Now Reading
Evaluation of root resorption after open bite treatment with and without extractions

Evaluation of root resorption after open bite treatment with and without extractions



American Journal of Orthodontics and Dentofacial Orthopedics, 2007-08-01, Volume 132, Issue 2, Pages 143.e15-143.e22, Copyright © 2007 American Association of Orthodontists


Introduction In this study, we evaluated the root resorption degree in open bite and normal overbite patients, treated with and without premolar extractions. Methods A sample of 120 patients was selected and divided into 4 groups. Group 1 comprised 32 patients treated with premolar extractions with an initial mean age of 14.01 ± 2.58 years and an initial mean overbite of –3.45 ± 0.23 mm. Group 2 included 28 open bite patients treated without extractions, with an initial mean age of 13.27 ± 2.75 years and an initial mean overbite of –3.10 ± 0.24 mm. Group 3 comprised 30 patients with normal overbite, treated with premolar extractions, having a mean age of 13.28 ± 1.79 years and a mean overbite of 1.09 ± 0.24 mm at the beginning of treatment. Group 4 consisted of 30 patients with normal overbite, treated without extractions, at a mean age of 12.87 ± 1.43 years and a mean overbite of 1.67 ± 0.24 mm at the beginning of treatment. The groups were matched by initial age, treatment time, and malocclusion type. Pretreatment and posttreatment periapical radiographs were used to evaluate the amount of root resorption. The groups were compared by using the Kruskal-Wallis and Dunn nonparametric tests. Correlations between the degree of root resorption and amount of tooth movement, usage time of anterior vertical elastics, and treatment time were investigated with the Spearman correlation coefficient. Results No statistically significant difference was found between the root resorption degrees of open bite vs normal overbite groups, but the extraction groups had statistically significant greater root resorption than the nonextraction groups. Significant correlations were observed in the extraction groups between root resorption degree and amount of overjet correction and retraction of maxillary incisor apex. Conclusions Root resorption was similar between open bite and normal overbite treatment protocols, but extraction treatment showed greater root resorption than nonextraction treatment. There was a statistically significant correlation of overjet correction and retraction of maxillary central incisor apices with the degree of root resorption.

Root resorption is frequently thought to be an iatrogenic consequence of orthodontic treatment. Several treatment factors have been investigated and related to loss of root length–eg, retraction and intrusion–which were considered the main causes of apical resorption. It was demonstrated that extraction patients have more incisor apical root resorption than nonextraction patients because of the great apical displacement during retraction of the anterior teeth. McNab et al found that the amount of root resorption is 3.72 times greater for extraction than nonextraction patients. DeShields also found a positive significant relationship between the amount of maxillary incisor retraction and root resorption, demonstrating that greater tooth displacement is associated with greater resorption.

In open bite patients, some factors increase the risk for root resorption–eg, tooth jiggling due to a high prevalence of tongue dysfunction. However, studies have not investigated the effects of extrusive vertical elastic forces on the anterior teeth during open bite treatment with extraction in the permanent dentition. Although some studies evaluated the root resorption degree in open bite patients after orthodontic treatment, they did not separate the sample into extraction and nonextraction subjects.

Therefore, the purpose of this study was to test the following null hypothesis: root resorption degree is similar between open bite and normal overbite patients, treated with and without premolar extractions. Additionally, correlations of resorption to the amount of time of anterior vertical elastic use, treatment time, overbite and overjet changes, and incisor apex displacements were investigated.

Material and methods

A sample of 120 patients was retrospectively and randomly selected from the files of the orthodontic department at Bauru Dental School, University of São Paulo, in Brazil. Open bite patients treated with standard fixed edgewise appliances were selected to compare with a control group of normal overbite patients treated with the same orthodontic appliance. Patients with incomplete orthodontic records and poor-quality radiographs were not included. Also, patients who had endodontic treatment, malformed roots, history of trauma, root resorption, incomplete root formation, impacted anterior teeth, dental anomalies of number, overjet greater than 7 mm at the pretreatment stage, or previous orthodontic or surgery treatment were excluded.

Regarding the amount of overbite and the extraction or nonextraction protocols, the sample was divided into 4 groups, whose characteristics are described below and summarized in Table I .

Table I
Characteristics of each group
Group 1 (OB-extraction) Group 2 (OB-nonextraction) Group 3 (no OB-extraction) Group 4 (no OB-nonextraction)
Patients (n) 32 28 30 30
Overbite (mm)
Mean ± SD −3.45 ± 0.23 −3.10 ± 0.24 1.09 ± 0.24 1.67 ± 0.24
Minimum −1.1 −1.3 0.1 0.1
Maximum −8.4 −5.7 3.5 3.5
Premolar extractions Yes No Yes No
Time of vertical elastics
Mean ± SD (mo) 5.94 ± 4.48 8.86 ± 4.73 No No
Initial age (y)
Mean ± SD 14.01 ± 2.58 13.27 ± 2.75 13.28 ± 1.79 12.87 ± 1.43
Minimum 9.41 9.33 11.16 10.02
Maximum 19.25 21.66 19.33 15.36
Malocclusion (n)
Class I 13 15 15 18
Class II 19 13 15 12
Sex (n)
Male 16 6 17 16
Female 16 22 13 14
OB , Open bite.

Group 1 consisted of 32 open bite patients (16 male, 16 female) with an initial mean age of 14.01 ± 2.58 years (range, 9.41-19.25 years) and an initial mean overbite of –3.45 ± 0.23 mm (minimum, –1.1 mm; maximum, –8.4 mm), corrected with 4-premolar extraction treatment (except for 3 patients who had only 2 maxillary premolar extractions) and edgewise appliances associated with anterior vertical elastics. Thirteen patients had Class I and 19 had Class II Division 1 malocclusions.

Group 2 consisted of 28 open bite patients (6 male, 22 female) with an initial mean age of 13.27 ± 2.75 years (range, 9.33-21.66 years) and an initial mean overbite of –3.10 ± 0.24 mm (minimum, –1.3 mm; maximum, –5.7 mm), corrected with nonextraction treatment and edgewise appliances associated with anterior vertical elastics. Fifteen patients had Class I and 13 had Class II Division 1 malocclusions.

Group 3 consisted of 30 patients (17 male, 13 female) with an initial mean age of 13.28 ± 1.79 years (range, 11.16-19.33) and an initial mean overbite of 1.09 ± 0.24 mm (minimum, 0.1; maximum, 3.5mm), corrected with 4-premolar extraction treatment (except for 3 patients who had only 2 maxillary premolar extractions) and edgewise appliances without anterior vertical elastics. Fifteen patients had Class I and 15 had Class II Division 1 malocclusions.

Group 4 consisted of 30 patients (16 male, 14 female) with an initial mean age of 12.87 ± 1.43 years (range, 10.02-15.36 years) and an initial mean overbite of 1.67 ± 0.24 mm (minimum, 0.1; maximum, 3.5 mm), corrected with nonextraction treatment and edgewise appliances without anterior vertical elastics. Eighteen patients had Class I and 12 had Class II Division 1 malocclusions.

Orthodontic mechanics included fixed edgewise appliances, with 0.022 × 0.028-in slot conventional brackets and extraoral headgear to correct the Class II relationship or to reinforce anchorage when necessary. The usual wire sequence was an initial 0.015-in Twist Flex (3M Unitek, Monrovia, Calif) or a 0.016-in nickel-titanium alloy, followed by 0.016, 0.018, 0.020, and 0.021 × 0.025-in stainless steel archwires (3M Unitek). Open bite in the extraction group was corrected by means of the “drawbridge principle,” due to the retraction of incisors resulting in uprighting and relative extrusion, and anterior vertical elastics. In the nonextraction group, the open bite was corrected with anterior vertical elastics, both with flat archwires. The treatments were performed by 104 graduate students over several years as part of their clinical program.

Root resorption was evaluated on the final periapical radiographs obtained with the long-cone paralleling technique. The periapical radiographs were scanned with the Sprint Scan 35 Plus scanner (version 2.7.2, Polaroid, Waltham, Mass) with resolution of 675 dpi in a scale of 1:1. The images were analyzed with Photoshop software (version 7.0, Adobe Systems, San Jose, Calif) at 300% enlargement, without image quality loss.

A modification of the method of Malmgren et al was used to evaluate the severity of apical root resorption, ranking it into 5 grades ( Fig 1 ). Eight scores, corresponding to each incisor, for each patient were obtained, totaling 262 scores in group 1, 224 scores in group 2, and 240 scores in groups 3 and 4 each.

Score system modified from Malmgren classification 24 (reprinted from: Beck B, Harris EF. Apical root resorption in orthodontically treated subjects. Analysis of edgewise and light wire mechanics. Am J Orthod Dentofacial Orthop 1994;105:350-61; with permission from the American Association of Orthodontists): grade 0, no root resorption; grade 1, mild resorption, root with normal length and irregular contour only; grade 2, moderate resorption with small area of root loss and apex exhibiting almost straight contour; grade 3, accentuated resorption with loss of almost one third of root length; grade 4, extreme resorption with loss of more than one third of root length.
Fig 1
Score system modified from Malmgren classification (reprinted from: Beck B, Harris EF. Apical root resorption in orthodontically treated subjects. Analysis of edgewise and light wire mechanics. Am J Orthod Dentofacial Orthop 1994;105:350-61; with permission from the American Association of Orthodontists): grade 0, no root resorption; grade 1, mild resorption, root with normal length and irregular contour only; grade 2, moderate resorption with small area of root loss and apex exhibiting almost straight contour; grade 3, accentuated resorption with loss of almost one third of root length; grade 4, extreme resorption with loss of more than one third of root length.

The cephalometric characteristics studied consisted of the changes in overbite and overjet, in the inclination of maxillary and mandibular incisors (Mx1.PP and IMPA), and in the vertical and horizontal displacements of maxillary and mandibular central incisor apices (Mx1apex vert, Mx1apex hor, Md1apex vert, and Md1apex hor) ( Table II , Fig 2 ). The pretreatment (T1) and posttreatment (T2) lateral cephalometric radiographs were traced by the same examiner (R.T.S.B.), and the landmarks were digitized on a digitizer (model 30TL.F, Numonics, Montgomeryville, Pa). Landmark coordinates were stored in a computer with Dentofacial Planner software (version 7.02, Dentofacial Planner, Toronto, Ontario, Canada). The cephalometric radiographs were obtained with 3 x-ray machines with magnification factors of 6.0%, 7.9%, and 9.8% of the image. The cephalometric software corrected these magnifications to standardize the cephalograms and allow their comparison. The length of each incisor on the pretreatment headfilm was assigned to the posttreatment headfilm to eliminate interference of the amount of apical resorption in the calculation of apical displacements. Treatment changes were calculated as T1 minus T2.

Table II
Definition of some planes and abbreviations of dental and skeletal variables
Lines and planes
Palatal plane: constructed from ANS to PNS
Functional occlusal plane: plane drawn through points of occlusal contact between first permanent molars and first and second premolars
Mandibular plane: constructed from Go to Me
ANSperp: line perpendicular to palatal plane, passing through ANS point
Meperp: line perpendicular to mandibular plane, passing through Me point
Dental cephalometric variables
Overbite: distance between incisal edges of maxillary and mandibular central incisors, perpendicular to occlusal plane ( Fig 2 )
Overjet: distance between incisal edges of maxillary and mandibular central incisors, parallel to occlusal plane ( Fig 2 )
Mx1.PP: maxillary incisor long axis to palatal plane angle ( Fig 2 )
IMPA: incisor mandibular plane angle ( Fig 2 )
Mx1apex vert: distance between maxillary central incisor and palatal plane ( Fig 2 )
Mx1apex hor: distance between maxillary central incisor and ANSperp ( Fig 2 )
Md1apex vert: distance between mandibular central incisor and mandibular plane ( Fig 2 )
Md1apex hor: distance between mandibular central incisor and Meperp ( Fig 2 )

Cephalometric variables: 1 , overbite; 2 , overjet; 3 , Mx1.PP; 4 , IMPA; 5 , Mx1apex vert; 6 , Mx1apex hor; 7 , Md1apex vert; 8 , Md1apex hor.
Fig 2
Cephalometric variables: 1 , overbite; 2 , overjet; 3 , Mx1.PP; 4 , IMPA; 5 , Mx1apex vert; 6 , Mx1apex hor; 7 , Md1apex vert; 8 , Md1apex hor.

Twenty-four randomly selected patients had their root resorption amounts reevaluated and their radiographs retraced, redigitized, and remeasured by the same examiner after a 30-day interval. In the root resorption evaluation, intraexaminer agreement was calculated with the kappa statistic. For the cephalometric evaluation, the casual error was estimated with Dahlberg’s formula (S 2 = Σd 2 /2n), where S 2 is the error variance and d is the difference between the 2 determinations of the same variable. The systematic error was calculated with the dependent t test at P < .05.

Statistical analysis

Compatibility between the groups was investigated with analysis of variance (ANOVA) for initial age and treatment time, and with the chi-square test for malocclusion type. The amount of root resorption was compared between the 4 groups by using the resorption degree median of the 8 incisors, by applying the Kruskal-Wallis and Dunn nonparametric tests.

Considering the cephalometric variables, the changes during treatment were compared between the groups by using ANOVA and Tukey tests. To apply ANOVA, a normal distribution of the samples was necessary. This was verified with the Kolmogorov-Smirnov test. Results of this test demonstrated that all variables had normal distribution. Correlation of the amount of overbite and overjet correction, amount of usage time of anterior vertical elastics, treatment time, and vertical and horizontal displacements of the maxillary and mandibular incisor apices with the amount of root resorption were investigated with the Spearman correlation coefficient. Correlation of resorption of the 8 incisors to treatment time and duration of elastics use was investigated, whereas only the central incisors were used to study the correlation to tooth movements. The groups were divided into extraction and nonextraction groups during the correlation because they had a statistically significant difference of root resorption degree.

All statistical analyses were performed with software (Statistica version 6.0, Statsoft, Inc, Tulsa, Okla). The results were considered significant at P < .05.

Results

Kappa statistics showed almost perfect agreement between the first and second root resorption evaluation (κ = 0.94). The systematic errors were statistically significant for 4 variables (overbite, P < .001; IMPA, P = .037; Md1apex vert, P = .020; and Md1apex hor, P = .006). The casual errors were within acceptable limits—0.337 mm and 1.753° for overjet and Mx1.PP, respectively.

The groups were matched for initial age, treatment time, and malocclusion type ( P = .228, .934, and .487, respectively). There was no statistically significant difference in root resorption degree when the open bite groups were compared with the normal overbite groups, considering the matched extraction or nonextraction protocols. However, the extraction groups had statistically significant greater root resorption than the nonextraction groups in the matched overbite groups ( Table III ).

Table III
Comparison of root resorption degree between groups (Kruskal-Wallis followed by Dunn tests)
Variable Group 1 (OB-extraction) Group 2 (OB-nonextraction) Group 3 (no OB-extraction) Group 4 (no OB-nonextraction) P
Mean SD Mean SD Mean SD Mean SD
Root resorption 1.57 A 0.51 1.17 B,C 0.39 1.55 A,B 0.59 1.08 C 0.39 .0009
Dunn test’s results: different letters represent statistically significant differences at P < .05.
OB , Open bite.

Statistically significant.

The cephalometric changes were statistically different among the 4 groups ( Table IV ). With the Spearman correlation coefficient, there were significant correlations between root resorption degree and amount of overjet correction and maxillary incisor apex horizontal displacement only in the extraction groups ( Table V ).

Table IV
Results of 1-way ANOVA and Tukey tests used to compare cephalometric changes between groups
Variable Group 1 (OB-extraction) Group 2 (OB-nonextraction) Group 3 (no OB-extraction) Group 4 (no OB-nonextraction) P
Mean SD Mean SD Mean SD Mean SD
Overbite −3.46 A 1.86 −3.87 A 1.50 −0.43 B 1.63 0.19 B 1.13 <.001
Overjet 3.57 A 2.78 1.63 B 1.84 2.44 A,B 2.78 1.23 B 1.80 <.001
Mx1.PP 8.65 A 7.56 3.73 A,B 7.21 1.49 B 8.98 −0.74 B 6.48 <.001
IMPA 3.79 A 6.15 2.31 A 5.56 3.24 A 4.97 −1.84 B 5.11 <.001
Mx1apex vert −0.82 A,C 1.66 −1.98 B 1.47 0.13 A 1.29 −1.01 B,C 1.28 <.001
Mx1apex hor −0.59 A 2.34 0.47 A 1.86 −2.41 B 2.09 −0.21 A 1.52 <.001
Md1apex vert −1.61 A,B 1.67 −2.65 A 1.79 −0.74 B 1.97 −1.83 A,B 1.41 <.001
Md1apex hor −1.07 A,B 1.61 −0.33 A 1.09 −1.96 B 2.10 −1.00 A,B 1.25 .001
Tukey test’s results: different letters represent statistically significant differences at P < .05.
OB , Open bite.

Statistically significant.

Table V
Correlation between root resorption and cephalometric variables, treatment time, and usage time of anterior vertical elastics (Spearman correlation coefficient)
Variable n R P
Extraction groups (1 and 3)
Resorption all incisors × treatment time 62 0.006 .960
Resorption MxMd1 × overbite 62 −0.166 .195
Resorption MxMd1 × overjet 62 0.273 .031
Resorption Mx1 × Mx1.PP 62 0.063 .625
Resorption Mx1 × Mx1apex vert 62 −0.247 .052
Resorption Mx1 × Mx1apex hor 62 −0.291 .021
Resorption Md1 × IMPA 62 0.097 .450
Resorption Md1 × Md1apex vert 62 −0.035 .783
Resorption Md1 × Md1apex hor 62 −0.029 .818
Nonextraction groups (2 and 4)
Resorption all incisors × treatment time 58 0.001 .990
Resorption MxMd1 × overbite 58 −0.138 .299
Resorption MxMd1 × overjet 58 0.097 .465
Resorption Mx1 × Mx1.PP 58 0.130 .327
Resorption Mx1 × Mx1apex vert 58 0.071 .596
Resorption Mx1 × Mx1apex hor 58 0.026 .845
Resorption Md1 × IMPA 58 0.124 .351
Resorption Md1 × Md1apex vert 58 0.213 .108
Resorption Md1 × Md1apex hor 58 −0.025 .849
Open bite groups (1 and 2)
Resorption all incisors × vertical elastics time 60 −0.202 .120

Statistically significant.

Discussion

Factors related to orthodontic treatment are primarily responsible for the high prevalence of root resorption. However, studies differ significantly regarding design, methodology, control group, and treatment characteristics. An inadequate or unstandardized radiographic technique and small sample size are common problems that cause questionable results. Additionally, several studies did not distinguish the variables related to patient and treatment; this can influence resorption degree. In this retrospective study, we sought to investigate root resorption in a homogeneous sample, treated with anterior vertical elastics to correct the open bite, compared with a normal overbite group.

In samples consisting of patients treated by various professionals, several variables should be considered. It is difficult to obtain large enough groups with adequate records treated by 1 operator to evaluate root resorption. It has been thought that treatment by various professionals prevents biased clinical procedures that could be systematically incorporated into treatment. Furthermore, other studies in the literature were conducted with samples from various sources, and it was stated that this factor should not interfere with the results.

Subjective methods, such as that of Malmgren et al, are predominantly used in root resorption studies, and therefore they seem to be reliable. Their primary advantage is that they do not depend on standardization of the initial radiographs, requiring only similar initial root status of the groups. This was assured by selecting only patients without root resorption at T1. The intraexaminer error demonstrated excellent concordance, attesting to the precision of the evaluation (κ = 0.94).

Many authors mentioned that extraction patients had more resorption than nonextraction patients because of greater tooth displacement during retraction. However, others did not find this relationship. Systematic studies on root resorption after treatment in open bite subjects did not separate extraction and nonextraction patients or state whether patients were treated with or without extractions. Because of this controversy, it was decided to subdivide the open bite and the control groups into extraction and nonextraction groups and then compare the matched groups regarding treatment protocols.

To increase the open bite extraction sample size, 3 patients with only 2 maxillary premolar extractions were included in group 1. Three similar patients were also included in the nonopen bite extraction group to match them (group 3). Both Class I and Class II patients were also accepted to increase the open bite sample. However, because the groups were similar regarding Class II types ( P = .487), the orthodontic mechanical difference would equally affect the groups and should not influence the results.

The open bite group treated with extractions did not show a statistically significant difference regarding root resorption degree when compared with the corresponding normal overbite extraction group, as shown in Table III . The same result was found when the open bite group treated without extractions was compared with the nonextraction normal overbite group. This result differs from other studies that found greater resorption in open bite patients. However, Harris and Butler did not state clearly whether the patients had been treated with or without extractions. Kuperstein and Frantz compared open bite groups with control groups, both treated with and without extractions, but they evaluated pooled extraction and nonextraction patients. Moreover, Kuperstein had an uneven distribution of this variable between the groups: the open bite group had 22 patients treated with extractions and 10 treated without extractions, whereas the control group had 14 patients treated with extractions and 18 treated without extractions. Thus, more patients treated with extractions in the open bite group probably contributed to greater root resorption in this group and not greater extrusion of the incisors during open bite treatment as stated. Another problem in these 3 studies was the radiographic method used to measure resorption. They did not use periapical radiographs, considered the best method to evaluate root resorption because of less image distortion than panoramic or lateral cephalometric radiographs.

The mean root resorption degrees were 1.57 in the open bite group and 1.55 in the normal overbite group. Because both groups were treated with extractions, it was speculated that the main contributing factor for resorption was not the vertical anterior elastics but the amount of incisor retraction.

When the extraction and nonextraction groups were compared, a statistically significant difference was observed. Between the open bite groups, the extraction group had more root resorption than the nonextraction group. The same result was observed between the normal overbite groups; the extraction group showed more resorption than the nonextraction group. Then it could be stated that extraction treatment causes more root resorption; this agrees with some authors, but contrasts with others. The inclusion of a few maxillary premolar extraction patients in each extraction group did not contribute to the decreased resorption amounts in these groups compared with the nonextraction groups.

With these results, it is obvious that a sample of open bite patients cannot be evaluated disregarding whether they were treated with or without extractions, since extraction is a significant factor for root resorption, consequent to the greater movement of the teeth. Probably the reason that no causative factor was found in root-resorption studies in open bite patients was that the authors pooled the extraction and nonextraction patients for comparison. Tooth retraction was the main resorption factor and not the open bite correction procedure.

Treatment time had no significant correlation with root resorption degree, agreeing with other studies but contrasting with some ( Table V ). However, treatment time should not be considered the main factor for root resorption. Perhaps the amount of tooth movement is the most important factor, and it is independent of treatment time because in some cases the appliances can be present with reduced action on the teeth, and, in other cases, patients can frequently miss appointments, or treatment can be delayed because of professional preference in prolonging the interval between activations.

Overbite correction was not significantly correlated with root resorption ( Table V ), probably because of the similar resorption degrees between the open bite group and the normal overbite group ( Table III ), which had different overbite changes during treatment ( Table IV ). No significant correlation was found with vertical apex displacement and root resorption, probably because extrusion is not an aggressive movement, since it does not cause vascular compression. Moreover, the forces applied to close the bite are less intense than those used for rapid forced eruption, with which root resorption is not associated. Regarding vertical movements, it seems that intrusion has more correlation with root resorption than extrusion.

There was a statistically significant correlation of root resorption and overjet change in the extraction groups ( Table V ). Many studies also attributed more resorption to greater overjet correction, suggesting that incisor retraction could be considered a predictive factor of more root resorption, and this was also confirmed in our study. The horizontal movement of the maxillary central incisor apices (Mx1apex hor) had statistically significant correlation to root resorption in the extraction groups. This result confirms that the greater the retraction of the incisor apices, the greater the resorption ( Table V ). This correlation was not significant in the nonextraction groups, probably due to less displacement that generally occurs when extractions are not performed during treatment ( Table IV ).

Studies have emphasized the risks that intermaxillary elastics could have on root resorption during Class II treatment, but they have not evaluated anterior vertical elastic risks during open bite treatment. The time of usage of anterior vertical elastics was investigated, and no significant correlation was found between this variable and root resorption, as shown in Table V . Because there was no correlation between extrusion movement and root loss (as justified above), it is understandable why a longer time of vertical elastics use was not significantly correlated to more root resorption during open bite treatment.

You're Reading a Preview

Become a DentistryKey membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here