0099-2399/96/2211-0603503.
00/0
JOURNALOF ENDODONTICS
Copyright 1996 by The American Association of Endodontists
Analysis of Nickel-Titanium Versus Stainless Steel
Instrumentation by Means of Direct Digital Imaging
Chris L. Coleman, DDS, MS, Timothy A. Svec, DDS, MS, Monty R. Rieger, PhD, John A. Suchina, DDS,
MS, M. Michael Wang, BDS, MS, and Gerald N. Glickman, DDS, MS
This study compared step-back preparations in
curved canals using nickel-titanium (Ni-Ti) K-files
and stainless steel K-files. Forty canals in mesial
roots of mandibular molars were embedded in
casting resin and cross-sectioned at three levels: 1
to 2 mm from the apical foramen, middle of the
curve, and coronal. Direct digital computer images
were recorded before and after instrumentation.
Superimposition of the images combined with dig-
ital subtraction computer software allowed direct
measurement of area instrumented, distance of
transportation, and shape analysis. The computer
software calculated absolute center of gravity for
each image analyzed to get a full 360-degree inter-
pretation of the canal transport. Time of instrumen-
tation was recorded. Results showed Ni-Ti files to
cause significantly less transportation and remain
more centered at the apical level (p < 0.05). Area
removed by Ni-Ti and stainless steel files was not
significantly different (p < 0.05). Time of instrumen-
tation was not significantly different for Ni-Ti and
stainless steel instruments (p < 0.05). Cross-sec-
tional shape of the instrumented canal was not
significantly different (p < 0.05).
The goals of root canal instrumentation are complete debride-
ment of pulp tissue, removal of microbes and affected dentin,
and proper cleansing and shaping of the root canal space before
obturation (1). Finely curved root canals provide a dilemma for
the endodontist. Instrumentation in curved canals often results
in apical transportation, apical zipping, and an hour-glass shape.
Zipping and stripping are types of canal transportation in which
one area of the canal is overprepared in relation to other areas.
The stated goals of root canal instrumentation are not attained
in a transported canal.
Lira and Webber (2) speculated that a more flexible file would
produce less, or perhaps no, apical transport. Nickel-titanium (Ni-
603
Printed in U.S.A.
VOL. 22, NO. 11, NOVEMBER1996
-Ti) alloy has been shown to be exceptionally elastic, having a
lower bending moment and lower permanent set after torsion,
compared with similar gauge stainless steel (SS) (3). Walia et al.
(4) reported on the bending and torsional properties of nitinol root
canal files. Nitinol files had two to three times more elastic flex-
ibility and superior resistance to torsional fracture in clockwise and
counterclockwise torsion, when compared with similarly manufac-
tured SS files (4). Hand files fabricated from Ni-Ti alloy are now
available. They are claimed to be superior in curved canals. Two
recently published reports have compared SS and Ni-Ti files using
similar instrumentation techniques. Glosson el al. (5) instrumented
60 canals with either K-Flex files (SS), Mity files (Ni-Ti K-files),
NT sensor files, Canal Master U hand instruments, or Lightspeed
engine-driven instruments (5). The Bramante method of evaluation
was used. It was found that hand filing with Mity or K-flex files in
a "quarter turn/pull" technique caused more transportation, was
less centered, removed more dentin, and produced more nonround
preparations than the other canal preparation systems. The more
flexible Ni-Ti hand files were not found to improve significantly
centering ability, compared with SS files. Esposito and Cunning-
ham (6) compared instrumentation by Ni-Ti hand files, Ni-Timatic
instruments, and SS K-Flex files. Curved roots were radiographed
at various master apical file sizes and superimposed. It was con-
cluded that Ni-Ti hand files and Ni-Timatic instruments were
significantly more effective in maintaining the original canal path
of curved roots when apical sizes of 35, 40, and 45 were attained.
Canal deviation was not significantly different at apical sizes of 25
and 30.
The Bramante and Berbert (7) model for evaluation of root canal
instrumentation was introduced in 1987. Superimposed photos of
the preinstrumentation and postinstrumentation canal cross-sec-
tions enabled Bramante to make a direct comparative analysis. The
original canal shape and area were known, so subsequent changes
would be caused by instrumentation. Parameters such as area
instrumented, centering ratio, distance of transport, shape, and time
have been calculated and analyzed (8-11). Parametric data were
gathered for statistical analysis.
The purpose of this in vitro study was to compare instrumen-
tation by Ni-Ti and SS K-files using step-back enlargement in
curved canals. Direct digital images were captured and combined
with the Bramante technique for an accurate and reproducible
analysis.
604 Coleman e t al.
MATERIALS AND METHODS
Twenty freshly extracted mandibular first molars were selected.
The mesial canals of the molars had a mean curvature of at least 30
degrees, as determined by Lim and W e b b e r ' s (12) modification of
the Schneider method (13). Each canal had a separate orifice and
a separate apical foramen. Access openings in the teeth were made
with a #4 round bur in a high-speed handpiece, and both mesial
canals were explored for patency with a Kerr Pathfinder K-I (Kerr,
Romulus, MI). The cusps were flattened to provide more easily
reproducible reference points. The distal root was resected to
facilitate embedding of the mesial root. Working lengths were
established and recorded with a Kerr Pathfinder K-1 by passing the
file to where it was barely visible at the apical foramen and then
subtracting 1 ram. The mesial root was marked on one side with an
indelible blue marker at the apical, middle, and coronal "target
points" for sectioning. Tbe apical target point was chosen --2 mm
coronal to the apical foramen. The middle target point was chosen
at the most curved portion of the root in the middle one-third. The
coronal target point was chosen at the straight portion of the root
in the coronal third. The apical foramina were then sealed with
sticky wax. A cotton pellet was placed into the chamber and
Cavit-G (Premier Dental Products, Norristown, PA) was used to
seal the access. Each root was embedded in Wards Bioplastic
casting resin (Wards Natural Science, Rochester, NY) using a
custom-made machined brass "muffle" (9). A total of four muffles
were used. The muffle functioned to realign the sectioned block in
exact orientation for instrumentation. After polymerization, the
block was removed from the muffle and cut perpendicular to the
long axis of the canal at the predetermined target points, creating
four pieces. An lsomet low-speed saw (Buehler Ltd., Evanston, IL)
with a 0.15-thick diamond watering blade (Buehler Ltd., Lake
Bluff, IL) was used. The mesiobuccal canal of each section was
labeled with a blue reference mark. Red rope wax was placed in the
canals of each section to be imaged. Direct computer images of the
root at the coronalmost side of each cut were made on a CUE 2
Image Analysis System (Olympus Corp., Tokyo, Japan) with an
attached Navitar zoom 6000 lens (Meyer Instruments, Houston,
TX) and stored on Bernoulli disks (Iomega Corp., Roy, UT). A
custom-made glass reference grid was constructed so that the
preinstrumentation and postinstrumentation images were in exactly
the same position in relation to the camera lens. The lens of the
image analysis system was fixed, and magnification was set. The
system was calibrated to millimeters before each imaging session
for both X and Y axes using a stage micrometer slide (VWR
Scientific, Sugarland, TX). After imaging, the wax was removed
from the canal spaces and the sections were reoriented in their
respective brass muffles. The canals were divided into two groups:
group 1 = roots instrumented with SS K-files and group 2 = roots
instrumented with Mity (Ni-Ti) K-files. One canal in each root was
prepared using Ni-Ti K-files (Mity File, JS Dental, Ridgefield,
CT), and the other was prepared with SS K-files (J.S. Dental,
Ridgefield, CT). The files were precurved by the operator before
instrumentation. Mesiobuccal and mesiolingual canals were alter-
nated between the file types. One half of the mesiobuccal canals
were prepared with Ni-Ti K-files, and one-half were prepared with
SS K-files. No canal irrigant was used. File sizes 8 to 25 were used
to prepare one canal only and discarded; sizes 30 to 60 were used
to prepare four canals and discarded. Each canal preparation time
was recorded in minutes and seconds. Timing was begun when a
#10 file could be negotiated to the working length. Timing ended
when the canal was prepared to a master apical file size 25, flared
Journal of Endodontics
mm 2
0.4 I
0.35 -
0.3-
0.25 - I SS-Root
0.2-
0.15 - I NT-Root
0.1-
0.05 -
0-
Apical Middle Coronal
FiG 1. Area removed by instrumentation. Apical s e c t i o n - - g r o u p 1
(SS) mean 0.15 _+ 0.10 mm 2, group 2 (Ni-Ti) mean 0.10 _+ 0.07 mm 2.
Middle s e c t i o n - - g r o u p 1 (SS) mean 0.22 -+ 0.09 mm 2, group 2
(Ni-Ti) mean 0.20 + 0.10 mm 2. Coronal s e c t i o n - - g r o u p 1 (SS) 0.36
_+ 0.18 mm 2, group 2 (Ni-Ti) 0.30 _+ 0.15 mm 2. There were no
statistically significant differences between the groups at any level
(p < 0.05).
with a step-back technique to size 60 in 1-mm increments and a
D- 11 T spreader (Hu Friedy, Chicago, IL) fit loosely at 1 mm from
working length. Each block was then removed from its brass mold,
wax was reinserted into the canal, and postinstrumentation images
were recorded.
The superimposed preinstrumentation and postinstrumentation
stored images were analyzed using the computer to measure
changes in areas and distances. Center of gravity, length of canal
transportation, shape factor number, and area of dentin removed
were assessed and calculated at each of the three levels of each
prepared canal. Time to instrument each canal fully was recorded.
Statistical analysis of the results was performed to find significant
differences between the groups. For all statistical tests used, sig-
nificance was set at p < 0.05. Time, area, shape, and distance of
transportation were analyzed using one-way analysis of variance to
find if there were differences between the groups. Student-New-
man-Keuls procedure was used to locate those differences.
RESULTS
Area
At the apical section, there was no significant difference be-
tween the groups (Fig. 1). Group 1 (SS) removed a mean of 0.15
0.10 mm 2, whereas group 2 (Ni-Ti) removed a mean of 0.10
0.07 mm 2. At the middle section, there was no significant
difference between the groups. Group 1 (SS) removed a mean of
0.22 0.09 m m 2, whereas group 2 (Ni-Ti) removed a mean of
0.20 0.10 mm 2. At the coronal section, there was no significant
difference between the groups. Group 1 (SS) removed a mean of
0.36 0.18 m m 2, whereas group 2 (Ni-Ti) removed a mean of
0.30 0.15 mm 2.
Distance of Transport/Centering Ability
At the apical section, there was a significant difference between
the groups (p < 0.05) (Fig. 2). Group 2 (Ni-Ti) was significantly
more centered than group I (SS). Group 1 had a mean distance of
0.24 +_ 0.14 mm, whereas group 2 (Ni-Ti) had a mean distance of
0.13 _+ 0.07 mm. At the middle section, there was no significant
difference between the groups. Group l (SS) had a mean distance
of 0.18 + 0.10 ram, whereas group 2 (Ni-Ti) had a mean distance
of 0.12 + 0.05 mm. At the coronal section, there was a significant
Vol. 22, No. 11, November 1996 Direct Digital Imaging 605
mm seconds
0.25
600
0.2
0.15 400 SS-Root
SS-Root I
[ ] NT-RootJ
0.1 [] NT-RootJ 200
0.05
0
0
FIG 4. Time of instrumentation. Group 1 (SS)--mean 343.4 + 78.5 s,
Apical Middle Coronal group 2 (Ni-Ti)--mean 415.4 _+ 127.5 s. There was not a statistically
FIG 2. Change in center of gravity (distance from preoperation to significant difference between the groups (p < 0.05).
postoperation). Apical section--group 1 (SS) mean 0.24 + 0.14 mm,
group 2 (Ni-Ti) mean 0.13 +_ 0.07 mm. Middle section--group 1 (SS)
of 5 min and 43 s _+ 1 rain and 18 s, whereas group 2 (Ni-Ti) took
mean 0.18 0.10 mm, group 2 (Ni-Ti) mean 0.12 _+ 0.05 mm.
a mean time of 6 rain and 55 s -+ 2 rain and 7 s.
Coronal section--group 1 (SS) mean 0.20 _+ 0.09 mm, group 2
(Ni-Ti) mean 0.13 + 0.07 mm. At the apical and coronal sections,
group 2 (Ni-Ti) remained significantly more centered than group 1
DISCUSSION
(SS) (p < 0.05).

Cleaning and shaping procedures have been emphasized more

Shape Factor
Number
0.7 I
0.65
SS-Root I
0.6
[ ] NT-RootJ
0.55
0.5
Apical Middle Coronal
FIG 3. Shape of the instrumented canal (4rr areaJperimeter2).
Apical section--group 1 (SS) mean 0.58 _+ 0.11, group 2 (Ni-Ti)
mean 0.59 _+ 0.15. Middle section--group 1 (SS) mean 0.61 _+ 0.16,
group 2 (Ni-Ti) mean 0.66 0.14. Coronal section--group 1 (SS)
mean 0.61 0.12, group 2 (Ni-Ti) mean 0.64 _+ 0.11. There were no
statistically significant differences between the groups at any level
(p < 0.05).
difference between the groups (p < 0.05). Group 2 (Ni-Ti) was
significantly more centered than group 1 (SS). Group 1 had a mean
distance of 0.20 _+ 0.09 mm, whereas group 2 (Ni-Ti) had a mean
distance of 0.13 _+ 0.07 ram.
Shape Analysis
Shape data were quantified by the computer as a shape factor
number. The formula for the shape factor number was ((4~r)
(areaJperimeter2)), wherein product one was a perfect circle, and
the product of 0 was a straight line. There was no significant
difference between the groups for postoperative shape factor num-
ber within each level of sectioning (Fig. 3). However, there was a
significant difference (p < 0.05) in the shape factor number when
comparing the apical versus the middle level and when comparing
the apical versus the coronal level. The canal was significantly less
circular in cross-section apically, compared with middle or coro-
nal.
Time
There was no significant difference in the time it took to
instrument the canals fully (Fig. 4). Group 1 (SS) took a mean time
and more in recent years. An instrument that is shown to enhance
the ability of the practitioner to maintain original canal curvature
and form while preparing the root canal space for obturation would
be an advantage. Ni-Ti files transported the canal significantly less
and remained more centered at the apical and coronal levels of
sectioning than SS files did (p < 0.05) (Fig. 2). At the middle
section, Ni-Ti files transported less than SS files; however, the
difference was not significant (p < 0.05). The greater flexibility of
the Ni-Ti file was found to improve significantly the centering of
the prepared canal. This was despite the inability to precurve the
Ni-Ti files to the degree that the SS instruments were. These
findings are different from Glosson et al. (5), who concluded that
the Ni-Ti Mity file did not significantly improve centering ability.
At the apical section, they found no significant difference in
centering ratio. There was a difference in the way transport and
centering were determined in the two studies. All previously pub-
lished Bramante methodology studies used the formula X1 -
X2/Y to produce a centering ratio (5, 8, 10, 11). Y is the greatest
diameter of the canal preparation. XI represents the maximum
extent of canal movement in one direction, and X2 represents the
movement in the opposite direction along the line Y. X1 then
becomes the distance of transport for that canal. This yields a
centering ratio that is valid for that one line of diameter only.
In this study, a center of gravity was recorded by the computer
as an X,Y single pixel coordinate for both preoperative and post-
operative canals. The center of gravity is the coordinate chosen by
the computer as the point equidistant from all points along the
perimeter of the canal cross-section. The distance that the center of
gravity moves from preinstrumentation to postinstrumentation co-
ordinates relates directly to the ability of the instrument to remain
centered in the original canal. The distance between these two
points is 0 if the canal preparation remains exactly centered.
Calculating the center of gravity takes into account 360 degrees of
the instrumentation of the canal for that particular section. Center
of gravity measurements are superior to centering ratio measure-
ments reported in previous Bramante analyses that measures only
one line of diameter through the postoperative canal.
The distance of transport data presented is measured from the
X,Y coordinate of the preoperative center of gravity to the X,Y
coordinate of the postoperative center of gravity in millimeters.
Previous research has used the greatest line of diameter of the
postoperative canal minus the preoperative distance along the same
line to deduce distance of transport (5, 8-11). Therefore, the
606 Coleman et al.
A mean shape factor numbers created by SS files were 0.61, 0.62, and
FIG 5. Preinstrumentation canal space (white) and additional area
removed by instrumentation (dark blue), apical section, extracted
molar root. SS files were used in the canal to the left; Ni-Ti files were
used in the canat to the right.
distance of transport reported in previous Bramante studies is the
maximum length of transport for each canal. The distance of
transport reported in this study is an average value, because all
canal radii are taken into account. Although the distance of trans-
port is measured in this manner, the value becomes a direct
measurement of how well centered a preparation has remained.
Area of dentin removed varied according to the type of file used.
At each level of sectioning, the trend was for the SS file (group 1)
to remove a greater area of dentin than the Ni-Ti file did (group 2)
in the paired canal (Fig. 1). However, this difference was not
significant at any of the measured levels (p < 0.05). At the apical
level in extracted roots, an area of 0.15 mm 2 was instrumented by
SS files and 0.10 mm 2 by Ni-Ti files. These values are similar to
what others have reported at the apical level of sectioning. Campos
and del Rio (9) removed 0.07 mm 2 of dentin at the apical section
Journal of Endodontics
using SS files, whereas Leseberg and Montgomery (10) removed
0.12 mm 2 at this level. Glosson et al. (5) removed 0.10 mm 2 of
dentin at the apical section, with both #30 SS K-Flex files and #30
Ni-Ti Mity files.
At the middle section, a mean area of 0.22 mm 2 of dentin was
removed by SS files and 0.20 mm 2 by Ni-Ti files. Other studies
reported quite similar values. Campos and del Rio (9) reported a
value of 0.2 mm 2 using SS files at this level. Leseberg and
Montgomery (10) recorded a value of 0.28 mm 2 at the middle level
of sectioning. Glosson et al. removed 0.17 mm 2 dentin, with #30
SS K-Flex files and 0.12 mm 2 dentin with #30 Ni-Ti Mity files.
Shape of the instrumented canal was quantified by the computer
as a shape factor number. No published report in the endodontic
literature to date has quantified shape in this manner. This allows
parametric statistical analysis of the shape factor number.
At each level of sectioning, the Ni-T,i file created a more circular
canal in cross-section. This difference was not significant at any
level. The mean shape factor numbers created by Ni-Ti files were
0.64, 0.66, and 0.59 for coronal, middle, and apical levels. The
0.58 for coronal middle and apical levels (Fig. 3). The trend for
each file type was to create a canal that was less circular in
cross-section as the apex was neared.
The time for instrumentation was greater for Ni-Ti filed canals,
compared with SS; however, the difference was not statistically
significant (Fig. 4). The increased flexibility of Ni-Ti files may
have led to some of the increase in time. However, the inability to
precurve the Ni-Ti files to the degree that the SS files were may
have been more important. The Bramante technique creates a
"kerf," which is the space left by the saw blade after sectioning.
The kerf feels similar to a ledge in clinical situations. The Ni-Ti
files were less efficient at passing the kerf at the middle and apical
levels of sectioning.
Strikingly similar preparation times were found in Glosson et
al.'s (5) study. However, in their analysis, the more flexible Mity
file prepared the canal significantly more quickly (536 -+ 98 s) than
the K-Flex file (651 +_ 119 s). This was despite their observation
that SS files were more easily negotiated into small, constricted
canals.
No canal irrigant was used during instrumentation. Most studies
of canal instrumentation have used saline as an irrigant (4, 14, 15).
Saline was used in those studies to carry debris away and prevent
canal blockage. It was found in a pilot study that the debris created
from instrumentation tended to be carried to the kerf area. There-
fore, debris did not pack into the canal and prevent it from being
instrumented. Frequent recapitulation with a #10 file was also
performed to help prevent blockage. Briseno and Sonnabend (16)
used no irrigant in their study of instrumentation by nine file types
in plastic blocks. They observed little debris along the root canals
if the files were removed and the flutes cleaned periodically. The
combination of recapitulation with small files, cleaning of the file
flutes, and the kerf acting as a reservoir for debris allowed the
canals to be instrumented without irrigant.
The modifications made fiom previously published Bramante
method studies worked well. Direct digital computer images were
captured and stored preoperatively and postoperatively. Measure-
ments were made via the computer software from the images.
Previous studies using the Bramante technique to evaluate canal
preparation used photographs that were then digitized to measure
changes from instrumentation (5). By recording direct images, the
Olympus CUE-2 image analysis system allowed the removal of
Vol. 22, No. 11, November 1996
any potential errors in photography and digitization of photo-
graphs.
Direct digital imaging of the cross-sections allowed storage,
enhancement, and superimposition of the preinstrumentation and
postinstrumentation canal. Enhancement involved creation of
black and white binary images, colorization (Fig. 5), dilation,
erosion, and edge detection. Using an enhanced image with a
defined edge, objects could be measured for area, shape, and center
of gravity from the specific object parameters readout.
This study showed that Ni-Ti files may offer some advantage
over SS files. Ni-Ti Mity files remained more centered in the
original canal, whereas enlarging the canal in a satisfactory manner
compared with SS files. The shape of the instrumented canal was
similar for both file types. However, the same increased flexibility
that may allow the Ni-Ti file to remain more centered also may
make it less efficient at passing ledges, canal obstructions, or filing
calcified canals. An ideal and practical technique would be to use
SS files to open the canal initially to sizes 15 or 20 before using
Ni-Ti instruments. In conclusion, based on the results of this study,
there seems to be a bright future for Ni-Ti files in endodontics.
We express our sincere gratitude to Dr. Gerald Pinero for his assistance in
the preparation of this manuscript. We would also like to thank J. S. Dental for
donation of all of the files used in this project.
Dr. Coleman is currently practicing in Webster, TX, and is a part-time
clinical assistant professor of endodontics; Drs. Svec, Suchina, and Wang are
associate professors, Division of Endodontics; and Dr. Rieger is an associate
professor, Section of Oral Biomaterials, University at Texas-Houston, Health
Sciences Center, Dental Branch, Houston, TX. Dr. Glickman is an associate
professor and director of endodontics, Department of Cariology, Restorative
Sciences, and Endodontics, University of Michigan, School of Dentistry, Ann
Arbor, MI. Address requests for reprints to Dr. Chris L. Coleman, University of
Texas-Houston, Health Sciences Center, Dental Branch, Department of Sto-
matology/Division of Endodontics, Room 202, 6516 John Freeman Avenue,
Houston, TX 77030-3402.
You Might Be Interested
The gloom merchants s e e m f u l l y in c o n t r o l . A r e c e n t a r t i c l e in t h e B r i t i s h M e d i c a l
example, is e n t i t l e d " P r o b a b i l i t y of adverse
theme that just because everything is all r i g h t d o e s n ' t
a file lately? That just means that the probability
Phooey! How about an article on "Probability
Direct Digital Imaging 607
References
1. Shilder H. Cleaning and shaping the root canal. Dent Clin North Am
1974;18:269-96.
2. Lim KC, Webber J. The validity of simulated root canals for the inves-
tigation of the prepared root canal shape. Int Endod J 1985;18:240-6.
3. Andreason GF, Morrow RE. Laboratory and clinical analysis of nitinol
wire. Am J Orthodont 1978;73:142-51.
4. Walia H, Brantley WA, Gerstein H. An initial investigation of the bending
and torsional properties of nitinol root canal files. J Endodon 1988;14:346-51.
5. Glosson CR, Hailer RH, Dove SB, del Rio CE. A comparison of root
canal preparations using Ni-Ti hand, Ni-Ti engine-driven, and K-flex endodon-
tic instruments. J Endodon 1995;21:146-51.
6. Esposito PT, Cunningham CJ. A comparison of canal preparation with
nickel-titanium and stainless steel instruments. J Endodon 1995;21:173-6.
7. Bramante CM, Berbert A. A methodology for evaluation of root canal
instrumentation. J Endodon 1987;13:243-5.
8. Calhoun G, Montgomery S. The effects of four instrumentation tech-
niques on root canal shape. J Endodon 1988;14:273-7.
9. Campos JM, del Rio C. Comparison of mechanical and standard hand
instrumentation techniques in curved root canals. J Endodon 1990;16:230-4.
10. 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.
11. Roig-Cayon M, Brau-Aguade E, Canalda-Sahli C, Moreno-Aguado V.
A comparison of molar root canal preparations using Flexofile, Canal Master
U, and Heliapical instruments. J Endodon 1994;20:495-9.
12. Lim KC, Webber J. The effect of root canal preparation on the shape
of the root canal. Int Endod J 1985;18:233-9.
13. Schneider SW. A comparison of canal preparation in straight and
curved root canals. Oral Surg 1971;32:271-5.
14. Alodeh MA, Doller R, Dummer PH. Shaping of simulated root canals in
resin blocks using the step-back technique with K-files manipulated in a
simple in and out filing motion. Int Endod J 1989;22:107-17.
15. El Deeb ME, Boraas JC. The effect of different files on the preparation
shape of severely curved canals. Int Endod J 1985;18:1-7.
16. Briseno BM, Sonnabend E. The influence of different root canal in-
struments on root canal preparation: an in vitro study. Int Endod J 1991;24:
15-23.
Journal ( B M J 31 1 : 6 1 9 ) , f o r
events that have not yet occurred." It g o e s o n t o e x p a n d on the
mean things can't go wrong...(e.g., Haven't broken
that you will soon is i n c r e a s i n g ) .
of being surprised by joy."
William Cornelius