Integrated System for Preparation of Bone Cement and Effects on Cement Quality and Environment

Per Mu llerWille, JianSheng Wang, Lars Lidgren

Department of Orthopedics, Lund University Hospital, S-221 85 Lund, Sweden

Received 9 September 1996; accepted 24 January 1997

Abstract: We developed a prepacked mixing system for the preparation of bone cement. The system is based on mixing and collection of bone cement under a vacuum and serves as both the storage and mixing device for the cement components, thereby minimizing the exposure of the operating staff to the monomer and the risk for contamination of the cement during preparation. We evaluated the system using Palacos R and Simplex P. The cement produced was compared with cement obtained from a commercially available mixing system. Temperature evolution during curing, handling characteristics, density, and porosity of the cement obtained were analyzed. The results showed that the experimental system produces cement with physical properties (i.e., setting times and temperature, porosity, and density) equal to or better than those obtained with commercially available systems. Reducing the amount of monomer in the experimental system led to a reduction of the curing temperature without compromising the physical properties of the cements. 1997 John Wiley & Sons, Inc. J Biomed Mater Res (Appl Biomater) 38: 135142, 1997 Keywords: vacuum mixing; bone cement; temperature; porosity; methyl methacrylate

INTRODUCTION

It is widely accepted that vacuum mixing of bone cement improves the mechanical properties of acrylic bone cements. Studies 1 3 have shown that pores within the cement are detrimental to the fatigue properties of bone cement and that removal of air inclusions can enhance the fatigue properties signicantly.1,4 Todays mixing systems require that cement components (liquid and powder) be transferred from their packages to the mixing system, and following mixing some even require transfer of the cement to a separate delivery system. The separate handling of bone cement components and the cementation system leads to the exposure of operating personnel to the unhealthy monomer 5 8 and introduces the risk of spilling either component during handling, which can result in inferior cement quality. Currently the procedure is to evacuate the air from the cement mixture after the cement components have been brought together. Storage, preparation, and mixing of cement in a continuous vacuum should further reduce the possibility of air inclusions. Thermal necrosis has been claimed to be a possible factor leading to later failure of a cemented arthroplasty.9,10 The temperature increase is governed by the amount of polymerizing monomer and the chemical composition of

the cement components.11,12 It could be of clinical importance if the preparation method allows a reduction of the monomer amount, thereby reducing the temperature during curing of the cement, without inuencing cement quality. The objective of this study was to create and evaluate an experimental integrated mixing and cementation system (EIMCS) for the preparation of conventional bone cements. The EIMCS serves as storage for the cement components, mixing system, and application device all in one unit. Mixing is conducted under a vacuum with components and mixing chamber initially free of air and with mechanical stirring of the cement mixture. The mixture is also collected under a vacuum.13 Curing properties and cement porosity were analyzed for Palacos R and Simplex P bone cement. The effects of reducing the monomer amount was also investigated.

MATERIALS AND METHODS

EIMCS

1997 John Wiley & Sons, Inc.

CCC 0021-9304/97/020135-08

The system (Fig. 1) consists of a exible monomer container (polymer coated Al foil) separated from the mixing chamber by a penetrable membrane (polymer coated Al foil). The powder is contained in the mixing chamber and a vacuum can be applied before the monomer and polymer are brought together. OPTIVAC, a commercially available system (MITAB, Sweden) was used as the reference system.
135

8h0d$$529b

04-25-97 21:42:10

jbma

W-JBM 529B

136

LLERWILLE, WANG, AND LIDGREN MU

A partial vacuum of 0.12 bar was applied and maintained throughout the mixing procedure. Ten seconds after the maximal vacuum level had been reached the monomer barrier was penetrated and mixing was conducted as with the reference system. The procedure is illustrated in Figure 2.
EIMCS with Reduced Monomer (EIMCS:red). The mixing chamber of the Prepack was lled with 40 g of polymer powder and the soft monomer container was lled with 18 mL of monomer. The procedure was the same as the EIMCS with the regular cement proportions.

Temperature and Setting Time

Figure 1. The Prepack system.

Experimental Groups

Cement component ratios as recommended by the manufacturer and with a 10% reduction of the monomer amount were used. The mixing of these two combinations in the EIMCS were compared with the mixing of the recommended ratio in an OPTIVAC conventional vacuum mixing system. The study was conducted using Palacos R (ScheringPlough) and Simplex P (Howmedica) bone cements; cements from the same lots were used for the experiments. Ten mixtures were performed for each of the three methods.
Mixing Procedure Reference System. The OPTIVAC reference system was chosen because it is one of the most efcient systems for porosity reduction of bone cement.14 The cement was added to the reference system according to the manufacturers recommendations. A partial vacuum (0.12 bar) was applied and mixing was started after 10 s. Mixing was conducted during 30 s at approximately 1 beat/s. Palacos R and Simplex P were injected into the temperature measuring molds 15,16 at 2 resp. 1.5 min after the start of mixing, and the temperature evolution of the curing cement was monitored. The excess cement from the temperature measurement was used to determine the dough time.16 The cement nozzle, lled with cement, was detached and used for cement porosity and density determinations. Palacos R was prechilled to 4 C for at least 24 h, according to the manufacturers instructions. Simplex P was stored at room temperature (21 { 2 C) for at least 24 h prior to use. EIMCS. The mixing chamber of the EIMCS was lled with 40 g of polymer powder and the soft monomer container was lled with 20 mL of monomer. The systems lled with Palacos R were prechilled to 4 C for at least 24 h and were removed from the refrigerator 2 min before mixing. The systems lled with Simplex were stored at room temperature as above.

Temperature measurements were performed at the center of a 6-mm thick and 60-mm diameter cement disk, according to ISO and ASTM 15,16 standards, and at the surface of the same disk. The temperature data were collected using a multichannel A/D board (DAS-TC, Keithley MetraByte) connected to a computer. Each specimen was inspected after curing; if voids were found at or near the measuring points, the temperature data were excluded from the statistical analysis. The maximum temperature reached at the surface and the center and the setting time were determined from the collected temperature data.
Dough Time

Dough time was determined by inserting a glass rod into the excess cement obtained from the temperature mold every 15 s. The time when the cement no longer adhered to the glass rod was dened as the dough time, according to the ISO standard.16
Porosity

Porosity was divided into two categories, macro- and microporosity. Macropores were dened as pores with a diameter exceeding 1 mm. The macropores were measured from radiographs of the cured cement left in the cement nozzles (hereafter referred to as cement cylinders) using an imaging system (Videoplan) to measure the area and diameter of the pores visible as radiolucent areas on the radiographs. The macroporosity was then expressed as area percentage, calculated as the total area of the macropores divided by the total area of the cement sample. Micropores were the pores with a diameter less than 1 mm. The microporosity was determined from 5-mm thick disks cut from the top of the cement cylinders with a diamond saw (Metronome). Two disks were cut from each cylinder; a liquid penetrant (Dinol, Ha ssleholm, Sweden) was used to detect the micropores on the cut surfaces. The four stained surfaces thus obtained were then examined for pores under a microscope at 401 magnication. The microporosity was then expressed as the average number of pores per surface (63.6 mm2 ). jbma W-JBM 529B

8h0d$$529b

04-25-97 21:42:10

INTEGRATED MIXING AND CEMENTATION SYSTEM

137

8h0d$$529b

04-25-97 21:42:10

jbma

W-JBM 529B

Figure 2. Mixing Procedure for the Prepack System

138

LLERWILLE, WANG, AND LIDGREN MU

TABLE I. Number of Measurements Used for Statistical Analysis

Temperature and Times Palacos R Reference system Prepack system Prepack reduced monomer 8 6 7 Simplex P 7 9 9

Porosity and Density Palacos R 10 10 10 Simplex P 10 9 10

Density

The density determination was performed with the same disks that were used for the microporosity determination using an electronic balance density kit (Mettler ME-33360, Mettler, Germany).

Porosity. There was threefold reduction of the macroporosity (Table VI) when comparing the EIMCS and EIMCS:red with the reference system. No statistically signicant difference in microporosity (Table VII) was found between the three methods. Density. There was no difference in cement density (Table VIII) between the mixing methods when used with Palacos R. Simplex P Temperature and Setting Time. The temperature reached in the center (Table II) of the curing cement was highest with the reference system and lowest with EIMCS:red. The surface temperatures (Table III) of the EIMCS and EIMCS:red experiments were signicantly lower than the temperature reached with the reference system. The setting times (Table IV) were again found to vary over a wide range (23 min). The variations could be connected to some extent to the ambient temperature at the time of mixing with temperatures at the upper temperature limit (i.e., 23 C) reducing the time. The EIMCS and reference system gave the longest times, while the EIMCS:red resulted in a signicantly shorter time. Dough Time. The longest dough time (Table V) was measured when using the reference (255 s) and EIMCS systems (238 s)and shortest with the reduced amount of monomer (207 s). Porosity. No differences with regard to macroporosity (Table VI) could be established. The values, however, indicate a reduction of the macroporosity when comparing the EIMCS and EIMCS:red with the reference system. Surface porosity was common in the cement nozzles from the reference and the EIMCS systems whereas no surface porosity could be found in the cement that cured in the temperature mold. The microporosity (Table VII) was found to be signicantly lower when using the EIMCS:red than with the EIMCS. Density. The density (Table VIII) obtained with the reference system and the EIMCS (both 1.243 g/cm3 ) was lower than when using the EIMCS (1.245 g/cm3 ). The

Statistical Analysis

The cement disks obtained from the temperature measurement were visually examined, and specimens with voids at or near the measuring points were eliminated from the data analysis. The results of the six different groups were obtained from at least six measurements (Table I). The signicance of the differences were analyzed by using the MannWhitney U test, considering p 0.05 as a signicant difference. The statistical analysis was conducted separately for the experiments done with Palacos R and Simplex P, thus comparing the effect of the preparation methods on the obtained cement, not the effect of cement type chosen.

RESULTS
Palacos R Temperature and Setting Time. The temperature reached in the center (Table II) of the curing cement was highest when using the EICMS and was signicantly higher than the temperature reached with the reference system, and the EIMCS:red experiment resulted in the lowest temperature. When looking at the surface temperature (Table III) the differences are not statistically signicant, but again the EIMCS:red yields the lower temperature when compared to the reference system and the EIMCS. The setting times (Table IV) varied over a wide range (2.53.5 min) and were signicantly different for the three methods. The EIMCS gave the longest time and the reference system resulted in the shortest time. Dough Time. The dough time (Table V) was longest when using the EIMCS (282 s) and shortest when reducing the amount of monomer (230 s).

8h0d$$529b

04-25-97 21:42:10

jbma

W-JBM 529B

INTEGRATED MIXING AND CEMENTATION SYSTEM

TABLE II. Temperature (C) in Center of Curing Cement

139

Palacos R OPTIVACTM Average Range Prepack Prepack Red. Monomer OPTIVACTM

Simplex P Prepack Prepack Red. Monomer

63.9 65.8 62.2 61.866.3 64.267.1 60.764.2 S (p 0.028) S (p 0.037) S (p 0.003)

68.0 66.0 63.2 63.472.2 62.669.6 59.565.3 NS (p 0.224) S (p 0.004) S (p 0.038)

S, signicant; NS, not signicant.

TABLE III. Temperature (C) at Surface of Curing Cement

Palacos R OPTIVACTM Average Range Prepack Prepack Red. Monomer OPTIVACTM

Simplex P Prepack Prepack Red. Monomer

48.9 48.8 46.9 47.151.0 47.550.2 43.848.7 NS (p 0.90) NS (p 0.09) NS (p 0.11)

48.9 47.3 46.6 47.949.8 45.650.3 45.147.8 S (p 0.044) S (p 0.001) NS (p 0.40)

S, signicant; NS, not signicant.

TABLE IV. Setting Time (s) of Curing Cement

Palacos R OPTIVACTM Average Range Prepack Prepack Red. Monomer OPTIVACTM

Simplex P Prepack Prepack Red. Monomer

745 897 785 691844 786960 741803 S (p 0.005) S (p 0.037) S (p 0.018)

612 644 498 558743 608719 463571 NS (p 0.13) S (p 0.002) S (p 0.001)

S, signicant; NS, not signicant.

TABLE V. Dough Time (s) of Curing Cement

Palacos R OPTIVACTM Average Range Prepack Prepack Red. Monomer OPTIVACTM

Simplex P Prepack Prepack Red. Monomer

263 282 230 255285 270300 210255 S (p 0.028) S (p 0.002) S (p 0.004)

255 238 207 240285 225285 195225 NS (p 0.09) S (p 0.001) S (p 0.002)

S, signicant; NS, not signicant.

8h0d$$529b

04-25-97 21:42:10

jbma

W-JBM 529B

140
TABLE VI. Macroporosity in Area %

LLERWILLE, WANG, AND LIDGREN MU

Palacos R OPTIVACTM Average Range Prepack Prepack Red. Monomer OPTIVACTM

Simplex P Prepack Prepack Red. Monomer

0.34 0.10 0.08 0.020.62 00.43 00.21 S (p 0.008) S (p 0.003) NS (p 0.97)

0.28 0.14 0.18 00.93 00.82 00.51 NS (p 0.082) NS (p 0.31) NS (p 0.50)

S, signicant; NS, not signicant.

TABLE VII. Microporosity: Pores per Unit Area (63.6 mm2)

Palacos R OPTIVACTM Average Range 0.43 01 Prepack Prepack Red. Monomer OPTIVACTM

Simplex P Prepack Prepack Red. Monomer

0.32 0.92 00.75 02.75 NS (p 0.50) NS (p 0.14) NS (p 0.06)

0.85 1.22 0.8 01.75 0.751.75 0.51.75 NS (p 0.12) NS (p 0.99) S (p 0.03)

S, signicant; NS, not signicant.

TABLE VIII. Density (g/cm3)

Palacos R Prepack Red. Monomer

Simplex P Prepack Red. Monomer

OPTIVACTM Average Range

Prepack

OPTIVACTM

Prepack

1.292 1.291 1.292 1.2861.296 1.2811.296 1.2771.296 NS (p 0.64) NS (p 0.09) NS (p 0.06)

1.243 1.243 1.245 1.2341.246 1.2351.246 1.2341.247 NS (p 0.63) S (p 0.002) S (p 0.001)

S, signicant; NS, not signicant.

surface porosity of the Simplex P samples also inuenced the results of the density measurements leading to differences between the methods that might not be relevant.

DISCUSSION

The recent development of cement and preparation techniques has focused on reducing the temperature reached by the curing cement and reducing the exposure of the / 8h0d$$529b 04-25-97 21:42:10

operational personnel to the unhealthy monomer.17,18 A mixing system addressing the problems of temperature and monomer exposure was demonstrated 19 and achieved by using a novel cement formulation and a new preparation procedure. However, the altered cement composition proved to be mechanically unsound, 20 23 leading to an unacceptably high clinical failure rate. Our aim was to develop a mixing system that would reduce the exposure of the operational personnel to the unhealthy monomer, eliminate the chance of contamination of the bone cement jbma W-JBM 529B

INTEGRATED MIXING AND CEMENTATION SYSTEM

141

during preparation, facilitate handling, and that, above all, was independent of cement formulation, so that it could be used with existing well-documented bone cements, and most importantly produce a high quality cement mixture. This was achieved by a system that contains the cement components and is based on the principals of mixing and collection of the cement under vacuum, an EIMCS. A secondary environmental effect of the combination of cement, mixing, and application systems in one unit is a reduction of waste. A slight reduction of the cement temperature was accomplished by a reduction of the monomer amount. We believe that the monomerpolymer ratio can be adjusted when the mixing conditions are improved, which also may reduce the amount of initial monomer exposure and resorption in the patient. However, changing the cement composition can inuence its mechanical properties (e.g., fatigue life), and these should be investigated before recommending a change in cement formulation. Prechilling the whole EIMCS when using Palacos R cement, as done in this study, prolonged the curing process of the cement. This was caused by the need to warm not only the cement components but also the mixing system before the polymerization process of the cement speeds up. The prechilling also seems to inuence the temperature evolution. No difference in curing temperature can be seen between the reference and EIMCS when using Simplex P, where both systems have the same initial temperature condition; Palacos R prechilled in the EIMCS results in higher temperatures compared with the reference system. The shorter setting times obtained for both cements when reducing the monomer amount can be due to a combination of the relative increase of the initiator concentration in the cement mixture and the decrease of available monomer for polymerization. As expected, a reduction of the monomer amount also leads to a reduction of temperature during curing. The reduction of the macroporosity when using Palacos R and the EIMCS can be attributed to the evacuation of air from the cement powder before it comes into contact with the liquid monomer, allowing for further elimination of air initially entrapped in the system. The degree of microporosity when using Palacos R, however, is already very low in all three methods, which may explain why no variation in cement density could be detected. The positive effect on porosity reduction, which was obtained by evacuating the polymer powder before the addition of the monome, was not as clearly shown when using Simplex P. This may be explained by the difference in viscosity between the two cements. The surface porosity of the Simplex P samples, obtained with the reference and EIMCS, inuenced the results of the density measurements, leading to differences between the methods that might not be relevant. The overall implications of our study are clear because we can show that it is possible to prepack and mix conventional bone cements in a closed system without impairing the cements physical quality (i.e., setting times and tem/ 8h0d$$529b 04-25-97 21:42:10

perature, porosity and density). A reduction of the monomer amount and setting temperature may be of clinical importance. The Prepack system also offers an obvious environmental advantage to the operating staff because no handling of the cement components is necessary and the exposure to the unhealthy monomer is minimized.
The study was funded by the MFR (Project 09509), Nutek, the Medical Faculty at Lund University, and the EC (CT950147).

REFERENCES
1. Davies, J. P.; Harris, W. H. Optimization and comparison of three vacuum mixing systems for porosity reduction of Simplex P cement, Clin. Orthop. Rel. Res. 254:261269; 1990. 2. Wixson, R. L. Do we need to vacuum mix or centrifuge cement? Clin. Orthop. Rel. Res. 285:8490; 1992. 3. James, S. P.; Jasty, M.; Davies, J.; Piehler, H.; Harris, W. H. A fractographic investigation of PMMA bone cement focusing on the relationship between porosity reduction and increased fatigue life, J. Biomed. Mater. Res. 26:651662; 1992. 4. Fritsch, E. W. Static and fatigue properties of two new lowviscosity PMMA bone cements improved by vacuum mixing, J. Biomed. Mater. Res. 31:451456; 1996. 5. Feith, R. Side-effects of acrylic cement implanted into bone, Acta Orthop. Scand. suppl. 161; 1975. 6. Willert, H.-G.; Ludwig, J.; Semlitsch, M. Reaction of bone to methacrylate after hip arthroplasty, J. Bone Joint Surg. 56A:13681382; 1974. 7. Mjo berg, B.; Pettersson, H.; Rosenqvist, R.; Rydholm, A. Bone cement, thermal injury and the radiolucent zone, Acta Orthop. Scand. 55:597600; 1984. 8. Darre, E.; Vedel, P.; Jensen, J. S. Efciency of bone cement mixing systemsma gas chromatographic study, Adv. Orthop. Surg. 1988:106108; 1988. 9. Reckling, F. W.; Dillon, W. L. The bonecement interface temperature during total joint replacement, J. Bone Joint Surg. 59A:8082; 1977. 10. Leeson, M. C.; Lippitt, S. B. Thermal aspects of the use of polymethylmethacrylate in large metaphyseal defects in bone, Clin. Orthop. Rel. Res. 295:239245; 1993. 11. Jefferiss, C. D.; Lee, A. J. C.; Ling, R. S. M. Thermal aspects of self-curing polymethylmethacrylate, J. Bone Joint Surg. 57B:511518; 1975. 12. Meyer, P. R.; Lautenschlager, E. P.; Moore, B. K. On the setting properties of acrylic bone cement, J. Bone Joint Surg. 55A:149156; 1973. n, H.; Jonsson, E.; Lidgren, L. Porosity 13. Wang, J.-S.; Franze of bone cement reduced by mixing and collecting under vacuum, Acta Orthop. Scand. 64:143146; 1993. n, H.; Toksvig 14. Wang, J.-S.; Mu llerWille, P.; Franze Larsen, S. Is there a difference between vacuum mixing systems in reducing bone cement porosity? J. Biomed. Mater. Res. (Appl. Biomater.) 33:115119; 1996. 15. ASTM. ASTM F 451-86. Standard Specication for Acrylic Bone Cement. Philadelphia, PA: The American Society for Testing and Materials; 1986. 16. ISO. ISO 5833/1-1979. Implants for SurgeryAcrylic Resin CementsPart 1: Orthopaedic Applications, Geneva, Switzerland: International Standardisation Organisation; 1979. 17. Nimb, L.; Stu rup, J.; Jensen, J. S. Improved cortical histology after cementation with a new MMA-DMA-IBMA bone cement: an animal study, J. Biomed. Mater. Res.; 27:565574; 1993. 18. Jensen, J. S.; Trap, B.; Skydsgaard, K. Delayed contact hy-

jbma

W-JBM 529B

142

LLERWILLE, WANG, AND LIDGREN MU

persensitivity and surgical glove penetration with acrylic bone cements, Acta Orthop. Scand. 62:2428; 1991. 19. KindtLarsen, T.; Smith, D. B.; Jensen, J. S. Innovations in acrylic bone cement and application equipment, J. Appl. Biomater. 6:7583; 1995. 20. RiegelsNielsen, P.; Srensen, L.; Andersen, H. M.; Lindequist, S. Boneloc cemented total hip prostheses. Loosening in 28/43 cases after 338 months, Acta Orthop. Scand. 66:215217; 1995. 21. Thanner, J.; FreijLarsson, C.; Ka rrholm, J.; Malchau, H.;

n, B. Evaluation of Boneloc. Chemical and mechaniWessle cal properties, and a randomized clinical study of 30 total hip arthroplasties, Acta Orthop. Scand. 66:207214; 1995. 22. Suominen, S. Early failure with Boneloc bone cement 4/8 femoral stems loose within 3 years, Acta Orthop. Scand. 66:13; 1995. 23. Thanner, J.; Malchau, H.; Ka rrholm, J.; Herberts, P.; Wes n, B. Increased micromotions of hip prostheses inserted sle with cold-curing cement. A stereoradiographic study, 41st ORS, Orlando, FL; 1995: 714.

8h0d$$529b

04-25-97 21:42:10

jbma

W-JBM 529B