Correlation between radiographic signs of third molar proximity with inferior alveolar nerve and postoperative occurrence of neurosensory disorders . A prospective , double-blind study 1

PURPOSE: To evaluate the interference of radiographic factors in the appearance of sensory deficit related to inferior alveolar nerve (IAN) after third molars (3Ms) removal. METHODS: A prospective, double-blind, observational, unicentric study was performed with 126 patients submitted to a surgical procedure of lower 3Ms removal in the period from March to October/2011. Collected data included gender, age, eruption stage of 3Ms, position/angle of 3Ms (Pell-Gregory and Winter classifications, respectively), presence/absence of radiographic signs of 3Ms proximity with the inferior alveolar canal and surgical technique. Occurrence evaluation of the IAN injury was performed on the seventh postoperative day through pin-prick, two-point discrimination and brush directional stroke tests. RESULTS: Predominant radiographic signs were: narrowing of the inferior alveolar canal (68.25%), darkening of root (46.82%) and diversion of the canal (31%). None of the patients presented sensory loss. Sixty-one (48.41%) of the cases had at least one or two radiographic signs of proximity with NAI. Forty-seven (37.3%) had 3 or more signs, and 18 (14.29%) did not have any radiographic signs of proximity to mandibular canal. CONCLUSION: There was not a positive correlation between presence of radiographic signs of 3Ms with IAN proximity and postoperative neurosensory disorders occurrence.


Introduction
In recent years, several aspects related to surgical removal of third molars (3Ms) have been discussed, especially the possibility of postoperative complications such as nerve injuries 1,2 .Clinically, neurosensory disorders related to inferior alveolar nerve (IAN) can be manifested as pain, anesthesia, paresthesia, or a combination of these conditions.Among the commonly performed dental procedures, IAN injury is not an uncommon complication 3 , being potentially related to anesthesia for nerve block, endodontic treatment and surgery to remove 3Ms 4 .
The preoperative radiographic evaluation has been considered as having a potential capacity to predict possible IAN injuries during a surgical procedure.Panoramic radiography is a commonly complementary exam used in the treatment plan for impacted teeth removal, being useful in the evaluation of surgical difficulty degree, third molars morphology and position, operative risks and proximity to adjacent vital structures, such as IAN 4 .
Certain radiographic signs often associated with damage to IAN can be observed in panoramic radiographs 5 .
Since prior knowledge about the 3Ms positioning is of great importance for surgical planning and because of the controversy about the predictability of IAN damage from the third molar position, the present study aimed to conduct a clinical and radiographic prospective study about the risk of IAN neurosensorial deficit after surgeries for lower 3Ms removal.To date, few studies have described an incidence of IAN injury less than 1% [6][7][8][9][10][11][12] .The following formula was used for sample calculation 13 :

Methods
Where: n: Calculated sample size dmax: maximun sampling error (set at 7%) Zα/2: Value extracted from the table of the standard normal distribution with 5% of significance level N: Population size (400 patients) P: Value of the sampling proportion (set to 0.5) Q: Complementary value of the sampling proportion (1 -P) Thus, after calculation we obtained a sample (n) of 132 patients, with a confidence interval of 95% and a sampling error of 7%.One right or left third molar was removed from each patient.
The study included patients having panoramic radiography who were candidates for removal of one third molar, with age up to 60 years old, without systemic problems, and who accepted to participate in this study after reading the free and clarified consent term.Patients with associated pain, pericoronitis, edema, hyperthermia, limited mouth opening, pre-existing systemic disease and any patologic process related to 3Ms, as well as pregnant or nursing women and surgical procedures with more than two hours of duration were excluded from this study.

Preoperative data collection
In addition to demographic variables (gender and age), radiographic data was collected from panoramic radiographs.Intraobserver agreement was obtained by computing kappa coefficient.
Radiographic interpretation was performed in dark using the same negatoscopy and one magnifying glass.The analyzed radiographic variables were: 1 -Eruption stage: total bone inclusion and partial bone inclusion; semi-inclusion and erupted.

Correlation between radiographic signs of third molar proximity with inferior alveolar nerve and postoperative occurrence of neurosensory disorders. A prospective, double-blind study
Acta Cirúrgica Brasileira -Vol.28 (3) 2013 -223

Evaluation of sensory defi cit after surgery
In the postoperative period of seven days the presence/ absence of neurosensory disorders, as well as any important information reported by the patient were investigated.Postoperative neurosensory disorders were measured through pin-prick (Figure 1), two-point discrimination (Figure 2) and brush directional stroke tests (Figure 3).Patients were instructed to remain with their eyes closed during tests 18,19 .Collection of postoperative data was performed by a professional who did not know if the removed tooth presented radiographic sign of IAN proximity.Furthermore, tests were considered of predictive value when positive results were less than 60% of testes 17 .Pin-prick test consisted of repeating ten touches with an explorer tip on the chin skin and perform how many times the patient accused the contact 18 .Brush directional stroke test consisted of performing ten horizontal and/or vertical movements with a microbrush tip on the chin area and ask for the patient to reproduce the movements 18 .Two-point discrimination test consisted in ten repetitions of the compass application with ends gradually but disorderly separated, so that patients can distinguish the sensitivity of one or two points 18 .

Data analysis
Data was encoded and tabulated through the software Statistical signifi cance adopted for all tests was 5% (p<0.05).

Results
From 132 patients enrolled in the sample, six failed to attend the follow-up appointment and were removed from this study.The fi nal sample consisted of 32 (25.39%) male patients and 94 (74.61%) female patients (Table 1).According to age, 45 patients (35.71%) were between 21 and 25 years and 35 patients (27.78%) were between 16 and 20 years (Table 1).The distribution of patients according to eruption stage and degree of bone inclusion of the 3Ms is present in Table 1.There was a higher prevalence of teeth with partial bone inclusion (57 patients, 45.24%), followed by erupted teeth (33 cases, 26.19%) and semi-included teeth (30 cases, 23.82%).
According to the classification of Pell-Gregory 14 in relation to horizontal position of 3Ms, most of the teeth was classified as Class 1 (50%), followed by teeth in Class 2 (47.62%) and only 2.38% in Class 3 (Table 2) In relation of its vertical position, the higher prevalence were teeth in position B (48.41%; p<0.001), followed by teeth in position A (36.51%).Position C was observed in 15.08% of cases (Table 2).According to angulation, as described by Ma'aita and Alwrikat 16 , most of the cases was in vertical position (46.04%),followed by teeth in mesioangular (35.71%) and horizontal position (17.46% ) (Table 2).
The root proximity relation of the 3Ms with mandibular canal was analyzed according to Félez-Gutiérrez et al. 6 modified by Gomes 17 .It was observed more frequently the "narrowing of the mandibular canal" (86 cases, 68.25%), followed by "obscuring of the teeth apices" (59 cases, 46.82%) (Figure 1).In the present study, 61 (48.41%) of the 126 3Ms had at least one or two radiographic signs of proximity with NAI.Forty-seven (37.3%) had three or more signs, and 18 (14.29%)did not have any radiographic signs of proximity to mandibular canal (Table 3).The surgical technique (osteotomy/odontoseccion) used for 3Ms removal did not influence the nerve injury occurrence.
From the 126 patients enrolled in the study, osteotomy was used in 83.3%, while in 28.56%, odontoseccion was added to this procedure (Table 4).

Discussion
Third molars position is very important for surgical planning and assessment of the procedure difficulty degree 19 .
For the present study, panoramic radiographs were used for topographic evaluation of the 3Ms, considering the effectiveness of this commonly used imaging exam as a tool in assessing risks and possible complications during the surgical removal of 3Ms, especially injury to NAI 20 .It allows general practitioners and maxillofacial surgeons to obtain an evaluation of the elementary preoperative conditions, such as teeth position and angulation (Pell-Gregory and Winter classifications), proximity of the 3Ms with the mandibular canal and assessment about root shape and number and bone quality 19 .
Patients with deficits related to NAI often suffer from paresthesia, anesthesia or dysesthesia of the lip, chin or vestibular gum in the affected side 3 .Prevalence of NAI injury has been reported between 0.5% and 5% according to some studies 20,21 .
Disturbances related to NAI were not observed in any patients after third molar surgery.Analyzing prospective studies involving the 3Ms removal that had NAI damage incidence inferior to 1%, it was observed only two studies with similar results to the present study (Table 6).
In a recent literature review about prospective studies evaluating risk factors for the occurrence of neurosensorial deficit after 3Ms surgical removal 3 , statistically significant association was found for some variables, such as eruption stage (p<0.0001),surgical technique (p<0.0001), and predictive radiographic signs (p<0.0001).Concerning to other investigated variables, the authors did not find any statistically significant association, such as gender (p=0.1648) and pattern of bone impaction (p=0.4632).
In this study, gender showed statistically significant influence.
Most of the third molars were partially erupted in the present research.Regarding the horizontal classification adopted, the highest found prevalence were teeth in Class 1 and Class 2, with close values (50% and 47.62%, respectively), confirming the study of Inaoka et al. 23 .According to vertical classification, there was a predominance of B position, different from the results of Szalma et al. 20 .
In relation to angulation as described by Ma'aita and Alwrikat1 16 , vertical position (46.04%) was followed by mesioangular angulation (35.71%), which were different from the results observed by Leung and Cheung 3 that observed a higher prevalence of mesioangular (45.8%), followed by horizontal (11.6%), vertical (21.3%) and distoangulares positions of teeth (21.3%).Blaeser et al. 21did not have association between 3Ms angulation and risk of IAN injury similar to the results of the present study.

A
prospective, double-blind, observational, unicentric pilot study was performed with patients from the postgraduate program in Oral and Maxillofacial Surgery of the Ceara Federal University in the period from March to October/2011.The study was approved by the ethics committee from UFC, Brazil under protocol number 130/11.

2 - 3 -
Horizontal and vertical positions according to Pell-Gregory classification14 .Angulation of the third molar according to Ma'aita and Alwrikat16 that reffers to the angle formed between dental long axis and occlusal plane: Horizontal <20°; Mesioangular = 20-80°, Vertical = 80-100°; Distoangular ≥ 100°.4 -Relation of third molar proximity to the mandibular canal according to Félez-Gutiérrez et al.5 classification modified by Gomes 17 : A) darkening of root tips; B) reflection of root tips, C) narrowing of root tips; D) bifid root tips resting on the mandibular canal; E) deviation of the mandibular canal; F) narrowing of the mandibular canal; G) root apex in island.Surgical procedure Regarding surgical technique, it was standardized and performed in an outpatient setting under local anesthesia with appropriate instruments, following strict standards of biosecurity control.All patients were operated by six dentists specialized in surgery with professional experience of about five years (mean 4.3 years) at 3Ms surgery.Practitioners were instructed to record the time of surgery. n
Microsoft ® Excel version 2010.Data was subjected to statistical analysis with the software Statistic Package for Social Sciences ® (SPSS), version 17.0 for Windows ® .Descriptive statistics (mean, median and standard deviation) and data frequency were performed.Kolmogorov-Smirnov test was used to check normality of the variables.Chi-square or Fisher's exact test was performedto analyze the association of the variables, when a variable had less than fi ve observations.Mann-Whitney and Krukal Wallis tests were used to compare the mean of ages.In addition, a logistic regression model was used to evaluate variables that could infl uence the proximity relation of lower third molar with mandibular canal.For this analysis, it was used the likelihood ratio test.In this test, the chi-square statistic was the difference in -2 log-likelihoods between the fi nal model and a reduced model formed by omitting an effect on the fi nal model.Furthermore, the null hypothesis was that all the parameters of this effect were 0.

TABLE 1 -
Distribution of patients according to gender, age and inclusion degree of the lower third molar.

TABLE 3 -
Distribution of the lower 3Ms according to Pell-Gregory classification and relation with mandibular canal.

TABLE 4 -
Distribution between surgical technique and lower 3M angulation.Correlation between radiographic signs of third molar proximity with inferior alveolar nerve and postoperative occurrence of neurosensory disorders.A prospective, double-blind studyThrough logistic regression, the influence of gender, bone inclusion degree of the third molars (Pell and Gregory classification) and angle in relation of proximity to the mandibular canal were evaluated.It was observed that only gender showed influence, with p<0.001 (Table5).

TABLE 6 -
Prospective studies about incidence <1% of sensorineural injury to IAN after surgery to lower 3Ms removal.