- Email: [email protected]
Physical equilibrium of the normal wrist and its relation to clinically defined “instability”
P H Y S I C A L E Q U I L I B R I U M OF T H E N O R M A L W R I S T A N D ITS R E L A T I O N TO C L I N I C A L L Y D E F I N E D " I N S T A B I L I T Y " V. ZDRAVKOVIC, H. A. C. JACOB and G. R. SENNWALD
From the Chirurgie St Leonhard, Clinicfor Hand and Outpatient Surgery, St Gallen, and the Department of Orthopaedic Surgery, Balgrist, University of Zurich, Switzerland The rotational stability of the proximal carpal row was tested on six unembalmed human cadaver hand specimens. The physiological load conditions were simulated by loading the wrist flexor and extensor tendons. Pure torque was introduced to the lunate, scaphoid and triquetrum, one at a time, by means of a dynamometer wrench, forcing the bones loaded to perform a flexion-extension motion. A truly stable state of equilibrium could be found in the normal wrist only under axial load. A uni-directional coupling was observed through the scapho-lunate ligament as a counteraction to a tendency for the lunate to extend and the scaphoid to flex. The triquetrum and lunate moved together, showing close coupling in both directions. As conclusion: a stable wrist can be defined as one which, while being loaded within a physiological range of stress, does not deviate from a stable state of equilibrium (the ability to return to a single position when disturbed) at any point within the whole physiological range of motion.
Journal of Hand Surgery (British and European Volume, 1995) 20B: 2:159-164 The stability of the wrist and a potential role of the scaphoid in preventing the "crumpling" deformity was first discussed by Gilford et al (1943). Linscheid et al (1972) established "instability" of the wrist as being a clinical entity, thus supporting usage of this technical term in hand surgery. In the case of the wrist the term "instability" is often loosely used to describe a number of wrist disorders including ligament injuries, scaphoid fractures and distal radius fractures (Cooney et al, 1990). Differences between "stable" and "unstable" wrists, as currently defined, are described more in the clinical rather than the technical sense. Pope and Panjabi (1985) pointed out that instability or stability is a mechanical entity, and should be treated likewise in medicine. Instability should be quantified by applying known external loads or motions and by observing through measurement the internal motions or displacements that follow. The terms "stable" and "unstable" are used in mechanics to define two states of equilibrium; a third state is that of neutral equilibrium (Duncan, 1958; Pope and Panjabi, 1985). A given object is in stable, unstable or neutral equilibrium depending on its further movement after being minimally displaced from an initial position, i.e. whether it returns to its initial position, remains at rest, or moves to a new position, respectively .(Fig 1). The equilibrium of the wrist in this sense, or of any of its bones, is not usually considered when discussing a diagnosis of wrist "instability". The importance of the equilibrium of the proximal carpal row issues from the fact that no tendons are attached to it, and therefore it is only stabilized through the surrounding joint surfaces and connecting ligaments (Jacob et al, 1992; Sennwald et al, 1993a). When in clinical practice a case of "wrist instability" is to be treated, it is important to understand the normal equilibrium of the wrist. "Instability" of the wrist is clinically often defined as a rotational malalignment of the scaphoid and lunate relative to the radius. Therefore,
it was decided to test physically, by disturbance of the physiological positions adopted by the scaphoid, lunate and triquetrum, how far this situation is counteracted by given constraints. In the present study pure torque is used to induce external destabilizing rotational displacements of the proximal carpal row, after which the following responses were observed. The reaction to pure torque is independent of location within the plane in which it is applied. This characteristic allows the application of torque also outside the body, as long as there is a mechanical connection (Fig 2A-C). Furthermore, motion caused by the torque depends only on the constraints applied to the object (Fig 2D and E). In this study torque was applied and measured strictly in the
_J
S Fig 1
159
N
U
The three possible states of equilibrium. Above--examples of equilibrium of a round object; below--examples of equilibrium of an intercalated object; S--stable equilibrium; N - neutral equilibrium; U--unstable equilibrium.
THE JOURNAL OF HAND SURGERY VOL. 20B No. 2 APRIL 1995
160
C
D Fig 2
E
m
t
Characteristics of pure torque important for the present study. A b o v e - - p u r e torque can be introduced with the same effect at any point of the object. B e l o w - - t h e reaction (R) of the pure torque depends on the constraints applied to the object.
flexion-extension plane, although the plane of rotation of the carpal bone would have certainly deviated somewhat from this, as might be expected from earlier motion studies (Sennwald et al, 1993b). The purpose of this study is to define the equilibrium of the wrist through an exact biomechanical experiment based upon clinical findings in carpal "instability".
MATERIALS AND METHODS Experiments were performed on six fresh, unembatmed, human cadaver forearm specimens, each including the elbow and the distal third of the humerus. Sex and age of specimens were unknown, but through X-rays and final dissection all were shown to be free of any wrist pathology. Two Steinmann pins were used to transfix the radius and ulna, 5 and 20 cm away from the radiocarpal joint, with the forearm in neutral rotation. Similar pins were used for fixation of the specimen in a specially designed fixation jig (Fig 3). The experiment was performed in a vertical position of the forearm, the hand being in the neutral position. The position of the hand was maintained and constrained by a 2 mm K-wire placed in the third metacarpal bone. The wire was free to move in an axial direction only, allowing distraction of the wrist without constraint. The skin, muscles, and finger flexor and extensor tendons were removed from the specimen, and the fingers were amputated at the level of the MP joints. Strong strings were sutured to the tendons of the flexor
carpi radialis, flexor carpi ulnaris, extensor carpi radialis longus and brevis, and extensor carpi ulnaris. One Steinmann pin of 2 mm diameter and one K-wire of 1 mm diameter were introduced from the dorsal side into each of the scaphoid, lunate, and triquetrum bones (Fig 3). To ensure better fixation, bone cement was injected into pre-drilled holes before introduction of the pins. Pure torque was applied by means of a wrench to
the 2 mm thick wires, one at a time. The 1 mm thick wire always remained unloaded and only served the purpose of observing angular movement of the bone to which it was attached. The influence of translations (or lateral forces) from the wrench were eliminated by two universal joints, leaving as an input to the bones only pure torque. The wrench holders were strictly kept in a plane parallel to the flexion-extension plane of the specimen (the plane observed on lateral X-rays). The magnitude of the torque was measured by a straingauge dynamometer built into the axis of the wrench (Fig 4). In this way controlled and reproducible torque could be applied by hand. The angle of rotational movement in the sagittal plane of each bone in the proximal carpal row, caused by the torque, was determined using a motion analysis system "ExpertVision, Version EV3D 3.1" (Motion Analysis Corporation, Santa Rosa, California). The hardware used consisted of an Analog Data Subsystem (ADS), a video processor (Type VPll0) and a Sun Workstation (SPARC IPS). Input data were obtained with an infra-red camera (TI-23A CCD, NEC Corporation Tokyo), which recorded the motion of reflective markers (8 mm alia). The markers were placed as follows: one pair on each of the 1 mm K-wires; one pair on an x-y plotter, defining the scale for the torque measurement; and one marker on the pen of the plotter measuring the actual torque. The markers applied to the plotter were used to determine the momentary torque when analyzing the motion data on the motion analysis system. An increasing torque of maximum 8 Ncm was applied to each of the bones in the proximal carpal row separately, during which the displacements of all three bones were measured. The motion of the bone to which the torque was applied was considered the primary motion, while the induced motion of the other two bones was defined as secondary. The starting position was maximum palmar flexion and maximum torque. This torque was released and the bones were left free to travel to their stable positions. A torque in the opposite direction was then applied, forcing the bones into maximum extension. After releasing this second destabilizing torque, the bones were left free to regain their stable positions. The initial starting position was reached with a final torque application in the direction of flexion. The results are shown as torque-angle curves, where the torque applied is related to the induced angular displacement for the scaphoid, lunate and triquetrum respectively. Also angle-angle curves are constructed in order to relate the angles of secondary motion to those of the primary motion. After the first measurement had been performed, without an axial load across the joint, the strings attached to the wrist flexor and extensor tendons were loaded with 10 N each, and the measurement repeated.
RESULTS In the unloaded wrist the torque-angle curves for the primary motion are S-shaped for all three bones. Only
WRIST EQUILIBRIUM
Fig 3
Fig 4
161
Experimental set-up.
The torque wrench with Hooke's universal joints which allow free lateral and angular shifts of the wrench holder relative to the distal segment.
at the end of the possible bone excursion was an increase of torque necessary in order to continue the motion (see the curves "unloaded" on Fig 5). The vertical, middle part of the curve shows a range of lunate positions within which the bones can be moved on application of a minimum amount of torque, remaining stationary, however, in the newly acquired position, therefore representing positions of neutral equilibrium. How much the bone suffers displacement by application of a given small amount of torque (e.g. 1 Ncm) depends on whether the wrist is loaded or not (Table 1), After loading the wrist the shape of the curves became oblique at the zero-torque crossing, showing that now much more torque is required to displace the bones (the curves "loaded" on Fig 5). Also, the starting position was immediately regained on release of torque, which corresponds to a state of stable equilibrium. Primary motion of each bone in the proximal carpal row causes the secondary motion of the other two. Coupling between the scaphoid and lunate depends on the direction of primary motion. When the scaphoid is subjected to primary motion in flexion, the secondary motion of the lunate follows it closely, even seemingly
THE JOURNAL OF HAND SURGERY VOL. 20B No. 2 APRIL 1995
162
30
2
-
.30
; 20
' 20
I to
10 -6
3O
4
I0
LO~I~
,~_
6
+
+
f, u~m
2O
30
3O FLEXION
Fig 5
L
T
PLI~ION
PLI~I~
Primary angular motion of the scaphoid (S), lunate (L), and triquetrum (T), in relation to applied torque.
Table 1 - - A n g l e s of primary motion, in degrees, achieved with the application of + 1 Ncm torque on the scaphoid, lunate and triquetrum Unloaded wrist
Scaphoid Lunate Triquetrum
Loaded wrist
Difference
Min
Max
Ave
Min
Max
Ave
Min
Max
Ave
17 32 39
60 54 45
33.4 42.8 42.3
4 5 13
15 33 22
8.9 20.4 18.2
13 17 2l
50 26 26
24.5 22.4 24.1
Min = minimum. Max = maximum. Ave = average.
overtaking it (quadrant PSF/SLF in Fig 6a). Palmar translation of the capitate was observed with the scaphoid moving into flexion. This probably supports gearingup of the secondary palmar flexion of the lunate by transferring load to its palmar horn. During primary scaphoid extension the lunate suddenly follows at a slower rate (arrow in quadrant PSE/SLE in Fig 6a), deviating from its initial course (broken line). In the case of primary lunate flexion the secondary flexion of the scaphoid follows at a slower rate (quadrant PLF/SSF in Fig 6a). During primary lunate extension the scaphoid suddenly begins to follow the lunate more closely (arrow in quadrant PLE/SSE in Fig 6a) than expected from its initial course (broken line). Relative movement between the two bones becomes more restricted in primary scaphoid flexion and primary lunate extension, i.e. where the primary motion would tend to increase the scapho-lunate angle (quadrants PSF/SLF versus PSE/SLE, and PLE/SSE versus PLF/SSF). Motions of the lunate and triquetrum are closely coupled bi-directionally with a lag of about 5 ° (Fig 6b). DISCUSSION There was a consistent return to the initial position of the bones of the proximal carpal row when the applied disturbing torque was released. For such a state of stable equilibrium to exist a load across the wrist was necessary. This finding is similar to an observation by Landsmeer
(1961) that the intercalated bone of the finger always adopts the position of least energy. In the present study the stabilizing effect of the axial load could be dramatically demonstrated with a total load of only 50 N applied through the five flexor and extensor tendons of the wrist. Our experiment confirms the tendency of the lunate to extend under capitate pressure, and the tendency for the scaphoid to flex under constraint of the trapezium and trapezoid (Kauer, 1980). These opposed motions are restrained by the scapho-lunate ligament which is, consequently, under tension, thereby affording tight coupling if relative motion is in the appropriate direction. This occurs by simultaneous flexion of the scaphoid and extension of the lunate. Conversely, the lunate and scaphoid could move relative to each other without constraint if the lunate is flexed and the scaphoid extended. According to Weber (1984), the structure of the scapho-lunate ligament can provide rotational constraints. It might therefore be deduced that the scapho-lunate angle can decrease and not increase under normal conditions, which is very important for wrist function. In flexion of the wrist the scaphoid is primarily forced into flexion, simultaneously driving the lunate into flexion through the action of the scapho-lunate ligament. In extension of the wrist the scaphoid and lunate are uncoupled, allowing independent extension: the scaphoid first, followed by the lunate. This is consistent with
WRIST EQUILIBRIUM
163 S.L.E.
.,"
30
./
20
] / //
/ ,' ]+
S.S.E.
20 / 10
,+/"
./ 20
]. /
/
10
~
o
3+0
30
P.S.E.
20
• . .
//
//
//
~1, +
//
/"
P•L•E.
/
20
30
30
/
S.L.F.
/
S.Lo~.'
$.S.F.
i - / / /"
30 / 20
].
10
20
J
/
10
/'
tO
/
./
].
10 / .'" 3O i P.S.F.
/
' 30
,'/
/
S.T.P+..-'//
'
/
30
2O
/
f
.~
// /]'/
/
/
•1 0 J ,/ P.toE•
30 P
....-
20 ~
10
f
17
27
P.L.E•
10
t0
20
20
]./
/ .
/
/
/ 1 ]
3O
f]S.L.F.
3,0
I
/" / .]" ]
S•T.F•
b Fig 6 Simultaneous motions of the lunate and scaphoid (a), and of the lunate and triquetrum (b). P.S.F./P.S.E.--primary scaphoid flexion/extension; S.L.F./S.L.E.--secondary lunate flexion/extension; P.L.F./P.L.E.--primary lunate flexion/extension; S.S.F./S.S.E.--secondary scaphoid flexion/extension; P.T.F./P.T.E.--primary triquetrum flexion/extension; S.T.F./S.T.E.--secondary triquetrum flexion/extension. Arrows indicate positions where the coupling relationship between scaphoid and lunate change.
the sequence of displacements in flexion-extension of the wrist as observed by Kauer (1980). Consequently, both the radio-carpal and the mid-carpal joints move in either flexion or extension, preventing the "zig-zag" deformity• This cannot be explained by the so-called slider-crank mechanical model of the wrist (Linscheid et al, 1972), in which free rotation in the scapho-lunate joint is assumed, and compressive forces across the wrist are ignored. Resistance to an increase of the scapho-lunate angle results from coupling of the lunate and scaphoid in wrist flexion. As a result, the main component of motion during wrist flexion takes place in the radio-carpal joint•
In extension, the scaphoid can be moved without constraint relative to the lunate, taking part in mid-carpal motion along with the capitate. This is in accordance with the findings of Sarrafian et al (1977)• Increased stress across the wrist caused by power grip induces an increase in the lunate uncovering index (Schuind et al, 1992)• This supports our findings that the stress across the wrist influences rotational stability of the lunate• It is conceivable that this extension of the lunate under stress could become excessive if it is unrestricted due to a torn scapho-lunate ligament. In this case the joint capsule and other ligaments could be overloaded, causing discomfort, pain, and loss of power.
164
A high level of coordination between the triquetrum and lunate (Fig 6b) confirms our kinematic findings, where the patterns of motion for the lunate and triquetrum were almost equal (Sennwald et al, 1993b), and Kauer's findings (1986) that capitate-lunate and hamatetriquetrum motion support one another. Ligaments attached to the triquetrum must have a significant effect on the motion of the lunate as well, especially in radial deviation. Sennwald et al (1993b) reported that the lunate and scaphoid have offset helical axes during the flexion-extension motion. These offset axes together with the peripheral coupling through the scapho-lunate ligament across the interface of the two bones explains the intriguing gearing-up/gearing-down effect as shown by the slopes of the curves in Figure 6a. We therefore suggest that a stable wrist should be defined as one which, while being loaded within a physiological range of stress, does not lose its stable state of equilibrium at any point within the whole physiological range of motion. Accordingly, treatment of an instable wrist should not be merely a restoration of anatomy, but re-establishment of the stable state of equilibrium of the proximal carpal row. The extent of instability that follows after cutting the scapho-lunate ligament in vitro is yet to be studied. Acknowledgements This study was partially supported by grant from "Stifttmg fiir Forschung, Welter- und Fortbildung in Handchirnrgie" of Chirurgie St Leonhard, St Gallen, Switzerland. We thank Professor Condamine, University of Caen, France, for his help in providing the anatomical material.
THE JOURNAL OF HAND SURGERY VOL. 20B No. 2 APRIL 1995
References COONEY, W. P., DOBYNS, J. H. and LINSCHEID, R. L. (1990). Arthroscopy of the wrist: Anatomy and classification of carpal instability. Journal of Arthroscopic and Related Surgery, 6: 2: 133-140. DUNCAN, J. Applied Mechanics for Engineers. London, Macmillan, 1958: 52-53. GILFORD, W. W., BOLTON, R. H. and LAMBRINUDI, C. (1943). The mechanism of the wrist joint: With special reference to fractures of the scaphoid. Guy Hospital Reports, 92: 52-59. JACOB, H. A. C., KUNZ, C. and SENNWALD, G. (1992). Zur Biomechanik des Carpus: Funktionelle Anatomie und Bewegtmgsanalyse der Karpalknochen. Orthop~ide, 21: 81-87. KAUER, J. M. G. (1980). Functional anatomy of the wrist. Clinical Orthopaedics and Related Research, 149: 9-20. KAUER, J. M. G. (1986). The mechanism of the carpal joint. Clinical Orthopaedics and Related Research, 202: 16-26. LANDSMEER, J. M. (1961). Studies in the anatomy of articulation: I. The equilibrium of the "intercalated" bone. Acta Morphologica NeerlandoScandinavica, 3: 287-303. LINSCHEID, R. L., DOBYNS, J. H., BEABOUT, J. W. and BRYAN, R. S. (1972), Traumatic instability of the wrist: Diagnosis, classification, and pathomechanics. Journal of Bone and Joint Surgery, 54A: 8: 1612-1632. POPE, H. M. and PANJABI, M. (1985). Biomechanical definitions of spinal instability. Spine, 10: 3: 255-256. SARRAFIAN, S. K., MELAMED, J. L. and GOSHGARIAN, (3. M. (1977). Study of wrist motion in flexion and extension. Clinical Orthopaedics and Related Research, 126: 153-159. SCHUIND, F. A., LINSCHEID, R. L., AN, K-N. and CHAO, E. Y-S. (1992). Changes in wrist and forearm configuration with grasp and isometric contraction of elbow flexors. Journal of Hand Surgery, 17A: 4: 698-703. SENNWALD, G. R., ZDRAVKOVIC, V., JACOB, H. A. C. and KERN, H. P. (1993a). Kinematic analysis of relative motion within the proximal carpal row. Journal of Hand Surgery, lSB: 5:609 612. SENNWALD, G. R., ZDRAVKOVIC, V., KERN, H-P., and JACOB, H. A. C. (1993b). Kinematics of the wrist and its ligaments. Journal of Hand Surgery, 18A: 5: 805-814. WEBER, E. R. (1984). Concepts governing the rotational shift of the intercalated segment of the carpus. Orthopedic Clinics of North America, 15: 2: 193-207.
Accepted: 29 September i994 Dr G. R. Sennwald, Cgirurgie St Leonhard. Pestalozzistrasse 2, CH-9000 St Gallen, Switzerland. © 1995 The British Societyfor Surgery of the Hand
From the Chirurgie St Leonhard, Clinicfor Hand and Outpatient Surgery, St Gallen, and the Department of Orthopaedic Surgery, Balgrist, University of Zurich, Switzerland The rotational stability of the proximal carpal row was tested on six unembalmed human cadaver hand specimens. The physiological load conditions were simulated by loading the wrist flexor and extensor tendons. Pure torque was introduced to the lunate, scaphoid and triquetrum, one at a time, by means of a dynamometer wrench, forcing the bones loaded to perform a flexion-extension motion. A truly stable state of equilibrium could be found in the normal wrist only under axial load. A uni-directional coupling was observed through the scapho-lunate ligament as a counteraction to a tendency for the lunate to extend and the scaphoid to flex. The triquetrum and lunate moved together, showing close coupling in both directions. As conclusion: a stable wrist can be defined as one which, while being loaded within a physiological range of stress, does not deviate from a stable state of equilibrium (the ability to return to a single position when disturbed) at any point within the whole physiological range of motion.
Journal of Hand Surgery (British and European Volume, 1995) 20B: 2:159-164 The stability of the wrist and a potential role of the scaphoid in preventing the "crumpling" deformity was first discussed by Gilford et al (1943). Linscheid et al (1972) established "instability" of the wrist as being a clinical entity, thus supporting usage of this technical term in hand surgery. In the case of the wrist the term "instability" is often loosely used to describe a number of wrist disorders including ligament injuries, scaphoid fractures and distal radius fractures (Cooney et al, 1990). Differences between "stable" and "unstable" wrists, as currently defined, are described more in the clinical rather than the technical sense. Pope and Panjabi (1985) pointed out that instability or stability is a mechanical entity, and should be treated likewise in medicine. Instability should be quantified by applying known external loads or motions and by observing through measurement the internal motions or displacements that follow. The terms "stable" and "unstable" are used in mechanics to define two states of equilibrium; a third state is that of neutral equilibrium (Duncan, 1958; Pope and Panjabi, 1985). A given object is in stable, unstable or neutral equilibrium depending on its further movement after being minimally displaced from an initial position, i.e. whether it returns to its initial position, remains at rest, or moves to a new position, respectively .(Fig 1). The equilibrium of the wrist in this sense, or of any of its bones, is not usually considered when discussing a diagnosis of wrist "instability". The importance of the equilibrium of the proximal carpal row issues from the fact that no tendons are attached to it, and therefore it is only stabilized through the surrounding joint surfaces and connecting ligaments (Jacob et al, 1992; Sennwald et al, 1993a). When in clinical practice a case of "wrist instability" is to be treated, it is important to understand the normal equilibrium of the wrist. "Instability" of the wrist is clinically often defined as a rotational malalignment of the scaphoid and lunate relative to the radius. Therefore,
it was decided to test physically, by disturbance of the physiological positions adopted by the scaphoid, lunate and triquetrum, how far this situation is counteracted by given constraints. In the present study pure torque is used to induce external destabilizing rotational displacements of the proximal carpal row, after which the following responses were observed. The reaction to pure torque is independent of location within the plane in which it is applied. This characteristic allows the application of torque also outside the body, as long as there is a mechanical connection (Fig 2A-C). Furthermore, motion caused by the torque depends only on the constraints applied to the object (Fig 2D and E). In this study torque was applied and measured strictly in the
_J
S Fig 1
159
N
U
The three possible states of equilibrium. Above--examples of equilibrium of a round object; below--examples of equilibrium of an intercalated object; S--stable equilibrium; N - neutral equilibrium; U--unstable equilibrium.
THE JOURNAL OF HAND SURGERY VOL. 20B No. 2 APRIL 1995
160
C
D Fig 2
E
m
t
Characteristics of pure torque important for the present study. A b o v e - - p u r e torque can be introduced with the same effect at any point of the object. B e l o w - - t h e reaction (R) of the pure torque depends on the constraints applied to the object.
flexion-extension plane, although the plane of rotation of the carpal bone would have certainly deviated somewhat from this, as might be expected from earlier motion studies (Sennwald et al, 1993b). The purpose of this study is to define the equilibrium of the wrist through an exact biomechanical experiment based upon clinical findings in carpal "instability".
MATERIALS AND METHODS Experiments were performed on six fresh, unembatmed, human cadaver forearm specimens, each including the elbow and the distal third of the humerus. Sex and age of specimens were unknown, but through X-rays and final dissection all were shown to be free of any wrist pathology. Two Steinmann pins were used to transfix the radius and ulna, 5 and 20 cm away from the radiocarpal joint, with the forearm in neutral rotation. Similar pins were used for fixation of the specimen in a specially designed fixation jig (Fig 3). The experiment was performed in a vertical position of the forearm, the hand being in the neutral position. The position of the hand was maintained and constrained by a 2 mm K-wire placed in the third metacarpal bone. The wire was free to move in an axial direction only, allowing distraction of the wrist without constraint. The skin, muscles, and finger flexor and extensor tendons were removed from the specimen, and the fingers were amputated at the level of the MP joints. Strong strings were sutured to the tendons of the flexor
carpi radialis, flexor carpi ulnaris, extensor carpi radialis longus and brevis, and extensor carpi ulnaris. One Steinmann pin of 2 mm diameter and one K-wire of 1 mm diameter were introduced from the dorsal side into each of the scaphoid, lunate, and triquetrum bones (Fig 3). To ensure better fixation, bone cement was injected into pre-drilled holes before introduction of the pins. Pure torque was applied by means of a wrench to
the 2 mm thick wires, one at a time. The 1 mm thick wire always remained unloaded and only served the purpose of observing angular movement of the bone to which it was attached. The influence of translations (or lateral forces) from the wrench were eliminated by two universal joints, leaving as an input to the bones only pure torque. The wrench holders were strictly kept in a plane parallel to the flexion-extension plane of the specimen (the plane observed on lateral X-rays). The magnitude of the torque was measured by a straingauge dynamometer built into the axis of the wrench (Fig 4). In this way controlled and reproducible torque could be applied by hand. The angle of rotational movement in the sagittal plane of each bone in the proximal carpal row, caused by the torque, was determined using a motion analysis system "ExpertVision, Version EV3D 3.1" (Motion Analysis Corporation, Santa Rosa, California). The hardware used consisted of an Analog Data Subsystem (ADS), a video processor (Type VPll0) and a Sun Workstation (SPARC IPS). Input data were obtained with an infra-red camera (TI-23A CCD, NEC Corporation Tokyo), which recorded the motion of reflective markers (8 mm alia). The markers were placed as follows: one pair on each of the 1 mm K-wires; one pair on an x-y plotter, defining the scale for the torque measurement; and one marker on the pen of the plotter measuring the actual torque. The markers applied to the plotter were used to determine the momentary torque when analyzing the motion data on the motion analysis system. An increasing torque of maximum 8 Ncm was applied to each of the bones in the proximal carpal row separately, during which the displacements of all three bones were measured. The motion of the bone to which the torque was applied was considered the primary motion, while the induced motion of the other two bones was defined as secondary. The starting position was maximum palmar flexion and maximum torque. This torque was released and the bones were left free to travel to their stable positions. A torque in the opposite direction was then applied, forcing the bones into maximum extension. After releasing this second destabilizing torque, the bones were left free to regain their stable positions. The initial starting position was reached with a final torque application in the direction of flexion. The results are shown as torque-angle curves, where the torque applied is related to the induced angular displacement for the scaphoid, lunate and triquetrum respectively. Also angle-angle curves are constructed in order to relate the angles of secondary motion to those of the primary motion. After the first measurement had been performed, without an axial load across the joint, the strings attached to the wrist flexor and extensor tendons were loaded with 10 N each, and the measurement repeated.
RESULTS In the unloaded wrist the torque-angle curves for the primary motion are S-shaped for all three bones. Only
WRIST EQUILIBRIUM
Fig 3
Fig 4
161
Experimental set-up.
The torque wrench with Hooke's universal joints which allow free lateral and angular shifts of the wrench holder relative to the distal segment.
at the end of the possible bone excursion was an increase of torque necessary in order to continue the motion (see the curves "unloaded" on Fig 5). The vertical, middle part of the curve shows a range of lunate positions within which the bones can be moved on application of a minimum amount of torque, remaining stationary, however, in the newly acquired position, therefore representing positions of neutral equilibrium. How much the bone suffers displacement by application of a given small amount of torque (e.g. 1 Ncm) depends on whether the wrist is loaded or not (Table 1), After loading the wrist the shape of the curves became oblique at the zero-torque crossing, showing that now much more torque is required to displace the bones (the curves "loaded" on Fig 5). Also, the starting position was immediately regained on release of torque, which corresponds to a state of stable equilibrium. Primary motion of each bone in the proximal carpal row causes the secondary motion of the other two. Coupling between the scaphoid and lunate depends on the direction of primary motion. When the scaphoid is subjected to primary motion in flexion, the secondary motion of the lunate follows it closely, even seemingly
THE JOURNAL OF HAND SURGERY VOL. 20B No. 2 APRIL 1995
162
30
2
-
.30
; 20
' 20
I to
10 -6
3O
4
I0
LO~I~
,~_
6
+
+
f, u~m
2O
30
3O FLEXION
Fig 5
L
T
PLI~ION
PLI~I~
Primary angular motion of the scaphoid (S), lunate (L), and triquetrum (T), in relation to applied torque.
Table 1 - - A n g l e s of primary motion, in degrees, achieved with the application of + 1 Ncm torque on the scaphoid, lunate and triquetrum Unloaded wrist
Scaphoid Lunate Triquetrum
Loaded wrist
Difference
Min
Max
Ave
Min
Max
Ave
Min
Max
Ave
17 32 39
60 54 45
33.4 42.8 42.3
4 5 13
15 33 22
8.9 20.4 18.2
13 17 2l
50 26 26
24.5 22.4 24.1
Min = minimum. Max = maximum. Ave = average.
overtaking it (quadrant PSF/SLF in Fig 6a). Palmar translation of the capitate was observed with the scaphoid moving into flexion. This probably supports gearingup of the secondary palmar flexion of the lunate by transferring load to its palmar horn. During primary scaphoid extension the lunate suddenly follows at a slower rate (arrow in quadrant PSE/SLE in Fig 6a), deviating from its initial course (broken line). In the case of primary lunate flexion the secondary flexion of the scaphoid follows at a slower rate (quadrant PLF/SSF in Fig 6a). During primary lunate extension the scaphoid suddenly begins to follow the lunate more closely (arrow in quadrant PLE/SSE in Fig 6a) than expected from its initial course (broken line). Relative movement between the two bones becomes more restricted in primary scaphoid flexion and primary lunate extension, i.e. where the primary motion would tend to increase the scapho-lunate angle (quadrants PSF/SLF versus PSE/SLE, and PLE/SSE versus PLF/SSF). Motions of the lunate and triquetrum are closely coupled bi-directionally with a lag of about 5 ° (Fig 6b). DISCUSSION There was a consistent return to the initial position of the bones of the proximal carpal row when the applied disturbing torque was released. For such a state of stable equilibrium to exist a load across the wrist was necessary. This finding is similar to an observation by Landsmeer
(1961) that the intercalated bone of the finger always adopts the position of least energy. In the present study the stabilizing effect of the axial load could be dramatically demonstrated with a total load of only 50 N applied through the five flexor and extensor tendons of the wrist. Our experiment confirms the tendency of the lunate to extend under capitate pressure, and the tendency for the scaphoid to flex under constraint of the trapezium and trapezoid (Kauer, 1980). These opposed motions are restrained by the scapho-lunate ligament which is, consequently, under tension, thereby affording tight coupling if relative motion is in the appropriate direction. This occurs by simultaneous flexion of the scaphoid and extension of the lunate. Conversely, the lunate and scaphoid could move relative to each other without constraint if the lunate is flexed and the scaphoid extended. According to Weber (1984), the structure of the scapho-lunate ligament can provide rotational constraints. It might therefore be deduced that the scapho-lunate angle can decrease and not increase under normal conditions, which is very important for wrist function. In flexion of the wrist the scaphoid is primarily forced into flexion, simultaneously driving the lunate into flexion through the action of the scapho-lunate ligament. In extension of the wrist the scaphoid and lunate are uncoupled, allowing independent extension: the scaphoid first, followed by the lunate. This is consistent with
WRIST EQUILIBRIUM
163 S.L.E.
.,"
30
./
20
] / //
/ ,' ]+
S.S.E.
20 / 10
,+/"
./ 20
]. /
/
10
~
o
3+0
30
P.S.E.
20
• . .
//
//
//
~1, +
//
/"
P•L•E.
/
20
30
30
/
S.L.F.
/
S.Lo~.'
$.S.F.
i - / / /"
30 / 20
].
10
20
J
/
10
/'
tO
/
./
].
10 / .'" 3O i P.S.F.
/
' 30
,'/
/
S.T.P+..-'//
'
/
30
2O
/
f
.~
// /]'/
/
/
•1 0 J ,/ P.toE•
30 P
....-
20 ~
10
f
17
27
P.L.E•
10
t0
20
20
]./
/ .
/
/
/ 1 ]
3O
f]S.L.F.
3,0
I
/" / .]" ]
S•T.F•
b Fig 6 Simultaneous motions of the lunate and scaphoid (a), and of the lunate and triquetrum (b). P.S.F./P.S.E.--primary scaphoid flexion/extension; S.L.F./S.L.E.--secondary lunate flexion/extension; P.L.F./P.L.E.--primary lunate flexion/extension; S.S.F./S.S.E.--secondary scaphoid flexion/extension; P.T.F./P.T.E.--primary triquetrum flexion/extension; S.T.F./S.T.E.--secondary triquetrum flexion/extension. Arrows indicate positions where the coupling relationship between scaphoid and lunate change.
the sequence of displacements in flexion-extension of the wrist as observed by Kauer (1980). Consequently, both the radio-carpal and the mid-carpal joints move in either flexion or extension, preventing the "zig-zag" deformity• This cannot be explained by the so-called slider-crank mechanical model of the wrist (Linscheid et al, 1972), in which free rotation in the scapho-lunate joint is assumed, and compressive forces across the wrist are ignored. Resistance to an increase of the scapho-lunate angle results from coupling of the lunate and scaphoid in wrist flexion. As a result, the main component of motion during wrist flexion takes place in the radio-carpal joint•
In extension, the scaphoid can be moved without constraint relative to the lunate, taking part in mid-carpal motion along with the capitate. This is in accordance with the findings of Sarrafian et al (1977)• Increased stress across the wrist caused by power grip induces an increase in the lunate uncovering index (Schuind et al, 1992)• This supports our findings that the stress across the wrist influences rotational stability of the lunate• It is conceivable that this extension of the lunate under stress could become excessive if it is unrestricted due to a torn scapho-lunate ligament. In this case the joint capsule and other ligaments could be overloaded, causing discomfort, pain, and loss of power.
164
A high level of coordination between the triquetrum and lunate (Fig 6b) confirms our kinematic findings, where the patterns of motion for the lunate and triquetrum were almost equal (Sennwald et al, 1993b), and Kauer's findings (1986) that capitate-lunate and hamatetriquetrum motion support one another. Ligaments attached to the triquetrum must have a significant effect on the motion of the lunate as well, especially in radial deviation. Sennwald et al (1993b) reported that the lunate and scaphoid have offset helical axes during the flexion-extension motion. These offset axes together with the peripheral coupling through the scapho-lunate ligament across the interface of the two bones explains the intriguing gearing-up/gearing-down effect as shown by the slopes of the curves in Figure 6a. We therefore suggest that a stable wrist should be defined as one which, while being loaded within a physiological range of stress, does not lose its stable state of equilibrium at any point within the whole physiological range of motion. Accordingly, treatment of an instable wrist should not be merely a restoration of anatomy, but re-establishment of the stable state of equilibrium of the proximal carpal row. The extent of instability that follows after cutting the scapho-lunate ligament in vitro is yet to be studied. Acknowledgements This study was partially supported by grant from "Stifttmg fiir Forschung, Welter- und Fortbildung in Handchirnrgie" of Chirurgie St Leonhard, St Gallen, Switzerland. We thank Professor Condamine, University of Caen, France, for his help in providing the anatomical material.
THE JOURNAL OF HAND SURGERY VOL. 20B No. 2 APRIL 1995
References COONEY, W. P., DOBYNS, J. H. and LINSCHEID, R. L. (1990). Arthroscopy of the wrist: Anatomy and classification of carpal instability. Journal of Arthroscopic and Related Surgery, 6: 2: 133-140. DUNCAN, J. Applied Mechanics for Engineers. London, Macmillan, 1958: 52-53. GILFORD, W. W., BOLTON, R. H. and LAMBRINUDI, C. (1943). The mechanism of the wrist joint: With special reference to fractures of the scaphoid. Guy Hospital Reports, 92: 52-59. JACOB, H. A. C., KUNZ, C. and SENNWALD, G. (1992). Zur Biomechanik des Carpus: Funktionelle Anatomie und Bewegtmgsanalyse der Karpalknochen. Orthop~ide, 21: 81-87. KAUER, J. M. G. (1980). Functional anatomy of the wrist. Clinical Orthopaedics and Related Research, 149: 9-20. KAUER, J. M. G. (1986). The mechanism of the carpal joint. Clinical Orthopaedics and Related Research, 202: 16-26. LANDSMEER, J. M. (1961). Studies in the anatomy of articulation: I. The equilibrium of the "intercalated" bone. Acta Morphologica NeerlandoScandinavica, 3: 287-303. LINSCHEID, R. L., DOBYNS, J. H., BEABOUT, J. W. and BRYAN, R. S. (1972), Traumatic instability of the wrist: Diagnosis, classification, and pathomechanics. Journal of Bone and Joint Surgery, 54A: 8: 1612-1632. POPE, H. M. and PANJABI, M. (1985). Biomechanical definitions of spinal instability. Spine, 10: 3: 255-256. SARRAFIAN, S. K., MELAMED, J. L. and GOSHGARIAN, (3. M. (1977). Study of wrist motion in flexion and extension. Clinical Orthopaedics and Related Research, 126: 153-159. SCHUIND, F. A., LINSCHEID, R. L., AN, K-N. and CHAO, E. Y-S. (1992). Changes in wrist and forearm configuration with grasp and isometric contraction of elbow flexors. Journal of Hand Surgery, 17A: 4: 698-703. SENNWALD, G. R., ZDRAVKOVIC, V., JACOB, H. A. C. and KERN, H. P. (1993a). Kinematic analysis of relative motion within the proximal carpal row. Journal of Hand Surgery, lSB: 5:609 612. SENNWALD, G. R., ZDRAVKOVIC, V., KERN, H-P., and JACOB, H. A. C. (1993b). Kinematics of the wrist and its ligaments. Journal of Hand Surgery, 18A: 5: 805-814. WEBER, E. R. (1984). Concepts governing the rotational shift of the intercalated segment of the carpus. Orthopedic Clinics of North America, 15: 2: 193-207.
Accepted: 29 September i994 Dr G. R. Sennwald, Cgirurgie St Leonhard. Pestalozzistrasse 2, CH-9000 St Gallen, Switzerland. © 1995 The British Societyfor Surgery of the Hand