PRIOR ART On implantation of foreign material into the body tissues, an inflammatory reaction always occurs which can either be part of the incorporation process, and thus entirely necessary for a successful result, or part of a rejection reaction or fibrous encapsulation of the material, which results in clinical failure.

Various studies have been made of the mechanisms which take place upon implantation of foreign material into the body tissues. Cell death and downward regulation of macrophage functions are described by Nygren et al. (J. Biomed. Mater. Res. 45,117-124,1999).

Examples of materials which are used in implants are silicon/quartz, aluminium/aluminium oxide and titanium/titanium oxide. Of these, titanium/titanium oxide has previously been used with great success as implantation material (Leventhal, G. S., Bone Joint Surg., 33A, 473-474, 1951).

It has already been found to be positive if the material which is implanted has pores. The advantageous effect of pores in that part of the material in

contact with tissue has been described by a number of researchers all concerned with the phenomenon of"foreign-body tumours". It is known that tight membranes of cellulose or silicone rubber give rise to tumours in rats more than a year after implantation. By providing pores with a diameter of at least 0.65 um, it was possible to avoid this tumour-inducing effect of the membrane (James et al., Biomaterials 18,1997,667-675).

However, despite extensive research in this field, there are still problems in improving the incorporation of implants in bone and soft tissue.

DISCLOSURE OF THE INVENTION According to the invention, the above problems are solved by constructing an implant which comprises a metal or a semiconductor and a surface layer of metal oxide or semiconductor oxide on the surface of the metal or the semiconductor. The oxide layer is hydrophobic and has pores. It has surprisingly been found that a synergistic effect is achieved by combining an oxide layer with the above two properties. According to the invention, a method for treating the surface of an implant according to the above comprises the steps of making the oxide porous and of making the oxide hydrophobic.

By studies of the early inflammatory reaction around the implant, and by means of subsequent work on the invention, it has been possible, according to the invention, to find a way of reducing the intensity of the initial inflammation, measured as cell death and downward regulation of macrophage functions.

By means of these studies, it has been discovered that this cell death and downward regulation of macrophage functions reflects the capacity for incorporation of the material such that increased cell death upon material

contact is a negative property. By means of systematic studies of which material properties influence incorporation, it is possible to distinguish three important parameters, of which two are essential to the invention, and of which the third affords a further advantage.

According to a preferred embodiment, the pores in the oxide layer have a pore size in the range of 0.2 am-3. 0 lim, preferably in the range of 0.2, um- 1.0 ßm. The pore size can be measured, for example, with the aid of an electron microscope, the lower limit of the range being defined as the minimum diametrical cross section across the pore, and the upper limit of the range being defined as the maximum diametrical cross section across the pore.

One feature which determines what pore size is most suitable is the draining effect which the pores have on undesired substances from the area of incorporation. The lower limit for this effect is ca. 0.2 J. m. The pore size is also dependent on the size of the cells in the implantation area. Cells must not enter the pores.

There are various ways of making the oxide porous. According to one embodiment, the oxide is made porous by anodization in an electrolyte consisting of 0.1 to 1% by volume of hydrofluoric acid. According to an alternative embodiment, the oxide is made porous by anodization in an electrolyte consisting of 0.1 to 1% by volume of hydrofluoric acid and 10% by volume of nitric acid.

According to one embodiment, the oxide layer is made hydrophobic by incubating the layer in an atmosphere of hexamethyl disilazane. Other substances can of course be used for the alkylation. Within the scope of the invention, the oxide layer can be made hydrophobic in ways other than by alkylation.

According to a preferred embodiment of the invention, the oxide layer has a layer of a substance which neutralizes oxygen radicals. This substance can be glutathione, for example. It has previously been found that a favourable effect is achieved by means of covalent bonding of glutathione to gold surfaces (Nygren et al., J. Biomed. Mater. Res. 45,117-124,1999). The arrangement proposed here differs from published data in that a porous oxide is used as substrate and in that it is hydrophobic. According to another preferred embodiment of the invention, it is possible to avoid coupling glutathione covalently to the material, and instead, according to this embodiment, a surface layer of substances which neutralize oxygen radicals is adsorbed or absorbed and evaporated onto the surface of the oxide.

The implant can consist of one material or can include different materials. It can be a metal such as titanium, or a semiconductor such as silicon.

According to a preferred embodiment, the implant consists of titanium with a surface layer of titanium dioxide.

In the case of implantation in bone, the later phase of development can be favourably influenced by coating with a layer of magnesium. Early studies of pure magnesium implants revealed increased callus formation in connection with the metal's dissolution and absorption in the body (E. D. McBride, Southern Medical Journal 31: (5) 508,1938). Later studies with ion- implanted magnesium in metal oxides show that small amounts of magnesium stimulate differentiation of bone-forming cells in contact with the material (Zreiqat et al., J. Biomed. Mater. Res. 44,389,1999). The capacity for incorporation achieved with an arrangement according to the invention is further improved by the oxide layer being coated with a layer of magnesium.

According to a preferred embodiment, the magnesium is applied electrolytically, using the implant as cathode. This method is not already known.

By means of the invention, it has been found that hydrophobicization, such as alkylation of metal surfaces or semiconductor surfaces, such as titanium surfaces, so that these obtain a hydrophobic surface, is favourable with regard to the survival of macrophages in contact with the material. The conclusion therefore is that an implant should be produced which comprises a metal oxide or semiconductor oxide with pores, preferably alkylated in order to obtain a hydrophobic surface and preferably provided with a layer of substance which neutralizes oxygen radicals. Ways of achieving each of these properties are described in the examples below. In summary, the invention affords an unexpected and advantageous effect, and an arrangement not previously described for implantation in bone and soft tissue.

According to the invention, the surface modification of the implant can be carried out using the following techniques: 1. Porous metal oxides and semiconductor oxides.

Porous metal oxides or semiconductor oxides are generally produced by two competing processes, etching of the oxide and oxide growth are allowed to act simultaneously. For example, it is known to etch silicon dioxide with hydrofluoric acid at the same time as new oxide is anodized on the material (Professor Hans Arwin IFM, LiT H, personal communication). A corresponding technique has been described for aluminium oxide which can be made porous by anodization in sulphuric acid (Gonzalez et al., J.

Electrochem. Soc. 147,984,2000). Titanium dioxide is soluble in hydrofluoric acid (Merck Index) and can be provided with a porous oxide layer in a corresponding manner.

2. Alkylation of metal oxides and semiconductor oxides.

A hydrophobic surface can be obtained by alkylation of metal oxides and semiconductor oxides. Titanium oxide can be made hydrophobic by incubation in butanol (Nygren, H. Colloids and Surfaces B: Biointerfaces 6, 329,1996) or by treatment with hexamethyl disilazane in gaseous phase or dissolved in suitable solvent. The latter method can also be used for alkylation of silicon oxide and aluminium oxide. The result is that the surface is covered with methyl groups (-CH3) and is hydrophobic. Titanium oxide with a hydrophobic surface has been found to have a less downwardly regulating effect on cells in contact with the implant than do corresponding hydrophilic materials (Figure 2). A higher proportion of cells also survive in contact with the material (Figure 1).

3. Addition of substances which neutralize oxygen radicals.

Substances which neutralize oxygen radicals can be substances which have a direct neutralizing action and are themselves destroyed in the reaction with oxygen radicals, for example tocopherol, glutathione and cysteine. Enzymes such as superoxide dismutase and catalase break down oxygen radicals to hydrogen peroxide and water and are not affected themselves by the reaction. All these different types of substances which neutralize oxygen radicals have been used in different contexts for acting on inflammation processes, and the use of these substances is known in the art. Glutathione has been coupled covalently to gold surfaces (Lestelius et al. J. Biomed.

Mater. Res. 28 871,1994) and in such cases has been shown to have an unexpected and favourable effect on cell survival in contact with implants (Nygren et al. J. Biomed. Mater. Res. 45,117-124,1999).

4. Inclusion of magnesium in metal oxides Implantation of magnesium in aluminium oxide for the purpose of stimulating bone formation has previously been done using ion implant techniques (Zreiqat et al., J. Biomed. Mater. Res. 44,389,1999). This is a possible way of applying magnesium in accordance with the invention.

BRIEF DESCRIPTION OF THE FIGURES Figure 1 shows that cell survival (FDA+) is higher on an alkylated titanium oxide surface.

Figure 2 shows that the stimulation capacity for PMA (phorbol myristate acetate) is better on an alkylated titanium oxide surface.

Figure 3 shows a scanning electron microscope image of porous titanium oxide which has been produced by anodization in 0.5% by volume of HF, at 50V for 90 minutes.

Figure 4 shows a scanning electron microscope image of porous titanium oxide which has been produced by anodization in 0.5% by volume of HF and 10% by volume of HNOs at 25V for 30 minutes.

ILLUSTRATIVE EMBODIMENTS Production of a porous surface layer of Tison titanium Pieces (8 x 8 x 1 mm) of commercially available titanium plate were mounted in a section of titanium and coupled as anode in an electrophoresis apparatus. The cathode was made of platinum wire, and the electrolyte used was hydrofluoric acid (HF) in a concentration of 0.5% by volume. A voltage of 50V was applied between the electrodes and a current of 2 mA was read off at the voltage sensor. The result after 90 minutes of anodization in 0.5% by volume of HF is shown in Figure 3.

A better result can be obtained by increasing the rate of oxidation by addition of a further acid. Electrolysis was carried out with a mixture of HF (0.5% by volume) and HN03 (10% by volume) as electrolyte. After 30 minutes at a voltage of 25V, the result shown in Figure 4 was obtained.

Production of a hydrophobic porous surface laver of Tison titanium The pieces were made hydrophobic by placing them in a saturated atmosphere of hexamethyl disilazane, or by washing them in a mixture of 90% by volume of butanol and 10% by volume of water.

Production of an oxygen-radical-neutralizing hydrophobic porous surface laver of TiO2 on titanium The hydrophobic pieces were immersed in an aqueous solution of 100 um of glutathione or an alcohol solution of 100 ug/ml of tocopherol, or a mixture of superoxide bismuthase (10 lig/ml) and catalase (90 llg/ml) and allowed to dry.

Inclusion of magnesium in the implant surface Porous titanium was placed as cathode in an electrolysis bath with 0.1 M tv) gCi2 and platinum as anode. At a voltage of 50V, a layer of magnesium is formed at the bottom of the pores after 10 minutes of electrolysis.

In summary, the invention relates to an implant intended for incorporation in bone and soft tissue and comprising a metal and/or a semiconductor and a layer of a metal oxide and/or a semiconductor oxide on the surface of the metal/semiconductor/metal oxide or the semiconductor/semiconductor oxide, where the oxide layer is hydrophobic and has pores, and to a method for treating the surface of an implant intended for incorporation in bone and soft tissue comprising a metal and/or a semiconductor and a layer of a metal oxide and/or a semiconductor oxide on the surface of the metal/semiconductor/metal oxide or the semiconductor/semiconductor oxide, where the oxide layer is made porous and hydrophobic.