The aim of this study was to compare a combination of a locking system with self-tapping (ST-L) or self-drilling-tapping (SDT-L) screws with a combination of conventional miniplates with self-tapping (ST) and self-forming (SF) screws. A standardized osteotomy and osteosynthesis with one of the above mentioned systems was performed in 24 sheep. Callus formation was measured with the help of CT scans assisted by a navigation system. Specimens of each osteotomy gap were taken and examined histologically. The best results were observed when self-tapping screws and the Mini-Locking-System (ST-L) were applied. The slowest healing was seen in animals treated with miniplates and SF screws. After 8 weeks an increase in bone formation could be observed in the ST, SF, SDT-L systems. The results after 8 weeks were comparable with those achieved by the ST-L system after 4 weeks. The improved stability of the osteosynthesis with the ST-L system resulted in early ossification of the osteotomy gap and the smallest amount of callus formation.
Over the last two decades, miniplate osteosynthesis has become an accepted method for the treatment of mandibular fractures . The advantages are that it is easier to apply the osteosynthesis material via the intra-oral approach, compared with the extra-oral approach with extensive exposure of bone, and that miniplates easily adapt to the surface of the bone .
A requirement of osteosynthesis of mandibular fractures is adequate fixation that withstands functional strain; especially as postoperative mandibulo-maxillary fixation is not intended. In general, the stability of the plate fixation relies on the pressure on the bone surface established by the screws , which may endanger the bone’s circulation. Pressure around the screws and under the plate may weaken the bone and endanger the stability of the osteosynthesis. A lack of stability of the osteosynthesis may cause screw loosening, plate fractures, impaired healing and a higher risk of infection. The complication rate for miniplate osteosynthesis is 4–31% . The region of the mandibular angle shows the highest complication rate, due to the dislocating forces of the masticator muscles.
The aim of this study was to compare the applicability and stability of different miniplate osteosynthesis systems. The plates were applied after defined osteotomies in sheep mandibles. The effects of different screw and plate designs on the stability of the osteosynthesis were evaluated by comparing the explantation torque, the amount of callus formation at the osteotomy site and the histological findings around the screws.
Materials and methods
This in vivo study was approved by the 131th assembly of the ethics committee of the animal trial council of the primary administrative division of the state of Baden-Württemberg, Germany. It was designed as an open, controlled, parallel grouped, in vivo study comparing 4 different osteosynthesis systems.
Osteosynthesis systems and application
System 1 used miniplates with self-tapping (ST) screws ( Fig. 1 a) . In this group, 2.0 mandible cortical screws from the COMPACT 2.0 system (Item No. 401.136, Synthes ® , Oberdorf, Switzerland) were inserted. They were manufactured from commercially pure titanium (ISO 5832-2). The external diameter was 2.0 mm, the core diameter was 1.4 mm. The screw head measured 3.5 mm in diameter and featured a cruciform recess. The thread pitch was 1.0 mm and the length was 3.5 mm. The screws were inserted manually after drilling a pilot hole into the bone (Item No. 317.760, Synthes ® , Oberdorf, Switzerland). A drill bit with a 1.5 mm diameter and a stop at 6 mm drilling depth was used. A new drill bit was used for every animal to maintain a similar quality of hole. A six-hole plate with a small bar ( Fig. 1 e) was inserted for fixation.
System 2 used miniplates with self-forming (SF) screws ( Fig. 1 b). 6 mm long cortical screws (Stardrive ® 2.0 from the COMPACT 2.0 system; Item No. 401.136, Synthes ® , Oberdorf, Switzerland) were inserted. The screws were made of a titanium-aluminium-niobium alloy (TiAl6Nb7, ISO 5832-11), which is harder than the commercially pure titanium (ISO 5832-2) used for the plates. The external diameter measured 2.0 mm, the core diameter was 1.35 mm. The thread pitch was 0.75 mm and the head diameter 3.3 mm. The screwdriver interface of the Stardrive ® system resembled a star-like socket. The screw tip was shaped conically like a wood screw. It was fluted with a notch to transport bone debris. Drilling of 1 mm diameter (0.35 mm less than the core diameter) with a new drill bit (Item No. 316.452, Synthes ® , Oberdorf, Switzerland) for each animal was performed to maintain a similar quality for the holes. The same six-hole plate with a small bar was inserted for fixation as in the ST group ( Fig. 1 c).
System 3 used a Mini-Locking-System with self-drilling-tapping (SDT-L) screws ( Fig. 1 d). The plates and screws of the Mini-Locking-System were produced as prototypes by the AO Development Institute (Davos, Switzerland). The dimensions of the plates and screws and the material used were similar to the miniplates. The difference was that the outer thread of the screw heads locked into the corresponding conical threaded plate holes. The corresponding screws had the same diameter (total 2.0 mm, core 1.4 mm) and length (6.0 mm) as the miniplate screws. They had a 2.9 mm long two-lipped spiral drill tip and were fluted for debris transport. Additionally a tapping function of two thread turns was integrated into the tip. The bone thread had a pitch of 0.6 mm per turn. The conical screw heads had a separate external thread to lock into the plate holes. The advancement per turn was 0.3 mm (two staged thread). The external diameter of the heads measured 3.0 mm. The screwdriver interface was the Stardrive ® system. The screws were made from a titanium–aluminium–niobium alloy (TiAl6Nb7, ISO 5832-11). The threads of the plates were designed to adapt to the threads of the screw heads as soon as they deviated from an exact orthogonal angle. The screws combined self-drilling and self-tapping, which made drilling and tapping unnecessary. Insertion took place at 300 rpm and at a maximum torque of 35 N cm by machine insertion.
System 4 used the Mini-Locking-System with self-tapping (ST-L) screws ( Fig. 1 e). In this experiment only manual insertion was performed after predrilling (by machine) with a 1.5 mm diameter, resulting in self-tapping behaviour (ST-L). The same material as for the SDT-L screws (titanium–aluminium–niobium alloy, TiAl6Nb7, ISO 5832-11) was used. They measured 6 mm in length and the screwdriver interface featured the Stardrive ® system. The bone thread pitch was 0.75 mm per turn and 0.375 mm per turn for the head thread pitch. The external diameter was 2.0 mm and the core diameter was 1.4 mm. The tip was slightly conical with a notch over three spiral turns. The plates were the same as the Mini-Locking plates which were used with the STD-L screws ( Fig. 1 f).
Animals and surgical technique
24 sheep were used in the experiment; 3 animals in each osteosynthesis system and observation period. 4 and 8 weeks were chosen as observation times. The average age of the 24 sheep was 2 years 6 months. Before the operation, all animals were examined by veterinary medical staff, weighed, de-wormed and kept under quarantine for 2 weeks. Each animal was kept under the same Good Laboratory Practice conditions and looked after by a specialist veterinary surgeon. Nutrition was the same for all sheep and their feeding behaviour was surveyed postoperatively. They were weighed weekly. The average weight at surgery was 65.1 kg (SD ± 3.2 kg).
The animals were anaesthetized using intubation anaesthesia carried out by a veterinary specialist. After the site of surgery had been shaved, the skin was disinfected thoroughly and a sterile drape was applied, an incision of about 10 cm was made in the submandibular median line. The muscles of the floor of the mouth could be reached after separating the subcutaneous tissue and the platysma muscle. On the right side of the mandible, the periosteum was separated to expose the surface of the bone. The surgeon took care not to injure the mental nerve and the oral mucosa. A standardized step-like osteotomy was performed between the first molar and the mental foramen ( Fig. 2 ). For the lateral and medial cut, narrow diamond-coated discs were used under permanent cooling with Ringer’s lactate. The delicate shape of these discs allowed the width of the osteotomy to be kept below 5 mm. The distance from the vertical, distal osteotomies to the first molars was 2 mm and the vertical dimension was slightly more than half of the mandible height. The horizontal split was 15 mm long. The angle of the vertical-cranial osteotomies (posterior) to the horizontal ones was 70°. The angle of the vertical-caudal incisions to the horizontal ones was 90°. After verifying the mobility of the fragments following the osteotomy the mandible osteotomy was anatomically reduced and fixed with three plates and 18 screws ( Fig. 2 ). Each animal was treated with one of the osteosynthesis systems. When inserting the screws, the authors applied 40.11 N cm in the ST system, 38.82 N cm in the SF system and 35 N cm in the Mini-Locking-System with ST-L and SDT-L screws. The torque values were preinstalled in the insertion machine. When the insertion was finished, the torque of each screw was quantified with a special torque measuring device (Model BGI, Mark-10 ® , Hicksville, USA). The wounds were closed carefully with resorbable suture material. After surgery, the animals were kept in separate boxes for 2 days. As sheep are ruminants they had to continue with their regular diet after they awoke from anaesthesia, so the osteosynthesis was loaded immediately. To assess ossification, polyfluorochrome labelling was performed by injecting bone-seeking vital-dyes subcutaneously into the groin region. Marking scheme and dosage protocols were applied according to R ahn & P erren . Animals with an observation time of 4 weeks received Calcein after the second week (Dosage 10 mg/kg, C 30 H 26 N 2 O 13 , Calcein, Item No. C 0875, Sigma–Aldrich). After the third week they were marked with Xylenol Orange in a dosage of 90 mg/kg (C 31 H 28 N 2 Na 4 O 13 S, Xylenol Orange, Item No. X 0127, Sigma–Aldrich). Sheep with an observation period of 8 weeks received Calcein after the second and the third weeks. After the sixth and the seventh weeks they were given Xylenol Orange. Animals with an observation period of 4 weeks were marked twice; sheep with an observation time of 8 weeks were marked four times.
Animals were killed after 4 and 8 weeks by an injection of 100 mg/kg phenobarbital (Narcoren ® , Merial GmbH, Halbergmoos, Germany) followed by 2 mg/kg potassium chloride (potassium chloride 7.45% Braun ® , Braun Melsungen AG, Melsungen, Germany). The head and the mandibles were examined with CT imaging.
Volume rendering of callus formation
1-mm thin-sliced, spiral CT scans of the skulls were taken (GE ProSpeed SX Power, GE Medical Systems, Milwaukee, USA) and extrapolated to 0.5 mm slices. The amount of callus formation was evaluated with a surgical tool navigator system and a modified version of the STP 3.5 software (Stryker-Leibinger, Freiburg, Germany). The region of osteosynthesis was measured and made visible in two- and three-dimensional aspects. On the single CT layers, callus formation on the osteotomy side and the control side were outlined. These data were used to create three-dimensional models and to calculate the volume. The difference between the osteotomy side and the control side was considered to present the callus volume. The volume of the control side was considered t be 100%.
Explantation torque of the screws
The soft tissue and the covering callus tissue were removed carefully to expose the heads of the screws. The torque of unscrewing was measured on eight defined screws (two screws per plate). The other screws were determined for histological survey. The same torque-measuring-device (Model BGI, Mark-10 ® , Hicksville, USA) used for tightening the screws was used. The values from the insertion and explantation of the screws were used to calculate the torque ratio in percent. To demonstrate the ratio of implantation-to-explantation torque the results of in vitro tests were chosen as the initial value.
The plate-bearing bone parts were separated from the mandible and stored in 4% formaldehyde solution (Merck, Darmstadt, Germany) over 2 weeks. Specimens were sliced and honed according to the method of S chenk . All 16 remaining screws were cut through the longitudinal axis on the slides. The specimens from the osteotomy sites were prepared parallel to the frontal level, which exposed both horizontal osteotomies (lateral and medial) on four slides. In addition, six slides with the four vertical osteotomy gaps in the transversal level were generated. Pictures of the samples were taken with a Axioscope microscope (Carl Zeiss, Wetzlar, Germany) and stored digitally. For light microscopy, the samples were stained using the method of Richardson-Levai-Laczko. The original bone was dyed light red and the newly built bone, purple. The imaging, optimization and evaluation were accomplished by the computer program, Axiovision 3.0 (Carl Zeiss, Wetzlar Germany): for the morphometric analysis of the osteotomy gaps, the total area of the osteotomy gaps was measured and the area of new bone formation was quantified. The ratio of these two values was regarded as representative of the quantity of the hard tissue formation. The amount of newly formed bone could be measured by evaluation of the fluorescence labelled areas.
The computer program, BMDP8 V (BDMP Statistical Software Inc., Los Angeles, USA) calculated the average, the standard deviation and the general mixed model analysis of variance. Tukey’s studentized range test was applied to examine the influence of the parameters of the different osteosynthesis systems and the screws. The Mann–Whitney U -test and the Wilcoxon test were used to compare the osteotomy side with the control side. A value of p < 0.05 was considered as the level of significance, p < 0.01 was regarded as very significant and a p < 0.001 was judged as highly significant.
Surgery and anaesthesia were tolerated well by the animals. In all sheep in the ST, SF and SDT-L groups and in one animal in the ST-L group a tumescence was palpable when the postoperative oedema faded. The post-mortem dissection of the mandibles showed callus formation.
The SF screws lacked sharpness. Despite predrilling, cracks occurred in the cortex of two animals due to radial compression. A part of these cracks extended from one screw hole to another. The STD-L screws also caused difficulties, when inserted by machine. Sometimes the cortical bone was so durable that the titanium screw’s drill tip was not sharp enough to enter deep enough for the tapping thread to get grip and propel. Histological analysis showed that this destroyed the bone close to the plate.
Loose screws were found in three animals and, in two animals, broken plates were found. The loose screws appeared only in systems without locking. The loose screws of the SF system occurred in the same sheep as the cracks during insertion. Each of the two animals with broken plates had lost all three plates. The failure of osteosynthesis was noted after 8 weeks with ST and after 4 weeks with ST-L screws. These five animals had to be killed before the planned time.
Most callus formed at the caudal and medial aspect of the mandible. Most callus formation occurred during the first 4 weeks. Evaluation of the volume of the non-operated left side of the mandible showed even distribution of new bone formation. There was no significant correlation between the weight and the mandible size. The actual volume of callus was evaluated based on the CT scan, by subtracting the volume of the side without osteotomy from the side of osteosynthesis ( Fig. 3 ). After 4 and 8 weeks, the highest quantity of callus was found in animals with SF screws. The lowest volume of callus was observed in the ST-L system. The difference between the SF group, ST and SDT-L was significant ( p = 0.042). There was no significant alteration between the ST and SDT-L systems. The discrepancy with the ST-L system was significant ( p = 0.021).
The ST screws presented an obvious loss of bone at the surface and in neighbouring cortical bone. The bone surrounding the ST-L screws was more compact than that in the vicinity of the ST screws with regular miniplates ( Fig. 4 ). The bone around the SF screws showed more porosity even in the cortical part, while directly at the surface of the screw, new bone formation compensated for the loss of hard tissue ( Fig. 5 ). In some of the SF screws a gap between the head of the screw and the plate was detected ( Fig. 5 ). For the SF screws, partial destruction of the bone thread was confirmed. Around the thread of the screw a new bone structure, 200 μm wide, had formed and contained remnants of old bone.