DEEP NECK FLEXORS IMPACT RECTUS ABDOMINIS MUSCLE ACTIVITY DURING ACTIVE STRAIGHT LEG RAISING

Abstract

Background:

Active straight leg raising (ASLR) is commonly performed to test fundamental movement competency. Head control or positioning can affect the abdominal muscle activity during movements.

Purpose:

To investigate whether abdominal muscle activity differs when the head is extended or when deep neck flexor (DNF) muscles are selectively activated during the ASLR.

Study Design:

Cross-sectional

Methods:

Participants were included based on the following criteria: 1) age>17 years; 2) no spinal or lower extremity pain in the prior month; 3) the vertical line of the malleolus in an elevated the lower limb resides below the knee joint line of a non-moving lower limb during ASLR and above during a passive straight leg raising in each lower limb; and 4) no history of diagnosed spinal deformities or central nervous system disorders. Participants with > 39% reference voluntary contraction in the sternocleidomastoid muscle during the craniocervical flexion test (CCFT) of 24 mmHg target were excluded from the analyses. Right ASLR was repeated in each of the following three head conditions in a random order: 1) neutral head position, 2) head extended by 25 °, and 3) CCFT maintained with a 24 mmHg target. Among the three head conditions, the relative latency for the onset of the right rectus femoris (RF) muscle during the right ASLR and the muscle activity amplitude for 50ms were compared after the onset of RF muscular activity in the following muscles: left rectus abdominis (RA), bilateral external obliques, bilateral internal obliques, and left gluteus maximus muscles.

Results:

Data from 31 participants (21 women and 10 men, mean age=22.5 years) were analyzed. The relative latency of the left RA (Hedges’ g = 0.39, p=.038) was higher in the CCFT condition (mean±SD=112.1ms±86.0ms) than that in the neutral head condition (82.9ms±58.6ms). However, no difference (all p>.05) was observed in other measures between the groups.

Conclusion:

In people with impaired movement competency in ASLR, head extension did not alter the abdominal muscle activities in ASLR. However, selective activation of the DNF muscles delayed the onset of RA muscle activity during the ASLR.

Level of Evidence:

4

INTRODUCTION

Active straight leg raising (ASLR) is a common movement test to evaluate multi-segmental control,¹ including the trunk, the hip raising a lower limb, and the hip supporting a base. Competency in ASLR is considered fundamental because of its wide support base to obtain movement control in a narrow base of support, such as that seen during standing.² Recently, a valid and reliable system³,⁴—the Functional Movement System (FMS™)—has been developed for grading movement competency,² most commonly used for injury risk assessment in professional athletes.⁵ In ASLR of the FMS™, a score of 1 indicates impaired movement competency and is assigned when the vertical line of the malleolus in an elevated the lower limb resides below the knee joint line of a non-moving lower limb that remains in the neutral position.² In those who score a 1 in the ASLR of the FMS™, there are two potential subgroups: 1) patients with limitations in both ASLR and passive straight leg raising (PSLR), and 2) those with limitation in ASLR but not in PSLR. In the latter subgroup, the mechanisms underlying the limited ASLR likely include stability or motor control dysfunction (SMCD) of the trunk and the hip muscles² and thus their ASLR was named ASLR-1-SMCD.

During movement, the muscle activity in one region of the human body can affect the muscles in other regions; such a relationship can be observed between the head and neck regions and the lumbopelvic region. The influence of trunk control based on head conditions has been observed in the hook-lying position⁶ and during lifting,⁷ abdominal hollowing,⁸ and prone bridging.⁹ One head position that potentially influences trunk control is head extension, and another is upper cervical flexion.^7-9 Selective activation of the deep neck flexor (DNF) muscles that contribute to the segmental control of the cervical spine results in upper cervical flexion and is often included in postural correction stability training during physiotherapy.¹⁰ Falla et al¹¹ demonstrated that the DNF muscles and lumbar multifidus muscles are coactivated with a specific postural correction instruction while sitting, a procedure designed for the selective activation of the DNF muscles. Among the trunk muscles, previous authors have consistently reported that abdominal muscle activities are altered with different head and neck conditions.^6-9 Therefore, selective activation of the DNF muscles was hypothesized to influence the activation of the abdominal muscles during movements.

Specific exercises are necessary to change the motor control of the abdominal muscles.¹² Participants with ASLR-1-SMCD require a specific exercise to change the feedforward control of the abdominal muscles and to enhance ASLR and other movement competencies (corrective exercise). If head and neck control /position influences the feedforward control of the abdominal muscles, head control may be important in corrective exercises. Therefore, the aim of this study was to investigate whether abdominal muscle activity differs when the head is extended or when deep neck flexor (DNF) muscles are selectively activated during the ASLR.

METHODS

Design

This cross-sectional study used superficial electromyography (EMG) to perform assessments in three head conditions (Figure 1) in a random order. The study protocol was approved by the institutional research ethics committee (Saitama Prefectural University, #29028). All the participants provided written consent before data collection.

Figure 1.

Schema of measurement conditions. ASLR, active straight leg raising; CCFT, craniocervical flexion test.

Participants

Using convenience sampling, participants with ASLR-1-SMCD were recruited via advertisements placed throughout the university. The inclusion criteria included: 1) age > 17 years; 2) no spinal or lower extremity pain in the previous month; 3) the vertical line of the malleolus in an elevated the lower limb resides below the knee joint line of a non-moving lower limb during ASLR and above during PSLR in each lower limb (an FMS™ ASLR score of “1”); and 4) no history of diagnosed spinal deformities or central nervous system disorders. The exclusion criteria included: 1) any pain during right or left ASLR and 2) right ASLR score of 0, 2, or 3 in the FMS™, as confirmed by an author (HT) certified for FMS™ level 1. The participants who exhibited > 39% of the reference voluntary contraction (RVC) in the sternocleidomastoid (SCM) muscle activity during the craniocervical flexion test (CCFT) of 24 mmHg target were excluded from the analyses because selective activation of the DNF muscles in the CCFT was not guaranteed, considering an upper threshold of 95% confidence intervals in healthy individuals (39.18 % RVC) in a previous study.¹³

Procedures

The participants’ arms were placed beside their trunk with palms facing up. Head movement was monitored with an inertial measurement unit (IMU) of myoMOTION™ system (Noraxon, U.S.A., Inc., AZ, USA), which was attached on the forehead. The participants raised their right lower limb without knee flexion to their end range of hip flexion from a relaxed supine lying position. We asked the participants to raise their right lower limb to their end range for 1 s immediately after they saw the lighting that was set in front of their face. The timing of the lighting was random, ranging from 10 to 15 s. The participants repeated the ASLR 20 times in each position so that the researchers could obtain at least 10 clear electromyography (EMG) datasets for optimal accuracy evaluation of muscle activity onset in each of the following three head conditions: 1) neutral head position, 2) head in the extension position, and 3) CCFT with 24 mmHg target.

In the neutral head condition, the frontal plane of the face was horizontal. In the head extension condition, the frontal plane of the face was tilted to 25 ° from the measurement table. The participants held this head position during measurement. The 25 ° extension was selected throughout pilot testing, considering its feasibility across the participants, minimum effort for the task, and negligible influence on the activity of the abdominal muscles. In the CCFT condition, the participants performed the craniocervical flexion and increased the pressure on the Chattanooga Stabilizer Pressure Biofeedback (DJO, LLC, USA), which was placed suboccipitally, from 20 mmHg to 24 mmHg with minimum activity of the superficial neck flexors. The lighting was shown while the participants maintained the 24 mmHg pressure.

The participants rested for 10 min in the supine position between performing the three head conditions. At the beginning of the measurement in each condition, the participants practiced ASLR for three to five times to establish the speed of raising the right lower limb. Furthermore, each participant practiced the CCFT for less than five minutes prior to the CCFT condition.

After the EMG data were collected during ASLR in the three head conditions, the participants performed three reference contractions for three seconds in each recorded muscle to be able to standardize the amplitude of the muscle activity. For the SCM, craniocervical flexion and the head-lift procedure were used as RVC, according to a previous study.¹³ For the other muscles, maximum voluntary contraction was performed according to the standard procedures.¹⁴

The CCFT

An author (HT) experienced in the CCFT conducted the test in accordance with an established procedure¹⁵ and gave verbal instructions as well as allowed visual feedback (from the pressure biofeedback) to the participants in order to attempt to maintain the 24 mmHg pressure. The CCFT is a reliable and valid procedure to assess performance of the DNF muscles,^13,16,17 thus, it was chosen as a condition to provide selective activation of the DNF muscles. The pressure target of 24 mmHg was selected, considering its feasibility in most of the participants.¹⁸

Measurements

The primary outcome measures included the following: 1) relative latency of the onset of the right rectus femoris (RF) muscle activation and 2) the muscle activity during an early phase of ASLR in the left rectus abdominis (RA) muscle, bilateral external oblique (EO) muscles, bilateral internal oblique (IO) muscles, and the left gluteus maximus (GM) muscle. Considering the asymmetry of movements and force production during right ASLR and limited sensors, the left RA and the left GM muscles as well as bilateral IO and EO muscles were selected. The onset of muscle activity was detected with raw EMG data using the visual detection method that has high inter-session reliability.¹⁹ In order to blind the assessor to the signal, a research assistant randomized the order of data presentation for each participant and masked the labels of the muscles in the EMG data during analyses. An assessor zoomed the data in order to identify the EMG onset based upon the following criteria: 1) a rise in the EMG amplitude above baseline levels, 2) recruitment of additional motor units, or 3) increased firing rate of active motor units. When the timing of onset was unclear because of overlap with other noises [e.g., electrocardiographic (ECG) complex] or earlier onsets of muscle activity than a trigger of a light, all EMG analysis for that trial was not undertaken. For the amplitude of muscle activity, amplitude data 50 ms after the onset of RF muscle activity were calculated, considering the potential differences among the three head conditions found during pilot testing. The mean of the first 10–15 datasets was used for statistical analyses of both primary measures.

The secondary outcome measures included the following: 1) demographics (age, sex, and body mass index), 2) time to reach 95% of the hip flexion range, and 3) amplitude of SCM activity during 50 ms (25 ms before and 25 ms after the trigger of lighting) for exclusion of the participants who did not selectively activate the DNF muscles. The IMU sensors of myoMOTION™ system were attached on the pelvis and the right thigh to monitor the movement of the right lower limb. Motion data of the head and right lower limb were captured and processed using myoMOTION™ system with sampling frequency of 100 Hz that was synchronized in the EMG system and an external trigger of a light.

EMG processing

Following standardized skin preparation as per the SENIAM recommendations, self-adhesive Ag/AgCl electrodes (ECG electrodes 2009111-150, CareFusion, Finland) were attached at standardized positions of the right SCM, left RA, bilateral EO, bilateral IO, right RF, and left GM muscles with 20-mm inter-electrode distance. In particular, for the SCM muscle, the electrodes were placed on the sternal portion of the muscle, with the electrode center 1/3 of the distance between the mastoid process and the sternal notch.²⁰ For the RA muscle, the electrode was placed 4 cm lateral to the navel and vertically with the lower border of the caudal electrode at navel level.²¹ For the EO, the electrodes were placed over a line extending from the most inferior point of the costal margin to the opposite pubic tubercle, where a lateral electrode was placed 14 cm lateral to the median line, lower the level of 1 cm above umbilicus.²² For the IO muscle, a lateral electrode was placed at 2 cm lower than the most prominent point of the anterior superior iliac spine (ASIS), and a medical electrode was placed inclined 6 ° inferomedially to the horizontal line.²² For the RF muscle, the electrodes were placed at 50% on the line between the ASIS and the superior part of the patella as per the SENIAM recommendations. For the GM muscle, the electrodes were placed at 50% on the line between the sacral vertebrae and the greater trochanter as per the SENIAM recommendations.

EMG data were collected and processed using myoMUSCLE™ system (Noraxon, U.S.A., Inc., AZ, USA) with a sampling frequency of 1500 Hz. The amplitude of the muscle activity was evaluated by reducing the ECG complex, filtering the EMG data with 20–500-Hz band pass filter,²³ and calculating the root mean square with 100 ms sliding window.

Analyses

Sample size estimation

Sample size was estimated using an internal pilot study²⁴ of 12 participants. One participant was excluded because of incorrect CCFT, and the data of 11 participants indicated potential differences in the relative latency of left RA onset (effect size f = 0.24). G*Power 3²⁵ estimated that 30 participants were required to detect statistical significance with α = .05 and β = .8. Considering a potential exclusion rate of 20%,¹⁸ 36 participants were needed, and an additional 24 participants were recruited because there was no methodological change from the pilot study.

Statistics

Descriptive analyses were used for demographic measures, and one-way repeated measures ANOVA was used for other variables with IBM SPSS version 25 (IBM Corp, Armonk, New York). Statistical significance was set at 5%, and the effect size of f was calculated from partial η²-values using G*Power 3.²⁵ The following criteria of f-value were used: .10 = small effect size, .25 = medium effect size, and .40 = large effect size.²⁶ Greenhouse-Geisser correction was undertaken when Mauchly's sphericity test denied the assumption of sphericity. Bonferroni corrections were used for post-hoc comparisons. Effect size of Hedges’ g was calculated, where the following criteria were used for interpretation: .2 = small effect, .5 = medium effect, and .8 = large effect.²⁶

Results

From the included 36 participants, five were excluded from the analyses because of incorrect CCFT (n = 3) or a technical problem during data collection (n = 2). The data of 31 participants (21 women and 10 men) were analyzed; their mean and standard deviation (SD) values for their age and body mass index were 22.5 (5.6) years and 21.1 (2.2) kg/m², respectively.

Table 1 presents the relative latency of the onset of muscles to the onset of the right RF muscle in ASLR. A statistically significant difference with a large effect size was detected only in the left RA (p = .043, f = 0.47). In the post-hoc analysis, the p-value (.038) was not less than an adjusted statistically significant level using the Bonferroni correction (.05/3). However, there was an effect size approaching medium, with a greater value in the CCFT than in the neutral condition (Hedges’ g = 0.39).

Table 1.

Relative latency of the onset of muscles to the onset of rectus femoris muscle in active straight leg raising. Values are presented with mean ± SD, positive values indicate delayed onset from the onset of rectus femoris muscle.

Muscle	Neutral (ms)	Extension (ms)	CCFT (ms)	p-value	Effect size f
Left RA	82.9 ± 58.6	86.0 ± 65.0	112.1 ± 86.0	.043*	0.47
Left EO	54.1 ± 44.0	41.5 ± 51.9	60.7 ± 54.8	.335	0.30
Right EO	33.6 ± 41.7	52.0 ± 39.3	43.4 ± 42.2	.458	0.17
Left IO	57.3 ± 42.1	41.9 ± 23.7	57.0 ± 60.5	.093	0.46
Right IO	31.5 ± 43.8	32.6 ± 61.4	32.2 ± 35.6	.958	0.03
Left GM	95.5 ± 115.6	93.1 ± 131.2	92.3 ± 107.8	.950	0.03

Neutral, head in neutral; Extension, head in 25 ° extension; CCFT, cranio-cervical flexion test; RA, rectus abdominis; EO, external oblique; IO, internal oblique; GM, gluteus maximus.

*Post-hoc analysis: Neutral vs CCFT, p = .038 (Hedges’ g = 0.39); Neutral vs Extension, p = 1.0 (Hedges’ g = 0.05); Extension vs CCFT, p = .253 (Hedges’ g = 0.34).

Table 2 presents the amplitude of the muscle activities for 50 ms after the onset of the right RF muscle in ASLR. There were no significant differences (all p > .05).

Table 2.

Amplitude of muscle activities for 50msec after the onset of rectus femoris muscle in active straight leg raising. Values are presented with mean ± SD.

Muscle	Neutral (%MVC)	Extension (%MVC)	CCFT (%MVC)	p-value	Effect size f
Left RA	2.4 ± 1.8	2.7 ± 1.7	2.2 ± 1.6	.094	0.37
Left EO	5.2 ± 7.2	4.7 ± 5.7	5.1 ± 6.8	.618	0.15
Right EO	3.2 ± 2.7	3.3 ± 2.6	3.3 ± 2.9	.912	0.08
Left IO	3.1 ± 2.1	3.2 ± 2.6	3.4 ± 2.4	.552	0.83
Right IO	6.2 ± 4.7	6.3 ± 4.7	6.4 ± 5.5	.895	0.10
Left GM	3.1 ± 2.7	3.1 ± 2.8	3.3 ± 3.1	.686	0.13

MVC, maximum voluntary contraction; Neutral, head in neutral; Extension, head in 25 ° extension; CCFT, cranio-cervical flexion test; RA, rectus abdominis; EO, external oblique; IO, internal oblique; GM, gluteus maximus.

The difference in the time taken to reach 95% of the hip flexion range among the three head conditions was not significantly different (p = .879, f = 0.05), with the mean (SD) time of 0.7 (0.2) sec in all head conditions, namely, neutral, head extension, and CCFT. The mean (SD) amplitude of the right SCM activity was 3.9 (2.8) %RVC, 4.4 (3.2) %RVC, and 8.1 (4.6) %RVC in the neutral head, head extension, and CCFT conditions, respectively.

Discussion

During movement, the muscle activity in one region of the human body can affect the muscles in other regions. Previous authors have reported that abdominal muscle activities were altered with different head and neck conditions.^6-9 Therefore, it was hypothesized that head extension or selective activation of the DNF muscles could change the firing pattern of the abdominal muscles during lower limb movements.

Contrary to the research hypothesis, abdominal muscle activities during ASLR were not altered during the head extension condition as compared to that in the neutral head condition. Previous authors have shown the influence of head extension on abdominal muscles during functional tasks^7,8 that included > 35 ° of head extension. Therefore, it could be possible that the 25 ° of head extension in this study was not enough to change the firing pattern of the abdominal muscles.

With almost medium effect size, the CCFT condition increased only the relative latency of the left RA muscle to the onset of the right RF muscle in ASLR as compared to that in the neutral head condition. A potential reason for the increased relative latency of the left RA muscle to the onset of right RF muscle during the CCFT may be associated with a hypothesized reciprocal relationship between the inhibited outer muscles (e.g., RA muscles) and the facilitated inner muscles (e.g., transversus abdominis); however, this hypothesis cannot be proved in the absence of EMG data for transversus abdominis. A previous study has demonstrated that the craniocervical flexion increased the amplitude of the inner abdominal muscles with a large effect size in the hook-lying position,⁶ suggesting facilitation of the transversus abdominis muscles by craniocervical flexion. Hodges and Richardson²⁷ demonstrated that a normal firing pattern during hip flexion of the onset of the transversus abdominis followed by the RA muscle was reversed in people with LBP. Further studies that assess the onset of the transverse abdominis muscles are necessary to confirm this hypothesis.

The primary measures in this study could have been affected by the speed of the lower limb movement. However, no difference was observed in the time to reach 95% of the hip flexion range, suggesting that a confounding factor was controlled. Further, the mean (SD) amplitude of the right SCM activity was 8.1 (4.6) %RVC in the CCFT condition. A previous study¹³ using a nasopharyngeal electrode for assessment of the DNF muscles demonstrated that the mean %RVC was 16.74% in the SCM and 45.46% in the DNF muscles during the CCFT with 24 mmHg target. Therefore, selective activation of the DNF muscles can be assumed during the CCFT with 24 mmHg target in this study, although direct evidence on the use of nasopharyngeal electrode in the DNF muscles was not attempted.

The results of the current study suggest the need for further investigations on the following points for better understanding of effective corrective exercises and reliable physical screening. First, whether the altered muscle firing pattern of the left RA due to the selective activation of the DNF muscles would be useful or harmful as a corrective exercise for the ASLR. When the selective activation of the DNF muscles would be useful as a corrective exercise for a low score on the ASLR, the relative latency in the left RA muscle would be more delayed during the right ASLR or the left RA amplitude would be less at an early phase of the right ASLR in participants with ASLR score of 3 in the FMS™ than those with ASLR-1-SMCD. This hypothesis should be investigated in a cross-sectional study and confirmed in a clinical trial. Second, whether the results of this study can be replicated in other activities, such as squatting and hurdle stepping, should be assessed. If so, selective activation of the DNF muscles could be included in the exercises to improve competency in other activities. Third, whether the hip flexion ranges during the ASLR according to the FMS™ procedure is altered among different head conditions should be investigated. Instantaneous hip flexion was used to identify clear onsets of muscles for achieving the purpose of this study. However, such an instantaneous movement is not used in the FMS™ procedure (i.e., smooth and natural movement of the lower limb) and speed of limb movement can be a contributing factor of the onset of abdominal muscle activity.²⁸ If there are differences in the hip flexion ranges, head conditions should be standardized in the ASLR testing of the FMS™.

One of the clinical implications of this study is the identification of the importance of head control in the exercise. Selective activation of the DNF muscles may be used to change activity pattern of the RA muscles. In relation to the head position, it would not be required to avoid head extension when the head extension is < 25 °.

Limitations

A limitation of this study is that limited muscles were evaluated using surface EMG and only the right side of ASLR was tested because of limited number of sensors. However, all the participants had ASLR-1-SMCD on both the sides and no limb asymmetries were present. Investigation of other abdominal muscles, such as the transversus abdominis (using intramuscular EMG) would be required for further understanding the effect of head condition on all abdominal muscles.

Conclusions

The results of the current study indicate that in people with ASLR-1-SMCD, head extension did not alter the abdominal muscle activities. However, selective activation of the DNF muscles delayed the onset of left RA muscle activity in right ASLR.