Biomechanical analysis of lumbar oblique manipulation for lumbar disc herniation with different protrusion subtypes

Abstract

Lumbar oblique manipulation (LOM) is one of the commonly used conservative treatments for lumbar disc herniation (LDH). Compared with surgical interventions, its non - invasive nature provides patients with greater comfort and reduces treatment resistance, resulting in its widespread clinical application. However, manual techniques may vary according to the type of disc herniation and could potentially lead to adverse outcomes. For which LDH patients is LOM truly suitable? This question deserves further consideration. To investigate the specific characteristics of LOM for different LDH types, this study constructed six representative finite element models of LDH and analyzed the biomechanical effects of LOM on each model. The findings suggest that for patients with herniations at the same level as the affected nerve root, LOM is safer when performed in the healthy lateral position, regardless of the presence of significant compression symptoms. Conversely, for patients with pronounced neurological compression symptoms, especially those with disc protrusions extending beyond the nerve root level, the technique may worsen disc and nerve root compression. This study provides the first bilateral LOM biomechanical evidence across six LDH subtypes, which can be used as a guide for safer clinical positioning.

Graphical abstract

Biomechanical analysis of Lumbar Oblique Manipulation for Lumbar Disc Herniation.

1. Introduction

Lumbar disc herniation (LDH) is a prevalent degenerative spinal disease that predominantly affects young and middle-aged individuals with a relatively high incidence rate at the L4/L5 and L5/S1 levels.^1,2 The annual incidence of LDH ranges from 0.5% to 1%, which indicates that substantial pain and financial strain are imposed on patients.³ Current LDH treatments are primarily divided into surgical and nonsurgical options with 80% to 90% of patients experiencing improved outcomes through nonsurgical interventions4, 5, 6. Trager RJ⁴ discovered, through a retrospective case series, that adult recipients of traditional treatment exhibited for newly diagnosed LDH or LSR would have reduced odds of lumbar discectomy over 1-year and 2-year follow-up compared with those receiving other care. Mo Z et al.⁵ demonstrated through a randomized controlled trial that, oblique pulling spinal manipulation had a favorable effect in alleviating pain, and modified oblique pulling manipulation had significant superiority in improving lumbar function. These studies suggest that the technique may involve reducing local muscle tension and altering small joint misalignment to stabilize the lumbar spine from the inside and outside, thereby exerting a positive effect on LDH.

Lumbar oblique manipulation (LOM) is a commonly used conservative technique for LDH and is characterized by practicality, high patient comfort, and minimal environmental requirements.⁷ During this maneuver, the patient lies in a lateral position with the upper lower limb flexed at the hip and knee, while the lower limb remains extended. At the same time, the upper body is slowly rotated, turning the chest toward the back. The surgeon places one hand on the posterior lateral aspect of the patient's upper iliac crest, applying moderate pressure to stabilize the area. The other hand supports the patient's upper shoulder or scapular region. When the practitioner feels significant tension in the lumbar soft tissues and increased resistance, applying force with both hands in an alternating manner completes the maneuver. LOM has been proposed to correct abnormal spatial positioning of the lumbar spine and restore its stability and offers significant clinical advantages over other conservative methods. However, owing to the diverse clinical presentations of LDH, LOM poses potential safety risk.^8,9 For example, inadequate control of torsion angles or applied forces during manipulation can exacerbate symptoms and result in severe complications including cauda equina syndrome, epidural haematoma, or vertebral fractures.¹⁰ Despite existing research on the biomechanical effects of spinal manipulation for specific LDH types using finite element analysis,^7,11,12 studies on the biomechanical impacts of manipulation techniques across different LDH types are lacking. For example, many studies have focused solely on unilateral diagonal manipulation, which differs from bilateral diagonal manipulation that is commonly applied in clinical practice.

This study provides the first bilateral LOM biomechanical evidence across six LDH subtypes, which can be used as a guide for safer clinical positioning.

2. Materials and methods

2.1. Establishment of normal lumbar spine FEM

The construction and simulation of the LDH model was achieved using three-dimensional finite element simulation technology. Previous research experiences indicate that the construction of the LDH model requires a normal spinal model as a foundation. To that end, a healthy adult male (28 years old, 176 cm tall, and 60 kg weight) was enrolled and had no history of spinal degeneration, fractures, tuberculosis, or severe spinal deformities. Informed consent was obtained from the patient. Afterwards, L1–S5 vertebral bodies were scanned using a GE Light Speed VCT 64-slice spiral CT scanner at 140 kV and 200 mA at a 0.625 mm layer thickness to obtain DICOM tomographic images. Clinical Trail Number: Not applicable. Ethics approval was obtained from the Shanghai University of Traditional Chinese Medicine Ethics Committee (approval number: 2024-1515-098).

DICOM data were imported into Mimics 21.0 (Materialise Inc., Leuven, Belgium) for three-dimensional cutting and Boolean operations to create a 3D lumbar spine model. The processed model was saved in STL format and imported into Geomagic Studio 2012 (Geomagic Inc., NC, USA) where it was refined using “forming entity,” “smoothening surfaces,” “spheroidizing,” and “removing nails” functions. Cortical and cancellous bones were differentiated using the Offset function. The Intervertebral discs (IVD), articular cartilages, spinal cord, and nerve roots were modelled using SolidWorks 2023 (SolidWorks Corp., Waltham, MA, USA). Specifically, each IVD consists of upper and lower endplates, an annulus fibrosus, and a nucleus pulposus.

The spinal cord comprises grey matter, white matter, pia mater, cerebrospinal fluid, and dura mater, which are attached to the proximal nerve root. The proportions of each structure were adjusted on the basis of previous studies and established as follows: the nucleus pulposus was set to 43% of the disc volume, the endplates were set to 1 mm thick, the dura mater was set to 0.4 mm, and the pia mater was set to 0.1 mm^11,13, 14, 15. In ANSYS 2021 R2 Workbench (ANSYS, Canonsburg, PA, USA), six intervertebral ligaments were modelled according to the following real human anatomy: the anterior longitudinal ligament (ALL), posterior longitudinal ligament (PLL), interspinous ligament (ISL), supraspinous ligament (SSL), intertransverse ligament (IL), and ligamentum flavum (LF). Stiffness values for these ligaments were assigned on the basis of prior finite element analysis.¹⁶ Some of the material parameter details are shown in Table 1.

Table 1.

Material properties of lumbar spine structures in the finite element model.

Component	Element type	E/MPa	v	Stiffness (N/mm)
Cortical bone	Tetra element	12,000	0.3	—
Cancellous bone	Tetra element	100	0.2	—
Normal nucleus pulposus	Hex element	1.0	0.49	—
Moderately degenerative nucleus pulposus	Hex element	2.0	0.49	—
Normal annulus fibrosus	Hex element	4.2	0.45	—
Moderately degenerative annulus fibrosus	Hex element	2.5	0.40	—
Endplate	Hex element	25	0.25	—
Articular cartilage	Hex element	24	0.3	—
SSL	Spring	—	—	15.38
ISL	Spring	—	—	0.19
ALL	Spring	—	—	8.74
PLL	Spring	—	—	5.83
IL	Spring	—	—	2.39
LF	Spring	—	—	15.75

2.2. Establishment of six classical LDH models

The L4-5 IVD was modified to represent different types of LDH. In general, the central type and paracentral/lateral recess type are more common in clinical practice, accounting for approximately 61% of all types.¹⁷ The central type mainly presents as the IVD protruding towards the posterior direction, compressing the dura mater, The lateral recess type mainly shows the IVD protruding towards the lateral recess, causing the lateral recess to become narrow. However, the type that causes more severe clinical symptoms is the intervertebral foramen type, characterized by the IVD protruding towards the intervertebral foramen position, resulting in the blockage of the intervertebral foramen and compression of the nerve roots. The extreme lateral type is a relatively rare variant, it's protrusion is located outside the intervertebral foramen and compresses the nerve roots that have already exited the foramen. Although this type is relatively uncommon, due to its unique anatomical location, it is prone to being missed during diagnosis, and the resulting nerve root pain symptoms are typical.^18,19 However, the movement of the protrusion is not just limited to the horizontal plane; it can also occur in the vertical plane. A retrospective study showed that among 1289 LDH patients, 4.57% of them presented the protrusion with migration above the nerve roots (shoulder type), 50.73% showed movement downward towards the nerve roots (axillary type).²⁰ The phenomenon that the protrusions extend freely to either the upper or lower sides of the nerve roots is a common but potentially risky pathological condition that may significantly affects clinical treatment options and prognosis assessment. Therefore, it requires special attention. Based on the most common types of protrusions encountered in clinical practice, the following six classical LDH types were constructed using clinical MRI reference images: central (M1), paracentral/lateral recess (M2), foraminal (M3), extreme lateral (M4), shoulder (M2a), and axillary (M2b). Because the paracentral (M2) pattern is clinically representative, M2a and M2b were constructed by shifting the sagittal protrusion position of the M2 herniation superiorly and inferiorly, respectively. In addition to copying the prominent shape, we adjusted the material parameters of the IVDs in the LDH herniation based on previous modeling experience, making them in a moderately degenerated state. Because in clinical practice, the protrusion of the intervertebral disc is often accompanied by a certain degree of degeneration.²¹ (Fig. 1).

Fig. 1.

Six classical types of LDH models. (M1) Central type; (M2) Lateral recess type; (M3) Intervertebral foramen type; (M4) Extreme lateral type; (M2a) shoulder type; (M2b) axillary type.

2.3. Model validation

The complete normal lumbar finite element model (FEM) comprises 167,777 elements and 337,239 nodes. To validate the model, the range of motion (ROM) of the model was compared with that of previous in vivo and in vitro studies22, 23, 24, 25. We constrained the ischium to six degrees of freedom, and a 500 N vertical downward force was applied to the L1 upper surface to simulate the body's self-weight in a standing position, whereas a 10 Nm moment was applied to the same surface to simulate lumbar flexion, extension, and rotation.^21,26 The validity of the model was confirmed if the degree of lumbar segment mobility was within the reference range, and ultimately, results confirmed the validity of the model (Fig. 2). Furthermore, the six LDH models accurately replicated the parameters and structures of the lumbar spine and ensured the natural nature and reliability of the models.

Fig. 2.

Comparison of range of motion of each segment under static load in normal lumbar spine model with the data from previous in vivo and in vitro studies. (A) Flexion and extension; (B) Left and right lateral bending; (c) Left and right axial rotation.

2.4. Lumbar oblique manipulation (LOM) simulation

In this study, LOM was simulated using manual mechanical parameters based on previous in vivo measurement studies.⁵ According to past experience, the weight borne by the spine when lying on its side is about 40% of the total body weight,²⁷ so a vertical downward pressure equivalent to 40% of the body weight (about 240 N) was applied to the L1 upper surface to simulate the gravitational force of the lumbar spine in the lateral recumbent position. And during the manipulation process, the patient's upper body is in a rotating position. In the finite element analysis, it is found that approximately 10Nm of torque needs to be applied for the lumbar spine to reach the extreme state of rotation, so a left/right rotational moment of 10Nm was applied to simulate lumbar rotation condition. Concurrently, During the actual operation, the force direction on the iliac region is horizontally forward, but due to limitations of the experimental conditions, the results of the previous study were used as a reference. Specifically, a 580 N horizontal force was added at the iliac crest to simulate the pushing and pulling force.²⁸ (Fig. 3).

Fig. 3.

Schematic diagram of manual force direction. (A) Manipulation in healthy lateral position; (B) Manipulation in affected lateral position.

2.5. Main evaluation parameters

In this study, the following parameters were evaluated: IVD stress, the stress differential between the annulus fibrosus and nucleus pulposus, the bilateral articular cartilage stress ratio, and the sagittal spacing between the L4-L5 IVD and the nerve roots. This study selected these four indicators because they are closely related to the core of the research on the treatment of LDH by the LOM method. They can quantitatively assess the effect and safety of the manipulation from multiple dimensions: The IVD stress reflects the risk of direct injury to the IVD. The stress differential between the annulus fibrosus and nucleus pulposus indicates the therapeutic potential for nucleus pulposus repositioning. The ratio of bilateral joint cartilage stress assesses the effect of adjusting the stability of the spinal segment, and the sagittal spacing between the IVD and the nerve root directly correlates with the core clinical symptom of nerve root compression in LDH. These four indicators form a complete assessment system, providing a scientific basis for clinical operations.

3. Results

3.1. Stress of the L4/5 IVD before and after LOM

Before LOM, the von Mises stress values of the IVDs for M1–M2b were 0.29 MPa, 0.29 MPa, 0.30 MPa, 0.30 MPa, 0.30 MPa and 0.30 MPa, and indicated a similar distribution of stresses across models. After the LOM in the healthy lateral position, stress values increased to 0.39 MPa, 0.40 MPa, 0.39 MPa, 0.40 MPa, 0.59 MPa and 0.58 MPa. In the affected lateral position, the stress values increased further and reached 0.43 MPa, 0.44 MPa, 0.43 MPa, 0.44 MPa, 0.41 MPa and 0.47 MPa. Notably, the IVD stress increased significantly with LOM at both positions. In the case of M1–M4, the increase in IVD stress is greater under the affected lateral position than under the healthy lateral position. In contrast, for M2a and M2b, the stress is greater under the healthy lateral position than under the affected position (Fig. 4).

Fig. 4.

IVD stress: Pre- and post-manipulation.

3.2. Difference in stress between the fibrous annulus and nucleus pulposus

Before LOM, the difference in stress between the annulus fibrosus and nucleus pulposus was 0.08 MPa, 0.07 MPa, 0.10 MPa, 0.09 MPa, 0.07 MP and 0.08 MPa for M1–M2b. After LOM in the healthy lateral position, values notably increased to 0.08 MPa, 0.10 MPa, 0.12 MPa, 0.11 MPa, 0.14 MPa, 0.11 MP and 0.12 MPa. After LOM in the affected lateral position, values were 0.08 MPa, 0.11 MPa, 0.15 MPa, 0.14 MPa, 0.11 MP and 0.12 MPa. The stress differential in M1 remained unchanged, whereas the other five types exhibited notable increases. For M2–M4, the increase was significantly more pronounced under the affected position, whereas M2a and M2b did not significantly differ between the healthy and affected lateral positions (Fig. 5).

Fig. 5.

Stress differential between the fibrous annulus and nucleus pulposus.

3.3. Stress in the bilateral articular cartilage

Before LOM, the bilateral facet cartilage stress ratios were 1:1.12, 1:1.17, 1:1.17, 1:1.17, 1:1.16 and 1:1.16 for M1–M2b. After LOM in the healthy lateral position, ratios notably shifted and reached 1:1.01, 1:1.01, 1:1.02, 1:1.01, 1:1.13 and 1:1.13. After LOM in the affected lateral position, ratios were adjusted to 1:1.06, 1:1.06, 1:1.06, 1:1.06, 1:1.02 and 1:1.03. Under both manipulation positions, the stress ratio approached 1, which indicated a more balanced distribution across the articular cartilage. Overall, M1–M4 showed greater balance under the healthy lateral position, whereas M2a and M2b were more balanced under the affected lateral position (Fig. 6).

Fig. 6.

Stress in bilateral articular cartilage during the manipulation.

3.4. Sagittal displacement of nerve roots in the IVD under LOM

Under the healthy lateral position, IVD moved anteriorly with displacements of 2.55 mm, 2.48 mm, 3.54 mm, 2.44 mm, 3.40 mm and 3.39 mm. The nerve roots in M1–M4 displaced posteriorly by 0.55 mm, whereas those in M2a and M2b displaced anteriorly by 4.95 mm and 4.94 mm, respectively. Consequently, the sagittal space between the nerve roots and IVD increased in M1–M4 and decreased in M2a and M2b. In the affected lateral position, both the IVD and nerve roots shifted anteriorly. IVD anterior displacements were 3.34 mm, 3.48 mm, 3.45 mm, 3.45 mm, 2.54 mm and 3.26 mm. Nerve root displacements were 3.76 mm, 3.76 mm, 3.75 mm, 3.75 mm, 4.09 mm and 4.26 mm. Because nerve root displacement exceeded IVD displacement, the sagittal gap between the two structures decreased in all models (Fig. 7).

Fig. 7.

Sagittal displacement of nerve roots in the IVD under LOM.

4. Discussion

LDH is a common musculoskeletal disorder in clinical settings that is characterized primarily by IVD degeneration and mechanical issues in affected spinal segments and is often accompanied by abnormal stress concentrations within the IVD29, 30, 31. However, these biomechanical alterations create a key opportunity for manual therapeutic interventions. Additionally, abnormal stress and positioning of the facet joints contribute significantly to LDH pathogenesis32, 33, 34. Traditional Chinese manipulation techniques are believed to alleviate clinical symptoms by minimizing adhesions, normalizing stress, and realigning spinal force transmission35, 36, 37. LOM is widely used in LDH treatment and has demonstrated considerable therapeutic benefits. However, owing to the variability in LDH presentations, applying LOM in different positions may produce unintended biomechanical effects, which underscores the need to further evaluate its safety. This study employed finite element simulation to analyse the biomechanical effects of LOM across six distinct LDH types and evaluated the safety of bilateral LOM.

We found that LOM resulted in a significant increase in IVD stress across all LDH models. With the exception of the central type, the stress differential between the annulus fibrosus and the nucleus pulposus increased in the other five types. This differential loading helps maintain negative pressure within the intervertebral discs and promotes the absorption of the protruding nucleus pulposus.³⁸ In M2–M4, both the magnitude of IVD stress and the annulus-nucleus stress differential were greater under the affected-side position than under the healthy lateral position. This may reflect altered biomechanical properties of the IVDs, such that normal IVDs are well hydrated, have good height and elasticity, can mostly bear the load of the spine, and can transfer the load through a relatively large displacement within a reasonable range.³⁹ When IVDs degenerate and protrude, their height is lost and their elasticity decreases, which results in a reduction in displacement and an increase in stress under force.⁴⁰ When lying on the affected side, the disc's center of loading shifts towards the protrusion and increases the local compressive stress. Furthermore, LOM amplifies this stress during the application of force. Similarly, the difference in stress between the annulus fibrosus and the nucleus pulposus in the affected lateral position is also amplified. This stress distribution may contribute to the restoration of favorable intradiscal pressure gradients that support nucleus pulposus retraction.³⁸ This also suggests that for LDH in which the protrusion does not exceed the level of the nerve root, affected-side LOM may offer greater biomechanical correction, although potentially at the cost of reducing safety. Although M2a and M2b produced the greatest IVD stress, they did not produce the maximum difference in annulus fibrosus/nucleus pulposus stress. These patterns suggest that M2a and M2b may respond more sensitively to LOM but may also carry elevated safety risks, particularly in the healthy-side position.

Notably, the biomechanical response of the central-type model is different from that of the other types. Although IVD stress increased with increasing LOM, the annulus-nucleus stress differential showed little to no change. This may reflect the distinct anatomical and pathological characteristics of central-type LDH. Because the protrusion lies centrally, manual forces may not effectively alter local stress transmission between the nucleus pulposus and annulus fibrosus compared with lateralized types. Therefore, from a safety standpoint, LOM should be applied with greater caution in central-type LDH. Overall, excessive stress may exacerbate this condition and potentially lead to further protrusion of the nucleus pulposus. This could increase the likelihood of neural tissue compression and trigger more severe neurological symptoms.

LOM produced a more balanced bilateral facet joint cartilage stress distribution across all models, including the central type. In M1–M4, this positive regulatory effect was more pronounced in the healthy lateral position. In contrast, M2a and M2b showed a more balanced facet loading under the affected-side position. These findings suggest that LOM can improve internal spinal stress distribution, reduce local stress concentrations, and correct facet joint asymmetry, which thereby enhances segmental stability. For central-type LDH, the therapeutic effect of manual techniques may represent the primary beneficial mechanism.

Additionally, the sagittal gap between the IVD and nerve root significantly changed after manipulation. In M1–M4 and under the healthy-side lateral position, LOM consistently increased this gap, which may be beneficial for reducing nerve root compression. Conversely, when positioned on the affected side, the gap narrowed. This pattern likely reflects the interaction between localized stress concentrations at the protrusion site and the direction of applied manual forces. Under the affected-side position, the IVD experiences lateral and posterior stress and leads to reduced anterior displacement. Additionally, the manipulative force acts horizontally backwards on the left iliac crest and above the midline, while the compressed nerve root is located below it. This generates anterior displacement of the compressed nerve root and thereby narrows the sagittal gap. These observations underscore the relative safety of the unaffected lateral position and emphasize the need for controlled force application in the affected position to prevent worsening nerve compression symptoms. In M2a and M2b, the sagittal space decreased under both positions, which suggests an increased risk of nerve root compression during LOM. This may have occurred because the IVD tissue had already detached from its original position and its stability was relatively poor.

In general, applying forces to a model that extend beyond the level of the nervous system is more dangerous. Biomechanically, the IVD tissue that exceeds the level of nerve roots no longer follows the normal stress distribution pattern of the IVD, and the effect of manual force also becomes less predictable. Regardless of the manipulation position, the applied forces tend to shift these protrusions towards the nerve roots. When manual techniques are applied, compressive or traction forces exerted on this area are more readily transmitted directly to the dural sac or cauda equina, increasing the risk of nerve tissue compression. Simultaneously, such herniations may have already invaded surrounding neural structures to some degree, further narrowing the already constricted spinal canal. Consequently, even minor displacement during manual manipulation may exacerbate symptoms of nerve compression. Furthermore, a protrusion extending beyond the nerve root level may indicate more severe disc degeneration or injury. The annulus fibrosus may have sustained significant damage, compromising its integrity and stability. Under manual manipulation, this weakened disc is more susceptible to further displacement or fragmentation of the protruding material.

These findings suggest that for LDH patients with protrusions exceeding the level of nerve roots, clinicians should exercise greater caution in selecting therapeutic approaches and consider individual factors, including protrusion morphology, severity, and patient condition, before selecting LOM.

This study has several limitations. First, finite element models were based on a single healthy adult male spine, which does not account for variability in age, sex, or differing degrees of disc degeneration, and the simulation simplifies the manipulation to a deterministic load. In actual clinical practice, there is individual variability in the force, velocity, and frequency of the applied manipulation, which may make it difficult to accurately predict the treatment outcome. Second, dynamic muscle forces, neuromuscular control, and physiological loading variations were not incorporated into the simulation, which potentially underrepresents the complexity of in vivo biomechanics. Third, long-term tissue remodelling and adaptive responses to manipulation were not modelled, which limits the ability to predict chronic or cumulative effects.

Future studies should incorporate patient-specific anatomical geometries, a broader spectrum of demographic and pathological variations, and dynamic loading conditions. Furthermore, stochastic factors, including variability in force application, velocity fluctuation models, and probability distributions of operation frequency should be integrated to evaluate their influence on treatment efficacy. Additionally, in vivo validation is needed to confirm the accuracy and clinical relevance of the simulated biomechanical responses.

5. Conclusion

For LDH patients with protrusions at the nerve root level, the LOM technique under the healthy lateral position appears to offer greater safety, whereas it may offer more direct therapeutic effects under the affected lateral position, albeit with an increased risk profile. With respect to LDH types in which protrusions exceed the level of nerve roots, LOM did not demonstrate biomechanical advantages in terms of reducing IVD stress or nerve root displacement. Instead, manipulation may actually exacerbate this condition. Therefore, the clinical application of LOM should prioritize a careful assessment of nerve root compression, joint space position, and overall disc stress patterns. In summary, For patients with protrusions at the nerve level, LOM can be prioritized in the healthy lateral position. When performing LOM on the affected lateral position, the force can be appropriately reduced to ensure the safety of the treatment. For patients with protrusions exceeding the level of the nerve roots, when there are no obvious symptoms of nerve root compression, It is appropriate to choose to perform the LOM operation in the affected lateral position to restore the stability of both joints. Additionally, the amount of force used during the procedure should be carefully controlled to avoid IVD injury. If the patient has obvious symptoms of nerve root compression, it is more appropriate to choose other gentle therapeutic methods, such as massage, heat therapy, and electrotherapy.

CRediT authorship contribution statement

Linling Zhang: Writing – original draft. Zhen Deng: Writing – original draft. Xuanzong Zhang: Methodology. Kuan Wang: Software. Maohua Lin: Data curation. Zhongxiang Yu: Data curation. Hongsheng Zhan: Data curation. Yongfang Zhao: Data curation. Frank Vrionis: Data curation. Huihao Wang: Writing – review & editing, Methodology.

Ethics approval

The study was approved by the ethics committee of Shanghai University of Traditional Chinese Medicine Ethics Committee (approval number: 2024-1515-098).

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Huihao Wang reports administrative support was provided by National Natural Science Foundation of China. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The data used to support the findings of this study are included within the article.