Ultrasonic Evaluation of Muscle Recovery Following Free Functioning Gracilis Transfer, A Preliminary Study

Background: Ultrasonic measurement has not been utilized to assess functional recovery of transplanted muscle. This study aimed to investigate the feasibility of B-ultrasound measurement for evaluation of muscle recovery following free functioning gracilis transfer. Methods: From January 2009 to January 2014, 35 patients receiving free functioning gracilis transfer due to total brachial plexus injury were enrolled. B-ultrasound was utilized to measure the cross-sectional area (CSA) of transplanted gracilis muscle at rest and contraction state. The ratio of pre- to post-transplant CSA value at rest state was dened as muscle bulk ratio (MBR). The ratio of CSA value at the contraction state to that at rest state was dened as contraction ratio (CR). Results: Patients with muscle strength M ≥ 4 had signicantly higher CR1 (post-transplant), CR2 (pre-transplant), and range of motion (ROM, joint mobility) than those with muscle strength M (cid:0) 4. The CR1>CR2 group had signicantly higher CR1, muscle strength, and ROM than the CR1 ≤ CR2 group. The MBR>1 group had signicantly higher muscle strength than the MBR ≤ 1 group. CR1 value was highly correlated with muscle strength and with ROM. CR2 value was moderately correlated with muscle strength and ROM. Multivariate linear regression analysis showed that higher CR1/CR2 value was associated with higher muscle strength and higher joint mobility. The CR1>CR2 group had better muscle strength and ROM than the CR1 ≤ CR2 groups. Conclusion: B-ultrasound measurement can quantitatively reect muscle strength following gracilis transfer, and CR but not MBR value could be a potential indicator for functional recovery of the transplanted gracilis muscle.


Background
Functional free gracilis muscle transfer is a surgical procedure for limb motor function in patients with Brachial plexus avulsion, brachial plexus nerve injuries, traumatic muscle loss, and tumor resection. [1][2][3][4][5][6][7] It transplants gracilis muscles with vascular nerve pedicles to the upper limbs, suturing the affected blood vessels and motor nerves to replace the function of lost muscles in the affected area. [8] The reinnervation after gracilis muscle transfer is a slow process, and according to the recovery status, auxiliary surgery (such as fusion of wrist joint) is usually required to promote functional recovery. [9,10] Therefore, timely assessing functional recovery of transplanted muscle can provide a reference for subsequent decisionmaking.
Currently, clinically available methods for assessing functional recovery of transplanted muscle include electromyography, [11] manual muscle test (MMT), [12] and magnetic resonance imaging (MRI). [13] However, all these evaluation methods have some defects. EMG results are not necessarily consistent with muscle strength recovery of the transplanted gracilis muscle. MMT is easy to conduct but subjective, Page 3/14 and muscle strength M4 cannot be quantitatively assessed. MRI cannot dynamically assess muscle contraction function, and it is expensive.
B-ultrasound is non-invasive, convenient, and cheap, which can re ect muscle strength by echo density, muscle thickness, cross-sectional area, and intermuscular volume. [14][15][16][17][18] Compared with the static observation of MRI, B-ultrasound can observe the dynamic contraction of muscle. Ultrasonic measurement can easily obtain morphological data of muscle, such as muscle thickness and crosssectional area. The cross-sectional area measured by B-ultrasound has been used to investigate the effect of resistance training on muscle strength. [15,19] Since muscle morphological parameters are closely related to muscle function, muscle strength and function can be re ected by measuring morphological parameters in ultrasonic measurement. [19] Although the correlation between morphologic parameters by ultrasonic measurement and muscle strength has been demonstrated. [15,20,21] However, there is no study on the application of B-ultrasound in the evaluation of functional recovery of transplanted muscle following free functioning gracilis transfer. In this study, therefore, we aimed to investigate the feasibility of B-ultrasound measurement for evaluation of muscle recovery following free functioning gracilis transfer.

Participants
From January 2009 to January 2014, 35 patients receiving free functioning gracilis transfer due to total brachial plexus injury at our hospital were enrolled. All patients received free functioning gracilis transfer for the rst time to reconstruct the functions of elbow exion -nger/thumb extension, and the ipsilateral accessory nerve was used as a donor motor nerve. The reconstruction of the transplanted gracilis muscle was located from the lateral side of the clavicle to the dorsal metacarpal extensor muscle and the tendon of long extensor muscle of thumb at the dorsal side of the distal forearm. This study was approved by the institutional review board of our hospital. Written informed consent was obtained from each patient.

Muscle strength measurement and range of motion (ROM)
The muscle strength was assessed by manual muscle strength test and the ROM of elbow exion was determined by a protractor tool. All assessment was performed by a trained physician who was not involved in the ultrasonic measurement. To avoid observation bias, the de nition of M4 strength was modi ed to " muscle can resist at least the examiner's one nger or at least 1 kg in weight" according to Lin et al.'s study [22].

Ultrasonic measurement and outcome analysis
All B-ultrasound examinations are performed using the Venue 40 Ultrasound (GE Healthcare, USA) and the cross-sectional area (CSA) of the with gracilis muscle were obtained based on the ultrasound images with the Scion Image software (National Institutes of Health, Bethesda MD). All the process was performed by the same physician and can be completed within 10 min. Before and after transplantation, the ultrasound examinations and corresponding CSA of transplanted gracilis muscle was measured at rest and at contraction state, respectively. Each ultrasound examination was continuously repeated ve times and the corresponding CSA value was averaged.
Before transplantation, the CSA of the gracilis muscle at rest state was measured at the neutral position of the hip joint, while and the CSA of the gracilis muscle at the contraction state was determined at 30°abduction of the hip joint.
For CSA measurement after transplantation, the patient was placed in the supine position, and the upper extremity was straightened. The midpoint of the transplanted gracilis muscle was determined by Bultrasound scan, and the plane was used to measure the CSA of the gracilis muscle at rest state. The CSA of the gracilis muscle at the contraction state was determined at 20 ° abduction of shoulder, wrist neutral position, 60 ° exion of elbow, which was the posture used to restore the original length of the transplanted muscle during surgery. Restoring the original length of the transplanted muscle in this position will not cause the elbow joint straightening obstacle, but also can achieve the best elbow exionnger extension effect. Therefore, patients received B-ultrasound examination in the same posture to make the pre-transplant and post-transplant results comparable.

Muscle bulk ratio (MBR) and contraction ratio (CR)
The ratio of pre-to post-transplant CSA value of gracilis muscle at rest state was de ned as MBR. The ratio of CSA value at the contraction state to that at rest state was de ned as contraction ratio (CR). The CR values of gracilis muscle before and after transplantation were de ned as CR-2 and CR-1, respectively ( Fig. 1). The patients were divided into following dichotomous groups for subgroup analysis of muscle strength or ROM: the MBR > 1 vs. MBR < 1, CR-1 > CR-2 group vs. CR-1 < CR-2 group, muscle strength ≥ M4 vs. muscle strength < M4 groups.

Statistical analysis
Continuous variables were presented as mean ± SD (standard deviation) while categorical data were indicated by number and percentage (%). Student's independent t-test was used to compare the differences between groups.  CR2), and MBR (MBR > 1 vs. MBR ≤ 1). It was found that the M ≥ 4 group had signi cantly higher CR1, CR2, and ROM values than the M 4 group (all P < 0.001, Table 2). The CR1 > CR2 group had signi cantly higher CR1, muscle strength, and ROM than the CR1 ≤ CR2 group (all P < 0.01, Table 3). The MBR > 1 group had signi cantly higher muscle strength than the MBR ≤ 1 group (P = 0.038, Table 4).

Associations of CR/MBR to M/ROM
To further evaluate the potential of ultrasonic outcomes as the indicator for muscle strength, correlation analysis between CR/MBR and muscle strength/ROM was performed. As shown in Table 5, CR1 had the highest correlation coe cient with muscle strength (0.808) and ROM (0.847) (both P < 0.01); CR2 had a medium correlation coe cient with M (0.491) and ROM (0.556) (both P < 0.01). However, MBR had no signi cant correlation with muscle strength and ROM (both P > 0.05). The coe cient between M and ROM was 0.903 (P < 0.001). The associations between CR/MBR and M/ROM were further evaluated by univariate and multivariate linear regression analysis. Multivariate linear regression analysis adjusted for patient's age and gender showed that higher CR1/CR2 value suggested higher muscle strength and ROM value (all P < 0.01, Table 6). Meanwhile, the CR1 > CR2 group had better muscle strength and ROM than the CR1 ≤ CR2 groups (all P < 0.01, Table 6). However, there were no signi cant results in the regression models of MBR (all P > 0.05).

Discussion
In this study, we investigated the feasibility of B-ultrasound measurement for evaluation of muscle recovery following free functioning gracilis transfer. The results showed that The M ≥ 4 group had signi cantly higher CR1, CR2, and ROM value (joint mobility) as compared with the M < 4 group. The CR1 > CR2 group had signi cantly higher CR1, muscle strength, and ROM than the CR1 ≤ CR2 group. The MBR > 1 group had signi cantly higher muscle strength than the MBR ≤ 1 group. CR1 value was highly correlated with muscle strength (r = 0.808) and ROM (r = 0.847), while CR2 value was moderately correlated with muscle strength (r = 0.491) and ROM (r = 0.556). Multivariate linear regression analysis showed that higher CR1/CR2 value was associated with higher muscle strength and higher joint mobility. Meanwhile, the CR1 > CR2 group had better muscle strength and ROM than the CR1 ≤ CR2 groups (all P < 0.01). Taken together, these results suggested that B-ultrasound measurement can quantitatively re ect muscle function following gracilis transfer, and CR but not MBR value could be a potential indicator for muscle function recovery.
Currently, ultrasonic parameters which could be used as the indicator for muscle contraction include echo intensity (EI), [20] muscle thickness, [23] muscle ber pennation, [24] measures of muscle architecture. [18] EI re ects muscle function by the density of high echo signals in the muscle but is susceptible to be affected by the connective tissue between the subcutaneous and muscle bundles. A study on the changes in muscle strength by Jacobs et al. has con rmed that EI is not an optimal indicator of muscle strength. [20] In addition, the transplanted gracilis muscle undergoes denervation and brosis, therefore EI is not suitable for function evaluation following free functioning gracilis transfer. Muscle ber pennation re ects the muscle contraction by the angle between the direction of muscle bers and the long axis of the muscle. [25] Gracilis muscle is a non-feathery muscle, and could not be measured by muscle ber pennation. The transplanted muscle is close to the skins, and the ultrasonic test is susceptible to be affected by probe-induced pressure, hence, muscle thickness measurement is also not applicable for muscle transplantation.
The gracilis muscle is non-feathery muscle, and its muscle bers are arranged in the same direction as the tendon. During the muscle contraction, all the muscle bers slide in parallel. Therefore, the histological cross-sectional area of the muscle bers of the gracilis muscle is substantially the same as the gross anatomical cross-sectional area. [18] Therefore, B-ultrasound can be used to measure CSA, and the dynamic contraction of muscles can be re ected by CSA changes (ie, CR) at rest and contraction. The transplanted gracilis muscle undergo denervation and nerve re-innervation, and the muscle volume changes during this process. The MBR and CR are standardized indicators calculated based on CSA, which can eliminate the impact of muscle volume change during denervation and nerve re-innervation. [26] Hence, we chose CR and MBR as the indicators to explore the recovery of gracilis muscle functional after free functioning gracilis transfer.
Our results showed a high correlation between CR value and muscle function indexes, indicating that CR can be used to dynamically evaluate the recovery of muscle function after transplantation. In addition, muscle strength and ROM were signi cantly higher in the CR1 > CR2 group than in the CR1 ≤ CR2 group, suggesting that when the muscle CR increases after transplantation, the patient has better recovery of muscle strength and joint mobility. We observed an increase in the CR value of the gracilis muscle after transplantation. One of the possible reasons is as follows: Since the motion range of elbow bowing is larger than hip adduction, the muscle bers of transplanted gracilis muscle need to be parallel sliding for a longer distance in the elbow bowing. Therefore, the CR value was elevated after transplantation. Our results showed that the mean CR1 values of the muscle strength ≥ M4 group and the CR1 > CR2 group were 1.35 ± 0.10 and 1.29 ± 0.15, respectively. Based on these results, we propose that CR1 of 1.3 might be an important reference value for transplanted muscle recovery. When the CR value of transplanted muscle reaches 1.3, the muscle might have satisfactory recovery. However, this reference value should be further validated in a large trial.
In this study, we also evaluated MBR, which re ects muscle volume change, ie atrophy (MBR < 1) or hypertrophy (MBR > 1). We found that patients who had no atrophy with gracilis muscle (MBR > 1) showed no signi cant advantage in muscle strength and joint mobility. Subgroup analysis of muscle strength also showed that MBR value was not signi cantly different between two muscle strength groups. Correlation analysis also revealed that there was no correlation between MBR and muscle strength/joint mobility. All these results suggested that MBR is not suitable as an indicator for evaluating muscle recovery following free functioning gracilis transfer.
Our preliminary ndings demonstrated the feasibility of utilizing B ultrasound to evaluate functional recovery after gracilis transfer. Early muscle contraction changes may not be easily detected by physical examination. B-ultrasound examination can dynamically detect the progress of muscle contraction recovery, as well as the tendon-gliding function. When a lower postoperative CR value is found in early postoperative period, the patients can seek the help of rehabilitation doctors for physical therapy to promote nerve reinnervation. If in the late postoperative period (e.g. at 1 year after surgery, there is no progress in muscle streagth for 3 consecutive months), a second femoral muscle transplantation could be performed to rebuild the exor. Even though this was a prospective study, however, the sample size was relatively small. However, In the future, a large trial should be conducted to validate the ndings of this study and de ne the reference values for functional recovery assessment.

Conclusions
In summary, our ndings demonstrated that B-ultrasound measurement can quantitatively re ect muscle function following gracilis transfer, and CR but not MBR could be a potential indicator for muscle function recovery. The CR value of transplanted muscle was highly correlated with the muscle function, which can be used to dynamically assess muscle recovery after muscle transplantation. Elevated CR value of gracilis muscle after transplantation suggested a better prognosis.