Effects of forearm rotation on wrist flexor and extensor muscle activities

Abstract

The forearm muscles coordinately control wrist motion, and their activity is affected by forearm rotation. Although forearm rotation has been implicated in the development of lateral and medial epicondylitis, its biomechanical background remains unknown. Therefore, the present study investigated the activity of wrist muscles in various forearm positions. Surface electromyography of the extensor carpi radialis brevis, extensor carpi ulnaris, flexor carpi radialis, and flexor carpi ulnaris was performed on 40 healthy upper limbs. We initially measured muscle strength and electromyographic activity (integrated electromyographic value per second) at maximum voluntary output towards wrist extension and flexion in a neutral position. We then assessed electromyographic activity under constant wrist torque (75% of maximum strength in the neutral position) in pronation, the neutral position, and supination. The percentage of maximum electromyographic activity was evaluated for each position. In wrist extension, the extensor carpi radialis brevis was activated during forearm pronation, while extensor carpi ulnaris activity did not change in any forearm position. In wrist flexion, the flexor carpi radialis was activated during forearm supination, while flexor carpi ulnaris activity was significantly lower with forearm pronation than in the neutral position. Since muscle activation increases traction force at the tendon origin, forearm positions that increase muscle activity may be a biomechanical risk factor for the development of tendinopathy. The present results are consistent with epidemiological and pathological findings on lateral and medial epicondylitis. These results provide insights into wrist biomechanics and the pathophysiology of lateral and medial epicondylitis.

Introduction

The multi-directional motion of the wrist compositely emerges from the midcarpal and radiocarpal joints [1–3], enabling precise hand positioning. The kinetic axis of the midcarpal joint, termed the dart thrower’s motion, shifts from radial extension to ulnar flexion [1, 2, 4]. This motion is predominantly active in flexion and significant for effectively transmitting upper limb force to objects [3, 5, 6]. Conversely, the kinetic axis perpendicular to the dart thrower’s motion minimizes midcarpal motion and emphasizes radiocarpal motion [1]. This kinetic axis shift from ulnar extension towards radial flexion is termed the opposite dart thrower’s motion and is dominant in wrist extension [1, 3]. The synergistic functions of these joints facilitate the wrist’s ability to move in diverse directions [7, 8].

However, the biomechanical mechanisms controlling synergistic wrist motion remain unclear, largely due to the involvement of forearm rotation, which affects the action vectors of wrist muscles in three dimensions [9]. Since no muscles attach to the proximal carpal bones, the radiocarpal joint moves subordinately to the midcarpal joint via ligamentous actions. This coordination further complicated by the influence of forearm rotation, making it difficult to fully comprehend wrist biomechanics.

Epidemiological findings on lateral and medial epicondylitis of the humerus may provide important information for elucidating the biomechanical effects of forearm rotation on wrist muscles. Labor tasks requiring forearm pronation may induce tendinopathic changes in the extensor carpi radialis brevis (ECRB), leading to the development of lateral epicondylitis [10, 11]. Similarly, tasks involving supination may affect the flexor carpi radialis (FCR), resulting in medial epicondylitis [12]. These tendinopathies arise from recurrent, significant traction forces at tendon origin [13–15]. Since muscle activation increases traction force at the tendon origin, forearm positions that increase muscle activity may be a biomechanical risk factor for the development of tendinopathy. Therefore, the present study hypothesized that the activity of the ECRB increases during extension in pronation, while that of the FCR increases during flexion in supination. To prove this hypothesis, we preliminarily investigated forearm positions and maximum muscle strength; wrist extensors were stronger in pronation, while wrist flexors were stronger in supination [16]. Besides these findings, a comprehensive understanding of muscle coordination necessitates further examinations of the activity of individual muscles. To provide insights into their coordinated motion, the present study investigated wrist muscle activity in three forearm positions chosen for their relevance to daily activities and clinical significance regarding injury risk: pronation, associated with a higher risk of lateral epicondylitis during pickup tasks [10, 11]; supination, linked to medial epicondylitis during heavy lifting [12]; and the neutral position, an intermediate posture.

Materials and methods

Study design and participants

This study protocol conformed to the principles outlined in the 1964 Declaration of Helsinki. Our Institutional Review Board approved this study (Approval No. KC-H30, Date 6. March. 2023). We obtained written informed consent from all participants.

This was a cross-sectional observational study involving healthy individuals (Level of Evidence III), which investigated the activity levels of muscles involved in wrist motion using surface electromyography (SEMG). SEMG measurements were conducted under consistent wrist torque across three forearm positions: pronation, a neutral position, and supination. To minimize the issue of crosstalk in SEMG, this study included young male participants (aged 20–40 years), as their larger forearm circumferences facilitate more reliable muscle activity recordings.

Participants comprised 20 healthy male volunteers, contributing 40 limbs bilaterally (age: 29.2 ± 5.8 years, height: 172.3 ± 5.1 cm, weight: 62.0 (59.0–71.3) kg, BMI: 21.1 (19.6–23.0), forearm circumference: 25.2 ± 1.5 cm, grip strength: 42.2 ± 5.6 kgf). This study recruited participants through public announcements at the research institution, focusing on those who voluntarily expressed their willingness to participate in the study. Exclusion criteria were individuals with a history of trauma to the upper limbs and those with symptoms of pain or sensory impairment in the upper limbs.

Preparation for wrist torque measurements

To accurately investigate the muscle activities involved in wrist motion for each forearm position, wrist torque was standardized across these positions. Wrist torque measurements were conducted using a self-developed device (Three-One Design, Inc., Tsukuba, Japan) [14–16] (Fig. 1). When participants performed isometric contractions, the sensor detected torque every 10 msec and displayed a muscle output graph on the monitor.

Fig. 1.

Wrist torque measurement device. This apparatus measures temporal wrist torque in forearm pronation, a neutral position, and supination. The device consists of a focal grip handle (*), a torque sensor integrated into the rotating center (**) (UTMII-20Nm, Unipulse Co., Tokyo, Japan), a platform and belt for forearm fixation (***), and a monitor to output measurement results. The gripping handle allows for adjustments of the position according to the lever arm from the participant’s wrist to the grip center. It may be oriented according to forearm rotational positions. (a) External view of the device, (b) measurement in the neutral position of the forearm, (c) measurement in forearm pronation, and (d) measurement in forearm supination

Participants were seated with their forearms resting on the hand support of the torque measuring device. The chair height was adjusted to ensure that the shoulder were not elevated. The elbow was maintained at a 90-degree flexion position to accurately assess forearm rotation. The position of the handle was adjusted so that the rotation center of the wrist coincided with that of the device. In consideration of the functional wrist position [2], participants gripped the handle with their wrists at a 20-degree extension and neutral radial/ulnar deviation. The extension angle was measured between the third metacarpal and the forearm’s long axis in the sagittal plane, while the radial/ulnar deviation angle was measured in the coronal plane. The distal forearm was securely fastened to the device’s platform using a band at a location where the muscle belly of the target muscles is absent, ensuring that it did not affect the EMG.

Preparation for SEMG measurements

The present study employed clinical electromyography (MEB9600, NIHON KOHDEN, Tokyo, Japan) to measure SEMG. Prior to electrode placement, we shaved the participants’ proximal forearms and cleaned the skin with alcohol. To maintain statistical power for the two-way ANOVA with the available sample size, this study selected two muscles from each group of wrist extensors and flexors for the measurement of SEMG. Muscle selection aimed primarily to elucidate the pathophysiology of lateral and medial epicondylitis. The ECRB and FCR were specifically targeted because they have been implicated in the development of lateral and medial epicondylitis, respectively [11, 12, 17, 18]. In addition, the extensor carpi ulnaris (ECU) and flexor carpi ulnaris (FCU) were selected to examine the effects of forearm rotation on muscle activities. As shown in Fig. 2, we used ultrasonography to identify the target muscles following the methodology of previous studies [19]. The ultrasonography was performed by an examiner with 12 years of experience using ultrasonography in daily practice to manage musculoskeletal disorders. To ensure consistency and minimize variability related to operator dependency, the same examiner conducted the imaging for all participants.

Fig. 2.

Method for identifying electrode placement positions. This figure shows electrode placement using ultrasonography for the ECRB as an example. We initially identified the ECRB and ECRL at a level anterior to that of the EDC, which glides during active finger motions through long-axis imaging (a). In short-axis imaging, we identified the ECRB by referencing the depth selected in the preceding long-axis image (b). At the level of the maximum forearm circumference, we then placed the negative electrode (depicted as a yellow square) at the central part of the muscle belly and attached the positive electrode 2 cm distally. ECRB, extensor carpi radialis brevis; ECRL, extensor carpi radialis longus; EDC, extensor digitorum communis; LE, lateral epicondyle; RH, radial head; EDC; yellow dotted line, the level of the maximum forearm circumference; yellow square, electrode

To minimize crosstalk interference from adjacent muscles, we positioned negative electrodes around the maximum circumference of the forearm and positive electrodes 2 cm distally [20, 21]. In short-axis ultrasound imaging, the central region of the target muscle belly was marked for each forearm position (pronation, the neutral position, and supination). The variation in marking positions due to changes in forearm position was less than the electrode diameter of 7 mm for all muscles and participants. As a result, no adjustments to the electrode placement were deemed necessary based on forearm position. Additionally, we ensured that the electrode attachment method was carefully managed to prevent the lead wires from interfering with the platform during forearm rotation. Based on previous studies [20–22], SEMG was set up as follows: input impedance was set at 10 GΩ, with a common-mode rejection ratio of 100 dB and sampling frequency of 1024 Hz [23, 24]. Considering the Nyquist frequency to avoid aliasing, a high-pass filter of 20 Hz and a low-pass filter of 500 Hz were applied to remove baseline drift and low-frequency noise from the EMG signals [25]. Simultaneous SEMG recordings were conducted on the ECRB, ECU, FCR, and FCU.

Measurement of wrist torques and SEMG during maximal voluntary muscle output

Before measurements, the participant maintained a relaxed forearm condition, and the examiner confirmed that the baseline of the SEMG waveform remained stable without noise-induced fluctuations. We then configured SEMG signals to be rectified and graphically displayed as the integrated value per second (iEMG), chosen for its robustness against noise contamination during quantification. Similarly, we displayed wrist torque as the average value per second.

After initiating wrist torque and SEMG measurements simultaneously, the participant exerted maximum muscle output in the direction of either wrist extension or flexion in the neutral position until the maximum torque was observed.

As shown in Fig. 3, wrist torque was averaged each second, with the highest value being identified as the maximum torque (T-max). Additionally, iEMG was calculated as maximum voluntary electrical activity (MVE) during the 1-second interval of recorded T-max for the ECRB and ECU in extension and for the FCR and FCU in flexion.

Fig. 3.

Measurement of torque and SEMG during maximal voluntary muscle output. This figure shows an example of wrist torque and SEMG during wrist extension. Wrist torque was graphically shown as the average value per second and SEMG as the integrated value per second. Wrist torque and SEMG were simultaneously recorded. We defined the peak of wrist extension torque as T-max (blue arrow) and the corresponding iEMG value as MVE (yellow arrow). SEMG, surface electromyography; T-max, maximum torque; iEMG, integrated electromyography; MVE, maximum voluntary electrical activity; ECRB, extensor carpi radialis brevis; ECU; extensor carpi ulnaris; FCR, flexor carpi radialis; FCU, flexor carpi ulnaris

Muscle activity by forearm positions

To standardize wrist torque across forearm positions for muscle activity comparisons, we need to consider that maximum wrist torque may vary by approximately 25% depending on the forearm position [16]. Therefore, we standardized wrist torque to 75% of T-max in the neutral position for the muscle activity evaluation. Torque graph specifications were changed to update results every 10 msec without averaging, thereby allowing participants to make fine torque adjustments (Fig. 4). By referring to the torque graph, participants aimed to achieve 75% of their T-max. We used a one-second period where participants exerted muscle output within 70–80% of their T-max for the assessment. Participants performed muscle output for the target torque in pronation, neutral, and supination positions. The order was randomized using a random number table, and a 3-minute rest was provided between positions to prevent fatigue [26]. The examiner measured average torque and iEMG for the targeted muscles during this period. These torque and iEMG values were then standardized as ratios to T-max and MVE, presented as %T-max and %MVE, respectively.

Fig. 4.

Method for target torque output. This figure shows an example with a target torque (75% of T-max) of 7 Nm. The participant referred to the real-time output torque and exerted muscle output for more than one second within 70–80% of the T-max range. The examiner measured the average torque and iEMG of the four muscles during the same one-second period. *, Target Torque; Double-headed arrows, one-second period in which muscle output was achieved within ± 5% of the target torque; T-max, maximum torque; iEMG, integrated electromyography

Composite of Muscle Action Vectors

We constructed vector diagrams to represent the muscle actions for each forearm position. Since the %MVE obtained in this study is a scalar quantity without directional information, we referenced the action directions of the four target muscles in different forearm positions from a previous study [9]. The magnitude of each vector was determined by the %MVE values obtained during 75% MVC wrist torque in this study. The direction of each vector, as influenced by forearm position, was derived from the literature [9]. The muscle action vectors for the four target muscles were then composited for each forearm position.

Statistical analysis

Continuous variables in the present study were assessed for normality using the Shapiro-Wilk test. Normally distributed data were presented as the mean ± standard deviation and non-normally distributed data as medians (interquartile ranges). In comparisons of three or more groups, Levene’s test for the homogeneity of variance was conducted before performing ANOVA. In the present study, the sphericity assumption was not violated. Post-hoc analyses were performed using Tukey’s method for significant ANOVA results.

%T-max was compared across different forearm positions using a repeated measures one-way ANOVA. In comparisons of %MVE during wrist extension, a repeated measures two-way ANOVA, considering muscles and forearm positions as factors, was conducted to evaluate the effects of forearm rotation on wrist extensor activity. Other non-normal or heteroscedastic %MVE were compared using Friedman’s test, followed by Scheffe’s method for post-hoc comparisons. We then performed a power analysis of the primary measure, a two-way repeated measures ANOVA of wrist extensor muscle activity. The significance level was set at p < 0.05.

We used G-power (version 3.1.9.7; Heinrich-Heine-Universität Düsseldorf) for a power analysis and Bellcurve for Excel version 3.20 (SSRI Co.) for other statistical analyses.

Results

T-max and MVE

In extension, T-max was 8.2 ± 1.4 Nm and MVE values were ECRB: 532 ± 180 µV and ECU: 416 ± 125 µV. In flexion, T-max was 13.6 ± 2.8 Nm and MVE values were FCR: 424 (327–590) µV and FCU: 564 (407–698) µV.

%T-max during muscle activity measurements

In wrist extension and flexion, output torques during muscle activity measurements did not significantly differ among the forearm positions tested (Table 1).

Table 1.

Wrist torque during muscle activity measurements in each forearm position

	pronation	neutral	supination	p-value
Extension torque [Nm] (%T-max)	6.17 ± 0.14 (74.9 ± 1.6)	6.18 ± 1.09 (74.9 ± 1.8)	6.21 ± 1.07 (75.3 ± 1.5)	0.378
Flexion torque [Nm] (%T-max)	10.21 ± 2.05 (75.2 ± 3.0)	10.22 ± 2.08 (75.1 ± 1.4)	10.24 ± 2.08 (75.3 ± 1.1)	0.859

%T-max, percentage of max torque

Muscle activity during wrist extension based on forearm rotational positions

%MVE during the target torque output for wrist extension is shown in Table 2; Fig. 5. There was a significant interaction effect between muscle activities and forearm rotation for the ECRB and ECU (F (2, 156) = 12.14, p < 0.001). Although the simple main effect of forearm rotational positions was significant for the ECRB (F (2, 156) = 22.22, p < 0.001), it was not significant for the ECU (F (2,156) = 0.09, p = 0.913). The post-hoc analysis revealed that the %MVE of the ECRB was significantly greater in the order of pronation, the neutral position, and supination (pronation vs. neutral: the mean difference was 4.6% (95% CI: 0.8–8.4%, p = 0.049), neutral vs. supination: the mean difference was 13.2% (95% CI: 9.3–17.0%, p < 0.001), pronation vs. supination: the mean difference was 17.8% (95% CI: 13.9–21.6%, p < 0.001).

Table 2.

%MVE during the target torque output for wrist extension

	pronation	neutral	supination	p-value
ECRB	72.3 ± 16.0	67.8 ± 12.0	54.6 ± 15.0	*
ECU	71.3 ± 22.1	72.3 ± 20.0	72.3 ± 12.8	*
FCR	12.0 (9.1–16.2)	11.7 (9.2–17.5)	8.8 (6.5–14.8)	0.014
FCU	10.7 (7.0–13.2)	12.7 (8.9–20.8)	12.7 (7.4–15.8)	0.007

*The results of a two-way ANOVA are detailed in the text

%MVE, percentage of maximum voluntary electrical activity; ECRB, extensor carpi radialis brevis; ECU; extensor carpi ulnaris; FCR, flexor carpi radialis; FCU, flexor carpi ulnaris

Fig. 5.

Interaction plot of forearm rotational positions and wrist extensor muscles. nteraction effect: F (2, 156) = 12.14, p < 0.001. Main effect of the muscles: F (1, 78) = 4.95, p = 0.029. Main effect of the forearm posture: F (2, 156) = 10.17, p < 0.001. ECRB, extensor carpi radialis brevis; ECU; extensor carpi ulnaris; %MVE, percentage of maximum voluntary electrical activity; *, p < 0.05; **, p < 0.01; ns, not significant

A power analysis of the two-way ANOVA showed a Mauchly’s sphericity test result of 0.99, effect size of 0.24, and %MVE correlation coefficient of 0.44. The test power at α = 0.05 was 0.88.

In antagonist flexors, the %MVE of the FCR was significantly higher in pronation than in supination (p = 0.014), but was not significantly different between pronation and the neutral position or between the neutral position and supination. Regarding the FCU, %MVE significantly varied across forearm positions (p = 0.007). It was significantly lower in pronation than in the neutral position (p = 0.009), but was not significantly different between pronation and supination or between the neutral position and supination.

Figure 6 shows muscle action vectors during wrist extension. The composite vector of the four target muscles shifted towards the ulnar side as the forearm pronated and towards the dorsal side as it supinated.

Fig. 6.

Action vector of each muscle in wrist extension output. The composite vector of the four muscles pointed in the dorsal-ulnar direction in forearm pronation and shifted dorsally in supination. %MVE, percentage of maximum voluntary electrical activity; ECRB, extensor carpi radialis brevis; ECU; extensor carpi ulnaris; FCR, flexor carpi radialis; FCU, flexor carpi ulnaris

Muscle activity during wrist flexion based on forearm rotational positions

Table 3 shows %MVE during the target torque for wrist flexion. The %MVE of the FCR significantly differed among forearm positions (p < 0.001). The post-hoc analysis showed that the %MVE of the FCR increased as forearm supinated (pronation vs. neutral: p < 0.001, neutral vs. supination: p = 0.016, pronation vs. supination: p < 0.001). The %MVE of the FCU also significantly varied across positions (p = 0.014), being significantly higher in the neutral position than in pronation (p = 0.023), but not significantly differing between supination and the neutral position or between supination and pronation.

Table 3.

%MVE during the target torque output for wrist flexion

	pronation	neutral	supination	p-value
ECRB	16.1 (11.1–21.9)	15.6 (10.4–20.1)	14.0 (11.1–17.4)	0.293
ECU	45.8 (31.2–57.0)	24.4 (18.1–32.6)	16.1 (13.7–22.7)	< 0.001
FCR	57.8 (51.5–71.6)	79.7 ± 21.9	93.1 (76.8–124.6)	< 0.001
FCU	62.4 ± 16.1	72.2 ± 16.0	73.1 ± 22.4	< 0.001

%MVE, percentage of maximum voluntary electrical activity; ECRB, extensor carpi radialis brevis; ECU; extensor carpi ulnaris; FCR, flexor carpi radialis; FCU, flexor carpi ulnaris

Among antagonist extensors, while the %MVE of the ECRB did not significantly differ among forearm positions, the %MVE of the ECU significantly varied (p < 0.001). The post-hoc analysis showed that the %MVE of the ECU increased as the forearm pronated (pronation vs. neutral: p < 0.001, neutral vs. supination: p = 0.007, pronation vs. supination: p < 0.001).

Figure 7 shows muscle action vectors during wrist flexion. The composite vector of the four target muscles pointed in the palmar-ulnar direction across all forearm positions, with the most significant ulnar deviation being observed in forearm supination.

Fig. 7.

Action vector of each muscle in wrist extension output. The composite vector of the four muscles pointed in the palmar-ulnar direction, with the greatest ulnar deviation observed in supination. During forearm pronation, the ECU was activated despite being an antagonist, contributing to the composite vector’s direction. %MVE, percentage of maximum voluntary electrical activity; ECRB, extensor carpi radialis brevis; ECU; extensor carpi ulnaris; FCR, flexor carpi radialis; FCU, flexor carpi ulnaris

Discussion

The present study examined the effects of forearm rotation on the activity of wrist muscles. During wrist extension, the effects of forearm rotation on wrist extensors varied by muscle: the ECRB was activated in forearm pronation, whereas ECU activity was unchanged. This difference arises from the relationship between each muscle’s course and the axis of forearm rotation, which extends from the radial head center to the ulnar fovea [27, 28]. The ECRB deviates radially from the axis of forearm’s rotation as it extends distally, affecting its action on the wrist to shift more towards extension during pronation and radial extension during supination [9]. In contrast, since the ECU runs in alignment with the rotation axis of the forearm, forearm rotation does not affect ECU running or activity. As a result, the composite vector of the four muscles acts towards the dorsal-ulnar side in forearm pronation and deviates towards the dorsal side in supination. These results provide a biomechanical rationale for previous findings showing that the opposite dart thrower’s motion predominated in wrist extension and that wrist extensor strength was improved in forearm pronation [1, 16].

During wrist flexion output, muscle coordination favored ulnar flexion, i.e., the direction of the dart thrower’s motion. Since the origin of wrist flexors offset to the ulnar side from the forearm’s rotational axis, forearm pronation changes the action direction of the FCR and FCU towards the radial side [9]. Under these conditions, the FCR and FCU significantly reduced their activities. The ECU was activated to compensate for reduced ulnar flexion despite being an antagonist muscle. Conversely, wrist supination changed the direction of the FCR and FCU’s actions towards the ulnar side. Under this condition, the FCR exhibited maximal activation, and the composite vector of the four targeted muscles pointed to the most significant degree of ulnar deviation across all examined forearm positions. These results provide a biomechanical rationale for previous findings showing that the dart thrower’s motion was prevalent in wrist flexion and wrist flexor strength improved in forearm supination [1–4, 16].

The present results are consistent with epidemiological and pathological findings on lateral and medial epicondylitis. The ECRB was activated during wrist extension in forearm pronation, introducing greater traction force to the tendon origin. The ECRB has a biomechanical predisposition to the development of lateral epicondylitis in forearm pronation. Similarly, the FCR was activated during wrist flexion in forearm supination. The FCR has a biomechanical predisposition to the development of medial epicondylitis in forearm supination. The results of this study provide valuable insights into the biological pathophysiology of lateral and medial epicondylitis. These findings may contribute to the further development of treatments aimed at controlling muscle activity. In the treatment of lateral epicondylitis, it is desirable to develop specialized orthotic devices that support the function of the ECRB in pronation while suppressing its excessive activity.

The present study has several limitations that need to be addressed. The use of SEMG to assess muscle activity cannot eliminate crosstalk from adjacent muscles [20]. Although ultrasound was employed to minimize crosstalk during electrode placement, eliminating the effects of superficial muscles on deeper muscles, e.g., the ECRL and ECRB, is challenging. Furthermore, a focus on specific muscles and forearm positions may limit the interpretation of results.

Conclusion

During wrist extension, the activity of the ECRB increased as the forearm pronated, while the activity of the ECU was not related to forearm rotation. Consequently, muscle coordination favored ulnar extension, i.e., the direction of the opposite dart thrower’s motion. During wrist flexion, the activity of the FCR increased in forearm supination, while that of the FCU decreased in forearm pronation. Consequently, muscle coordination favored ulnar flexion, i.e., the direction of the dart thrower’s motion. These results provide insights into the biological pathophysiology of lateral and medial epicondylitis.

Abbreviations

ECRB

Extensor carpi radialis brevis

ECU

Extensor carpi ulnaris

FCR

Flexor carpi radialis

FCU

Flexor carpi ulnaris

iEMG

Integrated electromyography

MVE

Maximum voluntary electrical activity

SEMG

Surface electromyography

T-max

Maximum torque

Author contributions

Kazuhiro Ikeda: research design, the acquisition of data, analysis and interpretation of data, drafting the paper. Koji Kaneoka: research design. Naoto Matsunaga: analysis and interpretation of data. Akira Ikumi: revising the paper criticallyMasashi Yamazaki: approval of the submitted and final versions. Yuichi Yoshii: conception, research design, revising the paper critically, approval of the submitted and final versions.

Funding

This work was supported by JSPS KAKENHI Grant Number 24K23521.

Data availability

The datasets used and/or analyzed in the present study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted according to the principles of the Declaration of Helsinki. The Institutional Review Board of Kikkoman General Hospital approved the study protocol.

Consent for publication

Not applicable.

Consent to participate

Written consent for publication was obtained from all study participants.

Clinical trial number

Not applicable.

Competing interests

The authors declare no competing interests.