Citation
Local tissue reaction to the release of  metal ions from spinal implants

Material Information

Title:
Local tissue reaction to the release of metal ions from spinal implants
Creator:
Miller, Mackenzie ( author )
Language:
English
Physical Description:
1 electronic file (57 pages) : ;

Subjects

Subjects / Keywords:
Implants, Artificial ( lcsh )
Biomedical materials ( lcsh )
Spine ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
Implant failure and metallosis are current issues in the field of orthopedics. The existing hypotheses used to explain this phenomenon, type IV metal hypersensitivity and ALVAL, are not sufficient and do not provide clinical relevance. Following the approved IRB protocol patients with scheduled revision surgeries that met the inclusion criteria were consented and enrolled into the study. During the revision surgery, tissue samples and removed hardware were collected and analyzed. From visual observation corrosion and wear were evident on all retrieved hardware. Metal ion concentrations were above normal values in all three patients with ICP-MS/AES data. These results indicate a cause and effect relationship between the corrosion of the hardware and the elevated metal ion concentrations. Using the ICP-MS/AES results, a voltage that was generated from the corrosion products of the metal instrumentation was calculated using the Nernst equation. These voltages were elevated above normal membrane potentials which could lead to adverse reactions in the tissue. From histological analysis two common themes were observed, foreign body giant cell reactions and histiocytes, laden with foreign material, trapped in the fibrous tissue. These are both pathological responses that indicate an abnormal reaction to the foreign material released from the metal hardware in the body. The results of this study connect important clinical factors with engineering principles and address the need for new collaboration, communication, and invention between material science, engineering, and medicine.
Thesis:
Thesis (M.S.)--University of Colorado Denver
Bibliography:
Includes bibliographical references.
System Details:
System requirements: Adobe Reader.
Statement of Responsibility:
by Mackenzie Miller.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
on10073 ( NOTIS )
1007343411 ( OCLC )
on1007343411

Downloads

This item has the following downloads:


Full Text
LOCAL TISSUE REACTION TO THE RELEASE OF METAL IONS FROM SPINAL
IMPLANTS
by
MACKENZIE MILLER B.S., University of Louisiana at Monroe, 2015 M.S., University of Colorado, Denver, 2017
A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Science Bioengineering Program
2017


This thesis for the Master of Science degree by Mackenzie Miller has been approved for the Bioengineering Program by
Cathy Bodine, Chair, Ph.D.
Reed Ayers, Ph D.
Christopher J. Kleck, MD
Date: May 13, 2017
11


Miller, Mackenzie (M.S., Bioengineering Program)
Local tissue reaction to the release of metal ions from spinal implants Thesis directed by Associate Professor Cathy Bodine
ABSTRACT
Implant failure and metallosis are current issues in the field of orthopedics. The existing hypotheses used to explain this phenomenon, type IV metal hypersensitivity and ALVAL, are not sufficient and do not provide clinical relevance. Following the approved IRB protocol patients with scheduled revision surgeries that met the inclusion criteria were consented and enrolled into the study. During the revision surgery, tissue samples and removed hardware were collected and analyzed. From visual observation corrosion and wear were evident on all retrieved hardware. Metal ion concentrations were above normal values in all three patients with ICP-MS/AES data. These results indicate a cause and effect relationship between the corrosion of the hardware and the elevated metal ion concentrations. Using the ICP-MS/AES results, a voltage that was generated from the corrosion products of the metal instrumentation was calculated using the Nernst equation. These voltages were elevated above normal membrane potentials which could lead to adverse reactions in the tissue. From histological analysis two common themes were observed, foreign body giant cell reactions and histiocytes, laden with foreign material, trapped in the fibrous tissue.
These are both pathological responses that indicate an abnormal reaction to the foreign material released from the metal hardware in the body. The results of this study connect important clinical factors with engineering principles and address the need for new collaboration, communication, and invention between material science, engineering, and
m
medicine.


The form and content of this form are approved. I recommend its publication.
Approved: Cathy Bodine
IV


ACKNOWLEDGEMENTS
To my biggest fans, greatest supporters, and most loving parents, thank you for your support and encouragement in every aspect of my life. To my loving and supportive boyfriend Matt, thank you for everything that you do! I love you guys!
A very special gratitude to my thesis committee, Dr. Cathy Bodine, Dr. Reed Ayers, and Dr. Christopher Kleck. It has been such an honor to work with all of you.
I am also grateful for the insight and support from the following University Health staff: Dr. Schowinsky, Dr. Patel, Susan Estes, Dr. Burger, Robert Cooley, Dr. Cain, Theresa Schroeder, and Dr. Ou-yang.
A special thanks to Claire Cofer and Emily Lindley for their help and guidance through the IRB and human subjects research process.
I couldnt be here without all of you! Thanks for everything!
v


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION............................................................9
II. REVIEW 01 THE LITERATURE............................................. 14
Metallosis and Spine Surgery............................................14
Corrosion..........................................................18
Metal Ions in the Body..................................................22
Metal Hypersensitivity..................................................25
Type IV cell-mediated response........................................26
Metal Hypersensitivity Testing........................................26
Aseptic Lymphocyte-Dominated Vasculitis Associated Lesion...............29
Electrochemical voltage shifts created as a result of implant corrosion.29
Conclusion.........................................................31
III. MATERIALS AM) METHODS................................................32
IV. RESULTS...............................................................42
V. DISCUSSION............................................................46
Limitations...........................................................47
Future Research.......................................................49
VI. CONCLUSION............................................................52
REFERENCES................................................................53
vi


LIST OF TABLES
TABLE
Table 1. Patient demographic information including gender, age, duration of previous hardware in situ, location of the revision, and indications for the revision based on the inclusion criteria...............................................................42
Table 2. From the retrieved instrumentation the type of metal and types of corrosion and wear seen on the hardware through visual inspection were recorded. Only the presence or absence of a type of corrosion or wear was recorded..............................43
Table 3. ICP-MS/AES data measuring the metal ion concentrations found in the retrieved tissue samples. Values below the detection of ICP-MS/AES are reported as < 0.1...46
vn


LIST OF FIGURES
FIGURE
Figure 1. (a) Intraoperative image of metallosis from a revision spine surgery, (b) Removed pseudotumor with metal staining from the same revision surgery (photos courtesy of Reed
Ayers, PhD.)........................................................................17
Figure 2. Circuit drawings representing a (a) multi-layer oxide and (b) a single-layer oxide (used with the permission of Dr. Rahul Bhola)..........................................18
Figure 3. Evidence of fretting (a) and pitting (b) corrosion on retrieved spinal instrumentation made of Ti.............................................................20
Figure 4. Schematic representation of the corrosion cell generated in vivo. As the metal instrumentation corrodes electrons, followed by positively charged metal ions, flow to the tissue which acts as the cathode. Negatively charged species, like oxygen and hydroxide, then move back to the metal instrumentation acting as the anode.......................40
Figure 5. The Nernst equation used for this calculation. R, T, and F are all know constants. The valency, z, used is the most common oxide state of each metal. The extracellular metal ion concentrations are the values from the experimental data. The intracellular metal ion concentrations are pulled from previous literature....................................41
Figure 6. (a) Evidence of pitting corrosion, (b) Fretting wear has the shiny appearance along the rod caused from micromotion between the connector and the rod. (c) Evidence of repassivation. The discoloration along the screw indicates reformation of the oxide layer at a different thickness, (d) Fracture between the threading and the head of the screw.....44
Figure 7. All images are of hematoxylin and eosin stained slides taken at 400x magnification. A) Foreign body giant cell reaction. B) Particle laden histiocytes trapped in fibrous tissue. C) Necrotic skeletal muscle and necrotic fibrovascular connective tissue. D) Typical look of metallosis............................................................45
vm


INTRODUCTION
Implant failure and metallosis are current issues in the field of orthopedics. Patients with these findings can experience severe pain, inflammation, and ultimately, implant failure leading to revision surgery. The cause and effect of metallosis has yet to be identified making the clinical diagnosis and prevention impossible. Current hypotheses include are metal hypersensitivity due to delayed type IV cell-mediated immune response and Aseptic Lymphocytic Vasculitis Associated Lesion (ALVAL). These reactions lead to the recruitment of macrophages leading to inflammation and pain. No experimental data has established a cause and effect relationship between these hypotheses and the clinical diagnosis of metallosis. Furthermore, there is no clinical relevance to these hypotheses because no tests currently exist to predict metal allergies due to implants or ALVAL. There is a significant need to further research the phenomenon of metallosis.
It is well documented that metals corrode in the body and release metal ions in the surrounding tissue. The electrochemical process of corrosion generates a voltage that can alter the physiological environment of local soft tissue. Previous in vitro experiments have shown that specific voltage ranges lead to either cell necrosis or apoptosis. Using failure analysis methods, we can identify corrosion products from orthopedic spine implants, identify the concentration of metal ions in the surrounding tissue, and calculate the voltage created by these corrosion products which may be linked to poor health outcomes and potential need for revision surgery.
The intellectual merit of this proposed activity is crucial in the advancement of knowledge and understanding by providing evidence to support the hypothesis that
9


corrosion is linked to poor health outcomes and the potential need for revision surgery. This will connect the fields of orthopedics, human physiology, and material science by identifying a link between the implant, the procedure, and the patient outcomes. The team is qualified to conduct this research because of the combined knowledge of materials and corrosion, human physiology, and clinical experience. The proposed concept is original in its entirety because it will connect the previous research of electrochemical gradients and implant failure due to metallosis to a clinical setting. The proposed activity will be conducted under an organized and chronological approach. First patients will be screened and enrolled to participate in the study. Once consented, the patients hardware and tissue samples will be collected during the revision surgery. Hardware will undergo destructive and non-destructive metallography testing to identify areas and types of corrosion. One tissue sample will be sent for ICP-MS/AES analysis and the other will be sent for histological analysis. The results of ICP-MS/AES will be used to calculate the voltage generated by the corrosion products using the Nernst equation. There is existing access to resources.
The broader impact of this proposed activity will advance the understanding and significance of metallosis in this field by supporting the hypothesis that corrosion leads to poor patient outcomes and need for revision surgery. The results of this study will be published in at least one scientific journal to enhance the scientific and technological understanding of metallosis. The results of the proposed activity will benefit society by developing a clinically relevant explanation to metallosis that may lead to better patient outcomes. Currently, patients that have experienced implant failure in the form of mechanical failure, pseudotumor, non-union, pain, and disk degeneration must undergo a
10


revision surgery. These surgeries are costly, painful, and may not always provide the best solution to the issue. A better understanding of metallosis and implant failure will improve future patient care and patient outcomes.
11


Advisor
Reed Ayers, Ph.D. Department of Orthopedics Committee Members Reed Ayers, Ph D.
Cathy Bodine, Ph.D.
Christopher Kleck, MD Title of Proposal
Local Tissue Reaction to the Release of Metal Ions from Spinal Implants Hypothesis
Corrosion products from orthopedic spine implants alters the physiologic environment of local soft tissue, by generating a voltage, leading to cell necrosis, identified through histology; propagating a pathological response that may lead to poor health outcomes.
Specific Aims
Specific Aim One: To collect two tissue samples from revision patients with and without evidence of a solid fusion and hardware failure; one sample adjacent to the metal implant (peri-hardware tissue) and one sample from either the vertebrae above or below the revision (control tissue). Tissue samples will be collected from peri-hardware and control areas during the revision surgery at the University of Colorado Hospital by Orthopedic Spine Surgeons. Control and peri-hardware samples will be stored in separate containers prior to sample preparation.
12


Specific Aim Two: To perform non-destructive and destructive analysis on the explanted instrumentation. All instrumentation will be photographed, and then analyzed under a scanning electron microscopy as well as energy dispersive spectroscopy to identify regions of corrosion and wear. Physical metallography will also be used for this analysis.
Specific Aim Three: To perform ICP-MS/AES on the collected tissue samples. At least two grams, but no more than 10 grams, of the peri-hardware and control samples will be stored in sterile sample jars and sent to Huffman Hazen Laboratories for analysis. These samples will be utilized to determine the presence and quantity of metal ions in the sample.
Specific Aim Four: To correlate the results of this research to clinical data acquired through the medical records and to prior in vitro studies. ICP-MS/AES results will be correlated to data acquired through the medical records. In another analysis, the results of ICP-MS/AES will be used to calculate the voltage generated by the instrumentation in-vivo and compared to prior in vitro studies.
13


REVIEW OF THE LITERATURE
Metallosis and Spine Surgery
An orthopedic implant is a device surgically placed into the body designed to restore function by replacing or reinforcing a damaged structure (Orthopedic Implants 2017). In orthopedic spine surgery, implants are used to treat deformity, mechanical back pain, stenosis, spondylolisthesis, fractures, and tumors. The varied requirements of these implants are diverse, to treat multiple pathologies, in individuals across the lifespan.
The number of orthopedic spine surgeries is on the rise. An epidemiological study conducted by Rajee observed a 137% increase in spinal fusion discharges in the U.S. between the years of 1998 to 2008, a much higher rate than any other orthopedic procedure (Rajaee et al. 2012). More patients are undergoing primary spine surgeries at a younger age which could later lead to revision surgeries. Revision surgeries are necessary for patients that experience degenerative disc disease, pseudoarthrosis, mechanical hardware failure, non-union, pain, and adjacent segment disease. These complications are often summarized as implant failure. A retrospective study conducted by Kelly et al. at an institution specific to spinal implants found that 21% of patients would undergo multiple revision surgeries (Kelly et al. 2013). That is double the percentage of revision surgeries seen in all of orthopedics. With the number of revision spine surgeries increasing it is important to define success in spinal surgery.
The success of a surgery is difficult to gauge since there is no clear definition of surgical or clinical success. In a study by Yee et al., almost 20% of patients were dissatisfied with the result of their surgery, with that number increasing in patients that had previous spinal intervention (Yee et al. 2008). Surgeons and patients have different
14


perspectives on successful procedures. Other factors contributing to the confusion of what defines a successful procedure include time and type of surgical procedure. In a retrospective study conducted by Maruenda and Barrios, 15-year follow-up data was collected for patients that underwent primary 1-3 level circumferential lumbar fusion. They reported good clinical results within the first year post surgery but from 2- to 15-year follow-up, the clinical outcome of patients worsened significantly. This resulted in 24 patients (37.5%) having a revision surgery (Maruenda et al. 2016). From this study it is apparent that time is a factor in patient success and rates of revision surgery. A retrospective cohort analysis conducted by Djurasovic and Glassman, measured quality of life improvements for patients undergoing a revision lumbar spine surgery (Djurasovic et al. 2011). They concluded that only modest improvements could be expected from a revision surgery as measured by patients reaching minimum clinically important differences (MCID) when comparing preoperative, 1 year postoperative, and 2 year postoperative Oswestry Disability Index (ODI) and MOS Short Form 36 (SF-36) data. Less than 50% of patients enrolled in the study met the MCID thresholds for ODI and SF-36. Only 24% of patients requiring revision for non-union met MCID thresholds for SF-36. The findings from this study suggest that the success of revision surgery when analyzed by quality of life improvements is poor. The patient centered outcomes for the success of revision surgery are low for such a costly procedure.
Economic costs are often reported as quality-added life years (QALYs) in healthcare. The concept behind QALYs is based on individuals going through different health states over time with each health state having a different value (Weinstein, Torrance, and McGuire 2009). This figure excludes costs such as lost work days and
15


other costs to the patient. A study by Adogwa and Parker calculated the 2 year comprehensive costs of revision lumbar fusion for pseudoarthrosis. They reported the procedure to only be marginally cost-effective at $118,945 per QALY gained (Adogwa et al. 2015). A survey study conducted by Shiroiwa et al., reported the previous conventional threshold for cost effectiveness of medical intervention, $50,000-$ 100,000 in the United States, is equivalent to the calculated willingness to pay (WTP) per QALY at $62,000 (Shiroiwa et al. 2010). From this reported threshold, the cost of lumbar fusion surgery for pseduoarthrosis is only marginally cost-effective because it is outside of the conventional threshold. Revision spine surgery is costly and only moderately effective. There is a need for further research to understand potential causes of implant failure leading to revision surgery.
One possible cause of implant failure could be metallosis. Metals have been used for orthopedic implants because of their toughness, elasticity, rigidity, and electrical properties (Hanawa, 2000). Many medical implants are made of titanium alloys because of these properties and biocompatibility (R. Bhola et al. 2011). However, the presence of foreign metallic debris in the body can cause local tissue damage and differences in tissue characteristics, called metallosis (Lohmann et al. 2014). The symptoms associated with metallosis are non-specific, ranging from a systemic response to asymptomatic (Oliveira et al. 2015). Metallosis is found during the revision surgery incidentally. This occurs because of the inconsistency of patient symptoms and the difficulty identifying this phenomenon through imaging techniques. The lack of clinical relevance could explain why this phenomenon is often used to describe the appearance of the tissue instead of a cause for implant failure leading to revision surgery.
16


Figure 1. (a) Intraoperative image of metallosis from a revision spine surgery, (b) Removed pseudotumor with metal staining from the same revision surgery (photos courtesy of Reed Ayers, PhD.).
(a)
(b)
In an attempt to document this under reported complication a systematic review, conducted by Goldenberg and Tee, found only three relevant articles regarding spinal metallosis (excluding all arthroplasty related studies) (Goldenberg et al. 2016). The patients included in these case reports displayed neurological symptoms after their initial surgery, leading to a revision procedure to relieve symptoms. Takahashi and Delecrin reported on two cases of intraspinal metallosis in symptomatic patients with spinal instrumentation (Takahashi, Delecrin, and Passuti 2001). They found stained, granulated tissue masses, described as metallosis that had formed in the spinal canal adjacent to the instrumentation leading to neurologic symptoms. This study introduced a possible cause of delayed neurologic symptoms due to metallosis. A case study by Tezer et al., identified an intraspinal metalloma due to crevice corrosion on stainless steel instrumentation leading to neurological symptoms. After removal of the instrumentation, the patient had complete symptom resolution at the three month checkup appointment (Tezer et al. 2005). These three case studies identify an existing issue and provide evidence for metallosis in spine, but do little to determine mechanisms.
17


Metallosis is an underreported phenomenon in spine surgery that is not well understood as evidenced by the lack of literature. It can cause neurological and nonspecific symptoms that may be alleviated with revision surgery and removal of metal instrumentation. Revision surgeries are only moderately successful, when measured by patient centered outcomes, and time is a factor contributing to patient success and rate of revision surgeries. Further, revision surgeries often require re-implantation of metal devices. There is an immediate need to understand the cause of metallosis in spine, starting with the metals used for orthopedic instrumentation.
Corrosion
Corrosion is an electrochemical process resulting in the cleaving of the chemical bonds that hold metal ions together (Lieberman 2014). This process of corrosion can weaken the implant leading to implant failure and the release of metal ions causing adverse reactions in the body (Lieberman 2014). Metals used in medical devices are coated by an oxide layer that acts as a barrier of protection between the metal surface and the body (Rahul Bhola and Mishra 2012).This layer coating the metal can be represented as an electrical circuit based on the composition of the oxide layer (R. Bhola et al. 2011).
CPEp R, CPE
V-FXq, 1 Rh R*
R
(a) (b)
Figure 2. Circuit drawings representing a (a) multi-layer oxide and (b) a single-layer oxide (used with the permission of Dr. Rahul Bhola).
This passive oxide layer will reform in solution or when exposed to air and electrolytes (R. Bhola et al. 2011), (Rahul Bhola and Mishra 2012). If this layer remains intact there
18


should be no corrosion on the implant, however, any imperfection in the oxide layer leads to corrosion of the underlying surface until this layer can reform through passivation.
Pitting corrosion is the most severe form of corrosion leading to a toxic release of metal ions. Corrosion is initiated by a chip or defect along the oxidative layer coating the metal implant (Lieberman 2014). Crevice corrosion occurs because of the construction and geometry of the metal. The corrosion is localized in the areas surrounding the crevice geometry like a weldment, bolted parts, or two interfacing pieces that can lead to the creation of an ionic gradient altering the localized area (Lieberman 2014). Galvanic corrosion is the result of two different metals forming a difference in electrochemical potential (Lieberman 2014). Fretting wear occurs due to repetitive micromotion when a load is imposed on them (Lieberman 2014). These types of corrosion and wear can be identified on retrieved orthopedic implants due to the constant load, various pieces used in the construction of the implant, and the possible damage created by surgical implantation and wear.
A retrieval study conducted by Kirkpatrick et al., identified three common modes of corrosion and wear that appear to happen simultaneously in-vivo; fretting wear crevice and galvanic corrosion (Kirkpatrick et al. 2005). In a retrieval study conducted by Villagraga et al., wear and corrosion were the most common types of damage seen on retrieved spinal hardware and they concluded that revision spine constructs contribute to this type of damage because of the additional segments and mobile pieces added (Villaragga et al. 2006).
19


(a) (b)
Figure 3. Evidence of fretting (a) and pitting (b) corrosion on retrieved spinal instrumentation made of Ti.
The motion created by the load on spinal constructs and the motions between multiple segments provide a source of mechanical damage to the oxide. The term mechanically assisted corrosion was coined by Gilbert et al., and describes the effect of motion between parts of a construct that lead to the initial damage of the oxide layer followed by other modes of corrosion (Gilbert, Buckley, and Jacobs 1993). The motion between various components of a construct damages the oxide layer leading to the release of metal ions into the body. This cycle of fracture to the oxide layer, repassivation, depletion of O2 in the surrounding environment, decrease in the pH of the surrounding environment, instability of the oxide, results in the exposure and corrosion of the underlying metal (Jacobs 2016). A retrieval study of 16 modular hips conducted by Kop and Swarts concluded that corrosion is most commonly seen at the interface between the pieces of the hip (Kop, 2009). Other sources of damage to the oxide layer can be due to manufacturing defects or damage during the surgical procedure. An in vitro study using rabbits by Mu et al., found the main sources of metal ion release to be handling during surgery, implantation, and wear and fretting during the time the metal was in the body
20


(Mu et al. 2002). The damage to the oxide layer can occur from mechanically assisted corrosion or handling during surgery leading to corrosion in the body.
It is well documented that corrosion occurs on any metal introduced to the body (Lieberman 2014), (N. Hallab, Merritt, and Jacobs 2001), (Cousen and Gawkrodger 2012), (Frigerio et al. 2011). Del Rio et al. conducted a study comparing the serum and urine levels of Ni and Cr as measured by atomic absorption spectrophotometry of three groups: no metal instrumentation; metal instrumentation and no radiographic signs of corrosion; and, metal instrumentation with radiographic signs of corrosion. The results showed that participants with metal in their body had elevated serum and urine Cr levels due to corrosion (del Rio, Beguiristain, and Duart 2007). A study by Hallab, Jacobs, and Skipor reported that metal ion products released from the corroding metal implant remain in the serum of the patients (N. J. Hallab et al. 2000). Using three patient groups they were able to determine metal ion concentrations between controls, patients with CoCr alloy implants, and patients with Ti alloy implants (N. J. Hallab et al. 2000). They found elevated concentrations of metals in the serum of patients with metal implants bound to proteins (N. J. Hallab et al. 2000). This study emphasized that corrosion happens in biological systems. Corrosion of metals in the body has been the focus of several studies that identified a number of biological reasons for this occurrence (del Rio, Beguiristain, and Duart 2007; Takahashi, Delecrin, and Passuti 2001; Rahul Bhola et al. 2011).
The body is a hostile environment. Serum and interstitial fluid chloride ion concentrations are 113 and 117 mEq respectively, making these environments very corrosive to any metal instrumentation introduced (Hanawa 2002). Takahashi reported on the normal process of the body once the metal is introduced and noted that the
21


construct is first surrounded by blood and interstitial fluid (Takahashi, Delecrin, and Passuti 2001). Once the metal is placed in the body it is surrounded by active metabolites, like chloride ions, that begin attacking the surface, pulling metal ions from the construct. There are other metabolites in the body that are known to cause corrosion of metal instrumentation. Active oxygen species produced by macrophages during the process of phagocytosis are known to induce the corrosion of metal instrumentation (Hanawa 2002). A study conducted by Mu et al. implanted cpTi constructs into rabbit tibias to identify the sources and causes of metal ion release (Mu et al. 2002). They noted that even the positive controls, where the implant was placed in the muscle tissue, with no obvious signs of wear or fretting still released metal ions into the surrounding tissue due to active oxygen species and other biochemical factors (Mu et al. 2002). The in-vivo environment has been shown to corrode metals in the body. The consequences and functions of these corrosion products in the body has been another area of research.
Metal Ions in the Body
Several studies have been conducted to identify normal concentrations of the metals used in metal implants in serum and muscle tissue. For individuals that have never been exposed to Co and Cr serum levels should be < 10 nmol/L and < 40 nmol/L respectively as reported by the UK SupraRegional Assay Service (SAS) (Bradberry, Wilkinson, and Ferner 2014). Recent MHRA guidelines report an appropriate level of Cr or Co to be 7 mg/L (Gill et al. 2012). Other metals such as Al, Mo, and, V have reported normal concentrations in muscle tissue at 8.4 4.8 ug/g, 1.61 1.41 ug/g, and 0.06 0.03 ug/g respectively((Vincent 2000), (Pasha et al. 2008), (Farah et al. 2010), (Harrington et al. 2014)). Only whole blood measurements have been reported for
22


normal Ti levels 0.00072 ug/g 0.1412 (Farah et al. 2010). At normal levels these metal ions are inert or benefit the body but at elevated levels there are several reactions and issues that arise.
Currently there is no known threshold or specific elevated metal ion level leading to a reaction in the body. A techniques study conducted by Hanawa, concluded that particles released from an implant in 100 ppb are toxic to the surrounding cells (Hanawa 2002). Hanawa went on to say that even trace elements in metal alloys cannot be overlooked as potential causes of patient symptoms (Hanawa 2002). There have been several studies focused specifically on Co and Cr toxicity. Bradberry et al., reported that patients with blood Co concentrations above 250 ug/L displayed peripheral neuropathy after a CoCr MoM hip arthroplasty (Bradberry, Wilkinson, and Ferner 2014). Other systemic issues were identified through this study and attributed to Co toxicity. In an in vitro study by Gill et al., Co and Cr were found to be mutagenic and genotoxic, damaging DNA in a variety of ways (Gill et al. 2012). The damaging effects of Co and Cr have raised concern over their safety and use in orthopedic implants. Ti is one of the most commonly used metals in orthopedics, yet not much is known about its effect in the body. A retrieval study by Lukina et al., found a 1,500 fold increase in Ti tissue concentrations in pediatric patients with sliding cpTi constructs when compared to patients with no metal implants (Lukina et al. 2016). This value is much higher than any previously reported findings. The concern of these elevated metal ion concentrations in the body as a result of orthopedic implants has led to several studies that aim to understand how the body processes these foreign particles.
23


The body has certain mechanisms to protect itself from foreign particles. The response and specific pathway can vary based on things like size, charge, and quantity of the particle. An in vitro study conducted by Yue et al., analyzed the cellular pathways used to internalize and digest metal particles of varying sizes. Phagocytosis was a crucial pathway for internalization of metal particles of all sizes, nanoparticles were digested in the lysosomal pathway, and micro particles were sequestered in the cytoplasm for digestion (Yue et al. 2010). Gill et al. expanded on this information explaining the cycle of phagocytosis, cell death, and release of cytokines in response to the metal particles could lead to the formation of a pseudotumor (Gill et al. 2012). The body tries to digest metal particles of all sizes yet chronic inflammation and pseudotumor formation still persist. Another study looked at the particular charge and shape of metal ions that allow them passage through receptors on the cell surface and accumulate in the cytoplasm.
This movement through nonspecific channels is described as quick and could lead to cell death due to mutagenic and genotoxic effects of specific metal ions (Gill et al. 2012).
The vicious cycle activated by the presence of foreign particle can be summarized as,
the activation of the osteolysis pathway, immune exhaustion, and chronic inflammation (Nadim J. Hallab, Cunningham, and Jacobs 2003). The exact pathway activated is variable and inconsistent and the amount of metallic material present could influence the physiological reaction. Lohman et al., found that the type of tissue response correlated with the concentration of metal ions,Co, Cr, and Ni, in the tissue (Lohmann et al. 2014). Mean metal concentrations of 222.2 52.9 ug/g displayed a strong lymphocytic tissue response while lower metal concentrations, 3.00.9 ug/g exhibited a macrophage response. These immune reactions can be observed in any healthy body yet only a
24


fraction of the population experiences complication such as metallosis with orthopedic implants.
As previously stated, there is no known concentration threshold of metal ions that trigger an immune response in the body. Lohman et al., hypothesizes that the bodys ability to clear the particles through the lymphatic system explains why this phenomenon does not occur in everyone with a metal implant (Lohmann et al. 2014). The amount of wear and corrosion particles coupled with the bodys ability to clear and digest this foreign material could be a potential cause of implant failure due to metallosis. Further research is necessary to validate this hypothesis and demonstrate the bodys clearing ability in-vivo. This hypothesis, similar to others, fails to address the cause of the reaction. Current research in the fields of metallosis and orthopedics has settled on two possible hypotheses for implant failure due to metallosis, focused on the patient physiology.
Metal Hypersensitivity
Metal hypersensitivity has been a topic of interest in orthopedics with the focus to characterize the adverse effects seen in some patients with orthopedic implants. Approximately 10% to 15% of the general population displays some type of dermal sensitivity to metal (N. Hallab, Merritt, and Jacobs 2001; Cousen and Gawkrodger 2012). Approximately 25% of patients with properly functioning implants and 60% of patients with implant failure show some type of metal sensitivity (Frigerio et al. 2011). While much is known about the dermal response to metal antigens, determining metal hypersensitivity prior to the implant procedure or implant failure due to metal hypersensitivity has proven to be a challenge in clinical settings.
25


Type IV cell-mediated response
The cellular response to this release of metal ions has been described as delayed Type IV cell-mediated responses (N. Hallab, Merritt, and Jacobs 2001). Type IV cell-mediated responses are characterized by the activation of T lymphocytes that release cytokines signaling for the recruitment of macrophages (N. Hallab, Merritt, and Jacobs 2001). The lymphocyte response has been linked to poor implant performance; yet this cause and effect relationship remains controversial (Frigerio et al. 2011). The assumption that metal hypersensitivity leads to implant failure has many in the field using clinical tests for allergies to solidify this relationship.
Metal Hypersensitivity Testing
Three tests exist clinically to determine metal sensitivity in patients. Patch testing in-vivo (N. Hallab, Merritt, and Jacobs 2001), lymphocyte transformation testing in vitro (LTT) (N. Hallab, Merritt, and Jacobs 2001), and leukocyte migration inhibition testing (LIF or MIF) are the current tests for determining metal sensitivity (N. Hallab, Merritt, and Jacobs 2001).
Patch testing. Patch testing is not an effective method to predict a stable or failed implant (Cousen and Gawkrodger 2012). Several concerns and issues surround this technique as a clinical determinant of metal hypersensitivity. The test is designed for short or acute exposures, unlike the long term, constant exposure seen with orthopedic implants (N. Hallab, Merritt, and Jacobs 2001). There is also concern with the accuracy of the test because of the difficulty knowing and acquiring appropriate metal agents to be used as antigens (N. Hallab, Merritt, and Jacobs 2001). Finally, patch testing involves sensitizing a previously insensitive patient (N. Hallab, Merritt, and Jacobs 2001). Second
26


exposure to the antigen, during the implantation of the orthopedic device, may trigger the bodys immune response leading to hypersensitivity.
A study by Elves, Wilson, Scales, and Kemp patch tested 50 patients who had a Total Joint Replacement (TJR) inserted anywhere from one to twelve years prior (Elves et al. 1975). Of the 50 patients, 19 reacted positive to the patch test and of the 26 patients with a failed implant only 15 were metal sensitive. They concluded that a significantly increased incidence of metal sensitivity appeared in patients with metallic implants (Elves et al. 1975). While this data may be convincing to some, questions about the methodology of this study remain. The challenge agents used were metal salts including nickel sulphate, cobalt chloride, and titanium oxalate. No one has yet to determine adequate challenge agents or a series of tests to determine metal sensitivity due to orthopedic implants (N. Hallab, Merritt, and Jacobs 2001). The next issue is that all of the patients have a metallic implant and are therefore already sensitized to metal. Second exposure from the patch test could be a reason for positive results. In a prospective study of 100 patients conducted by Frigerio, Pigatto, and Guzzi, similar results were reported (Frigerio et al. 2011). No participant showed signs of a metal allergy prior to surgery and all had undergone either a TKA or THA (Frigerio et al. 2011). Challenge agents used for patch testing included metal salts and were read 4 days later (Frigerio et al. 2011). One year after surgery 72 of the patients were brought back for postoperative testing (Frigerio et al. 2011). Results of the study showed a 6.5% increase in metal sensitization postoperative (Frigerio et al. 2011). The design of this study sensitized all participants to metal prior to surgery. This flaw in the study design could have provided researchers with several false positives or negatives upon retesting. Patch testing does not provide a
27


clear cause and effect relationship between the metal implant and implant failure because of the lack of appropriate challenge agents and the risk of sensitization upon secondary exposure.
Lymphocyte transformation testing. LTT measures the proliferation of lymphocytes following activation from a specific antigen in vitro (N. Hallab, Merritt, and Jacobs 2001). This test uses radioactive markers incorporated into lymphocyte DNA to quantify the proliferative response (N. Hallab, Merritt, and Jacobs 2001). In the study conducted by Frigerio et al., 12 of the participants were screened with patch testing and LTT (Frigerio et al. 2011). The small sample size they enrolled did not permit any significant findings to be reported however, they were able to conclude that LTT can provide some additional information (N. Hallab, Merritt, and Jacobs 2001; Frigerio et al. 2011)]. While this in vitro test does not sensitize the patient, like patch testing, little information surrounds its credibility in determining metal hypersensitivity.
Leukocyte migration inhibition testing. This in vitro test uses sensitizers, like metal ions, to measure the migration activity of a mixed population of leukocytes in culture (N. Hallab, Merritt, and Jacobs 2001). Alone, leukocyte testing is not good for identifying metal sensitivity (N. Hallab, Merritt, and Jacobs 2001). Findings from this test when paired with other information may provide some clinical relevance.
Type IV metal hypersensitivity does not completely explain the phenomenon of metallosis and with no ability to predict the occurrence of hypersensitivity, this definition appears to be clinically irrelevant. The FDA states that it is not possible to predict which patients will experience adverse reactions to MoM implants (Health 2015). Recently a new term has been used to describe this reaction through histological analysis.
28


Aseptic Lymphocyte-Dominated Vasculitis Associated Lesion
Aseptic, lymphocyte-dominated vasculitis-associated lesion or ALVAL, is now considered as a possible explanation for implant failure. This reaction is characterized by a dense perivascular inflammatory infiltrate that is only identified upon histopathologic analysis (Watters et al. 2010). In a case series conducted by Watters et al., the histological evidence did not support the hypothesis of a classic allergic reaction. The cells identified through histology are non-specific making it difficult to conclude any particular reaction is occurring (Watters et al. 2010). It is believed that the delayed-type hypersensitivity reaction is the cause of ALVAL yet evidence to support this has yet to be discovered. Research has not yet confirmed if the type IV metal hypersensitivity causes the failure of the implant or if the failure of the implant leads to metal hypersensitivity (Watters et al. 2010). This hypothesis does not provide sufficient clinical relevance because it cannot be identified prior to surgery. Diagnosis is only made upon histological analysis. Another issue with this hypothesis is the absence of a cause and effect relationship. Research has yet to determine why this is occurring or the order of reactions that takes place prior to implant failure. Alternative hypotheses, that address the cause and effect relationship, need to be established to better understand the phenomenon of metallosis and improve patient care and outcomes.
Electrochemical voltage shifts created as a result of implant corrosion
Corrosion in-vivo leads to electrochemical voltage shifts that may negatively impact the biocompatibility and effectiveness of the orthopedic implant (Haeri et al. 2012). When deviations from physiological voltages occur cell adhesion is impaired leading to implant loosening and decreased cell survival. Cell death can be initiated by
29


alterations in the movements of ions across cell membranes, the function of voltage-gated channels, and other pathways within the cell (Haeri and Gilbert 2013). Metal ions released through the process of corrosion generate a voltage and electrochemical changes that possible lead to altered cell physiology, apoptosis, and necrosis (Haeri et al. 2012).
An in vitro study conducted by Haeri, Wollert, and Langford identified a voltage viability range (Haeri et al. 2012). Using CoCrMo alloy plates as the working electrode they were able to determine the cell viability range from -300 mV +300 mV. Deviating outside this range they observed adverse cell reactions. Cathodic voltages led to cell apoptosis while anodic voltages led to cell necrosis (Haeri et al. 2012). The cause of this cell necrosis has been identified through another in vitro study by Haeri and Gilbert using time-lapse live-cell imaging to monitor the morphological changes of the cells (Haeri and Gilbert 2013). The corrosion products and reduction reactions occurring on the surface of the implant cause voltage changes in the surrounding tissue leading to cell death and implant failure (Haeri and Gilbert 2013; Haeri et al. 2012). These findings can be supported by histological studies. Watters et al., identified a necrotic layer of cells in 2 of the 3 patients in the study with failed MoM hips (Watters et al. 2010). Apoptosis is not easily identified through histological analysis.
A similar experiment using cpTi as the alloy component revealed similar outcomes. Ehrensberger et al., concluded that voltages from -lOOOmV to -600mV negatively impacted cell viability while voltages from -300mV- lOOOmV seemed to have little impact on cell morhpology (Ehrensberger, Sivan, and Gilbert 2010). In a case study conducted by Denaro et al., showed that a 12.1 X 10'6 T continuous EMF generated around Ti alloy hardware will inhibit the activity and proliferation of osteoblasts (Denaro
30


et al. 2008). Reduction reactions occurring on the surface of the metal can also influence the local tissue.
The oxide layer is another crucial piece to this reaction. Electrochemical properties and thickness of the oxide layer change through the process of corrosion and repassivation. The variability in the oxide layer can affect the cells that interact with the metal by changing the electrochemical properties along the metal surface (Ehrensberger, Sivan, and Gilbert 2010). The byproducts of corrosion and subsequent chemical reactions provide an explanation for the symptoms and reactions seen clinically. Cell viability is affected by the voltages generated from corrosion byproducts and the reduction reactions occurring on the surface of the metal that can potentially lead to implant failure.
Conclusion
Research has yet to identify the cause and effect relationship of implant failure potentially due to metallosis. Metal hypersensitivity has many flaws in its clinical testing and a strong correlation between implant failure and hypersensitivity has yet to be proven. Corrosion is known to occur when metals come into contact with the body and the release of these metal ions leads to electrochemical voltage shifts and reduction reactions that lead to changes in cell morphology and function resulting in cell death. This relationship between metal corrosion and altered cell physiology leading to implant failure requires further research to improve patient care and outcomes.
31


MATERIALS AND METHODS
Hypothesis
Corrosion products from orthopedic spine implants alters the physiologic environment of local soft tissue, by generating a voltage, leading to cell necrosis, identified through histology; propagating a pathological response that may lead to poor health outcomes.
Specific Aims
Specific Aim One: To collect two tissue samples from revision patients with and without evidence of a solid fusion and hardware failure; one sample adjacent to the metal implant (peri-hardware tissue) and one sample from either the vertebrae above or below the revision (control tissue). Tissue samples will be collected from peri-hardware and control areas during the revision surgery at the University of Colorado Hospital by Orthopedic Spine Surgeons. Control and peri-hardware samples will be stored in separate containers prior to sample preparation.
Specific Aim Two: To perform non-destructive and destructive analysis on the explanted instrumentation. All instrumentation will be photographed, and if needed, analyzed under a scanning electron microscopy as well as energy dispersive spectroscopy to identify regions of corrosion and wear. Physical metallography will also be used for this analysis.
Specific Aim Three: To perform ICP-MS/AES on the collected tissue samples. At least two grams, but no more than 10 grams, of the peri-hardware and control samples will be stored in sterile sample jars and sent to Huffman
32


Hazen Laboratories for analysis. These samples will be utilized to determine the presence and quantity of metal ions in the sample.
Specific Aim Four: To correlate the results of this research to clinical data acquired through the medical records and to prior in vitro studies. ICP-MS/AES results will be correlated to data acquired through the medical records. In another analysis, the results of ICP-MS/AES will be used to calculate the voltage generated by the instrumentation in-vivo and compared to prior in vitro studies.
33


Materials Specific Aim One
2 specimen containers
Sharpie
Methods Specific Aim One
This protocol has been reviewed and accepted by COMIRB, protocol ID number: 16-1176.
Patients to be enrolled in the study will meet the following inclusion and exclusion criteria:
Inclusion Criteria
Patients undergoing revision spine surgery with previous instrumentation in the thoracic or lumbar region with evidence of catastrophic hardware failure, pseudoarthrosis, implant loosening and non-fusion.
Age 18-85
* These inclusion criteria will be determined through radiographic and medical record review. While these criteria indicate a higher chance of observing metal staining it does not guarantee that it will be present. If the patients consent and do not have metal staining we will still proceed with our research as metal staining is not the only characteristic of interest. Metal ion concentrations in the peri-hardware tissue can be elevated in the absence of tissue staining and is therefore beneficial to this research.
Exclusion Criteria
Patients undergoing revision spine surgery without spinal implants
34


Patients undergoing revision spine surgery that are not suspected of having metal staining
Patients with an identified infection
Patients unable to provide consent
Pregnancy
Patient Population and Recruitment
Patients will be recruited through the Spine Center at the University of Colorado Hospital. Patients undergoing revision spine surgeries with evidence of catastrophic hardware failure, pseudoarthrosis, implant loosening and non-fusion with previous instrumentation will be offered enrollment. The enrolled patients will be identified prior to surgery by the surgeons and research staff through radiographs and images contained in the medical records. If they meet the inclusion and exclusion criteria, informed consent will be obtained by a member of the research staff.
Consent and Enrollment Procedure
The consent form will be administered by either the surgeon or research staff that have been trained in local IRB requirements prior to revision surgery. Patients that meet inclusion criteria will receive a phone call from the surgeon or research staff to schedule a time to discuss the research study and consenting process. If the patients agree a meeting will be set where they will have the purpose and methods of the study explained to them as well as risks and benefits. Consent will be obtained in a waiting room or office prior to surgery at the University of Colorado Hospital. They will receive a copy of the informed consent and be given time to review the document. The individual administering the consent will review it with the subject and they will determine if they
35


demonstrate adequate understanding to consent. The subject will be expected to explain the study purpose in his or her own words. Once consented, patients will be assigned a randomized and unique study identification number. After enrollment, patients will proceed with their scheduled surgery as planned.
The following information will be collected from the patients medical records:
Clinical Patient ID
Diagnosis
Indication for revision
Preoperative observations
Previous surgical intervention
Previous surgical instrumentation
Previous surgery type and number and previous implants used
Current laboratory test collected, including CBC, ESR, CRP, and Serum Ion levels
Age
Gender
Demographics
Pain
Radiographic imaging
Medical history
Pathology results
Histology results
Bacterial culture results
36


As part of the standard of care, tissue from the peri-hardware and control locations are normally removed through the course of surgery. A histological analysis is part of the standard of care for revision surgeries. The remainder of the tissue is normally discarded. We will collect this normally discarded tissue to perform ICP-MS/AES analysis. The normally discarded peri-hardware tissue and the control tissue will be collected at the time of the surgery. No more than ten grams (approx. 10 cm3) of the peri-hardware tissue, that is normally discarded, will be collected from the local area around the orthopedic implants. No more than ten grams (approx. 10cm3) of the control tissue, normally removed in the course of surgical exposure, will be collected from the subcutaneous and muscular tissue at least 1 spinal motion segment above or below the previous instrumentation placement. In the operating room, on a non- sterile surface covered by a medium drape, these samples will be separated into two containers immediately labeled with the study ID number, either the word control or peri-hardware and the destination of the sample. At least a two cubic centimeter sections of the control and peri-hardware samples will be placed in separate containers with the appropriate labelling and the destination Huffman- Hazen laboratories for ICP-MS/AES analysis. Per the standard of care tissue samples are sent for pathology and histology analysis. The results of pathology and histology will be accessible to the research staff through the medical records because this analysis is part of the standard of care.
Materials Specific Aim Two
70% Isopropyl Alcohol
4x4 gauze pads
Sharpie
37


LED Loupe
Nitrile Gloves
Medium Drape
DSLR Camera Methods Specific Aim Two
All explanted hardware will undergo non-destructive and destructive testing. For the nondestructive testing, photo-documentation of all components prior to removal, as well as after removal, in the as received condition will be taken. The research team will wear Nitrile gloves while handling the hardware. The hardware removed during the surgery will be placed on the nonsterile surface covered by the medium drape and cleaned using 4x4 gauze pads and 70% Isopropyl Alcohol. An initial examination of the hardware will be done in the operating room using the LED Loupe. The hardware will then be placed in a specimen container, labelled with the study ID number and the word Hardware. The sample will be taken for further testing. Using the DSLR camera the hardware will be reimaged and saved onto a password protected computer. In a certified lab, we will conduct optical microscopy and scanning electron microscopy on areas of interest as well as energy dispersive spectroscopy (EDS) for elemental composition estimates under the direction of the research staff. Destructive testing will include physical metallography.
Materials Specific Aim Three
Peri-hardware and Control Tissue Samples
Results from Huffman Hazen Laboratories
Laptop with Excel
38


Methods Specific Aim Three
Inductively coupled plasma mass spectroscopy atomic emission spectroscopy (ICP-MS/AES): The minimum two gram tissue samples will be sent to Huffman Hazen Laboratories for ICP-MS/AES analysis. Sample containers will be prepared by Huffman. The tissue sample will be placed in the specified sample container label only with the assigned patient ID and control or peri-hardware. Samples, method blanks, and method blank spikes will be digested in entirety with mixed acids including reflux with boiling perchloric acid to ensure complete oxidation of all organic material, plus hydrofluoric acid to insure solubilization and stabilization of titanium. Digestion solutions will be further diluted and analyzed by ICP-AES and ICP-MS. Metal concentrations will be reported on the as received weight basis and the dried sample weight basis. The process of ICP-MS/ AES will completely destroy the tissue sample.
Results from ICP-AES and ICP-MS will be formatted into an excel spreadsheet. Control data will be pulled from previous literature.
Materials Specific Aim Four
Computer
Excel Software Methods Specific Aim Four
The voltage generated by the corrosion products will be calculated using the Nernst equation, results from ICP-MS/AES, and normal ion concentration values found in the literature. For this model there are several assumptions that have to be made. This equation is only looking at voltage and therefore it is assumed that the charge carriers are moving independently of cells and tissue. For simplicity we assume no chemical
39


reactions are occurring between the charge carriers and physiological environment and we neglect the metal ions that would be lost due to repassivation. Finally this is assumed to be a localized response. This simple corrosion cell models the reaction that is assumed
in this analysis.
Metal Surrounding
Instrumentation Tissue
(Anode) (Cathode)
o2OHcr
Figure 4. Schematic representation of the corrosion cell generated in vivo. As the metal instrumentation corrodes electrons, followed by positively charged metal ions, flow to the tissue which acts as the cathode. Negatively charged species, like oxygen and hydroxide, then move back to the metal instrumentation acting as the anode.
The values used for z, valency, are the most common oxide form of the ions. R is the universal gas constant. T is body temperature in Kelvin. F is Faradays constant. The Nernst equation uses a ratio of ion concentration outside of the cell and inside the cell to determine voltage. In this set up we will assume that the metal ion concentrations collected from ICP-MS/AES data are corrosion byproducts from the metal implant and are therefore extracellular. All known values of metal ions found in the literature are assumed to make up the intracellular concentrations. Normal values for Al, Mo, V, Co, content in the surrounding tissue, are 8.4 4.8 ug/g; 1.61 1.41 ug/g; 0.06 0.03 ug/g; 1.35 1.97 ug/g respectively (Vincent 2000; Pasha et al. 2008; Farah et al. 2010; Harrington et al. 2014). A normal reference titanium level could only be found for whole blood and that was 0.00072 ug/g (Sarmiento-Gonzalez et al. 2008).
40


V =
R *T

[extracellular metal ion concentration]
)
z F
[intracellular metal ion concentration]
Figure 5. The Nernst equation used for this calculation. R, T, and F are all know constants. The valency, z, used is the most common oxide state of each metal. The extracellular metal ion concentrations are the values from the experimental data. The intracellular metal ion concentrations are pulled from previous literature.
Patient demographic data will be summarized into a table including, gender, age, duration of previous implant, location of the revision procedure, reason for revision based on inclusion criteria, and the presence or absence of metal staining. Mean age, duration in situ, and length of revision procedure (measured by the number of vertebral levels included in the surgery) will be calculated from this table.
All retrieved hardware will be documented and the type of metal and types of corrosion and wear identified will be summarized in a table. This summary will include the presence or absence of corrosion and wear. A quantity or measurement of corrosion and wear will not be reported.
Histological slides will be analyzed and a gross description of the tissue will be recorded. When applicable the quantity of lymphocytic infiltrate, types of inflammation, and quantity of necrosis will be reported. A description of the common themes seen in the histological slides will be presented.
41


RESULTS
Nine patients undergoing revision spine surgery in the lumbar and/or thoracic region were enrolled and consented into this study. Patients were enrolled and consented following the procedure defined in the approved IRB protocol ID 16-1176. In the study, 6/9 patients were male. The average age of participants in the study was 66 12.7 years. The duration of previous instrumentation in situ ranged from 6 weeks to 17 years with an average duration of 5.03 5.7 years. The size of the revision procedure, as measured by number of vertebral levels revised, ranged from 2-15 levels with an average length of 6 levels. Patients displayed a combination of indications for revision based on the inclusion criteria. All patients had pain of the spine, 3 patients had pseudoarthrosis, 3 patients had broken hardware, 2 patients had non-union, and 1 patient had loose hardware. In two of the patients with broken hardware the fractured instrumentation was not removed as it was part of a larger construct that did not need to be revised. This information is summarized in Table 1.
Table 1. Patient demographic information including gender, age, duration of previous hardware in situ, location of the revision, and indications for the revision based on the inclusion criteria.
Gender Age Duration in situ (in years) Location of revision Indications for Revision Metal Staining
M 39 2.5 L4-S1 Pain N
F 64 .115 T4-Pelvis Pain N
F 75 12 C2-T4* Pain Y
F 58 2 L4-S1 Pain N
M 77 17 T4-L1 Pain, Hardware Failure N
M 82 5 T 10-Pelvis Pain, Hardware Failure, Non-union, Pseudoarthrosis Y
M 70 .6667 L5-S1 Pain, Non-union, Pseudoarthrosis, Loosening of the instrumentation Y
M 67 3 L4-Iliac Pain N
M 62 3 L4-S2 Pain, Hardware Failure, Pseudoarthosis Y
42


*Hardware was removed from T4-T8 and the thoracic spine was part of the revision surgery therefore the inclusion criteria for the study were met.
All explanted hardware was documented and photographed. The type of metal and types of corrosion and wear were identified through visual inspection. Information about the retrieved hardware is summarized in Table 2. All explanted hardware was made from cpTi or Ti6A14V with one set having CoCr tulips on the screws. All of the hardware showed evidence of fretting wear, 8/9 samples showed evidence of pitting corrosion, and 8/9 showed evidence of the degradation of the oxide layer followed by repassivation as evidenced by discoloration of the oxide layer. Images of each mode of corrosion and wear are found in Fig 5.
Table 2. From the retrieved instrumentation the type of metal and types of corrosion and wear seen on the hardware through visual inspection were recorded. Only the presence or absence of a type of corrosion or wear was recorded._______________________________
Type of Metal Pitting Fretting/ Wear Spalling/ Galling Crevice Galvanic Degradation of the Oxide Layer / Repassivation Fracture
64Ti (screws) cpTi (rods) Y Y Y N N Y N
64Ti Y Y N N N Y N
64Ti Y Y N N N Y N
cpTi rods CoCr head and cpTi screw Y Y N N N N N
64Ti Y Y N N N Y N
64Ti Y Y N N N N Y lowest level screws fractured due to corrosion fatigue
64Ti Y Y N N N Y N
64Ti N Y N N N Y N
Ti Y Y N N N Y N
43


(c) (d)
Figure 6. (a) Evidence of pitting corrosion, (b) Fretting wear has the shiny appearance along the rod caused from micromotion between the connector and the rod. (c) Evidence of repassivation. The discoloration along the screw indicates reformation of the oxide layer at a different thickness, (d) Fracture between the threading and the head of the screw.
Histological slides were prepared for eight of the nine patients in the study. This lab is ordered by the surgeons and can be omitted if adverse reactions are not suspected or no abnormal tissue seen during the revision procedure. Results of this analysis were accessed from the patients medical record then reviewed with the pathologist. From the histological analysis quantity of inflammatory infiltrate was determined for two patients. Approximately 1-5% of the tissue contained inflammatory infiltrate with the appearance of chronic inflammation. The quantity of necrotic tissue was measured for two patients and ranged from 0.5-75% of the tissue sample. The results of the analysis revealed two common themes; foreign body giant cell reactions and histiocytes trapped in fibrous tissue (Fig 4). These observations are non-specific. The presence of particles and debris
44


was not visible in all of the samples; however, some foreign debris could have been removed during the processing and preparation of the histological slide.
Figure 7. All images are of hematoxylin and eosin stained slides taken at 400x magnification. A) Foreign body giant cell reaction. B) Particle laden histiocytes trapped in fibrous tissue. C) Necrotic skeletal muscle and necrotic fibrovascular connective tissue. D) Typical look of metallosis.
ICP-MS/AES data was collected for all nine sample however results are only available for three samples. The results of ICP-MS/AES are summarized in Table 3. Ti concentrations were elevated 10 -10 from normal serum levels, 0.00072 ug/g, in all three patients with Ti instrumentation (Sarmiento-Gonzalez et al. 2008). Alloying agents such as vanadium (V) were elevated 10-100x from normal lung tissue values of 0.06 ug/g (Farah et al. 2010). The mean voltage calculated using the Nernst equation was 329.6 95.7 mV
45


Table 3. ICP-MS/AES data measuring the metal ion concentrations found in the retrieved tissue samples. Values below the detection of ICP-MS/AES are reported as < 0.1.
As received weight (g) Dried weight (g) Solids (%w/w) Ca (%w/w) P (%w/w) Al (ug/g) Co (ug/g) Cr (ug/g) Fe (ug/g) Mo (ug/g) Ni (ug/g) Ti (ug/g) V (ug/g)
0.6848 0.2450 35.78 0.06 0.17 95 0.3 14 660 0.3 3.9 2024 39
1.8757 1.25645 66.92 0.1525 0.1625 2.75 <0.1 0.725 144.25 <0.1 0.2125 23.5 0.725
1.0051 0.4357 43.35 6.87 3.32 16 <0.1 8.3 364 0.5 4.6 491 5.8
DISCUSSION
Corrosion products from orthopedic spine implants alters the physiologic environment of local soft tissue, by generating a voltage, leading to cell necrosis, identified through histology; propagating a pathological response that may lead to poor health outcomes. While the results of this study did not prove this hypothesis, this work connects material science, engineering, and medicine. The corrosion of metals in vitro is a well-known fact among material scientists and engineers, yet there is no urgency to correct or solve this issue. Identifying a cause and effect relationships between corrosion and altered tissue physiology and associating these findings with clinical data and patient outcomes has created the need to find new materials for orthopedic implants. Based on this research there were trends identified that require further exploration.
Corrosion and wear were identified on all retrieved hardware, indicating a cause and effect relationship between the metal instrumentation and elevated metal ion concentrations in the surrounding tissue. All retrieved hardware was made of Ti alloy. The claims that Ti is corrosion resistant and biocompatible have increased the popularity and use of this alloy in recent years, yet research has proven these claims to be inaccurate (Rahul Bhola et al. 2011; Ehrensberger, Sivan, and Gilbert 2010). Damage to the oxide layer dramatically decreases biocompatibility and increases corrosion of the Ti alloy (Ehrensberger, Sivan, and Gilbert 2010).
46


This study provides a first step in connecting previous in vitro results to clinical applications. The voltage generated by corrosion products calculated using the Nernst equation is within the voltage viability range of Ti alloy, -300mV 1000mV, established by Ehrensberger, et al., therefore the absence of necrotic tissue in the histological analysis is not surprising (Ehrensberger, Sivan, and Gilbert 2010). However, these voltages are elevated 3x the resting membrane potential for muscle cells and could be a contributing factor to the altered physiology of the surrounding tissue (Hopkins 2006).
An abnormal appearance of the tissue was identified in all histological samples. The dense collagenous fibrous tissue, seen in all histological samples, is a pathological response not seen in healthy tissue. Foreign body giant cell reactions and histiocytes trapped in the fibrous tissue, sequestering black material in the cytoplasm, were the most common themes. The presence of these cells indicates a response to foreign material in the tissue. These observations are not described as metal hypersensitivity or ALVAL and no indication of these reactions was identified through this analysis.
The clinical consistency among all nine participants was the symptom of pain.
The variation within patient demographic information reflects the vast population needing surgical intervention and some of the difficulties understanding this phenomenon.
Limitations
The small sample size limited the analysis that could be completed in this study. The large standard deviations in the results can be attributed to the small patient population. Factors that could be used in correlation studies such as weight, height, BMI, comorbidities, serum and urine metal ion concentrations, and CBC were not included. A
47


larger sample size is critical in furthering the potential results of this research and is necessary to establish statistical relevance.
The lack of variation among the retrieved hardware samples limited the connections that could be made to previous in vitro work and the scope of the histological analysis. All retrieved instrumentation was made of Ti alloy which has a different cell viability voltage range than CoCrMo alloy (Ehrensberger, Sivan, and Gilbert 2010; Haeri and Gilbert 2013). The material of the construct may influence the voltage that is generated, therefore impacting the results of the histological analysis. Future research should separate the data by metals used in the construct to observe differences among voltages and tissue samples. A larger sample size and greater variability of hardware composition would be required to ensure statistical and clinical relevance in future research.
Quantifying the amount of corrosion and wear present on removed hardware cannot accurately be obtained without original manufacturing drawings. This factor limited the analysis to qualitative data through visual observation. Obtaining manufacturing drawings is difficult as these are often proprietary documents and may pose an issue in future research. Using different methods of observation with higher magnification to look at factors like grain shape and size would expand the data set to include things such as composition and manufacturing history of the metal.
There was difficulty collecting both the control and peri-hardware tissue samples. The comparative analysis between the two samples could not be completed however, a larger sample size would be necessary to establish statistical significance between these two. This issue was predicted before sample collection began. For future research studies
48


looking to compare the metal ion concentration a tissue biopsy of a patient with no metal instrumentation in their body would be needed for a control. In this study the unknown location of the tissue samples prevented the inclusion of a distance or diffusion factor.
To control for these factors in future research an exact distance from the metal instrumentation and location of the peri-hardware sample would be required.
There are several assumptions that were used to calculate the voltage generated by corrosion products. As an initial test, this simplified model was sufficient to connect in vitro results to clinical applications. Future research should look to more complex models that include a time or distance factor as well as certain chemical reactions that occur between the metal ions and surrounding physiological environment. Metal ions that repassivate were neglected in this model as well as metal ions and particles that are processed and removed from the body.
Future Research
There are several avenues for future research in this field. Takahashi reported that myelograms were the only useful images in predicting metallosis (Takahashi, Delecrin, and Passuti 2001). Developing a radiographic profile of patients with metallosis using x-ray, MRI, and CT images would improve the ability to predict and diagnose metallosis. An MRI profile would be beneficial for determining soft tissue appearance while CT and x-ray would useful for examining the metal, the metal bone interface, and the condition of the bone.
The occurrence of an infection in patients with metallosis is something that was observed throughout this study. The association of infection and metallosis could be useful for the future diagnosis of this physiological response. The altered physiology,
49


tissue fibrosis, and in some cases chronic inflammation indicate that the body is creating a pathological response. Patients with a known infection prior to revision surgery and patients that develop an infection post-operative with signs of metallosis during the revision procedure would be ideal candidates to better understand the relationship between these two reactions.
Other factors that could influence corrosion such as lactic acid build up after surgery, the manufacturing process of the implants, and potential damage to the hardware during implantation need to be researched further. As previously stated, the body is a hostile environment and things other than Cf and metabolites can interact with the oxide layer and underlying metal leading to corrosion. The manufacturing methods, industry standards, and testing methods used to produce and verify the instrumentation influence the integrity of the product. Understanding these methods and advancing them to reflect physiological conditions and real use situations could be beneficial in reducing implant failure. The oxide layer coating the metal instrumentation is a crucial piece of biocompatibility and corrosion resistance. Education about this oxide layer and improvements to surgical techniques could decrease the amount of damage to the hardware during implantation. This knowledge could also benefit future research into new materials for use in orthopedic instrumentation.
Research needs to focus on establishing a better understanding of metal toxicity and the cellular pathways used to process, digest, and defend against metal ions and particles. Bhola et al. stated that the stabilizing agents such as A1 and V in metal alloys could have negative, even toxic, effects on the body (Rahul Bhola and Mishra 2012). Hanawa et al., concluded that even metal particles in concentrations of 100 ppb cannot be
50


neglected when considered patient safety (Hanawa 2002). Metal ion concentration thresholds for metals used in orthopedic implants and the symptoms associated with these thresholds would benefit clinicians, engineers, and material scientists develop new patient diagnoses and new materials to be used in future orthopedic implants. While Yue et al., identified different cellular pathways used to digest particles of varying size an explanation of why and physiological consequences was lacking (Yue et al. 2010).
Results from this thesis, show cells containing metal particles often trapped in the fibrous tissue near the implant indicating that the normal process of digesting foreign particles is not possible. In vitro studies that investigate the process, limitations, and secondary defenses against these foreign particles could establish the basics of the physiological reaction seen from clinical samples. This information could also support or contradict the current hypotheses of type IV metal hypersensitivity and ALVAL.
51


CONCLUSION
The results of this study identified trends among patients undergoing revision surgery with previous instrumentation. The primary complaint of pain was consistent for all nine participants. Of the three patients that had ICP-MS/AES results, elevated metal ion concentrations and elevated voltages, calculated using the Nernst equation, were unanimous. These results build upon previous literature that established a cell viability voltage range for cells on metallic alloys. Trends identified through histological analysis indicate an abnormal and pathological response to foreign material in the body. None of these findings were consistent with the current hypotheses for metallosis including type IV metal hypersensitivity and ALVAL. The outcomes of this study connect clinicians and engineers and establish a need for research and invention of novel materials to be used in orthopedic instrumentation.
52


REFERENCES
Adogwa, Owoicho, Scott Parker, David Shau, Stephen Mendelhall, Oran Aaronson,
Joseph Cheng, Clinton Devin, and Matthew J. McGirt. 2015. Cost Per Quality-Adjusted Life Year Gained of Revision Fusion for Lumbar Pseudoarthrosis. Journal of Spinal Disorders and Techniques 28 (3): 101-5. doi: 10.1097/B SD ,0b013 e318269cc4a.
Bhola, R., S. M. Bhola, B. Mishra, R. A. Ayers, D. L. Olson, Donald O. Thompson, and Dale E. Chimenti. 2011. ELECTROCHEMICAL CHARACTERIZATION OF A LOW MODULUS Ti-35.5Nb-7.3Zr-5.7Ta ALLOY IN A SIMULATED BODY FLUID USING EIS FOR BIOMEDICAL APPLICATIONS. In 1184-91. doi:10.1063/l.3592069.
Bhola, Rahul, Shaily M Bhola, Brajendra Mishra, Reed Ayers, David L Olson, and Timothy Ohno. 2011. Cellular Response of Titanium and Its Alloys as Implants. Journal of Oral Implantology 37 (4): 387-99. doi: 10.1563/AAID-JOI-D-09-00075.1.
Bhola, Rahul, and Brajendra Mishra. 2012. Characterization of a Biomedical Titanium Alloy Using Various Surface Modifications to Enhance Its Corrosion Resistance and Biocompatibility. Materials Science Forum 706-709 (January): 105-12. doi :10.4028/www. scientific.net/MSF. 706-709.105.
Bradberry, S. M., J. M. Wilkinson, and R. E. Ferner. 2014. Systemic Toxicity Related to Metal Hip Prostheses. Clinical Toxicology 52 (8): 837-47. doi: 10.3109/15563650.2014.944977.
Cousen, Philippa J., and David J. Gawkrodger. 2012. Metal Allergy and Second-
Generation Metal-on-Metal Arthroplasties. Contact Dermatitis 66 (2): 55-62. doi: 10.1111/j. 1600-0536.2011.01970.x.
Denaro, Vincenzo, Nicola Papapietro, Alessandro Sgambato, Simona A. Bamaba, Laura Ruzzini, Barbara De Paola, Alessandro Rettino, and Achille Cittadini. 2008. Periprosthetic Electrochemical Corrosion of Titanium and Titanium-Based Alloys as a Cause of Spinal Fusion Failure. [Miscellaneous Article]. Spine 33 (1): 8-13. doi:10.1097/BRS.0b013e31815e3978.
Djurasovic, Mladen, Steven D. Glassman, Jennifer M. Howard, Anne G. Copay, and
Leah Y. Carreon. 2011. Health-Related Quality of Life Improvements in Patients Undergoing Lumbar Spinal Fusion as a Revision Surgery. Spine 36 (4): 269-76. doi:10.1097/BRS.0b013e3181cfl091.
Ehrensberger, Mark T., Shiril Sivan, and Jeremy L. Gilbert. 2010. Titanium Is Not the Most Biocompatible Metal under Cathodic Potential: The Relationship between Voltage and MC3T3 Preosteoblast Behavior on Electrically Polarized cpTi Surfaces. Journal of Biomedical Materials Research Part A 93 A (4): 1500-1509. doi: 10.1002/jbm.a.32622.
53


Elves, M W, J N Wilson, J T Scales, and H B Kemp. 1975. Incidence of Metal
Sensitivity in Patients with Total Joint Replacements. British Medical Journal 4 (5993): 376-78.
Farah, Ibrahim O., Phuong X. Nguyen, Zikri Arslan, Wellington Ayensu, and Joseph A. Cameron. 2010. Significance of Differential Metal Loads in Normal Versus Cancerous Cadaver Tissues. Biomedical Sciences Instrumentation 46: 404-9.
Frigerio, Elena, Paolo D. Pigatto, Gianpaolo Guzzi, and Gianfranco Altomare. 2011.
Metal Sensitivity in Patients with Orthopaedic Implants: A Prospective Study. Contact Dermatitis 64 (5): 273-79. doi: 10.1111/j. 1600-0536.2011.01886.x.
Gilbert, J. L., C. A. Buckley, and J. J. Jacobs. 1993. In Vivo Corrosion of Modular Hip Prosthesis Components in Mixed and Similar Metal Combinations. The Effect of Crevice, Stress, Motion, and Alloy Coupling. Journal of Biomedical Materials Research 27 (12): 1533-44. doi:10.1002/jbm.820271210.
Gill, Harindeijit S., George Grammatopoulos, Stephen Adshead, Evagellos Tsialogiannis, and Eleftherios Tsiridis. 2012. Molecular and Immune Toxicity of CoCr Nanoparticls in MoMHip Arthroplasty. Trends in Molecular Medicine 18 (3): 145-55. doi:http://dx.doi.org/10.1016/j.molmed.2011.12.002.
Goldenberg, Yoni, Jin W. Tee, Cesar M. Salinas-La Rosa, and Michael Murphy. 2016. Spinal Metallosis: A Systematic Review. European Spine Journal 25 (5): 1467-73. doi: 10.1007/s00586-015-4347-6.
Haeri, Morteza, and Jeremy L. Gilbert. 2013. Study of Cellular Dynamics on Polarized CoCrMo Alloy Using Time-Lapse Live-Cell Imaging. Acta Biomaterialia 9 (11): 9220-28. doi:10.1016/j.actbio.2013.06.040.
Haeri, Morteza, Torsten Wollert, George M. Langford, and Jeremy L. Gilbert. 2012.
Electrochemical Control of Cell Death by Reduction-Induced Intrinsic Apoptosis and Oxidation-Induced Necrosis on CoCrMo Alloy in Vitro Biomaterials 33 (27): 6295-6304. doi:10.1016/j.biomaterials.2012.05.054.
Hallab, Nadim J., Brian W. MS Cunningham, and Joshua J. Jacobs. 2003. Spinal
Implant Debris-Induced Osteolysis. [Miscellaneous Article] Spine, October.
Hallab, Nadim, Katharine Merritt, and Joshua J. Jacobs. 2001. Metal Sensitivity in Patients with Orthopaedic Implants. J Bone Joint SurgAm 83 (3): 428-428.
Hallab, N. J., J. J. Jacobs, A Skipor, K Mikecz, and JO Galante. 2000. Systemic Metal-Protein Binding Associated with Total Joint Replacement Arthroplasty. Journal of Biomedical Materials Research 49 (3): 353-61.
Hanawa, Takao. 2002. Evaluation Techniques of Metallic Biomaterials in Vitro
Science and Technology of Advanced Materials 3 (4): 289-295.
54


Harrington, James M., Daniel J. Young, Amal S. Essader, Susan J. Sumner, and Keith E. Levine. 2014. Analysis of Human Serum and Whole Blood for Mineral Content by ICP-MS and ICP-OES: Development of a Mineralomics Method. Biological Trace Element Research 160 (1): 132-42. doi: 10.1007/sl2011-014-0033-5.
Health, Center for Devices and Radiological. 2015. Metal-on-Metal Hip Implants -
Concerns about Metal-on-Metal Hip Implants. WebContent. U.S. Food &Drug Administration.
https://www.fda.gov/Medicaldevices/productsandmedicalprocedures/implantsand prosthetics/metalonmetalhipimplants/ucm241604.htm.
Hopkins, Philip M. 2006. Skeletal Muscle Physiology. Continuing Education in
Anaesthesia Critical Care & Pain 6 (1): 1-6. doi:10.1093/bjaceaccp/mki062.
Jacobs, Joshua. 2016. AAOS/AAHKS/The Hip Society Adverse Metal Reactions in Total Hip Arthroplasty. presented at the AAOS Webinar Series.
Kelly, Michael P., Lawrence G. Lenke, Keith H. Bridwell, Rashmi Agarwal, Jakub Godzik, and Linda Koester. 2013. The Fate of the Adult Revision Spinal Deformity Patient: A Single Institution Experience. Spine 38 (19): El 196-1200. doi: 10.1097/BRS. ObO 13 e31829e764b.
Kirkpatrick, John S., Ramakrishna Venugopalan, Preston Beck, and Jack Lemons. 2005. Corrosion on Spinal Implants. J Spinal Disor Tech 18 (3): 247-51.
Lieberman, Jay R. 2014. AAOS Comprehensive Orthopaedic Review. American Academy of Orthopaedic Surgeons 3.
Lohmann, Christoph H, Gurpal Singh, Hans-Georg Willert, and Gottfried H Buchhorn. 2014. Metallic Debris from Metal-on-Metal Total Hip Arthroplasty Regulates Periprosthetic Tissues. World Journal of Orthopedics 5 (5): 660-66. doi: 10.5312/wjo.v5.i5.660.
Lukina, Elena, Aleksandr Laka, Mikhail Kollerov, Mykhamad Sampiev, Peter Mason, Paul Wagstaff, Hilali Noordeen, Wai Weng Yoon, and Gordon Blunn. 2016. Metal Concentrations in the Blood and Tissues after Implantation of Titanium Growth Guidance Sliding Instrumentation. The Spine Journal 16 (3): 380-88. doi:10.1016/j.spinee.2015.11.040.
Maruenda, Jose I., Carlos Barrios, Felipe Garibo, and Borja Maruenda. 2016. Adjacent Segment Degeneration and Revision Surgery after Circumferential Lumbar Fusion: Outcomes throughout 15 Years of Follow-Up. European Spine Journal 25 (5): 1550-57. doi:10.1007/s00586-016-4469-5.
Mu, Y., T. Kobayashi, K. Tsuji, M. Sumita, and T. Hanawa. 2002. Causes of Titanium Release from Plate and Screws Implanted in Rabbits. Journal of Materials Science: Materials in Medicine 13 (6): 583-588.
55


Oliveira, Catarina A., Isabel S. Candelaria, Pedro B. Oliveira, Antonio Figueiredo, and Filipe Caseiro-Alves. 2015. Metallosis: A Diagnosis Not Only in Patients with Metal-on-Metal Prostheses. European Journal of Radiology Open 2: 3-6. doi:10.1016/j.ejro.2014.11.001.
Orthopedic Implants. 2017. Spine-Health, http://www.spine-health. com/glossary/orthopedic-implants.
Pasha, Qaisara, Salman A. Malik, Javed Iqbal, Nazia Shaheen, and Munir H. Shah. 2008. Comparative Evaluation of Trace Metal Distribution and Correlation in Human Malignant and Benign Breast Tissues. Biological Trace Element Research 125 (1): 30-40. doi: 10.1007/s 12011-008-8158-z.
Rajaee, Sean S., Hyun W. Bae, Linda E.A. Kanim, and Rick B. Delamarter. 2012.
Spinal Fusion in the United States: Analysis of Trends From 1998 to 2008. Spine 37 (1): 67-76. doi:10.1097/BRS.0b013e31820cccfb.
Rio, Javier del, Jose Beguiristain, and Julio Duart. 2007. Metal Levels in Corrosion of Spinal Implants European Spine Journal 16 (7): 1055-61. doi:10.1007/s00586-007-0311-4.
Sarmiento-Gonzalez, Alejandro, Juan Manuel Marchante-Gayon, Jose Maria Tejerina-Lobo, Jose Paz-Jimenez, and Alfredo Sanz-Medel. 2008. High-Resolution ICP-MS Determination of Ti, V, Cr, Co, Ni, and Mo in Human Blood and Urine of Patients Implanted with a Hip or Knee Prosthesis. Analytical andBioanalytical Chemistry 391 (7): 2583-89. doi:10.1007/s00216-008-2188-4.
Shiroiwa, Takeru, Yoon-Kyoung Sung, Takashi Fukuda, Hui-Chu Lang, Sang-Cheol Bae, and Kiichiro Tsutani. 2010. International Survey on Willingness-to-Pay (WTP) for One Additional QALY Gained: What Is the Threshold of Cost Effectiveness? Health Economics 19 (4): 422-37. doi:10.1002/hec,1481.
Takahashi, Shinobu, Joel Delecrin, andNorbertPassuti. 2001. Intraspinal Metallosis Causing Delayed Neurologic Symptoms After Spinal Instrumentation Surgery. Spine 26 (13): 1495-99.
Tezer, Mehmet, Unal Kuzgun, Azmi Hamzaoglu, Cagatay Ozturk, Fevziye Kabukcuoglu, and Mustafa Sirvanci. 2005. Intraspinal Metalloma Resulting in Late Paraparesis. Archives of Orthopaedic and Trauma Surgery 125 (6): 417-21. doi: 10.1007/s00402-005-0802-x.
Villaragga, Marta L., Peter A. Cripton, Stephanie D. Teti, Duane L. Steffey, Saki
Krisnamuthy, Todd Albert, Alan Hilibrand, and Alexander Vaccaro. 2006. Wear and Corrosion in Retrieved Thoracolumbar Posterior Internal Fixation. Spine 31 (21): 2454-62.
Vincent, John B. 2000. The Biochemistry of Chromium. The Journal of Nutrition 130 (4): 715-18.
56


Watters, Tyler Steven, Diana M. Cardona, K. Sunil Menon, Emily N. Vinson, Michael P. Bolognesi, and Leslie G. Dodd. 2010. Aseptic Lymphocyte-Dominated Vasculitis-Associated Lesion. American Journal of Clinical Pathology 134 (6): 886-93. doi: 10.1309/AJCPLTNEUAH8XI4W.
Weinstein, Milton C., George Torrance, and Alistair McGuire. 2009. QALYs: The Basics. Value inHealth 12 (March): S5-9. doi: 10.1111/j. 1524-4733.2009.00515.x.
Yee, Albert, Nana Adjei, Jennifer Do, Michael Ford, and Joel Finkelstein. 2008. Do
Patient Expectations of Spinal Surgery Relate to Functional Outcome? Clinical Orthopaedics and Related Research 466 (5): 1154-61. doi:10.1007/sll999-008-0194-7.
Yue, Hua, Wei Wei, Zhanguo Yue, Piping Lv, Lianyan Wang, Guanghui Ma, and Zhiguo Su. 2010. Particle Size Affects the Cellular Response in Macrophages. European Journal of Pharmaceutical Sciences 41 (5): 650-57. doi:10.1016/j.ejps.2010.09.006.
57


Full Text

PAGE 1

LOCAL TISSUE REACTION TO THE RELEASE OF METAL IONS FROM SPINAL IMPLANTS by MACKENZIE MILLER B.S., University of Louisiana at Monroe, 2015 M.S., University of Colorado, Denver, 2017 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Science Bioengineering Program 2017

PAGE 2

ii This thesis for the Master of Science degree by Mackenzie Miller h as been approved for the Bioengineering Program by Cathy Bodine Chair, Ph.D. Reed Ayers Ph.D. Christopher J. Kleck M D Date : May 13, 2017

PAGE 3

iii Miller, Mackenzie (M.S., Bioengineering Program ) Local tissue reaction to the release of metal ions from spinal implants Thesis directed by Associate Professor Cathy Bodine ABSTRACT Implant failure and metallosis are current issues in the field of orthopedics. The existing hypotheses used to explain this phenomenon, type IV metal hypersensitivity and ALVAL, are not sufficient and do not provide clinical relevance. Following the approved IRB protocol patients with scheduled revision surgeries that met the inclus ion criteria were consented and enrolled into the study. During the revision surgery tissue samples and removed hardware were collected and analyzed. From visual observation corrosion and wear were evident on all retrieved hardware. Metal ion concentra tions were above normal values in all three patients with ICP MS/AES data. These results indicate a cause and effect relationship between the corrosion of the hardware and the elevated metal ion concentrations. Using the ICP MS/AES results, a voltage tha t was generated from the corrosion products of the metal instrumentation was calculated using the Nernst equation. These voltages were elevated above normal membrane potentials which could lead to adverse reactions in the tissue. From histological analys is two common themes were observed, foreign body giant cell reactions and histiocytes, laden with foreign material, trapped in the fibrous tissue. These are both pathological responses that indicate an abnormal reaction to the foreign material released fr om the metal hardware in the body. The results of this study connect important clinical factors with engineering principles and address the need for new collaboration, communication, and invention between material science, engineering, and medicine.

PAGE 4

iv T he form and content of this form are approved. I recommend its publication. Approved: Cathy Bodine

PAGE 5

v ACKNOWLEDGEMENTS To my biggest fans, greatest supporters, and most loving parents, thank you for your support and encouragement in every aspect of my lif e. To my loving and supportive boyfriend Matt, thank you for everything that you do! I love you guys! A very special gratitude to my thesis committee, Dr. Cathy Bodine, Dr. Reed Ayers, and Dr. Christopher Kleck. It has been such an honor to work with all of you. I am also grateful for the insight and support from the following University Health staff: Dr. Schowinsky, Dr. Patel, Susan Estes, Dr. Burger, Robert Cooley, Dr. Cain, Theresa Schroeder, and Dr. Ou yang. A special thanks to Claire Cofer and Emily Lindley for their help and guidance through the IRB and human subjects research process.

PAGE 6

vi TABLE OF CONTENTS CHAPTER I. INTRODUCTION ................................ ................................ ................................ ........... 9 II. REVIEW OF THE LITERATURE ................................ ................................ .............. 14 Metallosis and Spine Surgery ................................ ................................ ........................ 14 Corrosion ................................ ................................ ................................ ....................... 18 Metal Ions in the Body ................................ ................................ ................................ .. 22 Metal Hypersensitivity ................................ ................................ ................................ .. 25 Type IV cell mediated response ................................ ................................ ................ 2 6 Metal Hypersensitivity Testing ................................ ................................ ................. 26 Aseptic Lymphocyte Dominated Vasculitis Associated Lesion ................................ ... 29 Electrochemical voltage shifts created as a result of implant corrosion ....................... 29 Conclusion ................................ ................................ ................................ ..................... 31 III. MATERIALS AND METHODS ................................ ................................ ................ 32 IV. RESULTS ................................ ................................ ................................ ................... 42 V. DISCUSSION ................................ ................................ ................................ .............. 46 Limitations ................................ ................................ ................................ ................. 47 Future Research ................................ ................................ ................................ ......... 49 VI. CONCLUSION ................................ ................................ ................................ ........... 52 REFERENCES ................................ ................................ ................................ ................. 53

PAGE 7

vii LI ST OF TABLES TABLE Table 1. Patient demographic information including gender, age, duration of previous hardware in situ, location of the revision, and indications for the revision based on the inclusion criteria. ................................ ................................ ................................ ............... 42 Table 2. From the retrieved instrumentation the type of metal and types of corrosion and wear se en on the hardware through visual inspection were recorded. Only the presence or absence of a type of corrosion or wear was recorded. ................................ ...................... 43 Table 3. ICP MS/AES dat a measuring the metal ion concentrations found in the retrieved tissue samples. Values below the detection of ICP MS/AES are reported as < 0.1. ....... 46

PAGE 8

viii LIST OF FIGURES FIGURE Figure 1. (a) Intraoperative image of metallosis from a revision spine surgery. (b) Removed pseudotumor with metal staining from the same revision surgery (photos courtesy of Reed Ayers, PhD.). ................................ ................................ ................................ ..................... 17 Figure 2. Circuit drawings representing a (a) multi layer oxide and (b) a single layer oxide (used with the permission of Dr. Rahul Bhola). ................................ ............................... 18 Figure 3. Evidence of fretting (a) and pitting (b) corrosion on retrieved spinal instrumentation made of Ti. ................................ ................................ .............................. 20 Figure 4. Schematic representation of the corrosion cell generated in vivo. As the metal instrumentation corrodes electrons, followed by positively charged metal ions, flow to the tissue which acts as the cathode. Negatively charged species, like oxygen and hydroxide, then move back to the metal instrumentation acting as the anode. ................................ ... 40 Figure 5. The Nernst equation used for this calculation. R, T, an d F are all know constants. The valency, z, used is the most common oxide state of each metal. The extracellular metal ion concentrations are the values from the experimental data. The intracellular metal ion concentrations are pulled from previous litera ture. ................................ .......................... 41 Figure 6. (a) Evidence of pitting corrosion. (b) Fretting wear has the shiny appearance along the rod caused from micromotion between the connector and the rod. (c) Evidence of repassivation. The discoloration along the screw indicates reformation of the oxide layer at a different thickness. (d) Fracture between the threading and the head of the screw. ......... 44 Figure 7. All images are of hematoxylin and eosin stained slides taken at 400x magnification. A) Foreign body giant cell reaction. B) Particle laden histiocytes trapped in fibrous tissue. C) Necrotic skeletal muscle and necrotic fibrovascular connective tissue. D) Typical look of metallosis. ................................ ................................ ................................ 45

PAGE 9

9 INTRODUCTION Implant failure and metallosis are current issue s in the field o f orthopedics. Patients with these findings can experience severe pain, inflammation, and ultimately implant failure leading to revision surgery. The cause and effect of metallosis has yet to be identified making the clinical diagnosis and prevention im possible. C urrent hypothese s include are metal hypersensitivity due to delayed type IV cell mediated immune response and Aseptic Lymphocytic Vasculitis Associated Lesion (ALVAL). These reacti ons lead to the recruitment of macrophages leading to inflammation and pain. No experimental dat a has established a cause and effect relationship between these hypotheses and the clinical diagnosis of metallosis Furthermore, there is no clinical relevan ce to th ese hypothese s because no test s currently exist to predict metal allergies due to implants or ALVAL There is a significant need to further research the phenomenon of metallosis It is well documented that m etals corrode in the body and release metal ions in the surrounding tissue The electrochemical process of corrosion generates a voltage that can alter the physiological environment of local soft tissue. Previous in vitro experiments have shown that specific voltage ranges lead to either cell necrosis or apoptosis. Using failure analysis methods, we can identify corrosion products from orthopedic spine implants, identify the concentration of metal ions in the surrounding tissue, and calculate the voltage created by these corrosion products whi ch may be linked to poor health outcomes and potential need for revision surgery. The intellectual merit of this proposed activity i s crucial in the advancement of knowledge and understanding by providing evidence to support the hypothesis that

PAGE 10

10 corrosio n is linked to poor health outcomes and the potential need fo r revision surgery This will co nne ct the field s of orthopedics human physiology, and material science by identifying a link between the implant, the procedure, and the patient outcomes The team is qualified to conduct this research because of the combined knowledge of materials and corrosion, human phys iology, and clinical experience The proposed concept is original in its entirety because it will connect the previous research of electroc hemical gradients and imp lant failure due to metallosis to a clinical setting. The proposed activity will be conducted under an organized and chronological approach. First patients will be screened and enrolled to participate in the study Once consented the hardware and tissue samples will be collected during the revision surgery. Hardware will undergo destructive and non destructive metallography testing to identify areas and types of corrosion. One tissue sample will be sent for ICP MS/AES analysis and the other will be sent for histological analysis. The results of ICP MS/AES will be used to calculate the voltage generated by the corrosion products using the Nernst equation. There is existing access to resources The broader impact of this proposed activity will advance the understanding and significance of metallosis in this field by supporting the hypothesis that corrosion lead s to poor patient outcomes and need for revision surgery The results of this study will be published in at least one scientific journal to enhance the scientific and technological understanding of metallosis The results of the proposed activity will benefit society by developing a clinically relevant explanation to metallosis that may lead to better patient o utcomes. Currently, p atients that have experience d implant failure in the form of mechanical failure, pseudotumor, non union, pain, and dis k degeneration must undergo a

PAGE 11

11 revision surgery These surgeries are costly, painful, and may not always provide the best solution to the issue A better understanding of metallosis and implant failure will improve future patient care and patient outcomes.

PAGE 12

12 Advisor Reed Ayers, Ph.D. Department of Orthopedics Committee Members Reed Ayers, Ph.D. Cathy Bodine, Ph.D. C hristopher Kleck, MD Title of Proposal Local Tissue Reaction to the Release of Metal Ions from Spinal Implants Hypothesis Corrosion products from orthopedic spine implants alters the physiologic environment of local soft tissue, by generating a voltage, leading to cell necrosis, identified through histology; propagating a pathological response that may lead to poor health outcomes. Specific Aims Specific Aim One: To collect two tissue samples from revision patients with and without evidence of a solid fu sion and hardware failure; one sample adjacent to the metal implant (peri hardware tissue) and one sample from either the vertebrae above or below the revision (control tissue). Tissue sam ples will be collected from peri hardware and control areas during t he revision surgery at the University of Colorado Hospital by Orthopedic Spine Surgeons. Control and peri hardware samples will be stored in separate containers prior to sample preparation.

PAGE 13

13 Specific Aim Two: To perform non destructive and destructive analysis on the explanted instrumentation. All instrumentation will be photographed, and then analyzed under a scanning electron microscopy as well as energy dispersive spectroscopy to identify regions of corrosion and wear. Physical metallography will als o be used for this analysis. Specific Aim Three: To perform ICP MS/AES on the collected tissue samples. At least two grams, but no more than 10 grams, of the peri hardware and control samples will be stored in sterile sample jars and sent to Huffman Hazen Laboratories for analysis. These samples will be utilized to determine the presence and quantity of metal ions in the sample. Specific Aim Four: To correlate the results of this research to clinical data acquired through the medical records and to prior in vitro studies. ICP MS/AES results will be correlated to data acquired through the medical records. In another analysis, the results of ICP MS/AES will be used to calculate the voltage generated by the instrumentation in vivo and compared to prior in vitro studies.

PAGE 14

14 REVIEW OF THE LITERA TURE Metallosis and Spine Surgery A n orthopedic implant is a device surgically placed into the body designed to 2017) In orthope dic spine surge ry, implants are used to treat deformity, mechanical back pain, stenosis, spondylolisthesis, fractures, and tumors. The varied requirements of these implants are diverse, to treat multiple pathologies, in individuals across the lifespan. Th e number of orthopedic spine surgeries is on the rise. An epidemiological study conducted by Rajee observed a 137% increase in spinal fusion discharges in the U.S. between the years of 1998 to 2008, a much higher rate than any other orthopedic procedure (Rajaee et al. 2012) More patients are undergoing primary spine surgeries at a younger age which could later lead to revision surgeries Revision surgeries are necessary for patients that experience degenerative disc disease, pseudoarthrosis, mechanical hardware failure, non union, pain, and adjacent segment disease These compl ications are often summarized as implant failure A retrospectiv e study conducted by Kelly et al at an institution specific to spinal implants found that 21% of patien ts would undergo multiple revision surgeries (Kelly et al. 2013) That is double the percentage of revision surgeries seen in all of orthopedics. With the number of revision sp ine surgeries increasing it is important to define success in spinal surgery. The success of a surgery is difficult to gauge since there is no clear definition of surgical or clinical success In a study by Yee et al. almost 20% of patients were dissati sfied with the result of their surgery, with that number increasing in patients that had previous spinal intervention (Yee et al. 2008) Surgeons and patients have different

PAGE 15

15 perspectives on successful procedures Other factors contributing to the confusion of what defines a successful procedure include time and type of surgical procedure In a retrospective study conducted by Maruenda and Barrios, 15 year follow up data was collected for patients that underwent primary 1 3 level circumferential lumbar fusion. They reported good clinical results within the f irst year post surgery but from 2 to 15 year follow up the clinical outcome of patients worsened significantly This result ed in 24 patients (37.5%) having a revision surgery (Maruenda et al. 2 016) From this study it is apparent that time is a factor in patient success and rates of revision surger y. A retrospective cohort analysis conducted by Djurasovic and Glassman, measured quality of life improvements for patients undergoing a revision lumbar spine surgery (Djurasovic et al. 2011) They concluded t hat only modest improveme nts could be expected from a revision surgery as measured by patients reaching minimum clinically important differences (MCID) when comparing preoperative, 1 year postoperative, and 2 year postoperative Oswestry Disability Index (ODI) and MOS Short Form 36 (SF 36) data. Less than 50% of patients enrolled in the study met the MCID thresholds for ODI and SF 36. Only 24% of patients requiring revision for non union met MCID thresholds for SF 36. The findings from this study suggest that the success of revis ion surgery when analyzed by qualit y of life improvements is poor. The patient centered outcomes for the success of revision surgery are low for such a costly procedure Economic costs are often reported as quality added li fe years in healthcare hrough different health states over time with each health state havi ng a different value (Weinste in, Torrance, and McGuire 2009) This figure excludes costs such as lost work days and

PAGE 16

16 other costs to the patient. A study by Adogwa and Parker calculated the 2 year comprehensive costs of revision lumbar fusion for pseudoarthrosis. They reported the procedure to only be marginally cost effective at $118,945 per QALY gained (Adogwa et al. 2015) A survey study conducted by Shiroiwa et al., reported the previous conventional threshold for cost effecti veness of medi cal intervention, $50,000 $100,000 in the United States is equ ivalent to the calculated willingness to pay (WTP) per QALY at $62,000 (Shiroiwa et al. 2010) From this reported threshold, the cost of lumbar fusion surgery for pseduoarthrosis is only marginally cost effective because it is outside of the conventional threshold. Revision spine surgery is costly and only moderately effective. There is a need for further research to un derstand potential causes of implant failur e leading to revision surgery. One possible cause of implant failure could be metallosis. Metal s have been used for orthopedic implants because of their toughness, elasticity, rigidity, and electrical properties (Hanawa, 2000). Many medical implants are made of titanium alloys because of these properties and biocompatibility (R. Bhola et al. 2011) However, the presence of foreign metallic debris in the body can cause local tis sue damage and differ enc es in tissue characteristics, called metallosis (Lohmann et al. 2014) The symptoms associated with metallosis ar e non specifi c, ranging from a systemic re sponse to asymptomatic (Oliveira et al. 2015) Metallosis is found during the revision surgery incidentally. This occurs because of t he incons istency of patient symptoms and the difficulty identifying this phenomenon through imaging techniques The lack of clinical relevance could explain why this phenomenon is often used to describe the appearance of the tissue instead of a cause for implant f ailure leading to revision surgery.

PAGE 17

17 In an attempt to document this under reported complication a systematic review, conducted by Goldenberg and Tee, found only three relevant articles regarding spinal metallosis ( excluding all arthroplasty related studies ) (Goldenberg et al. 2016) The patients included in these case reports displayed neurological symptoms after their ini tial surgery leading to a revision procedure to relieve symptoms. Takahashi and Delcrin reported on two cases of intraspinal metallosis in symptomatic patients with spinal instrumentation (Takah ashi, Delecrin, and Passuti 2001) They found stained, granulated tissue masses described as metallosis that had formed in the spinal canal adjacent to the instrumentation leading to neurologic symptoms. This study introduced a possible cause of delayed neurologic symptoms due to metallosis. A case study by Tezer et al. identified an intraspinal metalloma due to crevice corrosion on stainless steel instrumentation leading to neurological symptoms. After removal of the instrumentation, the patient had complete symptom resolution at the three month checkup appointment (Tezer et al. 2005) These three case studies identify an existing iss ue and provide evidence for metallosis in spine but do little to determine mechanisms Figure 1. (a) Intraoperative image of metallosis from a revision spine surgery. (b) Removed pseudotumor with metal staining from the same revision surgery (photos courtesy of Reed Ayers, PhD.). (a) (b)

PAGE 18

18 Metallosis is an underreported phenomenon in spine surgery that is not well understood as evidenced by the lack of literature It can cause neurological and non specif ic symptoms that may be alleviated with revision surgery and removal of metal instrumentation. Revision surgeries are only moderately successful, when measured by patient centered outcomes, and time is a factor contributing to patient success and rate of revision surgeries. Further, revision surgeries often require re implantation of metal devices. There is an immediate need to understand the cause of metallosis in spine starting with the metals used for orthopedic instrumentation Corrosion Corrosion is an electr ochemical process resulting in the cleaving of the chemical bonds that hold metal ions together (Lieberman 2014) This process of corrosion can weaken the implant leading to implant failure and the release of metal ion s causing adverse reactions in the body (Lieberman 2014) Metals used in medical devices are coated by an oxide layer that acts as a barrier of protection between the metal surface and the body (Rahul Bhola and Mishra 2012) This layer coating the metal can be represented as an electrical circuit based on the composition of the oxide layer (R. Bhola et al. 2011) This passive oxide layer will reform in solution or when exposed to air and electrolytes (R. Bhola et al. 2011) (Rahul Bhola and Mishra 2012) If this layer remains intact there Figure 2. Circuit drawings representing a (a) multi layer oxide and ( b) a single layer oxide (used with the permission of Dr. Rahul Bhola). (a) (b)

PAGE 19

19 should be no corrosion on the implant, however, any imperfection in the oxide layer leads to corrosion of the underlying surface until this layer can reform through passivation. Pitting corrosion is the most severe form of corrosion leading to a toxic release of metal ions. Corrosion is initiated by a chip or defect along the oxidative layer coating the metal implant (Lieberman 2014) Crevice corrosion occurs because of the construction and geometry of the metal. The corrosion is localized in the areas surrounding the crevice geometry like a weldment, bolted parts, or two interfacing pieces that can lead to the creation of an ionic gradient altering the localized area (Lieberman 2014) Galvanic corrosion is the result of two different metals forming a difference in electrochemical potential (Lieberman 2014) Fretting wear occurs due to repetitive micromotion when a load is imposed on them (Lieberman 2014) These types of corrosion and wear can be identified on retrieved orthopedic implants due to the constant l oad, various pieces used in the construction of the implant, and the possible damage created by surgical implantation and wear. A retrieval study conducted by Kirkpatrick et al. identified three common modes of corrosion and wear that appear to happen si multaneously in vivo ; fretting wear crevice and galvanic corrosion (Kirkpatrick et al. 2005) In a retrieval study conducted by Villagraga et al. wear and corrosion were the most common types of damage seen on retri eved spinal hardware and they concluded that revision spine constructs contribute to this type of damage because of the additional segments and mobile pieces added (Villaragga et al. 2006)

PAGE 20

20 Figure 3. Evidence of fretting (a) and pitting (b) corrosion on retrieved spinal instrumentation made o f Ti. The motion created by the load on spinal constructs and the motions between multiple segments provide a source of mechanical damage to the oxide The term mechanically assisted corrosion was coined by Gilbert et al. and describes the effect of motion bet ween parts of a construct that lead to the initial damage of the oxide layer followed by other modes of corrosion (Gilbert, Buckley, and Jacobs 1993) The motion between various components of a construct damages the oxide layer leading to the release of metal ions in to the body. This cycle of fracture to the oxide layer, repassivation, depletion of O 2 in the surrounding environment, decrease in the pH of the surrounding environment, ins tability of the oxide, results in the exposure and corrosion of the underlying metal (Jacobs 2016) A retrieval study of 16 modular hips conducted by Kop and Swarts co ncluded that corrosion is most commonly seen at the interface between the pieces of the hip (Kop, 2009). Other sources of damage to the oxide layer can be due to manufacturing defects or damage during the surgical procedure An in vitro study using rabbit s by Mu et al. found the main sour ces of metal ion release to be handling during surgery implantation and wear and fretting during the time the metal was in the body (a) (b)

PAGE 21

21 (Mu et al. 2002) The damage to the oxide layer can occur from mechanically assisted corrosion or handling during surgery leading to corrosion in t he body. It is well documented that corrosion occurs on any metal introduced to the body (Lieberman 2014) (N. Hallab, Merritt, and Jacobs 2001) (Cousen and Gawkrodger 2012) (Frigerio et al. 2011) Del Rio et al. conducted a study com paring the serum and urine levels of Ni and Cr as measured by atomic absorption spe ctrophotometry of three groups: no metal instrumentation; metal instrumentation and no radiographic signs of corrosion; and metal instrumentation with radiographic signs of corrosion. The results showed that participants with metal in their body had elevated serum and urine Cr levels due to corrosion (del Rio, Beguiristain, and Duart 2007) A study by Hallab, Jacobs, and Skipor reported that metal ion products released from the corroding metal implant remain in the serum of the patients (N. J. Hallab et al. 2000) Using three patient groups they were able to de termine metal ion concentrations between c ontrols, patients with CoCr alloy implants, and patients with Ti alloy implants (N. J. Hallab et al. 2000) They found elevated concentrations of metals in the serum of patients with metal implants bound to proteins (N. J. Hallab et al. 2000) This study emphasized that corrosion happens in biological systems. Corrosion of metals in the body has been the focus of several studies that identified a numb er of biological r easons for this occurrence (del Rio, Beguiristain, and Duart 2007; Takahashi, Delecrin, and Passuti 2001; Rahul Bhola et al. 2011) The body is a hostile environment. Serum and interstitial fluid chloride ion concentrations are 113 and 117 mEq respectively, making these environments very corrosive to any metal instrumentation introduced (Hanawa 2002) Takahashi reported on the normal process of the body once the metal is introduced and noted that the

PAGE 22

22 construct is first surrounded by blood and interstitial fluid (Takahashi, Delecrin, and Passuti 2001) Once the metal is placed in the body it is surrounded by active metabolites, like chloride ions, that begin attacking the surface, pulling metal ions from the construct. There are other metabolites in the body that are known to cause corrosion of metal instrumentation. Active oxygen species produced by macrophages during the process of phagocytosis are known to induce the corrosion of metal instrumentation (Hanawa 2002) A study conducted by Mu et al. implanted cpTi constructs into rabbit tibias to identify the sources and causes of metal ion release (Mu et al. 2002) They noted that even the positive controls where the implant was placed in the muscle tissue, with no obvious signs of wear or fre tting still released metal ions into the surrounding tissue due to active oxygen species and other biochemical factors (Mu et al. 2002) The in vivo environment has been shown to corrode metals in the body. The consequences and functions of these corrosion products in the body has been another are a of research. Metal Ions in the Body concentrations of the metals used in metal implants in serum and muscle tissue For individuals that have never been exposed to Co and Cr serum levels should be < 10 nmol/L and < 40 nmol/L respectively as reported by the UK SupraRegional Assay Service (S AS) (Bradberry, Wilkinson, and Ferner 2014) Recent MHRA guidelines report an appropriate level of Cr or Co to be 7 mg/L (Gill et al. 2012) Other metals such as Al, Mo, and, V have reported ncentrations in muscle tissue at 8.4 4.8 u g/g, 1.6 1 1.41 ug/g, and 0.06 0.03 u g/g respectively ( (Vincent 2000) (Pasha et al. 2008) (Farah et al. 2010) (Harrington et al. 2014) ). O nly whole blood measurements have been reported for

PAGE 23

23 0.00072 ug /g 0.1412 (Farah et al. 2010) A t normal levels these metal ions are inert or benefit the body but at elevated levels there are several reactions and issues that arise. Currently there is no known threshold or specif ic elevated metal ion level leading to a reaction in the body. A techniques study conducted by Hanawa, concluded th at particles released from an implant in 100 ppb are toxic to the surrounding cells (Hanawa 2002) Hanawa went on to say that even trace elements in metal alloys cannot be overlooked as potential causes o f patient symptoms (Hanawa 2002) There have been several studies focused specifically on Co and Cr toxicity. Bradberry et al. reported that pa tients with blood Co concentrations above 250 ug/L displayed peripheral neuropathy after a CoCr MoM hip arthroplasty (Bradberry, Wilkinson, and Ferner 2014) Other systemic issues were identified through this study and attributed to Co toxicity. In an in vitro study by Gill et al. Co and Cr were found to be mutagenic and genotoxic, damaging DNA in a variety of ways (Gill et al. 2012) The damaging effects of Co and Cr have raised concern over their safe ty and use in orthopedic implants. Ti is one of the most commonly used metals in orthopedics yet not much is known about its effect in the body. A retrieval study by Lukina et al. found a 1,500 fold increase in Ti tissue concentrations in pediatric pat ients with sliding cpTi constructs when compared to patients with no metal implants (Lukina et al. 2016) This value is much higher than any previously reported findings. The concern of these elevated metal ion concentrations in the body as a result of orthopedic implants has le d to several studies that aim to understand how the body processes these foreign particles.

PAGE 24

24 The body has certain mechanisms to protect itself from foreign particles. The response and specific pathway can vary based on t hings like size, char ge, and quantity of the particle An in vitro study conducted by Yue et al. analyzed the cellular pathways used to internalize and digest metal particles of varying sizes. Phagocytosis was a crucial pathway for internalization of metal particles of all sizes, nanoparticles were digested in the lysosomal pathway, and micro particles were sequestered in the cytoplasm for digestion (Yue et al. 2010) Gill et al. expanded on this information explaining the cycle of phagocytosis, cell death, and release of cytokines in r esponse to the metal particles could lead to the formation of a pseudotumor (Gill et al. 2012) The body tries to digest metal particles of all sizes yet chronic inflammation and pseudotumor formation still persist. Anoth er study looked at the particular charge and shape of metal ions that allow them passage through receptors on the cell surface and accumulate in the cytoplasm. This movement through nonspecific channels is described as quick and could lead to cell death d ue to mutagenic and genotoxic effects of specific metal ions (Gill et al. 2012) The vicious cycle activated by the presence of foreign particle can be summarized as, (Nadim J. Hallab, Cunningham, and Jacob s 2003) The exact pathway activated is variable and inconsistent and the amount of metallic material present could influence the physiological reaction Lohman et al. found that the type of tissue response correlated with the concentration of metal i ons ,Co, Cr, and Ni, in the tissue (Lohmann et al. 2014) Mean metal c oncentrations of 222.2 52.9 ug/g displayed a strong lymphocytic tissue response while lower metal concentrations, 3.0 0.9 ug/g exhibited a macrophage response. These immune reactions can be observed in any healthy body yet only a

PAGE 25

25 fraction of the population experiences complication such as metallosis with orthopedic implants. As previously stated, there is no known concentration threshold of metal ions that trigger an immune response in the body. Lohman et al. ability to clear the particles through the lym phatic system explains why this phenomenon does not occur in everyone with a metal implant (Lohmann et al. 2014) The amount of foreign mater ial could be a potential cause o f implant failure due to metallosis. Further earing ability in vivo This hypothesis, similar to others, fails to address the cause of the reaction. Current research in the fields of metallosis and orthopedics has settled on two possible hypotheses for implant failure due to metallosis, focused on the patient physiology Metal Hypersensitivity Metal hypersensitivity has been a topic of interest in orthopedics with the focus to characterize the adverse effects seen in some patients with orthopedic implants. Approximately 10% to 15% of the general population displays some type o f dermal sensitivity to metal (N. Hallab, Merritt, and Jacobs 2001; Cousen and Gawkrodger 2012) Approximately 25% of patients with properly functioning implants and 60% of patients with implant failure show som e type of metal sensitivity (Frigerio et al. 2011) While much is known about the dermal response to metal antigens d etermining metal hypersen sitivity prior to the implant procedure or implant failure due to metal hypersensitivity h as proven to be a challenge in clinical setting s

PAGE 26

26 Type IV cell mediated response The cellular response to this release of metal ions has been described as delayed T y pe IV cell mediated responses (N. Hallab, Merritt, and Jacobs 2001) Type IV cell mediated responses are char acterized by the activation of T lymphocytes that release cytokines signaling for the recruitment of macrophages (N. Hallab, Merritt, and Jacobs 2001). The lymphocyte response has been linked to poor implant performance ; yet this cause and effect relation ship remains controversial (Frigerio et al. 2011) The assumption that metal hypersensitivity leads to implant fail ure has many in the field using clinical test s for allergies to solidify this relationship Metal Hypersensitivity Testing Three tests exist clinically to d e termine metal sensitivity in patients. Patch testing in vivo (N. Hallab, Merritt, and Jacobs 2001) lymphocyte transformation testing in vitro (LTT) (N. Hallab, Merritt, and Jacobs 2001) and leukocyte migration inhibition testing (LIF or MIF) are the current tests for determining metal sensitivity (N. Hallab, Merritt, and Jacobs 2001). Patch test ing. Patch testing is not an effective method to predict a stable or failed implant (Cousen and Gawkrodger 2012) Several c oncerns and issues surround this technique as a clinical determinant of metal hyp ersensitivity. The test is designed for short or acute exposures, unlike the long term, constant exposure seen wi th orthopedic implants (N. Hallab, Merritt, and Jacobs 2001). There is also concern with the accuracy of the test because of the difficulty knowing and acquiring appr opriate metal agents to be used as antigens (N. Hallab, Merritt, and Jacobs 2001). Finally patch testing involves sensitizing a previously insensitive patient (N. Hallab, Merritt, and Jacobs 2001). Second

PAGE 27

27 exposure to the antigen, during the implantation of the orthopedic device, may trigger the e leading to hypersensitivit y. A study by Elves, Wilson, Scales, and Kemp patch tested 50 patients who had a Total Joint Replacement ( TJR ) inserted anywhere from one to twelve years prior (Elves et al. 1975) Of the 50 patients, 19 reacted positive to the patch test and of the 26 patients with a failed implant only 15 were metal sensitive They conclud ed that a significantly increased incidence of metal sensitivity appeared in patients with metallic implan ts (Elves et al. 1975). While this data may be convincing to some, questions about the methodology of this study remain The challenge agents used were metal salts including nickel sulphate cobalt chloride, and titanium oxalate. No one has yet to determine adequate challenge agents or a series of tests to determine metal sensitivity due to orthopedic implants (N. Hallab, Merritt, and Jacobs 2001). The next issue is that all of the patients have a metallic implant and are therefore already sensitized to metal. Second exposure from the patch test could be a reason for positive results. In a prospective study of 100 patients conducted by Frigerio, Pigatto, and Guzzi, si milar results were rep orted (Frigerio et al. 2011). No participant showed signs of a metal allergy prior to surgery and all had undergone either a T KA or THA (Frigerio et al. 2011). Challenge age nts used for patch testing included metal salts and were read 4 days later (Frigeri o et al. 2011). One year after surgery 72 of t he patients were brought back for postoperative testing (Frigerio et al. 2011). Results of the study showed a 6.5% increase in metal sensitization postoperative (Frigerio et al. 2011). The design of this study sensitized all participants to metal prior to surgery. This flaw in the study design could have provided researchers with several false positives or negatives upon retesting. Patch testing does not provide a

PAGE 28

28 clear cause and effect relationship between th e metal implant and implant failure because of the lack of appropriate challenge agents and the risk of sensitization upon secondary exposure Lymphocyte transformation testing L TT measures the proliferation of lymphocytes following activation from a spe cific antigen in vitro (N. Hallab, Merritt, and Jacobs 2001). This test uses radioactive markers incorporated into lymphocyte DNA to quantify the proliferative response (N. Hallab, Merritt, and Jacobs 2001). In the study conducted by Frigerio et al. 12 of the participants were screened with patch testing and LTT (Frigerio et al. 2011). The small sample size they enrolled did not permit any significant findings to be reported however, they were able to conclude that LTT can provid e some additi onal informati on (N. Hallab, Merritt, and Jacobs 2001; Frigerio et al. 2011) ]. While this in vitro test does not sensitize the patient, like patch testing, little information surrounds its credibility in determining metal hypersensitivity. Leukoc yte migration inhibition testing. T his in vitro test uses s ensitizers like metal ions to measure the migration activity of a mixed population of leukocytes in culture (N. Hallab, Merritt, and Jacobs 2001). A lone, leukocyte testing is not good for ide ntifying metal sensitivity (N. Hallab, Merritt, and Jacobs 2001). Findings from this test when paired with other information may provide some clinical relevance. Type IV metal hypersensitivity does not completely explain the phenomenon of metallosis and with no ability to predict the occurrence of hy persensitivity, this definition appears to b e clinically irrelevant. T he FDA states that it is not possible to predict which patients will experience adverse re actions to MoM implants (Health 2015) Recently a new term has been used to describe this reaction through histological analysis.

PAGE 29

29 Aseptic Lymphocyte Dominated Vasculitis Associated Lesion Aseptic, lymphocyte dominated vascu litis associated lesion or ALVAL, is now considered as a possible explanation for implant failure This reaction is characterized by a dense perivascular inflammatory infiltrate that is only identified upon histopathologic analysis (Watters et al. 2010) In a case series conducted by Watters et al. the histological evidence did not support the hypothesis of a classic allergic reaction. The cells identified through h istology are non specific making it difficult to conclude any particular reaction is occurring (Watters et al. 2010) It is believed that the delayed type hypersensitivity reaction is the cause of ALVAL yet evidence to support this has yet to be discovered. Research has not yet confirmed if the type IV metal hypersensitivity causes the failure of the implant or if the failure of the implant leads to metal hypersensitivity (Watters et al. 2010) This hypothesis does not provide sufficient clinical relevance because it cannot be identified prior to surgery. Diagnosis is only made upon histological analysis. Another issue wit h this hypothesis is the absence of a cause and effect relationship. Research has yet to determine why this is occurring or the order of reactions that takes place prior to implant failure. Alternative hypotheses, that address the cause and effect relati onship, need to be established to better understand the phenomenon of metallosis and improve patient care and outcomes. Electrochemical voltage shifts created as a result of implant corrosion Corrosion in vivo leads to electrochemical voltage shifts that may negatively impact the biocompatibility and effectiveness of the orthopedic implant (Haeri et al. 2012) When deviations from physiological voltages occur cell adhesion is impaired leading to implant loosening and decrease d cell survival Cell death can be initiated by

PAGE 30

30 alterations in the movements of ions across cell membranes, the function of voltage gated channels, and other pathways within the cell (Haeri and Gilbert 2013) Metal ions released through the process of corrosion generate a voltage and electrochemical changes that possible lead to alt ered cell physiology apoptosis and necrosis (Haeri et al. 2012) An in vitro study conducted by Haeri, Wollert, and Langford identified a voltage viability range (Haeri et al. 2012). Using CoCrMo alloy plates as the working electrode they were able to determine the cell viabi lity range from 300 mV +300 mV. Deviating outside this range they observed adverse cell reactions Cathodic voltages led to cell apoptosis while anodic vol tages led to cell necrosis (Haeri et al. 2012). The cause of this cell necrosis has been identif ied through another in vitro study by Haeri and Gilbert using time lapse live cell imaging to monitor the morphological changes of the cells (Haeri and Gilbert 2013) The corrosion products and reduction reactions occurring on the surface of the implant cause voltage changes in the surrounding tissue leading to cell death and implant failure (Haeri and Gilbert 2013; Haeri et al. 2012) These findings can be supported by histological studies. Watters et al. identified a necrotic layer of cells in 2 of the 3 patients in the study with failed MoM hips (Watters et al. 2010) Apoptosis is not easily identified through histological analysis. A similar experiment using cpTi as the alloy component revealed similar outcomes. Ehrensberger et al. concluded that voltages from 1000mV to 600mV negatively impacted cell viability while voltages from 300mV 1000mV seemed to have little impact on cell morhpology (Ehrensberger, Sivan, and Gilbert 2010) In a case study conducted by Denaro et al. showed that a 12.1 X 10 6 T continuous EMF generated around Ti alloy hardware will inhibit the activity and proliferation of os teoblasts (Denaro

PAGE 31

31 et al. 2008) Reduction reactions occurring on the surface of the metal can also influence the local tissue. The oxide layer is a nother crucial piece to this reaction. Electrochemical properties and thickness of the oxide layer change through the process of corrosion and repassivation. The variability in the oxide layer can affect the cells that interact with the metal by changing the electrochemical properties along the metal surface (Ehrensberger, Sivan, and Gilbert 2010) The byproducts of corrosion and subsequent chemical reactions provide an ex planation for the symptoms and reactions seen clinically Cell viability is affected by the voltages generated from corros ion byproducts and the reduction reactions occurring on the surface of the metal that can potentially lead to implant failure. Conclusion Research has yet to identify the cause and effect relationship of implant failure potentially due to metallosis Metal hypersensitivity has many flaws in its clinical testing and a strong correlation between implant failure and hypersensitivity has yet to be proven Corrosion is known to occur when metals come into conta ct with the body and the release of these me tal ions leads to electrochemical voltage shift s and reduction reaction s that lead to changes in cell morphology and function resulting in cell death. This relationship between metal corrosion and altered cell physiology leading to implant failure require s further research to improve patient care and outcomes

PAGE 32

32 MATERIALS AND METHOD S Hypothesis Corrosion products from orthopedic spine implants alters the physiologic environment of local soft tissue, by generating a voltage, leading to cell necrosis, iden tified through histology; propagating a pathological response that may lead to poor health outcomes. Specific Aims Specific Aim One: To collect two tissue samples from revision patients with and without evidence of a solid fusion and hardware failure; one sample adjacent to the metal implant (peri hardware tissue) and one sample from either the vertebrae above or below the revision (control tissue). Tissue samples will b e collected from peri hardware and control areas during the revision surgery at the Univ ersity of Colorado Hospital by Orthopedic Spine Surgeons. Control and peri hardware samples will be stored in separate containers prior to sample preparation. Specific Aim Two: To perform non destructive and destructive analysis on the explanted instrumentation. All instrumentation will be photographed, and if needed, analyzed under a scanning electron microscopy as well as energy dispersive spectroscopy to identify regions of corrosion and wear. Physical metallography will also be used for this a nalysis. Specific Aim Three: To perform ICP MS/AES on the collected tissue samples. At least two grams, but no more than 10 grams, of the peri hardware and control samples will be stored in sterile sample jars and sent to Huffman

PAGE 33

33 Hazen Laboratories for ana lysis. These samples will be utilized to determine the presence and quantity of metal ions in the sample. Specific Aim Four: To correlate the results of this research to clinical data acquired through the medical records and to prior in vitro studies. ICP MS/AES results will be correlated to data acquired through the medical records. In another analysis, the results of ICP MS/AES will be used to calculate the voltage generated by the instrumentation in vivo and compared to prior in vitro studies.

PAGE 34

34 Materials Specific Aim One 2 specimen containers Sharpie Methods Specific Aim One This protocol has been reviewed and accepted by COMIRB protocol ID number: 16 1176 Patients to be enrolled in the study will meet the following inclusion and exclusion criteria: Inclusion Criteria Patients undergoing revision spine surgery with previous instrumentation in the thoracic or lumbar region with evidence of catastrophic hardware failure, pseudoarthrosis, implant loosening and non fusion. Age 18 85 These inclusion criteria will be determined through radiographic and medical record review. While these criteria indicate a higher chance of observing metal staining it does not guarantee that it will be present. If the patients consent and do not have met al staining we will still proceed with our research as metal staining is not the only characteristic of interest. Metal ion concentrations in the peri hardware tissue can be elevated in the absence of tissue staining and is therefore beneficial to this res earch. Exclusion Criteria Patients undergoing revision spine surgery without spinal implants

PAGE 35

35 Patients undergoing revision spine surgery that are not suspected of having metal staining Patients with an identified infection Patients unable to provide consent Pregnancy Patient Population and Recruitment Patients will be recruited through the Spine Center at the University of Colorado Hospital. Patients undergoing revision spine surgeries with evidence of catastrophic hardware failure, pseudoarthrosis, implant loosening and non fusion with previous instrumentation will be offered enrollment. The enrolled patients will be identified prior to surgery by the surgeons and research staff through radiographs and images contained in the medical records. If they meet t he inclusion and exclusion criteria, informed consent will be obtained by a member of the research staff. Consent and Enrollment Procedure The consent form will be administered by either the surgeon or research staff that have been trained in local IRB req uirements prior to revision surgery. Patients that meet inclusion criteria will receive a phone call from the surgeon or research staff to schedule a time to discuss the research study and consenting process. If the patients agree a meeting will be set whe re they will have the purpose and methods of the study explained to them as well as risks and benefits. Consent will be obtained in a waiting room or office prior to surgery at the University of Colorado Hospital. They will receive a copy of the informed c onsent and be given time to review the document. The individual administering the consent will review it with the subject and they will determine if they

PAGE 36

36 demonstrate adequate understanding to consent. The subject will be expected to explain the study purpo se in his or her own words. Once consented, patients will be assigned a randomized and unique study identification number. After enrollment, patients will proceed with their scheduled surgery as planned The following information will be collected from the Clinical Patient ID Diagnosis Indication for revision Preoperative observations Previous surgical intervention Previous surgical instrumentation Previous surgery type and number and previous implants used Current laboratory test collected, including CBC, ESR, CRP, and Serum Ion levels Age Gender Demographics Pain Radiographic imaging Medical history Pathology results Histology results Bacterial culture results

PAGE 37

37 As part of the sta ndard of care, tissue from the peri hardware and control locations are normally removed through the course of surgery. A histological analysis is part of the standard of care for revision surgeries. The remainder of the tissue is normally discarded. We will collect this normally discarded tissue to perfo rm ICP MS/AES analysis. The normally discarded peri har dware tissue and the control tissue will be collected at the time of the surgery. No more than ten grams (approx. 10 cm3) of the peri hardware tissue, that is normally discarded, will be collected from the local area around the orthopedic implants. No more than te n grams (approx. 10cm3) of the control tissue, normally removed in the course of surgical exposure, will be collected from the subcutaneous and muscular tissue at least 1 spinal motion segment above or below the previous instrumentation placement. In the operating room, on a non sterile surface covered by a medium drape, these samples will be separated into two containers immediately labeled with the study ID number either the word control or peri hardware and the destination of the sample. At least a two cu bic centimeter sections of the control and peri hardware samples will be placed in separate containers with the appropriate labelling and the destination Huffman Hazen laboratories for ICP MS/AES analysis. Per the standard of care tissue samples are sent for pathology and histology analysis. The results of pathology and histology will be accessible to the research staff through the medical records because this analysis is part of the standar d of care. Materials Specific Aim Two 70% Isopropyl Alcohol 4x4 gauze pads Sharpie

PAGE 38

38 LED Loupe Nitrile Gloves Medium Drape DSLR Camera Methods Specific Aim Two All explanted hardware will undergo non destructive and destructive testing. For the nondest ructive testing, photo documentation of all components prior to removal, as well as after removal, in the as received condition will be taken. The research team will wear Nitrile gloves while handling the hardware. The hardware removed during the surgery w ill be placed on the nonsterile surface covered by the medium drape and cleaned using 4x4 gauze pads and 70% Isopropyl Alcohol. An initial examination of the hardware will be done in the operating room using the LED Loupe. The hardware will then be place d in a specimen container, labelled with the study ID number and the word Hardware The sample will be taken for further testing. Using the DSLR camera the hardware will be reimaged and saved onto a password protected computer. In a certified lab, we wi ll conduct optical microscopy and scanning electron microscopy on areas of interest as well as energy dispersive spectroscopy (EDS) for elemental composition estimates under the direction of the research staff. Destructive testing will include physical met allography. Materials Specific Aim Three Peri hardware and Control Tissue Samples Results from Huffman Hazen Laboratories Laptop with Excel

PAGE 39

39 Methods Specific Aim Three Inductively coupled plasma mass spectroscopy atomic emission spectroscopy (ICP MS/AES): The minimum two gram tissue samples will be sent to Huffman Hazen Laboratories for ICP MS/AES analysis. Sample containers will be prepared by Huffman. The tissue sample will be placed in the specified sample container label only with the assi gned patient ID a nd control or peri hardware. Samples, method blanks, and method blank spikes will be digested in entirety with mixed acids including reflux with boiling perchloric acid to ensure complete oxidation of all organic material, plus hydrofluori c acid to insure solubilization and stabilization of titanium. Digestion solutions will be further diluted and analyzed by ICP AES and ICP MS. Metal concentrations will be reported on the as received weight basis and the dried sample weight basis. The proc ess of ICP MS/ AES will completely destroy the tissue sample. Results from ICP AES and ICP MS will be formatted into an excel spreadsheet. Control data will be pulled from previous literature Materials Specific Aim Four Computer Excel Software Metho ds Specific Aim Four The voltage generated by the corrosion products will be calculated using the Nernst equation, results from ICP MS/AES and normal ion concentration values found in the literature. For this model there are several assumptions that h ave to be made. This equation is only looking at voltage and therefore it is assumed that the charge carriers are moving independently of cells and tissue. For simplicity we assume no chemical

PAGE 40

40 reactions are occurring between the charge carriers and physi ological environment and we neglect the metal ions that would be lost due to repassivation. Finally this is assumed to be a localized response. This simple corrosion cell models the reaction that is assumed in this analysis. The values used for z, valency, are the most common oxide form of the ions. R is the universal constant. The Nernst equation uses a ratio of ion concentration outside of the cell and inside the cell to determine voltage. In this set up we will assume that the metal ion concentrations collected from ICP MS/AES data are corrosion byproducts from the metal implant and are therefore extracellular. All known values of metal ions found in the literature are assumed to make up the intracellular concentrations. Al, Mo, V, Co, content in the surrounding tissue, are 8.4 4.8 ug/g; 1.61 1.41 ug/g; 0.06 0.03 ug/g; 1.35 1.97 u g/g respectively (Vincent 2000; Pasha et al. 2008; Farah et al. 2010; Harrington et al. 2014) whole blood and that was 0 .00072 ug/ g (Sarmiento Gonzlez et al. 2008) Surrounding Tissue (Cathode) Metal Instrumentation (Anode) e M + O 2 OH Cl Figure 4. Schematic representation of the corrosion cell generated in vivo. As the metal instrumentation corrodes electrons followed by positively charged metal ions flow to the tissue which acts as the cathode. Negatively charged species like oxygen and hydroxide, then m ove back to the metal instrumentation acting as the anode.

PAGE 41

41 Figure 5. The Nernst equation used for this calculation. R, T, and F are all know constants. The valency, z, used is the most common oxide state of each metal. The extracellular metal ion concentra tions are the values from the experimental data. The intracellular metal ion concentrations are pulled from previous literature. Patient demographic data will be summarized into a table including, gender, age, duration of previous implant, location of th e revision procedure, reason for revision based on inclusion criteria, and the presence or absence of metal staining. Mean age, duration in situ, and length of revision procedure (measured by the number of vertebral levels included in the surgery) will be calculated from this table. All retrieved hardware will be documented and the type of metal and type s of corrosion and wear identified will be summarized in a table. This summary will include the presence or absence of corrosion and wear. A quantity o r measurement of corrosion and wear will not be reported. Histological slides will be analyzed and a gross description of the tissue will be recorded. When applicable the quantity of lymphocytic infiltrate, types of inflammation, and quantity of necrosis will be reported A description of the common themes seen in the histological slides will be presented.

PAGE 42

42 R ESULTS N ine patients undergoing revision spine surgery in the lumbar and/ or thoracic r egion were enrolled and consented into th is study. Patients were enrolled and consented following the procedure defined in the approved IRB protocol ID 16 1176 In th e study 6/9 patients were male. The average age of participants in the study was 66 12.7 years The duration of previous instrumentation in situ ranged from 6 weeks to 17 years with an average duration of 5.03 5. 7 years. The size of the revision procedure as measure d by number of vertebral levels revised ranged from 2 15 levels with an average length of 6 levels Patients displayed a combination of indications for revision based on the inclusion criteria. All patients had pain of the spine, 3 patients had pseudoarthrosis, 3 patients had broken hardware, 2 patients had non union, and 1 patient had loose hardware. In two of the patient s with broken hardware the fractured instrumentation was not removed as it was part of a larger construct that did not need to be revised. This information is summarized in Table 1. Table 1. Patient demographic information including gender, age, duration of previous hardware in situ, location of the revision, and indications for the revision based on the inclusion criteria. Gender Age Duration in situ (in years) Location of revision Indications for Revision Metal Staining M 39 2.5 L4 S1 Pain N F 64 .115 T4 Pelvis Pain N F 75 12 C2 T4* Pain Y F 58 2 L4 S1 Pain N M 77 17 T4 L1 Pain, Hardware Failure N M 82 5 T10 Pelvis Pain, Hardware Failure, Non union, Pseudoarthrosis Y M 70 .6667 L5 S1 Pain, Non union, Pseudoarthrosis, Loosening of the instrumentation Y M 67 3 L4 Iliac Pain N M 62 3 L4 S2 Pain, Hardware Failure, Pseudoarthosis Y

PAGE 43

43 *Hardware was removed from T4 T8 and the thoracic spine was part of the revision surgery therefore the inclusion criteria for the study were met. All explanted hardware was documented and photographed. The type of metal and types of corrosion and wear were identified through visual inspection. Information about the retrieved har dwa re is summarized in Table 2. All explanted hardware was made from cpTi or Ti6Al4V with one set having CoCr tulips on the screws All of the hardware showed evidence of fretting wear, 8/9 samples showed evidence of pitting corrosion, and 8/9 showed evidence of the degradation of the oxide layer followed by repassivation as evidenced by discoloration of the oxide layer Images of each mode of corrosion and wear are found in Fig 5. Table 2. From the retrieved instrumentation the type of metal and types of corrosion and wear seen on the hardware through visual inspection were recorded Only the presence or absence of a type of corrosion or wear was recorded. Type of Metal Pitting Fretting/ Wear Spalling/ Galling Crevice Galvanic Degradation of the Oxide Layer / Repassivation Fracture 64Ti (screws) cpTi (rods) Y Y Y N N Y N 64Ti Y Y N N N Y N 64Ti Y Y N N N Y N cpTi rods CoCr head and cpTi screw Y Y N N N N N 64Ti Y Y N N N Y N 64Ti Y Y N N N N Y lowest level s crews fractured due to corrosion fatigue 64Ti Y Y N N N Y N 64Ti N Y N N N Y N Ti Y Y N N N Y N

PAGE 44

44 Figure 6 (a) E vidence of pitting corrosion. (b) Fretting wear has the shiny appearance along the rod caused from micromotion between the connector and the rod. (c) Evidence of repassivation The discoloration along the screw indicates reformation of the oxide layer at a different thickness. (d) Fracture between the threading and the head of the screw. Histological slides were prepared for eight of t he nine patients in the study. This lab is ordered by the surgeons and can be omitted if adverse reactions are not suspected or no abnormal tissue seen during the revision procedure. Results of this analysis were From the histological analysis quantity of inflammatory infiltrate was determined for two patients. Approximately 1 5% of the tissue contained inflammatory infiltrate with the appearance of chronic inflammation. The quantity of necrotic tissue was measur ed for two patients and ranged from 0.5 75% of the tissue sample. The results of the analysis revealed two common themes; foreign body giant cell reactions and histiocytes trapped in fibrous tissue (Fig 4). These observations are non specific. The prese nce of particles and debris (a) (b) (c) (d)

PAGE 45

45 was not visible in all of the samples; however, some foreign debris could have been removed during the processing and preparation of the histological slide. a) b) c) d) Figure 7 All images are of hematoxylin and eosin stained slides taken at 400x magnification. A) Foreign body giant cell reaction. B) Particle laden histiocytes trapped in fibrous tissue. C) N ecrotic skeletal muscle and necrotic fibrovascular connective tissue D) T ypical look of metallosis. ICP MS/AES data was collected for all nine sample however results are only available for three samples. The results of ICP MS/AES are summarized in Table 3. Ti concentrations were elevated 10 5 10 7 0.00072 ug/g, in all three patients with Ti instrumentation (Sarmiento Gonzlez et al. 2008) Alloying agents such as vanadium (V) were eleva ted 10 lung tissue values of 0.06 ug/g (Farah et al. 2010) The mean voltage calculated using the Nernst equation was 329.6 95.7 mV

PAGE 46

46 Table 3. ICP MS/AES data measuring the metal ion concentrations found in the retrieved tissue samples. Values below the detection of I CP MS/AES are reported as < 0.1. As received weight (g) Dried weight (g) Solids (%w/w) Ca (%w/w) P (%w/w) Al (ug/g) Co (ug/g) Cr (ug/g) Fe (ug/g) Mo (ug/g) Ni (ug/g) Ti (ug/g) V (ug/g) 0.6848 0.2450 35.78 0.06 0.17 95 0.3 14 660 0.3 3.9 2024 39 1.8757 1.25645 66.92 0.1525 0.1625 2.75 <0.1 0.725 144.25 <0.1 0.2125 23.5 0.725 1.0051 0.4357 43.35 6.87 3.32 16 <0.1 8.3 364 0.5 4.6 491 5.8 D ISCUSSION Corrosion products from orthopedic spine implants alters the physiologic environment of local soft tissue, by generating a voltage, leading to cell necrosis, identified through histology; propagating a pathological response that may lead to poor health outcomes. While the results of this study d id no t prov e th is hypothesis, this work connects material scien ce engineer ing and medicine The corrosion of metals in vitro is a well known fact among material scientists and engineers yet there is no urgency to correct or solve this issue I dentify ing a cause and effect relationships between corrosion and altered tissue physiology and a ssociating these findings with clinical data and patient outcomes has created the need to find new materials for orthopedic implants. Based on this research there were trends identified that require further exploration. Co rrosion and wear were identified on all retrieved hardware, indicating a cause and effect relationship between the metal instrumentation and elevated metal ion concentrations in the surrounding tissue. All retrieved hardware was made of Ti alloy. The cla ims that Ti is corrosion resistant and biocompatible have increased the popularity and use of this alloy in recent years yet research has proven these claims to be inaccurate (Rahul Bhola et al. 2011 ; Ehrensberger, Siva n, and Gilbert 2010) Damage to the oxide layer dramatically decreases biocompatibility and increase s corrosion of the Ti alloy (Ehrensberger, Sivan, and Gilbert 2010)

PAGE 47

47 T his study provide s a first step in connecting previous in vitro results to clinical applications. The voltage generated by corrosion products calculated using the Nernst equation is within the voltage viability range of Ti alloy 3 00mV 1000mV, established by Ehrensberger, et al., therefore the absence of necrotic tissue in the histological analysis is not surprising (Ehrensberger, Sivan, and Gilbert 2010) Howe ver, these voltages are elevated 3x the resting membrane potential for muscle cells and could be a contributing factor to the altered physiology of the surrounding tissue (Hopkins 2006) A n abnormal appeara nce of the tissue was identified in all histological samples. The dense collagenous fibrous tissue seen in all histological samples, is a pathological response not seen in healthy tissue F o reign body giant cell reactions and histiocytes trapped in the fibrous tissue sequestering black material in the cytoplasm were the most common themes The presence of these cells indicates a response to foreign material in the tissue. These observations are not described as metal hypersensitivity or ALVAL and no indication of these reactions was identified through this analysis The clinical consistenc y among all nine parti cipants was the symptom of pain. The variation within patient demographic information reflects the vast population needin g surgical intervention and some of the difficulties understanding this phenomenon. Limitations The small sample size limited the analysis that could be completed in this study The large standard deviations in the results can be attributed to the small pa tient population. F actors that could be used in correlation studies such as weight, height, BMI, comorbidities, serum and urine metal ion concentrations, and CBC were not include d A

PAGE 48

48 larger sample size is critical in furthering the potential results of t his research and is necessary to establish statistical relevance. The lack of variation among the retrieved hardware samples limited the connections that could be made to previous in vitro work and the scope of the histological analysis. All retrieved instrumentation was made of Ti alloy which has a different cell viability voltage range than CoCrMo alloy (Ehrensberger, Sivan, and Gilbert 2010; Haeri and Gilbert 2013) The material of the construct may influence the voltage that is genera ted, therefore impacting the results of the histological analysis. Future research should separate the data by metals us ed in the construct to observe differences among voltages and tissue samples. A larger sample size and greater variability of hardware composition would be required to ensure statistical and clinical relevance in future research Quantifying the amount of corrosion and wear present on removed hardware cannot accurately be obtained without original manufacturing drawings. This factor lim ited the analysis to qualitative data through visual observation Obtaining manufacturing drawings is difficult as these are often proprietary docume nts and may pose an issue in future research Using different methods of observation with higher magnific ation to look at factors like grain shape and size would expand the data set to include things such as composition and manufacturing history of the metal. There was difficulty collecting both the control and peri hardware tissue samples. The comparative analysis between the two samples could not be completed however, a larger sample size would be necessary to establish statistical significance between the se two This issue was predicted before sample c ollection began. For future research studies

PAGE 49

4 9 looking to comp are the metal ion concentration a tissue biopsy of a patient with no metal instrumentation in their body would be needed for a control. In this study t he unknown location of the tissue sample s prevented the inclusion of a distance or diffusion factor To control for these factors in future research an exact distance from the metal instrumentation and location of the peri hardware sample would be required. There are several assumptions that were used to calculate the voltage generated by corro sion products. As an initial test this simplified model was sufficient to connect in vitro results to clinical application s Future research should look to more complex models that include a time or distance factor as well as certain chemical reactions that occur between the metal ions and surrounding physiological envir onment Metal ions that repassivate were neglected in this model as well as metal ions and particles that are processed and removed from the body. Future Research There are several ave nues for future research in this field. Takahashi reported that myelograms were the only useful images in predicting metallosis (Takahashi, Delecrin, and Passuti 2001) Developing a radiograph ic profile of patients with metallosis using x ray, MRI, and CT images would improve the ability to predict and diagnose metallosis An MRI profile would be beneficial for determining soft tissue appearance while CT and x ray would useful for examining th e metal the metal bone interface and the condition of the bone. The occurrence of an infection in patients wi th metallosis is something that was observed throughout this study The association of infection and metallosis could be useful for the future diagnosis of this physiological response. The altered physiology,

PAGE 50

50 tissue fibrosis, and in some cases chronic inflammation indicate that the body is creating a pathological respon se. Patients with a known infection prior to revision surgery and patients that develop an infection post operative with signs of metallosis during the revision procedure would be ideal candidates to better understand the relationship between these two re actions Other factors that could influence corrosion such as lactic acid build up after surgery, the manufacturing process of the implants, and potential damage to the hardware during implantation need to be researched further. As previously stated, th e body is a hostile environment and things other than Cl and metabolites can interact with the oxide layer and underlying metal leading to corrosion. The m anufacturing methods, industry standards and testing methods used to produce and verify the instru mentation influence the integrity of the product. Unders tanding these methods and advanc ing them to reflect physiological conditions and real us e situations could be beneficial in reducing implant failure. The oxide layer coating the metal instrumentatio n is a crucial piece of biocompatibility and corrosion resistance. Education about this oxide layer and improvements to surgical techniques could decrease the amount of damage to the hardware during implantation. This knowledge could also benefit future researc h into new materials for use in orthopedic instrumentation. Research needs to focus on establishing a better understanding of m etal toxicity and the cellular pathways used to process, digest, and defend against metal ions and particles. Bhola et a l. stated that the stabilizing agents such as Al and V in metal alloys could have negative, even toxic, effects on the body (Rahul Bhola and Mishra 2012) Hanawa et al., concluded that even metal particles in concentrations of 100 ppb cannot be

PAGE 51

51 neglected when considered patient safety (Hanawa 2002) Metal ion concentration thresholds for metals used in orthopedic implants and the symptoms associated with these thresholds would benefit clinicians, engineers, and material scientists develop new patient diagnoses and new materials to be used in future orthopedic implants. While Yue et al., identified different cellular pathways used to digest particles of va rying size an explanation of why and physiological consequences was lacking (Yue et al. 2010) Results from this thesis, show cells containing metal particles often trapped in the fibrous tissue near the implant indicating that the normal process of digesting foreign particles is not possible. In vitro studies that investigate the process, limitations, and secondary defenses against these foreign particles could establish the basics of the physiological reaction seen from clinical samples. This information could also support or cont radict the current hypotheses of type IV metal hypersensitivity and ALVAL.

PAGE 52

52 C ONCLUSION The results of this study identified trends among patients undergoing revision surgery with previous instrumentation. The primary complaint of pain was consistent for all nine participants. Of the three patients that had ICP MS/AES results, elevated metal ion concentrations and elevated voltages, calculated using the Nernst equation, were unanimous These results build upon previous literature that established a cell viability voltage range for cells on metallic alloys. Trends identified through histological analysis indicate an abnormal and pathological response to foreign material in the body. None of these findings were consistent with the current hypotheses for metallosis including type IV metal hypersensitivity and ALVAL. The outcomes of this study connect clinicians and engineers and establish a need for research and invention of novel materials to be used in orthopedic instrumentation.

PAGE 53

53 REFERENCES Adogwa, Owoicho, Scott Parker, David Shau, Stephen Mendelhall, Oran Aaronson, Adjusted Life Year Gained of Revision Fusion for Lum Journal of Spinal Disorders and Techniques 28 (3): 101 5. doi:10.1097/BSD.0b013e318269cc4a. Bhola, R., S. M. Bhola, B. Mishra, R. A. Ayers, D. L. Olson, Donald O. Thompson, and OF A LOW MODULUS Ti 35.5Nb 7.3Zr 5.7Ta ALLOY IN A SIMULATED BODY 91. doi:10.1063/1.3592069. Bhola, Rahul, Shaily M Bhola, Brajendra Mishra, Reed Ayers, David L Olson, and r Response of Titanium and Its Alloys as Journal of Oral Implantology 37 (4): 387 99. doi:10.1563/AAID JOI D 09 00075.1. Alloy Using Various Surface Modificatio ns to Enhance Its Corrosion Resistance Materials Science Forum 706 709 (January): 105 12. doi:10.4028/www.scientific.net/MSF.706 709.105. Metal Clinical Toxicology 52 (8): 837 47. doi:10.3109/15563650.2014.944977. Generation Metal on Contact Dermatitis 66 (2): 55 62. doi:10.1111/j. 1600 0536.2011.01970.x. Denaro, Vincenzo, Nicola Papapietro, Alessandro Sgambato, Simona A. Barnaba, Laura Ruzzini, Barbara De Paola, Alessandro Rettino, and Achille Cittadini. 2008. Based Spine 33 (1): 8 13. doi:10.1097/BRS.0b013e31815e3978. Djurasovic, Mladen, Steven D. Glassman, Jennifer M. Howard, Anne G. Copay, and Related Quality of Li fe Improvements in Patients Spine 36 (4): 269 76. doi:10.1097/BRS.0b013e3181cf1091. der Cathodic Potential: The Relationship between Voltage and MC3T3 Preosteoblast Behavior on Electrically Polarized cpTi Journal of Biomedical Materials Research Part A 93A (4): 1500 1509. doi:10.1002/jbm.a.32622.

PAGE 54

54 Elves, M W, J N Wilson, J T Sca British Medical Journal 4 (5993): 376 78. Farah, Ibrahim O., Phuong X. Nguyen, Zikri Arslan, Wellington Ayensu, and Joseph A. of Differential Metal Loads in Normal Versus Biomedical Sciences Instrumentation 46: 404 9. Frigerio, Elena, Paolo D. Pigatto, Gianpaolo Guzzi, and Gianfranco Altomare. 2011. Contact Dermatitis 64 (5): 273 79. doi:10.1111/j.1600 0536.2011.01886.x. Prosthesis Components in Mixed and Similar Metal Combinations. The Effect of Journal of Biomedical Materials Research 27 (12): 1533 44. doi:10.1002/jbm.820271210. Gill, Harinderjit S., George Grammatopoulos, Stephen Adshead, Evagellos Tsialogiannis, Trends in Molecul ar Medicine 18 (3): 145 55. doi:http://dx.doi.org/10.1016/j.molmed.2011.12.002. Goldenberg, Yoni, Jin W. Tee, Cesar M. Salinas La Rosa, and Michael Murphy. 2016. European Spine Journal 25 (5): 1467 73. doi:10.1007/ s00586 015 4347 6. CoCrMo Alloy Using Time Lapse Live Acta Biomaterialia 9 (11): 9220 28. doi:10.1016/j.actbio.2013.06.040. George M. Langford, and Jeremy L. Gilbert. 2012. Induced Intrinsic Apoptosis and Oxidation Induced Necrosis on CoCrMo Alloy in Vitro Biomaterials 33 (27): 6295 6304. doi:10.1016/j.biomaterials.2012.05 .054. Implant Debris Spine October. Patients w J Bone Joint Surg Am 83 (3): 428 428. Journal of Biomedical Materials R esearch 49 (3): 353 61. in Vitro Science and Technology of Advanced Materials 3 (4): 289 295.

PAGE 55

55 Harrington, James M., Daniel J. Young, Amal S. Essader, Susan J. Sumner, and Keith E. Levin by ICP MS and ICP Biological Trace Element Research 160 (1): 132 42. doi:10.1007/s12011 014 0033 5. Health, Center for Devices and Radiologica on Metal Hip Implants Concerns about Metal on U.S. Food & Drug Administration https://www.fda.gov/Medicaldevices/productsandmedicalprocedures/implantsand prosthetics/metalonmetalhipimplants/ucm241604.htm. Continuing Education in Anaesthesia Critical Care & Pain 6 (1): 1 6. doi:10.1093/bjaceaccp/mki062. presented at the AAOS Webinar Series. Kelly, Michael P., Lawrence G. Lenke, Keith H. Bridwell, Rashmi Agarwal, Jakub Spine 38 (19): E1196 1200. doi:10.1097/BRS.0b013e31829e764b. Kirkpatrick, John S., Ramakrishna Venugopalan, Preston Beck, and Jack Lemons. 2005. J Spinal Disor Tech 18 (3): 247 51. American Academy of Orthopaedic Surgeons 3. Lohmann, Christoph H, Gurpal Singh, Hans Georg Willert, and Gottfried H Buchhorn. on Metal Total Hip Arthroplasty Regulates World Journal of Orthopedics 5 (5): 660 66. doi:10.5312/wjo.v5.i5.660. Lukina, Elena, Aleksandr Laka, Mikhail Kollerov, Mykhamad Sampiev, Peter Mason, Paul Wagstaff, Hilali Noordeen, Wai Weng Yoon, and Gordon Blunn. 2016. r Implantation of Titanium The Spine Journal 16 (3): 380 88. doi:10.1016/j.spinee.2015.11.040. Segment Degeneration and Revision Surgery after Circumferential Lumbar Fusion: Outcomes throughout 15 Years of Follow European Spine Journal 25 (5): 1550 57. doi:10.1007/s00586 016 4469 5. Release fro Journal of Materials Science: Materials in Medicine 13 (6): 583 588.

PAGE 56

56 Oliveira, Catarina A., Isabel S. Candelria, Pedro B. Oliveira, Antonio Figueiredo, and Filipe Caseiro Only in Patients with Metal on European Journal of Radiology Open 2: 3 6. doi:10.1016/j.ejro.2014.11.001. Spine Health http://www.spine health.com/glossary/orthopedic implants. Pasha, Qaisara, Salman A. Mali k, Javed Iqbal, Nazia Shaheen, and Munir H. Shah. 2008. Biological Trace Element Research 125 (1): 30 40. doi:10.1007/s12011 008 8158 z. Raja ee, Sean S., Hyun W. Bae, Linda E.A. Kanim, and Rick B. Delamarter. 2012. Spine 37 (1): 67 76. doi:10.1097/BRS.0b013e31820cccfb. Rio, Javier del, Jose Beguiristain, and Julio Duart European Spine Journal 16 (7): 1055 61. doi:10.1007/s00586 007 0311 4. Sarmiento Gonzlez, Alejandro, Juan Manuel Marchante Gayn, Jos Mara Tejerina Lobo, Jos Paz Jimnez, and Alfredo Sanz Medel. 2 Resolution ICP MS Determination of Ti, V, Cr, Co, Ni, and Mo in Human Blood and Urine of Analytical and Bioanalytical Chemistry 391 (7): 2583 89. doi:10.1007/s00216 008 2188 4. Shiroiwa, Takeru, Yoon Kyoung Sung, Takashi Fukuda, Hui Chu Lang, Sang Cheol to Pay (WTP) for One Additional QALY Gained: What Is the Threshold of Cost Health Economics 19 (4): 422 37. do i:10.1002/hec.1481. Spine 26 (13): 1495 99. Tezer, Mehmet, Unal Kuzgun, Azmi Hamzaoglu, Cagatay Ozturk, Fevziye Kabukcuoglu, Archives of Orthopaedic and Trauma Surgery 125 (6): 417 21. doi:10.1007/s00402 005 0802 x. Villaragga, Marta L., Peter A. Cripton, Stephanie D. Teti, Duane L. Steffey, Saki Spine 31 (21): 2454 62. The Journal of Nutrition 130 (4): 715 18.

PAGE 57

57 Watters, Tyler Steven, Diana M. Cardona, K. Sunil Menon, Emily N. Vinson, Michael P. Dominated Vasculitis American Journal of Clinical P athology 134 (6): 886 93. doi:10.1309/AJCPLTNEUAH8XI4W. Value in Health 12 (March): S5 9. doi:10.1111/j.1524 4733.2009.00515.x. Yee, Albert, Nana Adjei, Jennifer Do, Mi Clinical Orthopaedics and Related Research 466 (5): 1154 61. doi:10.1007/s11999 008 0194 7. Yue, Hua, Wei Wei, Zhanguo Yue, Piping Lv, Lianyan Wang, Guanghui Ma, and Zhiguo European Journal of Pharmaceutical Sciences 41 (5): 650 57. doi:10.1016/j.ejps.2010.09.006.