When a decision has been made to administer a bisphosphonate to a patient before or after a cemented joint replacement, there are two delivery options: (1) systemic (oral ingestion or intravenous injection) or (2) local. Although the first option has been used,3,17 to our knowledge, the second has only been proposed for use in humans by Peter et al19 who grafted zoledronate to a hydroxyapatite (HA) coating deposited on titanium implants inserted in the condyles of rats.
Local delivery of a bisphosphonate in total joint replacements must be tested in three distinct ways. The first is through clinical studies to determine the most efficacious set of conditions (specific bisphosphonate, mode of delivery, and dosage). The second is through biochemical studies to determine the mechanism(s) of action of the bisphosphonate (when incorporated into a cement) on cultures of osteoblast cells on human bone slices. The third is through in vitro studies to determine the influence of the bisphosphonate (when incorporated into the cement) on properties of the curing and the cured cement. The details regarding elution of the bisphosphonate from the cured cement into a biosimulating solution also must be evaluated. We are not aware of any clinical studies, but there are some biomechanical and in vitro studies.22,30 Mixing etidronate (EHDP) powder with a commercially available acrylic bone cement (CMW) (5-20 mg EHDP per gram of liquid monomer of CMW) inhibited bone resorption and macrophage-osteoclast differentiation induced by cement wear particles.22 Zenios et al30 found that when 15 mg liquid pamidronate was mixed with the powder and liquid monomer of a commercially available acrylic bone cement, there were reductions in the quasistatic four-point bending strength (39%) and modulus (53%) of the cured cement. “The use of liquid pamidronate mixed with the acrylic bone cement Palacos® R in order to reduce osteolysis is not recommended ….” This suggests that the physical form of the bisphosphonate may be important. To date, this issue has not been addressed. There have been no studies on preparing a bisphosphonate-impregnated bone cement and the fatigue performance of such a cement. The majority of total joint replacements are cemented.16 The impact of cement preparation conditions (eg, the method used to mix the powder and liquid monomer, and the temperature at which the powder is stored before mixing) on an array of in vitro properties of the cement and on the clinical outcomes of cemented arthroplasties have been documented.12,16 Given the nature of the loading that the cement experiences in vivo, dynamic mechanical properties of the cement likely will affect outcome. Of these properties, fatigue performance is one of the most important. There is evidence that compromised mechanical integrity of cemented total joint replacements by fatigue microcrack initiation and propagation in the cement zones can cause aseptic loosening.9,20,21 Aseptic loosening is the most common cause for revision of cemented arthroplasties.16,23
We postulated that it is comparatively easier to blend a bisphosphonate in powder form into acrylic bone cement than it is when the drug is in liquid form, and that the cement's fatigue life is decreased when the bisphosphonate is added in liquid rather than in solid form.
MATERIALS AND METHODS
Three sets of specimens were prepared using Cemex® XL acrylic bone cement (Tecres SpA, Verona, Italy). In the first set (control set), no modification was made to the cement composition. In the second set, (alend powder set), three Fosamax® tablets (70 mg; Merck & Co Inc, Whitehouse Station, TX) were ground to a fine powder using a commercially available ceramic mortar and pestle. That powder was blended thoroughly with the powder of the Cemex® XL (50 g) using a polyethylene spatula. In the third set (alend liquid set), 9 mL Fosamax® Oral Solution (Merck & Co Inc) was added to the Cemex® XL (50 g). In the second and third sets, the amount of alendronate used relative to the powder and liquid monomer of the cement was within the range cited by investigators of relevant studies.22,30 For all sets, the liquid monomer (18.3 mg) was in the Cemex® XL aliquot. Also for all sets, the blended powder was mixed with the liquid monomer of the cement to prepare dough used for fabricating solid circular cross-sectioned fatigue specimens.
The powder and monomer liquid for all sets were mixed using an open-bowl technique (hand/manual mixing), and the dough was injected into a four-celled silicone mold using a cement gun. Each cell had internal dimensions (Fig 1) that conformed to those given in ASTM F 2118-01a.1 After curing in the mold in ambient laboratory conditions (22° ± 1°C; relative humidity, 55 ± 2%) for approximately 20 minutes, the finished specimens were pressed out from the mold using a plunger. Specimens were accepted only if surface defects were not greater than 0.25 mm in major diameter in the gauge and/or transition sections, and if they had no internal defects with major diameter greater than 1 mm in those sections (as determined from radiographic examination in two orthogonal planes). The ratios of the accepted to the fabricated specimens were 72%, 70%, and 25% for the control, alend powder, and alend liquid sets, respectively. Among these specimens, 12 were selected randomly as the fatigue test specimens. Although exclusion of specimens with obvious flaws is likely to reduce the variability in the fatigue test results, there were two reasons why we used this method of selection. First, the selection means that differences in the specimen sets (eg, number of macropores) that may confound the comparison of the fatigue test results, as compared with the parameter of interest (eg, the presence or absence of alendronate in the cement as part of its composition), were eliminated or minimized. Second, the fatigue tests with the selected specimens measured intrinsic fatigue performance, rather than fatigue crack initiation and propagation from macropores. Some research groups5,6,10 have separated specimens with flaws before fatigue testing the others.
All specimens that were selected were lightly sanded using 600-grit emery paper (when necessary) and cleaned of all debris. All the dimensions of each specimen were measured with a digital caliper (Type Digi Kanon FMA-20, Nakamura Mfg Co Ltd, Tokyo, Japan). Only specimens that had each dimension within 5% of the nominal value (Fig 1) were used for fatigue testing. Selected specimens were stored for least 48 hours in ambient laboratory conditions before fatigue testing. In a fatigue test, the specimen was gripped in a custom fatigue test machine.
Eight hundred milliliters phosphate buffered saline (PBS) (pH 7.4; Gibco Invitrogen Corp, Grand Island, NY), at 37 ± 1°C, was contained in an environmental test chamber. During the test, the specimen was immersed in the chamber and subjected to fully reversed tension compression loading corresponding to a stress of ± 15 MPa at a frequency of 2 Hz until it fractured at a number of cycles (Nf). During the test, a waxed paper skirt was placed snug on top of the chamber to reduce the amount of PBS that evaporated. When necessary, more PBS was added to the chamber to ensure that the volume of PBS was always 800 mL. Only one stress level was used because the intention was not to obtain an S-Nf curve for each set, but to compare their indices of fatigue performance at a stress level used previously.13
The Nf results were treated using a statistical method to analyze extreme value data (the three-parameter Weibull method)28 with the aid of a commercially available software package (Mathlab® Version 6.5; The MathWorks, Inc, Natick, MA). This was done to yield estimates of three parameters: (1) b (which is a measure of the dispersion of the Nf results, such that the smaller b is, the larger the dispersion is); (2) No (the minimum or guaranteed fatigue life, which is the number of fatigue stress cycles above which all specimens have 100% probability of survival); and (3) Na (the characteristic fatigue life, which is the number of stress cycles below which lie 63.2% of the Nf results). These estimates then were combined to compute an overall index of the material's fatigue performance known as the Weibull mean (NWM).24
The method of fatigue testing used was an amalgam of some elements of ASTM F 2118-01a1 (test specimen configuration and dimensions; stress of ± 15 MPa at a frequency of 2 Hz; testing in PBS, at 37 ± 1°C; statistical analysis of the Nf results after logarithmic transformation; and analysis of the Nf results using the three-parameter Weibull method28) together with a recommendation presented by Lewis and Sadhasivini14 regarding the minimum number of test specimens (11, when the Nf results are analyzed using the three-parameter Weibull method) and specimen-aging conditions that have been used previously.13 Because we did a parametric study, all details of the methods were the same for testing all three sets.
After the fatigue tests, the porosities of the fatigue test specimens were determined using polished and stained discs (nominal height and diameter were 5 mm) cut from the central reduced section of the specimens, an optical microscope, a video digitizer, and image analysis software housed in a personal computer. Six discs were cut from each specimen, and disc measurements were made on eight fields. Each was approximately 3 mm2 and was selected at random.
The particle size distribution and mean particle size of the alendronate powder used in the alend powder set were determined using a laser diffraction system (Sympatec Particle Size Analyzer, Model HDD200, Sympatec GmbH, Golar, Germany), whereas its morphologic features and those of the fracture surfaces of the fatigue test specimens in each set, were obtained using an environmental scanning electron microscope (Model XL30, Philips, Achtsewed, The Netherlands) operated at an acceleration voltage of 15 kV.
For pair-wise comparison of the fatigue results (ln Nf) and the porosities of the sets, a one-way analysis of variance (ANOVA) was used in each case. Significance was set at the < 0.05 level.
No problems were encountered in preparing the specimens in either the control or alend powder sets. The polymerization reaction between the cement powder and the liquid monomer was completed within 10 minutes, with all the powder having reacted with the monomer. In contrast, while preparing the specimens in the alend liquid set, the liquid monomer remained immiscible with the cement powder for approximately 20 minutes, the dough was inhomogeneous (with segregation of the cement particles clearly visible in it), and the liquid alendronate interfered with the polymerization reaction which led to a polymerization process that took approximately 30 minutes to complete. The appearances of the final specimens in each set (Fig 2) reflect these points. The quality of the fabricated specimens was a concomitant of the extent to which the prepared cement powder reacted completely with the liquid monomer.
The fatigue test results (when expressed as ln Nf) were similar for the control and the alend powder sets (Table 1; Fig 3). In contrast, the results for the control and alend powder sets were greater (p < 0.0001) than the alend liquid set (Table 1; Fig 3). These trends are consistent with those for the estimates of the overall index of fatigue performance, NWM, with these estimates being 144,995; 88,803; and 4888 cycles for the control, alend powder, and alend liquid sets, respectively.
The trends in the specimen preparation and fatigue results paralleled the porosity results (9.32 ± 0.75; 9.66 ± 0.81; and 17.58 ± 0.45% for the control, alend powder, and alend liquid sets, respectively) and the microstructural evidence. There were fewer fatigue fractures initiated from stress risers (clusters of unreacted powder in the matrix) in the control and alend powder specimens than in the alend liquid specimens (Fig 4).
The alendronate powder used in the alend powder set contained a large number of small particles and a small number of larger ones, with the mean particle size being 34.5 ± 3.5 μm (Fig 5).
The literature on the in vitro properties of bisphosphonateimpregnated bone cements22,30 is limited by the lack of use of solid and liquid forms and reporting of only bending strength and bending modulus (quasistatic properties).30 We used powder and liquid forms of a bisphosphonate and determined the dynamic mechanical properties (fatigue). Our results supported both hypotheses. These findings combined with the results of Zenios et al30 suggest that if a bisphosphonate is to be incorporated into a bone cement for use in anchoring a total joint replacement, the bisphosphonate should be in powder form rather than liquid form. This suggestion is made cautiously because it is based on the results of two studies on two bisphosphonates. The bisphosphonate family is large and each has its own characteristics, so they must be considered individually.
Our study has four limitations. First, we used only one method of preparing the alendronate powder, yielding a specific particle size distribution and morphologic features. Second, only one method was used to mix the cement powder (with or without the alendronate) and the liquid monomer. The influence of mixing method on the properties of curing and cured acrylic bone cement is well known.12 Investigations of the influence of preparing the alendronate powder, the alendronate powder particle size distribution and morphologic features, and the cement mixing method are areas for future research. The third limitation was that the fatigue results were obtained at only one stress level. Of particular interest would be obtaining the S-Nf curve for each set, which would involve determining fatigue life at lower stress levels such as 12.5 or 10 MPa. The fourth limitation was that the porosities of the specimens were determined using a two-dimensional method. Because most cracks propagate in the bulk of the cement specimen,18 a nondestructive three-dimensional method of imaging the bulk of the specimen (such as microcomputed tomography11 or high resolution synchrotron xray computed tomography15) would be a better method for determining porosity. Given the ambit of the study, these limitations do not degrade our findings.
When alendronate was added to Cemex® XL acrylic bone cement, the fatigue performance of the cured cement was reduced when the bisphosphonate was added as a liquid, but not when it was incorporated as a powder that was blended with the cement powder. We think the trend in these results is a consequence of the difference in the quality of the prepared specimens; specifically, in the number of stress risers (clusters of unreacted powder in the matrix) in the specimens. There were more of these in the alend liquid set specimens than in the control or alend powder specimens. Stress risers have a negative impact on the in vitro fatigue life of acrylic bone cement.7,18 Our findings should guide orthopaedic surgeons when they choose to use alendronate-impregnated acrylic bone cement to anchor a total joint arthroplasty.
We thank Dr. Scott Sadler (Lexington, TN) for donating the alendronate sodium.
1. American Society for Testing and Materials (ASTM). Standard F 2118-01a. Test method for constant amplitude of force controlled fatigue testing of acrylic bone cement materials. 2002 Annual Book of ASTM Standards. Medical Devices and Emergency Medical Devices
. Vol 13.01. West Conshohocken. PA: ASTM International
2. Anderson JJ, Woelffer KE, Holtzman JJ, Jacobs AM. Bisphosphonates for the treatment of Charcot neuroarthropathy. J Foot Ankle Surg
3. Bhandari M, Bajammal S, Guyatt GH. Griffith L, Busse JW, Schunemann H, Einhorn TA. Effect of bisphosphonates on periprosthetic bone mineral density after total joint arthroplasty. J Bone Joint Surg
4. Cremers SC, Lodder MC, Den Hartigh J, Vermeij P, Van Pelt P, Lems WF, Papapoulos SE, Dijkmans BA. Short term whole body retention in relation to rate of bone resorption and cartilage degradation after intravenous bisphosphonate (pamidronate) in rheumatoid arthritis. J Rheumat
5. Cristofolini L, Minari C, Viceconti M. A methodology and criterion for acrylic bone cement fatigue tests. Fatigue Fract Eng Mater Struct
6. Cristofilini L, Minari C, Viceconti M. Reply to Letter to Editor. Fatigue Fract Eng Mater Struct
7. Dunne NJ, Orr JF, Mushipe M, Eveleigh R. The relationship between porosity and fatigue characteristics of bone cements. Biomaterials
8. Fleisch H. Bisphosphonates in osteoporosis. Eur Spine J
9. Jasty M, Maloney WJ, Bragdon CR, O'Connor DO, Haire T, Harris WH. The initiation of failure in cemented femoral components of hip arthroplasties. J Bone Joint Surg
10. Krause WR, Grimes LW, Mathis RS. Fatigue testing of acrylic bone cements: statistical concepts and proposed test methodologies. J Biomed Mater Res
11. Kurtz SM, Villarraga ML, Zhao K, Edidin AA. Static and fatigue mechanical behavior of bone cement with elevated barium sulfate content for treatment of vertebral compression fractures. [Erratum in: Biomaterials
. 2005;26:5926.] Biomaterials
. 2005;26:3699- 3712.
12. Lewis G. Properties of acrylic bone cement: state of the art review. J Biomed Mater Res
13. Lewis G. Fatigue testing and performance of acrylic bone-cement materials: state-of-the-art review. J Biomed Mater Res B Appl Bio-mater
14. Lewis G, Sadhasivini A. Estimation of the minimum number of test specimens for fatigue testing of acrylic bone cement. Biomaterials
15. Maire E, Buffiere J-Y, Salvo L, Blandin JJ, Ludwig W, Letang JM. On the application of x-ray microtomography in the field of materials science. Adv Eng Mater
16. Malchau H, Herberts P, Eisler T, Garellick G, Soderman P. The Swedish Total Hip Replacement Register. [Erratum in: J Bone Joint Surg Am
. 2004;86:363.] J Bone Joint Surg
. 2002;84(suppl 2):2-20.
17. Morris CD, Einhorn TA. Bisphosphonates in orthopaedic surgery. J Bone Joint Surg
18. Murphy BP, Prendergast PJ. On the magnitude and variability of the fatigue strength of acrylic bone cement. Int J Fatigue
. 2000;22:855- 864.
19. Peter B, Pioletti DP, Laib S, Bujoli B, Pilet P, Janvier P, Guicheux J, Zambelli PY, Bouler JM, Gauthier O. Calcium phosphate drug delivery system: influence of local zoledronate release on bone implant osteointegration. Bone
20. Prendergast PJ. The functional performance of orthopaedic bone cement. Key Eng Mater
21. Race A, Miller MA, Ayers DC, Mann KA. Early cement damage around a femoral stem is concentrated at the cement/bone interface. J Biomech
22. Sabokbar A, Fujikawa Y, Murray DW, Athanasou NA. Bisphosphonates in bone cement inhibit PMMA particle induced bone resorption. Ann Rheum Dis
23. Shannon BD, Klassen JF, Rand JA, Berry DJ, Trousdale RT. Revision total knee arthroplasty with cemented components and uncemented intramedullary stems. J Arthroplasty
. 2003;18(7 suppl 1):27-32.
24. Shigley JE, Mischke CR. Mechanical Engineering Design
. Ed 6. New York, NY: McGraw-Hill; 2001:165-172.
25. Shimshi M, Abe E, Fisher EA, Zaidi M, Fallon JT. Bisphosphonates induce inflammation and rupture of atherosclerotic plaques in apolipoprotein-E null mice. Biochem Biophys Res Commun
. 2005;328: 790-793.
26. Soltau J, Drevs J. The role of bisphosphonates in oncology. Drugs Future
27. Toussirot E, Wendling D. Bisphosphonates as anti-inflammatory agents in ankylosing spondylitis and spondylarthropathies. Expert Opin Pharmacother
28. Weibull W. A statistical distribution function of wide applicability. J Appl Mech
29. Wilkinson JM, Eagleton AC, Stockley I, Peel NF, Hamer AJ, East- ell R. Effect of pamidronate on bone turnover and implant migration after total hip arthroplasty: a randomized trial. J Orthop Res
. 2005; 23:1-8.
© 2006 Lippincott Williams & Wilkins, Inc.
30. Zenios M, Nokes L, Galasko CS. Effect of a bisphosphonate, disodium pamidronate, on the quasi-static flexural properties of Palacos R acrylic bone cement. J Biomed Mater Res BAppl Biomater
. 2004; 71:322-326.