Table 3

Summary of results for studies that investigated intervention effects on muscle activity patterns (specific muscle activity)
Muscle activity patterns (specific muscle activity)
Study and intervention type (experimental vs comparison)

Movement pattern characteristics assessed

Was there a statistically significant difference (p < 0.05) in physical parameters between groups at the end of the intervention period? (blank cell = not measured)

No. of Subjects TA thickness TA slide* TA & IO feedfoward timing Multifidus (LM) thickness Ratio of specific muscle activitiy Baseline differences between groups? SMD and 95%CIs (negative values favour experimental/motor control group) Pain Activity SMD and 95%CIs (negative values favour experimental group)
Akbari 2008Motor control exercise vs general exercise 49 No No No (TA & LM) Pain: Yes Activity: Yes Multifidus thickness −0.21 (−0.74 to 0.33) TA thickness −0.30 (−0.86 to 0.26) Yes Yes Pain −1.06 (−1.66 to −0.46) Activity −0.70 (−1.27 to −0.12)
Hides 1996Motor control exercise vs control 39 Yes,|| Insufficient data Insufficient data No No Insufficient data
Ferreira 2010Motor control exercise(MCE) vs general ex (GE) vs spinal manipulative therapy (SMT) 34 Yes†† No TA thickness ratio (contraction vs rest) MCE vs GE −0.29 (−0.44 to 0.57)†† MCE vs SMT −0.70 (−0.42 to 0.12)†† No No Pain −0.32 (−0.44 to 0.54) MCE vs GE −0.51 (−0.42 to 0.30)MCE vs SMT Activity −0.25 (−1.11 to 0.61)MCE vs GE −0.63 (−0.42 to 0.19)MCE vs SMT
Marshall 2008Swiss ball vs general exercise 50 No No Right feedforward activation of TA + IO −0.77 (−1.59 to 0.04 ) Left feedforward activation of TA + IO −0.46 (−1.25 to 0.34) No Yes Activity −0.77 (−1.34 to −0.19)
O’Sullivan 1997Motor control exercise vs general exercise 44 Yes No Ratio of TA + IO to RA −0.84 (−1.47 to −0.21) Yes No** Pain −1.29 (−1.96 to −0.62) Activity −0.56 (−1.18 to 0.06)
Vasseljen 2010, 2012 & Unsgaard-Tonsel 2010Motor control (ultrasound guided exercise (US)) vs motor control (high load, sling exercise (SE)) vs general exercise (GE) 109 No No No No§ TA slide* 0.47 (−0.18 to 0.75) TA thickness ratio (contraction vs rest)#: TA 0.16 (−0.53 to 0.85) US vs GE IO 0.13 (−0.55 to 0.80) US vs GE EO 0.23 (−0.48 to 0.95) US vs GE TA feedforward timing:§§ Minimal or no effect size for most comparisons No significant feedforward differences of clinical relevance No No Pain −0.46 (−1.09 to 0.18) US vs GE −0.28 (−0.90 to 0.35) US vs SE Activity −0.54 (−1.16 to 0.10) US vs GE-0.34 -0.98 to 0.30-0.01) US vs SE

TA = Transversus Abdominus, LM = Lumbar Multifidus, EO = External Oblique, IO = Internal Oblique.

* TA slide = amount of distance (mm) lateral translation of musculotendinous junction present on contraction vs relaxation.

As reported by the authors, but insufficient data for verification.

Our calculations show a statistically significant difference between groups for pain and activity, however the groups showed a significant difference at baseline which diminishes the strength of any conclusion about relative effectiveness of the intervention.

§\ No difference between groups at baseline was noted with the following exceptions: Left versus right differences were noted for the ultrasound guided group for IO ratio and TA lateral slide which created a statistically significant decrease in slide distance (reduced activation) and IO ratio post intervention for the left side only.

|| A statistically significant increase in favour of the experimental group for% size of Multifidus was reported by authors but insufficient data for verification.

Pain data obtained from Marshall 2008b, p331-332.

# Data for US versus SE groups similar.

**Our calculations of p value differ from those reported in the study, where we calculate p = 0.076 for post intervention activity levels (difference between groups post intervention) whereas the study reports p < 0.0001. However the six-month post intervention scores do reach significance (SMD = −0.73, 95%CI −1.35 to −0.11, p = 0.021).

†† Authors present ANOVA data (F2,31 = 4.09; p = 0.026) in favour of MCE vs GE (p = 0.043) and vs SMT (p = 0.053).

§§ Side to side differences (nondominant versus dominant side) produced significant, small between-group differences favouring the SE group for the dominant side only (SEvs MCE and SEvs GE) after adjusting for baseline difference.

Laird et al.

Laird et al. BMC Musculoskeletal Disorders 2012 13:169   doi:10.1186/1471-2474-13-169

Open Data