The Significance of Metal Staining on Alumina Femoral Heads in Total Hip Arthroplasty☆

Materials and Methods 

In 56 ceramic-on-ceramic THAs (Foundation Hip, Encore Orthopedics, Inc, Austin, Tex) performed between February 2001 and March 2002 by one of the authors, the alumina femoral head touched the acetabular cup in 4 cases during reduction of the total hip. This occurred in part because of a design feature specific to the shell, consisting of an elevated metal rim at the periphery of the cup. This feature was designed to prevent impingement of the metal neck against the ceramic liner [8]. We did not implant the metal-streaked alumina heads in these 4 cases but retrieved them for analysis instead because the significance of the surface metal staining was unknown. These 4 alumina heads comprised group 1 of the specimens reported here.

Group 2 consisted of 10 additional alumina heads that were retrieved from hips undergoing revision surgery for recurrent instability. Group 2 femoral heads were divided into groups 2A and 2B (Table 1). Group 2A consisted of 4 alumina heads from total hips with 2 or 3 dislocations or subluxations. Two heads in group 2A were from the same series of patients as group 1 alumina heads, and 2 were from patients who had the initial hip surgery done at another institution.

Table 1. Clinical Data for Specimens in Groups 2A and 2B

	Group	Specimen no.	Alumina type (28-mm heads)	Patient age (y)	No. of dislocations or subluxations before retrieval	Damage to metal cup	Damage to alumina liner
	2A	1	Biolox Forte	47	3	No	No
		2	Biolox Forte	67	2	No	Light stain
		3	Biolox Forte	51	3	No	No
		4	Biolox Forte	48	3	No	No
	2B	1	Biolox (32-mm size)	53	>3 (numerous)	Yes	Chipped
		2	Biolox (32-mm size)	74	4	Yes	No
		3	Biolox Forte	67	4	Yes	No
		4	Biolox Forte	67	>3 (numerous)	Yes	No
		5	Biolox Forte	65	>3 (numerous)	Yes	Fragmented
		6	Biolox	52	5	Yes	Chipped

Group 2B consisted of 6 alumina femoral heads retrieved from total hips with 4 or more dislocations or subluxations. These came from a series of 3813 ceramic-on-ceramic THAs performed between June 1990 and September 2004 by one of the authors. The 6 patients from whom these alumina heads were retrieved chose to delay revision surgery beyond the fourth episode of instability for a variety of reasons.

All except 2 femoral heads in this study were 28 mm in diameter, marketed under the trade name of Biolox or Biolox Forte (CeramTec AG, Plochingen, Germany). Biolox Forte represents a third-generation alumina ceramic that differs from the earlier generation Biolox in that it is treated by hot isostatic pressing and has a smaller grain size with less impurity [9].

All specimens were inspected for any evidence of gross damage to the femoral head surface or to the acetabular components (alumina liners and titanium metal shells). Alumina femoral head surface roughness was measured in the area of metal discoloration by a technician who was blinded to the origin of the specimens and not associated with this study. A 3-dimensional noncontact profilometry system was used to measure surface roughness using white light and a chromatic aberration technique (Micromeasure Plus 3D, Micro Photonics, Inc, Irvine, Calif). Roughness was measured over a surface of 1.5 × 1.5 mm with a step size of 3 μm. It was reported in micrometers using the unfiltered parameter Sa defined as the arithmetic mean of the deviation from the mean. Sa was used instead of Ra to eliminate the effect of surface texture directionality in the roughness value. Control measurements were made on new Biolox Forte alumina heads that were provided by the manufacturer. To detect possible differences in the mean surface roughness among groups, we used exact Kruskal-Wallis test, followed by a 2-way exact Wilcoxon rank sum test to compare pairs of groups. Data were analyzed by an independent statistician. Statistical significance was set at a P value of less than .05 (Table 2).

Table 2. Comparison of Surface Roughness of Alumina Femoral Heads With and Without Metal Staining

		Group			P
		1	2A	2B
	Alumina surface	New, unused	Metal-stained	Metal-stained and scratched
	n	4	4	6
	Surface roughness of alumina⁎ (Sa)	0.064 (0.053-0.074)	0.210 (0.075-0.381)	–	.0286
		0.064 (0.053-0.074)	–	0.352 (0.197-0.488)	.0286
		–	0.210 (0.075-0.381)	0.352 (0.197-0.488)	.1143

⁎ Mean values are reported with range specified in parentheses.

After the profilometry measurements were made, the metal-stained alumina heads were wiped with a cotton cloth dipped in dilute nitric acid (0.1 N), rinsed twice with ethanol, and dried with a jet of air. This treatment is known to dissolve metal stains without affecting the alumina. After removing the metal in this manner, the underlying alumina surface on each femoral head was examined with light microscopy, followed by scanning electron microscopy (SEM; JEOL T330A) with an energy dispersive x-ray analysis (4pi Revolution using KEVEX Quantum detector, Durham, NC) attachment for microchemical analysis of the metal stain. Specimens were coated with carbon before SEM analysis to prevent electrostatic charging. Adjacent nonstained areas of each specimen that also received the dilute nitric acid treatment served as internal controls.

Light microscopy was used to provide a lower power magnification analysis of the alumina surface. This was followed by SEM for a high-power analysis of the surface. Energy dispersive x-ray analysis was used for information on the composition of the metal stains on the alumina surface.

Results 

None of the 14 alumina heads in this study had any gross damage or irregularity on the alumina surface. Dark stains or streaks were the only evidence of contact with metal. Group 1 alumina heads had 0.5- to 1.0-mm-wide isolated linear streaks of metal staining (Fig. 1) that could be removed with dilute nitric acid, leaving a smooth surface (Fig. 2). The surface roughness of the metal stained areas was greater than that of adjacent nonstained areas. However, once the metal deposits were removed with nitric acid, SEM analysis of previously stained surfaces was comparable to adjacent unstained areas. Analysis of the metal stain before its removal showed that its composition was that of a Ti-6Al-4V alloy, transferred from the acetabular component.

View Large Image Download to PowerPoint

Fig. 1. Metal staining of alumina head after contact with the metal acetabular rim during hip arthroplasty.

View Large Image Download to PowerPoint

Fig. 2. SEM photomicrograph of an alumina head surface showing superficial metal deposition (left) that is removed with a chemical etch, leaving the undamaged alumina surface (right).

Group 2A alumina heads had metal staining associated with 2 or 3 episodes of hip instability; each of these had a sharply defined area of metal discoloration that was several millimeters in width and length (Fig. 3). Metal stains on these heads could not be wiped clean with dilute nitric acid. Scanning electron microscopy analysis of the stained areas demonstrated a range of features, such as uneven cracks, embedded metal particles, deep grooves, and pits in the alumina surface (Fig. 4), whereas adjacent areas remained smooth and undamaged. The acetabular shells and alumina liners from these total hips were undamaged, although the liners had faint, inconsistent metal streaks on the articulating surface.

View Large Image Download to PowerPoint

Fig. 3. Heavier metallic staining on the alumina head that had 3 episodes of dislocation of the prosthetic bearing.

View Large Image Download to PowerPoint

Fig. 4. SEM photomicrographs showing surface irregularities on an alumina head that had 3 dislocations.

Group 2B alumina heads were from total hips with 4 or more episodes of instability; metal stains on these heads were more extensive compared with those of group 2A (Fig. 5). Although all 14 alumina heads in this study were intact, significant macroscopic damage manifested as fragmentation and chipping of the alumina liners was present in 3 of 6 group 2B retrievals. In each case, fretting and damage to the metal acetabular shell were present, with metal and ceramic debris present during revision surgery (Fig. 6).

View Large Image Download to PowerPoint

Fig. 5. Extensive metal staining, but no obvious damage to an alumina head retrieved from a patient with multiple, that is, more than 3 dislocations of the prosthetic hip.

View Large Image Download to PowerPoint

Fig. 6. Acetabular liner and shell retrieved from a patient showing gross damage to the liner and metal.

The surface roughness of alumina heads with metal staining but no surface damage (group 1) was greater than that of control alumina; these measurements were made before the stain was removed (mean Sa, 0.210 vs 0.064; P = .0286, Table 2). Similarly, the surface roughness of alumina heads with metal staining and surface damage (group 2) was greater than that of control alumina (mean Sa, 0.352 vs 0.064; P = .0286). However, surface roughness did not differ significantly between the metal-stained groups 1 and 2 (mean Sa, 0.210 vs 0.352; P > .05).

Discussion 

Metal staining of alumina femoral heads has been reported during intraoperative reduction of total hip components 6, 7, during subluxations and relocations of the prosthetic hip joint 7, 10, 11, and after erosion of alumina heads through metal shells with failed polyethylene liners [12]. It has been speculated that metal transfer into the articulation could explain the occasional cases of excessive wear of alumina-on-alumina and alumina-on-polyethylene bearings 13, 14, 15, 16, 17, 18. Alumina discoloration occurs because particles from a surface metal oxide layer detach from the underlying titanium surface and are transferred to the alumina [19]. The titanium alloy Ti-6Al-4V, which is widely used in metal acetabular shells, is a relatively soft material that is associated with tissue discoloration and the release of particulate metallic debris when used as a bearing 20, 21. In the present study, none of the specimens had evidence of stripe wear, which is also associated with scratching and damage to alumina from edge loading of ceramic-on-ceramic articulations 11, 22.

Our results show that a one-time contact with metal results in metal streaking of the alumina, but the underlying ceramic surface remains undamaged. Thomsen and Breusch [23] observed metal streaks in 7 of 20 alumina heads explanted from 3 to 13 years after implantation; none was associated with pitting or scratching on SEM analysis. Metal streaking in these cases probably occurred during reduction of the total hip and went unrecognized by the surgeon [23]. Although metal streaks could be removed experimentally in our study, we are unaware of any method that will clean a metal-stained alumina head during surgery.

Our results show that metal staining increases the surface roughness of alumina. Although alumina-on-alumina bearings are relatively resistant to 3-body wear, interposed metal particles could theoretically contribute to this mode of wear. Although comparable data are not available for alumina, 3-body abrasion wear testing of zirconia ceramics using Ti-6Al-4V particles as third-body debris have shown no evidence of abrasion, material removal, or subsurface damage 24, 25. Third-body abrasive damage on yttria-doped alumina ceramic femoral heads has been reported, but the damage could have been from yttria-based material released from the bearing [26].

Although the retrievals in this study were from ceramic-on-ceramic articulations, alumina metal staining may also be a concern in alumina-on-polyethylene articulations. Damage to a highly cross-linked polyethylene liner because of scratching and pitting of a metal-stained alumina head in a total hip with instability has been reported [27]. Metal staining of alumina heads was associated with increased surface roughness and increased polyethylene wear in a series of revision THAs with alumina-on-polyethylene articulations [10].

Although not investigated in this study, metal transfer probably occurs when cobalt-chrome heads contact the acetabular shell, but the discoloration is difficult to detect [7]. Experimental transfer of titanium particles onto cobalt-chrome surfaces results in increased surface roughness and abrasive wear [24]. Retrieved cobalt-chrome femoral heads have shown microscopic surface damage, probably because of third-body wear from metal particles entrapped in the articulation [28].

Our results suggest that linear metal streaks on alumina from a 1-time contact with metal during primary THA are superficial. As few as 2 total hip dislocations can result in metal discoloration, surface alterations, and increased roughness of alumina femoral heads. The 6 total hips with 4 or more dislocations had more extensive metal discoloration and surface roughness. These hips also had gross damage to the acetabular rim and the alumina liner, resulting in the accumulation of ceramic and metal debris in the joint space. A limitation of this investigation is the small sample numbers available to us. Nonetheless, these data suggest that the greater the number of adverse events such as contact with metal, the higher the likelihood of damage to the alumina surface. Further data are needed to clarify the effect of interposed metal particles on the wear of ceramic-on-ceramic articulations. Until then, it appears prudent to avoid using a metal-streaked alumina head in primary THA and to revise unstable alumina-on-alumina hips early.