Journal of Orthopaedic Research 94%53 Raven Press, Ltd., New York 0 1991 Orthopaedic Research Society

Biomechanical Evaluation of a Biodegradable Composite as an Adjunct to Internal Fixation of Proximal Femur Fractures Pierre M. Witschger, Tobin N. Gerhart, Janet B. Goldman, Laura E. Edsberg, and Wilson C. Hayes Orthopaedic Biomechanics Laboratory, Department of Orthopaedic Surgery, Charles A . Dana Research Institute, Beth Israel Hospital and Harvard Medical School, Boston, Massachusetts, U.S.A.

Summary: Internal fixation of comminuted unstable fractures of the severely osteoporotic proximal femur is sometimes supplemented with polymethylmethacrylate (PMMA). We here report an in vitro biomechanical evaluation of a biodegradable particulate composite that might be used for similar purposes. The composite includes a matrix phase consisting of a hydrolyzable prepolymer [polypropylene fumarate (PPF)] cross-linked with methacrylate monomer, and a particulate phase consisting of tricalcium phosphate and calcium carbonate. We implanted dynamic hip screws in 22 cadaveric proximal femora and measured the yield load for an oblique force applied to the femoral head. The hip screws were then reinforced with either PMMA or the PPF composite and tested again. On the basis of analysis of variance, the average increases in yield load for PMMA and PPF reinforcement of 1,750 and 1,130 N were statistically significant (p < 0.00005), suggesting that both materials enhance congruence between implant and bone and thereby increase the projected load-bearing area of the implant. The increase in yield force with PMMA was slightly higher than the increase with PPF (p < 0.05), but both values after reinforcement were close (3,790 2 561 N for PMMA vs. 3,240 2 669 N for PPF). If we can demonstrate that appropriate rates of degradation, bony ingrowth, and static and fatigue properties can be achieved in vivo with this system, our data suggest that this PPF composite may have potential as an adjunct to the internal fixation of unstable fractures of the osteoporotic hip. Key Words: Hip fracture-Osteoporosis-Internal fixation-Biomaterials.

devices has generally resulted in fewer complications than conservative treatment (1,12,15,16,21). Nevertheless, in patients over 65 years of age, the mortality rate after hip fracture has been as high as 40% during the first postoperative year, with substantial functional losses 6 months after injury (1,2,6,16,19,20,22).This high complication rate is in part due to the failure of internal fixation devices to maintain purchase in the osteoporotic bone. Polymethylmethacrylate (PMMA) is widely used as an adjunct for internal fixation of pathologic hip fractures. Good results have been reported (13,21) with this method, stressing the benefits of pain re-

In the elderly patient suffering from severe osteoporosis, comminuted unstable intertrochanteric fractures remain a great therapeutic challenge. Operative stabilization with modern internal fixation Received May 12, 1989; accepted June 1, 1990. The present address of Dr. P. M. Witschger is Experimental Orthopaedics, Department of Orthopaedic Surgery, University of Bern, CH-3010 Bern, Switzerland. The present address of Dr. T. N. Gerhart is Boston Orthopaedic Group, Inc., 1269 Beacon Street, Brookline, MA 02146, U.S.A. Address correspondence and reprint requests to Dr. W. C. Hayes at Orthopaedic Biomechanics Laboratory, Beth Israel Hospital, 330 Brookline Avenue, Boston, MA 02215, U.S.A.

48

BIODEGRADABLE REINFORCEMENT OF PROXIMAL FEMUR FRACTURES

lief, improved mobility, and ease of nursing care. Some authors have also recommended PMMA reinforcement to enhance stability of internal fixation devices used for hip fracture patients in severely osteoporotic bone, particularly when distal interfragmentary support is lacking. Harrington et al. (13) reported a series of 42 unstable intertrochanteric fractures that were treated by open reduction and internal fixation supplemented with polymethylmethacrylate. Schatzker et al. (25) used PMMA as an adjunct to fixation in a series of 28 unstable intertrochanteric fractures in patients without tumors but with severe osteoporosis. The implants included Jewett nail plates, Smith-Peterson nails, McLaughlin plates, and AO/ASIF condylar plates. There was only one nonunion and two patients required reoperation due to non-cement-related causes. All of the patients were pain free early after surgery and began full weight-bearing the day after the operation (25). Bartucci et al. (1) compared a group of 49 hip fracture patients surgically fixed by a compression screw to a group of 39 patients treated with both a compression screw and adjunctive methylmethacrylate placed in the proximal fragment. Of those patients followed, the rates of fixation complications were the same for the 18 stable fractures in both groups, while for the 38 unstable, comminuted fractures, the rate of fixation complications was about ten times lower when adjunctive methylmethacrylate cement was used. However, for reasons that were not explained, the methylmethacrylate group showed a lower Iowa hip score. Despite these encouraging reports, the use of PMMA as an adjunct to internal fixation is not free from potential problems. Since polymethylmethacrylate is not biodegradable, the fixation devices cannot be removed or easily replaced if revision surgery becomes necessary. Moreover, despite the fact that fracture healing can proceed normally in the immediate vicinity of PMMA, the presence of PMMA in the femoral neck and its extrusion to more peripheral locations may well compromise healing of proximal femoral fractures (1,13,25). Furthermore, an appropriately immobilized fracture should heal within several months, after which the methylmethacrylate becomes unnecessary. A biodegradable cement would have the same advantages as PMMA during the initial postoperative period if it could adequately augment implant fixation in osteoporotic trabecular bone. However, with the progression of healing, the biodegradable

49

cement could slowly degrade and thereby avoid interference with fracture healing. Thus, with time, the increasing rigidity of the healing bone would take over the function of the implant. In cases with complications of fracture healing, removal or replacement of the implant by a prosthesis would also be easier with a biodegradable material. Based on these considerations, we tested the mechanical behavior of dynamic hip screws in cadaveric proximal femora, first without cement and then after augmentation with polymethylmethacrylate and with a biodegradable composite bone cement under development in our laboratory (7-11). Our objective was to compare screw reinforcement provided by PMMA with that provided by the biodegradable composite. MATERIALS AND METHODS

Eleven pairs of femora were obtained within 2 days of death from cadavers aged 62 to 90 years of age (mean of 78 f 9 years). There were six females and five males, and no evidence of previous fractures or bone abnormalities other than osteoporosis. Using standard surgical technique, a 0.125 in. guide pin was inserted at an angle corresponding to the cervicodiaphyseal angle of the femur. The guide pin was controlled radiologically for its centricity in the femoral neck. The guide pin was then overdrilled with a lag screw channel reamer until the tip of the reamer was within 2.0 cm of the subchondral bone of the femoral neck. A dynamic compression hip screw (Free-lock 1181-15, Zimmer, Inc., Warsaw, IN, U.S.A.) with a 15.8 mm diameter thread, a 25.4 mm thread length, and a 100 mm total length was inserted in each proximal femur and the position checked radiographically. On the anteropostenor roentgenographic view, the proximal tip of the screw was located at or inferior to the center of the femoral head and within 1.5 cm of the subchondral bone of the head. On the lateral view, the screw was either central or slightly posterior. We then performed an osteotomy perpendicular to the longitudinal axis of the femoral neck, leaving the total length of the proximal fragment at 65 mm. The distal end of the screw was held in a fixture that allowed the load to be applied at an angle of 45" in the frontal plane, and 0" in the sagittal plane (Fig. 1). These angles were chosen assuming that the diaphyseal axis of the femur is inclined at 10" to the vertical, the cervicodiaphyseal axis is 110", and the force on the femoral head is inclined at 15" to the

J Orthop Res, Vol. 9, No. I , 1991

P . M . WITSCHGER ET AL.

50

SCREW lOOrnrn

FORCE F

1

FIG. 1. Setup for load application.

vertical. Using an INSTRON 1331 servohydraulic materials test system, the specimen was loaded under displacement control at 2 mm/s against an acetabular component until yield occurred. Yield was defined as the force at which the load-deformation curve deviated by 1.0 mm from linearity. After yield, the screw was removed and 6 g of one of the two cements was injected into the proximal end of the hole. The particulate composite cement was used in 12 of the femurs and polymethylmethacrylate in 10. The original screws were then replaced in their initial positions before the bone cements set by polymerization. After allowing 24 h of setting time at 100% humidity, the specimens were retested as previously described. The biodegradable cement used in these experiments is a modified version of the particulate composites described by Gerhart et al. (7-10). Formerly, the composites were bonded together by a biodegradable matrix phase that consisted of crosslinked gelatin. The matrix used in the current experiments is a hydrolyzable prepolymer (polypropylene fumarate, PPF) which can be cross-linked in vivo with a free radical reaction similar to that which is used with conventional PMMA. The particulate phase consists of tricalcium phosphate and

J Orthop Res, Vol. 9, No. 1, 1991

calcium carbonate. After implantation, the ester linkages of the prepolymer slowly hydrolize, thus weakening the matrix, and permitting bioerosion of the cement by the host. In vivo data on rates of degradation and bony ingrowth as well as data on in vitro and in vivo fatigue properties for this material are not yet available but are the subject of ongoing experimentation. For these in vitro studies, a single batch of PPF cement was prepared and divided into individual 6 g portions for materialsltesting and for fixation of the 12 femoral specimens. After activation by adding two drops of dimethyl-p-toludine (DMT) to each 6 g portion, the PPF cement was packed into the cavity left by the hip screw in the femoral specimen, and the hip screw replaced in position using the length of the distal protruding end to determine the depth of insertion. The cement was tested for material properties by making six 6 by 12 mm cylindrical specimens in Teflon molds and performing destructive, unconfined compression testing according to ASTM standard F45 1-76 for the compressive testing of polymethylmethacrylate. As with the testing of reinforced hip screws, specimens for materials testing were allowed to set for 24 h under 100% relative humidity prior to mechanical testing. The yield force before and after cement reinforement was analyzed by a two-factor repeatedmeasures analysis of variance. In the analysis, the grouping factor was the cement type (PMMA vs. PPF) and the repeated factor was the absence or presence of cement reinforcement. A test of the interaction between cement type and cement reinforcement was included in the analysis.

RESULTS From the compressive testing of standardized cylindrical specimens, the compressive strength of the PMMA used for reinforcement was 103 -+ 2.5 (SD) MPa. The compressive strength of the PPF particulate composite was 28.7 -+ 5.3 MPa. Figure 2 shows typical load-displacement curves for femoral specimens tested to failure both before and after reinforcement with either PMMA (Fig. 2A) or PPF (Fig. 2B). As was typical of the load-displacement curves for all tested specimens, there was an initial nonlinear region (associated with settling of the specimen in the loading fixtures), followed by a linear region, and then finally by a pronounced postyield regime indicating collapse at approximately constant force. For the 11 pairs of proximal femurs,

51

BIODEGRADABLE REINFORCEMENT OF PROXIMAL FEMUR FRACTURES

-

- 032x1

03212 PMMA Reinforced

-

ameinforced

@%a cement Remforced

Control

50001

m

53 -P

4000

3000

w

1000

1

1

,//

p

2000

n

1000

P -I

FIG. 3. Average yield force __ 4000

TLZ

TLI

-

PPP

Reinforced

r

I

3000

z " n 4

2000

PPF ( 2 SD) for hip screws without

cement and after reinforcement with methylmethacrylate (PMMA) and with a biodegradable particulate composite (PPF). Using a repeated-measures analysis of variance, the increase in yield force with cement reinforcementwas highly significant (p < 0.00005). This increase in yield force was not the same for the two types of cement (p < 0.01), indicating that the average increase of 1,750 426 N with PMMA was significantly greater than the average increase of 1,130 573 N with PPF.

*

I

I

*

0 4

DISCUSSION

1000

0

0

B

2

4

6

8

10

DISPLACEMENT (mm)

FIG. 2. Load-displacement curves before and after reinforcement with PPMA (A) and PPF (6).

there were no significant differences in yield force between right and left sides ( t = -0.03, p > 0.4), resulting in a pooled average yield force of 2,080 & 270 N. Figure 3 shows the yield loads for the two groups of specimens both before and after reinforcement with PMMA ( N = 10) and PPF ( N = 12). Before adding PMMA, the 10 femurs yielded at 2,050 f 316 N. After PMMA reinforcement, the average yield force increased to 3,790 2 561 N , with an average increase of 87%. To test reinforcement with the PPF particulate composite, the remaining 12 of the initially tested femurs were reinforced with PPF. Before reinforcement, the average yield force was 2,110 f 236 N. After PPF reinforcement, the yield force was 3,240 5 669 N, with an average increase of 53%. The increase in yield force with cement reinforcement was highly significant by the analysis of variance (p < 0.00005). This increase in force was not the same for the two types of cement (p < 0.05), indicating that the average increase of 1,750 f 426 N with PMMA was significantly greater than the average increase of 1,130 +- 573 N with PPF .

The force across the hip joint during single-legged stance can be three times the body weight and during normal gait this can increase to 4.5 times the body weight (24). For a 60 kg male, the latter corresponds to a load of approximately 2,700 N. During our tests, the average yield force for the unreinforced femora was 2,050 N , a value only slightly greater than that which occurs in single-legged stance and below that which occurs during normal gait. Our model system isolated the biomechanical interaction between hip screw and trabecular bone of the femoral head by eliminating the interfragmentary support that would be provided by distal fragments. While this relatively severe testing configuration does not represent the load sharing between implant and bone that occurs in well-reduced fractures, it most likely represents the situation for many severely comminuted fractures and for fractures where there is a lack of distal fragment support. In such cases, especially when there is concern about the density and strength of the trabecular bone in the femoral head, our results suggest that, until interfragmentary continuity is reestablished, hip loads should be limited to well below those involved in the normal activities of daily living. If this is not possible, then consideration should be given to reinforcing fixation of the hip screw (1,13,21,25).

J Orthop Res,

Val. 9,NO. 1, 1991

52

P . M . WITSCHGER ET AL.

As noted above, the principal disadvantage of using PMMA for reinforcement is that it is inert (thereby accumulating fatigue damage) and potentially can interfere with fracture healing. Recently, a number of biodegradable substitutes for PMMA have been investigated but most exhibit mechanical properties that are far inferior to PMMA (7,17, 23,26). We measured a compressive strength of about 100 MPa for the PMMA used in our experiments. By contrast, the compressive strength of our biodegradable PPF particulate composite was 28.7 MPa. However, despite this more than threefold difference in compressive strengths, the reinforcement provided by PPF was close to the reinforcement of PMMA under these testing conditions. This finding further suggests that trabecular bone is the weakest link in the implant-bone construct. Moreover, the load-displacement curves (both before and after reinforcement with either PPF or PMMA) for these implant constructs are quite similar to those reported previously for trabecular bone (5,14). We conclude, therefore, that in the absence of distal support, the failure characteristics of trabecular bone are the dominant factor in the failure of both unreinforced and reinforced hip screws. Indeed, our measured values for the yield force with the unreinforced hip screws bear this out. Assuming a conservative estimate for the projected loading area of the hip screw of 1.6 by 2.5 cm (thread diameter by thread length) and our measured average yield force of about 2,000 N results in a compressive strength value for trabecular bone of about 5 MPa, a value that is in the midrange of reported values (3,4,17,18). When used to augment fixation of hip screws in fractures of the osteoporotic proximal femur, both PMMA and the PPF composite most likely act by enhancing congruence between implant and bone, by directly reinforcing trabecular bone, and by distributing the weight-bearing loads over larger trabecular volumes. However, it is probably the surrounding trabecular bone that controls failure of the system. Prior to clinical testing, considerable work remains to test the biocompatibility and in vivo resorption rates of this PPF-based particulate composite system since a rapid degradation rate would compromise the usefulness of the material for this application. However, if appropriate in vivo rates of resorption, bony ingrowth, and static and fatigue properties can be demonstrated, our findings suggest that this material has potential for augmenting internal fixation of osteoporotic fractures of the proximal femur.

J

Orthop Res, Vol. 9, No. I , 1991

Acknowledgment: These experiments were supported by grants from the Centers for Disease Control (CDC CR102550), by the Mueller Foundation (PMW), and by the Maurice E. Mueller Professorship in Biomechanics at Harvard Medical School (WCH). We thank Dr. Elizabeth Myers for conducting the statistical analyses, Jeanine Goodwin for assistance in manuscript preparation, and Zimmer, Inc. for providing the hip fixation screws.

REFERENCES 1. Bartucci EJ, Gonzalez MH, Cooperman DR, Freedberg HI, Barmada R, Laros GS: The effect of adjunctive methylmethacrylate on failures of fmation and function in patients with intertrochanteric fractures and osteoporosis. J Bone Joint Surg [Am] 67:1094-1 107, 1985 2. Brotman HP: An analysis for the chairman of the select committee on aging. Every Ninth American, House of Representatives 97th Congress, Census Bureau and National Center for Health Statistics. Washington, D.C., National Center for Health Statistics, U.S. Government Printing Office, 97-332, 1982 3. Brown TD, Ferguson AB: Mechanical property distributions in the cancellous bone of the human proximal femur. Acta Orthop Scand 51:429-437, 1980 4. Carter DR, Hayes WC: Bone compressive strength: The influence of density and strain rate. Science 194:11741176, 1976 5. Carter DR, Hayes WC: The compressive behavior of bone as a two-phase porous structure. J Bone Joint Surg [Am] 59:954-962, 1977 6. Cobey JC, Cobey JH, Conant L, Weil UH, Greenwald WF, Southwick WO: Indicators of recovery from fractures of the hip. Clin Orthop 117:258-262, 1976 7. Gerhart TN, Hayes WC, Stem SH: Biomechanical optimization of a model particulate composite for orthopaedic applications. J Orthop Res 4:7fS5, 1986 8. Gerhart TN, Miller RL, Hayes WC: A biodegradable particulate composite bone cement. Trans Orthop Res Soc 1258, 1987 9 Gerhart TN, Miller RL, Kleshinski SJ, Hayes WC: In-vitro characterization and biomechanical optimization of a biodegradable particulate composite bone cement. J Biomed Muter Res 22:1071-1082, 1988 10. Gerhart TN, Miller RL, Noecker RJ, Renshaw AA, Hayes WC: In-vivo histologic and biomechanical characterization of a biodegradable particulate composite bone cement. J Biomed Muter Res 23:l-16, 1989 11. Gerhart TN, Roux RD, Horowitz G, Miller RL, Hadf P, Hayes WC: Antibiotic release from an experimental biodegradable bone cement. J Orthop Res 6585-592, 1988 12. Harrington KD, Jonston JO: The management of comminuted intertrochanteric fractures of the hip. J Bone Joint Surg [Am] 55:1367-1376, 1973 13. Harrington KD, Sim FH, Enis JE, Johnston JO, Dick FM, Gristina AG: Methylmethacrylate as an adjunct in internal fixation of pathological fractures. J Bone Joint Surg [Am] 58:1047-1054, 1976 14. Hayes WC, Carter DR: Postyield behaviour of subchondral trabecular bone. J Biomed Muter Res [Symp] 7537-544, 1976 15. Jensen JS, Sonne-Holm S, Tondevold E: Unstable trochanteric fractures. A comparative analysis of four methods of internal fixation. Acta Orthop Scand 51:94%962, 1980 16. Jensen JS, Tondevold E: Mortality after hip fractures. Acta Orthop Scand 50:161-167, 1979

BIODEGRADABLE REINFORCEMENT OF PROXIMAL FEMUR FRACTURES 17. Knauss P: Materials properties and strength behaviour of spongy bone tissue at the coxal human femur (part I). Biomed Tech 26:200-210, 1981 18. Knauss P: Material properties and strength behaviour of spongy bone tissue at the coxal human femur (part 11). Biomed Tech 26:311-315, 1981 19. Levin LF: Alterations in trabecular structure accompanying decreased bone density in the human proximal femur. Harvard University Bachelor’s Thesis, 1989 20. Miller CW: Survival and ambulation following hip fracture. J Bone Joint Surg [Am]60:930-933, 1978 21. Muhr C, Tscherne H, Thomas R Comminuted trochanteric femoral fractures in geriatric patients: The results of 231

22. 23. 24. 25. 26.

53

cases treated with internal fixation and acrylic cement. Clin Orthop 138:41-44, 1979 Muller ME, Allgower M, Willenegger H: Manual ofznternal Fixation, Philadelphia, Lea & Febiger, 160-161, 1973 Murray WR: Use of antibiotic containing bone cement. Clin Orthop 190:89-95, 1984 Paul JP: Approaches to design. Proc R. SOCLond 192:163172, 1976 Schatzker J, Haeri GB, Chapman M: Methylmethacrylate as an adjunct in the internal fixation of intertrochanteric fracture of the femur. J Trauma 18:732-735, 1978 Weber SC, Chapman MW: Adhesives in orthopaedic surgery. Clin Orthop 191:249-261, 1984

J Orthop Res, Vol. 9, No. 1, 1991