Content uploaded by Vasiliki-Christina Panagiotopoulou
Author content
All content in this area was uploaded by Vasiliki-Christina Panagiotopoulou on Sep 25, 2018
Content may be subject to copyright.
Assessment of Corrosion in Retrieved Spine Implants
V. C. Panagiotopoulou,
1,2
H. S. Hothi,
1
H. A. Anwar,
2
S. Molloy,
2
H. Noordeen,
2
K. Rezajooi,
2
J. Sutcliffe,
3
J. A. Skinner,
1,2
A. J. Hart
1,2
1
Institute of Orthopedics and Musculoskeletal Science, University College London, Stanmore, UK
2
The Royal National Orthopedic Hospital, Stanmore, UK
3
The London Spine Clinic, London, UK
Received 8 August 2016; revised 29 November 2016; accepted 16 January 2017
Published online 00 Month 2017 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.b.33858
ABSTRACT: Recently the use of dissimilar metals in spine
instrumentation has increased, especially in the case of adult
deformities, where rods made from Cobalt Chrome alloys
(CoCr) are used with Titanium (Ti) screws. The use of dissimi-
lar metals increases the risk of galvanic corrosion and
patients have required revision spine surgery due to severe
metallosis that may have been caused by corrosion. We
aimed to assess the presence of corrosion in spine implant
retrievals from constructs with two types of material combi-
nations: similar (Ti/Ti) and dissimilar (CoCr/Ti). First, we
devised a grading score for corrosion of the rod-fixture junc-
tions. Then, we applied this score to a collection of retrieved
spine implants. Our proposed corrosion grading score was
proven reliable (kappa >0.7). We found no significant differ-
ence in the scores between 4 CoCr and 11 Ti rods (p5
0.0642). There was no indication that time of implantation
had an effect on the corrosion score (p50.9361). We recom-
mend surgeons avoid using implants designs with dissimilar
metals to reduce the risk of corrosion whilst a larger scale
study of retrieved spine implants is conducted. Future studies
can now use our scoring system for spine implant corrosion.
V
C2017 Wiley Periodicals, Inc. J Biomed Mater Res Part B: Appl Bio-
mater 00B: 000–000, 2017.
Key Words: corrosion, spinal implant, cobalt–chromium
(alloys), titanium (alloys), implant retrieval
How to cite this article: Panagiotopoulou, VC, Hothi, HS, Anwar, H, Molloy, S, Noordeen, H, Rezajooi, K, Sutcliffe, J, Skinner,
J, Hart, A 2017. Assessment of Corrosion in Retrieved Spine Implants. J Biomed Mater Res Part B 2017:00B:000–000.
INTRODUCTION
Approximately 500,000 surgical procedures per year are
performed for spine instrumentation in USA,
1
exceeding hip
replacement surgeries by 100,000.
2
However, research in
retrieval analysis of spine instrumentation is surprisingly
underdeveloped compared to the research of hip retrievals.
3–5
Research on corrosion of orthopedic implants required
grading scores for: (1) corrosion on the taper interfaces
6
and the stem surfaces
7
of hip implants; and (2) wear of the
polyethylene of knee replacements.
8
In spine instrumenta-
tion, only one grading score exists for fretting and crevice
corrosion at the screw-rod junctions, but it applies only for
stainless steel constructs.
9
The use of dissimilar metals is
common in spine surgery to take advantage of the physical
and mechanical properties of different materials.
10
However,
there are recent concerns raised due to corrosion and
wear
5,11–17
and tissue reactions such as metallosis.
13,14,18
In the case of instrumentation that uses dissimilar met-
als, only preclinical testing has been performed
19
and nei-
ther post market surveillance studies using blood metal
ions, nor implant retrieval studies.
This is the first retrieval study of comparing similar and
dissimilar constructs at the rod-fixing junctions, focusing
both on the surface of the rod and on the surface of the
rest of the instrumentation.
Our aim was to better understand the clinical signifi-
cance of spine implants that use dissimilar metals. Our
objectives were, first to develop a visual grading method for
corrosion severity, secondly correlate surface damage on
rods to screws, nuts, hooks, and connectors; and thirdly
relate corrosion findings to clinical and implant data.
MATERIALS AND METHODS
This is a study of retrieved spine implants from seven patients
and two hospitals Figure 1. We included rods, screws, set
screws, hooks, and connectors from six different manufacturers
(DePuy Synthes, Globus, K2M, Medtronic, Orthofix, and Stryker).
Patients
The cohort was composed of four females and three males
with a median age of 43 (minimum 16–maximum 64) years
at primary with a median time of implantation of 21.5 months
(minimum 4 – maximum 68) (see Table I). Patients were
revised for infection, rod fracture, fracture and loosening of
screws, correction of kyphosis, and prevention of fusion.
Correspondence to: V.C. Panagiotopoulou; e-mail: v.panagiotopoulou@ucl.ac.uk
Contract grant sponsor: British Orthopedic Association through an industry consortium of nine manufacturers: DePuy International, Zimmer
GmbH, Smith & Nephew, UK, Biomet, UK, JRI, Finsbury Orthopaedics, Corin Group, Mathys Orthopedics, and Stryker, UK.
V
C2017 WILEY PERIODICALS, INC. 1
Macroscopic inspection
All rods were visually assessed for signs of discoloration
and/or scratching on the contact areas between (1) rods
and screws; (2) rods and nuts; (3) rods and hooks; and (4)
rods and connectors. The same applied for the contact sur-
faces on the screws, setscrews, hooks, and connectors.
Details about the number of components and contact areas
can be found in Table II.
Microscopic inspection
Optical microscopy was performed on all components for
signs of fretting, pitting, and/or corrosion using a Keyence
VHX-700F series (Keyence Co., Japan). The magnification
ranged from 203to 2003according to the inspected
component.
Scanning electron microscope (SEM)
We used Scanning Electron Microscope (Hitachi S-3400 N)
to determine surface changes on the areas of interest, espe-
cially to find signs of fretting and pitting on the contact
areas on both rods and the rest of the components.
Energy dispersive X-ray spectroscope (EDX)
In combination with SEM, we performed elemental analysis
using Energy Dispersive X-ray Spectroscope (Oxford Instru-
ments) at a working distance of 10 mm, to determine
whether the black debris seeing on the components are
either biological deposits or corrosion products.
Creation and evaluation of grading score
After performing macroscopic and microscopic inspection in
all contact areas between the components, we combined the
findings in different groups. These groups served as a base-
line for the proposed grading score for rods (see Figure 2)
and for the adaptation of the score proposed for the tapers
of hip replacement by Goldberg (see Figure 3). For the eval-
uation of the objectivity of the grading scores, all compo-
nents were scored by two independent examiners with
expertise on corrosion of implants. The scores were then
compared for any statistical differences.
Statistical analysis
Statistical analysis, using SPSS software, was performed to
investigate (1) any significant difference or trends between
the scores for CoCr rods versus the scores for Ti rods; (2)
any correlation between score severity and time of implan-
tation; (3) the objectivity and repeatability of the proposed
grading score systems; and (4) correlation between the
score of the rod and the score of the screws, setscrews,
hooks, and connectors.
RESULTS
Macroscopic inspection and grading – rods
From the macroscopic inspection of the spine rods, we
noticed four main trends on the surfaces of the rods: (1) no
visible corrosion and/or fretting; (2) light discoloration
TABLE I. Demographic Data for the Patient Cohort, Including Gender, Age at Primary, Reason for Primary and Revision Sur-
geries, Months of Implantation, Material Combination, and Implant Manufacturer
Patient Gender
Age at
Primary
Initial
Diagnosis
Reason for
Revision
Months of
implantation
Material
Combination Manufacturer
1 F 64 Scoliosis Infection 68 CoCr/Ti K2M
2 F 69 Scoliosis Rod fracture 27 CoCr/Ti K2M
3 M N/A Fractured
Vertebra
Fractured Screw N/A Ti/Ti Stryker
4 M 30 Fractured
Vertebra
Correction of
Kyphosis
9 Ti/Ti DP Synthes/
Stryker
5 F 18 Scoliosis Infection 4 Ti/Ti Medtronic
6 F 16 Fractured
Vertebra
Prevent adjacent
segment
degeneration
16 Ti/Ti Orthofix
7 M 56 Fractured
vertebra
Screw Loosening 5 Ti/Ti Globus
TABLE II. Number of Components and Contact Areas Per Patient. Different Patterns on the Rods Revealed Different Contact
Areas From Different Components, Thus Attributing to the Large Number of Contact Areas. Retrieved Fixtures were Less
Than the Contact Areas on the Rods, Common Side Effects of the Increased Number of Components Used in Spinal Surgery
Patient
Number
of Rods
Number of
Fixtures
Number of Contact
Areas on the Rods
Number of Contact
Areas on the Fixtures
1 2 36 60 36
2 2 18 32 18
324 4 4
4 4 15 16 15
5 2 17 28 17
628 8 8
714 4 4
2 PANAGIOTOPOULOU ET AL. ASSESSMENT OF CORROSION IN RETRIEVED SPINE IMPLANTS
(affecting <30% of the junction point) but without presence
of pits or scratching; (3) moderate discoloration (affecting
>30% of the junction point), usually accompanied by the
presence of some pits, scratches and/or black debris; and
(4) severe discoloration (affecting >50% of the junction
point), accompanied by pits, scratching and/or black debris.
Using these observations, we composed a visual grading
score for the effect of screws, setscrews, hooks, and
connectors on the surface of spinal rods (see Figure 3 for
score and criteria).
Table III includes the number of junctions that share the
same score per material combination. Only 5% of junctions
on the CoCr rods showed no evidence of surface damage.
About 46% of the CoCr rod junctions were discolored, but
with no evidence of scratching, pits or black debris. We
found that 44% of junctions had evidence of moderate
FIGURE 1. Samples of components used in this study; where components labelled 1a–1d are Ti nuts, 2a–2d, 2f, and 2g are Ti rods, 2e is CoCr
rod, 3a–3c are Ti connectors, and 4a–4g are Ti screws.
FIGURE 2. Different grades of the visual scoring of the junctions on the rods, with corrosion score from 1 to 4, under macroscopic inspection
(blue circles mark the areas of interest), microscopic inspection of the areas of interest using Optical Microscope and Scanning Electron Micro-
scope (SEM), as scored according to the criteria mentione d. Blue arrows identify the damaged areas with fretting.
ORIGINAL RESEARCH REPORT
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS |MONTH 2017 VOL 00B, ISSUE 00 3
discoloration and some scratching, pits or black debris,
while 5% of the CoCr/Ti junctions had the highest corrosion
score,
4
with severe discoloration and/or increased presence
of scratching, pits, or black debris. For the Ti rods, more
than half of the junctions (67%) were discolored only (no
scratching, pitting or black debris), while the rest of the
junctions were moderately discolored (33%), with scratching,
pits or black debris. No junctions on Ti rods were severely
damaged (Grade 4).
In cases of junction scoring, where the percentage of
discoloration resulted in a score different that the one that
the percentage of surface marks proposed, the junctions
were scored according to the severity of scratching, pitting,
and black debris.
Macroscopic inspection and grading – fixtures
The macroscopic inspection of the contact areas of the rest
of the components revealed similarities with the assessment
of corrosion at the taper junctions of hip replacements.
Thus, we adopted the corrosion score proposed from Gold-
berg et al.
6
for the visual evaluation of the contact areas on
screws, setscrews, hooks, and connectors (see Figure 3).
From the macroscopic grading of the retrieved screws,
setscrews, hooks, and connectors by first examiner (see
Table IV), the majority of Ti components attached to CoCr
rods (88%) were either lightly or moderately discolored,
with half of them having signs of black debris too. 6% of
components were found with no apparent signs of damage,
while only 6% had severe discoloration and excess amount
of black debris. In the case of using similar materials, most
of the Ti components (48%) were moderately damaged
with <30% of their surface discolored but no black debris,
scratching, and pits. Rest of the Ti fixtures had no black
debris, with 40% being slightly discolored and 12% showed
no signs of discoloration.
20
Microscopic inspection and grading – rods
From the microscopic inspection of the rods, we found that
the more severe the visual score, the more severe signs of
fretting and pitting were present under the microscope. In
the case of CoCr, Grade 1 was accompanied with initial
marks of intergranular corrosion. Grade 2 and Grade 3
FIGURE 3. Different grades of the visual scoring on the contact areas on the fixtures. Screw with corrosion score 1, showed no signs of fretting
during macroscopic inspection, while under the optical microscope and SEM the area of interest (blue circles) showed slight fretting perpendicular
to the manufacturing lines. Nuts with corrosion score 2 and 3 presented some circular marks (blue arrows), resulting to discoloration affecting less
or more than 30% of the contact area, respectively. Both optical microscope and SEM confirmed circular fretting, probably during implantation.
Connector with corrosion score 4, had signs of black debris, which was confirmed to be of biological origin from Energy Dispersive X-ray
Spectroscopy.
TABLE III. Summary of Scores for the Spinal Rods in
Constructs with Similar or Dissimilar Materials
Visual
Score
CoCr/Ti Junctions
(n592)
Ti/Ti Junctions
(n560)
1 5 (5%) 0
2 42 (46%) 40 (67%)
3 41 (45%) 20 (33%)
4 4 (4%) 0
4 PANAGIOTOPOULOU ET AL. ASSESSMENT OF CORROSION IN RETRIEVED SPINE IMPLANTS
shared signs of fretting and pitting, as well as severe signs
of intergranular corrosion. The severity of fretting and pit-
ting escalated with the severity of visual scoring grade. On
the Ti rods, Grade 1 showed no signs of damage, while fret-
ting and pitting was present in Grades 2 and 3. The higher
the grade of the junction, the more fretting and pitting was
observed under the microscope.
Microscopic inspection and grading –fixtures
In all Ti components, microscopic inspection of grade 1
showed no signs of fretting or pitting. In components scored
as grade 2, fretting was present, while in grades 3 and 4,
severe fretting was combined with some level of pitting.
Scanning electron microscope and grading – rods
Similarly, as the microscopic inspection, SEM revealed that
the severity of the visual score was accompanied by the
severity of fretting, pitting, and presence of black debris. As
seen in Figure 2, in Grade 1 for CoCr rods there is presence
of intergranular corrosion initiation, although not visible
during macroscopic inspection. For grades higher than 2,
fretting and pitting are present. The more severe the fret-
ting and pitting, the higher the grade. Black debris were
also found while performing SEM.
Scanning electron microscope and grading – fixtures
We performed SEM on Ti components from both CoCr/Ti
constructs and Ti/Ti constructs. Fretting and pitting corro-
sion were present on moderately damaged fixtures, while
fixtures with score 1 showed no significant signs of fretting
(see Figure 3). In fixtures with grade 4, a large surface of
the contact area was covered with black debris.
Energy dispersive X-ray spectroscope and
grading – rods
On the junctions of CoCr rods, elemental analysis on the
black debris revealed transfer from the Ti components to
the CoCr rods. The severity of the grade was reflected on
the presence of oxygen, as well as titanium, aluminum, and
vanadium elements from the Ti components. Corrosion
debris also included sulphur and phosphorus.
On the contrary, black debris on the junctions of Ti rods
were of biological origin, since the elemental composition
revealed potassium, calcium, sodium, chloride, iron, and
oxygen among the elements detected.
Comparing EDX analysis data between CoCr rods and Ti
rods, the amount of oxygen was increased on the CoCr rods
(27% on average) compared to Ti rods (17% on average).
However, we were not able to confirm whether galvanic
coupling accelerated the corrosion process.
Energy dispersive X-ray spectroscope and
grading – fixture
Elemental composition analysis on Ti components from
both CoCr/Ti and Ti/Ti constructs revealed that black
debris was mostly biological, rather than corrosive debris.
This was suggested by the presence of oxygen with sodium,
potassium, chloride, and iron. No transfer from CoCr rods to
Ti fixtures was observed.
Evaluation of corrosion grading score – rods
All rod junctions were evaluated independently by two
examiners, expert in corrosion of medical devices. We then
performed kappa analysis in order to assess the agreement
of the two examiners, and thus determine the repeatability
and objectivity of the proposed score for corrosion on rod
junctions. Analysis in SPSS revealed that kappa was 0.737
for the scoring of CoCr rods, while in the case of Ti rods
kappa was 0.775. Both values are described as substantial
agreement between the two examiners in grading the junc-
tions of the rods.
Evaluation of corrosion grading score – fixtures
Same principal was applied to the scoring of the fixtures.
Once agreed to the adaptation of the already published
Goldberg score to the contact areas of the fixtures, the two
examiners scored the areas independently. SPSS analysis
TABLE IV. Summary of Scores for the Retrieved Screws,
Nuts, Hooks, and Connectors in Constructs with Similar or
Dissimilar Materials
Visual Score CoCr/Ti Ti/Ti
1 3 (6%) 6 (12%)
2 23 (44%) 19 (40%)
3 23 (44%) 23 (48%)
4 3 (6%) 0
Total number of contact
areas of retrieved fixtures
52 48
TABLE V. Mean Values of Scores Per Rod and Fixtures Per
Patient
Component Mean Score
Patient 1 Rod 1 2.5
Rod 2 2.3
Fixtures 2.5
Patient 2 Rod 1 2.8
Rod 2 2.4
Fixtures 2.5
Patient 3 Rod 1 2.0
Rod 2 2.0
Fixtures 2.5
Patient 4 Rod 1 2.0
Rod 2 2.0
Rod 3 2.3
Rod 4 2.3
Fixtures 1.8
Patient 5 Rod 1 2.3
Rod 2 2.8
Fixtures 2.9
Patient 6 Rod 1 2.0
Rod 2 2.0
Fixtures 2.1
Patient 7 Rod 1 2.5
Fixtures 2.0
ORIGINAL RESEARCH REPORT
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS |MONTH 2017 VOL 00B, ISSUE 00 5
revealed that the kappa coefficient was 0.753 suggesting
substantial agreement between the examiners.
Statistical analysis
Using SPSS software, we performed statistical analysis in
order to find any significant differences between the use of
different materials and the time of implantation. We calcu-
lated the mean value for all junctions per rod and the mean
value for all contact areas of fixtures per patients (see Table
V for the values). Using the ANOVA non-parametric test, we
found no significant difference for the corrosion scores
between using similar or dissimilar materials (p50.0642).
We also tested whether there is correlation between time of
implantation and corrosion score of the rod junctions, but
the non-parametric ttest showed no significant difference
(p50.9361). Comparison of the scores between fixtures
from different patients, revealed no significant difference
when correlated with time of implantation.
DISCUSSION
In this study, we describe grading scores and criteria for the
assessment of fretting, pitting, and corrosion on rod junc-
tions and contact areas of fixtures used in spinal implants.
Statistical kappa analysis showed the scoring systems to be
reproducible between different examiners. These scoring
methods may now be used in future studies investigating
retrieved spinal implants.
Our findings suggest that there is no significant differ-
ence in the severity of corrosion between the CoCr and Ti
rod junctions investigated in this study. It is of note howev-
er that only the CoCr components showed evidence of
severe corrosion (grade 4), in the case of rods revised for
fracture. This difference between the two metal alloy combi-
nations was however only marginally insignificant
(p50.0642).
Comparing the results from the forensic analysis of the
different materials, similar macroscopic patterns between
the two different types of constructs with the same score
resulted in similar trends during the microscopic inspection,
highlighting the similarities due the implantation damage
regardless the material. The use of different materials in
spine instrumentation self-indicates the importance of a uni-
versal score, which can be used regardless material combi-
nation or material consistency. Based on the hip retrieval
research, the widely-used Goldberg score
6
has been estab-
lished by comparing fretting and corrosion on both retriev-
als made of CoCr and Ti alloys.
We did not find a significant correlation between time of
implantation and overall rod score, suggesting that some
indentations and marks on the rod junctions might occur
during the fixing process of the implantation rather than
during use in situ. The mean corrosion scores for CoCr rods
revised for fracture were higher than the mean corrosion
scores for CoCr rods revised for infection, but no significant
difference was documented. This finding might suggest that
metallosis is rather a patient specific factor instead of a
result of combining different materials. However, it is
acknowledged that this study may be under-powered and
future studies involving great numbers of retrievals are
important.
Ti fixtures had similar levels of fretting and pitting
regardless of whether they were fixed with CoCr or Ti rods.
Black debris on the contact areas of the fixtures were most-
ly biological, while no transfer from the CoCr rods to Ti
fixtures was observed. Comparing the mean values of corro-
sion scores between rod junctions and fixtures of the same
patient showed that both components were almost equally
damaged, revealing no significant difference on the surface
damage between the more active metal part and the more
noble part of the junction. This suggests that galvanic cou-
pling of the metals in this instance did not accelerate the
corrosion process and was rather stable, similar to that sug-
gested by preclinical testing.
19
Up-to-date, most spine retrieval papers used compo-
nents made of Stainless Steel and Ti alloys,
9,12,18
both of
which have been used for several years in treating spine
deformities. Evidence of corrosion and wear has been docu-
mented, including case reports of metallosis, which is linked
to surface deterioration of the implanted devices.
20–24
How-
ever, there are more recent suggestions that the physical
and mechanical properties of CoCr rods are more appropri-
ate for the increased correctional forces of the deformed
spine than rods made of different material.
10
This resulted
in the introduction and promotion of CoCr rods from sever-
al manufacturers, especially in the cases of extensive adult
deformities. Based on the relevant published research on
retrieved hip replacements, there is evidence of increased
corrosion in mixed materials, putting the galvanic corrosion
into the spot light. The presence or not of this phenomenon
needs to be determined, in order to secure the safety of the
material combination used in spine surgery.
This work is the first retrieval study including compari-
son between similar and dissimilar use of materials on rod-
screw junctions, using components made of CoCr and Ti
alloys. This is a first step toward investigating and deter-
mining the in vivo performance of spine instrumentation;
future work will involve correlating the material loss and
change of surface roughness of the contact areas with the
corrosion scores described here. Equally important is the
correlation of the junction score in relation to the location
at the spine column in order to evaluate whether increased
loading results in increased corrosive damage. Finally, the
inclusion of more material combinations will offer a better
understanding of the processes that are taking place during
the implantation of spine instrumentation in the cases of
similar or dissimilar material combination.
CONCLUSION
We have described corrosion scores for the assessment of
retrieved spinal implants. Using this score, we found no evi-
dence of increased corrosion when two different materials
are galvanically coupled in spine instrumentation. This sug-
gests that metallosis may due more than just implant risk
6 PANAGIOTOPOULOU ET AL. ASSESSMENT OF CORROSION IN RETRIEVED SPINE IMPLANTS
factors, and the impact of the patient and surgeon should
be considered.
ACKNOWLEDGMENTS
We are grateful for the support of the theatre staff at the Royal
National Orthopedic Hospital, Anna Di Laura, Arianna Cerqui-
glini, Ilona Swiatkowska, and Akramul Hoque for their coordi-
nation of the retrieval center, Tom Gregory, SEM technician, at
UCL Archaeology Institute, Amir Amiri, Ali Najefi, and Adam
Benton for providing patient details and consents.
REFERENCES
1. Rajaee S, Kanim L, Bae H. National trends in revision spinal
fusion in the USA. Bone Joint J 2014;96(6):807–816.
2. Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary
and revision hip and knee arthroplasty in the United States from
2005 to 2030. J Bone Joint Surg Am 2007;89(4):780–785.
3. Gilbert JL, Jacobs JJ. The mechanical and electrochemical pro-
cesses associated with taper fretting crevice corrosion: A review.
Modularity of orthopedic implants. ASTM International; 1997.
4. Jacobs JJ, Gilbert JL, Urban RM. Current concepts review-
corrosion of metal orthopaedic implants. J Bone Joint Surg Am
1998;80(2):268–282.
5. Oladok un A, Pettersson M, Bryant M, Engqvist H, Persson C, Hall
R, Neville A. Fretting of CoCrMo and Ti6Al4V alloys in modular
prostheses. Tribol Mater Surface Interfac 2015;9(4):165–173.
6. Goldberg JR, Gilbert JL, Jacobs JJ, Bauer TW, Paprosky W,
Leurgans S. A multicenter retrieval study of the taper interfaces of
modular hip prostheses. Clin Orthop Relat Res 2002;401:149–161.
7. Bryant M, Ward M, Farrar R, Freeman R, Brummitt K, Nolan J,
et al. Failure analysis of cemented metal-on-metal total hip replace-
ments from a single centre cohort. Wear 2013;301(1):226–233.
8. Hood RW, Wright TM, Burstein AH. Retrieval analysis of total
knee prostheses: A method and its application to 48 total condylar
prostheses. J Biomed Mater Res 1983;17(5):829–842.
9. Akazawa T, Minami S, Takahashi K, Kotani T, Hanawa T, Moriya H.
Corrosion of spinal implants retrieved from patients with scoliosis.
J Orthop Sci 2005;10(2):200–205.
10. Serhan H, Mhatre D, Newton P, Giorgio P, Sturm P. Would CoCr
rods provide better correctional forces than stainless steel or titani-
um for rigid scoliosis curves? Clin Spine Surg 2013;26(2):E70–E74.
11. Serhan H, Slivka M, Albert T, Kwak SD. Is galvanic corrosion
between titanium alloy and stainless steel spinal implants a clini-
cal concern? Spine J 2004;4(4):379–387.
12. Vieweg U, Van Roost D, Wolf HK, Schyma CA, Schramm J. Corro-
sion on an internal spinal fixator system. Spine 1999;24(10):946–
951.
13. Kasai Y, Iida R, Uchida A. Metal concentrations in the serum and
hair of patients with titanium alloy spinal implants. Spine 2003;
28(12):1320–1326.
14. del Rio J, Beguiristain J, Duart J. Metal levels in corrosion of spi-
nal implants. Eur Spine J 2007;16(7):1055–1061.
15. Villarraga ML, Cripton PA, Teti SD, Steffey DL, Krisnamuthy S,
Albert T, et al. Wear and corrosion in retrieved thoracolumbar
posterior internal fixation. Spine 2006;31(21):2454–2462.
16. Kirkpatri ck JS, Venugopalan R, Beck P, Lemons J. Corrosion on
spinal implants. J Spinal Disord Tech 2005;18(3):247–251.
17. Prikryl M, Srivastava S, Viviani G, Ives M, Purdy G. Role of corro-
sion in Harrington and Luque rods failure. Biomaterials 1989;
10(2):109–117.
18. Wang JC, Warren DY, Sandhu HS, Betts F, Bhuta S, Delamarter
RB. Metal debris from titanium spinal implants. Spine 1999;24(9):
899–903.
19. Mali SA, Singh V, Gilbert JL. Effect of mixed alloy combinations
on fretting corrosion performance of spinal screw and rod
implants. J Biomed Mater Res Part B 2016;00B:000–000.
20. Takahashi S, Delecrin J, Passuti N. Intraspinal metallosis causing
delayed neurologic symptoms after spinal instrumentation sur-
gery. Spine 2001;26(13):1495–1498.
21. Tezer M, Juzgun U, Hamzaoglu A et al. Intraspi nal metalloma
resulting in late paraparesis. Arch Orthop Trauma Surg 2005;
125(6):417–421.
22. Gaine WJ, Andrew SM, Chadwick P, et al. Late operative site pain
with isola posterior instrumentation requiring implant removal:
Infection or metal reaction? Spine 2001;26(5):583–587.
23. Goldenberg Y, Tee JW, Salinas-La Rosa CM, Murphy M. Spinal
metallosis: A systematic review. Eur Spine J 2016;25(5):1467–1473.
24. Senaran H, Atilla P, Kaymaz F, et al. Ultrastructural analysis
of metallic debris and tissue reaction around spinal implants
in patients with late operative site pain. Spine 2004;29(15):1618–1623.
ORIGINAL RESEARCH REPORT
JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS |MONTH 2017 VOL 00B, ISSUE 00 7
