0099-2399/92/1806-0294/$03.00/0 JOURNAL OF ENDODONTICS Copyright © 1992 by The American Association of Endodontists
Printed in U.S.A. VOL. 18, NO. 6, JUNE 1992
A Three-Dimensional Study of Canal Curvatures in the Mesial Roots of Mandibular Molars Cary J. Cunningham, DDS, and E. Steve Senia, DDS, MS, BS, FACD
The degree and configuration of canal curvature was studied in the mesial roots of 100 randomly selected mandibular first and second molars. The teeth were radiographed in buccolingual (clinical) and mesiodistal (proximal) directions with # 8 K files in place. One hundred percent of the specimens demonstrated curvature in both views. No correlation in degree of curvature was found to exist between the clinical and proximal views. Secondary curvature, in a direction opposite to that of the principle curve, was seen more frequently in the proximal view. In the proximal view, canals exhibited greater mean curvature than in the clinical view 38% of the time. Weine type II morphology (two canals, one foramen) demonstrated the greatest range in canal curvature when viewed from the proximal. Coronal flaring with Canal Master rotary instruments to a level just coronal to the curve significantly reduced the severity of curvatures in both views for most cases.
Green (3) described the mesial root of the mandibular first molar as: "two canals invariably present--one buccal and one lingual. The two canals usually diverge and then converge slightly and end in two apical foramina. The apical third of the root curves slightly distally." Green stated that in the proximal view, root canals exhibited the largest number of variations. Few studies have actually measured the degree of curvature in root canals. Schneider (4) was one of the first to describe a reliable method of determining canal curvatures from clinical view radiographs. He did not investigate curvatures seen from the proximal view. Pineda and Kuttler (5) used a roentgenographic method for evaluating root canals from both clinical and proximal views. No discussion was included regarding the degree of curvature or if one could predict proximal curves from the curvatures seen in the clinical view radiograph. Vertucci (6) cleared 2000 permanent teeth, 100 of which were mandibular first molars and 100 second molars. He examined canal number, classification, apical foramina locations, and frequency of apical deltas. The classification, number of foramina, and percentages of canals per root that he found agreed with the findings of Okumura (7) and Pineda and Kuttler (5). The transparent specimens gave an excellent three-dimensional view of the pulp cavity. Vertucci's study (6) did not report on the degree or configuration of the curvatures or if they related to the type of canal classification. Fisher et al. (8) fabricated three-dimensional, morphologically accurate reproductions of human pulpal canal anatomy by injecting colored resin into pulp chambers and then embedding each tooth in transparent resin blocks. Skidmore and Bjorndal (9) created plastic casts to reproduce the root canal anatomy of human mandibular molars. These reproductions provided an excellent elucidation of canal anatomy, frequency of separate or joined foramina, and presence of anastomoses. No description or measurement of actual canal curvature was given, however. Slowey (10) described the mesiolingual canal of the mandibular first molar as usually straighter than the mesiobuccal canal. The mesiobuccal canal usually had a greater curvature toward the buccal in the coronal half of the root as seen from a proximal view radiograph. He felt that many endodontic failures were related to the presence of undetected canal configurations and totally missed canals. He emphasized the necessity of taking angled radiographs in multirooted teeth,
To enhance clinical success, dental practitioners must be aware of root canal morphology, including the configuration and degree of canal curvatures. This information is necessary not only in a mesial to distal direction, as seen in a clinical view radiograph, but also in a buccal to lingual direction (proximal view radiograph). Although canal curvature in the proximal view is unseen by the clinician with routine radiographic techniques, it can play a significant role in the cleaning and shaping process. How often the dentist will encounter this unseen curvature, its configuration, and amount of curvature has not been well documented. The morphology of human mandibular molars has been widely studied since the 1870's. Hess and Zurcher (1), in 1925, forced rubber into the pulp chambers and canals of teeth, vulcanized the rubber, then decalcified the specimens. Mueller (2) further reported on the subject in 1936. He described five vulcanite models of mandibular molars from the work of Hess and Zurcher of 1925. "They are grotesque looking figures, which represent decidedly irregular canals, irregular as to the fact that they are not conical nor straight." They aptly described them as the "most complicated of all canals." 294
Vol. 18, No. 6, June 1992
or any time unusual anatomy was suspected, to aid in detecting these anomalies. Painstaking efforts are described in the literature in attempts to teach students of endodontics the subtleties of root canal morphology (1-10). With advances in computer imaging, radiographic analysis with a video image processing program has been used to obtain three-dimensional anatomy of root canals, volume, cross-sectional views, and line drawings (11, 12). Tang and Stock (13) described pre- and postoperative instrumentation anatomy of root canals via enlarged photographic prints obtained from radiographs. Davis et al. (14) made models from the injection of silicone impression material. Photographs of the models demonstrated the clinical and proximal view canal curvatures, but no attempt at description or quantification was made. Weine et al. (15) categorized the most common canal configurations of 75 extracted mandibular second molars using radiographs from two directions with files in place at the working length. This method was chosen to provide a clinically oriented investigation. Review of the literature did not reveal a similar clinically oriented investigation to determine the frequency and degree of canal curvatures in mandibular molars. The purpose of this study was to determine the frequency, degree of curvature, and the configuration of mesiobuccal (MB) and mesiolingual (ML) root canals of mandibular molars. The effect of coronal flaring on canal curvatures in both dimensions was then investigated. A radiographic approach was used to evaluate clinical view (CV) as well as proximal view (PV) curvatures.
Three-Dimensional Canal Curvature
295
axis of the root perpendicular to the central X-ray beam. A line between the MB and ML canal orifices was aligned perpendicular to the X-ray film. Exposures were made (S. S. White Pennwalt Marksman I, Holmdel, N J) from a buccal to lingual direction with a constant source to object distance at 70 kVp, 15 mA, and 14 impulses. The same procedure was followed for the ML canal on the other half of the film. It was then developed, fixed, washed, and dried in an automatic processor according to the manufacturer's directions. Next, the teeth were cut through the furcation with an Isomet saw, separating the two roots. A line between the MB and ML canal orifices was aligned parallel to the film and the PV radiograph was exposed from a mesial to distal direction with #8 K files in both canals as previously described. Figures 1 and 2 demonstrate representative radiographs.
Determination of Curvature All radiographs were mounted in 2 x 2 plastic slide mounts and projected onto a built-in screen of a Bell & Howell Ring Master II projector (Bell & Howell, Laguna Miguel, CA) at a magnification of x7.8. The root outlines were traced onto a piece of white paper secured to the screen. The canal curvatures were calculated for both views using the technique described by Schneider (4). Point a was marked with a 0.3mm lead pencil on the white paper at the middle of the file at the level of the canal orifice (Fig. 3). A line was drawn with a straight edge aligned parallel to the file image from point a to a point where the instrument deviated from the straight edge, point b. A third point (c) was made at the apical foramen
MATERIALS AND METHODS
Specimen Selection Random selection of 100 mandibular first and second molars was performed from a pool of several thousand teeth representing a completely mixed age and race population. This number was chosen to reduce variation to a negligible level through stratified randomization. Selection criteria eliminated teeth with incompletely formed apices, third molars, previous endodontic therapy, and teeth with gross decay or large restorations that would make identification impossible. All molars had been previously fixed in 10% formalin. The root surfaces were debrided with hand scalers, washed, and stored in individually numbered vials containing distilled water. The crowns were removed just below the roof of the pulp chamber with an Isomet low-speed saw (Buehler Ltd., Evanston, IL) in preparation for radiographic evaluation.
FIG 1. Weine type II configuration. A, CV of MB canal. B, CV of ML canal. C, PV of MB and ML canals.
f!
Radiographic Technique Radiographs were made in the following manner. For the CV, a #8 K file was introduced into the MB canal orifice and gently advanced until visible at the apical foramen. The specimen was attached to one-half of a Kodak 02 D speed film, DF-58 (Eastman Kodak Co., Rochester, NY) with soft wax, then mounted onto a Plexiglas jig for alignment with the X-ray tube. A small tri-square was used to align the long
FIG 2. Weine type III configuration. A, CV of MB canal. B, CV of ML canal. C, PV of MB and ML canals.
296
Cunningham and Senia
Journal of Endodontics
Coronal Flaring
M
D
Twenty randomly selected teeth were chosen for coronal flaring with Canal Master rotary (CMR) instruments (Brasseler USA, Savannah, GA). This was done to determine the effects coronal flaring had on the original curvature as seen in both views. A custom matrix (3M Express HP Putty Impression Material, St. Paul, MN) was fabricated for each root then secured to an X-ray film packet so that the roots could be radiographed in the same orientation before and after CMR instrumentation. Preoperative radiographs were made using the previous settings. Rotary length was determined by measuring from a coronal reference point to the level of the primary curve (point b) in the CV radiographs. Following filing with #8 through #15 K files to rotary length (point b), rotary instrumentation was initiated with a CMR #50 and completed with a CMR #80 to the predetermined rotary length or until the instrument would no longer move apically using a gentle force (Fig. 5). In no instance was the CMR taken past the predetermined length. Irrigation was performed with 3 ml of 5.25% NaOC1 after each rotary instrument. Postoperative radiographs were made as before and the degree of curvature was recalculated for both views. Pre- and postrotary curvatures were compared (Fig. 5). RESULTS
Frequency of Curvature FiG 3. Technique used for determining primary root canal curvature in the clinical view. Points a to b, long axis of root canal to point of canal deviation from long axis. Point c, apical foramen. Angle is measured at the intersection of lines a and b and b and c. Vector 1 is the direction of transport during instrumentation.
and a line was drawn from this point to point b. The angle formed by the intersection of the two lines was measured as the canal curvature. The first curve encountered in a canal was the primary curve. A secondary curve was one that deviated in a direction opposite to the primary curve. When more than one curve was present in the canal, the primary curve was measured as previously described to its most apical extent (point c) prior to the deviation away from the central axis of the tooth. The secondary curvature was then measured from point c to the apical foramen (d) (Fig. 4). This separation of curvatures was made because if they had been combined, an apparent straightening of the curve would have resulted. Figure 2C demonstrates a secondary curvature as seen in a proximal view radiograph. The distance from the level of the canal orifice (point a) to the initiation of the curve (point b) was measured in millimeters for each canal from both views. Twenty percent of the specimens were then randomly chosen by an independent investigator to verify the method of curvature determination. A stereomicroscope (Zeiss, Oberkochen, FRG) was used to examine apical foramina in cases where it was difficult to discern, either visually or radiographically, the number of apical foramina and the Weine type classification.
All canals in both views had some curvature. Mean degrees of primary canal curvature are summarized in Tables 1 to 3. The greatest curvature was observed in the CV of the MB canal. The total sample of 100 teeth contained 48 right and 52 left molars; 53% were a Weine type II and 47% type III. Differences in canal curvature means (in degrees) obtained for Weine type II (Table 2) and III (Table 3) configurations were analyzed with Student's t-tests. In Weine type II canals, the CV of the MB canal curvature was significantly greater (p < 0.01) than the CV of the MB canal for the Weine type III molars. None of the other curvatures from any view were significantly different; however, all type II canal curvature means were greater than type III means. To determine if the degree of curvature of a canal seen in a CV correlated with its degree of curvature in a PV, Pearson correlation coefficients were calculated. The calculations were determined for all 100 roots, then for Weine type II (n = 53) and type III (n = 47). No statistically significant correlation was noted (p > 0.15). Clinically, one cannot estimate the degree of proximal curvature in either the MB or ML canal by examining the CV radiograph. Pearson correlation coefficients were then calculated for the CV of the MB canal compared with the CV of the ML canal for each tooth. As anticipated, the entire sample was significantly correlated (r = 0.81, p < 0.001), as was the correlation for Weine type II (r = 0.85, p < 0.001) and Weine type III (r = 0.68, p < 0.001). The correlation demonstrated the direct relationship in degree of curvature between the MB and ML canals when viewed from a clinical radiograph. Correlation coefficients were also calculated for the PV of the MB canal compared with the PV of the ML canal. The total sample of 100 teeth was significantly correlated (r =
Vol. 18, No. 6, June 1992
LI
Three-Dimensional Canal Curvature
/.
/
297
/B
2
12 °
0.75mm ' t
dfPrimary curve
d
Secondary y 2 curve FIG 4. Technique used for determining secondary root canal curvature in the proximal view. Angle is measured at the intersection of lines b and c and c and d. Vector 2 is the direction of transport during instrumentation. Note shorter ML root (average, 0.75 mm).
FiG 5. Comparison of canal curvature before and after the Canal Master rotary instrument was used in the coronal portion of the canal. A, Preoperative clinical view (/eft), CMR at rotary length (center), and postoperative view (right) with a reduction in curvature of 13 degrees. ML instrument was removed view for clarity. B, Preoperative proximal view (/eft), CMR at rotary length (center), and postoperative view (right) with a reduction in curvature of 17 degrees (MB) and 6 degrees (ML). Notch is on the buccal surface.
TABLE 1. Mean primary canal curvatures: clinical and proximal views of 100 roots
0.26, p < 0.01), as was the correlation for Weine type II canals (r = 0.34, p < 0.015). A strong relationship in mean degree of curvature between the MB and ML canals was demonstrated in the PV. However, the Weine type III correlation coefficient was not statistically significant (r = 0.1 l, p > 0.4). The primary PV curves were equal to or greater than CV curves in 25% of the specimens. When secondary curvatures were included, 38 % of the PV curves equaled or exceeded the clinical. Sixteen percent were a Weine type II and 22% Weine type III classification. The number and type of canals exhibiting secondary curvature in a plane opposite that o f the direction o f the primary curve are shown in Table 4. In a PV radiograph, this curvature was always away from the central axis of the root (Fig. 2C). The highest mean curvature was in the MB canal. The PV curvatures demonstrated the greatest range and number of secondary curves. Examination revealed 60 of 200 canals (30%) seen in the PV contained secondary curves, while only 5 o f 200 (2.5%) of the CV had them. Figure 6 demonstrates the range and means of the distance from the level of the canal orifice (point a) to the initiation of the curve (point b) for each canal from both views. The level of the secondary curve was measured from the foramen to its point of deviation (point c) as seen in the PV. This distance ranged from 0.9 to 3.8 m m with a mean of 2.2 ram. Of the 200 canals, 94% were patent to a #8 K file to the foramen. The remaining 6% were measured to the most apical
Canal (view)
Mean (degrees)
SD
Range (degrees)
MB (clinical) ML (clinical) MB (proximal) ML (proximal)
28.7 27.2 21.0 19.7
6.4 6.9 7.6 8.0
17-43.5 10-45 6-40 7.5-48
TABLE 2. Mean primary canal curvatures: clinical and proximal views (Weine type II, 53 roots) Canal (view)
Mean (degrees)
SD
Range (degrees)
MB (clinical) ML (clinical) MB (proximal) ML (proximal)
30.3 28.4 22.1 20.8
7.3 7.7 8.0 8.4
18.5-43.5 10-45 6.5-40 7.5-48
extent of the tip of the file. This was normally within 2 m m of the foramen. Twenty-one percent of the MB canals were longer than the ML canals by an average of 0.75 m m (Fig. 4). None of the ML canals were longer; the remaining 79% were approximately equal in length. The method o f curvature determination was verified by the independent investigator for 20% of the sample and found to be within +_ 2 degrees.
298
Cunningham and Senia
Journal of Endodontics
TABLE 3. Mean primary canal curvatures: clinical and proximal views (Weine type III, 47 roots)
Mean (degrees)
SD
(view)
Range (degrees)
MB (clinical) ML (clinical) MB (proximal) ML (proximal)
26.9 25.8 19.7 18.4
4.5 5.5 7.1 7.5
17-35 15-38 6-38 7.5-35.5
Canal
Coronal Flaring
from the proximal was approximately 1 mm greater than the MB canal. This could lead to an impression of a straighter canal. No statistical difference in degree of curvature was noted, but the MB mean curvature was greater than the ML (Table l). Secondary curves were seen in 30% of the PV radiographs with a mean distance of 2.2 mm from the foramen. This unseen curvature may account for loss of working length as canals are progressively instrumented with larger files. The larger instruments are unable to negotiate this additional curve and loss of length in the apical 2 mm of the canal Occurs.
Reduction of the arc of curvature, both in CV and PV, was achieved using Canal Master rotary instruments (Table 5). Figure 5 illustrates before and after coronal flaring with rotary instruments in a typical case. DISCUSSION The methodology used in this study followed that of Weine et al. (15) who used files in canals to determine canal configuration of the mandibular second molar. Using a radiographic technique with files to canal length inherently introduces errors in measuring canal curvature. The file will approximate the actual canal shape but may not conform exactly, especially where a canal is large and the file does not remain centered. The object of this study was not to measure exact curvatures, but to measure the curve that the endodontic instrument must negotiate to reach the apical foramen. By combining first and second molars (100 teeth), the Weine type classification correlated well with earlier studies by Vertucci (6), Skidmore et al. (9), and Weine et al. (15) when their results were also combined. Vertucci (6) and Skidmore et al. (9) found 40% of the mesial roots of mandibular first molars were a Weine type II, 60% type III. Weine et al. (15) found in 75 mandibular second molars that the mesial root consisted of 52% type II and 40% type III canals. When the results of all of these studies are combined for first and second molars, an approximate 50:50 ratio of Weine type II and III canals is seen which approximates the ratio seen in this investigation (53:47). The results of this study agreed with those reported by Weine (16) in regard to the direction of the PV curvatures. The MB canal initially progresses buccally from the orifice, then lingually afte the curve, terminating at the foramen. The ML canal initially progresses lingually then buccally after the curve. Weine (16) did not mention secondary curvatures in his textbook. However, the results of this study found that 30% of PV canals exhibited secondary curves (32% MB, 28% ML). These current findings disagree with Pineda and Kuttler (5) who found only 68.1% of the mesial roots in first molars and 58.3% in second molars with curvatures in both the CV and PV. Curvatures were found in 100% of the canals examined in this study. The results of this investigation agree with those of Green (3) who also reported a greater variety of canal configurations in the PV. Slowey's findings (10) that the ML canal was generally straighter than the MB when viewed from the proximal could not be confirmed. It was observed that the distance from the ML canal orifice to the level of the curve as viewed
Results of coronal flaring confirmed earlier reports by Roane et al. (17) that canal curvature is decreased by altering the entry angle of the instrument (Fig. 5). Coronal and middle flaring with the Canal Master rotary significantly reduced the magnitude of curvatures in both views. This was accomplished in a conservative fashion with CMR #50 through #80. The CMR #50 is approximately the size of a #1 Gates Glidden drill and smaller than a #1 Peeso reamer. The CMR #80 corresponds to a #2 to #3 Gates Glidden or a # 1 to #2 Peeso reamer. Preservation of the distal wall of the canal coronal to the level of the curve was possible with conservative rotary instrumentation. This avoided potential strip perforations in dumbbell-shaped roots with deep distal concavities. Sharp curves and secondary curves in the apical third were not altered to a significant degree by coronal flaring. The CMR was designed not to enter curves (18). Its rigid noncutting pilot prevents the instrument from gaining access to an area where breakage or strip perforations may occur. Leseberg and Montgomery (19) studied canal transportation at the level of the curve and documented the distal and axial (toward the midline) movement of the original canal. This canal transportation is caused by a combination of forces resulting from CV and PV curvatures which produce a vector distally and axially (Fig. 7). From their study it would appear that the greater the proximal curvature, the faster the transportation would progress toward the distal concavity. This could result in strip perforations. Further studies are needed to corroborate this observation. The presence of a dumbbell-shaped mesial root in mandibular molars with severe distal concavities creates difficulties in properly instrumenting in three dimensions. Evaluation of instrumentation techniques in curved canals of molars must include both clinical and proximal curvatures. Human teeth used as their own controls appear to be the ideal model (19) since the third dimension of curvature is built-in. Studies using plastic blocks with only one curvature, or human teeth, when the proximal curvatures are ignored, are not realistic because they neglect the unseen curvature. Only by considering the three-dimensional nature of canal curvatures can instruments and techniques be fairly evaluated and compared. Successful endodontic treatment of mandibular molars requires considerable knowledge of canal curvature if they are to be debrided and obturated successfully. Even armed with this knowledge, the irregularly shaped canals and additional curvatures not apparent on radiographs make treatment difficult at best. No single method of instrumentation and obturation can be adequately applied to all cases. The dental practitioner must be able to anticipate canal morphology in order to best select treatment modalities. Recent advances in digitized, computer-enhanced radiographic techniques may,
Vol. 18, No. 6, June 1992
Three-Dimensional Canal Curvature
299
TABLE 4. Number of canals with secondary curvatures--mean secondary canal curvatures: clinical and proximal views (100 roots) Canal (view)
Weine Type II
Weine Type III
Secondary Curves
Mean (degrees)
SD
Range (degrees)
MB (clinical) ML (clinical) MB (proximal) ML (proximal) Total
2 1 18 13 34
0 2 14 15 31
2 3 32 28 65
16.5 19.4 26.3 23.5
6.4 5.2 13.5 9.4
12-21 13.5-23.5 7-57 7-44
Start of Curve (Measured from Canal Orifice) 16
12 10 E 8 E
I
[ ] Mean I = Range
24 6
0
5.1
ML (CV)
MB (CV)
MB (PV)
ML (PV)
FIG 6. Range and means (in mm) of the distance from the level of the canal orifice (point a) to the initiation of the curve (point b) for each canal from both views.
TABLE 5. Reduction in degree of curvature after Canal Master rotary instruments (40 canals) Canal (view)
Average (degrees)
Range (degrees)
MB (clinical) ML (clinical) MB (proximal) ML (proximal)
8.9 5.9 9.8 7.8
1-18 1-9.5 1-16 0-16
D L
~
1
~
2
/
~
/•f•-Combinedvector ~ C a n a l ~
~
B ranspOrtatiOn
' - , , ~ Original canal
M FIG 7. Cross-sectional drawing of the mesial root at the level of the primary clinical curve demonstrating the combined vector of transportation from clinical and proximal curvatures. Vectors 1 and 2 correspond to the vectors shown in Figs. 3 and 4.
in the future, greatly aid in the visualization of three-dimensional canal morphology. Until then, knowledge of the threedimensional curvature in canals during cleaning and shaping and of the ability of instruments to accurately follow and clean the complex system of root canals is paramount for successful endodontics.
CONCLUSIONS 1. One hundred percent of the specimens demonstrated curvatures in the MB and ML canals, both in clinical and proximal view radiographs. 2. Clinical view curvatures of the MB and ML canals in the same tooth were similar and directly correlated for all specimens. 3. No correlation was found between clinical and proximal view curvatures in the same tooth. Proximal view curvatures cannot be predicted or estimated from examining a CV radiograph. 4. Proximal view curvatures were equal to or greater than those in the clinical view 38% of the time in either the MB or ML canals. 5. Weine type II morphology demonstrated a greater mean curvature with a wider range and a larger number of secondary curves than Weine type III roots when viewed from the proximal. 6. Secondary canal curvatures were only seen in 2.5% of the clinical view radiographs compared with 30% in the proximal views. 7. The mean distance from the level of the canal orifice (point a) to the initiation of canal curvature (point b) was slightly greater with a slightly wider range in the proximal view for both the MB and ML canals. 8. Twenty-one percent of the MB canals were longer than the ML canals by an average of 0.75 mm. 9. The degree of canal curvature was significantly reduced in both views after coronal flaring with Canal Master rotary instruments. This article is a work of the United States government and may be reprinted without permission. Dr. Cunningham is an employee of the United States Air Force, Lackland Air Force Base, TX. Opinions expressed therein, unless otherwise specifically indicated, are those of the authors. They do not purport to express views of the Department of the Air Force or any other department or agency of the United States government. We would like to thank Mr. John Schoolfield for the statistical and computer analysis and Dr. Steve Montgomery for his valuable assistance in preparing this manuscript. Dr. Cunningham is a resident, Department of Endodontics, University of Texas Health Science Center at San Antonio, San Antonio, TX and Wilford Hall USAF Medical Center, Lackland Air Force Base, TX. Dr. Senia is professor and director, Advanced Education Program in Endodontics, University of Texas Health Science Center at San Antonio.
References 1. Hess W, Zurcher E. The anatomy of the root canals of the teeth of the permanent and deciduous dentitions. London: John Bale, Sons and Danielson Ltd., 1925. 2. Mueller AH. Morphology of root canals. J Am Dent Assoc 1936;23:16981706.
300
Cunningham and Senia
3. Green D. Morphology of the pulp cavity of the permanent teeth. Oral Surg 1955;8:743-59. 4. Schneider SW. A comparison of canal preparations in straight and curved root canals. Oral Surg 1971 ;32:271-5. 5. Pineda F, Kuttler Y. Mesiodistal and buccolingual roentgenographic investigation of 7,275 root canals. Oral Surg 1972;33:101-10. 6. Vertucci FJ. Root canal anatomy of the human permanent teeth. Oral Surg 1984;58:589-99. 7. Okumura T. Anatomy of the root canals. J Am Dent Assoc 1927;14:6326. 8. Fisher DE, Ingersoll N, Bucher JF. Anatomy of the pulpal canal: threedimensional visualization. J Endodon 1975;1:22-5. 9. Skidmore AE, Bjorndal AM. Root canal morphology of the human mandibular first molar. Oral Surg 1971 ;32:778-84. 10. Slowey RR. Root canal anatomy road map to successful endodontics. Dent Clin North Am 1979;23:555-73. 11. Mayo VC, Montgomery S, del Rio C. A computerized method for evaluating root canal morphology. J Endodon 1986;12:2-7.
Journal of Endodontics 12. Gullickson DC, Montgomery S. The study of root canal morphology using a digital image processing technique. J Endodon 1987;13:158-63. 13. Tang MP, Stock CJ. An in vitro method for comparing the effects of different root canal preparation techniques on the shape of curved root canals. Int Endod J 1989;22:49-54. 14. Davis SR, Brayton SM, Goldman M. The morphology of the prepared root canal: A study utilizing injectable silicone. Oral Surg 1972;34:642-8. 15. Weine FS, Pasiewicz RA, Rice RT. Canal configuration of the mandibular second molar using a clinically oriented in vitro method. J Endodon 1988; 14:207-13. 16. Weine FS. Endodontic therapy. 4th ed. St. Louis: CV Mosby, 1989:3145. 17. Roane JB, Sabala CL, Duncanson MG. The "balanced force" concept for instrumentation of curved canals. J Endodon 1985;11:203-11. 18. Wildey WL, Senia ES. A new root canal instrument and instrumentation technique: a preliminary report. Oral Surg 1989;67:198-207. 19. Leseberg DA, Montgomery S. The effects of Canal Master, Flex-R, and K-Flex instrumentation on root canal configuration. J Endodon 1991 ;17:59-65.
T h e W a y It W a s Bismarck, the German statesman, is known for his observation to the effect that "it is better that men do not know how their laws or their sausages are made." Sausage figured in another, less publicized event in his life. Virchow, the famous pathologist, was also a politician and served in the opposite party to Bismarck in the Reichstag. Virchow proposed a law requiring inspection of slaughtered hogs for trichinosis parasites. Bismarck opposed and, becoming enraged at Virchow's legislative tactics, challenged him to a duel. Knowing Bismarck to be an accomplished swordsman, and having the choice of weapon, Virchow chose a duel consisting of eating sausages, one of which would be identified as loaded with trichinosis ova. Since Bismarck would get to choose first--and given his scorn of trichinosis--he would be forced to choose the infected one or be ridiculed. Bismarck wisely withdrew his challenge. In 1878 a bill requiring meat inspection passed. Frank Suis
Printed in U.S.A. VOL. 18, NO. 6, JUNE 1992
A Three-Dimensional Study of Canal Curvatures in the Mesial Roots of Mandibular Molars Cary J. Cunningham, DDS, and E. Steve Senia, DDS, MS, BS, FACD
The degree and configuration of canal curvature was studied in the mesial roots of 100 randomly selected mandibular first and second molars. The teeth were radiographed in buccolingual (clinical) and mesiodistal (proximal) directions with # 8 K files in place. One hundred percent of the specimens demonstrated curvature in both views. No correlation in degree of curvature was found to exist between the clinical and proximal views. Secondary curvature, in a direction opposite to that of the principle curve, was seen more frequently in the proximal view. In the proximal view, canals exhibited greater mean curvature than in the clinical view 38% of the time. Weine type II morphology (two canals, one foramen) demonstrated the greatest range in canal curvature when viewed from the proximal. Coronal flaring with Canal Master rotary instruments to a level just coronal to the curve significantly reduced the severity of curvatures in both views for most cases.
Green (3) described the mesial root of the mandibular first molar as: "two canals invariably present--one buccal and one lingual. The two canals usually diverge and then converge slightly and end in two apical foramina. The apical third of the root curves slightly distally." Green stated that in the proximal view, root canals exhibited the largest number of variations. Few studies have actually measured the degree of curvature in root canals. Schneider (4) was one of the first to describe a reliable method of determining canal curvatures from clinical view radiographs. He did not investigate curvatures seen from the proximal view. Pineda and Kuttler (5) used a roentgenographic method for evaluating root canals from both clinical and proximal views. No discussion was included regarding the degree of curvature or if one could predict proximal curves from the curvatures seen in the clinical view radiograph. Vertucci (6) cleared 2000 permanent teeth, 100 of which were mandibular first molars and 100 second molars. He examined canal number, classification, apical foramina locations, and frequency of apical deltas. The classification, number of foramina, and percentages of canals per root that he found agreed with the findings of Okumura (7) and Pineda and Kuttler (5). The transparent specimens gave an excellent three-dimensional view of the pulp cavity. Vertucci's study (6) did not report on the degree or configuration of the curvatures or if they related to the type of canal classification. Fisher et al. (8) fabricated three-dimensional, morphologically accurate reproductions of human pulpal canal anatomy by injecting colored resin into pulp chambers and then embedding each tooth in transparent resin blocks. Skidmore and Bjorndal (9) created plastic casts to reproduce the root canal anatomy of human mandibular molars. These reproductions provided an excellent elucidation of canal anatomy, frequency of separate or joined foramina, and presence of anastomoses. No description or measurement of actual canal curvature was given, however. Slowey (10) described the mesiolingual canal of the mandibular first molar as usually straighter than the mesiobuccal canal. The mesiobuccal canal usually had a greater curvature toward the buccal in the coronal half of the root as seen from a proximal view radiograph. He felt that many endodontic failures were related to the presence of undetected canal configurations and totally missed canals. He emphasized the necessity of taking angled radiographs in multirooted teeth,
To enhance clinical success, dental practitioners must be aware of root canal morphology, including the configuration and degree of canal curvatures. This information is necessary not only in a mesial to distal direction, as seen in a clinical view radiograph, but also in a buccal to lingual direction (proximal view radiograph). Although canal curvature in the proximal view is unseen by the clinician with routine radiographic techniques, it can play a significant role in the cleaning and shaping process. How often the dentist will encounter this unseen curvature, its configuration, and amount of curvature has not been well documented. The morphology of human mandibular molars has been widely studied since the 1870's. Hess and Zurcher (1), in 1925, forced rubber into the pulp chambers and canals of teeth, vulcanized the rubber, then decalcified the specimens. Mueller (2) further reported on the subject in 1936. He described five vulcanite models of mandibular molars from the work of Hess and Zurcher of 1925. "They are grotesque looking figures, which represent decidedly irregular canals, irregular as to the fact that they are not conical nor straight." They aptly described them as the "most complicated of all canals." 294
Vol. 18, No. 6, June 1992
or any time unusual anatomy was suspected, to aid in detecting these anomalies. Painstaking efforts are described in the literature in attempts to teach students of endodontics the subtleties of root canal morphology (1-10). With advances in computer imaging, radiographic analysis with a video image processing program has been used to obtain three-dimensional anatomy of root canals, volume, cross-sectional views, and line drawings (11, 12). Tang and Stock (13) described pre- and postoperative instrumentation anatomy of root canals via enlarged photographic prints obtained from radiographs. Davis et al. (14) made models from the injection of silicone impression material. Photographs of the models demonstrated the clinical and proximal view canal curvatures, but no attempt at description or quantification was made. Weine et al. (15) categorized the most common canal configurations of 75 extracted mandibular second molars using radiographs from two directions with files in place at the working length. This method was chosen to provide a clinically oriented investigation. Review of the literature did not reveal a similar clinically oriented investigation to determine the frequency and degree of canal curvatures in mandibular molars. The purpose of this study was to determine the frequency, degree of curvature, and the configuration of mesiobuccal (MB) and mesiolingual (ML) root canals of mandibular molars. The effect of coronal flaring on canal curvatures in both dimensions was then investigated. A radiographic approach was used to evaluate clinical view (CV) as well as proximal view (PV) curvatures.
Three-Dimensional Canal Curvature
295
axis of the root perpendicular to the central X-ray beam. A line between the MB and ML canal orifices was aligned perpendicular to the X-ray film. Exposures were made (S. S. White Pennwalt Marksman I, Holmdel, N J) from a buccal to lingual direction with a constant source to object distance at 70 kVp, 15 mA, and 14 impulses. The same procedure was followed for the ML canal on the other half of the film. It was then developed, fixed, washed, and dried in an automatic processor according to the manufacturer's directions. Next, the teeth were cut through the furcation with an Isomet saw, separating the two roots. A line between the MB and ML canal orifices was aligned parallel to the film and the PV radiograph was exposed from a mesial to distal direction with #8 K files in both canals as previously described. Figures 1 and 2 demonstrate representative radiographs.
Determination of Curvature All radiographs were mounted in 2 x 2 plastic slide mounts and projected onto a built-in screen of a Bell & Howell Ring Master II projector (Bell & Howell, Laguna Miguel, CA) at a magnification of x7.8. The root outlines were traced onto a piece of white paper secured to the screen. The canal curvatures were calculated for both views using the technique described by Schneider (4). Point a was marked with a 0.3mm lead pencil on the white paper at the middle of the file at the level of the canal orifice (Fig. 3). A line was drawn with a straight edge aligned parallel to the file image from point a to a point where the instrument deviated from the straight edge, point b. A third point (c) was made at the apical foramen
MATERIALS AND METHODS
Specimen Selection Random selection of 100 mandibular first and second molars was performed from a pool of several thousand teeth representing a completely mixed age and race population. This number was chosen to reduce variation to a negligible level through stratified randomization. Selection criteria eliminated teeth with incompletely formed apices, third molars, previous endodontic therapy, and teeth with gross decay or large restorations that would make identification impossible. All molars had been previously fixed in 10% formalin. The root surfaces were debrided with hand scalers, washed, and stored in individually numbered vials containing distilled water. The crowns were removed just below the roof of the pulp chamber with an Isomet low-speed saw (Buehler Ltd., Evanston, IL) in preparation for radiographic evaluation.
FIG 1. Weine type II configuration. A, CV of MB canal. B, CV of ML canal. C, PV of MB and ML canals.
f!
Radiographic Technique Radiographs were made in the following manner. For the CV, a #8 K file was introduced into the MB canal orifice and gently advanced until visible at the apical foramen. The specimen was attached to one-half of a Kodak 02 D speed film, DF-58 (Eastman Kodak Co., Rochester, NY) with soft wax, then mounted onto a Plexiglas jig for alignment with the X-ray tube. A small tri-square was used to align the long
FIG 2. Weine type III configuration. A, CV of MB canal. B, CV of ML canal. C, PV of MB and ML canals.
296
Cunningham and Senia
Journal of Endodontics
Coronal Flaring
M
D
Twenty randomly selected teeth were chosen for coronal flaring with Canal Master rotary (CMR) instruments (Brasseler USA, Savannah, GA). This was done to determine the effects coronal flaring had on the original curvature as seen in both views. A custom matrix (3M Express HP Putty Impression Material, St. Paul, MN) was fabricated for each root then secured to an X-ray film packet so that the roots could be radiographed in the same orientation before and after CMR instrumentation. Preoperative radiographs were made using the previous settings. Rotary length was determined by measuring from a coronal reference point to the level of the primary curve (point b) in the CV radiographs. Following filing with #8 through #15 K files to rotary length (point b), rotary instrumentation was initiated with a CMR #50 and completed with a CMR #80 to the predetermined rotary length or until the instrument would no longer move apically using a gentle force (Fig. 5). In no instance was the CMR taken past the predetermined length. Irrigation was performed with 3 ml of 5.25% NaOC1 after each rotary instrument. Postoperative radiographs were made as before and the degree of curvature was recalculated for both views. Pre- and postrotary curvatures were compared (Fig. 5). RESULTS
Frequency of Curvature FiG 3. Technique used for determining primary root canal curvature in the clinical view. Points a to b, long axis of root canal to point of canal deviation from long axis. Point c, apical foramen. Angle is measured at the intersection of lines a and b and b and c. Vector 1 is the direction of transport during instrumentation.
and a line was drawn from this point to point b. The angle formed by the intersection of the two lines was measured as the canal curvature. The first curve encountered in a canal was the primary curve. A secondary curve was one that deviated in a direction opposite to the primary curve. When more than one curve was present in the canal, the primary curve was measured as previously described to its most apical extent (point c) prior to the deviation away from the central axis of the tooth. The secondary curvature was then measured from point c to the apical foramen (d) (Fig. 4). This separation of curvatures was made because if they had been combined, an apparent straightening of the curve would have resulted. Figure 2C demonstrates a secondary curvature as seen in a proximal view radiograph. The distance from the level of the canal orifice (point a) to the initiation of the curve (point b) was measured in millimeters for each canal from both views. Twenty percent of the specimens were then randomly chosen by an independent investigator to verify the method of curvature determination. A stereomicroscope (Zeiss, Oberkochen, FRG) was used to examine apical foramina in cases where it was difficult to discern, either visually or radiographically, the number of apical foramina and the Weine type classification.
All canals in both views had some curvature. Mean degrees of primary canal curvature are summarized in Tables 1 to 3. The greatest curvature was observed in the CV of the MB canal. The total sample of 100 teeth contained 48 right and 52 left molars; 53% were a Weine type II and 47% type III. Differences in canal curvature means (in degrees) obtained for Weine type II (Table 2) and III (Table 3) configurations were analyzed with Student's t-tests. In Weine type II canals, the CV of the MB canal curvature was significantly greater (p < 0.01) than the CV of the MB canal for the Weine type III molars. None of the other curvatures from any view were significantly different; however, all type II canal curvature means were greater than type III means. To determine if the degree of curvature of a canal seen in a CV correlated with its degree of curvature in a PV, Pearson correlation coefficients were calculated. The calculations were determined for all 100 roots, then for Weine type II (n = 53) and type III (n = 47). No statistically significant correlation was noted (p > 0.15). Clinically, one cannot estimate the degree of proximal curvature in either the MB or ML canal by examining the CV radiograph. Pearson correlation coefficients were then calculated for the CV of the MB canal compared with the CV of the ML canal for each tooth. As anticipated, the entire sample was significantly correlated (r = 0.81, p < 0.001), as was the correlation for Weine type II (r = 0.85, p < 0.001) and Weine type III (r = 0.68, p < 0.001). The correlation demonstrated the direct relationship in degree of curvature between the MB and ML canals when viewed from a clinical radiograph. Correlation coefficients were also calculated for the PV of the MB canal compared with the PV of the ML canal. The total sample of 100 teeth was significantly correlated (r =
Vol. 18, No. 6, June 1992
LI
Three-Dimensional Canal Curvature
/.
/
297
/B
2
12 °
0.75mm ' t
dfPrimary curve
d
Secondary y 2 curve FIG 4. Technique used for determining secondary root canal curvature in the proximal view. Angle is measured at the intersection of lines b and c and c and d. Vector 2 is the direction of transport during instrumentation. Note shorter ML root (average, 0.75 mm).
FiG 5. Comparison of canal curvature before and after the Canal Master rotary instrument was used in the coronal portion of the canal. A, Preoperative clinical view (/eft), CMR at rotary length (center), and postoperative view (right) with a reduction in curvature of 13 degrees. ML instrument was removed view for clarity. B, Preoperative proximal view (/eft), CMR at rotary length (center), and postoperative view (right) with a reduction in curvature of 17 degrees (MB) and 6 degrees (ML). Notch is on the buccal surface.
TABLE 1. Mean primary canal curvatures: clinical and proximal views of 100 roots
0.26, p < 0.01), as was the correlation for Weine type II canals (r = 0.34, p < 0.015). A strong relationship in mean degree of curvature between the MB and ML canals was demonstrated in the PV. However, the Weine type III correlation coefficient was not statistically significant (r = 0.1 l, p > 0.4). The primary PV curves were equal to or greater than CV curves in 25% of the specimens. When secondary curvatures were included, 38 % of the PV curves equaled or exceeded the clinical. Sixteen percent were a Weine type II and 22% Weine type III classification. The number and type of canals exhibiting secondary curvature in a plane opposite that o f the direction o f the primary curve are shown in Table 4. In a PV radiograph, this curvature was always away from the central axis of the root (Fig. 2C). The highest mean curvature was in the MB canal. The PV curvatures demonstrated the greatest range and number of secondary curves. Examination revealed 60 of 200 canals (30%) seen in the PV contained secondary curves, while only 5 o f 200 (2.5%) of the CV had them. Figure 6 demonstrates the range and means of the distance from the level of the canal orifice (point a) to the initiation of the curve (point b) for each canal from both views. The level of the secondary curve was measured from the foramen to its point of deviation (point c) as seen in the PV. This distance ranged from 0.9 to 3.8 m m with a mean of 2.2 ram. Of the 200 canals, 94% were patent to a #8 K file to the foramen. The remaining 6% were measured to the most apical
Canal (view)
Mean (degrees)
SD
Range (degrees)
MB (clinical) ML (clinical) MB (proximal) ML (proximal)
28.7 27.2 21.0 19.7
6.4 6.9 7.6 8.0
17-43.5 10-45 6-40 7.5-48
TABLE 2. Mean primary canal curvatures: clinical and proximal views (Weine type II, 53 roots) Canal (view)
Mean (degrees)
SD
Range (degrees)
MB (clinical) ML (clinical) MB (proximal) ML (proximal)
30.3 28.4 22.1 20.8
7.3 7.7 8.0 8.4
18.5-43.5 10-45 6.5-40 7.5-48
extent of the tip of the file. This was normally within 2 m m of the foramen. Twenty-one percent of the MB canals were longer than the ML canals by an average of 0.75 m m (Fig. 4). None of the ML canals were longer; the remaining 79% were approximately equal in length. The method o f curvature determination was verified by the independent investigator for 20% of the sample and found to be within +_ 2 degrees.
298
Cunningham and Senia
Journal of Endodontics
TABLE 3. Mean primary canal curvatures: clinical and proximal views (Weine type III, 47 roots)
Mean (degrees)
SD
(view)
Range (degrees)
MB (clinical) ML (clinical) MB (proximal) ML (proximal)
26.9 25.8 19.7 18.4
4.5 5.5 7.1 7.5
17-35 15-38 6-38 7.5-35.5
Canal
Coronal Flaring
from the proximal was approximately 1 mm greater than the MB canal. This could lead to an impression of a straighter canal. No statistical difference in degree of curvature was noted, but the MB mean curvature was greater than the ML (Table l). Secondary curves were seen in 30% of the PV radiographs with a mean distance of 2.2 mm from the foramen. This unseen curvature may account for loss of working length as canals are progressively instrumented with larger files. The larger instruments are unable to negotiate this additional curve and loss of length in the apical 2 mm of the canal Occurs.
Reduction of the arc of curvature, both in CV and PV, was achieved using Canal Master rotary instruments (Table 5). Figure 5 illustrates before and after coronal flaring with rotary instruments in a typical case. DISCUSSION The methodology used in this study followed that of Weine et al. (15) who used files in canals to determine canal configuration of the mandibular second molar. Using a radiographic technique with files to canal length inherently introduces errors in measuring canal curvature. The file will approximate the actual canal shape but may not conform exactly, especially where a canal is large and the file does not remain centered. The object of this study was not to measure exact curvatures, but to measure the curve that the endodontic instrument must negotiate to reach the apical foramen. By combining first and second molars (100 teeth), the Weine type classification correlated well with earlier studies by Vertucci (6), Skidmore et al. (9), and Weine et al. (15) when their results were also combined. Vertucci (6) and Skidmore et al. (9) found 40% of the mesial roots of mandibular first molars were a Weine type II, 60% type III. Weine et al. (15) found in 75 mandibular second molars that the mesial root consisted of 52% type II and 40% type III canals. When the results of all of these studies are combined for first and second molars, an approximate 50:50 ratio of Weine type II and III canals is seen which approximates the ratio seen in this investigation (53:47). The results of this study agreed with those reported by Weine (16) in regard to the direction of the PV curvatures. The MB canal initially progresses buccally from the orifice, then lingually afte the curve, terminating at the foramen. The ML canal initially progresses lingually then buccally after the curve. Weine (16) did not mention secondary curvatures in his textbook. However, the results of this study found that 30% of PV canals exhibited secondary curves (32% MB, 28% ML). These current findings disagree with Pineda and Kuttler (5) who found only 68.1% of the mesial roots in first molars and 58.3% in second molars with curvatures in both the CV and PV. Curvatures were found in 100% of the canals examined in this study. The results of this investigation agree with those of Green (3) who also reported a greater variety of canal configurations in the PV. Slowey's findings (10) that the ML canal was generally straighter than the MB when viewed from the proximal could not be confirmed. It was observed that the distance from the ML canal orifice to the level of the curve as viewed
Results of coronal flaring confirmed earlier reports by Roane et al. (17) that canal curvature is decreased by altering the entry angle of the instrument (Fig. 5). Coronal and middle flaring with the Canal Master rotary significantly reduced the magnitude of curvatures in both views. This was accomplished in a conservative fashion with CMR #50 through #80. The CMR #50 is approximately the size of a #1 Gates Glidden drill and smaller than a #1 Peeso reamer. The CMR #80 corresponds to a #2 to #3 Gates Glidden or a # 1 to #2 Peeso reamer. Preservation of the distal wall of the canal coronal to the level of the curve was possible with conservative rotary instrumentation. This avoided potential strip perforations in dumbbell-shaped roots with deep distal concavities. Sharp curves and secondary curves in the apical third were not altered to a significant degree by coronal flaring. The CMR was designed not to enter curves (18). Its rigid noncutting pilot prevents the instrument from gaining access to an area where breakage or strip perforations may occur. Leseberg and Montgomery (19) studied canal transportation at the level of the curve and documented the distal and axial (toward the midline) movement of the original canal. This canal transportation is caused by a combination of forces resulting from CV and PV curvatures which produce a vector distally and axially (Fig. 7). From their study it would appear that the greater the proximal curvature, the faster the transportation would progress toward the distal concavity. This could result in strip perforations. Further studies are needed to corroborate this observation. The presence of a dumbbell-shaped mesial root in mandibular molars with severe distal concavities creates difficulties in properly instrumenting in three dimensions. Evaluation of instrumentation techniques in curved canals of molars must include both clinical and proximal curvatures. Human teeth used as their own controls appear to be the ideal model (19) since the third dimension of curvature is built-in. Studies using plastic blocks with only one curvature, or human teeth, when the proximal curvatures are ignored, are not realistic because they neglect the unseen curvature. Only by considering the three-dimensional nature of canal curvatures can instruments and techniques be fairly evaluated and compared. Successful endodontic treatment of mandibular molars requires considerable knowledge of canal curvature if they are to be debrided and obturated successfully. Even armed with this knowledge, the irregularly shaped canals and additional curvatures not apparent on radiographs make treatment difficult at best. No single method of instrumentation and obturation can be adequately applied to all cases. The dental practitioner must be able to anticipate canal morphology in order to best select treatment modalities. Recent advances in digitized, computer-enhanced radiographic techniques may,
Vol. 18, No. 6, June 1992
Three-Dimensional Canal Curvature
299
TABLE 4. Number of canals with secondary curvatures--mean secondary canal curvatures: clinical and proximal views (100 roots) Canal (view)
Weine Type II
Weine Type III
Secondary Curves
Mean (degrees)
SD
Range (degrees)
MB (clinical) ML (clinical) MB (proximal) ML (proximal) Total
2 1 18 13 34
0 2 14 15 31
2 3 32 28 65
16.5 19.4 26.3 23.5
6.4 5.2 13.5 9.4
12-21 13.5-23.5 7-57 7-44
Start of Curve (Measured from Canal Orifice) 16
12 10 E 8 E
I
[ ] Mean I = Range
24 6
0
5.1
ML (CV)
MB (CV)
MB (PV)
ML (PV)
FIG 6. Range and means (in mm) of the distance from the level of the canal orifice (point a) to the initiation of the curve (point b) for each canal from both views.
TABLE 5. Reduction in degree of curvature after Canal Master rotary instruments (40 canals) Canal (view)
Average (degrees)
Range (degrees)
MB (clinical) ML (clinical) MB (proximal) ML (proximal)
8.9 5.9 9.8 7.8
1-18 1-9.5 1-16 0-16
D L
~
1
~
2
/
~
/•f•-Combinedvector ~ C a n a l ~
~
B ranspOrtatiOn
' - , , ~ Original canal
M FIG 7. Cross-sectional drawing of the mesial root at the level of the primary clinical curve demonstrating the combined vector of transportation from clinical and proximal curvatures. Vectors 1 and 2 correspond to the vectors shown in Figs. 3 and 4.
in the future, greatly aid in the visualization of three-dimensional canal morphology. Until then, knowledge of the threedimensional curvature in canals during cleaning and shaping and of the ability of instruments to accurately follow and clean the complex system of root canals is paramount for successful endodontics.
CONCLUSIONS 1. One hundred percent of the specimens demonstrated curvatures in the MB and ML canals, both in clinical and proximal view radiographs. 2. Clinical view curvatures of the MB and ML canals in the same tooth were similar and directly correlated for all specimens. 3. No correlation was found between clinical and proximal view curvatures in the same tooth. Proximal view curvatures cannot be predicted or estimated from examining a CV radiograph. 4. Proximal view curvatures were equal to or greater than those in the clinical view 38% of the time in either the MB or ML canals. 5. Weine type II morphology demonstrated a greater mean curvature with a wider range and a larger number of secondary curves than Weine type III roots when viewed from the proximal. 6. Secondary canal curvatures were only seen in 2.5% of the clinical view radiographs compared with 30% in the proximal views. 7. The mean distance from the level of the canal orifice (point a) to the initiation of canal curvature (point b) was slightly greater with a slightly wider range in the proximal view for both the MB and ML canals. 8. Twenty-one percent of the MB canals were longer than the ML canals by an average of 0.75 mm. 9. The degree of canal curvature was significantly reduced in both views after coronal flaring with Canal Master rotary instruments. This article is a work of the United States government and may be reprinted without permission. Dr. Cunningham is an employee of the United States Air Force, Lackland Air Force Base, TX. Opinions expressed therein, unless otherwise specifically indicated, are those of the authors. They do not purport to express views of the Department of the Air Force or any other department or agency of the United States government. We would like to thank Mr. John Schoolfield for the statistical and computer analysis and Dr. Steve Montgomery for his valuable assistance in preparing this manuscript. Dr. Cunningham is a resident, Department of Endodontics, University of Texas Health Science Center at San Antonio, San Antonio, TX and Wilford Hall USAF Medical Center, Lackland Air Force Base, TX. Dr. Senia is professor and director, Advanced Education Program in Endodontics, University of Texas Health Science Center at San Antonio.
References 1. Hess W, Zurcher E. The anatomy of the root canals of the teeth of the permanent and deciduous dentitions. London: John Bale, Sons and Danielson Ltd., 1925. 2. Mueller AH. Morphology of root canals. J Am Dent Assoc 1936;23:16981706.
300
Cunningham and Senia
3. Green D. Morphology of the pulp cavity of the permanent teeth. Oral Surg 1955;8:743-59. 4. Schneider SW. A comparison of canal preparations in straight and curved root canals. Oral Surg 1971 ;32:271-5. 5. Pineda F, Kuttler Y. Mesiodistal and buccolingual roentgenographic investigation of 7,275 root canals. Oral Surg 1972;33:101-10. 6. Vertucci FJ. Root canal anatomy of the human permanent teeth. Oral Surg 1984;58:589-99. 7. Okumura T. Anatomy of the root canals. J Am Dent Assoc 1927;14:6326. 8. Fisher DE, Ingersoll N, Bucher JF. Anatomy of the pulpal canal: threedimensional visualization. J Endodon 1975;1:22-5. 9. Skidmore AE, Bjorndal AM. Root canal morphology of the human mandibular first molar. Oral Surg 1971 ;32:778-84. 10. Slowey RR. Root canal anatomy road map to successful endodontics. Dent Clin North Am 1979;23:555-73. 11. Mayo VC, Montgomery S, del Rio C. A computerized method for evaluating root canal morphology. J Endodon 1986;12:2-7.
Journal of Endodontics 12. Gullickson DC, Montgomery S. The study of root canal morphology using a digital image processing technique. J Endodon 1987;13:158-63. 13. Tang MP, Stock CJ. An in vitro method for comparing the effects of different root canal preparation techniques on the shape of curved root canals. Int Endod J 1989;22:49-54. 14. Davis SR, Brayton SM, Goldman M. The morphology of the prepared root canal: A study utilizing injectable silicone. Oral Surg 1972;34:642-8. 15. Weine FS, Pasiewicz RA, Rice RT. Canal configuration of the mandibular second molar using a clinically oriented in vitro method. J Endodon 1988; 14:207-13. 16. Weine FS. Endodontic therapy. 4th ed. St. Louis: CV Mosby, 1989:3145. 17. Roane JB, Sabala CL, Duncanson MG. The "balanced force" concept for instrumentation of curved canals. J Endodon 1985;11:203-11. 18. Wildey WL, Senia ES. A new root canal instrument and instrumentation technique: a preliminary report. Oral Surg 1989;67:198-207. 19. Leseberg DA, Montgomery S. The effects of Canal Master, Flex-R, and K-Flex instrumentation on root canal configuration. J Endodon 1991 ;17:59-65.
T h e W a y It W a s Bismarck, the German statesman, is known for his observation to the effect that "it is better that men do not know how their laws or their sausages are made." Sausage figured in another, less publicized event in his life. Virchow, the famous pathologist, was also a politician and served in the opposite party to Bismarck in the Reichstag. Virchow proposed a law requiring inspection of slaughtered hogs for trichinosis parasites. Bismarck opposed and, becoming enraged at Virchow's legislative tactics, challenged him to a duel. Knowing Bismarck to be an accomplished swordsman, and having the choice of weapon, Virchow chose a duel consisting of eating sausages, one of which would be identified as loaded with trichinosis ova. Since Bismarck would get to choose first--and given his scorn of trichinosis--he would be forced to choose the infected one or be ridiculed. Bismarck wisely withdrew his challenge. In 1878 a bill requiring meat inspection passed. Frank Suis
