INTRODUCTION

Orthodontically induced inflammatory root resorption (OIIRR) is a common consequence of orthodontic tooth movement.12 Because it is usually asymptomatic, imaging techniques are the only way to diagnose and measure its severity.3–12 The periapical radiographic technique is most commonly used for this matter.3–7 This radiographic technique hides several factors that might affect the image seen on the developed film, ie, magnification error, angular changes between the tooth and the film, difficulties in identifying consistently the same points on the film (the CEJ, for example), different laboratories, and different technicians.13–15 Studies that calculate the amount of the OIIRR usually ignore these inherent errors. Therefore, unusual reports such as root elongation after orthodontic treatment in nongrowing patients are not surprising.316

To overcome the problematic usage of reference points on the tooth itself, an external jig with a radiopaque object as a referent length in consecutive exposures was recommended.7 A special individualized jig with a metal wire parallel to the long axis of the tooth (as verified by a cephalometric film) was built for each patient before orthodontic treatment. The exact length of the wire was known. The tooth was radiographed before and after treatment using the parallel periapical technique. The metal wire served as the reference length for all calculations.7

The objectives of this study were to

• Examine the accuracy of using an external jig in different locations and different lengths for measuring root length differences due to angular changes between the tooth and the film.

• Examine the possibility of directly attaching the external jig to the crown surface to overcome the need for a cephalometric radiograph in order to verify parallelism between the jig and the long axis of the tooth.

MATERIALS AND METHODS

A human maxillary central incisor, extracted because of periodontal reasons, was used in this study. The tooth was placed in a special jig built to imitate the actual clinical conditions of the paralleling periapical radiographic technique. The film was placed 25 mm from the edge of the tooth crown, and the cone was positioned 100 mm from the center of the tooth. A special holding device and the jig kept the film always parallel to the cone, meaning that the central X-ray beam was always perpendicular to the film. The tooth was exposed at angles of 10°, 20°, 30°, and 40° to the film. The films were all developed under the same conditions (Dentax 810 Basic) and later scanned using an Umax Astra 2400S scanner. The image of the tooth was enlarged seven times and was analyzed using Adobe Photoshop 5 software (Adobe, San Jose, Calif).

An 11.8-mm wire was attached to the tooth in the following ways:

• Group a. In contact with the palatal crown surface.

• Group b. In contact with the buccal crown surface.

• Group c. At a five-mm distance from the buccal surface parallel to the tooth long axis.

• Group d. At a nine-mm distance from the buccal surface parallel to the tooth long axis.

• Group e. At a 13-mm distance from the buccal surface parallel to the tooth long axis.

• Group f. Another 16-mm wire was attached at a nine-mm distance from the buccal surface parallel to the tooth long axis.

Because this was an in vitro study, there was no need to verify the parallelism of the attached wire to the long axis of the tooth with a dedicated radiograph (Figure 1).

View Figure

• Download as Powerpoint
• Download Figure

Once a week for four consecutive weeks, using a dial caliper with ground tips (Dentarum, Pforzheim, Germany) and the tools of Photoshop software, each of two senior residents measured the tooth length and the wire length on the model and directly on the computer monitor.

The two lengths served as a database for

• Calculating the effect of the angular changes of the tooth to the film, (all other parameters remaining constant), on the ratios of wire/tooth lengths at different angles.

• Calculating the tooth length at different angles using the rule of three mathematical method.

RESULTS

Table 1 presents the wires and tooth length and their respective ratios. The ratio between the 11.8 ± 0.1 mm wire and the tooth length is 0.53 ± 0.02, and the ratio of a 16.0 ± 0.1 mm wire to the tooth length is 0.75 ± 0.02. All ratios that are different from both ratios in Tables 2 through 7 represent magnification errors that lead to inaccurate calculations of the tooth lengths and their changes.

TABLE 1.  The Model Real Tooth and Real Wire Lengths and Their Respective Ratios

View Fullscreen

TABLE 2.  The Measurements on the Monitor of the Tooth and Wire Lengths, the Ratios, and the Calculated Tooth Length for 11.8 ± 0.10 mm Wire Placed on the Palatal Side of the Crown (Group a)

View Fullscreen

The results in Tables 2 and 3 using an 11.8 ± 0.1 mm wire intimately attached to the palatal or facial tooth surface demonstrate that this method is inaccurate in measuring tooth length in consecutive films. For example, when the wire is attached to the palatal side, the ratios increased from 0.56 ± 0.01 at 10° to 0.65 ± 0.01 at 40°. Similar results, but in the opposite direction, occur when the wire is attached to the buccal surface of the tooth.

TABLE 3.  The Measurements on the Monitor of the Tooth and Wire Lengths, the Ratios, and the Calculated Tooth Length for 11.8 ± 0.10 mm Wire Placed on the Buccal Side of the Crown (Group b)

View Fullscreen

The ratios and the calculated tooth length in groups c to f (Tables 4 through 7) demonstrate that the best results for measuring tooth length changes in consecutive films occur with this method when the wire is placed nine mm from the tooth surface. The average changes in both the 11.8 ± 0.1 mm wire and the 16 ± 0.1 mm wire is less than one mm (Table 8) with statistical significance (P < .05). Table 8 presents the differences between the calculated tooth length and the real tooth length for all groups encountered in the study. Again, it is obvious that when the wire is placed nine mm from the tooth, the results are more accurate, meaning that the differences are closer to zero. Table 9 presents the effects of angular changes on the calculations and finds all changes with statistically significant differences other than the expected value of zero. There is no statistically significance difference between the results of groups d and f.

TABLE 4.  The Measurements on the Monitor of the Tooth and Wire Lengths, the Ratios, and the Calculated Tooth Length for 11.8 ± 0.10 mm Wire Placed Five mm from the Tooth, Parallel to its Long Axis (Group c)

View Fullscreen

TABLE 8.  The Differences Between the Calculated and the Real Tooth Lengths at Four Different Angles of the Six Groups (Mean ± SD) in Millimeters. The Average is Related to the Absolute Values of the Difference^a

View Fullscreen

TABLE 9.  The Angular Changes Effect on the Calculated Tooth Length in the Different Groups (Mean ± SD) in Millimeters. The Average is Related to the Absolute Value of the Difference^a

View Fullscreen

DISCUSSION

The amount of root shortening during orthodontic treatment is the issue of many publications.3–12 The accuracy of the method presented lately by Costopulus and Nanda7 tries to overcome the two problems encountered in analyzing consecutive periapical radiographs, ie, the consistent identification of the CEJ and the magnification factor. The study describes a special jig that was individually built for each patient. Unfortunately, the wire length and its distance from the tooth were not reported in their article; therefore, it was impossible to exactly repeat their experiment.

The basic assumption in the article of Costopulus and Nanda7 is that a radiopaque wire attached to the tooth and parallel to its long axis can serve as an external tool to make length measurements easy and accurate. Because both bodies move during treatment with harmony, their magnification factor will be identical. In other words, it was hypothesized that the ratio between the wire length (which is easily identified on the radiograph) and the tooth length will always remain constant. Therefore, accurate tooth or root length changes in consecutive films can be calculated. Furthermore, the method of Costopulus and Nanda7 demands that the wire in the special jig that was individually built for each patient will be parallel to the long axis of the tooth. This can be verified only by using a cephalometric radiograph, meaning that another exposure to unnecessary radiation is needed.

Tables 2 and 3 demonstrate that using wires that are intimately attached to either the palatal surface or the buccal surface of the crown of a tooth cannot serve as an accurate tool to measure tooth length changes due to angular changes between the tooth and the film. For example, in Table 3, we observe that the calculated tooth length for the wire attached to the buccal side of the crown becomes longer by 1.83 ± 0.4 mm because of angular changes from 10° to 20° without any change in its actual length. The calculations would have demonstrated shortening of the tooth if the angular changes between the tooth and the film were reversed (from 20° to 10°). Furthermore, the results reveal that the larger the angular change, the larger the inaccuracy in calculating the tooth length from the film images. Hence, it can be concluded that intimate wire attachment does not solve the issue of accurately measuring changes of tooth length and hence the amount of OIIRR.

Our results demonstrate that there are statistically significant differences in calculating tooth lengths when different wire lengths or different wire locations parallel to the long axis of the tooth are used. It is clear that only when the wire was attached nine mm buccal to the crown, (groups d and f) the measurements became more accurate when compared with the other groups (a, b, c, and e). No statistical difference was found between group d and group f.

Several explanations can elucidate these results. (1) It is expected that in the parallel periapical radiographic technique, the X-rays are parallel and perpendicular to the film and the objects. However, there is always a small angle between the rays, which causes some magnification. Figure 2a,b depicts the different magnification changes that affect the tooth and the external wire (see below). (2) The changes in the shadows of an elliptical body (the tooth) due to angular changes between this body and the film are different from the changes in the shadows of a long narrow body (the wire) due to similar angular changes. In the extreme situation one can imagine that there will be no changes in the shadows of a round body when it rotates relative to the film. (3) The radiopacity differences between the tooth and the metal wire make the consistent identification of the apex and crown edge more difficult than the wire edges. This by itself can lead to some calculation errors.

View Figure

• Download as Powerpoint
• Download Figure

In Figure 2a, the tooth and the wire are parallel to the film, where a and b are the measured tooth and wire shadows on the film, respectively. The ratio of their images, b/ a, is close to one (about 0.8). In Figure 2b, a and b represent the same parameters as in Figure 1a and the only change is that the tooth and the wire rotated with harmony. The ratio of their images, b/a, is much less than one (about 0.2). One can conclude that the tooth became shorter during this angular change, although the tooth did not go through any morphological change. The reason that the best results were achieved when the wire was attached nine mm from the tooth surface is probably because of the fact that during angular changes done in this experiment, the location of the wire maintains the best ratio relative to the tooth changes in the overall X-ray dispersion.

CONCLUSIONS

• Neither palatal nor buccal metal wires attached directly to the crown of the tooth can serve as a tool to accurately measure tooth length changes in consecutive radiographs.

• A metal wire, attached nine mm from the tooth crown surface and parallel to its long axis serves as the best tool for measuring the amount of tooth length changes compared with five- and 13 mm distances. This was determined with statistical significance.

• No statistical significance was found between 11.8 ± 0.1 mm and 16 ± 0.1 mm wire lengths when both were nine mm from the buccal surface of the tooth and parallel to its long axis.

• Unfortunately, when accuracy is important, there is no way to skip the extracephalometric radiograph to verify the parallelism between the wire and the long axis of the tooth using this jig method.