Thenar muscle atrophy secondary to palsy of the motor branch of the median nerve is often observed in patients with severe carpal tunnel syndrome (CTS). In clinical practice, tip pinch strength and electrophysiological assessment are commonly performed for evaluating the disability of the thenar muscle. However, such examinations are not entirely effective for assessing the palsy of the motor branch of the median nerve and the recovery of the thenar muscle.1–4 Anatomic studies regarding the motor nerve supply of the thenar muscle have shown that, although the abductor pollicis brevis (APB) is almost always innervated only with the median nerve, the flexor pollicis brevis (FPB) and opponens pollicis may have not only either single innervation with the median nerve or single innervation with the ulnar nerve but also dual innervation with the median and ulnar nerves.5–8 This variability in muscle innervation may explain why some patients with severe CTS and severe thenar muscle atrophy maintain pinch function and strength.2–4,9 For these reasons, pinch strength may be inadequate for assessing the palsy of the motor branch of the median nerve.
According to the studies of the motor nerve supply of the thenar muscle, the APB is the best target muscle for assessment of palsy of the motor branch of the median nerve. The compound muscle action potential (CMAP) measured from the APB represents the most suitable electrophysiological assessment and has become a popular tool for diagnosing CTS. However, CMAP cannot directly reflect the disability of thumb motion caused by the atrophy of the thenar muscles. Moreover, the examination is painful for some patients, who typically hesitate to undergo repeat examinations. Because the APB acts as the primary palmar abductor of the thumb,10,11 the authors hypothesized that the angle of thumb palmar abduction may serve as a good indicator of APB function and palsy of the motor branch of the median nerve.
Thumb palmar abduction relies on the thumb trapeziometacarpal joint. Several kinematics studies have described the range of motion of the trapeziometacarpal joint,12–16 but such studies typically used subjective axes and planes for range of motion assessment and employed special devices, such as the surface registration system and video capture, to track bone motion, making it difficult to adopt such approaches in daily clinical practice. In outpatient settings, it is important to be able to measure thumb active palmar abduction both easily and accurately. In this study, the authors address both of these challenges by developing a suitable measurement protocol and determining the normal maximum angle of thumb active palmar abduction in healthy adults.
Materials and Methods
This study was approved by the ethics committee at the authors' institution and informed consent was obtained for each participant. Twenty-five women (20 to 21 years old; all right-handed) with no disability in the hand were enrolled. The authors chose young participants because they expected their joints would have less osteoarthritis, which would facilitate determining the maximum angle of active thumb palmar abduction in healthy normal subjects.
Three methods for measuring active thumb palmar abduction were compared. All measurements were performed twice. Each method was assessed in terms of inter- and intraobserver reliability.
In the Japanese Orthopaedic Association (JOA) method, active palmar abduction of the thumb was defined according to the JOA as the angle between the first distal phalanx and the second metacarpal and was measured while subjects maximally abducted their thumbs perpendicularly to the palm (Figure 1A). The goniometer-based measurements were performed by 2 experienced hand therapists (K.S., N.K.). Within the JOA method, each hand was measured 4 times (twice by each observer).
Photograph showing the angle between the first distal phalanx and the second metacarpal, as measured by the Japanese Orthopaedic Association method (A). Photograph (B) and lateral radiograph (C) of a wrist and a hand showing the angle between the first and the second metacarpal, as measured by the American Society of Hand Therapists method and the radiographic method, respectively.
In the American Society of Hand Therapists (ASHT) method,17 active palmar abduction of the thumb was defined by the ASHT as the angle between the first and second metacarpal and was measured while subjects held their hands in the same position described in the JOA method (Figure 1B). The goniometer-based measurements were performed by the 2 experienced hand therapists, and each hand was measured 4 times (twice by each observer).
In the radiographic method, active palmar abduction of the thumb was defined as the angle between the first and second metacarpal and was measured using a hand radiograph (Figure 1C). Specifically, the radiograph was obtained with the subject keeping the bilateral thumbs maximally abducted to the palmar direction. The following protocol was applied. To guarantee that the thumb was abducted perpendicularly to the palm, the authors asked the participants to hold a box, with the forearms and palms straight along the side of the box, and to abduct the thumbs while maintaining contact with the upper side of the box (Figure 2A). The authors did not restrict the rotation of the thumb nail.
Photograph showing the position of the box and the subject's thumbs (A). Lateral radiograph of the bilateral wrists and hands showing the hand position during the first measurement (B). Anteroposterior radiograph of the bilateral hands showing the transition position (C). Lateral radiograph showing the position during the second measurement (D).
To evaluate intrasubject reliability, radiographs were performed 3 times as follows. The first radiograph was obtained once the subject had abducted the thumb as described above (ie, using the box). Next, a frontal radiograph of the hand was obtained in the absence of the box. Finally, the box was used again to achieve appropriate thumb abduction, and a second radiograph was obtained (Figure 2B–D). The measurements were performed by 2 experienced orthopedic surgeons (I.T., S.S.).
One-way analysis of variance was used to evaluate the difference in the measurements provided by the 3 methods, with P<.05 indicating statistical significance. Intra- and interobserver reliability were assessed by means of one- and two-way intraclass correlation coefficients (ICCs); those greater than 0.7 were considered to indicate strong reliability. To evaluate the intrasubject reliability, the paired t test and ICC were used. Statistical software was used for analysis.
The measurement results reported by the 2 experienced hand therapists for the JOA and ASHT methods are summarized in Table 1, whereas those reported by the 2 experienced orthopedic surgeons for the radiographic method are summarized in Table 2. The authors used the paired t test to assess the agreement between the means of the first and second measurements performed by the same observer and found significant differences only regarding the observations reported by one of the experienced orthopedic surgeons (I.T.) for left hands (first radiograph, P=.02; second radiograph, P=.03). Nevertheless, because the absolute differences between the first and second radiograph measurements were small (<2°; Table 2), the authors considered that these statistically significant differences will not be relevant in clinical practice.
Measurements Obtained via the JOA and ASHT Methods
Measurement Results Obtained via the Radiographic Method
Regarding the JOA method, the ICCs for intraobserver reliability were 0.95 for one of the experienced hand therapists (K.S.) and 0.91 for the other experienced hand therapist (N.K.) (Table 3), indicating strong reliability. The interobserver reliability of the JOA method was also strong, with the ICC being greater than 0.7 for both the first and second measurements (Table 4).
Intraobserver Reliability of the 3 Methods Tested
Interobserver Reliability of the 3 Methods Tested
Regarding the ASHT method, the ICCs for intraobserver reliability were 0.93 for one of the experienced hand therapists (K.S.) and 0.92 for the other experienced hand therapist (N.K.) (Table 3), indicating strong reliability. However, the ICCs for interobserver reliability were less than 0.7 for both measurements (first measurement, ICC=0.56; second measurement, ICC=0.48), indicating moderate interobserver reliability for the ASHT method (Table 4).
Regarding the radiographic method, the ICCs for intraobserver reliability were greater than 0.7 for both observers and for both radiographs (experienced orthopedic surgeon I.T., ICC=0.94 for both radiographs; experienced orthopedic surgeon S.S., ICC=0.86 and ICC=0.87 for the first and second radiograph, respectively; Table 3). Therefore, the intraobserver reliability of the radiographic method was at least as strong as those of the JOA and ASHT methods. The interobserver reliability of the radiographic method was also strong, with ICCs greater than 0.7 in all pairs of measurements evaluated (Table 4).
The authors used one-way analysis of variance to compare the measurement results provided by the 3 methods. For each method, they considered as representative the left-hand data reported by the first observer in the first measurement (ie, by one of the experienced hand therapists [K.S.] for the JOA and ASHT methods, and by one of the experienced orthopedic surgeons [I.T.] in the first radiograph for the radiographic method). The values obtained via the JOA method were significantly larger than those obtained via the ASHT and radiographic methods (P<.001 for the JOA method vs the ASHT method and for the JOA method vs the radiographic method; Figure 3).
The measurement results obtained via the 3 methods tested. The left hand data collected by the first observer (I.T.) at the first measurement were considered representative of each method. One-way analysis of variance was used for the statistical analysis.
To assess the reproducibility of the subject's motion during the measurements involved in the radiographic method, the authors used the paired t test and ICC to quantify intra-subject agreement and reliability, respectively. For each radiograph, the authors considered as representative the left-hand data collected by one of the experienced orthopedic surgeon [I.T.] in the first measurement. No statistically significant difference was found between the means of the first and the second radiographs performed in the same subject (P=.16; Figure 4). Intra-subject reliability was strong, with an ICC of 0.90 (95% confidence interval, 0.79–0.95). Therefore, the authors concluded that reproducibility of the subject's motion when performing measurements via the radiographic method was good.
Intrasubject agreement of measurements performed in the radiographic method. The left hand data collected by the first observer (I.T.) at the first and second measurements were considered representative. The paired t test was used for statistical analysis. There were no significant differences.
Finally, the authors used the paired t test to compare the measurement results reported for the left and right hands on the first radiograph analyzed as part of the radiographic method. They found significant differences only for data reported by one of the experienced orthopedic surgeon [I.T.], in which the left hand was larger than the right hand.
In their clinical practice involving patients with severe CTS, the authors often wished to evaluate not only the sensory disability, but also the disability of thumb motion. Pinch strength is commonly used for this purpose. However, it is debatable whether tip pinch strength can really reflect thumb disability in patients with CTS.
As described above, APB, which is the prime palmar abductor of the thumb, is the best target of concern. During palmar abduction of the thumb, the 3 muscles of the thenar (ie, APB, FPB, and opponens) cooperate to keep the first metacarpophalangeal joint away from the palm. On the other hand, during tip pinch, these muscles must cooperate to bring the first metacarpophalangeal joint close to the palm (Figure 5). Although the 3 thenar muscles arise mainly from the flexor retinaculum, the APB arises from the most proximal of the flexor retinaculum.18 Therefore, it seems reasonable to assume that the APB is activated in palmar abduction more than tip pinch action. In this context, tip pinch strength is not appropriate for evaluating the disability of thumb motion in patients with CTS.
Photographs showing the distinction between tip pinch (A) and thumb palmar abduction (B) in terms of the direction of the thumb. The red arrow indicates the direction of the contracture of the thenar muscles.
The JOA defined the angle between the first distal phalanx and the second metacarpal as the angle of palmar abduction of the first carpometacarpal joint, ranging from 0° to 90°. This definition was adopted as the JOA method (Table 1), which had strong intra- and interobserver reliability (Tables 3–4). The maximum abduction angle (given as the mean of 4 measurements in each hand) was 96.1°±9.9° and 90.9°±10.7° for the left and right hand, respectively. This angle included the metacarpophalangeal joint and interphalangeal joint, in addition to the trapeziometacarpal joint. Therefore, the values obtained via the JOA method were significantly larger (almost double) than those obtained via the ASHT and radiographic methods (Figure 3). The maximum range of the trapeziometacarpal joint seems to be below 90° considering the bone morphology of this joint. Therefore, despite the strong intra- and interobserver reliability, the authors did not consider the JOA method as the appropriate measuring method for the palmar abduction angle of the trapeziometacarpal joint.
The ASHT defined the angle between the first and the second metacarpal as the angle of palmar abduction of the first carpometacarpal joint.17 This definition was adopted as the ASHT method (Table 1). In the ASHT method, the values (expressed as the mean of 4 measurements in each hand) were 48.0°±6.1° and 48.1°±5.8° for the left and right hand, respectively. The ASHT method had strong intraobserver reliability (ICC >0.7) but moderate interobserver reliability (ICC <0.7; Tables 3–4). Therefore, the authors considered the ASHT method inappropriate.
In the radiographic method, the angles between the first and second metacarpal were measured on hand radiographs. The authors used a box to ensure that thumb abduction was perpendicular to the palm. The values (expressed as the mean of 8 measurements in each hand) were 45.3°±6.4° and 44.0°±7.0° for the left and right hand, respectively. To evaluate intraobserver agreement, the authors used the paired t test to compare the data from the first and second measurements performed by the same observer on the same radiograph. They found statistically significant differences regarding data measured on both radiographs by one of the experienced orthopedic surgeons (I.T.), but these differences were small (within 2°) and unlikely to be clinically relevant (Table 2). Moreover, the radiographic method had strong intra- and interobserver reliability (Tables 3–4). Finally, the reproducibility of subject posture during these measurements was good (Figure 4). These findings suggest that, among the 3 methods tested, the radiographic method is the most suitable for measuring the palmar abduction angle of the trapeziometacarpal joint.
The current authors refer to the case of a 63-year-old woman with severe CTS in the right hand who underwent the modified Camitz opponensplasty. The authors found that the recovery of palmar abduction of the right thumb was easily quantifiable between the pre- and postoperative radiographs when applying the radiographic method (Figure 6).
Lateral preoperative (A) and postoperative (B) radiographs of the bilateral wrists and hands of a 63-year-old woman showing severe carpal tunnel syndrome in the right hand evaluated by the radiographic method. The recovery of palmar abduction of the right thumb was observed convincingly.
The orientation of the trapezium is oblique to the plane of the hand in 3 directions: abduction-adduction, flexion-extension, and pronation-supination. Cooney et al12 reported that the trapezius was oriented in 38° of abduction, 48° of flexion, and 81° of pronation with respect to the coronal plane of the hand. Therefore, it is difficult to standardize the axis of motion of the trapeziometacarpal joint.
Most studies evaluating the kinetics of the trapeziometacarpal joint12–16 used special devices to define the coordination and measurement angles, and there is substantial discrepancy in the definition of the axis of this particular joint. In the current study, the palmar abduction angles were not measured with respect to a single axis of motion of the trapeziometacarpal joint, but rather reflected the motion about all 3 axes of this joint. Although this inconsistency with the axes of the trapeziometacarpal joint represents a limitation of the current study, the authors believe it represents an advantage in clinical practice, where it is not feasible to use special devices and measurement setups. They did not intend to measure the movement about the proper axis of the trapeziometacarpal joint because their aims were to determine the maximum angles of active palmar abduction of healthy thumbs and to develop an approach that can be immediately adopted in clinical practice. Therefore, the rotation of the thumb nail was not fixed while taking the radiographs involved in the radiographic method.
Bamberger et al19 and Dormitorio et al20 also measured the palmar abduction angle of the trapeziometacarpal joint with normal participants in a similar manner without the box, such as in the radiographic method, and reported 32° and 41°±5°, respectively. However, those studies did not report the inter- and intraobserver reliability of those methods.
Another limitation is the exposure to radiation during radiographs of the hand. It amounts to approximately 0.08 mGy, which is half of the exposure experienced during a chest radiograph. On the other hand, the authors can evaluate and compare radiographs in future clinical practice, which is the current practice for evaluating the Cobb angle in patients with scoliosis. The cost of the radiograph of the hand may also be a demerit of the radiographic method.
The authors have established an approach to measure the active palmar abduction angle of the thumb that can be immediately adopted in clinical practice for patients with CTS. With this measurement, the maximum active palmar abduction angle of the thumb in healthy young women was 45.3°±6.4° and 44°±7° for the left hand and the right hand, respectively.
The proposed measurement approach and the reference ranges for the maximum angle of active palmar abduction of the thumb can be used not only for patients with CTS but also for patients with osteoarthritis of the thumb carpometacarpal joint.
- Geere J, Chester R, Kale S, Jerosch-Herold C. Power grip, pinch grip, manual muscle testing or thenar atrophy: which should be assessed as a motor outcome after carpal tunnel decompression? A systematic review. BMC Musculoskelet Disord. 2007;8(1):114. https://doi.org/10.1186/1471-2474-8-114 PMID: doi:10.1186/1471-2474-8-114 [CrossRef]
- Jerosch-Herold C, Shepstone L, Miller L, Chapman P. The responsiveness of sensibility and strength tests in patients undergoing carpal tunnel decompression. BMC Musculoskelet Disord. 2011;12(1):244. https://doi.org/10.1186/1471-2474-12-244 PMID: doi:10.1186/1471-2474-12-244 [CrossRef]22032626
- Li ZM, Harkness DA, Goitz RJ. Thumb strength affected by carpal tunnel syndrome. Clin Orthop Relat Res. 2005;441(441):320–326. https://doi.org/10.1097/01.blo.0000181143.93681.0d PMID: doi:10.1097/01.blo.0000181143.93681.0d [CrossRef]16331021
- Chen PT, Jou IM, Lin CJ, et al. Is the control of applied digital forces during natural five-digit grasping affected by carpal tunnel syndrome?Clin Orthop Relat Res.2015;473(7):2371–2382. https://doi.org/10.1007/s11999-015-4189-x PMID: doi:10.1007/s11999-015-4189-x [CrossRef]25690168
- Ajmani ML. Variations in the motor nerve supply of the thenar and hypothenar muscles of the hand. J Anat. 1996;189(pt 1):145–150. PMID:8771405
- Olave E, Prates JC, Del Sol M, Sarmento A, Gabrielli C. Distribution patterns of the muscular branch of the median nerve in the thenar region. J Anat. 1995;186(pt 2):441–446. PMID:7649846
- Rowntree T. Anomalous innervation of the hand muscles. J Bone Joint Surg Br. 1949; 31B(4):505–510. https://doi.org/10.1302/0301-620X.31B4.505 PMID: doi:10.1302/0301-620X.31B4.505 [CrossRef]15397131
- Xie P, Jiang Y, Zhang X, Yang S. The study of intramuscular nerve distribution patterns and relative spindle abundance of the thenar and hypothenar muscles in human hand. PLoS One. 2012;7(12):e51538. https://doi.org/10.1371/journal.pone.0051538 PMID: doi:10.1371/journal.pone.0051538 [CrossRef]23251569
- Marquardt TL, Nataraj R, Evans PJ, Seitz WH Jr, Li ZM. Carpal tunnel syndrome impairs thumb opposition and circumduction motion. Clin Orthop Relat Res.2014;472(8):2526–2533. https://doi.org/10.1007/s11999-014-3651-5 PMID: doi:10.1007/s11999-014-3651-5 [CrossRef]24777728
- Cooney WP, Linscheid RL, An KN. Opposition of the thumb: an anatomic and biomechanical study of tendon transfers. J Hand Surg Am. 1984;9(6):777–786. https://doi.org/10.1016/S0363-5023(84)80048-9 PMID: doi:10.1016/S0363-5023(84)80048-9 [CrossRef]6512185
- Richer RJ, Peimer CA. Flexor superficialis abductor transfer with carpal tunnel release for thenar palsy. J Hand Surg Am. 2005;30(3):506–512. https://doi.org/10.1016/j.jhsa.2004.12.010 PMID: doi:10.1016/j.jhsa.2004.12.010 [CrossRef]15925160
- Cooney WP III, Lucca MJ, Chao EY, Linscheid RL. The kinesiology of the thumb trapeziometacarpal joint. J Bone Joint Surg Am. 1981;63(9):1371–1381. https://doi.org/10.2106/00004623-198163090-00002 PMID: doi:10.2106/00004623-198163090-00002 [CrossRef]7320028
- Crisco JJ, Halilaj E, Moore DC, Patel T, Weiss AP, Ladd AL. In vivo kinematics of the trapeziometacarpal joint during thumb extension-flexion and abduction-adduction. J Hand Surg Am. 2015;40(2):289–296. https://doi.org/10.1016/j.jhsa.2014.10.062 PMID: doi:10.1016/j.jhsa.2014.10.062 [CrossRef]
- Gehrmann SV, Tang J, Li ZM, Goitz RJ, Windolf J, Kaufmann RA. Motion deficit of the thumb in CMC joint arthritis. J Hand Surg Am. 2010;35(9):1449–1453. https://doi.org/10.1016/j.jhsa.2010.05.026 PMID: doi:10.1016/j.jhsa.2010.05.026 [CrossRef]20807622
- Ladd AL, Crisco JJ, Hagert E, Rose J, Weiss AP. The 2014 ABJS Nicolas Andry Award. The puzzle of the thumb: mobility, stability, and demands in opposition. Clin Orthop Relat Res. 2014;472(12):3605–3622. https://doi.org/10.1007/s11999-014-3901-6 PMID: doi:10.1007/s11999-014-3901-6 [CrossRef]25171934
- Miura T, Ohe T, Masuko T. Comparative in vivo kinematic analysis of normal and osteoarthritic trapeziometacarpal joints. J Hand Surg Am. 2004;29(2):252–257. https://doi.org/10.1016/j.jhsa.2003.11.002 PMID: doi:10.1016/j.jhsa.2003.11.002 [CrossRef]15043898
- Gibson G. Goniometry. In: Clinical Assessment Recommendations. Mt Laurel, NJ: American Society of Hand Therapists; 2015:77–116.
- Standring S, Borley NR, Collins P, et al. Gray's Anatomy: The Anatomical Basis of Clinical Practice (ed. 50). New York, NY: Churchill Livingstone; 1989;882–884.
- Bamberger HB, Stern PJ, Kiefhaber TR, McDonough JJ, Cantor RM. Trapeziometacarpal joint arthrodesis: a functional evaluation. J Hand Surg Am. 1992;17(4):605–611. https://doi.org/10.1016/0363-5023(92)90302-6 PMID: doi:10.1016/0363-5023(92)90302-6 [CrossRef]1629538
- Dormitorio B, Hattori Y, Yukata K, Sakamoto S, Doi K. The use of dynamic radiographs in trapeziometacarpal joint arthrodesis for accurate range of motion evaluation. J Orthop Sci. 2018;23(1):75–80. https://doi.org/10.1016/j.jos.2017.09.022 PMID: doi:10.1016/j.jos.2017.09.022 [CrossRef]
Measurements Obtained via the JOA and ASHT Methodsa
|Parameter||JOA Method||ASHT Method|
|Left Side||Right Side||Left Side||Right Side|
|Measurement by hand therapist K.S.||1st||2nd||1st||2nd||1st||2nd||1st||2nd|
|Measurement by hand therapist N.K.||1st||2nd||1st||2nd||1st||2nd||1st||2nd|
Measurement Results Obtained via the Radiographic Methoda
|Parameter||First Radiograph||Second Radiograph|
|Left Side||Right Side||Left Side||Right Side|
|Measurement by orthopedic surgeon I.T.||1st||2nd||1st||2nd||1st||2nd||1st||2nd|
|Measurement by orthopedic surgeon S.S.||1st||2nd||1st||2nd||1st||2nd||1st||2nd|
Intraobserver Reliability of the 3 Methods Testeda
|Observer||ICC||95% CI||ICC||95% CI|
|JOA Method||ASHT Method|
|Hand therapist K.S.||0.95||0.92–0.97||0.93||0.87–0.96|
|Hand therapist N.K.||0.91||0.85–0.95||0.92||0.86–0.95|
|Radiographic Method (1st Radiograph)||Radiographic Method (2nd Radiograph)|
|Orthopedic surgeon I.T.||0.94||0.90–0.97||0.94||0.90–0.97|
|Orthopedic surgeon S.S.||0.86||0.76–0.91||0.87||0.78–0.92|
Interobserver Reliability of the 3 Methods Testeda
|Method||First Measurement||Second Measurement||Mean|
|ICC||95% CI||ICC||95% CI|
| 1st radiograph||0.83||0.71–0.90||0.88||0.80–0.93|
| 2nd radiograph||0.81||0.56–0.91||0.88||0.78–0.93||0.85b|