Sequential intravital labeling of the regenerating bone was performed postoperatively with 3 % alizarin (0.83 ml/kg body weight) in week one (given on the day of surgery), 1 % calcein green (5 ml/kg body weight) in weeks two and three (7 and 14 days postoperatively), and 6 % xylenol orange (1.5 ml/kg body weight) in weeks four and five (21 and 28 days postoperatively).

Preparation technique

One of the twelve animals died five days postoperatively and was shown descriptively as additional information about the repair processes five days postoperatively. Five animals were analyzed histologically after six weeks and six animals after twelve weeks. The specimens were prepared using Donath and Breuner’s cutting and grinding technique [2, 3]. The specimens were prepared along the implant axis in 40–60 μm sections. Following histomorphometric analysis of the fluorochrome labeling, the specimens were stained with toluidine blue and examined histologically.

Histomorphometric analysis

In order to analyze peri-implant bone regeneration, the fluoro-chrome labeling of the peri-implant bone was always measured at the same three sites per implant: at implant threads 1 and 3 counted from the crestal end and at the last thread in the apical region (caudolateral corner) of the implants, with the areas to be measured directly next to the implant surface (Fig. 2). Four photos were taken of each of these three areas (Leica DMRX, Leica; CCD color video camera, Sony; 100x magnification): the first photo was taken under transmitted light microscopy without a specific filter and this was followed by three fluorescence photos to analyze the fluorochromes alizarin, calcein green and xylenol orange (Fig. 2). These four photos were stored digitally and analyzed histomorphometrically using an image analysis system (Image Tool for MS Windows, University of Texas Health Science Center, San Antonio, TX, USA). With this system the number of fluorochrome-labeled pixels (as % of the total area) was determined separately for alizarin, calcein green and xylenol orange at three regions per implant. The mean alizarin staining of the corresponding three implant regions was analyzed as a measure of bone regeneration in the first postoperative week, the mean calcein green as a measure of regenerative processes in weeks two and three and the mean xylenol orange fluorescence as a measure for weeks four and five.

After staining the specimens with toluidine blue, a further digital photo of the entire implant together with the surrounding bone was taken at 16x magnification. The bone-implant contact rate (BIC) was determined as the percentage of the implant surface covered with bone, using an image analysis program (Adobe Photoshop CS and Paint Shop Pro 7 for MS Windows) (Fig. 3).

Statistical methods

All quantitative measurements were described using descriptive statistics (n, mean, standard deviation, median, minimum, maximum and other quartiles). To examine whether the surface modifications of the implants influence bone regeneration, the animals in the 6-week and 12-week groups were analyzed separately: 1. Branemark, 2. Osseotite, 3. Xive and 4. Compress.

The mean of the three fluorochrome measurements per implant (measurement of the labeled bone surface in % at the 1^st thread, 3^rd thread and last thread) was interpreted as a combined parameter for bone regeneration in the peri-implant region. The four associated bone regeneration means (four implants per animal) were shown graphically for each of the four implant types using box plots.

To examine the temporal sequence of possible implant surface effects on bone regeneration, each fluorochrome labeling was analyzed separately and shown side by side for the four implant types: the alizarin measurements represent bone formation during week one, calcein green in weeks two and three and xylenol orange in weeks four and five.

In order to compare overall bone activity during the osseointegration process, the total fluorochrome labeling of each implant was calculated (alizarin + calcein green + xylenol orange) and shown in a box plot for the four implant surfaces.

Box plots of bone regeneration and BIC were plotted separately for the different implant types after a healing period of six weeks (n = 5 animals) and twelve weeks (n = 6 animals) (for production reasons, number of specimens n = 4–6 implants/time-group), although the median was also used for the groups, though it has only limited validity. For the four implant types, the sign test for non-normally distributed linked data was calculated and the p values were shown.

Results

Intravital fluorochrome labeling

In the 6-week (n = 5 animals) and 12-week (n = 6 animals) specimens, the box plots showed only small differences in the fluorochrome-labeled bone surface between the four implant types for alizarin, calcein green and xylenol orange (Fig. 4–6). In the 6-week group, median alizarin and calcein green labeling was higher with the Branemark and Xive implants than with the other two implant types, whereas the Xive and Compress implants demonstrated higher fluorochrome staining for xylenol orange. In most cases, the median fluorochrome labeling was the same or slightly lower in the 12-week specimens than in the 6-week specimens. The means and standard deviation of the fluorescence-labeled bone surfaces are shown in Table 1.