- Email: [email protected]
Intrinsic stability of an unconstrained metacarpophalangeal joint implant
Clinical Biomechanics 18 (2003) 119–125 www.elsevier.com/locate/clinbiomech
Intrinsic stability of an unconstrained metacarpophalangeal joint implant P.L. Kung, P. Chou, R.L. Linscheid, L.J. Berglund, W.P. Cooney III, K.N. An
*
Orthopedic Biomechanics Laboratory, Department of Orthopedics, Division of Orthopedic Research, Mayo Clinic/Mayo Foundation, 200 First Street SW, Rochester, MN 55905 USA Received 8 May 2002; accepted 5 November 2002
Abstract Objective. To compare the intrinsic stability of an unconstrained resurfacing metacarpophalangeal arthroplasty to that of a normal human cadaveric joint. Design. Cadaveric joints and metacarpophalangeal prostheses were studied in a mechanical testing machine at different angles and axial loads to determine the stability ratio in eight directions of movement. Background. An unconstrained resurfacing arthroplasty was designed to replicate the normal anatomy with the exception of the proximal component having a greater arc of curvature on its dorsal aspect. Methods. Eight fresh-frozen cadaveric joints and five different sizes of the A V A N T A metacarpophalangeal prosthesis were studied at 0°, 45° and 90° angles of flexion and at eight different directions of motion with three different axial loads (0, 20, 40 N). A 6component load cell measured the force needed to sublux the joint. The stability ratio was the measured outcome and is defined as ratio of the force of subluxation to the axial force. Results and conclusions. The unconstrained resurfacing arthroplasty has more intrinsic stability than the cadaveric metacarpophalangeal joint in all eight directions tested. Relevance A major complication of metacarpophalangeal implants is ulnopalmar subluxation. The A V A N T A implant is designed to decrease the risk of ulnopalmar subluxation by having a greater arc of curvature on the dorsal aspect of the proximal component. This study shows that the designed implant has greater stability due to the geometry of the implant compared to that of the anatomical joint. Ó 2003 Elsevier Science Ltd. All rights reserved. Keywords: Intrinsic stability; Unconstrained prosthesis; Metacarpophalangeal joint; Arthroplasty; In vitro
1. Introduction The metacarpophalangeal (MCP) joint is a diarthrodial joint that permits flexion and extension, abduction and adduction, pronation and supination of the proximal phalangeal base on the metacarpal head. The stability is derived from the shape of the articular surfaces, the capsuloligamentous structures and the traversing musculotendinous units including sagittal bands and interconnected extensor tendon mechanism.
*
Corresponding author. E-mail address: [email protected] (K.N. An).
The MCP joint is the most common hand joint affected by rheumatoid arthritis (RA). RA causes inflammation and deformation of the joint due to diminished constraint of the capsule and ligaments, especially the radial collateral ligament which leads to impairment and instability (Ruther et al., 1995). One treatment option for arthritic joints is replacement arthroplasty, which was introduced in the 1960s and early 1970s (Swanson, 1972; Swanson, 1973; Niebauer et al., 1969). Silicone prosthetic implants for the MCP have been designed as hinged, flexible implant replacements. Newer implants were designed more on their mechanical stability (Adams et al., 1990; Beevers and Seedhom, 1995; Beevers and Seedhom, 1993; McGovern et al., 2001). Previous studies of the MCP arthroplasties have
0268-0033/03/$ - see front matter Ó 2003 Elsevier Science Ltd. All rights reserved. PII: S 0 2 6 8 - 0 0 3 3 ( 0 2 ) 0 0 1 7 8 - X
120
P.L. Kung et al. / Clinical Biomechanics 18 (2003) 119–125
Fig. 1. Picture of the A V A N T A unconstrained MCP implant. The metacarpal head is a cobalt chrome alloy. The proximal phalanx is molded from ultra-high molecular weight polyethylene.
reported limited range of motion, recurrence of ulnar drift, implant failures, fractures and palmer subluxation due to instability (Beevers and Seedhom, 1993; Beckenbaugh et al., 1976; Hagert et al., 1986; Levack et al., 1987). The purpose of this study is to investigate the intrinsic geometric constraints of a resurfacing MCP implant in comparison to the human joint analyzing the shape of the joint without the soft tissue supporting structures. This study takes into account the resistance to displacement in multiple directions at different angulations of joint flexion-extension contact areas, and different compressive loads. The resurfacing implant is an articulating, unconstrained, two-component design with a semi-hemispherical head that melds with lateral volar contours flared to increase the lateral stability during flexion. The arc of curvature of the head is greater than that of an
anatomical joint in hopes of increasing the stability to decrease the risk of ulnopalmar subluxation. The stems are designed to fit within the intermedullary canal of the metacarpal and proximal phalanx and are secured with bone cement (polymethylmethacrylate). The metacarpal component is made of a cobalt chrome alloy, while the proximal phalanx is molded from ultra high molecular weight polyethylene (UHMWPE) (Fig. 1). Repair of joint collateral ligaments and joint capsule are an important element of the joint replacement procedure.
2. Methods In this experiment, the MCP implants and cadaveric human MCP joints were incorporated into fixtures and secured to a custom built material testing device (Fig. 2). The metacarpal head was allowed one degree of freedom in the axis parallel to the stem of the proximal phalanx. The proximal phalanx was then translated along one of eight different directions. A six degrees of freedom load cell holding the phalangeal component measured the forces between the two components. Five different sizes of the A V A N T A MCP prostheses (XL, L, M, S, and XS) (A V A N T A orthopedics, San Diego, CA, USA) and eight fresh-frozen cadaver joints were tested. The cadaver specimens were radiographed prior to dissection to ensure that there were no gross degenerative changes in the MCP joint. The soft tissues and muscles were removed from the metacarpal and
Fig. 2. Schematic of mechanical testing machine. Weights are added to the carriage, which is free to slide along the phalangeal axial direction. The proximal phalanx base component allows movement in eight different directions: volar, dorsal, ulnar, radial, dorsal–ulnar, ulnar–volar, volar–radial and dorsal–radial. Direction is defined as movement of the proximal phalange in the direction of motion with respect to the metacarpal head.
P.L. Kung et al. / Clinical Biomechanics 18 (2003) 119–125
proximal phalanx including the collateral ligaments and volar plate. Care was taken to keep the articular cartilage intact. Of the eight cadaveric joints, five were long fingers and three were index fingers. We chose to test the index and long (second and third) fingers for the experiment since osteoarthritis and RA commonly affect both. They are the largest of all the digits. This diminished variability by keeping the anatomical group consistent in size and geometry. The prostheses were potted in either cylindrical or rectangular aluminum sleeves with bone cement such that the implant stems were parallel to the sides and perpendicular to the bottom of the fixture (Fig. 2). Likewise, the human MCP joints were potted such that the shafts of the proximal phalanx and the metacarpal were parallel to the aluminum sleeves. To ensure that the specimens were potted accurately, an X-ray fluoroscope (FluroScan, model 50200 ADR, Northbrook, IL, USA) was used to center the joints in the sleeves prior to fixation. The sleeves were attached to the test fixture allowing for consistent and accurate flexion angle positions. A clamp held the proximal mounted component in positions at various joint angles (0°, 45° and 90°), while the distal end was fixed perpendicularly on the load cell. The metacarpal head was unconstrained along the vertical axis. The load cell was mounted on a motorized
121
X –Y stage (DCI Design Components, Franklin, MA, USA) that allowed simultaneous motion in the X and Y direction. To maximize the congruity of the joint, the center of the metacarpal head was aligned symmetrically by eye and then by fluoroscope with the center of the proximal phalangeal base. The centering of the components on the fixture was further facilitated by unlocking of the X –Y stage and the cylindrical fixture allowing the two components to interlock. The stages and fixture were then re-locked. Axial loads were applied to the proximal component through the center of articulation of the base of the proximal phalanx by adding dead weights through the slide on the Z-axis. Each prosthesis and each specimen underwent a series of translation tests in eight directions with three different axial loads and at three joint angles. Movement of the proximal phalanx component occurred with respect to the metacarpal head component in the volar, dorsal, radial, and ulnar direction. While one axis of the X –Y stage moved in the volar–dorsal direction, the other axis produced displacement in the radial–ulnar direction enabling movement in the volar–radial, volar–ulnar, dorsal–radial, and dorsal–ulnar direction for a total of 8 displacement directions. The extra large, large prostheses and human specimens were displaced 3.0 mm in the major directions. The medium, small and extra-small
Fig. 3. Sample graph of force vs. displacement. Force is in direction of motion and is measured by the load cell. In this particular graph, the positive displacement represents movement of the proximal phalanx of the implant in the radial direction. The negative displacement represents movement in the ulnar direction, both with three different axial loads.
122
P.L. Kung et al. / Clinical Biomechanics 18 (2003) 119–125
prostheses were displaced 2.5 mm in the radial, ulnar, volar and dorsal directions. The displacement distances were determined as the distance at which the prostheses or the human specimens would definitely sublux. Movement occurred at 0.5 mm/s in the major directions. Since this experiment is looking at the intrinsic stability of the implant versus the cadaveric joint, the outcome, the stability ratio (SR), is independent of velocity and 0.5 mm/s was arbitrarily chosen. Both the implants and the human specimens were tested at flexion angles of 0°, 45°, and 90° with three different loads on the carriage (0, 20 and 40 N) in the eight directions mentioned above. The implant tests were run with fetal bovine serum as lubricant, while the human specimens were tested with saline as lubricant. Data was collected on a standard PC at 10 Hz with an A–D card (DAS-16, Keithly Metrabyte, Cleveland, OH, USA). Displacement, load, and time were stored for later analysis. Force–displacement curves were plotted for each test (Fig. 3). The load cell measured maximum forces in the direction of motion and the constant axial loads in the Z direction due to the axial loads of 0, 20 and 40 N. These two forces were plotted to provide direct measurement of the stability of the prosthesis. Linear regression was used to find the line of best fit for the force/force curves. The slope is the translating force of subluxation divided by the compressive axial load. This slope multiplied by 100 is then defined as the SR. The higher the SR the greater the force required to sublux the joint. Though
this may seem like the coefficient of friction, the SR is not only dependent on frictional forces, but also on the geometry of the joint.
3. Results The maximum forces in the direction of motion versus the axial loads applied were linear. The slope representing the SR had an average correlation coefficient (R2 ) value of 0:989 0:065 SD. The SR in each direction is defined for the movement of the proximal phalanx in that particular direction of motion. The average SR of each of the MCP prostheses at flexion angles of 0°, 45° and 90° in each particular direction was greater than or comparable to the average SR of the human joints in their respective direction. The anatomic specimens, at all flexion angles of the MCP joint, are more stable in the radial–ulnar direction than that in the dorsal–volar direction (Figs. 4–6). At 0° and 45° of flexion, the volar direction is the least stable, while at 90°, it is the dorsal direction that has the lowest SR. For all angles, and for all sizes, the prosthesis was most stable when the proximal phalangeal base is moved in the volar, ulnar/volar and volar/radial directions and least stable in either the dorsal, dorsal–radial or dorsal–ulnar direction. There is no difference between radial and ulnar stability of the implant (Figs. 4–6). For the different sizes of the implant, the SR changed. In general, the larger the implant, the higher the SR (Table 1).
Fig. 4. Average SR of both human and implant MCPÕs at 0° of flexion. Shaded area represents one standard deviation. The SR is defined as the max force at subluxation/axial load. Directions indicate movement of the proximal phalanx base in the direction of motion.
P.L. Kung et al. / Clinical Biomechanics 18 (2003) 119–125
123
Fig. 5. Average SR of both human and implant MCPÕs at 45° of flexion. Shaded area represents one standard deviation. The SR is defined as the max force at subluxation/axial load. Directions indicate movement of the proximal phalanx base in the direction of motion.
Fig. 6. Average SR of both human and implant MCPÕs at 90° of flexion. Shaded area represents one standard deviation. The SR is defined as the max force at subluxation/axial load. Directions indicate movement of the proximal phalanx base in the direction of motion.
4. Discussion The MCP is normally constrained by its capsuloligamentous structures, by tension in the traversing musculotendinous units, and by the shape of the articular surfaces (Tamai et al., 1988; Minami et al., 1984).
It is important to maintain stability of the MCP joint following prosthetic replacement to obtain satisfactory clinical function. Previous studies have shown that instability is usually associated with palmer subluxation of the proximal phalangeal component alone or along with ulnar drift angulation (Wise, 1975). This study examined
124
P.L. Kung et al. / Clinical Biomechanics 18 (2003) 119–125
Table 1 The average and standard deviation of the SR of five differently sized implants (XL, L, M, S and XS) Dorsal (%)
Dorsal/ulnar (%)
Ulnar (%)
Ulnar/volar (%)
Volar (%)
Volar/radial (%)
Radial (%)
Dorsal/radial (%)
33.3 39.2 32.2 45.0 38.2
38.6 38.0 37.3 53.3 43.0
71.9 76.5 66.3 84.2 65.2
104.7 93.7 105.9 82.5 82.8
105.9 95.4 95.7 77.8 78.3
101.9 90.9 87.7 79.3 72.7
75.9 69.9 65.6 73.9 54.4
37.9 39.5 34.1 46.2 29.2
Average SD
37.6 5.1
42.0 6.7
72.8 7.8
93.9 11.3
90.6 12.2
86.5 11.2
67.9 8.5
37.4 6.3
45° flexion Prosthesis XL Prosthesis L Prosthesis M Prosthesis S Prosthesis XS
34.0 41.9 31.9 41.4 34.5
35.7 36.4 27.3 50.1 39.6
78.1 77.3 74.9 76.1 64.9
106.5 95.2 97.7 79.9 84.4
103. 0 91.1 97.1 75.7 85.5
109.4 91.1 94.6 79.5 74.9
75.2 67.7 68.9 64.6 55.2
37.7 37.0 34.5 43.6 26.1
Average SD
36.7 4.1
37.8 7.4
74.3 4.8
92.7 9.5
90.5 9.4
89.9 12.1
66.3 6.5
35.8 5.7
90° flexion Prosthesis XL Prosthesis L Prosthesis M Prosthesis S Prosthesis XS
6.0 28.9 23.0 28.2 25.6
21.2 34.3 23.2 14.9 13.0
68.8 76.2 62.7 42.7 35.6
99.1 97.6 85.9 50.7 69.0
88.3 85.6 78.1 49.5 63.4
98.5 84.0 80.1 54.7 60.5
62.1 67.2 51.6 38.6 24.3
18.8 21.7 10.7 18.5 16.3
Average SD
22.3 8.4
21.3 7.5
57.2 15.5
80.4 18.4
73.0 14.6
75.6 16.0
48.7 15.7
17.2 3.7
0° flexion Prosthesis Prosthesis Prosthesis Prosthesis Prosthesis
XL L M S XS
Movement is of the proximal phalanx base with respect to the metacarpal head. The stability ratio is defined as the max force at subluxation/axial load.
the contributions of the intrinsic morphology of the MCP prostheses as compared to the normal joint. We were particularly interested in seeing whether extending the dorsal rim of the distal component would provide additional stability against the volar subluxation seen in RA where the capsuloligamentous system is compromised. Early motion which is desirable in obtaining optimum movement of the joint places ulnopalmar subluxation stresses on the weakened ligamentous constraints during the early vulnerable period of capsular recovery. By extending the curvature of the proximal phalangeal component we hope to divert the subluxation stresses from the capsule to the dorsal aspect of the prosthetic surfaces. This does raise the possibility of restricting hyperextension if the dorsal lip impinges onto the dorsal cortex of the metacarpal metaphysis. This is unlikely to affect extension before the neutral position is reached. If subluxation is prevented for six to twelve weeks after surgery the likelihood of the collateral ligaments regaining their constraint function appears to be enhanced and the compromise on extension well justified. As shown from the graphs, the prostheses have a higher average SR than that of the human specimens in all their respective directions. The dorsal rim indeed leads to a high SR in the volar direction at all three
angles of joint flexion. Compared to the other directions, movement of the proximal phalanx base in the volar direction is the most stable. This increased stability is much higher than that of a normal joint. In the human joint, the intrinsic stability was similar between the radial and ulnar direction. The alignments of the surrounding soft tissues and tendons position the hand toward ulnar displacement which would most likely explain why angulation and subluxation in the ulnar direction occurs more frequently in clinical cases. There was no difference in the stability of the implant comparing ulnar and radial displacements. The design of the implant is symmetric so that it can be used in either hand. The lowest average stability occurs at 90° flexion when the distal component is moved in the dorsal, dorsal–ulnar and dorsal–radial direction with respect to the proximal component. In previous clinical studies of other implants, palmer subluxation occurs more often than subluxation in the dorsal direction. The collateral ligaments play a major role in stabilizing the joint in flexion and preventing joint subluxation or dislocation (Minami et al., 1984). The flair of the metacarpal head tightens the collateral ligaments in flexion (Minami et al., 1984; Minami et al., 1985). Preservation or repair of the
P.L. Kung et al. / Clinical Biomechanics 18 (2003) 119–125
collateral ligaments is essential to assist the prostheses to resist volar subluxation during flexion. The different sizes of the implants had different SRs. Since the prostheses are geometrically and materially the same, one would expect similar ratios for each size. The differences may be explained by the properties of the UHMWPE metacarpal base, under the different loads. Greater pressure in the smaller implants could have deformed the plastic. Deformation might cause the plastic to ‘‘give,’’ leading to a slightly higher ratio. Though the forces used in this experiment are smaller than the maximum force attainable in grip, they are still comparable to the mean force sustained on the MCP (An et al., 1985). The load witnessed by the MCP during normal function is not just axial but also has component vectors tangential to the joint contact area. The capability of a prosthetic joint to resist such subluxing forces is the rationale for this experiment. No differences were noted in the SRs between the index and long finger for this small sample size despite their slight morphological differences. A possible source of error is the use of fetal bovine serum for lubrication of the prosthesis and saline to lubricate the cadaver specimen, neither of which fully mimics the action of the hyaluronic acid usually coating articular cartilage. This difference is unlikely to alter the results since the frictional forces of cartilage on cartilage or ultra high molecular weight polyethelyene on cobalt chrome with lubricant is orders of magnitude smaller than the forces due to the geometry of the joint. Acknowledgements This study was supported by a grant from the National Institutes of Health, AR17172. References Adams, B.D., Blair, W.F., Shurr, D.G., 1990. Schultz metacarpophalangeal arthroplasty: a long-term follow-up study. J. Hand Surg. (Am.) 15, 641–645.
125
An, K., Chao, E., Cooney, W.I., Linscheid, R., 1985. Forces in the normal and abnormal hand. J. Orthop. Res. 3, 202–211. Beckenbaugh, R.D., Dobyns, J.H., Linscheid, R.L., Bryan, R.S., 1976. Review and analysis of silicone-rubber metacarpophalangeal implants. J. Bone Joint Surg. Am. 58, 483–487. Beevers, D., Seedhom, B., 1993. Metacarpophalangeal joint prostheses: a review of past and current designs. Proc. Inst. Mech. Eng. [H] 207, 195–206. Beevers, D., Seedhom, B., 1995. Metacarpophalangeal joint prostheses: a review of the clinical results of past and current designs. J. Hand Surg. (Br.) 20, 125–136. Hagert, C.G., Branemark, P.I., Albrektsson, T., Strid, K.G., Irstam, L., 1986. Metacarpophalangeal joint replacement with osseointegrated endoprostheses. Scand. J. Plast. Reconstr. Surg. 20, 207– 218. Levack, B., Stewart, H.D., Flierenga, H., Helal, B., 1987. Metacarpophalangeal joint replacement with a new prosthesis: description and preliminary results of treatment with the Helal flap joint. J. Hand Surg. (Br.) 12, 377–381. McGovern, R.M., Shin, A.Y., Beckenbaugh, R.D., Linscheid, R.L., 2001. Long-term results of cemented Steffee arthroplasty of the thumb metacarpophalangeal joint. J. Hand Surg. (Am.) 26, 115– 122. Minami, A., An, K.N., Cooney 3rd, W.P., Linscheid, R.L., Chao, E.Y., 1984. Ligamentous structures of the metacarpophalangeal joint: a quantitative anatomic study. J. Orthop. Res. 1, 361– 368. Minami, A., An, K.N., Cooney 3rd, W.P., Linscheid, R.L., Chao, E.Y., 1985. Ligament stability of the metacarpophalangeal joint: a biomechanical study. J. Hand Surg. (Am.) 10, 255–260. Niebauer, J.J., Shaw, J.L., Doren, W.W., 1969. Silicone-dacron hinge prosthesis. Design, evaluation, and application. Ann. Rheum. Dis. 28 (Suppl.), 56–58. Ruther, W., Verhestraeten, B., Fink, B., Tillmann, K., 1995. Resection arthroplasty of the metacarpophalangeal joints in rheumatoid arthritis. Results after more than 15 years. J. Hand Surg. (Br.) 20, 707–715. Swanson, A.B., 1972. Flexible implant arthroplasty for arthritic finger joints: rationale, technique, and results of treatment. J. Bone Joint Surg. Am. 54, 435–455. Swanson, A.B., 1973. Flexible Implant Resection Arthroplasty in the Hand and Extremities. C.V. Mosby, St. Louis. Tamai, K., Ryu, J., An, K., Linscheid, R., Cooney, W.I., Chao, E., 1988. Three-dimensional geometric analysis of the metacarpophalangeal joint. J. Hand Surg. (Am.) 13, 521–529. Wise, K., 1975. The anatomy of the metacarpophalangeal joints, with observations of the etiology of ulnar drift. J. Bone Joint Surg. Br. 57, 485–490.
Intrinsic stability of an unconstrained metacarpophalangeal joint implant P.L. Kung, P. Chou, R.L. Linscheid, L.J. Berglund, W.P. Cooney III, K.N. An
*
Orthopedic Biomechanics Laboratory, Department of Orthopedics, Division of Orthopedic Research, Mayo Clinic/Mayo Foundation, 200 First Street SW, Rochester, MN 55905 USA Received 8 May 2002; accepted 5 November 2002
Abstract Objective. To compare the intrinsic stability of an unconstrained resurfacing metacarpophalangeal arthroplasty to that of a normal human cadaveric joint. Design. Cadaveric joints and metacarpophalangeal prostheses were studied in a mechanical testing machine at different angles and axial loads to determine the stability ratio in eight directions of movement. Background. An unconstrained resurfacing arthroplasty was designed to replicate the normal anatomy with the exception of the proximal component having a greater arc of curvature on its dorsal aspect. Methods. Eight fresh-frozen cadaveric joints and five different sizes of the A V A N T A metacarpophalangeal prosthesis were studied at 0°, 45° and 90° angles of flexion and at eight different directions of motion with three different axial loads (0, 20, 40 N). A 6component load cell measured the force needed to sublux the joint. The stability ratio was the measured outcome and is defined as ratio of the force of subluxation to the axial force. Results and conclusions. The unconstrained resurfacing arthroplasty has more intrinsic stability than the cadaveric metacarpophalangeal joint in all eight directions tested. Relevance A major complication of metacarpophalangeal implants is ulnopalmar subluxation. The A V A N T A implant is designed to decrease the risk of ulnopalmar subluxation by having a greater arc of curvature on the dorsal aspect of the proximal component. This study shows that the designed implant has greater stability due to the geometry of the implant compared to that of the anatomical joint. Ó 2003 Elsevier Science Ltd. All rights reserved. Keywords: Intrinsic stability; Unconstrained prosthesis; Metacarpophalangeal joint; Arthroplasty; In vitro
1. Introduction The metacarpophalangeal (MCP) joint is a diarthrodial joint that permits flexion and extension, abduction and adduction, pronation and supination of the proximal phalangeal base on the metacarpal head. The stability is derived from the shape of the articular surfaces, the capsuloligamentous structures and the traversing musculotendinous units including sagittal bands and interconnected extensor tendon mechanism.
*
Corresponding author. E-mail address: [email protected] (K.N. An).
The MCP joint is the most common hand joint affected by rheumatoid arthritis (RA). RA causes inflammation and deformation of the joint due to diminished constraint of the capsule and ligaments, especially the radial collateral ligament which leads to impairment and instability (Ruther et al., 1995). One treatment option for arthritic joints is replacement arthroplasty, which was introduced in the 1960s and early 1970s (Swanson, 1972; Swanson, 1973; Niebauer et al., 1969). Silicone prosthetic implants for the MCP have been designed as hinged, flexible implant replacements. Newer implants were designed more on their mechanical stability (Adams et al., 1990; Beevers and Seedhom, 1995; Beevers and Seedhom, 1993; McGovern et al., 2001). Previous studies of the MCP arthroplasties have
0268-0033/03/$ - see front matter Ó 2003 Elsevier Science Ltd. All rights reserved. PII: S 0 2 6 8 - 0 0 3 3 ( 0 2 ) 0 0 1 7 8 - X
120
P.L. Kung et al. / Clinical Biomechanics 18 (2003) 119–125
Fig. 1. Picture of the A V A N T A unconstrained MCP implant. The metacarpal head is a cobalt chrome alloy. The proximal phalanx is molded from ultra-high molecular weight polyethylene.
reported limited range of motion, recurrence of ulnar drift, implant failures, fractures and palmer subluxation due to instability (Beevers and Seedhom, 1993; Beckenbaugh et al., 1976; Hagert et al., 1986; Levack et al., 1987). The purpose of this study is to investigate the intrinsic geometric constraints of a resurfacing MCP implant in comparison to the human joint analyzing the shape of the joint without the soft tissue supporting structures. This study takes into account the resistance to displacement in multiple directions at different angulations of joint flexion-extension contact areas, and different compressive loads. The resurfacing implant is an articulating, unconstrained, two-component design with a semi-hemispherical head that melds with lateral volar contours flared to increase the lateral stability during flexion. The arc of curvature of the head is greater than that of an
anatomical joint in hopes of increasing the stability to decrease the risk of ulnopalmar subluxation. The stems are designed to fit within the intermedullary canal of the metacarpal and proximal phalanx and are secured with bone cement (polymethylmethacrylate). The metacarpal component is made of a cobalt chrome alloy, while the proximal phalanx is molded from ultra high molecular weight polyethylene (UHMWPE) (Fig. 1). Repair of joint collateral ligaments and joint capsule are an important element of the joint replacement procedure.
2. Methods In this experiment, the MCP implants and cadaveric human MCP joints were incorporated into fixtures and secured to a custom built material testing device (Fig. 2). The metacarpal head was allowed one degree of freedom in the axis parallel to the stem of the proximal phalanx. The proximal phalanx was then translated along one of eight different directions. A six degrees of freedom load cell holding the phalangeal component measured the forces between the two components. Five different sizes of the A V A N T A MCP prostheses (XL, L, M, S, and XS) (A V A N T A orthopedics, San Diego, CA, USA) and eight fresh-frozen cadaver joints were tested. The cadaver specimens were radiographed prior to dissection to ensure that there were no gross degenerative changes in the MCP joint. The soft tissues and muscles were removed from the metacarpal and
Fig. 2. Schematic of mechanical testing machine. Weights are added to the carriage, which is free to slide along the phalangeal axial direction. The proximal phalanx base component allows movement in eight different directions: volar, dorsal, ulnar, radial, dorsal–ulnar, ulnar–volar, volar–radial and dorsal–radial. Direction is defined as movement of the proximal phalange in the direction of motion with respect to the metacarpal head.
P.L. Kung et al. / Clinical Biomechanics 18 (2003) 119–125
proximal phalanx including the collateral ligaments and volar plate. Care was taken to keep the articular cartilage intact. Of the eight cadaveric joints, five were long fingers and three were index fingers. We chose to test the index and long (second and third) fingers for the experiment since osteoarthritis and RA commonly affect both. They are the largest of all the digits. This diminished variability by keeping the anatomical group consistent in size and geometry. The prostheses were potted in either cylindrical or rectangular aluminum sleeves with bone cement such that the implant stems were parallel to the sides and perpendicular to the bottom of the fixture (Fig. 2). Likewise, the human MCP joints were potted such that the shafts of the proximal phalanx and the metacarpal were parallel to the aluminum sleeves. To ensure that the specimens were potted accurately, an X-ray fluoroscope (FluroScan, model 50200 ADR, Northbrook, IL, USA) was used to center the joints in the sleeves prior to fixation. The sleeves were attached to the test fixture allowing for consistent and accurate flexion angle positions. A clamp held the proximal mounted component in positions at various joint angles (0°, 45° and 90°), while the distal end was fixed perpendicularly on the load cell. The metacarpal head was unconstrained along the vertical axis. The load cell was mounted on a motorized
121
X –Y stage (DCI Design Components, Franklin, MA, USA) that allowed simultaneous motion in the X and Y direction. To maximize the congruity of the joint, the center of the metacarpal head was aligned symmetrically by eye and then by fluoroscope with the center of the proximal phalangeal base. The centering of the components on the fixture was further facilitated by unlocking of the X –Y stage and the cylindrical fixture allowing the two components to interlock. The stages and fixture were then re-locked. Axial loads were applied to the proximal component through the center of articulation of the base of the proximal phalanx by adding dead weights through the slide on the Z-axis. Each prosthesis and each specimen underwent a series of translation tests in eight directions with three different axial loads and at three joint angles. Movement of the proximal phalanx component occurred with respect to the metacarpal head component in the volar, dorsal, radial, and ulnar direction. While one axis of the X –Y stage moved in the volar–dorsal direction, the other axis produced displacement in the radial–ulnar direction enabling movement in the volar–radial, volar–ulnar, dorsal–radial, and dorsal–ulnar direction for a total of 8 displacement directions. The extra large, large prostheses and human specimens were displaced 3.0 mm in the major directions. The medium, small and extra-small
Fig. 3. Sample graph of force vs. displacement. Force is in direction of motion and is measured by the load cell. In this particular graph, the positive displacement represents movement of the proximal phalanx of the implant in the radial direction. The negative displacement represents movement in the ulnar direction, both with three different axial loads.
122
P.L. Kung et al. / Clinical Biomechanics 18 (2003) 119–125
prostheses were displaced 2.5 mm in the radial, ulnar, volar and dorsal directions. The displacement distances were determined as the distance at which the prostheses or the human specimens would definitely sublux. Movement occurred at 0.5 mm/s in the major directions. Since this experiment is looking at the intrinsic stability of the implant versus the cadaveric joint, the outcome, the stability ratio (SR), is independent of velocity and 0.5 mm/s was arbitrarily chosen. Both the implants and the human specimens were tested at flexion angles of 0°, 45°, and 90° with three different loads on the carriage (0, 20 and 40 N) in the eight directions mentioned above. The implant tests were run with fetal bovine serum as lubricant, while the human specimens were tested with saline as lubricant. Data was collected on a standard PC at 10 Hz with an A–D card (DAS-16, Keithly Metrabyte, Cleveland, OH, USA). Displacement, load, and time were stored for later analysis. Force–displacement curves were plotted for each test (Fig. 3). The load cell measured maximum forces in the direction of motion and the constant axial loads in the Z direction due to the axial loads of 0, 20 and 40 N. These two forces were plotted to provide direct measurement of the stability of the prosthesis. Linear regression was used to find the line of best fit for the force/force curves. The slope is the translating force of subluxation divided by the compressive axial load. This slope multiplied by 100 is then defined as the SR. The higher the SR the greater the force required to sublux the joint. Though
this may seem like the coefficient of friction, the SR is not only dependent on frictional forces, but also on the geometry of the joint.
3. Results The maximum forces in the direction of motion versus the axial loads applied were linear. The slope representing the SR had an average correlation coefficient (R2 ) value of 0:989 0:065 SD. The SR in each direction is defined for the movement of the proximal phalanx in that particular direction of motion. The average SR of each of the MCP prostheses at flexion angles of 0°, 45° and 90° in each particular direction was greater than or comparable to the average SR of the human joints in their respective direction. The anatomic specimens, at all flexion angles of the MCP joint, are more stable in the radial–ulnar direction than that in the dorsal–volar direction (Figs. 4–6). At 0° and 45° of flexion, the volar direction is the least stable, while at 90°, it is the dorsal direction that has the lowest SR. For all angles, and for all sizes, the prosthesis was most stable when the proximal phalangeal base is moved in the volar, ulnar/volar and volar/radial directions and least stable in either the dorsal, dorsal–radial or dorsal–ulnar direction. There is no difference between radial and ulnar stability of the implant (Figs. 4–6). For the different sizes of the implant, the SR changed. In general, the larger the implant, the higher the SR (Table 1).
Fig. 4. Average SR of both human and implant MCPÕs at 0° of flexion. Shaded area represents one standard deviation. The SR is defined as the max force at subluxation/axial load. Directions indicate movement of the proximal phalanx base in the direction of motion.
P.L. Kung et al. / Clinical Biomechanics 18 (2003) 119–125
123
Fig. 5. Average SR of both human and implant MCPÕs at 45° of flexion. Shaded area represents one standard deviation. The SR is defined as the max force at subluxation/axial load. Directions indicate movement of the proximal phalanx base in the direction of motion.
Fig. 6. Average SR of both human and implant MCPÕs at 90° of flexion. Shaded area represents one standard deviation. The SR is defined as the max force at subluxation/axial load. Directions indicate movement of the proximal phalanx base in the direction of motion.
4. Discussion The MCP is normally constrained by its capsuloligamentous structures, by tension in the traversing musculotendinous units, and by the shape of the articular surfaces (Tamai et al., 1988; Minami et al., 1984).
It is important to maintain stability of the MCP joint following prosthetic replacement to obtain satisfactory clinical function. Previous studies have shown that instability is usually associated with palmer subluxation of the proximal phalangeal component alone or along with ulnar drift angulation (Wise, 1975). This study examined
124
P.L. Kung et al. / Clinical Biomechanics 18 (2003) 119–125
Table 1 The average and standard deviation of the SR of five differently sized implants (XL, L, M, S and XS) Dorsal (%)
Dorsal/ulnar (%)
Ulnar (%)
Ulnar/volar (%)
Volar (%)
Volar/radial (%)
Radial (%)
Dorsal/radial (%)
33.3 39.2 32.2 45.0 38.2
38.6 38.0 37.3 53.3 43.0
71.9 76.5 66.3 84.2 65.2
104.7 93.7 105.9 82.5 82.8
105.9 95.4 95.7 77.8 78.3
101.9 90.9 87.7 79.3 72.7
75.9 69.9 65.6 73.9 54.4
37.9 39.5 34.1 46.2 29.2
Average SD
37.6 5.1
42.0 6.7
72.8 7.8
93.9 11.3
90.6 12.2
86.5 11.2
67.9 8.5
37.4 6.3
45° flexion Prosthesis XL Prosthesis L Prosthesis M Prosthesis S Prosthesis XS
34.0 41.9 31.9 41.4 34.5
35.7 36.4 27.3 50.1 39.6
78.1 77.3 74.9 76.1 64.9
106.5 95.2 97.7 79.9 84.4
103. 0 91.1 97.1 75.7 85.5
109.4 91.1 94.6 79.5 74.9
75.2 67.7 68.9 64.6 55.2
37.7 37.0 34.5 43.6 26.1
Average SD
36.7 4.1
37.8 7.4
74.3 4.8
92.7 9.5
90.5 9.4
89.9 12.1
66.3 6.5
35.8 5.7
90° flexion Prosthesis XL Prosthesis L Prosthesis M Prosthesis S Prosthesis XS
6.0 28.9 23.0 28.2 25.6
21.2 34.3 23.2 14.9 13.0
68.8 76.2 62.7 42.7 35.6
99.1 97.6 85.9 50.7 69.0
88.3 85.6 78.1 49.5 63.4
98.5 84.0 80.1 54.7 60.5
62.1 67.2 51.6 38.6 24.3
18.8 21.7 10.7 18.5 16.3
Average SD
22.3 8.4
21.3 7.5
57.2 15.5
80.4 18.4
73.0 14.6
75.6 16.0
48.7 15.7
17.2 3.7
0° flexion Prosthesis Prosthesis Prosthesis Prosthesis Prosthesis
XL L M S XS
Movement is of the proximal phalanx base with respect to the metacarpal head. The stability ratio is defined as the max force at subluxation/axial load.
the contributions of the intrinsic morphology of the MCP prostheses as compared to the normal joint. We were particularly interested in seeing whether extending the dorsal rim of the distal component would provide additional stability against the volar subluxation seen in RA where the capsuloligamentous system is compromised. Early motion which is desirable in obtaining optimum movement of the joint places ulnopalmar subluxation stresses on the weakened ligamentous constraints during the early vulnerable period of capsular recovery. By extending the curvature of the proximal phalangeal component we hope to divert the subluxation stresses from the capsule to the dorsal aspect of the prosthetic surfaces. This does raise the possibility of restricting hyperextension if the dorsal lip impinges onto the dorsal cortex of the metacarpal metaphysis. This is unlikely to affect extension before the neutral position is reached. If subluxation is prevented for six to twelve weeks after surgery the likelihood of the collateral ligaments regaining their constraint function appears to be enhanced and the compromise on extension well justified. As shown from the graphs, the prostheses have a higher average SR than that of the human specimens in all their respective directions. The dorsal rim indeed leads to a high SR in the volar direction at all three
angles of joint flexion. Compared to the other directions, movement of the proximal phalanx base in the volar direction is the most stable. This increased stability is much higher than that of a normal joint. In the human joint, the intrinsic stability was similar between the radial and ulnar direction. The alignments of the surrounding soft tissues and tendons position the hand toward ulnar displacement which would most likely explain why angulation and subluxation in the ulnar direction occurs more frequently in clinical cases. There was no difference in the stability of the implant comparing ulnar and radial displacements. The design of the implant is symmetric so that it can be used in either hand. The lowest average stability occurs at 90° flexion when the distal component is moved in the dorsal, dorsal–ulnar and dorsal–radial direction with respect to the proximal component. In previous clinical studies of other implants, palmer subluxation occurs more often than subluxation in the dorsal direction. The collateral ligaments play a major role in stabilizing the joint in flexion and preventing joint subluxation or dislocation (Minami et al., 1984). The flair of the metacarpal head tightens the collateral ligaments in flexion (Minami et al., 1984; Minami et al., 1985). Preservation or repair of the
P.L. Kung et al. / Clinical Biomechanics 18 (2003) 119–125
collateral ligaments is essential to assist the prostheses to resist volar subluxation during flexion. The different sizes of the implants had different SRs. Since the prostheses are geometrically and materially the same, one would expect similar ratios for each size. The differences may be explained by the properties of the UHMWPE metacarpal base, under the different loads. Greater pressure in the smaller implants could have deformed the plastic. Deformation might cause the plastic to ‘‘give,’’ leading to a slightly higher ratio. Though the forces used in this experiment are smaller than the maximum force attainable in grip, they are still comparable to the mean force sustained on the MCP (An et al., 1985). The load witnessed by the MCP during normal function is not just axial but also has component vectors tangential to the joint contact area. The capability of a prosthetic joint to resist such subluxing forces is the rationale for this experiment. No differences were noted in the SRs between the index and long finger for this small sample size despite their slight morphological differences. A possible source of error is the use of fetal bovine serum for lubrication of the prosthesis and saline to lubricate the cadaver specimen, neither of which fully mimics the action of the hyaluronic acid usually coating articular cartilage. This difference is unlikely to alter the results since the frictional forces of cartilage on cartilage or ultra high molecular weight polyethelyene on cobalt chrome with lubricant is orders of magnitude smaller than the forces due to the geometry of the joint. Acknowledgements This study was supported by a grant from the National Institutes of Health, AR17172. References Adams, B.D., Blair, W.F., Shurr, D.G., 1990. Schultz metacarpophalangeal arthroplasty: a long-term follow-up study. J. Hand Surg. (Am.) 15, 641–645.
125
An, K., Chao, E., Cooney, W.I., Linscheid, R., 1985. Forces in the normal and abnormal hand. J. Orthop. Res. 3, 202–211. Beckenbaugh, R.D., Dobyns, J.H., Linscheid, R.L., Bryan, R.S., 1976. Review and analysis of silicone-rubber metacarpophalangeal implants. J. Bone Joint Surg. Am. 58, 483–487. Beevers, D., Seedhom, B., 1993. Metacarpophalangeal joint prostheses: a review of past and current designs. Proc. Inst. Mech. Eng. [H] 207, 195–206. Beevers, D., Seedhom, B., 1995. Metacarpophalangeal joint prostheses: a review of the clinical results of past and current designs. J. Hand Surg. (Br.) 20, 125–136. Hagert, C.G., Branemark, P.I., Albrektsson, T., Strid, K.G., Irstam, L., 1986. Metacarpophalangeal joint replacement with osseointegrated endoprostheses. Scand. J. Plast. Reconstr. Surg. 20, 207– 218. Levack, B., Stewart, H.D., Flierenga, H., Helal, B., 1987. Metacarpophalangeal joint replacement with a new prosthesis: description and preliminary results of treatment with the Helal flap joint. J. Hand Surg. (Br.) 12, 377–381. McGovern, R.M., Shin, A.Y., Beckenbaugh, R.D., Linscheid, R.L., 2001. Long-term results of cemented Steffee arthroplasty of the thumb metacarpophalangeal joint. J. Hand Surg. (Am.) 26, 115– 122. Minami, A., An, K.N., Cooney 3rd, W.P., Linscheid, R.L., Chao, E.Y., 1984. Ligamentous structures of the metacarpophalangeal joint: a quantitative anatomic study. J. Orthop. Res. 1, 361– 368. Minami, A., An, K.N., Cooney 3rd, W.P., Linscheid, R.L., Chao, E.Y., 1985. Ligament stability of the metacarpophalangeal joint: a biomechanical study. J. Hand Surg. (Am.) 10, 255–260. Niebauer, J.J., Shaw, J.L., Doren, W.W., 1969. Silicone-dacron hinge prosthesis. Design, evaluation, and application. Ann. Rheum. Dis. 28 (Suppl.), 56–58. Ruther, W., Verhestraeten, B., Fink, B., Tillmann, K., 1995. Resection arthroplasty of the metacarpophalangeal joints in rheumatoid arthritis. Results after more than 15 years. J. Hand Surg. (Br.) 20, 707–715. Swanson, A.B., 1972. Flexible implant arthroplasty for arthritic finger joints: rationale, technique, and results of treatment. J. Bone Joint Surg. Am. 54, 435–455. Swanson, A.B., 1973. Flexible Implant Resection Arthroplasty in the Hand and Extremities. C.V. Mosby, St. Louis. Tamai, K., Ryu, J., An, K., Linscheid, R., Cooney, W.I., Chao, E., 1988. Three-dimensional geometric analysis of the metacarpophalangeal joint. J. Hand Surg. (Am.) 13, 521–529. Wise, K., 1975. The anatomy of the metacarpophalangeal joints, with observations of the etiology of ulnar drift. J. Bone Joint Surg. Br. 57, 485–490.