Subperiosteal preparation using a periosteal elevator leads to disturbances of local periosteal microcirculation. Soft-tissue damage can be considerably reduced using piezoelectric technology. In the study presented here, a novel device for the preparation of the periosteum was compared with a conventional periosteal elevator in an animal model.
Subperiosteal preparation using a periosteal elevator leads to disturbances of local periosteal microcirculation. Soft-tissue damage can be considerably reduced using piezoelectric technology. For this reason, we investigated the effects of a novel piezoelectric device on local periosteal microcirculation and compared this approach with the conventional preparation of the periosteum using a periosteal elevator. Material and methods. In the first part of the study, twenty Lewis rats were randomly assigned to one of two groups. Subperiosteal preparation was performed using either a piezoelectric device or a conventional periosteal elevator. Intravital microscopy was performed immediately after the procedure as well as three and eight days postoperatively. In the second part of the study, a further 50 Lewis rats underwent subperiosteal preparation with a piezoelectric device or a conventional elevator.
Specimens were obtained from these animals and examined immunohistochemically and histologically at the aforementioned time points. Statistical analysis of microcirculatory parameters was performed offline using analysis of variance (ANOVA) (p < 0.05).
Results. At all time points investigated, intravital microscopy demonstrated significantly higher levels of periosteal perfusion in the group of rats that underwent piezosurgery than in the group of rats that underwent treatment with a periosteal elevator. Immunohistochemical and histological assessments confirmed the superiority of the piezoelectric device.
Discussion. The use of a piezoelectric device for subperiosteal preparation is associated with better periosteal microcirculation than the use of a conventional periosteal elevator. As a result, piezoelectric devices can be expected to have a positive effect on bone metabolism.
The periosteum is a membrane that consists of connective tissue and covers bone. Morphologically, the periosteum can be divided into three zones, each of which contains highly specific cells. The inner zone is the osteogenic layer that contains cells similar to those of the endosteum. Among these cells are mesenchymal stem cells, osteoprogenitor cells, active and resting osteoblasts, and/or active and resting osteoclasts. The middle zone is a translucent layer that is characterised by a large number of capillaries. The outer zone is a typical fibrous layer that contains collagen fibres. 
The specific structure of the periosteum is seen not only in children but also in adults and allows bones to remodel themselves over time, for example during bone fracture healing. [2–4] Periosteal cells play a major role in the supply of blood to the bone. The importance of intact periosteal tissue is underlined by the substantial contribution of periosteal blood cells to the supply of blood to cortical bone (70–80% of arterial supply and 90–100% of venous return) when compared to intraosseous blood vessels.  The periosteum is closely attached to bone by collagen fibres in the bone matrix and by hemidesmosomes.  Surgical procedures, especially those directly involving bone, often have adverse effects on the osteogenic potential of the periosteum since they are associated with the detachment of periosteal tissue from the bone. Periosteal damage can either be caused by the deliberate separation of the periosteum from the bone during surgery or it can be the result of a disease or trauma.
The preparation of the periosteum is a routine procedure in trauma surgery, reconstructive surgery and especially dentoalveolar surgery. [7–10] It is commonly performed with a periosteal elevator that is used for manually lifting and separating periosteal tissue from the bone. This procedure causes damage to the morphological structure of the periosteum and especially to the cells of the osteogenic layer. The result is a complete or partial loss of periosteal function. [11, 12] It is currently impossible for surgeons to prepare the periosteum between the osteogenic layer and the underlying bone in such a way that the periosteum remains intact. The use of a periosteal elevator leads to the disruption of tissue at the periosteum-bone interface. The destruction of the connection between bone and periosteum damages the regenerative cells of the periosteum and reduces their osteogenic potential. [13–16] Successful osteoinduction and osteoconduction, however, require the preservation of cell vitality in the periosteum. 
Periosteal cells provide nutrition to the underlying bone by free diffusion. Adequate functioning of the periosteum is of far greater importance to patients who have underlying diseases such as diabetes mellitus or undergo tumour treatment and receive chemotherapeutic agents than it is to healthy people since the periosteum plays an important role in promoting rapid bone healing. If these patients undergo surgery involving bone, particular care must be taken to cause no damage or as little damage as possible to the periosteum with a view to ensuring subsequent bone healing without dehiscences or necrosis.  If the bone is damaged without compromising local periosteal microcirculation, good bone healing can be expected. If, by contrast, local periosteal microcirculation is compromised, the regenerative potential of the periosteum will be reduced. Good periosteal microcirculation is of paramount importance for bone modelling and remodelling.  In the literature, there is only a paucity of chronic studies on periosteal perfusion during and after subperiosteal preparation.
Whereas (piezoelectric) ultrasonic instruments have been available since 1988, devices utilizing the piezoelectric effect have been used for medical purposes only since 1998. Applications of piezoelectric devices include hard-tissue surgery, periodontal surgery, the removal of impacted teeth, apical surgery [18, 19], and bone expansion [20, 21].
The piezoelectric effect is based on physical interactions in crystalline materials. The application of an electric field creates nanoscale deformations in a crystal. This dynamic effect can be used to transform longitudinal or transverse movements of a ferroelectric material into a surgical cutting action. Piezoelectric devices are operated at different frequencies depending on the density of the tissue to be cut.
The tip of the ultrasonic device vibrates within a range of 20–200 µm at a frequency of 20,000 Hz. Piezoelectric devices are permanently cooled with sterile physiological saline during use so that heat-induced trauma can be ruled out [22, 23] and the risk of bacterial contamination is minimised.
The essential difference between piezoelectric devices and conventional preparation instruments is that piezoelectric devices operate in a tissue-specific manner. Every tissue has a specific frequency range at which it can be cut. A piezoelectric device can therefore cut a specific type of tissue without causing damage to adjacent tissues. Damage to the soft tissues (e.g. nerves) that surround bone, for example, is caused only at frequencies above 50 kHz. [24, 25] In addition, piezoelectric devices have the advantage that they cause minimal bleeding when they are used to cut bone. The extent to which piezoelectric devices adversely affect periosteal microcirculation has not yet been investigated. While there are a few studies that address the behaviour of bone when it is being cut by piezoelectric devices, there are no studies that examine local microcirculation within the periosteum during and after the cutting operation. We conducted this study in order to investigate the effects of piezoelectric surgery on local periosteal microcirculation and compared the use of a piezoelectric device and a conventional periosteal elevator for the preparation of the periosteum. This issue is of particular importance in the military setting since it is not uncommon for soldiers to sustain injuries to the facial skeleton and to the extremities during attacks and similar incidents. The periosteum plays a key role in the healing of these injuries and its function must be preserved as far as possible since an extensive loss of soft tissue and the separation of periosteal tissue lead to compromised vascularity and are thus associated with poor bone healing. 
Material and Methods
This study is based on animal experiments involving Lewis rats. Microcirculatory parameters were assessed and histological sections were examined.
All procedures were approved by the responsible authority (Ref. 12/0861) and were performed in accordance with the German Animal Protection Act and the Guide for the Care and Use of Laboratory Animals . The study involved 70 adult male Lewis rats with a body weight between 300 g and 330 g (Harlan-Winkelmann, Borchen, Germany). The rats were housed singly in cages at a room temperature of 22–24°C and a relative humidity of 60–65% with a 12-hour day/night cycle. They received water and dry food (Altromin, Lage, Germany) at libitum during the entire investigation.
Study Design and Experimental Groups
Microcirculatory parameters were assessed on day 0 immediately after subperiosteal preparation with the different instruments and on days 3 and 8 after the procedure. The experiments were performed on the basis of a model established by Stuehmer et al. . The rats (n=20) were divided into two experimental groups.
Group 1 n=10, subperiosteal preparation with a periosteal elevator, intravital microscopy
Group 2 n=10, subperiosteal preparation with a piezoelectric device, intravital microscopy
Immuno-histochemical and histological sections were examined after the animals had been killed. This part of the study involved 50 rats that were divided into five experimental groups.
Group 1 n=10, control group
Group 2 n=10, subperiosteal preparation with a piezoelectric device, immuno-histochemistry and histology after three days of healing
Group 3 n=10, subperiosteal preparation with a periosteal elevator, immunohistochemistry and histology after three days of healing
Group 4 n=10, subperiosteal preparation with a piezoelectric device, immunohistochemistry and histology after eight days of healing
Group 5 n=10, subperiosteal preparation with a periosteal elevator, immuno-histochemistry and histology after eight days of healing.
The animals were anaesthetised using an intraperitoneal injection of ketamine (Ketavet®, 75 mg per kg bodyweight, Parke-Davis, Germany) and xylazine (Rompun®, 25 mg per kg bodyweight, Bayer HealthCare, Germany). A surgical blade was used to make an incision through the skin and periosteum in the occipital region in order to expose the calvaria. Depending on the group, either a periosteal elevator or a piezoelectric device was used for the preparation procedure. The skin was then repositioned and secured in place with sutures (Ethicon Vicryl® sutures 4–0, Johnson & Johnson, Germany). The procedure took approximately ten minutes. Intravital microscopy was performed subsequently. Periosteal vascularisation was analysed by intravital microscopy on the following days at the time points indicated above. Every microscopic examination took approximately thirty minutes. After either three or eight days of healing, the animals were killed using an overdose of anaesthetics. Specimens were obtained and prepared for histological and immuno-histochemical analyses.
Intravital Fluorescence Microscopy of the Periosteum
Under anaesthesia with intraperitoneal ketamine (Ketavet®, 75 mg per kg bodyweight) and xylazine (25 mg per kg bodyweight), intravital fluorescence microscopy was performed immediately after the preparation of the periosteum and on days 3 and 8 after the procedure. Fluorescein-isothiocyanate-labelled dextran (FITC-dextran, molecular weight: 150,000 Da, Sigma, Taufkirchen, Germany, 5% in 0.9% NaCl solution, 0.1 ml) was injected into the tail vein of each animal for contrast enhancement of blood plasma. This technique permitted the imaging of microcirculation. All examinations were recorded on-line using a highly sensitive video camera and quantitatively analysed (off-line) with computer assistance at a later time in order to minimise examination times. Reflected light fluorescence microscopy was performed using a Zeiss Axiotech microscope (Zeiss, Oberkochen, Germany) at 20x magnification. A blue filter block (450–490 nm) permitted the visualisation of blood plasma. Microscopic images were recorded using a highly sensitive video camera (FK 6990 IQ-S, Pieper, Schwerte, Germany) and transferred to a DVD system (LQ-MS 800, Panasonic, Hamburg, Germany) for off-line evaluation.
Histology and Immunohisto-Chemistry
Formalin-fixed and paraffin-embedded specimens were cut into 5-µm-thick sections, stained with haematoxylin and eosin (H&E) and examined by microscopy (DM4000B Leica Mikrosysteme, Wetzlar, Germany). Formalin-fixed and paraffin-embedded specimens were also cut into 5-µm-thick sections for immunohistochemical analysis. The following antibodies were used: rabbit anti-collagen type I (1:800, BIOLOGO, Kronshagen, Germany), rabbit anti-collagen type IV (1:400, Acris Antibodies GmbH, Hiddenhausen, Germany), rabbit anti-collagen type VI (1:200, Acris Antibodies GmbH, Hiddenhausen, Germany), mouse anti-osteocalcin (1:200, QED Bioscience Inc., San Diego, USA), and mouse anti-SPARC (1:200, Santa Cruz Biotechnology, Santa Cruz, USA). A biotin-conjugated goat anti-rabbit antibody (1:600, Dianova, Hamburg, Germany) or a biotin-conjugated goat anti-mouse antibody (1:200, Dianova, Hamburg, Germany) was used as a secondary antibody. Incubation with streptavidin-horseradish peroxidase (Dianova, Hamburg, Germany) was followed by colour development with aminoethylcarbazole (AEC) substrate (Axxora Deutschland GmbH, Loerrach, Germany) at room temperature. Colour development was stopped under microscopic control by washing with water. The sections were counterstained with haematoxylin (Merck, Darmstadt, Germany) and examined by light microscopy (DM4000B Leica Mikrosysteme, Wetzlar, Germany).
Analysis of Intravital Fluorescence Microscopy
Computer-assisted quantitative image analysis was performed off-line using CapImage image analysis software (Zeintl, Heidelberg, Germany). Functional capillary density, micro vessel diameters and volumetric blood flow were determined in the venules. Functional vessel density was assessed on the basis of the length of perfused micro vessels per observation area. Diameters (d) were measured perpendicular to the vessel path and are expressed in mm. Volumetric blood flow was calculated using the formula: π x (d/2)2 x v/K, where K represents the Baker-Wayland factor to correct for the parabolic velocity profile in micro vessels with a diameter > 20 µm.
Histological analysis was performed using analySIS software (Soft Imaging System GmbH, Muenster, Germany) and a Leica DM4000B light microscope (Leica Camera AG, Solms, Germany). H&E staining was used for a descriptive analysis of the periosteum-bone interface. The specimens were stained for collagen type I and type IV and osteocalcin for immunohistochemical analysis. The piezoelectric device groups were then compared with the periosteal elevator groups.
Normal distribution and homogeneity of variance were assessed. Results are expressed as means and standard errors of measurement (SEM). Differences between groups were evaluated with a one-way analysis of variance (ANOVA) on ranks. Differences within groups were also analysed by ANOVA. Student-Newman-Keuls or Duncan post-hoc tests were used to isolate specific differences. A p-value
< 0.05 was considered significant. Data was collected and analysed using Microsoft Office Excel 2007 and IBM SPSS (Statistics 21, IBM Deutschland GmbH, Germany).
Intravital fluorescence microscopy
Periosteal microcirculation was imaged in detail using intravital fluorescence microscopy. The group of rats whose periosteum had been prepared with a piezoelectric device was compared with the group of rats whose periosteum had been prepared with a periosteal elevator.
Functional capillary density
The periosteal elevator groups showed an increase in functional capillary density from day 0 to day 8. This density, however, was always lower in the periosteal elevator group than in the piezoelectric device group. In the piezoelectric device group, mean functional capillary density decreased by 11.73 cm/ cm2, from 80.69 cm/cm2 on day 0 to 68.96 cm/cm2 on day 3. It then increased to 127.95 cm/cm2 on day 8 and was higher than in the periosteal elevator group at all time points. These results show that a major increase in mean functional capillary density occurred no earlier than after eight days of healing. From day 3 to day 8, mean functional capillary density increased by 387% in the periosteal elevator group and by 185% in the piezoelectric device group. The difference between the rats whose periosteum had been prepared with a periosteal elevator and the rats whose periosteum had been prepared with a piezoelectric device was significant postoperatively as well as after three and eight days of healing (p > 0.05). During the entire observation period, mean functional capillary density was higher in the piezoelectric device group than in the periosteal elevator group. Capillary density in the periosteal elevator groups was less than half as high as that observed for the piezoelectric device group on days 0 and 3. Densities were more similar after eight days of healing. At this time point, the difference between the two groups in mean capillary density was only 7.8 cm/cm2. Means and standard deviations are given in the next table.
Red blood cell velocity
On day 0, red blood cell velocity was 0.31 mm/s (± 0.12) in the periosteal elevator group and 0.69 mm/s (± 0.43) in the piezoelectric device group. During eight days of healing, both groups showed an increase in mean red blood cell velocities. The highest increase was noted for both groups on day 8. At this time point, mean red blood cell velocity was 1.76 times higher in the piezoelectric device group than in the periosteal elevator group. During the entire observation period, the differences between the two groups were significant (p > 0.05). At all time points, mean red blood cell velocities were significantly higher in the piezoelectric device group than in the periosteal elevator group. The highest velocity (2.93 mm/s) was measured on day 8 for an animal in the piezoelectric device group and was 2.25 times higher than the mean red blood cell velocity. Red blood cell velocities were almost constant in the piezoelectric device group and increased only moderately in the periosteal elevator group during the observation period. Means and standard deviations are shown in the next table.
The periosteal elevator group showed a significant increase in mean vessel diameter from day 0 to day 3 (p > 0.05). After eight days of healing, the mean vessel diameter was smaller than on day 3 but not as small as that measured postoperatively. In the piezoelectric device group, the mean diameter of perfused vessels decreased from day 0 to day 3 and then increased until day 8. After eight days of healing, the mean vessel diameter was similar to that measured on day 0. The difference was only 0.33 µm. A comparison of the two groups showed that the mean diameters in the periosteal elevator group were significantly smaller than those in the piezoelectric group during the entire observation period (p > 0.05). Means and standard deviations are given in the next table.
Light microscopical examinations of haematoxylin-eosin-stained specimens showed various degrees of changes in bone morphology and histo-morphology.
There were considerable qualitative differences at the periosteum-bone interface between the periosteal elevator group and the piezoelectric device group. When the periosteum was prepared using the novel piezoelectric device, the bone surface was smooth and showed no evidence of mechanical damage. The different histological layers of the periosteum were clearly demarcated. The outer fibrous layer and the inner cambium layer were clearly visible. Fat vacuoles and collagenous connective tissue were identified as well.
When the periosteum was prepared with a conventional periosteal elevator, the bone surface showed clear evidence of mechanical damage resulting from the use of the instrument. The inner zone of the periosteum was torn from the bone and there was no clear demarcation between the different layers. This type of damage did not occur when the piezoelectric device was used.
Collagen type I
In both groups, collagen type I levels were approximately identical on day 3 but different on day 8. At the latter time point, collagen type I levels were significantly higher in the piezoelectric device group than in the periosteal elevator group.
Collagen type IV
After the surgical intervention, collagen type IV levels in the piezoelectric device group were similar to those obtained for the periosteal elevator group and significantly (six times) higher than those of the control group. After eight days of healing, collagen type IV levels had almost returned to those of the control group. At this time point, collagen type IV levels were still (five times) higher in the periosteal elevator group.
During the observation period of eight days, there was a significantly more pronounced increase in osteocalcin levels in the piezoelectric device group than in the periosteal elevator group.
In the study presented here, a novel device for the preparation of the periosteum was compared with a conventional periosteal elevator in an animal model. The technique that was used in this study is an established method. It has been used by Menger et al. in the past twenty years and allows us to compare the various groups.
Microvascular perfusion of different types of tissues can be investigated in vivo by a variety of methods such as laser Doppler flowmetry and polarographic oximetry. [3, 27, 28] The main disadvantage of these methods is that tissue perfusion can be imaged only indirectly and that no information about the perfusion of individual micro vessels can be obtained. By contrast, intravital microscopy offers the possibility of studying the perfusion of individual micro vessels even over a prolonged period of time. [27, 29] This method has been shown to be suitable for investigating periosteal perfusion in other studies. [30, 31] We determined functional capillary density, blood flow within micro vessels and the diameters of micro vessels in the periosteum in order to investigate whether a piezoelectric device causes less irritation to micro vessels than a conventional periosteal elevator.
Our results show that the use of the piezoelectric device for the preparation of the periosteum was associated with a considerably higher post-procedural periosteal blood flow than the conventional method with a periosteal elevator.
One possible explanation is that the use of a piezoelectric device leads to the formation of fewer micro thrombi during subperiosteal preparation than a periosteal elevator. Functional capillary density was significantly higher after preparation with a piezoelectric device. As a result, a considerably higher number of perfused vessels were available for periosteal supply. In addition, the piezoelectric device was associated with a significantly higher microvascular blood flow than the periosteal elevator. Histological assessments of the effects of trauma on tissue and the immuno-histochemical staining of tissue specimens are common methods for examining tissue.  In the study presented here, the analysis of histological sections shows that a piezoelectric device is superior to a conventional periosteal elevator in preparing the periosteum.
Vessel density in the periosteum plays an important role in the supply of blood to bone.  Every surgical procedure that leads to subperiosteal exposure results in a decrease in periosteal perfusion.  Several studies reported that piezosurgery is an atraumatic process that causes only minimal tissue damage.  In the future, this technique can play a key role in the management of compromised patients since the vascular layer of the periosteum is largely preserved. This is one of the principles of biological osteosynthesis, which is used in the fields of orthopaedics and trauma surgery. Periosteal preparation with a piezoelectric device can be an option especially in the treatment of fractures that soldiers sustain during attacks and similar incidents and that are associated with severely compromised tissue. The use of piezoelectric devices for periosteal preparation may considerably improve the outcome of patients with injury patterns similar to those seen in military operational settings.
The results reported here show that the use of a piezoelectric device for the preparation of the periosteum has considerable advantages. Further studies are required to investigate possible effects in patients who have comorbidities and, for example, are treated with bisphosphonates, chemotherapeutic agents or other medications and in soldiers who sustained blast injuries that are challenged by poor soft-tissue quality and may include thermal injuries. Such studies are underway but results are not yet available.