The laboratory development of a new metacarpophalangeal prosthesis

The Laboratory Development of a New MetacarpophalangealProsthesis B. Weightman, D. M. Evans and D. Light THE L A B O R A T O R Y D E V E L O P M E N T OF A NEW M E T A C A R P O P H A L A N G E A L PROSTHESIS

B. W E I G H T M A N , D. M. EVANS and D. L I G H T , London SUMMARY

The paper describes the design and laboratory testing o f a new metacarpophalangeal prosthesis. The novel features o f the new design include the joining o f pairs o f prostheses by common hinge-pins and the inclusion o f bearings within the proximal phalanges. The common hinge-pins are designed to prevent the recurrence o f ulnar drift by sharing the forces on one finger with a neighbouringjoint, and together with the intramedullary bearings to prevent potentially loosening torques f r o m acting on the fixation o f the components. The metacarpophalangeal joint together with its surrounding structures is one of the principal targets of the rheumatoid process. In some cases drug treatment or early surgery can limit or postpone the destructive process, but many patients present for treatment with advanced deformities comprising instability, subluxation or dislocation, ulnar drift, and loss of function often associated with pain. Resection arthroplasty has achieved widespread acceptance in the management of these cases, with replacement of the metacarpal head by one of a number of available prosthetic devices. These fall into two basic categories; those designed to provide a spacer with some stabilising function and those designed to act as a more or less rigid hinge. Early post-operative results (i.e. up to 4 years) with the first group, which includes the Swanson and the Niebauer, are generally good, especially when their insertion is accompanied by soft tissue reconstruction for stability, but fracture of the devices has been a problem, and there is some concern about the longer-term recurrence of deformity. Haber (1979) showed reasonable correction o f deformity and adequate range of motion, but no improvement in pinch strength, in a relatively short (average 21 months, maximum 4 years) follow-up of 116 Swanson prostheses in 29 hands. Three hands (11.5%) showed 10 to 45 degrees of recurrent ulnar drift, but in each case the patient failed to wear splints as directed post-operatively. Three implants (2.6%) fractured but these were in the same hand of a patient who refused to wear her splints. Beckenbaugh (1976), in a 12 to 65 month follow-up (mean 32 months) o f 186 Swanson and 68 Niebauer prostheses, reported the recurrence of moderate to severe ulnar drift in 9.7 per cent. With the Swanson prosthesis the fracture rate was 26.6 per cent and clinical deformity, including subluxation, ulnar drift or rotation, was noted in 11.3 per cent of joints. With the Niebauer prosthesis the fracture rate was 38.2 per cent and clinical deformity was noted in 44.1 per cent o f joints. Other studies have found fracture rates for the Swanson prosthesis o f just under 2 per cent (Swanson, 1972), 2.8 per cent (Mannerfelt, 1975), 9 per cent (Ferlic, 1975) and 10 per cent (Hagert, 1975), and Ferlic (1975) reported 25 to 45 degrees of recurrent ulnar drift in 4 hands (10 per cent of series) followed for up to 60 months. B. WeightmanPh.D., Departmentof MechanicalEngineering,ImperialCollege,ExhibitionRoad, London SW7. D. M. Evans, F.R.C.S., Department of Plastic Surgery, WexhamPark Hospital, Slough TheHand-- Volume15

No. 1

1983

57

The Laboratory Development o f a New Metacarpophalangeal Prosthesis B. Weightman, D. M. Evans and D. Light

Fig. 1. The current design of the prosthesis (The metal components have a stepped intramedullary stem located in polyethylenebearing sleevesinserted in the proximalphalanges.) Prostheses such as the Swanson and Niebauer rely on the formation of thickened scar tissue radial to each " j o i n t " to provide a restraining force (Swanson, 1972, Madden, 1977, Haber,, 1979). But, since scar tissue always matures and stretches with time, late relaxation of this "ligament" must be anticipated, with thumb pinch and the ulnar pull of the long flexors and extensors tending to lead again to ulnar drift, which the implant is powerless to resist. GiUespie (1979) has reported that flexion and extension produces distortions in the Swanson hinge other than those intended, and this, together with any eventual tendency to recurrent ulnar drift, probably explains the incidence of disintegration of the prosthesis. Although the short-term incidence of fracture will undoubtedly be reduced by the recent introduction of a stronger and more tear-resistant silicon elastomer, the incidence of clinical deformity leading to fracture is likely to increase with time. Follow-up series longer than five years are needed to show the true incidence of implant fracture and recurrent deformity. The second more rigid group of prostheses widely used in the form of the Steffee, Schultz, Strickland, and St. Georg-Bucholz, require cement fixation which may loosen. Clinical experience with rigid-hinge knee and elbow replacements has shown that intramedullary stem fixation with cement is particularly susceptible to torsional loads, and after clinical trials of a metacarpophalangeal joint prosthesis Attenborough (1977) concluded that"cement will probably never be a satisfactory method of fixing prostheses in the distal metacarpal shaft or the proximal phalanges." Individual hinges which attempt to restrain lateral motion impose excessive stress on the hinge due to the large force involved in lateral pressure on the finger, and the St Georg cup has been shown to open under lateral pressure (Gillespie 1979). There may therefore be some limitation of the life of this generation of prostheses also, with removal problems due to the use of cement. We have designed a prosthesis whose aim is to overcome these problems, and present here the early development and laboratory testing. 58

The Hand-- Volume 15

No. 1

1983

The Laboratory Development o f a New Metacarpophalangeal Prosthesis B. Weightman, D. M. Evans and D. Light

~

X•Y

Fy Mx

I

F

Vz "

_%.

"~My

Ivy, Fig. 2.

normal compression force, Fz=medial-lateral shear force, Fx=dorsal-palmar shear force, My=internal-external torque, Mz= flexion-extension moment, Mx = abduction-adduction moment. Fig. 3 The forces and moments acting on the prosthesis. Fy =

DESCRIPTION OF THE PROSTHESIS

Figure 1 shows the prosthesis in its current form. Basically, it consists of four individual hinge joints connected in pairs by common hinge-pins, so that forces on one finger are shared with a neighbouring joint. The phalangeal component of each joint is made from a single piece of implantable metal alloy. At its proximal end is a bifurcated annular region lined with u.h.m.w, polyethylene bushes. Distally the component has a stepped intramedullary stem of circular cross-section which is designed to run in an u.h.m.w, polyethylene sleeve secured within a proximal phalanx. The u.h.m.w, polyethylene metacarpal component consists of a single annular region and an intramedullary stem. The annular region of this component fits between the bifurcated annular region of the phalangeal component and the two are connected by the curved common hinge-pin. Intramedullary fixation of the phalangeal sleeve and the metacarpal component is achieved by the use of finned u.h.m.w, polyethylene pegs introduced into precision reamed holes of diameter intermediate between the central core and outer fin diameters. The common hinge-pins are prevented from sliding laterally by snap fit polyethylene caps mounted after assembly. The development and testing of this prosthesis, which we now describe, has to date included (1) stress analysis of the device, as a whole and in parts, in relation to the forces acting on the normal metacarpophalangeal joint, (2) strength-tests of various components and of the method of fixation, and (3) experimental insertion into cadaveric hands, preceded and followed by measurement of the tendon forces required for flexion and extension. The Hand--Volume l5

No. 1

1983

59

The Laboratory Development o f a New Metacarpophalangeal Prosthesis B. Weightman, D. M. Evans and D. Light

~

S h e a r = 2.5 B e n d i n g =110

S h e a r = 6.8 B e n d i n g = 509

~

Comp.=13.8

o m p . = 25.8

C a m p . = 8"8

Camp. = 8"8

S h e a r = 6"8

S h e a r =12.0

UNITS:-

MN/ 2 /m

Fig. 4 The magnitude of the stresses in the prosthesis. STRESS ANALYSIS

The loads acting across any joint can be resolved into six components; forces along each o f three mutually perpendicular axes, and moments about the same three axes. This concept is illustrated in Figure 2 for the proximal phalanx at the metacarpophalangeal joint. In general it is these forces which act upon, and hence induce stresses in, a joint prosthesis. Figure 3 shows, schematically, two views (dorsal-palmar and medial-lateral) of one of the prostheses being acted upon by these forces. The torque My and the flexion-extension moment M z are not shown because they do not stress this prosthesis. That is, M produces rotation of the proximal phalanx on the intramedullary shaft, until the torque is carried by the soft tissue surrounding the joint, and M z is the flexion-extension moment. Force F in the diagram represents one of the principal effects of the second of each pair of linked prostheses; that is, it balances the abduction-adduction moment, M x, and hence prevents M x being transmitted to the proximal component. Linking the prostheses also tends to distribute the dorsal-palmar shear force F over two proximal components, and more importantly as far as proximal stem fixation is concerned, prevents torsional loads being applied to the proximal component. In order to ensure that the prosthesis will have adequate strength it is necessary to (i) have a reasonably accurate estimate of the magnitude of the forces shown in Figure 3, and (ii) calculate the stresses produced by them. Unfortunately, the literature contains only a limited number of studies of the forces transmitted across the metacarpophalangeal joint, and the results from these differ considerably. After Y

60

.

.

.

.

The Hand-- Volume 15

No. 1

1983

The Laboratory Development o f a New Metacarpophalangeal Prosthesis B. Weightman, D. M. Evans and D. Light

a careful review of the literature and our own computer-aided analysis of index finger forces during pulp pinch (Weightman and Amis, 1982) it was decided to use the results obtained by Berme, (1977) in the design of the prosthesis. These workers calculated average values for F , F , F and M of 46.5N, 170N 13N and 1.35Nm* respectwely, in the index fingers of four young healthy females when applying maximum effort to turning an instrumented tap. The magnitudes of the stresses produced by these forces are shown in Figure 4 for the proposed two sizes of prosthesis. Details of the dimensions of the prostheses and of the stress analysis calculations are given in an appendix. Compared with the yield stresses of u.h.m.w, polyethylene and cast cobaltchromium alloy (approximately 22 and 450 M N/ m 2 respectively), the stresses likely to occur in the large size prosthesis appeared generally acceptable (but see the following section). The only stress which initially caused concern was the maximum compressive stress in the polyethylene bearing bushes (13.8 MN/m2). It seemed unlikely that high wear rates will occur in metacarpophalangeal joint prostheses since normal functioning usually involves motion under low load f o l l o w e d b y loading, rather than motion under high load as in joints such as the hip, but permanent plastic deformation might be caused by compressive stresses of this magnitude. For this reason it was decided to study experimentally the ability of the bushes to resist permanent deformation and these tests are described in the following section of the paper. The bending stress in the intramedullary stem of the small prosthesis, and the compressive stress in its bearing bushes clearly indicated this size implant to be unsuitable for use in index fingers. Our current thinking is that if the large size prosthesis proves to be too large for all index (and middle) fingers, it will be necessary to introduce a medium sized version so that the small size can be restricted to ring and little fingers. 9 9

x

y

z

x

TESTS OF RESISTANCE TO LATERAL LOADING The previously described theoretical stress analysis of the prosthesis suggested that the bearing bushes in the phalangeal component might suffer plastic deformation if large lateral forces were applied to the finger. When lateral loads are applied to a finger distal to the proximal interphalangeal joint with this joint flexed, as in radial pinch, the lateral force produces a torque about the proximal phalanx as well as an abduction-adduction moment about the metacarpophalangeal joint. With the prosthesis under discussion such loading will cause rotation of the proximal phalanx (and bearing sleeve) about the intramedullary stem and this should limit the magnitude of the lateral force that can be applied. However lateral loading will not produce a torque about the proximal phalanx if the finger is fully extended or if theload is applied proximal to the proximal i,nterphalangeal joint, and the prosthesis must be capable of withstanding this form of loading. Walker (1975) estimated the maximum likely lateral moment about the metacarpophalangeal joint by slowly applying an increasing lateral force to the proximal interphalangeal joint of 20 male and 20 female volunteers until discomfort *value estimated from radial interosseous, lumbrical and radial collateralligament forces. The Hand-- Volume 15

No. 1

1983

6/

The Laboratory Development o f a New Metacarpophalangeal Prosthesis B. Weightman, D. M. Evans and D. Light

(a)

(b) increasing toad

cyclic load

It

' 'ii' ,,

\ \\\\\\\\

\ \\

\\\\\\\\

Fig. 5 Laboratory tests of the phalangeal components.

was felt. The average moments were 2.32 Nm (S.D.___0.69) for males, and 1.21 Nm (S. D. -F 0.37) for females. The following experimental work was carried out tO see if the large size phalangeal component could withstand lateral loads of this magnitude. After trial insertions of complete sets of prosthesis in cadaveric hands (see later section) implanted phalangeal components were subjected to a gradually increasing load applied to the distal head of the phalanx as shown schematically in Figure 5(a). Yielding was detected when the load-deflection curve became non-linear. Initial tests showed that the weakest part of the phalangeal component was the proximal end of the reduced diameter part of the intramedullary stem. Two prototype large components made of annealed stainless steel yielded here when the abductionadduction moment about the metacarpophalangeal joint reached values of 1.19 and 1.93 Nm. Increasing th.r diameter of this section of the stem from 2mm to 2.5mm gave values of moment at yield of 2.15 and 2.65 Nm, again for stainless steel. On the basis of these figures it was decided to make the metal phalangeal component of the prosthesis of titanium alloy (6% AI, 4%V) rather than of cast cobalt-chromium alloy, as had originally been planned (the stainless steel components were prototypes made only for laboratory testing). Since titanium alloy has a tensile yield stress of more than twice that of annealed stainless steel (approximately 825 MN/m 2 compared with approximately 300 MN/m 2) the above tests indicate that the maximum stress in the intramedullary stems of large size titanium alloy phalangeal components, subjected to the maximum likely lateral loads, will be of the order of half the yield stress of the material. In order to test the ability of the bearing bushes in the phalangeal component to resist permanent deformation dummy components were fitted with bushes and subjected to cyclic applications of a 50N lateral load as illustrated in Figure 5(b). The length of the dummy component was such that this force produced a moment at the bearing of 3.0 Nm. At the end of the test any angular " p l a y " resulting from deformation of the bearing bushes was measured. The tests produced a maximum angular " p l a y " of 0.5 degrees immediately after 1000 load cycles which completely disappeared ~/fter one hour. We conclude that inv i v o the bushes are unlikely to experience permanent deformation. JOINT SIMULATOR TESTING Although high wear in the phalangeal bearing bushes was thought unlikely to be a problem, and repetitive applications of an abduction-adduction moment had shown no significant permanent deformation of the bushes after 1000 cycles, it was considered prudent to test a complete phalangeal component (including bearing sleeve) in a joint simulator. 62

The Hand-- Volume 15

No. 1

1983

The Laboratory Development o f a New Metacarpophalangeal Prosthesis B. Weightman, D. M. Evans and D. Light

Fig. 6 A phalangeal component mounted in the joint simulator. Fig. 8 Experimentalassessment of flexion-extensionmotion. Figure 6 shows the machine constructed for this purpose. The phalangeal component was mounted on a bar, representing the common hinge-pin o f the prostheses, in a lubricant bath. The intramedullary stem was located in a polyethylene sleeve cemented into a metal tube representing the proximal phalanx. Two cords were attached medially and laterally to a cross-bar mounted on the tube at its distal end. These cords ran almost parallel to the long axis of the tube, around pulleys mounted on extensions of the hinge-pin, and then longitudinally to springbalances attached to the frame of the machine. The mounting tube was oscillated (flexion-extension) at 1 cycle per second through an arc of 90 degrees, by an electric motor connected to the distal end of the tube via a crank and connecting rod. During each cycle of the motion the tension in each loading cord varied as the angle of lap of the cords round the pulleys varied. Different diameter pulleys resulted in different tensions being produced in the two cords; the sum of the tensions produced a fluctuating compressive force across the joint, and the difference in tensions produced a fluctuating abduction-adduction moment. Test to date have. been run with Ringer's solution as lubricant, with pulleys giving a resultant joint compressive force fluctuating between 0 and 100N and an abduction-adduction moment between 0 a n d 0.85Nm. Total deformation of the bearing bushes (permanent deformation plus wear) was assessed periodically throughout the tests by measuring the angular play The Hand-- Volume 15

No. 1

1983

63

/O

The Laboratory Development o f a New Metacarpophalangeal Prosthesis B. Weightman, D. M. Evans and D. Light 0"3-

/

~

n*"

~

o

o

0.2

,o

k.0

9

~

0

0'1 ~ ~'e 9 ~ 99

0 ,e

....

0

0"2 0~ 016 0"8 1~0 112 BONESTRENGTH(kN)

Fig. 7 The results of fixation strength tests on two designs of phalangeal bearing sleeves.

between the component and an extended hinge-pro in a specially constructed apparatus. The component was rigidly held in a mounting jig and tare loads were applied to each end o f the extended hinge-pin in turn. Angular displacements of the pin were measured with clock-gauge indicators, and to date have shown a maximum increase in total angular play of 0.25 degrees after one million cycles. We conclude that repeated flexion and extension o f the prosthesis under load (even if it is experienced) will not lead to unacceptable wear. FIXATION STRENGTH

TESTS

The prosthesis was designed from its inception for use without cement; fixation was to be achieved using a miniaturised version of the finned-peg method of cementless fixation previously developed at Imperial College for the tibial component o f the I C L H total knee prosthesis. These pegs are made o f uhmw polyethylene and consist of a number of thin circumferential fins protruding from a central core. The fins are slit radially to form small cantilevers, so that when the peg is driven into a hole drilled in cancellous bone (of diameter intermediate between the outside and core diameters of the peg) they flex backwards to form a saw-tooth profile interlocked in trabeculae. The original design envisaged an all-polyethylene phalangeal component with an integral finned-peg for intra medullary fixation. However, preliminary tests of a miniaturised finned-peg showed that although a considerable tensile force was required to extract it from cancellous bone, the fixation had insufficient torsional strength to withstand the internal-external torque (M., in Figure 2) produced by normal activity. Thus, the finned-peg concept appeared su'itable for the metacarpal component, since this is protected against torsional loads by the pairing of the prosthesis, but unsuitable as an integral part of the phalangeal component. 64

The Hand-- Volume l 5

No. 1

1983

The Laboratory Development o f a New Metacarpophalangeal Prosthesis B. Weightman, D. M. Evans and D. Light

In order to solve this problem the phalangeal component was redesigned to include a metal intramedullary shaft which would run in a uhmw polyethylene bearing sleeve inserted in the phalanx. The phalanx and sleeve are then free to rotate on the metal shaft until the internal-external torque is carried by the soft tissue surrounding the joint. In this way the interface between the phalanx and sleeve is not subjected to torsion. Although it is extremely difficult to envisage the phalangeal bearing sleeve being subjected to extraction forces i n - v i v o , an attempt was made to optimise the outside profile of the sleeve by maximising the force necessary to extract it from tibial plateau cancellous bone. The initial idea was a conical sleeve withcircumferential identations to be pushed into a prepared (undersize) conical hole. A number of tests were carried out with different shape and size indentations. Figure 7 shows the fixation strengths achieved with the best of these (plotted against bone strength assessed by measuring the force necessary to extract a bone screw), together with the results for our current design--a "stepped finned sleeve". At a later stage of the development program complete sets of metacarpophalangeal prostheses were implanted in cadaver hands (see Cadaveric Testing) in order to develop the surgical technique and study the range of motion achieved. After these experiments the fixation strengths of the metacarpal components were measured. In the first cadaveric trial all four metacarpal components had pegs with a core diameter of 3.0mm, an outside diameter of 5.0mm, a fin thickness of 0.3mm, and a fin pitch of 1.4mm. They had been inserted into drilled holes of 4.0mm diameter, and it required forces of 28, 42, 110, and 42N to extract them from the index, middle, ring and little metacarpals respectively. By indicating that the fixation strength of small pegs in large metacarpals was less than that of small pegs in small metacarpals, this test suggested the need for a second, larger, size of peg for use in index and middle fingers. In the second trial the original size pegs were used for the ring and little fingers and pegs with a 4.0mm core diameter, a 6.0mm outside diameter, and the same fin thickness and pitch as before were inserted into 5.0mm diameter holes in the index and middle finger metacarpals. The pull-out forces measured after this trial were 75, 75, 80 and 80 N for the index, middle, ring and little fingers respectively. On the basis of the above tests we concluded that the finned-peg technique of cementless fixation is suitable for both the phalangeal bearing sleeve and the metacarpal component of the prosthesis. JOINT KINEMATICS In order to function satisfactorily prostheses must attempt to reproduce the kinematics of the natural joint. As far as the present design of prosthesis is concerned it was important to establish (1) the validity of having a fixed flexion-extension axis, (2) the location of this axis relative to the metacarpal and phalanx, and (3) the functional consequences of eliminating lateral movement. The weight of present evidence (e.g. Flatt, 1969, Youm, 1978, and Unsworth, 1979), indicates that the natural metacarpophalangeal joint does have a fixed axis of rotation in the saggital plane, although there are opposing views (e.g. Walker, 1975). This conclusion is obviously shared by the designers of other prostheses with fixed axes of rotation (e.g. Steffee, St. Georg-Bucholz and Strickland). The Hand--Volume l5

No. 1

1983

65

The Laboratory Development o f a New Metacarpophalangeal Prosthesis B. Weightman, D. M. Evans and D. Light r 90~ / Angle of / Flexion _...~80 o

o without prosthesis 9

with prosthesis

30~ 2O I

10

I

I

I

I

8

6

4

2

Extensor

force

( N )

I

2

I

i

I

4 6 8 Flexor force { N )

I

I0

Fig. 9 The results of the flexion-extensionstudies9 Thus, in relation to the present design it is important only to determine the optimum position o f this axis in the saggital plane. Flatt, 1969 demonstrated the importance of correct positioning of the axis in the dorsal-volar direction but, unfortunately, gave no quantitative data. The design of the present prosthesis has therefore been based on the results published by Unsworth (1979). In their sample (36 male and 24 female cadaveric metacarpophalangeal joints) these workers found that on average the flexion-extension axis was 2.63mm below the centre of the metacarpal intramedullary cavity and 1.19mm below the centre of the phalangeal cavity. The prosthesis has been designed with values of 2.5 and 1.0mm for both sizes, in the belief that the surgical population will be more biased towards females than the above sample. A cadaver test has been carried out to study the effect on the relative strengths of the flexors and extensors of inserting such a prostheses, and this is described in the following section. The elimination of lateral movement will prevent recurrence of ulnar drift, provided the intramedullary stem fixation is adequate, but the likely effects of this on hand function are open to doubt and will require in v i v o analysis. While loss of lateral finger movement in the normal hand would seriously limit function, we suggest that the extent of functional loss and instability in the rheumatoid hand in which such surgery is indicated justifies this compromise for the sake of long-term stability. CADAVERIC TESTING At various stages in the development of the prosthesis prototypes have been inserted into cadaver hands, and on each occasion design modifications have been made, culminating in the present design. 66

The Hand-- Volume 15

No. 1

1983

The Laboratory Development o f a New Metacarpophalangeal Prosthesis B. Weightman, D. M. Evans and D. Light

Difficulty was experienced in assembly of all four components on a single transverse bar, but only while attempting to pass the bar through the ring and little finger joints. Furthermore, the use o f a single bar introduced the possibility of the " h a n d shake" phenomenon where the metacarpals could be compressed together, allowing the bar to protrude at each end. To overcome these two difficulties the joints are now assembled in pairs on two separate short bars, without losing the benefit of force F (Figure 3) which is required to resist the adduction-abduction moment (Mx). This arrangement will also allow some independent flexion of the fourth and fifth carpo-metacarpal joints for cupping of the palm. One cadaver hand was prepared for extension and flexion motion assessment as follows. The extensor tendons were dissected out and held by a mass suture, and all eight digital flexor tendons similarly treated. The hand was rigidly mounted on a wooden frame using two lag screws passed through the carpus on either side, and including the base of the second metacarpal. Lines from the flexor and extensor tendons were passed over pulleys so that weights could be used to apply flexion and extension forces (Figure 8). In order to concentrate flexion forces at the metacarpophalangeal joints it was necessary to splint the interphalangeal joints in extension. Force was applied initially to the extensor tendons in 100 gram increments up to 1kg. Joint angles were measured after each increment. Then similar increments were applied to the flexor tendons while maintaining the full extensor force. Finally, the extensor force followed by the flexor force, were incrementally reduced to zero. The open circles in Figure 9 show the results obtained, plotted as angle of metacarpophalangeal joint flexion (the mean of four) versus tendon force. The current design of prosthesis was inserted and the experiments repeated. The results of this test are shown by the closed circles in Figure 9. In general, the relative effectiveness of the flexor and extensor tendon forces was largely unchanged by insertion of the prosthesis. That is, the slopes of the angular displacement versus tendon force curves, with and without the prosthesis, were similar. This was particularly true for the extensors (the curves on the left of Figure 9); dearly for angles of flexion between approximately 20 and 80 degrees the moment arms of the extensor tendons about the axes of the prosthesis were almost identical to those about the axes of the natural joints. There was, however, some difference between the effectiveness of the flexors with and without the prosthesis (the curves on the right of Figure 9). In the normal hand, with an extensor force of 9.8N, an increase in flexor tendon force from 6N to 10N produced an increase in angle of flexion of 8 degrees (i.e. from 32 degrees to 40 degrees). After the prostheses had been fitted the same increase in flexor force produced an increase in angle of flexion of 33 degrees (i.e. from 32 degrees to 65 degrees). Clearly, in the middle of the flexion range, the moment arms o f the flexors were increased by the insertion of the prosthesis. This, together with the previous finding that the extensor moment arms were unaffected by the insertion of the prosthesis, suggests that the flexor tendons were bowstringing away from the joint after the insertion of the prosthesis. We conclude that in this model the prosthesis allows at least as free metacarpophalangeal joint movement as the normal joint, and that any limitation of excursion which may occur in v i v o will be due to soft tissue factors. TheHand--Volume l5

No. 1

1983

67

The Laboratory Development o f a New Metacarpophalangeal Prosthesis B. Weightman, D. M. Evans and D. Light

FUTURE WORK

A set of special instruments (mainly reamers and introducing tools) has been developed during cadaveric testing. Further post-mortem work may be necessary in order to refine these instruments, but apart from this we feel justified, on the basis of the laboratory studies reported in this paper, in starting a pilot series of clinical trials in the near future. That is, the tests indicate, as far as such tests ever can, that the prosthesis should be capable of withstanding in-vivo conditions. ACKNOWLEDGEMENT The work described in this paper was supported by a grant from the Science and Engineering Research Council.

REFERENCES ATTENBOROUGH, C. G. (1977) Joint replacement in the upper limb. In, Joint Replacement in the Upper Limb, published by Mechanical Engineering Publications Ltd. for the Institution of Mechanical Engineers, London. BECKENBAUGH, R. D., DOBYNS, J. H., LINSCHEID, R. L. and BRYAN, R.S. (1976) Review and Analysis of Silicone-Rubber Metacarpophalangeal Implants. The Journal of Bone and Joint Surgery, 58A: 483-487. BERME, N., PAUL, J. P. and PURVES, W. K. (1977). A Biomechanical Analysis Of The Metacarpophalangeal Joint. Journal of Biomechanics, 10: 409:412. FERLIC, D. C., CLAYTON, M. L. and HOLLOWAY, M. (1975) Complications of Silicone Implant Surgery in the Metacarpophalangeal Joint. The Journal of Bone and Joint Surgery, 57A: 991-994. FLATT, A. E., and FISCHER, G. W. (1969) Biomechanical Factors in the Replacement of Rheumatoid Finger Joints. Annals of the Rheumatic Diseases 28: Supplement, No. 5: 36-41. GILLESPIE, T. E., FLATT, A. E., YOUM, Y., and SPRAGUE, B. L. (1979). Biomechanical evaluation of metacarpophalangeal joint prosthesis designs. The Journal of Hand Surgery, 4: 508-521. HABER, J., BOSWlCK, J. A., and PHELPS D. B. (1979). The Role of Soft Tissue Reconstruction in Flexible Implant Arthroplasty of the Metacarpophalangeal Joint. Clinical Orthopaedics and Rel~ited Research, 140: 178-183. HAGERT, C-G., EIKEN, O., OHLSSON, N-M., ASCHAN, W., and MOVIN, A. (1975). Metacarpophalangeal Joint Implants. Scandinavian Journal of Plastic and Reconstructive Surgery 9:147-157. MADDEN, J; W., DE VORE, G., and AREM, A. J. (1977) A rational postoperative management, program for metacarpophalangeal joint implant arthroplasty. The Journal of Hand Surgery, 2: 358-366. MANNERFELT, L. and ANDERSSON, K. (1975). Silastic Arthroplasty of the Metacarpophalangeal Joints in Rheumatoid Arthritis. The Journal of Bone and Joint Surgery 57A: 484-489. PURVES, W. K., BERME, N., and PAUL, J. P. (1979). Finger joint biomechanics. In Disability, edited by Kenedi, R. M., Paul, J. P. and Hughes, J. published by MacMillan. SWANSON, A. B. (1972). Flexible Implant Arthroplasty for Arthritic Finger Joints. The Journal of Bone and Joint Surgery, 54A: 435-455. UNWORTH, A. and ALEXANDER, W. J. (1979). Dimensions of the metacarpophalangeal joint with particular reference to joint prostheses. Engineering in Medicine, 8" No. 2., 75-80. WALKER, P. S. and ERKMAN, M. J. (1975). Laboratory Evaluation of a Metal-plastic Type of Metacarpophalangeal Joint Prosthesis. Clinical Orthopaedics and Related Research, 112: 349-356. WEIGHTMAN B., and AMIS, A. A. (1982). Finger joint force predictions related to the design of joint replacements. Journal of Biomedical Engineering, 4: 197-205. YOUM, Y., GILLESPtE, T. E., FLATT, A. E. and SPRAGUE, B. L. (1978). Kinematic Investigation Of Normal MCP Joint. Journal of Biomechanics, 11: 109-118.

68

The Hand-- Volume 15

No. 1

1983

The Laboratory Development o f a New Metacarpophalangeal Prosthesis B. Weightman, D. M. Evans and D. Light

l~._~

b7

d1

~"

"ram) d1 d2

d2

G b

d3

Large 5"0 4.0 4,0 5,0 2,5

Small 3-0 3-0 4-0 3.0 2-5

121 Fig. 10The dimensions of the prostheses, (two sizes). APPENDIX STRESS ANALYSIS

The stresses in Figure 4 were calculated using the following expressions:-(1) Shear stress in phalangeal stern = 4(F2 + Fz2)'~/nda2 (2) Bending stress in phalangeal stem = 32Mx/ndl 3 (3) Maximum compressive stress in phalangeal bearing bushes = (Fy2 -~ Fx2)v2/2ad3 + 3Mx(a + b)/2ad3(a 2 + 3ab + 319) (4) Shear stress in metacarpal stem = 4(F,2 + F2)V'/xd22 (5) Compressive stress in metacarpal component = (Fy2+ F2)/2bd3 In these expressions:--

(i) Fx, Fy, F z and

lVlx are shown in Figure 3, and have the values 46.5N, 170N, 13N and 1.35Nm respectively.

(ii) F , = E~cos0, where 0 is the angle of metacarpophalangeal joint flexion, and has the value of 43.75 degrees (mean value from Berme, Paul and Purves, 1977). (iii) dl, d2, d 3, a and b have the meanings and values shown in Figure 10. In expression (3) the first term represents the direct compression of the bearing bushes in the phalangeal component produced by a combination of F and Fy. The second term represents the maximum compression produced (at the outer edges o f the bushes) by the moment M . This term was derived by assuming a linear distribution o f stress across the bearing. In both parts of the expression the projected area of the bushes (diameter x length) has been used. The Hand-- Volume 15

No. 1

1983

69