Intended for healthcare professionals


Treatment of displaced intracapsular hip fractures in older patients

BMJ 2010; 340 doi: (Published 11 June 2010) Cite this as: BMJ 2010;340:c2810
  1. Jan Erik Madsen, professor of orthopaedic trauma
  1. 1Orthopaedic Department, Oslo University Hospital, Ullevaal, 0407 Oslo, Norway
  1. j.e.madsen{at}

    Total hip arthroplasty is preferable to hemiarthroplasty in healthy patients

    In the linked systematic review (doi:10.1136/bmj.c2332), Hopley and colleagues compare outcomes after total hip replacement versus hemiarthroplasty when treating displaced femoral neck fractures in older patients.1 Hip fractures cause considerable death and disability in elderly people. Worldwide, 1.6 million new hip fractures occurred in 2000, and these accounted for the loss of 2.35 million disability adjusted life years (DALYs) annually and 1.4% of the burden of disease in women in the Western world.2 The incidence of hip fractures is estimated to rise to more than six million in 2050, around half of which will be femoral neck fractures.3

    Displaced femoral neck fractures in elderly people are treated by internal fixation or prosthetic replacement, with either a hemiarthroplasty replacing the femoral head and neck or a total hip arthroplasty which also includes acetabular replacement. These treatments are associated with different complication rates, function, and independency of living. Hemiarthroplasty produces consistently better function than internal fixation,4 5 6 and lower reoperation rates (10% v 40%).7 No significant differences have been seen for mortality, perhaps because individual studies and reviews lack sufficient statistical power to detect them.8 Internal fixation is therefore mainly used for undisplaced fractures, whereas arthroplasties are favoured for displaced fractures. Hemiarthroplasty has been more popular than total hip arthroplasty, perhaps because total hip arthroplasty had inferior results in some early reports.9 However, over the past few years, an increasing body of evidence supports the use of total hip arthroplasty instead of hemiarthroplasty in a selected group of physically and mentally fit patients.10

    In Hopley and colleagues’ systematic review and meta-analysis, data from 15 studies, seven of which were randomised, and 1890 arthroplasty procedures showed a lower risk of reoperation after total hip arthroplasty compared with hemiarthroplasty (relative risk 0.57, 95% confidence interval 0.34 to 0.96, risk difference 4.4%).1 Furthermore, total hip arthroplasty showed better hip function after one to four years (mean difference 5.4/100 points in Harris hip score). No significant difference was seen for the risk of dislocation (1.48, 0.89 to 2.46) and other general complications (1.14, 0.87 to 1.48).

    Ten of the 15 studies included only physically and mentally fit patients, and the rest did not specify patient related inclusion criteria. Thus, the results may not be representative of the average patient with a femoral neck fracture. In addition, only seven of the 15 studies were randomised, and proper concealment of randomisation was reported in only four. Even if a selection bias in the non-randomised studies could not be shown, the observational and randomised studies differed. The lower risk for reoperation after total hip arthroplasty was mainly driven by the observational studies, the benefit disappeared when the high quality studies were analysed separately. The same tendency was seen for dislocations, whereas hip function was consistently better after total hip arthroplasty.

    A wide variety of prosthetic implants were used in the studies. In the total hip arthroplasty group important details like femoral head size and surgical approach may have affected the dislocation rates. Hemiarthroplasties were unipolar or bipolar and uncemented or cemented. When modern hemiarthroplasties are used, cemented and uncemented stems seem to perform equally well.11 Bipolar designs, with dual articulation between the large femoral head and the acetabular cartilage and between the head and femoral stem, were introduced to improve hip movement and function, but their benefits have yet to be proved.12

    Although many different hemiarthroplasties have been used with good results, it is generally accepted that old monobloc cementless types, like the Austin Moore prosthesis, should be avoided because early loosening and subsidence lead to impaired function.7 The use of this prosthesis may have skewed the functional results in favour of total hip arthroplasty.

    Nevertheless, the review shows that total hip arthroplasty is a safe procedure in fit patients with femoral neck fracture, and that it produces better functional results than hemiarthroplasty. However, total hip arthroplasty may be associated with a higher rate of complications and a reoperation rate of 0-9% is still reported in the included studies. Thus, there is room for future improvement for both hemiarthroplasty and total hip arthroplasty. Prosthetic design, with use of larger heads or (semi-) constrained joints, and the choice of surgical approach, may reduce dislocation rates and other complications. Along with patient selection, these factors should be included in further studies. Long term results are also needed; total hip arthroplasty may result in mechanical complications such as dislocations and acetabular loosening, which level out early functional benefits.

    While awaiting such evidence, both forms of hip replacement should be considered complementary treatment modalities. Surgeons should tailor treatment individually and in particular evaluate comorbidities, ambulatory status, and cognitive function. Physically and mentally fit patients may have better functional results with total hip arthroplasty than with hemiarthroplasty. Frailer patient should still be treated with a modern type hemiarthroplasty.


    Cite this as: BMJ 2010;340:c2810


    • Research, doi:10.1136/bmj.c2332
    • Competing interests: The author has completed the Unified Competing Interest form at (available on request from the corresponding author) and declares: (1) No support from any company for the submitted work; (2) He has received lecture honorariums and travel expenses from Synthes, Stryker, and Smith & Nephew; (3) No spouse or children with financial relationships that may be relevant to the submitted work; and (4) No non-financial interests that may be relevant to the submitted work.

    • Provenance and peer review: Commissioned; not externally peer reviewed.


    View Abstract

    Log in

    Log in through your institution


    * For online subscription