Physical Characteristics Associated With Low Back Pain in Japanese Collegiate Female Rhythmic Gymnasts

Abstract

Background:

Few studies have explored relationships between low back pain (LBP) and physical characteristics (physique, muscle strength, tightness, and flexibility) in female rhythmic gymnasts (RGs).

Hypothesis:

Due to their extreme flexibility, modifiable physical factors for LBP in RGs are not related to muscle tightness.

Study Design:

Cross-sectional study.

Level of Evidence:

Level 4.

Methods:

College female RGs (n = 95) were categorized into LBP and non-LBP groups based on questionnaires and orthopaedic surgeon interview. Physical assessments included whole-body bone mineral content and density, trunk lean body mass, and scoliosis presence by dual-energy X-ray absorptiometry scans, presence of abnormal findings and spine alignment by magnetic resonance imaging scans, hip muscle strength testing (flexion, extension, and abduction), range of motion (ROM), and flexibility testing. LBP and non-LBP groups were compared, and multivariate regression analysis performed.

Results:

RGs with LBP exhibited significantly longer practice time, lower hip flexion muscle strength on the nondominant side versus dominant side, lower active straight leg raise on the nondominant side versus dominant side, and lower hip external rotation (ER) ROM on the nondominant side versus dominant side. A history of LBP was also associated with current LBP. Multivariate logistic regression analysis revealed that a history of LBP (odds ratio [OR], 6.33; 95% CI, 1.56-25.62), differences in hip flexion strength (OR, 0.96; 95% CI, 1.00-1.08), and differences in hip ER ROM (OR, 1.12; 95% CI, 1.03-1.19) were factors associated with LBP.

Conclusion:

LBP was experienced by 27.3% RGs and associated not only with a history of LBP and a deficit in hip flexor strength on the nondominant side but also a deficit in hip ER ROM on the nondominant side.

Clinical Relevance:

Focusing on hip flexor strength and hip ER ROM on the nondominant side may be of value for LBP rehabilitation and prevention in RGs.

Keywords

artistic athlete lower back injury physical function rhythmic gymnastics

Rhythmic gymnastics is characterized by combining classical ballet movements with strength and fitness derived from artistic gymnastics and modern dance. Rhythmic gymnastics has a shorter history than artistic gymnastics, having been adopted as an official event for individual competitions at the 1984 Los Angeles Games and for team competitions at the 1996 Atlanta Games. Rhythmic gymnasts (RGs) perform programs with music that include intricate body maneuvers coordinated with manipulation of handheld apparatus, such as ropes, hoops, balls, clubs, and ribbons. These programs feature leaps, turns, balances, and acrobatic maneuvers, demanding exceptionally high flexibility and muscle strength.³⁰ Gymnastics, including rhythmic gymnastics, is characterized in specialized sports from an early period, with training typically commencing before adulthood.^4,16,22,32 To attain elite status, RG athletes undergo gymnastics-specific interventions, such as extreme stretching, from an early age to acquire advanced flexibility and balance skills.⁸

RG athletes have been reported to be at risk of low back pain (LBP) due repetitive hyperextension of the lumbar spine during competition.²⁹ Hutchinson¹² reported that 86% of elite-level RGs experience LBP during practice and competition, with LBP being the sole recorded injury necessitating time away from the sport. Cupisti et al⁶ found that injuries occurred in 70% of RG athletes, with back injuries constituting 22.2%, after ankle and foot injuries (38.9%). Cugusi et al⁵ summarized risk factors for LBP in female artistic gymnasts, identifying increased age, weight, and gymnastics practice time as factors associated with LBP. Although RGs experiencing more overuse injuries than artistic gymnasts and therefore would benefit from reducing RGs’ practice time, it is unclear whether this is practical.²⁸ Therefore, identifying modifiable factors associated with LBP in RGs is essential. Mizoguchi et al²⁴ reported that decreased flexibility of hip and shoulder flexors on the dominant side was associated with current LBP of high school volleyball players in case control setting. In a prospective study, Enoki et al⁷ also found that lower passive hip flexion angle was an associated factor for chronic LBP in pole vaulters and decathletes. However, it remains unclear whether these factors apply to RGs with their heightened flexibility.

This study aimed to identify modifiable physical characteristics associated with LBP in RGs, particularly the range of motion (ROM) and muscle strength of the lumbar region and adjacent joints to the lumbar, which are clinically closely associated with LBP. Given their flexibility, we hypothesized that factors other than muscle tightness are associated with the characteristics of LBP in RG athletes.

Methods

Participants

This study included 95 collegiate female RGs from 3 universities. All participants had started rhythmic gymnastics before entering college and were competitive RGs who competed in national to international competitions. Detailed explanation of the experimental procedures was provided to the participants before obtaining written informed consent. The study received approval from the ethics committee of Waseda University (2022-094).

Assessment Procedures

The assessment comprised a questionnaire survey, dual-energy X-ray absorptiometry (DXA) scan, magnetic resonance imaging (MRI) scan, hip muscle strength tests, muscle tightness and flexibility tests, and spinal alignment evaluation, all conducted on the same day. Participants first completed the questionnaire, followed by measurements of height and weight (taken clothed and without shoes) using a stadiometer (YG200DN, Yagami Co) to the nearest 0.1 cm and a digital scale (BC622, TANITA Co) to the nearest 0.1 kg, respectively. The questionnaire survey covered age at beginning competition, age at menarche, training hours per day, days per week, and current and previous history of LBP. All participants reported no lumbar pain during testing.

DXA Measurement

DXA (Horizon, Hologic Inc) was used to measure whole-body bone mineral content and density as well as the regional lean body mass of the trunk. The participants were placed in the supine position so that the midline of the body coincided with the centerline of the examination table. The supine posture involved the following: the shoulder joint at 45°; the forearm pronated; the hip and knee fully extended; and the feet fixed in way that the toes were in contact and did not move during scanning. The trunk region was defined as extending from the first cervical vertebra to the femoral neck, excluding the arm region from the axillary fossa to the fingertips.³⁵ Scoliosis presence was determined from the DXA image by an orthopaedic surgeon with >40 years of experience, considering the low radiation dose following the method of Jamaludin et al.¹⁵

MRI Scan and Image Acquisition

Spinal alignment and lumbar pathologies, such as lumbar spondylolysis, lumbosacral facet syndrome, and herniated discs, were assessed using short tau inversion recovery (STIR) and 3-dimensional (3-D) LAVA imaging on a 3T MRI scanner (SIGNA Premier 3.0T; GE Healthcare). STIR scanning parameters were as follows: repetition time/echo time (TR/TE), 1800/minimum (60) ms; slice thickness, 2.0 mm; field of view (FOV), 24.0 cm; matrix, 320 × 224; inversion time, 150 ms. The lumbar spine, including the lumbosacral vertebrae, was scanned using 3-D LAVA based on the previous study with the following parameters²: TR/TE, 4.5/2.1 ms; slice thickness, 0.80 mm; FOV, 24.0 cm; matrix, 384 × 38. Participants were supine during scanning, with their ankles in 90° dorsiflexion and knees extended using a fixture.

Hip Muscle Strength Testing

Maximal isometric hip muscle strength was measured for hip flexion, extension, and abduction using handheld dynamometry (HHD; Mobie, SAKAI Medicine). Two examiners, one a physical therapist with >5 years of experience operating the HHD and the other ensuring proper posture, conducted the test. Hip flexion strength was measured in a half-sitting position (adjusted for participant height), with the hip on the measurement side in flexion and the opposite hip and knee in extension (Figure 1a). Hip extension strength was performed in the prone position, with the hip on the measurement side in extension and the contralateral hip and knee in flexion (Figure 1b). Hip abduction strength was conducted in the lateral decubitus, with the hip and knee on the measurement side in extension and the contralateral hip and knee joints in mild flexion (Figure 1c). Participants were instructed to exert maximum force for 3 s, with the HHD device placed proximal to the knee joint with maximum force without leg and pelvis rotation. Maximal hip muscle torque was calculated as hip muscle strength by multiplying force by lever arm and normalized using body weight (Nm/kg): force (kg) × lever arm (m)/body mass (kg). Lever arm length was measured from the superior aspect of the greater trochanter to the lateral epicondyle on the DXA image.

Muscle Tightness and Flexibility Testing

Modified active and passive straight leg raise (SLR), hip extension, and hip external rotation (ER) and internal rotation (IR) were evaluated using a standard goniometer. Considering that RGs reach their maximum ROM in normal SLR, modified active/passive SLR was measured to assess hamstring tightness in RGs with excessive flexibility, with participants flexing the hip on the measurement side to maximum and actively/passively extending the knee from that position (Figure 1d). Hip extension ROM was measured by Thomas²¹ test in which the participant positioned at the edge of the bed with the hip in extension on the measurement side (Figure 1e). Specifically, the participant supine on the testing table, actively and maximally flexing 1 hip joint, bringing the knee toward the chest to flatten the lumbar spine and stabilize the pelvis, which indicates the hip extension ROM. Hip IR and ER were performed with the hip in 90° of flexion (Figure 1f,g).

The shoulder mobility test, which includes a complex function of thoracic extension and shoulder flexion, required participants to rotate their upper limbs from front to back while holding a bar with both hands (Figure 1h).²³ Participants aimed to narrow the width between their hands holding the bar (100 cm in length and 2.5 cm in diameter), which had an attached scale for measuring the between-hands distance. The finger-floor distance (FFD) task involved participants standing on a platform, measuring the distance from the platform surface to the fingertips. FFD indicates that the higher the value shows the higher the flexibility. Measurements for shoulder mobility and FFD were collected in 0.1 cm increments.

Spinal Alignment Evaluation

The thoracic kyphosis angle, measured in a standing position by an examiner using an inclinometer (Bubble Inclinometer; Baseline Inc) over the spinous processes of the first and second thoracic spines and over the twelfth thoracic and first lumbar spines, was recorded once for each angle. The overall thoracic kyphosis angle was calculated by summing all angles.²⁷ The intratester and intertester reliability of this measurement was established (intraclass correlation coefficient, 0.95; 95% CI, 0.91-0.97).²⁰ Lumbar lordosis and sacral slope angles were obtained from the midsagittal plane in MRI scans between the superior edge of L1 and the inferior edge of L5 and between the superior edge of the sacrum and the horizontal line, respectively.³⁶

Assessment of LBP and MRI Findings

Inclusion criteria for LBP included selecting “yes” in the questionnaire survey to either of the questions “Have you often felt pain in your lower back during the past month?” or “Has the degree of pain interfered with your performance during competition (practice or games)?”, or responding “yes” in a direct interview conducted by an orthopaedic surgeon with >10 years of experience to questions regarding experiencing lumbar region pain. In addition, the presence or absence of abnormal findings in L1-L5 was recorded by the same orthopaedic surgeon from MRI scans, including STIR and 3-D LAVA mode. Determination of MRI scans was deidentified by a third party and the surgeon was blinded so as not to allow matching with other data.

Statistical Analysis

A statistical power analysis was conducted for sample size estimation. The study required a minimum of 18 athletes with LBP to achieve 90% power, an alpha of 0.05, based on a reference value of 20% prevalence of LBP as shown in previous studies.^5,10 In addition, since this study is a cross-sectional design, we set the power higher than that of a prospective cohort study.³⁶

Descriptive statistics (means and standard deviations) were used to analyze the participants’ age, height, weight, arm span, arm span-height ratio, athletic careers, age at menarche, practice time, hip muscle strength, ROM, flexibility, and spinal alignment. Left and right hip muscle strength and ROM data were categorized by defining the dominant side (ie, the superior side for holding a 1-leg support posture). After confirming data normality, t tests were used to compare variables between the LBP and non-LBP groups. Effect size was calculated, with criteria for interpretation set at >0.2 (small), >0.6 (moderate), >1.2 (large), and >2.0 (very large). ¹¹ The chi-square test was used to examine the relationships between current LBP and a history of LBP as well as scoliosis prevalence. Multivariate logistic regression, using variables with P < 0.05 from the 2-group comparisons, was employed to identify predictive variables for LBP. Only significant variables (P < 0.05) were retained in the model for each dataset. Models were developed using forward stepwise (conditional) regression. The significance was set at P < 0.05 in all cases, and data were analyzed using SPSS for Windows Version 28.0 (SPSS Inc).

Results

Among the 95 participants, (27.3%) were classified in the LBP group. The MRI findings determined by the orthopaedic surgeons are shown in Table 1. A comparison of basic characteristics between the LBP and non-LBP groups is presented in Table 2. The LBP group exhibited a significantly higher arm span-height ratio (1.03 ± 0.18 vs. 1.01 ± 0.02; P = 0.013) and longer practice time (6.7 ± 2.7 vs 5.7 ± 2.3 h/day; P = 0.05). In addition, a significant association was observed between the prevalence of LBP and a history of LBP (χ² = 13.8; P < 0.001). No significant differences were noted in age, height, weight, bone mineral content, bone mineral density, trunk lean mass, athletic careers, age at menarche, or scoliosis between the 2 groups.

Table 1.

MRI findings in RGs with LBP

Diagnosis	Level	LBP (n = 26)	Non-LBP (n = 69)
Lumbosacral facet syndrome	L2/3	1	1
	L3/4	2	1
	L4/5	3	3
	L5/S1	2	3
Lumbar spondylolysis	L4	0	1
	L5	5	2
Disk herniation	L4/L5	0	1
	L5/S1	2	1
No abnormal image findings		11	56

LBP, low back pain; MRI, magnetic resonance imaging; RGs, rhythmic gymnasts.

Table 2.

Characteristics of the LBP and non-LBP groups

Variables	LBP (n = 26)	Non-LBP (n = 69)	P Value
Age, y	20.8 ± 1.1	19.6 ± 1.1	0.09
Height, cm	161.4 ± 5.6	160.7 ± 5.4	0.59
Weight, kg	53.8 ± 5.2	54.3 ± 6.0	0.34
Arm span, cm	166.3 ± 6.5	163.6 ± 7.1	0.11
BMC, g	2151.6 ± 267.7	2156.5 ± 212.7	0.46
BMD, g/cm³	1.12 ± 0.78	1.12 ± 0.70	0.50
Trunk lean mass, kg	20.18 ± 1.75	20.14 ± 1.97	0.47
Arm span-height ratio	1.03 ± 0.18	1.01 ± 0.02	0.01
Athletic careers, y	13.9 ± 2.2	13.4 ± 2.9	0.22
Age at menarche, y	15.8 ± 1.5	15.2 ± 2.2	0.15
Practice time, h/day	6.7 ± 2.7	5.7 ± 2.3	0.05
LBP history
With	10	5
Without	16	64	<0.001
Scoliosis
With	11	24
Without	15	45	0.63

LBP, low back pain; BMC, bone mineral content; BMD, bone mineral density. Bold P values indicate significance levels below 5%.

Table 3 presents the differences in hip muscle strength between the LBP and non-LBP group. The LBP group exhibited lower hip flexion muscle strength on the nondominant side (97.2 ± 18.1 vs. 105.7 ± 17.3 Nm/kg; P = 0.02), and the difference between the nondominant and dominant sides was greater compared with the non-LBP group (−9.1 ± 10.3 vs. −3.0 ± 16.3 Nm/kg; P = 0.04).

Table 3.

Differences in hip muscle strength between the LBP and non-LBP groups

Variables		LBP	Non-LBP	Mean Difference	Effect Size	P Value
Hip muscle strength, Nm/kg
Flexion	Dominant	106.3 ± 19.2	108.8 ± 19.1	−2.5 ± 4.4 (−11.2 to 6.3)	0.13	0.29
	Nondominant	97.2 ± 18.1	105.7 ± 17.3	−8.5 ± 4.0 (−16.6 to −0.5)	0.49	0.02
	Difference	−9.1 ± 10.3	−3.0 ± 16.3	−6.1 ± 3.4 (−12.9 to 0.8)	0.41	0.04
Extension	Dominant	135.0 ± 33.7	144.7 ± 36.2	−9.7 ± 8.2 (−26.0 to 6.6)	0.27	0.11
	Nondominant	139.5 ± 39.1	148.3 ± 40.6	−8.8 ± 9.3 (−27.2 to 9.6)	0.22	0.17
	Difference	4.4 ± 16.0	3.5 ± 19.9	0.9 ± 4.4 (−7.8 to 9.6)	0.40	0.83
Abduction	Dominant	144.0 ± 39.8	156.1 ± 39.1	−12.1 ± 9.1 (−30.0 to 5.9)	0.31	0.09
	Nondominant	150.6 ± 56.8	157.0 ± 39.9	−6.4 ± 10.4 (−27.1 to 14.3)	0.14	0.27
	Difference	6.5 ± 26.8	0.9 ± 29.7	5.7 ± 6.7 (−7.6 to 18.9)	0.19	0.20

Data are mean ± SD. LBP, low back pain. Differences were calculated by subtracting the nondominant side value from the dominant side value.

Differences in ROM, flexibility, and spinal alignment between the LBP and non-LBP groups are shown in Table 4. The LBP group had an increased difference between nondominant and dominant sides on modified active straight leg raise (m-ASLR) compared with the non-LBP group (−7.4° ± 11.2° vs −3.8° ± 7.7°; P = 0.04), a decreased hip ER ROM on the nondominant side compared with the dominant side (47.2° ± 5.3° vs 53.3° ± 7.7°; P < 0.001), and an increased difference between nondominant and dominant sides on hip ER ROM (−5.6° ± 6.1° vs −1.2° ± 6.6°; P = 0.002) than non-LBP group.

Table 4.

Differences in ROM, flexibility, and spine alignment between the LBP and non-LBP groups

Variables		LBP	Non-LBP	Mean Difference	Effect Size	P Value
ROM, deg
m-ASLR	Dominant	171.5 ± 4.3	171.8 ± 4.9	−0.2 ± 1.1 (−2.4 to 2.0)	0.05	0.42
	Nondominant	164.2 ± 12.7	168.0 ± 9.2	−3.8 ± 2.4 (−8.5 to 0.9)	0.37	0.06
	Difference	−7.4 ± 11.2	−3.8 ± 7.7	−3.6 ± 2.0 (−7.6 to 0.4)	0.41	0.04
m-PSLR	Dominant	158.8 ± 13.9	161.3 ± 12.6	−2.6 ± 3.0 (−8.5 to 1.2)	0.19	0.20
	Nondominant	152.7 ± 14.1	157.12 ± 14.1	−4.4 ± 3.5 (−11.5 to 2.6)	0.20	0.11
	Difference	−6.1 ± 12.4	−4.2 ± 10.1	−1.9 ± 2.5 (−6.8 to 3.1)	0.29	0.23
Hip extension	Dominant	11.1 ± 4.6	12.5 ± 4.7	−1.4 ± 1.1 (−3.5 to 0.8)	0.29	0.24
	Nondominant	10.6 ± 4.8	12.6 ± 5.5	2.0 ± 1.2 (−4.4 to 0.5)	0.37	0.06
	Difference	−0.5 ± 4.3	0.1 ± 3.3	−0.6 ± 0.8 (−2.2 to 1.0)	0.17	0.24
Hip ER	Dominant	52.9 ± 7.4	54.5 ± 6.6	−1.6 ± 1.6 (−4.8 to 1.5)	0.24	0.15
	Nondominant	47.2 ± 5.3	53.3 ± 7.7	−6.0 ± 1.6 (−9.3 to −2.8)	0.85	<0.001
	Difference	−5.6 ± 6.1	−1.2 ± 6.6	−4.4 ± 1.5 (−7.4 to −1.5)	0.68	0.002
Hip IR	Dominant	55.6 ± 10.4	52.9 ± 10.2	2.7 ± 2.4 (−2.0 to 8.6)	0.26	0.13
	Nondominant	60.0 ± 11.8	56.7 ± 10.4	3.3 ± 2.5 (−1.6 to 8.2)	0.31	0.11
	Difference	4.4 ± 8.0	3.8 ± 8.8	0.6 ± 2.0 (−3.3 to 4.5)	0.07	0.38
Flexibility		25.4 ± 5.3	25.4 ± 6.0	0.5 ± 1.3 (−2.6 to 2.7)	0.01	0.49
Shoulder flexibility, cm		28.9 ± 17.3	35.0 ± 29.3	−6.1 ± 6.1 (−18.2 to 6.1)	0.23	0.16
FFD, cm		25.4 ± 5.3	25.4 ± 6.0	0.5 ± 1.3 (−2.6 to 2.7)	0.00	0.49
Spine alignment, deg
Thoracic kyphosis angle		18.6 ± 7.7	19.0 ± 7.4	−0.5 ± 1.8 (−4.0 to 3.0)	0.05	0.40
Lumbar lordosis angle		29.5 ± 7.4	29.3 ± 11.8	0.2 ± 3.2 (−6.3 to 6.8)	0.02	0.47
Sacral slope angle		38.3 ± 7.0	37.4 ± 8.1	0.9 ± 1.9 (−2.8 to 4.6)	0.12	0.31

Data are means ± SD. Bold P values indicate significance levels below 5%. Differences were calculated by subtracting the nondominant side value from the dominant side value. FFD, finger-floor distance; LBP, low back pain; m-ASLR, modified active straight leg raise; m-PSLR, modified passive straight leg raise; ROM, range of motion.

In multivariate logistic regression analysis (Table 5), 3 variables, namely LBP history (odds ratio [OR], 6.33; 95% CI, 1.56-25.62; P = 0.011), differences in hip flexion strength (OR, 1.04; 95% CI, 1.00-1.08; P = 0.040), and differences in hip ER ROM (OR, 1.12, 95% CI, 1.03-1.19; P = 0.013), emerged as the most significant factors associated with LBP in RGs.

Table 5.

Multivariate logistic regression analysis of variables associated with LBP

Variables	Regression Coefficient	P Value	OR	95% CI
LBP history	1.85	0.011	6.33	1.56-25.62
Practice time	0.23	0.06	1.26	0.99-1.61
Hip flexion strength difference	0.42	0.04	1.04	1.00-1.08
Hip ER ROM difference	0.12	0.013	1.12	1.03-1.19
m-ASLR difference	0.05	0.12	1.26	0.99-1.11
Constant	3.01

Model chi-square test (P < 0.001). Overall percentage correctly predicted, 78.3%. Pseudo-R² (McFadden) = 0.406. Differences were calculated by subtracting the nondominant side value from the dominant side value. Bold P values indicate significance levels below 5%. ER, external rotation; LBP, low back pain; m-ASLR, modified active straight leg raise; OR, odds ratio.

Discussion

Joint hypermobility and lumbar spine hyperextension, which are specific physical characteristics, are recognized as related factors affecting LBP in RGs. However, limited findings on joint mobility, flexibility, muscle tightness, and muscle strength in RGs hinder identifying modifiable LBP risk factors. The present study aimed to determine the modifiable physical characteristics of RGs with LBP, with a focus on side-to-side differences. We hypothesized that characteristics other than flexibility are associated with LBP. Our results revealed associations with a history of LBP and side-to-side differences in both hip flexion strength and hip ER ROM.

In our study, LBP prevalence in RGs was 27.3%, with 57.7% of these participants exhibiting abnormal findings on MRI scans. The rate is slightly higher than the approximately 20% reported in previous studies,^5,10 indicating a high LBP rate among RGs. Abnormal MRI scan findings were observed mostly in the lower lumbar spine, including lumbar spondylolysis at L5, intervertebral arthritis at L3/4, L4/5, and L5/S1, and lumbar disc herniation at L5/S1. Previous studies have shown that 90% of patients with LBP are diagnosed with nonspecific LBP, where the cause cannot be clinically identified.¹⁸ In RGs, LBP is often associated with mechanical stress to the lumbar region. ^12,30,33 In addition, a previous history of LBP was associated strongly with current LBP. Although the timing of lumbar spondylolysis found on MRI scans remains unclear, managing lumbar injuries during the growth period may be crucial for preventing LBP in collegiate RGs. Previous studies indicated that lumbar bone stress injury and spondylolysis are more likely to occur during skeletal immaturity,^19,36 and a history of lumbar spondylolysis is an associated factor for LBP.¹⁴

Frutuoso et al⁹ reported that the nondominant leg predominately acts as the propulsion leg during most jumps in rhythmic gymnastics, indicating functional asymmetry. The physical characteristics of the LBP group had an increased difference between nondominant and dominant sides on m-ASLR, a decreased hip ER ROM on the nondominant side compared with the dominant side, and an increased difference between nondominant and dominant sides on hip ER ROM than the non-LBP group. Wu et al³⁷ also observed higher bone mass density at the proximal femur on the take-off side (ie, nondominant side) compared with the landing side (ie, dominant side) in female RGs. Given the higher load on the nondominant side, it is conceivable that the dysfunction of the nondominant side may be related to the LBP in RGs.

In the present study, the hip flexion muscle test involved extending the contralateral hip to place the hip on the measurement side in the final range of flexion. A previous study showed a significant contribution of the iliopsoas muscle with a greater hip flexion angle.¹⁷ Therefore, RGs with LBP may have reduced iliopsoas muscle strength on the nondominant side. The iliopsoas muscle serves as a stabilizer linking the trunk, pelvis, and legs.¹ Bachrach³ also suggested that weakness in the psoas muscles—part of the hip flexors—could lead to laxity and mechanical instability in the lumbosacral region,¹³ stimulating nociceptors in the surrounding soft tissues and inducing pain. Modifying the function of the nondominant iliopsoas muscle is preferable for maintaining trunk stability in preventing and treating LBP in RGs. Notably, the peak torque of hip flexion in female RGs has been shown to be lower on the nondominant side compared with the dominant side,⁹ as observed in the non-LBP group in the present study. Given the demonstrated association between hip flexion muscle strength and LBP,²⁶ understanding the degree of asymmetry in hip flexor strength is needed for LBP prevention and treatment in RGs.

The difference in hip ER ROM between nondominant and dominant side emerged as a factor associated with LBP in RGs. In addition, since the nondominant hip ER ROM of the LBP group was lower than that of the non-LBP group, this result would indicate a relative limitation of hip ER ROM on the nondominant side. Sadeghisani et al³¹ identified a greater lumbopelvic rotation ROM during passive hip ER in patients with LBP regularly involved in rotation-related sports. In addition, Scholtes et al³⁴ suggested that patients with LBP who participate in rotation-related sports may exhibit greater and earlier lumbopelvic movement during active hip lateral rotation compared with those without LBP. Thus, limited hip ER ROM could potentially lead to compensatory lumbosacral movement. Therefore, given the limited hip ER ROM in in hip flexion, it would be recommended to assess and improve the tightness of the nondominant lateral hip external rotators, such as the gluteus maximus and deep external rotators, to address LBP in RG.

In the LBP group, the difference between the nondominant and dominant sides was more pronounced than that in the non-LBP group, indicating a relative lack of m-ASLR on the nondominant side. However, m-ASLR was not identified as an associated factor for LBP in the multivariate analysis. In addition, no difference in modified passive straight leg raise was observed between the LBP and non-LBP groups, confirming greater muscle flexibility in the posterior femoral muscles of RGs. Moltubakk et al²⁵ demonstrated that RGs with high hamstring flexibility had an advantage in flexion torque production in the knee-extended position compared with non-LBP group. In other words, m-ASLR may be a suitable assessment of hamstring flexibility in highly flexible RGs, indicating functional ROM reflecting muscle function in the terminal ROM of the joint angle. RGs are required to maintain a straight posture with straightened lower extremities for an attractive appearance during competition. Based on these observations, we speculate that the LBP group developed compensatory movements in the lumbar region due to the lack of m-ASLR.

This study had several limitations. First, given the cross-sectional nature of the study, we could not determine whether effects were due to LBP-associated disorders or characteristics of RGs who develop LBP. Second, LBP may not entirely align with abnormal findings in the lumbar region identified on MRI scans. In addition, as MRI coverage was limited to the first to the fifth lumbar spine, MRI-based findings for the thoracic spine, considered a potentially susceptible region in highly flexible RGs, could not be determined. This study provides valuable knowledge for RGs with high flexibility.

Conclusion

RGs had a high rate of LBP (27.3%), and LBP history was associated as a nonmodifiable parameter for LBP. In addition, deficiencies not only in nondominant hip flexor muscle strength but also in nondominant hip ER ROM were associated with LBP as a modifiable physical function. These results may suggest the usefulness of incorporating flexor strength exercises into training routines as well as implementing daily assessments and stretching for nondominant hip ER ROM for even highly flexible RGs to effectively manage LBP.

Footnotes

Acknowledgements

The authors sincerely thank the players who participated in this study.

The authors report no potential conflicts of interest in the development and publication of this article.

ORCID iDs

Toshiharu Tsutsui

Yuka Tsukahara

References

Andersson

Oddsson

Grundström

Thorstensson

The role of the psoas and iliacus muscles for stability and movement of the lumbar spine, pelvis and hip. Scand Med Sci Sports. 1995;5(1):10-16.

Ang

Robertson

Malara

, et al. Diagnostic accuracy of 3-T magnetic resonance imaging with 3D T1 VIBE versus computer tomography in pars stress fracture of the lumbar spine. Skeletal Radiol. 2016;45(11):1533-1540.

Bachrach

RM.

Psoas dysfunction/insufficiency, sacroiliac dysfunction and low back pain. In: A

Vleeming

, et al, eds. Movement Stability and Low Back Pain. The Essential Role of the Pelvis. New York, NY: Churchill Livingstone. 1997;309-318.

Camargo

CTA

Gomez-Campos

Cossio-Bolaños

Barbeta

VJDO

Arruda

Guerra-Junior

. Growth and body composition in Brazilian female rhythmic gymnastics athletes. J Sports Sci. 2014;32(19):1790-1796.

Cugusi

Manca

Sarritzu

, et al. Risk factors associated with low back pain in competitive female gymnasts: a meta-analytic approach. J Sports Sci. 2020;38(22):2543-2552.

Cupisti

D’Alessandro

Evangelisti

, et al. Injury survey in competitive sub-elite rhythmic gymnasts: results from a prospective controlled study. J Sports Med Phys Fitness. 2007;47(2):203-207.

Enoki

Kuramochi

Murata

Tokutake

Sakamoto

Shimizu

Internal risk factors for low back pain in pole vaulters and decathletes: a prospective study. Orthop J Sports Med. 2021;9(2):232596712098561.

Feeley

Agel

LaPrade

RF.

When is it too early for single sport specialization?

Am J Sports Med. 2016;44(1):234-241.

Frutuoso

Diefenthaeler

Vaz

de la Rocha Freitas

Lower limb asymmetries in rhythmic gymnastics athletes. Int J Sports Phys Ther. 2016;11(1):34-43.

10.

Gram

MCD

Clarsen

Bø

. Injuries and illnesses among competitive Norwegian rhythmic gymnasts during preseason: a prospective cohort study of prevalence, incidence and risk factors. Br J Sports Med. 2021;55(4):231-236.

11.

Hopkins

Marshall

Batterham

Hanin

Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc. 2009;41(1):3-12.

12.

Hutchinson

MR.

Low back pain in elite rhythmic gymnasts. Med Sci Sports Exerc. 1999;31(11):1686-1688.

13.

Huxel Bliven

Anderson

BE.

Core stability training for injury prevention. Sports Health. 2013;5(6):514-522.

14.

Iwamoto

Abe

Tsukimura

Wakano

Relationship between radiographic abnormalities of lumbar spine and incidence of low back pain in high school and college football players: a prospective study. Am J Sports Med. 2004;32(3):781-786.

15.

Jamaludin

Fairbank

Harding

, et al. Identifying scoliosis in population-based cohorts: automation of a validated method based on total body dual energy X-ray absorptiometry scans. Calcif Tissue Int. 2020;106(4):378-385.

16.

Jayanthi

Pinkham

Dugas

Patrick

LaBella

Sports specialization in young athletes: evidence-based recommendations. Sports Health. 2013;5(3):251-257.

17.

Jiroumaru

Kurihara

Isaka

Measurement of muscle length-related electromyography activity of the hip flexor muscles to determine individual muscle contributions to the hip flexion torque. SpringerPlus. 2014;3(1):624.

18.

Koes

Van Tulder

Thomas

Diagnosis and treatment of low back pain. BMJ. 2006;332(7555):1430-1434.

19.

Kountouris

Sims

Beakley

, et al. MRI bone marrow oedema precedes lumbar bone stress injury diagnosis in junior elite cricket fast bowlers. Br J Sports Med. 2019;53(19):1236-1239.

20.

Lewis

Valentine

RE.

Clinical measurement of the thoracic kyphosis. A study of the intra-rater reliability in subjects with and without shoulder pain. BMC Musculoskelet Disord. 2010;11(1):39.

21.

Magee

DJ.

Orthopedic Physical Assessment. 4th ed. Philadelphia, PA: Saunders; 2002.

22.

Malina

Baxter-Jones

Armstrong

, et al. Role of intensive training in the growth and maturation of artistic gymnasts. Sports Med. 2013;43:783-802.

23.

Mise

Mitomi

Mouri

, et al. Hypomobility in males and hypermobility in females are risk factors for shoulder pain among young swimmers. J Sport Rehab. 2022;31(1):17-23.

24.

Mizoguchi

Akasaka

Otsudo

Hall

Physical function characteristics in Japanese high school volleyball players with low back pain: a case-controlled study. Medicine. 2020;99(46):e23178.

25.

Moltubakk

Eriksrud

Paulsen

Seynnes

Bojsen-Møller

Hamstrings functional properties in athletes with high musculo-skeletal flexibility. Scand Med Sci Sports. 2016;26(6):659-665.

26.

Nourbakhsh

Arab

AM.

Relationship between mechanical factors and incidence of low back pain. J Orthop Sports Phys Ther. 2002;32(9):447-460.

27.

O’Gorman

Jull

Thoracic kyphosis and mobility: the effect of age. Physiotherapy Practice. 1987;3(4):154-162.

28.

Paxinos

Mitrogiannis

Papavasiliou

, et al. Musculoskeletal injuries among elite artistic and rhythmic Greek gymnasts: a ten-year study of 156 elite athletes. Acta Orthop Belg. 2019;85(2):145-149.

29.

Purcell

Micheli

Low back pain in young athletes. Sports Health. 2009;1(3):212-222.

30.

Roberts

Spine injuries in rhythmic gymnastics. Sport Health. 2009;27(3):27-29.

31.

Sadeghisani

Sobhani

Kouchaki

Bayati

Ashari

Mousavi

Comparison of lumbopelvic and hip movement patterns during passive hip external rotation in two groups of low back pain patients with and without rotational demand activities. Ortop Traumatol Rehabil. 2015;17(6):611-618.

32.

Sands

Caine

Borms

Scientific Aspects of Women’s Gymnastics. Vol 45. Basel: Karger Medical and Scientific Publishers; 2003.

33.

Sands

McNeal

Penitente

, et al. Stretching the spines of gymnasts: a review. Sports Med. 2016;46(3):315-327.

34.

Scholtes

Gombatto

Van Dillen

LR.

Differences in lumbopelvic motion between people with and people without low back pain during two lower limb movement tests. Clin Biomech (Bristol). 2009;24(1):7-12.

35.

Takei

Taketomi

Tanaka

Torii

Growth pattern of lumbar bone mineral content and trunk muscles in adolescent male soccer players. J Bone Miner Metab. 2020;38(3):338-345.

36.

Tsutsui

Iizuka

Takei

Maemichi

Torii

Risk factors for symptomatic bilateral lumbar bone stress injury in adolescent soccer players: a prospective cohort study. Am J Sports Med. 2023;51(3):707-714.

37.

Ishizaki

Kato

Kuroda

Fukashiro

The side-to-side differences of bone mass at proximal femur in female rhythmic sports gymnasts. J Bone Miner Res. 1998;13(5):900-906.