Innervation Patterns of Thumb Trapeziometacarpal Joint Ligaments

SCIENTIFIC ARTICLE Innervation Patterns of Thumb Trapeziometacarpal Joint Ligaments

Elisabet Hagert, MD, PhD, Julia Lee, MD, Amy L. Ladd, MD

Purpose The human thumb trapeziometacarpal (TM) joint is a unique articulation that allows stability during pinch and grip and great degrees of mobility. Because the saddle-shaped articulating surfaces of the TM joint are inherently unstable, joint congruity depends on the action of restraining ligaments and periarticular muscles. From other joints, it is known that proprioceptive and neuromuscular joint stability depend on afferent information from nerve endings within ligaments. We hypothesize that the TM joint ligaments may similarly be innervated, indicating a possible proprioceptive function of the joint. Methods We harvested 5 TM joint ligaments in entirety from 10 fresh-frozen cadaver hands with no or only minor signs of osteoarthritis and suture-marked them for proximal-distal orientation. The ligaments harvested were the dorsal radial, dorsal central, posterior oblique, ulnar collateral, and anterior oblique ligaments. After paraffin-sectioning, we stained the ligaments using a triple-antibody immunofluorescent technique and analyzed them using immunofluorescence microscopy. Results Using the triple-stain technique, mechanoreceptors could be classified as Pacinian corpuscles, Ruffini endings, or Golgi-like endings. The 3 dorsal ligaments had significantly more nerve endings than the 2 volar ligaments. Most of the nerve endings were close to the bony attachments and significantly closer (P ϭ .010) to the metacarpal insertion of each ligament. The anterior oblique ligament had little to no innervation in any of the specimens analyzed. Discussion The TM joint ligaments had an abundance of nerve endings in the dorsal ligaments but little to no innervation in the anterior oblique ligament. The Ruffini ending was the predominant mechanoreceptor type, with a greater density in the mobile metacarpal portion of each ligament. Clinical relevance Presence of mechanoreceptors in the dorsal TM joint ligaments infers a proprioceptive function of these ligaments in addition to their biomechanical importance in TM joint stability. (J Hand Surg 2012;37A:706–714. Copyright © 2012 by the American Society for Surgery of the Hand. All rights reserved.) Key words Innervation, ligaments, mechanoreceptors, proprioception, thumb.

From the Department of Clinical Science and Education, Karolinska Institutet, Stockholm, Sweden; and Foundation for purchase of equipment and support of E.H. and J.L., the OREF/RJOS/DePuy Career the Department of Orthopaedic Surgery, Robert A. Chase Hand and Upper Extremity Center, Stanford DevelopmentAwardforpurchaseofequipment,andtheHirschFoundation(KarolinskaInstitutet)for University,PaloAlto,CA. support of E.H. Received for publication October 12, 2011; accepted in revised form December 21, 2011. Corresponding author: Elisabet Hagert, MD, PhD, Department of Clinical Science and Education, No beneﬁts in any form have been received or will be received related directly or indirectly to the Hand and Foot Surgery Center, Karolinska Institutet, Storängsv.10, 11542 Stockholm, Swe- subject of this article. den; e-mail: [email protected]. TheauthorsthankAlexSox-Harris,PhD,forinvaluablehelpinthestatisticalanalysisofthemanuscript. 0363-5023/12/37A04-0012$36.00/0 doi:10.1016/j.jhsa.2011.12.038 This study was supported by a National Institute of Health (NIH) grant NIH/SBIR grant# 2 R44 EB003067-02A1 for the purchase of cadaver specimens, the Ronald and Ann Williams Charitable

HE ORIGIN OF THE HUMAN thumb trapeziometa- MATERIALS AND METHODS carpal (TM) joint with its complex saddle shape Specimens has been traced to the advent of terrestrial bi- T We dissected 10 fresh-frozen human cadaver hands for pedalism, when primates evolved from an arboreal an- this study: 3 female and 5 male (mean age, 65 y; range, imal with long fingers and redundant thumbs to a ter- 53–74 y; age unknown in 2 specimens). Using loupe restrial primate with more equal digital lengths and the 1,2 magnification, standard hand surgical instruments, and ability to perform thumb opposition and handle tools. meticulous dissection, we identified the nerve branches The human TM joint appears to be a morphological to the dorsal and volar TM joint capsule, as we did the compromise from that of our primate ancestors and is ligaments stabilizing the joint, which we then prepared. flat enough to allow for a high grade of mobility while Of the 10 specimens, 9 had macroscopic signs of mild being curved enough to support precise and powerful osteoarthritic (OA) changes involving 25% or less of 3 grips. The combination of mobility and strength places the proximal and distal carpometacarpal joint surfaces, unusually high demands on the stability of the TM joint. whereas 1 of the 10 samples showed no signs of OA This depends on a balance between compressive and according to the Outerbridge classification.17 The lig- tensile forces across the joint. Articular congruence aments harvested were the dorsal radial (DRL), permits stability in compression, whereas ligamentous dorsal central (DCL), posterior oblique (POL), an- restraint permits stability in tension.4 However, the terior oblique (AOL), and ulnar collateral (UCL) unique shape and looseness of the TM joint renders it (Fig. 1). We harvested each ligament at the insertion into inherently unstable in all positions, with the exception bone and suture-marked it for proximal-distal orientation. of the final screw-home torque seen in full opposition, In addition, we excised sections of the median nerve as Edmunds5 purported. As such, this joint is highly from 2 hands to use for control staining. dependent on the secondary and dynamic stability that The local institutional review board granted approval is created by periarticular muscles. for this project and the handling of human remains was Recent studies on the scapholunate ligament propose in strict accordance with ethical and practical protocols. that the dynamic neuromuscular stability of the scapholu- Slide preparation and immunohistochemistry nate joint depends on the innervation of mechanoreceptors and nerve endings in the joint capsule and We immediately fixed all harvested specimens in 4% ligaments,6,7 which transmit afferent information formaldehyde, placed them in paraffin wax, and sec- ␮ through the nerves innervating the wrist joint,8 tioned them into 8- m-thick slices before mounting resulting in alterations of wrist muscle function.9,10 them on glass slides. We deparaffinized specimens us- ϫ This suggests the intact proprioceptive function is ing serial xylene washes (3 3 min) followed by gradual rehydration. We used a microwave antigen important in maintaining stability of the scapholunate retrieval method to expose antibody-binding sites. joint. Numerous studies on the knee, shoulder, and Specimens were then made permeable with 1% Triton- ankle joints support similar findings, where an intact X100 and blocked with Image-iT FX (Invitrogen, joint innervation and preserved proprioceptive func- Carlsbad, CA). Afterward, we stained the specimen tion is essential for a normal joint homeostasis.11–14 samples with primary antibodies for 1 hour at 37°C in There are 2 studies describing the precise macro- ϫ 15,16 a humid chamber, rinsed them for 3 5 minutes in 0.1 scopic innervation of the TM joint but no publica- mol/L phosphate-buffered solution (PBS), and stained tions on the microscopic innervation and possible pres- them with secondary antibodies under the same condi- ence of mechanoreceptors in TM joint ligaments. Both tions. We executed a final 3 ϫ 5-minute rinse in 0.1 macroscopic studies describe numerous branches from mol/L PBS before using ProLong Gold Anti-Fade Re- the palmar cutaneous and thenar branches of the median agent with 4=,6=-diamidino-2-phenylindole (DAPI) (In- nerve, lateral cutaneous nerve of the forearm, and su- vitrogen) to mount the slides. perficial branch of the radial nerve terminating at or around the joint. Given the extensive innervation to the Antibodies TM joint, we hypothesized that the ligaments stabiliz- We used 2 antibodies in conjunction with DAPI to yield ing the joint may be innervated with mechanoreceptors a dual-color stain: p75 and PGP9.5. Rabbit anti-nerve and sensory nerve endings and that the pattern of in- growth factor (NGF) receptor 75 (p75) is an antibody nervation may add knowledge to the role of various directed against the low-affinity NGF receptor p75, also ligaments in the neuromuscular stability of the TM known as p75NTR or CD271. This transmembrane joint. protein is expressed on the cell membrane of nerve cells

JHS ᭜ Vol A, April  708 INNERVATION PATTERNS OF THUMB TMJ LIGAMENTS

FIGURE 1: A The 5 ligaments harvested from the dorsal aspect of the TM joint were the DRL, the DCL, and the POL. B Those harvested from the volar TM joint were the UCL and the AOL. MC1, first metacarpal; MC2, second metacarpal; APL, abductor pollicis longus. Asterisk indicates the dorsal tubercle of the trapezium (Tz). and responsible for signaling related to neuronal goat anti-mouse Alexa Fluor 647 (Invitrogen) as the growth, migration, differentiation, and cell death. p75 is secondary antibody to emphasize PGP9.5. This anti- considered the primary antibody in marking and identify- body was optimally diluted to 1:200 with 0.1 mol/L ing the Pacinian corpuscle.18 We optimized rabbit anti- PBS in a volume of 100 ␮L per ligament sample. NGF receptor p75 (Code AB1554; Millipore, Bil- lerica, MA) to a dilution of 1:100 in 0.1 mol/L PBS Immunofluorescence imaging in a volume of 100 ␮L per ligament sample. We imaged the immunohistochemical sections using We also used rabbit monoclonal anti-Protein Gene fluorescent microscopy (Observer.Z1; Carl Zeiss Mi- Product (PGP) 9.5 (Code AB1761; Millipore). This croImaging, Thornwood, NY) and AxioVision soft- antibody is directed against PGP9.5, a pan-neuronal ware (release 4.7.1; Carl Zeiss MicroImaging). We marker present in neuronal cytoplasm in all mammals imaged the median nerve control first, evaluating pos- including humans, and especially in the metabolically itive and negative controls individually for p75 and dynamic regions of the cell. PGP9.5 is used to identify PGP9.5 stains as well as background, nonspecific, and the afferent axons of mechanoreceptors and the bulbous p75 and PGP9.5 combination staining in the dual-color terminals of the Ruffini endings. For use, we optimized samples. After ensuring staining success, we evaluated anti-PGP9.5 to a dilution of 1:400 in 0.1 mol/L PBS in ligaments of the TM joint. a volume of 100 ␮L per ligament sample. We used a multidimensional acquisition setting to To visualize p75 and PGP9.5 simultaneously, we capture images of nerve fascicles and mechanorecep- employed a triple-stain technique.19 Use of monoclonal tors. This setting captured pictures at 488 and 596 nm, antibodies derived from different species allowed for and the DAPI wavelength separately, and for different simultaneous use of multiple secondary antibodies to amounts of time at each wavelength; then it automati- generate, in conjunction with DAPI, a triple-stained cally merged the 3 images into 1 comprehensive pic- specimen sample. ture. We adjusted exposure time for each individual wavelength to optimize resolution and maximum por- Secondary antibodies trayal of immunofluorescence detail. We used the Alexa Fluor series for the secondary antibody. For dual-staining, we used goat anti-rabbit Alexa Ordinal analysis of innervation Fluor 488 (Invitrogen) as the secondary antibody to We employed an ordinal grading system previously highlight both p75 and PGP9.5, only emphasizing p75 used in analysis of ligament innervation to quantify the in the triple stain. In addition, in triple-staining, we used degree of innervation and assess for mechanoreceptor

JHS ᭜ Vol A, April  INNERVATION PATTERNS OF THUMB TMJ LIGAMENTS 709

FIGURE 2: The macroscopic innervation of the thumb carpometacarpal joint capsule. A The dorsal capsule frequently received branches (arrows) from the dorsal sensory radial nerve. B The innervation of the volar capsule was more intermittent with branches from the thenar branch of the median nerve (arrows). presence.6,18 The grading system ranged from (ϩϩϩ) (P ϭ .010) with a greater number of sensory nerve to (–), where (ϩϩϩ) represents a presence of several endings in the distal halves near the first metacarpal nerve fascicles and mechanoreceptors, (ϩϩ) single insertion rather than the proximally and near the trape- nerve fascicle and receptor, (ϩ) nerve fascicles but no zial origins. receptors, and (–) no nerve fibers, fascicles, or mechanoreceptors. Mechanoreceptors and nerve fascicles Mechanoreceptors and nerves were distinguished Statistical analysis through specific immunofluorescent (IF) patterns. The We used mixed-effects linear regression to estimate the Ruffini ending, a slowly adapting receptor that is con- differences in innervation grade between ligaments stantly active in the joint, is believed to be important in while accounting for dependencies in the data, specifi- the monitoring static joint position and changes in am- cally 2 locations per ligament and 5 ligaments per plitude and velocity of joint rotations.20,21 It was the specimen. From this model, we made simultaneous most commonly found mechanoreceptor, and was iden- pairwise comparisons between the ligaments, with a tified by its characteristic terminal dendritic branches family-wise 95% confidence level producing estimates, and partial encapsulation. In the triple-stain technique, Z statistics, P values, and simultaneous 95% confidence the Ruffini ending was seen with regions of individual intervals for the mean differences between each liga- p75 and PGP9.5 IF but also regions with overlapping IF ment pair. This model was elaborated to estimate the (Fig. 3). effect of location (proximal vs distal) as well as an Rarely, we saw a series of grouped dendritic recep- interaction term to evaluate whether the effect of loca- tors, which has previously been suggested to represent tion varied by ligament. We performed all statistical the so-called “Golgi-like ending” in ligaments (Fig. analyses in collaboration with our institute’s statistician. 4).22 They were characterized by groups of Ruffini endings found in the same location. The Golgi-like RESULTS ending signals only the extremes of joint motion and is General observations more common in ligaments of large joints (ie, the Macroscopically, we saw branches from the dorsal sen- cruciate ligaments of the knee) than in small joints.23 sory radial nerve and occasionally from the thenar Finally, the Pacinian corpuscle is a rapidly adapting branches of the median nerve entering the dorsal and receptor that is thought to be active only in joint accel- volar TM joint capsule, respectively (Fig. 2). Micro- eration and deceleration, signaling sudden joint pertur- scopically, nerve fascicles and mechanoreceptors—Pa- bations.21 Although common in the skin of the palm,24 cinian corpuscles, Ruffini endings, and Golgi-like end- we rarely found it in the TM joint ligaments. The Pacini ings—were successfully imaged and mapped primarily corpuscle was identified primarily through its onion- within the epifascicular layer, close to the ligament like, lamellar capsule with distinct p75 IF (Fig. 5). The insertion into bone, and in close proximity to nerve nerve fascicle leading to the mechanoreceptor, the par- fascicles and arterioles. We found statistically signifi- ent axon, expresses p75 IF in the perineurium and cant differences when comparing degree of innervation PGP9.5 IF in the proper axon, but only p75 was ob- in the distal and proximal portions of each ligament served in the lamellar layers of the corpuscle.

JHS ᭜ Vol A, April  710 INNERVATION PATTERNS OF THUMB TMJ LIGAMENTS

FIGURE 3: A Ruffini ending from a dorsal central ligament as seen when stained for p75 A and PGP9.5 B. The terminal bulbous ending of the Ruffini ending (arrowhead). Note the afferent parent axon (N) ending in a characteristic dendritic receptor (arrowhead) and the areas of overlapping p75 and PGP9.5 immunofluorescence D. C The DAPI stain marks the nuclei of the surrounding fibrocytes in the collagen.

paring the 3 dorsal ligaments as a group (DRL, DCL, and POL) with the 2 volar ligaments as a group (AOL and UCL), there was a statistically greater distribution of nerve endings in the dorsal ligament group (P Ͻ .001). Comparing the mean grade of innervation between each ligament (Fig. 7; see Fig. 8, available on the Journal’s Web site at www.jhandsurg.org for linear hypotheses), the dorsal ligaments (DRL, DCL, and POL) had a statistically significant greater degree of innervation than both the AOL (P Ͻ .001) and the UCL (P Ͻ .001 for DRL-UCL and DCL-UCL, respectively (P ϭ .009 for POL-UCL). There was no significant FIGURE 4: A grouped Ruffini ending, frequently called a Golgi- difference when comparing degree of innervation like tendon organ, as seen in a dual p75 (green) and DAPI (blue) within the dorsal ligaments or within the volar liga- stain from a dorsal radial ligament. The parent axons (N) are seen ments (P Ͼ .100). ending as a cluster of 4 spray-endings (arrowheads). The DAPI stains the nuclei of an adjacent arteriole (asterisk). DISCUSSION The TM joint presents a challenge to clinicians because of its inherent predisposition for OA disease. Thumb Distribution of innervation trapeziometacarpal OA has a prevalence of 21% to 35% Figure 6 maps the mean grade of innervation in the 5 in postmenopausal women,25–27 with a rapid increase in TM joint ligaments from the 10 specimens. When com- prevalence after the age of 50,28 and is in fact the most

JHS ᭜ Vol A, April  INNERVATION PATTERNS OF THUMB TMJ LIGAMENTS 711

strengthen our methodology, we devised a triple- staining IF technique that enhanced our ability to accu- rately depict and differentiate among sensory nerve endings.19 The consistent presence of mechanoreceptors and free nerve endings, with a significantly greater innervation in the dorsal ligaments (P Ͻ .001), suggests a proprioceptive function of these ligaments in stabilizing the TM joint. Because the ligaments we analyzed in this study were harvested from 10 cadaver specimens with no or minor signs of TM OA, and given the rich innervation and minor OA, we postulate that these subjects had intact and protective proprioceptive function. Studies on patients with knee OA have revealed a general proprioceptive impairment with loss of knee sensorimotor function,33,34 and animal studies have shown an increase in knee OA associated with joint denervation and loss of proprioception.35,36 Further studies will be required to investigate whether the innervation pattern is similarly altered in patients with FIGURE 5: A Pacinian corpuscle from a dorsal radial ligament advanced TM OA. visualized in an IF triple-stain with simultaneous capture of p75 Our findings of mechanoreceptors in the epifascicu- (green), PGP9.5 (red/yellow), and DAPI (blue). The parent axon lar regions of the TM ligaments and predominately (N) is primarily seen with p75 immunofluorescence (IF) but also close to the ligament insertion into bone correspond with occasional markings of the axonal fibers in PGP9.5 (arrow). with previous reports on mechanoreceptors in human Note the characteristic lamellar layers (arrowheads) of the shoulder, wrist, knee, and ankle ligaments.37–40 The Pacinian corpuscle which are only p75 IF. The nuclei of the surrounding collagen fibers are marked by DAPI. function of various mechanoreceptor types has been delineated using microneurography of cutaneous and joint afferents.41,42 Owing to their differing character- common cause of surgical intervention due to osteoar- istics, we infer different proprioceptive functions of thritis in the upper extremity.29 In the past decade, ligaments depending on the predominance of mechan- 18 causes of OA have expanded beyond cartilaginous wear oreceptor type. Ligaments that are continuously sub- due to abnormal biomechanical stresses on a joint. A jected to high compressive loads (ie, knee and ankle current hypothesis suggests the loads placed on a joint ligaments) have been shown to have a predominance of are most likely normal, but the biomaterials or connec- Pacinian corpuscles, which are rapidly adapting recep- tive tissues supporting the joint are abnormal, creating tors activated primarily by joint compression and sud- an inability to handle even physiological joint loads and den perturbations.12,40 thus causing OA.30 These authors propose that the joint On the other hand, the Ruffini ending is a slowly should be viewed as a synovial organ, in which any part adapting and constantly active receptor sensitive to ax- of that organ—the cartilage, subchondral bone, syno- ial stress and tensile strains.43,44 It has been found to be vium, ligament, nerve, or periarticular muscle—may be the dominant mechanoreceptor ending in the wrist lig- involved in the development of OA.31 One of the de- aments6 and was similarly the predominant type found scribed potential contributors to progressive OA is thus in the dorsal TM ligaments. Studies on the kinematic a failure of proprioception and neuromuscular control and kinetic functions of the TM joint have shown that of the joint.32 motion occurs mostly from translation of the first meta- Our study is based on IF analysis of fresh-frozen carpal on the trapezium,45 with large compressive cadaveric ligaments, which may negatively influence forces transferred across the joint during pinch and grip the quality of the nerve endings during analysis. Har- loading.46 These compressive forces are located primar- vest from fresh specimens is generally considered to be ily in the volar-ulnar region of the trapezium during more reliable; however, because we consistently found functional tasks47 and result in a tendency for dorsora- nerve endings in the dorsal ligaments, we are confident dial subluxation of the first metacarpal on the trape- that our specimens were of adequate quality. To zium. Based on this, maintained integrity of the dorsal

JHS ᭜ Vol A, April  712 INNERVATION PATTERNS OF THUMB TMJ LIGAMENTS

FIGURE 6: Semiquantitative distribution of nerves and mechanoreceptors in each of the 5 carpometacarpal ligaments from 10 cadaveric specimens. Specimens 1, 2, 4, 5, and 8 represent the male cadavers, and 3, 6, 7, 9, and 10 the female cadavers. Number 6 was the only specimen with no macroscopic signs of OA. ϩϩϩ (3.0) richly innervated with several nerve fascicles (N) and mechanoreceptors (R); ϩϩ (2.0) single N and R; ϩ (1.0) N but no R; – (0.0), no signs of innervation.

FIGURE 7: Comparative analysis of the mean differences in ligament innervation between the 5 TM ligaments. The DRL, DCL, and POL ligaments have signiﬁcantly more innervation than the volar ligaments (P Ͻ .001 for DRL-AOL, DCL-AOL, POL-AOL, DRL-UCL, and DCL-UCL, respectively; P ϭ .009 for POL-UCL). (See Fig. 8, available on the Journal’s Web site at www.jhandsurg.org, for linear hypotheses.)

JHS ᭜ Vol A, April  INNERVATION PATTERNS OF THUMB TMJ LIGAMENTS 713 ligament complex is important in maintaining joint con- ous ligament afferent proprioceptive pathway: a human in vivo gruity,48 and dorsoradial subluxation has also been im- experimental study. J Hand Surg 2011;36A:37–46. 49 9. Hagert E, Persson JK. Desensitizing the posterior interosseous nerve plied as a risk factor in the development of TM OA. alters wrist proprioceptive reflexes. J Hand Surg 2010;35A:1059– The presence of Ruffini endings—signaling axial 1066. strains and tensile forces—in the dorsal ligament com- 10. Hagert E, Persson JK, Werner M, Ljung BO. Evidence of wrist proprioceptive reflexes elicited after stimulation of the scapholunate plex, with predominance in the mobile metacarpal por- interosseous ligament. J Hand Surg 2009;34A:642–651. tion of the ligaments, appears logical because this cor- 11. Sjolander P, Johansson H, Djupsjobacka M. Spinal and supraspinal responds with the restraining function of the dorsal effects of activity in ligament afferents. J Electromyogr Kinesiol ligament complex. That the dorsal ligaments are signif- 2002;12:167–176. 12. Johansson H, Sjolander P, Sojka P. A sensory role for the cruciate icantly more densely innervated than the volar liga- ligaments. Clin Orthop 1991;268:161–178. ments (P Ͻ .001) infers their importance as primary 13. Diederichsen LP, Norregaard J, Krogsgaard M, Fischer-Rasmussen stabilizers of the TM joint, both from a sensory and a T, Dyhre-Poulsen P. Reflexes in the shoulder muscles elicited from the human coracoacromial ligament. J Orthop Res 2004;22:976– biomechanical perspective. From a clinical standpoint, 983. the overall predominance of sensory nerve endings in 14. Michelson JD, Hutchins C. Mechanoreceptors in human ankle liga- the dorsal ligaments may also explain the dorsal joint ments. J Bone Joint Surg 1995;77B:219–224. pain frequently encountered in patients with basal 15. Lorea DP, Berthe JV, De Mey A, Coessens BC, Rooze M, Foucher G. The nerve supply of the trapeziometacarpal joint. J Hand Surg thumb OA. 2002;27B:232–237. A recently proposed explanation for why TM OA is 16. Poupon M, Duteille F, Cassagnau E, Leborgne J, Pannier M. [Ana- predominantly a female disease has been hormonal tomical study of the nerve supply of the trapeziometacarpal joint. influences50 and the presence of relaxin receptors in the Fifteen dissections]. Rev Chir Orthop Reparatrice Appar Mot 2004; 90:346–352. thumb ligaments, which contribute to increased liga- 17. Outerbridge RE. The etiology of chondromalacia patellae. J Bone ment laxity and which, in turn, may predispose to Joint Surg 1961;43B:752–757. progression of OA.51 Ligament laxity has previously 18. Hagert E, Forsgren S, Ljung BO. Differences in the presence of 52 mechanoreceptors and nerve structures between wrist ligaments may been correlated with reduced joint proprioception, imply differential roles in wrist stabilization. J Orthop Res 2005;23: and the suggested treatment to counteract ligamentous 757–763. failure is a conscious neuromuscular training to pro- 19. Lee J, Ladd AL, Hagert E. Immunofluorescent triple-staining tech- mote dynamic joint stability.53 We believe that early- nique to identify sensory nerve endings in human thumb ligaments. Cells Tissues Organs 2011;Aug 10. [Epub ahead of print]. onset basal thumb OA should similarly be treated 20. Maeda T, Ochi K, Nakakura-Ohshima K, Youn SH, Wakisaka S. through a careful rehabilitation of muscles that reduce The Ruffini ending as the primary mechanoreceptor in the periodon- dorsoradial subluxation and instability. Details regard- tal ligament: its morphology, cytochemical features, regeneration, and development. Crit Rev Oral Biol Med 1999;10:307–327. ing the true nature of muscle action on the TM joint and 21. Johansson H, Pedersen J, Bergenheim M, Djupsjobacka M. Periph- knowledge about which muscles protect or destabilize eral afferents of the knee: their effects on central mechanisms reg- the joint represent challenges for our research team. ulating muscle stiffness, joint stability and proprioception and coor- dination. In: Lephart SM, Fu FH, eds. Proprioception and neuromuscular control in joint stability. Champaign, IL: Human REFERENCES Kinetics, 2000:5–22. 1. Almécija S, Moyà-Solà S, Alba DM. Early origin for human-like 22. Halata Z. The mechanoreceptors of the mammalian skin ultrastruc- precision grasping: a comparative study of pollical distal phalanges ture and morphological classification. Adv Anat Embryol Cell Biol in fossil hominins. PLoS One 2010;5:e11727. 1975;50:3–77. 2. Alba DM, Moyà-Solà S, Kohler M. Morphological affinities of the 23. Johansson H, Sjolander P, Sojka P. Receptors in the knee joint Australopithecus afarensis hand on the basis of manual proportions ligaments and their role in the biomechanics of the joint. Crit Rev and relative thumb length. J Hum Evol 2003;44:225–254. Biomed Eng 1991;18:341–368. 3. Marzke MW, Tocheri MW, Steinberg B, Femiani JD, Reece SP, 24. Stark B, Carlstedt T, Hallin RG, Risling M. Distribution of human Linscheid RL, et al. Comparative 3D quantitative analyses of trape- Pacinian corpuscles in the hand. A cadaver study. J Hand Surg ziometacarpal joint surface curvatures among living catarrhines and 1998;23B:370–372. fossil hominins. Am J Phys Anthropol 2010;141:38–51. 25. Armstrong AL, Hunter JB, Davis TR. The prevalence of degenera- 4. Linscheid RL, Dobyns JH. Dynamic carpal stability. Keio J Med tive arthritis of the base of the thumb in post-menopausal women. 2002;51:140–147. J Hand Surg 1994;19B:340–341. 5. Edmunds JO. Current concepts of the anatomy of the thumb trape- 26. Dahaghin S, Bierma-Zeinstra SM, Ginai AZ, Pols HA, Hazes JM, ziometacarpal joint. J Hand Surg 2011;36A:170–182. Koes BW. Prevalence and pattern of radiographic hand osteoarthritis 6. Hagert E, Garcia-Elias M, Forsgren S, Ljung BO. Immunohisto- and association with pain and disability (the Rotterdam study). Ann chemical analysis of wrist ligament innervation in relation to their Rheum Dis 2005;64:682–687. structural composition. J Hand Surg 2007;32A:30–36. 27. Wilder FV, Barrett JP, Farina EJ. Joint-specific prevalence of osteo- 7. Mataliotakis G, Doukas M, Kostas I, Lykissas M, Batistatou A, Beris arthritis of the hand. Osteoarthritis Cartilage 2006;14:953–957. A. Sensory innervation of the subregions of the scapholunate in- 28. Sonne-Holm S, Jacobsen S. Osteoarthritis of the first carpometacar- terosseous ligament in relation to their structural composition. pal joint: a study of radiology and clinical epidemiology. Results J Hand Surg 2009;34A:1413–1421. from the Copenhagen Osteoarthritis Study. Osteoarthritis Cartilage 8. Vekris MD, Mataliotakis GI, Beris AE. The scapholunate interosse- 2006;14:496–500.

JHS ᭜ Vol A, April  714 INNERVATION PATTERNS OF THUMB TMJ LIGAMENTS

29. Van Heest AE, Kallemeier P. Thumb carpal metacarpal arthritis. mechanoreceptors in joints, muscle and skin in human subjects. Clin J Am Acad Orthop Surg 2008;16:140–151. Exp Pharmacol Physiol 2005;32:135–144. 30. Brandt KD, Radin EL, Dieppe PA, van de Putte L. Yet more 42. Vallbo AB, Hagbarth KE. Activity from skin mechanoreceptors evidence that osteoarthritis is not a cartilage disease. Ann Rheum Dis recorded percutaneously in awake human subjects. Exp Neurol 1968; 2006;65:1261–1264. 21:270–289. 31. Brandt KD, Dieppe P, Radin E. Etiopathogenesis of osteoarthritis. 43. Grigg P. Properties of sensory neurons innervating synovial joints. Med Clin North Am 2009;93:1–24, xv. Cells Tissues Organs 2001;169:218–225. 32. Sharma L. Proprioceptive impairment in knee osteoarthritis. Rheum 44. Grigg P, Hoffman AH. Stretch-sensitive afferent neurons in cat knee Dis Clin North Am 1999;25:299–314, vi. joint capsule: sensitivity to axial and compression stresses and 33. Barrack RL, Skinner HB, Cook SD, Haddad RJ, Jr. Effect of artic- strains. J Neurophysiol 1996;75:1871–1877. ular disease and total knee arthroplasty on knee joint-position sense. 45. Imaeda T, Niebur G, Cooney WP III, Linscheid RL, An KN. Kine- J Neurophysiol 1983;50:684–687. matics of the normal trapeziometacarpal joint. J Orthop Res 1994; 34. Hurley MV, Scott DL, Rees J, Newham DJ. Sensorimotor changes 12:197–204. and functional performance in patients with knee osteoarthritis. Ann 46. Cooney WP III, Chao EY. Biomechanical analysis of static forces in Rheum Dis 1997;56:641–648. the thumb during hand function. J Bone Joint Surg 1977;59A:27–36. 35. O’Connor BL, Vilensky JA. Peripheral and central nervous sys- 47. Ateshian GA, Ark JW, Rosenwasser MP, Pawluk RJ, Soslowsky LJ, tem mechanisms of joint protection. Am J Orthop 2003;32:330– Mow VC. Contact areas in the thumb carpometacarpal joint. J Or- 336. thop Res 1995;13:450–458. 36. O’Connor BL, Visco DM, Brandt KD, Myers SL, Kalasinski LA. 48. Van Brenk B, Richards RR, Mackay MB, Boynton EL. A biome- Neurogenic acceleration of osteoarthrosis. The effects of previous chanical assessment of ligaments preventing dorsoradial subluxation neurectomy of the articular nerves on the development of osteoar- of the trapeziometacarpal joint. J Hand Surg 1998;23A:607–611. throsis after transection of the anterior cruciate ligament in dogs. 49. Hunter DJ, Zhang Y, Sokolove J, Niu J, Aliabadi P, Felson DT. J Bone Joint Surg 1992;74A:367–376. Trapeziometacarpal subluxation predisposes to incident trapezio- 37. Morisawa Y. Morphological study of mechanoreceptors on the cora- metacarpal osteoarthritis (OA): the Framingham Study. Osteoarthri- coacromial ligament. J Orthop Sci 1998;3:102–110. tis Cartilage 2005;13:953–957. 38. Tomita K, Berger EJ, Berger RA, Kraisarin J, An KN. Distribution 50. Wolf JM. The influence of ligamentous laxity and gender: implica- of nerve endings in the human dorsal radiocarpal ligament. J Hand tions for hand surgeons. J Hand Surg 2009;34A:161–163. Surg 2007;32A:466–473. 51. Wolf JM, Schreier S, Tomsick S, Williams A, Petersen B. Radio- 39. Del Valle ME, Harwin SF, Maestro A, Murcia A, Vega JA. Immu- graphic laxity of the trapeziometacarpal joint is correlated with nohistochemical analysis of mechanoreceptors in the human poste- generalized joint hypermobility. J Hand Surg 2011;36A:1165–1169. rior cruciate ligament: a demonstration of its proprioceptive role and 52. Sharma L. The role of proprioceptive deficits, ligamentous laxity, clinical relevance. J Arthroplasty 1998;13:916–922. and malalignment in development and progression of knee osteoar- 40. Moraes MR, Cavalcante ML, Leite JA, Ferreira FV, Castro AJ, thritis. J Rheumatol Suppl 2004;70:87–92. Santana MG. Histomorphometric evaluation of mechanoreceptors 53. Lee HM, Cheng CK, Liau JJ. Correlation between proprioception, and free nerve endings in human lateral ankle ligaments. Foot Ankle muscle strength, knee laxity, and dynamic standing balance in patients Int 2008;29:87–90. with chronic anterior cruciate ligament deficiency. Knee 2009;16:387– 41. Macefield VG. Physiological characteristics of low-threshold 391.

JHS ᭜ Vol A, April  INNERVATION PATTERNS OF THUMB TMJ LIGAMENTS 714.e1

FIGURE 8: Comparative analysis of mean grade of distribution between the CMC ligaments.

JHS ᭜ Vol A, April 