- Email: [email protected]
Sodium-azide versus ProClin 300: influence on the morphology of UHMWPE particles generated in laboratory tests
ARTICLE IN PRESS
Biomaterials 25 (2004) 835–842
Sodium-azide versus ProClin 300: influence on the morphology of UHMWPE particles generated in laboratory tests S. Affatatoa,*, G. Bersagliaa, D. Emiliania, I. Foltrana, A. Tonia,b a
Laboratorio di Tecnologia Medica, Istituti Ortopedici Rizzoli, Via di Barbiano 1/10, Bologna 40136, Italy b I Divisione di Ortopedia e Traumatologia, Istituti Ortopedici Rizzoli, Bologna, Italy Received 7 March 2003; accepted 14 July 2003
Abstract Ultra-high molecular-weight polyethylene (UHMWPE) has been used in total joint replacement for the last three decades and is currently the best polymer available for this use. Nevertheless, the wear of UHMWPE remains a serious clinical problem. Polyethylene wear debris has been identified as a cause of osteolysis and a major factor reducing the life of the total hip arthroplasty. Debris generated at the articular surfaces enters the periprostethic tissue where it is phagocyted by macrophages. Studies have shown that particles in the 0.1–10 mm size range are particularly important in causing adverse cellular reactions resulting in osteolysis. The morphology, size, mass, and number of wear particles produced in a hip joint simulator are influenced by the tribological conditions used during the experiment. This paper shows that the morphology of the UHMWPE particles generated in vitro is influenced by the type of lubricant used. This study compared, quantitatively and qualitatively, particles generated in vitro using bovine calf serum as lubricant with two different preservatives: sodium azide and ProClin 300. No significant difference was observed with regards to wear between the two types of lubricant used. Quantitative analysis of the wear particles showed that particles generated in serum with sodium azide were morphologically different from those produced in serum with ProClin 300. Furthermore, the particles produced in serum with sodium azide were more similar to those found in retrieved acetabular cups. r 2003 Elsevier Ltd. All rights reserved. Keywords: Sodium-azide; ProClin 300; Serum with sodium-azide; Serum with ProClin; Polyethylene wear particles; Contact angle; Wettability
1. Introduction A natural, human joint is a complex mechanism with low friction. Joints allow movement and transmit large loads from bone to bone. Joint components in contact with each other are lubricated by a combination of protein fluids that are squeezed from the synovial tissue after loading [1]. Synovial fluid (SF) is a dialysate of blood plasma containing electrolytes, cells, protein and mucopolysaccaride [2]. SF has two tasks: lubrication and nutrition of the cartilage. The protein content of SF is about 20 mg/ml (2%) and the proteins of SF are identical to the proteins of blood plasma. The principal proteins are albumin, gamma-globulin, transferring and seromucin [2]. The lubricating ability of SF is important in joints and its properties are such as to make the best *Corresponding author. Tel.: +39-051-636-6864; fax: +39-051-6366863. E-mail address: [email protected] (S. Affatato). 0142-9612/$ - see front matter r 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0142-9612(03)00603-3
type of lubricant currently known. It is not possible to reproduce a fluid with the same features in the laboratory. At present, serum is the most popular lubricant in wear testing of prosthetic joints and their materials because its composition is close to that of SF. Polyethylene particles isolated from the used serum lubricant of hip simulator tests [3] and periprosthetic tissues [4] have been found to be similar, about 0.2–2 mm in size. These kinds of particles are particularly responsible for adverse cellular reaction that leads to osteolysis. Morphology and the number of wear particles produced at joint interfaces are influenced by tribological conditions. The purpose of a tribological study of prosthetic joints is to develop joint couplings that minimize wear and friction in order to improve the long-term performance of these prostheses. The need for a systematic study of wear is evident in order to improve the knowledge of the tribological characteristics of a hip
ARTICLE IN PRESS 836
S. Affatato et al. / Biomaterials 25 (2004) 835–842
prosthesis. Currently it is believed that the laboratory determination of wear rate typical of joint pairing constitutes an important aspect in pre-clinical validation of a prosthesis. To study the tribology of prosthetic joints in conditions similar to those prevailing in the human body, joint simulators are necessary. The simulator study should produce wear mechanisms, wear rates and wear debris similar to those seen clinically [5]. The lubricant is a crucial parameter in the tribological studies of prosthetic joints, and is one consideration in the present study. We investigated the wear properties of polyethylene acetabular cups with respect to the lubricant. In particular, we compared the effects of lubricant with sodium-azide as preservative versus the same lubricant with ProClin 300.
2. Materials and methods
two types of lubricant at a concentration of 20 mmol to bind any calcium in solution and minimize precipitation of calcium phosphate onto the bearing surfaces. The latter event has been shown to strongly affect friction and wear properties [13]. All solutions were filtered using a 0.20 mm filter (Millipore, Bedford, USA). UHMW Polyethylene wear was calculated by the weight loss method using a microbalance (SARTORIUS AG, Germany) with a sensitivity of 0.01 mg and an uncertainty of 70.10 mg. The evolution of polyethylene wear was evaluated both in terms of weight loss versus number of motion cycles (rotation), and also as wear rate versus number of motion cycles (rotation). The wear rate was calculated at each weighing interval as the weight lost since the previous interval control over the number of cycles between the two controls, as recommended by ISO 14242-Part 2. The average wear rate was calculated with a linear regression between 0.5 million cycles and 2.5 million cycles, as recommended by the aforementioned ISO recommendation.
2.1. Materials 2.3. Lubricant preparation The wear behaviour of the CoCrMo/UHMWPE coupling was investigated using a hip joint simulator. Twelve UHMWPE commercial acetabular cups were tested against 12, 28-mm CoCrMo femoral heads. All specimens tested were manufactured and supplied by Wright Cremascoli ORTHO (Milano, Italy). The UHMWPE acetabular cups were components for surgical implants (ISO 5834-2) and were manufactured by Chirulen GUR 1050 PE rods supplied by POLY HI SOLIDUR (France). 2.2. Experimental procedure The study was carried out using a 12-station hip joint wear simulator (Shore Western, USA). The simulator and the test procedure have been described in detail elsewhere [6,7] in particular, was used. A frequency of 1 Hz, according to the rotation test frequency, was applied with a sinusoidal load having a peak magnitude of about 2 kN under room temperature conditions (23 C). At regular time intervals of 500,000 cycles, the lubricants were collected and analysed immediately to characterise the morphology of the wear particles. The whole wear test lasted 2.5 million cycles, according to other authors [8–11], which represents a clinical followup of approximately 2.5 years. The lubricant was 30% sterile bovine calf serum (SIGMA, St. Louis, USA), balanced with deionised water, and with a protein content of 18 mg/ml that is close to physiological range [12]. To retard bacterial degradation during the wear tests, two types of preservatives were chosen in order to compare their performance with regard to wear behaviour. Ethylenediaminetetraacetic acid (EDTA) was also added to the
In order to compare the performance of the two types of preservatives with regard to wear behaviour. A first solution was composed of bovine calf serum plus sodium azide (E. Merck, Darmstadt, Germany) and a second solution was composed of ProClin 300 (Supelco, Bellefonte, USA) were chosen. A 0.2% sodium azide solution (hereinafter referred as LSA), prepared with 500 ml of phosphate buffer saline 20 mm and 1 g of sodium azide was added to one as an antimycotic; whereas a 15 ppm ProClin 300 solution (hereinafter referred as LPC), prepared with 500 ml of phosphate buffer saline 20 mm and 1 ml of ProClin 300 was added to the other. The amounts of preservative were chosen in order to prevent microbe growth and compare the performance of the two types of preservatives with regard wear behaviour. The samples of preservatives to be tested were inoculated into BACTEC culture vial types Plus Aerobic/F (enriched Soybean-Casein Digest broth with CO2) and BACTEC Plus Anaerobic/F (reproduced enriched soybean-casein digest broth with CO2) for culture and recovery of microorganisms aerobic, anaerobic and yeast. The vials were inserted into the BACTEC 9050 fluorescent series. The sensor is monitored by the instrument for incubation and periodic reading. Each vial contains a chemical sensor which can detect increases in CO2 produced by the growth of microorganisms. Sodium azide is primarily biostatic, not a biocide, at levels normally used, thus it might not effectively control microorganisms in a product; its use imposes restrictions on product disposal, and it is banned in some places. As
ARTICLE IN PRESS S. Affatato et al. / Biomaterials 25 (2004) 835–842
a powder, sodium azide is both hazardous and inconvenient to work with. ProClin 300 preservative is a highly effective biocide offered for the control of microorganisms in reagents and products intended for in vitro diagnostic use [14]. With broad-spectrum activity, excellent compatibility and stability, and low toxicity at use levels, ProClin 300 is the ideal choice for an effective preservative in diagnostic reagents. At unusually low concentrations, ProClin 300 preservative eradicates bacteria, fungi, and yeasts in reagents for prolonged periods, thereby increasing a product’s shelf life.
837
2.6. Contact angle measurements The contact angle of the polyethylene was measured using a SkyScan-1072 micro-CT system (Belgium). A microsyringe firmly attached to the push plate was used to measure out 5 ml drops. The tip was positioned above the surface of polyethylene at such a height that the growing pendant drop touched the surface and detached before it fell. All measurements were made within 45 s of the drop formation. For all measurements the set-up used a magnification of 27.12, an exposure of 5.4 s, a source of 100 kVp–98 mA, and a filter of 1 mm aluminium.
2.4. Debris isolation The serum samples were vigorously stirred for 15 min to allow a homogeneous particle distribution. Although particles have a density less than water, they could be embedded in proteins and thus settle on the bottom of the serum container. Therefore, serum samples of 5 ml were taken from the middle of the container in order to have a representative solution. The procedure to digest the serum samples follows an internal procedure. The digestion procedure was needed because it was very difficult to separate the particles from the lubricant. It is known that serum adheres strongly to polyethylene, and cannot be directly filtered. The two different samples of lubricant were diluted with distilled water and then digested with NaOH 6 m for 24 h at 60 C in an ultrasonic bath, centrifuged at 4000 rpm for 70 min, and a final isolation by filtering the solution. 2.5. SEM and morphologies analysis The morphologic features and the distribution of the particles on the filters were studied with scanning electron microscopy. All filters were allowed to air dry and subsequently attached to standard 25 mm scanning electron microscope (SEM, Jeol JSM 5400, Tokyo, Japan) stubs. The edges of the filters were trimmed to fit the SEM stub, which was mounted using a thin layer of glue. The stub was then sputter coated with gold and imaged using a Jeol SEM operating at 10 kV. Particles in the SEM micrographs were outlined and filled manually using an image processing program. Only separate particles were selected. Agglomerates and particles partially obscured by other particles were omitted. Then the micrograph was converted to a binary picture; afterwards, the binary picture was processed by an image analysis program (Image ProPlus vers. 4.5, Media Cybernetics, USA) that counted the particles and provided a list of the area (A), elongation, and roundness of each particle.
3. Results After 2.5 million cycles, the total wear of the UHMWPE acetabular cups tested in a hip joint simulator was 66711 mg (lubricant LSA) and 67718 mg (lubricant LPC). The evolution of total volumetric wear for all the polyethylene acetabular cups tested under these conditions is plotted in Fig. 1. Table 1 reports the gravimetric wear values of all UHMWPE specimens (weight loss versus number of cycles) and the wear rates of the two different preservatives added to the bovine calf serum (slopes of the regression lines). No significant differences (Kruskall–Wallis non-parametric test) were observed in the wear behaviour of the UHMWPE cups tested by two different preservatives. 3.1. Wear particle study Quantification of PE wear debris represents a challenging task in the characterisation of orthopaedic devices. In the first part of this study, the amount of debris was quantified by the weight loss method, the only accepted technique for this purpose. This method allows the amount of wear in the orthopaedic device to be determined, but provides no information on debris particulate size distribution. In order to gain more insight into the wear mechanisms, information on the wear debris morphology and distribution must be available. The first step of this analysis is debris isolation; a requirement of the isolation technique is that wear debris must be devoid of serum proteins in order to effectively image these particles. 3.2. SEM analysis and size distribution of particle debris The particles in the pictures were categorized into rounded granules, elongated, fibrillar, and occasional larger fragments based on the following criteria. Granules of smooth particles were those with
ARTICLE IN PRESS S. Affatato et al. / Biomaterials 25 (2004) 835–842
838
Volumetric wear trend
SODIO AZIDE
PROCLEAN
100 90
volumetric wear (mm3)
80 70 60 50 40 30 20 10 0 0
0.5
1
1.5
2
2.5
3
cycles (millions)
Fig. 1. Volumetric wear trend of UHMWPE acetabular cups versus number of cycles in bovine calf serum additivated with two different preservatives: sodium azide and ProClin 300.
Table 1 Summary of the gravimetric wear values for UHMWPE acetabular cups tested in bovine calf serum with sodium azide and with ProClin 300 Number of cycles (millions)
Weight loss of UHMWPE in LSA (mg)
Weight loss of UHMWPE in LPC (mg)
0.5 1 1.5 2 2.5 Wear rate (slope of the regression line)
571 24710 39717 53720 72711 28
1875 3475 49714 60713 71719 26
Fig. 3. View of fibrils elongated particles as much as 3 mm long and approximately less than 0.25 mm wide.
sional (Figs. 4 and 6). Finally, the size distributions were visualized by preparing histograms and box-plots (Figs. 7 and 8).
4. Discussion
Fig. 2. SEM micrograph of granules smooth particles ranging from approximately 0.1 to 1.0 mm.
R > 1:0 mm (Fig. 2), elongated particles those with R > 3:0 mm (Fig. 3). Fibrillar particles were those with a width of several micrometers (Fig. 5). Particles that remained in the pictures after the aforementioned classification were categorized as irregular and occa-
At the interface of the two joint surfaces wear particles are generated and their size and morphology are dependent on tribological factors such as the material properties, type of loading, and the kind of lubricant regime [15,16]. An elucidation of the exact nature of wear particle formation will provide useful information for such studies, and the particles themselves, retrieved by such a high-performance method, could be available and more appropriate for experimental studies 1996 [17]. So it has been interesting to evaluate if there were some differences on the size ranges, in two different
ARTICLE IN PRESS S. Affatato et al. / Biomaterials 25 (2004) 835–842
Fig. 4. Occasional larger fragments of particles such as flakes. These particles more than 10.0 mm of dimension could be the result of the comminution of the fibrils.
Fig. 5. Fibrillar particles obtained from LSA lubricant are ranged from less than 1.0 mm to several micrometers in length.
Fig. 6. SEM micrograph of occasionally elongated particles with a length ranged approximately from 1.0 to 5.0 mm.
839
in vitro test conditions, because these could be critical determinants to biological reactions. In fact, the characteristics of wear debris are significant in terms of macrophage response, for example, what the relative significance is of the type of debris as opposed to the differing size and shapes of different particulates. Experimental studies with various types of particulate biomaterials have shown that size is an important determinant of tissue response; particles between 10 and 100 mm cause a florid histiocytic response [18]. The debris generated by the progressive wear of total hip replacement devices are considered as the primary cause of osteolysis, bone resorption, and premature failure of artificial hips [17,19]. Modern investigations show that the particle concentration is dependent upon induction of osteolysis in tissues (more than 10 109 particles per gram wet tissue) [20,21]. The precise mechanisms of UHMWPE wear debris induced osteolysis have not been fully explained. However the main message from such investigations is that it is not the wear amount that determines the biological response to the debris but the wear amount within the critical size range between 0.2 and 0.8 mm that is needed to activate macrophages [22]. Another biological study indicates a phagocytable size range of 0.3–10 mm to be the most biologically active [23]. A further in vitro study on the effect of size and dose on bone resorption activity of macrophages by clinically relevant ultra high molecular weight polyethylene particles showed that particles of 0.24 mm are about ten times more active than particles of 0.45 and 1.71 mm [24]. The objective of our study was the comparability between the in vivo and in vitro tests with the bovine serum with two different preservatives: sodium azide and ProClin 300. The use of ProClin 300 has some advantages over sodium azide; in fact previous studies [14] indicate that ProClin 300 offers better protection than sodium azide, without the handling and disposal concerns of traditional preservatives. Moreover, ProClin 300 presents no toxicological problems or health hazards at recommended use levels while sodium azide is very toxic and ingestion or inhalation may be fatal. The active ingredients in ProClin 300 have an established history of successful use as preservatives. However, there are some circumstances under which ProClin 300 may degrade in the presence of high pH, high temperature, reducing agents and aggressive nucleophiles. In particular, the presence of strong nucleophiles such as glutathione and cysteine in serum can have a deleterious effect on ProClin 300 preservative. Quantification and characterisation of polyethylene wear particles represents a challenging task in the characterisation of orthopaedic devices. In the present
ARTICLE IN PRESS S. Affatato et al. / Biomaterials 25 (2004) 835–842
840
LSA Debris Roundness 50%
800
45%
700
40% 600
Frequency
30% 25%
400
20%
300
Percentage %
35% 500
15% 200 10% 100
5%
>6
32. 5
12. 3
2.
2.
92. 1
71. 9
1.
1.
31. 5
51. 7 1.
1.
91. 1
11. 3 1.
0.
70. 9 0.
50. 7 0.
0.
10. 3 0.
0. 05 10
30. 5
0% .1
0
Size range (microns)
Fig. 7. Particle size distribution ranged from 0.015 to 12.35 mm using LSA as lubricant.
LPC Particles Roundness 800
50% 45%
700
40% 600
Frequency
30%
400
25% 20%
300
Percentage
35% 500
15% 200 10% 100
5%
>2 .5 0
2. 10 -2 .5 0
1. 90 -2 .1 0
1. 70 -1 .9 0
1. 50 -1 .7 0
1. 30 -1 .5 0
1. 30 1. 1-
91. 0 0.
0. 90 0. 7-
0. 70 0. 5-
0. 50
0%
0. 3-
0. 10 -0 .3 0
0
Size range (microns)
Fig. 8. Particle size distribution ranged from 0.015 to 1.70 mm using LPC as lubricant.
study, some information on debris particulate size, distribution, and morphology has been given. The first step of this analysis was, as had already been done in a similar study [6,8], debris isolation. If the serum proteins are absent then it is effectively possible to image wear debris. After debris isolation, the morphologic features and the distribution of the particles on the filters were studied with scanning electron microscopy. The particles recovered from the lubricant LPC could be classified in three kinds of configurations:
Fig. 9. Schematic drawing of contact angle measurement.
(1) It is possible to observe granules of smooth particles ranging from approximately 0.1 to 1.0 mm (Fig. 2). (2) SEM micrograph shows fibrils elongated particles as much as 3 mm long and approximately less than 0.25 mm wide (Fig. 3). Often the fibrils show a
rounded head attached to one end with a thin taper at the other. (3) Occasional larger fragments of particles such as flakes are seen in Fig. 4. These particles, more than
ARTICLE IN PRESS S. Affatato et al. / Biomaterials 25 (2004) 835–842
841
Fig. 10. Wettability of LSA versus LPC lubricants.
10.0 mm in size, could be the result of the comminution of the fibrils. Agglomerates were often seen in the micrographs. Agglomerates consisted of clusters of rounded particles, clumps of spherical particles, or patches of fibrous particles. In addition, the agglomerates were often combinations of various particle types (build-ups consisting of flakes, fibrils and rounded particles). The particles recovered from the lubricant LSA could be classified in two kinds of configurations: (1) Fibrillar particles ranged from less than 1.0 mm to several micrometers in length. The width of these particles was generally only tenths of micrometers (Fig. 5). (2) Elongated particles (Fig. 6) were occasionally detected, and they appeared to have smooth surface texture. The length of these particles ranged approximately from 1.0 to 5.0 mm. In addition, rounded bead particles appeared to be joined together, often with attached fibrils. A typical particle size distribution using LSA and LPC as lubricants are shown, respectively, in Figs. 7 and 8. The majority of the particles ranged from 0.1 mm up to 6 mm for the two different preservatives.
5. Conclusions Our studies demonstrated that the vast majority of the numerous wear particles were submicron-sized (lubricant LSA) and either rounded or elongated in shape. The morphological features and size distribution of the polyethylene particles produced in these wear tests were in agreement with clinical findings [4,25,26]. The different tests suggest that some factors have influenced on the particle morphology. Further analyses should be done to search for the reasons for the difference in dimensions of debris particles produced by the two kinds of lubricant. Supposing that the adding of the two preservatives could have changed the physical
properties of two lubricants we performed also some studies on the wettability of the polyethylene. Contact angles of liquids on polymer surfaces are widely used to predict wetting and adhesion properties of these solids by calculating their solid-vapour surface tension. Contact angle is a measure of the wetting of a liquid on a solid surface. In particular, contact angle describes the shape of a liquid drop resting on a solid surface. By drawing a tangent line from the drop shape to the touch of the solid surface, contact angle is defined as the angle between the tangent line and the solid surface (Fig. 9). It is expressed in degrees, with 0 being complete wetting and 180 being absolute non-wetting. The measurement provides information to study the bonding energy of the solid surface and surface tension of the liquid droplet. Because of the simplicity in technique and measurement, it has been broadly accepted in various research environments and industries for material surface analysis related to wetting, adhesion and absorption. Contact angles on polymer surfaces are not only influenced by the interfacial tensions but also by many other phenomena, such as surface roughness, chemical heterogeneity, sorption layers, molecular orientation, swelling, and partial solution of the polymer or lowmolecular constituents in the polymer material [27,28]. These effects have to be considered when contact angle measurements are used to calculate the solid surface tension of polymers. Generally, as the average contact angle increases, the wetting decreases. This phenomenon applied at our polyethylene absorption report that we found no significant difference in their behaviour from a point of view of wear. Infact, as shown in Fig. 10 it is possible to note that the contact angle on the lubricant LPC (part A) is comparable with that on the LSA one (part B). Our results failed to show any significant difference between the two different preservatives used in combination with the bovine calf serum. In our opinion further tribological in vitro studies are necessary to have a better understanding and so to validate or not the use of the ProClin 300 as preservative in wear tests.
ARTICLE IN PRESS 842
S. Affatato et al. / Biomaterials 25 (2004) 835–842
Acknowledgements The authors would like to thank V. Sassoli (Servizio Farmacia Interna, Istituti Ortopedici Rizzoli, Bologna, Italy) and A. Andollina (Laboratorio di Patologia clinica analisi chimico-cliniche e microbiologiche, Istituti Ortopedici Rizzoli, Bologna, Italy) for her support during this experiment.
References [1] Jin ZM, Dowson D, Fisher J. The effect of porosity of articular cartilage of a normal human hip joint. Proc Inst Mech Eng 1992;206:117–24. [2] Unsworth A. Tribology of human, artificial joints. Proc Inst Mech Eng 1991;205(3):163–72. [3] Wang A, Sun D, Stark C, Dumbleton J. Wear and morphology of UHMWPE in total joint replacements. Wear 1995;181–183:241–9. [4] Mckellop HA, Campbell P, Park SH. The origin of submicron polyethylene wear debris in total hip arthroplasty. Clin Orthop Rel Res 1995;311:3–20. [5] Bos I. Histological investigation of polyethylene particles in total hip replacement: ceramic versus metal heads. Hip Int 2000; 10(3):151–60. [6] Affatato S, Fernandez B, Tucci A, Esposito L, Toni A. Isolation and morphological characterisation of UHMWPE wear debris generated in vitro. Biomaterials 2001;22:2325–31. [7] Affatato S, Bordini B, Fagnano C, Taddei P, Tinti A, Toni A. Effects of the sterilisation method on the wear of the UHMWPE acetabular cups tested in a hip joint simulator. Biomaterials 2002;23:1439–46. [8] Affatato S, Bersaglia G, Foltran I, Taddei P, Fini G, Toni A. The performance of gamma- and EtO-sterilised UHMWPE acetabular cups tested under severe conditions Part 1: role of the third-body wear process. Biomaterials 2002;23:4839–46. [9] Saikko VO, Ahlroos T, Calonius O, Keranen O. Wear simulation of total hip prostheses with polyethylene against CoCr, alumina and diamond-like carbon. Biomaterials 2001;22:1507–14. [10] Maher S, Prendergast PJ, Lyons C. Measurement of the migration of a cemented hip prosthesis in an in vitro test. Clin Biomech (Bristol, Avon) 2001;16:307–14. [11] Smith S, Ash H, Unsworth A. A tribologiacl study of UHMWPE acetabular cups and polyurethane compliant layer acetabular cups. J Biomed Mater Res 2000;53:710–6. [12] Schmidig G, Essner A, Wang A. Knee simulator wear of crosslinked UHMWPE 46th Annual Meeting. Orlando, FL: Orthopaedic Research Society; 2000. p. 555. [13] McKellop H, Lu B, Benya P. Friction, lubrication and wear of cobalt–chromium, alumina and zirconia hip prostheses compared on a joint simulator. Proceedings of the 38th Annual Meeting. Washington, DC: Orthopaedic Research Society, 1992; p. 402. [14] Supelco ProClin Preservatives: mechanism and stability, USA, 1996. p. 1–4.
[15] Besong AA, Tipper JL, Mathews BJ, Ingham E, Stone MH, Fisher J. The influence of lubricant on the morphology of ultrahigh molecular weight polyethylene wear debris generated in laboratory tests. Proc Inst Mech Eng 1999;213:155–8. [16] Maccauro G, Piconi C, Proietti L, Timpanaro M, De Santis V, Magliocchetti G, De Santis E. Analysis of the catastrophic failure of a THR: the role of the acetabular component. Hip Int 2001; 11(4):201–8. [17] Revell PA, Saffar NAL, Kobayashi A. Biological reaction to debris in relation to joint prostheses. Proc Inst Mech Eng 1996;211:187–97. [18] Campbell P, Nasser S, Kossovsky N, Amstutz HC. Histopathological effects of ultrahigh-molecular-weight polyethylene and metal wear debris in porous and cemented surface replacements. In: St. John KR, editor. Particulate debris from medical implants: mechanisms of formation and biological consequences ASTM STP 1144:3851. Philadelphia, PA: American Society for Testing and Materials; 1992. [19] Lombardi AVJ, Eugh GA, Volz RG, Albrigo JL, Brainard BJ. Fracture/dissociation of the polyethylene in metal-backed patellar components in total knee arthroplasty. J Bone Jt Surg Am 1998;70(6):675–9. [20] Kobayashi A, Kadoya Y, Bonfield W, Revell Pay T, Scott G, Freeman MA. Quantitiative analysis of polyethylene wear particles in osteolytic regions around failed total prostheses. ORS, 1997; p. 68. [21] Campbell P, Klausner B, Schmalzried TP, Amstutz HC. PE wear particle morphology in capsule vs. osteolytic lesions. ORS, 1997; p. 67. [22] Ingham E, Fisher J. Biological reactions to wear debris in total joint replacement. Proc Inst Mech Eng [H] 2000;214(1): 21–37. [23] Green TR, Fisher J, Stone M, Wroblewski BM, Ingham E. Polyethylene particles of a ‘critical size’ are necessary for the induction of cytokines by macrophages in vitro. Biomaterials 1998;19(24):2297–302. [24] Green TR, Fisher J, Matthews JB, Stone MH, Ingham E. Effect of size and dose on bone resorption activity of macrophages by in vitro clinically relevant ultra high molecular weight polyethylene particles. J Biomed Mater Res 2000;53(5):490–7. [25] Green TR, Fisher J, Stone M, Wroblewski BM, Ingham E. Polyethylene particles of a ‘critical size’ are necessary for the induction of cytokines by macrophages in vitro. Biomaterials 1998;19:2297–302. [26] Rader CP, Sterner T, Jakob F, Schutze N, Eulert J. Cytokine response of human macrophage-like cells after contact with polyethylene and pure titanium particles. J Arthroplasty 1999; 14:840–8. . [27] Augsburg A, Grundke K, Poschel K, Jacobasch H-J, Neumann AW. Determination of contact angles and solid surface tensions of poly(4-X-styrene) films. Acta Polymer 1998;49:417–26. . [28] Kwok DY, Lam CNC, Wu R, Zschoche S, Poschel K, Gietzelt T, Grundke K, Jacobasch H-J, Neumann AW. Low-rate dynamic contact angles on poly[styrene-alt-(hexy1/10-carboxydecyl (90/ 10)maleimide)] and the determination of solid surface tensions. Macromol Chem Phys 1999;200:1121–33.
Biomaterials 25 (2004) 835–842
Sodium-azide versus ProClin 300: influence on the morphology of UHMWPE particles generated in laboratory tests S. Affatatoa,*, G. Bersagliaa, D. Emiliania, I. Foltrana, A. Tonia,b a
Laboratorio di Tecnologia Medica, Istituti Ortopedici Rizzoli, Via di Barbiano 1/10, Bologna 40136, Italy b I Divisione di Ortopedia e Traumatologia, Istituti Ortopedici Rizzoli, Bologna, Italy Received 7 March 2003; accepted 14 July 2003
Abstract Ultra-high molecular-weight polyethylene (UHMWPE) has been used in total joint replacement for the last three decades and is currently the best polymer available for this use. Nevertheless, the wear of UHMWPE remains a serious clinical problem. Polyethylene wear debris has been identified as a cause of osteolysis and a major factor reducing the life of the total hip arthroplasty. Debris generated at the articular surfaces enters the periprostethic tissue where it is phagocyted by macrophages. Studies have shown that particles in the 0.1–10 mm size range are particularly important in causing adverse cellular reactions resulting in osteolysis. The morphology, size, mass, and number of wear particles produced in a hip joint simulator are influenced by the tribological conditions used during the experiment. This paper shows that the morphology of the UHMWPE particles generated in vitro is influenced by the type of lubricant used. This study compared, quantitatively and qualitatively, particles generated in vitro using bovine calf serum as lubricant with two different preservatives: sodium azide and ProClin 300. No significant difference was observed with regards to wear between the two types of lubricant used. Quantitative analysis of the wear particles showed that particles generated in serum with sodium azide were morphologically different from those produced in serum with ProClin 300. Furthermore, the particles produced in serum with sodium azide were more similar to those found in retrieved acetabular cups. r 2003 Elsevier Ltd. All rights reserved. Keywords: Sodium-azide; ProClin 300; Serum with sodium-azide; Serum with ProClin; Polyethylene wear particles; Contact angle; Wettability
1. Introduction A natural, human joint is a complex mechanism with low friction. Joints allow movement and transmit large loads from bone to bone. Joint components in contact with each other are lubricated by a combination of protein fluids that are squeezed from the synovial tissue after loading [1]. Synovial fluid (SF) is a dialysate of blood plasma containing electrolytes, cells, protein and mucopolysaccaride [2]. SF has two tasks: lubrication and nutrition of the cartilage. The protein content of SF is about 20 mg/ml (2%) and the proteins of SF are identical to the proteins of blood plasma. The principal proteins are albumin, gamma-globulin, transferring and seromucin [2]. The lubricating ability of SF is important in joints and its properties are such as to make the best *Corresponding author. Tel.: +39-051-636-6864; fax: +39-051-6366863. E-mail address: [email protected] (S. Affatato). 0142-9612/$ - see front matter r 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0142-9612(03)00603-3
type of lubricant currently known. It is not possible to reproduce a fluid with the same features in the laboratory. At present, serum is the most popular lubricant in wear testing of prosthetic joints and their materials because its composition is close to that of SF. Polyethylene particles isolated from the used serum lubricant of hip simulator tests [3] and periprosthetic tissues [4] have been found to be similar, about 0.2–2 mm in size. These kinds of particles are particularly responsible for adverse cellular reaction that leads to osteolysis. Morphology and the number of wear particles produced at joint interfaces are influenced by tribological conditions. The purpose of a tribological study of prosthetic joints is to develop joint couplings that minimize wear and friction in order to improve the long-term performance of these prostheses. The need for a systematic study of wear is evident in order to improve the knowledge of the tribological characteristics of a hip
ARTICLE IN PRESS 836
S. Affatato et al. / Biomaterials 25 (2004) 835–842
prosthesis. Currently it is believed that the laboratory determination of wear rate typical of joint pairing constitutes an important aspect in pre-clinical validation of a prosthesis. To study the tribology of prosthetic joints in conditions similar to those prevailing in the human body, joint simulators are necessary. The simulator study should produce wear mechanisms, wear rates and wear debris similar to those seen clinically [5]. The lubricant is a crucial parameter in the tribological studies of prosthetic joints, and is one consideration in the present study. We investigated the wear properties of polyethylene acetabular cups with respect to the lubricant. In particular, we compared the effects of lubricant with sodium-azide as preservative versus the same lubricant with ProClin 300.
2. Materials and methods
two types of lubricant at a concentration of 20 mmol to bind any calcium in solution and minimize precipitation of calcium phosphate onto the bearing surfaces. The latter event has been shown to strongly affect friction and wear properties [13]. All solutions were filtered using a 0.20 mm filter (Millipore, Bedford, USA). UHMW Polyethylene wear was calculated by the weight loss method using a microbalance (SARTORIUS AG, Germany) with a sensitivity of 0.01 mg and an uncertainty of 70.10 mg. The evolution of polyethylene wear was evaluated both in terms of weight loss versus number of motion cycles (rotation), and also as wear rate versus number of motion cycles (rotation). The wear rate was calculated at each weighing interval as the weight lost since the previous interval control over the number of cycles between the two controls, as recommended by ISO 14242-Part 2. The average wear rate was calculated with a linear regression between 0.5 million cycles and 2.5 million cycles, as recommended by the aforementioned ISO recommendation.
2.1. Materials 2.3. Lubricant preparation The wear behaviour of the CoCrMo/UHMWPE coupling was investigated using a hip joint simulator. Twelve UHMWPE commercial acetabular cups were tested against 12, 28-mm CoCrMo femoral heads. All specimens tested were manufactured and supplied by Wright Cremascoli ORTHO (Milano, Italy). The UHMWPE acetabular cups were components for surgical implants (ISO 5834-2) and were manufactured by Chirulen GUR 1050 PE rods supplied by POLY HI SOLIDUR (France). 2.2. Experimental procedure The study was carried out using a 12-station hip joint wear simulator (Shore Western, USA). The simulator and the test procedure have been described in detail elsewhere [6,7] in particular, was used. A frequency of 1 Hz, according to the rotation test frequency, was applied with a sinusoidal load having a peak magnitude of about 2 kN under room temperature conditions (23 C). At regular time intervals of 500,000 cycles, the lubricants were collected and analysed immediately to characterise the morphology of the wear particles. The whole wear test lasted 2.5 million cycles, according to other authors [8–11], which represents a clinical followup of approximately 2.5 years. The lubricant was 30% sterile bovine calf serum (SIGMA, St. Louis, USA), balanced with deionised water, and with a protein content of 18 mg/ml that is close to physiological range [12]. To retard bacterial degradation during the wear tests, two types of preservatives were chosen in order to compare their performance with regard to wear behaviour. Ethylenediaminetetraacetic acid (EDTA) was also added to the
In order to compare the performance of the two types of preservatives with regard to wear behaviour. A first solution was composed of bovine calf serum plus sodium azide (E. Merck, Darmstadt, Germany) and a second solution was composed of ProClin 300 (Supelco, Bellefonte, USA) were chosen. A 0.2% sodium azide solution (hereinafter referred as LSA), prepared with 500 ml of phosphate buffer saline 20 mm and 1 g of sodium azide was added to one as an antimycotic; whereas a 15 ppm ProClin 300 solution (hereinafter referred as LPC), prepared with 500 ml of phosphate buffer saline 20 mm and 1 ml of ProClin 300 was added to the other. The amounts of preservative were chosen in order to prevent microbe growth and compare the performance of the two types of preservatives with regard wear behaviour. The samples of preservatives to be tested were inoculated into BACTEC culture vial types Plus Aerobic/F (enriched Soybean-Casein Digest broth with CO2) and BACTEC Plus Anaerobic/F (reproduced enriched soybean-casein digest broth with CO2) for culture and recovery of microorganisms aerobic, anaerobic and yeast. The vials were inserted into the BACTEC 9050 fluorescent series. The sensor is monitored by the instrument for incubation and periodic reading. Each vial contains a chemical sensor which can detect increases in CO2 produced by the growth of microorganisms. Sodium azide is primarily biostatic, not a biocide, at levels normally used, thus it might not effectively control microorganisms in a product; its use imposes restrictions on product disposal, and it is banned in some places. As
ARTICLE IN PRESS S. Affatato et al. / Biomaterials 25 (2004) 835–842
a powder, sodium azide is both hazardous and inconvenient to work with. ProClin 300 preservative is a highly effective biocide offered for the control of microorganisms in reagents and products intended for in vitro diagnostic use [14]. With broad-spectrum activity, excellent compatibility and stability, and low toxicity at use levels, ProClin 300 is the ideal choice for an effective preservative in diagnostic reagents. At unusually low concentrations, ProClin 300 preservative eradicates bacteria, fungi, and yeasts in reagents for prolonged periods, thereby increasing a product’s shelf life.
837
2.6. Contact angle measurements The contact angle of the polyethylene was measured using a SkyScan-1072 micro-CT system (Belgium). A microsyringe firmly attached to the push plate was used to measure out 5 ml drops. The tip was positioned above the surface of polyethylene at such a height that the growing pendant drop touched the surface and detached before it fell. All measurements were made within 45 s of the drop formation. For all measurements the set-up used a magnification of 27.12, an exposure of 5.4 s, a source of 100 kVp–98 mA, and a filter of 1 mm aluminium.
2.4. Debris isolation The serum samples were vigorously stirred for 15 min to allow a homogeneous particle distribution. Although particles have a density less than water, they could be embedded in proteins and thus settle on the bottom of the serum container. Therefore, serum samples of 5 ml were taken from the middle of the container in order to have a representative solution. The procedure to digest the serum samples follows an internal procedure. The digestion procedure was needed because it was very difficult to separate the particles from the lubricant. It is known that serum adheres strongly to polyethylene, and cannot be directly filtered. The two different samples of lubricant were diluted with distilled water and then digested with NaOH 6 m for 24 h at 60 C in an ultrasonic bath, centrifuged at 4000 rpm for 70 min, and a final isolation by filtering the solution. 2.5. SEM and morphologies analysis The morphologic features and the distribution of the particles on the filters were studied with scanning electron microscopy. All filters were allowed to air dry and subsequently attached to standard 25 mm scanning electron microscope (SEM, Jeol JSM 5400, Tokyo, Japan) stubs. The edges of the filters were trimmed to fit the SEM stub, which was mounted using a thin layer of glue. The stub was then sputter coated with gold and imaged using a Jeol SEM operating at 10 kV. Particles in the SEM micrographs were outlined and filled manually using an image processing program. Only separate particles were selected. Agglomerates and particles partially obscured by other particles were omitted. Then the micrograph was converted to a binary picture; afterwards, the binary picture was processed by an image analysis program (Image ProPlus vers. 4.5, Media Cybernetics, USA) that counted the particles and provided a list of the area (A), elongation, and roundness of each particle.
3. Results After 2.5 million cycles, the total wear of the UHMWPE acetabular cups tested in a hip joint simulator was 66711 mg (lubricant LSA) and 67718 mg (lubricant LPC). The evolution of total volumetric wear for all the polyethylene acetabular cups tested under these conditions is plotted in Fig. 1. Table 1 reports the gravimetric wear values of all UHMWPE specimens (weight loss versus number of cycles) and the wear rates of the two different preservatives added to the bovine calf serum (slopes of the regression lines). No significant differences (Kruskall–Wallis non-parametric test) were observed in the wear behaviour of the UHMWPE cups tested by two different preservatives. 3.1. Wear particle study Quantification of PE wear debris represents a challenging task in the characterisation of orthopaedic devices. In the first part of this study, the amount of debris was quantified by the weight loss method, the only accepted technique for this purpose. This method allows the amount of wear in the orthopaedic device to be determined, but provides no information on debris particulate size distribution. In order to gain more insight into the wear mechanisms, information on the wear debris morphology and distribution must be available. The first step of this analysis is debris isolation; a requirement of the isolation technique is that wear debris must be devoid of serum proteins in order to effectively image these particles. 3.2. SEM analysis and size distribution of particle debris The particles in the pictures were categorized into rounded granules, elongated, fibrillar, and occasional larger fragments based on the following criteria. Granules of smooth particles were those with
ARTICLE IN PRESS S. Affatato et al. / Biomaterials 25 (2004) 835–842
838
Volumetric wear trend
SODIO AZIDE
PROCLEAN
100 90
volumetric wear (mm3)
80 70 60 50 40 30 20 10 0 0
0.5
1
1.5
2
2.5
3
cycles (millions)
Fig. 1. Volumetric wear trend of UHMWPE acetabular cups versus number of cycles in bovine calf serum additivated with two different preservatives: sodium azide and ProClin 300.
Table 1 Summary of the gravimetric wear values for UHMWPE acetabular cups tested in bovine calf serum with sodium azide and with ProClin 300 Number of cycles (millions)
Weight loss of UHMWPE in LSA (mg)
Weight loss of UHMWPE in LPC (mg)
0.5 1 1.5 2 2.5 Wear rate (slope of the regression line)
571 24710 39717 53720 72711 28
1875 3475 49714 60713 71719 26
Fig. 3. View of fibrils elongated particles as much as 3 mm long and approximately less than 0.25 mm wide.
sional (Figs. 4 and 6). Finally, the size distributions were visualized by preparing histograms and box-plots (Figs. 7 and 8).
4. Discussion
Fig. 2. SEM micrograph of granules smooth particles ranging from approximately 0.1 to 1.0 mm.
R > 1:0 mm (Fig. 2), elongated particles those with R > 3:0 mm (Fig. 3). Fibrillar particles were those with a width of several micrometers (Fig. 5). Particles that remained in the pictures after the aforementioned classification were categorized as irregular and occa-
At the interface of the two joint surfaces wear particles are generated and their size and morphology are dependent on tribological factors such as the material properties, type of loading, and the kind of lubricant regime [15,16]. An elucidation of the exact nature of wear particle formation will provide useful information for such studies, and the particles themselves, retrieved by such a high-performance method, could be available and more appropriate for experimental studies 1996 [17]. So it has been interesting to evaluate if there were some differences on the size ranges, in two different
ARTICLE IN PRESS S. Affatato et al. / Biomaterials 25 (2004) 835–842
Fig. 4. Occasional larger fragments of particles such as flakes. These particles more than 10.0 mm of dimension could be the result of the comminution of the fibrils.
Fig. 5. Fibrillar particles obtained from LSA lubricant are ranged from less than 1.0 mm to several micrometers in length.
Fig. 6. SEM micrograph of occasionally elongated particles with a length ranged approximately from 1.0 to 5.0 mm.
839
in vitro test conditions, because these could be critical determinants to biological reactions. In fact, the characteristics of wear debris are significant in terms of macrophage response, for example, what the relative significance is of the type of debris as opposed to the differing size and shapes of different particulates. Experimental studies with various types of particulate biomaterials have shown that size is an important determinant of tissue response; particles between 10 and 100 mm cause a florid histiocytic response [18]. The debris generated by the progressive wear of total hip replacement devices are considered as the primary cause of osteolysis, bone resorption, and premature failure of artificial hips [17,19]. Modern investigations show that the particle concentration is dependent upon induction of osteolysis in tissues (more than 10 109 particles per gram wet tissue) [20,21]. The precise mechanisms of UHMWPE wear debris induced osteolysis have not been fully explained. However the main message from such investigations is that it is not the wear amount that determines the biological response to the debris but the wear amount within the critical size range between 0.2 and 0.8 mm that is needed to activate macrophages [22]. Another biological study indicates a phagocytable size range of 0.3–10 mm to be the most biologically active [23]. A further in vitro study on the effect of size and dose on bone resorption activity of macrophages by clinically relevant ultra high molecular weight polyethylene particles showed that particles of 0.24 mm are about ten times more active than particles of 0.45 and 1.71 mm [24]. The objective of our study was the comparability between the in vivo and in vitro tests with the bovine serum with two different preservatives: sodium azide and ProClin 300. The use of ProClin 300 has some advantages over sodium azide; in fact previous studies [14] indicate that ProClin 300 offers better protection than sodium azide, without the handling and disposal concerns of traditional preservatives. Moreover, ProClin 300 presents no toxicological problems or health hazards at recommended use levels while sodium azide is very toxic and ingestion or inhalation may be fatal. The active ingredients in ProClin 300 have an established history of successful use as preservatives. However, there are some circumstances under which ProClin 300 may degrade in the presence of high pH, high temperature, reducing agents and aggressive nucleophiles. In particular, the presence of strong nucleophiles such as glutathione and cysteine in serum can have a deleterious effect on ProClin 300 preservative. Quantification and characterisation of polyethylene wear particles represents a challenging task in the characterisation of orthopaedic devices. In the present
ARTICLE IN PRESS S. Affatato et al. / Biomaterials 25 (2004) 835–842
840
LSA Debris Roundness 50%
800
45%
700
40% 600
Frequency
30% 25%
400
20%
300
Percentage %
35% 500
15% 200 10% 100
5%
>6
32. 5
12. 3
2.
2.
92. 1
71. 9
1.
1.
31. 5
51. 7 1.
1.
91. 1
11. 3 1.
0.
70. 9 0.
50. 7 0.
0.
10. 3 0.
0. 05 10
30. 5
0% .1
0
Size range (microns)
Fig. 7. Particle size distribution ranged from 0.015 to 12.35 mm using LSA as lubricant.
LPC Particles Roundness 800
50% 45%
700
40% 600
Frequency
30%
400
25% 20%
300
Percentage
35% 500
15% 200 10% 100
5%
>2 .5 0
2. 10 -2 .5 0
1. 90 -2 .1 0
1. 70 -1 .9 0
1. 50 -1 .7 0
1. 30 -1 .5 0
1. 30 1. 1-
91. 0 0.
0. 90 0. 7-
0. 70 0. 5-
0. 50
0%
0. 3-
0. 10 -0 .3 0
0
Size range (microns)
Fig. 8. Particle size distribution ranged from 0.015 to 1.70 mm using LPC as lubricant.
study, some information on debris particulate size, distribution, and morphology has been given. The first step of this analysis was, as had already been done in a similar study [6,8], debris isolation. If the serum proteins are absent then it is effectively possible to image wear debris. After debris isolation, the morphologic features and the distribution of the particles on the filters were studied with scanning electron microscopy. The particles recovered from the lubricant LPC could be classified in three kinds of configurations:
Fig. 9. Schematic drawing of contact angle measurement.
(1) It is possible to observe granules of smooth particles ranging from approximately 0.1 to 1.0 mm (Fig. 2). (2) SEM micrograph shows fibrils elongated particles as much as 3 mm long and approximately less than 0.25 mm wide (Fig. 3). Often the fibrils show a
rounded head attached to one end with a thin taper at the other. (3) Occasional larger fragments of particles such as flakes are seen in Fig. 4. These particles, more than
ARTICLE IN PRESS S. Affatato et al. / Biomaterials 25 (2004) 835–842
841
Fig. 10. Wettability of LSA versus LPC lubricants.
10.0 mm in size, could be the result of the comminution of the fibrils. Agglomerates were often seen in the micrographs. Agglomerates consisted of clusters of rounded particles, clumps of spherical particles, or patches of fibrous particles. In addition, the agglomerates were often combinations of various particle types (build-ups consisting of flakes, fibrils and rounded particles). The particles recovered from the lubricant LSA could be classified in two kinds of configurations: (1) Fibrillar particles ranged from less than 1.0 mm to several micrometers in length. The width of these particles was generally only tenths of micrometers (Fig. 5). (2) Elongated particles (Fig. 6) were occasionally detected, and they appeared to have smooth surface texture. The length of these particles ranged approximately from 1.0 to 5.0 mm. In addition, rounded bead particles appeared to be joined together, often with attached fibrils. A typical particle size distribution using LSA and LPC as lubricants are shown, respectively, in Figs. 7 and 8. The majority of the particles ranged from 0.1 mm up to 6 mm for the two different preservatives.
5. Conclusions Our studies demonstrated that the vast majority of the numerous wear particles were submicron-sized (lubricant LSA) and either rounded or elongated in shape. The morphological features and size distribution of the polyethylene particles produced in these wear tests were in agreement with clinical findings [4,25,26]. The different tests suggest that some factors have influenced on the particle morphology. Further analyses should be done to search for the reasons for the difference in dimensions of debris particles produced by the two kinds of lubricant. Supposing that the adding of the two preservatives could have changed the physical
properties of two lubricants we performed also some studies on the wettability of the polyethylene. Contact angles of liquids on polymer surfaces are widely used to predict wetting and adhesion properties of these solids by calculating their solid-vapour surface tension. Contact angle is a measure of the wetting of a liquid on a solid surface. In particular, contact angle describes the shape of a liquid drop resting on a solid surface. By drawing a tangent line from the drop shape to the touch of the solid surface, contact angle is defined as the angle between the tangent line and the solid surface (Fig. 9). It is expressed in degrees, with 0 being complete wetting and 180 being absolute non-wetting. The measurement provides information to study the bonding energy of the solid surface and surface tension of the liquid droplet. Because of the simplicity in technique and measurement, it has been broadly accepted in various research environments and industries for material surface analysis related to wetting, adhesion and absorption. Contact angles on polymer surfaces are not only influenced by the interfacial tensions but also by many other phenomena, such as surface roughness, chemical heterogeneity, sorption layers, molecular orientation, swelling, and partial solution of the polymer or lowmolecular constituents in the polymer material [27,28]. These effects have to be considered when contact angle measurements are used to calculate the solid surface tension of polymers. Generally, as the average contact angle increases, the wetting decreases. This phenomenon applied at our polyethylene absorption report that we found no significant difference in their behaviour from a point of view of wear. Infact, as shown in Fig. 10 it is possible to note that the contact angle on the lubricant LPC (part A) is comparable with that on the LSA one (part B). Our results failed to show any significant difference between the two different preservatives used in combination with the bovine calf serum. In our opinion further tribological in vitro studies are necessary to have a better understanding and so to validate or not the use of the ProClin 300 as preservative in wear tests.
ARTICLE IN PRESS 842
S. Affatato et al. / Biomaterials 25 (2004) 835–842
Acknowledgements The authors would like to thank V. Sassoli (Servizio Farmacia Interna, Istituti Ortopedici Rizzoli, Bologna, Italy) and A. Andollina (Laboratorio di Patologia clinica analisi chimico-cliniche e microbiologiche, Istituti Ortopedici Rizzoli, Bologna, Italy) for her support during this experiment.
References [1] Jin ZM, Dowson D, Fisher J. The effect of porosity of articular cartilage of a normal human hip joint. Proc Inst Mech Eng 1992;206:117–24. [2] Unsworth A. Tribology of human, artificial joints. Proc Inst Mech Eng 1991;205(3):163–72. [3] Wang A, Sun D, Stark C, Dumbleton J. Wear and morphology of UHMWPE in total joint replacements. Wear 1995;181–183:241–9. [4] Mckellop HA, Campbell P, Park SH. The origin of submicron polyethylene wear debris in total hip arthroplasty. Clin Orthop Rel Res 1995;311:3–20. [5] Bos I. Histological investigation of polyethylene particles in total hip replacement: ceramic versus metal heads. Hip Int 2000; 10(3):151–60. [6] Affatato S, Fernandez B, Tucci A, Esposito L, Toni A. Isolation and morphological characterisation of UHMWPE wear debris generated in vitro. Biomaterials 2001;22:2325–31. [7] Affatato S, Bordini B, Fagnano C, Taddei P, Tinti A, Toni A. Effects of the sterilisation method on the wear of the UHMWPE acetabular cups tested in a hip joint simulator. Biomaterials 2002;23:1439–46. [8] Affatato S, Bersaglia G, Foltran I, Taddei P, Fini G, Toni A. The performance of gamma- and EtO-sterilised UHMWPE acetabular cups tested under severe conditions Part 1: role of the third-body wear process. Biomaterials 2002;23:4839–46. [9] Saikko VO, Ahlroos T, Calonius O, Keranen O. Wear simulation of total hip prostheses with polyethylene against CoCr, alumina and diamond-like carbon. Biomaterials 2001;22:1507–14. [10] Maher S, Prendergast PJ, Lyons C. Measurement of the migration of a cemented hip prosthesis in an in vitro test. Clin Biomech (Bristol, Avon) 2001;16:307–14. [11] Smith S, Ash H, Unsworth A. A tribologiacl study of UHMWPE acetabular cups and polyurethane compliant layer acetabular cups. J Biomed Mater Res 2000;53:710–6. [12] Schmidig G, Essner A, Wang A. Knee simulator wear of crosslinked UHMWPE 46th Annual Meeting. Orlando, FL: Orthopaedic Research Society; 2000. p. 555. [13] McKellop H, Lu B, Benya P. Friction, lubrication and wear of cobalt–chromium, alumina and zirconia hip prostheses compared on a joint simulator. Proceedings of the 38th Annual Meeting. Washington, DC: Orthopaedic Research Society, 1992; p. 402. [14] Supelco ProClin Preservatives: mechanism and stability, USA, 1996. p. 1–4.
[15] Besong AA, Tipper JL, Mathews BJ, Ingham E, Stone MH, Fisher J. The influence of lubricant on the morphology of ultrahigh molecular weight polyethylene wear debris generated in laboratory tests. Proc Inst Mech Eng 1999;213:155–8. [16] Maccauro G, Piconi C, Proietti L, Timpanaro M, De Santis V, Magliocchetti G, De Santis E. Analysis of the catastrophic failure of a THR: the role of the acetabular component. Hip Int 2001; 11(4):201–8. [17] Revell PA, Saffar NAL, Kobayashi A. Biological reaction to debris in relation to joint prostheses. Proc Inst Mech Eng 1996;211:187–97. [18] Campbell P, Nasser S, Kossovsky N, Amstutz HC. Histopathological effects of ultrahigh-molecular-weight polyethylene and metal wear debris in porous and cemented surface replacements. In: St. John KR, editor. Particulate debris from medical implants: mechanisms of formation and biological consequences ASTM STP 1144:3851. Philadelphia, PA: American Society for Testing and Materials; 1992. [19] Lombardi AVJ, Eugh GA, Volz RG, Albrigo JL, Brainard BJ. Fracture/dissociation of the polyethylene in metal-backed patellar components in total knee arthroplasty. J Bone Jt Surg Am 1998;70(6):675–9. [20] Kobayashi A, Kadoya Y, Bonfield W, Revell Pay T, Scott G, Freeman MA. Quantitiative analysis of polyethylene wear particles in osteolytic regions around failed total prostheses. ORS, 1997; p. 68. [21] Campbell P, Klausner B, Schmalzried TP, Amstutz HC. PE wear particle morphology in capsule vs. osteolytic lesions. ORS, 1997; p. 67. [22] Ingham E, Fisher J. Biological reactions to wear debris in total joint replacement. Proc Inst Mech Eng [H] 2000;214(1): 21–37. [23] Green TR, Fisher J, Stone M, Wroblewski BM, Ingham E. Polyethylene particles of a ‘critical size’ are necessary for the induction of cytokines by macrophages in vitro. Biomaterials 1998;19(24):2297–302. [24] Green TR, Fisher J, Matthews JB, Stone MH, Ingham E. Effect of size and dose on bone resorption activity of macrophages by in vitro clinically relevant ultra high molecular weight polyethylene particles. J Biomed Mater Res 2000;53(5):490–7. [25] Green TR, Fisher J, Stone M, Wroblewski BM, Ingham E. Polyethylene particles of a ‘critical size’ are necessary for the induction of cytokines by macrophages in vitro. Biomaterials 1998;19:2297–302. [26] Rader CP, Sterner T, Jakob F, Schutze N, Eulert J. Cytokine response of human macrophage-like cells after contact with polyethylene and pure titanium particles. J Arthroplasty 1999; 14:840–8. . [27] Augsburg A, Grundke K, Poschel K, Jacobasch H-J, Neumann AW. Determination of contact angles and solid surface tensions of poly(4-X-styrene) films. Acta Polymer 1998;49:417–26. . [28] Kwok DY, Lam CNC, Wu R, Zschoche S, Poschel K, Gietzelt T, Grundke K, Jacobasch H-J, Neumann AW. Low-rate dynamic contact angles on poly[styrene-alt-(hexy1/10-carboxydecyl (90/ 10)maleimide)] and the determination of solid surface tensions. Macromol Chem Phys 1999;200:1121–33.