Lumbar spine repositioning sense in adolescents with and without non-specific chronic low back pain – An analysis based on sub-classification and spinal regions

BACK-to-MOVE: Machine learning and computer vision model automating clinical classification of non-specific low back pain for personalised management

BACKGROUND

Low back pain (LBP) is a major global disability contributor with profound health and socio-economic implications. The predominant form is non-specific LBP (NSLBP), lacking treatable pathology. Active physical interventions tailored to individual needs and capabilities are crucial for its management. However, the intricate nature of NSLBP and complexity of clinical classification systems necessitating extensive clinical training, hinder customised treatment access. Recent advancements in machine learning and computer vision demonstrate promise in characterising NSLBP altered movement patters through wearable sensors and optical motion capture. This study aimed to develop and evaluate a machine learning model (i.e., 'BACK-to-MOVE') for NSLBP classification trained with expert clinical classification, spinal motion data from a standard video alongside patient-reported outcome measures (PROMs).

METHODS

Synchronised video and three-dimensional (3D) motion data was collected during forward spinal flexion from 83 NSLBP patients. Two physiotherapists independently classified them as motor control impairment (MCI) or movement impairment (MI), with conflicts resolved by a third expert. The Convolutional Neural Networks (CNNs) architecture, HigherHRNet, was chosen for effective pose estimation from video data. The model was validated against 3D motion data (subset of 62) and trained on the freely available MS-COCO dataset for feature extraction. The Back-to-Move classifier underwent fine-tuning through feed-forward neural networks using labelled examples from the training dataset. Evaluation utilised 5-fold cross-validation to assess accuracy, specificity, sensitivity, and F1 measure.

RESULTS

Pose estimation's Mean Square Error of 0.35 degrees against 3D motion data demonstrated strong criterion validity. Back-to-Move proficiently differentiated MI and MCI classes, yielding 93.98% accuracy, 96.49% sensitivity (MI detection), 88.46% specificity (MCI detection), and an F1 measure of .957. Incorporating PROMs curtailed classifier performance (accuracy: 68.67%, sensitivity: 91.23%, specificity: 18.52%, F1: .800).

CONCLUSION

This study is the first to demonstrate automated clinical classification of NSLBP using computer vision and machine learning with standard video data, achieving accuracy comparable to expert consensus. Automated classification of NSLBP based on altered movement patters video-recorded during routine clinical examination could expedite personalised NSLBP rehabilitation management, circumventing existing healthcare constraints. This advancement holds significant promise for patients and healthcare services alike.

Questionnaires that assess disability in children and adolescents with low back pain adhere to the concepts of the International classification of functioning, disability and health (ICF), but lack validity for this population: a systematic review

PURPOSE

We aimed to identify Patient-Reported Outcome Measures (PROMs) that assess disability in children and adolescents with low back pain (LBP), analyzing their adherence to the International Classification of Functioning, Disability and Health (ICF) biopsychosocial model; and to describe the measurement properties of these PROMs.

METHODS

We searched Pubmed, Embase and CINAHL databases. The review included searches up to March 2022. Meaningful concepts of the PROMs were linked to ICF domains, and we manually searched for the measurement properties of each included PROM.

RESULTS

We included 23 studies, of which eight PROMs were analyzed. We retrieved 182 concepts in total. Activities was the domain with the highest number of linked concepts, whereas personal factors had no linked concepts. The modified Hannover Functional Ability Questionnaire (mHFAQ) and the Micheli Functional Scale (MFS) had measurement properties tested in children and adolescents, but had no information about construct validity.

CONCLUSION

Although most of the identified PROMs had broad coverage of their concepts in the ICF, only two PROMs had measurement properties tested in the population of interest in this review, in which the mHFAQ presented wide coverage in relation to the ICF. Further studies are needed to investigate content validity of these PROMs.

The effects of proprioceptive weighting changes on posture control in patients with chronic low back pain: a cross-sectional study

Introduction

Patients with chronic low back pain (CLBP) exhibit changes in proprioceptive weighting and impaired postural control. This study aimed to investigate proprioceptive weighting changes in patients with CLBP and their influence on posture control.

Methods

Sixteen patients with CLBP and 16 healthy controls were recruited. All participants completed the joint reposition test sense (JRS) and threshold to detect passive motion test (TTDPM). The absolute errors (AE) of the reposition and perception angles were recorded. Proprioceptive postural control was tested by applying vibrations to the triceps surae or lumbar paravertebral muscles while standing on a stable or unstable force plate. Sway length and sway velocity along the anteroposterior (AP) and mediolateral (ML) directions were assessed. Relative proprioceptive weighting (RPW) was used to evaluate the proprioception reweighting ability. Higher values indicated increased reliance on calf proprioception.

Results

There was no significant difference in age, gender, and BMI between subjects with and without CLBP. The AE and motion perception angle in the CLBP group were significantly higher than those in the control group (JRS of 15°: 2.50 (2.50) vs. 1.50 (1.42), JRS of 35°: 3.83 (3.75) vs. 1.67 (2.00), p_JRS < 0.01; 1.92 (1.18) vs. 0.68 (0.52), p_TTDPM < 0.001). The CLBP group demonstrated a significantly higher RPW value than the healthy controls on an unstable surface (0.58 ± 0.21 vs. 0.41 ± 0.26, p < 0.05). Under the condition of triceps surae vibration, the sway length (p_stable < 0.05; p_unstable < 0.001), AP velocity (p_stable < 0.01; p_unstable < 0.001) and ML velocity (p_unstable < 0.05) had significant group main effects. Moreover, when the triceps surae vibrated under the unstable surface, the differences during vibration and post vibration in sway length and AP velocity between the groups were significantly higher in the CLBP group than in the healthy group (p < 0.05). However, under the condition of lumbar paravertebral muscle vibration, no significant group main effect was observed.

Conclusion

The patients with CLBP exhibited impaired dynamic postural control in response to disturbances, potentially linked to changes in proprioceptive weighting.

Intra-rater and inter-rater reliability of neutral and target lumbar positioning tests in subjects with and without non-specific lower back pain

BACKGROUND

Proprioception, one's sense of movement and position, is a common term used in back rehabilitation. Kinesthetic rehabilitation may be useful in managing lower back pain; however, reliable measures are required to quantify lumbar proprioception sense.

OBJECTIVE

To investigate intrarater and interrater reliability of neutral lumbar positioning (NLP) and target lumbar positioning (TLP) tests and compare the position sense errors in subjects with non-specific low back pain and healthy controls.

METHODS

Intrarater (between-day) and interrater (within-day) reliability of NLP and TLP tests were assessed in 30 subjects with low back pain and 30 healthy subjects using a digital inclinometer. NLP is evaluated when the subject is repositioned to neutral from flexion, while TLP is evaluated in lumbar flexion, by bending laterally left and right.

RESULTS

Intrarater reliability for NLP tests had ICC values of 0.85 and 0.89 and TLP tests had 0.78 and 0.92. Likewise, interrater reliability for NLP had ICC values of 0.75 and 0.85, and for the TLP test, the interrater reliability had 0.78 and 0.93. Subjects with back pain had significantly larger neutral and target lumbar proprioceptive errors compared to healthy controls (p< 0.001).

CONCLUSIONS

Intrarater and interrater reliability showed good agreement for both NLP and TLP tests of lumbar proprioception. Subjects with nonspecific low back pain have impaired lumbar proprioceptive sense.

Lumbo-pelvic proprioception in sitting is impaired in subgroups of low back pain-But the clinical utility of the differences is unclear. A systematic review and meta-analysis

BACKGROUND

Altered spinal postures and altered motor control observed among people with non-specific low back pain have been associated with abnormal processing of sensory inputs. Evidence indicates that patients with non-specific low back pain have impaired lumbo-pelvic proprioceptive acuity compared to asymptomatic individuals.

OBJECTIVE

To systematically review seated lumbo-pelvic proprioception among people with non-specific low back pain.

METHODS

Five electronic databases were searched to identify studies comparing lumbo-pelvic proprioception using active repositioning accuracy in sitting posture in individuals with and without non-specific low back pain. Study quality was assessed by using a modified Downs and Black's checklist. Risk of bias was assessed using an adapted tool for cross-sectional design and case-control studies. We performed meta-analysis using a random effects model. Meta-analyses included subgroup analyses according to disability level, directional subgrouping pattern, and availability of vision during testing. We rated the quality of evidence using the GRADE approach.

RESULTS

16 studies met the eligibility criteria. Pooled meta-analyses were possible for absolute error, variable error, and constant error, measured in sagittal and transverse planes. There is very low and low certainty evidence of greater absolute and variable repositioning error in seated tasks among non-specific low back pain patients overall compared to asymptomatic individuals (sagittal plane). Subgroup analyses indicate moderate certainty evidence of greater absolute and variable error in seated tasks among directional subgroups of adults with non-specific low back pain, along with weaker evidence (low-very low certainty) of greater constant error.

DISCUSSION

Lumbo-pelvic proprioception is impaired among people with non-specific low back pain. However, the low certainty of evidence, the small magnitude of error observed and the calculated "noise" of proprioception measures, suggest that any observed differences in lumbo-pelvic proprioception may be of limited clinical utility.

PROSPERO-ID

CRD42018107671.

Abdominal muscle activity and pelvic motion according to active straight leg raising test results in adults with and without chronic low back pain

BACKGROUND

Active Straight Leg Raise (ASLR) is a clinical test that challenges lumbopelvic stability and have been used in low back pain patients. There are reports of positive tests in asymptomatic individuals, however, it is not clear if they are false positives or a consequence of performing the test differently.

OBJECTIVES

To compare abdominal muscles activity and pelvic motion during an ASLR between adults with chronic low back pain and asymptomatic participants with a positive and negative ASLR test.

DESIGN

Cross-sectional study design.

METHOD

Nineteen volunteers with chronic nonspecific low back pain (LBP) and 19 asymptomatic, 12 with a negative ASLR (AG-) and 7 with a positive ASLR test (AG+) were assessed while performing an ASLR (dynamic postural challenge). Pelvic rotation and medio-lateral center of pressure displacement (COPml), as well as bilateral EMG abdominal muscles activity were assessed. Muscles asymmetry index were also analyzed. Kruskal-wallis test was used to compare groups (α = 0.05).

RESULTS

Both LBP and AG + displayed significantly less contralateral internal oblique/transversus abdominis (IO/TrA) muscle activation than AG- (p = 0.003 and p = 0.005, respectively) and also more asymmetry between sides in the IO/TrA than AG- (p = 0.022 and p = 0.004, respectively). No significant differences between LBP and AG+ were found (p > 0.05).

CONCLUSION

A positive ASLR test in an asymptomatic person could be more than a false positive, since IO/TrA muscle activation appears to be lower and more asymmetrical in people with a positive ASLR test, both with and without low back pain. Future studies may consider removing asymptomatic individuals with a positive ASLR.

A preliminary study on the influence of high heel footwear on lumbar spine proprioception

Study aim: High heel footwear (HHF) alters posture and kinetics of the lower limbs and spine. Various musculoskeletal disorders in the lower quadrant can occur as a consequence of HHF use. This study aimed to investigate the effect of habitual use of HHF on lumbar spine proprioception.

Material and methods: Thirty young healthy women (mean age: 21.6 ± 1.6 years; height: 157.6 ± 5.9 cm; weight: 57 ± 10.2 kg) participated in this study. Participants were randomly allocated to one of the two sequences of walking with HHF or walking barefoot for five minutes on a predetermined walkway consisting of an even surface, ramp and stairs. A 24-hour washout period was incorporated between each walking condition. Lumbar spine repositioning error was measured using a Back Range of Motion II (BROM II) device before and after walking for lumbar spine forward flexion, right, and left lateral flexion.

Results: Measurements showed that repositioning error was significantly higher in all spinal movements after walking with HHF (forward flexion p = 0.0044; right lateral flexion p = 0.0243 left lateral flexion p = 0.0379) as compared to barefoot walking.

Conclusion: The findings of this study suggest that HHF alters lumbar spine proprioception. This study provides evidence to support early assessment of proprioception and incorporation of retraining exercises in habitual users of HHF.

Effect of sustained experimental muscle pain on joint position sense

Introduction:

Joint position sense (JPS) is impaired in clinical musculoskeletal pain conditions, but when this impairment develops in the transition from initial to prolonged pain is not known.

Objectives:

This study assessed whether progressively developing sustained experimentally induced muscle pain impacts JPS in healthy individuals.

Methods:

Twenty-eight healthy individuals received injection of nerve growth factor (NGF) into the right extensor carpi radialis brevis muscle on days 0 and 2 to induce sustained pain and hyperalgesia. Wrist JPS was assessed 2 days before day 0 (day −2), before the injection on days 0 and 2, and on days 4 and 14. Joint position sense was quantified as the ability to return the wrist to a neutral position following movements in the direction of radial and ulnar deviation. A 3-dimensional motion analysis system was used to calculate absolute, relative, and joint-angle repositioning errors. Numerical rating scale scores of pain intensity, body chart pain drawings, and pressure pain thresholds (PPTs) were recorded on each day.

Results:

Compared with baseline, pressure pain thresholds decreased while pain intensity and area increased at day 2 (P < 0.001) and day 4 (P < 0.001) before returning to baseline on day 14 (P > 0.13). Relative to day 0, there was no change in wrist JPS at day 2, 4, and 14 following movements in either target direction (P > 0.05).

Conclusion:

Despite the presence of sustained muscle pain and hyperalgesia for 4 days at the elbow, no statistical change in wrist joint position error was observed. These findings suggest that pain and hyperalgesia lasting as long as 4 days does not impair JPS.

Comparison of proprioceptive acuity of the cervical spine in healthy adults and adults with chronic non-specific low back pain: A cross-sectional study

Background

It has been suggested that patients with chronic non-specific low back pain (CNSLBP) perform poorly in postural tasks when compared to healthy individuals. Despite its importance in posture and alignment of the trunk in relation to the head, neck proprioception has not been examined in patients with low back pain. The purpose of this study was to compare neck proprioception in patients with CNSLBP with healthy individuals.

Methods

Cervical joint reposition error was measured five times consecutively in the neutral head position, 30° and 60° left and right head rotation. The main outcome measure was the mean cervical joint repositioning error of the head.

Results

Forty-six participants with (n = 24, 54 ± 16yrs SD, 14 females) and without (n = 22, 36 ± 13yrs SD, 13 females) CNSLBP were included in the study. Comparison of mean cervical joint repositioning error between patients and healthy controls showed no statistically significant group difference in any of the applied positions. The range of deviation in CNSLBP patients was between 1.57° and 3.27° compared to 1.46° to 2.26° in healthy controls. An overshooting tendency for both groups was found in the neutral head position.

Conclusion

The ability to accurately position the head does not seem to be impaired in patients with CNSLBP. This may suggest that sensorimotor control is affected on other levels of the movement system and future research should focus on methods to identify the source of these aberrations.

The amount of postural change experienced by adolescent computer users developing seated -related upper quadrant musculoskeletal pain

BACKGROUND

Improved techniques of measuring sitting posture have not led to a more comprehensive understanding of poor posture, nor its association with pain. There is also an evidence gap regarding critical thresholds of sitting postural change over time related to pain production. This paper describes postural angle changes over a 12-month period, and describes the process of placing defensible cut-points in the angle change data, to better understand associations between posture change over time, and onset of upper quadrant musculoskeletal pain (UQMP).

METHODS

This paper reports on data captured at baseline and 12-month follow-up, in adolescents in school using computers. Four sitting postural angles, head flexion (HF), neck flexion (NF), craniocervical angle (CCA) and trunk flexion (TF), and self-reported seated UQMP in the previous month were captured at each time-point. Research questions were: 1) What is the magnitude and direction of change in each postural angle over 12 months? 2) What are best cut-points in the continuous posture change distribution to most sensitively test the association between posture change and UQMP? 3) Is gender-specific cut-points required? The 12-month posture angle change data was divided into quintiles (0-20th%; 21-40^th%, 41-60^th%, 61-80^th%, >80^th%), and the odds of UQMP occurring in each posture change quintile were calculated using logistic regression models.

RESULTS

Two hundred and eleven students participated at baseline, of which 153 were followed-up at one year. Both males and females with postural change into extension (which represents lesser flexion range) were more at risk for the development of UQMP, than any other group. The best cut-point for HF was 40^th% (≤-3.9°), NF was 20th% (≤-2.9°) and TF was 40^th% (≤-1.1°). For CCA however, change at or beyond 40^th % for extension or beyond 60% for flexion was associated with UQMP.

CONCLUSIONS

Identification of critical postural angle change cut-points assists in considering the pain-producing mechanisms for adolescents using desk top computers.

Repeated end range spinal movement while seated abolishes the proprioceptive deficit induced by prolonged flexed sitting posture. A study assessing the statistical and clinical significance of spinal position sense

BACKGROUND

Sustained spinal flexion has been proposed to affect the properties of spinal tissues, increase postural muscle's activation latency and act detrimentally on proprioception.

OBJECTIVES

This study evaluated the effect of flexed posture (FP) on spinal proprioception and assessed the immediate effect of spinal movement on the presumable flexion-induced proprioceptive deficit.

DESIGN

Clinical measurement study.

METHODS

Marker-based kinematic analyses of the head, spine, and pelvis were conducted on 50 individuals. Subjects were educated in a lordotic sitting posture (IOSP) that they reproduced immediately; after 10 and 30 min in FP; and after sagittal spinal movement. Nine sagittal angles were calculated. Absolute error (AE) and constant error (CE) were used to evaluate repositioning accuracy. Repeated measures ANOVA was used to test for significant differences in angles obtained among postures, as well as for the AE and CE calculated from the trials.

RESULTS

No significant differences were found in reposition error (RE) after immediate reproduction of IOSP (all p > 0.0083). Following FP AEs presented significant differences for head (4.1°), head protraction (1.9°), head tilt (2.1°), lumbar (3.2°) and pelvis angle (2.1°). CEs revealed significant differences for head protraction (-1.8°) and lumbar angle (-3.5°). No significant differences were found for AE and CE after spinal sagittal movement (all p > 0.0083).

CONCLUSIONS

Prolonged FP can affect spinal position sense, but sagittal spinal movement can abolish the proprioceptive deficit. The significant differences documented, may be of limited clinical utility given their magnitude, and the reliability data presented may be of use in reinterpreting previously documented proprioceptive analyses.

Understanding Adolescent Low Back Pain From a Multidimensional Perspective: Implications for Management

Synopsis Low back pain (LBP) is the leading cause of disability worldwide. It often begins in adolescence, setting a course for later in life. We have tracked the course of LBP in the Raine Study cohort from the age of 14 years into early adulthood. Our work has found that LBP is already prevalent in individuals at 14 years of age and increases throughout adolescence and into early adulthood. It is often comorbid with other musculoskeletal pain. For some adolescents, LBP has little impact; for others, its impact includes care seeking, taking medication, taking time off from school and work, as well as modifying physical and functional activity. Of concern is the increasing prevalence of LBP with impact across adolescence, reaching adult rates by 22 years of age. The predictors of disabling LBP in adolescence are multidimensional. They include female sex, negative back pain beliefs, poor mental health status, somatic complaints, involvement in sports, and altered stress responses. Genetics also plays a role. Ironically, the factors that we have historically thought to be important predictors of LBP, such as "poor" spinal posture, scoliosis, carrying school bags, joint hypermobility, and poor back muscle endurance, are not strong predictors. This challenges our clinical beliefs and highlights that adolescent LBP needs a flexible and targeted multidimensional approach to assessment and management. In most cases, we recommend a cognitive functional approach that challenges negative LBP beliefs, educates adolescents regarding factors associated with their LBP, restores functional capacity where it is impaired, and encourages healthy lifestyle habits. J Orthop Sports Phys Ther 2017;47(10):741-751. Epub 12 Sep 2017. doi:10.2519/jospt.2017.7376.

Is There a Relationship Between Lumbar Proprioception and Low Back Pain? A Systematic Review With Meta-Analysis

OBJECTIVE

To systematically review the relationship between lumbar proprioception and low back pain (LBP).

DATA SOURCES

Four electronic databases (PubMed, EMBASE, CINAHL, SPORTDiscus) and reference lists of relevant articles were searched from inception to March-April 2014.

STUDY SELECTION

Studies compared lumbar proprioception in patients with LBP with controls or prospectively evaluated the relationship between proprioception and LBP. Two reviewers independently screened articles and determined inclusion through consensus.

DATA EXTRACTION

Data extraction and methodologic quality assessment were independently performed using standardized checklists.

DATA SYNTHESIS

Twenty-two studies (1203 participants) were included. Studies measured lumbar proprioception via active or passive joint repositioning sense (JRS) or threshold to detection of passive motion (TTDPM). Data from 17 studies were pooled for meta-analyses to compare patients with controls. Otherwise, descriptive syntheses were performed. Data were analyzed according to measurement method and LBP subgroup. Active JRS was worse in patients compared with controls when measured in sitting (standard mean difference, .97; 95% confidence interval [CI], .31-1.64). There were no differences between groups measured via active JRS in standing (standard mean difference, .41; 95% CI, -.07 to .89) or passive JRS in sitting (standard mean difference, .38; 95% CI, -.83 to 1.58). Patients in the O'Sullivan flexion impairment subgroup had worse proprioception than the total LBP cohort. The TTDPM was significantly worse in patients than controls. One prospective study found no link between lumbar proprioception and LBP.

CONCLUSIONS

Patients with LBP have impaired lumbar proprioception compared with controls when measured actively in sitting positions (particularly those in the O'Sullivan flexion impairment subgroup) or via TTDPM. Clinicians should consider the relationship between sitting and proprioception in LBP and subgroup patients to guide management. Further studies focusing on subgroups, longitudinal assessment, and improving proprioception measurement are needed.

An exploration of familial associations of two movement pattern-derived subgroups of chronic disabling low back pain; a cross-sectional cohort study

BACKGROUND

Altered movement patterns with pain have been demonstrated in children, adolescents and adults with chronic disabling low back pain (CDLBP). A previously developed classification system has identified different subgroups including active extension and multidirectional patterns in patients with CDLBP. While familial associations have been identified for certain spinal postures in standing, it is unknown whether a familial relationship might exist between movement pattern-derived subgroups in families with CDLBP.

OBJECTIVES

This study explored whether familial associations in movement pattern-derived subgroups within and between members of families with CDLBP existed.

DESIGN

Cross-sectional cohort study.

METHOD

33 parents and 28 children with CDLBP were classified into two subgroups based on clinical analysis of video footage of postures and functional movements, combined with aggravating factors obtained from Oswestry Disability Questionnaire. Prevalence of subgroups within family members was determined, associations between parent and child's subgroup membership was evaluated using Fisher's exact test, and spearman's correlation coefficient was used to determine the strength of association between familial dyads.

RESULTS

The majority of parents were classified as active extenders, sons predominately multidirectional and daughters were evenly distributed between the two subgroups. No significant association was found when comparing subgroups in nine parent-child relationships.

CONCLUSIONS

The exploration of a small cohort of family dyads in this study demonstrated that children's movement pattern-derived subgroups could not be explained by their parents' subgroup membership. These results cannot be generalised to the CLBP population due to this study's small sample. Larger sample studies are needed to further elucidate this issue.

The effects of gait with use of smartphone on repositioning error and curvature of the lumbar spine

[Purpose] This study evaluated the effect of smartphone use on lumbar spine repositioning error and lumbar curvature while walking on a treadmill. [Subjects] A total of 20 healthy individuals (18 males and 2 females) volunteered for this study. [Methods] The subjects walked for 20 min on a treadmill while using a smartphone. To determine the effect of smartphone use, lumbar repositioning error was measured using an electronic goniometer while lumbar curvature was assessed using a Spinal Mouse before and immediately after treadmill use. Differences in the lumbar repositioning error and lumbar curvature data between the pre- and post-walking were compared using the paired t-test. [Results] The lumbar spine repositioning error was significantly greater post-walking compared with pre-walking (6.70±2.91° vs. 3.02±1.79°). There was no significant difference in lumbar curvature between pre- and post-walking (14.24±3.18° vs. 13.94±3.12°). [Conclusion] These findings indicate that the lumbar repositioning error increased immediately after walking while using a smartphone, but that the lumbar curvature was unchanged.

Measuring lumbar reposition accuracy in patients with unspecific low back pain: systematic review and meta-analysis

STUDY DESIGN

Systematic review and meta-analysis.

OBJECTIVE

To evaluate if patients with nonspecific chronic low back pain (NSCLBP) show a greater lumbar reposition error (RE) than healthy controls.

SUMMARY OF BACKGROUND DATA

Studies on lumbar RE in patients with NSCLBP present conflicting results.

METHODS

A systematic review and meta-analysis of the available literature were performed to evaluate differences in RE between patients with NSCLBP and healthy controls. Data on absolute error, constant error (CE), and variable error were extracted and effect sizes (ESs) were calculated. For the CE flexion pattern and active extension pattern, subgroups of patients with NSCLBP were analyzed. Results of homogeneous studies were pooled. Measurement protocols and study outcomes were compared. The quality of reporting and the authors' appraisal of risk of bias were investigated.

RESULTS

The original search revealed 178 records of which 13 fulfilled the inclusion criteria. The majority of studies showed that patients with NSCLBP produced a significantly larger absolute error (ES, 0.81; 95% confidence interval [95% CI], 0.13-1.49) and variable error (ES, 0.57; 95% CI, 0.05-1.09) compared with controls. CE is direction specific in flexion and active extension pattern subgroups of patients with NSCLBP (ES, 0.39; 95% CI, -1.09 to 0.3) and ES, 0.18; 95% CI, -0.3 to 0.65, respectively). The quality of reporting and the authors' appraisal of risk of bias varied considerably. The applied test procedures and instrumentation varied between the studies, which hampered the comparability of studies.

CONCLUSION

Although patients with NSCLBP seemed to produce a larger lumbar RE compared with healthy controls, study limitations render firm conclusions unsafe. Future studies should pay closer attention to power, precision, and reliability of the measurement approach, definition of outcome measures, and patient selection. We recommend a large, well-powered, prospective randomized control study that uses a standardized measurement approach and definitions for absolute error, CE, and variable error to address the hypothesis that proprioception may be impaired with CLBP.