We report two cases of surface deterioration of a

Transcription

1 Phase transformation of a zirconia ceramic head after total hip arthroplasty K. Haraguchi, N. Sugano, T. Nishii, H. Miki, K. Oka, H. Yoshikawa From Osaka University Medical School, Japan We report two cases of surface deterioration of a zirconia ceramic femoral head associated with phase transformation after total hip arthroplasty. One head was retrieved at revision due to recurrent dislocation after six years and the other because of failure of the locking mechanism of the polyethylene liner after three years. The monoclinic content of the zirconia ceramics rose from 1% to about 30% on the surface of the heads. SEM revealed numerous craters indicating extraction of the zirconia ceramics at the surface. Surface roughness increased from an initial value of m up to 0.12 m. This is the first report to show that phase transformation of zirconia ceramics causes deterioration of the surface roughness of the head in vivo after total hip arthroplasty. J Bone Joint Surg [Br] 2001;83-B: Received 31 January 2001; Accepted after revision 9 April 2001 The use of zirconia ceramics may reduce the rate of failure in total hip arthroplasty (THA) since they have a higher mechanical strength than alumina ceramics. The crystal structure, however, is unstable and low-temperature ageing can occur in vitro 1 because of phase transformation. There is no description of phase transformation of zirconia ceramics after THA in the English literature. We report two cases of deterioration and increase in surface roughness of a zirconia ceramic femoral head associated with phase transformation after THA. K. Haraguchi, MD, Research Fellow N. Sugano, MD. DMSc, Assistant Professor T. Nishii, MD, DMSc, Assistant Professor H. Miki, MD, Assistant Professor K. Oka, MD, Resident H. Yoshikawa, MD, DMSc, Professor Department of Orthopaedic Surgery, Osaka University Medical School, 2-2 Yamadaoka, Suita , Osaka, Japan. Correspondence should be sent to Dr N. Sugano British Editorial Society of Bone and Joint Surgery X/01/ $2.00 Patients and Methods Between 1992 and 1994, in a clinical trial directed at obtaining Japanese regulatory approval for general usage, 18 patients had 21 cementless THAs using a Kobelco system 2 (H-5; Kobe Steel Ltd, Kobe, Japan) with a modular zirconia femoral head. Sterilisation was by ethylene oxide gas and autoclaving was not used. Two patients (two hips) required a revision procedure and the retrieved heads were analysed. The ceramic used was yttria tetragonal zirconium oxide polycrystal (Y-TZP) containing 3 mol% yttrium oxide for stabilisation. The commercial starting powder was produced by a coprecipitation method. The material was processed by cold isostatic pressing, followed by pressureless sintering. Hot isostatic pressing was then used to achieve a density of more than 99.9% with polishing of the surface as the final process. The chemical composition and the physical and mechanical properties of the Y-TZP are shown in Table I. The surface roughness (measured in Ra) of a new Y-TZP head was m of Ra (arithmetic mean of the absolute value of the measured profile height deviation as measured from the graphical centre line), and the monoclinic content, calculated by x-ray diffraction, was 1%. The Y-TZP meets the criteria of the International Organisation for Standardisation (ISO) and the American Society for Testing and Materials (ASTM) F1873 (Table I). The surfaces were observed by the naked eye and light microscopy. The surface roughness was measured using a contact surface profilometer (ST-501; Mitutoyo Corporation, Kawasaki, Japan) at the pole of the femoral head, at the equator and 5 mm below the equator, and the Ra values were calculated. The surface and inner bulk structure were observed by SEM. The monoclinic content of the surface of the Y-TZP head was calculated by x-ray diffraction at the same sites. Case 1. A 58-year-old woman with bilateral osteoarthritis of the hip had a right cementless THA in January 1994 and a similar operation on the left in February 1994 using the H-5 system. She had relief from pain and was able to walk without assistance. Five years later she developed recurrent dislocation of the left hip. Computer-aided radiological measurements showed polyethylene wear to be 1.6 mm (Fig. 1). Revision hip arthroplasty was undertaken using a femoral head of 46 mm diameter to improve stability. 996 THE JOURNAL OF BONE AND JOINT SURGERY

2 PHASE TRANSFORMATION OF A ZIRCONIA CERAMIC HEAD AFTER TOTAL HIP ARTHROPLASTY 997 Table I. Chemical composition and physical and mechanical properties of the Y-TZP Requirement for Requirement for Property Y-TZP 2 ISO ASTM standard F1873 Chemical composition (weight %) ZrO 2 + HfO 2 + Y 2 O 3 >99.0 > Y 2 O to to to 5.4 A1 2 O 3 <0.5 < Other total oxides <0.5 < Physical properties Bulk density (g/cm 3 ) Grain size (µm) Monoclinic phase (%) 1-5 Mechanical properties Flexural strength by four point bend (MPa) >800 Elastic modulus (GPa) Vickers hardness (HV) >1200 Fracture toughness (Kgf/mm 2/3 ) Compressive strength (MPa) Impact strength (MPa) Fig. 1 Case 1. Radiograph before revision showing osseointegration of the acetabular and femoral components. The polyethylene wear was 1.6 mm. Fig. 2 Case 2. Anteroposterior radiographs of the right hip. Figure 2A Preoperative film showing osteoarthritis of the right hip. Figure 2B One month after primary THA using the H-5 system. Figure 2C Three years after primary THA there is dislodgement of the ring of the polyethylene liner. Case 2. A 49-year-old man sustained a fracture-dislocation of the left hip in a traffic accident in He was treated by open reduction and internal fixation, but developed posttraumatic osteoarthritis and underwent cementless THA in March 1994 using the H-5 system (Figs 2A and 2B). Although there was relief from pain and he could walk without assistance, he developed discomfort and instability in January 1997 without any obvious cause. Radiographs showed dislodgement of the ring of the polyethylene liner (Fig. 2C). At revision both the acetabular cup and the femoral stem were found to be stable. The polyethylene liner and the femoral head were exchanged. VOL. 83-B, NO. 7, SEPTEMBER 2001

3 998 K. HARAGUCHI, N. SUGANO, T. NISHII, H. MIKI, K. OKA, H. YOSHIKAWA Fig. 3 Case 1. Photomicrographs of the retrieved zirconia head. Diagram showing A) the pole of the head, B) at the equator of the head, C) 5 mm below the equator and D) the lower border of the head. The distribution of the craters varied with the site. The surface near the pole (A) was relatively smooth while that near the equator was rough (C). The arrows on the diagram indicate the sites of observation. Results The retrieved heads appeared glossy to the naked eye, but photomicrographs showed surface roughening (Fig. 3) which increased at the pole of the femoral head, at the equator, and 5 mm below the equator (Table II). It was lowest at the pole and highest 5 mm below the equator. SEM of the surface of the heads showed numerous craters delineated by the cracks indicating extraction of the surface layers of Y-TZP (Fig. 4), while SEM of the surface of an unused head revealed a smooth polished surface with some small voids of less than 0.5 m (Fig. 5). SEM of the surface of the crater revealed a stratified crystal structure with some voids suggesting transformation from the tetragonal phase to the monoclinic phase (Fig. 6A) and SEM of the inner bulk of the heads showed a tetragonal crystal structure and no abnormality suggesting phase transformation (Fig. 6B). The monoclinic contents on the surface were about 30% in case 1 and 20% in case 2 (Table II). They were similar regardless of the site of measurement. Table II. Surface roughness in Ra and monoclinic content of retrieved zirconia heads Case 1 Case 2 Site of Monoclinic Monoclinic measurement Ra ( m) content (%) Ra ( m) content (%) Pole Equator Below the equator Discussion Zirconia ceramics were introduced in 1985 as a promising new material for THA. 3 The superior mechanical strength, in comparison with alumina ceramics, could reduce the rate of fracture of femoral heads while allowing the use of Fig. 4 Case 1. SEM of the retrieved zirconia head. The craters delineated by the cracks indicate extraction of the surface layers of the zirconia ceramics. THE JOURNAL OF BONE AND JOINT SURGERY

4 PHASE TRANSFORMATION OF A ZIRCONIA CERAMIC HEAD AFTER TOTAL HIP ARTHROPLASTY 999 Fig. 5 SEM of the surface of the unused zirconia head. Figure 5A Some white bands about 30 m in length were observed. Figure 5B At the site of a white band, consolidation of the ceramic grains was relatively poor and small voids less than 0.5 m were observed. Fig. 6 Figure 6A SEM inside the crater revealed a stratified crystalline structure containing some voids indicating transformation from the tetragonal phase to the monoclinic phase. Figure 6B SEM of the inner bulk of the zirconia head showed a tetragonal crystal structure and no abnormality suggesting phase transformation. heads 4 of smaller diameter and retaining the excellent wear performance against polyethylene. 5,6 These characteristics were considered to be important for improving the longevity of a THA. However, recently published results of THA using a zirconia head have not been encouraging. 7,8 One cause is thought to be phase transformation in vivo. 9 Zirconia ceramics have three phases of crystal structure, which transform according to temperature. The monoclinic phase transforms into a tetragonal phase at less than 1100 C, while the tetragonal phase transforms into the cubic phase at 2370 C. Since the tetragonal phase is unstable, but shows the greatest mechanical strength of the three phases, Y-TZP was used for the surgical grade zirconia ceramics. Transformation from the tetragonal phase into the monoclinic phase brings a 3% increase in the volume of ceramics. While this phase transformation plays an important role in increasing the mechanical strength in Y-TZP, when it occurs extensively it may cause an increase in the surface roughness. Phase transformation in Y-TZP is induced at a relatively low temperature in the presence of water and pressure. 1 Therefore, sterilisation of Y-TZP in an autoclave is contraindicated. VOL. 83-B, NO. 7, SEPTEMBER 2001

5 1000 K. HARAGUCHI, N. SUGANO, T. NISHII, H. MIKI, K. OKA, H. YOSHIKAWA In case 1 the cause of the recurrent dislocation appeared to be wear of the polyethylene. 10 The retrieved polyethylene liner showed a linear wear pattern superolaterally, but no abnormal abrasive wear of the rim suggesting impingement. SEM failed to show the scratches caused by dislocation on the surface of the retrieved head. The distribution of the craters on the surface of the Y-TZP head was circumferential around the axis of the neck and depended on the site on the head. These findings suggest that recurrent dislocation may not have been the cause of the surface roughening. In case 2, the head was retrieved because of dislodgement of the ring of the polyethylene liner. Laboratory tests carried out by the manufacturer have shown that impingement of the femoral neck on the polyethylene liner caused dislodgement of the ring and the locking mechanism has now been changed. The cause of phase transformation of the Y-TZP heads is still speculative, but heat and pressure contribute. Lu and McKellop 11 reported that the surface temperature of the polyethylene of a Y-TZP head rose to 99 in a hip simulator. An interesting finding is that the monoclinic content is similar all over the surface of the head, while surface roughness varies. This implies that phase transformation may not always cause uniform surface roughening, and that the manufacturing process of Y-TZP may influence surface deterioration after phase transformation. SEM of an unused Y-TZP head has shown small voids on the surface, but they are so small that they do not affect the Ra value. Stress concentration on the voids may cause acceleration of phase transformation of the head. Further analysis regarding phase transformation of other Y-TZP heads with a different manufacturing process is required, and careful follow-up is essential after THA using this head. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. References 1. Chevalier J, Cales B, Drouin JM. Low temperature aging of Y-TZP ceramics. J Am Ceram Sec 1999;82: Yamamuro T, Nakamura T, Iida H. Bone-conservative cementless hip prosthesis made of high-tech materials: Kobelco H-5. Rinshou Seikei Geka 1999;34: Piconi C, Meccauro G. Zirconia as a ceramic biomaterial. Biomaterials 1999;20: Cales B. Zirconia as a sliding material: histologic, laboratory and clinical data. Clin Orthop 2000;379: Piconi C, Burger W, Richter HG, et al. Y-TZP ceramics for artificial joint replacements. Biomaterials 1998;19: Hamada Y, Horiuchi T, Sano Y, Usui I. The wear of a polyethylene socket articulating with a zirconia ceramic femoral head in canine total hip arthroplasty. J Biomed Mater Res 1999;48: Allain J, Le Mouel S, Goutallier D, Voisin MC. Poor eight-year survival of cemented zirconia-polyethylene total hip replacements. J Bone Joint Surg [Br] 1999;81-B: Hernigou P, Babrami T. Osteolysis and acetabular wear of polyethylene against zirconia ceramic in comparison to alumina ceramic in total hip arthroplasty. Procs 67th Annual Meeting of the American Academy of Orthopaedic Surgeons, Orlando, Hamadouche M, Sedel I. Ceramics in orthopaedics. J Bone Joint Surg [Br] 2000;82-B: Orozco F, Hozack WJ. Late dislocations after cementless total hip arthroplasty resulting from polyethylene wear. J Arthroplasty 2000;15: Lu Z, McKellop H. Frictional heating of bearing materials tested in a hip joint wear simulator. Proc Inst Mech Eng [H] 1997;211: THE JOURNAL OF BONE AND JOINT SURGERY