ORIGINAL RESEARCH

International Journal of Surgery 9 (2011) 310e313

Contents lists available at ScienceDirect

International Journal of Surgery

journal homepage: www.theijs.com

Original Research

Girdlestones excision arthroplasty: Current updateq

I. Basu*, M. Howes, C. Jowett, B. Levack
Orthopaedic Department, Queens Hospital, Rom Valley Way, Romford, RM7 0AG, United Kingdom

a r t i c l e i n f o a b s t r a c t

Article history: Girdlestones procedure has become a salvage operation reserved for patients with significant co-mor-
Received 28 September 2010 bidities. Recent literature addresses this infrequently used intervention inadequately. This observational
Received in revised form study aims to update current literature and review the modern role of this intervention in orthopaedic
24 November 2010
practice. Twenty-four records were obtained from which patient demographics, indications and co-
Accepted 27 January 2011
Available online 16 February 2011
morbidities were investigated. Seventeen patients completed an abridged Harris Hip Scoring questionnaire
and commented on satisfaction. The average age was 78 years and patients had multiple co-morbidities.
Dementia was the most frequent condition but several patients suffered from cardiovascular and respi-
Keywords:
Girdlestones
ratory disease. The most common operative indication was persistent prosthetic infection with Staphylo-
Resection arthroplasty coccus aureus, the most common pathogen. Overall mortality was 41% but all surviving patients had
Hip complete resolution of infection and 65% had adequate pain control. No patients mobilised without aids
Prosthetic infections although 29% of patients were able to manage stairs and 29% were able to mobilise outdoors. Only 29% were
Revision surgery unsatisfied with the outcome. This study demonstrates that Girdlestones candidates are an ageing high-
Orthopaedics risk group and shows that the Girdlestones procedure can, in select cases, provide good functional
outcomes. However such intervention comes at the expense of high mortality and should therefore only be
used as a last resort.
Ó 2011 Surgical Associates Ltd. Published by Elsevier Ltd. All rights reserved.

1. Introduction undergoing the procedure and to investigate the functional

Excision arthroplasty of the hip was first documented in surgical
practice over a century ago. However it was made popular in 1943
by Gathorne Girdlestone who used the technique for the treatment
for septic arthritis of the hip.1 Thus his name has been synonymous
with the procedure, which is currently used to address problems
arising from failed hip surgery including peri-prosthetic infections
or recurrently dislocating prostheses. With antimicrobial therapy
and one and two stage revision procedures becoming increasingly
effective, definitive excision arthroplasty has become a salvage
operation, reserved for those with significant co-morbidities or for
situations in which revisions and washouts have repeatedly failed.
Although the procedure is a well-documented surgical interven-
tion, there is little current evidence identifying the patient groups
involved or the functional outcome following the procedure. This
study aims to update current literature, to assess the patient group
q Part of the data has previously been presented at the IOS Conference 2010 with
the abstract due to be published in the JBJS(Br).
* Corresponding author. 10 High View Close, Loughton, Essex IG10 4EG, United
Kingdom. Tel.: þ44 7795175634.
E-mail addresses: ibasu@doctors.org.uk (I. Basu), Matthew.Howes@bhrhospitals.
nhs.uk (M. Howes), charliejowett@hotmail.com (C. Jowett), brian.levack@ntlworld.
com (B. Levack).
outcome of patients following this procedure.
2. Methods
To include sufficient patients to be consistent with previously published liter-
ature, the study was carried out in two phases. The study was started in 2002 and
completed in 2010. All patients undergoing definitive Girdlestones procedure were
identified from theatre records firstly between 1995 and 2002 and later between
2005 and 2010. Forty-one separate patients were identified of which 27 sets of notes
were obtainable, 3 of which were temporary notes and did not include documen-
tation of operative intervention. The remaining 24 sets were used to investigate
operative indications and patient demographics. Sixteen sets had clearly docu-
mented co-morbidities at the time of their admission clerking and these were used
for analysis.
Of the 41 patients initially identified 17 were deceased at time of follow-up and
were therefore excluded from functional analysis, 7 were unable or unwilling to
participate or could not be contacted which left 17 participants suitable for functional
analysis follow-up. The reasons for being unable to complete a questionnaire
predominantly included those suffering with severe mental illness or profound
dementia with no consistent carers to discuss functional outcome on their behalf.
Follow-up time from time of operation ranged from 1 year to 7 years with an average
time to follow-up of 22 months. Patients were contacted via telephone and asked to
complete the functional component of the Harris Hip Scoring Questionnaire. Patients
were also asked to comment on their overall satisfaction following the procedure. For
patients with dementia for whom a relative or carer could be contacted who was
aware of the patients’ functional status before and after the operation, the relative or
carer was asked to complete the functional aspect of the questionnaire on the
patients’ behalf.
Data was subsequently collated and analysed in Excel.

1743-9191/$ e see front matter Ó 2011 Surgical Associates Ltd. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijsu.2011.01.012
 ORIGINAL RESEARCH

I. Basu et al. / International Journal of Surgery 9 (2011) 310e313 311

3. Results Table 1
The most frequent organisms.

Of the 41 cases identified 17 patients had died at the time of
follow-up suggesting an overall mortality of 41%. Of the 24 sets of
notes obtained there were 16 females and 8 males with an average
age of 78 years at the time of operation. The ages ranged from 42 to
96 years. The most common indication was infected prosthesis
which made up over 60% of cases. Dislocation of total hip replace-
ment was the next frequent making up 29% of cases (Fig. 1). Of the
infected prostheses Staphylococcus aureus and Enterococcus faecalis
were the most common causative organisms however no growth
was the most common finding (Table 1).
Sixteen patients had clear documentation of co-morbidities at the
time of operative intervention. The average number of co-morbidities
was 3.6. Dementia was the most frequent independent medical
condition, affecting 5 patients. Six patients had respiratory co-
morbidities including asthma and COPD. Cardiovascular co-morbid-
ities including past myocardial infarctions, atrial fibrillation or
congestive cardiac failure affected 8 patients. Four patients were found
to have diabetes, 3 patients were known to have had TIA’s or CVA’s and
3 were known to have recurrent DVT’S and/or PE’s (Table 2).
Functional analysis was conducted on 17 patients. In all sur-
viving patients the infection had been controlled and their wounds
had healed. 29% of these patients reported no pain and 65% were
thought to have adequate pain control reporting only mild pain,
slight pain or no pain at all. 41% of patients were left with a severe
limp yet 23% reported no limp. All patients were either severely
restricted in their walking or required significant support. No
patients were able to mobilise without aids although 29% of
patients were able to manage stairs following the procedure and
29% were able to mobilise outdoors with walking aids. 59% of
patients were able to bend to put on their socks and shoes and over
70% of patients were able to sit comfortably in a chair for longer
than 5 h. Having scaled up the abridged Harris Hip Scores the
average score was 51, which is considered unsatisfactory by normal
standards. Despite this only 29% of patients were unsatisfied with
the procedure (Fig. 2).
4. Discussion
Over the last few decades revision surgery has become the
surgical option of choice for patients suffering from failed pros-
theses. Girdlestones procedure is now used exclusively for high-
risk surgical patients with poor post-operative prognosis.2 As
demonstrated in this study patients who are offered Girdlestones
procedure today are elderly with an average age of 78 years. This is
inline with average ages found in recent studies from Sharma et al.
in both 2005 and 2006.3,4 Earlier studies from Petty and Goldsmith
Organism Frequency
No growth 6
Staphylococcus aureus 3
Enterococcus faecalis 2
Coliforms 1
Pseudomonas 1
Streptococcus (unspecified) 1
Not done 1
in 1980s and Grauer in 1989 involved much younger patients1,5 and
such discrepancies may reflect advancements in technologies and
techniques restricting the number of eligible candidates to those
with limited physiological reserve.
The average number of co-morbidities noted from this study is
3.6. Several patients had significant cardiovascular and respiratory
disease and several suffered with dementia and other psychologi-
cally detrimental conditions. Such patients would be expected to
have poor surgical outcomes and unsatisfactory results from pro-
longed post-operative rehabilitation. With increasingly frail patient
groups associated with more recent studies one may expect
increasingly poor post-operative outcomes. Indeed mortality rates
seem significantly higher in later studies. Recent literature quotes
figures between 58% and 68% not dissimilar from the figure of 41%
noted in this study.3,4 However these are all significantly higher
than the majority of figures quoted over a decade ago. Despite the
apparent relationship between mortality rates, functional outcome
is not so closely related. Marchetti et al. demonstrated an associa-
tion between increasing age and improved functional outcomes.6
However this may simply be a reflection of lower expectations in
more elderly co-morbid patients as throughout the literature there
have been mixed reports on functional outcomes with no obvious
correlation between ages or pre-morbid state and functional out-
comes.1,5,7e14 This study would suggest a high level of satisfaction
in those undergoing the procedure today.
There is a well-documented difficulty in finding conventional
assessment tools which accurately reflect outcome for this
intervention. Several studies have highlighted the inappropri-
ateness of traditional assessment tools used in this context.6,15
Assessments validated for use in Hip Surgery such as the full
Harris Hip Score consider factors such as range of motion, leg
length discrepancy and fixed flexion contractures as part of the
overall assessment. This is clearly inappropriate for assessing
excision arthroplasty which unavoidably reduces muscle bulk,
causes leg length discrepancy and leads to inherent and
unavoidable limitations to range of motion. Similarly more
generic assessments such as the SF-36 fail to consider concurrent

4.17% Table 2
4.17% Cases affected by co-morbid conditions.
INFECTED Co-morbididty No. of cases
DISLOCATION Respiratory dysfunction (COPD, ASTHMA) 6
AVASCULAR NECROSIS Cardiac impairment 4
LOOSE PROSTHESIS Atrial fibrillation 4
29.17% Cerebral ischaemic events 3
Dementia 5
Diabetes 4
Dvt’s/PE’s 3
62.50% Hypertension 5
Anaemia 2
Neurological dysfunction (Epilepsy, MS, Learning difficulties) 3
Carcinoma 1
Venous insufficiency 1
Osteoporosis 2
Fig. 1. Pie chart demonstrating the indication for procedure.
 ORIGINAL RESEARCH

312 I. Basu et al. / International Journal of Surgery 9 (2011) 310e313

0.00%
a 17.65% b
23.53%
29.41%

No Pain No Limp
41.18%
Slight Pain Slight Limp
Mild Pain Moderate Limp
17.65% Moderate Pain 5.88%
Severe Limp
Marked Pain
Crippled

11.76% 23.53%
29.41%

0.00%
17.65%
c d
No Support
35.29% Cane for Long Walks
41.18%
Cane most of time
Easily
With 1 Crutch
With Difficulty
With 2 Canes
Unable to do
With 2 Crutches/Frame

82.35%
23.53%
0.00%
e 23.53%
f 17.65%
23.53%
Unlimited

6 Blocks
5.88%
3 Blocks
5.88%
5.88% Indoors
Normally
Bed to Chair
70.59% Using Railing
2 rails or Sitting
Unable to Manage

5.88% 47.06%
g h 23.53%

23.53% Sit comfortably >5hrs Satisfied

In High Chair for <30mins Unsatisfied

Unsure
Unable to Sit
58.82%
17.65%
70.59%

Fig. 2. Pie charts demonstrating functional status of the patient group: (a) Pain; (b) Limps; (c) Putting on shoes and socks; (d) Sitting comfortably; (e) Stair assessment; (f) Support
when walking; (g) Walking distance; (h) Overall satisfaction.

disabling co-morbidities shared by the majority of Girdlestone
candidates which results in a sub-optimal assessment. Several
studies have suggested more simple assessments which focus on
resolution of infection, reduction in pain and the ability to
manage activities of daily living comfortably are more appro-
priate.15 However, due to such small number of patients
undergoing this procedure it has been almost impossible to
validate such systems in any meaningful sense. This study uses
an abridged version of the Harris Hip Score which considers pain
and activities of daily living. This tool gives a fair representation
of functional outcome when taken in context and produces
results which can be easily compared to historical studies.
 ORIGINAL RESEARCH
I. Basu et al. / International Journal of Surgery 9 (2011) 310e313
The questions of limp and of support whilst walking consis-
tently scored poorly. All mobile patients required walking aids
when mobilising independently. Such high dependency on walking
aids is not unique to this study and is thought to be primarily due to
the leg length discrepancy and muscle weakness inherent to the
operative technique.2 Dependence on walking aids is a necessary
consequence rather than a specific reflection of poor outcome.
Irrespective of this approximately 76% were able to mobilise
indoors and 29% were able to mobilise outdoors suggesting that for
such an elderly group of patients adequate mobility was obtained.
After recovery from operative intervention 65% of patients
experienced mild pain to no pain at all and were considered to have
adequate pain control. Past literature suggests that adequate pain
control is obtained in anything from 35%1 to 100%4 of patients
making our findings consistent with previous studies.
All patients involved in this study had complete resolution of
local infection which is consistent with studies published by
Sharma et al.3,4 Grauer1 and Ahlgren et al.12 The most frequent
causative organisms in this study were found to be staphylococci
followed by Enterococci. The same organisms were found to be
commonly responsible in studies from Castellanos et al.11 and more
recently from Rittmeister et al.16 However the most common
finding in this study, which is not routinely supported by the
literature, was sterile culture. In such cases aggressive antibiotic
therapy administered pre-operatively is likely responsible for
eradicating organisms from puss samples.
Ultimately 59% were happy with the outcome, 18% of patients
were unhappy with the long-term result, and 23% were indifferent.
These findings are similar to those published by Esenwein et al. in
2001.10 Of those who committed to yes or no answers 77% were
happy with the overall outcome in spite of the consistently “sub-
optimal” functional outcome scores calculated from the abridged
Harris Hip Scoring System. This observation is consistent with the
general consensus that traditional tools used independently are
inappropriate for assessing outcome following this procedure.2,6
5. Conclusion
In conclusion this paper demonstrates that the Girdlestone
procedure is being reserved for an increasingly elderly population
with multiple co-morbidities. This study confirms that for such
group of patients the procedure can still offer acceptable functional
outcomes and can be a worthwhile surgical option when employed
appropriately.
Despite modern technology pushing the Girdlestones excision
arthroplasty into surgical small print it remains a valuable addition
313
to the surgical arsenal for challenging patients with difficult
pathology. In such cases, and when all other avenues have been
exhausted the Girdlestones excision arthroplasty can be the kindest
choice for significantly unwell patients and is therefore well worth
remembering in modern day practice.
Ethical approval
None.
Sources of funding
None.
Conflicts of interest
None.
References
1. Grauer JD, Amstutz HC, O’Carroll PF, Dorey FJ. Resection arthroplasty of the hip.
J Bone Joint Surg Am 1989;71(5):669e78.
2. Sharma H, Dreghorn CR, Gardner ER. Girdlestone resection arthroplasty of the
hip: current perspectives. Curr Orthopaedics 2005;19(5):385e92.
3. Sharma H, De Leeuw J, Rowley DI. Girdlestone resection arthroplasty following
failed surgical procedures. Int Orthop 2005;29(2):92e5.
4. Sharma H, Kakar R. Outcome of Girdlestone’s resection arthroplasty following
complications of proximal femoral fractures. Acta Orthop Belg 2006;72(5):
555e9.
5. Petty W, Goldsmith S. Resection arthroplasty following infected total hip
arthroplasty. J Bone Joint Surg Am 1980;62(6):889e96.
6. Marchetti PG, Toni A, Baldini N, Binazzi R, D’Elia L, Sudanese A, et al. Clinical
evaluation of 104 hip resection arthroplasties after removal of a total hip
prosthesis. J Arthroplasty 1987;2(1):37e41.
7. Clegg J. The results of the pseudarthrosis after removal of an infected total hip
prosthesis. J Bone Joint Surg Br 1977;59(3):298e301.
8. Bittar ES, Petty W. Girdlestone arthroplasty for infected total hip arthroplasty.
Clin Orthop Relat Res 1982;170:83e7.
9. McElwaine JP, Colville J. Excision arthroplasty for infected total hip replace-
ments. J Bone Joint Surg Br 1984;66(2):168e71.
10. Esenwein SA, Robert K, Kollig E, Ambacher T, Kutscha-Lissberg F, Muhr G. Long-
term results after resection arthroplasty according to Girdlestone for treatment
of persisting infections of the hip joint. Chirurg 2001;72(11):1336e43.
11. Castellanos J, Flores X, Llusa M, Chiriboga C, Navarro A. The Girdlestone
pseudarthrosis in the treatment of infected hip replacements. Int Orthop
1998;22(3):178e81.
12. Ahlgren SA, Gudmundsson G, Bartholdsson E. Function after removal of a septic
total hip prosthesis. A survey of 27 Girdlestone hips. Acta Orthop Scand
1980;51(3):541e5.
13. Mallory TH. Excision arthroplasty with delayed wound closure for the infected
total hip replacement. Clin Orthop Relat Res 1978;137:106e11.
14. Bohler M, Salzer M. Girdlestone’s modified resection arthroplasty. Orthopedics
1991;14(6):661e6.
15. Molfetta L, Bassetti M, Benvenuti M, Caldo D. Analysis of clinical data in
patients with Girdlestone arthroplasty: a new score. Hip Int 2007;17(3):170e5.
16. Rittmeister ME, Manthei L, Hailer NP. Prosthetic replacement in secondary
Girdlestone arthroplasty has an unpredictable outcome. Int Orthop
2005;29(3):145e8.