Predisposing factors to severe external root resorption associated to orthodontic treatment

» The authors report no commercial, proprietary or financial interest in the products or companies described in this article. Objective: The aim of this study was to evaluate predisposing factors among patients who developed moderate or severe external root resorption (Malmgren’s grades 3 and 4), on the maxillary incisors, during fixed orthodontic treatment in the permanent dentition. Methods: Ninety-nine patients who underwent orthodontic treatment with fixed edgewise appliances were selected. Patients were divided into two groups: G1 – 50 patients with no root resorption or presenting only apical irregularities (Malmgren’s grades 0 and 1) at the end of the treatment, with mean initial age of 16.79 years and mean treatment time of 3.21 years; G2 – 49 patients presenting moderate or severe root resorption (Malmgren’s grades 3 and 4) at the end of treatment on the maxillary incisors, with mean initial age of 19.92 years and mean treatment time of 3.98 years. Periapical radiographs and lateral cephalograms were evaluated. Factors that could influence the occurrence of severe root resorption were also recorded. Statistical analysis included chi-square tests, Fisher’s exact test and independent t tests. Results: The results demonstrated significant difference between the groups for the variables: Extractions, initial degree of root resorption, root length and crown/ root ratio at the beginning, and cortical thickness of the alveolar bone. Conclusion: It can be concluded that: Presence of root resorption before the beginning of treatment, extractions, reduced root length, decreased crown/root ratio and thin alveolar bone represent risk factors for severe root resorption in maxillary incisors during orthodontic treatment.


introduction
The external root resorption (ERR) is frequently observed by orthodontists and are usually diagnosed in clinical practice when radiographs (panoramic or periapical) are performed. ERR are usually asymptomatic, and when the loss of root structure by resorption become severe the physiology and retention of the afected teeth may be compromised. 14 The anterior teeth are more afected by root resorption, probably because they are single-rooted with tapered roots, conducting the orthodontic force directly to the apex. They are also constantly moved during orthodontic treatment. Moreover, they are more exposed to external factors, such as trauma, making these teeth a good reference for the magnitude of root resorption during orthodontic treatments. 22 The literature about root resorption is extensive but very controversial in relation to the factors that actually inluence the occurrence and severity of resorption during orthodontic treatment. In this context, this study was conducted in order to evaluate which factors are commonly observed in patients who develop moderate or severe degrees of resorption in maxillary incisors during orthodontic treatment.

Material
The sample consisted of 99 patients from the Paulo Picanço Advanced Orthodontics Center, Fortaleza-CE, who underwent orthodontic treatment with fixed appliance (Edgewise technique) in the permanent dentition, and met the following requirements: Good oral and systemic health, no tooth loss in the region from canine to canine in the maxillary arch, absence of vertical bone loss and periodontal disease, and no prosthesis. Patients who developed grade 2 resorption 26 at the end of treatment was excluded from the study. Only patients who had complete treatment records, medical records, initial cephalogram and initial and final periapical radiographs of maxillary incisors in good conditions were included in the sample.
Teeth with endodontic treatment, incomplete development of the root apex, tooth agenesis, supernumerary teeth and patients with incomplete initial records were excluded from the sample. Radiographs with distortion or blur were also eliminated.

Methods
In order to evaluate the variables, initial and final periapical radiographs and initial cephalograms were used of each patient. The radiographs were scanned with a scanner (Microtek ScanMaker i800, Microtek International, Inc., Carson, USA) and coupled to a Pentium computer. The images were transferred to the software Dolphin Imaging Premium 10.5 (Dolphin Imaging & Management Solutions, Chatsworth, USA) through which the images were digitized, the points were marked and cephalometric measurements were performed automatically by the software Dolphin.
The periapical radiographs were analyzed as follows: The degree of initial and final, based on the classification proposed by Malmgren: Grade 0 (no resorption), grade 1 (presence of apical irregularities), grade 2 (presence of resorption by 2 mm), grade 3 (presence of resorption between 2 mm and a third of the original length), grade 4 (the presence of root resorption greater than one third of the original length of the root) (Fig 1). 7,8 The bone crest was classified based on the observation of periapical radiographs as follows: Flat (width greater than 1 mm, representing the crest of a rectangular shape) and sharp (width less than or equal to 1 mm, representing the crest of triangular shape) (Fig 2).
Predisposing factors to severe external root resorption associated to orthodontic treatment original article The evaluation of the root shape was performed based on the classiication proposed by Consolaro 9 as follows: Triangular (T), rhomboid (R), pipette (P) and dilacerated (D) (Fig 3).
The root length was obtained by measuring the distance from the apex to the cementoenamel junction, following the long axis of the incisor (Fig 4). This measurement was performed in both initial and inal periapical radiographs.
Cephalometric variables used are shown in Figures 5, 6 and 7, and described in Table 1.

Method error
For the evaluation of the intraexaminer error, the measurements were performed on 20 patients randomly selected, after a month interval. It was applied the dependent t test to obtain the systematic      root/crown proportion error and Dahlberg 11 formula to estimate the random error. To evaluate the error of the score of root resorption, the Kappa test was used.

Statistical analysis
It was used the following statistical tests: Chisquare test for intergroup comparison of gender, type of malocclusion, treatment with or without extraction, shape of the root and bone crest; independent t test for comparison of the ages, treatment time and cephalometric variables between the two groups. All tests were performed with Statistica software (Statistica for Windows, versão 7.0, Statsoft). Results were considered significant when p < 0.05.

rESuLtS
The largest linear random error found, was for the variable 1-PTV apical (1.36 mm), and angular for the measure 1.NA (1.41°) and the largest systematic error was 0.12 mm of the variable Overjet. The percentage of agreement was substantial (87%) and Kappa coeficient was 0.794.
Increased age and longer treatment was signiicantly related to the occurrence of severe root resorption (Table 2).
Gender, type of malocclusion, morphology of the root and the bone crest are not risk factors for developing severe root resorption ( Table 3). The treatment protocol with extractions increases the risk of severe apical root resorption ( Table 3). The short root length and the decreased proportion of the crown/root ratio at the beginning of treatment increases the chance of developing severe resorption (Table 4).
Patients with thin maxillary alveolar cortical bone are more likely to develop severe resorption than patients with good bone thickness (Table 5). Wits (mm) Distance between the points A and B projected perpendicularly to the functional occlusal plane.
Overjet (mm) Distance between the incisal edges of the maxillary and mandibular central incisors projected perpendicularly to the occlusal plane.
Overbite (mm) Distance between the incisal edges of the maxillary and mandibular central incisors measured perpendicularly to the occlusal plane.
H-11 (mm) Total length of the maxillary central incisor, including crown and root.

UA+UP (mm)
Sum of the thickness of the anterior (buccal) and posterior (palatal) alveolar crest bone.

Variables
Group 1 absent or light resorption (n = 50) Group Table 4 -Intergroup comparison of the root length and the crown/root proportion in T 1 , T 2 and T 2 -T 1 (independent t test).

Variables
Group 1 absent or light resorption (n = 50) Group   The sample of this study was obtained from Paulo Picanço Orthodontics Center. Initially, periapical radiographs were examined to obtain the degree of resorption and then the patients were divided into two groups. At this time, it was excluded patients with absences in the region from canine to canine in the maxilla, presence of vertical bone loss, patients with resorption grade 2, patients with incomplete, damaged or without inal orthodontic records. Ater the selection, 165 patients from all subjects had completed iles available. Ater checking the periapical radiographs, the evaluation of the orthodontic records was made, noting that many patients had incomplete records, absence of inal orthodontic records, lack of signed informed consent, incomplete permanent dentition. At the end, a inal sample of 99 patients attended the inclusion criteria.
Although the control group (group 1, with resorption grade 0 or 1 at the end of treatment) have more patients, this was the most diicult to collect, especially patients with grade 0 who fulilled all the requirements of research, since the literature airms that the resorption in orthodontically treated patients is 100%. 3,4,19,22,36 To collect the ideal sample was the greatest challenge of this study, however, the number of subjects was considered good, because of the strict selection criteria.

Method
In this study only the maxillary incisors were evaluated, because previous studies showed that these teeth have more susceptibility to develop resorption during orthodontic treatment. Incisors are the more constantly moved, for example, during retraction and intrusion. 8,12,22 Moreover, incisors are single-rooted elements and easier for obtaining images without distortion or image overlays. 15,18,23,33 The most common radiographic examination used for the detection of resorption is periapical xray. For Sameshima and Asgarifar 33 this type of radiograph shows finer detail, allowing visualization of anatomical details such as the cementoenamel junction and have less distortion and overlap when compared to panoramic and cephalometric radiographs. 26 In this study the standardized technique for periapical radiographs used was the parallelism, and the fact that all patients in the study underwent radiographic follow-up at the Paulo Picanço Orthodontics Center, ensuring greater standardization of radiographs. The choice for this technique is the fact that it allows greater standardization of the image for pre-and post-treatment, what is not possible by the bisection technique, since the average angle of incidence of x-rays is more difficult to reproduce even hindering over the accuracy of quantitative measurements. 32

Age
The results were signiicant in relation to age, because the group 1 presented younger patients when compared to group 2. It can be airmed that older patients have a higher risk of developing moderate or severe root resorption during orthodontic treatment (Table 2).
Some authors mentioned that age does not inluence root resorption. 1,9,19 However, corroborating the results of this study, Sameshima and Sinclair 34,35 observed that the resorption is most prevalent in adults than in children.
Adults seem more susceptible to resorption because, with aging, the periodontal membrane becomes less vascularized, inelastic, more narrow and the cementum becomes thicker, and also the fact that the apical third of the root is more irmly anchored in adult patients, creating a diiculty in tooth movement and predisposing to resorption. 3,4,6 However, the results of this study must be interpreted with caution, since the age diference between the groups 1 and 2 was only about three years, and both patients in group 1 and group 2 were considered "young adults" ( Table 2).

Treatment time
The results showed that a longer treatment is a risk factor to the occurrence of severe root resorption, as the group 1 showed a signiicantly shorter treatment time compared to group 2 ( Table 2).
These results are opposite to those authors who suggest no relation between the duration of treatment and the degree of resorption. 2,13,37 According to Sameshima and Sinclair, 34,35 the duration of treatment and the amount of horizontal displacement of the apical root of the maxillary incisors had strong correlation with root resorption.
Brin et al 5 evaluated the root resorption in patients with Class II malocclusion treated in only one phase or with treatment divided into two phases, observed that patients undergoing a single phase of treatment had a proportion of moderate to severe resorption slightly larger than the group with two phases of treatment.

Gender
In this study no signiicant relationships were found between severe resorption and the gender (Table 3), agreeing with the results of most authors. 9,19,20,31,34,35,36

Type of malocclusion
Regarding the type of malocclusion and orthodontic techniques, many studies have shown that there is no relationship between root resorption and type of malocclusion. 6,19,30 The results of this study also found no signiicant relationship (Table 3).

Type of treatment (with or without extraction)
The results of this study showed that patients treated with extraction were more likely to develop severe root resorption than patients whose treatment did not include extraction (Table 3).
Many studies showed that patients treated with extractions showed more resorption and with more severe degree, because mechanical retraction of anterior teeth cause greater movement of the root apex and the need for longer treatment. 12,16,29,34,35 Root and bone crest morphology The results of this study showed no relationship between the morphology of the root and the bone crest to the occurrence of severe resorption during orthodontic treatment (Table 3).
Most authors consider, regarding the root morphology, that teeth with atypical root have a higher risk of root resorption. 25,29,30,34,35 It is likely that this result is due to the fact that most of the sample presented rhomboid root and tapered bone crest, which according to most authors, reduce the risk of resorption.

Degree of initial resorption
The results have shown that patients with some degree of root resorption at the beginning of treatment have a greater predisposition and an increased risk of developing severe root resorption during orthodontic treatment (Table 3).
Some authors believe that patients with minimal or no resorption present little risk to severe resorption, patients with moderate resorption have regular risk to severe and extreme resorption, while patients starting orthodontic treatment with severe resorption has a high risk of extreme resorption at the end of treatment. 6,36 Root length and crown/root proportion In this study, the root length decreased significantly more during treatment in patients with severe Predisposing factors to severe external root resorption associated to orthodontic treatment original article resorption compared to the group with mild resorption (Table 4), as expected, because of the criterion for group division. However, the root length of group 2 was already smaller than the group 1 before the beginning of treatment (Table 4). This can be considered an indication that a smaller root length at the beginning of treatment is a risk factor for increased occurrence of root resorption during orthodontic treatment. Likewise, the crown/root proportion at the beginning of treatment was lower in group 2 than in group 1, and continued lower at the end of treatment (Table 4). Furthermore, the crown/ root proportion suffered a greater decrease in group 2 (Table 4), as expected.
The root length and crown/root proportion seem to inluence the tendency to resorption. A large crown will tend to concentrate force on certain focal points, thus short roots tend to sufer more resorption during orthodontic movements.
In cases of patients with severe resorption the professional should be very careful with the root length, teeth with roots length less than or equal to 9 mm have a higher risk of teeth mobility. 27 According to Kalkwarf et al, 24 the reduction of 3 mm in apical tooth structure corresponds to the loss of alveolar bone crest of 1 mm.

Incisor position
Variables related to the position of the maxillary incisor showed no statistically signiicant diference between the groups, indicating no relationship among the protrusion and buccal tipping of the incisors and the occurrence of severe root resorption ( Table 5).
The results showed no statistically signiicant diference between the groups for the protrusion and initial inclination of the incisors, indicating that they are not risk factors to the occurrence of severe root resorption, although group 2 showed a greater buccal inclination of the incisors at the beginning of treatment, but not statistically signiicant (Table 5).

FMA and PFH/AFH
The FMA angle showed no statistically significant difference between the groups, indicating no relationship between the vertical pattern of the patients and the occurrence of severe root resorption ( Table 5). The proportion PFH/AFH showed no statistically significant difference between the groups, indicating no relationship between the facial pattern of the patients and the occurrence of severe root resorption (Table 5).
Handelman 17 analyzed the variable SN-MP that, represents the relationship between the cranial base and the mandibular plane and found positive association between this variable and root resorption.
Harris, Kineret and Tolley 19 also conducted a study to evaluate the relationship between FMA and root resorption and found a relatively high correlation between them.

ANB and WITS
The results of this study demonstrate that the maxillomandibular relationship was not signiicantly different between the groups, indicating that this variable is not a risk factor for the occurrence of severe root resorption (Table 5). It is likely that this result is due to the fact that the present sample does not show large maxillomandibular discrepancies.
In a study by Harris, Kineret and Tolley 19 , these two variables (ANB and Wits) were evaluated and it was observed that both have strong relationship with the occurrence of resorption, as higher maxillomandibular discrepancies tend to require greater retraction of anterior teeth and therefore enhance the risk of resorption.

Overjet and Overbite
The results of this study showed no statistically signiicant diference between the groups for overjet and overbite, indicating that, at the beginning of treatment, these variables are not a risk factor to the occurrence of severe root resorption (Table 5). These indings, however, contradict most of the authors and the justiication for this is the absence of great skeletal discrepancies and also the absence of a signiicant number of cases with open bite and/or deep bite.
There is a consensus in considering the overjet as a risk factor for resorption, because the correction requires the retraction of anterior teeth, and the greater the magnitude of this malocclusion, the greater the amount of movement, increasing the risk and severity of resorption. 2,5,28,31 Freitas et al 12 observed a great degree of resorption for correction of great amount of overjet.