The effects of Canal Master, Flex-R, and K-Flex instrumentation on root canal configuration

The effects of Canal Master, Flex-R, and K-Flex instrumentation on root canal configuration

0099-2399/91/1702-0059[$02.00/0 JOURNAL OF ENDODONTICS Copyright 9 1991 by The American Association of Endodontists Printed in U.S.A. VOL. 17, NO. 2,...

3MB Sizes 1 Downloads 23 Views

0099-2399/91/1702-0059[$02.00/0 JOURNAL OF ENDODONTICS Copyright 9 1991 by The American Association of Endodontists

Printed in U.S.A. VOL. 17, NO. 2, FEBRUARY1991

The Effects of Canal Master, Flex-R, and K-Flex Instrumentation on Root Canal Configuration Dennis A. Leseberg, DDS, and Steve Montgomery, DDS, FACD

The purpose of this study was to compare the Canal Master instrument and instrumentation technique (CM/CMI) to Flex-R files used with a "balanced force" technique (FR/BF) and K-Flex files used with a step-back filing technique (KF/SB). Thirty-six root canals in extracted human mandibular molars were equally divided into three groups. The roots were sectioned, then photographed and evaluated before and after instrumentation. Evaluation included canal shape, direction and extent of transportation, amount of dentin removed, and centering ability. Canal Master instrumentation and Flex-R/balanced force techniques both produced more round preparations than the K-Flex/step-back. All three techniques resulted in canal transportation, although to varying degrees. Flex-R/balanced forces and Canal Master instrumentation transported very little up to a #30, and Canal Master instrumentation transported significantly less after the # 4 5 instrument at mid-root. K-Flex/step-back removed the most dentin while Flex-R/balanced force removed the least in the apical sections and Canal Master instrumentation removed the least at mid-root. The ability of Canal Master instrumentation to keep instruments centered in the canals was significantly better than for K-Flex/step-back.

configuration and three specific characteristics: (a) No preparation was completely funnel-shaped from the orifice to the apex. (b) The narrowest portion of the canal was not at the apex, but near the middle portion of the curve. (c) Every file, whether precurved or straight, tended to straighten within the canal. The net results were canal preparations that were not completely debrided and were difficult to seal. Jungmann et al. (4) also concluded that neither filing nor reaming would predictably produce a round preparation in the apical portion. They and Vessey (2) found that the roundest preparations were obtained using a reaming action, whereas a filing action produced significant dimensional changes. In response to the difficulties encountered using conventional instruments, numerous techniques have been devised. Mullaney (5) and Goerig (6) recommended step-back procedures. Montgomery (7) advocated opening the cervical portion of the canal first to facilitate direct access to the apical area. He also suggested that new hand instruments may have to be designed specifically to clean and shape the remaining apical canal. In 1985, Roane et al. (8) introduced the "balanced force" concept for instrumentation of curved canals. To facilitate this technique, they designed a file with a modified parabolic tip (Flex-R). The first cutting flute at the tip was blunted so the instrument would transport root canals less than conventional files. More recently, Wildey and Senia (9) developed the Canal Master, a new instrument that varies considerably from conventional file design (Fig. 1). Features of the instrument are (a) a noncutting pilot (which is intended to guide the instrument through the canal), (b) a 1- to 2-mm long cutting head with a minimal cutting surface (which is to provide maximum control and limit indiscriminate cutting), and (c) a smooth, round, flexible shaft (which is to allow the instrument to accurately follow curved canals). The three features combined are said to greatly reduce canal transportation and to minimize ledging, zipping, and strip perforations. To date, no study has compared this new instrument to conventional files. The purpose of this study was to compare canal preparation using the Canal Master instrument and the technique recommended by Wildey and Senia (CM/CMI) with Flex-R files used with the balanced force technique (FR/BF), and K-Flex files used with a modified step-back filing technique (KF/SB). Instrumented canals were evaluated for shape, direction and extent of transportation, amount of dentin removed, and the ability of instruments to remain centered in the canal.

According to Ingle (1), careful canal preparation and obturation are the cornerstones of successful root canal therapy. The first criterion for success, an apical moisture-proof seal, is not possible unless the space to be filled is properly cleaned and shaped to receive the filling material. Since the fit of the guttapercha cone in the apical portion of the root canal and its proper seal in that area determine the adequacy of obturation, one of the primary goals of canal instrumentation should be the precise preparation of the apical portion of the canal (2). Until recently, the design of endodontic files and reamers did not facilitate the preparation of canals in such a manner, especially those having any degree of curvature. Weine et al. (3) found that regardless of the instrument or enlarging technique, all preparations showed the same general 59

60

Leseberg and Montgomery

Journal of Endodontics

in an Isomet low-speed saw (Buehler Ltd., Evanston, IL). The cuts were made perpendicular to the root surface on the two lines scribed prior to embedding the roots in resin. After cutting, the three root sections could be exactly repositioned within the locking jig (Fig. 2).

Photography

FIG 1. A, Canal Master rotary instrument; B, Canal Master hand instrument; C, Flex-R file; D, K-Flex file.

A stereomicroscope with camera attachment (Zeiss, Oberkochen, West Germany) was used to photograph the coronal surface of each section at x5 magnification. Blue inlay wax was placed into each canal and photographs were taken using Kodak Technical Pan print film (Eastman Kodak, Rochester, NY). Express Impression Material (3M. Minneapolis, MN) was placed onto a microscope slide and each section of the root was placed so that it made a custom form for that specific section (Fig. 3). This allowed the sections to be photographed in the same position for pre- and postinstrumentation photographs. Thus, each canal served as its own control.

M A T E R I A L S AND M E T H O D S

Canal Instrumentation

Specimen Selection

Each canal was then randomly assigned to one of three groups. In group A (KF/SB), the canals were instrumented with K-Flex files (Kerr, Romulus, MI) using a modified stepback filing technique ( l 2). The canals were prepared primarily using a rasping or filing motion. When necessary to advance the instrument, the filing technique was modified so that the files were "set" in dentin with a clockwise quarter-turn movement and then withdrawn with a filing motion. All instruments were precurved by using the buccal view radiographs as guides. The canals were enlarged to a size #30 file 0.5 mm short of CL followed by a sequential 1.0-mm step-back to a #45 file. The first postinstrumentation photographs were taken at this time and the root sections were repositioned in the locking jig. The canals were then instrumented to a #45 file 0.5 mm short of CL and the second postinstrumentation photographs were taken. Group B (FR/BF) was instrumented using Flex-R files (Union Broach, New York, NY) and a balanced force technique (13). The canals were enlarged to a #30 file 0.5 mm

Extracted human mandibular first and second molars with fully formed apices were used. They were stored in 10% formalin after extraction, Thirty-six mesial canals with demonstrable separate apical foramina were selected. The occlusal surfaces were flattened on a model trimmer to provide even reference points for instrumentation. The distal roots were removed and access into the pulp chamber was made with a 557 bur in a high-speed handpiece with water spray. Canal length (CL) was determined by placing a #10 file into each canal until it was just visible at the apical foramen. Radiographs were made with buccal and mesial views. Canal curvature was determined by using Schneider's method (10).

Preparation of Teeth Each selected root was scored with a longitudinal groove 0.5 mm in depth along the mesiobuccal line angle for later orientation. A drop of soft boxing wax was used to seal the two apical foramina. A wet cotton pellet was placed into the chamber and the access opening was closed with Cavit. A point 2.0 m m above the apical foramina was marked on each root. A second mark was made 1.0-mm apical to the point where the canal began to deviate from a straight line. This point was determined by overlaying the buccal view radiograph onto the buccal surface of the tooth. At these two levels a line was scribed circumferentially around each root. Each tooth was then embedded in clear casting resin (Chemco Resin Crafts, Dublin. CA) and sectioned using the methodology introduced by Bramante et al. (11). When set, the resin blocks were trimmed and orientation grooves were placed on two opposing faces. The blocks were lightly coated with Vaseline and a locking jig was made by embedding one-half of the resin block in die stone. When set, the stone was indexed, Vaseline, and the second half of the jig poured. The resin block was then removed from the stone locking jig and sectioned using a 0.15-mm diamond wafering blade

FIG 2. Locking jigs. The open jig shows the indexing grooves. The repositioned root sections are seen in the bottom half of the locking jig.

Vol. 17, No. 2, February 1991

Effects of Instrumentation

short of CL and the first postinstrumentation photographs taken. After repositioning the sections, the canals were enlarged to a #35 file 1.0 m m short of CL and then preparation continued using Gates Glidden (GG) drills. A #2 G G was used 4 mm short of CL, a #3 G G 6 mm short of CL, and a #4 GG 8 mm short of CL. The canals were then instrumented to a #45 file 1.5 m m short of CL and the second postinstrumentation photographs taken. This preparation corresponded to a #45 apical control zone recommended (13) for mesial canals of mandibular molars. In group C (CM/CMI), canals were instrumented using the new Canal Master rotary and hand instruments. Because of the radical design change of these instruments, instructions and a video tape were provided with the introductory kit. These instructions were followed in the study. The canals were instrumented using # 15 Flex-R files with a filing motion to CL. They were then prepared per the manufacturer's recommendation with #50 through #100 Canal Master rotary instruments to the level of the curve. This was followed by instrumentation of the apical portion of the canals with hand instruments. As in the previous groups, photographs were taken after enlargement with #30 and #45 instruments. New hand instruments were used in each canal, while each rotary instrument was used in six canals. The pulp chambers were flooded with 2.5% NaOC1 during all instrumentation and the canals were irrigated with 5.0 ml after each file size. The stone locking jig was coated with rubber base adhesive to seal the system and prevent loss of the NaOCI. The locking jigs were held securely together with a vice during instrumentation.

61

Evaluation Black and white prints (8 x 10 inches) were made of each photographed root section. Total magnification was x40. A canal, for example, which was 1 m m in diameter was enlarged on a photograph so that it was 40 m m in diameter. This permitted very precise and accurate measurement. Each print was coded so the instrumentation technique was unknown to the evaluator. Analysis was accomplished with the Sigma Scan Program, version 3.9 (Jandell Scientific, Corte Madera, CA), a model 2210 Digitizing Tablet (Numonics, Montgomeryville, PA), and a CompuAdd 286/16 computer (CompuAdd, Dallas, TX). A Xerox 1:1 transparency of each preinstrumentalion photograph was made and was overlaid onto the corresponding postinstrumentation photograph to ensure exact positioning on the Digitizing Tablet. A photograph ofa Fixott Everett Grid (Union Broach, New York, NY) with the same magnification as the other photographs was used to calibrate the computer. Canal shapes were subjectively classified as round or irregular. Direction of canal transportation was recorded as buccaI, lingual, mesial, or distal. The extent of canal transportation was determined by measuring the greatest distance between the periphery of the postinstrumented canal and the corresponding periphery of the preinstrumented canal which was overlaid on it. The methodology was similar to that recently described by McCann eta]. (14) except that their measurements were made with a microscopic reticle instead of a computer. The amount of dentin removed was calculated by comparing pre- and postinstrumented canal areas. Centering ratio, the ability of the instrument to remain centered in the original canal, was calculated using the formula X 1 - X 2 / Y ( 15). X 1 represented the maximum extent of canal movement in one direction; X2 was the movement in the opposite direction; and Y was the largest diameter of the final canal preparation. The data were evaluated statistically using analysis of variance. RESULTS

Canal Curvature There were no significant differences in root canal curvatures among the three groups. The mean curvature of all three groups was 24 degrees in the mesiodistal plane and 16 degrees in the buccolingual plane.

Canal Shape FiG 3. Express impression material on a glass slide made a custom form for each root section so they could be photographed in exactly the same position before and after instrumentation.

Of the 36 canals, only 5 in the apical sections and 2 in the mid-root sections were judged to be round prior to instru-

TABLE 1. Number of round canals Apical Section

Mid-root Section

Group A: K-Flex step-back B: Flex-R balanced force C: Canal Master instrumentation * Total number of canals in each g r o u p

Pre

#30

#45

Total

Pre

#30

#45

Total

n*

3 0 2

1 4 4

3 4 7

4 8 11

0 2 0

0 3 0

0 8 3

0 11 3

12 11 12

62

Leseberg and Montgomery

Journal of Endodontics

FIG 4. A, A mid-root section showing the canal prior to instrumentation. The distal wall was apparently transported (arrowhead) with a #10 Flex-R file while determining canal length (original magnification, x5). B, The same section as A after canal instrumentation to #45 with FR/BF. The preparation is fairly round but has transported and invaded sound dentin (original magnification •

mentation. In the apical sections more round canals were seen after instrumentation with FR/BF and CM/CMI than with KF/SB. In the mid-root sections no round canals were observed after KF/SB instrumentation (Table 1). With FR/ BF more canals were round but sometimes roundness was obtained at the expense of transportation and invasion of sound dentin (Fig. 4). Only three canals instrumented with CM/CMI were judged to be round. Direction and Extent of Canal Transportation Compared with the preinstrumented locations (controls), the postinstrumented canals were moved mesially in the apical sections and distally in the mid-root sections in all three groups, although to varying degrees. Canals instrumented to a size #30 with CM/CMI and FR/ BF transported significantly less than KF/SB in both apical and mid-root sections (Fig. 5) (Table 2). After instrumentation to a size #45 at mid-root, CM/CMI transported significantly less than the other two techniques (Fig. 6).

FIG 5. Apical sections showing canals instrumented to size #30 instruments. There was significantly less transportation with CM/CMI and FR/BF than with KF/SB (original magnification x5). A, CM/CMI. Very little transportation. B, FR/BF. Little transportation. C, KF/SB. Note some mesial transportation (arrowhead).

Amount of Dentin Removed After instrumenting to size #30, KF/SB removed more dentin in the apical areas than the FR/BF. After size #30, KF/SB removed significantly more dentin at mid-root than CM/CMI or FR/BF. When instrumentation was continued

Vol. 17, No. 2, February 1991

Effects of Instrumentation

63

Ability of the Instruments to Stay Centered Centering ratios were significantly better after instrumentation with CM/CMI and FR/BF than after KF/SB in three of the four sections evaluated (Table 4). The smaller the instrument size, the better the instruments remained centered in the canals. To verify accuracy and reproducibility of the computer analysis, five sections were chosen at random and reevaluated. The maximum error in linear measurements was 0.02 m m while the maximum error in area determination was 0.02 mm. 2 These errors were considered to be insignificant. The canals were reevaluated for shape by the same operator with 97.2% accuracy. One specimen in the FR/BF group was lost due to instrument separation, and no specimens were lost in the other two groups. DISCUSSION

FIG 6. Mid-root sections with canals instrumented to size #45 showing that CM/CMI transported significantly less and removed significantly less dentin than FR/BF or KF/SB (original magnification, • A, KF/SB. The canals were irregularly shaped and were transported distally. B, FR/BF. There is some transportation and extensive dentin removed. C, CM/CMi. Little transportation or dentin removed.

to a size #45, there was no difference in the amount of dentin removed from the apical region among the three techniques. In the mid-root sections, however, CM/CMI removed significantly less than the other two techniques (Fig. 6) (Table 3).

When a new root canal instrument is introduced, several characteristics need to be investigated including cutting efficiency, torsional and flexibility properties, debridement effectiveness, and effects on root canal configuration. Instruments can be compared by using them with the same or different instrumentation techniques. For this study, the effects of root canal configuration were evaluated when the instruments were used with the techniques for which they were originally designed. Files were designed to be used primarily with filing motions while the Flex-R and Canal Master instruments were designed to be used with rotary motions. The other characteristics of this new instrument await further investigations. Direct comparisons among the group results of this study must be done with caution because the optimal instrumentation size at working length for KF/SB is #30 (12) while the other two techniques advocate larger canal preparation (9, 13). In the apical sections, if the results of canal preparation to #30 with KF/SB are compared with enlargement to #45 with the other two techniques, the results are similar. In midroot sections, however, the results of CM/CMI enlarged to #45 were significantly better than those for FR/BF enlarged to #45 or KF/SB enlarged to #30; transport (0.18 versus 0.41 and 0.48 ram), dentin removed (0.15 versus 0.42 and 0.28 mm2), and centering ratio (0.21 versus 0.36 and 0.45). The experimental model (11) used in this study proved to be an excellent way to compare uninstrumented canals (controls) with their configuration following preparation. After two cuts with the 0.15-ram wafering blade, less than 0.5 m m of canal structure was removed and the die stone locking jigs maintained each section in such a solid and exact alignment that instrumenting the sections had the same feel as an intact tooth. Another advantage of this technique was the capability to evaluate the canals after every instrument size, Significant changes in canal configuration occurred between the #30 and 45 instruments, particularly in the mid-sections. CM/CMI transported significantly less and removed less dentin than the other techniques. This was probably due to the small smooth shaft of the instrument and the short area of cutting flutes. Despite the fact that all three techniques transported the original canals, there were considerable differences in the extent of transport among the groups. Both the FR/BF and

64

Leseberg and Montgomery

Journal of Endodontics TABLE 2. Mean length of transport (mm)

Apical Section (Mesial Direction)

Mid-root Section (Distal Direction)

Group 12 11 12

A: K-Flex step-back B: Flex-R balanced force C: Canal Master instrumentation

#30

SD

#45

SD

#30

SD

#45

SD

0.25* 0.11 0.17

+0.07 +0.03 9 +0.12 9

0.32 0.26 0.26

-+0.10 _+0.10 _+0.10

0.48t 0.11 0.12

-+0.17 _+0.04 _+0.06

0.55 0.41 0.181:

-+0.19 _+0.08 _+0.06

* A > B, C, p < 0.001. t A > B, C, p < 0.0001. ~:C < A, B; B < A , p < 0.0001.

TABLE 3. Mean area of dentin removed (postinstrumented area less preinstrumented area in mm 2)

Apical Section

Mid-root Section

Group 12 11 12

A: K-Flex step-back B: Flex-R balanced force C: Canal Master instrumentation

#30

SD

#45

SD

#30

SD

#45

SD

0.12 0.0510.09

_+0.04 _+0.02 +0.05

0.20 0.15 0.19

-+0.06 _+0.04 _+0.07

0.28* 0.06 0.07

-+0.08 _+0.02 _+0.05

0.37 0.42 0.15:[:

-+0.10 _+0.15 _+0.04

* B < A , p < 0.003. t A > B, C, p < 0.0001. ~ C < A, B, p < 0.0001.

TABLE 4. Mean centering ratio (Xl - X2/Y)*

Apical Section

Mid-root Section

Group 12 11 12

A: K-Flex step-back B: Flex-R balanced force C: Canal Master instrumentation

#30

SD

#45

SD

#30

SD

#45

SD

0.3810.19 0.22

_+0.16 +0.09 _+0.15

0.37 0.33 0.23

_+0.19 _+0.15 _+0.15

0.455 0.16 0.15

_+0.15 _+0.07 +0.10

0.47 0.36 0.21w

_+0.16 _+0.06 -+0.11

* The smaller the number, the better the instruments remained centered in the canals. t A > B,C, p < 0.005. :l: A > B, C, p < 0.0001. w B; B < A , p < 0.0001.

the CM/CMI produced significantly less transportation than did the KF/SB. Less apical transportation with Flex-R files and the balanced force technique corroborates the findings of Calhoun et al. (15) and Sabala et al. (16). A valid comparison with these studies is difficult, however, because Sabala et al. (16) used canals in resin blocks. Calhoun et al. (15) used only circumferential filing motions with K-Flex files and instrumented to #45 0.5 mm from CL with the balanced force technique. The current study used extracted human molars, used some rotary motions with the K-flex files, and instrumented to #30 0.5 mm from CL when preparing the #45 apical control zone. FR/BF and CM/CMI may provide better apical canal debridement because the inner curve of the canal wall is more likely to be contacted by the instruments. Debridement efficiency of these techniques should be studied. Green (17) found the average diameter of uninstrumented mesial canals of mandibular molars varied between 0.18 and 0.30 mm at a level 1.0 mm from the root apex. If an instrument remained centered, adequate debridement of the larger canals would likely require instrumentation to a #45 at that level. In this study, CM/CMI had the best overall centering ratio when canals were taken to a size #45. If small curved canals could be instrumented to a #45 with minimal canal transportation, then perhaps one could expect better debridement, irrigation, and obturation (18). Studies need to be done to evaluate these aspects. An interesting observation made during this study was the

indiscriminate removal of dentin with KF/SB, especially in the mid-root sections. CM/CMI removed less dentin, and remained centered to a greater degree, which should be desirable clinically. Despite efforts to direct the filing motion toward the bulkier mesial or "safety" zone, the K-Flex files used in a rasping manner resulted in one strip perforation and two near perforations. This study did not determine if this was because of the file design or the filing motions used. Canal roundness (Fig. 4B) was sometimes observed in midroot sections in association with transportation and removal of sound dentin. Canal roundness is of value if achieved by instrument centering so that the instruments touch all wall surfaces evenly, thus avoiding transportation. In mid-root sections enlarged to #45, with CM/CMI, the instruments remained centered in the canals significantly better than the other two techniques (Table 4). A surprising finding was the transportation observed in two mid-root sections after using only a #10 Flex-R file (Fig. 4A). No attempt was made to enlarge the canals; the instruments were carefully manipulated until they were visible at the apical foramina as a means of determining canal length (controls). Yet, these small and very flexible files were apparently capable of causing some transportation when used in two curved canals. There was, however, no transportation apparent in any of the apical sections. The degrees of curvature were measured for all canals in this study. Clinically the buccolingual curvature can cause

Vol. 17, No. 2, February 1991

problems if not taken into account. Experimentally, especially in resin blocks, the buccolingual curvature must be included in order to more accurately simulate clinical conditions. Although only one instrument separated in this study, both the Flex-R files and the Canal Master instruments were used in a reaming fashion and, when used in curved canals, the possibility of breakage must be considered. Studies need to be undertaken to determine the safety of these instruments when used in such a manner. CONCLUSIONS Conclusions drawn from this study are: 1. The experimental model proved to be an excellent way to compare instrumented canals with their original shapes. 2. Under conditions of this study, Flex-R files used with the "balanced force" technique and the Canal Master instruments and technique performed significantly better than KFlex files with a modified step-back filing method. 3. In the mid-root sections instrumented to size #45, Canal Master instrumentation transported significantly less, removed less dentin, and remained centered better than the other two techniques. We wish to thank Dr. Cheng Yuan for his statistical assistance. Dr. Walter Redmond for the use of his computer program, Phred Petersen for his assistance with the photography, and Dr. David Cames for his assistance with the manuscript. Dr. Leseberg is a graduate student, University of Texas Health Science Center at San Antonio, San Antonio, TX. Dr. Montgomery is professor, Graduate Division of Endodontics, University of Texas Health Science Center at San Antonio.

Effects of Instrumentation

65

References 1. Ingle JI. Endodontics. 3rd ed. Philadelphia: Lea & Febiger, 1985:37-8. 2. Vessey RA. The effect of filing versus reaming on the shape of the prepared root canal. Oral Surg 1969;27:543-7. 3. Weine FS, Kelly RF, Lio PJ. The effect of preparation procedures on original canal shape and on apical foramen shape. J Endodon 1975;1:255-62. 4. Jungmann eL, Uchin RA, Bucher JF. Effect of instrumentation on the shape of the root canal. J Endodon 1975;1:66-9. 5. Mullaney TP. Instrumentation of finely curved canals. Dent Clin North Am 1979;23:575-92. 6. Goerig AC, Michelich RJ, Schultz HH. Instrumentation of root canals in molars using the step-down technique. J Endodon 1982;8:550-4. 7. Montogomery S. Root canal wall thickness of mandibular molars after biomechanical preparation. J Endodon 1985;11:257-63. 8. Roane JB, Sabala eL, Duncanson MG. The "balanced force" concept for instrumentation of curved canals. J Endodon 1985;11:203-11. 9. Wildey WL, Senia ES. A new root canal instrument and instrumentation technique: a preliminary report. Oral Surg 1989;67:198-207. 10. Schneider WS. A comparison of canal preparations in straight and curved root canals. Oral Surg 1971 ;32:271-5. 11. Bramante CM, Berbert A, Borges AP. A methodology for evaluation of root canal instrumentation. J Endodon 1987;13:243-5. 12. Grossman LI, Oliet S, del Rio CE. Endodontic practice. 11th ed. Philadelphia: Lea & Febiger, 1988:202-10. 13. Roane JB, Benenati FW, Sabala CL, Biggs JT. Endodontics in a single visit. Lecture manual, University of Oklahoma Health Science Center, 1986:100-22. 14. McCann JT, Keller DL, LaBounty GL. Remaining dentin/cementum thickness after hand or ultrasonic instrumentation. J Endodon 1990;16:10913. 15. Calhoun G, Montgomery S. The effects of four instrumentation techniques on root canal shape. J Endodon 1988;14:273-77. 16. Sabala eL, Roane JB, Southard LZ. Instrumentation of curved canals using a modified tipped instrument: a comparison study. J Endodon 1988;14:59-64. 17. Green EN. Microscopic investigation of root canal diameters. J Am Dent Assoc 1958;57:636-44. 18. Southard DW, Oswald RJ, Natkin E. Instrumentation of curved molar root canals with the Roane technique. J Endodon 1987;13:479-89.