Participants
Forty healthy subjects participated in this study. Inclusion criteria for the current study were: (a) ages between 20 and 45 years; (b) no history of LBP or lower extremity pain in the last 6 months; (c) able to perform the improved Biering-Sorensen test (BST) in the prone position, and Abdominal Drawing-in Maneuver (ADIM) in the supine position for more than 30s; (d) right-handed; and (e) body mass index (BMI) within ±10% of international standards. Exclusion criteria were: LBP patients; a history of abdominal or spinal surgery; previous experience of real-time ultrasound imaging on trunk muscles; any medical condition affecting the spine (eg, ankylosing spondylitis, scoliosis, rheumatoid arthritis, osteoporosis, systemic disease, and severe neurological disorders), or pregnancy. All subjects reviewed and signed informed consent that was approved by the Institutional Research Ethics Committee of the First Affiliated Hospital, Sun Yat-sen University. We confirmed that all methods were performed following the ethical standards of the Declaration of Helsinki.
The healthy volunteers were recruited from posted notices and social network platform, both genders were included and were randomly divided into one of two groups. Group 1 (real-time ultrasound visual feedback group, experimental group, Exp) contained 16 females and 4 males and group 2 (verbal instructed group, control group, CG ) contained 15 females and 5 males.
Procedures
Images of the LM and TrA were acquired in B-mode with a portable ultrasound machine (SonoSite M-Turbo, Seattle, USA) with a 2-5MHz curvilinear-array (for LM) or 6-13MHz linear-array transducer (for TrA), automatically adjusted by the scanning depth.
All participants were firstly given a verbal explanation regarding the purpose and operation procedure of the experiment and the anatomical structure and function of the multifidus and abdominal muscles before the test. Besides, the locations, the morphological and structural changes of the TrA and LM during contraction were shown and identified using ultrasound imaging in the experimental group.
Image acquisition for each condition and each time point (Trest, Tc-max, Tc-15s, Tc-30s) was repeated three times. To maximize time efficiency, one operator recorded the time and saved the image, whereas the other operator positioned the transducer and optimized the quality of the image. To avoid potential fatigue associated with the detection sequence, the muscles (TrA and LM) being tested and imaged were randomly assigned.
Assessment of LM Thickness
Images of the right LM at rest and during maximum isometric contraction (MIC) were acquired following procedures outlined by Zhang et al. [20] using ultrasound. Participants underwent measurements in randomly-assigned order. Participants were firstly in a prone lying position, with a pillow placed under the abdomen to diminish the lumbar lordosis. The lumbar spinous processes from L1 to S1 were palpated and marked on the skin with an indelible marker, which was then identified by real-time ultrasound.
The thickness of the right LM was imaged in a longitudinal section at the L4-5 level. Three separate resting ultrasound images were collected immediately after exhalation. To avoid affecting the muscle shape, the operator must be careful not to compress the skin with the transducer. Then, subjects were asked to maintain back muscle contractions for 30 seconds, and performed three times with a two-minute interval between tasks. To obtain the maximum isometric contraction data, subjects were positioned in the standard prone position with the upper limbs positioned overhead, shoulders abducted to approximately 120°. Subjects were then asked to lift their head, trunk, and upper extremities with maximum effort, which had been used previously to activate the contraction of LM [20-22]. Oral encouragement was given to the individuals in the control group during MIC. Besides the verbal instruction, subjects in the experimental group were required to watch the real-time ultrasound imaging and maintain continuous contraction of the LM with maximum effort. Image acquisitions were collected at three time points (Tc-max, Tc-15s, Tc-30s) during contraction, and the experiment was repeated three times. A single practice and two preliminary acquisitions were executed before images being recorded. Linear measurements of the LM thickness were taken in all subjects using on-screen calipers from the superior border of the LM to the tip of the L4-5 zygapophyseal joint (Fig. 1a, b).
Assessment of TrA Thickness
Images of the right TrA muscle were acquired at rest and during the ADIM maneuver [23, 24], with subjects in a supine hook-lying position and their arms crossed over the chest. US images were collected with the transducer placed transversely just along the midaxillary line at the level of the umbilicus, in which the middle of the TrA muscle belly was positioned within the field of view [25]. All images were obtained at the end of exhalation to avoid the influence of respiration.
The ADIM is a fundamental motor control training used to preferentially activate the TrA muscle contraction in comparison with the more superficial abdominal muscles [25]. In this study, the ADIM was used to assess the altered muscle thickness associated with a voluntary contraction of the TrA muscle. To perform the ADIM, participants were instructed to “draw-in your umbilicus toward the spine without moving back or pelvis, while comfortably breathing in and out.” The angle of the transducer was adjusted properly to collect the best quality images while keeping the transducer perpendicular to the surface of the skin in the same position. During the ADIM, the participants in the experimental group were asked to view the changes of muscle thickness on the ultrasound monitor as visual feedback in an attempt to maximize a preferential and continuous TrA contraction. Image acquisitions were obtained at three time points (Tc-max, Tc-15s, Tc-30s) during ADIM, and were performed 3 times separately with a 2mins interval between each ADIM task. The thickness of the TrA muscle was assessed from the superior fascial border to the inferior fascial border in centimeters (Fig. 2a, b).
Statistical analysis
Statistical analysis was performed using SPSS 22.0 (SPSS Inc. Chicago, IL, USA). Shapiro-Wilk test was used to identify the normality of data distribution. Baseline demographic characteristics such as age, weight, height, and BMI were reported with descriptive statistics (mean±SD) and were compared between groups with unpaired t-test. Separate 2-by-4 repeated measures analyses of variance (ANOVAs) were used to examine intervention effects (dependent variables), with the group (control or experimental) as between-subject variable and time (Trest, Tc-max, Tc-15s, Tc-30s) as the within-subject variable. The LM and TrA thickness at each time point was reported as a mean with the standard deviation of each group. Post-hoc tests for multiple comparisons with Bonferroni adjustments were adopted when a significant interaction effect existed. The significance level was set at a priori alpha level of 0.05.
Table 1 Participant demographics (mean±SD)
group
|
Participants (n)
|
Gender (F/M)
|
Age (years)
|
Weight ( kg)
|
height(m)
|
BMI(kg/m2)
|
Exp
|
20
|
16/4
|
25.95±2.89
|
55.65±8.45
|
1.62±0.06
|
21.08±1.85
|
CG
|
20
|
15/5
|
26.60±2.58
|
56.25±9.14
|
1.63±0.06
|
21.02±2.00
|
P
|
|
0.705
|
0.458
|
0.803
|
0.601
|
0.922
|
Exp=experimental group, CG=control group.
Results of the ANOVA for intervention effects on LM thickness showed significant interaction effect between the factors of time and group (F = 3.137, P = 0.027), also statistically significant main effects due to time (F = 31.45, P < 0.001) and group (F = 12.34, P < 0.001). The results showed that no significant differences were found in the thickness of LM at rest, Tc-max and T15s between groups (P > 0.999, P > 0.999, and P >0.414, respectively). However, the ability to recruit LM muscle contraction differed at T30s between groups, with subjects in the experimental group that obtained visual ultrasound biofeedback maintaining a relative maximum contraction compared to the CG group (P = 0.006). At the time point of Tc-max, T15s, and T30s, the results of the intra-group comparison showed that LM contraction thickness in the Exp group was superior to Trest (P < 0.001). While, in the CG group, the thickness of LM had no significant statistical significance in T30s compared with Trest (P > 0.999). (table 2 and Fig.1c)
Table 2 Comparison of LM thickness at different time points between two groups (mean ± SD, cm)
Time point
|
Exp (n=20)
|
CG (n=20)
|
Trest
|
2.432±0.328
|
2.460±0.285
|
Tc-max
|
3.162±0.334#
|
3.058±0.342#
|
T15s
|
3.101±0.332#
|
2.849±0.350#
|
T30s
|
3.052±0.326#,*
|
2.662±0.282
|
Fgroup / P value
|
12.34 / < 0.001
|
Ftime / P value
|
31.45 / < 0.001
|
Fgroup*time / P value
|
3.137 / 0.027
|
NOTE: Exp=experimental group, CG=control group. # Compared with the resting state, P < 0.05 ;
* Compared with the control group, P < 0.05.
Results of the ANOVA for intervention effects on TrA thickness showed significant interaction effect between the factors of time and group (F = 3.583, P = 0.015) and statistically significant main effects due to time (F = 40.54, P < 0.001) and group (F = 14.01, P < 0.001). Post-hoc analysis showed that there were no significant differences in TrA thickness between the two groups at Trest and Tc-max (P > 0.999 and P > 0.999, respectively). While, at 15s and 30s of continuous contraction, TrA contracted thickness in the Exp group was significantly better than that in the CG group (P = 0.031 and P= 0.010, respectively). At the time point of Tc-max, T15s, and T30s, the results of the intra-group comparison showed that TrA contraction thickness in the Exp group was superior to Trest (P < 0.001). While, in the CG group, the thickness of TrA had no significant statistical significance in T30s compared with Trest (P > 0.999). (table 3 and Fig.2c)
Table 3 Comparison of TrA thickness at different time points between two groups (mean ± SD, cm)
Time point
|
Exp (n=20)
|
CG (n=20)
|
Trest
|
0.281±0.076
|
0.285±0.081
|
Tc-max
|
0.539±0.122#
|
0.518±0.103#
|
T15s
|
0.521±0.118#,*
|
0.412±0.099#
|
T30s
|
0.468±0.120#,*
|
0.348±0.098
|
Fgroup / P value
|
14.01 / < 0.001
|
Ftime / P value
|
40.54 / < 0.001
|
Fgroup*time / P value
|
3.583 / 0.015
|
NOTE: Exp=experimental group, CG=control group. # Compared with the resting state, P < 0.05;
* Compared with the control group, P < 0.05.