Synovial Fluid Microanalysis Allows Early Diagnosis of Ceramic Hip
Prosthesis Damage
Susanna Stea,1 Francesco Traina,2 Alina Beraudi,1 Monica Montesi,1 Barbara Bordini,1 Stefano Squarzoni,3
Alessandra Sudanese,2 Aldo Toni1,2
1
Medical Technology Lab, Istituto Ortopedico Rizzoli, Bologna, Italy, 2Dept of Orthopedic-Traumatology and Prosthetic Surgery, Istituto
Ortopedico Rizzoli, Bologna, Italy, 3CNR, National Research Council of Italy, Institute of Molecular Genetics, Bologna, Italy
Received 29 April 2011; accepted 3 January 2012
Published online 27 January 2012 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jor.22077
ABSTRACT: The first clinical sign of ceramic hip prosthesis failure is hip noise. We therefore investigated whether isolation, observation at scanning electron microscopy, and chemical identification with microanalysis of particles from synovial fluid of ‘‘noisy hip’’ could
be predictive of ceramic damage. Firstly, the level of ‘‘physiological wear’’ of well functioning ceramic-on-ceramic hip prostheses was
assessed with this method, then the test was validated as diagnostic method for liner fracture. Twelve asymptomatic patients were
enrolled to demonstrate the first aim; 39 cases of noisy hip (GROUP 1), and 7 cases of pending failure not related to ceramic (GROUP
2) were enrolled for the second aim. The analysis of the synovial fluid of the 12 asymptomatic patients allowed to set the ‘‘physiological
wear’’ threshold. The analysis of GROUP 1 hips demonstrated the presence of ceramic particles (2 physiological, 12 mild, and 25
strong). The analysis of GROUP 2 showed a physiological presence of ceramic particles in all cases. Revision surgery in GROUP 1 was
performed in 16 hips out 25 with strong ceramic particle presence and 2 out of 12 with mild ceramic particle presence. Failure of the
ceramic component was evident in all but one of these cases, while the integrity of components was demonstrated in all seven hips
of GROUP 2. Synovial fluid microanalysis can be a useful surrogate in predicting ceramic failure particularly when a strong presence
of ceramic particles is observed. On the contrary there is not enough evidence to predict ceramic failure in presence of mild positivity.
ß 2012 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 30:1312–1320, 2012
Keywords: hip prosthesis; ceramic components; synovial fluid; failure; microanalysis
Ceramic on ceramic hip prosthesis has shown a good
long-term resistance to wear,1 however, this advantage
is counterbalanced by the risk of ceramic fracture,
which could lead to a catastrophic failure of the
implant if it is not early recognized.2 In case of late
diagnosis, material deterioration occurs and massive
destruction of bone and tissues may be observed, mostly
caused by the metallic debris originating from the
abrasion of alumina particles on the metallic stem,
neck, or cup metal back.3
While the clinical and radiological signs of ceramic
head fractures are clear, as the ceramic epiphysis typically breaks in big fragments, and the patients usually
complain of persistent and severe hip pain,4 on the
other hand a ceramic liner fracture could easily go undetected.5 For this reason, ceramic liner fractures are
usually underestimated and seldom reported. Thus, while
the incidence of ceramic head fractures is reported
with a reasonable variance ranging from 0.004% to
0.05%,1,6 the incidence of ceramic liners fracture varies
from 0.013% up to 1.1%.1,7 Conventional radiographs do
not allow an early diagnosis as only macroscopic damage
can be detected. Computed axial tomography (CT scan),
although useful to detect implant malpositioning and
ceramic fragments in the articular space, cannot show
small ceramic particles and initial damage of the liner
since the metal back shadows the liner itself. Thus CT
scan is not an effective tool to make an early diagnosis
of ceramic fracture.
Correspondence to: Susanna Stea (T: þ39 (0)51 6366861; F: þ39
(0)51 6366863; E-mail: [email protected])
ß 2012 Orthopaedic Research Society. Published by Wiley Periodicals, Inc.
1312
JOURNAL OF ORTHOPAEDIC RESEARCH AUGUST 2012
Early recognition of clinical signs of ceramic liner
damage is essential to avoid the wide spread of ceramic particles in the periarticular space. Probably the
first clinical sign that the patient himself reports to
the clinician is hip noise.8 Patients can report it as
clicking, squeaking, or grinding. However, not all
noisy ceramic hips are broken;9–12 in fact the noise
often results from the damage of the lubrication fluid
film at the interface and can be due to edge load, impingement, metal transfer onto the bearing, alteration
of the lubricating fluid properties, debris interposition
bearing, or finally to surface damage.13 Synovial fluid
analysis has been previously proposed as diagnostic
test for the early diagnosis of ceramic fracture,8 but
the number of patients enrolled in the study was too
small to statistically evaluate the effectiveness of
needle aspiration and synovial fluid analysis as a diagnostic test for ceramic hip fracture.8 Moreover there is
no published data on the quantification of number of
ceramic wear particles in the synovial fluid of well
functioning ceramic-on-ceramic hip prosthesis and this
datum is essential for set the level of ‘‘pathological’’ wear.
Thus, the aims of the present study are: firstly to
define the level of ‘‘physiological wear’’ in well functioning ceramic-on-ceramic hip prosthesis and secondly is to
validate synovial fluid microanalysis as diagnostic test
for early recognition of ceramic liner fracture.
PATIENTS AND METHODS
After the Institutional Review Board approval was obtained,
and all patients gave their written consent, hip needle aspirate and synovial fluid analysis were performed according to
a published guideline.8
EARLY DIAGNOSIS OF HIP PROSTHESIS DAMAGE
Twelve patients with well-functioning ceramic-on-ceramic
implants were enrolled to define the amount of particles that
can be ‘‘physiologically’’ isolated in the synovial fluid. They
had no pain, correct range of motion, the implants were
correctly implanted and osteointegrated.
From January 2004 to March 2009, 38 consecutive
patients with a ceramic hip prosthesis who referred to our
department complaining of noise and discomfort (39 hips)
were enrolled in the diagnostic protocol (GROUP 1).
In addition, seven ‘‘non-noisy’’ patients, who were scheduled for a revision of their ceramic hip prosthesis for causes
not related to the ceramic coupling, were enrolled as control
group (GROUP 2). All patients, regardless of the group, had
a follow-up of at least 24 months.
Details of the types of prosthesis implanted in the
patients are listed in Table 1, while patients demographics of
the three groups are listed in Tables 2–4.
C-arm fluoroscopy or ultrasound-guided synovial fluid
aspiration was performed in regional anaesthesia and in
aseptic conditions. All patients had a short-term antibiotic
prophylaxis with a second generation cephalosporin.
Particles were isolated in synovial fluid by hypochlorite
digestion. Then they were measured and chemically identified by scanning electron microscopy (SEM) and microanalysis following a published procedure.8 Briefly, 200 ml of fluid
were dropped onto a polycarbonate filter (25 mm diameter)
with a pore size of 0.2 mm (Millipore, Isopore TM, Ireland).
Synovial fluid proteins were digested by at least three subsequent additions of sodium hypochlorite (Sigma–Aldrich, St
Luis, MO). Filters were then washed out with distilled water
and air dried. Both hypochlorite and water were previously
0.2 mm filtered. Filters were then mounted on SEM holders
by bi-adhesive tape, gold sputtered, and examined in a
Cambridge Stereoscan 200 electron microscope operated at
20 kV. Micrographs were taken at 10,000X magnification.
Quantification was defined by 20 fields capturing 90 m2 each,
representing all the regions of the filter. The particles were
counted, measured in their major diameter and their chemical composition was verified by Energy dispersive X-ray
Spectroscopy; analysis was performed at 25 mm WD with an
Oxford INCA Energy 200 apparatus. A CT scan and a two
projection radiographs of the affected hip were also performed during the same admission.
Results of synovial fluid analysis were matched with the
clinical–radiological evaluation of each case, thus some
patients having a high probability of ceramic damage underwent revision surgery, while the others were scheduled for a
closer clinical follow-up.
Statistical Analysis
Threshold of physiological wear in ceramic-on-ceramic
couplings was set on the basis of results observed in the 12
well functioning patients.
The sensitivity and specificity of synovial fluid analysis as
diagnostic test of ceramic fracture were calculated from all
revised patients with mild or strong presence of ceramic particles in GROUP 1 (16 out of 25 who had a strong positive
result and 2 out of 12 who had mild), and on all hips of
GROUP 2. Sensitivity is the proportion of actual positive
cases correctly identified, and it is estimated from the
number of positive synovial fluid ceramic tests that correctly
predict prosthesis damage. Specificity is the proportion of
actual negative cases correctly identified, and it is estimated
from the number of negative synovial fluid ceramic tests that
1313
correctly predict no prosthesis damage. Confidence interval
at 95% is calculated with score method.9
RESULTS
Definition of Physiological Levels of Wear
In 6 out of 12 asymptomatic patients no ceramic
particles were detected in the 20 observed fields, in 3
patients 2 particles each were observed, in 2 patients
3 particles each were detected, and in the last patient
5 particles were observed. All the particles were less
than 3 m in size.
The amount and the dimension of the particles observed in these patients allowed us to set the threshold
of ‘‘physiological amount of ceramic particles,’’ consisting in a maximum of 5 particles, with less than 3 m as
maximum size, observed in 20 fields.
Validation of the Method
GROUP 1 (noisy hips) and 2 (non-noisy hips) patients
were then observed and classified. Three levels of positivity were defined: physiological, mild, and strong
(Table 5).
All patients with noisy ceramic hips (GROUP 1),
resulted positive: 12 had a mild and 25 a strong presence of ceramic particles in the synovial fluid. In the
remaining two cases a physiological presence of ceramic particles was observed in addition to the presence of
metal particles, deriving from the titanium implant
components (Table 6). All patients in GROUP 2 resulted
to have a ‘‘physiological damage.’’
After the clinical and radiological evaluation, in 16
out of 25 hips (64%) where a strong presence of ceramic
particles was observed, a revision surgery was performed. Reason for revision was the noise and the discomfort at their hip due to the ceramic failure, septic
and aseptic loosening of the THA was excluded on clinical and radiographic evaluation and by bone scan and
leuko-scan. The nine remaining patients underwent a
closer clinical follow up due to unwillingness to be
treated or to comorbidity.
In all the 16 cases in which needle aspirate had
shown a strong presence of ceramic particles and a revision surgery was performed, a ceramic component
presented macroscopic damage. In 11 cases there was
a fracture of the rim of the liner, in 4 cases a crescent
lesion of the head and in 1 case a fracture of the dome
of the liner.
In the group of 12 hips with mild presence of ceramic particles in the synovial fluid, only 2 (16%) were
revised because at the clinical–radiological evaluation
there was a strong suspect of ceramic failure, the
remaining 10 (84%) underwent a closer follow up. In
the two cases of revision surgery, a mild damage of the
rim of the liner was found in one case and a mild head
wear was found in the other. In that case we observed
stripe wear without alteration of head sphericity
visible at naked eye, so it has considered negative for
ceramic failure.
JOURNAL OF ORTHOPAEDIC RESEARCH AUGUST 2012
JOURNAL OF ORTHOPAEDIC RESEARCH AUGUST 2012
Wagner Cone Stem-Conus
(Centerpulse Orthopedics,
Winterthur, Switzerland,
now Zimmer, USA)
PROFEMUR1 Z (Wright
Medical Technology,
Arlington, TN, USA)
EP-FIT PLUS and SL-PLUS
stem (Smith&Nephew,
Andover, MA, USA)
APTA-FIXA (Ala Ortho, Milano,
Italy)
An.C.A. (Cremascoli, Milano,
Italy)
Anca Fit (Wright Medical
Technology, Arlington, TN)
Prosthesis
Design characteristics
Stem: Dual tapered, titanium-alloy with a rectangular cross-section, grit-blasted across the entire
implant to create a uniform 6 m surface coating.
Cup: Titanium alloy modular component, Triple
Radius Profile, additional circumferential ribs,
coated with high-grade open-pored titanium
plasma sprayed and HA coated.
Stem: Protasul1-64 titanium alloy 58 tapered stem
with a circular cross-section, with 8 longitudinal
ribs and corundum blasted surface.
Stem: Modular neck, titanium alloy (Ti-6A1-4V)
anatomically shaped, grit blasted, and coated with
HA coating in the proximal third.
Cup: Titanium alloy (Ti-6A1-4V) modular
component, having a titanium porous coating and
two peripheral fins.
Stem: Chrome-cobalt alloy anatomically shaped,
porous-coated in the proximal part, and fully
coated with HA.
Cup: Dense alumina (Biolox1), with a threaded
titanium alloy (Ti-6A1-4V) ring, and with a 3-D
porous alumina beads coating (Poral1) at the
dome.
Stem: Modular neck, titanium alloy (Ti-6A-14V)
anatomically shaped, porous coated in the
proximal half and fully coated with HA
Cup: Titanium alloy (Ti-6A1-4V) modular
component, having a titanium porous coating.
Stem: Rectangular, dual-taper straight stem, forged
titanium alloy, grit blasted (roughness 4–6 m).
Table 1. Details of the Prostheses Implanted
Cup: Press fit after line to line reaming, secondary
fixation by in-growth on the porous coating
Stem: Press-fit for initial fixation with a cortical
multicontact anchorage, ongrowth for secondary
fixation.
Cup: Primary stability with full contact of the rim of
the acetabulum with a gradient until the pole,
secondary fixation by in-growth on the porous
coating.
Stem: Press-fit for initial fixation, the longitudinal
ribs intended for engaging the femoral cortex,
allowing rotational stability and bone on-growth
for secondary stability.
Stem: Press-fit for initial fixation, the tapered,
rectangular cross-section provides proximal
fixation and rotational stability, and bone
on-growth for secondary stability.
Stem: Press-fit for initial fixation, proximal porus
coating for methaphyseal secondary fixation.
Cup: Primary stability by screwing in the cup,
secondary stability by ongrowth on Poral1 coating.
Cup: Press fit after line to line reaming, rotational
stability through fins, secondary fixation by
in-growth on the porous coating.
Stem: Press-fit for initial fixation, HA, and porous
coating for secondary fixation.
Stem: Press-fit for initial fixation, proximal HA
coating for methaphyseal secondary fixation
Fixation
1314
STEA ET AL.
JOURNAL OF ORTHOPAEDIC RESEARCH AUGUST 2012
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
M
F
F
M
M
F
M
F
F
F
M
F
F
M
F
F
F
F
M
M
57
53
60
62
60
54
65
62
36
47
53
41
31
58
39
52
67
65
35
68
56
47
41
41
68
53
39
44
41
45
68
67
67
30
44
67
55
45
79
Arthritis
DDH
Arthritis
DDH
Arthritis
DDH
Arthritis
Arthritis
DDH
Arthritis
Arthritis
DDH
Post traum arthritis
Arthritis
Cup revision
Arthritis
Arthritis
Arthritis
Arthritis
Arthritis
Arthritis
DDH
Arthritis
Arthritis
Arthritis
Arthritis
Necrosis
Arthritis
DDH
Revision due to dislocation
Arthritis
Arthritis
Arthritis
Arthritis
Arthritis
DDH
DDH
Arthritis
Arthritis
Head diameter
(Biolox Forte)
AnCA FIT/Dual Fit
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/Dual Fit
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/Profemur Z
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/Conus
28 mm
AnCA FIT/Dual Fit
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/AnCA FIT
28 mm
Fixa/Apta
36 mm
Epfit/SL
28 mm
AnCA FIT/Profemur Z
28 mm
Epfit/SL
36 mm
Fixa/Apta
32 mm
AnCA FIT/AnCA FIT
28 mm
Fixa/Apta
36 mm
AnCA FIT/AnCA FIT
28 mm
Fixa/Apta
32 mm
Lineage/AnCA FIT
28 mm
Tipor/SL plus
36 mm Biolox DELTA
AnCA FIT/AnCA
28 mm
AnCA FIT/AnCA
28 mm
Fixa/Apta
36 mm
Tipor/Apta
32mm Biolox DELTA
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/AnCA FIT
28 mm
AnCA FIT/AnCA FIT
28 mm
Fixa/Apta
36 mm
AnCA FIT/AnCA FIT
28 mm
Implants (cup/stem)
Samples 31 and 32 are from the same patient (bilateral). DDH, developmental dysplasia of the hip.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Preoperative
diagnosis
Demographics of GROUP 1 Patients, Reporting Hip Noise
Age at
No. Gender surgery
Table 2.
4.4
4.6
1.1
3.8
6.9
1.3
5
2.5
2.7
1.7
1.9
4.2
2
2.8
3.3
8
1.3
6.4
6.9
1.4
1.8
3
1.5
2.1
4.2
0.1
6.3
2.1
4
0.4
10.5
11.2
1.4
0.3
6
6.5
5.1
3.1
4.6
3
2
21
2
4
104
16
2
5
2
5
1
10
2
4
2
45
4
6
1
2
4
7
2
6
7
6
1
13
3
8
10
2
4
27
7
24
1
4
Time between
onset of noise
Onset of
noise from and synovial fluid
THR (years) analysis (months)
Squeaking/no pain
Grinding/pain
Grinding/pain
Clicking/no pain
Grinding/pain
Grinding/pain
Clicking/no pain
Clicking/no pain
Clicking/no pain
Squeaking/no pain
Clicking/no pain
Clicking/no pain
Grinding/pain
Grinding/pain
Grinding/dislocation
Grinding/little pain
Grinding/no pain
Clicking and ginding/pain
Grinding/no pain
Squeaking/no pain
Squeaking/no pain
Clicking/no pain
Grinding/no pain
Squeaking/no pain
Clicking/no pain
Squeaking and clicking/no pain
Grinding/pain
Squeaking/no pain
Grinding/pain
Grinding/pain
Grinding/no pain
Grinding/no pain
Squeaking/no pain
Clicking/no pain
Clicking/no pain
Clicking/no pain
Grinding/no pain
Squeaking/no pain
Clicking/no pain
Symptoms
EARLY DIAGNOSIS OF HIP PROSTHESIS DAMAGE
1315
1316
STEA ET AL.
Table 3. Demographics of GROUP 2 Patients, ‘‘Non-Noisy’’ Patients
Preoperative
No. Gender Age
diagnosis
Time between
Head diam. surg. and collection
of fluid (years)
(Biolox Forte)
Implants (cup/stem)
40
M
33
DDH
Delta PF/C2 Lima
41
M
45
AnCA FIT/AnCA FIT
42
44
45
46
47
M
M
F
F
F
70
75
78
68
70
Post traum.
arthritis
Arthritis
Arthritis
Arthritis
Arthritis
Arthritis
36 mm Biolox
delta
28 mm
AnCA FIT/Dual FIT
Fixa/Apta
AnCA FIT/AnCA FIT
Delta PF Lima/CPF Link
AnCA FIT/AnCA FIT
28
32
28
36
28
mm
mm
mm
mm
mm
Symptoms
2.5
Stem aseptic loosening
8
Stem aseptic loosening
10
3.5
3.3
2
6.5
Stem aseptic loosening
Septic loosening
Pain without loosening
Cup aseptic loosening
Stem aseptic loosening
Table 4. Demographics of the Asymptomatic Patients with a Well Functioning Ceramic-on-Ceramic Prosthesis
No.
Gender
Age
48
49
50
M
F
M
59
53
50
51
52
53
54
55
56
57
58
59
M
F
F
F
F
F
F
F
F
52
40
60
55
50
51
58
68
57
Preoperative
diagnosis
Implants
(cup/stem)
Head diameter
(Biolox Forte)
Time between
surgery and
collection of synovial
fluid (yrs)
Arthritis
DDH
Sequelae
epiphysiolysis
Arthritis
DDH
Arthritis
DDH
DDH
Arthritis
Arthritis
DDH
DDH
AnCA FIT/AnCA FIT
AnCA FIT/AnCA FIT
AnCA FIT/AnCA FIT
28 mm
28 mm
28 mm
1
6.5
2
AnCA FIT/AnCA FIT
AnCA FIT/Conus
AnCA FIT/Dual Fit
Fixa/Apta
AnCA FIT/Conus
Fixa/Apta,
AnCA FIT/AnCA FIT
Fixa/Apta,
Fixa/Apta
28
28
28
32
28
28
28
32
32
2
1.6
7
0.9
1
1
6
1
2
mm
mm
mm
mm
mm
mm
mm
mm
mm
DDH, developmental dysplasia of the hip.
The ceramic components of the seven patients of
GROUP 2, who underwent revision surgery for causes
not related to the coupling, were not damaged
(Table 7).
During all revision surgeries the synovial fluid was
harvested for laboratory analysis to exclude secondary
infection, and none was positive; furthermore no hip
became infected as a result of the hip aspirate.
Table 5. Levels of Ceramic Damage at Synovial Fluid
Analysis (200 ml)
Level
Number of particles
1. Physiological
0–5 ceramic particles
2. Mild
6–10 ceramic particles
3. Strong
At least 11 ceramic particles
At least 1 ceramic particle
JOURNAL OF ORTHOPAEDIC RESEARCH AUGUST 2012
Dimensions
<3
<3
<3
>3
m
m
m
m
In summary, considering the 18 patients who underwent revision surgery, having a strong or mild
presence of ceramic particles (GROUP 1), and the 7
revised patients without symptoms of ceramic fractures (GROUP 2), the statistical analysis showed that
synovial fluid microanalysis had a 100% sensitivity
(95% CI 83–100) and 88% specificity (95% CI 66–97) in
predicting ceramic liner fracture. Sensitivity was
determined on the basis of 17 positive synovial fluids
out of 17 macroscopic damaged component at revision;
specificity was determined on the basis of 7 negative
synovial fluids out of 8 undamaged component at
revision.
DISCUSSION
The present study had two aims: firstly to set the
threshold of ‘‘physiological’’ wear in ceramic-on-ceramic
hip prosthesis, secondly to validate synovial fluid analysis
Result for
cer wear
Strong
Strong
Strong
Strong
Strong
Strong
Mild
Strong
Mild
Strong
Mild
Strong
Strong
Mild
Strong
Strong
Strong
Strong
Mild
Mild
Strong
Mild
Strong
Strong
Strong
Strong
Strong
Neg (pos metal)
Strong
Mild
Strong
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Uncertain
Positive
Negative
Uncertain
Positive
Positive
Positive
Positive
Negative
Positive
Positive
Positive
Positive
Negative
Positive
Positive
Positive
No CT
Positive
Uncertain
Positive
Positive
Positive
Positive
Negative
Uncertain
Positive
Uncertain
Uncertain
Positive
Yes
No
No
Yes
Yes
Yes
No
Yes
No
Yes
No
Yes
Yes
Yes
No
No
No
Yes
Yes
No
No
No
Yes
Yes
No
Yes
Yes
No
No
Yes
Yes
Impingment
Revision
(evidence from CT) surgery
Mild wear of head
Mild damage of the head.
Head and modular neck
Head and modular neck
(liner not revised even if
slightly damaged)
Noise worsened (34)
Unchanged: Still noising for 15 months,
then slightly improved
Recurrence of noise 6 months after
revision surgery
Unchanged: Still noising (26)
No recurrence of noise post-surg. (40)
No recurrence of noise post-surg. (30)
No recurrence of noise post-surg. (28)
Unchanged: Still noising (32)
Unchanged: Still noising (19)
Unchanged: Still noising (19)
No recurrence of noise post-surg. (20)
Improved: No noise (74)
No recurrence of noise post-surg. (58)
No recurrence of noise post-surg. (63)
Unchanged: Still noising (63)
No recurrence of noise post-surg. (50)
No recurrence of noise post-surg. (65)
Improved: reduced noise (57)
Unchanged: Still noising (33)
No recurrence of noise post-surg. (60)
Recurrence of noise 36 months after revision
surgery. Then second revision (cup)
Clinical evaluation at follow-up
(months from wear analysis or revision)
(Continued)
Improved significantly (18)
No recurrence of noise post-surg. (16)
Unchanged: Still noising (18)
Head and modular neck
Damage due to subluxation. Recurrence of noise 12 months after
revision surgery
Liner, head and modular neck
Liner fracture
Improved but not completely post surgery,
then lost to follow-up
Liner, head and modular neck
Damage of the head
No recurrence of noise post-surg. (12)
due to subluxation.
Liner, head and modular neck
Modular neck fracture
No recurrence of noise post-surg. (12)
Unchanged: Still noising (18)
Unchanged: Still noising (12)
Liner and head
Liner chipping
No recurrence of noise post-surg. (12)
Liner chipping
Severe wear of head and
bottom of liner
Cup, liner and head
Liner and head
Liner chipping
Mild damage of the head.
Head fracture
Cup, liner and head
Liner, head and modular neck
Liner and head
Liner chipping
Liner chipping
Liner chipping
Liner, head and modular neck
Liner and head
Liner and head
Liner chipping
Liner chipping
Appearance of the
revised component
Liner, head and modular neck
Liner, head and modular neck
Revised component
Table 6. Results Synovial Fluid analysis, Cup Malposition at CT (Possible Impingement), and follow-up of Patients
EARLY DIAGNOSIS OF HIP PROSTHESIS DAMAGE
1317
JOURNAL OF ORTHOPAEDIC RESEARCH AUGUST 2012
Mild
In case of revision surgery revised components and their macroscopic appearance are listed.
Unchanged: Still noising (6)
No
Liner chipping
Liner, head and
modular neck
39
Positive
Liner chipping
Liner and head
Yes
No
No
No
No
No
Yes
Strong
Mild
Neg (pos. metal)
Strong
Mild
Strong
Mild
32
33
34
35
36
37
38
Positive
Negative
Positive
Positive
Positive
Positive
Negative
Appearance of the
revised component
Revised component
Impingment
Revision
(evidence from CT) surgery
Result for
cer wear
No.
Table 6. (Continued)
No recurrence of noise post-surg. (12)
Slightly worstened Still noising (11)
Improved (16)
Unchanged: Still noising (10)
Unchanged: Still noising (9)
Unchanged: Still noising (8)
No recurrence of noise post-surg. (8)
STEA ET AL.
Clinical evaluation at follow-up
(months from wear analysis or revision)
1318
JOURNAL OF ORTHOPAEDIC RESEARCH AUGUST 2012
as diagnostic test for early recognition of ceramic liner
fracture.
Physiological wear is usually determined on retrieved samples, or, possibly, through X-rays analysis
of the component, even if this method is not accurate,
particularly for hard bearings. The evaluation of the
number of wear particles in synovial fluid is not a common procedure; therefore, it was necessary to validate
the method.
A published guideline was used to recruit patients
eligible for synovial fluid analysis and possibly for
revision surgery. In all 39 cases in which the needle
aspirate was performed, particles of ceramic or metal
were detected in the synovial fluid (in 37 cases ceramic
particles were found, in 2 cases metal particles were
found). We matched the results of the needle aspirate
with clinical symptoms and CT scans in 18 patients
that underwent a revision surgery: 16 of them had
strong presence of ceramic particles at synovial fluid
examination, and at revision surgery they all presented a damage of the ceramic components. In two
cases with mild presence of ceramic particles at synovial fluid examination, one fracture of the liner rim
was found at revision surgery in one case, while in the
other only mild wear of the head was observed. These
results show a high predictive value of needle aspirate
for an early diagnosis of ceramic failure when a strong
ceramic presence was observed. Conversely when a
mild presence of ceramic particles was detected in the
synovial fluid, a revision surgery could be questionable. In our experience in one case out of two, no important fracture of ceramic component was found.
A possible complication of needle aspirate could be
secondary infection of the total hip arthroplasty. In
our study we performed the procedure in asepsis and
avoided cases of infections, or local comorbidities such
as hematoma or neural damage.
Needle aspirate could be an important test to perform in case of noisy ceramic hips, in which patients
complain of hip pain and persistent hip noise. In the
literature there is not a consensus to perform a revision surgery in case of noisy hips. Revision rates
reported for squeaking hips varies from 0% to 4%;10,13
furthermore, in revised hips for squeaking, a higher
ceramic wear rate has been found, but without signs of
ceramic failure.14 Whether this justify a revision surgery is not clear, nor it is clear if these ceramic hips
are more subject to catastrophic failure of the ceramic
components. On the other hand, a ceramic fracture
even if of limited dimensions could be a trigger for
ceramic catastrophic failure.
In the two cases in which there was a physiological
presence of ceramic particles in the aspirate, but a
high presence of metallic debris, a modular neck failure occurred in one case and a stem-cup impingement
was highlighted by CT scan evaluation in the other
case. This would confirm that hip noise not necessarily
means ceramic fracture, even if a ceramic fracture
always cause hip noise.
EARLY DIAGNOSIS OF HIP PROSTHESIS DAMAGE
1319
Table 7. Summary of Relationship between Results of Debris Analysis and Outcome of Patients
Revision surgery
Macroscopic analysis
of retrievals
GROUP 1 (noisy)
2 Physiological
1 (strong metal wear) Modular neck fracture
ceramic wear
(but presence of metal)
12 Mild damage
1 fracture of the liner rim
2
1 no ceramic fracture, only
mild wear of head
25 Strong damage
16
11 fracture of the liner rim
4 crescent lesion of the ceramic
head, 1 fracture of the
dome of the liner
No revision
surgery
Follow-up
(12–62 months)
1 (mild metal wear)
1 improved
10
2 improved
2 worstened
9
6 unchanged
1 improved
0 worstened
8 unchanged
GROUP 2 (non-noisy)
7 Physiological
7
No ceramic damage
Treatment of ceramic fracture is troublesome and
must be adequately performed to avoid additional
problems. To avoid early failure of the revision surgery, it is best to retrieve all component fragments, to
meticulously clean the joint from ceramic debris with
an extensive synovectomy, and to replace the damaged
material with a new ceramic bearing specifically
designed for revision, if possible.3 However, sometimes
a revision surgery of the cup and occasionally of the
stem is required. On the other hand, in case of early
recognition of a ceramic failure, where ceramic fragments are few and metal component are intact, the revision of the bearings is an easier and straightforward
procedure.
While ceramic head fracture is usually catastrophic
and requires urgent revision,15 ceramic liner fracture
is often a consequence of repeated microtraumas
that lead to chipping of the rim of the liner and then
to a complete fracture. In these cases, the revision
surgery could be performed before a wide spread of
ceramic fragments in the articular space occurs,
leaving the surgeons free to choose any possible bearing option without risks of early failure for third body
particles.16–18
In addition, it must be considered that in our
patients, the clinical symptoms of eight out of nine
non-operated patients with strong positive result were
unchanged at an average of 12 months follow up; only
in one case there was complete resolution of the symptomatology. No clinical worsening or sudden failure of
the ceramic components were observed. Also in six out
ten patients with a mild presence of ceramic particles
the symptoms unchanged, in two cases the clinical
symptoms worsen and in two there was complete resolution (Table 6). Thus the decision to go for a revision
surgical procedure should be taken not only on the
basis of needle aspirate positivity, but matching the
test with the clinical symptoms of each patient.
A limitation of the present study is the relative
small number of patients enrolled, due to the limited
number of patients complaining of hip pain and noise
after a ceramic hip prosthesis. Nevertheless, the number of patients was statistically relevant, and the statistical evaluation of synovial fluid analysis, as
diagnostic test for early recognition of a ceramic liner
failure, has shown that it has a very high specificity
and sensitivity.
Beside this, anything can be said about the nine
patients who showed ceramic wear in synovial fluid
but did not undergo to revision surgery, because no
confirmation of ceramic failure was definitively demonstrated in the components.
Figure 1. Flow-chart for the treatment of ‘‘noisy hip’’ patients.
JOURNAL OF ORTHOPAEDIC RESEARCH AUGUST 2012
1320
STEA ET AL.
To our knowledge, there are no other diagnostic tests
as predictive as needle aspirate for early diagnosis of
ceramic liner failure; on the basis of our results we
propose to introduce it in the therapeutic protocol of
noisy ceramic hips (Fig. 1).
In conclusion needle aspirate of noisy ceramic hips
has been proven in our hands to be a predictive diagnostic test for early recognition of ceramic failure, in
particular when it is strongly positive. Validation in
other centers will definitely allow us to propose it as
reference test.
ACKNOWLEDGMENTS
Dr. Milva Battaglia, Diagnostic Sonography of Istituto
Rizzoli is gratefully acknowledged for her contribution to synovial fluid sample collection. Financial support for research
was given by Istituto Rizzoli and by Ministry of Health,
grant RF-2009-1472961
REFERENCES
1. D’Antonio JA, Sutton K. 2009. Ceramic materials as bearing
surfaces for total hip arthroplasty. J Am Acad Orthop Surg
17:63–68.
2. Fritsch EW, Gleitz M. 1996. Ceramic femoral head fractures
in total hip arthroplasty. Clin Orthop Relat Res 328:129–
136.
3. Hannouche D, Nich C, Bizot P, et al. 2003. Fractures of
ceramic bearings: history and present status. Clin Orthop
Relat Res 417:19–26.
4. Hummer CD III, Rothman RH, Hozack WJ. 1995. Catastrophic failure of modular zirconia-ceramic femoral head
components after total hip arthroplasty. J Arthroplasty 10:
848–850.
5. Sariali E, Stewart T, Mamoudy P, et al. 2010. Undetected
fracture of an alumina ceramic on ceramic hip prosthesis.
J Arthroplasty 25:e1–e5.
JOURNAL OF ORTHOPAEDIC RESEARCH AUGUST 2012
6. Willmann G. 2000. Ceramic femoral head retrieval data.
Clin Orthop Relat Res 379:22–28.
7. Hamilton WG, McAuley JP, Dennis DA, et al. 2010. THA
with Delta ceramic on ceramic: results of a multicenter
investigational device exemption trial. Clin Orthop Relat
Res 468:358–366.
8. Toni A, Traina F, Stea S, et al. 2006. Early diagnosis of
ceramic liner fracture. Guidelines based on a twelve-year
clinical experience. J Bone Joint Surg Am 88:55–63.
9. Newcombe RG. 1998. Two-sided confidence intervals for the
single proportion: comparison of seven methods. Stat Med
17:857–872.
10. Walter WL, O’Toole GC, Walter WK, et al. 2007. Squeaking
in ceramic-on-ceramic hips: the importance of acetabular
component orientation. J Arthroplasty 22–24:496–503.
11. Choi IY, Kim YS, Hwang KT, et al. 2010. Incidence and
factors associated with squeaking in alumina-on-alumina
THA. Clin Orthop Relat Res 468:3234–3239.
12. Restrepo C, Matar WY, Parvizi J, et al. 2010. Natural history
of squeaking after total hip arthroplasty. Clin Orthop Relat
Res 468:2340–2345.
13. Keurentjes JC, Kuipers RM, Wever DJ, et al. 2008. High
incidence of squeaking in THAs with alumina ceramic-onceramic bearings. Clin Orthop Relat Res 466:1438–1443.
14. Walter WL, Yeung E, Esposito C. 2010. A review of squeaking hips. J Am Acad Orthop Surg 18:319–326.
15. Sharma V, Ranawat AS, Rasquinha VJ, et al. 2010. Revision
total hip arthroplasty for ceramic head fracture: a long-term
follow-up. J Arthroplasty 25:342–347.
16. Kempf I, Semlitsch M. 1990. Massive wear of a steel ball
head by ceramic fragments in the polyethylene acetabular
cup after revision of a total hip prosthesis with fractured
ceramic ball. Arch Orthop Trauma Surg 109:284–287.
17. Allain J, Roudot-Thoraval F, Delecrin J, et al. 2003. Revision
total hip arthroplasty performed after fracture of a ceramic
femoral head. A multicenter survivorship study. J Bone
Joint Surg Am 85-A:825–830.
18. Hasegawa M, Sudo A, Uchida A. 2006. Cobalt-chromium
head wear following revision hip arthroplasty performed
after ceramic fracture-a case report. Acta Orthop 77:833–835.