercise had significantly less disability (MD: 11.37 (95% CI, 8.68 to 14.07) and back pain (MD: 1.98 (95% CI, 1.62 to 2.34) compared to exercise alone. At 12 mo, CBT + exercise had significantly less disability (MD: 11.1 (95% CI, 8.72 to 13.81), back pain (MD: 2.77 (95% CI, 2.41 to 3.13), and leg pain (MD: 1.13 (95% CI, 1.03 to 1.65) compared to exercise alone. A small proportion of participants in both groups reported minor transitory pain worsening and mood alterations.

Comments: The panel determined a low certainty in the evidence, with minor and transitory undesirable effects and no major adverse events reported.

SSS, Swiss spinal stenosis questionnaire; ZCQ, Zurich claudication questionnaire; SF-36, Short Form 36; NPRS, The Numeric Pain Rating Scale; RMDQ, Roland Morris Disability Questionnaire; ODI, Oswestry Disability Index; CBT, cognitive-behavioral therapy; MD, mean difference; MCID, minimal clinically important difference; RCT, randomized controlled trial; RR, relative risk.

PICO questions, recommendations, definitions of interventions, supporting evidence, comments and remarks regarding LSS.

Key question/intervention	Recommendation	Strength of recommendation	Quality of evidence
PICO 1. For patients with lumbar spinal stenosis, should multimodal rehabilitation interventions versus another treatment be used to decrease pain, and improve function, quality of life, and return to function?
Multimodal therapy	For patients with LSS and neurogenic claudication with or without LBP, we suggest offering a combination of education and advice, manual therapy and home-based exercise for improvement in walking capacity and symptoms/physical function in the short and long term.	Conditional/Weak	Moderate (⊕⊕⊕O)
Definition: Multimodal rehabilitation interventions may include sedentary and nutrition lifestyle changes, ⁷¹^,¹²⁵ behavioral change techniques in conjunction with manual therapy, exercise and/or rehabilitation, and ancillary non-pharmacological treatment.
Included studies: We identified 3 RCTs ³^,⁸⁶^,¹⁰⁸ in which a comprehensive program, including various combination of self-management strategy, with or without cognitive behavioral approach, patient education and advice to stay active, supervised and home exercises (strengthening, stretching, and conditioning exercises, and stationary cycling), and manual therapy (thrust and non-thrust manipulation, manual spine stretching) was compared to home exercises or to medical care plus exercise (Appendix 6, Table 1).
Primary outcomes: Functional disability (ODI), leg pain (NRS), physical performance scale of the Zurich Claudication Questionnaire (ZCQ) or Swiss Spinal Stenosis (SSS) questionnaire, physical function (SF-36), and walking distance (Self-Paced Walk Test (SPWT))/gait disturbance (Japanese Orthopaedic Association Back Pain Evaluation Questionnaire).
Key results: In one RCT (Ammendolia 2018) ³, the adjusted mean difference (MD) in walking distance in the comprehensive group vs the self-directed group was 304.1 m (95% CI, 77.9 to 530.3) at 3 mo and 421.0 m (95% CI, 181.4 to 660.6) at 6 mo. At 6 mo, 82% of participants in the comprehensive group and 63% in the self-directed group achieved the MCID, (adjusted RR 1.3; 95% CI, 1.0 to 1.7). Both primary treatment effects persisted at 12-mo favoring the comprehensive program. At 6-mo, the comprehensive program showed significantly greater improvements in the ODI walk scale (-0.8; 95% CI, -1.3 to -0.4) and at 12-mo in the ZCQ, SF-36 physical function and bodily pain scores.
In one other RCT (Schneider 2019), ¹⁰⁸ manual therapy/individualized exercise had greater, but non-clinically important improvement of symptoms/physical function (Swiss Spinal Stenosis (SSS) questionnaire) at 2 mo, compared to medical care (adjusted mean difference -2.0; 95% CI: -3.6 to -0.4) or group exercise (-2.4; 95% CI: -4.1 to -0.8). Using the >30% responder criterion (secondary responder analyses), manual therapy/exercise had a greater proportion of responders in symptoms/physical function (20%; omnibus P = .002) and walking capacity (Self-Paced Walking Test) (65.3%; omnibus P = .04) at 2-mo compared to medical care (7.6% and 48.7%) or group exercise (3% and 46.2%). Group exercise also had greater improvement in average daily physical activity (armband accelerometer) at 2 -mo compared to medical care (28.7; 95% CI: 2.7 to 54.7). At 6-mo, there were no between-group differences in mean outcome scores or responder rates.
In the third RCT (Minetama (2019),⁸⁶ the supervised physical therapy group showed significant greater improvement at 6 wk vs home exercise in ZCQ symptom severity and physical function (mean difference (MD) −0.4; 95% CI: −0.6 to −0.2), walking distance on the SPWT (MD 455.9 m; 95% CI: 308.5 to -603.2), leg pain (MD −1.4; 95%CI: −2.5 to −0.3), gait disturbance (MD 16.0; 95%CI: 5.4 to -26.7), and physical functioning (MD 9.2; 95%CI: 2.1 to -16.3).
Comment: The panel determined a moderate certainty in the evidence, with minor and transitory undesirable effects and no major adverse events reported.
Remarks: Multimodal rehabilitation intervention was delivered twice weekly over 6 wk. It included individualized instruction on exercise and self-management strategies using a cognitive behavioral approach. At the end of the program, daily home exercise (30 min cycling plus 30 min of structured exercises) and self-care strategies should be maintained.³^,¹⁰⁸
PICO 2. For patients with lumbar spinal stenosis, should acupuncture versus another treatment be used to decrease pain, and improve function, quality of life, and return to function?
Acupuncture	For patients with LSS and neurogenic claudication with or without LBP, we suggest considering traditional acupuncture on a trial basis to improve pain and physical function in the short-term.	Conditional/Weak	Very low (⊕000)
Definition: Needle acupuncture (eg, Hwato Acupuncture, Suzhou, China; 0.30×40 mm/0.30×75 mm) at various sites (eg, Acupoints of Shenshu (BL23), Dachangshu (BL25), Weizhong (BL40), Chengshan (BL57), and Taixi (KI3)) ¹⁰³ or outward from the spinous process bilaterally at L2, L4, S2, and S4, middle of the popliteal fossa, inferior recess in the fibular head, lower end of the groove of the inner and outer head of the gastrocnemius).⁸⁷
Included studies: We identified 2 RCTs ⁸⁷^,¹⁰³ investigating the effect of acupuncture in patients with NC caused by LSS (Appendix 6 Table 2). Primary outcomes: Physical function (RMDQ) and physical performance (ZCQ)
Key results: A RCT by Qin et al (2020) ¹⁰³ compared acupuncture to noninsertive sham acupuncture for 24 treatments over 8-wk in patients LSS with NC. The acupuncture group showed significant greater improvement in disability at 8 wk (adjusted mean difference (MD) -2.6 [95% CI, -3.7 to -1.4]) and at 3 mo (MD -2.3 [95% CI, -3.9 to -0.7], but not at 6-mo. The acupuncture group also showed greater improvement in leg and buttock pain intensity (NRS) at 8 wk (MD -2.9 [95% CI, (-3.8 to -2.0)], 3 mo MD -2.4 [95% CI, -3.3 to -1.4)] and 6 mo (MD -2.1 [95% CI, -3.0 to -1.2]), and back pain (NRS) at 8 wk (MD -2.3 [95% CI, -3.0 to -1.5]) and 3 mo [95% CI -1.7 (-2.6 to -0.8]).
A low-quality comparative study by Oka et al (2018) ⁸⁷ assigned 119 Japanese patients with LSS and L5 radiculopathy (mixed population) to receive either acupuncture (5 sessions in one month), back flexion exercises and an educational manual or pain medication (acetaminophen). Significant reduction in symptom severity was observed in all 3 groups, while improved physical function was found in the acupuncture group only (MD − 2.1, 95% CI − 0.40 to − 0.01). The acupuncture group also demonstrated better physical function compared to exercise group at 1 month (between-group difference in ZCQ least-square mean = 2.17, P = .02).
Comment: The panel determined a moderate certainty in the evidence, with minor and transitory undesirable effects and no major adverse events reported. The most frequently reported transient minor adverse events were worsening of symptoms, general discomfort, pain at the treated areas, and body ache. ⁵⁷ The resources required for an acupuncture intervention are relatively small (cost of care and equipment needed), with the exception of training and certification to provide the technique.
Remarks: There is very low quality evidence from 2 small trials that acupuncture provides marginal short-term improvement in pain and functional recovery for degenerative LSS. Current evidence provides borderline clinically important short-term improvement and is insufficient to suggest long-term benefit.
PICO 3. In patients who underwent spinal fusion with or without decompression, should supervised training after surgery versus another treatment be used to decrease pain, and improve function, quality of life, and return to function?
Supervised training after surgery	For patients with LSS and neurogenic claudication, we suggest offering post-operative rehabilitation with CBT to reduce pain and improve function at 1 month and 12 mo postsurgery.	Conditional/Weak	Low (⊕⊕00)
Definition: Post-operative rehabilitation was defined as a supervised program of exercises and/or educational materials encouraging activity 12 wk after surgery. Supervised exercise may include active spinal mobilization, strengthening of spinal deep muscles, stretching of lower limb and low back, functional exercise, walking, and ergonomic advice.
Included studies: A RCT by Monticone et al (2014) ⁸⁷ compared individual 60-min sessions twice/wk of cognitive-behavioral therapy (CBT) for 4 wk combined with exercise (90-min session 5 times/wk for 4 wk) to exercise therapy alone in of patients with post-operatively following lumbar fusion due to LSS with NC (Appendix 6, Table 3).
Outcomes: Functional disability (ODI), back and leg pain intensity (NRS)
Key results: At 1 month, CBT + exercise had significantly less disability (MD: 11.37 (95% CI, 8.68 to 14.07) and back pain (MD: 1.98 (95% CI, 1.62 to 2.34) compared to exercise alone. At 12 mo, CBT + exercise had significantly less disability (MD: 11.1 (95% CI, 8.72 to 13.81), back pain (MD: 2.77 (95% CI, 2.41 to 3.13), and leg pain (MD: 1.13 (95% CI, 1.03 to 1.65) compared to exercise alone. A small proportion of participants in both groups reported minor transitory pain worsening and mood alterations.
Comments: The panel determined a low certainty in the evidence, with minor and transitory undesirable effects and no major adverse events reported.