1. Introduction
External fixation is associated with a risk of pin-tract infection in skin and bone tissues, although external fixation is a minimally invasive and useful method for treating bone fractures and deformities [
1–
7]. Mechanical instability of external fixation pins that leads to loose anchorage at the bone-pin interface can lead to pin-tract infections [
8,
9]. In the worst-case scenario, chronic osteomyelitis develops because of bacterial infection in a pin tract [
8]. A promising strategy for reducing the infection risk is to improve fixation strength at the bone-pin interface, especially for long-term implantation.
Apatite coatings on external fixation pins have improved the fixation strength at the bone-pin interface [
10–
16]. Plasma-sprayed apatite coating has increased the extraction torque of various external fixation pins by 53%–124% compared with that of uncoated pins [
10,
11]. Other approaches to form an apatite coating include those using supersaturated calcium phosphate (CaP) solutions [
15–
21]. Conventional supersaturated solution methods have been generally two-step techniques composed of a physicochemical modification step of substrates and an immersion step in a supersaturated CaP solution prepared from chemical reagents [
17–
19]. We developed a method wherein an apatite layer was formed on anodically oxidized titanium (Ti) pins just by immersing the pins in an infusion fluid-based supersaturated CaP solution at 37 °C for 48 h [
15,
16]. This particular solution has the advantage of biological safety over other supersaturated CaP solutions prepared from chemical reagents. The resulting apatite layer on Ti pins led to the formation of dense bone in the cortical and medullary cavity regions after implantation. As a result, the extraction torque values of Ti pins coated with the apatite layer were significantly higher (by 29.9%–46.5%) than those of uncoated pins after percutaneous implantation for four weeks in rabbits [
15,
16]. However, the extraction torque data used were those for unloosened pins corresponding to “no redness” and “skin infection” cases diagnosed by macroscopic visual inspection. Cases with screw loosening were diagnosed as “osteomyelitis,” and their data were excluded from the extraction torque analysis. Hence, the extraction torque information used previously consisted of data for cases of no infection, skin infection without osteomyelitis, and skin infection with osteomyelitis but without screw loosening. Therefore, it is not clear whether apatite-coated Ti pins have an advantage in fixation strength over uncoated Ti pins under different osteomyelitic conditions.
A method that immerses Ti pins at room temperature (25 °C) is clinically more useful than the previous method at 37 °C. Although the room temperature methods for apatite layer formation using infusion fluid-based supersaturated CaP solutions have been reported for Ti rods, they were two-step techniques [
20]. The Ti rods required a pretreatment step prior to an immersion step in an infusion fluid-based supersaturated CaP solution [
20].
The first purpose of the present study was to develop a one-step immersion method at 25 °C for coating apatite on Ti pins using an infusion fluid-based supersaturated CaP solution. The second purpose was to evaluate the extraction torque of apatite-coated Ti pins in comparison with that of uncoated Ti pins. The special interest here was to clarify the relation between the degree of osteomyelitis and the extraction torque of apatite-coated and uncoated Ti pins. We therefore assessed the degree of osteomyelitis histologically.
3. Discussion
The most important finding of the present study was that the apatite layer formed by the one-step method at room temperature increased the fixation strength of the Ti screw even in the presence of partial osteomyelitis. The experiment was performed with a four-week percutaneous implantation of the Ti pin in the proximal tibial metaphysis of rabbits. The extraction torque of the uncoated Ti pins classified as Grade 1b (partial osteomyelitis) was lower than that with Grade 0b (no osteomyelitis). On the other hand, in the Ap group, the extraction torque for Grade 1b was almost identical to that for Grade 0b. Furthermore, dense bone formation was observed in the cortical and medullary cavity regions in the Ap group, even in the presence of partial osteomyelitis, whereas no or very little bone formation was observed in the medullary cavity in the UN group.
The reason for the increase in extraction torque in the group with Grade 0b + 1b in the Ap group is that there was dense bone formation in the cortical and medullary cavity regions caused by the osteoconductivity or osteointegrating activity of apatite [
10,
11,
15,
16]. Thus, the pin and surrounding bone tissue were likely to integrate when the pin was coated with an apatite layer [
15,
16,
22]. Based on these considerations, even when partial osteomyelitis is clinically observed in the bone-pin interface the apatite-coated Ti screw is likely to retain its bone fixation strength for a longer period than the uncoated Ti screw. This effect should be advantageous for long-term implantation of external fixation, such as for bone transport, treatment of an open fracture, or deformity correction. Plasma-sprayed apatite coatings also increase the extraction torque of various external skeletal fixation pins by 53%–124% compared with that of uncoated pins [
10,
11]. However, apatite-coated dental implants have an increased risk of bacterial colonization with an increasing ailing period and larger peri-implant defects [
23,
24]. Therefore, long-term implantation studies of the apatite-coated Ti pin to clarify such an effect on bacterial colonization are necessary.
Despite the improvement in fixation, the apatite layer formed by the 25 °C immersion process showed no infection-reducing effect in either soft or bone tissues. These results were similar to those of our previous studies [
15,
16]. In this study,
S. aureus was detected in the infected pin tract at similar frequencies for both the uncoated and apatite-coated Ti screws. The causal bacterium (
S. aureus) can be a major problem not only for humans but for animals as well [
1–
9]. To reduce pin tract infection, antibiotic-containing wound devices and/or fibroblast growth facter-2 (FGF-2) to accelerate wound healing could be necessary [
16,
20]. It has been reported that FGF-2 can be immobilized within an apatite layer by supplementing a supersaturated CaP solution with FGF-2 in the previous method at 37 °C [
16]. The resulting FGF-2-apatite composite layer on the Ti pin reduced pin tract infection rate in the same animal model [
16]. When a sponge pad made of poly(ɛ-caprolactone) containing cefazolin sodium (an antibiotic) was put on the skin around this Ti pin, the infection rate was further reduced [
25]. Such approaches would also be effective in reducing infection rate for the apatite-coated Ti pins prepared by the present room temperature method.
An apatite layer was wholly and homogeneously formed on the Ti screw even at room temperature (25 °C) within 48 h without pretreatment of the Ti pin if it is performed in a CaP solution using increased concentrations of calcium and phosphate ions compared with the previous conditions [
15,
16]. The experimental results suggested that the new CaP solution, with increased calcium and phosphate ion concentrations, is effective in forming a low-crystalline apatite layer even at room temperature. Extraction torque values are the same for the pins prepared by the one-step method at 25 °C and those prepared by the previous method at 37 °C regardless of whether there is Grade 0b osteomyelitis (none) or Grade 1b (partial osteomyelitis).
Based on these facts, the one-step method is more useful than the previous method in terms of the immersion temperature [
15,
16], with the chemistry and crystallinity of the resulting CaP different from those in the previous method. The difference in chemistry was clear as the molar ratio of the CaP prepared by the one-step method at 25 °C was 1.40–1.44, whereas the previous molar ratio was 1.575 ± 0.005 [
15]. A difference in crystallinity would be present because the lower synthetic temperature causes lower crystallinity of apatite [
26]. In addition, the layer formed in the one-step method had a submicron-scale porous structure, whereas that formed using the previous method was an aggregate of dense and nano-sized particles [
15]. Thus, the lower-temperature process has the advantage of preserving the biological activity of signal molecules if the molecules are intended to be contained in the solution for co-precipitation with apatite [
16,
19,
27].
4. Materials and Methods
4.1. Preparation of an Infusion Fluid-Based Supersaturated CaP Solution
A supersaturated CaP solution was aseptically prepared by mixing five clinically available infusion fluids: Ringer’s solution (Ca
2+ 2.25 mM) (Otsuka Pharmaceuticals, Tokushima, Japan) and calcium chloride corrective injection 1 mEq/mL (Ca
2+ 500 mM) (Otsuka Pharmaceuticals, Tokushima, Japan) as calcium sources; Klinisalz
® (PO
43− 10 mM) (I’rom Pharmaceuticals, Tokyo, Japan) and dipotassium phosphate corrective injection 1 mEq/mL (PO
43− 500 mM) (Otsuka Pharmaceuticals, Tokushima, Japan) as phosphorus sources; and Meylon
® Injection 7% (NaHCO
3 833 mM) (Otsuka Pharmaceuticals, Tokushima, Japan) as an alkalinizer. The chemical compositions of the CaP solution are summarized in
Table 2. The calcium and phosphate ion concentrations in the CaP solution were increased by 2.43 and 1.62 times, respectively, over those in our previous infusion fluid-based CaP solution to maintain a sufficient degree of supersaturation even at 25 °C by compensating for the retrograde solubility of apatite with temperature [
15,
16,
22,
28].
4.2. Immersion of Ti Pins in the Supersaturated CaP Solution
The Ti pins used were commercially available, gamma ray-sterilized titanium cancellus screws (#407-030; Synthes, West Chester, PA, USA) with an anodically oxidized surface. They were 4.0 mm diameter and 30 mm length [
15,
16,
25,
29]. Each Ti pin was immersed in 10 mL of the infusion fluid-based supersaturated CaP solution at 25 °C for 48 h followed by immersion in 2 mL of distilled water for injection (Wasser “Fuso”; Fuso Pharmaceuticals Industries, Osaka, Japan) twice for rinsing. The rinsed Ti pins were freeze-dried for later characterization of the surface layer. They were used without drying for animal experiments.
4.3. Characterization of the Surface Layer
The surfaces of Ti pins were observed using an SEM (XL30; FEI Company Ltd., Tokyo, Japan) equipped with an EDX (Genesis 2000; EDAX Japan K.K., Tokyo, Japan). The Ti screws were coated with a thin carbon film before observation. To identify the crystalline phase of the surface layer, the layers were scraped off the Ti screw and mounted on a silicon-zero-background plate for analysis using XRD (Rint 2250; Rigaku, Tokyo, Japan).
The amounts of calcium and phosphorus deposited on the Ti screws were determined by chemical analysis. Each Ti screw was immersed in 2 mL of a 10 mM citric acid-sodium citrate buffer (pH 5.43) at 25 °C for more than 3 h to dissolve the surface layer completely. The resulting solutions were analyzed quantitatively for calcium and phosphorus using ICP (SPS7800; Seiko Instruments Inc., Chiba, Japan).
4.4. Animal Experiments
The surgical technique was the same as that described in our previous studies [
15,
16,
25,
29]. Ti screws were implanted into 20 skeletally mature male Japanese white rabbits, weighing approximately 3.0 kg. The rabbits were divided into two groups: 10 rabbits in the apatite-coated Ti screw (Ap) group and the other 10 in the uncoated Ti screw (UN) group. Percutaneous implantation of Ti screws in the rabbits’ proximal tibial metaphyses (
Figure 1) was carried out following the method described previously [
15,
16,
25,
29]. Briefly, small (10 mm) incisions were made in the skin at the medial proximal tibia aseptically after an intravenous injection of barbiturate (40 mg/kg body weight). Then, a hole, 2.5 mm in diameter, was drilled in the tibial metaphysis and with individual taps for each screw. The Ti screws were then manually inserted into these holes. Two Ti screws in the same group were implanted individually in bilateral proximal tibial metaphyses of the rabbit. Hence, the total number of implanted Ti screws was 20 for each group. After implantation, the skin was apposed with two 3–0 nonabsorbable sutures. Postoperatively, each rabbit was allowed free activities in its own cage. The rabbits did not receive any antibiotics or treatment for their wounds. The rabbits did not receive any postoperative medication against pain either. All of the rabbits were sacrificed four weeks after the operation.
All animal experiments and breeding were performed under the conditions approved by the ethics committees of both the University of Tsukuba and the National Institute of Advanced Industrial Science and Technology (AIST). All activities were done in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals (
http://grants.nih.gov/grants/olaw/Guide-for-the-care-and-use-of-laboratory-animals.pdf).
4.5. Classification of Pin Tract Infections by Visual Inspection
Four weeks after implantation, pin tract infections were evaluated using a modified Checketts classification before sacrifice [
16,
25]. Grade 0v corresponds to “no redness”, in which no redness, discharge, or pin loosening was observed. Grade 1v corresponds to infections only in the soft tissue, characterized by redness and discharge around the pin without pin loosening. Grade 2v corresponds to infections in both soft and bone tissues, characterized by redness and discharge around the pin associated with pin loosening caused by osteomyelitis. A single physician who was blinded to the group identification of pins evaluated the rabbits for pin tract infections. The result was analyzed by χ
2 test for independence. The significance level was set at
p < 0.05 for each analysis.
4.6. Histological Analysis
After collection of exudate in the pin tracts, the proximal tibial metaphyses were fixed in 10% neutral buffered formalin, decalcified, and embedded in paraffin. The sections were sliced, 5 μm thick, perpendicular to the tibial longitudinal axis and parallel to the hole of the screw. They were stained with hematoxylin-eosin. The specimens were observed histologically using a light microscope (BX-51; Olympus Optical Co., Ltd., Tokyo, Japan) to evaluate the grade of pin-tract inflammation in soft and bone tissues.
For the soft tissue, Grade 0s corresponds to “no inflammation”, where no inflammation is observed in the surrounding soft tissue along the whole length of boundary lines between the pin and soft tissue. Grade 2s corresponds to “severe inflammation”, where inflammation is observed in the surrounding soft tissue along the whole length of boundary lines between the pin and soft tissue [
25,
29]. Grade 1s is a status between Grades 0s and 2s and corresponds to “partial inflammation”, where inflammation is observed in the surrounding soft tissue along only a part of the length of boundary lines between the pin and soft tissue.
Similarly, for the bone tissue, Grade 0b corresponds to “no osteomyelitis”, where no inflammation is observed in the surrounding bone tissue along the whole length of the boundary line between the screw and bone tissue. Grade 2b corresponds to “severe osteomyelitis”, where inflammation is observed in the surrounding bone tissue along the whole length of the boundary line between the screw and bone tissue [
25]. Grade 1b is a status between Grades 0b and 2b, and corresponds to “partial osteomyelitis”, where inflammation is observed in the surrounding bone tissue along only a part of the length of boundary line between the screw and bone tissue.
A single physician who was blinded to the results of the histological examination evaluated the soft and bone tissues for pin-track inflammation. The results were analyzed by χ2 test for independence. The significance level was set at p < 0.05 for each analysis.
4.7. Reanalysis of Osteomyelitis Status and Extraction Torque Data
Reanalysis for osteomyelitis status and extraction data were carried out using previous data [
16]. Evaluations of the grade of pin-tract inflammation in the soft and bone tissues were performed using histological sections, as described in Section 4.6. Extraction torque data were analyzed on the basis of the grade of pin-tract inflammation.
4.8. Bacterial Culture and Identification
After complete removal of Ti screws, exudate around each Ti screw was collected with a cotton swab. The swabs with exudate were consigned to a company for clinical laboratory testing (SRL. Inc. Tachikawa, Tokyo, Japan) to detect major bacterial species: S. aureus and E. coli.
4.9. Biomechanical Analysis
After sacrificing the rabbits, the extraction torque of the Ti screw was measured using a torque-measuring apparatus (HTG2-5N; Imada Co., Ltd., Toyohashi, Japan). The extraction torque data for the Ap and UN groups were compared using Student’s t-test at a significance level of p < 0.05.