ARTICLE IN PRESS

Current Orthopaedics (2005) 19, 385–392

www.elsevier.com/locate/cuor

HIP

Girdlestone resection arthroplasty of the hip:

Current perspectives
H. Sharma, C.R. Dreghorn, E.R. Gardner

Victoria Infirmary, Glasgow G42 9TY, UK

KEYWORDS Summary Modern technological advancements in revision hip arthroplasty have

Girdlestone revolutionised the treatment of failed primary total hip replacements. The decision
operation; to perform a Girdlestone operation is taken as a last resort, particularly for
Resection medically sub-optimal and functionally compromised patients, who have a high
arthroplasty; anaesthetic and operative risk at one-stage and two-stage reimplantations.
Excision Girdlestone resection arthroplasty should be considered as a salvage procedure,
arthroplasty; primarily aimed at pain relief and infection control. Such patients must be warned to
Hip expect 2–3 in of limb shortening and reliance upon a walking aid postoperatively.
& 2005 Elsevier Ltd. All rights reserved.

Introduction There are several retrospective studies published

on long-term outcome of this salvage procedure,
With an increase in life expectancy, the number of mainly infected total hip arthroplasty with variably
patients with primarily replaced and revised hips is reported results.1–10 The indication for the Girdle-
increasing dramatically. Although, revision total hip stone operation has now become a salvage proce-
arthroplasty has revolutionised the treatment of dure. This article is an overview of Girdlestone
failed primary total hip replacements, medically resection arthroplasty of the hip with special
sub-optimal and functionally compromised patients, regard to indications, patient selection, surgical
who have a high anaesthetic and operative risk, may technique, mortality and morbidity characteristics,
not be suitable for any further major interventions, outcome analysis and prognostic factors influencing
especially one-stage and two-stage reimplantations. outcome.
In such cases, Girdlestone resection arthroplasty
(i.e. removal of prosthesis and/or cement) is
considered to be an acceptable salvage option.
Historical perspectives
Corresponding author. Tel.:+44 141 639 3697;
The first report of resection arthroplasty (removal of
fax: +44 141 201 5082.
E-mail addresses: [email protected] (H. Sharma), the head and the neck of femur) as a treatment for
[email protected] (C.R. Dreghorn), septic arthritis of the hip was published by White in
[email protected] (E.R. Gardner). 1849.11 In 1928, Gathorne Robert Girdlestone

0268-0890/$ - see front matter & 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.cuor.2005.06.005
 ARTICLE IN PRESS
386 H. Sharma et al.

(1881–1950), an orthopaedic surgeon from Oxford, Secondary or modified

described a similar procedure for the tuberculous
hip joint.11 In 1943, Girdlestone popularised it for Used for failed hip replacements or failed con-
the treatment of late septic arthritis. Girdlestone struction after hip trauma (Figs. 1 and 2).
himself did not report any long-term functional For all practical purposes, any Girdlestone
results. procedure performed nowadays can be considered
In 1945, Batchelor recommended dividing the as a secondary or modified Girdlestone procedure.
femoral neck distally, flush with the lesser trochan-
ter, to achieve smooth surfaces. He observed that
irregular osseous spurs or prominences might cause Why resection, why not revision?
considerable pain with motion. He also advocated
performing an abduction osteotomy in conjunction The modern Girdlestone operation involves the
with the resection of the femoral head and neck to removal of the prosthesis and/or cement following
improve stability.11 The modern Girdlestone proce- septic or aseptic loosening of a total hip prosthesis,
dure predominantly consists of removal of the hemi-prosthesis or a failed osteosynthesis. It has
prosthesis and/or cement following septic and proved to be an effective salvage procedure, for
aseptic loosening of total hip arthoplasty or hemi- controlling pain and infection. With the advance-
prosthesis.11 ments in revision hip technique and technology,
Girdlestone resection arthroplasty is rarely indi-
cated as a primary procedure. It is a salvage
procedure, and it should not be considered as an
Types of girdlestone resection alternative to one- or two-stage reimplantations.
arthroplasty Indications include infective and aseptic loosen-
ing of total or hemi-arthroplasties, recurrent
The Girdlestone procedure can be described into dislocations and un-united operated femoral neck
two main categories: fractures.
Girdlestone pseudarthrosis may also be consid-
ered as the first stage of a two-stage revision. The
Primary decision to perform a resection arthroplasty with-
out reimplantation of a second prosthesis is based
Performed for primary hip disorders like septic hip, upon multiple factors. Important considerations
tuberculous hip, and rarely for osteoarthritis and include infection with multiple organisms or bac-
rheumatoid arthritis. teria resistant to antibiotic therapy, poor quality

Figure 1 Early Girdlestone for septic total hip arthroplasty.

 ARTICLE IN PRESS
Girdlestone resection arthroplasty of the hip: Current perspectives 387

Surgical technique of girdlestone

procedure
A standing anteroposterior radiograph of the pelvis
should be taken to plan the resection. The role of
preoperative traction is debatable. A direct lateral or
posterior approach through the previous scar should
include the excision of any sinus or scar. Iliopsoas
tenotomy and/or adductor tenotomy may be re-
quired. Femoral preparation comprises removal of
the prosthesis (with or without trochanteric osteot-
omy), removal of cement and smoothing of the
transected femoral surface. Utmost care should be
observed to prevent femoral shaft fractures intrao-
peratively. Tissue specimens should be sent for both
bacteriology and histology. Acetabular preparation
consists of the removal of the prosthesis and cement.
Primary wound closure is preferred even in infected
cases.14 Intra- and immediate postoperative blood
loss may vary from 1 to 5 l.

Radiographic classification
Grauer et al.11 described four possible levels of
proximal femoral resection:
Figure 2 Late Girdlestone with proximal migration of
greater trochanter nearly abutting against lower ilium.
Type I—a substantial portion (41.5 cm) of the
femoral neck remains, usually performed for
failed resurfacings,
Table 1 Indications for resection. Type II—a small portion of the femoral neck
Infection with multiple organisms or bacteria remains (1.5 cm or less),
resistant to antibiotic therapy Type III—intertrochanteric resection,
Poor quality local soft tissues Type IV—sub-trochanteric resection.
Unacceptable complexity of any possible
reconstruction The obvious clinical implication of this classifica-
Refusal by the patient to have another operation tion is that the more proximal the resection, the
after removal of the implant better is the overall function, walking and activity
Patients with systemic disease or poor overall of the patient. Contrary to this observation, no
health
correlation could be found between the radiologi-
Inadequate bone stock
Or combinations of these factors
cal appearances and the quality of the result in
some studies.15,16

local soft tissues, unacceptable complexity of the

reconstruction, refusal by the patient to have Overall outcome
another operation after removal of the implant,
and patients with systemic disease, poor overall Girdlestone arthroplasty patients cannot be as-
health, inadequate bone stock or combinations of sessed with the usual parameters routinely applied
these factors.12,13 The problem of repeated opera- for hip evaluation.3 The functional outcome in
tions, prolonged morbidity, intercurrent illnesses previous studies on Girdlestone arthroplasty is
and repeated invasive investigations after an illustrated in Table 2. The reported results of
infected prosthesis leads to depression and dis- Girdlestone arthroplasty are not uniform. Satisfac-
satisfaction. Table 1 therefore summarises the tory results have been reported by Campbell
circumstances when resection may be the appro- et al.,19 Mallory,21 Ahlgren et al.,17 Grauer
priate option. et al.,11 Bohler and Salzer18 and Castellanos
 ARTICLE IN PRESS
388 H. Sharma et al.

Table 3 Favourable outcome determinants with

Girdlestone resection arthroplasty.

Sharma
et al.23

88.88%

77.77%
Old patient (Marchetti et al.,3 Bittar and Petty22

94.4%
100%
found poor results in younger patients)

18

—
Male
Unilateral16
Healed wound5
Esenwein

Non-diabetic (Rittmeister et al.27 found poor

et al.2

results in diabetics)

81.5

59.3
100
5.2
27
— Smooth inter-trochanteric line5
Summary of the functional outcome with Girdlestone resection arthroplasty of the hip in the previously published studies.

No cement left in the femoral canal1,27. Contrary

paper by Ballard et al.15
Castellanos

Less limb shortening/conservative proximal

et al.20

femoral resection11
100

Reduced level of expectation

4.1
78
83
86

83
Unfavourable preoperative condition with strong
pain or persistent infection
et al.12
Bourne

et al.,20 while poor outcomes were recorded Clegg,1

33
91
97
85

79
4

Petty and Goldsmith,5 Bittar and Petty,22 McElwaine

and Colville4 and Esenwein et al.2 In Table 3, the
favourable outcome determinants with Girdlestone
resection arthroplasty have been listed.
Clegg1

4–7.5
100
30
90
80

Mortality analysis
Mallory21

Intra- or postoperative mortality is reported to be

100
100
3.8

between 7% and 62% following Girdlestone opera-

10
90

90

tion for infected total hip replacements.20 A higher

postoperative mortality follows the Girdlestone
procedure for failed hip fractures than after
Ahlgren
et al.17

revision for failed arthroplasty.23

100
100

100
4–5
27
89

Morbidity analysis
Campbell
et al.19

Complications include infection with persistent

Good

100

discharging sinuses and fistulae, haemorrhage and

45

73

88
—

hypovolaemia, proximal femoral fracture, traction-

related problems (pin-site infection, common per-
Goldsmith5
Petty and

oneal nerve compression, joint stiffness, contrac-

tures), the effects of immobilisation (decubitus
100
5.4

ulcers, urinary infection, chest infection, disuse

21

76

15
0

osteoporosis, muscle wasting), persistent pain,

thrombo-embolism, psychological disturbances (de-
Leg length discrepancy
Need for walking aid

pression, psychosis, suicidal tendency), a Trende-

Control of infection

Patient satisfaction

lenburg gait and generalised fatigue.

No. of patients
Relief of pain

Age, sex, side and body mass index

Table 2

(in cm)

Younger patients have higher functional demands

and expectation so they are likely to be dissatisfied
 ARTICLE IN PRESS
Girdlestone resection arthroplasty of the hip: Current perspectives 389

with the operation. The results are reported to be (83%—Parr et al.; 86%—Castellanos; 97%—Bourne
significantly poorer in women, particularly older et al.).12,20,24
ones as reported by Grauer et al.11 The patients Castellanos et al.20 found no correlation between
were satisfied with the operation in unilateral cases the type of organisms and the persistence of
as a secondary operation, but it was generally infection, although, Kantor et al. identified worse
unsatisfactory as a primary procedure or when functional results in patients with chronic drai-
performed bilaterally.16 Grauer et al.11 also ob- nage.10 Clegg1 advocated a complete removal of all
served that body weight, height and body habitus the cement in order to achieve an eradication of
pose no statistically significant influence on pain, infection. Practically, it can be quite difficult to get
walking or function. rid of all cement remnants (Figs. 3–5). We agree
with the views of Petty and Goldsmith,5 Ahlgren
et al.,17 Bourne et al.12 that small amounts of
retained cement do not seem to influence wound
Pain relief healing after resection arthroplasty.
The primary goal of the Girdlestone procedure is
pain relief. Adequate pain relief was observed in
60% (Scalvi et al.), 77% (Ballard et al.), 80% (Parr
Leg length discrepancy and need for
et al.), 83% (Castellanos), 91% (Bourne et al.) and walking aids
89% (Sharma et al.).6,12,15,20,23,24
Most of the studies reported limb shortening of
approximately 4–6 cm.2,12,20,25 The degree of short-
ening is often dependent on the amount of bone
Infection control lost from the proximal femur and the quality of the
scar tissues at the time of surgery.11 Associated
Control of infection has been reported in 73–100% gluteus medius insufficiency magnifies the need for
cases postoperatively following the Girdlestone walking aids. McElwaine and Colville4 noted that
operation.1,5,11,18 Sharma et al.23 achieved 100% calipers were found to be unacceptable in the
infection control in the surviving patients, similar majority of patients. Grauer et al.11 reported a
to Mallory21 and Ahlgren et al.17. Infection control positive correlation between shortening and level
was achieved in the majority of the studies of resection, patients with less shortening walking

Figure 3 Septic failure of revision total hip arthroplasty with recurrent dislocation is a common indication for
Girdlestone procedure.
 ARTICLE IN PRESS
390 H. Sharma et al.

Figure 4 Pre-Girdlestone radiograph showing aseptic Figure 5 Post-Girdlestone radiograph of the same pa-
recurrent dislocation of total hip arthroplasty. tient. Note that stable components, cement and circlage
wires could be left alone in these relatively high
better, although the difference was not statistically anaesthetic risk patients.
significant.
Electromyocinesigraphic26 examination per-
formed in Girdlestone patients in order to study Delayed reimplantation following
the automatic function of the muscles with patients girdlestone arthroplasty
standing and walking confirmed that there was no
innervation of the hip abductors but high activity in Rittmeister et al.27 reported greater patient satis-
the rectus femoris during standing. The contral- faction and better function if Girdlestone hips were
ateral abductor group and the trunk muscles were converted to a hip arthroplasty rather than being
hyperactive during standing and walking. left with the excisional procedure. The incidence of
postoperative complications and revisions were
similar for both groups. Charlton et al.28 in retro-
Overall satisfaction spective study showed a high rate of dislocation
(11.4%) and persistent limp (39%) following delayed
Subjective satisfaction varies between 14% and conversion. A high dislocation rate following con-
100%.5,12,18,23 in the reported series. It is hard to version of the Girdlestone procedure to secondary
attach much credibility to this overall measurement total hip arthroplasty relates to soft tissue con-
in view of this major discrepancy in reported results. tracture, limb length discrepancy, deficient bone
stock and malpositioning of the components. A
constrained acetabular component should be con-
Failed total hip replacements versus sidered to reduce the dislocation rate.
failed hip fractures Schroder et al.29 followed two groups of patients:
32 patients had a long-standing pseudarthrosis; in
The mortality was higher in the failed fracture the other group of 16 patients, a total hip
group (68%) in comparison to those with failed replacement was reimplanted at an average of 3
arthroplasty23 (48%) (Figs. 6 and 7). years after a pseudarthrosis. The improvement in
 ARTICLE IN PRESS
Girdlestone resection arthroplasty of the hip: Current perspectives 391

Figure 7 While performing Girdlestone operation in the

Figure 6 Aseptic loosening of Thompson’s uncemented same patient, an intraoperative femoral shaft fracture
hemi-arthroplasty subsequently underwent Girdlestone occurred. It was successfully treated non-operatively.
procedure.

hip function after the reimplantation was marginal resection arthroplasty following failed operated hip
and the results were comparable to a well- trauma do considerably worse than after failed total
functioning pseudarthrosis. Personal satisfaction hip arthroplasty. There is no significant difference
and the activities of daily living were marginally between the long-term outcomes of Girdlestone
better in the reimplantation group, (Harris hip procedures performed at a District General Hospital
score 64 compared to 58 in those with a pseudar- compared to a University Teaching Hospital.
throsis). Brandt et al.30 stated that prosthesis The decision to perform a Girdlestone operation
removal and delayed reimplantation arthroplasty is mostly taken as a last resort, as all the suitable
is an effective treatment to limit the recurrence of candidates are filtered off for revision surgery,
Staphylococcus aureus prosthetic joint infection, before reaching the stage of this salvage operation.
provided there is no evidence of infection at the The decision between revision and resection should
time of reimplantation arthroplasty. not be taken lightly and it should not be considered
as an alternative to one-stage or two-stage
reimplantations. Such patients must be warned to
Conclusions expect 2–3 in of limb shortening and reliance upon
a walking aid postoperatively. This operation can be
Various studies have confirmed that the Girdlestone made acceptable by proper explanation to the
procedure is very effective in achieving its primary patient with realistic expectations.
goals of infection control and pain relief for
irreversibly failed total hip joints and to salvage
failed operated hip trauma. A high mortality and Practice points
a poor functional outcome could be attributed to
a higher age group, poor general health and highly  The primary goals of the Girdlestone opera-
selected group of patients, who were unfit for tion are pain relief with infection control. It
reimplantation surgery. Patients who have had
 ARTICLE IN PRESS
392 H. Sharma et al.

is a salvage procedure, and should not be prosthesis; a retrospective study of 40 patients. Acta Orthop
considered as an alternative to one-stage or Belg 1991;57:109–13.
two-stage reimplantations 10. Kantor GS, Osterkamp JA, Dorr LD, Fischer D, Perry J,
Conaty JP. Resection arthroplasty following infected total
 A high mortality is observed in the literature hip replacement arthroplasty. J Arthroplasty 1986;1:83–9.
with such groups of patients which can be 11. Grauer JD, Amstutz HC, O’Carroll PF, Dorey FJ. Resection
attributed to higher age group, poor general arthroplasty of the hip. J Bone Joint Surg Am 1989;71:
health and highly selected group of patients, 669–78.
who are unfit for reimplantation surgery 12. Bourne RB, Hunter GA, Rorabeck CH, Macnab JJ. A six-year
follow-up of infected total hip replacements managed by
 Patients must be warned to expect 2–3 in of
Girdlestone’s arthroplasty. J Bone Joint Surg Br 1984;66:
limb shortening and invariable assisted 340–3.
mobility postoperatively 13. Muller RT, Schlegel KF, Konermann H. Long-term results of
the Girdlestone hip. Arch Orthop Trauma Surg 1989;108:
359–62.
Research directions 14. Horan FT. Robert Jones, Gathorne Girdlestone and excision
arthroplasty of the hip. J Bone Joint Surg Br 2005;87:104–6.
15. Ballard WT, Lowry DA, Brand RA. Resection arthroplasty of
 Future studies with larger number of pa- the hip. J Arthroplasty 1995;10:772–9.
tients utilising results from multiple centres 16. Haw CS, Gray DH. Excision arthroplasty of the hip. J Bone
may reveal more meaningful results with Joint Surg Br 1976;58:44–7.
primary pathology diagnosis stratification 17. Ahlgren SA, Gudmundsson G, Bartholdsson E. Function after
 Need for a separate questionnaire for removal of a septic total hip prosthesis. A survey of 27
Girdlestone hips. Acta Orthop Scand 1980;51:541–5.
assessing Girdlestone patients 18. Bohler M, Salzer M. Girdlestone’s modified resection
 To compare the long-term outcome of arthroplasty. Orthopedics 1991;14:661–6.
patients with septic retained implants with 19. Campbell A, Fitzgerald B, Fisher WD, Hamblen DL. Girdle-
those treated by excisional arthroplasty stone pseudarthrosis for failed total hip replacement. J Bone
 To assess the role of Girdlestone resection Joint Surg Br 1978;60:441–2.
20. Castellanos J, Flores X, Llusa M, Chiriboga C, Navarro A. The
arthroplasty for MRSA (Methicillin Resistant
Girdlestone pseudarthrosis in the treatment of infected hip
Staphylococcus Aureus) infected hips replacements. Int Orthop 1998;22:178–81.
21. Mallory TH. Excision arthroplasty with delayed wound
closure for the infected total hip replacement. Clin Orthop
1978;137:106–11.
References 22. Bittar ES, Petty W. Girdlestone arthroplasty for infected
total hip arthroplasty. Clin Orthop 1982;170:83–7.
23. Sharma H, De Leeuw J, Rowley DI. Girdlestone resection
1. Clegg J. The results of the pseudarthrosis after removal of
arthroplasty following failed surgical procedures. Int Orthop
an infected total hip prosthesis. J Bone Joint Surg Br
2005;29:92–5.
1977;59:298–301.
24. Parr PL, Croft C, Enneking WF. Resection of the head and
2. Esenwein SA, Robert K, Kollig E, Ambacher T, Kutscha-Lissberg
neck of the femur with and without angulation osteotomy.
F, Muhr G. Long-term results after resection arthroplasty
A follow-up study of thirty-eight patients. J Bone Joint Surg
according to Girdlestone for treatment of persisting infec-
Am 1971;53:935–44.
tions of the hip joint. Chirurg 2001;72:1336–43.
3. Marchetti PG, Toni A, Baldini N, Binazzi R, D’Elia L, Sudanese 25. Herzog T, Link W, Engel S, Beck H. Resection arthroplasty:
A, et al. Clinical evaluation of 104 hip resection arthroplas- middle- and long-term results. Arch Orthop Trauma Surg
ties after removal of a total hip prosthesis. J Arthroplasty 1989;108:279–81.
1987;2:34–41. 26. Chamay A, Taillard W, Kritsikis N, Burckhardt A. Girdle-
4. McElwaine JP, Colville J. Excision arthroplasty for infected stone’s hip resection. Clinical and electromyocinesigraphic
total hip replacements. J Bone Joint Surg Br 1984;66:168–71. study. Orthopade 1987;16:287–94.
5. Petty W, Goldsmith S. Resection arthroplasty following 27. Rittmeister M, Manthei L, Muller M, Hailer NP. Reimplanta-
infected total hip arthroplasty. J Bone Joint Surg Am 1980; tion of the artificial hip joint in girdlestone hips is superior
62:889–96. to girdlestone arthroplasty by itself. Z Orthop Ihre Grenzgeb
6. Scalvi A, Campacci A, Marcer M, Cassini M, Guerra C, 2004;142:559–63.
Ferraresi M, et al. Girdlestone arthroplasty for loosening of 28. Charlton WP, Hozack WJ, Teloken MA, Rao R, Bissett GA.
the total hip prosthesis: evaluation and results. Chir Organi Complications associated with reimplantation after girdle-
Mov 1995;80:279–85. stone arthroplasty. Clin Orthop 2003;407:119–26.
7. Vatopoulos PK, Diacomopoulos GJ, Demiris CS, Gorgolis J, 29. Schroder J, Saris D, Besselaar PP, Marti RK. Comparison of
Papathanassiou BT. Girdlestone’s operation: a follow-up the results of the Girdlestone pseudarthrosis with reimplan-
study. Acta Orthop Scand 1976;47:324–8. tation of a total hip replacement. Int Orthop 1998;22:
8. Renvall S, Einola S. Girdlestone operation. An acceptable 215–8.
alternative in the case of unreconstructable hip arthro- 30. Brandt CM, Duffy MC, Berbari EF, Hanssen AD, Steckelberg
plasty. Ann Chir Gynaecol 1990;79:165–7. JM, Osmon DR. Staphylococcus aureus prosthetic joint
9. De Laat EA, van der List JJ, van Horn JR, Slooff TJ. infection treated with prosthesis removal and delayed
Girdlestone’s pseudarthrosis after removal of a total hip reimplantation arthroplasty. Mayo Clin Proc 1999;74:553–8.