Reliability of quantifying the spatial distribution of fatty infiltration in lumbar paravertebral muscles using a new segmentation method for T1-weighted MRI

Mhuiris, Aine Ni, Volken, Thomas, Elliott, James M., Hoggarth, Mark, Samartzis, Dino and Crawford, Rebecca J. (2016) Reliability of quantifying the spatial distribution of fatty infiltration in lumbar paravertebral muscles using a new segmentation method for T1-weighted MRI. BMC Musculoskeletal Disorders, 17 234: 1-7. doi:10.1186/s12891-016-1090-z


Author Mhuiris, Aine Ni
Volken, Thomas
Elliott, James M.
Hoggarth, Mark
Samartzis, Dino
Crawford, Rebecca J.
Title Reliability of quantifying the spatial distribution of fatty infiltration in lumbar paravertebral muscles using a new segmentation method for T1-weighted MRI
Journal name BMC Musculoskeletal Disorders   Check publisher's open access policy
ISSN 1471-2474
Publication date 2016-05-27
Year available 2016
Sub-type Article (original research)
DOI 10.1186/s12891-016-1090-z
Open Access Status DOI
Volume 17
Issue 234
Start page 1
End page 7
Total pages 7
Place of publication London, United Kingdom
Publisher BioMed Central
Collection year 2017
Language eng
Formatted abstract
Background
To our knowledge, there are no methods allowing for quantification of the spatial distribution of lumbar paravertebral muscle fatty infiltration (FI) in the transverse plane. There is an increasing emphasis on muscle tissues as modifiable factors in lumbar spine health. Population datasets based on conventional T1-weighted (T1-W) magnetic resonance imaging (MRI) represent a valuable resource for examining all spinal tissues, and methods with reliability are needed. The aim of our study was to determine the reliability of a novel method quantifying lumbar paravertebral muscle fat content based on conventional T1-W MRI.

Methods
Axial 3-Tesla T1W MRIs from ten adult subjects (3W, 7M; mean age 52.8 ± SD 7.2 years) were randomly selected from the large prospective cross-sectional Hong Kong Population-based Disc Degeneration Cohort study examining lumbar spine degeneration. The selected sample included subjects with mixed imaging-determined disc degeneration and low back pain history. Two raters with MRI lumbar paravertebral muscle analysis experience (R1 > 250 h and R2 > 1000 h) repeat-measured the image-set a week apart. Multifidus and erector spinae (spinalis, longissimus and iliocostalis) were manually outlined together on a single-slice from the inferior vertebral end-plates of L1 to L5 using a semi-automated, quartile-defining (Q1-4 (medial to lateral) and Qmean) MatLab-based programme. Bland-Altman plots and intra-class correlation coefficients (ICC) with 95 % confidence intervals (CI) describe intra- and inter-rater reliability according to lumbar level, quartile, and side, and combined level and quartile.

Results
There was good intra- (ICC = 0.88; CI: 0.87–0.90) and inter-rater agreement (ICC = 0.82; CI: 0.80–0.84). Intra-rater values for Qmean (ICC; CI) were higher at L5 (0.89; 0.79–0.94) than L1 (0.61; 0.37–0.78). Higher intra-rater values for L1-5 were shown at Q1 (0.93; 0.91–0.95) than Q3 (0.83; 0.78–0.87) or Q4 (0.81; 0.76–0.85), and on the right (0.91; 0.90–0.93) than left (0.85; 0.83–0.88). Similar observations were made for inter-rater values in terms of lumbar level and quartile, with no differences between sides shown.

Conclusions
In our study of ten cases we demonstrate a reliable method to quantify the spatial distribution of fat content in lumbar paravertebral muscles based on T1W MRI. Understanding the geography of fat content in these muscles may offer additional insight in determining and improving spinal health. The clinical relevance and application of this method require testing across various populations to build on the early feasibility established in this study.
Keyword Reliability
Paravertebral muscle
Fatty infiltration
T1-weighted MRI
Lumbar spine
Q-Index Code C1
Q-Index Status Provisional Code
Institutional Status UQ

Document type: Journal Article
Sub-type: Article (original research)
Collections: HERDC Pre-Audit
School of Health and Rehabilitation Sciences Publications
 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 0 times in Thomson Reuters Web of Science Article
Scopus Citation Count Cited 0 times in Scopus Article
Google Scholar Search Google Scholar
Created: Sun, 26 Jun 2016, 00:31:29 EST by System User on behalf of School of Health & Rehabilitation Sciences