Evaluation of Sagittal Root Positions and Bone Perforation in Anterior Teeth Using Cone Beam Computed Tomography: An Observational Study Open Access

The dimensions of facial and lingual/palatal aspects of the alveolar process depend on the eventual site and the size and inclination of the roots of the erupted teeth.¹  Cone beam computed tomography (CBCT) has definitely widened the use of computed tomography images in implant dentistry due to visualization, particularly in the buccolingual direction and reduced radiation dosage, which previously was not possible with conventional radiographic techniques.^2–4  The accuracy and application of CBCT in implant planning and placements were investigated in several studies.⁵  Consequently, high accuracy concerning linear and volumetric measurements of osseous defects were established for CBCT in several in vitro studies.⁶ 

Virtual implants can be used to plan the placement of the implant before the actual surgical procedure. Currently, the virtual implant, essential to computer-guided surgery, helps clinicians assess implant positioning with nearby vital structures and plan implant surgical procedures without perforations.⁷  Clinical trials demonstrated satisfactory outcomes for computer-guided implant treatment plans, and a systematic review found that the estimated deviations measured at entry point and apex were 1.07 and 1.63 mm, respectively.⁷ 

The likelihood of immediate implant perforation over the labial bone plate may lead to future implant failure, an esthetically compromised condition, or both.⁷  Hence, analyses of preoperative angulation of the anterior teeth are critical for the long-term survival of the implant in the esthetic areas. Although there are many classifications suggesting the different types of sagittal root positions (SRPs) in the maxillary arch, data concerning the mandibular arch are lacking.

Generally, the mandibular anterior region is considered safe for implant surgical procedures.⁸  However, some studies have demonstrated medial lingual foramen⁹  or lingual plate perforation leading to atypical hemorrhage.^8,10  Data are currently lacking concerning implant position in relation to the lingual plate and the significance of bone morphology to the incidence of lingual perforation.

Nevertheless, data related to SRP and labial perforations are lacking in the Indian population about the maxillary and mandibular anterior teeth for the placement of immediate implants. Hence, the current study aims to find the prevalence of SRP, the predictive incidence of labial bone perforation (LBP) using virtual implant placement, and the correlation between SRP and LBP in maxillary and mandibular anterior teeth.

The study was carried out using CBCT scans obtained from December 2016 to June 2018 from the dental college and hospital database after approval from the institutional research board and ethical committee. The study design and protocol were reviewed by an independent biostatistician not involved in data acquisition. The CBCT scans that were used for diagnostic or therapeutic procedures (not specifically acquired for this research) were acquired from a centralized CBCT center. The scans met study inclusion and exclusion criteria. Based on a random number generator, an initial 140 samples were selected and analyzed without stratifying for age and gender.

The CBCT scans were included in the study based on the following criteria. Scans showing permanent central incisor, lateral incisor, or canine with fully formed apexes were included. Conversely, orthognathic surgically treated cases, grafted alveolar ridge cases, patients with supernumerary or impacted teeth, history of periodontal disease or traumatic injury, presence of artefacts or faulty CBCT images, presence of pathological lesions or evidence of root resorption, and images with dental implant in the region of interest were excluded.

All images were acquired with a CBCT machine (Planmeca) with specifications as follows: 85kV 8mA 6.15s, 908 mGy/cm², and the appropriate sectional slice of 200-μm thickness. To reduce the distortion of digital images, all images were analyzed on screen in a darkened glare-free environment, with a downward gaze, appropriate image blackening and contrast, and a viewing distance between 50 and 70 cm.¹⁰  To improve intra- and interpersonal reliability, all investigators were tested for intra- and interexaminer variability. All measurements were analyzed twice by both examiners with an interval of 1 week between each measurement. If the variability between the 2 examiners was found up to 10%, then the average was considered. However, if the variability was more than 10%, then reassessment was done by another investigator. The scans were saved in a Digital Imaging and Communications in Medicine (DICOM) format, and the data were encrypted in files that were confidentially protected, yet retrievable if needed only by investigators. The DICOM files were imported into Carestream 3D Imaging software (3.3.11 version, Kodak) imaging for analysis. A standardized orientation was established using the following steps for the assessment of study parameters¹¹:

1. Orientation of image was done so that the occlusal plane was parallel to the axial plane in all views and the mid-sagittal plane bisected the head's midline (Figure 1).

2. The crown of the selected tooth was intersected with the axial plane (Figure 2).

3. The sagittal plane was adjusted perpendicular to the subject's arch form in the axial, and sagittal planes were adjusted to pass through the center of the selected crown and root (Figure 3).

The SRP and LRB parameters were analyzed as described in the sections that follow.

Viewing a sagittal sectioned image of the region of interest and the center section of each investigated tooth, the sagittal root positions were classified according to Kan et al¹¹  (Figure 4). They classified the relationship of the SRPs of the maxillary anterior teeth to their respective osseous housings using CBCT. This approach considers implant stability and allows clinicians to appropriately reorganize implant sites as favorable, technically sensitive, or contraindicated for immediate implant placement. The Kan et al¹¹  classification was also applied to the mandibular arch in which the lingual cortical plate was assessed in place of the palatal cortical plate. (Figure 5).

The classification of SRP in the maxillary and mandibular anterior teeth is describes as follows:

• Class I: The full root portion of anterior tooth is in contact with the labial cortical plate in maxillary or mandibular anterior teeth.

• Class II: The apical third of the root portion is located in the middle of the alveolar housing but is not in contact with either the labial or palatal/lingual cortical plate in maxillary or mandibular anterior tooth.

• Class III: The full root portion of the anterior tooth is in contact with the palatal cortical plate in the maxillary or lingual cortical plate in mandible.

• Class IV: Two thirds or more of the root portion of maxillary or mandibular anterior tooth is in contact with both labial and palatal/lingual cortical plates.

Root form implants for maxillary and mandibular anterior teeth were selected from an implant database available in the virtual implant software to replace the investigated tooth root with appropriate diameter, that is, 4.2 mm (maxillary central incisor), 3 mm (maxillary lateral incisor), 4.2 mm (maxillary canine teeth), and 3.75 mm (mandibular anterior teeth)^12,13  as well as length according to the available bone in the particular case, parallel to the long axis of the alveolar bone buccolingually, with 3–4 mm of the implant anchor in native bone apically and/or 1 mm subcrestal to the labial cortical plate and in the alveolar housing in the cingulum area, taking care not to contact the lingual/palatal alveolar bone. The implant position was verified from different cross-sectional and 3-dimensional views. LBP was said to occur when the virtual implant contact/extruded the outline of the cortical bone in the sagittal sectioned images (Figure 6).

An independent statistician not involved in data acquisition performed all statistical tests. The data were analyzed using the SPSS software version 21.0 (IBM SPSS Statistics for Windows version 21.0, IBM Corp, Armonk, NY). Frequency analysis was done for SRP and LBP. The mean rank and sum of ranks were used for right and left side comparison. The correlation between SRP and LBP was assessed by the Pearson correlation test. A P value less than .05 was considered as statistically significant.

From the scans of 1680 teeth available for this study, 1338 teeth qualified for further analysis. Scans were from 81 men aged 19–84 years (mean age = 50.37 ± 14.63 years) and 57 women aged 21–74 years (mean age = 48.33 ± 15.80 years). Scans represented 218 (16.29%) maxillary central incisors, 213 (15.92%) maxillary lateral incisors, 216 (16.14%) maxillary canines, 215 (16.07%) mandibular central incisors, 228 (17.04%) mandibular lateral incisors, and 248 (18.54%) mandibular canines.

The overall and arch-wise frequency distribution of SRP in maxillary and mandibular arches was analyzed (Figures 7 through 9). In maxillary teeth overall, Class 1 was highest at 81.48%, and Class III was least prevalent at 1.08%. Similar distributions were seen in mandibular teeth. Maxillary anterior teeth were the most commonly positioned in Class I (74.18% to 87.96%) followed by Class IV (8.33% for canines; 13.15% for lateral incisors). Class III was least prevalent (0.14% to 0.93%). Similarly, in the mandibular arch, the canines showed the highest frequency of Class I SRP (56.45%), and both incisors were in Class I SRP (approximately 28.00%). The Class IV SRP was more common in both incisors (37.28% to 39.53%) compared with canines. However, there was no statistically significant difference in SRP between right and left sides in maxillary and mandibular arches (Table 1).

The cortical LBP was calculated in the maxillary and mandibular anterior teeth. The maxillary incisors showed a slightly increased prevalence of LBP (2.8% to 3.6%) compared with canines (1.8% to 1.9%). Similar results were observed in mandibular anterior sextant, with the central incisors showing the highest rate of perforation (8.5% to 11.93%) and canines showing lowest rate (1.6% to 3.2%). The maxillary right side teeth showed slightly higher predilection than the left side. The overall probability of LBP was 4.26%, and the perforation was more likely to occur in the mandibular arch (5.64%) than in the maxillary arch (2.8%) (Table 2). The teeth most likely affected were mandibular central incisors (n = 22; 20.42%) followed by mandibular lateral incisors (n = 11; 6.61%). Compared with the ridge classification, LBP was most likely to occur at the SRP Class IV (n = 35; 15.3%) compared with other SRP types (Table 3). There was no significant difference in the LBP compared with both sides (Table 4). The Pearson r correlation coefficient test showed statistically significant results between overall SRP and LBP (r = 0.17; P < .01) (Table 5).

This study presented the frequency of the SRP in the maxillary and mandibular arches. assessed according to the classification by Kan et al.¹¹  Lau et al¹⁴  considered root angulation and root position in the sagittal view. Xu et al¹⁵  classified SRP in alveolar bone based on position of incisor apex in the 3 partitions considering anatomical factors in maxillary esthetic zone. Chung et al¹⁶  classified SRP of maxillary central incisors based on the relationship between the maxillary central incisor and the alveolar bone in the sagittal plane. Wang et al¹⁷  classified the root position into 3 types based on the position of apices of maxillary central incisors, that is, buccal, middle, and palatal SRP. Gluckman et al¹⁸  classified root position based on vertical axial inclination, buccopalatal orientation of tooth in ridge, and thickness of bone walls. These classifications are mainly considered for maxillary incisors but have disadvantages; without provide guidelines for clinicians complex to classify^14,15  or providing guidelines for clinicians¹⁶  or appears complex to define the root position of maxillary anterior teeth.¹⁸ 

This study adopted the step-by-step protocol for analysis of CBCT from a previous study.¹⁹  Previous studies reported that the type 1 SRP or buccal type maxillary incisor root apice was in the range of 78.8%¹⁴  to 95.4%.¹⁵  Type 4 SRP or middle third placement of incisors was 4.4%¹⁶  to 13.0%.¹¹  Type 3 or palatal placement ranged from 0.2%¹³  to 1.8%.¹⁴  In our study, SRPs in all maxillary anterior teeth were 81.48%, 7.1%, 1.08%, and 10.34% for Class I, II, III, and IV, respectively; these values are similar to those of Kan et al.¹¹  However, maxillary canines showed slightly higher frequency 87.96% of Class 1 SRP and lower frequency of Class II SRP compared with the findings of Kan et al.¹¹  These findings are suggestive of available bone for immediate implant placement and primary stability. Approximately 81% of implant were anchored more on palatal bone, and in the remaining cases implant engagement was limited to apical and trabecular labial bone, particularly in SRP Class IV.

Furthermore, to our knowledge this is the first study to use the same classification to classify mandibular SRPs. Class I and IV seemed to be the most prevalent. As the mandibular anterior region is usually D1 and/or D2 bone density and has higher compressive strength, the primary implant stability is achieved easily in Class I SRP type. However, Class IV has the highest risk of osseous destruction due to the limited amount of residual bone and inherent characteristics of higher bone density. In this study, SRP Class II and III were almost in the same range. Although it is higher than maxillary teeth, the ability of engaging labial bone in achieving primary implant stability without lingual perforation is critical, and additional regenerative methods might be indicated.

The dental implants causing notable bleeding complications were most often located in the mandibular canine and incisors region due to the presence of the sublingual and submental arteries in these areas. The median distance from the sublingual and submental arteries to the alveolar crest was 15 mm in the region of the incisors and canines.²⁰  Damage to these vessels during the surgical procedure would lead to hemorrhage, requiring emergency management.^9,21  Thus, the immediate implant placement should be carried out cautiously in the mandibular anterior.

This computer simulation study was designed to estimate the predicted incidence of bone perforation using virtual implant software. The lingual bone and medial lingual foramen were handled carefully so as not to perforate, so the management of hemorrhage becomes a challenging task. In the present study, overall LBP was 4.26% with no significant difference in the right and left sides. This is the first study that reports LBP in the mandibular anterior teeth using a virtual implant. The frequency of LBP was mostly in Class IV SRP and in the mandibular central incisors. These findings are contradictory to other studies. Chung et al¹⁶  reported overall LBP to be 81.7%. The maxillary central incisor was 2.37 times more likely to have LBP than the canine. Class I SRP was 4.9 times more likely to be associated with LBP than wad a Class IV SRP. Most of the maxillary incisor roots were positioned close to the buccal cortical plate and had a thin buccal bone wall.²²  However; Chan et al²³  reported that 18.75% of fenestration defects occurred in maxillary anterior teeth using parallel type implants. The lower occurrences of buccal bone perforations in this study were probably due to use of root form implants compared with other studies. In our opinion, the high chance of cortical perforation may be prevented when an implant is placed along the long axis of the ridge, or immediate implants should be avoided in Class IV cases as the highest predictive incidence of LBP was observed in Class IV SRP, and the correlation was statistically significant.

Small voxel sizes and segmental accuracy are important in CBCT image accuracy. It was reported that full field of view results in lower noise levels, resulting in fewer artifacts.²⁴  In the current study, CBCT images of acceptable diagnostic quality was analyzed with voxel size of 200 μm while following radiation dose-sparing protocols. The elimination of implant-related artifacts, such as beam-hardening through the virtual implant placement, is another advantage of this study. The virtual implant placement could serve as a tool for better implant placement using the standardized assessment protocol as presented in this study. This study follows a predefined protocol, with which the CBCT data were obtained and analyzed, and the virtual implants were placed. The clinical relevance with the CBCT data can be comparable with direct measurments.⁷  This is important in that the protocol must be followed to get a data that can be compared with other studies using the same methods.

The current study has some limitations, however. Commercially available CBCT machines need to be calibrated repeatedly according to the manufacturer, affecting the diagnostic image quality. Hence, the image quality depends on the standardization of the dose calibration in different machines. The implants selected were all the same sizes for the same type of teeth across all the patients. This is not present clinically, where different companies offer different implant sizes that are chosen based on various parameters. The same holds true for the type of implant. Moreover, the sample size to determine the accuracy of this protocol is small. Hence, further studies with larger sample size (multicentric) and different types and sizes of implants should be carried out, and the data can then be compared to make the protocol even more accurate. Most importantly studies comparing presurgical and surgical analysis would provide more evidences to the clinician.

The current study is a preliminary report that could contribute to an enormous database that might be created with the help of more studies to standardize the implant protocol to achieve the best results for dental implantation. With increasing use of CBCT in dental implantology in recent times, patient-specific and other specific indications for CBCT analysis should be considered in addition to complying with the ALADIP (As Low As Diagnostically Acceptable being Indication-oriented and Patient-specific) principles. Computer-guided implant planning is known to improve the predictability of the implant placement and risk management. However, the accuracy of transferring the virtual implant placement into clinical scenario is challenging.

Class I SRP was the most prevalent in maxillary and mandibular arches, followed by Class IV type. Significantly less bone perforations occurred when the virtual root form implants were placed. The highest number of bone perforations occurred in Class IV SRP and in mandibular anterior teeth. The SRPs and LBPs showed a statistically significant correlation in the maxillary and mandibular arches. Immediate implant placement in mandibular anterior teeth and Class IV SRP (approximately 10% to 30%) should be interpreted cautiously, and advanced regenerative approaches should be planned well before treating them; alternatively, delayed implant placement may be considered.

CBCT:

cone beam computer tomography

DICOM:

Digital Imaging and Communications in Medicine

LBP:

labial/lingual bone perforation

SRP:

sagittal root position

We would like to acknowledge Sinhgad Dental College and Hospital for providing access to cone beam computed tomography software. We also acknowledge Dr Vineet Vinay, Department of Public Healrth Dentistry, Sinhgad Dental College and Hospital, Maharashtra, India, for analyzing and interpreting study results.

The authors report no conflicts of interest.