In-vivo Corrosion and Local Release of Metallic Ions from Vascular Stents into Surrounding Tissue

ABSTRACT: Objectives. To evaluate retrieved bare metal vascular stents and surrounding tissue. Background. Limited information is available regarding the condition of stent surfaces and their interaction with vascular tissue following implantation. Corrosion of stents presents two main risks: release of metallic ions into tissue and deterioration of the mechanical properties of stents which may contribute to fracture. Release of heavy metal ions could alter the local tissue environment leading to up-regulation of inflammatory mediators and promote in-stent restenosis. Methods. Nineteen cases were collected from autopsy, heart explants for transplant, and vascular surgery (23 vessels containing 33 bare metal stents). A method was developed for optimal tissue dissolution and separation of the stent/tissue components without inducing stent corrosion. When available, chemical analysis was performed to assess metallic content in both the control and dissolved tissue solutions. Electron microscopy and digital optical microscopy imaging were used to evaluate stents. Results. Twelve of the 33 stents showed varying degrees of corrosion. Metallic levels in the tissue surrounding the corroded stents were significantly higher (0.5–3.0 µg/cm² stent) than in control solutions (0–0.30 µg/cm² stent) and in tissue surrounding stents that did not undergo corrosion (0–0.20 µg/cm² stent). Conclusions. Corrosion of some retrieved stents is described which leads to transfer of heavy metal ions into surrounding tissue. The contribution of this metallic ion release to the mechanisms of in-stent restenosis as well as its effect on the mechanical properties of stents is unknown and requires further investigation. J INVASIVE CARDIOL 2010;22:528–535 Key words: stent fracture, intimal proliferation, inflammation, in-stent restenosis, stent corrosion, bare metal stents ———————————————————— The use of stents has lead to a reduction in the incidence of restenosis after percutaneous vascular intervention. Yet, clinical limitations persist, including in-stent restenosis (ISR) and thrombosis. In addition, angiographic stent fractures are detected in 1–3% of coronary cases and in up to 37% of peripheral cases.^1,2 Given the large number of stents implanted annually, investigating the potential factors involved in the mechanisms of ISR and stent fracture is of significant importance. Current stents are manufactured from corrosion resistant alloys which form protective oxide films insulating the bulk material from the corrosive physiologic fluid.^3–5 These properties result from optimal stent surface treatments during manufacturing that produce a uniform thin film. Stents are increasingly deployed in overlapping configurations as a treatment of ISR, leading to the risk of mechanical damage of the oxide film caused by micro-motion at points of overlap. Once the film is removed in a localized region, the underlying alloy becomes exposed to the physiologic fluid and may undergo corrosion. In addition, with several alloys available, overlapping stents of dissimilar materials is possible, leading to the risk of galvanic corrosion. Stent fracture may be a result of in vivo mechanisms that alter the mechanical properties of stent alloys such as: biomechanical effects and/or biocorrosion. Biomechanical forces can include overlapping stents under pulsatile flow, high vessel curvature, and presence of heavy calcification. Metal fatigue, with nickel-titanium strut fracture, has been reported for both thoracic and abdominal aortic stent grafts.⁶ Since fatigue is known to accelerate under the combined action of corrosion and cyclic loading,⁷ it is essential to understand the role that corrosion plays in potentially promoting such fractures. The corrosion resistance of bare metal stents (BMS) has been determined in vitro but information on the condition of stent surfaces and their interaction with surrounding tissues following implantation is scarce. This report describes stented vascular segments retrieved from patients, with the goal of assessing stent surface condition and its effect on the chemistry of the surrounding tissue. All the stents described in this report are BMS.

Methods

Retrieved specimens. Nineteen cases, consisting of 23 vessels with a total of 33 BMS, were obtained from autopsy, heart explants for transplant, and vascular surgery. The specimens were fixed in 10% neutral buffered formalin, radiographs were obtained, and available medical records were reviewed. Analysis of tissue and evaluation of medical records was approved by the University of Alabama at Birmingham Institutional Review Board (IRB Protocol: X060928009). Retrieved stents (n = 33) were manufactured from stainless steel (SS, n = 15), cobalt-chromium (CoCr, n = 10), nickel-titanium (NiTi, n = 4), a cobalt-based alloy called Elgiloy-tantalum (Elgiloy-Ta, n = 3) and tantalum (Ta, n = 1). Clinical case descriptions are presented in Table 1. Controls. Commercial “off-the-shelf” Multi-link Zeta^® SS, Radius^® NiTi, and Multi-link Vision^® CoCr stents (Abbott Vascular, Santa Clara, California) were used as controls to assess the effect of the chemical used for tissue dissolution and to ensure that it had no corrosive effect prior to its use on the retrieved stents. Control stents from each type were exposed to 1M solution of NaOH for 24 hours. Metallic levels in the solution were then measured and expressed in mass of leached metallic element per stent surface area (µg/cm). These data were compared to the data from dissolved retrieved tissue. All control stents had similar nominal oxide film chemistry as their corresponding retrieved stents (An Elgiloy-Ta control was not available and CoCr was substituted as a control due to its similar oxide composition). There was no available control match for the Ta stent, however, evidence shows that Ta does not corrode in concentrations of NaOH less than 3.0 M.⁸ For considerations such as polishing and surface finish, it was assumed that both retrieved and control stents followed the surface preparation guidelines described in the American Society for Testing and Material (ASTM) standard practice for surface preparation and marking of metallic surgical implants ASTM F-86-01.⁴Tissue dissolution. Retrieved vessels were placed in 1M solution of NaOH for 24 hours to dissolve away the tissue. Macroscopic observations. Control and retrieved stents were evaluated using scanning electron microscopy (SEM; Phillips 515, Eindhoven, The Netherlands) and digital optical microscopy (KEYENCE VHX-600). Chemical analysis. Metallic levels of titanium, chromium, cobalt, and nickel in control and dissolved tissue solutions were measured using a high resolution inductively coupled mass spectrometer (ICP-MS, Finnigan MAT) at the Trace Elements Laboratory, London Health Sciences Center, London, Ontario, Canada. In order to monitor metallic contamination throughout all steps, metallic levels in the blank tubes used to store the tissue as well as in samples of 1M NaOH solution were measured. Raw measurements were obtained in units of ppb (µg/L) and data are expressed in units of µg/cm² stent and µg/g tissue.

Results

A summary of the results for all 19 cases (23 vessels containing 33 BMS) is presented in Table 1. Analysis of eight cases is presented in this study. Chemical analysis of tissue. The levels of metallic ions measured in dissolved tissue (0.5–3.0 µg/cm² stent) were orders of magnitude higher than those measured in controls (0–0.30 µg/cm² stent) after normalization, as shown in Figure 1A. The dissolution protocol was considered a successful non-corrosive approach for the stent alloys analyzed. When enough tissue was available around significantly corroded stents, metallic ion levels were normalized to tissue weight and measured up to 39 µg/g dry tissue compared to the zero values measured from non-corroded stents (Figure 1B).

SEM Observations

Controls. All stents (Figure 2) showed uniform surfaces following exposure to NaOH. Retrieved specimens. Twelve of the 33 retrieved stents showed varying degrees of surface corrosion. Representative examples of corroded and non-corroded cases are described below. Case 1. The NiTi stent exhibited a pattern of numerous small pits (pitting corrosion) within the alloy on the majority of the surface (Figures 3A and B). The overlapping SS stent was corrosion-free; however, rough irregular edges were observed (Figures 3C and D). Surrounding tissue contained significant levels of titanium and nickel (Figure 1). Case 2. The NiTi stent demonstrated regions of pitting corrosion on the majority of the surface (Figure 3E). Surrounding tissue contained significant levels of titanium and nickel (Figure 1). Case 3. The SS stent demonstrated localized pitting corrosion on its luminal surface (Figures 4A and B). The remainder of the stent surface was unaltered. Surrounding tissue contained a significant level of nickel (Figure 1). Case 4. The SS stent demonstrated localized mechanical fretting at the overlap region. This type of damage occurs when there is a shearing force (against overlapping stent) and results in surface wear removal of the stent material along with the oxide film (Figure 4C). Higher magnification showed initiation of a chemical reaction demonstrated by the formation of small pits (Figure 4D). The Elgiloy-Ta stent exhibited two corrosion features: localized pitting at random regions (Figure 4E) and fretting at the crossing points of the braided stent wires (Figure 4F). Surrounding tissue did not contain measurable levels of metallic elements, although minimal surrounding tissue was removed during surgery (Figure 1A). The tissue analyzed was primarily composed of the thrombus that occluded the vessel. Case 5. The Ta stent demonstrated significant alterations consistent with galvanic corrosion due to overlap with the dissimilar SS stent (Figure 5A). Localized wear possibly due to fretting, was observed on the overlapped portion of the SS stent (Figure 5B) while the non-overlapped region was unaltered (Figure 5C). The tissue surrounding the Ta/SS stent overlap showed chromium levels three times higher than the non-overlap region (Figure 1A). Cases 6 & 7. No corrosion was observed on all stents (Figure 6). Metallic levels in surrounding tissue were within the range of control values (Figure 1A) and data normalization to tissue weight showed zero/negligible levels (Figure 1B). Case 8. Both stents were surrounded by heavy calcification deposits (Figures 7A and C). The left anterior descending (LAD) coronary artery stent had features of abrasion wear and corrosion on the surfaces overlapping the calcification (Figure 7B). The right coronary artery (RCA) stent did not undergo corrosion at the regions which were not covered by calcification however it underwent multiple single strut fractures within heavily calcified areas (Figure 7D). Information on tissue chemistry was not available.

Discussion

Corrosion has been found on other metallic medical devices. Reports of bio-corrosion of NiTi in explanted aortic endografts raised the concern of the stability of stent wires in-vivo.⁹ This corrosion varied from the formation of micro-size pits on the surface, to large irregular surface alterations, and fracture. Significant corrosion has also been demonstrated on orthopedic and dental devices, affecting their durability and the choice of implants.^10,11 This study demonstrates that stent corrosion occurs in vivo and is associated with release of heavy metal ions into adjacent tissue. Twelve of the 33 retrieved BMS demonstrated corrosion and 2 fractured. When available, tissue chemistry analysis data revealed increased levels of metallic elements as a result of corrosion. One stent in case 1 exhibited a roughness on some of its surfaces (Figures 3C and D) indicating incomplete surface polishing during manufacturing. Types of corrosion. Corrosion on stent alloys occurs when the protective oxide film breaks down.¹² This can be a result of: 1) a local chemical/electrochemical attack or 2) mechanical damage to the surface. It has been shown that the dissolution rate of the oxide film is accelerated by the presence of amino acids, proteins, and chloride ions.^13,14 The oxide film has the capability of reforming once broken down; however this is dependent on the amount of oxygen available. The concentration of dissolved oxygen in the body is one fourth of that in the air, and therefore can lead to a delay in the repair of this film in vivo.¹⁵ The data in our study show that, in the absence of mechanical damage to the oxide film, corrosion appeared to be diffuse on NiTi stents (Figures 3 A, B and E) and more localized on SS and Elgiloy-Ta stents (Figures 4 A, B and E). When there is micro-motion between the stent and an overlapping hard surface, mechanical damage of the oxide film occurs. This can happen either pre-implantation (by scratches or machine marks during manufacturing) or post-implantation (by overlapping stents or surrounding hard calcifications). This damage can expose the underlying metal to corrosion.¹⁶ Accordingly, fretting damage on the SS stents found in Cases 4 & 5 (Figures 4C and D, 5B) and the Elgiloy-Ta stent (Figure 4F) from Case 4 resulted in initiation of corrosion. Four types of stent alloys were analyzed in this study; SS (68% iron, 18% chromium, 14% nickel), NiTi (55% nickel, 45% titanium), Ta (100% tantalum), and Elgiloy-Ta (Elgiloy: 40% cobalt, 20% chromium, 17% iron, 16% nickel, 7% molybdenum; with a 100% tantalum core. Metallic alloys are characterized by the degree of their resistance to corrosion according to their voltage potential. When metals with different voltage potentials are in contact in the same electrolyte, galvanic corrosion may occur, characterized by ion migration from the more active metal.¹⁷ This concept is represented in case 5; according to the galvanic series of metals in sea water,¹⁸ Ta is at opposite poles from passive SS. This difference leads to accelerated corrosion of the more active metal (Ta), consistent with findings shown in Figure 5A. Ideally, stents in-vivo should exhibit the same smooth surfaces seen on both the control (Figure 2) and retrieved stents which did not undergo corrosion (Figure 6). Considering the miniature size of stents, any type of surface alteration is considered significant and may potentially alter surrounding tissue chemistry and/or modify the mechanical properties of stents. Altered tissue chemistry and local responses. Increased levels of metallic elements (as high as 39 µg/g dry tissue) were measured in tissue surrounding corroded stents compared to negligible levels in tissue surrounding non-corroded stents (Figure 1). Corroded SS stents mainly released nickel and chromium while corroded NiTi stents released nickel and titanium. Iron was detected in all retrieved tissue but was not reported due to its presence in blood hemosiderin. Once corrosion byproducts are released, they may react with oxygen or form complexes with bio-molecules.¹⁹ Metal ions modulate biological responses including inflammation at implant sites.^20–22 For example, analysis of soft tissue collected from around titanium spinal implants showed an average of 33.65 µg/g of dry tissue from patients that developed pseudarthrosis, and metal particles were observed in the tissue coupled with a macrophage cellular response.²³ In addition, animal data show that nickel concentrations greater than 25 µg/g dry tissue are associated with severe inflammation and necrosis.²¹ In-vitro studies have shown that very low concentrations of nickel and cobalt ions which show no influence on cell morphology, cause significant expression of endothelial cell adhesion molecule as well as adhesion of polymorphonuclear neutrophil granulocytes to endothelial cells in vitro.^24,25 Similarly they may activate at least two endothelial cell signal transduction pathways by up-regulating cytokines, and induce expression of adhesion molecules.²⁶ A recent study identified a molecular pathway in which exposure of vascular smooth muscles cells to stainless steel ions causes stimulation of their synthetic phenotype via an increase in the expression of thrombospondin-1 combined with a dependant increase in transforming growth factor-b activity, thus suggesting that stent corrosion may be a key contributor to the mechanisms of IRS.²⁷ The present report suggests that when stents corrode in-vivo, it generates an active microenvironment possibly leading to ISR. Ion release is not a concern for the newly developed biodegradable iron and magnesium stents since both elements have a high tissue tolerance, are biocompatible with endothelial and smooth muscle cells, and may be designed to slowly degrade in body fluids. Therefore, this type of metallic contamination poses no notable risks for inflammatory reactions. Clinical implications. The three cases (cases 1, 2 and 4) that demonstrated diffuse corrosion had symptoms of ISR or stent occlusion which led to stent removal during vascular surgery. Corrosion and metallic ion release has been implicated in pre-clinical and clinical ISR. Recent studies have shown that the placement of gold-coated SS stents in coronary arteries was associated with a considerable increase in the risk of ISR.²⁸ Severe surface defects on the gold coating were reported on “as received” stents, causing galvanic corrosion of the underlying SS during electrochemical in-vitro tests. It has been suggested that these findings may have contributed to the significant restenosis associated with these stents in-vivo.²⁹ There is also evidence that late ISR occurs beyond 6 months of BMS placement characterized by prominent infiltration of lipid-laden macrophages with strong collagen-degrading matrix metalloproteinase immunoreactivity around stent struts.³⁰ Data from clinical series^1,31,32 and autopsy studies³³ suggest a greater frequency of stent fracture associated with longer duration of implant, vessel motion, bifurcation deployment and complex lesions such as those containing multiple overlapping stents and diffuse calcification; all of which are risk factors for causing mechanical damage to the oxide film leading to corrosion. These fractures have been associated with restenosis, thrombosis, or vessel occlusion. All stents surrounded by heavy calcification, similar to the ones shown in Figures 7A & C, either underwent localized wear and corrosion (n = 3) as in Figure 7B or experienced fracture (n = 2) as shown in Figure 7D. The hardness of calcium deposits obtained from abdominal aortic aneurysms is similar to that of metals such as nickel and iron (both elements present in stent alloys).³⁴ Assuming coronary calcification has comparable hardness, these deposits could cause the fretting corrosion observed. Although the current data show no direct link between stent corrosion and fracture, formation of surface alterations is considered a fatigue-inducing factor. When combined with other biomechanical stresses (e.g., high curvature), corrosion could accelerate the fracture process. Stent fracture may lead to the transfer of metallic debris into tissue since it occurs as a result of fatigue which causes irregular fracture surfaces (Figure 7D). Further studies are required to determine whether these present findings are true of the broader population of stents after implantation and assess their incidence according to stent type (bare metal versus polymer-coated drug eluting). The Food and Drug Administration specifies consistent testing criteria of all manufactured stents for corrosion resistance according to guidelines in the ASTM F2129-06 standard.³⁵ If the present findings are confirmed, additional testing guidelines should be considered such as assessing stent surface conditions following implantation in animals for 6–12 months. Study limitations. Since this study involved a small number of stents solely obtained from autopsy, heart explants for transplants, or vascular surgery cases, these results may not be representative of all devices produced by all manufacturers. Complete medical records were not available for all cases. When precise matching of control stents to explanted stents was not available, stents with closely similar surfaces were used. Due to technical limitations, measurements of metallic ion levels were not obtained from all dissolved tissues and will be the topic of a follow-up study. Corrosion was not assessed on the two drug eluting stents reported in the Table due to presence of the polymer coating. Lastly, since tissue was dissolved for stent retrieval and examination, histopathologic assessment of the stent-tissue interface was not possible.

Conclusion

These results provide direct evidence that corrosion of stents occurs in vivo, and is associated with release of heavy metal ions into vascular wall. The pathophysiologic significance of these specific findings is not known, however the orthopedic implant literature suggests a link to inflammation. Further evaluation of explanted stents is currently being performed. A better understanding of the influence of metallic ions/particulates on local tissue responses is needed.

References

1. Aoki J, Nakazawa G, Tanabe K, et al. Incidence and clinical impact of coronary stent fracture after sirolimus-eluting stent implantation. Catheter Cardiovasc Interv 2007;69:380–386. 2. Scheinert D, Scheinert S, Sax J, et al. Prevalence and clinical impact of stent fractures after femoropopliteal stenting. J Am Coll Cardiol 2005;45:312–315. 3. Singh R, Dahotre NB. Corrosion degradation and prevention by surface modification of biometallic materials. J Mater Sci Mater Med 2007;18:725–751. 4. ASTM-F86. 1995. Standard Practice for Surface Preparation and Marking of Metallic Surgical Implants. Annual Book of ASTM Standards: Medical Devices and Services. Volume 13.01. Philadelphia: American Society for Testing and Materials, p 6–8. 5. Thierry B, Tabrizian M, Trepanier C, et al. Effect of surface treatment and sterilization processes on the corrosion behavior of NiTi shape memory alloy. J Biomed Mater Res 2000;51:685–693. 6. Jacobs TS, Won J, Gravereaux EC, Faries PL, et al. Mechanical failure of prosthetic human implants: A 10-year experience with aortic stent graft devices. J Vasc Surg 2003;37:16–26. 7. Morita M, Sasada T, Nomura I, et al. Influence of low dissolved oxygen concentration in body fluid on corrosion fatigue behaviors of implant metals. Ann Biomed Eng 1992;20:505–516. 8. Gad Allah AG, Badawy WA, Rehan HH. Stability of anodic tantalum oxide films in NaOH solutions J Appl Electrochem 1989;19:768–776. 9. Heintz C, Riepe G, Birken L, et al. Corroded nitinol wires in explanted aortic endografts: An important mechanism of failure? J Endovasc Ther 2001;8:248–253. 10. MacDonald SJ, Brodner W, Jacobs JJ. A consensus paper on metal ions in metal-on-metal hip arthroplasties. J Arthroplasty 2004;19(8 Suppl 3):12–16. 11. Lemons JE, Louis PJ, Beck P, et al. Biocompatibility of nanostructured diamond surfaces for TMJ articulation. 2003; San Antonio, TX. p 0846. 12. Rondelli G, Vicentini B. Localized corrosion behaviour in simulated human body fluids of commercial Ni-Ti orthodontic wires. Biomaterials 1999;20:785–792. 13. Williams RL, Brown SA, Merritt K. Electrochemical studies on the influence of proteins on the corrosion of implant alloys. Biomaterials 1988;9:181–186. 14. Brown SA, Farnsworth LJ, Merritt K, Crowe TD. In vitro and in vivo metal ion release. J Biomed Mater Res 1988;22:321–338. 15. Black J. 1984. Biological Performance of Materials. Plenum, New York. 16. Brown SA, Hughes PJ, Merritt K. In vitro studies of fretting corrosion of orthopaedic materials. J Orthop Res 1988;6:572–579. 17. Barbosa MA. Corrosion mechanisms of metallic biomaterials. Barbosa, M.A. Porto: Elsevier Science Publishers B.V. 1991:227–258. 18. RS-TR-67-11. Army Missile Command Report: Practical Galvanic Series. 19. Kocijan A, Milosev I, Pihlar B. The influence of complexing agent and proteins on the corrosion of stainless steels and their metal components. J Mater Sci Mater Med 2003;14:69–77. 20. Santin M, Mikhalovska L, Lloyd AW, et al. In vitro host response assessment of biomaterials for cardiovascular stent manufacture. J Mater Sci Mater Med 2004;15:473–477. 21. Wataha JC, O'Dell NL, Singh BB, et al. Relating nickel-induced tissue inflammation to nickel release in vivo. J Biomed Mater Res 2001;58:537–544. 22. Messer RL, Mickalonis J, Lewis JB, et al. Interactions between stainless steel, shear stress, and monocytes. J Biomed Mater Res A 2008;87:229–235. 23. Wang JC, Yu WD, Sandhu HS, et al. Metal debris from titanium spinal implants. Spine 1999;24:899–903. 24. Klein CL, Kohler H, Kirkpatrick CJ. Increased adhesion and activation of polymorphonuclear neutrophil granulocytes to endothelial cells under heavy metal exposure in vitro. Pathobiology 1994;62:90–98. 25. Klein CL, Neider P, Wagner M, et al. The role of metal corrosion in inflammatory processes: Induction of adhesion molecules by heavy metal ions. Journal of Materials Science: Materials in Medicine 1994;5:798–807. 26. Wagner M, Klein CL, van Kooten TG, Kirkpatrick CJ. Mechanisms of cell activation by heavy metal ions. J Biomed Mater Res 1998;42:443–452. 27. Pallero MA, Talbert Roden M, Chen YF, et al. Stainless Steel Ions Stimulate Increased Thrombospondin-1-Dependent TGF-Beta Activation by Vascular Smooth Muscle Cells: Implications for In-Stent Restenosis. J Vasc Res 2009;47:309–322. 28. Pache J, Dibra A, Schaut C, et al. Sustained increased risk of adverse cardiac events over 5 years after implantation of gold-coated coronary stents. Catheter Cardiovasc Interv 2006;68:690–695. 29. Shih CC, Shih, CM, Chou, KY, et al. Electrochemical and SEM evaluation of gold-coated stents in vitro. J Electrochem Soc 2007;154:C326–C330. 30. Inoue K, Abe K, Ando K, et al. Pathological analyses of long-term intracoronary Palmaz-Schatz stenting; Is its efficacy permanent? Cardiovasc Pathol 2004;13:109–115. 31. Lee MS, Jurewitz D, Aragon J, et al. Stent fracture associated with drug-eluting stents: Clinical characteristics and implications. Catheter Cardiovasc Interv 2007;69:387–394. 32. Sianos G, Hofma S, Ligthart JM, et al. Stent fracture and restenosis in the drug-eluting stent era. Catheter Cardiovasc Interv 2004;61:111–116. 33. Nakazawa G, Finn AV, Vorpahl M, et al. Incidence and predictors of drug-eluting stent fracture in human coronary artery: A pathologic analysis. J Am Coll Cardiol 2009;54:1924–1931. 34. Marra SP, Daghlian CP, Fillinger MF, Kennedy FE. Elemental composition, morphology and mechanical properties of calcified deposits obtained from abdominal aortic aneurysms. Acta Biomater 2006;2:515–520. 35. 2005.CDRH Guidance. Non-clinical tests and recommendations for interventional stents and associated delivery devices. US Food and Drug Administration: https://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/ucm071863.htm. p 10–11.

———————————————————— From the University of Alabama at Birmingham, Birmingham, Alabama and ^†Cyprus University of Technology, Nicosia, Cyprus. Disclosure: The authors received a Cyprus Research Promotion Foundation Grant IPE/STOXOS/0308/04 for partial support of the project. Manuscript submitted April 8, 2010, provisional acceptance given June 4, 2010, final version accepted August 3, 2010. Address for correspondence: Brigitta C. Brott, MD, The University of Alabama at Birmingham, Interventional Cardiology, FOT 907 510 20th Street So, Birmingham, AL 35294. Email: ­bbrott@uab.edu