Modifying lumbar flexion pain thresholds in chronic low back pain through visual-proprioceptive manipulation with virtual reality: a cross-sectional study

Key points

1. 01Understating visual feedback of lumbar flexion (F- condition) delayed pain onset, allowing 5% more movement than the control and 7% more than the overstated condition
2. 02The overstated VR condition did not significantly change pain-free range of motion compared to no VR
3. 03Participants with higher pain intensity and pain interference scores were more responsive to the understated VR condition (VR responders vs non-responders, both p around 0.05)
4. 04Kinesiophobia and catastrophizing showed a trend toward greater response in the F- condition but differences were not statistically significant, likely due to small subgroup sizes
5. 05Participants remained unaware of the visual manipulation and no motion sickness was reported

How it was conducted

Design

Double-blind, within-subject cross-sectional study with randomised condition order and repeated measures

Participants

50 adults (28 women, 22 men) with non-specific chronic low back pain (pain duration mean 11 months, mean age 52 years), recruited from hospital orthopaedic and rehabilitation services

Conditions

Three: (1) no VR control (F), (2) VR understating arm length by 20% to suggest less flexion than real (F-), (3) VR overstating arm length by 20% to suggest more flexion than real (F+), each repeated 3 times in randomised order

Primary outcome

Maximum pain-free lumbar range of motion (degrees) measured by electro-goniometer, expressed as percentage of control condition

Secondary analysis

Whether pain intensity, pain interference, kinesiophobia (TSK), and catastrophizing (PCS) moderated response to the F- condition

What they found

• One-way ANOVA showed significant differences among the three conditions (p < 0.001, moderate effect size eta-p2 = 0.11)
• F- condition mean pain-free ROM was 105% (SD 2.1) vs 100% (SD 0) for control and 98% (SD 2.3) for F+
• F- vs control: 5% increase in pain-free ROM (p = 0.04, 95% CI [0.6%, 10.7%])
• F- vs F+: 7% increase in pain-free ROM (p < 0.001, 95% CI [2.6%, 11.6%])
• F vs F+: no significant difference (p = 1.00, 95% CI [-4.1%, 7.6%])
• VR responders (F- ROM above mean) had higher pain intensity than non-responders: 5.9 (SD 1.9) vs 4.8 (SD 1.6), p = 0.05, mean difference -1.0 (95% CI -2.1 to 0.02), Cohen d = 0.6
• VR responders had higher pain interference: 5.5 (SD 2.4) vs 4.1 (SD 2.1), p = 0.04, mean difference -1.4 (95% CI -2.7 to 0.1), Cohen d = 0.6
• Kinesiophobia subgroup (TSK > 37, n = 21) showed 111% (SD 12.1) F- ROM vs 104% (SD 15.3) in those without kinesiophobia, difference not significant (p = 0.22, Cohen d = 0.5)
• Catastrophizing subgroup (PCS >= 30, n = 21) showed 112% (SD 16) F- ROM vs 104% (SD 14.4), difference not significant (p = 0.15, Cohen d = 0.5)
• Perceived task difficulty did not differ between F- and F+ conditions (p = 0.55)

Limitations

• Single-session cross-sectional design cannot determine whether short-term perceptual effects persist or translate into lasting clinical benefit
• Recruited from specialist hospital services, which may have excluded the most severe or treatment-resistant cases and limits generalisability
• Small subgroups of participants with kinesiophobia (n = 21) and catastrophizing (n = 9 above PCS threshold of 30) reduced statistical power for those secondary analyses
• Non-specific cLBP is clinically heterogeneous (e.g., discogenic vs motor control origins), meaning underlying pain mechanisms varied and could affect results

Why it matters

For patients

If you have chronic low back pain that is worse during bending, VR-based visual feedback that makes your movement look smaller than it really is may allow you to bend a little further before pain starts, though the effect is modest and currently only demonstrated in a single lab session.

For clinicians

This study provides preliminary evidence that visual-proprioceptive manipulation via VR can shift movement-evoked pain thresholds in cLBP, with the strongest response in patients carrying higher pain burden, which may inform graded exposure and fear-avoidance rehabilitation protocols.

For readers

The findings support pain-as-perception models by showing that what patients see about their own movement, not only tissue nociception, influences when pain is felt during lumbar flexion.