Introduction
Amyotrophic lateral sclerosis (ALS) is a rare neurodegenerative disease that involves motor neurons in the cerebral cortex, brainstem, and spinal cord. Reference Oskarsson, Gendron and Staff1 Fasciculation and progressive muscle paresis are the characteristics of ALS. Reference Rosa Silva, Santiago Júnior and Dos Santos2 Unlike other neuromuscular diseases, the paresis of muscle groups in patients with ALS tends to present a dissociated pattern. Split hand, characterized by preferential weakness and wasting of the abductor pollicis brevis (APB) and first dorsal interosseous (FDI) muscles with relative sparing of the abductor digiti minimi muscle (ADM), has showed good sensitivity and specificity in the differential diagnosis of ALS (Figure 1). Reference Hu, Wang and Liu3 Besides, split plus sign, Reference Menon, Bae, Mioshi, Kiernan and Vucic4 split pinch grip strength, Reference Lee, Heshmat, Heggie, Thorpe, McCombe and Henderson5 and split elbow Reference Khalaf, Martin and Ellis6 have been observed by both clinical assessment of muscle strength and electrophysiological studies in ALS patients. Through comparing pairs of muscle groups in a large ALS cohort, Ludolph et al found that the muscles known to receive the more pronounced corticomotoneuronal (CM) connections was significantly weaker. Reference Ludolph, Emilian and Dreyhaupt7 These split phenomena may have suggestive effects on the etiology of ALS, which deserves further exploration.
Picture of split hand sign in an ALS patient.
Since the Awaji criteria Reference de Carvalho, Dengler and Eisen8 in 2008 emphasized the significance of fasciculation potential (FP), fasciculation has been widely considered to be a vital biomarker for the early diagnosis of ALS. These apparently random, spontaneous twitching of muscle fibers can be detected in the whole duration of ALS, possibly related to the overexcitation of motor neurons. Reference de Carvalho and Swash9 The origin of fasciculation remained controversial. Both the upper motor neuron (UMN) and lower motor neuron (LMN) may be involved in the generation of fasciculation, Reference de Carvalho, Kiernan and Swash10 and the elucidation of fasciculation may be critical in understanding the underlying disease process of ALS. Compared to electromyography (EMG), muscle ultrasonography (MUS) has the advantages of noninvasiveness and easy operation in the observation of fasciculation. Misawa et al reported that fasciculations were much more frequently detected by MUS than by EMG in the tongue among ALS patients, and the application of MUS could markedly increase diagnostic sensitivity of ALS. Reference Misawa, Noto and Shibuya11 Our previous study also showed significant differences in fasciculation grade and distribution pattern of fasciculation between ALS than non-ALS patients. Reference Liu, Li and Niu12 Besides, increasing studies emphasized the significance of MUS in the monitoring ALS progression. Reference Arts, van Rooij and Overeem13–Reference Tsugawa, Dharmadasa, Ma, Huynh, Vucic and Kiernan15 To the best of our knowledge, research on differences in fasciculation between antagonistic muscle groups among ALS patients was scarce.
Though real-time ultrasound imaging, we aim to explicit whether the fasciculation intensity in pairs of antagonistic muscle groups tends to present split phenomena as muscle strength in ALS patients and explore whether clinical factors affect its occurrence. The possible origin of the split phenomenon of fasciculation will also be discussed.
Methods
Subjects
Consecutive patients with ALS according to the Awaji criteria Reference de Carvalho, Dengler and Eisen8 and non-ALS patients who exhibited symptoms resembling ALS were recruited from March 2017 to May 2021. Non-ALS patients included those mimic disorders of ALS, such as peripheral neuropathy (PN) including multifocal motor neuropathy (MMN), cervical spondylosis or lumbar spondylosis, and myopathy. All enrolled patients were recorded with their name, gender, age, body mass index (BMI), disease duration, region of onset, and clinical symptoms with detailed physical examination. ALS patients were assessed using the ALS Functional Rating Scale-Revised (ALSFRS-R). Reference Cedarbaum, Stambler and Malta16 Muscle strength was measured using the Medical Research Council (MRC) score, including bilateral assessment of the following limb muscle actions: shoulder abduction, elbow flexion, elbow extension, wrist flexion, wrist extension, finger flexion, finger extension, thumb abduction, little finger abduction, hip flexion, knee flexion, knee extension, ankle dorsal extension, ankle plantar flexion, toe dorsal extension, and toe plantar flexion. The total MRC score was 160.
This study was approved by the Ethics Committee of the Peking Union Medical College Hospital (PUMCH) (JS1210). All enrolled patients provided written informed consent to be included in the study.
Ultrasound Study
MUS examination was performed using an 8–12 MHz linear array transducer (LOGIQ e; General Electric company, Wuxi, China). The initial settings were kept constant during all examinations. The gain was set to automatic mode, and the depth and focus were adjusted depending on the muscle and individual patient variations. The patients were asked to relax for a minimum of 30 min before the MUS examination was initiated.
Four major pairs of antagonistic muscles/muscle groups including elbow flexors-extensors (biceps brachii and triceps brachii), wrist flexors-extensors (flexors and extensors of the forearm), knee flexors-extensors (biceps femoris and quadriceps femoris), and ankle flexors- extensors (gastrocnemius and tibialis anterior) were selected for fasciculation detection. Muscles in bilateral limbs were tested. Each muscle was imaged transversely using the B-mode. The transducer was adjusted to be perpendicular to the belly of the muscle groups, which also was the standard insertion site for the needle used for EMG assessment. This specific orientation allowed the maximal cross-sectional image of the muscles. The transducer was held in the same position for 60 s. The presence of fasciculation was recorded for each muscle group. The whole process was recorded in videos for all muscles tested. The patient kept the muscles relaxed and silent during the MUS examination.
The intensity of fasciculation was divided into five grades (score 0–4) based on the firing frequency and site number in the specific muscle group involved in each assessment Reference Liu, Li and Niu12 (Table 1).
The criteria for fasciculation grade
Outcome indicators
According to previous studies, Reference Ludolph, Emilian and Dreyhaupt7,Reference Liu, Wang, Shen, Yang, Liu and Cui17 the split phenomenon of muscle strength was defined as that elbow flexor was stronger than extensor, wrist flexor stronger than extensor, knee extensor stronger than flexor, and ankle flexor stronger than extensor. Similarly, we defined the split phenomenon of fasciculation as the fasciculation intensity of elbow/wrist/ankle flexor higher than that of extensor, while the fasciculation intensity of knee extensor higher than that of flexor. We further defined the anti-split phenomenon as the fasciculation intensity of elbow/wrist/ankle extensor higher than that of flexor, while the fasciculation intensity of knee flexor higher than that of extensor.
We found the intensity of fasciculation in MUS was significantly influenced by the muscle strength, and it has been reported that FP intensity displayed a significant negative relation with the muscle strength among ALS patients. Reference Arts, van Rooij and Overeem13 Progression rate has also been reported to correlate with the fasciculation. Reference Tsugawa, Dharmadasa, Ma, Huynh, Vucic and Kiernan15 In this study, ALSFRS-R progression rate (48-ALSFRS-R score/duration) was used to represent the disease progression. Reference Labra, Menon, Byth, Morrison and Vucic18 In some patients with ALS, only LMN was involved in the early course of the disease. The measurement and analysis of fasciculation in these patients might be helpful to explain the origin of ALS fasciculation. Therefore, subgroup analyses based on muscle strength, progression rate, and clinical subtypes [patients with symptoms of UMN and LMN involvement/pure LMN involvement, UMN + LMN/LMN] were conducted.
Statistical analysis
The Shapiro–Wilk test was used to assess whether demographic data exhibited a normal distribution. Normally distributed variables including age were expressed as means (standard deviation, SD) while nonnormally distributed variables, including BMI, disease duration, total MRC score, and ALSFRS-R score, were expressed as medians (interquartile range, IQR), and comparisons between ALS and non-ALS patients were assessed using the Mann–Whitney U test. The χ2 test was used to assess comparisons between the frequency of categorical variables. Paired t test was used for comparison in the fasciculation intensity of the pair of muscle groups. Two-sided P-values were calculated for all analyses. A value of P < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS, version 23.0.
Result
Clinical Characteristics
A total of 206 participants were recruited, including 140 ALS patients and 66 non-ALS patients. All ALS patients were followed for at least 6 months and were diagnosed with probable or definite ALS according to the Awaji criteria. Reference de Carvalho, Dengler and Eisen8 The non-ALS group included 40 PN patients including 9 with MMN, 18 cervical spondylosis or lumbar spondylosis patients, and 7 myopathy patients. The clinical characteristics of the patients were shown in Table 2. Significant differences in age (P = 0.01) and BMI (P = 0.01) were observed between ALS and non-ALS patients.
Clinical characteristics of included patients
ALS = amyotrophic lateral sclerosis; BMI = body mass index; MRC = Medical Research Council; ALSFRS-R = ALS Functional Rating Scale-Revised; UMN = upper motor neuron; LMN = lower motor neuron.
P-values with significant differences are listed in bold type.
Differences in Muscle Strength and Fasciculation Intensity Between Antagonistic Muscles
For each pair of antagonistic muscles, the dissociation in muscle strength and fasciculation intensity was presented in Figure 2. The frequency of split phenomenon of fasciculation intensity was significantly higher than those of muscle strength, (26.1% vs. 7.1% for elbow flexor-extensor, 38.3% vs. 5.7% for wrist flexor-extensor, 37.9% vs. 3.0% for knee extensor-flexor, and 33.6% vs. 14.4% for ankle flexor-extensor) (P < 0.01).
Comparison of muscle strength and fasciculation intensity between antagonistic muscles in ALS patients.
Analysis of mean (±SD) muscle strength showed that elbow flexors, wrist flexors, and ankle flexors were significantly stronger than extensors (elbow flexors/extensors: 3.93 ± 1.21/3.89 ± 1.24 P < 0.05; wrist flexors/extensors: 4.06 ± 1.15/4.01 ± 1.15 P < 0.05; ankle flexors/extensors 4.38 ± 1.12/4.14 ± 1.45 P < 0.05), while there was no significant difference between knee extensors and flexors (4.32 ± 1.01/4.38 ± 0.95 P = 0.06). For fasciculation intensity, our results showed significant split phenomena in all antagonistic muscles: the mean (±SD) fasciculation intensity of elbow flexors/extensors 2.13 ± 1.52/1.92 ± 1.56 (P < 0.01), wrist flexors/extensors 2.10 ± 1.53/1.74 ± 1.53 (P < 0.01), knee flexors/extensors 1.55 ± 1.42/1.88 ± 1.57 (P = 0.02), and ankle flexors 1.51 ± 1.22/1.33 ± 1.29 (P < 0.01) were significantly higher than that of extensors (P < 0.01).
As for non-ALS patients, no significant difference was observed in fasciculation intensity between antagonistic muscles (Supplementary Table 2).
Subgroup fasciculation analysis
Muscle strength
We divided the antagonistic muscles into three subgroups (MRC 0–1, MRC 2–4, and MRC 5) according to the muscle strength of the weaker ones. The results of fasciculation analysis were presented in Figure 3. For muscles with MRC 0-1 level, the fasciculation intensity of ankle flexor was significantly higher than that of ankle extensor (P = 0.02), while there was insignificant difference between other antagonistic muscles. For muscles with MRC 2–4, the fasciculation intensity of wrist flexor (P < 0.01) and ankle flexor (P < 0.01) was significantly higher than that of wrist extensor and ankle extensor, while the fasciculation intensity of knee extensors was higher than that of knee flexors (P < 0.01). For muscle groups with MRC 5, there were marked differences in mean split fasciculation intensity between elbow extensor and flexor (P = 0.02), wrist extensor and flexor (P < 0.01), and knee extensor and flexor (P < 0.01). Therefore, the split phenomenon of fasciculation intensity was common in antagonistic muscles with MRC 2–5 level of muscle strength.
Comparison of fasciculation intensity between antagonistic muscles with different muscle strength.
Progression rate
We defined ALSFRS-R progression rate ≥ 1 as the rapid progression group. For patients with rapid progression rate, the fasciculation intensity of wrist flexor was significantly higher than that of wrist extensor (P < 0.01), as presented in Figure 4. For patients with slow progression rate, all four pairs of antagonistic muscles showed obviously split phenomena in fasciculation intensity (elbow flexor > extensor, P < 0.01; wrist flexor > extensor, P < 0.01; knee extensor > flexor, P < 0.01; ankle flexor > extensor, P = 0.03). Therefore, the split phenomenon of fasciculation intensity was more common in patients with slow progression rate.
Comparison of fasciculation intensity between antagonistic muscles in ALS patients with different progression rate.
To eliminate the influence of muscle strength, we further subdivided the patients with different progression rate into three subgroups according to the muscle strength, and the results was showed in Supplementary Table 1. In slow progression group, significant dissociation of fasciculation intensity was observed in muscles with MRC 2–4 level (wrist flexor > extensor, P < 0.05; knee extensor > flexor, P < 0.0001; ankle flexor > extensor, P < 0.01) and MRC 5 level (wrist flexor > extensor, P < 0.01; knee extensor > flexor, P < 0.05). In patients with rapid progression rate, the fasciculation intensity of wrist flexor was significantly higher than that of wrist extensor (P < 0.05) at the MRC 5 level.
Clinical subtypes
For patients with pure LMN involvement, the fasciculation intensity of wrist flexor was higher than that of wrist extensor (P = 0.01), so was knee extensor higher than knee flexor (P = 0.01). For patients with UMN and LMN involvement, the split phenomena of fasciculation intensity were observed between elbow flexor-extensor (P < 0.01), wrist flexor-extensor (P < 0.01), knee extensor-flexor (P < 0.01), and ankle flexor-extensor (P = 0.04) (Figure 5). Therefore, the split phenomenon of fasciculation intensity was more common in patients with both UMN and LMN involvement.
Comparison of fasciculation intensity between antagonistic muscles in ALS patients with different subtypes.
Further hierarchical analysis based on muscle strength showed that there was significant differences in fasciculation intensity only between wrist flexor and extensor (P < 0.05) with MRC 5 level in LMN subgroup. In UMN and LMN subgroup, there were significant difference in fasciculation intensity between wrist flexor-extensor (P < 0.05), knee extensor-flexor (P < 0.05), and ankle flexor-extensor (P < 0.001) with MRC 2–4 group, between elbow flexor-extensor (P < 0.05), wrist flexor-extensor (P < 0.01), and knee extensor-flexor (P < 0.05) at MRC 5 group (Supplementary Table 1).
Discussion
In the study, we compared the muscle strength and fasciculation intensity of four pairs of antagonistic muscles in totally 140 ALS patients. Our results showed that the muscle strength of antagonistic muscles in ALS population tended to dissociate, which was consistent with prior studies. Reference Ludolph, Emilian and Dreyhaupt7,Reference Liu, Wang, Shen, Yang, Liu and Cui17,Reference Simon, Lee and Bae19 In non-ALS patients, similar split phenomenon was not prominent. However, more than 85% pair of antagonistic muscles did not present split phenomenon of muscle strength, which might be due to the fact that our subjects were examined at the early stage of the disease, and the mean MRC scores of tested muscles were around 4 level.
From Figure 2, we could see that the frequency of split phenomenon of fasciculation intensity was significantly higher than that of muscle strength, indicating the potential role of split fasciculation in diagnosis and differential diagnosis of ALS. The occurrence of pathogenic fasciculation in ALS patients was considered as the subclinical evidence of motor neuron overexcitation, which was earlier than the decline of muscle strength. Reference Eisen and Vucic20–Reference Juan, Fang and Qi22 With the progression of the disease, the split phenomenon of fasciculation might gradually present firstly, then the split of muscle strength. Follow-up data were needed to expound the dynamic changes of split phenomena of fasciculation and muscle strength.
Fasciculation was more likely to be detected in muscles with relatively preserved muscle strength, especially for the high-grade one. Reference Bokuda, Shimizu and Kimura21,Reference Avidan, Fainmesser, Drory, Bril and Abraham23 As the disease progressed, the muscle strength of ALS patients presented a continuous decline. Thus, the muscle strength could partly reflect the clinical stage of the disease course. Reference Abdul Aziz, Toh and Loh24 Subgroup analysis based on muscle strength revealed that the mean fasciculation intensity of muscles with MRC 0–1 level was relatively lower than those with MRC 2–4 level. Also, the significant difference in mean fasciculation intensity was detected only between ankle flexor and ankle extensor with MRC 0–1 level, while for antagonistic muscles with MRC 2–4 level, all four pairs showed significant differences in mean fasciculation intensity.
For patients with rapid progression, the difference in fasciculation intensity was detected only between wrist flexor and extensor, while for those with slow progression, there were significant differences in fasciculation intensity between all antagonistic muscles. The further subgroups analysis based on muscle strength revealed similar results. The pathogenesis of progression rate on split phenomena needed to be further studied. Rapid progression reflected the more severe degeneration of LMNs. Patch clamp studies showed that among SOD1-deficit mice, the excitability of LMNs increased before the onset of ALS and decreased with the development of the disease. Reference Tsugawa, Dharmadasa, Ma, Huynh, Vucic and Kiernan15 In patients with rapid progression, the excitability of motor neurons might decline faster so that it might be difficult to show the split phenomena. Besides, our results showed that the mean fasciculation intensity of rapid progression group was relatively higher than that of slow progression group. The easily presented high grade of fasciculation in both flexors and extensors might partly cover the split phenomenon, and more detailed grading system of fasciculation was needed.
Stratified analysis on clinical subtypes showed that the frequency of split phenomenon of fasciculation was higher in patients with both UMN and LMN involvement than those with only LMN involvement. Also, the mean fasciculation intensity of the former was higher than that of the latter. Although the origin of fasciculation remained a source of debate, increasing studies has proven that both UMN and LMN played a role in the generation of fasciculation. Reference de Carvalho and Swash9,Reference de Carvalho and Swash25,Reference Kleine, Stegeman, Schelhaas and Zwarts26 The functional status of the LMN was the basis of the fasciculation, especially near the soma of motor neurons, which was always believed to be the origin of these random flickering. Reference Roth27,Reference de Carvalho, Turkman, Pinto and Swash28 Consistent with our results, muscles groups with MRC 0–1 level of muscle strength showed no split phenomenon of fasciculation, regardless of the progression rate and clinical subtypes. When the disease progressed to the latest stage, the loss of LMNs has been very severe, Reference Cleveland, Bruijn and Wong29 which disabled the appearance of high grade and split phenomenon of fasciculation. The influence of UMN might play a vital role in the split phenomenon of fasciculation in ALS. The simulation of UMN could facilitate the generation of fasciculation and caused increase of fasciculation intensity. Reference de Carvalho, Miranda, Lourdes Sales Luís and Ducla-Soares30,Reference Mills31 It has been reported that the differences in density of direct CM innervation might be the main cause of the specific pattern of paresis in ALS patients. Reference Eisen and Weber32–Reference Eisen, Turner and Lemon35 Just like the post-stroke patients, the damage of pyramidal tract caused the weakness of elbow extensors, wrist extensors, ankle extensors, and knee flexors with relative sparing of their antagonistic muscles.
We also noticed the frequency of anti-split phenomenon of fasciculation intensity was higher than that of muscle strength. More work should be done to explicit the potential pathogenesis of anti-split phenomenon and its relationship with split phenomenon of fasciculation intensity. We hypothesized that the intensity of fasciculation might be affected by the different degree of involvement of upper and LMN exciting and degeneration process and the complicated interactive between upper and LMN during the course of the disease. A follow-up study on the changes of fasciculation could give more information.
To the best of our knowledge, this is the first study on split phenomenon of fasciculation among ALS patients. Through MUS, we elucidated the dissociation of fasciculation intensity in pairs of antagonistic muscle groups among the ALS population and found out the possibly influential factors by subgroup analysis. However, this study has several limitations. First, we only analyzed the fasciculation intensity at the initial visit and did not carry out longitudinal observations of the changes in the split phenomena that occurred during the disease progression. Secondly, due to the limitation of the included patients, we could not conduct the multilevel stratified analysis or multiple-factor analysis, for example, analyzing the dissociation of fasciculation intensity in ALS patients with slow progression and symptoms of UMN and LMN involvement. Besides, the stratification of population was almost experimental, which needed future studies to confirmed.
In conclusion, split phenomenon of fasciculation between antagonistic muscles was not uncommon in ALS patients. Muscle strength, progression rate, and UMN involvement were influence factors for this split phenomenon. Our results suggested that in the early stages of ALS, when the muscle strength was relatively preserved, the fasciculation of antagonistic muscles tended to present split phenomenon. We hypothesized that the occurrence of the split phenomenon needed relative preservation of the LMN function and mainly caused by the differences in density of CM projections. As the disease progressed, the loss of LMN gradually increased, and the split phenomenon of fasciculation would finally disappear.
Supplementary Material
To view supplementary material for this article, please visit https://doi.org/10.1017/cjn.2023.62
Acknowledgments
The authors would like to thank all the authors of the original articles and Dr Zongmuyu Zhang (Peking Union Medical College Hospital) for guidance of data analysis.
Funding
CAMS Innovation Fund for Medical Sciences (CIFMS 2021-I2M-1-003) and Beijing Natural Science Foundation (7202158).
Conflict of Interest
No conflict declared.
Statement of Authorship
Nan Hu (First author) conducts the data analysis and writes of the main manuscript; Yi Li and Jingwen Liu are contributed to data collection; Liying Cui (Senior author) and Mingsheng Liu (Corresponding author) lead all authors contributing to all aspects of the development, evidence and data analysis, writing, editing, and final approval of this manuscript. All authors have read and approved the manuscript.
