Introduction - Materials and methods.
C. Bellucci & N. PerriniPrivate practice, Rome, Italy.
Accademia Italiana Endodonzia, Pistoia, Italy. Introduction.
The survival of root filled teeth may depend on the amount of residual dentine (Hess & Keller 1928,Weeler 1965, Lloyd DuBrul 1980).M any studies demonstrate a direct relationship between the loss of tooth structure and the possibility of fracture of the tooth crown or root (Trabert et al.1978, Guzy & Nicholls1979, Mondelli et al. 1980, Sorensen & Martinoff 1984, Morfis 1990, Felton et al.1991,Testori et al.1993,Assif&Gorfil1994).However, little information about the thickness of canal walls is available.
There is an appreciable loss of dentine whilst preparing the access cavity and during canal preparation. The root thickness may be thin especially where grooves are present or between two fused roots. Attention must be paid when selecting endodontic instrumentation, with regard to their diameter and flexibility so that they are compatible with the anatomic shape of the root canal.
It is generally recommended to prepare the longest possible post whilst leaving at least 4 mm of intact filling. However, although roots often appear straight in standard radiographic observations, one study reported straight roots in only 2.7% of cases (Zillich & Yaman 1986).The purpose of this study was to measure the root wall, including both dentine and cementum, in extracted human incisors, canines, and premolars to develop geometric average models. The average dimension of the radicular wall for each cut surface indicated the areas at risk of stripping, perforation and radicular fracturing during endodontic treatment and during post-preparation. Materials and methods.
A total of 220 human teeth extracted primarily for periodontal reasons from adult subjects aged between 35 and 55 years and not previously root treated were collected. T he following teeth were included: 40 maxillary incisors, made up of 15 lateral incisors and 25 central incisors; 20 maxillary canines; 45 maxillary premolars; 50 mandibular incisors; 30 mandibular canines; and 35mandibular premolars. Teeth were stored in a buffered formalin solution or a 2%sodiumhypochlorite solution. The teeth were then cleaned with an ultrasonic instrument to remove calculus and placed in 2%sodiumhypochlorite solution to remove remnants of the periodontal ligament. The teeth were dried and submitted for sectioning and measuring.
The buccal surface of each sample was used to measure the entire length of the tooth and the length of the root alone; all measurements were repeated. Three reference points were established and three horizontal parallel planes for sectioning the root were designated:
- through the buccal cemento-enamel junction
- 4 mm coronal to the anatomic apex, and
- half-way between the other two planes.
All of these measurements were made with a precision mechanical caliper, and all horizontal planes were perpendicular to the long axis of the tooth. All roots had a slight curvature, when a large curvature was present the planes were not parallel but perpendicular to the surface of the root at each position.
The teeth were sectioned by the means of an ultra thin, silicon-carbide separating disk (Dedeco International Inc., Long Eddy, NY, USA), with a thickness of 0.2mm. Measurements of the dentinal width were made using anOPMI1-FC optical microscope (Carl Zeiss, Oberkochen, Germany) with focal lense of 220 mm and with the aid of an object micrometer. All of the measurements were made to an accuracy of 0.01mm. For each cut surface, the following were recorded: maximal bucco-lingual (B-L) diameter, and maximal mesio-distal (M-D) diameter; buccal root thickness, measured from the external buccal limit of the root canal to the buccal surface of the root; lingual root thickness, measured from the external lingual limit of the root canal to the lingual surface of the root; mesial root thickness, measured from the external mesial limit of the root canal to the mesial surface of the root; and distal root thickness, measured from the external distal limit of the root canal to the distal surface of the root.
No distinction was made between mandibular central or lateral incisors. All mandibular premolars were included in a single group. The presence of two canals in one root was rare. In these cases, analysis was carried out in the following manner: the buccal wall measurement was made in relation to the buccal root canal and the measurement of the thickness of lingual wall was made in relation to the lingual root canal; in the mesial and distal zones, respectively, two measurements were made corresponding to the two canals. Therefore, for sections with two canals there were six measurements of wall thickness rather than four; anarithmetic average was made between the two mesial measurements and two distal measurements. For this reason, only one canal roots appear in the graphic model. Maxillary premolars were placed in two groups depending whether they had one single root or two distinct roots.
The data obtained were processed using SPSS software (Statistical Package for the Social Sciences).Mea ns and standard deviations of wall thicknesses were calculated for each position (buccal, lingual, mesial, and distal), each cut surface and each group of teeth. A Matlab software graphic tool (The Mathworks, Natick, MA, USA) was used to obtain the representation of the average thicknesses. Effects, and interactions of tooth type (mandibular or maxillary incisor, canine or premolar), location (buccal, lingual, mesial or distal), and section (surfaces 1-3) on the variation of wall thickness were examined by subjecting the data to a three-way analysis of variance (anova). For the statistical analysis, we considered the thickness as the dependent variable and the tooth type, section and location as independent variables. A mixed, three-way anova model was applied to analyze one single independent factor (tooth type) and two repeated factors (section and location).We considered all the differences tested as significant when the experimental F had a significance equal to P < 0.05. From this analysis, the group of maxillary premolars with two roots was excluded since it was not homogeneous with respect to the other groups of teeth. A separate analysis was conducted for this group. More over data were subjected to one-way anova to determine which group of teeth presented the greatest wall thickness and whether the difference was significant with respect to the other groups, the Duncan post hoc test was performed with significance level of 0.01. Furthermore, the wall thicknesses were compared, to determine whether the canals were central; the means of the buccal and lingual thicknesses, and the mesial and distal thicknesses for each of the three cut surfaces were subjected to paired t-test. The test was considered statistically significant if the two-tailed significance level was less than 0.01.