Ability to Discriminate Between Healthy and Low Back Pain Sufferers Using Ultrasound During Maximum Lumbar Extension

Archives of Physical Medicine and Rehabilitation Home

Antonio I. Cuesta-Vargas, PhD
Archives of Physical Medicine and Rehabilitation
Volume 95, Issue 6, Pages 1093–1099, June 2014

Journal Reference
Full Text PDF

Abstract
Objective
To analyze the ability to discriminate between healthy individuals and individuals with chronic nonspecific low back pain (CNLBP) by measuring the relation between patient-reported outcomes and objective clinical outcome measures of the erector spinae (ES) muscles using an ultrasound during maximal isometric lumbar extension.

Design
Cross-sectional study with screening and diagnostic tests with no blinded comparison.

Setting
University laboratory.

Participants
Healthy individuals (n=33) and individuals with CNLBP (n=33).

Interventions
Each subject performed an isometric lumbar extension. With the variables measured, a discriminate analysis was performed using a value ≥6 in the Roland and Morris disability questionnaire (RMDQ) as the grouping variable. Then, a logistic regression with the functional and architectural variables was performed. A new index was obtained from each subject value input in the discriminate multivariate analysis.

Main Outcome Measures
Morphologic muscle variables of the ES muscle were measured through ultrasound images. The reliability of the measures was calculated through intraclass correlation coefficients (ICCs). The relation between patient-reported outcomes and objective clinical outcome measures was analyzed using a discriminate function from standardized values of the variables and an analysis of the reliability of the ultrasound measurement.

Results
The reliability tests show an ICC value >.95 for morphologic and functional variables. The independent variables included in the analysis explained 42% (P=.003) of the dependent variable variance.

Conclusions
The relation between objective variables (electromyography, thickness, pennation angle) and a subjective variable (RMDQ ≥6) and the capacity of this relation to identify CNLBP within a group of healthy subjects is moderate. These results should be considered by clinicians when treating this type of patient in clinical practice.

Comments Are Closed