Implant Materials
One hundred and eight stainless-steel (ASTM F138) 5/6-millimeter-diameter tapered pins (Orthofix, Bussolengo, Italy) were divided into three groups of thirty-six each. The pins in Group A remained uncoated, the pins in Group B were plasma-sprayed with hydroxyapatite (Biocoatings, Fornovo Taro, Parma, Italy), and the pins in Group C were plasma-sprayed with commercially pure titanium (ASTM F67-89) (Biocoatings).
The crystallinity of hydroxyapatite, calculated with use of x-ray spectroscopy, was greater than 70 per cent, and the purity, determined with mass spectroscopy, was greater than 97 per cent. The calcium-to-phosphate ratio was 1.67 ± 0.01. The hydroxyapatite powder contained small traces of heavy elements in amounts that were below the limits set by the American Society for Testing and Materials F1185-88 standard test.
The thickness of the coatings in Groups B and C was measured metallographically at ten different sites on six pins from each group with use of a Versamet II microscope (Union Optical, Tokyo, Japan) at 200 times magnification. The surface roughness of the pins was measured at three different sites on six pins from each group with use of a roughness meter (Perthometer S5P; Feinpruf Perthen, Göttingen, Germany) and a five-micrometer-diameter probe (Perthen RHT6-50; Feinpruf Perthen).
Eighteen adult female sheep of similar size and body mass (mean, 45 ± 5 kilograms) were included in the study. Maturity was confirmed by radiographic evidence of closure of the proximal tibial and distal femoral physes.
Operative Technique
The operative procedure was performed with use of assisted general anesthesia according to a standardized protocol. Each animal was premedicated with an intramuscular injection of ketamine (ten milligrams per kilogram of body weight), xylazine (0.2 milligram per kilogram of body weight), and atropine sulphate (0.0125 milligram per kilogram of body weight) as well as with intravenous administration of a 2.5 per cent solution of thiopental sodium (eight milligrams per kilogram of body weight). General anesthesia was maintained with use of nitrous oxide and 1 to 2.5 per cent halothane with assisted ventilation. An antibiotic (cefazolin, twenty-two milligrams per kilogram of body weight) was given, starting the day before the operation, for five days.
The left hindlimb was shaved, and the area over the tibia was aseptically prepared. Six small (six-millimeter) incisions were made in the skin, and a trocar-centered drill-guide was passed through the fascia and muscle to the tibia. A 4.8-millimeter-diameter drill was used to predrill a hole in the tibial cortices, and six pins from the same group were inserted manually into the anterolateral part of the tibia. Three pins were inserted in the proximal end of the tibia and three, in the distal portion. The pins were numbered 1 to 6 from proximal to distal.
After implantation of the pins, a unilateral external fixator (Orthofix) was mounted on them. The medial portion of the mid-part of the tibial diaphysis then was exposed, and a five-millimeter-long cylinder of bone between the proximal and distal groups of pins was removed so that load would be borne by the bone-pin interfaces.
Postoperative Treatment
After the operation, the sheep were placed in heated cages in the recovery room, and they were closely monitored for twenty-four hours. Unrestricted weight-bearing was allowed immediately postoperatively. Ketoprofen (500 milligrams per day) was given for pain for two days. After the twenty-four-hour period of recovery, the sheep were returned to regular housing facilities, where they were monitored twice a day and allowed normal activity. The sheep were killed six weeks after the operation.
Radiographic Analysis
Standardized radiographs were made immediately after the operation and at six weeks, after the sheep had been killed. At six weeks, the tibia was removed and placed on a radiographic plate with the longitudinal axis of the pin parallel to the plate. Anteroposterior radiographs then were made with a distance of one meter between the x-ray source and the film and an exposure of 0.15 second, fifty kilovolts, and sixty milliamperes. Rarefaction of the pin track, defined as any radiolucency around a pin where it entered or exited the cortex, or both, was determined in a blinded manner by an independent investigator who evaluated the anteroposterior radiographs of the tibia. Furthermore, the area of resected bone was examined on anteroposterior and lateral radiographs and was classified as united or non-united. It was considered united if the gap between the bone ends had disappeared completely, and it was classified as non-united if there was any visible gap.
Biomechanical Analysis
Software (Moon System, Bologna, Italy) for measuring and recording the torque resistance of the pins was developed. An adapter (BLM; Cusano Milanino, Milan, Italy) was manufactured to connect the pins to a torque-meter, which was connected to a personal computer. After a pin had been inserted to the desired depth in the tibia, torque was applied through the torque-meter, in the direction of tightening the pin, and the final insertion torque was measured. Immediately after the animal was killed, extraction torque was measured during removal of five pins (pins 1, 2, 3, 5, and 6) from each tibia with use of the same system as that for measuring the insertion torque.
Histological, Scanning Electron Microscopy, and Histomorphometric Analyses
A two-centimeter-long segment of diaphyseal bone containing pin 4 was obtained from each sheep and was processed for morphological analysis. Each segment was isolated and was fixed in a 10 per cent formalin solution buffered at pH 7.2. Dehydration was performed with warm (37 to 40 degrees Celsius) methanol under vacuum, which was followed by impregnation and embedding in methylmethacrylate (Technovit 7200; Kulzer System, Norderstedt, Germany). Transverse sections were obtained with use of an Exact cutting system (boron-nitride-blade saw; Kulzer System, Norderstedt, Germany). The sections were made thin (fifty micrometers) with pieces of sandpaper with progressively smaller granules; care was taken to avoid damaging the bone-pin interface. Non-deplasticized sections were microwave-stained with basic fuchsin and light green in order to distinguish between mineralized bone (green) and osteoid or connective tissue (red-orange).
Infection was defined histologically by the presence of neutrophils or lymphocytes in the soft tissue around the pin.
The specimen containing pin 4 also was used for histomorphometric analysis. Bone-pin contact was evaluated in a blinded manner, by an independent investigator, on photomicrographs made at sixty times magnification with use of a light microscope (Leica Aristoplan; Leitz Wetzler, Heidelberg, Germany) connected to an image analyzer (ASM 68; Leitz Wetzler). The percentage of bone-pin contact where the pin entered and exited the cortex was measured, and the mean of the two percentages was calculated.
Each specimen with a hydroxyapatite-coated or titanium-coated number-4 pin was analyzed with scanning electron microscopy (Scan 200; Cambridge Stereo, Oxford, United Kingdom) to confirm bone-pin contact at higher magnification. To accomplish this, we selected one area, at the cortical entry site, in which there had appeared to be complete bone-pin contact at sixty times magnification. This area was studied at 100, 1000, and 10,000 times magnification to identify accurately the bone and the sides of the pin. Then, the interface was examined at 10,000 times magnification to determine whether bone-pin contact, which was defined as the absence of any measurable gap between the bone and the pin, was indeed present along the entire visible interface.
One randomly chosen specimen of cortical bone containing a hole from which a pin had been extracted was removed from each sheep. The specimen was embedded in polymethylmethacrylate and processed for sectioning without decalcification to evaluate the mechanism of failure during extraction of the pin.
Selected hydroxyapatite-coated pins were processed for histological analysis after extraction to evaluate the mechanism of failure at the bone-pin interface.
Statistical Analysis
The data in the three groups were compared with use of analysis of variance with the Duncan test and the Student t test (SPSS/PC+ software; SPSS, Chicago, Illinois). The level of significance was set at p < 0.05.
The mean thickness of the hydroxyapatite coating (Group B) was 56 ± 19 micrometers, and the mean thickness of the titanium coating (Group C) was 47 ± 11 micrometers. The mean surface roughness was 0.3 ± 0 micrometer in Group A (uncoated pins), 4.9 ± 0.3 micrometers in Group B, and 6.2 ± 0.8 micrometers in Group C (p < 0.001 for Group A compared with Group B, for Group A compared with Group C, and for Group B compared with Group C).
All sheep recovered from the operation and could walk within forty-eight hours. No major postoperative complications occurred.
At six weeks, rarefaction of the pin track was observed on the radiographs of twenty-nine specimens in Group A, fifteen in Group B, and thirty in Group C (p = 0.021 for Group A compared with Group B and p = 0.016 for Group B compared with Group C) (Table I and Figs. 1-A, 1-B, and 1-C. The site of the osteotomy healed in only two sheep, both of which were in Group B.
The mean final insertion torque was 4360 ± 1050 newton-millimeters in Group A, 3420 ± 676 newton-millimeters in Group B, and 3740 ± 643 newton-millimeters in Group C. With the numbers available, no significant differences were found among these values. The mean extraction torque was 253 ± 175 newton-millimeters in Group A, 3360 ± 1260 newton-millimeters in Group B, and 1720 ± 1030 newton-millimeters in Group C (p = 0.002 for Group A compared with Group B, p = 0.017 for Group A compared with Group C, and p = 0.03 for Group B compared with Group C). The extraction torque was significantly lower than the corresponding insertion torque in Group A (p < 0.001) and Group C (p = 0.003). No significant difference could be detected between the extraction and insertion torques in Group B (hydroxyapatite-coated pins), with the numbers available (Table I).
Histological examination at sixty times magnification showed direct bone-pin contact along 16 ± 9 per cent of the interface in Group A, 30 ± 12 per cent in Group B, and 28 ± 15 per cent in Group C (p = 0.042 for Group A compared with Group C) (Table II). In addition to the decreased percentage of bone-pin contact, the specimens in Group A were found to have many areas of bone resorption and fibrous encapsulation (Fig. 2). Better osteointegration and many areas of direct bone-pin contact were seen in Group B (Fig. 3). The osteointegration in Group C looked very similar to that in Group B (Fig. 4), but at 10,000 times magnification gaps of one to three micrometers—that is, no real bone-pin contact—were seen around all of the titanium-coated pins (Fig. 5). In contrast, contact was found between bone and the hydroxyapatite-coated pins even at 10,000 times magnification (Fig. 6).
No evidence of pin-track infection was found on histological examination of any specimen. There was no breaking or sloughing of the hydroxyapatite or titanium coating of any pin.
Histological observations of the holes after removal of the pins showed no fractures in the surrounding bone in any of the three groups. Neither the uncoated nor the titanium-coated pins left metallic particles in the bone. The hydroxyapatite-coated pins left small fragments of hydroxyapatite, approximately ten micrometers thick, in direct contact with the pin track.
The hydroxyapatite coating of the extracted Group-B pins looked intact, and no metallic substrate was visible. Histologically, very few small fragments of bone were seen between the threads of these pins, and there was no detachment of the hydroxyapatite coating from the metallic core of the pin.
The bone-pin interface is considered by some investigators to be the weakest link in the stability of external fixation1. Lack of stability of the bone-pin interface can lead to loosening of the pin and infection and thus affect the clinical results of treatment1.
The biomechanical and histomorphometric analyses that we used to examine the bone-pin interface are more precise than radiographic or clinical observations, which are more subjective.
The extraction torque for the hydroxyapatite-coated pins was significantly higher than that for the titanium-coated pins (p = 0.03) or the uncoated pins (p = 0.002), and the extraction torque for the titanium-coated pins was significantly higher than that for the uncoated pins (p = 0.017). These results demonstrate the effectiveness of a coating that provides a favorable environment for osteointegration. The greater surface roughness of the coated pins (p < 0.001 for both Group B and Group C) also may have helped to increase the initial mechanical stability of the pins. The higher extraction torque for the hydroxyapatite-coated pins compared with that for the titanium-coated pins appears to be due to the superior ability of hydroxyapatite to osteointegrate. Histological studies of retrieved titanium porous-coated prostheses have demonstrated a lack of biological fixation by direct adhesion of bone to the surface of the implant4,7. However, many studies4,7,12,13 have shown that a hydroxyapatite coating can bond directly with bone. This ability was confirmed by the histological and histomorphometric analyses in the present study. At sixty times magnification histomorphometric analyses showed the same percentage of bone-pin contact in the specimens with a hydroxyapatite-coated pin and those with a titanium-coated pin, but at 10,000 times magnification direct contact was observed only in the specimens with a hydroxyapatite-coated pin.
The extraction torque was significantly lower than the insertion torque in Groups A and C (p < 0.001 for Group A and p = 0.003 for Group C). This progressive deterioration of strength had already been observed, in other studies of animals1,3,14-16, at the interfaces of uncoated pins of various shapes, including the tapered pins used in the present study. It should be noted that the results regarding the extraction torque of uncoated pins, reported at comparable periods after implantation, have been similar, regardless of the size or shape of the pin or the technique of insertion1,3,10,15. It is likely that there is a more-or-less severe deterioration of strength at the interface of any metallic pin as a result of mechanical and thermal damage of cortical bone during insertion of the pin, with subsequent formation of fibrous tissue10. Osseous stress at the pin track may subsequently increase the fibrous encapsulation and cause bone resorption, loosening of the pin, and infection1,3,10,15,16. The ability of a hydroxyapatite coating to improve osteointegration of the implant prevented any mechanical deterioration. The better osteointegration facilitated bone-remodeling and direct osseous coverage of the surface of the pin. The biomechanical results in our study confirmed those of previous studies of hydroxyapatite-coated pins of a different shape2,3,10,11. Although a direct comparison of the studies is not possible because of differences with regard to the shape of the pin, the technique of insertion, and the type of fixator, it should be emphasized that the extraction torque of the tapered hydroxyapatite-coated pins in the present investigation is the highest for hydroxyapatite-coated pins ever reported under similar experimental conditions2,3,10, to our knowledge.
The improvement in the stability of the interface between bone and hydroxyapatite-coated pins should reduce the frequency of loosening of pins and the prevalence of infection. Clinically, it is important to achieve a more stable bone-pin interface early in the course of treatment with external fixation, a crucial period because the fracture is unstable1,3,5,9,10. A stable bone-pin interface also may be particularly useful for treatment of long duration or when the surgeon decides to change from static to dynamic fixation.
Histological analysis after the hydroxyapatite-coated pins had been removed showed no fractures of the surrounding bone and some small particles of hydroxyapatite in direct contact with the pin track. These particles do not appear to be a disadvantage, as particles of hydroxyapatite have been shown to stimulate the healing of small osseous defects18. The presence of these small particles and the fact that the coating was not detached from the metallic substrate of the extracted pins demonstrates that the failure occurred in the outer layers of the hydroxyapatite coating.
Although additional force was necessary to remove the hydroxyapatite-coated pins, it was always possible to remove them manually, without the use of a special tool. In a clinical study, the extraction of hydroxyapatite-coated bicylindrical pins was more painful than the extraction of uncoated pins11. However, neither general nor local anesthesia was needed and pain occurred only at the beginning of the process, lasting just for the moment needed to loosen the bone-pin link11. The advantage afforded by the better stability of the pin certainly outweighs the disadvantage of the brief period of discomfort during extraction.
In conclusion, we demonstrated that a hydroxyapatite coating improved the strength of the interface between bone and tapered pins. Titanium-coated pins also yielded better results than uncoated pins. The stronger fixation and the better stability at the interface of hydroxyapatite-coated tapered pins should lead to a substantial decrease in the frequency of pin-loosening and infection and consequently to better clinical results after treatment with external fixation.