The morphology of the lumbar paraspinal, gluteal and piriformis muscles in chronic sacroiliac joint pain

Abstract

The sacroiliac joint (SIJ) can be identified as the source of pain in approximately 10-45% of patients suffering from chronic low back pain. It is hypothesised that SIJ pain may be associated with an imbalance of force transfer from the trunk to the SIJ, and functional studies have demonstrated altered local muscle activity in patients with SIJ pain. The lumbar erector spinae and multifidus and some hip joint muscles are known to be important stabilisers of the SIJ. However, there is little evidence about potential changes in the morphology of these muscles in individuals experiencing SIJ pain. Therefore, the primary aim of this study was to compare the volume, anatomical cross-sectional area (ACSA) and physiological cross-sectional area (PCSA), and fatty infiltration of the lumbar erector spinae, multifidus, gluteal and piriformis muscles in patients with chronic SIJ pain and healthy controls.

Sixteen participants with chronic SIJ pain (13 with unilateral and 3 with bilateral pain, mean age 43.5 9.9 and 54.0 15.1 years, respectively) and 16 sex-matched controls (mean age 39.5 16.8 and 58.7 16.2 years, for the matched unilateral and bilateral control participants, respectively) underwent magnetic resonance imaging (MRI) of the lumbopelvic region. Anatomical cross-sectional areas and muscle volumes were estimated from axial scans, and PCSA was calculated for the lumbar erector spinae, multifidus, gluteal and piriformis muscles. Fatty infiltration was assessed using quantitative (percentage of fatty infiltration and the quartile classification system) and qualitative (Goutallier classification system) methods. Differences in the mean measurements of the muscle parameters between sides and study groups (unilateral SIJ pain and controls) were compared, and measurement reliability assessed using intra-class correlation (ICC) coefficient and Cohen’s kappa (). Each morphological parameter was compared to participant demographics (age, pain duration, numerical rating score) to determine the relationship (Spearman’s ) between the measures.

Reliability measurements for all parameters were moderate to strong (ICC 0.63 to 0.99 for muscle volume; 0.30 to 0.90 and ICC 0.50 to 0.98 for assesments of fatty infiltration). When considering unilateral pain, no significant differences in muscle volume, PCSA and ACSA were found between sides or groups in all the muscles investigated. Additionally, there were no differences in fatty infiltration within the gluteal and piriformis muscles. In patients with bilateral pain, there were no observable differences found in the parameters between sides on a qualitative level. The inter-individual variation within the data, particularly for muscle volume, were large.

In participants with unilateral SIJ pain, the grade and percentage of fatty infiltrate in the lumbar erector spinae and multifidus muscles increased from proximal (L3/4) to distal (L5/S1). Fatty infiltration within erector spinae (L3/L4, L5/S1) and multifidus (L4/L5) was significantly greater on the symptomatic side of those with unilateral SIJ pain compared to controls.

With regard to relationship between muscle measurements and participant demographics, a decline in piriformis muscle volume was correlated with increasing age (p=0.02) in those with unilateral SIJ pain. Increased pain duration correlated with a bilateral decrease in ACSA of the lumbar erector spinae at L3/L4 and at L4/L5 on the asymptomatic side and increased fatty infiltration on the asymptomatic side at L5/S1 ( 0.67; p=0.01).

Morphological assessment of volume, ACSA and PCSA of lumbar erector spinae, multifidus, gluteal and piriformis muscles did not enable differentiation of individuals with chronic unilateral SIJ pain from healthy controls. This differs from other lumbopelvic pain conditions such as non-specific low back pain or hip osteoarthritis whereby decreases in muscle volume are accompanied by increased fatty infiltration on the symptomatic side. This study was limited by a relatively small sample size, which may have contributed to high inter-individual variation in the data. In addition, other factors such as functional changes in muscle activity might be a potential reason for altered loading, and consequently could be studied in more detail alongside in-vivo muscle fibre typing or further compartmentalisation of muscle data.