Evid Based Med 18:22-23 doi:10.1136/ebmed-2012-100678
  • Therapeutics
  • Systematic review

Insufficient evidence to recommend a specific bearing surface in implantable hip devices: comparative studies find little difference in outcomes

  1. Sam Oussedik2
  1. 1Department of Orthopaedics, University College Hospital, London, UK
  2. 2Department of Trauma and Orthopaedics, Northwick Park Hospital, Harrow, UK
  1. Correspondence to: Fares Haddad
    Department of Orthopaedics, University College Hospital, 235 Euston Road, London NW1 2BU, UK; fsh{at}

Commentary on: [Abstract/FREE Full text]


Total hip replacement surgery (THR) is a highly successful procedure for relieving painful degenerative conditions of the hip and restoring function. The articulating surfaces of the prosthetic joint can be manufactured from a number of different materials. Traditionally, metal (cobalt-chromium-molybdenum alloys) has been employed for the femoral head bearing surface and polyethylene for the acetabular bearing surface. Polyethylene wear particles have been implicated in loosening of the components and the need for revision surgery. More recently, the focus is on alternative bearing couples (combinations of materials) in an attempt to improve the longevity of THR and reduce the rate with which revision surgery is required. To this end, metal on metal, ceramic on ceramic and ceramic on metal bearings have been employed. Possible advantages include improved longevity and the ability to use a larger diameter articulation, thus reducing risk of dislocation. Possible disadvantages include exposure to high concentrations of metal ions and resultant adverse tissue reaction with metal on metal, risk of bearing fracture and squeaking with ceramic on ceramic and increased cost.


A thorough literature search of Medline, Embase, the Cochrane Controlled Trails Register and the National Joint Registry was conducted. In order to assess unpublished literature, the authors worked with the Food and Drug Administration (FDA) in the USA to identify all safety and effectiveness summaries and mandatory postmarket studies submitted to the agency regarding hip prostheses.

Three authors abstracted data from the identified publications and examined methodological quality and publication bias before carrying out quantitative and qualitative analyses of the reported results.


The authors identified 18 randomised trials and comparative observational studies which were included in the analysis. Methodological quality was found to be moderate to high in four studies, moderate in five studies and low in six studies. There was no significant heterogeneity. Comparisons of functional results between types of bearing surfaces revealed a mean difference of 2.4 points (on the 100-point Harris hip score) favouring metal on polyethylene over metal on metal bearing couples. No other significant differences were found.

Comparisons of the need for revision surgery between types of bearing surfaces revealed no significant differences in the comparative literature. Analysis of data from National Joint Registers showed a higher revision rate for metal on metal designs in the Australian, New Zealand and England and Wales National Registers. Ceramic on ceramic and ceramic on polyethylene revision rates differed across registries with no bearing couple clearly favoured.


This systematic review is very appropriate at a time when the need for joint replacement surgery continues to rise and comes in the wake of recently well-publicised problems with metal on metal designs.1 The methodology is scientifically robust for analysing the findings of prospective randomised trials, of which there are very few, and for looking at registries which produce valuable hypothesis generating data.

Of the 18 comparative studies identified in the English language literature, only four were found to be of moderate to high quality. The evidence from the joint registries is, unfortunately, undermined by the heterogeneity in the type of data reported (eg, the Swedish register does not stratify results by bearing surface).

Classes of bearing surfaces have been analysed without the ability to take into account subclass differences. There are different polyethylenes, metals, ceramics and each implant type may perform differently. This must be acknowledged as a potential weakness of the evidence base and influences the interpretation of any results. The review makes clear that there is a need for further study. The lack of comparative data in the literature also precludes subgroup analysis by age or gender. The findings are also limited to use in osteoarthritis and avascular necrosis. Although the review appropriately addressed a focused question on bearing surfaces, other questions are relevant. For example, other prosthetic design features may have a role in functional results and longevity, such as femoral and acetabular component design and fixation type.

We must use caution in interpreting the results of this systematic review and meta-analysis because their applicability is limited, in the registry data, to failing implants that are revised. No assumptions can be made about those that are not revised. These could produce better, more functional results for longer than the traditional technology. Thus, if a device is seen to fail early in a young population it may be because this population is more active. When functioning well, the device may allow a greater and more satisfactory activity level. Moreover, some interventions, if applied in expert hands and in the optimal patient population, may be very successful and have advantages over traditional therapies. If applied inappropriately, however, they are highly likely to fail and be captured by registries.

The authors call for harmonisation of the methods and data reported across all registers and to this end the FDA has initiated the International Consortium of Orthopaedic Registries. An additional call for a high-quality prospective randomised controlled trials is made.

The current evidence does not demonstrate superiority of any bearing couple. This paper clearly highlights the need for more rigour in the way that new bearing surfaces are introduced and that any future changes must be introduced in a stepwise fashion through phase 1 trials, prospective randomised studies and, ultimately, careful registry review.


  • Competing interests None.


Free Sample

This recent issue is free to all users to allow everyone the opportunity to see the full scope and typical content of EBM.
View free sample issue >>

Don't forget to sign up for content alerts so you keep up to date with all the articles as they are published.