Patent Application
20090088858
The present invention relates to a method of surface finishing a bone implant for particular use in orthopaedic or dental procedures.
Such an implant is preferably manufactured so that in use it becomes integrated into the bone tissue. The implant must also be made of a material that is non-corrosive and that does not produce any immunological reaction causing rejection by the body. Typically, therefore such implants are metallic in nature and usually made from titanium, zirconium, niobium, or tantalum, or an alloy based on any of the aforesaid elements; medical grade stainless steel, or a cobalt-chromium alloy could also be used.
Events that lead to the integration of an implant into bone, and hence that determine the long-term performance of the implant, take place largely at the interface formed between the tissue and the implant. The development of this interface is complex and is influenced by numerous factors, including surface chemistry, surface charge, surface topography and surface contamination of the implant. To improve bone tissue integration, various techniques have been used to increase the surface roughness of titanium implants, and include: machining/micromachining, particle blasting, titanium plasma-spraying, chemical/electrochemical etching, particle blasting and chemical etching, electrochemical anodization, and pulsed laser ablation. Among these methods, abrasive blasting, also called sand blasting or grit blasting, is one of the most commonly found for metallic implants having surfaces that come into contact with bone that is to inter-grow therewith. Implants of this kind are used as prostheses in orthopaedics, for replacing broken or diseased bone, and in dentistry, for building artificial teeth. Nowadays, abrasive blasting is widely used to produce micro-rough surfaces on implantable devices, with typically 4-6 micrometers of Ra (average roughness, according to ISO 4287-1997 and ASME B46.1-1995). This process is widely used because it is efficient for roughening a surface while being cost effective and shows excellent clinical results. In a histological study of well-fixed grit-blasted Ti-6A1-7Nb stems of hip replacements, [A. Zweymüller, F. K. Lintner and M. F. Semlitsch, Journal of Clinical Orthopedics 235, 195-206 (1988)], it was found that excellent osseo-integration had been obtained as a result of the micro roughness of the implant surfaces. It has also been found that osseo-integration occurs even under osteoporotic conditions: a grit-blasted Ti-6Al-7Nb prosthesis retrieved from a 100-year-old patient 3 years after arthroplasty showed sound bony fixation [. K. Lester and P. Campbell, American Journal of Orthopaedics 25(1), 30-34 (1996)]. It has also been reported that an examination of 17 acetabular components that had been in situ for between 10 days and 2.5 months found that, during the insertion of the treated socket the rough cp titanium surface becomes covered with a slurry of bone particles and blood plasma. This mixture of living cells, inorganic substances, and native bone-growth promoting agents (BMPs, growth factors, etc.) adheres to the surface and allows osteogenesis to occur, regardless of whether or not the implant is in direct contact with the host bone, [F. K. Lintner, M. Huber, G. Böhm, J. Attems and R. Wais, in 15 Jahre Zweymüller-Hüftendoprothese, III. Wiener Symposium, K Zweymüller, Eds. (Verlag Hans Huber, Bern, 1996) pp. 131-137]. It has been reported by Delaunay & Kapandji [C. P. Delaunay and A. I. Kapandji, Journal of Arthroplasty 11(6), 643-652 (1996)] that they obtained a cumulative 6-7 year survival rate of approximately 98% when using a “Zweymüller Stem”. Extended clinical follow-up by these authors also found that grit-blasted cp titanium threaded cups to have a 9-10 year survival rate of approximately 99% [C. P. Delaunay and A. I. Kapandji, Journal of Clinical Orthopaedics 340, 130-141 (1997); and C. P. Delaunay and A. I. Kapandji, Acta Orthop Scand 69(4), 379-383 (1998)].
U.S. Pat. No. 5,456,723 discloses a method of producing such a metallic bone implant with a micro-roughness of 2 μm or less by subjecting the surface of the implant to pickling In a reducing acid. Such a micro-roughness may be applied directly to the surface of the implant by the pickling process or be super-imposed upon a micro-roughness having an Rt (peak to valley height, according to ISO 4287-1997 and ASME 1346.1-1995) in excess of 10 μm produced by sand-blasting. In the latter case, the pickling step is so long and aggressive that the amount of the base material removed enables direct release and detachment of all the particles left behind by the sand-blasting, while producing an additional porous micro-topography smaller than 2 μm.
Many studies have analyzed the wear particles generated in joint replacement articulations and their effects on periprosthetic tissues. Additional third-body wear has been reported in cobalt-based alloy bearing surfaces for total hip replacement (THR) endoprotheses, originating for example from radio-pacifying agents in bone cements and from hydroxyapatite coatings [M. Rokkum, M. Brandt, K. Bye et al., The Journal of Bone and Joint Surgery 81-B(4), 582-589 (1999)]. Extensive surface inclusions from silicon- and/or aluminium-containing materials were found when five retrieved cementless implants with grit blasted surfaces were analyzed [J. L. Ricci, F. J. Kummer, H. Alexander and R. S. Casar, Journal of Applied Biomaterials 3, 225-230 (1992)]. Evidence of the same contaminants was found in soft-tissue samples adjacent to these implants, and also in a variety of new, non-implanted devices from different manufacturers. While in one experimental study alumina-blasting particles were considered to be a possible cause of tissue breakdown [B. W. Darvell, N. Samman, W. K. Luk, R. K. F. Clark and H. Tideman, Journal of Dentistry 23(5), 319-322 (1995)], in other experimental studies these surface contaminants on blasted titanium implants seemed not to have any negative effects on the rate of bone on-growth [V. M. Goldberg, S. Stevenson, J. Feighan and D. Davy, Clinical Orthopaedics 319, 122-129 (1995); and A. Piattelli, L. Manzon, A. Scarano, M. Paolantonio and M. Piattelli, International Journal of Oral & Maxillofacial Implants 13, 805-810 (1998)]. More recently, Plenk et al. [H. Plenk Jr, M. Boehler, I. Steffan and A. Walter, European Cells and Materials 7(1), 78 (2004)]] showed that particle impaction of blasted implant surfaces leads to contamination of peri-implant tissues and metallic wear, but seems not directly to hinder bony anchorage. However, alumina particles produce increased third-body wear on both, the low and high carbon cobalt-based alloys studied. As a consequence, cobalt-based alloy wear particles lead to a pronounced foreign body reaction with all signs of an immune response. This work completes a previous study describing such adverse tissue reaction to metallic wear particles [M. Boehler, F. Kanz, B. Schwartz, I. Steffan, A. Walter, H. Plenk Jr and K. Knahr, The Journal of Bone and Joint Surgery 84-B, 128-136 (2002)].
It will thus be appreciated that there is a need to develop a method which can reduce or substantially eliminate the hard particles contaminating grit-blasted surfaces without affecting too much the overall topography because a combination of topographies in the micrometre and nanometre range has shown to influence significantly the biological performance of implantable titanium devices in orthopaedic medicine and in dentistry.
One such approach for achieving this, is disclosed in US2004/0016651, wherein the use of blasting particles, in particular iron, is described for roughening the surface of the implant. Once the surface has been blasted and patterned, the blasting particles are selectively etched from the implant, in a stripping step. The etch used to remove these contaminants, is selective to the material of the blasting particles, wherein a stripping time of at least 60 minutes, preferably 90 minutes, has to be observed to achieve an implant surface which is substantially free of residual iron.
The object of the present invention is therefore to provide such a method of surface finishing a bone implant that produces a rough surface whilst reducing contamination caused by blasting media.
According to a first aspect of the present invention there is provided a method of surface finishing a bone implant comprising the steps of roughening a surface of the implant by blasting with abrasive particles; pickling the surface-roughened implant in a pickling or etching solution; and characterised in that the pickling step loosens any partially embedded abrasive blasting particles that may be contaminating the surface of the implant, by performing a short etch of the surface of the implant, in such a manner that the surface roughness remains substantially the same as before the pickling step, and in that the method comprises the additional step of cleaning the roughened surface of the implant by mechanical action to detach the loosened blasting particles there from. The term short etch means an etch of substantially less than 60-90 minutes, preferably of only some seconds, certainly a maximum of about 1-2 minutes.
This combination of steps fulfils the object of the invention. The pickling of the implant in the pickling solution loosens any partially embedded abrasive blasting particles that may be contaminating the surface of the implant. Those loosened abrasive particles as well as the firmly adhering abrasive blasting particles are then detached by the mechanical cleaning action. Hence, the pickling process should not be designed to clean the surface of the implant or to produce additional pitting of the surface of the implant, as in some prior art procedures, but only to unlock any partially embedded abrasive blasting particles. Similarly, the mechanical action also should not be designed to provide any additional roughening of the surface, all of which is carried out during the initial abrasive blasting, but only to remove the loosened and firmly adhering abrasive particles.
Preferably, the abrasive particles used to roughen the surface of the implant are ceramic and/or metal particles. If ceramic particles are used, these preferably comprise at least one of oxide particles, nitride particles and carbide particles.
Preferably also, the abrasive particles are propelled in a gaseous or liquid blasting medium.
Preferably also, the step of roughening the surface of the implant produces a surface roughness ranging from 3 to 7 μm of Ra inclusive and from 20 to 70 μm of Rt inclusive.
Preferably also, the pickling solution comprises one or a mixture of: a mixture of ammonium bi-fluoride and nitric acid; ammonium fluoride—ammonium bi-fluoride in acid mixtures; hydrofluoric acid based mixtures; sodium fluoride in acid mixtures; ammonium bi-fluoride and ammonium acetate in water; hydrochloric acid based mixtures; mixture of sulphuric and hydrochloric acid; and at least one fluoride salt, at least one acid and water.
Advantageously, the step of pickling the surface-roughened implant takes place by immersion of the implant in a mixture of ammonium bi-fluoride ([NH4)]HF2, 50 g of powder for 1 litre), nitric acid (65% HNO3, 400 ml for 1 litre), and water to make up the solution to 1 litre. Preferably, the pickling takes place at room temperature, preferably a temperature of between 20° C. and 25° C., in particular at 22° C.±2° C., for between 15 and 30 seconds.
Preferably also, the mechanical action comprises with the use of ultrasounds when the implant is immersed in a liquid medium after pickling. Alternatively, the mechanical action comprises the blasting of the implant with substantially non-abrasive or only slightly-abrasive particulate media. In this latter case, preferably the particulate media comprise at least one of dry-ice pellets, crystalline particles of water soluble material, and particles of bioactive blasting material. The blasting medium can be gaseous, for example filtered air or nitrogen, or liquid, for example water.
In cases where dry-ice pellets are used these preferably comprise carbon dioxide snow flakes. Preferably also, they have an average diameter of 3 mm and are propelled in compressed air at an average pressure of 11 bars. The rate of supply of dry-ice pellets into the compressed air is preferably substantially of the order of 100 kilograms per hour. The time for blasting the surface lies in the range 20 seconds up to 3 minutes.
In cases where crystalline particles of water soluble material are used, these preferably comprise particles of at least one of a sugar, sodium chloride, sodium sulfate and a mixture of any of the aforesaid materials. Alternatively, where particles of bioactive blasting material are used these preferably comprise particles of calcium phosphate and/or calcium carbonate.
Preferably also, the bone implant is comprised of one of titanium, zirconium, niobium, tantalum, an alloy based on any of the aforesaid elements, medical grade stainless steel, and a cobalt-chromium alloy.
According to a second aspect of the present invention there is provided a bone implant defining a surface with a surface roughness ranging from 3 to 7 μm of Ra inclusive that has been produced by blasting said surface with abrasive particles and characterised in that a substantial proportion of any abrasive blasting particles embedded therein have been loosened there from by pickling of the implant in a pickling solution and subsequently detached by mechanical action.
The present invention will now be described by way of example with reference to the accompanying drawings, in which:
To obtain the results from BSE micrographs the implant sample to be analyzed was introduced without any further treatment in a scanning electron microscope equipped with a BSE (Back-Scattered-Electron) detector. Five Back-Scattered-Electrons (BSE) micrographs were taken per sample at different randomly selected positions on the surface of the sample. The conditions used are the following: acceleration voltage: 20 kV; spot-size: large (10); magnification: 100×; working distance: 25 mm; detector adjustment: chemical contrast; amplification of the BSE signal: medium. Under such conditions, alumina appears black whereas the titanium surface appears white. An image analysis is then performed for each BSE micrograph with the contrast adjusted so that black corresponds to alumina. The sum of all the black areas compared to the entire surface of the micrograph analyzed gives the surface contamination in percent. The mean value as well as the standard deviation per sample are calculated from 5 different measurements performed on 5 different BSE micrographs. The accuracy of the method was controlled by Energy-Dispersive X-Ray (EDX) mapping performed in the same conditions (distance to detectors and magnification).
To obtain the results from optical micrographs the implant sample to be analyzed was chemically etched for 20 seconds at room temperature (22±2° C.) in a mixture of ammonium bifluoride ([NH4)]HF2, 50 g of powder for 1 litre) and nitric acid (65% HNO3, 400 ml for i litre) completed with water to obtain 1 litre of solution. The sample was then carefully rinsed with osmotic or purified water and left to dry in air. This enables a satisfactory optical contrast to be obtained. Six pictures were taken randomly of the surface of the implant using a digital camera coupled to a microscope equipped with a polarizing filter. The conditions used were the following: magnification of 60×, polarizing filter adjusted to obtain the maximum dark-field. Under such conditions, alumina appears white whereas the titanium surface appears black. An image analysis was then performed for each optical micrograph with the contrast adjusted so that white corresponds to alumina. The sum of all the white areas compared to the entire surface of the micrograph analyzed gives the surface contamination in percent. The mean value as well as the standard deviation per sample were then calculated from the 6 different measurements performed on the 6 different optical micrographs.
In contrast,
The difference in magnification, in surface sensitivity and in precision between the two methods explains this discrepancy. The BSE method is very accurate and can measure very fine particles but it also measures, as deep as 3-5 microns underneath the surface, the alumina particles that are completely embedded in the substrate. On the other hand, the optical method is only sensitive to large alumina particles which are very slightly embedded.
In addition, it can be seen from
Grit blasting is a stochastic process leading to a rough topography. The fact is that the coarser contributions to roughness often hide the fine surface roughness features when considering the standard “integral” roughness parameters such as Ra or Rt, which are defined according to ISO 4287-1997 and ASME B46.1-1995 such that Ra is arithmetic average of the absolute values of all points of the profile and Rt is the maximum peak-to-valley height of the entire measurement trace. These “integral” roughness parameters are scale dependent and depend also on the cut-off wavelength applied when measuring same. Typically, the roughness parameter Ra obtained after alumina grit blasting lies between 3 to 7 micrometres. However, it is known that such blasted surfaces are characterized by numerous surface roughness features ranging from the 100 micrometres scale to the nanometre scale.
In the present method, a bone implant made by a conventional method from a biocompatible material, such as any of titanium, zirconium, niobium, tantalum, an alloy based on any of the aforesaid elements, medical grade stainless steel, and a cobalt-chromium alloy is surface roughened. The roughening is produced by blasting with abrasive particles to produce micro and submicro topography. The blasting particles are preferably ceramic particles, such as oxides (for example: alumina, zirconia, titania, fused titanium dioxide, fused aluminium oxide), nitrides (for example carbon nitride, silicon nitride or boron nitride) or carbides (for example chromium carbide, silicon carbide or boron carbide), or metal particles. The medium use to propel the blasting particles can be gaseous, such as air or nitrogen (that may or may not have been dried), or liquid, for example water. After this blasting, the surface contamination by the blasting material has been typically found to be between 15 and 40% when measured by BSE and 10 to 30% when measured optically. The Ra and Rt roughness parameters are typically 3≦Ra≦7 μm and 20≦Rt≦70 μm respectively.
After surface blasting, the implant is subjected to a pickling process in a pickling or corrosive solution in order to loosen any partially embedded blasting particles from its surface. The exposure of the implant to the pickling solution must be carried out under controlled conditions for a controlled time in order to provide sufficient loosening of the partially-embedded blasting particles while minimizing topographical modifications to the surface of the implant. After pickling, the implant should be rinsed clean using osmotic or purified water and may be dried.
The pickling process loosens and unlocks the blasting particles from the surface of the implant, by performing a brief etch of the implant. Placement of the implant within the pickling (or etching) solution causes an etch, which is preferably isotropic, of the surface of the implant. It is clear, that this etch will also effect the regions around any embedded or partially embedded blasting particles. This etch thus causes the implant surface to loosen its hold on the blasting particles, and in places where the particles are only slightly embedded: will actually be etched away to completely free the blasting particle from the surface.
Preferably, the pickling or etching process is performed rapidly, with the corresponding solution attacking the surface of the implant in an appropriately aggressive manner to allow this. When the etch is performed quickly, the topography of the surface before and after the etch remains substantially unmodified. In other words, the surface roughness achieved by the blasting process, is substantially the same after the pickling step. Preferably, the substrate is etched by no more than 20 μm; more preferably, the surface is etched by between 4 μm and 10 μm, and in further situations, the surface is etched by between 1 μm and 2 μm. It is, of course, possible for etches of greater than 20 μm to be performed, and etches lower than 1 μm to be performed, in order to remove blasting particles which are significantly deeply embedded, or in cases when the particles are very loosely bound or embedded.
There are significant advantages to be seen from etching the surface in this manner. As has been stressed, the etch is specifically tailored to rapidly etch the surface of the implant, and in this manner the roughness of the surface which is defined by the blasting step remains substantially unaltered. Certainly, from the point of view of a useful surface with regard to micrometre and nanometre patterning and subsequent tissue integration into the host body, this is the case. Additionally, as the etch is performed on the implant, rather than on the material making up the blasting particles, it allows for any of the above mentioned particles to be used as the blasting media. This is a significant advantage, as dependent upon the final requirements of the implant, different blasting media could be required or useful for giving different surface finishes. Furthermore, certain blasting materials are not compatible with living tissue, and therefore, any contamination of the implant with such materials could lead to complications within the patient after implant.
The pickling or etching solution may comprise any of the following compositions:
ammonium fluoride—ammonium bi-fluoride in acid mixtures (such as hydrochloric acid, or sulphuric acid, or nitric acid);
Various controlled conditions may be applied during the pickling process. The temperature of the pickling bath can be between 5° C. and boiling point, which may be above 100° C. The pickling bath may also be agitated. The agitation may be achieved by mechanical means or by bubbling an inert gas through the pickling solution. The pickling bath may also be aerated or kept under inert atmosphere, for example an argon or nitrogen atmosphere. The immersion time of the implant can be from a few seconds to several minutes depending on the aggressivity or etching rate of the pickling solution. The immersion time must be adjusted in order to keep the topographical parameters of the roughened implant surface before etching substantially the same. Preferably, however, the etch should be performed rapidly, as this will leave the final surface after the etch with substantially the same surface roughness as was obtained with the blasting step. It has also been found that he efficacy of the pickling process can be enhanced electrochemically by anodic polarization of the implant in the bath.
After the pickling step, the implant undergoes a careful rinsing in water, preferably delonised and filtered water, in order to completely remove the pickling solution. Following this rinse, the method in accordance with the invention comprises a mechanical cleaning step in order to detach the blasting particles there from which have been loosened or loosened by the pickling process. Various forms of mechanical cleaning action including dry-ice blasting, as mentioned above, may be employed as follows:
Preferably, however, the surface of the implant is cleaned by blasting with a substantially non-abrasive particulate media such as one of the following:
After the cleaning step, the surface contamination on the implant is further reduced but still present and is typically between 1% and 10% when measured by BSE and between 0.1% and 5% when measured optically. The Ra and Rt roughness parameters are typically 3≦Ra≦7 μm and 20≦Rt≦70 μm respectively, which are the same as after the blasting to produce the roughened surface, It will thus be appreciated that the processing conditions used in the pickling step and the cleaning step are adjusted in order to keep the topographical parameters approximately the same.
With a conventional implant such as a hip-joint comprised of a Ti-6AL-7NB alloy in accordance with ISO 5832-11, it has been found that a surface treatment using the following method produces optimal results.
First, the surface of the implant is subjected to an abrasive blasting procedure using a conventional alumina particulate (Al2O3, Biloxit Type K20 or K24). Dependent on the blasting apparatus used the pressure applied for blasting can range between 3 and 8 bars inclusive. Such a procedure produces a surface roughness ranging from 4 to 6 μm of Ra (average roughness, according to ISO 4287-1997 and ASME B46.1-1995) Next, the implant is pickled by immersion in a mixture of ammonium bi-fluoride ([NH4)]HF2, 50 g of powder for 1 litre) and nitric acid (65% HNO3, 400 ml for 1 litre) in water. The pickling bath should be maintained in the temperature range 20° C. to 25 C, in particular at 22° C. i 2° C., i.e. approximately at room temperature. The immersion time of the implant in the bath should be for between 15 and 30 seconds. After this pickling step, the implant is carefully rinsed in water, preferably deionised and filtered water, in order to remove the pickling solution. Finally, the implant is subjected to dry-ice blasting wherein it is blasted directly using 3 mm (average diameter) CO2 pellets at an average of 11 bars with compressed air, using a dry-ice pellet supply of 100 kilograms per hour. The duration of ice blasting is from between 20 seconds and up to 3 minutes, depending upon the previous parameters.
It has been found that the pickling step produces a maximum reduction of 40% of the alumina contamination (when considering mean values of contamination measured by the BSE method, and compared to a grit blasted surface). A mean reduction of 25% is measured on Ti-6Al-7Nb. The etching depth measured after 20 seconds of etching at 27° C. on polished Ti-6Al-7Nb samples corresponds to 2.7 μm for cp Ti (Grade 2) and to 1.4 μm for Ti-6Al-4V (Grade 5, ELI) when measured with an optical profilometer FRT-MicroProf, equipped with a chromatic sensor CWL 0.3 mm using the following method:
The improvement of the results over conventional surface treatment methods can be appreciated by a comparison of
As previously stated, the pickling process used in the method in accordance with the invention is not designed to clean or structure the surface of the implant significantly with regard to the values of the alumina contamination as well as the roughness parameters before and after the pickling process. Rather, the pickling process is designed to loosen any partially embedded blasting particles. This is achieved by a preferably isotropic and rapid etch of the implant itself, which leaves the resulting surface structure substantially the same as before the etch. This etch, however, results in the implant surface surrounding the embedded blasting particles being etched, and the grip/hold or physical bond between the blasting particle and implant being lessoned such that the particles are loosened, unlocked or even completely freed from the surface. Also, the mechanical cleaning step is not designed to induce any additional roughening of the treated surface with regard to the roughness parameters before and after the combined process but to detach the abrasive particles from the surface which have been loosened by the pickling process. Hence, the abrasive blasting procedure used initially should be designed to produce the degree of surface roughness required in accordance with the type of implant in question and its proposed use. This can be appreciated by a consideration of the roughness parameters shown in tabular form in the following Table 1. Here, the roughness parameters of the surface of an implant after different surface treatments are shown having been measured with a non-contact optical profilometer FRT-MicroProf, equipped with a chromatic sensor CWL 0.3 mm (length of measurement=5.6 mm; 1000 pts/mm; cut-off=0.8 mm, 0.8 mm was ignored at the beginning and the end of the profile; the calculation was performed using a Gaussian filter and an attenuation factor of 50% at the cut-off wavelength). Mean values and standard deviations (STDEV) are calculated from 6 measurements. The results for Ra and Rt are also shown graphically in
| Number | Date | Country | Kind |
|---|---|---|---|
| 04030665.6 | Dec 2004 | EP | regional |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP2005/013905 | 11/22/2005 | WO | 00 | 5/20/2008 |
