Wear and corrosion of sliding counterparts of stainless-steel hip screw-plates

Wear and corrosion of sliding counterparts of stainless-steel hip screw-plates

Injury, Int. J. Care Injured 31 (2000) 85±92 www.elsevier.com/locate/injury Wear and corrosion of sliding counterparts of stainless-steel hip screw-...

931KB Sizes 0 Downloads 7 Views

Injury, Int. J. Care Injured 31 (2000) 85±92

www.elsevier.com/locate/injury

Wear and corrosion of sliding counterparts of stainless-steel hip screw-plates B.F. Shahgaldi*, J. Compson Orthopaedic Research (King's College), Rayne Institute, St Thomas' Hospital, Lambeth Palace Road, London SE1 7EH, UK Accepted 9 September 1999

Abstract Wear of the lag screw and barrel of stainless-steel sliding hip screw-plates (SHSP) produces particles and corrosion that jam the sliding screw. This alters the mechanics of the SHSP causing failure and diculty in removal. We examined 15 surgicallyremoved SHSP for the extent of wear and corrosion of metal, and damage to the surrounding tissue. SHSP implants were in place from 2 months to 11 years (average 38.9 months) and they were removed because of fracture non-union, implant fracture or pain. None was infected. Electron microscopy was used to examine the implant surfaces and to determine the chemistry of corrosion products both on the metal and in the tissue. The distal ends of all the lag screws and inside walls of the barrels showed wear and corrosion. The severity of corrosion increased with increasing duration of implant use. Surfaces of the inside walls of the barrels were rough from the manufacturing process. This had contributed to wear. Better manufacturing practice to improve surface smoothness of the lag screws and inside walls of the barrels is needed. Use of cobalt±chromium alloys would improve hardness and resistance to corrosion compared with the stainless steel presently used, which would be an advantage for sliding hip screw-plates in younger patients. Deliberate texturing of screws is counter productive and should be avoided. # 2000 Elsevier Science Ltd. All rights reserved. Keywords: Sliding hip screw-plate; Wear; Corrosion; Scanning electron microscopy; Bone resorption

1. Introduction Sliding hip screw-plates (SHSP) are popular devices for ®xation of hip fractures [1±7]. These devices are treated as permanent in the elderly and are only removed in young patients if there is pain or when failure occurs. Wear and corrosion of the sliding parts of dynamic hip screws present a unique problem: accumulated debris causes jamming of the lag screw inside the barrel. This alters the mechanics of the dynamic hip screw causing failure and diculty in removal. Wear and corrosion of lag screws and inside walls of SHSP barrels have not, to our knowledge, been reported previously, although corrosion of screw heads and their screw-hole counterparts of fracture plates are well documented [8±12]. Tissue necrosis, pain, in¯am* Corresponding author.

mation, allergic reaction and carcinogencity are attributed to the corrosion products of stainless steel [13± 17].

2. Materials and methods Fifteen surgically retrieved SHSP and tissue samples from eight cases were examined. SHSP implants were removed between 1996 and 1998. They were in place from 2 months to 11 years (average 38.9 months) and were removed because of fracture non-union (two, at 2 and 5 months), implant fracture (one lag screw fractured at 7 months and one plate fractured at 1 year), femoral head collapse or lag screw cut out (two, at 21 and 54 months). The remaining nine, that were in place for a minimum of 1 year and a maximum of

0020-1383/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 0 - 1 3 8 3 ( 9 9 ) 0 0 2 4 1 - 7

86

B.F. Shahgaldi, J. Compson / Injury, Int. J. Care Injured 31 (2000) 85±92

11 years, were removed because of persistent pain. None was infected. After removal each implant was washed in 100% alcohol and left on a clean surface to dry. Visible wear and corrosion of each component were recorded. All barrels were cut from their plates at the barrel±plate junction. Each barrel was then sawed longitudinally into two halves. As cutting progressed, debris generated from sawing was regularly brushed away and cut surfaces were washed with a jet of 100% alcohol, to avoid spread of debris into the barrel. The barrel halves were then macroscopically examined for corrosion and wear. All loose wear and corrosion products were carefully picked up on pieces of double sided adhesive tape and saved for subsequent examination. One half of each barrel was then mounted on an aluminium stub, coated with carbon for examination under a scanning electron microscope (SEM). The second half was carefully scraped with a soft nylon brush and rinsed in alcohol, then dried and examined by scanning electron microscopy to characterise its surface roughness. Five lag screws were also scraped, washed and examined microscopically to determine their surface roughness. The remaining lag screws were carbon coated and examined for corrosion. Bone and soft tissue samples still attached to the plates after alcohol washing were carefully removed from the metal. Surfaces that were in contact with the metal were then coated with carbon and examined under a scanning electron microscope. In addition tissue from the proximity of the implants were processed

for histology and 10 mm sections were examined by scanning electron microscopy [18]. The same SEM procedure was used for examination of metal surfaces and tissue samples. Secondary electron mode was used to examine the morphology of metal surfaces or tissue. Backscatter electron mode was used for locating corrosion products while their chemical composition was determined by Energy Dispersive X-ray Analysis (EDXA). An acceleration voltage of 15 keV, constant working distance, specimen tilt and X-ray detector position were maintained for all specimens. In all cases, X-ray spectra were obtained for 100 s. 3. Results All devices examined were of keyed type, all lag screws but one had two parallel ¯at surfaces which keyed with similar ¯at surfaces within the barrel. The plate±barrel angles of all devices were 1358. The three manufacturers devices that were examined by analysis had chemical compositions consistent with 316L stainless steel (Fig. 1). In ®ve implants the surfaces of the screws and outside of the barrels were textured. The outside of nontextured barrels were polished to a mirror ®nish (Fig. 1), but the inside walls of all barrels showed roughness consistent with machining. This was clearly observed in the specimens that had been in place for 2 and 5 months and in areas where no wear or corrosion had occurred. Lag screws of non-textured implants were smooth and appeared polished, but trade marks and

Fig. 1. Type of sliding hip screw-plate examined.

B.F. Shahgaldi, J. Compson / Injury, Int. J. Care Injured 31 (2000) 85±92

Fig. 2. (a) Corrosion and wear at the distal end of a 10 year old screw. (b) Morphologies of corrosion and wear (secondary electron micrograph, scale is marked by the 500 mm dotted line). Note proximity of manufacturers marking to the worn and corroded areas.

87

numbers were inscribed on the ¯at surfaces interlocking with the barrels. This made an otherwise smooth surface rough and had contributed to the wear and corrosion (Fig. 2). Gross wear, either all round or localised, was seen on the distal ends of all lag screws (Fig. 2). The wear was more extensive on textured surfaces. In early specimens localised pitting corrosion within otherwise clean wear tracks was seen on the distal end of lag screws. The extent and depth of wear, which appeared to have increased with increasing age of implants, could not be accurately determined in older specimens because of extensive corrosion of the worn areas. Inside the barrels, wear tracks were seen at the proximal end on and around ¯at keying surfaces (Fig. 3). Corrosion increased with increasing ages of implants: in early specimens localised corrosion in and around worn regions was seen, but in 10 and 11 year-old specimens a layer consisting of corrosion products and wear particles embedded in proteinous membrane lined the inside of the barrel. Surfaces underlying the corrosion layers were extensively roughened and contained numerous large pits (Fig. 3b). One lag screw of an implant had fractured leaving its distal half behind in the barrel. A 59 year-old woman with coxa vera of the hip su€ered a fracture neck of femur. The fracture did not heal and the implant fractured 7 months later. The lag screw half was welded to the inside wall of the barrel by accumulation of wear particles and corrosion at the interface between the screw and the barrel wall (Fig. 4). The lag screw was extensively worn all round at its distal end, but the barrel was worn mainly on one of its keying surfaces. Corrosion products consisted of separate layers with di€erent chemical composition. 316L stainless steel alloy contains around 58% iron, 18% chromium, 13% nickel and 2.5% molybdenum. Corrosion products consisted of separate iron or chromium rich layers in association with oxygen, calcium, phosphorous, and sulphur. Large amounts of nickel were detected in the corrosion ®lm lining the inside of the barrel and on the surface of the lag screw which fractured at 7 months (Fig. 4c). In other specimens nickel was persistently absent from the corrosion products. Corrosion debris was present both free and within cells in all tissue types. The size of particles were variable ranging from submicron to 50 mm (Figs. 5 and 6b). Chemically, the composition of the debris in all locations was similar to the composition of the corrosion ®lm on the metal. The surfaces of bone samples from two specimens that had been in contact with the metal contained bone resorbing cells burdened with chromium rich corrosion products (Fig. 6).

Fig. 3. (a) Worn and corroded areas on the ¯at key and the inside wall of barrel of a 4.5 year-old specimen (bottom of picture). (b) Morphology of wear on the ¯at key (top left) and the wall of the barrel (bottom right). (Electron micrograph, scale is marked by the 150 mm dotted line).

88 B.F. Shahgaldi, J. Compson / Injury, Int. J. Care Injured 31 (2000) 85±92

B.F. Shahgaldi, J. Compson / Injury, Int. J. Care Injured 31 (2000) 85±92

4. Discussion Three devices in our series were removed because of ®xation failure and non-union at 2, 5 and 7 months. Gross wear and corrosion were seen on these devices, as was expected. We were, however, surprised by the extent of wear and corrosion of older specimens, 1 to 11 years. In these cases the fractures had healed successfully and sliding of the screw had presumably ceased.

89

We observed gross wear at the distal end of the lag screws and proximal end of the barrels of all devices examined. This suggested that wear of sliding counterparts had occurred early during fracture healing and before complete engagement of lag screws in the barrels. Our observations are consistent with the results of the laboratory study of Kyle et al. [19]. In our series the rough surfaces of sliding counterparts had contributed to wear. The presence of machining roughness on the inside walls of

Fig. 4. (a) Corrosion inside the barrel and on the surface of the lag screw. (b) Corrosion morphology and (c) EDXA spectrum of corrosion in boxed area in (c). (Electron micrpgraph, scale is marked by the 430 mm dotted line).

90

B.F. Shahgaldi, J. Compson / Injury, Int. J. Care Injured 31 (2000) 85±92

Fig. 4 (continued)

barrels and texturing on the surfaces of lag screws would have inevitably increased the frictional forces between lag screws and barrels causing wear. Although the devices we examined came from three di€erent manufacturers they were all of the same design and they were all made from contemporary stainless-steel alloys. Austenitic stainless steel,

Fig. 5. Corrosion particles in tissue. (Backscatter electron micrograph, scale is marked by the 43 mm dotted line).

because of its high strength and low cost, is the material of choice for manufacture of sliding hip screw-plates. However, stainless-steel devices are more likely to get jammed and their sliding surfaces damaged before complete engagement of the screw in the barrel than similar devices made from cobalt±chromium alloys [19,20]. In vitro studies have shown that fretting corrosion of stainless-steel fracture-®xation plates and screws starts within seconds of initiation of relative motion and the corrosion rate is three times greater than similar devices made from cobalt±chromium alloys [21±24]. The con®ned interface between the barrel and lag screw promotes crevice corrosion [25], although in our specimens crevice corrosion was secondary to fretting corrosion. Crevice corrosion may take as long as 6 months to occur in vivo due to the time required for an oxygen gradient to become established [11,24]. Wear induced corrosion was present in our specimens as early as 2 months after insertion of SHSP implants. Particles generated from corrosion have di€erent sizes, shapes and chemical composition compared with particles generated by mechanical wear between articulating surfaces of, for example, a hip joint. Corrosion particles are of submicron size within cells, but they may be as large as 50 mm while in the matrix of the tissue. Chemically, the proportions of elements making up the corrosion products are di€erent from those of the bulk alloy. Often they are present as an oxide of a single element or bound with elements such as phosphorous and sulphur that are normally present in the tissue [26]. The presence of a large amount of nickel in the fractured device and its absence in the intact

B.F. Shahgaldi, J. Compson / Injury, Int. J. Care Injured 31 (2000) 85±92

91

Fig. 6. (a) The surface of bone specimen in contact with metal showing areas of resorption pits. (b) A bone resorbing cell on the surface and (c) EDXA spectrum of tissue and bright particles in the boxed area. (Backscatter electron micrographs, scales are marked by dotted mm lines).

devices con®rms previous reports of nickel solubility and its systemic dissemination [27].

5. Conclusions Wear and corrosion of sliding counterparts of sliding hip screw-plates produce debris. Corrosion changes the chemical environment around the implant inducing

acidic pH [28]. Corrosion may, therefore, inhibit bone healing and enhance bone resorption. In a fully consolidated fracture the implant supports about 25% of the forces on the hip joint [29]. Corrosion damages the structural integrity of an implant reducing its mechanical strength [30]. Use of cobalt± chromium alloys would improve hardness and resistance to corrosion compared with stainless steel presently used which would be an advantage for sliding hip screw-plates in younger people. Better manufactur-

92

B.F. Shahgaldi, J. Compson / Injury, Int. J. Care Injured 31 (2000) 85±92

ing practice to improve surface smoothness of lag screws and inside walls of barrels is needed. Deliberate texturing of screws is counter productive and should be avoided.

[12] [13]

Acknowledgements

[14]

The authors are grateful to Professor FW Heatley FRCS, Director of Orthopaedic Academic Unit of Guy's, King's and St Thomas' (GKT) Medical School, King's College, London, for his support. This project was funded by the Orthopaedic Clinic Research Fund of St Thomas' Hospital.

[15]

References

[19]

[1] Pugh WL. Self-adjusting nail-plate for fractures about the hip joint. J Bone Jt Surg 1955;37A:1085±93. [2] Massie WK. Fracture of the hip: an instructional course lecture, the American Academy of Orthopaedic Surgery. J Bone Jt Surg 1964;46A:658±90. [3] Brown TIS, Court-Brown C. Failure of sliding nail-plate in subcapital fractures of the femoral neck. J Bone Jt Surg 1979;61B:342±6. [4] Hayse-Moore GH, MacEachern AG, Jamson Evans DC. Treatment of introchanteric fractures of the femur: a comparison of the Richard's screw-plate with the Jewett nail-plate. J Bone Jt Surg 1983;65B:262±7. [5] Leung KS, So WS, Shen WY, Hui PW. Gamma Nails and Dynamic Hip Screws for peritrochanteric fracture: a randomised prospective study in elderly patients. J Bone Jt Surg 1992;74B:345±51. [6] Mainds CC, Newman RJ. Implant failure in patient with proximal fractures of the femur treated with a sliding screw device. Injury 1989;20:98±100. [7] Ecker ML, Joyce JJ, Kohl EJ. The treatment of trochanteric hip fractures using a compression screw. J Bone Jt Surg 1975;57A:23±7. [8] Cater WH, Hicks JH. Corrosion in metal used for internal ®xation. Lancet 1956;Oct 27:871±3. [9] Scale JT, Winter GD. Corrosion of orthopaedic implants. J Bone Jt Surg 1959;41B:810±20. [10] Harding AF, Cook SD, Thomas KA, Collins CL, Haddad RJ, Milicic M. A clinical and metallurgical analysis of retrieved Jewett and Richards hip plate devices. Clin Orthop 1985;195:261±7. [11] Cook SD, Renz EA, Barrack RL, Thomas KA, Harding AF,

[16] [17] [18]

[20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30]

Haddad RJ, Milicic M. Clinical and metallurgical analysis of retrieved internal ®xation devices. Clin Orthop 1985;194:236±47. Cook SD, Thomas KA, Harding AF, Collins CL, Haddad RJ, Milicic M, Fischer WL. The in vivo performance of 250 internal ®xation devices: a follow-up-study. Biomaterials 1987;8:177±84. Black J. Does corrosion matter? J Bone Jt Surg 1988;70B:517± 20. Williams DF, Meachim G. A combined metallurgical and histological study of tissue prosthesis interaction in orthopaedic patients. J Biomed Materials Res 1974;8(3):1±9. French HG, Cook SD, Haddad RJ. Correlation of tissue reaction to corrosion in osteosynthetic devices. J Biomed Materials Res 1984;18:817±28. Delgado ER. Sarcoma following a surgically treated fractured tibia. Clin Orthop 1958;12:315±8. Dube VE, Fisher DE. Hemangiondothelioma of the leg following metallic ®xation of the tibia. Cancer 1972;30:1260±6. Shahgaldi BF, Heatley FW, Dewar A, Corrin B. In-vivo corrosion of cobalt-chromium and titanium wear particles. J Bone Jt Surg 1995;77B:962±9. Kyle RF, Wright TM, Burstein AH. Biomechanical analysis of the sliding characteristics of compression hip screw. J Bone Jt Surg 1980;62A:1308±14. Simpson AHRW, Varty K, Dodd CAF. Sliding hip screws: modes of failure. Injury 1989;20:227±31. Cohen J, Lindenbaum B. Fretting corrosion in orthopaedic implants. Clin Orthop 1968;61:167±75. Syrett BC, Wing SS. An electrochemical investigation of fretting corrosion of surgical materials. Corrosion 1978;34:138±42. Brown SA, Simpson J. Crevice and fretting corrosion of stainless steel plates and screws. J Biomed Materials Res 1981;15:867±78. Brown SA, Hughes PJ, Merrit K. In vitro studies of fretting corrosion of orthopaedic materials. J Orthopaedic Res 1988;6(4):572±9. Steinemann S. Corrosion of surgical implants Ð in vivo and in vitro tests. In: Winter G, Leray J, de Groot K, editors. Evaluation of Biomaterials. Chichester: Wiley, 1980. p. 1±34. Walczak J, Shahgaldi F, Heatley F. In vivo corrosion of 316L stainless steel hip implants: morphology and elemental compositions of corrosion products. Biomaterials 1998;19:229±37. Samitz MH, Katz SA. Nickel dermatitis hazards from prostheses: in vivo and in vitro solubilization studies. Br J Dermatol 1975;92:287±90. Willer HG, Broback LG, Buchhorn GH, Ing D. Crevice corrosion of cementer titanium hip implants. Clin Orthop 1996;333:51±7. Frankel VH, Burstein AH. Orthopaedic Biomechanics: the application of engineering to musculoskeletal system. Philadelphia: Lea and Febiger, 1970. Rose RM, Schiller AL, Radin EL. Corrosion-accelerated mechanical failure of vitallium nail-plate. J Bone Joint Surg 1972;54A:854±62.