- Email: [email protected]
In vivo surface analysis of titanium and stainless steel miniplates and screws
Int. J. Oral Maxillofac. Surg. 1996; 25:463-468 Printed in Denmark. All rights reserved
Copyright 9 Munksgaard 1996 InternationalJournal of
Oral&
Maxilloficid Surgery ISSN 0901-5027
In vivo surface analysis of titanium and stainless steel miniplates and screws
I. R. Matthew 1, J. W. Frame 1, R. M. Browne =, B. G. Millar 1 1Units of Oral Surgery and 2Oral Pathology, University of Birmingham,School of Dentistry, St Chad's Queensway,Birmingham, UK
I. R. Matthew, J. W. Frame, R. M. Browne, B. G. Millar: In vivo surface analysis of titanium and stainless steel miniplates and screws. Int. J. Oral Maxillofac. Surg. 1996; 25: 463-468. 9 Munksgaard, 1996 Abstract. This study was undertaken to characterize the surfaces of Champy| nium and stainless steel miniplates and screws that had been used to stabilize fractures of the mandible in an animal model. Miniplates and screws were retrieved at 4, 12, and 24 weeks after surgery. Low-vacuum scanning electron microscopy (SEM) of autoclaved unused (control) and test miniplates from the same production batches was undertaken. Energy-dispersive X-ray (EDX) analysis was used to identify compositional variations of the miniplate surface, and Vickers hardness testing was performed. At autopsy, clinical healing of all fractures was noted. SEM analysis indicated no perceptible difference in the surface characteristics of the miniplates at all time intervals. Aluminium and silicon deposits were identified by EDX analysis over the flat surfaces. There was extensive damage to some screw heads. It is concluded that there were no significant changes in the surface characteristics of miniplates retrieved up to 24 weeks after implantation in comparison with controls. Damage to the screws during insertion due to softness of the materials may render their removal difficult. There was no evidence to support the routine removal of titanium or stainless steel miniplates because of surface corrosion up to 6 months after implantation.
Accepted for publication 8 May 1996
Metal miniplates are used to immobilize fractures of the facial skeleton, but should nonfunctional miniplates and screws be removed routinely after bony healing? The decision to leave miniplates in situ may be influenced by many factors such as the known biocompatibility of the implant, poor access, and patient choice. The financial and resource implications of a second procedure to remove miniplates may also be a constraint. Screws may be damaged, and may fracture during insertion or removal, thus complicating miniplate retrieval. Some clinicians may thus feel it is justifiable to leave implants in situ if there have been no clinical symptoms from the fracture site during healing. However, an international
stainless steel and pure titanium cylindrical implants retrieved from 7 to 10.5 months after implantation into bone defects created in the tibia of rabbits. No evidence of corrosion was found. MOBERG et al. 8 examined miniplates made of cobalt-chromium alloy, nickelchromium alloy, and titanium, retrieved from the mandibles of monkeys up to 6 months after surgery. The miniplates had been inserted via an external skin incision, and were not used to stabilize an experimental fracture. Using light microscopy, the authors found no visible corrosion sites or corrosion products on the surface of the implants. Retrieval studies provide useful information about the surface characteristics of miniplates that have been removed
study group has recommended that all nonfunctional implants should be evaluated for removal (ToR~ERSEN & GJ~ROETn). This proposal was based upon a lack of consistent information about the long-term effects of metal implants on the human body. A major concern about the permanent retention of metal miniplates and screws after osteosynthesis is the effect on the implant surface of general corrosion, and fretting corrosion between miniplate and screws. Experimental studies on animals have shown that the surface of implant materials may retain their original surface characteristics after a period of implantation. LtNDER & LUNDSKOG6 used scanning electron microscopy (SEM) to study the surface of
Key words: jaw fractures; materials; corrosion.
biocompatible
464
M a t t h e w et al.
Fig. 1. Typical surface appearance of autoclaved stainless steel control miniplate. There are numerous surface irregularities in form of pits and craters with flat bases (low-vacuum SEM, •
from patients after osteosynthesis 2'3'11'12. However, for retrieval studies to be informative, the miniplate surfaces must b e e x a m i n e d before insertion. T h e present study undertook a controlled comparison in an animal model o f unused miniplates and miniplates f r o m the same production batch that h a d been retrieved after osteosynthesis in a clinically analogous situation. T h e objectives were to characterize the surface appearance of stainless steel and titan i u m C h a m p y miniplates after osteosynthesis of a fracture o f the mandible, and to compare those miniplates with other unused miniplates. A n additional objective was to c o m p a r e the surface composition of each flat surface of the miniplate by energy-dispersive analysis.
Material and methods The study was undertaken with the support of the Head of Biomedical Science and Ethics at the School of Medicine, University of Birmingham. Twelve mature female beagle dogs were housed under the same conditions for 6 weeks before the study commenced. For each dog, surgery was performed by the same surgeon and assistant. A veterinary surgeon undertook the anaesthetic care of each animal during surgery. For premedication, acepromazine maleate BP (ACE C-Vet Ltd, UK: 0.125-0.25 mg/kg) was administered subcutaneously 30 min before surgery. Intravenous pentobarbitone sodium (Sagatal,
RMB Animal Health Ltd, UK: 26.4mg/kg i.v.) was administered on induction, and an oroendotracheal tube was placed. Anaesthesia was maintained with halothane (0.5%) in oxygen (2.0 l/rain) and nitrous oxide (1.5 1/ min), and vital signs were monitored throughout the procedure. The surgical site was cleansed with aqueous chlorhexidine gluconate (0.05% w/v) and isolated with sterile drapes. A two-sided mucoperiosteal flap was raised to expose the buccal aspect of the second right mandibular premolar tooth and body of the mandible. Two autoclaved Champy miniplates (Martin Medizin-Technik, Germany) were positioned inferior to adjacent root apices. Miniplates were inserted by the surgical technique described by CHAMPVet al. 4. One short fourhole (code 25-250--04 for stainless steel, 253 5 0 ~ 4 for titanium) miniplate and one spaced four-hole (code 25-252-04 for stainless steel, 25-3524)4 for titanium) miniplate were used to stabilize each fracture. For each time interval, half the animals received titanium miniplates, and half received stainless steel. Holes were prepared for the autoclaved screws, and the miniplates were removed from the mouth. A fracture of the body of the mandible was created between the inner screw holes with a sterile, stainless steel, fiat, fissure bur and sterile saline coolant. A horizontal step at the midpoint of the fracture line ensured good alignment of the bone edges to avoid postoperative occlusal interferences. The fracture was reduced and the upper miniplate was secured with ll-mm-long screws (code 25090-11 for stainless steel, 25-092-11 for titanium) in the outer holes and 5-mm screws (code 25-090--07 for stainless steel, 25-092-
07 for titanium) in the inner holes. The lower miniplate was secured with 11-mm screws. Buprenorphine (Temgesic, Reckitt & Colman, UK: 0.01 mg/kg i.m.) was administered every 12 h for 36 h postoperatively to prevent any discomfort. A bolus dose of procaine penicillin (150 mg/ml) and benzathine penicillin (112.5 mg/ml) was administered immediately after surgery (Duplocillin LA, Mycofarm UK Ltd, UK: 1 ml/mg i.m.) for antibiotic prophylaxis. When the dogs had recovered from the procedure, they were returned to their original pens. Each animal was observed closely during the postoperative period, and none of them displayed signs of distress. Four animals were killed at 4, 12, and 24 weeks after surgery by rapid injection of pentobarbitone sodium (Euthatal, RMB Animal Health Ltd, UK: 143mg/kg i.v.). The jaws were removed and the soft tissues were reflected gently to expose the miniplates. The miniplates and screws were removed, and immersed in a mixture of 1% thymol in normal saline and stored at -20~ before examination. The miniplates and screws were examined with a JEOL 5300LV low-vacuum SEM (Jeol Technics Ltd, Japan). After a general scan of each miniplate at x75, 10 photographic images of each miniplate were taken at x750 and • 1500 magnifications. Higher magnifications (up to • were used to examine surface irregularities. Control (unused autoclaved) miniplates from the same production batches as the test samples were examined. Unused miniplates that were stored in the thymol/saline solution under the same conditions as the test samples were also analysed. Energy-dispersive X-ray (EDX) analysis was undertaken with a PGT system 4 plus apparatus (Princeton Gamma-Tech, Inc, USA), and a minimum of five randomly selected sample sites were analysed semiquantitatively from the surface of each miniplate. The miniplates were then sectioned perpendicular to the long axis with Champy miniplate cutters between the screw holes at one end of the miniplate. The small portion was embedded in acrylic resin (LR White hard grade, London Resin Company Ltd, UK). The cut surface of each miniplate was polished on a Struers rotary polishing apparatus (DAP-7, Struers, Denmark) using waterproof silicon carbide papers (grade 80 first, then 220, and finally 1200) and a soft lubricated polishing cloth. The polished surfaces were cleansed with 70% ethanol solution before EDX examination. Sampling of the polished surface was undertaken at four sites along the sectioned region adjacent to the medial and lateral surfaces. The remaining portion of each miniplate was then subjected to Vickers hardness testing with a Vickers M41 photoplan microscope (Vickers Ltd, UK). Ten readings were taken at random on both surfaces of the miniplates, a 200 g force load for 30 s being applied. Data were processed with the Minitab
Metal miniplates and screws
465
Topographic SEM analysis
Fig. 2. Typical surface appearance of autoclaved titanium control miniplate showing groups of randomly arranged polymorphous defects, and small number of surface cracks (low-vacuum SEM, x 1500).
Fig. 3. Metallic deposit present on surface of autoclaved control titanium miniplate. Deposit is darker than surrounding titanium, and was confirmed by EDX to be composed of aluminium (low-vacuum SEM, x3000). (Minitab, Inc, PA, USA) statistical package (Version 9) on an IBM-compatible PC.
Results Miniplate retrieval Clinical and radiographic evidence of satisfactory healing was found at all 12 fracture sites. All of the l l - m m - l o n g screws were secured firmly within bone.
In three dogs, both of the 5-mm titanium screws were loose, and a single 5nun titanium screw had become loose in one dog. Of the stainless steel components, a single 5-mm screw was loose in three dogs. After 24 weeks, bone had formed over the lateral surface of one stainless steel miniplate. A sharp chisel was used to remove the bone with minimal force before miniplate retrieval.
At x75 magnification, intrinsic surface irregularities and a small number of score marks were visible on the surfaces of autoclaved stainless steel and titanium control (unused) miniplates. At x750 magnification, surface irregularities in the form of small pits or craters with a flat base were found on the control stainless steel miniplates (Fig. 1). The diameter varied from -1 to 50 I.tm. Approximately 85% of irregularities sampled at random were 1-5 ~m wide, 10% were 5-10 Ixm wide, and 5% were greater than 10 ~tm wide, and of variable outline. The irregularities were dispersed randomly across each surface. Shallow scratch marks with a random orientation and distribution were also found on the flat surfaces. Groups of polymorphous defects were interspersed randomly over the surfaces of the control titanium miniplates (Fig. 2). The depressions were generally shallower but wider than the irregularities seen on the stainless steel miniplates. Some of the depressions had a crenated outline, and others had a jagged edge with steep vertical surfaces and a fiat base. Linear cracks were seen adjacent to the depressions on both surfaces of control titanium miniplates. The cracks varied from 1 to 30 ~tm long and 1-5 I.tm in width, and were distributed randomly but generally parallel to each other. Shallow scratch marks were also present on titanium control miniplates, and these were orientated randomly. At • 1500 magnification, dark irregularities with a smooth surface were found on control stainless steel and titanium miniplates (Fig. 3). The irregularities were subjected to EDX examination, and proved to be composed of either aluminium or silicon. The linear cracks found on the control titanium miniplates were also analysed by EDX, and these were of similar composition to the adjacent surface. The storage of autoclaved control miniplates in 1% thymol and normal saline solution under similar conditions to the test plates did not influence the surface appearance of either the titanium or stainless steel miniplates. Soft-tissue accretions were identified on areas of the implanted stainless steel miniplates retrieved 4 weeks after surgery. At other sites where the metal sur-
466
Matthew et al. was similar to that of the controls. The same was also true of the miniplates examined 24 weeks after surgery (Figs. 4 and 5). Deposits of aluminium and silicon on the miniplates at 24 weeks were again identified. Most of the control screw heads examined (at x75 magnification) had a well-defined cruciform outline, but on some the vertical walls were flared rather than parallel. Some of the retrieved screws had been damaged during manipulation, and loose metal fragments were seen on the screw heads (Fig. 6). Although surface damage was seen to be due to screw manipulation, there was no evidence of surface corrosion defects on any. of the implanted screws.
Fig. 4. Surface appearance of titanium miniplate retrieved 24 weeks after surgery. Score mark extends across lower left comer, caused during manipulation of miniplate (low-vacuum SEM, x 1500).
The principal elements identified on EDX spectra for stainless steel miniplates were iron, chromium, molybdenum, and nickel. Other elements identified included carbon, nitrogen, and oxygen. Titanium was the predominant element on spectra for titanium miniplates. Minor variations in element ratios were noted between spectra from the surface and from the deeper layer, but the differences were not great enough to justify further analysis. Vickers hardness testing
Fig. 5. Surface appearance of stainless steel miniplate retrieved 24 weeks after surgery. There is wide pit in centre of image (low-vacuum SEM, x2000).
face was exposed, the surface appearance of the test miniplates at x750 magnification was similar to that of the controis. EDX examination of titanium miniplates confirmed the presence of soft-tissue deposits at the base of the irregularities. Aluminium and silicon deposits were also identified on both surfaces of the titanium miniplates, in a random distribution.
Twelve weeks after surgery, there was greater coverage of the miniplate surfaces by soft tissue, but characterization of exposed areas of miniplate surface was still possible. EDX examination revealed surface defects containing organic material, and aluminium or silicon deposits similar to those found on the controls. However, the overall surface appearance of the test miniplates
Analysis of variance was undertaken on the 10 Vickers hardness readings for each miniplate. There was no significant difference in Vickers hardness for autoclaved control miniplates before and after immersion in 1% thymol/normal saline solution for 6 months (ANOVA, F=0.30: d f = l : P>0.05). There was an overall significant difference between Vickers hardness readings for titanium and stainless steel miniplates (ANOVA, F=50.58: d f = l : P<0.001), but no significant difference between medial and lateral surfaces of the miniplates for each material (ANOVA, F=2.16: df= 1: P>0.05). The overall mean value for titanium miniplates was 221.23 (SD=23.49), and 238.40 (SD=24.08) for stainless steel miniplates. Discussion In the present study, there was little evidence of changes in the surface appearance of miniplates retrieved up to 24 weeks after implantation, other than the adherence of soft tissue on the miniplate
Metal miniplates and screws
Fig. 6. Cruciform screw head of titanium screw at retrieval 4 weeks after surgery. Note loose metal particles on surface of screw head (low-vacuum SEM, x75).
surfaces and handling defects. The surface irregularities of the control stainless steel and titanium miniplates appear to have occurred during implant manufacture. The craters, pits, surface cracks, and depressions probably arose during production of the sheets from which the miniplates are cut. The countersinks for the screw holes had a smooth finish and a consistent bevel angle, indicating that a precision milling apparatus was used to machine the screw holes. The fine scratches on the surfaces of all miniplates probably occurred during the final polish. The aluminium and silicon deposits appear to become embedded within pre-existing surface irregularities. These surface deposits may have been caused by aluminium oxide or silicon carbide abrasives used to polish the miniplates during manufacture. Metal release from maxillofacial plating systems has been reported. ROSENBERGet al.10 identified metallic deposits by light microscopy in the soft tissues adjacent to Champy miniplates that had been retrieved after osteosynthesis. These deposits may have been displaced into the tissues during miniplate fixation. MOBERGet al. 8 examined metal release from miniplates in vivo by fixing miniplates on to the lateral surface of the mandible via an extraoral incision in monkeys. After 3 and 6 months, the titanium concentration within soft tissues
adjacent to titanium miniplates and screws was not significantly raised. Tissue samples adjacent to Champy stainless steel miniplates and screws yielded significantly higher concentrations of nickel than the controls. Titanium is resistant to the general corrosion, pitting attack, and crevice corrosion that may occur in other metal implants as a result of attack from aggressive organic fluids I. Bessho & ILZUKA2 examined 113 Champy titanium miniplates that had been retrieved after miniplate fixation of mandibular fractures, and identified surface depressions that were considered to be caused by pitting corrosion. Stress corrosion cracking was identified in areas that were likely to receive tensile stress. However, the authors did not describe the appearance of the miniplates before insertion. The same paper reported the results of a study of Champy miniplate osteosynthesis in Japanese white rabbits. Miniplates were retrieved 6 months after implantation, and the authors identified stress corrosion cracking more frequently between the central screw holes on the under surface of the miniplate. Stress cracking of commercially pure titanium is virtually unknown 13. In a follow-up study with the same experimental method, BESSHO et al.3 identified stress cracking up to 24 months after implantation of the mini-
467
plates. No Variation in the appearance of the pitting corrosion was found in either group over the 2-year study period. The authors may have described the appearance of manufacturing defects rather than pitting or stress corrosion on the miniplate surface. Cracking was observed only in the fracture group in areas where the miniplate surface was subjected to stress, and there was no difference in the appearance of these socalled stress cracks over the 2-year period. In the present study, cracks were found on lateral and medial surfaces of the titanium miniplates, and the distribution of surface cracks was random. LOUKOTA & SrmLTON7 analysed the effects of compressive and tensile forces on different types of miniplates and found that all miniplates (including Champy) performed beyond the requirements of the clinical situation. BESSHOet al. 3 identified corrosion pits on the surface of unused miniplates before insertion, but these miniplates had been subjected to sand blasting, stone polishing, and natural passivation before examination. This form of pre-treatment of miniplates before insertion is not a standard clinical practice. In the present study, the consistent identification by EDX analysis of aluminium and silicon deposits on the surface of miniplates was unexpected. However, MOBERG et al. 8 reported that the titanium miniplates (Nobelpharma, Sweden) used in their experimental study released significantly large amounts of aluminium into the adjacent tissues (mean 18 ~tg/g wet weight, SD 16 I.tg/g). The concentration of aluminium within muscle tissue has been reported to be 0.15-6.0 I.tg/g9. Aluminium has been linked to several disorders in man including Alzheimer's disease, Parkinson's disease, and osteomalacia 5. It seems advantageous that metal plates should be free of any surface contaminants such as aluminium or silicon. No aluminium or silicon deposits were identified on the chamfered edges of the screw hole, but polishing of this machined surface would in any case be unnecessary. TORGERSEN t~ GJERDET11 did not report the presence of alumini u m or silicon deposits on Champy miniplates retrieved from patients after osteosynthesis, but their SEM surface studies were conducted at low magnifications, and any deposits were therefore unlikely to have been identified. TORG-
468
Matthew et al.
ERSEN et al. 12 subsequently identified by EDX analysis aluminium and titanium within the soft tissues adjacent to Champy titanium miniplates that had been retrieved from patients after osteosynthesis.9 In conclusion, there was little evidence in this study of changes in the surface characteristics of the miniplates retrieved up to 24 weeks after implantation when compared with the controls. It is essential in retrieval studies to compare the findings with unused miniplates from the same batch. The surface characteristics of implanted specimens may be the result of the manufacturing process. Damage to the cruciform screw heads during insertion and retrieval may render screw removal difficult. There is no evidence from this study to support the view that Champy miniplates should be removed routinely up to 24 weeks after insertion because of the risk of surface corrosion.
Acknowledgments. We wish to acknowledge the financial support of West Midlands Health Authority, the generous donation of Champy components from the UK importers Albert Waeschle, technical assistance from Miss Sangita Bhatt, and the veterinary advice and assistance of Professor Morton and
Paul Townsend, Unit of Biomedical Science and Ethics at the University of Birmingham. 9.
References 1. BANNONBP, MILD EE. Titanium alloys for biomaterial application: an overview. In: LUCKEYHA, KUBLIF, eds.: Titanium alloys in surgical implants ASTM STP. Baltimore, MD: American Society for Testing and Materials, 1983: 7-15. 2. BESSHOK, IIZUKAT. Clinical and animal experiments on stress corrosion of titanium miniplates. Clin Mater 1993: 14: 223-7. 3. BEssrto K, FUnMURAK, hZUKAT. Experimental long-term study of titanium ions eluted from pure titanium miniplates. J Biomed Mater Res 1995: 29: 901-4. 4. CHAMPYM, LODDKIP,SCHMITrR, JAEGER 9 JH, MUSTERD. Mandibular osteosynthesis by miniature screwed miniplates via a huccal approach. J Maxillofac Surg 1978: 6: 14-21. 5. EXLEYC, BIRCHALLJD. The cellular toxicity of aluminium. J Theor Biol 1992: 159: 83-98. 6. LL~DERL, LUNDSKOGJ. Incorporation of stainless steel, titanium and Vitatlium in bone. Injury 1975: 6: 277-85. 7. LOOKOrARA, SnELTONJC. Mechanical analysis of maxillofacial miniplates. Br J Oral Maxillofac Surg 1995: 33: 174-9. 8. MOBERGLE, NORDENRAMA, KJELLMAN O. Metal release from miniplates used in jaw fracture treatment. A pilot study. Int
10.
11.
12.
13.
J Oral Maxillofac Surg 1989: 18: 31114. MAEUSU P-A, BLOCH PR, GERET V, STEmEMANN SG. Surface characterisation of titanium and Ti-alloys. In: CmusTEL P~ MEUNER A, LEE AJC, eds.: Biological and biomechanical performance of biomaterials. Amsterdam: Elsevier, 1986: 57-62. ROSENBERGA, GRATZ KW, SAILER HF. Should titanium miniplates be removed after bone healing is complete? Int J Oral Maxillofac Surg 1993: 22: 185-8. TORGERSEN S, GJERDET NR. Retrieval study of stainless steel and titanium miniplates and screws used in maxillofacial surgery. J Mater Sci: Mater Med 1994: 5: 256-62. TORGERSENS, GJERDETNR, ERaCHSENES, BANG G. Metal particles and tissue changes adjacent to miniplates. A retrieval study. Acta Odont Scand 1995: 53: 65-71. W1LLIAMSDE Titanium and titanium alloys. In: WILL]AMSDF, ed.: Biocompatibility of clinical implant materials. Vol. 1. Boca Raton, FL: CRC Press, 1981: 944.
Address: L R. Matthew University'of Birmingham School of Dentistry St Chad's Queensway Birmingham 84 6NN UK
Copyright 9 Munksgaard 1996 InternationalJournal of
Oral&
Maxilloficid Surgery ISSN 0901-5027
In vivo surface analysis of titanium and stainless steel miniplates and screws
I. R. Matthew 1, J. W. Frame 1, R. M. Browne =, B. G. Millar 1 1Units of Oral Surgery and 2Oral Pathology, University of Birmingham,School of Dentistry, St Chad's Queensway,Birmingham, UK
I. R. Matthew, J. W. Frame, R. M. Browne, B. G. Millar: In vivo surface analysis of titanium and stainless steel miniplates and screws. Int. J. Oral Maxillofac. Surg. 1996; 25: 463-468. 9 Munksgaard, 1996 Abstract. This study was undertaken to characterize the surfaces of Champy| nium and stainless steel miniplates and screws that had been used to stabilize fractures of the mandible in an animal model. Miniplates and screws were retrieved at 4, 12, and 24 weeks after surgery. Low-vacuum scanning electron microscopy (SEM) of autoclaved unused (control) and test miniplates from the same production batches was undertaken. Energy-dispersive X-ray (EDX) analysis was used to identify compositional variations of the miniplate surface, and Vickers hardness testing was performed. At autopsy, clinical healing of all fractures was noted. SEM analysis indicated no perceptible difference in the surface characteristics of the miniplates at all time intervals. Aluminium and silicon deposits were identified by EDX analysis over the flat surfaces. There was extensive damage to some screw heads. It is concluded that there were no significant changes in the surface characteristics of miniplates retrieved up to 24 weeks after implantation in comparison with controls. Damage to the screws during insertion due to softness of the materials may render their removal difficult. There was no evidence to support the routine removal of titanium or stainless steel miniplates because of surface corrosion up to 6 months after implantation.
Accepted for publication 8 May 1996
Metal miniplates are used to immobilize fractures of the facial skeleton, but should nonfunctional miniplates and screws be removed routinely after bony healing? The decision to leave miniplates in situ may be influenced by many factors such as the known biocompatibility of the implant, poor access, and patient choice. The financial and resource implications of a second procedure to remove miniplates may also be a constraint. Screws may be damaged, and may fracture during insertion or removal, thus complicating miniplate retrieval. Some clinicians may thus feel it is justifiable to leave implants in situ if there have been no clinical symptoms from the fracture site during healing. However, an international
stainless steel and pure titanium cylindrical implants retrieved from 7 to 10.5 months after implantation into bone defects created in the tibia of rabbits. No evidence of corrosion was found. MOBERG et al. 8 examined miniplates made of cobalt-chromium alloy, nickelchromium alloy, and titanium, retrieved from the mandibles of monkeys up to 6 months after surgery. The miniplates had been inserted via an external skin incision, and were not used to stabilize an experimental fracture. Using light microscopy, the authors found no visible corrosion sites or corrosion products on the surface of the implants. Retrieval studies provide useful information about the surface characteristics of miniplates that have been removed
study group has recommended that all nonfunctional implants should be evaluated for removal (ToR~ERSEN & GJ~ROETn). This proposal was based upon a lack of consistent information about the long-term effects of metal implants on the human body. A major concern about the permanent retention of metal miniplates and screws after osteosynthesis is the effect on the implant surface of general corrosion, and fretting corrosion between miniplate and screws. Experimental studies on animals have shown that the surface of implant materials may retain their original surface characteristics after a period of implantation. LtNDER & LUNDSKOG6 used scanning electron microscopy (SEM) to study the surface of
Key words: jaw fractures; materials; corrosion.
biocompatible
464
M a t t h e w et al.
Fig. 1. Typical surface appearance of autoclaved stainless steel control miniplate. There are numerous surface irregularities in form of pits and craters with flat bases (low-vacuum SEM, •
from patients after osteosynthesis 2'3'11'12. However, for retrieval studies to be informative, the miniplate surfaces must b e e x a m i n e d before insertion. T h e present study undertook a controlled comparison in an animal model o f unused miniplates and miniplates f r o m the same production batch that h a d been retrieved after osteosynthesis in a clinically analogous situation. T h e objectives were to characterize the surface appearance of stainless steel and titan i u m C h a m p y miniplates after osteosynthesis of a fracture o f the mandible, and to compare those miniplates with other unused miniplates. A n additional objective was to c o m p a r e the surface composition of each flat surface of the miniplate by energy-dispersive analysis.
Material and methods The study was undertaken with the support of the Head of Biomedical Science and Ethics at the School of Medicine, University of Birmingham. Twelve mature female beagle dogs were housed under the same conditions for 6 weeks before the study commenced. For each dog, surgery was performed by the same surgeon and assistant. A veterinary surgeon undertook the anaesthetic care of each animal during surgery. For premedication, acepromazine maleate BP (ACE C-Vet Ltd, UK: 0.125-0.25 mg/kg) was administered subcutaneously 30 min before surgery. Intravenous pentobarbitone sodium (Sagatal,
RMB Animal Health Ltd, UK: 26.4mg/kg i.v.) was administered on induction, and an oroendotracheal tube was placed. Anaesthesia was maintained with halothane (0.5%) in oxygen (2.0 l/rain) and nitrous oxide (1.5 1/ min), and vital signs were monitored throughout the procedure. The surgical site was cleansed with aqueous chlorhexidine gluconate (0.05% w/v) and isolated with sterile drapes. A two-sided mucoperiosteal flap was raised to expose the buccal aspect of the second right mandibular premolar tooth and body of the mandible. Two autoclaved Champy miniplates (Martin Medizin-Technik, Germany) were positioned inferior to adjacent root apices. Miniplates were inserted by the surgical technique described by CHAMPVet al. 4. One short fourhole (code 25-250--04 for stainless steel, 253 5 0 ~ 4 for titanium) miniplate and one spaced four-hole (code 25-252-04 for stainless steel, 25-3524)4 for titanium) miniplate were used to stabilize each fracture. For each time interval, half the animals received titanium miniplates, and half received stainless steel. Holes were prepared for the autoclaved screws, and the miniplates were removed from the mouth. A fracture of the body of the mandible was created between the inner screw holes with a sterile, stainless steel, fiat, fissure bur and sterile saline coolant. A horizontal step at the midpoint of the fracture line ensured good alignment of the bone edges to avoid postoperative occlusal interferences. The fracture was reduced and the upper miniplate was secured with ll-mm-long screws (code 25090-11 for stainless steel, 25-092-11 for titanium) in the outer holes and 5-mm screws (code 25-090--07 for stainless steel, 25-092-
07 for titanium) in the inner holes. The lower miniplate was secured with 11-mm screws. Buprenorphine (Temgesic, Reckitt & Colman, UK: 0.01 mg/kg i.m.) was administered every 12 h for 36 h postoperatively to prevent any discomfort. A bolus dose of procaine penicillin (150 mg/ml) and benzathine penicillin (112.5 mg/ml) was administered immediately after surgery (Duplocillin LA, Mycofarm UK Ltd, UK: 1 ml/mg i.m.) for antibiotic prophylaxis. When the dogs had recovered from the procedure, they were returned to their original pens. Each animal was observed closely during the postoperative period, and none of them displayed signs of distress. Four animals were killed at 4, 12, and 24 weeks after surgery by rapid injection of pentobarbitone sodium (Euthatal, RMB Animal Health Ltd, UK: 143mg/kg i.v.). The jaws were removed and the soft tissues were reflected gently to expose the miniplates. The miniplates and screws were removed, and immersed in a mixture of 1% thymol in normal saline and stored at -20~ before examination. The miniplates and screws were examined with a JEOL 5300LV low-vacuum SEM (Jeol Technics Ltd, Japan). After a general scan of each miniplate at x75, 10 photographic images of each miniplate were taken at x750 and • 1500 magnifications. Higher magnifications (up to • were used to examine surface irregularities. Control (unused autoclaved) miniplates from the same production batches as the test samples were examined. Unused miniplates that were stored in the thymol/saline solution under the same conditions as the test samples were also analysed. Energy-dispersive X-ray (EDX) analysis was undertaken with a PGT system 4 plus apparatus (Princeton Gamma-Tech, Inc, USA), and a minimum of five randomly selected sample sites were analysed semiquantitatively from the surface of each miniplate. The miniplates were then sectioned perpendicular to the long axis with Champy miniplate cutters between the screw holes at one end of the miniplate. The small portion was embedded in acrylic resin (LR White hard grade, London Resin Company Ltd, UK). The cut surface of each miniplate was polished on a Struers rotary polishing apparatus (DAP-7, Struers, Denmark) using waterproof silicon carbide papers (grade 80 first, then 220, and finally 1200) and a soft lubricated polishing cloth. The polished surfaces were cleansed with 70% ethanol solution before EDX examination. Sampling of the polished surface was undertaken at four sites along the sectioned region adjacent to the medial and lateral surfaces. The remaining portion of each miniplate was then subjected to Vickers hardness testing with a Vickers M41 photoplan microscope (Vickers Ltd, UK). Ten readings were taken at random on both surfaces of the miniplates, a 200 g force load for 30 s being applied. Data were processed with the Minitab
Metal miniplates and screws
465
Topographic SEM analysis
Fig. 2. Typical surface appearance of autoclaved titanium control miniplate showing groups of randomly arranged polymorphous defects, and small number of surface cracks (low-vacuum SEM, x 1500).
Fig. 3. Metallic deposit present on surface of autoclaved control titanium miniplate. Deposit is darker than surrounding titanium, and was confirmed by EDX to be composed of aluminium (low-vacuum SEM, x3000). (Minitab, Inc, PA, USA) statistical package (Version 9) on an IBM-compatible PC.
Results Miniplate retrieval Clinical and radiographic evidence of satisfactory healing was found at all 12 fracture sites. All of the l l - m m - l o n g screws were secured firmly within bone.
In three dogs, both of the 5-mm titanium screws were loose, and a single 5nun titanium screw had become loose in one dog. Of the stainless steel components, a single 5-mm screw was loose in three dogs. After 24 weeks, bone had formed over the lateral surface of one stainless steel miniplate. A sharp chisel was used to remove the bone with minimal force before miniplate retrieval.
At x75 magnification, intrinsic surface irregularities and a small number of score marks were visible on the surfaces of autoclaved stainless steel and titanium control (unused) miniplates. At x750 magnification, surface irregularities in the form of small pits or craters with a flat base were found on the control stainless steel miniplates (Fig. 1). The diameter varied from -1 to 50 I.tm. Approximately 85% of irregularities sampled at random were 1-5 ~m wide, 10% were 5-10 Ixm wide, and 5% were greater than 10 ~tm wide, and of variable outline. The irregularities were dispersed randomly across each surface. Shallow scratch marks with a random orientation and distribution were also found on the flat surfaces. Groups of polymorphous defects were interspersed randomly over the surfaces of the control titanium miniplates (Fig. 2). The depressions were generally shallower but wider than the irregularities seen on the stainless steel miniplates. Some of the depressions had a crenated outline, and others had a jagged edge with steep vertical surfaces and a fiat base. Linear cracks were seen adjacent to the depressions on both surfaces of control titanium miniplates. The cracks varied from 1 to 30 ~tm long and 1-5 I.tm in width, and were distributed randomly but generally parallel to each other. Shallow scratch marks were also present on titanium control miniplates, and these were orientated randomly. At • 1500 magnification, dark irregularities with a smooth surface were found on control stainless steel and titanium miniplates (Fig. 3). The irregularities were subjected to EDX examination, and proved to be composed of either aluminium or silicon. The linear cracks found on the control titanium miniplates were also analysed by EDX, and these were of similar composition to the adjacent surface. The storage of autoclaved control miniplates in 1% thymol and normal saline solution under similar conditions to the test plates did not influence the surface appearance of either the titanium or stainless steel miniplates. Soft-tissue accretions were identified on areas of the implanted stainless steel miniplates retrieved 4 weeks after surgery. At other sites where the metal sur-
466
Matthew et al. was similar to that of the controls. The same was also true of the miniplates examined 24 weeks after surgery (Figs. 4 and 5). Deposits of aluminium and silicon on the miniplates at 24 weeks were again identified. Most of the control screw heads examined (at x75 magnification) had a well-defined cruciform outline, but on some the vertical walls were flared rather than parallel. Some of the retrieved screws had been damaged during manipulation, and loose metal fragments were seen on the screw heads (Fig. 6). Although surface damage was seen to be due to screw manipulation, there was no evidence of surface corrosion defects on any. of the implanted screws.
Fig. 4. Surface appearance of titanium miniplate retrieved 24 weeks after surgery. Score mark extends across lower left comer, caused during manipulation of miniplate (low-vacuum SEM, x 1500).
The principal elements identified on EDX spectra for stainless steel miniplates were iron, chromium, molybdenum, and nickel. Other elements identified included carbon, nitrogen, and oxygen. Titanium was the predominant element on spectra for titanium miniplates. Minor variations in element ratios were noted between spectra from the surface and from the deeper layer, but the differences were not great enough to justify further analysis. Vickers hardness testing
Fig. 5. Surface appearance of stainless steel miniplate retrieved 24 weeks after surgery. There is wide pit in centre of image (low-vacuum SEM, x2000).
face was exposed, the surface appearance of the test miniplates at x750 magnification was similar to that of the controis. EDX examination of titanium miniplates confirmed the presence of soft-tissue deposits at the base of the irregularities. Aluminium and silicon deposits were also identified on both surfaces of the titanium miniplates, in a random distribution.
Twelve weeks after surgery, there was greater coverage of the miniplate surfaces by soft tissue, but characterization of exposed areas of miniplate surface was still possible. EDX examination revealed surface defects containing organic material, and aluminium or silicon deposits similar to those found on the controls. However, the overall surface appearance of the test miniplates
Analysis of variance was undertaken on the 10 Vickers hardness readings for each miniplate. There was no significant difference in Vickers hardness for autoclaved control miniplates before and after immersion in 1% thymol/normal saline solution for 6 months (ANOVA, F=0.30: d f = l : P>0.05). There was an overall significant difference between Vickers hardness readings for titanium and stainless steel miniplates (ANOVA, F=50.58: d f = l : P<0.001), but no significant difference between medial and lateral surfaces of the miniplates for each material (ANOVA, F=2.16: df= 1: P>0.05). The overall mean value for titanium miniplates was 221.23 (SD=23.49), and 238.40 (SD=24.08) for stainless steel miniplates. Discussion In the present study, there was little evidence of changes in the surface appearance of miniplates retrieved up to 24 weeks after implantation, other than the adherence of soft tissue on the miniplate
Metal miniplates and screws
Fig. 6. Cruciform screw head of titanium screw at retrieval 4 weeks after surgery. Note loose metal particles on surface of screw head (low-vacuum SEM, x75).
surfaces and handling defects. The surface irregularities of the control stainless steel and titanium miniplates appear to have occurred during implant manufacture. The craters, pits, surface cracks, and depressions probably arose during production of the sheets from which the miniplates are cut. The countersinks for the screw holes had a smooth finish and a consistent bevel angle, indicating that a precision milling apparatus was used to machine the screw holes. The fine scratches on the surfaces of all miniplates probably occurred during the final polish. The aluminium and silicon deposits appear to become embedded within pre-existing surface irregularities. These surface deposits may have been caused by aluminium oxide or silicon carbide abrasives used to polish the miniplates during manufacture. Metal release from maxillofacial plating systems has been reported. ROSENBERGet al.10 identified metallic deposits by light microscopy in the soft tissues adjacent to Champy miniplates that had been retrieved after osteosynthesis. These deposits may have been displaced into the tissues during miniplate fixation. MOBERGet al. 8 examined metal release from miniplates in vivo by fixing miniplates on to the lateral surface of the mandible via an extraoral incision in monkeys. After 3 and 6 months, the titanium concentration within soft tissues
adjacent to titanium miniplates and screws was not significantly raised. Tissue samples adjacent to Champy stainless steel miniplates and screws yielded significantly higher concentrations of nickel than the controls. Titanium is resistant to the general corrosion, pitting attack, and crevice corrosion that may occur in other metal implants as a result of attack from aggressive organic fluids I. Bessho & ILZUKA2 examined 113 Champy titanium miniplates that had been retrieved after miniplate fixation of mandibular fractures, and identified surface depressions that were considered to be caused by pitting corrosion. Stress corrosion cracking was identified in areas that were likely to receive tensile stress. However, the authors did not describe the appearance of the miniplates before insertion. The same paper reported the results of a study of Champy miniplate osteosynthesis in Japanese white rabbits. Miniplates were retrieved 6 months after implantation, and the authors identified stress corrosion cracking more frequently between the central screw holes on the under surface of the miniplate. Stress cracking of commercially pure titanium is virtually unknown 13. In a follow-up study with the same experimental method, BESSHO et al.3 identified stress cracking up to 24 months after implantation of the mini-
467
plates. No Variation in the appearance of the pitting corrosion was found in either group over the 2-year study period. The authors may have described the appearance of manufacturing defects rather than pitting or stress corrosion on the miniplate surface. Cracking was observed only in the fracture group in areas where the miniplate surface was subjected to stress, and there was no difference in the appearance of these socalled stress cracks over the 2-year period. In the present study, cracks were found on lateral and medial surfaces of the titanium miniplates, and the distribution of surface cracks was random. LOUKOTA & SrmLTON7 analysed the effects of compressive and tensile forces on different types of miniplates and found that all miniplates (including Champy) performed beyond the requirements of the clinical situation. BESSHOet al. 3 identified corrosion pits on the surface of unused miniplates before insertion, but these miniplates had been subjected to sand blasting, stone polishing, and natural passivation before examination. This form of pre-treatment of miniplates before insertion is not a standard clinical practice. In the present study, the consistent identification by EDX analysis of aluminium and silicon deposits on the surface of miniplates was unexpected. However, MOBERG et al. 8 reported that the titanium miniplates (Nobelpharma, Sweden) used in their experimental study released significantly large amounts of aluminium into the adjacent tissues (mean 18 ~tg/g wet weight, SD 16 I.tg/g). The concentration of aluminium within muscle tissue has been reported to be 0.15-6.0 I.tg/g9. Aluminium has been linked to several disorders in man including Alzheimer's disease, Parkinson's disease, and osteomalacia 5. It seems advantageous that metal plates should be free of any surface contaminants such as aluminium or silicon. No aluminium or silicon deposits were identified on the chamfered edges of the screw hole, but polishing of this machined surface would in any case be unnecessary. TORGERSEN t~ GJERDET11 did not report the presence of alumini u m or silicon deposits on Champy miniplates retrieved from patients after osteosynthesis, but their SEM surface studies were conducted at low magnifications, and any deposits were therefore unlikely to have been identified. TORG-
468
Matthew et al.
ERSEN et al. 12 subsequently identified by EDX analysis aluminium and titanium within the soft tissues adjacent to Champy titanium miniplates that had been retrieved from patients after osteosynthesis.9 In conclusion, there was little evidence in this study of changes in the surface characteristics of the miniplates retrieved up to 24 weeks after implantation when compared with the controls. It is essential in retrieval studies to compare the findings with unused miniplates from the same batch. The surface characteristics of implanted specimens may be the result of the manufacturing process. Damage to the cruciform screw heads during insertion and retrieval may render screw removal difficult. There is no evidence from this study to support the view that Champy miniplates should be removed routinely up to 24 weeks after insertion because of the risk of surface corrosion.
Acknowledgments. We wish to acknowledge the financial support of West Midlands Health Authority, the generous donation of Champy components from the UK importers Albert Waeschle, technical assistance from Miss Sangita Bhatt, and the veterinary advice and assistance of Professor Morton and
Paul Townsend, Unit of Biomedical Science and Ethics at the University of Birmingham. 9.
References 1. BANNONBP, MILD EE. Titanium alloys for biomaterial application: an overview. In: LUCKEYHA, KUBLIF, eds.: Titanium alloys in surgical implants ASTM STP. Baltimore, MD: American Society for Testing and Materials, 1983: 7-15. 2. BESSHOK, IIZUKAT. Clinical and animal experiments on stress corrosion of titanium miniplates. Clin Mater 1993: 14: 223-7. 3. BEssrto K, FUnMURAK, hZUKAT. Experimental long-term study of titanium ions eluted from pure titanium miniplates. J Biomed Mater Res 1995: 29: 901-4. 4. CHAMPYM, LODDKIP,SCHMITrR, JAEGER 9 JH, MUSTERD. Mandibular osteosynthesis by miniature screwed miniplates via a huccal approach. J Maxillofac Surg 1978: 6: 14-21. 5. EXLEYC, BIRCHALLJD. The cellular toxicity of aluminium. J Theor Biol 1992: 159: 83-98. 6. LL~DERL, LUNDSKOGJ. Incorporation of stainless steel, titanium and Vitatlium in bone. Injury 1975: 6: 277-85. 7. LOOKOrARA, SnELTONJC. Mechanical analysis of maxillofacial miniplates. Br J Oral Maxillofac Surg 1995: 33: 174-9. 8. MOBERGLE, NORDENRAMA, KJELLMAN O. Metal release from miniplates used in jaw fracture treatment. A pilot study. Int
10.
11.
12.
13.
J Oral Maxillofac Surg 1989: 18: 31114. MAEUSU P-A, BLOCH PR, GERET V, STEmEMANN SG. Surface characterisation of titanium and Ti-alloys. In: CmusTEL P~ MEUNER A, LEE AJC, eds.: Biological and biomechanical performance of biomaterials. Amsterdam: Elsevier, 1986: 57-62. ROSENBERGA, GRATZ KW, SAILER HF. Should titanium miniplates be removed after bone healing is complete? Int J Oral Maxillofac Surg 1993: 22: 185-8. TORGERSEN S, GJERDET NR. Retrieval study of stainless steel and titanium miniplates and screws used in maxillofacial surgery. J Mater Sci: Mater Med 1994: 5: 256-62. TORGERSENS, GJERDETNR, ERaCHSENES, BANG G. Metal particles and tissue changes adjacent to miniplates. A retrieval study. Acta Odont Scand 1995: 53: 65-71. W1LLIAMSDE Titanium and titanium alloys. In: WILL]AMSDF, ed.: Biocompatibility of clinical implant materials. Vol. 1. Boca Raton, FL: CRC Press, 1981: 944.
Address: L R. Matthew University'of Birmingham School of Dentistry St Chad's Queensway Birmingham 84 6NN UK