Ten normal adult complete cervical spine formalin-embalmed specimens (7 male, 3 female, age: 36-54 years, average age: 43 years) without neck deformities, lesions or a surgical history were used for this study. The paravertebral muscles were routinely removed, and the integrity of the cervical nerve roots and the ligaments outside of the cervical intervertebral foramina was preserved (Fig. 1). A unilateral opening was made in the window in the posterior cervical spine, as follows: the lamina, the yellow ligament and the medial facet of the corresponding facet joint were removed to form a square bone window approximately 1.5 cm in length, exposing the cervical nerve roots via a posterior approach. To preserve the integrity of the posterior cervical column as much as possible, the C5 and C7 nerve roots were exposed on one side of the specimen, and the C6 and C8 nerve roots were exposed on the other side. Anterior cervical plate and screw fixation was performed to limit vertebral motion. A denaturing powder was embedded in the lower end of the cervical spine specimen, and the specimen was fixed on an apparatus allowing three-dimensional motion of the spine.
The distal end of the nerve root was tied with silk thread, and a weight was connected to an attached pulley. The angle at which the nerve is pulled is based on the angle at which it emerges from the intervertebral foramen. Weights were used to gradually apply a load to the nerve root (50 g/weight, 60 weights in total). A positioning needle (24 mm in length, 0.55 mm in diameter, and 2 mm in diameter at the red pearl head) was fixed to the vertebral lamina, and this point was denoted as point e. Two positioning needles were fixed to the nerve roots in the bone window, and these points were recorded as points a and b, respectively. Two other positioning needles were fixed to the nerve stem outside the intervertebral foramen, and these points were denoted as c and d, respectively. A straight line was made with the red pearl heads of the five positioning needles parallel to the axis of the nerve root. At this time, it should be noted that the intraspinal positioning needle was not inserted deeply, since the tip of the needle could enter other tissues after passing through the nerve root, which could seriously affect the experimental results. A scale (accurate to 1 mm) was placed next to the finished model for calibration during the subsequent measurements. The finished specimen model is shown in Figure 2.
A Canon EOS 50D digital camera fixed on a tripod was used, and the camera position was adjusted such that the lens plane was parallel to the axis of the nerve root (f/5.6; 1/60 s; ISO: 200, focal length: 110 mm). After each loading, the camera recorded the position of each positioning needle head. Photoshop was used to measure the displacement of the positioning needles in the image.
The measurement scale was defined as follows: a mark line was drawn on the upper horizontal scale and placed on the scale line, with a length of 1 cm on the steel ruler. We ensured that the mark line intersected the edge vertex of the scale line (intersection point P1). Furthermore, another line was drawn and placed on the scale line, reaching 2 cm; we ensured that this line also intersected the vertex of the scale line (P2). Two lines were drawn from the left vertical ruler to intersect P1 and P2. After selecting the ruler tool on the left menu bar, P1 and P2 were connected, and the pixel count M between P1 and P2 was recorded. The number of millimeters per pixel was calculated as W=10/M (mm) (Fig. 3a).
Once the center point was located, line 1 was drawn tangential to the upper edge of the positioning needle, and line 2 was drawn tangential to the left edge of the needle on the left vertical ruler. The intersection of lines 1 and 2 was defined as the origin O. Then, line 3 was drawn from the top horizontal ruler tangential to the lower edge of the needle, and the value Y was recorded in the information window. Then, line 4 was moved from the top of the horizontal ruler and dragged to a Y value of Y/2. Similarly, line 5 was drawn from the left vertical ruler tangential to the right edge of the needle, and the value of X was recorded in the information window. Then, line 6 was drawn from the left vertical ruler and dragged to an X value of X/2. The intersection of lines 4 and 6 was defined as the center "Q" of the circular needle (Fig. 3b). After defining these measurements, the distances between a and b, c and d, e and b were measured and are labeled in the figure as Lab, Lcd, and Leb, respectively (Fig. 3c, 3d).
SPSS 20.0 statistical software was used for statistical processing of the measured data.