Bilateral Activity of Spine Extensors and Rotators during Asymmetric Lumbar Stabilization Exercises Executed in Prone, Quadruped, and Standing-Prone Positions

Abstract

(1) Background: Most daily activities and sport gestures involve asymmetric movement patterns of the upper and lower extremities, transferring asymmetric mechanical loadings to the spine. Therefore, asymmetric lumbar stabilization exercises are frequently prescribed in athletic programs and preventive/rehabilitation interventions. This study analyzed the bilateral activity of the thoracic erector spinae (ES), lumbar multifidus (MF), external oblique (EO), and internal oblique (IO) during asymmetric lumbar stabilization exercises executed in prone, quadruped, and standing-prone positions, rising an upper and/or lower limb in all possible combinations. A limited subset of these data has been previously published in earlier studies. (2) Methods: Surface EMG signals were bilaterally recorded from the selected muscles using wireless EMG sensors. (3) Results: ES, MF, and oblique muscles’ activity was significantly higher in prone, standing-prone, and quadruped positions, respectively, and was maximized by specific limb rise combinations (up to 69%, 64%, 34%, and 24% maximum-voluntary-isometric-contraction for ES, MF, EO, and IO). The bilateral difference in muscle activation was significantly higher in the quadruped position and revealed different strategies used to stabilize the body in response to the different exercise conditions. (4) Conclusions: The study results can provide deeper insights into the stabilizing function of the lumbar and oblique muscles and aid in designing optimal progressions for lumbar stabilization exercises.

1. Introduction

A number of research studies have consistently provided evidence that low back pain is correlated with weakness and reduced endurance of lumbar extensor muscles [1,2,3,4,5,6,7,8]. For these reasons, lumbar stabilization exercises are frequently prescribed in preventive and rehabilitation interventions for the lumbo-pelvic-hip complex, particularly for low back pain [9]. Additionally, lumbar stabilization exercises are used in training programs for various sports to prevent injuries and improve performance [10,11]. Stiffening the torso, the lumbar, and other core muscles enables safer and more efficient transmission of the power generated by the lower limbs to the upper body during explosive movements [11,12].

In these contexts, asymmetric lumbar stabilization exercises, which engage the two sides of the body differently, are routinely used alongside symmetric exercises [9,10]. Indeed, most daily activities and sporting gestures involve asymmetric movement patterns in the upper and lower extremities. Consequently, during the execution of these tasks, the vertebral column often experiences asymmetric mechanical loadings, which are primarily engendered by load transfers among the spine, arms, pelvis, and legs. To maintain a safe neutral position of the spine in such situations, the spine-stabilizing muscles on one side of the body need to be engaged differently than those on the other side. Furthermore, asymmetries in lumbar muscle strength between the left and right sides are frequently observed and have been suggested as potential indicators of lumbar pathology [13,14,15]. Addressing these imbalances may require targeted corrective exercises [13,14,15]. Careful progressions of asymmetric lumbar stabilization exercises, appropriately dosed between the sides of the body, can gradually accustom the spinal joint structures to asymmetric stimuli. This approach can also be utilized to manage and treat imbalances in lumbar muscles between the body’s two sides.

Asymmetric lumbar stabilization exercises are frequently performed in prone and quadruped positions, involving lifting the upper and lower limbs in various combinations through hip extension and shoulder flexion movements. Elevating either an upper or lower limb off the ground from these body positions reduces the extension of the base of support and shifts the position of the body’s center of mass. Asymmetric combinations of limb lifts further modify the base of support and its symmetry, resulting in trunk instability in the transverse and frontal planes of the body. This further challenges the lumbar stabilizers and the oblique muscles, which can reach considerable levels of activity avoiding the use of external overloads and the related mechanical loading induced on the spine’s structures [16,17].

Previous studies have recorded electromyographic (EMG) activity bilaterally from lumbar and oblique muscles during asymmetric lumbar stabilization exercises, including elevation of an upper limb and the contralateral lower limb (contralateral arm–leg rise) from the quadruped [18,19,20,21,22,23,24,25,26,27], prone [22,27], and plank positions [24], and elevation of a lower limb (leg rise) [18,19,20,21,26] or an upper limb (arm rise) [26] from the quadruped position. Currently, there is a lack of EMG data for isolated arm rises and leg rises from the prone position, as well as for the simultaneous rise of an upper limb and the ipsilateral lower limb (ipsilateral arm–leg rise) from the prone and quadruped positions. A recent EMG study from our laboratory analyzed lumbar multifidus and erector spinae activity during the ipsilateral and contralateral arm–leg rise from the standing-prone position [28]. However, the oblique muscle activity was not recorded. Furthermore, isolated arm or leg rises from the same position were not investigated. Knowledge of all the missing information listed above would provide a more comprehensive EMG characterization and a deeper understanding of asymmetric lumbar stabilization exercises. Additionally, due to the frequent observed lumbar muscle asymmetries between the left and right sides of the body [15], it is also of interest to analyze each asymmetric exercise performed twice in a mirrored manner relative to the sagittal plane of the body (for example, left arm rise and right arm rise).

Here, we examine the bilateral activity of the thoracic erector spinae (ES), lumbar multifidus (MF), external oblique (EO), and internal oblique (IO) during asymmetric lumbar stabilization exercises performed in prone, quadruped, and standing-prone positions rising an upper and/or a lower limb in all possible combinations. Specifically, this study aims to:

• compare the levels of activation of each muscle across the different body positions and limb elevation combinations, and between the two sides of the body;
• compare the difference of activation between the two sides of the body among the muscles, and across the different body positions and limb elevation combinations;
• identify the different muscular strategies employed to stabilize the body in response to the changes in the base of support and the position of the body’s center of mass induced by the different exercise conditions.

The results of this investigation offer a better understanding of the stabilizing function of the lumbar and oblique muscles during the execution of asymmetric tasks. Furthermore, by grading muscle activation across various body positions and combinations of limb elevation, this information can assist in designing optimal progressions for lumbar stabilization exercises in rehabilitation interventions and athletic programs.

2. Materials and Methods

2.1. Participants

Ten male and ten female volunteers (average age: 29 y, range: 21–57 y; average height: 1.77 m, range: 1.58–1.90 m; and average body mass: 75 kg, range: 52–89 kg) were enrolled from a nearby fitness center. The subjects considered for inclusion possessed a track record of expertise in bodyweight resistance training (intermediate-to-advanced level) and demonstrated a full active range of shoulder flexion and hip extension motion, specifically at least 180 degrees of shoulder flexion and about 20 degrees of hip extension beyond the neutral position. Exclusion criteria included musculoskeletal injuries; musculoskeletal disorders affecting the spine or limbs; and the inability to perform lumbar stabilization exercises with correct form and technique, without experiencing any discomfort. All participants provided written informed consent for their inclusion in the study after receiving comprehensive information on all relevant aspects of the experiment. The investigation was carried out following the principles outlined in the Declaration of Helsinki and received approval from the ethics committee of the University of Perugia.

2.2. Selected Exercises and Muscles

Each participant performed eight isometric limb rise exercises: unilateral limb rises (right leg, left leg, right arm, and left arm), contralateral arm–leg rises (right leg and left arm, left leg and right arm), and ipsilateral arm–leg rises (right leg and right arm, and left leg and left arm). These movements were executed from three different body positions: prone position with relaxed upper limbs at the sides of the head; quadruped position; and prone-standing position, i.e., standing with the chest facing the ground, longitudinal trunk axis parallel to the ground, and relaxed upper limbs perpendicular to the ground (hip and shoulder at 90° flexion, and knee and elbow near full extension). Participants were instructed to elevate their limbs until full shoulder flexion and hip extension were gently approached without forcing against the joint structures that limit the range of motion. Each static position was held for 10 s while maintaining the spine close to the neutral zone. Minor deviations from the neutral position were allowed to accommodate pelvic and scapular movements and facilitate the tasks. During each trial, surface EMG signals were bilaterally recorded from the ES, MF, EO, and IO muscles.

2.3. Familiarization Session

One week before the testing session, all participants underwent a pretest familiarization session despite already being familiar with core stability exercises. In this session, a professional trainer detailed the testing protocol and demonstrated the exercises, which were then practiced by the subjects under the trainer’s guidance. Only the quadruped ipsilateral arm–leg rise and the standing-prone ipsilateral/contralateral arm–leg rise demanded considerable effort during familiarization.

2.4. Testing Session

After a 10 min general warm-up and specific hip and shoulder mobility exercises, we recorded the EMG activity of the eight selected muscles (four bilateral muscle pairs) during their maximum voluntary isometric contractions (MVIC). The body positions and resisted-movement conditions adopted in the MVIC tests for each muscle are detailed in Appendix A. These tests were conducted following a procedure previously outlined by Biscarini et al. [29]. The EMG signals collected during the MVIC tests were utilized for offline normalization of the EMG signals recorded during the limb rise trials (refer to the subsequent section for specifics). After the MVIC tests and a recovery phase lasting more than 5 min, each participant executed the 24 isometric exercises resulting from the combination of eight limb rising combinations and three body positions previously outlined. To mitigate the impact of fatigue on muscle activity, the 24 exercises were completed in a randomized sequence, with a minimum 2 min rest between successive trials.

2.5. Data Recording and Processing

Surface EMG signals were bilaterally recorded from the ES, MF, EO, and IO muscles with wireless EMG sensors (FreeEMG 1000; BTS Bioengineering, Milano, Italy) and Ag/AgCl surface electrodes (Covidien Kendall, Minneapolis, MN, USA) placed on the muscle belly, parallel to the muscle fibers, 2 cm apart from each other. Preceding electrode placement, the skin was shaved, gently abraded via sandpaper, and cleaned with an alcoholic solution. Precise electrode locations for each muscle are reported in Appendix A. The raw EMG signals were differentially amplified (933 gain), bandpass filtered from 10 to 500 Hz, digitalized (16-bit resolution, 1 kHz sampling frequency), and transformed into amplitude envelopes through a point-to-point moving root mean square filter (500 ms time interval). The EMG amplitude envelopes recorded from each muscle during the trials were then normalized to the highest 1 s average of the EGM amplitude envelope obtained during the MVIC of the same muscle [30]. Finally, the mean value of each normalized EMG signal was computed for statistical comparison between the task conditions.

2.6. Statistical Analysis

ANOVA designs were used to compare data samples. Mauchly’s sphericity test proved the sphericity assumption, and probabilities were corrected based on Greenhouse–Geisser and Huynh–Feldt epsilon when appropriate. The normality and homogeneity of the variance between populations were assessed by the Shapiro–Wilk test and Levene’s test, respectively. When necessary to meet these ANOVA assumptions, data were transformed using the logarithm (ln) function. Descriptive statistics, presented as mean ± SD, consistently refer to the untransformed data, even if the analysis was performed on transformed data.

In each body position, the eight combinations of isometric limb rise exercises were grouped into four pairs, each pair comprising two specular exercises relative to the sagittal plane of the body. No significant statistical differences emerged between the two specular exercises of each pair. For example, muscle activation during the right arm rise did not significantly differ from the corresponding muscle activation on the opposite side of the body during the left arm rise. Consequently, data recorded from the two specular exercises within a pair were collapsed by averaging their values. As a result, only 4 limb rise combinations were finally considered: arm raise, leg raise, contralateral arm–leg rise, and ipsilateral arm–leg rise.

EMG data samples were analyzed separately for each muscle (MF, MF, EO, and IO) by a 3-way repeated-measures ANOVA with body position (three levels: prone, quadruped, and standing-prone), limb rise combination (four levels: arm raise, leg raise, contralateral arm–leg rise, and ipsilateral arm–leg rise), and side of the body (two levels) as independent within-subject factors. To enable a meaningful comparison, for each muscle and within each exercise condition, the body sides were classified as EMG-dominant (the side with higher EMG averaged activity across the 20 participants) and non-dominant. The difference in muscle activation between the EMG dominant and non-dominant sides was analyzed by a 3-way repeated-measures ANOVA with body position, limb rise combination, and muscle (ES, MF, EO, and IO) as independent within-subject factors. In cases of significant main effects or interactions, the observed power (ω) and partial eta squared (${\eta }_{\mathrm{p}}^2$) coefficients were used to assess the statistical power and effect size. Post hoc analysis was run via the Scheffè test. The significance level for all statistical tests was set at p < 0.05.

3. Results

The activity level of each muscle was significantly affected by body position, limb rise combination, and body side (ES: $p<{10}^{-3}$, $0.71\le {\eta }_p^2\le 0.81$, $\mathsf{\omega }>0.99$; MF: $p<{10}^{-3}$, $0.73\le {\eta }_p^2\le 077$, $\mathsf{\omega }>0.99$; EO: $p<{10}^{-3}$, $0.57\le {\eta }_p^2\le 0.84$, $\mathsf{\omega }>0.99$; IO: $p\le 0.002$, $0.25\le {\eta }_p^2\le 0.65$, $\mathsf{\omega }\ge 0.93$). The activity of the superficial muscles (ES and EO) was also significantly affected by all factor interactions (ES: $p<{10}^{-3}$, $0.33\le {\eta }_p^2\le 0.55$, $\mathsf{\omega }>0.99$; EO: $p<{10}^{-3}$, $0.40\le {\eta }_p^2\le 0.54$, $\mathsf{\omega }>0.99$).

The difference in activation between body sides showed significant dependence on body position, limb rise condition, and muscle ($p<{10}^{-3}$, $0.53\le {\eta }_p^2\le 0.70$, $\mathsf{\omega }>0.99$), and all their interactions ($p<{10}^{-3}$, $0.24\le {\eta }_p^2\le 0.46$, $\mathsf{\omega }\ge 0.97$).

The key findings from the post-hoc analysis are summarized in the following sections and illustrated in Figure 1, Figure 2, Figure 3 and Figure 4.

Table 1. Muscle activations (%MVIC) in the different body positions and limb rise combinations. Data from previous studies are also included for comparison, along with the vertebral level (ranging from T9 to L3) where the thoracic ES electrodes were positioned (bold data: body side with higher activation).

Unilateral leg raise (rLs = raised-leg side; cs = contralateral side)
body position	Reference	ES		MF		EO		IO
		rLs	cs	rLs	cs	rLs	cs	rLs	cs
prone	current study (T12-L1)	16.9	40.0	40.4	34.0	11.9	6.9	9.3	19.8
quadruped	current study (T12-L1)	9.2	20.1	34.2	23.0	19.6	9.4	12.2	22.2
	Callaghan (T9) [18]	5.1	14.3	22.1	11.3	8.7	5.1	11.6	12.1
	Drake (T9) [20]	4.5	19.9	23.3	15.8	12.4	5.7	19.3	22.5
	Kavcic (T9) [19]	4	18	15	8	29	14	29	28
	Stevens (L1) [21]	12	18	21	15	22	17	8	26
	Yoon (L1) [26]	22	20	32	16	23	21	15	18
standing-prone	current study	29.5	24.8	50.7	43.0	11.0	5.4	11.2	11.3
Arm raise (rAs = raised-arm side; cs = contralateral side)
body position	Reference	ES		MF		EO		IO
		rAs	cs	rAs	cs	rAs	cs	rAs	cs
prone	current study (T12-L1)	69.3	33.7	35.7	40.1	7.1	15.2	13.4	6.7
quadruped	current study (T12-L1)	53.9	11.5	11.8	23.3	10.2	32.3	22.7	10.1
	Yoon (L1) [26]	26	18	11	18	21	25	18	8
standing-prone	current study (T12-L1)	61.3	34.3	41.8	42.0	7.8	10.9	8.2	7.2
Contralateral arm–leg rise (rLs = raised-leg side; rAs = raised-arm side).
body position	reference	ES		MF		EO		IO
		rLs	rAs	rLs	rAs	rLs	rAs	rAs	rLs
prone	current study (T12-L1)	31.9	66.0	45.8	49.8	14.5	7.5	8.8	17.5
	Ekstrom (L1) [22]	36	49	40	45	-	-	-	-
	Kelly (L1-L2) [27]	82.5	104	86.5	88.5	-	-	-	-
quadruped	current study (T12-L1)	18.6	57.3	45.3	38.0	34.4	12.3	12.4	24.0
	Callaghan (T9) [18]	11.0	44.8	32.6	16.3	16.0	5.7	15.4	17.3
	Drake (T9) [20]	12.0	56.4	22.8	23.1	21.3	10.4	17.3	25.1
	Ekstrom (L1) [22]	36	35	41	29	-	-	-	-
	Imai (L3) [23]	16	22	32	22	36	33	-	-
	Kavcic (T9) [19]	7	48	25	14	38	16	29	31
	Kelly (L1-L2) [27]	26	37	44.5	24.5	-	-	-	-
	Masaki (L1) [25]	15.1	22.5	28.5	19.3	-	-	-	-
	Okubo (L3) [24]	17	22	30	19	39	32	-	-
	Stevens (L1) [21]	18	32	27	21	30	20	8	29
	Yoon (L1) [26]	26	41	43	25	28	24	14	20
standing-prone	current study (T12-L1)	36.8	54.4	60.1	55.4	17.1	8.2	11.4	12.3
Ipsilateral arm–leg rise (rALs = raised-arm-and-leg side; cs = contralateral side)
body position	Reference	ES		MF		EO		IO
		rALs	cs	rALs	cs	rALs	cs	rALs	cs
prone	current study (T12-L1)	66.7	48.0	52.5	63.6	8.7	10.9	11.0	10.5
quadruped	current study (T12-L1)	41.6	20.6	36.5	32.9	14.7	8.6	21.4	14.4
standing-prone	current study (T12-L1)	56.9	32.9	57.0	53.3	14.6	7.3	15.6	10.6

presents the muscle activation levels recorded in each of the 12 exercise conditions, along with values available in prior studies. The color-coded map in Figure 5 displays a simplified overview of the muscular activation distribution.

3.1. Erector Spinae

ES activity was significantly higher in the prone position ($p\le 0.001$) during exercises involving upper limb elevation (prone arm rise, contralateral arm–leg rise, and ipsilateral arm–leg rise) ($p<{10}^{-3}$) and at the side of the raised arm ($p<{10}^{-3}$). In these conditions, the ES reached activation levels of 66–69% MVIC (Figure 2a, Figure 3a, and Figure 4a). Isolated leg rise led to higher activation in the contralateral ES in prone ($p<{10}^{-3}$) and quadruped ($p<{10}^{-3}$) positions, and in the ipsilateral ES in the standing-prone position ($p<{10}^{-3}$).

The bilateral difference in ES activation was significantly higher during arm rise and contralateral arm–leg rise ($p<{10}^{-3}$) and significantly lower in the standing-prone position ($p<{10}^{-3}$), reaching maximum disparity (42% MVIC) during quadruped arm rise (Figure 2e). Overall, the difference in ES activation between the sides of the body was higher than that of the other muscles ($p<{10}^{-3}$).

3.2. Multifidus

The MF activity was significantly higher in the standing-prone position ($p\le 0.016$) and during contralateral/ipsilateral arm–leg rise ($p<{10}^{-3}$). It peaked at 57–64% MVIC during prone ipsilateral arm–leg rise and standing ipsilateral/contralateral arm–leg rise (Figure 3c and Figure 4a,c). Isolated leg rise significantly heightened ipsilateral MF activity across all body positions ($p<{10}^{-3}$), while isolated arm rise significantly heightened contralateral MF activity in prone ($p=0.001$) and quadruped ($p<{10}^{-3}$) positions. During contralateral arm–leg rise in the quadruped position, MF activity was significantly higher on the side of the raised leg ($p<{10}^{-3}$), whereas during ipsilateral/contralateral arm–leg rise in the prone position, it was significantly higher on the side contralateral to the raised leg ($p=0.001$).

The bilateral difference in MF activation was significantly lower in standing-prone than in the quadruped position ($p=0.004$). It peaked at approximately 11% MVIC during leg rise and arm rise from the quadruped position, as well as during prone ipsilateral arm–leg rise (Figure 1e, Figure 2e, and Figure 4d).

3.3. Internal and External Obliques

Arm raise and contralateral arm–leg rise in the quadruped position exhibited the highest levels of activity of EO (32–34% MVIC) and IO (23–24% MVIC), along with a significantly higher bilateral difference in EO activation (22% MVIC) compared to all other exercise conditions ($p<{10}^{-3}$) (Figure 2b,e and Figure 3b,e). Both EO activation and its bilateral activation difference were significantly higher in the quadruped position ($p<{10}^{-3}$) and during arm rise and contralateral arm–leg rise ($p\le 0.002$). The bilateral difference in IO activation reached a maximum value of 13% MVIC, was significantly lower in the standing-prone position ($p<{10}^{-3}$), and was not significantly influenced by the limb rise combination. Overall, the EO activation difference between the sides of the body was higher than those of the IO ($p=0.039$) and MF ($p=0.015$), which were not significantly different from each other.

4. Discussion

This study analyzes the bilateral activity of spine extensors and rotators during 12 distinct lumbar stabilization exercises resulting from 4 asymmetric limb elevation combinations (leg raise, arm raise, contralateral arm–leg rise, ipsilateral arm–leg rise) performed in 3 different body positions (prone, quadruped, and standing-prone position). This comprehensive dataset can provide deeper insights into the stabilizing function of the lumbar and oblique muscles during the execution of asymmetric tasks.

The oblique muscles displayed characteristic activation patterns modulated by the different exercise conditions. In the prone position, a leg rise that brings the hip close to full hyperextension is naturally accompanied by a slight elevation from the ground of the ipsilateral pelvic side, indicating a small transversal pelvic rotation to the same side. With the chest in contact with the ground, this pelvic rotation triggers a compensatory spine rotation in the opposite direction. Maintaining this pelvic position against gravity requires an isometric effort involving closed-kinetic-chain spine rotation toward the side opposite the lifted leg. This is accomplished by pressing against the ground with the side of the chest ipsilateral to the lifted leg (the position cannot be sustained with the assistance of the upper limbs, which must remain relaxed in line with the body at the sides of the head throughout the entire exercise). The EO ipsilateral and the IO contralateral to the lifted leg generate the stabilizing effort. This explains the higher activation of these two muscles than the contralateral ones (Figure 1a). An arm rise from the prone position results in the opposite effect: a slight spine rotation towards the side of the lifted arm driven by the activation of the EO contralateral and the IO ipsilateral to the lifted arm, which exhibit higher activation compared to the contralateral muscles (Figure 2a). Hence, the elevation of a lower limb and the simultaneous elevation of the contralateral upper limb yield similar effects: a higher activation of the EO ipsilateral to the lifted leg and the IO ipsilateral to the lifted arm (Figure 3a). Conversely, the elevation of a lower limb and the simultaneous elevation of the ipsilateral upper limb yield contrasting effects on the oblique muscles, which display no significant difference in activation between the two sides of the body (Figure 4a).

Similar patterns of oblique muscle activations are obtained in the quadruped position (Figure 1b, Figure 2b and Figure 3b), except for the ipsilateral arm–leg rise (Figure 4b). In this condition, the body zones in contact with the ground (ipsilateral forefoot, knee, and hand) are nearly aligned, posing a challenge to body equilibrium. To maintain balance, the body’s center of mass should be shifted laterally towards the supporting side, reaching a position above the base of support. This was naturally achieved by participants through two rigid trunk displacements driven by closed-kinetic-chain (CKC) movements of the hip and shoulder of the supporting limbs: (1) a slight lateral trunk translation towards the supporting side, induced by transverse-plane CKC hip adduction and shoulder flexion; and (2) a slight trunk rotation that rises the non-supporting side of the trunk upwards and towards the supporting side, induced by transverse-plane CKC hip abduction and shoulder extension. These rigid trunk displacements maintain the spine in a neutral position; however, the rotated trunk and the weight of the raised limbs expose the spine to gravity-induced lateral flexion toward the supporting side. Counteracting this movement entails activating the ipsilateral external and internal oblique muscles on the non-supporting side of the body. The results confirm higher activation of these two muscles than the contralateral ones (Figure 4b).

In the standing-prone position, the leg rise reduces the base of support to the contralateral foot plant, leading to two effects on the oblique muscles: (1) a movement-facilitating effort of spine rotation towards the side of the raised leg, with greater activation of the EO ipsilateral and the IO contralateral to the supporting leg; (2) a rigid trunk rotation and lateral translation that bring the body center of mass above the base of support, with higher activation of the oblique muscles ipsilateral to the raised leg (as in the case of the ipsilateral arm–leg rise from the quadruped position). The result is a heightened activation of the OE ipsilateral to the raised leg compared to the contralateral OE, with a minimal difference in the activation of the IO muscles (Figure 1c). In contrast, the arm rise does not impact the base of support, resulting in minimal imbalance and a negligible difference in the activation of contralateral oblique muscles (Figure 2c). Consequently, the leg mainly determines the effect of contralateral arm–leg rise on the oblique muscles (Figure 3c). In the ipsilateral arm–leg rise, only the second effect influences the difference in oblique muscle activation between the sides of the body, as the arm rise and leg rise tend to induce opposite spine rotations that counterbalance each other, maintaining the spine in a neutral position in the transverse plane. This led to increased activation of the oblique muscles ipsilateral to the raised limbs (Figure 4c).

The ES consistently exhibited higher activation on the side of the raised arm. This is likely due to the posterior scapular tilt and slight spine extension linked with full shoulder flexion. Isolated leg rise resulted in higher activation in the contralateral ES in prone and quadruped positions, to compensate for the dominant activity of the ipsilateral MF (Figure 1a,b). Conversely, standing leg rise led to higher activation of the ES ipsilateral to the raised leg (Figure 1c). In the standing-prone position and single leg support, the gravity-induced lateral spine flexion towards the side of the supporting leg is resisted not only by the EO and IO on the side of the raised leg (as previously discussed), but also by the ES and MF of the same side.

In all body positions, leg rise enhanced the ipsilateral MF activity due to the posterior pelvic tilt and slight spine extension associated with full hip extension. Conversely, arm rise enhanced the contralateral MF activity to compensate for the opposite dominant behavior of the ES. The MF activity during contralateral or ipsilateral arm–leg raise was mainly influenced by the effect of the raised leg in the quadruped and standing-prone positions. However, during the prone contralateral/ipsilateral arm–leg raise, the MF activity was higher on the side contralateral to the raised leg. This unexpected behavior observed in the prone position might stem from the participant’s attempt to maintain ground contact on the anterior pelvic side opposite the raised leg (counteracting the effect of gravity), with the aim of maximizing the base of support and stabilizing the body.

Previous studies examined the bilateral activity of ES, MF, EO, and IO during only 4 of the 12 exercise conditions investigated in the current study: contralateral arm–leg rise in the quadruped [18,19,20,21,22,23,24,25,26,27] and prone [22,27] positions, and isolated lower limb [18,19,20,21,26] or upper limb [26] rise from the quadruped position (Table 1). Overall, these prior findings are in good agreement with our results, particularly regarding the differences in muscle activation between the body’s sides. The 12 exercise conditions analyzed in this study generated a wide spectrum of muscle activations, reaching up to 69%, 64%, 34%, and 24% MVIC for ES, MF, OE, and OI, respectively. This extensive dataset of muscle activity levels can aid in optimizing the planning and progress of asymmetric lumbar stabilization exercises. Notably, ES, MF, and oblique muscles exhibited significantly higher activity in the prone, standing-prone, and quadruped positions, respectively, and reached maximum activation with specific limb rise combinations. Hence, employing a mix of diverse exercises could be a successful approach for building strength in the lumbar and oblique muscles through asymmetric stabilization exercises.

Despite the extensive EMG dataset presented in this study, further research is needed to achieve a complete understanding of asymmetric lumbar stabilization exercises. For example, assessing the mechanical loading acting on specific joint structures during various exercise conditions is crucial for rehabilitative interventions [18,19,31]. Additionally, exploring the activation timing of lumbar and oblique muscles before and during limb lifting could provide further insights into the motor control strategies used in different body positions and limb rise combinations [32,33,34,35].

The main limitation of this study stems from the relatively small sample size (20 participants) and the specific inclusion criteria. This study involves exercises that require a certain level of expertise in bodyweight resistance training. For this reason, participants with an intermediate-to-advanced fitness level were selectively recruited from local fitness centers. This may limit the generalizability of the findings to a broader population. Furthermore, the long-term effects or sustainability of the exercises over time were not investigated. This could be crucial for understanding the practicality of incorporating these exercises into long-term training or rehabilitation programs.

5. Conclusions

This study analyzed the bilateral activity of spinal extensors and rotators during asymmetric lumbar stabilization exercises executed in prone, quadruped, and standing-prone positions, raising an upper and/or lower limb in all possible combinations. The findings uncovered different strategies used to stabilize the body in reaction to the changes in the base of support and the position of the body’s center of mass induced by the different exercise conditions. This information can provide deeper insights into the stabilizing function of the lumbar and oblique muscles during the execution of asymmetric tasks. Moreover, the gradation of the levels of muscle activation across different body positions and limb elevation combinations could assist in designing optimal progressions for lumbar stabilization exercises in athletic programs and rehabilitation interventions, avoiding the use of external overloads and the corresponding mechanical loadings exerted on the spine structures.