Only a few of the more than 200,000 total hip arthroplasties performed
annually in the United States are done after an infection of the
hip joint or the proximal aspect of the femur.
Although some infections produce severe early destruction of the
hip joint, most patients, if treated promptly, regain good hip function
and do not present again until much later in life, when secondary
degenerative changes have occurred1.
It is therefore necessary to be aware of the possibility of a previous
infection and to inquire about it specifically.
The first recurrence of bone infection may be delayed for many
decades. Gallie2 reported a case
of femoral osteomyelitis in a ten-year-old girl that did not recur
until after nearly eighty years. The risk that a previous infection
of the hip region poses to a hip prosthesis is multifactorial. The
type of infection (osteomyelitis or septic arthritis), the level
of activity of the infection (active or quiescent), the time since
the infection (recent or historical), the organism (pyogenic, tuberculous,
or fungal), and the reconstruction technique all contribute to the outcome.
Osteomyelitis
Ever since Staphylococcus aureus was first isolated
from osteomyelitis by Pasteur3 in
the late nineteenth century, it has remained the predominant infecting
organism, implicated in approximately 90% of infections
where an organism is isolated (range, 88% [140
of 159] to 95% [392 of 411])4,5. The infecting organisms in the remaining
cases are largely Staphylococcus epidermidis and
streptococci.
More recently, there has been a shift in the prevalence of certain causative
organisms, with fewer infections caused by Staphylococcus
aureus. There has also been a significant increase (from
eight of sixty-five cases to eight of nineteen cases; p < 0.001)
in the proportion of cases that are subacute and that have a greater
tendency to recur4,6.
With modern antibiotic management, most recurrences of osteomyelitis
are apparent within the first year5.
Gillespie and Mayo5 found that,
of 18% (119) of 655 osteomyelitis cases that recurred, 66% (seventy-nine)
did so by six months; 16% (nineteen), between six and twelve
months; 12% (fourteen), between one and five years; and
only 6% (seven), after more than five years. In 50% of
these cases, there was only one episode of recurrence.
Septic Arthritis
As in osteomyelitis, in septic arthritis the most common infecting
organism is Staphylococcus aureus. Kelly7 found Staphylococcus aureus in
77% of 141 adult hips with septic arthritis, gram-negative organisms
in 16%, and streptococci in 4%. Haemophilus
influenzae, which is rare in adults, has a peak prevalence
in two-year-old children, whereas gram-negative organisms, which
generally have a worse outcome, have a higher prevalence in adults7.
While the clinical picture, with pain localized to the involved joint,
suggests the diagnosis of septic arthritis in older children and
in adults, in infants the diagnosis is often difficult and delayed.
The outcome can vary from complete resolution with little damage
to the joint to a coxa magna, a spontaneous osseous or fibrous ankylosis,
or complete destruction of the proximal aspect of the femur with dislocation.
Trochanteric overgrowth or acetabular dysplasia may occur, and both
can have an adverse effect on future arthroplasty.
Infection of the hip joint, compared with that of other joints,
is particularly incapacitating. Morrey et al.8 found
that eighteen of thirty-seven infected hips in children had a clinically
unsatisfactory result; in comparison, only six of thirty-one other
joints with infection had an unsatisfactory result.
Tuberculous Bone and Joint Infections
Although the incidence of pulmonary tuberculosis has fallen dramatically
over the last few decades, the incidence of extrapulmonary tuberculosis
has remained steady, with almost 400 cases of skeletal tuberculosis per
year in the United States; 13% (ninety) of 676 cases reported
in the literature have involved the hip9,10.
Infection rates in the developed world have been increasing in subgroups
of the population because of various host factors (such as immunosuppression
and malnourishment) as well as the development of increasingly resistant
strains. Although patients who are infected with the human immunodeficiency virus
do not appear to have an increased risk of contracting tuberculosis,
increased rates of reactivation have been observed in these patients11,12. Radiographically, tuberculous bone
infections normally have a destructive appearance, with adjacent
osteopenia and minimal sclerosis; on the other hand, they may also
simulate pyogenic arthritis, with a florid periosteal reaction,
sclerosis, and sequestrum formation. The main feature distinguishing
tuberculous bone infection from pyogenic infection is the marked
destruction, to the point that one may suspect the presence of a destructive
neoplasm; however, this is not seen in all cases.
Tuberculosis can be easily overlooked when the differential diagnosis
is made, and it should always be considered in cases of hip destruction.
Hecht et al.13 described four
cases of reactivation of previously unsuspected tuberculosis in
patients who had a normal erythrocyte sedimentation rate, level
of C-reactive protein, and leukocyte count preoperatively and for
whom no intraoperative cultures had been performed.
Fungal Infections
Fungal infections of the joint differ from pyogenic or tuberculous infections
in that there is less destruction of the joint14-16.
Spontaneous infection of the natural hip joint is usually caused
by noncandidal fungi, whereas candidal infection at the site of
a prosthetic joint is usually associated with immunosuppression,
intravenous therapy, drug abuse, or direct inoculation. Candida
is the only fungus reported to have caused infections at the site
of hip arthroplasties17.
Other Organisms
While a wide variety of other organisms such as brucella18 and syphilis are known to infect bone,
isolation of these organisms from bone can be difficult, and infections
with these organisms at the site of a joint arthroplasty have not
yet been described, to our knowledge.
Intravenous Drug Abuse
The high risk of joint infection is related to the activity of
drug injection and does not appear to be increased by the presence
of the human immunodeficiency virus. Munoz-Fernandez et al.19 demonstrated that the prevalence
of joint infection in intravenous drug abusers who were positive
for the human immunodeficiency virus (twenty-five [5%] of
482) was similar to the prevalence in intravenous drug abusers who
were not (six [7%] of eighty-five). They
also found that, when they compared the thirty-one joint infections
in these 567 intravenous drug abusers (irrespective of whether they had
the immunodeficiency virus) with twenty-one joint infections in
616 patients who were not intravenous drug abusers, there was a
fourfold increase in the prevalence of infection of the hip, sacroiliac
joint, and sternocostal joint (64.5% compared with 16.6%)
among the intravenous drug abusers. It is therefore reasonable to
assume that, while active intravenous drug abuse presents an unacceptable
risk of additional hip infection after total hip arthroplasty, the
risks for a patient who has truly stopped abusing drugs should be acceptable.
Hemoglobinopathies
Patients with hemoglobinopathies such as sickle-cell disease have
an increased risk of bone and joint infection. Engh et al.20 reported that twenty of seventy patients
with complete hemoglobinopathies who were followed for ten years
required treatment for osteomyelitis. All ten fully documented infections
were caused by gram-negative organisms; eight of them were due to Salmonella
species. With collapse of the femoral head, the radiographic features
of sickle-cell disease give little indication as to whether an infection
is present. Bone-remodeling in sickle-cell disease usually produces
a positive technetium bone scan; however, the degree of concordance
between this study and an indium leukocyte scan may help to differentiate
between infection and infarction as a cause of pain. If in doubt,
a joint aspiration may be necessary.
Ongoing Sources of Infection
Although treatable sources of ongoing infection remote from the
hip should obviously be addressed, the actual threat to the success
of a total hip arthroplasty is unclear. Del Sel and Charnley21 described the results of primary total
hip arthroplasty in thirty-one patients with an actively infected
hip joint on the contralateral side (at the site of a hip implant,
a Girdlestone resection arthroplasty, or septic arthritis); five
of the infected hips had a draining sinus. Although antibiotics
were rarely used perioperatively, at a mean follow-up of 3.5 years
(range, eight months to ten years) after the arthroplasty there were
no infections on the "clean" side. Surin et al.22 compared the risk factors in 803 patients
in whom a deep infection had occurred with those in patients in
whom an infection had not occurred. They found that, while the risk
increased threefold (p < 0.05) in the presence of a remote
infection, there was no association between the infecting organisms
at the two different sites and therefore no evidence of a causal
relationship.
History and Examination
While systemic signs such as a fever or local signs such as a
discharging sinus would alert the clinician to the presence of active infection,
in the vast majority of patients with a previously infected hip
that is later treated with arthroplasty, any remaining activity
of the infection is low and these features are absent. Hardinge
et al.1 emphasized the importance
of ensuring that the onset of hip symptoms is slowly progressive. Even
after many decades of quiescence, rapidly increasing pain from a
joint that had previously been infected raises the suspicion that
the infection may have been reactivated, particularly when the patient
had had tuberculosis.
Hematological Investigations
The white blood-cell count is usually noncontributory23 and can often be normal even in patients
with an actively infected hip. For this reason, it cannot reliably
be used to rule out infection. The erythrocyte sedimentation rate
has been extensively investigated in the context of suspected infection
at the site of a total hip arthroplasty. When the erythrocyte sedimentation
rate is 30 mm/hr, the sensitivity for detecting infection
is 60% to 94% and the specificity is 65% to
85%23-26. In the case
of tuberculous infection, however, the preoperative erythrocyte
sedimentation rate has not been found to be predictive of the risk
of reactivation of infection after total hip arthroplasty27. The level of C-reactive protein23,28 is also a useful marker for the diagnosis
of ongoing infection.
Diagnostic accuracy may be improved by considering the erythrocyte
sedimentation rate and the level of C-reactive protein together.
In a study of 202 hip replacements, Spangehl et al.23 demonstrated that all thirty-five that
were complicated by deep infection were in patients who had at least
an erythrocyte sedimentation rate of >30 mm/hr
or a C-reactive protein level of >10 mg/L. They
suggested that a normal erythrocyte sedimentation rate and C-reactive
protein level effectively excludes the possibility of infection
at the site of a total hip arthroplasty, and we use the same criteria
to exclude the possibility of ongoing infection before performing
a primary arthroplasty of a previously infected hip (Table I).
Plain Radiographs
The radiographic appearance of bone infection can be confusing. Initially,
infection is usually more extensive than is radiographically apparent.
However, during the early stages of resolution, the osseous destruction
may appear to progress radiographically. Chronic osteomyelitis may appear
as sclerosis with adjacent osteoporosis, a fusiform thickening of
the cortex, or even an extensive thickened involucrum, but it is
very difficult to ascertain the activity of the infection. Osteoarthritis
secondary to healed septic arthritis should evolve slowly with subchondral sclerosis
and marginal osteophytes. Conversely, when there is reactivation
of an infection, there is usually a more sudden clinical deterioration
with rapid bone destruction, often associated with local osteopenia.
Computerized tomography may demonstrate the anatomy better than
plain radiographs do, but again it gives little indication of the
activity of the infection unless adjacent soft-tissue changes indicate
edema or abscess formation. Even when there is no macroscopic evidence of
infection, scattered microabscesses29 representing
areas of focal acute and chronic osteomyelitis may persist and are
very difficult to identify with standard radiographic techniques.
Magnetic Resonance Imaging
Active osteomyelitis typically appears dark on T1 images and bright
on T2 images, and with chronicity the bright T2 marrow signal becomes
more heterogeneous. In his review of eleven published series in
which magnetic resonance imaging was used to diagnose bone infection, Schauwecker30 found a cumulative sensitivity and
specificity of 95% (188 of 197) and 88% (103 of
117), respectively, for the presence of active infection.
Tang et al.31 concluded that
magnetic resonance imaging was particularly useful for demonstrating
the presence of active infection in patients with previous chronic osteomyelitis,
with a sensitivity of ten of ten. In the presence of abnormal anatomy
as a result of long-standing infection, Mason et al.32 found magnetic resonance imaging
to be more sensitive (eleven of eleven cases) than indium-labeled
white blood-cell scintigraphy (five of eight cases) for demonstrating
active infection.
Scintigraphy
Unfortunately, no nuclear medicine scan is specific for infection,
as such scans simply show different elements of the inflammatory
and metabolic responses of the bone to infection. A technetium-99
methylene diphosphonate scan is a very sensitive test for excluding
the diagnosis of active infection in patients with a chronic or
subacute bone infection; however, it is not very specific33. A negative scan is therefore helpful
for excluding the diagnosis of active infection, but a positive
scan is of little help. In an attempt to increase the specificity
of nuclear scans, gallium citrate and indium-labeled white blood-cell
scans have been used30,33-35.
These scans are best used sequentially, with the technetium-99 scan
used first. If the uptake on the gallium or indium scan is more
intense than it is on the technetium scan, an active infection should
be suspected. Indium-labeled white blood-cell scans, when used sequentially with
technetium scans, are more specific than gallium scans; therefore,
gallium scans are of little use in the diagnosis of active
infection30,34. In a further attempt
to refine the role of nuclear scans, indium-labeled polyclonal antibodies have
been used36. While this technique
may hold promise for the future, its utility has not yet been established.
Aspiration Biopsy
In a study of thirty-one hip arthroplasties, Cherney and Amstutz37 reported no growth on culture of the
preimplantation aspirate from any of seven hips that had a recurrence
of infection after the arthroplasty, and the true value of preoperative
biopsy of joint aspirate has been questioned. In our experience
with hip arthroplasties complicated by deep infection, aspiration
has proven to be a useful diagnostic tool, with a sensitivity of
86% and a specificity of 94% for thirty-five infections
in 180 hips23. Other authors have
reported similar results25,38;
however, we have not yet accumulated sufficient experience with
primary arthroplasty of previously infected hips to support a similar
statement regarding those hips. If doubt persists after a negative
culture of aspirate from the hip joint, we perform multiple core
needle biopsies of the affected bone to obtain material for both
bacteriological and histological analysis to try to distinguish
between osteonecrosis and chronic osteomyelitis. However, we do
not have data to support the efficacy of this practice.
Mantoux Test
Shrivaram et al.10 found that
the Mantoux test (the cell-mediated response to an intradermal inoculation
of purified protein derivative of Mycobacterium tuberculosis)
was strongly positive in all of ten patients with active tuberculous
osteomyelitis of the spine. While, to the best of our knowledge,
such data do not exist for the hip, we still recommend that the
Mantoux test be performed in cases of suspected tuberculous infection
of the hip.
After all nonoperative measures have been exhausted, the surgical options
(other than total hip arthroplasty) after the onset of painful secondary
arthritis in a hip with an active or previous infection are Girdlestone
resection arthroplasty and arthrodesis of the hip, except in the
few cases where an osteotomy may be of benefit.
Girdlestone Resection Arthroplasty
Girdlestone40 popularized resection
of the femoral head and neck, which he first performed in 1923.
He later reported that his first patient could still walk 10 miles
(16 km) a day at the age of seventy-six years, twenty-one years
after this operation40. However,
the reported outcomes of Girdlestone resection arthroplasty have
been variable.
Haw and Gray41 reported patient
satisfaction with thirty-one of forty hips at a mean of ten years;
this increased to twenty-two of twenty-four hips if the resection
had been performed as a secondary or salvage procedure. Other authors42-44 have often found a much less satisfactory
outcome of the procedure. Petty and Goldsmith43 described
moderate-to-severe pain in sixteen of twenty-one patients, while
Bittar and Petty42 described a
poor functional outcome, with only fourteen of thirty-three patients
satisfied at a mean of six years (range, three to thirteen years).
This procedure results in a slow, lurching gait because the abductors
of the hip are at a mechanical disadvantage and the energy efficiency
is less than half that of a normal hip45.
The functional result of a Girdlestone resection arthroplasty, except
when it is performed as a salvage procedure, is generally considered
unacceptable today. There are, however, cases in which this procedure should
be considered, such as with an uncontrollable infection of the hip
or in an intravenous drug abuser in whom repeated episodes of bacteremia
put the hips at considerable risk for additional infection. In addition, it
may be the procedure of choice in some underdeveloped regions of
the world, where implants are not affordable and hip fusion is unacceptable.
Arthrodesis
Following hip arthrodesis, most patients are able to walk at
least 1 mile (1.6 km) and to continue with their occupation46,47. Callaghan et al.46 reported that, after a follow-up of
more than seventeen years, eighteen of twenty-eight patients had
a slight limitation of daily activities and eight had moderate or
severe restrictions. Similarly, after a follow-up of more than twenty
years, Sponseller et al.47 found
that eighteen of fifty-three patients had no substantial limitation;
forty-one of the fifty-three were satisfied, particularly when they
had had the procedure at a younger age. This may no longer apply
with the increased expectations of today’s patients.
The long-term deleterious effects of an arthrodesis of the hip
on other areas, particularly the lumbar spine, ipsilateral knee,
and contralateral hip, are well recognized46-48.
In the study by Sponseller et al.47,
thirty of fifty-three patients who were followed for more than twenty
years had low-back pain, although in none was it considered to be
incapacitating. There is also often a substantial limb-length
inequality, and most patients have a moderate-to-severe limp, a
decreased walking speed, and a gait efficiency of only 53%45,49,50.
While performance of a primary hip arthrodesis after a joint
infection is unusual today, it is not unusual to convert a hip that
was fused or ankylosed following a previous infection to a total
hip arthroplasty, to provide relief of symptoms in other joints.
Arthrodesis of the hip does, however, retain its role in the treatment
of the younger patient after a unilateral hip infection, particularly when
the clinical setting indicates that durability is more important
than mobility.
Outcome of Arthroplasty After Pyogenic Infection
Recurrence of Infection
When reviewing many of the series of total hip arthroplasties, we
found it difficult to determine which cases were true recurrences
of infection and which were newly acquired infections. In most cases,
the isolated organisms are those that are commonly found infecting
total hip arthroplasty sites with no history of sepsis. Therefore,
without phage-typing of the organisms from both episodes, the true
origin of the infection remains uncertain.
The prevalence of deep infection of the hip after one-stage primary
arthroplasty for the treatment of a quiescent infection of the hip
is reported to range from zero of twenty-seven to four of forty-two1,27,51. Hardinge et al.1 found no recurrence of infection in
any of thirteen hips with a quiescent infection in which a cemented
Charnley total hip prosthesis had been implanted without the use
of antibiotics systemically or in the cement. They suggested several
criteria to help rule out active infection before implantation;
these included no wound discharge for at least twenty years and
quiescence for a definite (though unspecified) period. They also recommended
that the patient be in good general health and have had a gradual
deterioration of hip function over a number of years. Serial radiographs
should show slow progression of the degenerative changes with sclerosis
and osteophytes, and both the white blood-cell count and the erythrocyte
sedimentation rate should be normal. Laforgia et al.27 reported deep infection after four
of forty-two arthroplasties in hips with a quiescent infection;
two of the deep infections were related to poor wound-healing. At
the time of arthroplasty, cultures were positive for six of the
forty-two hips, although none of the six had a subsequent deep infection.
In a retrospective review of forty-three infections at the sites
of 3051 hip replacements, Schmalzried et al.52 demonstrated
a twelvefold increase in the rate of infection for hips following
a history of hematogenous septic arthritis (8.2%) compared
with previously uninfected osteoarthritic hips (0.7%).
In their series of total hip arthroplasties performed in hips that
had had a previous infection, Cherney and Amstutz37 demonstrated
a higher rate of recurrent infection in those that had had hematogenous
septic arthritis (four of nine hips) or an infection at the site
of a hip replacement (five of eight hips) than in those
that had had an infection at the site of a hemiarthroplasty (one
of six hips) or internal fixation (two of eight hips). Hardinge
et al.53 reported no cases of
reactivation of infection (tuberculous or pyogenic) in seventeen
fused hips in which osseous union had been present for ten or more
years. In their survival analysis of recurrence of infection after
revisions of hip replacements that had been complicated by deep
infection, Went et al.54 found
that seventy of ninety-eight hips had initially been thought to
be cured of infection but seven of the seventy became infected within
three years and eighteen became infected within ten years.
Reinfecting Organism
When infection occurs at the site of a primary or revision arthroplasty
in a previously infected hip, there is a variable association between
the original organism and that causing the subsequent infection.
When both the original infecting organism and that subsequently
infecting the site of the prosthesis have been known, they have
generally been found to be the same in approximately two-thirds
of cases (Table II)37,55,56. McDonald et al.56 isolated more than one organism in
eight of eleven hips that became reinfected. In a study of two-stage
exchange total hip arthroplasties performed with an antibiotic-loaded
interval spacer in thirty infected hips, Younger et al.57 found that, of five hips with a positive
culture at the second stage, only one had evidence of the same organism
that was grown on a culture of tissue taken from the site of a recurrent infection.
In another study58, of sixty-one
two-stage exchange arthroplasties, they found that, of three recurrences, only
one was caused by the same organism as the one that had caused the
primary infection.
Function After Total Hip Arthroplasty
Following primary arthroplasty in previously infected hips, mean Charnley
hip scores59 (with pain, walking
function, and range of movement each assigned 1 to 6 points) improved from
8.3 to 16.0 points in a series of thirteen hips1 and
from 9.3 to 15.7 points in a series of twenty-eight hips27. In the latter study, there was
a reduction in the need for a cane or crutches from sixteen to five patients27. Jupiter et al.51 found
that, in a series of twenty-four hips with quiescence of infection
for at least one year, the mean Harris hip score60 had
increased from 36 to 86 of a possible 100, equating to 75% excellent
results, 17% good results, and 8% fair results.
The numbers in these studies are small, and while the functional result
does not appear to be as good as that in hips without prior infection,
this may in part reflect the severity of the preexisting hip disease.
In reviewing the radiographs in their series of hip replacements performed
in hips with previous septic arthritis, Laforgia et al.27 found that at a mean of five years
(range, two to seventeen years) there were radiolucencies adjacent
to six of forty-two cups and seven of forty-two stems, results comparable
with those of arthroplasty in hips without prior sepsis.
Outcome of Arthroplasty After Tuberculous Infection
Recurrence of Infection
Laforgia et al.27 reported
reactivation of infection in one of seventy-two hips that had undergone
total hip arthroplasty at a mean of forty years (range, seventeen
to seventy-two years) after the onset of infection. Only eleven
of the seventy-two hips had been treated with perioperative antituberculous
therapy, which was started three weeks before the operation and
was continued for three months afterward. Jupiter et al.51 found no reactivation of infection
in any of seven tuberculous hips that had been treated and had healed.
Similarly, Hardinge et al.1 found
no reactivation in twenty-seven patients and Eskola et al.61 found none in eighteen patients.
Johnson et al.62 reported two
cases of reactivation of tuberculosis after total hip arthroplasty.
In both cases, the childhood infection had not been treated with
antituberculous therapy and, although intraoperative specimens of
the capsule, femur, and acetabulum were all negative on culture
and histological analysis, the tuberculous infection recurred by
one year, after a disease-free interval of thirty-seven and forty-two
years. The density of bacilli in infected bone is much less than
it is in infected sputum, making the diagnosis of tuberculous osteomyelitis
particularly difficult27. Kim
et al.63 reported reactivation
of tuberculosis in one of twenty patients; the patient had discontinued
antituberculous therapy against medical advice. Nonetheless, the infection
fully resolved on completion of the therapy. Opinion differs about
the use of additional perioperative antituberculous treatment to
prevent this small risk of recurrence. Most authors51,62 have suggested that a full course of
perioperative antituberculous therapy is necessary only if it has not
been given previously. This is our own practice. Eskola et al.61, however, recommended that antituberculous
therapy be given perioperatively in all patients, even those who
have been treated previously.
Function After Total Hip Arthroplasty
While there has been good relief of pain and improved walking ability
following total hip arthroplasty in previously infected hips, there
has been a more modest improvement in range of movement63. The complication rate was not unusually
high after seventy-two arthroplasties in one study, but a marked
leg-length discrepancy persisted after ten of them27. Kim et al.63 reported
improvement in the mean Charnley hip score from 7.1 to 13 for twenty
hips that were treated with total hip arthroplasty after previous
tuberculous infection. In a series of seventy-two hips that were treated
with total hip arthroplasty after tuberculous infection, thirty-eight
had been mobile and thirty-four had had a previous fusion27. The mean Charnley hip score for
the thirty-eight mobile hips improved from 8.9 to 15.8, with a corresponding
reduction in cane dependence from sixteen to six patients. For the
thirty-four hips that had had a previous fusion (which was spontaneous in
eleven and operative in twenty-three), the mean Charnley hip score
increased from 10.2 to 14.9 and the use of a cane decreased from
eight to five patients.
In a study by Jupiter et al.51,
the mean Harris hip score increased from 57 to 91, representing
five excellent and two good results.
Eskola et al.61 reported an
improvement in the mean flexion arc of the hip from 40° (range,
0° to 90°) to 86° (range, 55° to 120°). They reported fifteen good
or excellent results, two fair results, and only one poor result
(in a patient with a subsequent deep infection treated with a Girdlestone
resection arthroplasty).
Laforgia et al.27 found radiographic
evidence of cup loosening in seven of seventy-two hips and evidence
of loosening of the femoral component in thirteen at a mean of five years
(range, two to seventeen years).
Preoperative Assessment
In the preoperative clinical assessment, the presence of any sign
of active infection should be sought and excluded. There is often
contracture of the flexors or adductors, which should be quantified,
and the power of the abductor muscles should be assessed, particularly
if they have been violated by a previous operation. Any leg-length
inequality should be determined, and the neurovascular status of the
limbs should be recorded.
In the preoperative radiographic assessment, Judet radiographs should
be made, in addition to the usual radiographs, to assess the bone
stock unless the acetabulum is very well preserved.
Operative Details
The technical difficulty of the arthroplasty depends on the degree
of soft-tissue contracture and the extent of bone loss. Often, when
treatment was early and effective, such that the patient made a
good functional recovery, the anatomy is well preserved and the
technique is not substantially different from that of a routine
primary hip replacement. In other cases, particularly those of neonatal
sepsis, there may be extensive distortion of the pelvic and femoral
anatomy, which may require modified techniques with special components
and sometimes bone graft.
The first decision is whether to perform the reconstruction in one
stage or to delay replacement of the hip until after a thorough
débridement and an interim excision arthroplasty, with
or without the use of an antibiotic depot in the form of antibiotic-loaded
cement beads or an antibiotic-loaded functional spacer57,58,64-67. The second decision is
whether to implant the hip prosthesis with or without cement and,
if cement is used, whether to use plain or antibiotic-loaded cement.
While a detailed discussion of infection after total hip arthroplasty
is not within the scope of this paper, some important points can
be learned from the literature pertaining to the management of infection
at the site of a total hip arthroplasty28,37,55,56,68-76 and,
by inference, can be applied to hip replacement after infection.
It is important to rule out the diagnosis of active infection
in any hip with a previous infection. If active infection
has been ruled out, it appears safe to proceed to hip replacement
in one stage, with anticipation of infection-recurrence rates ranging from
zero of twenty-seven to four of twenty-four1,27,51. After performing primary hip arthroplasty
on hips with quiescent infection, Laforgia et al.27 found
that, at a mean of five years (range, two to nineteen years), infection
or reactivation had occurred in none of nine hips in which gentamicin-loaded bone
cement had been used but in four of thirty-three in which it had
not been used. However, the use of systemic antibiotic therapy had
been erratic, and it had been used in only one of the four patients
who had an infection.
In their review of twenty-nine reported series of infections
after total hip arthroplasty, Garvin and Hanssen69 demonstrated
that, in the absence of antibiotic-loaded cement, a one-stage revision
led to resolution of the infection in 58% (thirty-five)
of sixty hips, whereas a two-stage procedure led to resolution in
82% (130) of 158 hips. Similarly, with antibiotic-loaded
cement, the rate of resolution after the one-stage procedures was
82% (976 of 1189) compared with 91% (385 of 423)
after the two-stage procedures. The studies differ greatly in exact
management details, which precludes statistical analysis. There
is, however, an apparent benefit of using antibiotic-loaded cement
in a one-stage arthroplasty, whereas the benefit, if any, is much
reduced when the reconstruction is two-stage (Table III).
If active infection cannot be excluded, it is prudent to refrain from
performing a primary hip replacement. Instead, a two-stage reconstruction,
not unlike a two-stage exchange arthroplasty57,58,69,77 for
the treatment of infection at the site of a hip replacement, should
be considered.
If the operation is performed in one stage, previous incisions should
be used whenever appropriate. Severe contractures may require release
of the adductor or iliopsoas tendons. Antibiotics should be withheld
until synovial specimens have been taken for culture, which should
include studies for tuberculosis and fungi, if suspected.
The original acetabulum may be filled with scar tissue, and in cases
with marked bone loss and a proximally displaced greater trochanter
the hip center may be difficult to ascertain. In this situation,
an intraoperative anteroposterior radiograph of the pelvis with
the reamer in situ can be very useful to ensure
correct placement of the cup. Following long-standing infection of
the hip, the proximal aspect of the femur may be distorted, with a
narrow medullary canal and an excessively anteverted neck, which
may cause difficulty with reaming and may increase the risk of femoral
fracture. If the previous infection led to a spontaneous fusion,
there is an increased risk of heterotopic ossification, and prophylactic measures
such as anti-inflammatory medication or radiotherapy should be considered.
Although the operation may in some cases be technically demanding,
the reported complication rates with respect to nerve injury and
dislocation (although based on only small series) do not appear
to be greatly increased27,61.
If the operation is performed in two stages, the role of antibiotic-loaded
cement at the second stage is not as clear. However, Lai et al.78 demonstrated a marked benefit of
using antibiotic-impregnated beads between stages, with recurrence
of infection in two of thirty-four hips with beads and in three
of six without beads.
In reconstruction of the hip damaged by infection, on occasion there
will be such extensive loss of bone from the acetabulum or the proximal
aspect of the femur that augmentation with an allograft may be desirable.
We are not aware of any reports on the outcome of the use of bulk allografts
in primary hip arthroplasty after infection. However, their use
in reconstruction for the treatment of infection at the site of
a total arthroplasty in a hip with bone loss has been well described
and can provide some guidance. Wang and Chen79 reported
their experience with allografts in a two-stage revision for infection
at the site of a total hip arthroplasty. At a mean of four years
(range, two to seven years), they found recurrent infection in two
of sixteen hips with morselized grafts and in zero of six hips in
which a combination of bulk and morselized grafts had been used.
Both recurrences were with virulent organisms: one was a methicillin-resistant
strain of Staphylococcus aureus, and the other
was a pseudomonas. Similarly, Berry et al.80 found
that one of thirteen hips reconstructed with bulk allograft was
complicated by deep infection at a mean of forty-two months (range,
twenty-four to ninety-seven months) after a two-stage revision.
At a mean of forty-eight months (range, twenty-four to seventy-two months),
Alexeeff et al.81 found no recurrent
infection in eleven hips reconstructed with bulk allograft after
deep infection. However, the series by Alexeeff et al.
differed in that an antibiotic-loaded cement spacer was used between
stages. It can therefore be assumed that the selective use of a
bulk allograft is not contraindicated by a history of infection
that is no longer active.
We recommend that primary hip replacement be performed in two stages
when resistant organisms caused the initial infection, the primary
treatment appears to have been inadequate, or there is a sinus or
other evidence of a recent flare-up. Similarly, a two-stage procedure
is indicated in the presence of an elevated erythrocyte sedimentation
rate or level of C-reactive protein, a positive biopsy of hip aspirate,
positive bone biopsy (on histological examination or culture), or
positive sequential technetium-indium scans, which suggest ongoing
infection. Magnetic resonance imaging may indicate active infection,
but it is not routinely used in our assessment.
Since 1986, we have used the PROSTALAC implant (prosthesis of
antibiotic-loaded acrylic cement; DePuy, Warsaw, Indiana) in staged
procedures (Figs. 2, 3-A, 3-B, and 3-C). With the use
of a system of molds, a femoral component is fashioned by encasing
a narrow-stemmed femoral component in antibiotic-impregnated cement. A
constrained acetabular component is also cemented with antibiotic-loaded
cement. As the PROSTALAC is a temporary, low-demand implant, the
concerns about high antibiotic concentrations weakening the cement
can be largely disregarded. We typically use 2.4 to 3.6 g of tobramycin
and 1.0 to 1.5 g of vancomycin per 40 g of cement powder. The actual choice
of antibiotic is dependent on the historical or preoperative cultures;
however, the concentrations used are much higher than those employed
in definitive cementation of an implant. Appropriate parenteral
antibiotics are administered for six weeks. If a repeat aspiration
of the hip four weeks after the completion of antibiotic therapy
is negative for bacterial growth and the erythrocyte sedimentation rate
and the C-reactive protein level have returned to normal, the second
stage is performed with use of hybrid or cementless fixation depending
on the age of the patient and the preference of the surgeon. Before
the PROSTALAC was available, the first stage consisted of either
a Girdlestone resection arthroplasty and placement of an antibiotic-loaded
cement depot in the acetabulum (Figs. 4-A and 4-B) or a technique, with which we have
been successful, of forming a surface hemiarthroplasty from antibiotic-loaded
cement (Figs. 5-A,
5-B, and 5-C). These earlier
techniques had the theoretical advantage of not violating the femoral
canal. However, the spread of infection simply as a result of use
of the PROSTALAC stem has not been noted, and walking is greatly improved
by use of this implant.
As we previously reported, use of this two-stage regimen
led to eradication of infection and a successful outcome after six
of six primary replacements in hips with an active pyogenic infection and
after forty-five of forty-eight revisions for infection at the site of
total hip arthroplasty in a study with a minimum duration of follow-up
of two years58.
If the history of deterioration is slowly progressive, there
are no clinical features of active infection, and the inflammatory
markers, aspiration, and nuclear medicine scans are negative, we advocate
a one-stage arthroplasty with use of antibiotic-loaded cement for
the femoral component and either an all-polyethylene acetabular
component inserted with antibiotic-loaded cement or a porous-coated
acetabular component inserted without cement.
Hardinge K; Cleary J; and Charnley J: Low-friction arthroplasty for healed septic and tuberculous arthritis. J Bone Joint Surg Br,1979.61: 144-7, 61144
1979
[PubMed]
Gallie WE: First recurrence of osteomyelitis eighty years after infection. J Bone Joint Surg Br,1951.33: 110-1, 33110
1951
[PubMed]
Pasteur L: De l’extension de la theorie des germes a l’etiologie
de quelques maladies communes. Bull Acad Med (Paris),1880.9: 435-47, 9435
1880
Craigen MA; Watters J; and Hackett JS: The changing epidemiology of osteomyelitis in children. J Bone Joint Surg Br,1992.74: 541-5, 74541
1992
[PubMed]
Gillespie WJ, and Mayo KM: The management of acute haematogenous osteomyelitis in
the antibiotic era: a study of the outcome. J Bone Joint Surg Br,1981.63: 126-31, 63126
1981
[PubMed]
Jones NS; Anderson DJ; and Stiles PJ: Osteomyelitis in a general hospital. A five-year study
showing an increase in subacute osteomyelitis. J Bone Joint Surg Br,1987.69: 779-83, 69779
1987
[PubMed]
Kelly PJ: Bacterial arthritis in the adult. Orthop Clin North Am,1975.6: 973-81, 6973
1975
[PubMed]
Morrey BF; Bianco AJ Jr; and Rhodes KH: Septic arthritis in children. Orthop Clin North Am,1975.6: 923-34, 6923
1975
[PubMed]
Farer LS; Lowell AM; and Meador MP: Extrapulmonary tuberculosis in the United States. Am J Epidemiol,1979.109: 205-17, 109205
1979
[PubMed]
Shivaram U; Wollschlager C; Khan F; and Khan A: Spinal tuberculosis revisited. South Med J,1985.78: 681-4, 78681
1985
[PubMed]
Chaisson RE, and Slutkin G: Tuberculosis and human immunodeficiency virus infection. J Infect Dis,1989.159: 96-100, 15996
1989
[PubMed]
Watson JM, and Gill ON: HIV infection and tuberculosis. BMJ.,1990.300: 63-5, 30063
1990
[PubMed]
Hecht RH; Meyers MH; Thornhill-Joynes M; and Montgomerie JZ: Reactivation of tuberculous infection following total
joint replacement. A case report. J Bone Joint Surg Am.,1983.65: 1015-6, 651015
1983
[PubMed]
Fitzgerald E; Lloyd-Still J; and Gordon SL: Candida arthritis. A case report and review of the literature. Clin Orthop,1975.106: 143-7, 106143
1975
[PubMed]
Noyes FR; McCabe JD; and Fekety FR Jr: Acute candida arthritis. Report of a case and use of amphotericin B. J Bone Joint Surg Am,1973.55: 169-76, 55169
1973
[PubMed]
Svirsky-Fein S; Langer L; Milbauer B; Khermosh O; and Rubinstein E: Neonatal osteomyelitis caused by Candida tropicalis. Report
of two cases and review of the literature. J Bone Joint Surg Am,1979.61: 455-9, 61455
1979
[PubMed]
Evans RP, and Nelson CL: Staged reimplantation of a total hip prosthesis after
infection with Candida albicans. A report of two cases. J Bone Joint Surg Am,1990.72: 1551-3, 721551
1990
[PubMed]
Samra Y; Hertz M; Shaked Y; Zwas S; and Altman G: Brucellosis of the spine. A report of three cases. J Bone Joint Surg Br,1982.64: 429-31, 64429
1982
[PubMed]
Munoz-Fernandez S; Macia MA; Pantoja L; Cardenal A; Pena JM; Martin Mola E; Balsa A; Barbado FJ; Vazquez JJ; and Gijon Banos J: Osteoarticular infection in intravenous drug abusers:
influence of HIV infection and differences with non drug abusers. Ann Rheum Dis,1993.52: 570-4, 52570
1993
[PubMed]
Engh CA; Hughes JL; Abrams RC; and Bowerman JW: Osteomyelitis in the patient with sickle-cell disease. J Bone Joint Surg Am,1971.53: 1-15, 531
1971
[PubMed]
del Sel HJ, and Charnley J: Total hip replacement following infection in the opposite
hip. Clin Orthop,1979.141: 138-42, 141138
1979
[PubMed]
Surin VV; Sundholm K; and Bäckman L: Infection after total hip replacement. With special reference
to a discharge from the wound. J Bone Joint Surg Br,1983.65: 412-8, 65412
1983
[PubMed]
Spangehl MJ; Masri BA; O’Connell JX; and Duncan CP: Prospective analysis of preoperative and intraoperative
investigations for the diagnosis of infection at the sites of two
hundred and two revision total hip arthroplasties. J Bone Joint Surg Am,1999.81: 672-83, 81672
1999
[PubMed]
Forster IW, and Crawford R: Sedimentation rate in the infected and uninfected total
hip arthroplasty. Clin Orthop,1982.168: 48-52, 16848
1982
[PubMed]
Levitsky KA; Hozack WJ; Balderston RA; Rothman RH; Gluckman SJ; Maslack MM; and Booth RE Jr: Evaluation of the painful prosthetic joint. Relative value
of bone scan, sedimentation rate and joint aspiration. J Arthroplasty,1991.6: 237-44, 6237
1991
[PubMed]
Magnuson JE; Brown ML; Hauser MF; Berquist TH; Fitzgerald RH Jr; and Klee GG: In-111-labeled leukocyte scintigraphy in suspected orthopedic prosthesis
infection: comparison with other imaging modalities. Radiology,1988.168: 235-9, 168235
1988
[PubMed]
Laforgia R; Murphy JC; and Redfern TR: Low friction arthroplasty for old quiescent infection
of the hip. J Bone Joint Surg Br,1988.70: 373-6, 70373
1988
[PubMed]
Sanzén L; Carlsson AS; Josefsson G; and Lindberg LT: Revision operations on infected total hip arthroplasties.
Two- to nine-year follow-up study. Clin Orthop,1988.229: 165-72, 229165
1988
[PubMed]
Coutts RD; Schiller AL; and Harris WH: Subclinical osteomyelitis of the femoral head. Two cases
illustrating a special problem in the use of total hip replacement. Clin Orthop,1972.86: 68-72, 8668
1972
[PubMed]
Schauwecker DS: The scintigraphic diagnosis of osteomyelitis. AJR Am J Roentgenol,1992.158: 9-18, 1589
1992
[PubMed]
Tang JS; Gold RH; Bassett LW; and Seeger LL: Musculoskeletal infection of the extremities: evaluation
with MR imaging. Radiology,1988.166: 205-9, 166205
1988
[PubMed]
Mason MD; Zlatkin MB; Esterhai JL; Dalinka MK; Velchik MG; and Kressel HY: Chronic complicated osteomyelitis of the lower extremity: evaluation
with MR imaging. Radiology,1989.173: 355-9, 173355
1989
[PubMed]
Al-Sheikh W; Sfakianakis GN; Mnaymneh W; Hourani M; Heal A; Duncan RC; Burnett A; Ashkar FS; and Serafini AN: Subacute and chronic bone infections: diagnosis using
In-111, Ga-67 and Tc-99 MDP bone scintigraphy, and radiography. Radiology,1985.155: 501-6, 155501
1985
[PubMed]
Merkel KD; Brown ML; Dewanjee MK; and Fitzgerald RH: Comparison of indium-labeled-leukocyte imaging with sequential
technetium-gallium scanning in the diagnosis of low-grade musculoskeletal
sepsis. A prospective study. J Bone Joint Surg Am,1985.67: 465-76, 67465
1985
[PubMed]
Datz FL, and Thorne DA: Effect of chronicity of infection on the sensitivity of
the In-111-labeled leukocyte scan. AJR Am J Roentgenol,1986.147: 809-12, 147809
1986
[PubMed]
Rubin RH; Fischman AJ; Needleman M; Wilkinson R; Callahan RJ; Khaw BA; Hansen WP; Kramer PB; and Strauss HW: Radiolabeled, nonspecific, polyclonal human immunoglobulin in
the detection of focal inflammation by scintigraphy: comparison
with gallium-67 citrate and technetium-99m-labeled albumin. J Nucl Med,1989.30: 385-9, 30385
1989
[PubMed]
Cherney DL, and Amstutz HC: Total hip replacement in the previously septic hip. J Bone Joint Surg Am,1983.65: 1256-65, 651256
1983
[PubMed]
Lachiewicz PF; Rogers GD; and Thomason HC: Aspiration of the hip joint before revision total hip
arthroplasty. Clinical and laboratory factors influencing attainment
of a positive culture. J Bone Joint Surg Am,1996.78: 749-54, 78749
1996
[PubMed]
Taylor RG: Pseudarthrosis of the hip joint. J Bone Joint Surg Br,1950.32: 161-5, 32161
1950
[PubMed]
Girdlestone GR; Watson-Jones R; McFarland B; Stamm TT; and Pridie KH: Discussion on treatment of unilateral osteoarthritis of
the hip-joint. Proc R Soc Med,1945.38: 363-8, 38363
1945
Haw CS, and Gray DH: Excision arthroplasty of the hip. J Bone Joint Surg Br,1976.58: 44-7, 5844
1976
[PubMed]
Bittar ES, and Petty W: Girdlestone arthroplasty for infected total hip arthroplasty. Clin Orthop,1982.170: 83-7, 17083
1982
[PubMed]
Petty W, and Goldsmith S: Resection arthroplasty following infected total hip arthroplasty. J Bone Joint Surg Am,1980.62: 889-96, 62889
1980
[PubMed]
Shepherd MM: A further review of the results of operations on the hip
joint. J Bone Joint Surg Br.,1960.42: 177-204, 42177
1960
[PubMed]
Waters RL; Perry J; Conaty P; Lunsford B; and O’Meara P: The energy cost of walking with arthritis of the hip and
knee. Clin Orthop.,1987.214: 278-84, 214278
1987
[PubMed]
Callaghan JJ; Brand RA; and Pedersen DR: Hip arthrodesis. A long-term follow-up. J Bone Joint Surg Am,1985.67: 1328-35, 671328
1985
[PubMed]
Sponseller PD; McBeath AA; and Perpich M.: Hip arthrodesis in young patients. A long-term follow-up study. J Bone Joint Surg Am,1984.66: 853-9, 66853
1984
[PubMed]
Perugia L; Santori FS; Mancini A; Manili M; and Falez F: Conversion of the arthrodesed hip to a total hip arthroplasty. Indications
and limitations. Ital J Orthop Traumatol,1992.18: 145-53, 18145
1992
[PubMed]
Amstutz HC, and Sakai DN: Total joint replacement for ankylosed hips. Indications,
technique, and preliminary results. J Bone Joint Surg Am,1975.57: 619-25, 57619
1975
[PubMed]
Kilgus DJ; Amstutz HC; Wolgin MA; and Dorey FJ: Joint replacement for ankylosed hips. J Bone Joint Surg Am,1990.72: 45-54, 7245
1990
[PubMed]
Jupiter JB; Karchmer AW; Lowell JD; and Harris WH: Total hip arthroplasty in the treatment of adult hips
with current or quiescent sepsis. J Bone Joint Surg Am,1981.63: 194-200, 63194
1981
[PubMed]
Schmalzried TP; Amstutz HC; Au MK; and Dorey FJ: Etiology of deep sepsis in total hip arthroplasty. The
significance of hematogenous and recurrent infections. Clin Orthop,1992.280: 200-7, 280200
1992
[PubMed]
Hardinge K; Williams D; Etienne A; MacKenzie D; and Charnley J: Conversion of fused hips to low friction arthroplasty. J Bone Joint Surg Br,1977.59: 385-92, 59385
1977
[PubMed]
Went P; Krismer M; and Frischhut B: Recurrence of infection after revision of infected hip
arthroplasties. J Bone Joint Surg Br,1995.77: 307-9, 77307
1995
[PubMed]
Carlsson AS; Josefsson G; and Lindberg L: Revision with gentamicin-impregnated cement for deep infections
in total hip arthroplasties. J Bone Joint Surg Am,1978.60: 1059-64, 601059
1978
[PubMed]
McDonald DJ; Fitzgerald RH Jr; and Ilstrup DM: Two-stage reconstruction of a total hip arthroplasty because
of infection. J Bone Joint Surg Am,1989.71: 828-34, 71828
1989
[PubMed]
Younger AS; Duncan CP; and Masri BA: Treatment of infection associated with segmental bone
loss in the proximal part of the femur in two stages with use of
an antibiotic-loaded interval prosthesis. J Bone Joint Surg Am,1998.80: 60-9, 8060
1998
[PubMed]
Younger AS; Duncan CP; Masri BA; and McGraw RW: The outcome of two-stage arthroplasty using a custom-made
interval spacer to treat the infected hip. J Arthroplasty,1997.12: 615-23, 12615
1997
[PubMed]
Charnley J: The long-term results of low-friction arthroplasty of
the hip performed as a primary intervention. J Bone Joint Surg Br,1972.54: 61-76, 5461
1972
[PubMed]
Harris WH: Traumatic arthritis of the hip after dislocation and acetabular fractures:
treatment by mold arthroplasty. An end-result study using a new
method of result evaluation. J Bone Joint Surg Am,1969.51: 737-55, 51737
1969
[PubMed]
Eskola A; Santavirta S; Konttinen YT; Tallroth K; Hoikka V; and Lindholm ST: Cementless total replacement for old tuberculosis of the
hip. J Bone Joint Surg Br,1988.70: 603-6, 70603
1988
[PubMed]
Johnson R; Barnes KL; and Owen R: Reactivation of tuberculosis after total hip replacement. J Bone Joint Surg Br,1979.61: 148-50, 61148
1979
[PubMed]
Kim YY; Ko CU; Lee SW; and Kwak BM: Replacement arthroplasty using the Charnley prosthesis
in old tuberculosis of the hip. Int Orthop,1979.3: 81-8, 381
1979
[PubMed]
Harris WH, and Aufranc OE: Mold arthroplasty in the treatment of hip fractures complicated
by sepsis. A report of nine cases. J Bone Joint Surg Am,1965.47: 31-42, 4731
1965
[PubMed]
Hughes PW; Salvati EA; Wilson PD Jr; and Blumenfeld EL: Treatment of subacute sepsis of the hip by antibiotics
and joint replacement. Criteria for diagnosis with evaluation of
twenty-six cases. Clin Orthop,1979.141: 143-57, 141143
1979
[PubMed]
Tsukayama DT; Estrada R; and Gustilo RB: Infection after total hip arthroplasty. A study of the
treatment of one hundred and six infections. J Bone Joint Surg Am,1996.78: 512-23, 78512
1996
[PubMed]
Wilson PD Jr; Aglietti P; and Salvati EA: Subacute sepsis of the hip treated by antibiotics and
cemented prosthesis. J Bone Joint Surg Am,1974.56: 879-98, 56879
1974
[PubMed]
Buchholz HW; Elson RA; Engelbrecht E; Lodenkämper H; Röttger J; and Siegel A: Management of deep infection of total hip replacement. J Bone Joint Surg Br,1981.63: 342-53, 63342
1981
[PubMed]
Garvin KL, and Hanssen AD: Current concepts review. Infection after total hip arthroplasty. Past,
present, and future. J Bone Joint Surg Am,1995.77: 1576-88, 771576
1995
[PubMed]
Hope PG; Kristinsson KG; Norman P; and Elson RA: Deep infection of cemented total hip arthroplasties caused
by coagulase-negative staphylococci. J Bone Joint Surg Br,1989.71: 852-5, 71852
1989
Josefsson G; Lindberg L; and Wiklander B: Systemic antibiotics and gentamicin-containing bone cement
in the prophylaxis of postoperative infections in total hip arthroplasty. Clin Orthop,1981.159: 194-200, 159194
1981
[PubMed]
Josefsson G; Gudmundsson G; Kolmert L; and Wijkström S: Prophylaxis with systemic antibiotics versus gentamicin
bone cement in total hip arthroplasty. A five-year survey of 1688
hips. Clin Orthop,1990.253: 173-8, 253173
1990
[PubMed]
Marra F, Robbins GM, Duncan CP, Masri
BA, Wasan KM, Kwong EH, Jewesson PJ. Amphotericin B-loaded
bone cement for the treatment of Candida albicans osteomyelitis. Unpublished
data
Masri BA; Duncan CP; and Beauchamp CP: Long-term elution of antibiotics from bone-cement: an
in vivo study using the prosthesis of antibiotic-loaded acrylic
cement (PROSTALAC) system. J Arthroplasty,1998.13: 331-8, 13331
1998
[PubMed]
Masri BA; Duncan CP; Jewesson P; Ngui-Yen J; and Smith J: Streptomycin-loaded bone cement in the treatment of tuberculous
osteomyelitis: an adjunct to conventional therapy. Can J Surg,1995.38: 64-8, 3864
1995
[PubMed]
Raut VV; Siney PD; and Wroblewski BM: One-stage revision of infected total hip replacements
with discharging sinuses. J Bone Joint Surg Br,1994.76: 721-4, 76721
1994
[PubMed]
Nestor BJ; Hanssen AD; Ferrer-Gonzalez R; and Fitzgerald RH Jr: The use of porous prostheses in delayed reconstruction
of total hip replacements that have failed because of infection. J Bone Joint Surg Am,1994.76: 349-59, 76349
1994
[PubMed]
Lai KA; Shen WJ; Yang CY; Lin RM; Lin CJ; and Jou IM: Two-stage cementless revision THR after infection. 5 recurrences
in 40 cases followed 2.5-7 years. Acta Orthop Scand,1996.67: 325-8, 67325
1996
[PubMed]
Wang JW, and Chen CE: Reimplantation of infected hip arthroplasties using bone allografts. Clin Orthop,1997.335: 202-10, 335202
1997
[PubMed]
Berry DJ; Chandler HP; and Reilly DT: The use of bone allograft in two-stage reconstruction
after failure of hip replacements due to infection. J Bone Joint Surg Am,1991.73: 1460-8, 731460
1991
[PubMed]
Alexeeff M; Mahomed N; Morsi E; Garbuz D; and Gross A: Structural allograft in two-stage revisions for failed
septic hip arthroplasty. J Bone Joint Surg Br,1996.78: 213-6, 78213
1996
[PubMed]